Interpretable machine learning model for early prediction of acute kidney injury in patients with rhabdomyolysis

Abstract

Background: Rhabdomyolysis (RM) is a complex set of clinical syndromes. RM-induced acute kidney injury (AKI) is a common illness in war and military operations. This study aimed to develop an interpretable and generalizable model for early AKI prediction in patients with RM.

Methods: Retrospective analyses were performed on 2 electronic medical record databases: the eICU Collaborative Research Database and the Medical Information Mart for Intensive Care III database. Data were extracted from the first 24 hours after patient admission. Data from the two datasets were merged for further analysis. The extreme gradient boosting (XGBoost) model with the Shapley additive explanation method (SHAP) was used to conduct early and interpretable predictions of AKI.

Results: The analysis included 938 eligible patients with RM. The XGBoost model exhibited superior performance (area under the receiver operating characteristic curve [AUC] = 0.767) compared to the other models (logistic regression, AUC = 0.711; support vector machine, AUC = 0.693; random forest, AUC = 0.728; and naive Bayesian, AUC = 0.700).

Conclusion: Although the XGBoost model performance could be improved from an absolute perspective, it provides better predictive performance than other models for estimating the AKI in patients with RM based on patient characteristics in the first 24 hours after admission to an intensive care unit. Furthermore, including SHAP to elucidate AKI-related factors enables individualized patient treatment, potentially leading to improved prognoses for patients with RM.

Keywords

Acute kidney injury / eICU-CRD / MIMIC-III / Rhabdomyolysis / XGBoost

Conflict of interest statement

The authors declare no conflict of interest.

Author contributions

All authors contributed to the design of the study concept. Zhang X, Liang X, Fu Z, Zhou Y, Fang Y, and Liu X contributed to collecting data, developing models, and drafting the manuscript. Zhang X, Liang X, Fu Z, Zhou Y, and Yuan Q contributed to data analysis and interpretation. Zhang X, Liang X, Yuan Q, Liu R, Hong Q, and Liu C contributed to cleaning the data and algorithm programming. Fu Z, Yuan Q, Liu R, Hong Q, and Liu contributed to the statistical analysis. Zhang X, Liang X, Zhou Y, Hong Q, and Liu C contributed to the editing and approved the manuscript. All authors revised the manuscript drafts and approved the final version for submission.

Funding

This study was supported by National Natural Science Foundation of China (82200780) and China Postdoctoral Science Foundation (2022 M723899).

Ethical approval of studies and informed consent

The study followed the principles of the Declaration of Helsinki as revised in 2013. To apply for permission to access the database, researchers must complete the National Institutes of Health’s Web-based course known as Protecting Human Research Participants (certification number 29493483). Research using the eICU-CRD is exempt from institutional review board approval due to the retrospective design, lack of direct patient intervention, and the security schema, for which the reidentification risk was certified as meeting safe harbor standards by an independent privacy expert (Privacert, Cambridge, MA, Health Insurance Portability and Accountability Act Certification number 1031219-2). The data in the MIMIC-III database have been previously deidentifed, and the institutional review boards of the Massachusetts Institute of Technology (number 0403000206) and Beth Israel Deaconess Medical Center (number 2001-P-001699/14) both approved the use of the database for research. Without a requirement for individual patient, written informed consent because data were deidentified and publicly available.

Acknowledgments

None.

References

[1] Petejova N, Martinek A. Acute kidney injury due to rhabdomyolysis and renal replacement therapy: a critical review. Crit Care. 2014; 18(3):224. doi:10.1186/cc13897

[2] Boutaud O, Roberts LJ 2nd. Mechanism-based therapeutic approaches to rhabdomyolysis-induced renal failure. Free Radic Biol Med. 2011; 51(5):1062-1067. doi:10.1016/j.freeradbiomed.2010.10.704

[3] Abu-Zidan FM, Idris K, Cevik AA. Prehospital management of earthquake crush injuries: a collective review. Turk J Emerg Med. 2023; 23(4):199-210. doi:10.4103/tjem.tjem_201_23

[4] Aslan M, Bilgi DO. Management of kidney injury in critically ill patients with earthquake-induced crush syndrome: a case series of 18 patients. Ther Apher Dial. 2023; 28:314-320. doi:10.1111/1744-9987.14082

[5] Luo Y, Liu C, Li D, et al. Progress in the diagnostic and predictive evaluation of crush syndrome. Diagnostics (Basel). 2023; 13(19):3034. doi:10.3390/diagnostics13193034

[6] Legrand M, Clark AT, Neyra JA, Ostermann M. Acute kidney injury in patients with burns. Nat Rev Nephrol. 2024; 20(3):188-200. doi:10.1038/s41581-023-00769-y

[7] Adnan BH. A systematic review on outcomes of patients with heatstroke and heat exhaustion. Open Access Emerg Med. 2023; 15:343-354. doi:10.2147/OAEM.S419028

