Changes in serum alanine aminotransferase levels in telbivudine versus lamivudine treatment for chronic hepatitis B: A meta-analysis

Abstract

Objective

A meta-analysis to compare the efficacy and safety of telbivudine (TBV) and lamivudine (LAM) in patients with chronic hepatitis B (CHB), assessed via changes in serum alanine aminotransferase (ALT) levels.

Method

The electronic literature databases PubMed®, Embase®, Web of Science, Cochrane Library, CISCOM, CINAHL, Google Scholar, China BioMedicine and China National Knowledge Infrastructure were searched for relevant studies. The effect of TBV and LAM treatment on serum ALT was assessed using standard mean differences (SMDs) and 95% confidence intervals (CI).

Results

The meta-analysis included six studies (TBV n = 202; LAM, n = 208). Post-treatment ALT levels were significantly lower than pretreatment values for both TBV and LAM (TBV: SMD = 3.00, 95%CI 1.91, 4.09; LAM: SMD = 2.33, 95%CI 1.58, 3.07). Post-treatment ALT was significantly lower after treatment with TBV than LAM (SMD = 0.58, 95%CI 0.21, 0.94).

Conclusion

Both LAM and TBV are effective in normalizing ALT levels in patients with CHB, but TBV may be a better choice due to its lower rates of drug resistance.

Keywords

Chronic hepatitis B alanine aminotransferase telbivudine lamivudine meta-analysis

Introduction

Chronic hepatitis B (CHB) is a major public health threat worldwide,¹ and is one of the most frequent causes of hepatocellular carcinoma.² CHB is caused by hepatitis B virus (HBV), which is transmitted in bodily fluids (for example, via used hypodermic needles, blood transfusion, vertical transmission from mother to child during birth, and sexual transmission).³ Untreated patients are vulnerable to liver-related sequelae.⁴ An estimated 5% of the total worldwide population is infected with HBV, and approximately 500 000 patients annually die of HBV-related complications, including CHB.⁵ The incidence of CHB varies widely, with the highest rates in the Asia–Pacific and sub Saharan Africa.⁶ Risk factors for the development of CHB include vertical transmission of HBV from mother to infant,⁷ age, alcohol use and HIV infection.⁸

The antiviral nucleoside analogues telbivudine (TBV) and lamivudine (LAM) are promising treatments for CHB.⁹ HBV DNA levels are the most critical clinical criterion for the judgement of antiviral treatment efficacy, and patients with persistently elevated HBV DNA have a heightened risk of cancers including hepatocellular carcinoma.^10,11 The serum alanine aminotransferase (ALT) concentration is an important marker of CHB¹² and is commonly used to assess hepatocellular injury and liver health. ¹² LAM effectively reduces serum viral load, improves ALT levels and liver fibrosis, and accelerates hepatitis B e antigen (HBeAg) seroconversion rates.^13–15 Treatment with LAM reduces ALT levels in patients who were positive for hepatitis B surface antigen (HBsAg), suggesting that LAM can inhibit HBV transmission and normalize ALT.¹⁶ TBV has been shown to be more effective than LAM in the treatment of patients with CHB,^17–19 as well having a superior therapeutic response and resistance profile.^18,20–23

The aim of the present meta-analysis was to compare the efficacy of TBV and LAM in the treatment of CHB (assessed via ALT levels and HBV DNA stratum).

Materials and methods

Data sources and keywords

Eligible studies published up to 30 June 2014 were identified by searching the electronic literature databases PubMed®, Embase®, Web of Science, Cochrane Library, CISCOM, CINAHL, Google Scholar, China BioMedicine (CBM) and China National Knowledge Infrastructure (CNKI). The following keywords and search strategies were used: (“hepatitis B surface antigens” OR “hepatitis B surface antigen” OR “Australia antigen” OR “HBsAg” OR “hepatitis B antigen” OR “hepatitis B virus”) AND (“alanine transaminase” OR “glutamic alanine transaminase” OR “alanine 2 oxoglutarate aminotransferase” OR “alanine aminotransferase” OR “glutamic pyruvic transaminase” OR “SGPT” OR “ALT” OR “GPT”) AND (“telbivudine” OR “beta-L-2′-deoxythymidine” OR “Tyzeka®” OR “TBV”) AND (“lamivudine” OR “2′,3′ Dideoxy 3′ thiacytidine” OR “3TC” OR “Epivir®” OR “GR109714X” OR “BCH189”). In addition, reference lists of relevant articles were searched to identify additional relevant papers. No language restrictions were set in this meta-analysis. When articles provided unclear additional data in their original publications, the first authors were contacted for clarification.

