Pericardiocentesis practice in the United Kingdom

Pericardiocentesis practice in the United Kingdom

S. Balmain,1 N. M. Hawkins,2 M. R. MacDonald,3 F. G. Dunn,2 M. C. Petrie3

Introduction

In individual centres pericardiocentesis is a relatively

infrequent procedure. Whether or not to drain an

effusion is often a difficult clinical decision (1,2).

Methods of drainage depend on individual physi-

cian’s experience and preference, as well as available

facilities. Assessment of ‘real-life’ practice is limited.

Who performs pericardiocentesis, their procedural

preferences, and the perceived indications for drain-

age are unknown. We performed a survey to evaluate

the practice of pericardiocentesis among UK cardiol-

ogists.

Methods

In March 2003, 640 questionnaires (Appendix 1) were

sent to all cardiology consultants, specialist registrars

and staff grades in the UK Directory of Cardiology

(2003 edition). District general hospitals (DGHs) and

tertiary referral centres (TRCs) received similar num-

bers of questionnaires (344 vs. 296 respectively). Car-

diologists were asked to report their pericardiocentesis

experience within the preceding 2 years, focussing

particularly on indications for pericardiocentesis and

procedural preferences. We also asked respondents to

report any procedure-related complications and to

provide details of the aetiology of pericardial effusion

if available. Data were collated and analysed using

Access and Excel software (Windows XP Professional;

Microsoft Corporation, Seattle, WA).

Results

RespondentsA total of 274 (43%) completed questionnaires were

returned: 88% consultants, 9% specialist registrars

and 3% staff grades. Respondents were equally dis-

tributed between TRCs and DGHs (49% and 51%

respectively). A maximum of three questionnaires

SUMMARY

Background: Pericardial effusions frequently present challenging clinical dilem-

mas. Whether or not to drain an effusion, and if so by what method, are two

common decisions facing cardiologists. We performed a survey to evaluate pericar-

diocentesis practice in the United Kingdom (UK). Methods: A total of 640 ques-

tionnaires were sent to all cardiologists in the UK Directory of Cardiology in March

2003. Results: A total of 274 (43%) completed questionnaires were returned,

88% from consultants, equally distributed between tertiary referral centres and dis-

trict general hospitals. More than 1500 procedures were performed, largely using

a paraxiphoid approach (89%). Clinical tamponade was the commonest indication

for pericardiocentesis (83%). However, the majority of respondents (69%) consid-

ered echocardiographic features alone an indication for pericardiocentesis, even in

the absence of clinical tamponade. The commonest perceived indications for drain-

age were right ventricular diastolic collapse and right atrial collapse (69% and

33% of respondents respectively). For guidance, 82% use echocardiography, either

alone or with fluoroscopy or the electrocardiogram (ECG) injury trace. 11% employ

fluoroscopy alone or with the ECG injury trace. The remaining 11% stated that

they would use the ECG injury trace alone or use no guidance. Using the ECG

injury trace alone is said by the European Society of Cardiology (ESC) guidelines to

offer an inadequate safeguard. Reported complications included ventricular punc-

ture (n ¼ 12, 0.8%) and hepatic damage (n ¼ 4, 0.3%). Conclusion: Pericardio-

centesis practice varies substantially in the UK. Many cardiologists would perform

pericardiocentesis based on echocardiographic features alone. 11% of cardiologists

use guidance that is considered inadequate by the ESC guidelines.

What’s knownVery little is known about this topic. Although

there are ESC guidelines on the management of

pericardial disease, there are no published data on

pericardiocentesis practice in the UK.

What’s newThis article adds insight into the variability of

current pericardiocentesis practice in the UK, and

that available practice guidelines are not adhered

to.

1Department of Cardiology,

New Royal Infirmary of

Edinburgh, Edinburgh, UK2Department of Cardiology,

Stobhill Hospital, Glasgow, UK3Department of Cardiology,

Royal Infirmary, Glasgow, UK

Correspondence to:

Dr Sean Balmain,

Department of Cardiology, New

Royal Infirmary of Edinburgh,

51 Little France Crescent,

Edinburgh EH16 4SA, UK

Tel.: + 447811113015

Fax: + 448717142621

Email:

[email protected]

Disclosures

The authors have stated that

they have no interests that

might be perceived as posing a

conflict or bias.

doi: 10.1111/j.1742-1241.2007.01536.x

OR IG INAL PAPER

ª 2007 The AuthorsJournal compilation ª 2007 Blackwell Publishing Ltd Int J Clin Pract, October 2008, 62, 10, 1515–1519 1515

were returned from any single centre, reflecting the

large number of hospitals which responded.