[8] Finke SR, Janig C, Deschler A, et al.Medical emergencies during running events. Notf Rett Med. 2023; 26(3):189-198. doi:10.1007/s10049-021-00959-w

[9] Wang Z, Shen J, Zhang L. Omeprazole was safely reused in a rhabdomyolysis patient associated with proton pump inhibitors: a case report. Clin Case Rep. 2023; 11(11):e8189. doi:10.1002/ccr3.8189

[10] Preger A, Wei R, Berg B, Golomb BA. COVID-19-associated rhabdomyolysis: a scoping review. Int J Infect Dis. 2023; 136:115-126. doi:10.1016/j.ijid.2023.09.002

[11] Matsuki Y, Sugihara T, Kihara T, et al. COVID-19-triggered acute liver failure and rhabdomyolysis: a case report and review of the literature. Viruses. 2023; 15(7):1445. doi:10.3390/v15071445

[12] Stahl K, Rastelli E, Schoser B. A systematic review on the definition of rhabdomyolysis. J Neurol. 2020; 267(4):877-882. doi:10.1007/s00415-019-09185-4

[13] Melli G, Chaudhry V, Cornblath DR. Rhabdomyolysis: an evaluation of 475 hospitalized patients. Medicine (Baltimore). 2005; 84(6):377-385. doi:10.1097/01.md.0000188565.48918.41

[14] Candela N, Silva S, Georges B, et al. Short- and long-term renal outcomes following severe rhabdomyolysis: a French multicenter retrospective study of 387 patients. Ann Intensive Care. 2020; 10(1):27. doi:10.1186/s13613-020-0645-1

[15] Nielsen FE, Cordtz JJ, Rasmussen TB, Christiansen CF. The association between rhabdomyolysis, acute kidney injury, renal replacement therapy, and mortality. Clin Epidemiol. 2020; 12:989-995. doi:10.2147/CLEP.S254516

[16] Chavez LO, Leon M, Einav S, Varon J. Beyond muscle destruction: a systematic review of rhabdomyolysis for clinical practice. Crit Care. 2016; 20(1):135. doi:10.1186/s13054-016-1314-5

[17] Safari S, Yousefifard M, Hashemi B, et al. The value of serum creatine kinase in predicting the risk of rhabdomyolysis-induced acute kidney injury: a systematic review and meta-analysis. Clin Exp Nephrol. 2016; 20(2):153-161. doi:10.1007/s10157-015-1204-1

[18] El-Abdellati E, Eyselbergs M, Sirimsi H, et al. An observational study on rhabdomyolysis in the intensive care unit. Exploring its risk factors and main complication: acute kidney injury. Ann Intensive Care. 2013; 3(1):8. doi:10.1186/2110-5820-3-8

[19] Wen T, Mao Z, Liu C, Wang X, Tian S, Zhou F. Association between admission serum phosphate and risk of acute kidney injury in critically ill patients with rhabdomyolysis: a retrospective study based on MIMIC-III. Injury. 2023; 54(1):189-197. doi:10.1016/j.injury.2022.10.024

[20] Rodriguez E, Soler MJ, Rap O, Barrios C, Orfila MA, Pascual J. Risk factors for acute kidney injury in severe rhabdomyolysis. PloS One. 2013; 8(12):e82992. doi:10.1371/journal.pone.0082992

[21] McMahon GM, Zeng X, Waikar SS. A risk prediction score for kidney failure or mortality in rhabdomyolysis. JAMA Intern Med. 2013; 173(19):1821-1828. doi:10.1001/jamainternmed.2013.9774

[22] Simpson JP, Taylor A, Sudhan N, Menon DK, Lavinio A. Rhabdomyolysis and acute kidney injury: creatine kinase as a prognostic marker and validation of the McMahon score in a 10-year cohort: a retrospective observational evaluation. Eur J Anaesthesiol. 2016; 33(12):906-912. doi:10.1097/EJA.0000000000000490

[23] Pollard TJ, Johnson AEW, Raffa JD, Celi LA, Mark RG, Badawi O. The eICU collaborative research database, a freely available multi-center database for critical care research. Sci Data. 2018; 5:180178. doi:10.1038/sdata.2018.178

[24] Johnson AE, Stone DJ, Celi LA, Pollard TJ. The MIMIC code repository: enabling reproducibility in critical care research. J Am Med Inform Assoc. 2018; 25(1):32-39. doi:10.1093/jamia/ocx084

[25] Johnson AE, Pollard TJ, Shen L, et al. MIMIC-III, a freely accessible critical care database. Sci Data. 2016;3:160035. doi:10.1038/sdata.2016.35

[26] Liu C, Liu X, Mao Z, et al. Interpretable machine learning model for early prediction of mortality in ICU patients with rhabdomyolysis. Med Sci Sports Exerc. 2021; 53(9):1826-1834. doi:10.1249/MSS.0000000000002674

[27] Liu C, Yuan Q, Mao Z, et al. Development and validation of a model for the early prediction of the RRT requirement in patients with rhabdomyolysis. Am J Emerg Med. 2021; 46:38-44. doi:10.1016/j.ajem.2021.03.006

