Netherlands Journal of Critical Care - NJCC 2021-2 Vol... · 2021. 3. 19. · Netherlands Journal of Critical Care 76 NETH J CRIT CARE- VOLUME 29 - NO 2 - MARCH 2021 Submitted February

Netherlands Journal of Critical CareBi-monthly journal of the Dutch Society of Intensive Care

Volume 29 - No 2 - March 2021

SPECIAL REPORTOne year of COVID-19 in the Netherlands a Dutch narrativeOn behalf of the Dutch COVID-19 Research Consortium

ORIGINAL ARTICLEREMAP-CAP: delivering research in the pandemicJ.L.G. Haitsma Mulier, E.R. Rademaker, M.J.M. Bonten, L.P.G. Derde

SPECIAL REPORTSharing is caring: how COVID-19 led to large-scale collaboration for icudata.nlP. Elbers, P. Thoral, T. Dam, L. Fleuren on behalf of the Dutch ICU Data Warehouse Collaborators

Netherlands Journal of Critical Care

NETHERLANDS JOURNAL OF CRITICAL CARE

EXECUTIVE EDITORIAL BOARDD.W. Donker, editor in chiefI. van Stijn, managing editorH. Dupuis, language editorD. van Dijk, associate editorM.M.J. van Eijk, associate editorN. Kusadasi, associate editorC.L. Meuwese, associate editor

[email protected]

C O P Y R I G H TNetherlands Journal of Critical CareISSN: 1569-3511NVIC p/a Domus MedicaP.O. Box 2124, 3500 GC UtrechtT: +31-(0)30 - 760 74 44

© 2021 NVIC. All rights reserved. Except as outlined below, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the publisher. Permission may be sought directly from NVIC.

DERIVATIVE WORKSSubscribers may reproduce tables of contents or prepare lists of articles including abstracts for internal circulation within their institutions. Permission of the publisher is required for resale or distribution outside the institution. Permission of the publisher is also required for all other derivative works, including compilations and translations.

ELEC TRONIC STORAGEPermission of the publisher is required to store or use electronically any material contained in this journal, including any article or part of an article.

SUBSCRIPTIONSAn annual subscription to the Netherlands Journal of Critical Care consists of 6 issues. Issues within Europe are sent by standard mail and outside Europe by air delivery. Cancellations should be made, in writing, at least two months before the end of the year.The annual subscription fee for the Netherlands is 5 175, for Europe 5 295, for the rest of the world 5 390. Subscriptions are accepted on a prepaid basis only and are entered on a calendar year basis.Please transfer the fee to ABNAMRO NL55 ABNA 0524 5618 93, swift-code: ABNANL2A. Do not forget to mention the complete address for delivery of the Journal.

CLAIMSClaims for missing issues should be made within two months of the date of dispatch. Missing issues will be mailed without charge. Issues claimed beyond the two-month limit must be prepaid at back copy rates.

ADVER TISING/ BUSINESS CONTAC TSFor advertising, please contact Eventex. EventexSint Bernulphusstraat 13D6861GS OosterbeekT: +31 (0)26 – 333 75 77E: [email protected]

Netherlands Journal of Critical Care is indexed in:

NETH J CRIT CARE - VOLUME 29 - NO 2 - MARCH 2021 75

CONTENTS

EDITORIAL76 COVID-19, a remarkable year M.M.J. van Eijk

SPECIAL REPORT78 One year of COVID-19 in the Netherlands - a Dutch narrative On behalf of the Dutch COVID-19 Research Consortium

SPECIAL REPORT85 Sharing is caring: how COVID-19 led to large-scale collaboration for icudata.nl P. Elbers, P. Thoral, T. Dam, L. Fleuren on behalf of the Dutch ICU Data Warehouse Collaborators

ORIGINAL ARTICLE87 REMAP-CAP: delivering research in the pandemic J.L.G. Haitsma Mulier, E.R. Rademaker, M.J.M. Bonten, L.P.G. Derde

ORIGINAL ARTICLE91 Effect of high-dose methylprednisolone in mechanically ventilated ICU patients

with COVID-19: a retrospective observational study D. Hoogeveen, T.J.P. Ketels, T.J. Wilbers, A.C. Strang

ORIGINAL ARTICLE96 Neurological complications in COVID-19 patients admitted to a general ICU in the Netherlands E.A. Krijnen, S.P.P. Matthijs, D.H.T. Tjan

CASE REPORT102 Endotracheal tube obstruction in patients diagnosed with COVID-19 E.A. van Boven, S. van Duin, H.H. Ponssen

CASE REPORT104 Cerebral microbleeds in a COVID-19 patient B. Maatman, E. Aronica, S.D. Roosendaal, D.C. Velseboer

CASE REPORT107 ECMO as rescue for COVID-19 related ARDS: the pros and cons J.J. van der Heijden, C.L. Meuwese, S.A. Braithwaite

LETTER TO THE EDITOR113 Fraction of inspired oxygen (FiO2) to modulate respiratory drive in a mechanically

ventilated patient A. Osinski, J. van Rosmalen, R.P.J. Baltussen

LETTER TO THE EDITOR116 When to start spontaneous breathing in mechanically ventilated COVID-19 patients?

Oxygenation index and PaO2/FiO2 ratio can help I. Bakker, P.H. Egbers, E.C. Boerma, C. Bethlehem

120 NVIC conference and course agenda121 Editorial board 121 International advisory board122 Information for authors


76 NETH J CRIT CARE - VOLUME 29 - NO 2 - MARCH 2021

Submitted February 2021; Accepted February 2021

E D I T O R I A L

COVID-19, a remarkable year

M.M.J. van EijkIntensive Care Centre, University Medical Centre Utrecht, Utrecht University, Utrecht, the Netherlands

Correspondence

M.M.J. van Eijk - [email protected]

Keywords - COVID-19; ventilation; research

The past year will forever be remembered as the year of the outbreak of a novel coronavirus. From the moment the first Dutch patient was admitted to hospital, healthcare providers received a crash course in recognising and treating these patients. Intensivists, ICU nurses and all others working with the most critically ill of these patients have since gained enormous insight into this novel virus, its treatment, dangers and pitfalls. We were confronted with a serious new medical condition at a scale not seen before; during the so-called ‘first-wave’, the number of ICU beds in the Netherlands was almost doubled. Dutch ICUs managed this high influx of patients by working and cooperating with each other, and with the help of numerous other departments and specialities.

To commemorate this past year, the editorial board of the Netherlands Journal of Critical Care (NJCC) has chosen to publish a special edition of the Journal, and asked the Dutch ICU community for their input. This resulted in a variety of articles, covering all aspects of the past year. The national ICU evaluation foundation (Nationale Intensive Care Evaluatie (NICE)) has presented all the data collected during the past year and up to the deadline of this edition of the NJCC; their article forms a comprehensive overview of one year COVID-19 in data.[1]

Several colleagues, from all corners of the country, have submitted interesting papers on this subject. Hoogeveen et al. report on a retrospective study conducted in a large ICU cohort, investigating the use of high-dose methylprednisolone in COVID-19 patients.[2] Although it is only observational, it emphasises the positive effect of high-dose methylprednisolone and that this, as we now know, beneficial treatment was already administered in our ICUs during the ‘first wave’.

As all those working in the ICU in the past year are very much aware, ventilation of the COVID-19 patient can be very challenging, especially concerning the transition from a

controlled mode of ventilation to an assisted mode. Bakker et al. propose the oxygenation index in order to estimate when the patient is ready for this transition and to prevent self-inflicted lung injury (SILI).[3] Osinski et al. report on a case in which the respiratory drive is based primarily on arterial oxygen, postulating that this, in some cases, may be a better way of regulating the respiratory drive.[4] Although both proposed methods are interesting and possibly useful, the prevention of SILI remains very difficult in COVID-19 patients and poses a major challenge for the future.

Van Boven et al. share a very common but potentially fatal complication.[5] Although not uncommon during non-COVID-19 times, with the increased workload and the number of patients in prone position this can be a serious problem in the ICU, and warrants vigilance of all those working with ventilated ICU patients. To avoid the above-mentioned problems (SILI, tube obstructions et cetera) extracorporeal membrane oxygenation (ECMO) seems a ‘simple’ solution. Van der Heijden et al. present two cases showing that, with very strict indications, ECMO treatment can be lifesaving, but unfortunately not always.[6] Krijnen et al. and Maatman et al. both describe the neurological effects and complications of a COVID-19 infection in ICU patients, both focussing on vascular problems.[7,8]

Although the clinical efforts of the Dutch ICUs have been in a very bright spotlight this past year, the scientific efforts are no less impressive. Both Elbers et al. (Dutch ICU Data Warehouse) and Haitsma Mulier (REMAP-CAP) report on their unique studies.[9,10] The results of these studies will follow soon, but both show that robust scientific evidence can only be gained by cooperation. Information on participation in these studies can be found via the links provided in the articles.

Hopefully, the measures imposed on society and the beneficial effect of the vaccination campaign will soon slow down the



COVID-19, a remarkable year

pandemic so that we all, including the NJCC, can return to a relatively normal situation. Until then, we hope you enjoy this overview of all the aspects of Dutch ICU work during this remarkable year and we would like to thank all authors and reviewers who, despite the ongoing COVID-19 infections, found time for the NJCC.

References

1. Dutch COVID-19 Research Consortium. One year of COVID-19 in the Netherlands - a Dutch narrative. Neth J Crit Care. 2021;29:78-84.

2. Hoogeveen D, Ketels TJP, Wilbers TJ, Strang AC. Effect of high-dose methylprednisolone in mechanically ventilated ICU patients with COVID-19: a retrospective observational study. Neth J Crit Care. 2021;29:91-95.

3. Bakker I, Egbers PH, Boerma EC, Bethlehem C. When to start spontaneous breathing in mechanically ventilated COVID-19 patients? Oxygenation index and PaO2/FiO2 ratio can help. Neth J Crit Care. 2021;29:116-117.

4. Osinski A, van Rosmalen J, Baltussen RPJ. Fraction of inspired oxygen (FiO2) to modulate respiratory drive in a mechanically ventilated patient. Neth J Crit Care. 2021;29:113-115.

5. Van Boven EA, van Duin S, Ponssen HH. Endotracheal tube obstruction in patients diagnosed with COVID-19. Neth J Crit Care. 2021;29:102-103.

6. Van der Heijden JJ, Meuwese CL, Braithwaite SA. ECMO as rescue for COVID-19 related ARDS: the pros and cons. Neth J Crit Care. 2021;29:107-112.

7. Krijnen EA, Matthijs SPP, Tjan DHT. Neurological complications in COVID-19 patients admitted to a general ICU in the Netherlands. Neth J Crit Care. 2021;29:96-101.

8. Maatman B, Aronica E, Roosendaal SD, Velseboer DC. Cerebral microbleeds in a COVID-19 patient. Neth J Crit Care. 2021;29:104-106.

9. Elbers P, Thoral P, Dam T, Fleuren L. Sharing is caring: how COVID-19 led to large-scale collaboration for icudata.nl. Neth J Crit Care. 2021;29:85-86

10. Haitsma Mulier JLG, Rademaker ER, Bonten MJM, Derde LPG. REMAP-CAP: delivering research in the pandemic. Neth J Crit Care. 2021;29:87-90.




S P E C I A L R E P O R T

One year of COVID-19 in the Netherlands - a Dutch narrative

On behalf of the Dutch COVID-19 Research ConsortiumDutch National Intensive Care Evaluation, Department of Clinical Informatics, Amsterdam University Medical Center, University of

Amsterdam, the Netherlands

See appendix for a list of the members of the Dutch COVID-19 Research Consortium

Correspondence

[email protected]

Keywords - COVID-19, mortality, risk factors, ICU, outcome

‘It is easy to be wise after the event’(Sir Arthur Conan Doyle)

A novel zoonotic corona virus hits the Netherlands In December 2019, China reported a cluster of cases of pneumonia in the city of Wuhan (Hubei Provence). In the beginning the biological agent was not known, but in January 2020 a corona virus was identified as the pathogen causing this pneumonia. Because of its resemblance to a previous corona virus epidemic, this new virus was called ‘severe acute respiratory coronavirus 2 (SARS-CoV-2) and the disease was called ‘corona virus disease 2019’ (COVID-19). By that time, however, SARS-CoV-2 was no longer contained to China and the first confirmed patients with COVID-19 were seen outside China in Thailand, Japan and South Korea. Despite the fact that the Chinese authorities locked down the entire city of Wuhan, the worldwide spread of COVID-19 was rapid and unstoppable.

On 14 February 2020, the French reported the first European death to COVID-19. An 80-year-old Chinese tourist died in a hospital in Paris. Two days later a major surge of patients was noticed in Italy. After more than 150 patients fell ill with COVID-19 in the Lombardy area, the Italians locked down ten North-Italian towns. A 56-year-old man, who had returned from Italy, was the first Dutch patient to be confirmed of having COVID-19 on 27 February 2020. In the days thereafter several patients were diagnosed with COVID-19 and all had recently visited Italy. One week later the first Dutch patient with confirmed COVID-19 died as a consequence of the disease. By that time, over 320 patients were diagnosed with COVID-19 and the Netherlands went into ‘intelligent’ lockdown. This, however, could not prevent the surge of infected patients. By the end of March over 10,000 Dutch patients were confirmed of having COVID-19 but since testing was mostly only done for the very sick and healthcare workers, the actual number of infections was likely much higher.[1]

The surge of critically ill patients in the NetherlandsMost people (about 80%) recover from the disease without needing special treatment and for the majority – especially children and young adults – illness due to COVID-19 is generally mild. However, for some people it can cause serious illness. Around 1 in every 10 people who are infected with COVID-19 develop difficulty in breathing and require hospital care. Particularly patients who are aged over 60 years, and people who have underlying medical conditions such as diabetes, heart disease, respiratory disease or hypertension, are among those who are at greater risk.[2] Of the patients needing hospital treatment roughly one fifth eventually needed ICU care.[3] In the Netherlands, 0.35% of the infected people needed ICU admission.[4] This resulted in a huge demand for ICU beds that were not available in the Netherlands in April 2020 (figure 1).

In the pre-COVID era the Netherlands had roughly 1050-1100 ICU beds available, although some were not ‘fully operational’ due to shortages of ICU personnel. Usually 80% of these ICU beds are occupied by critically ill patients with non-COVID reasons for ICU admission, resulting in only 400 available beds to allocate to COVID-19 patients (figure 2). This huge shortage of ICU beds and personnel necessitated the Dutch hospitals to suspend all non-essential treatments and redirect all the workforce to the COVID-19 wards and ICUs. Additional ICU beds were created at new locations, such as post-anaesthesia care units, recovery units, operating theatres and newly created ICUs. The first wave of critically ill COVID-19 patients was managed with the help of many non-ICU trained personnel and necessitated stricter selection of patients than prior to the pandemic.



One year of COVID-19 in the Netherlands

Despite the fact that non-essential medical care was downscaled, the majority of ICUs continued with regular care that could not be postponed. Figure 2 shows that the average 800 occupied ICU beds decreased to roughly 500 non-COVID beds during the height of the first wave.

Demographics of Dutch COVID-19 patientsIn the beginning of the pandemic, the impression was that especially obese, older men were overrepresented in the critically ill patients. Now, 12 months into the COVID-19 pandemic this has shown to be rather consistent over time. The mean age of

Figure 1. Patients with proven COVID admitted to the Dutch ICUs

Figure 2. Number of patients with and without COVID-19 admitted to Dutch ICUs




COVID patients is 63.8 (SD 11.5) years, which is only minimally older than the mean age of non-COVID, critically ill patients (60.3 years, SD 17.2 years). In comparison with historical patients with severe acute respiratory illnesses (so-called SARI, such as pneumonia, influenza, etc.), the COVID-19 patients appear slightly younger (mean age of SARI patients was 66.3 years), and they have less comorbidities (table 1).

The number of comorbidities in COVID-19 patients increases with increasing age and is depicted in figure 3. When the first wave subsided and ICU crowding was less imminent the, often subconscious, admission selection of patients became less strict. This is reflected in the fact that more albeit younger patients with comorbidities were admitted between the first and the second wave (data not shown).

Outcome of COVID-19 patientsThe mortality of critically ill COVID-19 patients differs enormously between publications. This is the result of different inclusion criteria, differences in healthcare settings, ICU crowding and in outcome definitions.[5-7] Especially healthcare systems and ICUs that were stretched to the maximum of their capacity (and beyond) potentially delivered suboptimal care, which could have resulted in higher mortality rates.[8] For example, in the Italian area of Lombardy the ICU mortality was 48.7% and hospital mortality was estimated to be 53.4%.[6] It has been suggested that a large part of these deaths could have been prevented if ICUs had been less congested and accessibility to hospitals and ICUs had been more immediate.[9] The same might be true in the Netherlands. We, at least, see that some areas in the Netherlands had higher crude mortality rates than other regions (figure 4).

Table 1. Comparison between COVID-19 patients and historical admission with severe acute respiratory infection (SARI), e.g. influenza pneumonia, community-acquired pneumonia, etc.

COVID-19 patients N (%)

SARI patients N (%)

COPD/respiratory insufficient 661 (12.4) 7549 (38.1)

Renal failure 236 (4.4) 1717 (8.7)

Cirrhosis 19 (0.4) 239 (1.2)

Cardiovascular insufficiency 77 (1.4) 760 (3.8)

Malignancy/haematological malignancy

142 (2.7) 1960 (9.9)

Immunological insufficiency 459 (8.6) 3807 (19.2)

Diabetes 1211 (22.7) 4006 (20.2)

Mechanical ventilation at ICU admission

2106 (39.5) 7941 (40)

Mechanical ventilation within the first 24 h of ICU admission

3648 (68.5) 11153 (56.2)

Total 7380 (100) 19835 (100)

Figure 3. The number of comorbidities increases with age: no comorbidities (green), one comorbidity (blue) or two or more comorbidities (red)




Apart from access to immediate ICU care, several other variables were associated with better or worse outcome of critically ill COVID-19 patients. Obviously, older patients have a higher risk of mortality than younger patients and the odds for death increase significantly for patients aged 60 years and above (table 2). Particularly the patients 70 years and older have a 10 to 25 times increased risk of hospital mortality.

Another striking feature is that women have a better outcome than men (odds ratio 0.68, confidence interval 0.58-0.8). This has been noted in various other publications and meta-analysis.[10,11] In a univariate analysis obesity was also associated with ICU admission.[12] However, despite the initial thought that obesity was associated with a worse outcome, the survival of overweight patients appeared to be better (e.g. patients with a BMI 30-35 had an odds ratio of 0.68 in comparison with patients with a BMI 18.5-25). This might also be explained by the younger age of those patients but it does contradict previous findings in COVID-19 patients.[12] However, previous research in general Dutch ICU patients showed a U-shaped survival curve in relation to BMI.[13] Whether this is a reflection of the normal outcome of ICU patients or the result of confounding needs to be elucidated. The diminished mortality in (slightly) overweight COVID-19 patients might have resulted from admission bias (only accepting the younger, healthiest obese patient onto the ICU). Again, like in every other disease, patients with comorbidities have a higher mortality than patients without such mortalities.

International publications suggest that some groups of patients (patients from non-Dutch descendants) might have a worse course of COVID-19 and are overrepresented within the critically ill subpopulation.[14] In the Netherlands, it is prohibited to collect these data due to privacy restraints. Whether this represents confounding or has a pathophysiological origin remains to be elucidated.

Survival trends over timeAfter the first wave, when the capacity of ICUs was stretched to the limit, the numbers of critically ill COVID-19 patients declined. This meant that ICUs that had diluted their personnel (sometimes to a nurse-to-patient ratio of >1:4) could return to normal care. This was paralleled by better treatment modalities of COVID-19 patients. During the first wave various randomised controlled trials proved that some hypothetical treatments were ineffective (e.g. chloroquine, lopinavir/ritonavir) while others were effective in critically ill patients (e.g. dexamethasone and potentially the prevention of thromboembolism by increased anticoagulation strategies).[15,16] Combined, these results have led to better outcomes over time in COVID-19 patients on hospital wards. The case-mix of COVID-19 patients that ends up on the ICU in the second wave is different than that in the first wave. This results in variations in the Kaplan-Meier curves over time (figure 5).

Figure 4. Kaplan-Meier survival curves of crude survival after ICU admission. Only 5 out of the 12 Dutch ICU regions are depicted here to illustrate the huge ranges in crude 60-day survival.




Outcome beyond survivalWe know that patients who survived COVID-19 have serious sequelae. Many people have not been able to perform their daily activities or resume their work at all or at least for a long time. More than half of the survivors report decreased functional capacity and many describe persistent functional limitations.[17-19] We know that a substantial number of the general ICU survivors suffer from cognitive, psychological and physical impairments, which combined are called the ‘post-intensive care syndrome’, although in the general ICU population it appears that some patients were already experiencing these symptoms prior to admission.[20] However, such information and long-term follow-up of surviving COVID-19 patients are not yet available, but studies are under way.[21,22] It is quite conceivable that overcrowding, restrictions in access to early mobilisation and physiotherapy as well as specific treatment modalities for COVID-19 (e.g. prolonged muscle paralysis and mechanical ventilation in prone position for ARDS) contribute to worse outcomes. Results for the Dutch ICU population could be obtained by combining databases of discharged critically ill COVID-19 patients with the databases of insurance companies (vektis.nl). Unfortunately, strict privacy regulations prevent swift analyses which could improve patient care for the next wave.

