1
|
Sun R, Li S, Wei Y, Hu L, Xu Q, Zhan G, Yan X, He Y, Wang Y, Li X, Luo A, Zhou Z. Development of interpretable machine learning models for prediction of acute kidney injury after noncardiac surgery: a retrospective cohort study. Int J Surg 2024; 110:2950-2962. [PMID: 38445452 DOI: 10.1097/js9.0000000000001237] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/01/2023] [Accepted: 02/15/2024] [Indexed: 03/07/2024]
Abstract
BACKGROUND Early identification of patients at high-risk of postoperative acute kidney injury (AKI) can facilitate the development of preventive approaches. This study aimed to develop prediction models for postoperative AKI in noncardiac surgery using machine learning algorithms. The authors also evaluated the predictive performance of models that included only preoperative variables or only important predictors. MATERIALS AND METHODS Adult patients undergoing noncardiac surgery were retrospectively included in the study (76 457 patients in the discovery cohort and 11 910 patients in the validation cohort). AKI was determined using the KDIGO criteria. The prediction model was developed using 87 variables (56 preoperative variables and 31 intraoperative variables). A variety of machine learning algorithms were employed to develop the model, including logistic regression, random forest, extreme gradient boosting, and gradient boosting decision trees. The performance of different models was compared using the area under the receiver operating characteristic curve (AUROC). Shapley Additive Explanations (SHAP) analysis was employed for model interpretation. RESULTS The patients in the discovery cohort had a median age of 52 years (IQR: 42-61 years), and 1179 patients (1.5%) developed AKI after surgery. The gradient boosting decision trees algorithm showed the best predictive performance using all available variables, or only preoperative variables. The AUROCs were 0.849 (95% CI: 0.835-0.863) and 0.828 (95% CI: 0.813-0.843), respectively. The SHAP analysis showed that age, surgical duration, preoperative serum creatinine, and gamma-glutamyltransferase, as well as American Society of Anesthesiologists physical status III were the most important five features. When gradually reducing the features, the AUROCs decreased from 0.852 (including the top 40 features) to 0.839 (including the top 10 features). In the validation cohort, the authors observed a similar pattern regarding the models' predictive performance. CONCLUSIONS The machine learning models the authors developed had satisfactory predictive performance for identifying high-risk postoperative AKI patients. Furthermore, the authors found that model performance was only slightly affected when only preoperative variables or only the most important predictive features were included.
Collapse
Affiliation(s)
- Rao Sun
- Department of Anesthesiology and Pain Medicine, Hubei Key Laboratory of Geriatric Anesthesia and Perioperative Brain Health, and Wuhan Clinical Research Center for Geriatric Anesthesia
| | - Shiyong Li
- Department of Anesthesiology and Pain Medicine, Hubei Key Laboratory of Geriatric Anesthesia and Perioperative Brain Health, and Wuhan Clinical Research Center for Geriatric Anesthesia
| | - Yuna Wei
- Yidu Cloud Technology Inc, Beijing, People's Republic of China
| | - Liu Hu
- Health Management Center, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei
| | - Qiaoqiao Xu
- Department of Anesthesiology and Pain Medicine, Hubei Key Laboratory of Geriatric Anesthesia and Perioperative Brain Health, and Wuhan Clinical Research Center for Geriatric Anesthesia
| | - Gaofeng Zhan
