Abstract:OBJECTIVE:To explore the high risk factors of adult complex appendicitis, and to provide a reference for the development of a reasonable treatment strategy for acute appendicitis.METHODS:A retrospective case-control study was conducted to collect clinical data of 312 adult patients with acute appendicitis confirmed by pathology undergoing appendectomy, including open and laparoscopic surgery, from May 2011 to August 2016 at Affiliated Hospital of Qingdao University. Age <14 years old, pregnant women, complicating abscess around the appendix, AIDS, blood system diseases, autoimmune diseases, inflammatory bowel disease or progressive cancer patients were excluded. According to the intra-operative findings and pathological types, patients were divided into complex appendicitis(112 cases, including gangrene and perforation) and non-complex appendicitis (200 cases, including simple and non-perforated appendicitis, ie suppurative appendicitis). After comparing the clinical data of these two groups, statistically significant variables were induded for multivariate logistic regression analysis to identify risk factors of complex appendicitis, and to establish a regression model. Enter method was applied to establish the regression equation: P=ExpiΣBiXi/1+ExpΣBiXi, and to calculate the relative risk of each variable. Meanwhile, retrospective and prospective verification was performed on this predictive model (cases of acute appendicitis from September 2016 to December 2017 were further collected). The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of complex appendicitis were calculated with the regression model.RESULTS:Comparison of the clinical data between the complex appendicitis group and the non-complex appendicitis group showed that differences of 10 preoperative indexes were statistically significant, including period from abdominal pain to admission [(59.1±42.6) hours vs. (47.5±34.4) hours, t=3.051, P=0.002], white blood cell count [(12.9±3.7)×109/L vs. (9.2±4.0)×109/L, t=9.755, P<0.001], neutrophil count [(9.8±4.0)× 109/L vs.(7.1±3.9)×109/L, t=6.020, P<0.001], neutrophil percentage[(84.5±8.7)% vs.(68.2±16.0)%, t=12.754, P<0.001], C-reactive protein levels [(86.0±45.4) μg/L vs. (55.9±35.8) μg/L, t=7.614, P<0.001], serum albumin levels [(334.0±4.8) g/L vs. ( 41.0±4.3) g/L, t=16.055, P<0.001], vomiting ratio [44.6%(50/112) vs. 23.5%(47/200), χ²=14.980, P<0.001], high fever(≥39°C) ratio [16.1%(18/112) vs. 7.5%(15/200), χ²=5.577, P=0.022], the proportion of patients ≥60 years old [22.3%(25/112) vs. 13.0%(26/200), χ²=4.562, P=0.038] and previous history of appendicitis [16.1%(18/112) vs. 7.5%(15/200), χ²=5.577, P=0.022]. The above 10 variables were included in the logistic regression model for multivariate analysis. The results showed that six variables were associated with complex appendicitis. According to their strength, they were old age (≥60 years old) X1(OR=5.094), high fever (≥39°C) X2(OR=4.464), neutrophil count X6 (OR=1.269), neutrophil percentage X4 (OR=1.077), C-reactive protein level X5 (OR=1.027), and serum albumin level X3 (OR=0.763). A predictive regression model was established: P=1/[1+e(0.557+1.628X1+1.496X2-2.7X3+0.74X4+0.27X5+0.238X6)], whose sensitivity and specificity of judging complex appendicitis were 76.8%(86/112) and 90.0%(180/200),respectively. Sensitivity and specificity for predictive value of complex appendicitis in further prospective validation of the model were 76.2%(48/63) and 81.1% (30/37), respectively.CONCLUSIONS:Age ≥ 60 years old, body temperature ≥39°C, increased neutrophil count, neutrophil percentage and C-reactive protein levels, and hypoalbuminemia are risk factors for complex appendicitis. The establishment of predictive model may help determine complex appendicitis.

Progress of prediction models related to acute appendicitis

Prediction of the Severity of Acute Pancreatitis Using Machine Learning Models.

Artificial Intelligence and Acute Appendicitis: A Systematic Review of Diagnostic and Prognostic Models

Construction of a clinical prediction model for complicated appendicitis based on machine learning techniques

Accurate Prediction of Biliary Atresia with an Integrated Model Using MMP-7 Levels and Bile Acids.

Preoperatively predicting the pathological types of acute appendicitis using machine learning based on peripheral blood biomarkers and clinical features: a retrospective study

Prognostic factors and prediction models for acute aortic dissection: a systematic review

Development and Validation of a Differential Diagnosis Model for Acute Appendicitis and Henoch-Schonlein Purpura in Children.

Comparison of MPL-ANN and PLS-DA models for predicting the severity of patients with acute pancreatitis: An exploratory study

Preoperative prediction of complicated appendicitis using machine learning method

Predicting severity of acute appendicitis with machine learning methods: a simple and promising approach for clinicians

Machine Learning Predictive Models for Acute Pancreatitis: A Systematic Review.

Artificial Intelligence Differentiates Abdominal Henoch‐Schönlein Purpura from Acute Appendicitis in Children

Prediction of complex acute appendicitis based on HGS-MSVM

Accurate diagnosis of acute appendicitis in the emergency department: an artificial intelligence-based approach

Diagnosis and classification of pediatric acute appendicitis by artificial intelligence methods: An investigator-independent approach

Analysis of High Risk Factors for Acute Complex Appendicitis in Adults

Bayesian Approach to Predicting Acute Appendicitis Using Ultrasonographic and Clinical Variables in Children

Predictive model for identification of gangrenous or perforated appendicitis in adults: a multicenter retrospective study

Development and Validation of a Novel Risk Prediction Model Using Recursive Feature Elimination Algorithm for Acute-on-Chronic Liver Failure in Chronic Hepatitis B Patients with Severe Acute Exacerbation

Discovery of Urinary Proteomic Signature for Differential Diagnosis of Acute Appendicitis.