PARCCS: A Machine Learning Risk-Prediction Model for Acute Peripartum Cardiovascular Complications During Delivery Admissions

Abstract

Background: Maternal mortality in the United States remains high, with cardiovascular (CV) complications being a leading cause.

Objectives: The purpose of this paper was to develop the PARCCS (Prediction of Acute Risk for Cardiovascular Complications in the Peripartum Period Score) for acute CV complications during delivery.

Methods: Data from the National Inpatient Sample (2016-2020) and International Classification of Diseases, Tenth Revision codes to identify delivery admissions were used. Acute CV/renal complications were defined as a composite of pre-eclampsia/eclampsia, peripartum cardiomyopathy, renal complications, venous thromboembolism, arrhythmias, and pulmonary edema. A risk prediction model, PARCCS, was developed using machine learning consisting of 14 variables and scored out of 100 points.

Results: Of the 2,371,661 pregnant patients analyzed, 7.0% had acute CV complications during delivery hospitalization. Patients with CV complications had a higher prevalence of comorbidities and were more likely to be of Black race and lower income. The PARCCS variables included electrolyte imbalances (13 points [p]), age (3p for age <20 years), cesarean delivery (4p), obesity (5p), pre-existing heart failure (28p), multiple gestations (4p), Black race (2p), gestational hypertension (3p), low income (1p), gestational diabetes (2p), chronic diabetes (6p), prior stroke (22p), coagulopathy (5p), and nonelective admission (2p). Using the validation set, the performance of the model was evaluated with an area under the receiver-operating characteristic curve of 0.68 and a 95% CI of 0.67 to 0.68.

Conclusions: PARCCS has the potential to be an important tool for identifying pregnant individuals at risk of acute peripartum CV complications at the time of delivery. Future studies should further validate this score and determine whether it can improve patient outcomes.

Keywords: CVD; cardiovascular disease; machine learning; preeclampsia; risk prediction.

Graphical abstract

Figure 1

Study Flow Diagram Variable ranking and transformation: The described machine learning method employs a random forest algorithm to generate decision trees, utilizing bootstrapping to reduce overfitting. Variable ranking based on mean decrease impurity measurement aids in selecting important variables that undergo a transformation, such as stratifying continuous variables like age into categories. Intermediate model analysis: The scoring model's performance was evaluated by varying the number of variables, and the optimal trade-off between model complexity and prediction accuracy was determined through a parsimony plot. Final model analysis: The performance of the chosen model was subsequently assessed using the testing dataset. ICD-10 = International Classification of Diseases-10th Revision.

Central Illustration

PARCCS: A Machine Learning Risk-Prediction Model A total of 14 variables were selected based on predictive ability as well as clinical relevance. On performance evaluation using the same testing dataset, an area under the curve of 0.68 (95% CI: 0.67-0.68) was achieved. The PARCCS has a total score range of 0 to 100.

Figure 2

Parsimony Plot In the initial stage, 20 variables were ranked. This was subsequently narrowed down to 14 variables, which led to the creation of a parsimonious model. Additional variables beyond the chosen 14 did not contribute significantly to improving the model's performance. Detailed information regarding this selection process can be found in the methods section. COPD = chronic obstructive pulmonary disease.

Figure 3

Receiver Operating Characteristics Curve AUC = area under the curve.