Reply

Abstract

We appreciate Dr Kawada’s interest and letter [1Kawada T. Recommendation for the independent use of specific biomarkers for various clinical settings (letter).Ann Thorac Surg. 2013; 96: 1126-1127Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar] related to our study [2Katagiri D. Doi K. Honda K. et al.Combination of two urinary biomarkers predicts acute kidney injury after adult cardiac surgery.Ann Thorac Surg. 2012; 93: 577-583Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar]. As the author pointed out, the optimal sensitivity and specificity should be determined based on the clinical situations in which each biomarker will be measured. A high-sensitivity test will be useful to rule out the disease when it yields negative results, whereas a positive result from a test with high specificity indicates a high probability of the presence of disease. Required characteristics of diagnostics depend on the features of therapeutics. In the case of postcardiac surgery acute kidney injury (AKI), it remains unclear which statistical measures (sensitivity or specificity) should be addressed for improvement of outcomes, because no therapeutic method has been demonstrated as effective for the prevention or treatment of postcardiac surgery AKI. Therefore, we became determined to pursue improvement of both sensitivity and specificity. We recognize that receiver operating characteristics (ROC) analysis is not the only means available to evaluate the performance of biomarkers. For that reason, we conducted additional analyses of the net reclassification improvement and the incremental discrimination improvement index [3Pencina M.J. D’Agostino Sr R.B. D’Agostino Jr., R.B. et al.Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond.Stat Med. 2008; 27: 157-172Crossref PubMed Scopus (4599) Google Scholar]. In addition, we demonstrated the improvement of AKI detection by adding the biomarker panel of urinary L-type fatty acid-binding protein (L-FABP) and N-acetyl-β-D-glucosaminidase (NAG) to the clinical model. This result potentially indicates that this biomarker panel can enable us to identify the optimal timing of treatment against AKI better than the current clinical practice can. We agree that any biomarker should be evaluated to assess its utility when it has been proven to refine clinical data and to improve process of patient care. Recently, Siew and colleagues [4Siew E.D. Ware L.B. Ikizler T.A. Biological markers of acute kidney injury.J Am Soc Nephrol. 2011; 22: 810-820Crossref PubMed Scopus (201) Google Scholar] suggested a potential randomized control trial design incorporating assessment of biomarker usefulness. In this design, biomarker levels were measured in all patients. If biomarkers are accurate in identifying patients at risk, then biomarker-positive patients in the treatment arm will have better outcomes than those of biomarker-positive patients in the control arm. However, similar responses will be observed in biomarker-negative patients irrespective of treatment. We emphasize that any biomarker should pass the last phase of development: disease control [5Pepe M.S. Etzioni R. Feng Z. et al.Phases of biomarker development for early detection of cancer.J Natl Cancer Inst. 2001; 93: 1054-1061Crossref PubMed Scopus (1179) Google Scholar]. This final phase addresses the issue of whether screening by biomarkers reduces the burden of disease on the population. Further investigation is ultimately necessary to demonstrate the biomarker panel of urinary L-FABP and NAG will certainly improve clinical management of AKI patients. Recommendation for the Independent Use of Specific Biomarkers for Various Clinical SettingsThe Annals of Thoracic SurgeryVol. 96Issue 3PreviewI read with interest the article by Katagiri and colleagues [1]. The authors used a combination of biomarkers for the diagnosis of acute kidney injury (AKI) after cardiac surgery. They mentioned that their strategy could be applied to biomarkers with different sensitivity and specificity values. By presenting a statistically significant increase of the area under the curve (AUC) in the receiver-operating-characteristics (ROC) curve by using two biomarkers, they concluded the advantage of using two biomarkers over the use of a single biomarker. Full-Text PDF