Continuing SARI surveillance in the futureSince the 2009 pandemic of influenza A/H1N1, the World Health Organisation (WHO) advised to establish Severe Acute Respiratory Illnesses (SARI) surveillance in each country to enable earlier detection of potential epidemics and pandemics.[23] In the Netherlands several initiatives aimed to change their quality registries into near ‘real-time’ surveillances for such severe infections. While the APACHE IV coding, which is used by the NICE registry, does not allow a very granular registration of pulmonary infections, it can be used to approximate the incidences of severe pulmonary infections.[24,25] The admissions to the ICU for pulmonary infections parallel the incidence curves of influenza-like illnesses throughout the years. However, in order to function as a proper SARI sentinel, the NICE database should be near real-time and needs to be able to combine ICU data with other registries to improve the microbiological diagnoses. Near real-time data processing via Fast Healthcare Interoperability Resource (FHIR) messages between hospitals and the central NICE registry will be implemented in 2021. Combining databases necessitates the use of a unique identifier for patients in all registries. However, the use of a (pseudonymised) social security number (BSN) is not allowed. In the COVID-19 pandemic, however, patients are frequently transferred between hospitals and regions. The Dutch government has allowed the use of the BSN specifically for this pandemic as this was the only option to properly follow up these patients. It is somewhat ironic that we needed another pandemic to improve a surveillance that was intended to detect this pandemic in the first place. A proposal for

Table 2. The odds for hospital mortality for various risk factors

NAS item COVID-19 survivors N (%)

COVID-19 deceasedN (%)

Odds ratio (95% CI)

COVID-19 still in hospital N

Age groups

<40 178 (92.7) 14 (7.3) Reference 9

40-45 101 (91) 10 (9) Reference 2

45-50 238 (89.5) 28 (10.5) Reference 11

50-55 383 (87.8) 53 (12.2) 1.38 (0.92-2.06) 19

55-60 520 (84) 99 (16) 1.9 (1.33-2.7) 35

60-65 596 (75.3) 195 (24.7) 3.26 (2.35-4.51) 50

65-70 549 (65.7) 287 (34.3) 5.21 (3.8-7.15) 48

70-75 557 (56.8) 423 (43.2) 7.57 (5.55-10.31) 54

75-80 290 (46.9) 328 (53.1) 11.27 (8.15-15.57) 27

80-85 65 (36.3) 114 (63.7) 17.47 (11.53-26.47) 7

>85 6 (26.1) 17 (73.9) 28.23 (10.67-74.65) 0

Gender

Men 2419 (67.2) 1181 (32.8) Reference 204

Woman 1075 (73.5) 338 (26.5) 0.74 (0.65-0.85) 59

BMI groups

<18.5 14 (53.8) 12 (46.2) 1.58 (0.72-3.44) 2

18.5-25 694 (65.2) 371 (34.8) Reference 61

25-30 1446 (68.6) 663 (31.4) 0.84 (0.73-0.98) 95

30-35 824 (73.5) 297 (26.5) 0.66 (0.56-0.79) 58

35-40 293 (69.8) 127 (30.2) 0.8 (0.63-1.01) 29

>40 135 (75) 45 (25) 0.61 (0.43-0.88) 11

Comorbidities

COPD & respiratory insufficiency (no)

3123 (70.3) 1317 (29.7) Reference 225

COPD & respiratory insufficiency (yes)

371 (59.6) 252 (40.4) 1.61 (1.36-1.91) 38

Renal failure (no) 3405 (670.3) 1441 (29.7) Reference 244

Renal failure (yes) 89 (41) 128 (59) 3.4 (258-4.48) 19

Cardiovascular insufficiency (no)

3464 (69.4) 1528 (30.6) 29

Cardiovascular insufficiency (yes)

30 (42.3) 41 (57.7) 3.1 (1.93-4.98) 6

Malignancy (no) 3433 (69.7) 1495 (30.3) Reference 256

Malignancy (yes) 61 (45.2) 74 (54.8) 2.79 (1.97-3.93) 7

Immunological insufficiency (no)

3247 (70.1) 1385 (29.9) Reference 235

Immunological insufficiency (yes)

247 (57.3) 184 (42.7) 1.75 (1.43-2.14) 28

Diabetes (no) 2783 (71.3) 1122 (28.7) Reference 210

Diabetes (yes) 711 (61.4) 447 (38.6) 1.56 (1.36-1.79) 53

Comorbidities

None 2261 (75.1) 750 (24.9) Reference 149

1 990 (63.8) 562 (36.2) 1.7 (1.49-1.94) 80

>1 243 (48.6) 257 (51.4) 3.18 (2.61-3.86) 34

If the 95% confidence interval (95% CI) does not contain 1.0 then the risk factor is statistically significantly associated with hospital mortality in comparison to the reference group. The definitions of the risk factors are derived from the APACHE IV model for severity of illness and can be found in the data dictionary available at www.stichting-nice.nl




legislative change is under construction in which quality registries such as NICE are better secured in the General Data Protection Regulations and the Dutch Healthcare Quality, Complaints and Disputes Act (in Dutch the WKKG), including the use of a pseudonymised unique identifier.

ConclusionsIn times of disasters, such as the current SARS-CoV-2 pandemic, when hospitals are flooded with patients and healthcare systems are in dire need of information, it is invaluable to have an up-and-running registry that swiftly adapts to the current need. In the Netherlands the NICE database was able to rise to the occasion because of the continuous effort of intensivists of all Dutch ICUs. Near real-time information and combining ICU data with other registries will improve early warning and give more insights into quality improvement possibilities in future pandemics.

DisclosuresAll authors declare no conflict of interest. No funding or financial support was received.

References

1 Coronacrisis in Nederland. (Accessed 24th January, 2021, at https://nl.wikipedia.org/wiki/Coronacrisis_in_Nederland.).

2 Risk groups and COVID-19. Rijksinstituut voor Volksgezondheid and Milieu, 2021. (Accessed 24th January 2021 at https://www.rivm.nl/en/novel-coronavirus-covid-19/risk-groups).

3 National Intensive Care Evaluation website. (Accessed 24th January 2021 at https://stichting-nice.nl/covid-19-op-de-ic.jsp).

4 Corona en nu - wat zijn tegenwoordig je kansen. Volkskrant. (Accessed 24th January 2021 at https://www.volkskrant.nl/columns-opinie/corona-en-nu-wat-zijn-tegenwoordig-je-kansen~bb5c00b5c).

5 Rosenthal N, Cao Z, Gundrum J, Sianis J, Safo S. Risk Factors Associated With In-Hospital Mortality in a US National Sample of Patients With COVID-19. JAMA Netw Open. 2020;3:e2029058.

6 Grasselli G, Greco M, Zanella A, et al. Risk Factors Associated With Mortality Among Patients With COVID-19 in Intensive Care Units in Lombardy, Italy. JAMA Intern Med. 2020;180:1345-55.

7 Ferrando C, Suarez-Sipmann F, et al. Clinical features, ventilatory management, and outcome of ARDS caused by COVID-19 are similar to other causes of ARDS. Intensive Care Med. 2020;46:2200-11.

8 Asch DA, Sheils NE, Islam MN, et al. Variation in US Hospital Mortality Rates for Patients Admitted With COVID-19 During the First 6 Months of the Pandemic. JAMA Intern Med. 2020 Dec 22:e208193.

9 Ciminelli G, Garcia-Mandicó S. How Healthcare Congestion Increases Covid-19 Mortality: Evidence from Lombardy, Italy* medRxiv. 2020.10.27.20221085.

10 Peckham H, de Gruijter NM, Raine C, et al. Male sex identified by global COVID-19 meta-analysis as a risk factor for death and ITU admission. Nat Commun. 2020;11:6317.

11 Schiffer VMMM, Janssen EBNJ, van Bussel BCT, et al. The ‘sex gap’ in COVID-19 trials: a scoping review. EClinicalMedicine. 2020;100652.

12 Hussain A, Mahawar K, Xia Z, Yang W, El-Hasani S. Obesity and mortality of COVID-19. Meta-analysis. Obes Res Clin Pract. 2020;14:295-300.

13 Pickkers P, de Keizer N, Dusseljee J, Weerheijm D, van der Hoeven JG, Peek N. Body mass index is associated with hospital mortality in critically ill patients: an observational cohort study. Crit Care Med. 2013;41:1878-83.

14 Intensive Care National Audit and Research Centre (ICNARC) report on covid. (Accessed 24th January 2021 at https://www.icnarc.org/Our-Audit/Audits/Cmp/Reports).

15 Horby P, Lim WS, Emberson JR, et al. Dexamethasone in Hospitalized Patients with Covid-19 - Preliminary Report. N Engl J Med. 2020: doi: 10.1056/NEJMoa2021436.

16 Lopes RD, Fanaroff AC. Anticoagulation in COVID-19: It Is Time for High-Quality Evidence. J Am Coll Cardiol. 2020;76:1827-9.

17 Taboada M, Moreno E, Cariñena A, et al. Quality of life, functional status, and persistent symptoms after intensive care of COVID-19 patients. Br J Anaesth. 2020: https://doi.org/10.1016/j.bja.2020.12.007.

18 Garrigues E, Janvier P, Kherabi Y, et al. Post-discharge persistent symptoms and health-related quality of life after hospitalization for COVID-19. J Infect. 2020;81:e4-e6.

19 Hosey MM, Needham DM. Survivorship after COVID-19 ICU stay. Nat Rev Dis Primers. 2020;6:60.

20 Geense WW, van den Boogaard M, Peters MAA, et al. Physical, Mental, and Cognitive Health Status of ICU Survivors Before ICU Admission: A Cohort Study. Crit Care Med. 2020;48:1271-9.

21 Tas J, van Gassel RJJ, Heines SJH, et al. Serial measurements in COVID-19-induced acute respiratory disease to unravel heterogeneity of the disease course: design of the Maastricht Intensive Care COVID cohort (MaastrICCht). BMJ Open. 2020;10:e040175.

22 Prescott HC, Girard TD. Recovery From Severe COVID-19: Leveraging the Lessons of Survival From Sepsis. JAMA. 2020;324:739-40.

Figure 5. Kaplan-Meier survival curves of crude survival after ICU admission in two month periods.



23 WHO Global epidemiological surveillance standards for influenza [March 13, 2018]. (Accessed 24th January 2021 at https://www.who.int/influenza/resources/documents/influenza_surveillance_manual/en/ (2013)).

24 Koetsier A, van Asten L, Dijkstra F, et al. Do intensive care data on respiratory infections reflect influenza epidemics? PloS One. 2013;8:e83854

25 van Asten L, Luna Pinzon A, de Lange DW, et al. Estimating severity of influenza epidemics from severe acute respiratory infections (SARI) in intensive care units. Crit Care. 2018;22:351

Appendix The Dutch COVID-19 Research Consortium collaborators

D.P. VerbiestL.F. te VeldeE.M. van DrielT. RijpstraP.H.J. ElbersA.P.I HouwinkL. GeorgievaE. VerweijR.M. de JongF.M. van IerselT.J.J. KoningE. Rengers

N. KusadasiM.L. ErkampR. van den BergC.J.M.G. JacobsJ.L. EpkerA.A. RijkeboerM.T. de BruinP. SpronkA. DraismaD.J. VersluisA.E. van den BergM. Vrolijk-de Mos

J.A. LensR.V. PruijstenH. KieftJ. RozendaalF. NooteboomD.P. Boer I.T.A. JanssenL. van GulikM.P. KoetsierV.M. SilderhuisR.M. SchnabelI. Drogt

W. de RuijterR.J. BosmanT. FrenzelL.C. Urlings-StropA. DijkhuizenI.Z. HenéA.R. de MeijerJ.W.M. HoltkampN. PostmaA.J.G.H. BindelsR.M.J. WesselinkS. Brinkman

E.R. van Slobbe-BijlsmaP.H.J. van der VoortB.J.W. EikemansD.J. Mehagnoul-SchipperD. GommersJ.G. LutisanM. HoeksemaM.G.W. BarnasB. Festen-SpanjerM. van LieshoutN.C. GrittersM. van Tellingen

G.B. BrunnekreefJ. VandeputteT.P.J. DormansM.E. HoogendoornM. de GraaffD. MoolenaarA.C. ReidingaJ.J. SpijkstraR. de WaalD.A. DongelmansD.W. de LangeN.F. de Keizer


For the most recent data, visit our website

stichting-nice.nl



Submitted January 2021; Accepted January 2021

S P E C I A L R E P O R T

Sharing is caring: how COVID-19 led to large-scale collaboration for icudata.nl

P. Elbers, P. Thoral, T. Dam, L. Fleuren on behalf of the Dutch ICU Data Warehouse CollaboratorsLaboratory for Critical Care Computational Intelligence, Department of Intensive Care Medicine, Amsterdam Medical Data Science, Amsterdam

UMC, Vrije Universiteit Amsterdam and University of Amsterdam, Amsterdam, the Netherlands

Correspondence

P. Elbers – [email protected]

Keywords - data sharing, collaboration, machine learning

For many intensivists worldwide, the pandemic will have created long-lasting memories. Some of them grim, such as massively overwhelmed ICUs, the dehumanising appearance of seemingly interchangeable and mostly proned patients, and healthcare professionals hidden behind personal protective equipment. Some of them energising, including surprising patient recoveries as well as the unprecedented praise and recognition for intensivists and intensive care medicine from both the media and society at large.

For the Laboratory for Critical Care Computational Intelligence[1] at Amsterdam UMC, the pandemic proved to be nothing short of a rollercoaster ride. And despite the many challenges imposed upon our profession by the pandemic, this ride was largely fuelled by excitement, in particular the rapidly expanding enthusiasm for large-scale data sharing and collaboration between Dutch ICUs.

Obviously, our story started long before the COVID-19 crisis. Intensive care medicine is a natural habitat for data science as large amounts of data are routinely collected during intensive care treatment, such as those from devices for monitoring and life support. Our laboratory was created with the primary aim to unite clinical and data science expertise to use these data to improve the care and treatment of future critically ill patients. We do so by developing and validating models, integrating these into clinical decision support tools to be used at the bedside and evaluating their effect on outcomes relevant for critically ill patients.

Three of the most prominent results from our philosophy are AmsterdamUMCdb, the first freely available European ICU database under the European Society of Intensive Care Medicine / Society of Critical Care Medicine joint data sharing initiative,[2] bedside decision support for personalised antibiotic dosing[3,4] and bedside decision support for preventing untimely patient discharge from intensive care units (ICUs).[5]

Because of these contributions to the field, our lab had the infrastructure and knowledge base to readily facilitate large-scale data sharing when the pandemic hit the Netherlands. Specifically, our expertise could facilitate sharing of high-frequency device data and most other clinical information from the electronic health record (EHR), with the goal to generate insights from ICU patient data as the pandemic was unfolding. These data were thought to reflect the large variation in COVID-19 related clinical practice resulting from the limited and rapidly evolving COVID-19 evidence base and possibly the large variation in patient characteristics and outcomes between centres. These variations may be leveraged by advanced statistics and machine learning to determine optimal individual patient management.

Right from the start of the project[6] we experienced an unprecedented momentum. Data protection officers immediately offered help to provide a legal framework for responsible data sharing. Within days, our medical ethics committee approved our protocols. Data sharing agreements were drafted ensuring equal possibilities for data access for all participating ICUs. All hospitals in the Netherlands with an ICU were approached and documentation was reviewed locally before permission to participate was granted. With full support of the Dutch Society for Intensive Care (NVIC) and their research network RCCnet, 66 out of 81 ICUs confirmed their participation within weeks.

Template Structured Query Language (SQL) queries were developed to automatically extract EHR data for each of the major EHR systems used in the Netherlands: MetaVision, ChipSoft and Epic. Collected data cover the entire ICU stay and include demographics, data from devices for vital signs monitoring and life support, data on administered medication, laboratory results and data entered by the treatment team including clinical observations. All data were pseudonymised in the delivering hospitals.

An extraction, transformation and load process was designed to combine raw data from the different EHR systems. All parameters from the collaborating ICUs were manually reviewed and mapped



Sharing is caring

to a common parameter ontology. Subsequently, a software data pipeline converted all parameter units as needed, filtered out data entry errors, calculated derived parameters and merged data into the data warehouse. Data quality control was a continuous process with internal validity checks from the providing hospitals and validation checkpoints throughout the software pipeline. Covering entire ICU admissions with highly granular data, the Dutch Data Warehouse is the largest COVID-19 dataset to date. It is a natural habitat for advanced statistics and machine learning, effectively extending the opportunities for high-frequency big data analysis provided by large ICU datasets such as MIMIC[5] and AmsterdamUMCdb[6] to the COVID-19 domain. So far, data from 23 ICUs on 1633 patients treated between March and October 2020 have been processed and added to the Dutch Data Warehouse, now containing over 120 million data points mapped to a common ontology of 875 parameter names.

Of course, data sharing among ICUs is far from new. In fact, intensivists in the Netherlands have been doing so for at least 25 years in the context of the National Intensive Care Evaluation (NICE). Their important contribution to the field should be applauded. However, their efforts are primarily focussed on benchmarking and quality improvement at the level of the ICU. The primary focus of the data warehouse is to improve the quality of care, treatment and outcome at the level of the individual patient, by leveraging high-frequency information during the entire ICU treatment.

While the first insights from the COVID-19 data are now being published and disseminated by webinars, we did discover that combining these data is a much slower process than we initially anticipated. This is partly related to the extensive data processing and quality control steps as described above, but most importantly to the fact that every Dutch ICU stores their data differently, even if they use the same type of EHR system.

Therefore, we are very excited that the NVIC, strongly supported by RCCnet, has now initiated a similar large-scale collaboration by Dutch ICUs to engage in large-scale data sharing on all critically ill patients. Again, this should help understand the timing and combination of treatments that may lead to better outcomes in a specific ICU patient. This collaboration is coordinated by our laboratory at Amsterdam UMC. There are strong ties with the NICE foundation, known for their experience in analysing data for ICU benchmarking. Machine learning partner Pacmed will use their expertise to combine and analyse the data. Zorgverzekeraars Nederland, uniting all major health insurance companies, will support the initiative with 2 million euros for the next five years.

This new collaboration between Dutch ICUs is called icudata.nl[7] and is expected to give rise to the Dutch ICU Data Warehouse (figure 1). We are thrilled that it was immediately received with great enthusiasm among a large majority of ICUs of all sizes, including those from

university medical centres, large teaching hospitals as well as smaller community hospitals. Their enthusiasm could make the Dutch Data Warehouse the largest of its kind worldwide.

It is our intention to extend the transparent collaboration as fuelled by the COVID pandemic to icudata.nl. We want to ensure equal possibilities for data access for all participating ICUs. The process to access the data for bona fide research and quality purposes to improve the care and treatment of future critically ill patients should be as easy and as little bureaucratic as possible. On the other hand, a governance structure with the participating hospitals, NVIC, NICE, representatives from patients, and other stakeholders is currently being designed. This structure should again ensure that data will be shared responsibly and in full compliance with all relevant laws and regulations. Participating ICUs will receive frequent updates and progress may also be monitored through www.icudata.nl.This is our story on how a severe crisis can give rise to a unique opportunity. Let’s keep up the momentum and join us if you have not done so already!


References

1. The Laboratory for Critical Care Computational Intelligence. https://icudata.nl/index-lccci.html.

2. Thoral P. AmsterdamUMCdb, the first freely accessible European ICU database under the ESICM/SCCM Joint Data Sharing Initiative [Internet]. Available from: https://www.amsterdammedicaldatascience.nl/#amsterdamumcdb

3. Roggeveen LF, Fleuren LM, Guo T, et al. Right Dose Right Now: bedside data-driven personalized antibiotic dosing in severe sepsis and septic shock—rationale and design of a multicenter randomized controlled superiority trial. Trials. BioMed Central. 2019;20:1-13.

4. Roggeveen LF, Guo T, Driessen RH, et al. Right Dose, Right Now: Development of AutoKinetics for Real Time Model Informed Precision Antibiotic Dosing Decision Support at the Bedside of Critically Ill Patients. Front Pharmacol. 2020;11:646.

5. Thoral PJ, Fornasa M, de Bruin DP, et al. Developing a Machine Learning prediction model for bedside decision support by predicting readmission or death following discharge from the Intensive Care unit. researchsquare.com; 2020; Available from: https://www.researchsquare.com/article/rs-12522/latest.pdf

6. CovidPredict [Internet]. Available from: https://covidpredict.org/7. The Dutch ICU Data Warehouse. https://icudata.nl/

Figure 1. Dutch ICUs will start collaborating by sharing large amounts of routinely collected data to improve the quality of care and treatment strategies for future critically ill patients




O R I G I N A L A R T I C L E

REMAP-CAP: delivering research in the pandemic

J.L.G. Haitsma Mulier1,2, E.R. Rademaker1,2, M.J.M. Bonten2,3, L.P.G. Derde1,2

1Department of Intensive Care, 2Julius Centre for Health Sciences and Primary Care, 3Department of Medical Microbiology, University

Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands

Correspondence

J.L.G. Haitsma Mulier – [email protected]

Keywords - REMAP-CAP, randomised, embedded, adaptive, platform, trial, research, pandemic

AbstractThe Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) pandemic is an unprecedented global health crisis. For hospitalised patients with Coronavirus Disease 2019 (COVID-19) mortality and morbidity is high. A case fatality rate of 49% for critically ill patients was reported in early studies. We urgently need effective treatments for these patients. In past pandemics, the clinical research response has largely failed. During the Influenza A(H1N1) pandemic, no randomised trials delivered results. Traditional randomised trials are not well suited for research in pandemics. They are robust, but lack the flexibility to adapt to changing circumstances and only investigate a single treatment against a control arm. Additionally, sample size calculations are almost impossible in new diseases. Adaptive platform trials, specifically REMAP-CAP, help overcome the challenges of pandemic research. We describe the key design principles of adaptive platform trials, the design of REMAP-CAP, and how this trial has delivered important results that contribute to the treatment of hospitalised and critically ill patients with COVID-19.

IntroductionSince the start of the current pandemic, over 100 million cases of COVID-19 and over 2.2 million deaths globally have been reported.[1] The disease presents as severe pneumonia with typical imaging abnormalities, hypoxaemia, an exaggerated immune response and thromboembolic complications. Initially, with no treatment available, mortality rates for critically ill patients were around 49%.[2] Although vaccination is commonly seen as the largest contributing factor to tackling this global crisis, rapid identification of effective treatments is crucial as morbidity and mortality from COVID-19 is significant, with a large proportion of hospital and intensive care unit (ICU) beds occupied by these patients. We describe how adaptive platform trials, specifically REMAP-CAP, can help overcome challenges and find effective treatments for patients.

Research during pandemicsIn past pandemics, the clinical research response has largely failed.[3] The influenza A(H1N1) (‘swine flu’) pandemic of 2009 is the most recent example. In response to that pandemic, 15 clinical trials were registered with anticipated enrolment of approximately 7000 patients. To date, only three trials, including 153 patients in total, have published results. None delivered results during the pandemic.[4]

Delivering research in a pandemic is challenging for several reasons. High-quality data are needed in the shortest time possible, and once a conclusion has been reached, that information should become publicly available immediately to let patients benefit from it. Secondly, most studies are designed to investigate the effect of a single treatment, but patients may receive multiple therapies. For COVID-19 this implies treatment with antiviral drugs, immune-modulators and anticoagulation, which may interact and may differ in subsequent disease states. A third challenge is that randomised controlled trials (RCTs) depend on assumptions about effect size to calculate a required sample size. Yet, effect sizes are notoriously difficult to estimate and this often leads to retrofitting the effect size to match the predicted recruitment.[5] Lastly, there is a need to immediately investigate newly available therapies, and the standard of care may change rapidly if effective treatments are found. In summary, pandemics call for a different approach in study design.