Who performs pericardiocentesis?More than 1500 pericardiocentesis procedures were

performed over the 2-year audit period, 56% in

TRCs and 44% in DGHs. Procedures were more fre-

quently performed by a trainee under supervision in

TRCs than DGHs (47% vs. 33% respectively). Most

cardiologists (81%) do not believe that on-site car-

diothoracic surgical support is necessary. Of those

not requiring surgical support, 42% work in TRCs

and 58% in DGHs.

Indications for pericardiocentesisThe commonest reason for pericardiocentesis was

clinical tamponade (83% of procedures). 10% were

undertaken for diagnostic purposes. Only 7% were

performed based on echocardiographic appearances

alone. However, the majority of respondents (69%)

regarded echocardiographic features an indication for

pericardiocentesis, even in the absence of clinical

tamponade. Features considered an indication for

drainage were: right ventricular diastolic collapse

(69% of respondents); right atrial collapse (33%);

respiratory variation of transvalvular Doppler flow

(7%) and large effusion size (3%).

Procedural preferencesA paraxiphoid approach was most popular (89%),

with the remaining operators preferring an apical

approach. For guidance (Table 1), 78% routinely use

echocardiography alone or in conjunction with fluo-

roscopy, the electrocardiogram (ECG) injury trace or

both. 11% of respondents employ fluoroscopy alone

or in combination with the ECG injury trace. The

remaining 11% use the ECG injury trace alone, or

use no guidance. 10% routinely inject contrast into

the pericardial space.

ComplicationsThe most commonly reported complications were

ventricular puncture [n ¼ 12 (0.8%), nine right ven-

tricular, one left ventricular and two unspecified]

and hepatic damage (n ¼ 4, 0.3%). Right coronary

artery laceration was reported once, as was subphren-

ic haematoma and splenic vein perforation. One

peri-procedure death was reported, but related to

progressive cardiogenic shock rather than pericardio-

centesis. Few procedures resulted in failed drainage

(n ¼ 10, 0.7%).

Aetiology of pericardial effusionAetiology was ascertained in only 26% of cases.

Malignancy predominated (45%), followed by car-

diac surgery (21%), viral infection (13%), autoim-

mune disorders (7%), myocardial infarction (5%),

uraemia (5%) and tuberculous infection (3%).

Discussion

This is the first national survey of pericardiocentesis.

The European Society of Cardiology (ESC) has since

published guidelines on the diagnosis and manage-

ment of pericardial disease, including pericardiocen-

tesis (3).

Indications for pericardiocentesisClinical tamponade was the primary indication for

pericardiocentesis. This is a consistent finding

worldwide (4–6). The ESC guidelines state that

pericardiocentesis is mandatory in tamponade (class

I indication). The guidelines are less clear when

considering whether or not large effusions should

be drained. They initially state that pericardiocente-

sis is optional in large or recurrent effusions, or if

previous tests are inconclusive (class IIa indication),

or in small effusions (class IIb indication) (all level

of evidence B) (3). The guidelines subsequently

contradict this statement, recommending pericardio-

centesis for effusions > 20 mm in diameter by echo-

cardiography (class IIa indication). The latter

recommendation cites a report by Sagrista-Sauleda

et al. (7) of 28 patients with large idiopathic

chronic pericardial effusion. However, this subgroup

represented just 2.2% of patients with pericardial

disease (7). The same authors do not advocate peri-

cardial drainage in the initial management of

patients with large pericardial effusion without clini-

cal tamponade (2). The ESC guidelines do not dis-

tinguish between different indications for

pericardiocentesis in acute and chronic effusions

(3). Clarification of this issue (i.e. whether or not

to drain large effusions) would be welcome when

the ESC revises the guidelines.

