Endostatin shows a useful value for predicting failure to recover from acute kidney injury: some confounders to consider

Jia and colleagues have concluded that plasma endostatin shows a useful value for predicting failure to recover from acute kidney injury (AKI) [1]. They studied two populations of patients with AKI following non-cardiac major surgery, with the primary endpoint of recovery or “non-recovery” from AKI. Patients classified as “non-recovery” from AKI in fact consisted of two groups, a cohort receiving renal replacement therapy (RRT) at day 7 and another cohort without RRT [1]. We would like to make some comments. Endostatin, the C-terminal fragment of collagen XVIII, is a cytokine with a molecular weight of 20 kDa [2]. It stands to reason that this small molecule can be easily removed by RRT as the cutoff point of filter membranes is about 35 kDa [3]. According to the authors, patients with renal recovery showed endostatin concentrations of 62.6 ng/ml, whereas patients failing to recover showed higher concentrations of 108.5 ng/ml [1]. Also, almost 20% of the AKI population received RRT for 7 days [1]. Considering that endostatin can be removed by RRT, the endostatin values in this group of patients may fall significantly [3]. This could give the clinician the false impression that the patient will recover from AKI. Accordingly, if endostatin is used as a predictive tool in the future, falsely low endostatin values in RRT patients could lead to a premature de-escalation of care for intensive care unit (ICU) patients. There has been no investigation of the performance of endostatin in patients who receive RRT. Therefore, we believe there is a critical need for a future study with a focus on the performance of the currently known sepsis biomarkers in patients who receive RRT [4]. As noted by experts in endostatin, there is not enough evidence to date to support the use of endostatin measurements in clinical practice [5]. RRT is a good example of one of those conditions [5].