Selection criteria

To be included, studies had to fulfil all of the following criteria: (i) patients with CHB; (ii) human clinical cohort studies or diagnostic tests focused on CHB diagnosis with administration of LAM or TBV; (iii) pre and post-treatment data for serum ALT levels; (4) report adjusted standard mean differences (SMDs) and 95% confidence intervals (CI) for ALT serum level; (5) number of patients stated; and (6) >20 patients. When studies had populations overlapping >50% with two or more papers, the one with the most comprehensive population was included. When the same authors published more than one study with the same population, only the latest or most complete was included. All authors approved the final list of included studies.

Data extraction

Data were extracted from the included studies by two investigators (M.D. and G.M.X.) independently and in duplicate. If necessary, consensus was achieved by discussion and re-examination. The following data were extracted from eligible studies: surname of first author; date of publication; source of publication; study type; study design; sample size; age; sex; ethnicity and country of subjects; source of samples; detection method for ALT serum level; time of post-treatment sampling; serum ALT levels; and HBV DNA levels.

Quality assessment

Quality of the included studies was assessed by two investigators (F.L.W. and J.S.Z.) independently and in duplicate, using predefined criteria based on Newcastle-Ottawa Scale (NOS) criteria.²⁴ The NOS criteria are based on three aspects: (i) subject selection (0–4); (ii) comparability of subjects (0–2); (iii) clinical outcome (0–3). Total NOS scores range from 0 (lowest) to 9 (highest). According to the NOS scores, the included studies were classified as low quality (0–6) or high quality (7–9). Discrepancies regarding NOS scores were resolved by discussion and consultation with an additional reviewer (D.J.X.).

Statistical analyses

The effect size of each study was calculated using Z test with summary SMD (95%CI). Heterogeneity was evaluated using Cochran’s Q-statistic (P < 0.05 was regarded as statistically significant),²⁵ and I² test (0%, no heterogeneity; 100%, maximal heterogeneity).²⁶ If between-study heterogeneity existed, the random-effects model (DerSimonian and Laird) was used to calculate the pooled ORs; otherwise the fixed-effects model was used (Mantel–Haenszel). Meta-regression and subgroup meta-analyses were performed for detection method and postreatment time. One-way sensitivity analysis was performed and publication bias was assessed using funnel plots and Egger’s linear regression test.²⁷ Data were inputted into STATA® software (version 12.0; Stata Corp, College Station, TX, USA) by two investigators (Y.M.L. and H.Z.Y.) independently and in duplicate.

Results

The study selection procedure is displayed in Figure 1. The literature search identified 132 papers on the basis of title and key words. After screening according to exclusion criteria, the final meta-analysis included six studies (TBV n = 202; LAM, n = 208),^28–33 of which three were low quality (NOS score = 6)^28,29,33 and three were of high quality (NOS score >6).^30–32 All studies were performed in Asians, with five in China and one in Thailand. The studies and their main characteristics are listed in Table 1.

Figure 1.

Flow diagram of the study selection process in a meta-analysis undertaken to compare the relative efficacy of telbivudine and lamivudine in the treatment of chronic hepatitis B infection (assessed via serum alanine aminotransferase levels).

Table 1.

Characteristics of studies included in a meta-analysis undertaken to compare the relative efficacy of telbivudine and lamivudine in the treatment of chronic hepatitis B infection (assessed via serum alanine aminotransferase levels).