Table 1 Procedural preferences

Guidance method Respondents (%)

Echocardiography alone or in combination 214 (78)

Echocardiography alone 86 (31)

With fluoroscopy 85 (31)

With ECG injury trace 27 (10)

With fluoroscopy and ECG injury trace 16 (6)

Fluoroscopy alone or with ECG injury trace 30 (11)

Fluoroscopy alone 24 (9)

With ECG injury trace 6 (2)

ECG injury trace alone 5 (2)

No guidance 25 (9)

1516 Pericardiocentesis practice in the UK

ª 2007 The AuthorsJournal compilation ª 2007 Blackwell Publishing Ltd Int J Clin Pract, October 2008, 62, 10, 1515–1519

The perceived and actual indications for pericar-

diocentesis differed substantially. The majority of

respondents (69%) considered echocardiographic fea-

tures an indication for drainage, even in the absence

of clinical tamponade. However, few were actually

performed on the basis of echocardiography alone

(7%). The discrepancy likely reflects uncertainty

regarding the indications for drainage, and caution

in ‘real life’ clinical practice. Cardiologists may be

concerned by ‘echocardiographic’ tamponade, but

only commit to an invasive procedure when clinical

compromise dictates.

Although the ESC guidelines list echocardiographic

and clinical criteria for the diagnosis of ‘cardiac

tamponade’, whether or not echocardiographic

appearances alone merit intervention is not discussed

(3). Some have reported strong correlations between

chamber collapse, particularly of the right ventricle

and clinical tamponade (8–14). However, these stud-

ies involved only small numbers of patients, often

with established clinical tamponade (10–14), or

experimental animal models (8,9). Most large series

fail to distinguish between echocardiographic and

clinical tamponade, performing pericardiocentesis in

both (5,15,16). Consequently, the evolution of con-

servatively managed patients with echocardiographic

but not clinical tamponade is uncertain. Pericardio-

centesis is of doubtful benefit in such patients. The

only group to address this question is Merce et al.

(1,17) They found a poor correlation between clinical

tamponade and right-sided chamber collapse (17),

and that clinical outcomes related to underlying aeti-

ology rather than whether or not pericardiocentesis

was performed (1). Indeed, irrespective of chamber

collapse or effusion size, no conservatively managed

patients with nonmalignant effusion developed clini-

cal tamponade, few effusions persisted, and no new

diagnoses occurred during follow-up (1,17). Fewer of

our respondents believed that right atrial (33%)

compared with right ventricular (69%) collapse was

an indication for drainage.

Procedural preferencesA paraxiphoid approach was most popular, as in

many centres (6,15,16). Echocardiography alone, or

in combination with fluoroscopy, was used most fre-

quently to guide pericardiocentesis. 48% of respon-

dents use fluoroscopy and 20% the ECG injury trace

to guide pericardiocentesis. Historically, these meth-

ods were associated with a high complication rate

when used without echo guidance (18). However, in

our survey, the majority of respondents combine

fluoroscopy or the ECG injury trace with echocardi-

ography. Surprisingly, 22% do not use echocardiog-

raphy: 11% prefer fluoroscopy, 2% employ the ECG

injury trace alone, while 9% reported routinely using

no guidance. The ESC guidelines advocate a subxi-

phoid approach guided by either echocardiography

or fluoroscopy, as employed by the majority of

respondents (3). They explicitly state that the ECG

injury trace alone is not an adequate safeguard (3).

19% of respondents consider on-site surgical back up

a necessity.

ComplicationsThe low-reported complication rate is comparable

with large series, including a similar frequency of

ventricular puncture (0.8%) (5,15,16).

AetiologyRespondents reported a low diagnostic yield of peri-

cardiocentesis (< 30%), which is consistent with pre-

vious reports (4,19,20), as is the predominance of

neoplastic aetiology (5,6,15,20,21). The aetiological

spectrum in different series reflects geographical loca-

tion, underlying disease prevalence, hospital charac-

teristics, effusion size and the applied study protocol

(2). Consequently, some series have cited acute idio-

pathic pericarditis (4) and cardiac surgery (5,16) as

the commonest cause. The low-reported frequency of

tuberculous effusion contrasts with studies from

areas with a higher prevalence of tuberculosis infec-

tion (6,15). The relatively low incidence of uraemic

effusion likewise differs from some reports

(20,22,23), reflecting the lower prevalence of end-

stage renal disease and on-site haemodialysis facilities

in many DGHs.