- Department of Anesthesiology and Pain Medicine, Hubei Key Laboratory of Geriatric Anesthesia and Perioperative Brain Health, and Wuhan Clinical Research Center for Geriatric Anesthesia
| | - Xu Yan
- Department of Anesthesiology and Pain Medicine, Hubei Key Laboratory of Geriatric Anesthesia and Perioperative Brain Health, and Wuhan Clinical Research Center for Geriatric Anesthesia
| | - Yuqin He
- Department of Anesthesiology and Pain Medicine, Hubei Key Laboratory of Geriatric Anesthesia and Perioperative Brain Health, and Wuhan Clinical Research Center for Geriatric Anesthesia
| | - Yao Wang
- Yidu Cloud Technology Inc, Beijing, People's Republic of China
| | - Xinhua Li
- Department of Anesthesiology and Pain Medicine, Hubei Key Laboratory of Geriatric Anesthesia and Perioperative Brain Health, and Wuhan Clinical Research Center for Geriatric Anesthesia
| | - Ailin Luo
- Department of Anesthesiology and Pain Medicine, Hubei Key Laboratory of Geriatric Anesthesia and Perioperative Brain Health, and Wuhan Clinical Research Center for Geriatric Anesthesia
| | - Zhiqiang Zhou
- Department of Anesthesiology and Pain Medicine, Hubei Key Laboratory of Geriatric Anesthesia and Perioperative Brain Health, and Wuhan Clinical Research Center for Geriatric Anesthesia
| |
Collapse
|
2
|
Shan Y, Lin M, Gu F, Ying S, Bao X, Zhu Q, Tao Y, Chen Z, Li D, Zhang W, Fu G, Wang M. Association between fasting stress hyperglycemia ratio and contrast-induced acute kidney injury in coronary angiography patients: a cross-sectional study. Front Endocrinol (Lausanne) 2023; 14:1300373. [PMID: 38155953 PMCID: PMC10753820 DOI: 10.3389/fendo.2023.1300373] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/23/2023] [Accepted: 11/30/2023] [Indexed: 12/30/2023] Open
Abstract
Aims Stress hyperglycemia ratio (SHR), an emerging indicator of critical illness, exhibits a significant association with adverse cardiovascular outcomes. The primary aim of this research endeavor is to evaluate the association between fasting SHR and contrast-induced acute kidney injury (CI-AKI). Methods This cross-sectional study comprised 3,137 patients who underwent coronary angiography (CAG) or percutaneous coronary intervention (PCI). The calculation of fasting SHR involved dividing the admission fasting blood glucose by the estimated mean glucose obtained from glycosylated hemoglobin. CI-AKI was assessed based on elevated serum creatinine (Scr) levels. To investigate the relationship between fasting SHR and the proportion of SCr elevation, piecewise linear regression analysis was conducted. Modified Poisson's regression analysis was implemented to evaluate the correlation between fasting SHR and CI-AKI. Subgroup analysis and sensitivity analysis were conducted to explore result stability. Results Among the total population, 482 (15.4%) patients experienced CI-AKI. Piecewise linear regression analysis revealed significant associations between the proportion of SCr elevation and fasting SHR on both sides (≤ 0.8 and > 0.8) [β = -12.651, 95% CI (-23.281 to -2.022), P = 0.020; β = 8.274, 95% CI (4.176 to 12.372), P < 0.001]. The Modified Poisson's regression analysis demonstrated a statistically significant correlation between both the lowest and highest levels of fasting SHR and an increased incidence of CI-AKI [(SHR < 0.7 vs. 0.7 ≤ SHR < 0.9) β = 1.828, 95% CI (1.345 to 2.486), P < 0.001; (SHR ≥ 1.3 vs. 0.7 ≤ SHR < 0.9) β = 2.896, 95% CI (2.087 to 4.019), P < 0.001], which was further validated through subgroup and sensitivity analyses. Conclusion In populations undergoing CAG or PCI, both lowest and highest levels of fasting SHR were significantly associated with an increased occurrence of CI-AKI.