A novel approachAdaptive platform trials (APTs) deliver answers in the shortest possible timeframe, investigate the best treatment regimen instead of comparing single treatments, and can adapt to the dynamic changes during a pandemic.[6]

Netherlands Journal of Critical CareThe REMAP-CAP trial


There are three key design features of APTs. First, as with all RCTs, an APT is a prospective randomised experiment — a trial — of different treatment strategies. Second, APTs are a platform, allowing testing of multiple hypotheses at the same time under a core (or master) protocol. Because APTs focus on the underlying disease, rather than individual interventions, the overarching design can be created before any specific experimental arm is defined. The third element, distinguishing APTs, is that they are adaptive. They use information generated during the trial conduct to alter subsequent operations of the trial in a pre-specified way. New treatments can be added to the trial during its lifetime. Both elements (master protocol and adaptive design features) add complexity to the trial, but with the intent of making it more efficient.

REMAP-CAPA Randomised, Embedded, Multifactorial, Adaptive Platform (REMAP) trial adds two important features to the APT design elements. First, it is embedded in clinical care, facilitating recruitment. This is especially relevant in a pandemic where time for research activities is limited. Second, it is multifactorial, allowing testing of multiple hypotheses and multiple interventions at the same time. Randomisation occurs to multiple aspects of the treatment regimen at the same time. This enables ever-increasing knowledge generation about the best treatment regimen for a disease or condition.

A REMAP for Community Acquired Pneumonia (REMAP-CAP) was set up after the swine flu pandemic with the explicit goal to deliver research in pandemics. Initially an ICU trial, it focused on CAP, as this was considered to have the highest likelihood of resembling a future pandemic disease relevant to ICUs. It was set up across multiple regions, acknowledging the global nature of pandemics. Seed funding was obtained from the European Union in 2014 (FP7-HEALTH-2013-INNOVATION-1 #602525), extending to Australia, New-Zealand, Canada and the United States afterwards. Recently, India, Nepal and Pakistan joined REMAP-CAP through the Critical Care Asia network.

Analysis of REMAP-CAPAPTs often use Bayesian statistical inference models, as these are well suited for handling the adaptive and multifactorial design aspects. In REMAP-CAP, the primary analysis is generated from one overarching Bayesian cumulative logistic model, which calculates posterior probability distributions for the primary outcome based on evidence accumulated in the trial and assumed prior knowledge in the form of a prior distribution. The primary model adjusts for location (site, nested within country), age, sex, and time period (two-week periods). The model contains treatment effects for each intervention within each domain and pre-specifies relevant treatment-by-

treatment interactions across domains. The model is fit using a Markov Chain Monte Carlo algorithm that draws iteratively (10,000 draws) from the joint posterior distribution, allowing calculation of odds ratios with their 95% credible intervals (CrI) and the probability that each intervention (including control) is optimal in the domain, that an intervention is superior to control, that two non-control interventions are equivalent, or that an intervention is futile.

The model for the primary analysis includes all patients enrolled in the trial, providing the most robust estimation of the coefficients of all covariates. Importantly, not all patients are eligible for all domains or interventions. Therefore, the model includes covariate terms reflecting each patient's domain eligibility. The estimate of an intervention’s effectiveness, relative to any other intervention within that domain, is thus generated from those patients that might have been eligible to be randomised to those interventions within the domain.

From theory to practiceThe first REMAP-CAP patient was randomised at the University Medical Center Utrecht in April 2018. At the start of the pandemic, in March 2020, 56 centres globally participated

Box 1. Key concepts of REMAP-CAP

Key concept: multifactorial

To allow investigation of multiple aspects of treatment, REMAP-CAP organises treatments into domains, each covering one common therapeutic area (e.g., antiviral therapy) and containing at least two interventions (including a control arm). Patients eligible for the platform are assessed for eligibility and potentially randomised to multiple interventions across multiple domains (but to only one intervention in each domain for which they are eligible). Thus, they are randomised to a regimen of treatments rather than a single treatment. Participating sites can choose the domains and interventions they offer locally.

Key concept: adaptive

The adaptive design allows interventions to be added or altered over the course of the trial. Accumulating evidence in the trial is used to update the statistical model that is then used to decide about termination for graduation (e.g. superiority or futility of one part of the trial) and for updating randomisation probabilities (response adaptive randomisation, RAR). With RAR, patients are more likely to receive more promising interventions. The weights of the randomisation probabilities are updated at each interim. The interim analyses for assessing if pre-specified platform triggers that lead to termination (graduation or futility) have been reached are pre-planned. Accumulating evidence is also used to re-estimate the optimal sample size and an experimental arm can leave the trial as soon as the data permit.



in the study (including three Dutch centres). The pandemic stratum, facilitating inclusion of patients suspected or known to have COVID-19, was opened on 3 March 2020, and the first patient included six days later. Before the start of the pandemic, four domains were approved and open for recruitment. Nine domains containing 17 active interventions have been added so far (table 1). The number of participating sites increased rapidly during the pandemic, with now 296 active sites across the world. Of note, more than 130 sites from the United Kingdom, joined the trial after the chief medical officer had urged British hospitals to participate in REMAP-CAP.[7] Currently, 5312 patients with suspected or proven COVID-19 have been randomised for more than 10,000 treatment options, and six platform conclusions have been drawn during the pandemic, all guiding patient care.

After publication of the RECOVERY results, convincingly showing benefit of dexamethasone for patients admitted with COVID-19, the Corticosteroid Domain of REMAP-CAP was closed for these patients and available data were analysed.[8] In REMAP-CAP, treatment with a seven-day fixed-dose course of hydrocortisone or shock-dependent dosing of hydrocortisone resulted, compared with no hydrocortisone, in 93% and 80% probabilities of superiority for the primary outcome.[9] This treatment has now been incorporated in national and international treatment guidelines.[10,11] Despite not reaching the predefined statistical trigger (i.e., 99%), the 93% posterior probability of benefit from hydrocortisone contributes more information than the traditional frequentist result, where the only conclusion would have been that the null-hypothesis could not be refuted. The next two platform conclusions were that the interleukin 6 (IL-6) receptor antagonists sarilumab and tocilizumab are superior to no-immune modulation. These drugs reduce mortality for critically ill patients with COVID-19 by about 24%, and reduce time spent on organ support in the ICU, on top of treatment with corticosteroids.[12] A fourth result is that the antiviral drug lopinavir/ritonavir is futile in the treatment of critically ill patients with COVID-19. These results will be published soon. In the COVID-19 Antiviral Domain, two arms containing hydroxychloroquine were previously closed as equipoise for randomisation was lost as a result of external evidence. The results have been included in a meta-analysis, which yielded that treatment with hydroxychloroquine was associated with increased mortality in COVID-19 patients, and there was no benefit of chloroquine.[13] Lastly, two platform conclusions about anticoagulation were reached within six months of finalising the protocol. For this domain, REMAP-CAP collaborated with the ACTIV-4 and ATTACC trials in a multi-platform RCT (mpRCT). Therapeutic anticoagulation appeared to be futile, and possibly harmful, in critically ill patients receiving organ support, compared with local standard thromboprophylaxis, but superior in patients with less severity of illness.

Table 1. Currently active and paused domains of the REMAP-CAP study

Active domains Method CAP COVID-19 severe§

COVID-19 moderate

Antibiotic Comparing 5 different empiric antibiotic strategies

Antiviral* Comparing no influenza antiviral agent to 5 or 10 days of oseltamivir

Corticosteroid† Comparing administration of no corticosteroids to shock dependent or fixed dose hydrocortisone dosing regimens

COVID-19 ACE2 RAS

Comparing 4 different RAS inhibition strategies (no RAS inhibition, ACE inhibition, angiotensin receptor blockers (ARB) and ARB+DMX-200, a chemokine receptor-2 inhibitor)

COVID-19 Antiplatelet

Comparing 3 different antiplatelet interventions (no antiplatelet, aspirin and P2Y12 inhibitors)

COVID-19 immune modulation

Comparing 5 different strategies targeting the immune response to COVID-19 (no immune modulation, interferon-beta, anakinra, tocilizumab and sarilumab)

COVID-19 Statin therapy

Comparing no statin versus simvastatin for the treatment of ARDS caused by COVID-19

Mechanical ventilation

Comparing guideline recommended mechanical ventilation strategies to clinician-preferred mechanical ventilation strategy

Macrolide duration‡

Aimed at finding the most optimal treatment duration (3-5 vs 14 days) for macrolides.

Vitamin C Comparing no vitamin C to high dose vitamin C

Paused domains

COVID-19 Antiviral

Comparing 4 different COVID-19 specific antiviral treatment strategies (no antiviral agent, lopinavir/ritonavir, hydroxychloroquine and a combination of lopinavir/ritonavir and hydroxychloroquine

COVID-19 Therapeutic Anticoagulation

Comparing therapeutic anticoagulation to local standard anticoagulation

COVID-19 Immunoglobulin therapy

Comparing no immunoglobulin administration to convalescent plasma

§patients on the ICU requiring organ support;*Influenza patients only; †this domain is no longer open for COVID-19 patients;‡only available for patients randomised into one of the beta-lactam plus macrolideinterventions in the antibiotic domain; ACE2 = angiotensin-converting enzyme 2; ARB = angiotensin receptor blocker; DMX =dexamethasone, RAS = renin-angiotensin system

active not active



ConclusionIn summary, adaptive platform trials are a ‘new and improved’ method for performing trials, in line with current practice. They combine a robust design with flexibility that makes them more efficient and safe for patients. They allow rapid identification of the best treatment regimen for a disease or condition, embedded in clinical practice. Thus, they facilitate ‘learning while doing’ and pave the way towards a more learning healthcare system.[3,14] REMAP-CAP is a fully Bayesian APT aimed at finding the best treatment for CAP and designed to adapt to pandemics. The results from this trial within a year of the start of the pandemic prove the practical value and the effectiveness of the design.


References

1. World Health Organisation. COVID-19 - Numbers at a glance [Internet]. 2021. Available from: https://covid19.who.int.

2. Wu Z, McGoogan JM. Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention. JAMA. 2020;323:1239-42.

3. Angus DC. Optimizing the Trade-off Between Learning and Doing in a Pandemic. JAMA. 2020;323:1895-6.

4. Rojek AM, Martin GE, Horby PW. Compassionate drug (mis)use during pandemics: lessons for COVID-19 from 2009. BMC Med. 2020;18:265.

5. Schulz KF, Grimes DA. Sample size calculations in randomised trials: Mandatory and mystical. Lancet. 2005;365:1348-53.

6. Angus DC, Alexander BM, Berry S, et al. Adaptive platform trials: definition, design, conduct and reporting considerations. Nat Rev Drug Discov. 2019;18:808.

7. Atherton F, Calderwood C, McBride M, et al. Novel Coronavirus: Clinical Trials [Public letter]. 2020. Alert reference: CEM/CMO/2020/012

8. RECOVERY Collaborative Group. Dexamethasone in Hospitalized Patients with Covid-19 — Preliminary Report. N Engl J Med. 2020;NEJMoa2021436.

9. The Writing Committee for the REMAP-CAP Investigators. Effect of Hydrocortisone on Mortality and Organ Support in Patients with Severe COVID-19: The REMAP-CAP COVID-19 Corticosteroid Domain Randomized Clinical Trial. JAMA. 2020;324(13).

10. Siemieniuk R, Rochwerg B, Agoritsas T, et al. A living WHO guideline on drugs for covid-19. BMJ. 2020;370:m3379.

11. Alhazzani W, Møller MH, Arabi YM, et al. Surviving Sepsis Campaign: Guidelines on the Management of Critically Ill Adults with Coronavirus Disease 2019 (COVID-19). Crit Care Med. 2020;48:e440-e469.

12. The REMAP-CAP Investigators. Interleukin-6 Receptor Antagonists in Critically Ill Patients with Covid-19 — Preliminary report. medRxiv. 2021.

13. Axfors C, Schmitt AM, Janiaud P, et al. Mortality outcomes with hydroxychloroquine and chloroquine in COVID-19: an international collaborative meta-analysis of randomized trials. medRxiv. 2020.

14. Berry SM, Connor JT, Lewis RJ. The platform trial: An efficient strategy for evaluating multiple treatments. JAMA. 2015;313:1619-20.



Submitted December 2020; Accepted January 2021


Effect of high-dose methylprednisolone in mechanically ventilated ICU patients with COVID-19: a retrospective observational study

D. Hoogeveen, T.J.P. Ketels, T.J. Wilbers, A.C. StrangThe first two authors contributed equally to this workDepartment of Intensive Care, Rijnstate Hospital, the Netherlands

Correspondence

T.J.P. Ketels - [email protected]

Keywords - COVID-19, corticosteroids, intensive care, methylprednisolone

AbstractBackground: COVID-19 is associated with clinical features that closely resemble acute respiratory distress syndrome and causes hypoxic respiratory failure requiring ventilator support. The primary purpose of this study is to investigate the effects of high-dose methylprednisolone on the respiratory condition of COVID-19 patients on the intensive care unit (ICU) in the absence of early treatment with hydrocortisoneMethods: This retrospective observational study reports on all patients who were hospitalised with COVID-19 and received mechanical ventilation while on the ICU in Rijnstate Hospital. These patients received intravenous methylprednisolone following the ‘Meduri protocol’. The primary outcome was defined as improvement of the respiratory conditions expressed by P/F ratio and PEEP.Results: Seventeen of the 42 COVID-19 patients admitted to our ICU received methylprednisolone, initiated at day 13 (median). The mean length of ICU stay of these patients was 34 days. The average P/F ratio improved significantly from 14.2 kPa (107 mmHg) to 17.9 kPa (135 mmHg) after one week of treatment and from 17.9 kPa to 20.8 kPa (158 mmHg) after two weeks of treatment. This also applies to the positive end-expiratory pressure (PEEP), which decreased significantly to 9 cmH2O and 7 cmH2O after one and two weeks of treatment, respectively, compared with 10 cmH2O before start of the protocol. In our population the 28-day mortality was 18% versus an overall in-hospital mortality of 29% in patients with COVID-19.Conclusion: In patients with COVID-19 who received mechanical ventilation on the ICU, the use of high-dose methylprednisolone seemed to provide a significant improvement in both oxygenation and ventilation.

IntroductionCurrently, the world is experiencing a pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2; COVID-19). In turn, COVID-19 was preceded by similar coronaviruses such as SARS and MERS, also known for causing severe respiratory failure.[1,2] Furthermore, the COVID-19 virus is also known to cause hypoxaemic respiratory failure requiring ventilatory support.[3-6] The clinical features of many of these patients suffering from SARS, MERS and COVID closely resemble those of acute respiratory distress syndrome (ARDS): patients require high oxygen ventilation of a low-compliant lung. Early reports suggest that as many as 30% of COVID-19 patients suffer from ARDS.[7] While other studies claim that up to 60% of the COVID-19 patients admitted to the intensive care unit (ICU) are eventually diagnosed with ARDS.[8] Based on the reports, it was hypothesised that this COVID-19 related ARDS is of a vascular endotype, in which endothelial and subendothelial inflammation appears to play a key role in its pathophysiology.[9]

As advised by the European Society of Intensive Care Medicine (ESICM) in 2017[10] and confirmed by a more recent meta-analysis,[11] corticosteroids should be considered in patients with early, moderate to severe ARDS.[10] The initial reports from China and Italy, however, suggested avoiding corticosteroids in patients with COVID-19.[3,6,8]This is in contrast to more recent studies showing a lower mortality in patients treated with methylprednisolone.[12] Due to the conflicting evidence, there was initially no consensus in the expert panel of clinicians in the Netherlands on whether to use corticosteroids as a rescue therapy for the severely affected ICU patient with COVID-19. In the desperate position of progressive oxygenation and ventilatory failure in our most affected patients, we started

Netherlands Journal of Critical CareHigh-dose methylprednisolone in ventilated ICU patients with COVID-19


treating that specific group of patients with methylprednisolone. Meanwhile, multiple clinical trials started to report the positive effects of corticosteroids on mortality, duration of mechanical ventilation and severity of the illness.[13-15] In this report we aim to assess the effect and safety of a long, high-dose methylprednisolone treatment in severely affected COVID-19 patients admitted to the ICU, in the absence of early treatment with hydrocortisone. This retrospective observational study reports on all of the 17 patients who received methylprednisolone in the first half of 2020, focusing mainly on their respiratory conditions.

MethodsPatientsFor this retrospective observational study we included all COVID-19 patients admitted to the ICU in Rijnstate Hospital who received methylprednisolone as rescue therapy between 30 March and 5 June 2020. In total 42 COVID-19 patients were admitted to the ICU in Rijnstate Hospital in this period; of these 17 patients eventually received methylprednisolone and all were included in this report. Informed consent for retrospective studies was obtained from all patients.

InterventionBefore treatment with methylprednisolone was considered, patients had to meet the following criteria; persistent need of mechanical ventilation with progressive oxygenation and ventilatory failure. This was defined as a P/F ratio <15 kPa, PEEP > 10 cmH2O and static compliance <50 ml/cm H20. All patients were screened for active co-infections by serological and deep bronchial cultures before treatment was given. When treatment with methylprednisolone was indicated, we used the Meduri[16] methylprednisolone protocol as seen in table 1.

Table 1. Meduri protocol

Daily dose of methylprednisolone

Day 1 to 14 2 mg/kg/day

Day 15 to 21 1 mg/kg/day

Day 22 to 25 0.5 mg/kg/day

Day 26 to 28 0.25 mg/kg/day

Days 29 and 30 0.125 mg/kg/day

Data collectionThe following data were collected from the electronic patient file: baseline parameters such as age, sex, comorbidity, days of ICU admission and date of start methylprednisolone. The following parameters were scored daily during the period from 14 days before start of the Meduri protocol until 14 days after treatment was started: co-infections, SOFA score, inflammation parameters and respiratory parameters such as PEEP and P/F ratio.

OutcomeThe primary outcome was defined as improvement in respiratory conditions expressed as P/F ratio and PEEP. For the secondary outcome analysis we looked at inflammation (expressed by CRP), total duration of mechanical ventilation, duration of the ICU admission, in-hospital and 28-day mortality and multi-organ failure (SOFA score).

Statistical analysisIBM SPSS statistics 25 was used for the statistical analysis. Descriptive statistics were used to compare baseline data. We used the Shapiro-Wilk test to determine whether the variable was normally distributed in our study population. For normally distributed variables we used the mean and 95% confidence interval (CI), for non-normally distributed variables the median and the standard error. For the analysis of our primary outcome parameters paired T-tests were performed as all three were normally distributed.

ResultsThe baseline characteristics of our study population are shown in table 2. A total of 17 patients were treated with the Meduri protocol. All patients were being mechanically ventilated when the methylprednisolone was initiated. The majority of the patients were male (82%), with an average age of 64 years. Cardiovascular comorbidity was present in 29% of the patients, with hypertension (4 patients) and acute coronary syndrome (2 patients) as most frequent diagnosis. Pulmonary comorbidity was present in 35% of our patients, with obstructive lung disease (5 patients) and lung cancer (1 patient) as the most frequently found diagnosis. The mean duration of ICU admission for this group was 34 days, all patients required mechanical ventilation from the day of ICU admission. The median day at which the methylprednisolone was started was day 13 of ICU admission.

Table 2. Baseline characteristics

Gender (male, %) 82%

Age (years, mean, 95% CI) 64 [59-70]

BMI (kg/m2, median, SE) 27,8 [1,3]

Duration of mechanical ventilation (days, mean, 95% CI)

34 [27-42]

Duration of ICU admission (days, mean, 95% CI) 34 [25-43]

28-day mortality 18%

In hospital mortality 29%

Start Meduri (days ICU admission, median, SE) 13 [2.7]

Cardiovascular comorbidity 29%

Pulmonary comorbidity 35%

CI = confidence interval; SE = standard error



In our population, the 28-day mortality was 18% compared with an in-hospital mortality of 29%. A total of five patients in the population died during treatment. Three patients died within five days of starting methylprednisolone: one from massive intracranial bleeding, the other two patients from progressive respiratory failure. One patient died 14 days after steroid induction, also due to progressive respiratory failure caused by COVID-19 and recurrent pneumothorax. Finally one patient died on the general hospital ward after being discharged from the ICU, due to the consequence of recurrent aspiration and pneumonia. Regarding adverse events: no new co-infections were found after start of the treatment with dexamethasone.

Methylprednisolone following Meduri protocolAfter treatment with methylprednisolone was initiated, both the ventilation and oxygenation of our patients improved gradually. As shown in figure 1, the average P/F ratio improved significantly after one week of treatment, from an average of 14.2 kPa (107 mmHg) to 17.9 kPa (135 mmHg) after one week and eventually to 20.8 kPa (158 mmHg) after two weeks of treatment. Likewise, we reported positive effects for the positive end-expiratory pressure (PEEP). The PEEP needed to maintain adequate ventilation decreased significantly to 9 cmH2O and 7 cmH2O after one and two weeks of treatment, respectively, compared with 10 before the start of the protocol (figure 2). In the same way, mean level of C-reactive protein

Figure 1. Average P/f ratio (kPa) two weeks before and after start of the Meduri protocol.

Figure 2. Average PEEP (cmH2O) two weeks before and after start of the Meduri protocol.

Figure 3. Average CRP (mg/L) two weeks before and after start of the Meduri protocol.

Figure 4. Average SOFA score two weeks before and after start of the Meduri protocol.



(CRP) improved significantly one week after the start of treatment, from 219 to 99 and eventually declining to a mean of 66 in week 2 (figure 3). The SOFA score also improved significantly from 6.2 initially to 4.7 after one week, to 3.8 in week 2 (figure 4)

DiscussionIn this observational study, we show that conditions of mechanical ventilation (FiO2 and PEEP level) significantly improved after initiation of high-dose methylprednisolone in 17 patients with pneumonia caused by COVID-19, meeting criteria of severe ARDS. In addition, CRP levels significantly decreased, as did the SOFA scores.

In the early phase of the COVID-19 pandemic in China, corticosteroids were believed to act counterproductively, by increasing viral replication and causing increased duration of viral shedding as reported in SARS and MERS-CoV infected patients[1,2] Corticosteroids thereby increase the risk of nosocomial infections and death.[17] Later on, accumulating evidence supported the role of corticosteroids in preventing immune-mediated damage, such as in severe ARDS.

In agreement with the positive effects of corticosteroids in ARDS, reports from China eventually encouraged the use of steroids in COVID-19 patients on the ICU. Wu and co-authors[14] describe lower mortality if these patients were treated with methylprednisolone, which is in line with evidence from non-viral ARDS.[18-20] The Recovery trial[13] even supports the use of corticosteroids before onset of ARDS-like syndrome in COVID-19 and introduced a 6 mg hydrocortisone once daily regimen as standard of care, inspired by the positive effect of hydrocortisone in non-COVID-19 ARDS as described by Villar et al.[21] Patients requiring oxygen support or mechanical ventilation showed decreased 28-day mortality compared with controls in the Recovery trial. Likewise, in non-mechanically ventilated patients, Salton and co-authors[15] show lower ICU referral, lower mechanical ventilation and lower mortality than in patients not treated with steroids. Although most trials were stopped early after the positive results from the Recovery trial, a recently published meta-analysis by Sterne et al.,[22] including seven randomised trials and over 1700 patients, showed a significantly lower 28-day all-cause mortality in patients who were treated with corticosteroids compared with usual care or placebo. Another meta-analysis[23] confirmed these results.

