Anti-proteinuric effect of losartan: statistical vs clinical significance.

Sir, We read with great interest the recent article on the antiproteinuric effect of losartan when compared with amlodipine [1]. Though we agree with the findings of the authors, we are unable to assess the clinical significance of this article from the data provided. A perusal of table 1 (Baseline characteristics) indicates that the patients receiving losartan had considerably more proteinuria compared with the amlodipine group (3.1 g vs 2.5 g), therefore potentially the decrease in proteinuria with treatment could be greater in the losartan group as the baseline risks were higher. In spite of the randomization process, details on the randomization procedure (i.e. random allocation sequence), allocation concealment and the implementation of allocation sequence (as required by the CONSORT statement [2]) have not been provided in the current study. The results of a decrease in proteinuria (32.4% at 4 weeks and 50.4% at 20 weeks), though impressive, have been averaged for the group and it is impossible to truly appreciate the clinical significance of this study. Numbers needed to treat (NNTs), which is the reciprocal of absolute risk reduction (ARR) are a better measure of clinical significance than relative risk reduction and statistical significance [3]. It would be meaningful to pre-determine a useful decrement of proteinuria and determine what proportion of patients treated with losartan when compared with amlodipine achieved this. A recent study, however, indicates that journals infrequently report NNTs despite recent wide practice of evidence-based health care [4]. The results of the current study, though statistically highly significant, would be more meaningful if the results were presented in such a way as to be able to calculate the NNT and ARR, thereby increasing the ability of the nephrologist to calculate the true clinical significance of the article. The long-term outcome of change of surrogate markers, i.e. a decrease in proteinuria and in urinary TGF, on overall morbidity due to renal insufficiency remains to be studied in long-term outcome-based studies.