ACE Inhibitor and Postoperative Renal Dysfunction

ACE Inhibitor and Postoperative Renal Dysfunction In Response: Our study aimed to clarify the actual incidence of postoperative renal impairment, assessed with an accurate GFR measure, and to assess the risk factors associated with a perioperative decrease in the nephron population. Therefore, renal function between day 0 and day 7 was not studied. Indeed, the rapidly changing situation that characterizes the postoperative period makes creatinine clearance measurement very difficult, since the patients must be in stable situation to obtain an accurate measurement. Moreover, to consider a meaningful outcome, we waited until the 7th postoperative day to evaluate renal function. It would have been even more interesting to assess renal function over a longer period, since long term prognosis of limited renal insults remains unknown (1). Concerning fluid loading, the amount given postoperatively after day 1 was not fully described. Of course, we checked that the difference observed between the two groups was not related to any difference in fluid loading, transfusion, and catecholamine requirement, in the immediate postoperative period. These data are described in Table 1. However, the main variation in blood volume occurs during the immediate perioperative period and our patients received appropriate care during the whole perioperative period, including close supervision of their requirements in fluid loading. Nevertheless, the perioperative period must be considered as a whole and thus renal dysfunction was related to all events that occurred during this period. Concerning Colson and Ryckwaert’s statement that the better glomerular filtration rate in patients with postoperative renal dysfunction (PORDF) are related to angiotensin converting enzyme inhibitors (ACEI), this point is not supported by the data. Indeed, preoperative creatinine clearance was significantly higher in the group developing PORDF but this may be explained by the individual variations of this value, related to the age of the patients (2,3). On the other hand, there were a little more females in the “No PORDF-group”. The difference may also be explained by the kidney compensatory mechanisms, i.e., when the nephron pool is lower, intact nephrons realize the function of the impaired ones. All these points may participate to the small difference between the two groups. Concerning the criticism about the risk factors identified in our study and in the literature, the difference is very clear and should be considered as one of the main information provided by our study: if we analyze the results with the paradigm used in former studies, the risk factors for a postoperative renal dysfunction are the same than in previously published studies: age and preoperative renal dysfunction. This is not surprising, since the chance of going below a preestablished value is higher, if the preoperative value is low, for whatever reason, such as advanced age or preoperative renal impairment. The main problem is not to affirm evidences, but to identify the possible renal insults in this period, and to improve our strategies to limit this complication. This point has been largely discussed in our paper. Briefly, some important recommendations for future research should be performed: 1) in the perioperative period, renal function must be assessed by using a test which estimates the nephron population as creatinine clearance does, and not by using only serum creatinine; 2) perioperative variation in renal function must be used to assess perioperative renal insults, and not the decrease in renal function under a given threshold; 3) short-term variation in renal function (4) is not the appropriate end-point. Finally, we do not agree with Colson and Ryckwaert’s conclusion about our data, considering that ACEIs does not contribute to a decrease in postoperative renal function, even if we should acknowledge that our study showed a relationship between ACEI and PORDF, and does not demonstrate a causal relation. Firstly, the impaired postoperative renal function related to ACEIs is not surprising since glomerular filtration rate is preserved during hypoperfusion through several compensatory mechanisms. One of the main compensatory responses is angiotensin II (5). The high incidence of hypotension in this group has been previously suggested, including by Ryckwaert and Colson (6) in a recent study, demonstrated that, in chronically ACEI-treated patients, mean arterial pressure after induction was further decreased. Secondly, Colson and Ryckwaert are mistaken in saying that it was surprising that only ACEI was a risk factor for PORDF. Indeed, compared to previous studies, we used a different end-point (variation in renal function), and the fact that ACEI was the only significant risk factor is explained by the fact that the end point was different; the power of our study was limited by its sample size to identify other risk factors, or because ACEI actually had a major deleterious effect on renal function. In conclusion, to date, the problem is to use a valid method to assess renal function in the perioperative period, and we have done it, to identify predictive risk factors in order to limit postoperative renal dysfunction. Further studies are needed to confirm that the association between ACEIs and renal dysfunction is causative. Short-term improvements in renal function, in small groups of patients, after a short administration of ACEIs (4), cannot be definitely considered as clinically relevant. Thus, we maintain our conclusion concerning the deleterious effect of ACEIs on renal function in this period. Clearly, the challenge or the anesthesiologists is now to assess long-term consequences of moderate perioperative renal insults, as pointed out by nephrologists (1). Marie-Laure Cittanova, MD