Editor,—We read with great interest the paper by Girgrah et al ((2000) Gut 46:114–120; OpenUrlCrossRefPubMed) . Their report suggests that the subtle sodium retention that is characteristic of preascitic cirrhosis is improved by administration of low dose losartan. This is despite the paradoxical observation of an angiotensin concentration that is significantly lower in patients compared with healthy volunteers (mean (SEM) patients 6 (2); controls 40 (10) pmol/l). Our results of angiotensin II measurements are at variance with those published by Girgrah et al and are summarised in fig 1.1 ,2 Our studies suggest that there is a progressive increase in angiotensin II concentrations with increasing severity of sodium retention. In fact, this increase in angiotensin II is evident before any measurable derangement in systemic haemodynamic characteristics.3 The values measured in healthy volunteers are also significantly higher than those reported in the literature.4 We are not sure if these differences in measured values are the result of different patient populations, differences in the method of collection of the sample (Girgrahet al—EDTA and aprotinin; Newbyet al and Helmy et al—0.5 ml of 0.45% …