The relationship between administration of high-dose corticosteroids and improvement in conditions of mechanical ventilation may be explained by decreased pulmonary inflammation, reflected by a significantly lower CRP. This mechanism has been proven in ARDS. [18]

To hypothesise, by relief of pulmonary inflammation, mechanical ventilation conditions improve, which presumably

translates to decreased organ failure as displayed in an almost halving of SOFA scores two weeks after initiation of high-dose methylprednisolone. This decrease in SOFA scores may lead to decreased mortality in patients with COVID-19 pneumonia meeting severe ARDS criteria. In our population no significant side effects were found. Comparably, no serious short-term side effects were reported in different dose regimens of corticosteroids in the recently published meta-analysis.[22] However, long-term consequences remain unknown.

Our study has many limitations, especially due to the observational, non-controlled, single-centre design in a small number of patients. However, to our knowledge, there are no trials investigating the potential benefit of a longer, high-dose treatment with corticosteroids for the more severely affected COVID-19 patients who fail to respond to the initial treatment. Our study shows that a long, high-dose treatment with corticosteroids is both effective and safe. The success of other clinical trials such as the Recovery trial with use of dexamethasone for ten days, is an invitation to reduce our steroid dose and duration in the future, especially to decrease side effects. Whether the much shorter and lower dosed corticosteroid treatments used in most of the previously mentioned clinical trials are sufficient for the more affected ICU patients remains to be investigated. As well as whether prolonged therapy with high-dose corticosteroids is effective in patients who show progression of the disease after the initial treatment. In learning more about treating severely ill COVID-19 patients, our observed beneficial effect of high-dose steroids may be biased by other unidentified contributing treatment factors.

ConclusionIn our 17 ICU patients suffering from COVID-19 pneumonia meeting criteria of severe ARDS, administering high-dose corticosteroids seemed to provide a significant improvement in both oxygenation and ventilation, as well as a rapid decline of the inflammation markers. Doing so also proved safe within our small population.

Disclosures All authors declare no conflict of interest. No funding or financial support was received.

References

1. Lee N, Allen Chan KC, Hui DS, et al. Effects of early corticosteroid treatment on plasma SARS-associated coronavirus RNA concentrations in adult patients. J Clin Virol. 2004;31:304-9.

2. Arabi YM, Mandourah Y, Al-Hameed F, et al. Corticosteroid therapy for critically ill patients with middle east respiratory syndrome. Am J Respir Crit Care Med. 2018;197:757-67.

3. Zhou F, Yu T, Du R, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet. 2020;395:1054-62.

4. Cao J, Tu WJ, Cheng W, et al. Clinical Features and Short-term Outcomes of 102 Patients with Corona Virus Disease 2019 in Wuhan, China. Clin Infect Dis. 2020;71:748-55.



5. Ruan Q, Yang K, Wang W, Jiang L, Song J. Clinical predictors of mortality due to COVID-19 based on an analysis of data of 150 patients from Wuhan, China. Intensive Care Med. 2020;46:846-8.

6. Phua J, Weng L, Ling L, et al. Intensive care management of coronavirus disease 2019 (COVID-19): challenges and recommendations. Lancet Respir Med. 2020;8:506-17.

7. Li X, Ma X. Acute respiratory failure in COVID-19: is it "typical" ARDS? Crit Care. 2020;24:198.

8. Immovilli P, Morelli N, Antonucci E, Radaelli G, Barbera M, Guidetti D. COVID-19 mortality and ICU admission: the Italian experience. Crit Care. 2020;24:228.

9. Mangalmurti NS, Reilly JP, Cines DB, Meyer NJ, Hunter CA, Vaughan AE. Covid-19-associated acute respiratory distress syndrome clarified: a vascular endotype? Am J Respir Crit Care Med. 2020;202:750-3.

10. Annane D, Pastores SM, Rochwerg B, et al. Guidelines for the diagnosis and management of critical illness-related corticosteroid insufficiency (CIRCI) in critically ill patients (Part I): Society of Critical Care Medicine (SCCM) and European Society of Intensive Care Medicine (ESICM) 2017. Intensive Care Med. 2017;43:1751-63.

11. Mammen MJ, Aryal K, Alhazzani W, Alexander PE. Corticosteroids for patients with acute respiratory distress syndrome: a systematic review and meta-analysis of randomized trials. Pol Arch Intern Med. 2020;130:276-86.

12. Chen N, Zhou M, Dong X, et al. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study. Lancet. 2020;395:507-13.

13. Horby P, Lim WS, Emberson JR, et al. for the RECOVERY Collaborative Group, Dexamethasone in hospitalized patients with covid-19 - preliminary report. N Engl J Med. 2020: doi: 10.1056/NEJMoa2021436. Online ahead of print.

14. Wu C, Chen X, Cai Y, et al. Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China. JAMA Intern Med. 2020. doi: 10.1001/jamainternmed.2020.0994.

15. Salton F, Confalonieri P, Meduri GU, et al. Prolonged low-dose methylprednisolone in patients with severe covid-19 pneumonia. Open Forum Infect Dis. 2020;7:ofaa421.

16. Meduri GU, Siemieniuk RAC, Ness RA, Seyler SJ. Prolonged low-dose methylprednisolone treatment is highly effective in reducing duration of mechanical ventilation and mortality in patients with ARDS. J Intensive Care. 2018;6:53.

17. Stockman LJ, Bellamy R, Garner P. SARS: Systematic review of treatment effects. PLoS Med. 2006;3:e343.

18. Meduri GU, Bridges L, Shih MC, Marik PE, Siemieniuk RAC, Kocak M. Prolonged glucocorticoid treatment is associated with improved ARDS outcomes: Analysis of individual patients' data from four randomized trials and trial-level meta-analysis of the updated literature. Intensive Care Med. 2016;42:829-40.

19. Peter JV, John P, Graham PL, Moran JL, George IA, Bersten A. Corticosteroids in the prevention and treatment of acute respiratory distress syndrome (ARDS) in adults: Meta-analysis. BMJ. 2008;336:1006-9.

20. Yang Z, Lei X, Li X. Early application of low-dose glucocorticoid improves acute respiratory distress syndrome: A meta-analysis of randomized controlled trials. Exp Ther Med. 2017;13:1215-24.

21. Villar J, Ferrando C, Martínez D, et al. Dexamethasone treatment for the acute respiratory distress syndrome: a multicentre, randomised controlled trial. Lancet Respir Med. 2020;8:267-76.

22. Sterne JAC, Murthy S, Diaz J, et al. for the WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group, Association between administration of systemic corticosteroids and mortality among critically ill patients with covid-19: a meta-analysis. JAMA. 2020;324:1330-41.

23. Siemieniuk RA, Bartoszko JJ, Ge L, et al. Drug treatments for covid-19: living systematic review and network meta-analysis. BMJ. 2020;370:m2980.





Neurological complications in COVID-19 patients admitted to a general ICU in the Netherlands

E.A. Krijnen1,2, S.P.P. Matthijs3, D.H.T. Tjan2

1Faculty of Medicine, University Medical Center Utrecht, Utrecht University, Utrecht, the Netherlands

Departments of 2Intensive Care Medicine and 3Neurology, Gelderse Vallei Hospital, Ede, the Netherlands

Correspondence

D.H.T. Tjan - [email protected]

Keywords - COVID-19, SARS-CoV-2, coronavirus, stroke, neurological complication

AbstractCoronavirus disease 2019 (COVID-19) predominantly affects the respiratory system. However, the viral infection has been associated with brain involvement. In the literature multiple routes are described through which this neurotrophic virus enters the central and peripheral nervous system, including direct and indirect pathways. This can cause various neurological complications, mostly in the advanced stages of the disease. In this case series we present four patients with COVID-19 and stroke out of 59 COVID-19 patients admitted to the intensive care unit in Ede, the Netherlands, before October 2020. Patients had two to five risk factors for cerebrovascular disease. In conclusion, COVID-19 itself might be a risk factor for neurological complications, particularly stroke in patients with cerebrovascular risk factors. Although the underlying pathophysiology remains to be fully understood, physicians should be aware of these neurological complications while treating critically ill patients with COVID-19.

BackgroundSevere acute respiratory syndrome coronavirus-2 (SARS-CoV-2), causing coronavirus disease 2019 (COVID-19), predominantly affects the respiratory system. Common symptoms of COVID-19 patients are fever, dry cough, shortness of breath, diarrhoea and fatigue.[1,2] Neurological symptoms are also frequently observed among COVID-19 patients. Headache and dizziness are considered non-specific minor symptoms associated with COVID-19.[3] Both the peripheral nervous system and the central nervous system (CNS) can be affected, mostly in advanced stages of the disease.[3] Possible CNS manifestations include meningitis, encephalitis, acute myelitis, stroke, and encephalopathy.[1,4-9] Manifestations of the peripheral nervous system may include Guillain-Barré syndrome, anosmia, chemosensory dysfunction and skeletal muscle damage.[5,10,11]

Nervous system disorders due to SARS-CoV-2 accompanied by the immune system are positively correlated with the degree of severity of COVID-19 symptoms.[6]

In the literature, multiple routes are reported through which SARS-CoV-2 may enter the brain, including direct and indirect pathways. The virus may gain access to the CNS through a synapse-connected route after invading peripheral nerve terminals of the respiratory or enteric network.[12-14] SARS-CoV-2 invades host cells through the angiotensin-converting enzyme 2 (ACE2) receptor, which is expressed in various cell types including the brain and capillary endothelium.[15] The interaction on capillary endothelium may damage the blood-brain barrier through which the virus gains bloodstream access into the CNS.[16] Together with hyperinflammatory responses and cytokine storms caused by the virus, this might result in acute myelitis, meningitis, encephalitis and severe encephalopathy.[6,17] As with other neurotrophic viruses, SARS-CoV-2 may spread transneuronally to distant brain targets.[18] Beside pathways to infiltrate the CNS, a procoagulant tendency may lead to CNS thromboembolisms resulting in ischaemic stroke.[9] Endothelial dysfunction and cytokine storm in combination with immobilisation leads to microangiopathy, hypercoagulability and blood stasis.[4,9,19] Haemorrhagic stroke has been reported in COVID-19 patients as well. Damaged intracranial arteries due to the interaction of SARS-CoV-2 and ACE2 receptors could lead to vessel wall rupture.[20] Coagulation disorders seen in COVID-19 patients result in disseminated intravascular coagulation, which accompanies this damage to intracranial arteries. Together these disorders could be the cause of haemorrhagic stroke.[20,21] COVID-19 related stroke may cause focal inflammation in the injured brain region causing substantial secondary brain injury.[22]

Netherlands Journal of Critical CareNeurological complications in COVID-19


Serious neurological complications of COVID-19 are increasingly reported, primarily in small case series and a few cohort studies. The aim of this study was to report the observed neurological complications of COVID-19 in the ICU of Gelderse Vallei Hospital during the first COVID-19 crisis.

MethodsConsecutive COVID-19 patients admitted to the ICU of Gelderse Vallei Hospital from 18 March to 12 October 2020 were retrospectively analysed. A confirmed case of COVID-19 was defined as a positive result on polymerase chain reaction (PCR) analysis of throat swab specimens or a CO-RADS score of at least 5 on chest CT scan with clinical symptoms of COVID-19.[23] Radiological assessments and laboratory testing were performed according to the clinical care needs of patients. We retrospectively reviewed medical records of all patients with confirmed SARS-CoV-2 infection. We collected data on

age, sex, duration of ICU admission, nervous system symptoms and risk factors for neurological complications of all patients. Risk factors include hypertension, diabetes mellitus, obesity, smoking, alcohol, dyslipidaemia, atrial fibrillation, history of cerebrovascular or cardiovascular events, and physical performance.[24] For patients with neurological symptoms, we collected laboratory findings (triglyceride level, low density lipoprotein cholesterol and non-fasting glucose) and radiological examinations. Neurological manifestations were reviewed and confirmed by a neurologist. TOAST classification was used to classify ischaemic stroke subtypes (table 1).[25] Written informed consent was obtained from each patient described.

ResultsFifty-nine patients with a confirmed COVID-19 diagnosis were admitted to the ICU. Of these patients, 19 were females and 40 males, with a median age of 69 years (range 29-83). Eleven patients died during admission to hospital, nine of whom died during their ICU stay. Five patients died after discharge from hospital. The median ICU stay was eight days (range 1-50). Cerebrovascular risk factors were present in 56 patients, with a median of two risk factors per patient (range 0-7). Four patients developed neurological symptoms during their ICU stay, all were diagnosed with ischaemic or haemorrhagic stroke (based on clinical examination and CT brain). Their characteristics including their risk factors are summarised in table 2. All cases are reviewed in detail below. At the beginning of the pandemic the normal doses of thromboprophylaxis based on body weight of 2850 IU or

Table 1. TOAST classification

Types of ischaemic stroke

Large-artery atherosclerosis

Cardioembolism3

Small-vessel occlusion

Stroke of other determined aetiology

Stroke of undetermined aetiology

*Adapted from Goldstein et al.[25]

Figure 1. CT brain of four ICU patients with coronavirus disease 2019 and stroke. 1A and 1B: patient 1 with ischaemic stroke, 1C and 1D: patient 2 with ischaemic stroke (white arrow indicates lesion), 1E and 1F: patient 3 with haemorrhagic stroke, 1G and 1H: patient 4 with haemorrhagic stroke.



5700 IU nadroparin once a day were given. After the thrombotic tendency in COVID-19 patients was established in the literature, we adopted an intermediate-intensity dose strategy of nadroparin twice daily 2850 IU or 5700 IU for thromboprophylaxis. If at the time of admission a haemorrhage was clinically suspected, the thromboprophylaxis strategy was individually adjusted to the patient after consultation with a neurologist.

Patient 1A 78-year-old male with a past medical history of stroke presented to the emergency room (ER) of another hospital with dyspnoea, headache, fever and a cough for two weeks. He was transferred to Gelderse Vallei Hospital for an ICU bed. Due to progressive respiratory failure, the patient was intubated and mechanically ventilated. After regular daily wakeup the sedation could be completely stopped on day 11. Due to severe ICU-acquired weakness, confirmed by electromyography on day

15, a tracheostomy was performed. On day 16, the patient still experienced persistent reduced consciousness (E1M4Vtube) and absent response to pain stimulus on his right side. CT brain showed semi-recent ischaemic changes in the left parieto-occipital lobe (figure 1A). No atrial fibrillation or thrombosis were observed during admission. EEG showed a diffuse slow background rhythm, with lowered left parieto-occipital amplitude, but no epileptic activity. On day 22, no haemorrhagic transformation or extension of the infarction was seen on second CT brain (figure 1B). During ICU admission the patient improved to a maximal EMV score with improving ICU-acquired weakness after which he could be decannulated on day 40. After 42 days, he could be transferred to the neurology ward. Upon discharge, the patient was able to get out of bed for one hour twice daily. He returned to his initial hospital for further rehabilitation. Duplex ultrasound was not performed to diagnose possible carotid stenosis as this would have no clinical consequences given the patient’s unstable condition.

Table 2. Characteristics of patients and cohort with coronavirus disease 2019 (COVID-19) who developed neurological symptoms during their stay at the ICU

Patient 1 Patient 2 Patient 3 Patient 4 Cohort

Age, years 78 74 62 68 73 [29-83]

Gender Male Male Female Male Male 37 (67.3%) / Female 18 (32.7%)

ICU stay, days 42 18 46 27 8 [1-50] 8 [1-50]

Duration of hospitalisation, days 46 30 83 28 16 [1-63]

COVID-19 diagnosis before onset of neurological complication, days

30 20 28 33 NA

Neurological complicationTOAST classification

Subtype haemorrhagic

Ischaemic strokeLarge-arteryatherosclerosisNA

Ischaemic strokeCardioembolism

NA

HaemorrhagicStrokeNA

HaemorrhagicStrokeNA

NA

HypertensionBlood pressure at admission, mmHg

No150/58

No146/91

De novo158/66

Yes112/65

Yes 24 (43.6%) / No 31 (56.4%)

DyslipidaemiaTriglyceride, mmol/lLDL-cholesterol, mmol/l

No1.7-

No1.1-

No1.22.69

Yes1.11.98

Yes 23 (41.8%) / No 32 (58.2%)

Diabetes mellitusNon-fasting glucose, mmol/l

No8.2

No7.3

No5.2

Type 2 diabetes10.3

Yes 16 (29.1%) / No 39 (70.9%)

Atrial fibrillation No No De novo duringadmission

No Yes 5 (9.1%) / No 50 (90.9%)

History of cerebrovascular or cardiovascular events

CVA (2 times) TIA No No Yes 17 (30.9%) / No 38 (69.1%)

Smoking Stopped at 40 years old

Unknown Unknown Stopped at 65 years Yes 15 (35.7%) / No 27 (64.3%)

Obesity, BMI 26.2 24.9 29.0 31.7 29.4 [22.5 – 56.9]

Physical status, Barthel score 100 100 90 80 100 [70 – 100]

Alcohol Incidental 2-3 units/day Unknown No Yes 23 (54.8%) / No 19 (45.2%)

Deceased No No No Yes Yes 15 (27.3%) / No 40 (72.7%)

Anticoagulation before event

Clopidogrel 75 mg po per day and nadroparin 2850 IU sc per day

Nadroparin 2850 IU sc twice daily till PE, then 7600 IU sc twice daily

Nadroparin 2850 IU sc per day till PE, then 7600 IU sc twice daily

Nadroparin 5700 IU sc twice daily till DVT, then 9500 IU sc twice daily

NA

Variables are shown as mean with range, or as number with percentage per subgroup. Note that the four patients reported were excluded from the cohort column. TOAST = Trial of Org 10172 in Acute Stroke Treatment; LDL = low-density lipoprotein; CVA = cerebrovascular accident; TIA = transient ischaemic attack; BMI = body mass index; PE = pulmonary embolism; DVT = deep venous thrombosis; NA = not applicable; po = orally; sc = subcutaneously



Patient 2A 74-year-old male with a medical history of sick sinus syndrome, transient ischaemic attack (TIA) and first-degree atrioventricular block, presented to the ER with progressive dyspnoea for three weeks, and possibly mild loss of strength of his right hand for one day. Because of acute respiratory failure he could not be properly neurologically evaluated. Brain imaging was postponed and patient was admitted to the ICU. The patient was intubated and mechanically ventilated in the prone position for two days. On day 2, CT angiography showed extensive subtotal and segmental pulmonary embolisms in all lung lobes, for which therapeutic anticoagulation was initiated. After discontinuation of sedation, the right-sided hemiparesis persisted. On day 7, CT brain showed multiple subacute infarction areas on both sides (figures 1C and 1D). Ultrasound showed an atrial septal defect potentially contributing to the cerebral embolisation of thrombi causing infarctions. No atrial fibrillation was observed during admission. He improved after two weeks, after which he could be discharged to the neurology ward on day 18. On the ward a normal carotid duplex was performed. On day 30, the patient could be discharged to a rehabilitation centre with a right-sided hemiparesis Medical Research Council (MRC) scale grade 3 to 4.

Patient 3A 62-year-old female with a medical history of COPD gold 2 and gastric reflux disease presented to the ER with fever, cough and dyspnoea for one week. Due to progressive respiratory failure, she was admitted to the ICU and was intubated. During admission the patient developed type 2 acute coronary syndrome with rapid atrial fibrillation and hypotension for which she was treated. On day 12, CT angiography showed pulmonary embolisms after which therapeutic anticoagulation was started. On day 17, the patient suddenly developed nystagmus, hypertension and a transient conjugated eye deviation to the right. Valproic acid was started because non-convulsive status epilepticus was diagnosed. CT brain showed a small haemorrhage in the left parietal lobe, possibly as a result of sinus thrombosis (figure 1E). Labetalol was started to control her blood pressure. Valproic acid was continued because it was effective. The following day a cerebral venous occlusion was excluded. On day 22, she developed a persistent conjugated eye deviation to the right. CT brain showed an increase in the size of the intraparenchymal haemorrhage (figure 1F). Therapeutic anticoagulation was continued at a lower dose. Tracheostomy was performed because of an accompanying ICU-acquired weakness. On day 43, a control CT brain showed that the intracerebral haemorrhage had decreased in size, after which full therapeutic anticoagulation was resumed. After decannulation on day 46, the patient was discharged to the nursing ward. Five weeks later, the patient was discharged to a rehabilitation centre. In the following three months, the patient presented multiple times at the ER with focal seizures. To date, she is not able to walk, and is experiencing memory and concentration deficits.

Patient 4A 68-year-old male with a past medical history including hypertension, type 2 diabetes and COPD gold 3 with reduced functional lung capacity presented to the ER with dyspnoea, cough and diarrhoea for one week. Due to respiratory failure, the patient was admitted to the ICU and was intubated. Flucloxacillin was initiated due to phlebitis after venous cannulation of the right arm. On day 10, he developed deep vein thrombosis of the right arm for which he received therapeutic nadroparin. Due to ICU-acquired weakness, a tracheostomy was performed on day 16, after which he could be weaned off mechanical ventilation. On day 26, the patient showed decreased consciousness (E1M4Vtube). CT brain showed parenchymal bleeding in the right hemisphere with ventricular breakthrough and mass effect with midline shift (figures 1G and 1H), most likely due to hypertension during admission. The consulting neurologist determined that the size and extent of the bleeding was severe. In combination with the patient’s impaired physical condition before admission, his prognosis was poor. After a family meeting where prognosis was discussed, palliative care was initiated, and patient died on day 28.