Collapse
Affiliation(s)
- Yu Shan
- Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
- Key Laboratory of Cardiovascular Intervention and Regenerative Medicine of Zhejiang Province, Hangzhou, China
| | - Maoning Lin
- Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
- Key Laboratory of Cardiovascular Intervention and Regenerative Medicine of Zhejiang Province, Hangzhou, China
| | - Fangfang Gu
- Department of Cardiology, The Affiliated Huzhou Hospital (Huzhou Central Hospital), College of Medicine, Zhejiang University, Huzhou, Zhejiang, China
| | - Shuxin Ying
- Department of Endocrinology and Metabolism, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
| | - Xiaoyi Bao
- Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
- Key Laboratory of Cardiovascular Intervention and Regenerative Medicine of Zhejiang Province, Hangzhou, China
| | - Qiongjun Zhu
- Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
- Key Laboratory of Cardiovascular Intervention and Regenerative Medicine of Zhejiang Province, Hangzhou, China
| | - Yecheng Tao
- Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
- Key Laboratory of Cardiovascular Intervention and Regenerative Medicine of Zhejiang Province, Hangzhou, China
| | - Zhezhe Chen
- Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
- Key Laboratory of Cardiovascular Intervention and Regenerative Medicine of Zhejiang Province, Hangzhou, China
| | - Duanbin Li
- Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
- Key Laboratory of Cardiovascular Intervention and Regenerative Medicine of Zhejiang Province, Hangzhou, China
| | - Wenbin Zhang
- Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
- Key Laboratory of Cardiovascular Intervention and Regenerative Medicine of Zhejiang Province, Hangzhou, China
| | - Guosheng Fu
- Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
- Key Laboratory of Cardiovascular Intervention and Regenerative Medicine of Zhejiang Province, Hangzhou, China
| | - Min Wang
- Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang, China
- Key Laboratory of Cardiovascular Intervention and Regenerative Medicine of Zhejiang Province, Hangzhou, China
| |
Collapse
|
3
|
Wang J, Yen F, Lin K, Shin S, Hsu Y, Hsu C. Epidemiological characteristics of diabetic kidney disease in Taiwan. J Diabetes Investig 2021; 12:2112-2123. [PMID: 34529360 PMCID: PMC8668071 DOI: 10.1111/jdi.13668] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/08/2021] [Revised: 09/02/2021] [Accepted: 09/12/2021] [Indexed: 12/17/2022] Open
Abstract
Diabetic kidney disease (DKD) is a critical microvascular complication of diabetes. With the continuous increase in the prevalence of diabetes since 2000, the prevalence of DKD has also been increasing in past years. The prevalence of DKD among individuals with type 2 diabetes in Taiwan increased from 13.32% in 2000 to 17.92% in 2014. The cumulative incidence of DKD among individuals with type 1 diabetes in Taiwan was higher than 30% during 1999-2012. DKD is the leading cause of end-stage renal disease (ESRD), with a prevalence of approximately 45% in a population on chronic dialysis in Taiwan. Among individuals with type 2 diabetes, the prevalence of ESRD in the receipt of dialysis also increased from 1.32% in 2005 to 1.47% in 2014. Risk factors for DKD development are age, race, family history, hyperglycemia, hypertension, dyslipidemia, dietary patterns, and lifestyles. Prognostic factors that aggravate DKD progression include age, family history, sex, glycemic control, blood pressure (BP), microvascular complications, and atherosclerosis. This review summarizes updated information on the onset and progression of DKD, particularly in the Taiwanese population. Translating these epidemiological features is essential to optimizing the kidney care and improving the prognosis of DKD in Asian populations.
Collapse
Affiliation(s)
- Jun‐Sing Wang
- Division of Endocrinology and MetabolismDepartment of Internal MedicineTaichung Veterans General HospitalTaichungTaiwan
- Faculty of MedicineSchool of MedicineNational Yang‐Ming UniversityTaipeiTaiwan
- Rong Hsing Research Center for Translational MedicineInstitute of Biomedical ScienceNational Chung Hsing UniversityTaichungTaiwan
- PhD Program in Translational MedicineNational Chung Hsing UniversityTaichungTaiwan
| | | | - Kun‐Der Lin
- Department of Internal MedicineKaohsiung Municipal Ta‐Tung HospitalKaohsiung Medical University HospitalKaohsiung Medical UniversityKaohsiungTaiwan
- Division of Endocrinology and MetabolismDepartment of Internal MedicineKaohsiung Medical University Hospital and College of MedicineKaohsiung Medical UniversityKaohsiungTaiwan
| | - Shyi‐Jang Shin
- Division of Endocrinology and MetabolismDepartment of Internal MedicineKaohsiung Medical University Hospital and College of MedicineKaohsiung Medical UniversityKaohsiungTaiwan