First author	Country	Patients, n		Sex, M/F		Age, years		Sample	ALT analysis method	Study Timepoints, weeks	Time of post-treatment sampling, weeks	NOS score
First author	Country	TBV	LAM	TBV	LAM	TBV	LAM	Sample	ALT analysis method	Study Timepoints, weeks	Time of post-treatment sampling, weeks	NOS score
Wang, 2013²⁸	China	20	18	17/3	17/1	41.6 ± 11.4	42.4 ± 10.7	Serum	ECLIA	1, 2, 3, 4, 5, 6, 7, 8	<24	6
Ji, 2012³²	China	33	26	38/21		26–59		Serum	Unknown	12, 24	≥24	7
Dong, 2012³¹	China	86	88	152/28		36 (17–68)		Serum	ACA	12, 24, 36, 48	≥24	8
Aung, 2011²⁹	Thailand	11	12	NA		NA		Serum	ECLIA	52	≥24	6
Li, 2010³³	China	22	21	17/5		18–60		Serum	Unknown	12 weeks	<24	6
Chen, 2009³⁰	China	30	43	62/11		28 (16–46)		Serum	ACA	4, 12, 24, 48 weeks	≥24	7

TBV, telbivudine; LAM, lamivudine; M, male; F, female; NOS, Newcastle–Ottawa Scale; ACA, automatic biochemical analyser; ECLIA, electrochemiluminescence immunoassay.

Meta-analyses for the effects of TBV and LAM on serum ALT are shown in Figure 2. There was significant heterogeneity (TBV pretreatment vs post-treatment, I²= 97.8%, P < 0.001; LAM pretreatment vs post-treatment, I²= 96.3%, P < 0.001; post-treatment LAM vs TBV, I²= 96.3%, P < 0.001), therefore the random-effects model was used. Post-treatment ALT levels were significantly lower than pretreatment values for both TBV and LAM (TBV: SMD = 3.00, 95%CI 1.91, 4.09, P < 0.001; LAM: SMD = 2.33, 95%CI 1.58, 3.07, P < 0.001). Post-treatment ALT was significantly lower after treatment with TBV than after LAM (SMD = 0.58, 95%CI 0.21, 0.94, P = 0.002). HBV DNA levels were significantly lower than baseline after LAM treatment (SMD = 1.75, 95%CI 1.14, 2.35, P < 0.001).

Figure 2.

Forest plots for the effect of telbivudine and lamivudine on serum alanine aminotransferase levels in patients with chronic hepatitis B. Studies: Wang, 2013;²⁸ Ji, 2012;³² Dong, 2012;³¹ Aung, 2011;²⁹ Li, 2010;³³ Chen, 2009;³⁰ a–g, different study timepoints.

Subgroup analyses based on detection method are shown in Figure 3. For both TBV and LAM, post-treatment ALT levels were significantly lower than pretreatment values for all detection methods (P < 0.05 for each comparison). For between-drug analyses, postreatment ALT levels were significantly lower in patients treated with TBV than in those treated with LAM, when samples were analysed using an automatic biochemical analyser (ACA) or unknown method, (ACA: SMD = 0.63, 95%CI 0.26,1.01, P = 0.001; unknown method: SMD = 2.77, 95%CI 1.20, 4.33, P = 0.001). There was no significant between-group difference in post-treatment ALT when samples were analysed using electrochemiluminescence immunoassay (ECLIA), however (Figure 3).

Figure 3.

Subgroup analyses by method of serum alanine aminotransferase (ALT) analysis and time of post-treatment sampling for the effect of telbivudine and lamivudine on serum ALT levels in patients with chronic hepatitis B. Studies: Wang, 2013;²⁸ Ji, 2012;³² Dong, 2012;³¹ Aung, 2011;²⁹ Li, 2010;³³ Chen, 2009;³⁰ a–g, different study timepoints.

Subgroup analyses based on time of post-treatment sampling are shown in Figure 3. Time of post-treatment sampling was classified as <24 or ≥24 weeks. Post-treatment ALT levels were significantly lower than pretreatment values for both TBV and LAM, in both the <24 and ≥24 weeks subgroups (P < 0.05 for each comparison). In the ≥24 weeks subgroup, post-treatment ALT levels were higher in patients treated with LAM than in patients treated with TBV (SMD = 1.38, 95%CI 0.76, 1.99, P < 0.001). There were no significant between-group differences in the <24 weeks subgroup.

Sensitivity analyses indicated that the meta-analysis was stable (Figure 4). Funnel plots and Egger’s test found strong evidence for publication bias in the comparison between pre-TBV and post-TBV treatment ALT levels (P < 0.001). There was no evidence of publication bias in the comparison between pre-LAM and post-LAM treatment ALT levels, or in TBV vs LAM treatment.

Figure 4.