DiscussionIn this case series we present four patients with COVID-19 and stroke in the COVID-19 ICU population. Neurological symptoms were present at presentation in only one patient. Patients had two to five risk factors for cerebrovascular disease, including hypertension, dyslipidaemia, atrial fibrillation, positive medical history for cerebrovascular or cardiovascular diseases, obesity and the use of alcohol or smoking. Patient characteristics correspond to the total COVID-19 ICU population.Mao et al. reported stroke in 5.7% of non-ICU patients (5/88) with severe SARS-CoV-2 infection admitted to a general hospital.[5] In a study by Li et al., stroke was reported in about 5% of non-ICU patients (11/219) admitted to a major tertiary hospital, with an average time of onset 12 days after COVID-19 diagnosis. These patients were associated with severe disease and had a higher incidence of cerebrovascular risk factors. Patients with stroke had an increased inflammatory response and higher D-dimer levels.[26] Retrospective observational analysis by Rifino et al. identified stroke in 53 of 1760 (3.0%) patients with COVID-19 admitted to a general hospital.[27] The only study in an ICU population is the study by Klok et al.[9] They reported a cumulative incidence of stroke of 3.7%. All patients received at least standard doses of thromboprophylaxis, which is comparable with the antithrombotic protocol of Gelderse Vallei Hospital. Moreover, higher age and hypercoagulable state, defined as a prolonged prothrombin time of more than three seconds or activated partial thromboplastin time over five seconds, were risk factors for thrombotic complications.[9] A prospective multicentre cohort study by Koh et al. included a total of 47,572 patients. Thirty-nine patients had neurological manifestations; of these patients 84.4% were asymptomatic or had mild symptoms, 2.2% had severe and 13.3% critical SARS-CoV-2



infection. Sixteen patients had ischaemic stroke, and three had a TIA, with a median age of 53 years. Only three of these 19 patients had critical COVID-19. Fifteen of them had cerebrovascular risk factors. Another two critically ill patients developed an intracerebral haemorrhage. Four cases of young men with cerebral venous thrombosis were identified, two of whom were asymptomatic and two had mild COVID-19.[28]

In the literature, neurological syndromes have been reported in patients with a SARS-CoV-2 infection and causality between both entities has been explored. COVID-19 is associated with a wide spectrum of neurological syndromes which affect the entire neuraxis, including cerebral vasculature and inflammatory CNS syndromes with central and peripheral nervous system involvement. The underlying mechanisms of these syndromes may be multifactorial, resulting from combined or independent effects of sepsis, hypoxia, thrombosis and cytokine storms. Valencia-Enciso et al. revealed a trend between the time of onset of ischaemic stroke and severity of the disease. In patients with severe COVID-19, stroke developed late, while in those with mild COVID-19, stroke presented early (mean 23.2 and 5.1 days, respectively).[4] These data are in line with the time of stroke onset in the four patients with severe COVID-19 infection presented here, who were admitted to the ICU. Inflammatory markers were related to the development of large vessel occlusion.[4] Therefore, together with an increased predisposition to thrombosis,[29] the hyperinflammatory state present in patients with severe COVID-19 might be a substantial contributor to the multifactorial underlying mechanisms resulting in stroke. However, data are still emerging and it is too early to draw conclusions.Besides stroke, other neurological manifestations of COVID-19, such as encephalopathy, encephalitis and epilepsy, have been studied as well. The emergence of a neurological disorder during SARS-CoV-2 infection must be assessed in its complex context and it must be determined whether it could be a direct or indirect effect of viral invasion in the nervous system or represent a random finding. The European Academy of Neurology core COVID‐19 Task Force initiated an online survey on neurological symptoms observed in patients with COVID‐19. All reported neurological disorders by in total 2343 responders were interpreted as being possibly related to COVID-19. The authors stated that the numbers provided by their survey represent relevant information for the European healthcare system to consider strengthening neurological services.[30] Timely neurological assessment of critically ill patients remains important in any case, especially during this COVID-19 pandemic, given the increased risk of neurological complications. A daily wakeup call is important to pursue because intra-arterial thrombectomy can be considered in selected cases of stroke in which the exact start of neurological symptoms is known or not exactly known.COVID-19 could have substantially contributed to the development of neurological complications of the patients described here. The infection might not have been the direct cause of the neurological complication. However, it could have exacerbated the risk for neurological complications in these critically ill patients, especially

stroke. Patient 1 could have had significant carotid stenosis. However, this was not investigated by duplex ultrasound. So, it is unknown whether this led to the ischaemic stroke or whether it was due to the prothrombotic state in the COVID 19 infection. The infarctions in both hemispheres in patient 2 might also have been exacerbated by a prothrombotic state but could also be caused by the atrial septum defect which was found. In patient 3 no other cause of haemorrhagic stroke was found; COVID-19 may therefore play a major role. The parenchymal haemorrhage in patient 4 was possibly due to a combination of previous chronic hypertension and a higher blood pressure during admission, so COVID-19 might be a lower risk factor. Compared with the total COVID-19 cohort, it is not possible to differentiate based on risk factors between ICU patients with and without neurological complications.Larger studies are needed to confirm clinical, radiological and laboratory characteristics of neurological complications in patients with COVID-19. Moreover, identifying patients at risk for neurological complications and how to control these risks should be elucidated as well, together with the underlying pathophysiological mechanisms. Given the hyperinflammatory state, it would be interesting to study whether the incidence of neurological complications has changed with standard use of dexamethasone in patients with COVID-19 during the second wave.

ConclusionCOVID-19, caused by SARS-CoV-2, is associated with neurological manifestations especially in the hyperinflammatory phase. In our small cohort of COVID-19 ICU patients, we experienced a high number of strokes (4/59). COVID-19 itself could be a risk factor for neurological complications. Whether pre-existing risk factors, as present in our patients, contribute to this high number of strokes is suggested in other studies but needs further investigation. So, despite the fact that the underlying pathophysiology remains to be fully understood, physicians should be aware of neurological complications while treating critically ill patients with COVID-19 in ICU.


References

1. Durrani M, Kucharski K, Smith Z, Fien S. Acute Transverse Myelitis Secondary to Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2): A Case Report. Clin Pract Cases Emerg Med. 2020;4:344-8.

2. Mao R, Qiu Y, He JS, et al. Manifestations and prognosis of gastrointestinal and liver involvement in patients with COVID-19: a systematic review and meta-analysis. Lancet Gastroenterol Hepatol. 2020;5:667-78.

3. Garg RK. Spectrum of Neurological Manifestations in Covid-19: A Review. Neurol India. 2020;68:560-72.

4. Valencia-Enciso N, Ortiz-Pereira M, Zafra-Sierra MP, Espinel-Gómez L, Bayona H. Time of Stroke Onset in Coronavirus Disease 2019 Patients Around the Globe: A Systematic Review and Analysis. J Stroke Cerebrovasc Dis. 2020;29:105325.

5. Mao L, Jin H, Wang M, et al. Neurologic Manifestations of Hospitalized Patients With Coronavirus Disease 2019 in Wuhan, China. JAMA Neurol. 2020;77:683-90.


6. Khatoon F, Prasad K, Kumar V. Neurological manifestations of COVID-19: available evidences and a new paradigm. J Neurovirol. 2020;26:619-30.

7. Poyiadji N, Shahin G, Noujaim D, Stone M, Patel S, Griffith B. COVID-19-associated Acute Hemorrhagic Necrotizing Encephalopathy: Imaging Features. Radiology. 2020;296:119-20.

8. Moriguchi T, Harii N, Goto J, et al. A first case of meningitis/encephalitis associated with SARS-Coronavirus-2. Int J Infect Dis. 2020;94:55-8.

9. Klok FA, Kruip MJHA, van der Meer NJM, et al. Incidence of thrombotic complications in critically ill ICU patients with COVID-19. Thromb Res. 2020;191:145-7.

10. Lechien JR, Chiesa-Estomba CM, De Siati DR, et al. Olfactory and gustatory dysfunctions as a clinical presentation of mild-to-moderate forms of the coronavirus disease (COVID-19): a multicenter European study. Eur Arch Otorhinolaryngol. 2020;277:2251-61.

11. Zhao H, Shen D, Zhou H, Liu J, Chen S. Guillain-Barré syndrome associated with SARS-CoV-2 infection: causality or coincidence? Lancet Neurol. 2020;19:383-4.

12. Li YC, Bai WZ, Hashikawa T. The neuroinvasive potential of SARS-CoV2 may play a role in the respiratory failure of COVID-19 patients. J Med Virol. 2020;92:552-5.

13. Toljan K. Letter to the Editor Regarding the Viewpoint "Evidence of the COVID-19 Virus Targeting the CNS: Tissue Distribution, Host-Virus Interaction, and Proposed Neurotropic Mechanism". ACS Chem Neurosci. 2020;11:1192-4.

14. Matschke J, Lütgehetmann M, Hagel C, et al. Neuropathology of patients with COVID-19 in Germany: a post-mortem case series. Lancet Neurol. 2020;19:919-29.

15. Baig AM, Khaleeq A, Ali U, Syeda H. Evidence of the COVID-19 Virus Targeting the CNS: Tissue Distribution, Host-Virus Interaction, and Proposed Neurotropic Mechanisms. ACS Chem Neurosci. 2020;11:995-8.

16. Wu D, Yang XO. TH17 responses in cytokine storm of COVID-19: An emerging target of JAK2 inhibitor Fedratinib. J Microbiol Immunol Infect. 2020;53:368-70.

17. Román GC, Spencer PS, Reis J, et al. The neurology of COVID-19 revisited: A proposal from the Environmental Neurology Specialty Group of the World Federation of Neurology to implement international neurological registries. J Neurol Sci. 2020;414:116884.

18. McGavern DB, Kang SS. Illuminating viral infections in the nervous system. Nat Rev Immunol. 2011;11:318-29.

19. Levi M, Thachil J, Iba T, Levy JH. Coagulation abnormalities and thrombosis in patients with COVID-19. Lancet Haematol. 2020;7:438-40.

20. Carod-Artal FJ. Neurological complications of coronavirus and COVID-19. Rev Neurol. 2020;70:311-22.

21. Vinayagam S, Sattu K. SARS-CoV-2 and coagulation disorders in different organs. Life Sci. 2020;260:118431.

22. Shi K, Tian DC, Li ZG, Ducruet AF, Lawton MT, Shi FD. Global brain inflammation in stroke. Lancet Neurol. 2019;18:1058-66.

23. Prokop M, van Everdingen W, van Rees Vellinga T, et al. CO-RADS: A Categorical CT Assessment Scheme for Patients Suspected of Having COVID-19-Definition and Evaluation. Radiology. 2020;296:97-104.

24. Flicker L. Cardiovascular risk factors, cerebrovascular disease burden, and healthy brain aging. Clin Geriatr Med. 2010;26:17-27.

25. Goldstein LB, Jones MR, Matchar DB, et al. Improving the reliability of stroke subgroup classification using the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) criteria. Stroke. 2001;32:1091-8.

26. Li Y, Li M, Wang M, et al. Acute cerebrovascular disease following COVID-19: a single center, retrospective, observational study. Stroke Vasc Neurol. 2020;5(3):279-84.

27. Rifino N, Censori B, Agazzi E, et al. Neurologic manifestations in 1760 COVID-19 patients admitted to Papa Giovanni XXIII Hospital, Bergamo, Italy. J Neurol. 2020:1-8.

28. Koh JS, De Silva DA, Quek AML, et al. Neurology of COVID-19 in Singapore. J Neurol Sci. 2020;418:117118.

29. Hess DC, Eldahshan W, Rutkowski E. COVID-19-Related Stroke. Translational stroke research. 2020;11:322-25.

30. Moro E, Priori A, Beghi E, et al. The international European Academy of Neurology survey on neurological symptoms in patients with COVID-19 infection. Eur J Neurol. 2020;27:1727-37.




Submitted November 2020; Accepted January 2021

C A S E R E P O R T

Endotracheal tube obstruction in patients diagnosed with COVID-19

E.A. van Boven, S. van Duin, H.H. PonssenDepartment of Intensive Care, Albert Schweitzer Hospital, Dordrecht, the Netherlands

Correspondence

E.A. van Boven - [email protected]

Keywords - COVID-19, intensive care unit, acute respiratory distress, airway obstruction

AbstractIn this case report, two COVID-19 patients with acute respiratory distress are presented. In both cases the acute respiratory distress was due to occlusion of the endotracheal tube. Patient A had an occlusion due to a blood clot. Patient B had an occlusion due to mucus secretions. In general, acute obstructions of the endotracheal tube are uncommon. The aim of this case report was to describe unusual complications in intubated COVID-19 patients. The increased risk of acute endotracheal obstruction in COVID-19 patients is due to multiple factors, such as hypercoagulopathy and the duration of the ventilatory support.

Introduction Since COVID-19 is a relatively new disease, treating physicians will be confronted with various complications. This case report describes two COVID-19 patients with unusual acute ventilation problems in the course of the ICU admission.

Case report 1Patient A (56 years) was admitted to the ward with COVID-19 and supported with oxygen therapy. After two days the patient deteriorated and developed hypoxaemia. He was admitted to the intensive care unit (ICU) and intubated with a size 8 mm tube. The patient’s clinical course on the ICU was characterised by high ventilation requirements, as is generally seen in COVID-19 patients, with a high positive end-expiratory pressure (PEEP) of ±16 cm H2O and low tidal volumes of 6 ml/kg on pressure control.[1] The patient was sedated and received neuromuscular blockage. Over time, the ventilator support was slowly reduced and the sedation and neuromuscular blockage could be stopped. On day 12, weaning from the mechanical ventilator was started. The weaning started three times a day for an hour and could be intensified in the following days. On day 15, the patient was hypertensive and agitated during weaning and the weaning had

to be stopped. The patient was restarted on ventilator support. On day 16, the patient was increasingly uncomfortable and sweating followed by acute respiratory failure, during which no ventilation was possible at all. The endotracheal tube was changed under suspicion of obstruction. Figure 1 shows the photo of the tube in which a blood clot is visible, blocking the tube. After changing the tube, the patient could resume his weaning program without any problems and the next day he was successfully extubated.

Case report 2Patient B (52 years) initially presented with dyspnoea due to COVID-19 infection. Two days after admission, the patient deteriorated and developed hypoxaemia. The patient was admitted to the ICU and intubated with a size 8 mm tube.

Figure 1. Endotracheal tube obstructed by a blood clot.



On the ICU the patient was supported on pressure control with high ventilator settings as is generally seen in COVID-19 patients with high PEEP (PEEP 15 cm H2O, FiO2 35%). However, on day 6 the ventilator settings had to be increased due to hypoxaemia. A chest X-ray showed an increase in bilateral pulmonary infiltrates. On day 8, the patient was hypoxic and the P/F ratio had decreased. The patient was turned into a prone position. The patient deteriorated further, with the pCO2 rising to 15 kPa. A thoracic CT scan was negative for pulmonary embolism and the infiltrates were stable. On further analysis of the respiratory problem, the peak pressure was found to be high (40 cm H2O), with normal plateau pressure generating a low tidal volume of <6 ml/kg. Based on these numbers, there was a suspicion of high resistance over the extrathoracic airways. We performed a fibreoptic bronchoscopy. A mucus plug was seen obstructing the endotracheal tube and the airway above the carina (figure 2). The mucus was removed, after which the CO2 level immediately decreased to 7 kPa, the peak pressure decreased and the tidal volumes normalised. The same problem occurred again a week later, and once again the respiratory problem was due to occlusion by mucus accumulation in the endotracheal tube.

Consideration Above we have described two COVID-19 patients with acute respiratory problems during mechanical ventilation, which were unusual for our hospital. In the literature, endotracheal obstruction by a blood clot or mucus is uncommon. In Australia, an analysis showed that only 2% of all problems with endotracheal tubes were caused by a blood clot or mucus accumulation.[2]

The first case describes a patient with an obstruction of the endotracheal tube due to a blood clot. It has recently been

established that COVID-19 patients have a high incidence of thrombotic complications.[3] This probably also increases the chance of formation of a blood clot inside the tube. The second case presents a patient with a tube obstruction due to mucus accumulation. A previous study showed that there is a negative relation between the duration of ventilator support and the open diameter of the endotracheal tube. Longer duration of ventilator support results in mucus accumulation in the tube, leading to a reduced functional diameter of the endotracheal tube.[4] Since the average duration of the ventilator support in COVID-19 patients is longer than in the general ICU population, it is more likely that tube obstruction due to mucus accumulation will occur. On the ICU, the standard care of the tube entails the use of a humidifier. This was not adjusted for COVID-19 patients. Endotracheal tube obstruction can be recognised by an increase in peak flow as was shown by Kawati et al., although this might be a late sign of gradual obstruction.[5]

Kawati et al. also tried to investigate whether there were early signs of gradual tube obstruction by simulation of endotracheal tube obstruction in pigs. Not one parameter to diagnose a partial endotracheal tube obstruction was found. It was identified that a decrease compared with the starting point in the expiratory flow could be a sign of partial endotracheal tube obstruction. However, even this one parameter is subject to many other changes such as compliance of the respiratory system.[6]

Hopefully this case report will help others working in ICUs to recognise acute airway obstructions in COVID-19.

DisclosuresAll authors declare no conflicts of interest. No funding or financial support was received.

Informed consentAn informed consent was obtained from both patients.

References

1. Fan E, Beitler JR, Brochard L, et al. COVID-19-associated acute respiratory distress syndrome: is a different approach to management warranted? Lancet Respir Med. 2020;8:816-21.

2. Szekely SM, Webb RK, Williamson JA, Russell WJ. The Australian Incident Monitoring Study. Problems related to the endotracheal tube: an analysis of 2000 incident reports. Anaesth Intensive Care. 1993;21:611-6.

3. Klok FA, Kruip MJHA, van der Meer NJM, et al. Confirmation of the high cumulative incidence of thrombotic complications in critically ill ICU patients with COVID-19: An updated analysis. Thromb Res. 2020;191:148-50.

4. Shah C, Kollef MH. Endotracheal tube intraluminal volume loss among mechanically ventilated patients. Crit Care Med. 2004;32:120-5.

5. Kawati R, Lattuada M, Sjostrand U, et al. Peak airway pressure increase is a late warning sign of partial endotracheal tube obstruction whereas change in expiratory flow is an early warning sign. Anesth Analg. 2005;100:889-93, table of contents.

6. Kawati R, Vimlati L, Guttmann J, et al. Change in expiratory flow detects partial endotracheal tube obstruction in pressure-controlled ventilation. Anesth Analg. 2006;103:650-7.

Figure 2. Endotracheal tube obstructed by mucus secrete .



Submitted November 2020; Accepted January 2021

C A S E R E P O R T

Cerebral microbleeds in a COVID-19 patient

B. Maatman1, E. Aronica2, S.D. Roosendaal3, D.C. Velseboer1

Departments of 1Intensive care , 2Pathology and 3Radiology, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands

Correspondence

B. Maatman - [email protected]

Keywords - cerebral microbleeds, COVID-19, coma

AbstractA 31-year-old male was admitted to the intensive care unit with respiratory failure due to SARS-CoV-2 infection. A persistent altered mental state, with agitation and subsequently persistent coma, was one of the complications he suffered during his illness. Brain MRI showed diffuse cerebral microbleeds (CMBs). The pattern of distribution resembled the CMBs that have been described earlier in critically ill patients. Post-mortem analysis of brain samples confirmed CMBs and additionally showed diffuse subcortical hypoxic white matter damage. We suggest that cerebral hypoxia may cause CMBs in COVID-19 patients in a similar way to other well-described causes of hypoxia. Furthermore, we want to bring to attention that coma in COVID-19 patients with CMBs may be associated with diffuse subcortical white matter injury not visible on brain MRI.

IntroductionCOVID‐19 is a novel viral disease in which knowledge regarding disease epidemiology and clinical presentation has been rapidly evolving in the past months since the initial identification. In addition to respiratory failure, severe neurological sequelae associated with COVID-19 have been described, including encephalitis, acute necrotising encephalopathy and cerebrovascular disease.[1,2] In this case report we present a distinct neurological complication in a COVID-19 patient, possibly associated with hypoxic episodes.

Case reportA 31-year-old male with a history of substance abuse was admitted to the emergency room of a general hospital with complaints of dyspnoea, cough and fever of one week duration. The polymerase chain reaction tests on throat and nasal swabs for SARS-CoV-2 upon admission were positive. Chest X-ray revealed bilateral consolidations. Treatment was initiated with ceftriaxone and azithromycin. Four days after admission his respiratory symptoms worsened and he was transferred to the intensive care unit where invasive mechanical

ventilation was initiated. At day 31, there was no sign of respiratory improvement and he was transferred to our referral hospital for expertise. During his stay in our department, deep venous thrombosis and pulmonary embolisms were diagnosed for which unfractionated heparin was initiated. He developed bilateral pneumothoraces and chest tubes were inserted. He had persistent fever and positive blood cultures with Staphylococcus epidermidis for which vancomycin was administered. Disseminated intravascular coagulation (DIC) was suspected due to widespread petechiae, yet formal DIC calculations were repeatedly inconclusive. Laboratory variables included a nadir platelet count of 103 x 109/l, a nadir fibrinogen level of 3.5 g/l, a peak D-dimer of 3.47 mg/l and a peak INR of 1.3. Despite proning, high PEEP level and intermittent neuromuscular blocking agents, he had recurrent episodes of severe hypoxaemia lasting for hours. During his stay, his altered mental state was a major concern, fluctuating between agitation and coma. Discontinuation of sedation for agitation initially improved his consciousness. However, at day 36 he relapsed into a coma (Glasgow Coma Score: E1M1Vnt (not testable in an intubated patient)) and after this there were no signs of neurological recovery. There were no abnormalities on repeated CT scans of the brain. An EEG showed diffuse slowing consistent with encephalopathy and no signs of epilepsy. A lumbar puncture was performed and analysis of cerebrospinal fluid showed no abnormalities. MRI of the brain demonstrated diffuse cerebral microbleeds (CMBs), with a high density in the corpus callosum and subcortical white matter, consistent with the pattern of distribution of critical illness-associated CMBs (figure 1).[3] Forty-six days after admission there was still no improvement in his neurological and respiratory condition, and his treatment was discontinued. Analysis of autopsy samples of the brain showed multiple CMBs with a dense concentration in the frontoparietal subcortical region and corpus callosum (figure 2). In addition, diffuse subcortical hypoxic white matter injury, an increased microglial activity with disruption of the blood-brain barrier and micro infarctions with micro calcifications were seen (not shown in figure).