Study limitationsThe modest response rate (43%) may misrepresent

pericardiocentesis practice throughout the UK. In

particular, inexperienced cardiologists or those with

serious complications may not respond. Anonymity

hopefully reduces such bias. The survey methodology

was limited in that we did not telephone or re-send

questionnaires to non-responders, which may have

improved the response rate. Information recalled

from memory is less robust than data from prospec-

tive studies or large registries. Self-reporting of com-

plications is particularly prone to bias.

Epidemiological data, such as effusion aetiology, is

subject to recall error, and not the focus of the study.

In this respect, the concordance of results with exist-

ing published series is surprising. The main interest

lies in the actual and perceived indications for drain-

age and procedural preferences. Given the relatively

small number of procedures per respondent, this

data may be accurate. We must acknowledge that

objectively reviewing patient records would improve

accuracy.

Pericardiocentesis practice in the UK 1517


Reporting subjective clinical decisions through

objective data has limitations. Risks are balanced

against the degree of haemodynamic compromise,

aetiology and anatomical position of the effusion

and potential benefits. Finally, some information that

may have proved interesting was not collected. We

failed to ask where pericardiocentesis was performed,

or whether intrapericardial pressure was routinely

measured.

Conclusion

Pericardiocentesis practice varies considerably in the

UK. Although clinical tamponade is the commonest

indication for pericardiocentesis, 69% of clinicians

would drain an effusion based on echocardiographic

findings alone. The value of this approach is uncer-

tain. 11% of cardiologists use techniques which are

considered unsafe by the ESC. Pericardiocentesis

practice should adhere to the ESC guidelines. This

survey predates these guidelines and should be

repeated to determine whether clinical practice has

changed.

References

1 Merce J, Sagrista-Sauleda J, Permanyer-Miralda G, Soler-Soler J.

Should pericardial drainage be performed routinely in patients

who have a large pericardial effusion without tamponade? Am J

Med 1998; 105: 106–9.

2 Soler-Soler J, Sagrista-Sauleda J, Permanyer-Miralda G. Manage-

ment of pericardial effusion. Heart 2001; 86: 235–40.

3 Maisch B, Seferovic PM, Ristic AD et al. Guidelines on the diagno-

sis and management of pericardial diseases executive summary;

The Task force on the diagnosis and management of pericardial

diseases of the European society of cardiology. Eur Heart J 2004;

25: 587–610.

4 Sagrista-Sauleda J, Merce J, Permanyer-Miralda G, Soler-Soler J.

Clinical clues to the causes of large pericardial effusions. Am J Med

2000; 109: 95–101.

5 Tsang TS, Enriquez-Sarano M, Freeman WK et al. Consecutive

1127 therapeutic echocardiographically guided pericardiocenteses:

clinical profile, practice patterns, and outcomes spanning 21 years.

Mayo Clin Proc 2002; 77: 429–36.

6 Gibbs CR, Watson RD, Singh SP, Lip GY. Management of pericar-

dial effusion by drainage: a survey of 10 years’ experience in a city

centre general hospital serving a multiracial population. Postgrad

Med J 2000; 76: 809–13.

7 Sagrista-Sauleda J, Angel J, Permanyer-Miralda G, Soler-Soler J.

Long-term follow-up of idiopathic chronic pericardial effusion. N

Engl J Med 1999; 341: 2054–9.

8 Leimgruber PP, Klopfenstein HS, Wann LS, Brooks HL. The he-

modynamic derangement associated with right ventricular diastolic

collapse in cardiac tamponade: an experimental echocardiographic

study. Circulation 1983; 68: 612–20.

9 Klopfenstein HS, Schuchard GH, Wann LS et al. The relative

merits of pulsus paradoxus and right ventricular diastolic

collapse in the early detection of cardiac tamponade: an

experimental echocardiographic study. Circulation 1985; 71: 829–

33.

10 Kronzon I, Cohen ML, Winer HE. Diastolic atrial compression: a

sensitive echocardiographic sign of cardiac tamponade. J Am Coll

Cardiol 1983; 2: 770–5.

11 Engel PJ, Hon H, Fowler NO, Plummer S. Echocardiographic

study of right ventricular wall motion in cardiac tamponade. Am J

Cardiol 1982; 50: 1018–21.

12 Armstrong WF, Schilt BF, Helper DJ et al. Diastolic collapse of the

right ventricle with cardiac tamponade: an echocardiographic

study. Circulation 1982; 65: 1491–6.