Sensitivity analysis of summary odds ratio coefficients for the effect of telbivudine and lamivudine on serum ALT levels in patients with chronic hepatitis B. Studies: Wang, 2013;²⁸ Ji, 2012;³² Dong, 2012;³¹ Aung, 2011;²⁹ Li, 2010;³³ Chen, 2009;³⁰ a–g, different study timepoints.

Discussion

The current meta-analysis compared the relative efficacy of TBV and LAM in the treatment of CHB infection (assessed via serum ALT levels), and found that both drugs significantly reduced ALT compared with pretreatment levels. In addition, TBV resulted in a significantly greater decrease in ALT levels than LAM.

Studies have shown that LAM can improve liver histology and suppress viral DNA replication in patients with CHB after 1 year of treatment, with reversal of cirrhosis and fibrosis observed after 3 years.³⁴ Although LAM has a good safety profile in the treatment of HBV infection, resistance (and accompanying viral breakthrough, hepatitis flare and liver decompensation) is a growing problem.³⁵ TBV has lower resistance rates, causes viral suppression by inhibiting HBV DNA synthesis,¹⁸ and has higher rates of survival, virological response, therapeutic response, HBeAg loss and ALT normalization than LAM.^19,36 In addition, TBV has a high seroconversion rate and no carcinogenic, mutagenic or embryonic toxic effects.³⁷ Other nucleoside analogues include adefovir dipivoxil fumarate (ADF), entecavir (ETV) and tenofovir disoproxil fumarate (TDF).³⁸ Although ETV and TDF are first-line drugs for CHB, we evaluated LAM due to its clinical relevance and TBV due to its known effect on ALT.^7,37,38

Although ALT levels are decreased by treatment with nucleoside analogues such as LAM and TBV, levels have been shown to increase after withdrawal of treatment.^39,40 Studies have found that TBV provides greater and more consistent suppression of HBV DNA than LAM, which may be reflected in normalization of ALT levels.^7,41 The between-drug difference in post-treatment ALT levels in the present analysis may be a result of LAM drug resistance. The current meta-analysis indicated that LAM treatment significantly decreased HBV DNA levels compared with baseline, which may lead to a reduction in HBV-related complications.⁴² Our data are in accordance with others who found that TBV resulted in significantly greater decreases in serum ALT than LAM, as well as a significantly higher survival rate in patients with CHB.²⁰

We conducted subgroup analyses to determine whether detection method had any effect on our findings. In all subgroups (ACA, ECLIA and unknown), both TBV and LAM resulted in significant decreases in ALT levels, but the superiority of TBV could not be demonstrated when samples were analysed using ECLIA. This could be explained by differences between detection methods, but may also be due to the small number of studies involved. In addition, time of post-treatment sampling had no effect on the finding that TBV and LAM decreased ALT levels, but there was no significant between-drug difference when ALT was assessed at <24 weeks post-treatment. This finding suggests that the difference between LAM and TBV is not apparent in the early stages of treatment. Both TBV and LAM are capable of normalizing ALT levels in patients with CHB, but TBV is preferred because of its relatively low resistance levels.

The present study had several limitations. Since it was a cross-sectional retrospective study, our analysis did not take into consideration changes in viral load or treatment efficacy over time. In addition, due to limited data, it was not possible to analyse phenotypic or genotypic resistance, restricting our results to a narrow clinical application. We did not include some unpublished articles and abstracts due to incomplete data. Additionally, three of the six included studies were of low methodological quality due to insufficient data and lack of adjustment for confounding factors. This may be a source of heterogeneity and could affect our findings. Differences in included criteria, follow-up periods, treatment efficacy and severity of disease are also potential sources of heterogeneity. Although each study population was well defined and had similar inclusion criteria, there may be unknown factors that could have influenced our results. The participants included in this study did not differ significantly from other Asian population-based or hospital-based patients in terms of age/sex distribution, HBeAg and HBsAg status, or HBV DNA levels. The fact that all of the studies investigated Asian populations limits the applicability of these findings to other ethnicities. Finally, these findings may be influenced by the high levels of heterogeneity in our study. Further, larger scale clinical trials are required to contribute to the screening and development of individualized treatment schemes for patients with CHB.

In conclusion, both LAM and TBV are effective in normalizing ALT levels in patients with CHB, but TBV may be a better choice due to its lower rates of drug resistance.

