There were 38 septic AKI patients in the present study population. Among 23 non-septic AKI patients, 21 (91.3%) were diagnosed with septic subclinical AKI (Table 1). Presepsin and PLR predicted the progression of septic subclinical AKI to septic AKI. In particular, the AUC, sensitivity, and specificity of presepsin on Day 1 for predicting progression were 0.69, 82%, and 52%, respectively, corresponding to a higher sensitivity than those observed for presepsin on Days 2, 3, and 5, as well as PLR (Table 2). These results demonstrate that presepsin on Day 1 is a good “rule out” test for predicting the progression of septic subclinical AKI to septic AKI.
Our results will contribute to the prevention and management of progression of septic subclinical AKI to septic AKI. Septic AKI is associated with poor patient prognoses, including death [5]. Serum creatinine is a delayed, low-sensitivity, potentially misleading biomarker of AKI [15, 16], as it is influenced by many confounding factors [17] and thus may fail to detect the progression of septic subclinical AKI to septic AKI in septic patients who are at increased risk of death. Presepsin, on the other hand, was found to predict the progression of septic subclinical AKI to septic AKI in the absence of diagnostic increases in serum creatinine (Table 2). Thus, elevated presepsin might help lead to earlier diagnosis and rapid introduction of conventional interventions to prevent the progression of septic subclinical AKI to septic AKI. The use of elevated presepsin as a predictor may thus increase the likelihood that a treatment will be successful, since changes in its levels occur rapidly and can be measured quickly (in less than 17 minutes), whereas it can take hours to detect/measure changes in NGAL and days for serum creatinine.
Presepsin cut-off values indicated in the present study for predicting the progression of septic subclinical AKI to septic AKI (Table 2), and prognosis in septic subclinical AKI patients (Table 3), were higher than those previously reported to predict severe sepsis and septic shock [13, 14]. Nakamura et al. reported a significant negative correlation between presepsin levels and estimated glomerular filtration rate in both non-sepsis and sepsis patients [18]. Presepsin values were significantly elevated in ICU patients with renal failure and end-stage kidney disease, irrespective of whether they had sepsis or not. Presepsin values in patients with sepsis ranged from 2,632 to 20,000 pg/ml, while those in patients without sepsis ranged from 2,134 to 19,633 pg/ml [18]; the presepsin cut-off values identified in the present study were lower than these. Our results demonstrate that a lower presepsin cut-off value should be adopted for predicting the progression of septic subclinical AKI to septic AKI compared to that used for ICU patients with renal failure and end-stage kidney disease.
The AUC, sensitivity, and specificity of PLR for predicting the progression of septic subclinical acute kidney injury to septic AKI were 0.67, 71%, and 62%, respectively (Table 2), and PLR was demonstrated to be a predictor of septic subclinical AKI in multivariate logistic regression analyses (Additional file 2: Table S2). Previously, Smith et al. reported that PLR is a significant prognostic marker in pancreatic cancer patients [19]. PLR was suggested to be useful for predicting the progression of septic subclinical AKI to septic AKI in the present study. PLR can be obtained at low cost and rapidly, and is convenient for bedside use. In small- and medium-sized hospitals where presepsin values cannot be easily measured, PLR provides information for aggressive and supportive therapy in septic subclinical AKI patients within the first few hours of ICU admission. Among the various variables which can be tested to assess inflammation, platelet and lymphocyte counts (which are used to calculate PLR) are important for predicting the progression of septic subclinical AKI to septic AKI.