13 Singh S, Wann LS, Schuchard GH et al. Right ventricular and right

atrial collapse in patients with cardiac tamponade–a combined

echocardiographic and hemodynamic study. Circulation 1984; 70:

966–71.

14 Singh S, Wann LS, Klopfenstein HS et al. Usefulness of right ven-

tricular diastolic collapse in diagnosing cardiac tamponade and

comparison to pulsus paradoxus. Am J Cardiol 1986; 57: 652–6.

15 Cho BC, Kang SM, Kim DH et al. Clinical and echocardiographic

characteristics of pericardial effusion in patients who underwent

echocardiographically guided pericardiocentesis: Yonsei Cardiovas-

cular Center experience, 1993–2003. Yonsei Med J 2004; 45: 462–8.

16 Lindenberger M, Kjellberg M, Karlsson E, Wranne B. Pericardio-

centesis guided by 2-D echocardiography: the method of choice

for treatment of pericardial effusion. J Intern Med 2003; 253: 411–

7.

17 Merce J, Sagrista-Sauleda J, Permanyer-Miralda G et al. Correla-

tion between clinical and Doppler echocardiographic findings in

patients with moderate and large pericardial effusion: implications

for the diagnosis of cardiac tamponade. Am Heart J 1999; 138:

759–64.

18 Wong B, Murphy J, Chang CJ et al. The risk of pericardiocentesis.

Am J Cardiol 1979; 44: 1110–4.

19 Permanyer-Miralda G, Sagrista-Sauleda J, Soler-Soler J. Primary

acute pericardial disease: a prospective series of 231 consecutive

patients. Am J Cardiol 1985; 56: 623–30.

20 Corey GR, Campbell PT, Van Trigt P et al. Etiology of large peri-

cardial effusions. Am J Med 1993; 95: 209–13.

21 Olsen PS, Sorensen C, Andersen HO. Surgical treatment of large

pericardial effusions. Etiology and long-term survival. Eur J Car-

diothorac Surg 1991; 5: 430–2.

22 Colombo A, Olson HG, Egan J, Gardin JM. Etiology and prognos-

tic implications of a large pericardial effusion in men. Clin Cardiol

1988; 11: 389–94.

23 Levine MJ, Lorell BH, Diver DJ, Come PC. Implications of echo-

cardiographically assisted diagnosis of pericardial tamponade in

contemporary medical patients: detection before hemodynamic

embarrassment. J Am Coll Cardiol 1991; 17: 59–65.

Paper received May 2007, accepted July 2007

Appendix 1: Survey questions

1. What grade are you?

2. Where do you work (district general hospital or

tertiary referral centre)?

3. How many pericardiocentesis procedures have you

performed in the past 2 years?

4. How many pericardiocentesis procedures have you

supervised in the past 2 years?

5. Do you feel that cardiothoracic surgical support is

necessary to perform pericardiocentesis?

6. Regarding the indication for pericardiocentesis,

how many procedures were performed for (a) diag-

nostic purposes, (b) echocardiographic appearances,

(c) clinical tamponade/haemodynamic compromise?

7. How many procedures yielded a diagnosis?

8. Of those procedures which yielded a diagnosis,

how many were due to: (a) trauma, (b) malignancy,

1518 Pericardiocentesis practice in the UK


(c) renal disease, (d) viral infection, (e) tuberculosis,

(f) autoimmune disease, (g) postmyocardial infarc-

tion, (h) endocrine disease, (i) other aetiology (please

specify)?

9. Which of the following do you routinely use to

guide pericardiocentesis? (a) echocardiography, (b)

fluoroscopy, (c) injection of contrast into the peri-

cardium, (d) the ECG injury trace, (e) others (please

specify)

10. What size of pericardial effusion (in mm) would

you consider to be unsuitable for drainage?

11. In the absence of clinical tamponade, which of the

following echocardiographic features would you con-

sider to be an indication for pericardiocentesis? (a) right

atrial collapse, (b) right ventricular diastolic collapse,

(c) other echocardiographic feature (please specify)

12. Which anatomical approach do you favour for

pericardiocentesis? (a) subxiphoid, (b) apical, (c)

other approach (please specify)

13. Did you experience any complications of pericar-

diocentesis over the past 2 years? If so please provide

details.

Pericardiocentesis practice in the UK 1519


Documents

Pericardiocentesis practice in the United Kingdom