Martin Schumacher

BACKGROUND: The heterogeneity statistic I(2), interpreted as the percentage of variability due to heterogeneity between studies rather than sampling error, depends on precision, that is, the size of the studies included. METHODS: Based on a real meta-analysis, we simulate artificially 'inflating' the sample size under the random effects model. For a given inflation factor M = 1, 2, 3,... and for each trial i, we create a M-inflated trial by drawing a treatment effect estimate from the random effects model, using s(i)(2)/M as within-trial sampling variance. RESULTS: As precision increases, while estimates of the heterogeneity variance tau(2) remain unchanged on average, estimates of I(2) increase rapidly to nearly 100%. A similar phenomenon is apparent in a sample of 157 meta-analyses. CONCLUSION: When deciding whether or not to pool treatment estimates in a meta-analysis, the yard-stick should be the clinical relevance of any heterogeneity present. tau(2), rather than I(2), is the appropriate measure for this purpose.

OBJECTIVE In the BI 10773 (Empagliflozin) Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG OUTCOME) trial involving 7,020 patients with type 2 diabetes and established cardiovascular (CV) disease, empagliflozin given in addition to standard of care reduced the risk of CV death by 38% versus placebo (hazard ratio [HR] 0.62 [95% CI 0.49, 0.77]). This exploratory mediation analysis assesses the extent to which treatment group differences in covariates during the trial contributed to CV death risk reduction with empagliflozin. RESEARCH DESIGN AND METHODS Effects of potential mediators, identified post hoc, on the HR for CV death with empagliflozin versus placebo were analyzed by Cox regression models, with treatment group adjusted for the baseline value of the variable and its change from baseline or updated mean (i.e., considering all prior values), each as a time-dependent covariate. HRs were compared with a model without adjustment for covariates. Multivariable analyses also were performed. RESULTS Changes in hematocrit and hemoglobin mediated 51.8% and 48.9%, respectively, of the effect of empagliflozin versus placebo on the risk of CV death on the basis of changes from baseline, with similar results in analyses on the basis of updated means. Smaller mediation effects (maximum 29.3%) were observed for uric acid, fasting plasma glucose, and HbA1c. In multivariable models, which incorporated effects of empagliflozin on hematocrit, fasting glucose, uric acid, and urine albumin:creatinine ratio, the combined changes from baseline provided 85.2% mediation, whereas updated mean analyses provided 94.6% mediation of the effect of empagliflozin on CV death. CONCLUSIONS In this exploratory analysis from the EMPA-REG OUTCOME trial, changes in markers of plasma volume were the most important mediators of the reduction in risk of CV death with empagliflozin versus placebo.

A common statistical problem is the assessment of the predictive power of a quantitative variable for some dependent variable. A maximally selected rank statistic regarding the quantitative variable provides a test and implicitly an estimate of a cutpoint as a simple classification rule. Restricting the selection to an arbitrary given inner part of the support of the quantitative variable, we show that the asymptotic null distribution of the maximally selected rank statistic is the distribution of the supremum of the absolute value of a standardized Gaussian process on an interval. The asymptotic argument holds also in the case of tied or censored observations. We compare Monte Carlo results with an approximation of the asymptotic distribution under the null hypothesis. In addition, we investigate the behaviour of the test procedure and of the familiar Spearman rank test for independence, under some alternatives. Moreover, we discuss some aspects of the problem of estimating an underlying cutpoint.

BACKGROUND: Stevens-Johnson syndrome (SJS) and its severe form, toxic epidermal necrolysis (TEN), are rare but life-threatening cutaneous adverse reactions to drugs, especially to allopurinol, carbamazepine, lamotrigine, phenobarbital, phenytoine, sulfamethoxazole, oxicam and nevirapine. Recently, a strong association between carbamazepine and allopurinol induced SJS or TEN has been described with respectively, HLA-B*1502 and HLA-B*5801 in a Han Chinese population from Taiwan and other Asian countries. OBJECTIVE: The objective is to further investigate the relationship between SJS/TEN and HLA-B in a large number of patients in a European population. METHODS: HLA-B genotyping was performed on 150 patients included in a European study (RegiSCAR) of SJS and TEN. We focused on patients related to 'high-risk' drugs including: 31 cases related to allopurinol, 28 to sulfamethoxazole, 19 to lamotrigine and 14 to oxicam. RESULTS: Sixty-one percent of 31 allopurinol-induced SJS/TEN patients carried the HLA-B*5801 allele and the figure was 55% for 27 patients of European ancestry [odds ratio=80 (34-187)], (P<10(-6)) as previously observed in Han Chinese. For other drugs, two rare alleles showed a weaker association with SJS/TEN in a limited number of patients: B*38 for sulfamethoxazole or lamotrigine-related patients, and B*73 for oxicam. CONCLUSION: At variance with prior results in Asia, this study shows that even when HLA-B alleles behave as strong risk factors, as for allopurinol, they are neither sufficient nor necessary to explain the disease. Further investigations are necessary to delineate the exact role of the HLA region in SJS/TEN, and to look for other associations in other regions of the genome.