Lactate in Drainage Fluid to Predict Complications in Robotic Esophagectomies-A Pilot Study in a Matched Cohort

Abstract

Background/Objectives: Despite advances in minimally invasive procedures, anastomotic leakages (ALs) after esophageal resections mark the most feared complication. Its early detection can lead to quick interventional treatment with improved survival. Nonetheless, early detection remains challenging, and scores are imprecise and complex. Methods: In our study we analyzed mediastinal drainage fluid to find parameters suggesting AL even before it became clinically evident and correlated them to routine biomarkers. All patients with AL after robotically assisted esophageal resections were included and matched 1:1 with uneventful controls. Additionally, transhiatal distal esophageal resections operated during this period were included. Drainage fluid was collected on postoperative days (PODs) 1-4 with consecutive blood gas analysis. Test quality was determined by the area under the curve (AUC) of the receiver operating characteristic curve (ROC). Results: In total, 40 patients were included, with 17 developing AL. There were no significant differences in gender, age, BMI or oncological treatment. The 30-day morbidity rate was 65.0%. The study was restricted to events in the first 12 days. While lactate value in drainage fluid differed significantly from POD 3 onwards in the two groups, serum CRP remained without significant differences. We developed the LacCRP score (CRP/30 + lactate/2). The AUC on POD 3 was 0.96, with a sensitivity and specificity of 100% and 75%, respectively. An estimator of 1.08 was found in multivariate analysis: one-point increase in the LacCRP score increases AL probability by 8%. Conclusions: This study demonstrates that postoperative lactate determinations in drainage fluid can predict AL after esophageal resection, and its combination with serum CRP results in a reliable LacCRP score.

Keywords: Ivor Lewis; anastomotic leakage; esophageal resection; lactate; prediction of complications.

Figure 1

(a) Serum C-reactive protein (CRP) and (b) lactate detected in drainage fluid early after operation in patients with and without prompt anastomotic leakage. The graphs show mean values; the error bars represent the 95% confidence interval.

Figure 2

(a) Noble und Underwood (NUn) score early after operations in patients with and without prompt anastomotic leakage. The NUn score is calculated for serum C-reactive protein (CRP), white blood cell count (WBC), and albumin [11.3894 + (0.005 × CRP) + (WBC × 0.186) − (0.174 × albumin)]. (b) Evolution of LacCRP score early after operations in patients with and without prompt anastomotic leakage. The LacCRP score is calculated for serum C-reactive protein (CRP) and lactate in drainage fluid [CRP/30 + lactate/2]. The graph shows mean values; the error bars represent the 95% confidence interval.

Figure 3

ROC analysis of the predictive quality of the LacCRP score on POD 3 (a) and POD 4 (b). The AUC is 0.96 and 0.87 on POD3 and POD 4, respectively, with a cutoff of 9.5.