A prediction model to identify patients without a concerning intraabdominal diagnosis

Machine learning based prediction models for analyzing risk factors in patients with acute abdominal pain: a retrospective study

Background

Acute abdominal pain (AAP) is a common symptom presented in the emergency department (ED), and it is crucial to have objective and accurate triage. This study aims to develop a machine learning-based prediction model for AAP triage. The goal is to identify triage indicators for critically ill patients and ensure the prompt availability of diagnostic and treatment resources.

Methods

In this study, we conducted a retrospective analysis of the medical records of patients admitted to the ED of Wuhan Puren Hospital with acute abdominal pain in 2019. To identify high-risk factors, univariate and multivariate logistic regression analyses were used with thirty-one predictor variables. Evaluation of eight machine learning triage prediction models was conducted using both test and validation cohorts to optimize the AAP triage prediction model.

Results

Eleven clinical indicators with statistical significance (p < 0.05) were identified, and they were found to be associated with the severity of acute abdominal pain. Among the eight machine learning models constructed from the training and test cohorts, the model based on the artificial neural network (ANN) demonstrated the best performance, achieving an accuracy of 0.9792 and an area under the curve (AUC) of 0.9972. Further optimization results indicate that the AUC value of the ANN model could reach 0.9832 by incorporating only seven variables: history of diabetes, history of stroke, pulse, blood pressure, pale appearance, bowel sounds, and location of the pain.

Conclusion

The ANN model is the most effective in predicting the triage of AAP. Furthermore, when only seven variables are considered, including history of diabetes, etc., the model still shows good predictive performance. This is helpful for the rapid clinical triage of AAP patients and the allocation of medical resources.

Thrombocytopenia on the first day of emergency department visit predicts higher risk of acute kidney injury among elderly patients

Background

Few studies have addressed risk factors for acute kidney injury (AKI) in geriatric patients. We investigated whether thrombocytopenia was a risk factor for AKI in geriatric patients with medical illnesses.

Methods

A prospective cohort study was conducted, by recruiting elderly (≥65 years) patients who visited the emergency department (ED) for medical illnesses during 2014. They all received hemogram for platelet count determination, and were stratified according to the presence of thrombocytopenia (platelets, <150 K/μL) during their initial ED evaluation. They were prospectively followed up during their ED stay. We analyzed the relationship between the diagnosis of thrombocytopenia and subsequent AKI after ED stay, using Cox proportional hazard modeling, with platelet count as a continuous variable or thrombocytopenia as a categorical variable.

Results

Of 136 elderly patients (mean age of 80.7 ± 8.2 years, 40% with chronic kidney disease, and 39% with diabetes) enrolled, 22.8% presented with thrombocytopenia, without differences in baseline renal function. After a mean ED stay of 4.4 ± 2.1 days, 41.9% developed AKI (52.6% Kidney Disease Improving Global Outcomes [KDIGO] grade 1, 24.6% grade 2, and 22.8% grade 3). Patients with higher AKI severity had stepwise lower platelet counts compared to those without AKI. The Cox proportional hazard model revealed that lower platelet count as a continuous variable (hazard ratio [HR] 0.984, 95% confidence interval [CI] 0.975–0.994) and as a categorical variable (presence of thrombocytopenia) (HR 1.86, 95% CI 1.06–3.27) increased the risk of AKI. The sensitivity analyses accounting for nephrotoxic medications use, including non-steroidal anti-inflammatory drugs, vancomycin, and contrast, yielded similar results.

Discussion

Thrombocytopenia is common among ED-visiting elderly, and the potential relationship between platelet counts and the risk of AKI suggests the utility of checking hemogram for those at-risk ofdeveloping adverse renal events.

Conclusion

Thrombocytopenia on initial presentation might indicate an increased risk of AKI among elderly patients with medical illnesses.