[28] Knaus WA, Wagner DP, Draper EA, et al. The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. Chest. 1991; 100(6):1619-1636. doi:10.1378/chest.100.6.1619

[29] Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-Related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the working group on sepsis-related problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996; 22(7):707-710. doi:10.1007/bf01709751

[30] Liu X, Hu P, Yeung W, et al. Illness severity assessment of older adults in critical illness using machine learning (ELDER-ICU): an international multicentre study with subgroup bias evaluation. Lancet Digit Health. 2023; 5(10):e657-e667. doi:10.1016/S2589-7500(23)00128-0

[31] Handelman GS, Kok HK, Chandra RV, Razavi AH, Lee MJ, Asadi H. eDoctor: machine learning and the future of medicine. J Intern Med. 2018; 284(6):603-619. doi:10.1111/joim.12822

[32] Meyer A, Zverinski D, Pfahringer B, et al. Machine learning for real-time prediction of complications in critical care: a retrospective study. Lancet Respir Med. 2018; 6(12):905-914. doi:10.1016/S2213-2600(18)30300-X

[33] Tseng PY, Chen YT, Wang CH, et al. Prediction of the development of acute kidney injury following cardiac surgery by machine learning. Crit Care. 2020; 24(1):478. doi:10.1186/s13054-020-03179-9

[34] Cour M, Bresson D, Hernu R, Argaud L. SOFA score to assess the severity of the post-cardiac arrest syndrome. Resuscitation. 2016; 102:110-115. doi:10.1016/j.resuscitation.2016.03.001

[35] Boudhabhay I, Poillerat V, Grunenwald A, et al. Complement activation is a crucial driver of acute kidney injury in rhabdomyolysis. Kidney Int. 2021; 99(3):581-597. doi:10.1016/j.kint.2020.09.033

[36] Cabral BMI, Edding SN, Portocarrero JP, Lerma EV. Rhabdomyolysis. Dis Mon. 2020; 66(8):101015. doi:10.1016/j.disamonth.2020.101015

[37] Okubo K, Kurosawa M, Kamiya M, et al. Macrophage extracellular trap formation promoted by platelet activation is a key mediator of rhabdomyolysis-induced acute kidney injury. Nat Med. 2018; 24(2):232-238. doi:10.1038/nm.4462

[38] Liu C, Yuan Q, Mao Z, et al. The top 100 most cited articles on rhabdomyolysis: a bibliometric analysis. Am J Emerg Med. 2020; 38(9):1754-1759. doi:10.1016/j.ajem.2020.05.031

[39] Liu C, Chi K, Geng X, et al. Exogenous biological renal support improves kidney function in mice with rhabdomyolysis-induced acute kidney injury. Front Med (Lausanne). 2021; 8:655787. doi:10.3389/fmed.2021.655787

[40] Youssef A, Pencina M, Thakur A, Zhu T, Clifton D, Shah NH. External validation of AI models in health should be replaced with recurring local validation. Nat Med. 2023; 29(11):2686-2687. doi:10.1038/s41591-023-02540-z

[41] Dvijotham KD, Winkens J, Barsbey M, et al. Enhancing the reliability and accuracy of AI-enabled diagnosis via complementarity-driven deferral to clinicians. Nat Med. 2023; 29(7):1814-1820. doi:10.1038/s41591-023-02437-x

[42] Topol EJ. As artificial intelligence goes multimodal, medical applications multiply. Science. 2023; 381(6663):adk6139. doi:10.1126/science.adk6139

[43] Voelker R. Google Health's chief clinical officer talks about incorporating AI in health care. JAMA. 2023; 330(14):1315-1317. doi:10.1001/jama.2023.19137

[44] Hernandez AF, Lindsell CJ. The future of clinical trials: artificial to augmented to applied intelligence. JAMA. 2023; 330(21):2061-2063. doi:10.1001/jama.2023.23822

[45] Perni S, Lehmann LS, Bitterman DS. Patients should be informed when AI systems are used in clinical trials. Nat Med. 2023; 29(8):1890-1891. doi:10.1038/s41591-023-02367-8

[46] Lawrentschuk N. Electronic medical records—a disappointing mirage for clinicians and research. BJU Int. 2023; 131(Suppl 4):4-5. doi:10.1111/bju.16082

[47] Kulkarni PA, Singh H. Artificial intelligence in clinical diagnosis: opportunities, challenges, and hype. JAMA. 2023; 330(4):317-318. doi:10.1001/jama.2023.11440

[48] Hunter DJ, Holmes C. Where medical statistics meets artificial intelligence. N Engl J Med. 2023; 389(13):1211-1219. doi:10.1056/NEJMra2212850

[49] Volovici V, Syn NL, Ercole A, Zhao JJ, Liu N. Steps to avoid overuse and misuse of machine learning in clinical research. Nat Med. 2022; 28(10):1996-1999. doi:10.1038/s41591-022-01961-6