1
|
Wang Y, Sun X, Lu J, Zhong L, Yang Z. Construction and evaluation of a mortality prediction model for patients with acute kidney injury undergoing continuous renal replacement therapy based on machine learning algorithms. Ann Med 2024; 56:2388709. [PMID: 39155811 DOI: 10.1080/07853890.2024.2388709] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/30/2024] [Revised: 04/09/2024] [Accepted: 06/24/2024] [Indexed: 08/20/2024] Open
Abstract
BACKGROUND To construct and evaluate a predictive model for in-hospital mortality among critically ill patients with acute kidney injury (AKI) undergoing continuous renal replacement therapy (CRRT), based on nine machine learning (ML) algorithm. METHODS The study retrospectively included patients with AKI who underwent CRRT during their initial hospitalization in the United States using the medical information mart for intensive care (MIMIC) database IV (version 2.0), as well as in the intensive care unit (ICU) of Huzhou Central Hospital. Patients from the MIMIC database were used as the training cohort to construct the models (from 2008 to 2019, n = 1068). Patients from Huzhou Central Hospital were utilized as the external validation cohort to evaluate the models (from June 2019 to December 2022, n = 327). In the training cohort, least absolute shrinkage and selection operator (LASSO) regression with cross-validation was employed to select features for constructing the model and subsequently established nine ML predictive models. The performance of these nine models on the external validation cohort dataset was comprehensively evaluated based on the area under the receiver operating characteristic curve (AUROC) and the optimal model was selected. A static nomogram and a web-based dynamic nomogram were presented, with a comprehensive evaluation from the perspectives of discrimination (AUROC), calibration (calibration curve) and clinical practicability (DCA curves). RESULTS Finally, 1395 eligible patients were enrolled, including 1068 patients in the training cohort and 327 patients in the external validation cohort. In the training cohort, LASSO regression with cross-validation was employed to select features and nine models were individually constructed. Compared to the other eight models, the Lasso regularized logistic regression (Lasso-LR) model exhibited the highest AUROC (0.756) and the optimal calibration curve. The DCA curve suggested a certain clinical utility in predicting in-hospital mortality among critically ill patients with AKI undergoing CRRT. Consequently, the Lasso-LR model was the optimal model and it was visualized as a common nomogram (static nomogram) and a web-based dynamic nomogram (https://chsyh2006.shinyapps.io/dynnomapp/). Discrimination, calibration and DCA curves were employed to assess the performance of the nomogram. The AUROC for the training and external validation cohorts in the nomogram model was 0.771 (95%CI: 0.743, 0.799) and 0.756 (95%CI: 0.702, 0.809), respectively. The calibration slope and Brier score for the training cohort were 1.000 and 0.195, while for the external validation cohort, they were 0.849 and 0.197, respectively. The DCA indicated that the model had a certain clinical application value. CONCLUSIONS Our study selected the optimal model and visualized it as a static and dynamic nomogram integrating clinical predictors, so that clinicians can personalized predict the in-hospital outcome of critically ill patients with AKI undergoing CRRT upon ICU admission.
