A nomogram to predict prolonged postoperative ileus after intestinal resection for Crohn's disease

Abstract

Purpose: We aim to identify the risk factors of PPOI in patients with CD and create a nomogram for prediction of PPOI for CD.

Methods: Data on 462 patients who underwent partial intestinal resection for CD in Jin-ling Hospital between January 2019 and June 2021 were retrospectively collected. Univariate and multivariate analyses were performed to determine the risk factors for PPOI and we used the risk factors to create a nomogram. Then we used the Bootstrap-Concordance index and calibration diagrams to evaluate the performance of the Nomogram. Decision curve analysis was performed to evaluate clinical practicability of the model.

Results: The incidence of PPOI was 27.7% (n of N). Course of CD ≥ 10 years, operation time ≥ 154 min, the lowest mean arterial pressure ≤ 76.2 mmHg, in-out balance per body weight ≥ 22.90 ml/kg, post-op day 1 infusion ≥ 2847 ml, post-op lowest K⁺ ≤ 3.75 mmol/L, and post-op day 1 procalcitonin ≥ 2.445 ng/ml were identified as the independent risk factors of PPOI in patients with CD. The nomogram we created by these risk factors presented with good discriminative ability (concordance index 0.723) and was moderately calibrated (bootstrapped concordance index 0.704). The results of decision curve analysis showed that the nomogram was clinically effective within probability thresholds in the 8 to 66% range.

Conclusion: The nomogram we developed is helpful to evaluate the risk of developing PPOI after partial intestinal resection for CD. Clinicians can take more necessary measures to prevent PPOI in CD's patients or at least minimize the incidence.

Keywords: Crohn’s disease; Nomogram; Prolonged postoperative ileus; Risk factors.