A nomogram to predict the occurrence of pseudocyst in patients with acute pancreatitis

Abstract

Background

Pseudocyst formation is common in many patients with acute pancreatitis during follow-up. Many risk factors have been proposed to be associated with the development of PP, but the predictive factors are still underexplored. The focus of this study was to investigate whether early laboratory indicators could effectively predict the occurrence of PP.

Methods

2811 AP patients hospitalized in the Second Affiliated Hospital of Soochow University between November 2008 and September 2020 were retrospectively studied. Univariate and multivariate analyses were used to screen the risk variables. The nomograms of those risk factors were validated and evaluated by logistic analysis.

Results

AP patients had a 6.1 % (172/2811) incidence of PP. In a univariate analysis, the development of PP was correlated with serum lactate dehydrogenase (LDH), albumin (ALB), calcium (Ca), hemoglobin (Hb), organ dysfunction, CT severity index (CTSI), etiology, age, etc. Further logistic regression analysis showed that the risk factors were different between hyperlipidemic pancreatitis patients (LDH, ALB and Ca) and non-hyperlipidemic pancreatitis patients (LDH, Hb, ALB and Ca). A nomogram based on the identified risk factors was developed. Our model showed good discrimination ability, with a boostrap - corrected C index of 0.905 (95 % CI = 0.875–0.935), and had well-fitted calibration curves. The area under the curve (AUC) of the nomogram were 0.905 (95 % CI = 0.875–0.935) and 0.933 (95 % CI = 0.890–0.975) in the training and validation groups, respectively. The results of DCA indicated that the nomogram may have clinic usefulness.

Conclusions

The nomogram that incorporates early laboratory data (LDH, Hb, ALB, and Ca) in AP patients is able to predict the incidence of PP with greater accuracy than the CTSI and AP severity.