Collapse
Affiliation(s)
- Yongbin Wang
- Department of Intensive Care Unit, Huzhou Central Hospital, Fifth School of Clinical Medicine of Zhejiang Chinese Medical University, Huzhou, China
- Huzhou Central Hospital, Affiliated Central Hospital of Huzhou University, Huzhou, China
| | - Xu Sun
- Huzhou Central Hospital, Affiliated Central Hospital of Huzhou University, Huzhou, China
- Department of General Surgery, Huzhou Central Hospital, Fifth School of Clinical Medicine of Zhejiang Chinese Medical University, Huzhou, China
| | - Jianhong Lu
- Department of Intensive Care Unit, Huzhou Central Hospital, Fifth School of Clinical Medicine of Zhejiang Chinese Medical University, Huzhou, China
- Huzhou Central Hospital, Affiliated Central Hospital of Huzhou University, Huzhou, China
| | - Lei Zhong
- Department of Intensive Care Unit, Huzhou Central Hospital, Fifth School of Clinical Medicine of Zhejiang Chinese Medical University, Huzhou, China
- Huzhou Central Hospital, Affiliated Central Hospital of Huzhou University, Huzhou, China
| | - Zhenzhen Yang
- Huzhou Central Hospital, Affiliated Central Hospital of Huzhou University, Huzhou, China
- Department of Nephrology, Huzhou Central Hospital, Fifth School of Clinical Medicine of Zhejiang Chinese Medical University, Huzhou, China
| |
Collapse
|
2
|
Thongprayoon C, Radhakrishnan Y, Cheungpasitporn W, Petnak T, Zabala Genovez J, Chewcharat A, Qureshi F, Mao MA, Kashani KB. Association of hypochloremia with mortality among patients requiring continuous renal replacement therapy. J Nephrol 2023; 36:161-170. [PMID: 35347649 DOI: 10.1007/s40620-022-01305-0] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/19/2021] [Accepted: 03/05/2022] [Indexed: 02/07/2023]
Abstract
BACKGROUND Serum chloride derangement is common in critically ill patients requiring continuous renal replacement therapy (CRRT). We aimed to assess the association between serum chloride levels before and during CRRT with mortality. METHODS This is a retrospective cohort study of critically ill patients receiving CRRT for acute kidney injury from December 2006 through November 2015 in a tertiary referral hospital in the United States. We used logistic regression to assess serum chloride before and mean serum chloride during CRRT as predictors for 90 days mortality after CRRT initiation. The normal reference range for serum chloride was 99-108 mmol/L. RESULTS Of 1282 eligible patients, 25%, 50%, and 25% had hypochloremia, normochloremia, and hyperchloremia, respectively. The adjusted odds ratio for 90 days mortality in patients with hypochloremia before CRRT was 1.82 (95% CI 1.29-2.55). During CRRT, 4%, 70%, 26% of patients had mean serum chloride in the hypochloremia, normochloremia, and hyperchloremia range, respectively. The adjusted odds ratio for 90 days mortality in patients with mean serum chloride during CRRT in the hypochloremia range was 2.96 (95% CI 1.43-6.12). Hyperchloremia before and during CRRT was not associated with mortality. The greater serum chloride range during CRRT was associated with increased mortality (OR 1.29; 95% CI 1.13-1.47 per 5 mmol/L increase). CONCLUSION Hypochloremia before and during CRRT is associated with higher mortality.
Collapse
Affiliation(s)
- Charat Thongprayoon
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Yeshwanter Radhakrishnan
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Wisit Cheungpasitporn
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Tananchai Petnak
- Division of Pulmonary and Critical Care Medicine, Department of Medicine, Mayo Clinic, Rochester, MN, USA.,Division of Pulmonary and Pulmonary Critical Care Medicine, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand
| | - Jose Zabala Genovez
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Api Chewcharat
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Fawad Qureshi
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Michael A Mao
- Division of Nephrology and Hypertension, Mayo Clinic, Jacksonville, FL, USA
| | - Kianoush B Kashani
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA. .,Division of Pulmonary and Critical Care Medicine, Department of Medicine, Mayo Clinic, Rochester, MN, USA.
| |
Collapse
|
3
|
Thongprayoon C, Cheungpasitporn W, Radhakrishnan Y, Zabala Genovez JL, Petnak T, Shawwa K, Qureshi F, Mao MA, Kashani KB. Association of Serum Potassium Derangements with Mortality among Patients Requiring Continuous Renal Replacement Therapy. Ther Apher Dial 2022; 26:1098-1105. [PMID: 35067000 DOI: 10.1111/1744-9987.13804] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/28/2021] [Revised: 01/05/2022] [Accepted: 01/19/2022] [Indexed: 12/01/2022]
Abstract
BACKGROUND We aimed to assess the association between serum potassium and mortality in patients receiving continuous renal replacement therapy (CRRT). METHODS We studied 1,279 acute kidney injury patients receiving CRRT in a tertiary referral hospital in the United States. We used logistic regression to assess the association of serum potassium before CRRT and mean serum potassium during CRRT with 90-day mortality after CRRT initiation, using serum potassium 4.0-4.4 mmol/L as reference group. RESULTS Before CRRT, there was a U-shaped association between serum potassium and 90-day mortality. There was a significant increase in mortality when serum potassium before CRRT was ≤3.4 and ≥4.5 mmol/L. During CRRT, progressively increased mortality was noted when mean serum potassium was ≥4.5 mmol/L. The odds ratio of 90-day mortality was significantly higher when mean serum potassium was ≥4.5 mmol/L. CONCLUSION Hypokalemia and hyperkalemia before CRRT and hyperkalemia during CRRT predicts 90-day mortality. This article is protected by copyright. All rights reserved.
