Correspondence

568 www.thelancet.com/infection Vol 13 July 2013

Authors’ replyLiao and Fielding raise an interesting point about the usefulness of our research examining public response to the infl uenza A H1N1 pandemic in fi ve countries. However, the intent of our study was somewhat diff erent from that which they suggest; it was to address three key questions related to future global pandemic policy that require multinational comparative data for multiple behaviours.

First, we examined whether the public’s adoption of protective behaviours across countries varied or had largely similar rates. This information can help to establish whether pandemic response policies can be quite uniform at the global level or whether responses need to be more country-specifi c. The data clearly show great variation in the adoption of behaviours and support the need for country-specifi c, contextualised policies and communication approaches. We thank Atlanti-Duault and colleagues for their complementary study showing the importance of contextualisation of communications using the example of France and Quebec. Our study also shows global patterns in that the public in all countries studied were more likely to adopt personal protective behaviours compared with social distancing ones and that vaccination levels were fairly low across the board. Such information provides insights into the challenges that countries share, even as they must contextualise their responses.

Second, we examined whether the adoption of non-pharmaceutical behaviours was a deterrent to vaccination, which can address con-cerns about the eff ect of promotion of potentially con tradictory behaviours. We found no evidence that non-pharmaceutical behaviours were a deterrent to vaccination in any country studied. Thus, the data suggest it is possible to develop policies supporting both kinds of protective behaviours, which might help to maximise

people’s opportunities to protect themselves in future pandemics.

Third, we examined public support for future government policies directly. We learned that all countries showed relatively high support for government policies that promote mask-wearing, school closures, and avoidance of public gathering sites like shopping centres. Such evidence can directly inform country policies in future pandemics.

These conclusions are crucially important and need the scale of our study’s design. Moreover, we could fi nd no studies that provide such data within the H1N1 or pandemic literature. Thus, we believe this study provides a unique contribution that we hope will help inform global and country-specifi c responses that help to reduce the burden of illness in the case of future pandemics. We also hope it spurs additional research like that of Atlanti-Duault and colleagues.GKS’s husband has consulted for Eli Lilly. We declare that we have no confl ict of interest.

Gillian K SteelFisher, Robert J Blendongsteel@hsph.harvard.edu

Harvard Opinion Research Program, Harvard School of Public Health, Department of Health Policy and Management, Kresge, Boston, MA 02115, USA

opposed immunisation, and family doctors—traditionally in charge of immunisation in France—were not involved in the H1N1 campaign.4 Govern ment credibility was low; callers to H1N1 hotlines often cited the 1986 Chernobyl catastrophe, in which the French government minimised dangers associated with radiation. However, whereas H1N1 immunisation rates were low, seasonal infl uenza immunisation rates were not notably aff ected.

Communication about pandemics such as infl uenza is not simply a technical challenge.5 Coordination among public health offi cials, govern-ments, and the media can be eff ective in promoting preventive behaviour as a crisis unfolds. However, additional, more complex strategies, such as contextualising information to enhance personal agency and local identifi cation of credible spokespeople in preparation of such events, should be explored.We declare that we have no confl icts of interest.

Laëtitia Atlani-Duault, Cécile Rousseau, Jean-Paul Moatti, Jean-François Delfraissy, Bernadette Murguelaetitia.atlani-duault@ird.fr

UMR 912 SESSTIM (Aix-Marseille University/ INSERM/IRD), Marseille, France (LA-D, J-PM); Hunter College, City University of New York, New York, NY, USA (LA-D); McGill University, Montreal, QC, Canada (CR); Université Paris Sud, Paris, France, and INSERM, Paris, France (J-FD); Institut de Microbiologie et des Maladies Infectieuses, Paris, France (BM)

1 SteelFisher GK, Blendon RJ, Ward JRM, Rapoport R, Kahn EB, Kohl KS. Public response to the 2009 infl uenza A H1N1 pandemic: a polling study in fi ve countries. Lancet Infect Dis 2012: 12: 845–50.

2 Rousseau C, Moreau N, Dumas M-P, Bost I, Lefebvre S, Atlani-Dualt L. Public media communications about H1N1, risk perceptions and immunisation behaviours: a Quebec–France comparison. Public Underst Sci (In press).

3 Washer P. Lay perceptions of emerging infectious diseases: a commentary. Pub Underst Sci 2011; 20: 506–12.

4 Verger P, et al. Pandemic infl uenza (A/H1N1) vaccine uptake among French private general practitioners: a cross sectional study in 2010. PLoS One 2012: 7: e41837.

5 Atlani-Duault L,Kendall C. Infl uenza, anthropology, and global uncertainties. Med Anthropol 2009; 28: 207–11.

Rifampicin and moxifl oxacin for tuberculous meningitis

Ravina Ruslami and colleagues presented a study assessing pharma-cokinetics, safety, and survival benefi t of diff erent treatment regimens con-taining high-dose rifampicin and moxi fl oxacin in patients with tuber-culous meningitis in a hospital setting.1

Their fi ndings that a treatment regimen containing a higher dose of rifampicin and standard-dose or high-dose moxifl oxacin during the fi rst 2 weeks is safe in patients with tuberculous meningitis, and that high-dose intravenous rifampicin could be associated with a survival

Correspondence

www.thelancet.com/infection Vol 13 July 2013 569

benefi t in patients with severe disease are important to note.

