Effects of losartan on diabetic maculopathy in type 2 diabetic patients – reply

Dear Sir, We thank you for your interest in our article [1]. It is correct that we studied several ophthalmological parameters as well as urinary albumin excretion rate (UAE) and the transcapillary escape rate of albumin (TERalb) in the present study. However, the predefined primary end-point of the study was the change in central retinal thickness as measured by OCT. The secondary OCT parameter, retinal thickness in the area of hard exudates actually showed a similar trend, although the change was not statistically significant. We did not expect to find a change in visual acuity or semiquantitative retinopathy grade in a study of this limited size and duration, but we found it mandatory to include these parameters in a study on diabetic retinopathy. As for UAE and TERalb, they were included as markers of vascular permeability. Patients were not selected on the basis of abnormalities in these parameters, and the vast majority of study subjects actually had normal baseline values of UAE and TERalb. Hence, no effect of treatment on these parameters was to be expected. The statement that only three patients in the losartan group showed increasing retinal thickness (RT) during the study is not correct. It is evident from Fig. 3 (B), that placebo patients have changes in RT that are distributed around zero (six decrease, two are unchanged, four increase), whereas most losartan patients seem to increase (one decreases, two are unchanged, eight increase). Of course, the three patients in the losartan groups that increase the most weigh heavily in the calculation of the mean changes but the general tendency seems to be the same for most patients in the group. Using non-parametric statistics based on rank sum tests, the difference between groups is still statistically significant. Gleim and Tressler compare our OCT data with data obtained in a study by Koozekanani et al. [2]. In this context it is important to realize that both the OCT technique and the regions evaluated by OCT differed in the two studies, thus rendering a direct comparison of retinal thickness measurements impossible. We have recently compared a group of type 2 diabetic patients with diabetic maculopathy with a group of age-, gender- and duration-matched type 2 diabetic patients without any sign of retinopathy on fundus photographs and flourescein angiograms [3]. The mean difference in central retinal thickness between patients with maculopathy and patients without retinopathy was 17 μm (P < 0.01). The difference between losartan and placebo-treated patients observed in the present study was of the same magnitude (15 μm). Therefore, we do believe that this change is clinically relevant. We have noticed the results from the RENAAL [4] and LIFE [5] studies (and we have actually reported safety data from the latter in our article). In these two large studies, patients were titrated to reach a specific blood pressure (BP) level. On average, they required higher losartan doses than 50 mg o.d. as well as the addition of other BP-lowering agents, thus making comparisons between these studies and the present study difficult. Importantly, losartan has a relatively short half-life [6], and studies employing 24-h ambulatory BP measurement support the notion that losartan 50 mg, when taken in the morning, has little or no effect on night BP [7, 8]. As our patients were studied in the morning, 24 h after the last losartan dose, the apparent adverse effect of treatment on retinal thickness may represent a rebound phenomenon. We are aware that the present study was a relatively small and short-term pilot study. We sincerely believe that we have tempered our conclusions accordingly, merely suggesting that type 2 diabetic patients with maculopathy do not seem to benefit from short-term treatment with losartan (50 mg o.d.) as far as retinal thickness is concerned, as this dose may increase retinal thickness in the central macular area. We also emphasize that long-term studies (possibly with another dosage) including a larger number of patients are required to assess the clinical implications of these findings [1]. Carl Erik Mogensen has a consulting function for Merck. He has given lectures at symposia sponsored by this company. The present research project has received an unconditional grant of 50 000 US$ from Merck.