Collapse
Affiliation(s)
- Charat Thongprayoon
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN
| | - Wisit Cheungpasitporn
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN
| | | | - Jose L Zabala Genovez
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN
| | - Tananchai Petnak
- Division of Pulmonary and Critical Care Medicine, Department of Medicine, Mayo Clinic, Rochester, MN.,Division of Pulmonary and Pulmonary Critical Care Medicine, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand
| | - Khaled Shawwa
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN
| | - Fawad Qureshi
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN
| | - Michael A Mao
- Division of Nephrology and Hypertension, Mayo Clinic, Jacksonville, FL, USA
| | - Kianoush B Kashani
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN.,Division of Pulmonary and Critical Care Medicine, Department of Medicine, Mayo Clinic, Rochester, MN
| |
Collapse
|
4
|
Thongprayoon C, Radhakrishnan Y, Cheungpasitporn W, Petnak T, Qureshi F, Mao MA, Kashani KB. Association of Serum Phosphate Derangement With Mortality in Patients on Continuous Renal Replacement Therapy. Can J Kidney Health Dis 2022; 9:20543581221114697. [PMID: 35923184 PMCID: PMC9340369 DOI: 10.1177/20543581221114697] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/23/2022] [Indexed: 11/17/2022] Open
Abstract
Background: There is limited evidence on the association of serum phosphate with
mortality in patients receiving continuous renal replacement therapy
(CRRT). Objective: To assess the association of serum phosphate with mortality in critically ill
patients requiring CRRT for acute kidney injury (AKI). Design: A cohort study. Setting: A tertiary referral hospital in the United States. Patients: Acute kidney injury patients receiving CRRT from 2006 through 2015 in
intensive care units. Measurements: (1) Serum phosphate before CRRT and (2) mean serum phosphate during CRRT were
categorized into 3 groups; ≤2.4 (hypophosphatemia), 2.5 to 4.5 (normal serum
phosphate group), and ≥4.6 (hyperphosphatemia) mg/dL. Methods: Multivariable logistic regression was used to assess the association between
serum phosphate and 90-day mortality. Results: A total of 1108 patients were included in this study. Of these, 55% died
within 90 days after CRRT initiation. Before CRRT, 3%, 30%, and 66% had
hypophosphatemia, normophosphatemia, and hyperphosphatemia, respectively.
Before CRRT, both hypophosphatemia and hyperphosphatemia were significantly
associated with higher 90-day mortality with the adjusted odds ratio (OR) of
2.22 (95% confidence interval [CI]: [1.03, 4.78]) and 1.62 (95% CI: [1.21,
2.18]), respectively. During CRRT, 3%, 85%, and 12% had mean serum phosphate
in hypophosphatemia, normophosphatemia, and hyperphosphatemia range. During
CRRT, hyperphosphatemia was significantly associated with higher 90-day
mortality with adjusted OR of 2.22 (95% CI: [1.45, 3.38]). Limitations: Single center, observational design, lack of information regarding causes of
serum phosphate derangement. Conclusion: Most CRRT patients had hyperphosphatemia before CRRT initiation but maintain
normal serum phosphate during CRRT. Before CRRT, hypo- and
hyperphosphatemia, and during CRRT, hyperphosphatemia predicted higher
mortality. Trial registration: Not registered.
Collapse
Affiliation(s)
- Charat Thongprayoon
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Yeshwanter Radhakrishnan
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Wisit Cheungpasitporn
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Tananchai Petnak
- Division of Pulmonary and Critical Care Medicine, Department of Medicine, Mayo Clinic, Rochester, MN, USA
- Division of Pulmonary and Pulmonary Critical Care Medicine, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand
| | - Fawad Qureshi
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| | - Michael A. Mao
- Division of Nephrology and Hypertension, Mayo Clinic, Jacksonville, FL, USA
| | - Kianoush B. Kashani
- Division of Nephrology and Hypertension, Department of Medicine, Mayo Clinic, Rochester, MN, USA
- Division of Pulmonary and Critical Care Medicine, Department of Medicine, Mayo Clinic, Rochester, MN, USA
| |
Collapse
|