Footnotes

Declaration of conflicting interest

The authors declare that there are no conflicts of interest.

Funding

The study was supported by the Clinical Research Scheme of Heat Clearing and Ddetoxicating Fuzheng Quxie Treatment of HBeAg Negative Chronic Hepatitis B (20121139, Guangdong Province Pharmaceutical Administration). We would like to acknowledge the reviewers for their helpful comments on this paper.

References

Iloeje

Yang

Chen

. Natural history of chronic hepatitis B: what exactly has REVEAL revealed? Liver Int 2012; 32: 1333–1341.

Göbel

Erhardt

Herwig

. High prevalence of significant liver fibrosis and cirrhosis in chronic hepatitis B patients with normal ALT in central Europe. J Med Virol 2011; 83: 968–973.

Trinh

Nguyen

. Prevalence, risk factors, and disease knowledge of chronic hepatitis B infection in Vietnamese Americans in California. J Cancer Educ 2013; 28: 319–324.

Museru

Vargas

Kinyua

. Hepatitis B virus infection among refugees resettled in the U.S.: high prevalence and challenges in access to health care. J Immigr Minor Health 2010; 12: 823–827.

Sorrell

Belongia

Costa

. National Institutes of Health consensus development conference statement: management of hepatitis B. Hepatology 2009; 49(5 suppl): S4–S12.

Kew

. Epidemiology of chronic hepatitis B virus infection, hepatocellular carcinoma, and hepatitis B virus-induced hepatocellular carcinoma. Pathol Biol (Paris) 2010; 58: 273–277.

Pan

Duan

Bhamidimarri

. An algorithm for risk assessment and intervention of mother to child transmission of hepatitis B virus. Clin Gastroenterol Hepatol 2012; 10: 452–459.

Falade-Nwulia

Seaberg

Rinaldo

. Comparative risk of liver-related mortality from chronic hepatitis B versus chronic hepatitis C virus infection. Clin Infect Dis 2012; 55: 507–513.

Wei

Ling

. Perioperative antiviral therapy for chronic hepatitis B-related hepatocellular carcinoma. Hepatobiliary Pancreat Dis Int 2013; 12: 251–255.

10.

Chan

Tse

. High viral load and hepatitis B virus subgenotype ce are associated with increased risk of hepatocellular carcinoma. J Clin Oncol 2008; 26: 177–182.

11.

Huang

Lai

Lau

. Posthepatectomy HBV reactivation in hepatitis B-related hepatocellular carcinoma influences postoperative survival in patients with preoperative low HBV-DNA levels. Ann Surg 2013; 257: 490–505.

12.

Kim

Flamm

Di Bisceglie

. Serum activity of alanine aminotransferase (ALT) as an indicator of health and disease. Hepatology 2008; 47: 1363–1370.

13.

Sueki

Maekawa

Miura

. Correlation between pretreatment viral sequences and the emergence of lamivudine resistance in hepatitis B virus infection. J Med Virol 2012; 84: 1360–1368.

14.

Leung

. Liver disease-significant improvement with lamivudine. J Med Virol 2000; 61: 380–385.

15.

Chen

Lee

. Comparison of sequence changes of precore and core promoter regions in HBeAg-positive chronic hepatitis B patients with and without HBeAg clearance in lamivudine therapy. J Hepatol 2006; 44: 76–82.

16.

Jiang

. The efficacy and safety of antiviral therapy with lamivudine to stop the vertical transmission of hepatitis B virus. Eur J Clin Microbiol Infect Dis 2012; 31: 2211–2218.

17.

Stein

. Phase I, open-label, single-dose study to evaluate the pharmacokinetics and safety of telbivudine in children and adolescents with chronic hepatitis B. Antimicrob Agents Chemother 2013; 57: 4128–4133.

18.

Hou

Yin

. Telbivudine versus lamivudine in Chinese patients with chronic hepatitis B: Results at 1 year of a randomized, double-blind trial. Hepatology 2008; 47: 447–454.

19.

Lai

Gane

Liaw

. Telbivudine versus lamivudine in patients with chronic hepatitis B. N Engl J Med 2007; 357: 2576–2588.

20.

Chan

Chen

Gane

. Randomized clinical trial: efficacy and safety of telbivudine and lamivudine in treatment-naïve patients with HBV-related decompensated cirrhosis. J Viral Hepat 2012; 19: 732–743.