Netherlands Journal of Critical CareCerebral microbleeds in a COVID-19 patient


DiscussionRecently, a distinct microbleed phenomenon has been described in the context of critical illness.[3] These CMBs involve the juxtacortical white matter and corpus callosum and spare the deep and periventricular white matter and the grey matter. There is a resemblance with the CMBs that are seen in patients with disease associated with hypo-oxygenation such as ARDS and high-altitude exposure. This suggests that there might be a common pathogenesis were hypoxia plays a significant role.[4,5] Hypoxia-induced hydrostatic or chemical effects on the blood-brain barrier could potentially account for extravasation of erythrocytes.[6] In our patient, oxygenation was compromised due to COVID-19 associated ARDS and recurrent pneumothoraces. The distribution pattern was consistent with the CMBs seen in critical illness and hypoxia-related disease, suggesting that cerebral hypoxia played a role. Although there are emerging data that severe COVID-19 can be complicated by significant coagulopathy, the evidence that

therapeutic anticoagulation is effective in these patients remains scarce.[7] Next to this, antithrombotic therapy, initiated after the occurrence of thrombotic complications, may have contributed to the expansion of CMBs. Another possible explanation for critical illness-associated cerebral microbleeds is DIC.[8] Our patient suffered from persistent fever and had petechiae. However, accurate diagnosis of DIC is challenging and in our case formal DIC score calculations were repeatedly inconclusive, despite evidence of increased fibrinogen degradation products and reduced platelet levels. In COVID-19 patients there are other possible explanations for the development of CMBs. A significant consumption coagulopathy can lead to thrombosis in small medullary veins leading to CMBs.[9] It is also worth noting that the spike protein of the SARS-CoV-2 virus has a strong affinity for the angiotensin-converting enzyme 2 (ACE-2) receptor which has widespread expression in endothelial cells.[10] Microscopic disruption of the endothelium in brain tissue may also be responsible for CMBs.[9] In a recent study in COVID-19 patients, with analysis of brain MRIs but no histological examination, CMBs were associated with critical illness, increased mortality, and worse functional outcome.[11]

Spontaneous intracerebral haemorrhage usually results in a focal neurological deficit and is easily diagnosed by CT scan. It is caused by arterial rupture, leads to haematoma formation in the lobar hemispheres or deep grey structures and is associated with high mortality.[12] CMBs, best visualised by MRI, result from disruption of the integrity of the blood-brain-barrier with subsequent erythrocyte extravasation and are usually clinically asymptomatic.[13] CMBs are recognised consequences of cerebral amyloid angiopathy and chronic hypertension and they have been associated with older age, hypertension, smoking, white matter disease, lacunar infarcts, previous ischaemic stroke or intracerebral haemorrhage.[14] Post-mortem studies have shown that CMBs detected on MRI result from the accumulation of blood in the vicinity of pathologically altered vessels.[15] They are associated with surrounding tissue damage, so concomitant brain dysfunction might be possible.[16] Although CMBs are only small in size, in the order of several millimetres, it is not

Figure 1. Axial susceptibility-weighted image (SWI) shows numerous punctiform susceptibility artifacts in the brain (in neuroradiological reports termed ‘microbleeds’). They are located mainly in the corpus callosum (genu and splenium are visible here) and subcortical white matter. In the close-up of the left frontal area, the ‘microbleeds’ are encircled; for clarity, veins have been highlighted in blue.

Figure 2. Histopathological specimens of the corpus callosum (A); frontal cortex (B); grey matter (C); stained with haematoxylin & eosin showing extravasation of red blood cells. There was no indication of vasculitis. Scale bars: A, C: 60 µM. B: 100 µM

Netherlands Journal of Critical CareCerebral microbleeds in a COVID-19 patient


unreasonable to think that they could cause symptoms if they form rapidly or in functionally strategic locations.[17] Another possibility is that, rather than disrupting function by the direct destruction of tissue, CMBs could disrupt the activity of the surrounding neurons, thus affecting local brain function or connectivity.[18] Moreover, most CMBs contain haemosiderin, a compound that may affect the electrical activity of the cortex and thereby possibly initiate focal seizures.[19] What is more likely than such local effects, is that the accumulation of multiple CMBs over time could have a more insidious effect on the brain functions that depend on the integrity of widespread anatomical networks, for example cognition or gait.[20] Although these are plausible mechanisms by which CMBs could cause clinical symptoms, the evidence that CMBs independently affect brain function or cause focal neurological symptoms through associated tissue damage remains limited.[21] The consciousness of our patient was fluctuating between agitation and coma, and was initially clouded by sedatives to decrease his agitation. An EEG showed diffuse slowing consistent with encephalopathy, which has been described in another COVID-19 patient with CMBs and coma.[22] Finally, persistent coma was the reason to perform an MRI of the brain, which revealed the presence of CMBs. The contribution of these CMBs to his coma is difficult to determine. The involvement of strategic locations, such as the basal ganglia, and extensive number of CMBs suggests that it played a role in the development of coma. More likely, the coexistent histological presence of diffuse subcortical white matter hypoxic injury and micro infarctions played a more significant role in the development of coma. This combination of CMBs and subcortical white matter injury, without typical features of viral or post-viral encephalitis, has recently been described in the context of COVID-19-associated neuropathology.[11, 23] A challenge for future research is studying how CMBs could affect the brain since they are closely linked to many manifestations of cerebrovascular disease. Moreover, in COVID-19 the independent contribution of CMBs to disability and death is uncertain. Fundamental research is needed to unravel the association between SARS-CoV-2 infection and cerebral CMBs, and provide answers whether the CMBs develop secondary to hypoxia or to other COVID-19-associated factors.

ConclusionIn this case report we present a patient with a fatal SARS-CoV-2 infection with severe neurological complications. The pattern of distribution of CMBs was similar to other disease associated with hypoxaemia. We suggest that the recurrent and prolonged hypoxic episodes may have played a role in the development of critical illness-associated CMBs. Furthermore, we show that SARS-CoV-2 infected patients with CMBs may also have diffuse subcortical hypoxic white matter lesions on post-mortem analyses, not detected by MRI. We recognise the limitations of this single case report and acknowledge that additional studies

are necessary to establish a connection between hypoxaemia, CMBs and coma in COVID-19 patients.

DisclosuresWritten informed consent was obtained from the patient’s family for the publication of this case report. All authors declare no conflict of interest. No funding or financial support was received.

References

1. Paterson RW, Brown RL, Benjamin L, Nortley R, Wiethoff S, Bharucha T, et al. The emerging spectrum of COVID-19 neurology: clinical, radiological and laboratory findings. Brain. 2020.

2. Schurink BP, Roos EP, TP Radonic, EP Barbe, Bouman CS et al. Viral presence and immunopathology in patients with lethal COVID-19: a prospective autopsy cohort study. Lancet Microbe: Elsevier Ltd.; 2020.

3. Fanou EM, Coutinho JM, Shannon P, Kiehl TR, Levi MM, Wilcox ME, et al. Critical Illness-Associated Cerebral Microbleeds. Stroke. 2017;48(4):1085-7.

4. Kallenberg K, Dehnert C, Dorfler A, Schellinger PD, Bailey DM, Knauth M, et al. Microhemorrhages in nonfatal high-altitude cerebral edema. J Cereb Blood Flow Metab. 2008;28(9):1635-42.

5. Riech S, Kallenberg K, Moerer O, Hellen P, Bartsch P, Quintel M, et al. The Pattern of Brain Microhemorrhages After Severe Lung Failure Resembles the One Seen in High-Altitude Cerebral Edema. Crit Care Med. 2015;43(9):E386-E9.

6. Bailey DM, Bartsch P, Knauth M, Baumgartner RW. Emerging concepts in acute mountain sickness and high-altitude cerebral edema: from the molecular to the morphological. Cell Mol Life Sci. 2009;66(22):3583-94.

7. Mondal S, Quintili AL, Karamchandani K, Bose S. Thromboembolic disease in COVID-19 patients: A brief narrative review. J Intensive Care. 2020;8:70.

8. Levi M, Toh CH, Thachil J, Watson HG. Guidelines for the diagnosis and management of disseminated intravascular coagulation. Brit J Haematol. 2009;145(1):24-33.

9. Varga Z, Flammer AJ, Steiger P, Haberecker M, Andermatt R, Zinkernagel AS, et al. Endothelial cell infection and endotheliitis in COVID-19. Lancet. 2020;395(10234):1417-8.

10. Cantuti-Castelvetri RO, Pedro LP, Djannatian M, Franz J, et al. Neuropilin-1 facilitates SARS-CoV-2 cell entry and provides a possible pathway into the central nervous system. bioRxiv; 2020.

11. Agarwal S, Jain R, Dogra S, Krieger P, Lewis A, Nguyen V, et al. Cerebral Microbleeds and Leukoencephalopathy in Critically Ill Patients With COVID-19. Stroke. 2020;51(9):2649-55.

12. Woo D, Broderick JP. Spontaneous intracerebral hemorrhage: epidemiology and clinical presentation. Neurosurg Clin N Am. 2002;13(3):265-79, v.

13. Fazekas F, Kleinert R, Roob G, Kleinert G, Kapeller P, Schmidt R, et al. Histopathologic analysis of foci of signal loss on gradient-echo T2*-weighted MR images in patients with spontaneous intracerebral hemorrhage: evidence of microangiopathy-related microbleeds. AJNR Am J Neuroradiol. 1999;20(4):637-42.

14. Viswanathan A, Chabriat H. Cerebral microhemorrhage. Stroke. 2006;37(2):550-5.15. Haller S, Montandon ML, Lazeyras F, Scheffler M, Meckel S, Herrmann FR, et al.

Radiologic-Histopathologic Correlation of Cerebral Microbleeds Using Pre-Mortem and Post-Mortem MRI. Plos One. 2016;11(12).

16. Schrag M, McAuley G, Pomakian J, Jiffry A, Tung S, Mueller C, et al. Correlation of hypointensities in susceptibility-weighted images to tissue histology in dementia patients with cerebral amyloid angiopathy: a postmortem MRI study. Acta Neuropathol. 2010;119(3):291-302.

17. Jeon SB, Kwon SU, Cho AH, Yun SC, Kim JS, Kang DW. Rapid appearance of new cerebral microbleeds after acute ischemic stroke. Neurology. 2009;73(20):1638-44.

18. Cianchetti FA, Kim DH, Dimiduk S, Nishimura N, Schaffer CB. Stimulus-evoked calcium transients in somatosensory cortex are temporarily inhibited by a nearby microhemorrhage. PLoS One. 2013;8(5):e65663.

19. Baumann CR, Schuknecht B, Lo Russo G, Cossu M, Citterio A, Andermann F, et al. Seizure outcome after resection of cavernous malformations is better when surrounding hemosiderin-stained brain also is removed. Epilepsia. 2006;47(3):563-6.

20. Baezner H, Blahak C, Poggesi A, Pantoni L, Inzitari D, Chabriat H, et al. Association of gait and balance disorders with age-related white matter changes: the LADIS study. Neurology. 2008;70(12):935-42.

21. Werring DJ. Cerebral microbleeds. Pathophysiology to clinical practice.: Cambridge Medicine; 2010.

22. De Stefano P, Nencha U, De Stefano L, Megevand P, Seeck M. Focal EEG changes indicating critical illness associated cerebral microbleeds in a Covid-19 patient. Clin Neurophysiol Pract. 2020;5:125-9.

23. Reichard RR, Kashani KB, Boire NA, Constantopoulos E, Guo Y, Lucchinetti CF. Neuropathology of COVID-19: a spectrum of vascular and acute disseminated encephalomyelitis (ADEM)-like pathology. Acta Neuropathol. 2020;140(1):1-6.




C A S E R E P O R T

ECMO as rescue for COVID-19 related ARDS: the pros and cons

J.J. van der Heijden¹, C.L. Meuwese¹, S.A. Braithwaite²Departments of ¹Intensive Care and ²Cardiothoracic Anaesthesiology, University Medical Centre, Utrecht University,

Utrecht, the Netherlands

Correspondence

J.J. van der Heijden – [email protected]

Keywords - ECMO, COVID-19, ARDS

AbstractUsing extracorporeal membrane oxygenation (ECMO) as rescue respiratory support for COVID-19 related ARDS may be seen as a controversial use of scarce and stretched resources in the context of a global pandemic. We present two cases of COVID-19 related ARDS both requiring ECMO support in an experienced ECMO tertiary centre. The cases illustrate the decision-making concerning the initiation of ECMO during life-threatening COVID-19 related respiratory failure. Both patients encountered the typical complications during extensive ECMO support, but with a completely different outcome. To identify patients that really benefit from this potentially life-saving support modality and attain meaningful functional outcome in addition to survival, remains a matter of debate.

IntroductionDespite the enormous number of patients with confirmed COVID-19, only few are supported with ECMO for obvious reasons such as lack of resources and expertise in combination with high demand of staffing and costs, plus the highly available alternative of mechanical ventilation. At the beginning of the pandemic there were several reports stressing these important factors[1,2] and the outcome of first published ECMO cases was very disappointing.[3] Overall, expert opinion was to reserve ECMO support for younger patients without comorbidity and multi-organ failure who had been on mechanical ventilation for less than a week.[4]

However, the confrontation of healthcare workers with many relatively young patients with severe ARDS and the publication of more case series from China and Italy with a better outcome[5,6] resulted in an increasing number of ECMO patients. Interim Extracorporeal Life Support Organization guidelines that appeared in April 2020 also shifted to a slightly more liberal approach, for example by including patients who had been on mechanical ventilation for up to two weeks.[7] In this context,

the following patients presented to our hospital during these early months of 2020, also known as ‘the first wave’.

Case 1A 46-year-old male with no previous medical history was referred via a nearby hospital because of progressive respiratory failure despite mechanical ventilation with high ventilator settings, including a 100% fraction of inspired oxygen. His symptoms started two weeks earlier, including fatigue, fever and dyspnoea. Polymerase chain reaction confirmed the presence of SARS-CoV-2 RNA. He was treated with hydroxychloroquine and ceftriaxone; bacterial blood and sputum cultures remained negative.

At presentation to our hospital the patient was deeply sedated in the prone position with ongoing neuromuscular blockade and he had been on mechanical ventilation for seven days.Ventilator settings had been set at: pressure control level 14 cm H₂O above PEEP 22 cm H₂O, frequency 20/min, I:E ratio 1:1, and fiO₂ 100%. Despite the extreme ventilator settings and prone positioning, an arterial blood gas showed severe respiratory acidosis (pH 7.12, PaCO₂ 95 mmHg and bicarbonate 30.0 mmol/l) in combination with a low P/F ratio (63). Chest X-ray showed diffuse shadowing with air bronchograms and bilateral consolidations (figure 1a).

After a short and unsuccessful trial of inhaled nitric oxide, we decided to initiate peripheral veno-venous (V-V) ECMO the same day. In order to perform cannulation, the patient was turned to the supine position, at which oxygen saturation dropped to 80% but the patient remained haemodynamically relatively stable with a sinus tachycardia of 120 beats/min and invasive blood pressure of 120/70 mmHg with a minimal dose of vasopressors.



Netherlands Journal of Critical CareECMO as rescue for COVID-19 related ARDS


Ultrasound-guided cannulation followed (23 and 19 French cannulas inserted in the right femoral and right internal jugular vein, respectively). Sweep gas flow via the ECMO system was started at 2 l/min, which was slowly increased to 10 l/min to prevent a rapid correction of arterial PaCO₂. Blood flow was kept at 4 l/min. Ventilator settings were adjusted to allow as much ‘lung rest’ as possible: pressure control level 12 cm H₂O above PEEP 14 cm H₂O, frequency 8/min, I:E ratio 1:2, and fiO₂ 80%.

A bronchoscopy was performed on day 1 which showed oedematous airways without purulent secretions. Bacterial cultures obtained during bronchoalveolar lavage showed no pathogenic micro-organisms and a negative Aspergillus antigen index (0.16). Antibiotics were stopped on admission until blood cultures demonstrated Enterococcus faecium on day 6 for which the central lines were changed and vancomycin was started. Vancomycin was continued until after decannulation because of a second positive blood culture with Staphylococcus epidermidis despite adequate levels of vancomycin. Kidney

function remained normal with an adequate clearance. During admission we regularly aimed for a negative fluid balance using intravenous furosemide.

Due to the lack of respiratory improvement, a CT scan of the chest was performed on day 13. It showed severe bilateral consolidations with ground glass opacities in combination with bilateral pleural effusion, possibly fitting empyema (figure 2). There were no signs of pulmonary embolism. After draining and culturing pleural effusion and repeating bronchoalveolar lavage, meropenem was started to anticipate the possibility of Pseudomonas species or other ceftriaxone-resistant micro-

Figure 1a and b. Chest X-rays of patients 1 and 2, respectively, showing diffuse shadowing with air bronchograms, bilateral consolidations and both ECMO cannulas in position

Figure 2 a, b and c. Chest CT of patient 1 at different levels showing bilateral consolidations with ground glass opacities in combination with bilateral pleural effusion



organisms. Bacterial cultures, however, remained negative and meropenem was stopped after five days. The Aspergillus antigen index also turned out to be negative again (0.25). As hyperinflammation seemed to be the cause of the stagnating clinical course, high-dose methylprednisolone (1 mg/kg/day for the first 2 weeks) was started.

Our patient improved in the following days allowing us to lower the ECMO settings for the first time. Ventilator settings were gradually increased in the weaning phase of the ECMO support, unfortunately resulting in a pneumothorax on day 19. After drainage, however, it was possible to switch to pressure support and continue ECMO weaning. Decannulation took place on day 21 which was immediately followed by fever (max 39.6 ºC). Because repetitive blood cultures remained negative after stopping vancomycin, we interpreted this as the frequently observed systemic inflammatory response syndrome after ECMO decannulation. Sedation was stopped on day 25 and the patient awoke the following day. Because of severe ICU-acquired weakness, a tracheostomy was performed on day 27 and after another week he was weaned off mechanical ventilation. A chest X-ray on day 30 showed clear improvement of the pulmonary abnormalities (figure 3). He was discharged to the medical ward on day 37 and transferred to a rehabilitation centre on day 56. Besides some remaining peripheral weakness of hands and feet, he recovered well. He lives at home, has good quality of life and resumed work completely.

Case 2 A 60-year-old fit and active male with normal stature and a recent medical history including bronchial hyperreactivity, type 2 diabetes mellitus and radiotherapy for prostate carcinoma was admitted to a referring hospital three weeks before presentation

at our centre. He was diagnosed with COVID-19 and treated with hydroxychloroquine and high-flow oxygen. After four days he was transferred to the ICU because of progressive respiratory failure. He was intubated and mechanically ventilated with intermittent prone positioning because of a low P/F ratio. Ceftriaxone was started on admission to the ICU and changed after two days to piperacillin/tazobactam due to persistent fever and increasing inflammatory parameters. Cultures, however, remained negative. Because of further respiratory deterioration on day 7, a CT angiogram of the chest was performed, which showed severe bilateral consolidations in combination with subsegmental pulmonary embolism. A therapeutic dose of

Figure 3. Chest X-ray of patient 1 before ICU discharge, showing clear improvement of pulmonary abnormalities

Figure 4 a, b and c. Chest CT of patient 2 at day 1 of our ICU admission showing bilateral consolidations and also subpleural reticulation and some traction bronchiectasis, possibly in the setting of (early) pulmonary fibrosis



low-molecular-weight heparin was started in combination with intravenous furosemide. The following days showed gradual clinical improvement, but after two weeks, ventilator settings had to be increased significantly and bronchoalveolar lavage was performed to search for ventilator-associated pneumonia. Piperacillin/tazobactam was restarted awaiting results and the patient was transferred to our centre for possible ECMO indication.

At presentation ventilator settings were as follows: pressure control level 24 cm H₂O above PEEP 6 cm H₂O, frequency 30/min, I:E ratio 1:1,5 and fiO2 60%. Arterial blood gas showed pronounced hypercapnia (pH 7.21, PaCO₂ 124 mmHg and bicarbonate 48.9 mmol/l) in combination with a low P/F ratio (112). Compliance was extremely low (11 ml/cm H₂O). We performed a new CT scan the same day which still showed severe bilateral consolidations and subsegmental pulmonary embolism. But now subpleural reticulation and some traction bronchiectasis could also be seen, possibly in the setting of pulmonary fibrosis (figure 4), matching low compliance. After extensive discussion in our ECMO team, taking into account the patient’s previous good functional status in combination with the fact he had yet to receive anti-inflammatory therapy, we decided to start V-V ECMO support. Ultrasound-guided cannulation (25 French left femoral vein and 21 French right internal jugular vein) was followed by a second bronchoalveolar lavage and high-dose methylprednisolone (1 mg/kg/day for the first two weeks). The ventilator settings were changed to pressure control level 14 cm H₂O above PEEP 10 cm H₂O, frequency 12/min, I:E ratio 1:5 and fiO₂ 40%. Sedation and neuromuscular blockade could be stopped soon after and the patient awoke, changing to pressure support with tidal volumes around 4 ml/kg and acceptable breathing frequency (20-25/min) using 3 l/min blood flow and 6 l/min gas flow.

Bacterial cultures remained negative as did the Aspergillus antigen index (0.09). Polymerase chain reaction on bronchoalveolar lavage material, however, was positive for Cytomegalovirus. Because of an increasing plasma Cytomegalovirus load the following week ganciclovir was started. A tracheostomy was performed in the setting of severe ICU-acquired weakness and after almost two weeks of physiotherapy gas flow could be slowly weaned, followed by decannulation on day 15. However, four days after removing the ECMO, he showed a rapid respiratory and haemodynamic decline. CT angiography showed a new pneumomediastinum and increase of bilateral consolidations without progression of pulmonary embolism or signs of right ventricular failure (figure 5). Ongoing ventilator-induced lung injury or even self-inflicted lung injury were suspected, possibly in combination with a new ventilator-associated pneumonia. Because hypoxaemia was not present at that time, we decided to go for a second ECMO run, but this time configured for

extracorporeal CO₂ removal (ECCO₂R). A 22 French dual-lumen cannula was inserted in the right internal jugular vein. Blood flow was started at 1.8 l/min but could easily be lowered to 1.1 l/min. Gas flow had to be increased from 5 to a maximum of 10 l/min to accomplish low respiratory drive. Cultures again remained negative.

A few days later he spontaneously developed a cerebellar bleed with a minimal hydrocephalus impeding mobilisation on ECCO₂R. The heparin ratio was in the lower normal range (1.5), but with a stable thrombocytopenia (70 x 10 ⁹/l) for which the oxygenator was replaced. He was treated conservatively and after a week recovered quite well. Physiotherapy and mobilisation could be restarted and gas flow could even be

Figure 5a, b and c. Chest CT of patient 2 showing new pneumomediastinum, developing subpleural cysts and an increase of bilateral consolidations

seems comparable with other causes of ARDS.[8-10] The presence of older age and important comorbidities, however, will surely reduce the probability of functional recovery after a complex ICU admission. Taking into account this uncertainty relating to eventual survival and functional outcome, the question should be asked in advance if such a long ICU trial is in line with the patient’s wishes.

Common ICU complications will inevitably occur such as central line associated bloodstream infections, ventilator-associated pneumonia, delirium and ICU-acquired weakness.[11] Continuous antibiotic therapy is often needed for Gram-positive bacteria in the context of colonisation of the ECMO circuit. Repeated changing of the oxygenator for this reason is not only



lowered to a minimum of 3 l/min. But during further episodes of central line associated bloodstream infections (Enterococcus faecium and Staphylococcus epidermidis) he remained highly gas flow dependent and delirious. This once again seemed to stop him from recovering further, so we decided to continue vancomycin as long as ECCO₂R was ongoing.

