An interpretable machine learning model for the prevention of contrast-induced nephropathy in patients undergoing lower extremity endovascular interventions for peripheral arterial disease

Meredith Cox; J C Panagides; John Di Capua; Anahita Dua; Sanjeeva Kalva; Jayashree Kalpathy-Cramer; Dania Daye

doi:10.1016/j.clinimag.2023.05.011

An interpretable machine learning model for the prevention of contrast-induced nephropathy in patients undergoing lower extremity endovascular interventions for peripheral arterial disease

Clin Imaging. 2023 Sep:101:1-7. doi: 10.1016/j.clinimag.2023.05.011. Epub 2023 May 24.

Authors

Meredith Cox¹, J C Panagides², John Di Capua³, Anahita Dua⁴, Sanjeeva Kalva⁵, Jayashree Kalpathy-Cramer⁶, Dania Daye⁷

Affiliations

¹ Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. Electronic address: mcox12@mgh.harvard.edu.
² Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. Electronic address: john_panagides@hms.harvard.edu.
³ Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. Electronic address: jdicapua@partners.org.
⁴ Department of Surgery, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. Electronic address: adua1@mgh.harvard.edu.
⁵ Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. Electronic address: skalva@partners.org.
⁶ Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. Electronic address: jkalpathy-cramer@mgh.harvard.edu.
⁷ Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. Electronic address: ddaye@mgh.harvard.edu.

PMID: 37247523
DOI: 10.1016/j.clinimag.2023.05.011

Abstract

Background: Contrast-induced nephropathy (CIN) is a postprocedural complication associated with increased morbidity and mortality. An important risk factor for development of CIN is renal impairment. Identification of patients at risk for acute renal failure will allow physicians to make appropriate decisions to minimize the incidence of CIN. We developed a machine learning model to stratify risk of acute renal failure that may assist in mitigating risk for CIN in patients with peripheral artery disease (PAD) undergoing endovascular interventions.

Methods: We utilized the American College of Surgeons National Surgical Quality Improvement Program database to extract clinical and laboratory information associated with 14,444 patients who underwent lower extremity endovascular procedures between 2011 and 2018. Using 11,604 cases from 2011 to 2017 for training and 2840 cases from 2018 for testing, we developed a random forest model to predict risk of 30-day acute renal failure following infra-inguinal endovascular procedures.

Results: Eight variables were identified as contributing optimally to model predictions, the most important being diabetes, preoperative BUN, and claudication. Using these variables, the model achieved an area under the receiver-operating characteristic (AU-ROC) curve of 0.81, accuracy of 0.83, sensitivity of 0.67, and specificity of 0.74. The model performed equally well on white and nonwhite patients (Delong p-value = 0.955) and patients age < 65 and patients age ≥ 65 (Delong p-value = 0.659).

Conclusions: We develop a model that fairly and accurately stratifies 30-day acute renal failure risk in patients undergoing lower extremity endovascular procedures for PAD. This model may assist in identifying patients who may benefit from strategies to prevent CIN.

Keywords: Acute renal failure; Endovascular intervention; Machine learning; Peripheral artery disease; Risk assessment.

MeSH terms

Acute Kidney Injury* / chemically induced
Acute Kidney Injury* / prevention & control
Endovascular Procedures* / adverse effects
Endovascular Procedures* / methods
Humans
Lower Extremity
Peripheral Arterial Disease* / etiology
Retrospective Studies
Risk Assessment / methods
Risk Factors
Treatment Outcome