The usefulness of presepsin in the diagnosis of postoperative infectious complications after gastrectomy for gastric cancer: a prospective cohort study

Abstract

This prospective study aimed to evaluate presepsin use as a biomarker of on postoperative infectious complications after gastrectomy, compared to C-reactive protein (CRP), white blood cells (WBCs), and neutrophils (Neuts). Overall, 108 patients were enrolled between October 2019 and December 2020. Presepsin, CRP, WBC, and Neut levels were measured preoperatively and on postoperative days (PODs) 1, 3, 5, and 7, using a postoperative morbidity survey. Grade II or higher infectious complications occurred in 18 patients (16.6%). Presepsin levels on all evaluated PODs were significantly higher in the infectious complication group than in the non-complication group (p = 0.002, p < 0.0001, p < 0.0001, and p = 0.025, respectively). The area under the curve (AUC) values were the highest for presepsin on PODs 3 and 7 (0.89 and 0.77, respectively) and similar to that of CRP, with a high value > 0.8 (0.86) on POD 5. For presepsin, the optimal cut-off values were 298 pg/mL (sensitivity, 83.3%; specificity, 83.3%), 278 pg/mL (sensitivity, 83.3%; specificity, 82.2%), and 300 pg/mL (sensitivity, 83.3%; specificity, 82%) on PODs 3, 5, and 7, respectively. Presepsin levels on PODs 3, 5, and 7 after gastrectomy is a more useful biomarker of postoperative infectious complications compared to CRP, WBCs, and Neuts, with a high sensitivity and specificity.

Diagnostic efficacy of serum presepsin for postoperative infectious complications: a meta-analysis

“…Then, after excluding 254 studies that did not meet the inclusion criteria, the remaining 29 studies were thoroughly reassessed via full-text screening. Additional 20 studies were excluded due to various reasons (e.g., lack of specific data), resulting in the inclusion of eight studies ( 1 , 15 , 16 , 19 , 21 – 24 ) in this meta-analysis. The inter-rater reliability for the assessment of study inclusion was 0.84, indicating a very good agreement between the two investigators.…”

“…This meta-analysis included eight studies published between 2015 and 2022 that evaluated presepsin as a biomarker for PICs ( Table 1 ). The studies were conducted in several countries, including Japan (n = 6) ( 1 , 15 , 21 – 24 ), Russia (n = 1) ( 19 ), and China (n = 1) ( 16 ). The total number of patients across all studies was 984, with sample sizes ranging from 13 to 490 patients.…”

“…All studies were found to possess a low risk of bias in terms of the reference standard, flow and timing, and applicability ( Figure 2 ) ( 1 , 15 , 16 , 19 , 21 – 24 ). However, as regards the index test domain, the risk of bias in all studies was unclear due to the lack of predefined cutoff values.…”

“…Regarding the diagnostic efficacy of presepsin for PIC, extraction of data from eight studies enabled the computation of pooled sensitivity of 76% (95% CI 68%–82%, I 2 = 2.9%) and specificity of 83% (95% CI 75%–89%, I 2 = 84.5%) ( Figure 3 ) ( 1 , 15 , 16 , 19 , 21 – 24 ). The HSROC curve showed an area under the curve (AUC) of 0.77 (95% CI 0.73–0.81) ( Figure 4 ).…”

“…These desirable characteristics potentiate its clinical utility for the early diagnosis of bacterial infections. While individual studies have evaluated presepsin for PICs, its diagnostic accuracy may widely vary depending on surgical procedure, measurement timing, and threshold values ( 1 , 15 , 18 , 19 ). A comprehensive synthesis of existing data regarding the diagnostic precision and clinical value of presepsin for PICs across surgical specialties is lacking.…”