On day 57 of our ICU admission a new CT scan was taken to re-evaluate the pulmonary status (figure 6). Unfortunately, there were now also signs of honeycombing and an increase in subpleural cysts, fitting a diagnosis of progressive pulmonary fibrosis. Bilateral consolidations were still extensive, and a new left-sided pleural effusion was seen, possibly empyema. Bronchoalveolar lavage showed Sphingomonas paucimobilis for which meropenem was started; pleural effusion, however, remained negative. As an incidental finding, a spontaneous haematoma was also seen in the left-sided musculus teres major and subscapularis. The case was extensively discussed with our lung transplantation physicians and the team decided to see if a workup towards lung transplantation would be feasible. Prerequisites were complete neurological recovery including resolution of delirium and a significant improvement of muscle strength leading to a chronic situation of home ventilation without ECMO. Although the delirium faded away and physiotherapy could be intensified, our patient slowly developed acute kidney injury and further lung transplantation screening revealed severe pulmonary hypertension (estimated systolic pulmonary pressure >100 mmHg) with echocardiographic signs of significant right-sided heart failure. Changing to veno-arterial-venous support by that time was found disproportionate. These findings were discussed with the family and after more than 100 days of ICU admission he died as a result of multi-organ failure.

DiscussionWe present two patients supported by ECMO for COVID-19 related ARDS. Both encountered multiple complications during a prolonged ICU admission. Their different outcomes represents the extreme results of such a challenging and expensive support modality. As reflected by our two cases, the most common reason for using ECMO in COVID-19 patients is as respiratory support for COVID-19 related ARDS meeting traditional criteria and not responding to conventional therapy. Prone positioning, neuromuscular blockade, optimising mechanical ventilation, recruitment manoeuvres and a trial of inhaled nitric oxide should all have been performed before initiating ECMO.[7]

Because of the severity of disease, a quick recovery is not to be expected and eligible patients must be in such a pre-hospital condition that they are able to recover from an ICU admission of weeks or even months. Looking at the limited data available on survival after ECMO support in COVID-19 patients, mortality

Figure 6a, b and c. Chest CT of patient 2 after almost 3 months of ICU admission, showing honeycombing and an increase in subpleural cysts, fitting a diagnosis of progressive pulmonary fibrosis



expensive but can also be insufficient because of the remaining cannulas. Diagnosing ventilator-associated pneumonia is often difficult requiring repeated bronchoalveolar lavage as in our patients, potentially exposing healthcare workers to SARS-CoV-2. Mobilisation on ECMO is normally feasible, but ongoing delirium significantly impeded the physiotherapy required for ICU-acquired weakness in our second patient. A more ECMO-specific complication is coagulopathy secondary to consumption of clotting factors, for example due to ECMO oxygenator thrombosis, which can result in serious spontaneous bleeding such as the cerebellar bleeding in our second patient. Underdosing of anticoagulant therapy or an ongoing inflammatory response, however, can cause clinically significant thrombosis, especially in the setting of COVID-19. Besides a small thrombus in the inferior caval vein in our first patient and pulmonary embolism in our second patient, we did not detect severe thrombosis; however, such a diagnosis can easily be missed.[12] Daily monitoring of coagulation tests (e.g. APTT/PT) or even thromboelastography is therefore mandatory in combination with a normal blood count with optional levels of anticoagulation (anti-Xa) and clotting factors (such as fibrinogen or antithrombin) on a regular basis.[13] Even with optimal anticoagulation therapy, oxygenator thrombosis cannot always be prevented and regular system changes quite often remain necessary, especially in the early stage of ICU admission and hyperinflammation.

Downregulation of the immune system with or without immune suppressive therapy can lead to opportunistic infections or viral reactivation, such as Cytomegalovirus in our second patient. Some specific antifungal or antiviral therapy carries a high risk for worsening acute kidney injury, although our patients did not suffer from such drug-induced renal complications. Ongoing respiratory failure requiring high pressures of mechanical ventilation, heightens the risk of lung fibrosis and prolonged permissive hypercapnia in this setting can lead to vascular changes causing pulmonary hypertension and eventually end-stage right ventricular failure with secondary multi-organ failure. Overcoming all these possible problems is already challenging but the bigger picture is to prevent the patient becoming chronically ECMO-dependent as seen in our second patient. Although lung transplantation in the setting of end-stage pulmonary failure after COVID-19 has been described,[14] the shortage of donors and a pre-existing waiting list for organ

donation plus the numerous risk factors in an ECMO-dependent recipient makes this a very undesirable option.

Conclusion The usage of ECMO in the setting of COVID-19 related ARDS can be a life-saving decision that leads to the desired good functional outcome in a patient who otherwise would have inevitably died. At the same time, it is a scarce and expensive resource in the context of a global pandemic and it coincides with a long and often complicated ICU admission. Identification of patients that benefit most from ECMO in this specific setting remains difficult as no patient is exactly the same and disease progression cannot always be stopped. A careful assessment before initiating ECMO and recurrent evaluation during ECMO support seems mandatory, taking every above mentioned aspect into account.


References

1. MacLaren G, Fisher D, Brodie D. Preparing for the Most Critically Ill Patients With COVID-19: The Potential Role of Extracorporeal Membrane Oxygenation. JAMA. 2020;323:1245-6.

2. Ramanathan K, Antognini D, Combes A, et al. Planning and provision of ECMO services for severe ARDS during the COVID-19 pandemic and other outbreaks of emerging infectious diseases. Lancet Respir Med. 2020;8:518-26.

3. Henry BM, Lippi G. Poor survival with extracorporeal membrane oxygenation in acute respiratory distress syndrome (ARDS) due to coronavirus disease 2019 (COVID-19): pooled analysis of early reports. J Crit Care. 2020;58:27-8.

4. Bartlett RH, Ogino MT, Brodie D, et al. Initial ELSO Guidance Document: ECMO for COVID-19 Patients with Severe Cardiopulmonary Failure. ASAIO J 2020;66:472-4.

5. Li X, Guo Z, Li B, et al. Extracorporeal membrane oxygenation for coronavirus disease 2019 in Shanghai, China. ASAIO J 2020;66:475-81.

6. Loforte A, Dal Checco E, Gliozzi G, et al. Veno-venous extracorporeal membrane oxygenation support in COVID-19 respiratory distress syndrome. ASAIO J. 2020;66:734-8.

7. Shekar K, Badulak J, Peek G, et al. Extracorporeal Life Support Organization Coronavirus Disease 2019 Interim Guidelines: A consensus document from an International Group of Interdisciplinary Extracorporeal Membrane Oxygenation Providers. ASAIO J. 2020;66:707-21.

8. Schmidt M, Hajage D, Lebreton G, et al. Extracorporeal membrane oxygenation for severe acute respiratory distress syndrome associated with COVID-19: a retrospective cohort study. Lancet Respir Med. 2020;8:1121-31.

9. Lorusso L, Combes A, Lo Coco V, et al. ECMO for COVID-19 patients in Europe and Israel. Intensive Care Med. 2021;9:1-5.

10. Shaefi S, Brenner S, Gupta S, et al. Extracorporeal membrane oxygenation in patients with severe respiratory failure from COVID-19. Intensive Care Med. 2021; https://doi.org/10.1007/s00134-020-06331-9.

11. Luyt C-E, Sahnoun T, Gautier M, et al. Ventilator-associated pneumonia in patients with SARS-CoV-2-associated acute respiratory distress syndrome requiring ECMO: a retrospective cohort study. Ann Intensive Care. 2020;10:158.

12. Sklar M, Sy E, Lequier L, et al. Anticoagulation practices during venovenous extracorporeal membrane oxygenation for respiratory failure. A systematic review. Ann Am Thorac Soc. 2016;13:2242-50.

13. Chlebowski M, Baltagi S, Carlson M, Levy JH, Spinella PC. Clinical controversies in anticoagulation monitoring and antithrombin supplementation for ECMO. Crit Care. 2020;24:19.

14. Bharat A, Querrey M, Markov N, et al. Lung transplantation for patients with severe COVID-19. Sci Transl Med. 2020;16:12



Submitted August 2020; Accepted January 2021

L E T T E R T O T H E E D I T O R

Fraction of inspired oxygen (FiO2) to modulate respiratory drive in a mechanically ventilated patient

A. Osinski, J. van Rosmalen, R.P.J. BaltussenDepartment for Critical Care, Máxima Medical Centre, Veldhoven, the Netherlands

Correspondence

J. van Rosmalen - [email protected]

Keywords - oxygen, respiratory drive, mechanical ventilation, COVID-19

Dear editor,With great interest we read the review by Jonkman, de Vries and Heunks, extensively explaining the physiology of respiratory drive in ICU patients, published recently in Critical Care.[1] We would like to present a recent clinical observation of a patient’s respiratory drive that was possibly influenced by subtle differences in arterial oxygen content.

A 71-year-old male was admitted to our ICU in the spring of 2020 with respiratory insufficiency due to COVID-19. He had no relevant medical history, especially no pulmonary diseases or tobacco use. According to our local protocol, he was treated with selective digestive tract decontamination and (after his approval) chloroquine. His admission was complicated by subsegmental pulmonary emboli, infection with Aspergillus fumigatus, a pneumothorax and a central line-associated bloodstream infection. Chronic lymphocytic leukaemia without secondary complications or indications for therapeutic intervention was diagnosed during his ICU stay. He was ventilated mainly in the prone position. After tracheotomising him, our patient was weaned from mechanical ventilation and discharged to the pulmonology ward after 51 days in the ICU.

When we observed bradypnoea during pre-oxygenation for bronchial suctioning, we hypothesised our patient’s respiratory drive might be influenced by the arterial oxygen content

(CaO2). Secondly, we hypothesised that we might be able to use this characteristic (instead of sedatives and/or neuromuscular blocking agents) to maintain a supposedly, safe manner of mechanical ventilation. Our observations and measurements were obtained in the fourth week of his ICU admission. At that moment, he was being mechanically ventilated (PEEP 5 cmH2O, 12 cmH2O pressure support), in a supine position and haemodynamically stable without vasopressors or inotropes (blood pressure 138/53 mmHg, heart rate 96 beats/min). His PaO2/FiO2 ratio was 139. He was sedated with midazolam (9 mg/hour), propofol (70 mg/hour), fentanyl (125 µg/hour) to a Richmond Agitation-Sedation Scale (RASS ) of -5. Neuromuscular blocking agents were ceased and his core temperature was 37.2 °C. The haemoglobin level was 6.1 mmol/l.

We performed three sets of measurements (table 1) in which we gradually decreased the FiO2 to obtain different oxygen saturation levels, as measured by pulse oximetry (SpO2). Apart from the FiO2, all of the ventilator settings and the sedatives remained unchanged. Each arterial blood gas analysis was performed after five minutes of stable SpO2. Directly after obtaining the arterial blood sample we performed advanced respiratory measurements (figure 1) as recently described by Bertoni et al.[2] Our first set of measurements was obtained at five levels of SpO2 in order to find a threshold of SpO2 above which our patient’s mechanical ventilation would remain



Figure 1. Measurements of peak airway pressure (Ppeak, left) and delta occlusion pressure (ΔPocc, right) through an expiratory hold. Predicted dynamic transpulmonary driving pressure (PL,dyn) is derived from the formula (Ppeak – PEEP) – 2/3 x ΔPocc. Predicted respiratory muscle pressure (Pmus) is derived from the formula -3/4 x ΔPocc. These formulas and methods were recently described by Bertoni et al. [2]

Netherlands Journal of Critical CareFiO2 to modulate respiratory drive


(predicted) lung-protective. We found this to be at an SpO2 of 96-97% and repeated our measurements twice. In all three sets of measurements, with decreasing FiO2 (and thus CaO2) we observed an increasing respiratory drive, measured as predicted respiratory muscle pressure (Pmus, an indicator of load-induced diaphragm trauma). We also found an increasing predicted dynamic transpulmonary driving pressure (PL,dyn, an indicator of lung stress) with decreasing FiO2. Average percentage changes in Pmus and PL,dyn (SpO2 targets 98-100% compared with 94-95%) were 168.3% and 18.3%. More interestingly, PL,dyn consequently shifted from values regarded as safe to unsafe (≥16-17 cmH2O), as stated by Bertoni et al.[2] Theoretically, this might increase lung stress and the risk of patient self-inflicted lung injury (P-SILI). As a result of - and not the cause of - the increasing respiratory drive, PaCO2 levels decreased with lower CaO2. Our measurements suggest that the respiratory drive in our patient was influenced by CaO2, even if the changes were small. More

remarkably, this relationship was seen even when SpO2 values remained in the normal range (≥94%).

Since excessive respiratory drive is regarded to possibly cause patient self-inflicted lung injury (P-SILI) in selected patients, strategies to modulate respiratory drive have gained recent interest. Modulation of ventilator support is proposed, but also more invasive procedures such as administration of sedation, neuromuscular blocking agents or even extracorporeal CO2 removal.

We appreciate the ongoing debate about oxygen dosing in mechanical ventilation[3] and we are aware of the drawbacks of hyperoxia. In treating ARDS patients, we target oxygenation goals as suggested in the ARDS network trial, i.e. SpO2 88-95.[4] In this patient however, aiming for a higher SpO2 seemed to effectively reduce respiratory drive. A little bit more oxygen supported our

Table 1. Three decremental FiO2 trials with their accompanying clinical parameters, arterial blood gas values and mechanical ventilation measurements, showing lower predicted dynamic transpulmonary driving pressure and predicted respiratory muscle pressure with higher FiO2

First trial Second trial Third trial

Time 16:41h 16:54h 17:09h 17:15h 17:26h 17:49h 18:09h 18:15h 18:20h 18:34h 18:43h

Clinical parameters

Targeted SpO2 (%) 98-100 96-97 95-94 92-93 90-91 98-100 96-97 94-95 98-100 96-97 94-95

Actual SpO2 (%) 99 97 95 93 90 99 97 94 99 97 94

Respiratory rate (breaths/min) 24 24 24 24 28 24 23 25 16 25 17

Tidal volume (ml) 352 372 418 452 421 384 376 426 414 431 477

Arterial blood pressure (mmHg) 138/53 145/56 145/57 146/56 153/57 151/53 146/51 152/53 165/51 119/50 121/50

Heart frequency (beats/min) 96 95 95 96 97 98 99 98 97 103 96

Arterial blood gas values

pH 7.40 7.42 7.43 7.47 7.48 7.42 7.46 7.47 7.43 7.45 7.46

PaCO2 (mmHg) 73 68 66 61 59 68 62 59 66 63 62

PaO2 (mmHg) 97 76 65 52 48 89 69 60 110 69 61

Bicarbonate (mmol/l) 45 44 44 44 44 44 43 43 44 44 44

SaO2 (%) 98 96 94 89 86 97 95 93 99 95 93

Arterial oxygen content (CaO2, ml/dl) 13.4 13.1 12.8 12.1 11.6 13.2 12.9 12.6 13.6 12.9 12.6

Mechanical ventilator measurements

FiO2 0.70 0.60 0.50 0.40 0.38 0.70 0.48 0.40 0.65 0.45 0.40

EtCO2 55 50 47 45 43 50 47 46 55 48 43

Peak (cmH2O) 17.6 15.8 16.5 18 17.7 18 17.6 18.4 17.8 17.9 17.9

PEEPtotal (cmH2O) 5 5 5 5 5 5 5 5 5 5 5

Predicted dynamic transpulmonary driving pressure (PL,dyn, cmH2O)

15.9 17.2 18.4 20.1 22.1 14.2 15.5 16.8 14.9 15.8 18.1

Predicted respiratory muscle pressure (Pmus, cmH2O)

3.8 7.2 7.8 8.0 10.6 1.4 3.3 3.8 2.4 3.2 7.9

Arterial oxygen content was derived from the formula (SaO2 x haemoglobin concentration) x 1.34 + (0.003 x PaO2). Haemoglobin concentration was 6.1 mmol/l

Netherlands Journal of Critical CareFiO2 to modulate respiratory drive


lung-protective strategy, probably with less risk and/or adverse effects then the above-mentioned alternatives.

We found no studies elaborating on the influence of arterial oxygen content on the respiratory drive in ICU patients. We do not know whether our findings are applicable to a larger population and if the reaction to oxygen might be a temporary phenomenon in this patient. More research is needed to answer these questions.

DisclosuresWritten informed consent was obtained from the patient for this publication.

All authors declare no conflict of interest. No funding or financial support was received.

References

1. Jonkman AH, de Vries HJ, Heunks LMA. Physiology of the Respiratory Drive in ICU Patients: Implications for Diagnosis and Treatment. Crit Care. 2020;24:104.

2. Bertoni M, Telias I, Urner M, et al. A novel non-invasive method to detect excessively high respiratory effort and dynamic transpulmonary driving pressure during mechanical ventilation. Crit Care. 2019;23:346.

3. Angus DC. Oxygen Therapy for the Critically Ill. N Engl J Med. 2020;382:1054-6. 4. Acute Respiratory Distress Syndrome Network; Brower RG, Matthay MA, Morris

A, Schoenfeld D, Thompson BT, Wheeler A. Ventilation with Lower Tidal Volumes as Compared with Traditional Tidal Volumes for Acute Lung Injury and the Acute Respiratory Distress Syndrome. N Engl J Med. 2000;342:1301-8.




L E T T E R T O T H E E D I T O R

When to start spontaneous breathing in mechanically ventilated COVID-19 patients? Oxygenation index and PaO2/FiO2 ratio can help

I. Bakker, P.H. Egbers, E.C. Boerma, C. BethlehemDepartment of Intensive Care, Medical Centre Leeuwarden, Leeuwarden, the Netherlands

Correspondence

C. Bethlehem – [email protected]

Keywords - COVID-19, oxygenation index, spontaneous breathing

Dear editor,Patients with COVID-19 induced respiratory failure are initially treated with lung-protective controlled mechanical ventilation to prevent ventilator-induced lung injury.[1] Adequate timing of the switch to a support ventilation mode seems critical. Prolonged sedation and neuromuscular blocking may increase ICU-acquired weakness, whereas premature transition to spontaneous breathing may result in self-inflicted lung injury (P-SILI).[2] Currently, there are no data to predict a successful switch from controlled mechanical ventilation to spontaneous breathing in COVID-19 patients.The oxygenation index (OI) was originally developed to measure severity of illness and predict outcome in neonatal respiratory failure. Previously, OI was also found to predict mortality in adult patients with ARDS.[3] The OI has not been used before to predict the moment of weaning from mechanical ventilation. In contrast to the PaO2/FiO2 (P/F) ratio, the OI is not only based on the fraction of inspired oxygen (FiO2) and partial pressure of oxygen in arterial blood (PaO2), but also integrates the mean airway pressure, which corresponds with static lung compliance.[4] We hypothesise that OI is a predictor for successful transition to spontaneous breathing.

Methods This retrospective single-centre study was performed in a closed-format 29-bed mixed ICU in a tertiary teaching hospital. All patients ≥18 years admitted to the ICU from 15 March to 30 April 2020 with PCR-confirmed COVID-19, and at least one attempt to switch from controlled mechanical ventilation to spontaneous breathing, were included in the study. The study was performed in accordance with the Declaration of Helsinki. According to applicable laws, the need for individual consent was waived by a local ethics committee.

ProtocolAll patients were treated in accordance with a local COVID-19 protocol, including haemodynamic monitoring with pulse-

contour analysis (PiCCO®, Getinge AB, Gothenburg, Sweden). In the first week of ICU admission pressure regulated volume control was the preferred ventilator setting. Sedation with midazolam/fentanyl was titrated to a Richmond Agitation-Sedation Scale level -4 to -5.[5] Additional muscle paralysis was used if deemed necessary to facilitate lung-protective ventilation with tidal volumes of 6 ml/kg ideal body weight and a driving pressure <15 cm H2O. Prone position was applied in patients with a P/F ratio <150. The decision to switch to spontaneous breathing was made by the attending physicians during office hours and only when an improvement was observed in a combination of compliance, P/F ratio and radiological abnormalities, together with a decrease in the extravascular lung water index (EVLWi). The attempt was regarded as unsuccessful if breathing frequency was >30/min, minute volume was >15 l/min, tidal volume >8 ml/kg ideal body weight, P0.1 >5 cm H2O, presence of physical signs of respiratory distress or a decrease in P/F ratio.

Data collectionDemographic characteristics, severity of illness scores over the first 24 hours following ICU admission and the maximum SOFA score during admission were extracted from the patient data management system (Epic®, Verona, Wisconsin, USA). C-reactive protein (CRP), EVLWi, Pulmonary Vascular Permeability Index (PVPi ) and ventilation parameters, including mode of mechanical ventilation, static compliance (Cstat) and P/F ratio, were recorded once daily. OI was calculated for the day of admission and prior to the moment that the first attempt to switch to spontaneous breathing was made. The formula FiO2 (%) x mean airway pressure (mmHg)/PaO2 (mmHg) was used for calculation.

Statistical analysis The Statistical Package for Social Sciences (SPSS 24 for Windows, Chicago, IL, USA) was used for statistical analysis. Comparison between the groups with a successful and an unsuccessful

Netherlands Journal of Critical CareWhen to start spontaneous breathing


first switch to spontaneous breathing was performed with the Mann-Whitney U test. A p value <0.05 was considered statistically significant. For distinctive variables a ROC analysis was executed to test performance and to determine an optimal cut-off value.

ResultsA total of 28 patients with COVID-19 were treated with mechanical ventilation in the ICU between 15 March and 30 April 2020. Twenty-three patients (64% male) had at least one attempt to switch from controlled mechanical ventilation to spontaneous breathing and were included in the study. The median age was 65 years [IQR 59-73] and median BMI 30.4 [IQR 25.6-33.1]. In 10 (44%) of the patients the first switch to spontaneous breathing was unsuccessful and return to pressure regulated volume control ventilation was necessary. On admission, there was no difference between groups in age, sex, BMI, comorbidities, APACHE III, SOFA score or OI and P/F ratio. Also, maximum SOFA scores during admission were similar between groups. Time from admission to first attempt to switch to spontaneous breathing did not differ between the group with a successful first attempt and the group with an unsuccessful first attempt (median 10 [IQR 7-13] versus 9 days [IQR 6-14]; p=0.693). On the day of the first switch, the OI and P/F ratio differed significantly between groups. The median OI in patients with a successful switch to spontaneous breathing was 5.8 [IQR 4.3-7.0] and 7.7 [IQR 5.8-8.4] in patients with an unsuccessful switch; p=0.015. Median P/F ratio was 214 [IQR 201-236] and 164 [IQR 147-206] respectively; p=0.008. The calculated AUC for the OI ROC curve was 0.80 (95% CI 0.61-0.99; p=0.016),

with an optimal cut-off value of <7.5 (figure 1). The AUC for the P/F ROC curve was 0.82 (95% CI 0.63-1.00, p=0.01), with an optimal cut-off value of ≥168 mmHg. There were no differences in Cstat, EVLWi, PVPi or CRP between groups. Furthermore, median Cstat did not change significantly over the first 21 days after admission. The number of days on mechanical ventilation was significantly lower in patients who had a successful first switch to spontaneous breathing (median 10 [IQR 9-19] versus 22 days [IQR 9-34]; p=0.004).

