To the Editor:
The recent article by Kancir et al.1 assessing nephrotoxicity of 6% hydroxyethyl starch (HES) 130/0.4 compared with 0.9% isotonic saline in the patients undergoing hip arthroplasty was of great interest. They showed no harmful effect of intraoperative infusion of 6% HES 130/0.4 on postoperative renal function. Many things of this study were well done. The authors used a prospective, double-blinded, placebo-controlled design and chose a sensitive and well-validated endpoint of acute kidney injury (AKI): neutrophil gelatinase-associated lipocalin (NGAL), which fulfills many characteristics of an appropriate “real-time” biomarker for AKI detection and is called as a troponin-like biomarker for human AKI.2,3 They had a consistent operation (elective hip arthroplasty). Also, they had tried to control most of the known factors affecting perioperative AKI, such as age, body mass index, preoperative comorbidities and medications, duration of surgery, intraoperative blood loss and transfusion, uses of vasoactive drugs.4,5 All these are strengths in the study design.
In this study, the urine and plasma NGAL levels were determined by a commercial enzyme-linked immunosorbent assay, with a minimal detection level of 1.6 pg/ml. We noted that median levels of urine NGAL at hospital discharge (urine 4) for two groups were about between 9 and 10 times of baseline values. Furthermore, measured values of urine NGAL at every observed time point had the highly variable ranges. In such a small sample study, therefore, only comparing median urine NGAL levels may have of limited clinical value. Most importantly, we were not provided with the cutoff value of urine or plasma NGAL with their enzyme-linked immunosorbent assay for diagnosis of postoperative AKI. Furthermore, we were very interested in knowing how many patients in each group had a higher NGAL level than the cutoff value. As a general rule, a level of more than 150 ng/ml can identify patients at high risk for AKI, and a level greater than 350 ng/ml, those at high risk for renal replacement therapy.6 Were the number of patients with a risk of AKI by NGAL measurement in the two groups comparable?
In fact, AKI is a low incidence event after noncardiac surgery. Kheterpal et al.7,8 demonstrate that in patients undergoing major noncardiac surgery with preoperatively normal renal function, incidence of AKI is approximately 1%, with AKI defined as an absolute level of estimated glomerular filtration rate less than 50 ml/min during the postoperative period. Assuming that this is a real incidence of AKI after noncardiac surgery and 6% HES 130/0.4 can result in a 100% increased risk of AKI, namely, a 2% incidence of AKI, 2,351 patients per group would have been required to have an 80% chance of finding a significant difference. Evidently, the study by Kancir et al. is not powered to show this difference.
Finally, follow-up period of this study only was 10 to 12 days. The median reported time to HES-induced acute renal failure is 16 days.9 According to the accumulated evidences, the U.S. Food and Drug Administration recently released a Safety Communication-Boxed Warning for HES solutions to increase mortality, severe renal injury, and risk of bleeding. Its recommendations include that need for renal replacement therapy has been reported up to 90 days after HES administration, and renal function monitor should last for at least 90 days in all patients.10 A short follow-up period in this study would have missed some of the adverse renal events. In addition, this study was also not designed to assess patient-relevant safety outcomes. Thus, this study fails to provide the robust evidence that HES 130/0.4 is safe for the kidney in noncardiac surgical patients. Here, we would like to echo the conclusion of a recent systematic review by Gattas et al.11 that there is no convincing evidence that third generation HES 130/0.4 is safe in surgical, emergency, or intensive care patients despite publication of numerous clinical studies.
Competing Interests
The authors declare no competing interests.
