Hsp72 is a novel biomarker to predict acute kidney injury in critically ill patients

Abstract

Background and objectives: Acute kidney injury (AKI) complicates the course of disease in critically ill patients. Efforts to change its clinical course have failed because of the fail in the early detection. This study was designed to assess whether heat shock protein (Hsp72) is an early and sensitive biomarker of acute kidney injury (AKI) compared with kidney injury molecule (Kim-1), neutrophil gelatinase-associated lipocalin (NGAL), and interleukin-18 (IL-18) biomarkers.

Methods: A total of 56 critically ill patients fulfilled the inclusion criteria. From these patients, 17 developed AKI and 20 were selected as controls. In AKI patients, Kim-1, IL-18, NGAL, and Hsp72 were measured from 3 days before and until 2 days after the AKI diagnosis and in no-AKI patients at 1, 5 and 10 days after admission. Biomarker sensitivity and specificity were determined. To validate the results obtained with ROC curves for Hsp72, a new set of critically ill patients was included, 10 with AKI and 12 with no-AKI patients.

Results: Urinary Hsp72 levels rose since 3 days before the AKI diagnosis in critically ill patients; this early increase was not seen with any other tested biomarkers. Kim-1, IL-18, NGAL, and Hsp72 significantly increased from 2 days before AKI and remained elevated during the AKI diagnosis. The best sensitivity/specificity was observed in Kim-1 and Hsp72: 83/95% and 100/90%, respectively, whereas 1 day before the AKI diagnosis, the values were 100/100% and 100/90%, respectively. The sensibility, specificity and accuracy in the validation test for Hsp72 were 100%, 83.3% and 90.9%, respectively.

Conclusions: The biomarker Hsp72 is enough sensitive and specific to predict AKI in critically ill patients up to 3 days before the diagnosis.

Conflict of interest statement

Competing Interests: Dr. Luis E. Morales Buenrostro is an speaker of Novartis, Roche, and Sanofi México. He also has participated in advisory boards of Boehringer Ingelheim México, Eli Lilly Mexico, Astra Zeneca México and Bristol-Myers Squibb, Mexico. Dr. Norma A. Bobadilla and Dr. Jonatan Barrera have presented patent applications in México Canada, China, Europe, and US (Diagnostic method for detecting acute kidney injury using Hsp72 as a sensitive biomarker, US 2013/0065239 A1). The rest of the authors declare that there are no financial competing interests. This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials.

Figures

Figure 1. Kim-1 in critically ill patients after ICU admission.

A) Urinary Kim-1 levels assessed by ELISA from AKI and no-AKI patients. In no-AKI patients, Kim-1 was measured at days 1, 5 and 10 of their stay in the ICU; whereas, in AKI patients Kim-1 was measured from ICU admission until two days after the diagnosis of AKI. Every point represents the biomarker value in each urine sample, and the lines depicted the mean and 95% confident interval. AVG = average of urinary Kim-1 levels in no-AKI patients determinated at 1, 5, and 10 days after ICU admission. B) Specificity and sensitivity of Kim-1 to detect AKI two days before the AKIN criteria, determined by ROC analysis. C) Specificity and sensitivity of Kim-1 to detect AKI one day before the AKIN criteria, determined by ROC analysis.

Figure 2. NGAL-1 in critically ill patients after ICU admission.

A) Urinary NGAL levels assessed by ELISA from AKI and no-AKI patients. Every point represents the biomarker value in each urine sample, and the lines depicted the mean and 95% confident interval. AVG = average of urinary NGAL levels in no-AKI patients determinated at 1, 5, and 10 days after ICU admission B) Specificity and sensitivity of NGAL to detect AKI two days before the AKIN criteria, as determined by ROC analysis. C) Specificity and sensitivity of NGAL to detect AKI one day before the AKIN criteria, determined by ROC analysis.

Figure 3. IL-18 in critically ill patients after ICU admission.

A) Urinary IL-18 levels assessed by ELISA from AKI and no-AKI patients. Every point represents the biomarker value in each urine sample, and the lines depicted the mean and 95% confident interval. AVG = average of urinary IL-18 levels in no-AKI patients determinated at 1, 5, and 10 days after ICU admission. B) Specificity and sensitivity of IL-18 to detect AKI two days before the AKIN criteria, determined by ROC analysis. C) Specificity and sensitivity of IL-18 to detect AKI one day before the AKIN criteria, determined by ROC analysis.

Figure 4. Hsp72 in critically ill patients after their admission at ICU.

A) Urinary Hsp72 levels assessed by ELISA from AKI and no-AKI patients. Every point represents the biomarker value in each urine sample, and the lines depicted the mean and 95% confident interval. AVG = average of urinary Hsp72 levels in no-AKI patients determinated at 1, 5, and 10 days after ICU admission. B) Specificity and sensitivity of Hsp72 to detect AKI two days before the AKIN criteria, determined by ROC analysis. C) Specificity and sensitivity of Hsp72 to detect AKI one day before the AKIN criteria, determined by ROC analysis.