InterpretationBoth the OI and P/F ratio are useful parameters to predict a successful transition from pressure regulated volume control ventilation to spontaneous breathing in patients with COVID-19 induced respiratory failure. Static compliance was not different between the groups and did not change over time in this small cohort of patients. Since there were no differences between groups in disease-severity scores or OI and P/F ratio at baseline, the longer duration of mechanical ventilation in the group with an unsuccessful first attempt might be caused by worsening of lung injury as a result of P-SILI. However, due to the retrospective nature of the study, a causal relationship between duration of mechanical ventilation and a successful switch to spontaneous breathing cannot be confirmed.

We conclude that both OI and P/F ratio may help to predict a successful switch from pressure regulated volume control ventilation to spontaneous breathing in patients with COVID-19 induced respiratory failure. Prospective confirmation of an optimal cut-off value is necessary.


References

1. Slutsky AS, Ranieri M. Ventilator Induced Lung Injury. N Engl J Med. 2013;369:2126-36.2. Cruces P, Retamal J, Hurtado DE, et al. A physiological approach to understand the role

of respiratory effort in the progression of lung injury in SARS-CoV-2 infection. Crit Care. 2020;24:494.

3. Monchi M, Bellenfant F, Cariou A, et al. Early predictive factors of survival in the acute respiratory distress syndrome. A multivariate analysis. Am J Respir Crit Care Med. 1998;158:1076-81.

4. Naik BI, Colquhoun DA, Shields IA, Davenport RE, Durieux ME, Blank RS. Value of the oxygenation index during 1-lung ventilation for predicting respiratory complications after thoracic surgery. J Crit Care. 2017;37:80-4.

5. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med. 2002;166:1338-44.

Figure 1. ROC curves for ventilatory parameters in relation to a successful switch from controlled to support ventilation

Netherlands Journal of Critical CareTitelKlein


COURSES AND CONFERENCES | NVIC

NVIC Basiscursus EchografieTuesday 15 June - Wednesday 16 June 2021Wednesday 8 September - Thursday 9 September 2021Tuesday 9 November - Wednesday 10 November 2021

NVIC Cursus Luchtwegmanagement op de ICMonday 19 April - Tuesday 20 April 2021Tuesday 16 November - Wednesday 17 November 2021

NVIC FCCS cursusTuesday 30 March - Wednesday 31 March 2021Wednesday 14 April - Thursday 15 April 2021Wednesday 19 May - Thursday 20 May 2021Monday 31 May - Tuesday 1 June 2021Wednesday 9 June - Thursday 10 June 2021Tuesday 29 June - Wednesday 30 June 2021Tuesday 6 July - Wednesday 7 July 2021Tuesday 20 July – Wednesday 21 July 2021Tuesday 24 August – Wednesday 25 August 2021Wednesday 22 September - Thursday 23 September 2021Monday 4 October - Tuesday 5 October 2021

Tuesday 12 October - Wednesday 13 October 2021Wednesday 3 November - Thursday 4 November 2021Wednesday 24 November - Thursday 25 November 2021Tuesday 14 December - Wednesday 15 December 2021

NVIC NajaarscongresThursday 16 September 2021

COURSES AND CONFERENCES | OTHER

EuroNeuro 2021 - ParisWednesday 14 April 2021 - Friday 16 April 2021

40th ISICEM - Brussels, BelgiumTuesday 31 August 2021 - Friday 3 September 2021

ESICM LIVES 2021 - Copenhagen, DenemarkSaturday 2 October 2021 - Wednesday 6 October 2021

N V I C C O N F E R E N C E A N D C O U R S E A G E N D A




Jan BakkerProfessor of Intensive CareColumbia University Medical Center, New York University Medical Center, New York, USAPontificia Universidad Católica de Chile, Santiago, ChileErasmus MC University Medical Center, Rotterdam, Netherlands

Charles GomersallDept. of Anaesthesia and IntensiveCareThe Chinese University of HongKong, Prince of Wales HospitalHong Kong, China

Frank van HarenA/ Professor, Australian NationalUniversity Medical SchoolDepartment of Intensive CareMedicineThe Canberra HospitalPO Box 11, Woden, ACT 2606Canberra, Australia

Charles HindsProfessor of Intensive CareMedicineSt. Bartholomew’s HospitalWest Smithfield, London, UK

Patrick HonoréProfessor of ICU MedicineDirector of Critical CareNephrology PlatformICU departmentUniversitair Ziekenhuis Brussel,VUB UniversityBrussels, Belgium

Alun HughesProfessor of Clinical PharmacologyUniversity College LondonLondon, UK

Manu MalbrainDept. of Intensive Care UnitHospital Netwerk AntwerpCampus StuivenbergAntwerp, Belgium

Paul MarikAssociate ProfessorDept. of Medicine and MedicalIntensive Care UnitUniversity of MassachusettsSt. Vincent’s Hospital, Worcester, USA

Greg MartinDept. of MedicineDivision of Pulmonary, Allergy andCritical CareEmory University School ofMedicineAtlanta, USA

Ravindra MehtaProfessor of Clinical MedicineAssociate Chair for ClinicalResearchDepartment of MedicineUCSD Medical CentreSan Diego, USA

Xavier MonnetService de réanimation médicaleCentre Hospitalier Universitairede BicêtreLe Kremlin-Bicêtre, France

Jean-Charles PreiserDept. Intensive CareErasme University HospitalBrussels, Belgium

Yasser SakrDept. of Anaesthesiology andIntensive CareFriedrich-Schiller UniversityHospitalJena, Germany

Hannah WunschDept. of AnaesthesiaNew York Presbyterian ColumbiaNew York, USA

International advisory board

Dirk Donker, Editor in ChiefDept. of Intensive Care Medicine,Div. of Anesthesiology, IntensiveCare and Emergency MedicineUniversity Medical Center UtrechtPO Box 855003508 GA Utrecht

Walter van den Bergh, Section Editor GeneralDept. of Critical CareUniversity of GroningenHanzeplein 19700 RB Groningen

Dennis Bergmans, Section Editor Infection and Inflammation Dept. of Intensive CareMaastricht University Medical Center+P. Debyelaan 256229 HX Maastricht

Frank Bosch, Section Editor ImagingDept. of Internal MedicineRijnstate HospitalPO Box 95556800 TA Arnhem

Diederik van Dijk, Section Editor CardioanesthesiaDept. of Intensive Care Medicine,Div. of Anesthesiology, IntensiveCare and Emergency MedicineUniversity Medical Center UtrechtPO Box 855003508 GA Utrecht

Maarten van Eijk, Associate Section Editor AnesthesiologyDept. of Intensive Care MedicineUniversity Medical Center UtrechtPO Box 855003508 GA Utrecht

Janneke Horn, Section Editor GeneralDept. of Intensive CareAmsterdam UMC location AMCUniversity of AmsterdamMeibergdreef 91105 AZ Amsterdam

Can Ince, Section Editor PhysiologyDept. of PhysiologyAmsterdam UMC location AMCUniversity of AmsterdamMeibergdreef 91105 AZ Amsterdam

Evert de Jonge, Section Editor Scoring and quality assessmentDept. of Intensive Care MedicineLeiden University Medical CenterPO Box 96002300 RC Leiden

Nicole Juffermans, Section Editor Hemostasis and ThrombosisDept. of Intensive CareAmsterdam UMC location AMCUniversity of AmsterdamMeibergdreef 91105 AZ Amsterdam

Jozef Kesecioglu, Section Editor EthicsDept. of Intensive Care MedicineUniversity Medical Center UtrechtPO Box 855003508 GA Utrecht

Michael Kuiper, Section Editor NeurologyDept. of Intensive Care MedicineMedical Center LeeuwardenPO Box 8888901 BR Leeuwarden

Nuray Kusadasi, Associate Section Editor Hemato-OncologyDept. of Intensive Care MedicineUniversity Medical Center UtrechtPO Box 855003508 GA Utrecht

Christiaan Meuwese, Associate Section Editor CardiologyDept. of Intensive Care MedicineUniversity Medical Center UtrechtPO Box 855003508 GA Utrecht

Marike van der Schaaf, Section Editor RehabilitationDept. of RehabilitationAmsterdam UMC location AMCUniversity of AmsterdamMeibergdreef 91105 AZ Amsterdam

Ilse van Stijn, Managing EditorDept. of Intensive Care MedicineOLVGPO Box 955001090 HM Amsterdam

Eleonora Swart, Section Editor Pharmacology Dept. of PharmacyAmsterdam UMC location AMCUniversity of AmsterdamMeibergdreef 91105 AZ AmsterdamDept. of Clinical Pharmacology and Pharmacy Amsterdam UMC location VumcDe Boelelaan 11171081 HV Amsterdam

Pieter Roel Tuinman, Section Editor GeneralDept. of Intensive Care MedicineAmsterdam UMC location VumcPO Box 70571007 MB Amsterdam

David van Westerloo, Section Editor GeneralDept. of Intensive Care MedicineLeiden University Medical CenterPO Box 96002300 RC Leiden

Job van Woensel, Section EditorPediatricsPediatrics Intensive Care UnitEmma children’s hospitalAmsterdam UMC location AMCUniversity of AmsterdamMeibergdreef 91105 AZ Amsterdam

Editorial board of the Netherlands Journal of Critical Care



Information for authors

The Netherlands Journal of Critical Care (NJCC) is the official journal of the Dutch Society of Intensive Care (Nederlandse Vereniging voor Intensive Care-NVIC). The journal has a circulation of about 1750 copies bimonthly in the Netherlands and Belgium.High-quality reports of research related to any aspect of intensive care medicine, whether laboratory, clinical, or epidemiological, will be considered for publication in the NJCC. This includes original articles, reviews, and meta-analyses. Case reports, clinical images, book reviews, editorials, letters to the editor, clinical problem solving, research news and correspondence are also welcome. All manuscripts pass through an independent review process managed by the editorial board. The journal does not have any publication fees, and colour figures are reproduced free of charge.The journal is indexed by Embase, Emcare and Scopus. A Medline annotation is in preparation.

Ethical standardsManuscripts reporting original research must contain a statement that all human and animal studies have been approved by the appropriate ethics committee and have therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Where appropriate, it should also be stated in the text that all persons gave their informed consent prior to their inclusion in the study. If possible, the Journal aims to also include informed consent of individuals described in case reports. Details that might disclose the identity of the subjects under study should be omitted. The editors reserve the right to reject manuscripts that do not comply with the above-mentioned requirements. The author will be held responsible for false statements or failure to fulfil the above-mentioned requirements. The editors adhere to the guidelines laid down by the International Committee of Medical Journalist Editors (www.icmje.org), concerning authorship and scientific conduct.

Types of papers The following manuscript types are considered for publication: original articles, review articles, systematic reviews, meta-analyses, editorials, case reports, clinical images, book reviews, letters to the editor, clinical problem solving, research news and correspondence.

Original articles Original articles should not exceed 3000 words (excluding abstract, references, tables and legends). The manuscript should be clear in outline (with subheadings) for maximum clarity. The text should follow the IMRAD format and contain an abstract, introduction, materials (patients) and methods, results, discussion section and references. This should be followed by tables and figures (maximum of 6 in total) with titles above and legends below these elements. The abstract should not exceed 250 words and should be structured: background, methods, results, conclusions. Do not include references in the abstract. Authors should provide a minimum of 3 keywords, a running title, and a list of not more than 30 references.Original articles must meet the following criteria:- the manuscript presents the results of primary scientific research; - the results have not been published in full elsewhere; - analyses are described in full in the manuscript; - conclusions are presented in a clear and concise manner and are supported

by the data; - the research meets all applicable ethical standards; - the article adheres to appropriate reporting guidelines and community

standards for full data disclosure; - when reporting the results of a randomised controlled trial, author(s)

should use the CONSORT statement as a guide to preparing the manuscript (http://www.consort-statement.org/);

- conflicts of interest should be clearly stated in the manuscript (see below).

The authors are encouraged to refer to national and international registries of trials in their papers (such as clinicaltrials.gov), where applicable.

Review articlesReview articles, systematic reviews and meta-analyses should not exceed 3000 words (excluding abstract, references, tables and legends). The manuscript should contain subheadings. A maximum of 6 tables and figures (in total) are allowed. Please provide titles above and legends below these elements. The abstract should not exceed 250 words and should be structured: background, methods, results, conclusions, with the exception of a non-systematic review, which may contain a non-structured abstract. No references should be included in the abstract. Authors should provide a minimum of 3 keywords, a running title, and a list of not more than 70 references. The authors are encouraged to refer to national and international registries of trials in their papers (such as clinicaltrials.gov), where applicable.

EditorialsEditorials are always commissioned by the Editors and comment on one or more articles in the same issue of the Journal or to a subject with high news value. Editorials should not exceed 1500 words and may include up to 15 references. Editorials have a maximum of 3 authors and no abstract. Please provide 2-3 key words.

Case reportsThe text of a case report should include an abstract, introduction, case report/case history, discussion, tables and figures (2 in total), and references. The main text may be up to 2000 words; the abstract should not exceed 150 words and may be unstructured. Please provide a minimum of 3 keywords and a list of not more than 30 references. Please include an informed consent statement from the patient described in the case.

Clinical problem-solving These manuscripts consider the step-by-step process of clinical decision-making. Information about a patient is presented to an expert clinician or clinicians in stages (indicated by boldface type in the manuscript) to simulate the way such information emerges in clinical practice. The clinician responds (in regular type) as new information is presented, sharing his or her reasoning with the reader. The text should not exceed 2500 words, and there should be no more than 15 references. Please include an informed consent statement from the patient described in the case.

Research newsResearch news should be a review of a manuscript which has appeared in the past two months. It contains sections on why this study was done, the research question, how this was investigated, conclusions and the impact of the study on clinical practice. The text should not exceed 800 words with a maximum of 5 references. Contributions for this section will be commissioned; however, inquiries about contributions can be sent to [email protected].

Clinical imagesA clinical image should contain one or two pictures with a legend and a short case history, and should preferably not be referenced. The manuscript should succinctly present relevant clinical information, including a short description of the patient’s history, relevant physical and laboratory findings, clinical course, response to treatment (if any), and condition at last follow-up. Please provide a minimum of 3 keywords. The text should not exceed 500 words. Please include an informed consent statement from the patient described in the case.

Photo quizIn this section relevant images for critical care medicine (e.g. flow and pressure curves of mechanical ventilation or haemodynamic indices, radiological images or laboratory results) will be accompanied by a short introduction of the context. The introduction will be followed by ‘what is your diagnosis?’. The answer will include a brief discussion of the literature. A photo quiz should not exceed 500 words and contain no more than two figures, and five references conform the Vancouver style. Abbreviations of measurements should be quoted in SI units.

Book reviewsA book review should not exceed 300 words. Please mention in the header: title, author, edition and year. Scan the cover in high resolution (300 dpi/1 mb) and send with the text. With an online review, the cover can usually be downloaded. Details with the cover: title, author, edition, year, publisher, number of pages, price and ISBN number. Conclude with the name and affiliation(s) of the reviewer.

Letters to the editorLetters to the editor provide an opportunity to present results of scientific value where a short format is most appropriate. They should not exceed 1000 words, 5 references and 1 figure or table.

CorrespondenceCorrespondence provides an opportunity to debate published articles. This should not exceed 500 words, 5 references and 1 figure or table. Correspondence is sent to the authors for rebuttal, and a final decision on publication is made at the end of this process, by the editor.

General informationEach manuscript should be accompanied by a cover letter stating the following: the complete postal address, email address and telephone number of the corresponding author and, if it is a re submission, the previous Neth J Crit Care number and year. The language of the journal is British English. Authors who are unsure of proper English usage should have their manuscript checked by someone proficient in the English language. All text should be double spaced. The manuscript pages, including references and legends, should be sequentially numbered throughout



General guidelines on house style- The title of the manuscript should be in typeface Times New Roman, size 20.

With the exception of the first word and proper nouns, initial capitals are not used in the title.

- The names of departments should be in typeface Times New Roman, size 12.- The names of hospitals should be written in English.- Write ‘the Netherlands’, without capitalising the t.- Generally, abbreviations should not be used in the title (see Table of standard

abbreviations for exceptions).- The corresponding author only provides his/her email address on the title page.- Please provide a minimum of three keywords and a running title.- The abstract of original and review articles should be written in a structured format.- Unstructured abstracts should take the form of a single paragraph.- Headings must be in bold. Use no more than two levels of headings. - Paragraphs starting immediately under headings and subheadings should begin

at the left margin. Subsequent paragraphs should be indented. - Non-standard abbreviations (see table of standard abbreviations) should always

be explained and their use kept to a minimum.- Use British English spelling – except in titles of institutions that have chosen to

use US spelling, e.g. Academic Medical Center, Amsterdam. Examples: anaemia (instead of anemia), oesophagus (instead of esophagus), litre (instead of liter), colour (instead of color), labelling (instead of labeling), practice (noun), and practise (verb). This should be used consistently. Use the s-form spelling, e.g. minimise, randomisation.

- Do not use full stops in initials, abbreviations and academic titles.- References are numbered sequentially in the text and placed in square brackets

after the punctuation. [..]- Genus names should be written in italics, e.g. Staphylococcus aureus, S. aureus.- Numbers under 10 are spelled out except for measurements with a unit (10

mmol/l) or age (4 weeks old), or when in a list with other numbers (5 mice, 6 rats, 12 gerbils).

- When referring to tables or figures in the text, use italics; do not use a capital letter, e.g. see table 2.

TablesTables are to be numbered independently of the figures with Arabic numbers and are uploaded as separate documents.- Tables should be laid out in Word, using the table function. Other tables (e.g.

in pdf format or PowerPoint) will not be accepted;- Do not use internal horizontal or vertical lines;- Do not use spaces, tabs or hard returns in tables;- Each piece of data must be contained in its own cell;- Numbers and percentages are presented in the same cell;- Tables should always be cited in the text in consecutive numerical order;- For each table, please supply a title explaining the components of the table;- Any abbreviations used in the table must be defined in a legend; - Tables should not exceed the printed area of the page (174 x 234 mm).

FiguresFigures should also be numbered with Arabic numbers and are uploaded in separate documents. Legends should be given in the document that contains the text, references, and tables. Authors wishing to include figures or tables that have already been published elsewhere are required to obtain permission from the copyright owner and provide evidence that such permission has been granted when submitting their paper. Colour figures can be published. Short, clear legends make additional description in the text unnecessary. Figures should be provided in electronic format (TIFF or JPEG).

Conflict of interest Authors must indicate any conflict of interest. This includes a financial relationship with an organisation that sponsored the research (funding, speakers fee, consultancy fee), management relations with the organisation that sponsored the research (consultant, member of board). All sources of funding obtained for the research should also be stated. A conflict of interest statement can be downloaded from the website. The completed and signed form should be uploaded as a separate document when submitting the manuscript. If no conflict exists, authors should state: All authors declare no conflicts of interest. No funding or financial support was received.

Author agreement All authors must certify they have seen and approved the manuscript being submitted. All authors warrant that the article is the authors’ original work, has not been published previously and is not under consideration for publication elsewhere. The author agreement form can be downloaded from the website. An article will only be published when this form is completed, signed and returned.

CopyrightCopyright ownership is to be transferred in a written statement, which must accompany all manuscript submissions and must be signed by all authors. The agreement should state: ‘The undersigned authors transfer all copyright

ownership of the manuscript (title of article) to the Netherlands Journal of Critical Care’. Any relevant papers that may be considered as duplicating in part the current submission should be reported.

ReferencesNeth J Crit Care uses the Vancouver style of referencing. Only articles cited in the text are to be listed. They should be arranged in order of appearance in the text and numbered consecutively. Only the reference number should appear in the text between brackets. […] Include all author names (unless there are more than six, in which case abbreviate to three and add ‘et al.’), and page numbers. Use the Medline abbreviation for names of journals.Article in journals: Calandra T, Cometta A. Antibiotic therapy for Gram-negative bacteremia. Infect Dis Clin North Am. 1991;5:817-34.Books (sections): Thijs LG. Fluid therapy in septic shock. In: Sibbald WJ, Vincent J L (eds). Clinical trials for the treatment of sepsis. (Update in intensive care and emergency medicine, volume 19). Berlin Heidelberg New York: Springer; 1995, pp 167-90.Conference meetings: Rijneveld AW, Lauw FN, te Velde AA, et al. The role of interferon gamma in murine pneumococcal pneumonia. 38th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC). San Diego, Ca.; 1998, pp 290.Article/ information on website: Samenwerken aan duurzame zorg, landelijke monitor proeftuinen: Rijksinstituut voor Volksgezondheid en Milieu, 2015. (Accessed 11 August 2015, at http://www.rijksoverheid.nl/documenten-en-publicaties/rapporten/2015/07/10/samen-werken-aan-duurzame-zorg.html.)

How to submitPlease submit manuscripts directly to the Editorial Office through our online submission system at www.njcc.nl.

Peer reviewAll papers are subject to a peer-review system handled by the editors. Authors are encouraged to resubmit, when invited, the revised paper within two weeks after the editorial decision. The changes made in the revised paper should be highlighted and the manuscript accompanied by a letter with a point-to-point rebuttal.

ProofsThe corresponding author will receive proofs of accepted papers by email. Corrected proofs should be returned within 48 hours of receipt.

Production processDecisions of the editors are final. All material accepted for publication is subject to copyediting. The Neth J Crit Care reserves the right to edit for house style, clarity, precision of expression, and grammar. Authors review these changes at the proof stage but must limit their alterations in the proof to correcting errors and to clarifying misleading statements.

Table of commonly used abbreviationsAIDS acquired immunodeficiency syndromeALI acute lung injuryARDS adult respiratory distress syndromeAPACHE acute physiology and chronic health evaluationBIPAP biphasic positive airways pressureCCU coronary care unitCOPD chronic obstructive pulmonary diseaseCPAP continuous positive airway pressureCT computed tomographyECG electrocardiogramECMO extracorporeal membrane oxygenationEEG ElectroencephalogramELISA enzyme-linked immunosorbent assayETCO2 end-tidal carbon dioxideHIV human immunodeficiency virusIC intensive careICU intensive care unitIM IntramuscularINR international normalised ratioIPPV intermittent positive pressure ventilationIV IntravenousMAP mean arterial pressureMODS multiorgan dysfunction syndromeMRI magnetic resonance imagingPACU post anaesthesia care unitPEEP positive end expiratory pressurePET positron emission tomographySARS severe adult respiratory syndromeSIRS systemic inflammatory response syndromeSOFA sequential organ failure assessmentSPECT single-photon emission computed tomography TIA transient ischaemic attackTRALI transfusion-related acute lung injury

Documents

Netherlands Journal of Critical Care - NJCC 2021-2 Vol... · 2021. 3. 19. · Netherlands Journal of Critical Care 76 NETH J CRIT CARE- VOLUME 29 - NO 2 - MARCH 2021 Submitted February