- Grander ClinicKaohsiungTaiwan
| | - Yueh‐Han Hsu
- Department of Internal MedicineDitmanson Medical Foundation Chia‐Yi Christian HospitalChia‐Yi CityTaiwan
- Department of NursingMin‐Hwei College of Health Care ManagementTainan CityTaiwan
| | - Chih‐Cheng Hsu
- Institute of Population Health SciencesNational Health Research InstituteZhunan, MiaoliTaiwan
- Department of Health Services AdministrationChina Medical UniversityTaichung CityTaiwan
- Department of Family MedicineMin‐Sheng General HospitalTaoyuanTaiwan
| | | |
Collapse
|
4
|
Yun JS, Park YM, Han K, Kim HW, Cha SA, Ahn YB, Ko SH. Severe hypoglycemia and the risk of end stage renal disease in type 2 diabetes. Sci Rep 2021; 11:4305. [PMID: 33619302 PMCID: PMC7900096 DOI: 10.1038/s41598-021-82838-5] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/21/2020] [Accepted: 01/15/2021] [Indexed: 12/17/2022] Open
Abstract
We investigated the association between the incidence of severe hypoglycemia and the risk of end-stage renal disease (ESRD) in patients with type 2 diabetes. Baseline and follow-up data for 988,333 participants with type 2 diabetes were retrieved from the National Health Insurance System database. The number of severe hypoglycemia episodes experienced from 2007 to 2009 was determined. The primary outcome was the development of ESRD after the baseline evaluation. Participants were followed from the baseline until death or December 31, 2016, during this period 14,545 participants (1.5%) developed ESRD. In the crude model, compared with those who experienced no severe hypoglycemia, the hazard ratios (95% confidential intervals) for developing ESRD were 4.96 (4.57–5.39), 6.84 (5.62–8.32), and 9.51 (7.14–12.66) in participants who experienced one, two, and three or more episodes of severe hypoglycemia, respectively. Further adjustment for various confounding factors attenuated the association between severe hypoglycemia and ESRD; the significance of the association between severe hypoglycemia and ESRD was maintained. Having three or more severe hypoglycemia episodes was associated with a nearly two-fold increased risk of developing ESRD. Prior episodes of severe hypoglycemia were associated with an increased risk of ESRD among Korean adults with type 2 diabetes.
Collapse
Affiliation(s)
- Jae-Seung Yun
- Division of Endocrinology and Metabolism, Department of Internal Medicine, St. Vincent's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
| | - Yong-Moon Park
- Department of Epidemiology, Fay W. Boozman College of Public Health, University of Arkansas for Medical Sciences, Little Rock, AR, USA
| | - Kyungdo Han
- Department of Preventive Medicine, College of Medicine, The Catholic University of Korea, Seoul, Korea
| | - Hyung-Wook Kim
- Division of Nephrology, Department of Internal Medicine, St. Vincent's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
| | - Seon-Ah Cha
- Division of Endocrinology and Metabolism, Department of Internal Medicine, St. Vincent's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
| | - Yu-Bae Ahn
- Division of Endocrinology and Metabolism, Department of Internal Medicine, St. Vincent's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
| | - Seung-Hyun Ko
- Division of Endocrinology and Metabolism, Department of Internal Medicine, St. Vincent's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea.
| |
Collapse
|
5
|
Severe Hypoglycemia as a Predictor of End-Stage Renal Disease in Type 2 Diabetes: A National Cohort Study. INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH 2019; 16:ijerph16050681. [PMID: 30813549 PMCID: PMC6427770 DOI: 10.3390/ijerph16050681] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 01/05/2019] [Revised: 02/21/2019] [Accepted: 02/22/2019] [Indexed: 12/16/2022]
Abstract
Aims: This study investigated whether there is a link between severe hypoglycemia and progression into end-stage renal disease (ESRD) in patients with type 2 diabetes. Methods: Tapping into Taiwan’s Health Insurance Research Database, we identified all type 2 diabetes patients between 1996 and 2013 and identified those diagnosed with a severe hypoglycemia episode during an emergency department visit and those who were not. Controls were then matched 1:1 for age, sex, index year, and medication. Results: We identified 468,421 type 2 diabetes patients diagnosed as having severe hypoglycemia in an emergency department visit. Compared with controls, these patients with SH had a higher risk of all-cause mortality (Hazard Ratio (HR), 1.76; 95% confidence interval, 1.61–1.94) and progressed into ESRD within a shorter period of time. Results were similar after controlling for competing risk. Conclusion: Severe hypoglycemia is significantly associated with worsening renal dysfunction in patients with type 2 diabetes and hastened progression into ESRD.
Collapse
|