Early postoperative changes in routine blood biomarkers and postoperative complications after lung resection: a retrospective cohort study

Abstract Background Early inflammatory and immune responses after lung resection may be associated with postoperative complications. However, whether changes in routine blood biomarkers over the perioperative period provide information on early postoperative risk beyond single time-point measurements remains unclear. This study evaluated the associations of changes in white blood cell count, lymphocyte percentage, albumin, and red cell distribution width from before surgery to 24 h after surgery with postoperative complications, and assessed whether these dynamic measures provided incremental discrimination beyond baseline clinical variables. Methods Patients undergoing lung resection were retrospectively included. Preoperative measurements were defined as the most recent laboratory results obtained within 7 days before surgery, and postoperative measurements were those obtained 24 h after surgery. Changes (Δ) were calculated as the postoperative value minus the preoperative value. The primary outcome was any postoperative complication; secondary outcomes included postoperative pulmonary complications (PPC), postoperative length of stay, and chest tube duration. Logistic regression was used to assess the associations of biomarker changes with overall postoperative complications and PPC. Three prespecified nested models were developed sequentially using clinical variables alone, clinical variables plus ΔWBC, and clinical variables plus ΔWBC and ΔLymphocyte. Incremental discrimination was assessed using receiver operating characteristic curves, DeLong tests, and likelihood ratio tests. Internal optimism correction was performed using 1000 bootstrap resamples, with additional calibration and decision curve analyses. Results Among 908 patients included in the primary analysis, 223 (24.6%) developed at least one postoperative complication. At 24 h after surgery, median WBC increased from 5.57 (IQR, 4.71–6.69) to 10.54 (8.40–12.91) ×10^9/L, lymphocyte percentage decreased from 31.90% (26.50–37.90) to 8.70% (6.00–12.80), and albumin decreased from 41.20 (39.05–43.40) to 37.60 (35.30–40.10) g/L. After adjustment, ΔWBC was associated with overall postoperative complications (OR, 1.13; 95% CI, 1.08–1.19; P < 0.001) and PPC (OR, 1.12; 95% CI, 1.07–1.18; P < 0.001). ΔLymphocyte was also independently associated with overall postoperative complications (OR, 0.97; 95% CI, 0.95–0.99; P = 0.001) and PPC (OR, 0.98; 95% CI, 0.96–0.99; P = 0.009). The AUC of the clinical model was 0.729, increasing to 0.761 after addition of ΔWBC (P = 0.002) and to 0.764 after further addition of ΔLymphocyte (P = 0.410). The optimism-corrected C-indexes from 1000 bootstrap resamples were 0.702, 0.734, and 0.737, respectively. Conclusions Changes in routine blood biomarkers at 24 h after lung resection were independently associated with postoperative complications, with ΔWBC showing the clearest incremental discrimination beyond baseline clinical variables. ΔWBC may have potential as an early adjunct to conventional clinical risk assessment, although its clinical utility requires validation in independent cohorts.