21.

Wang

Cai

. Telbivudine myopathy in a patient with chronic hepatitis B. Int J Clin Pharm 2012; 34: 422–425.

22.

Lü

Yang

Fan

. Serum HBV DNA level at week 12 is superior to viral response at week 24 in predicting long-term treatment outcome of telbivudine for chronic hepatitis B patients. Chin Med J (Engl) 2013; 126: 2333–2336.

23.

Lai

Leung

Teo

. A 1-year trial of telbivudine, lamivudine, and the combination in patients with hepatitis B e antigen-positive chronic hepatitis B. Gastroenterology 2005; 129: 528–536.

24.

Stang

. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010; 25: 603–605.

25.

Jackson

White

Riley

. Quantifying the impact of between-study heterogeneity in multivariate meta-analyses. Stat Med 2012; 31: 3805–3820.

26.

Peters

Sutton

Jones

. Comparison of two methods to detect publication bias in meta-analysis. JAMA 2006; 295: 676–680.

27.

Zintzaras

Ioannidis

. HEGESMA: genome search meta-analysis and heterogeneity testing. Bioinformatics 2005; 21: 3672–3673.

28.

Wang

Chen

Fan

. Efficacy and safety of telbivudine therapy in liver failure patients with chronic hepatitis B virus infection. J Med Virol 2013; 85: 1907–1912.

29.

Aung

Leowattana

Tangpukdee

. Observation of genotype C infected chronic hepatitis B patients in clinical practice. J Infect Dev Ctries 2011; 5: 882–889.

30.

Chen

Zhu

. The efficacy of Telbivudine treatment for HBeAg-positive chronic hepatitis B in 48 weeks. Infectious Disease Information 2009; 22: 216–218.

31.

Dong

Sun

Chen

. Efficacy of oral telbivudine treatment for Chronic hepatitis B. Shandong Medical Journal 2012; 52: 77–79.

32.

Feng

. Clinical observation of telbivudine for the treatment of chronic hepatitis B e antigen-positive. Inner Mongolia Medical Journal 2012, pp. 978–979.

33.

. Clinical efficacy of telbivudine treatment for chronic severe hepatitis B. Shandong Medical Journal 2010; 50: 71–72.

34.

Wang

Shao

. Histological improvement of long-term antiviral therapy in chronic hepatitis B patients with persistently normal alanine aminotransferase levels. J Viral Hepat 2013; 20: 328–335.

35.

Kwak

Choi

Lee

. Long-term efficacy of entecavir therapy in chronic hepatitis B patients with antiviral resistance to lamivudine and adefovir. J Viral Hepat 2011; 18: e432–e438.

36.

Jiang

Wang

Zhao

. Lamivudine versus telbivudine in the treatment of chronic hepatitis B: a systematic review and meta-analysis. Eur J Clin Microbiol Infect Dis 2013; 32: 11–18.

37.

Ghany

Feld

Zhao

. Randomised clinical trial: the benefit of combination therapy with adefovir and lamivudine for chronic hepatitis B. Aliment Pharmacol Ther 2012; 35: 1027–1035.

38.

Ridruejo

Adrover

Silva

. Virological breakthrough and resistance in patients with chronic hepatitis B receiving nucleos(t)ide analogues in clinical practice. Hepatology 2011; 54: 1104–1105.

39.

Deng

Yoshida

Jin

. Reversible phospho-Smad3 signalling between tumour suppression and fibrocarcinogenesis in chronic hepatitis B infection. Clin Exp Immunol 2014; 176: 102–111.

40.

Deng

Zhou

Gao

. The effects of telbivudine in late pregnancy to prevent intrauterine transmission of the hepatitis B virus: a systematic review and meta-analysis. Virol J 2012; 9: 185–185.

41.

Zhang

Wang

. Efficacy of telbivudine treatment for hepatitis B e antigen-positive chronic hepatitis B patients with poor response to adefovir dipivoxil. J Viral Hepat 2013; 20(Suppl 1): 46–51.

42.

Zheng

Jiang

Chen

. Serum HBV DNA level at week 24 as a proper predictor for the effect of 2-year lamivudine treatment. Chin Med J (Engl) 2011; 124: 1257–1260.