We agree with the authors that, on the basis of the small number of patients per group, clinical results should be interpreted carefully. To compensate for small group sizes, one could consider a diff erent strategy with drug exposure as a continuous variable. Additionally, isoniazid con-centrations should also be measured since isoniazid contributes to rapid culture conversion and penetrates well in cerebrospinal fl uid. Receiver operating characteristic analysis could show the extent to which cumulative drug exposures of rifampicin, moxifl oxacin, and isoniazid relate to outcome. Resultant potentially crucial values for positive treatment outcome could be detected and related to the antagonistic eff ect on cell kill as observed after co-administration of rifampicin and moxifl oxacin in in vitro and in vivo studies.2,3

Another consideration is the potential benefi t of a higher oral dosage to reach similar drug exposure as achieved with intravenous dosing. The proposed alternative strategy to analyse the data would also compensate for the diff erence in drug exposure due to intravenous administration compared with oral dosing, especially in the presence of predisposing factors for poor drug absorption like HIV co-infection.

We therefore would like to encourage the authors to do further analyses of their data to generate additional hypotheses.

Onno Akkerman, Arianna Pranger, Richard van Altena, Tjip van der Werf, Jan-Willem Alff enaar o.w.akkerman@umcg.nl

Department of Pulmonology and Tuberculosis (OA, RvA, TvdW), and Department of Hospital and Clinical Pharmacy (AP, J-WA), University of Groningen, University Medical Center Groningen, Groningen, Netherlands

1 Ruslami R, Ganiem AR, Dian S, et al. Intensifi ed regimen containing rifampicin and moxifl oxacin for tuberculous meningitis: an open-label, randomised controlled phase 2 trial. Lancet infect Dis 2013; 13: 27–35.

2 Drusano GL, Sgambati N, Eichas A, Brown DL, Kulawy R, Louie A. The combination of rifampin plus moxifl oxacin is synergistic for suppression of resistance but antagonistic for cell kill of Mycobacterium tuberculosis as determined in a hollow-fi ber infection model. mBio 2010; 1: e00139-10.

3 Balasubramanian V, Solapure S, Gaonkar S, et al. Eff ect of coadministration of moxifl oxacin and rifampin on Mycobacterium tuberculosis in a murine aerosol infection model. Antimicrob Agents Chemother 2012; 56: 3054–57.

Rovina Ruslami and colleagues showed that treatment containing a higher dose of rifampicin than standard-dose or high-dose moxi fl oxacin during the fi rst 2 weeks is safe in patients with tuberculous meningitis, and that high-dose intra venous rifampicin can lead to increased survival benefi ts. The study has many positive points and has reinforced the much needed optimism regarding management of tuberculous meningitis, especially in low-resource settings.

This study showed that doubling the dose of moxifl oxacin resulted in a proportional increase in the plasma and cerebrospinal fl uid concentrations of the drug.1 A previous study showed no signifi cant change in the plasma or cerebrospinal fl uid concentration of the drug after doubling the dose (from 400 mg to 800 mg) in patients with tuberculous meningitis who were on rifampicin.2 Another study involving patients with extrapulmonary or pulmonary tuberculosis indicated that moxi-fl oxacin showed great variability in plasma protein binding, area under the plasma–concentration–time curve (AUC0–24 h), and minimum inhibitory concentration, and that drug–drug interactions have a large eff ect on the AUC0–24 h/minimum inhibitory concentration ratio. After these fi ndings, the role of moxifl oxacin in the treatment of tuberculous meningitis remains unclear.

Thus, because of the high variability in the plasma and cerebrospinal fl uid concentrations of moxifl oxacin, use of the drug for treatment of tuberculous meningitis will necessitate moni toring

of therapeutic drug con cen tration to ensure optimum dosing. Monitoring might not be operationally feasible in resource-poor settings. Also, questions remain regarding the effi cacy of moxifl oxacin for tuberculous meningitis and its tolerability in the long term. Furthermore, its use as an additive or alternative therapy for pulmonary tuberculosis is still debated.4

A major drawback of the study is that the authors evaluated only high-dose intravenous rifampicin for tuberculous meningitis; this might be diffi cult to implement in developing countries with high burdens of disease. A previous study showed the pharmacokinetics and tolerability of a higher dose of oral rifampicin compared with the standard dose in patients with pulmonary tuberculosis,5 and the HIGHRIF studies are aiming to establish the right dose of oral rifampicin. Confi rmation of whether high plasma concentrations from higher dose oral rifampicin correlate with its cerebrospinal fl uid concentrations will be important. Although absorption of oral rifampicin in the gastrointestinal system might be aff ected in critically ill patients, the replication of the fi ndings of this study using higher doses of oral rifampicin for tuberculous meningitis is a potential game changer. We declare that we have no confl icts of interest.

*Kingsley N Ukwaja, Cajetan C Onyedumukwajakingsley@yahoo.co.uk

Department of Internal Medicine, Federal Teaching Hospital, Abakaliki, Ebonyi State, Nigeria (KNU); and Department of Medicine, College of Medicine, University of Nigeria Teaching Hospital, Enugu Campus, Enugu, Nigeria (CCO)

1 Ruslami R, Ganiem AR, Dian S, et al. Intensifi ed regimen containing rifampicin and moxifl oxacin for tuberculous meningitis: an open-label, randomised controlled phase 2 trial. Lancet Infect Dis 2013; 13: 27–35.

2 Alff enaar JW, van Altena R, Bökkerink HJ, et al. Pharmacokinetics of moxifl oxacin in cerebrospinal fl uid and plasma in patients with tuberculous meningitis. Clin Infect Dis 2009; 49: 1080–82.

3 Pranger AD, van Altena R, Aarnoutse RE, et al. Evaluation of moxifl oxacin for the treatment of tuberculosis: 3 years of experience. Eur Respir J 2011; 38: 888–94.

For the HIGHRIF studies see http://ae-tbc.eu/index.php?id=17074