Machine Learning-Based Mortality Prediction in Chronic Kidney Disease among Heart Failure Patients: Insights and Outcomes from the Jordanian Heart Failure Registry
Abstract
:1. Introduction
2. Methods
2.1. Study Design and Setting
2.2. Data Collection
2.3. Statistical Analysis
2.4. Machine Learning Analysis
3. Results
3.1. Characteristics of the Included Patients
3.2. The Association between Patients’ Characteristics and Kidney Function
3.3. Machine Learning Model Performance
3.4. Feature Importance Analysis
4. Discussion
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Yoo, B.S. Clinical Significance of B-type Natriuretic Peptide in Heart Failure. J. Lifestyle Med. 2014, 4, 34–38. [Google Scholar] [CrossRef] [PubMed]
- Bhagat, A.A.; Greene, S.J.; Vaduganathan, M.; Fonarow, G.C.; Butler, J. Initiation, Continuation, Switching, and Withdrawal of Heart Failure Medical Therapies During Hospitalization. JACC. Heart Fail. 2019, 7, 1–12. [Google Scholar] [CrossRef] [PubMed]
- Beezer, J.; Al Hatrushi, M.; Husband, A.; Kurdi, A.; Forsyth, P. Polypharmacy definition and prevalence in heart failure: A systematic review. Heart Fail. Rev. 2022, 27, 465–492. [Google Scholar] [CrossRef] [PubMed]
- Vijay, K.; Neuen, B.L.; Lerma, E.V. Heart Failure in Patients with Diabetes and Chronic Kidney Disease: Challenges and Opportunities. Cardiorenal Med. 2022, 12, 1–10. [Google Scholar] [CrossRef] [PubMed]
- Polonsky, T.S.; Bakris, G.L. Heart Failure and Changes in Kidney Function: Focus on Understanding, Not Reacting. Heart Fail. Clin. 2019, 15, 455–461. [Google Scholar] [CrossRef] [PubMed]
- GBD 2017 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990-2017: A systematic analysis for the Global Burden of Disease Study 2017. Lancet 2018, 392, 1789–1858. [Google Scholar]
- Szlagor, M.; Dybiec, J.; Młynarska, E.; Rysz, J.; Franczyk, B. Chronic Kidney Disease as a Comorbidity in Heart Failure. Int. J. Mol. Sci. 2023, 24, 2988. [Google Scholar] [CrossRef] [PubMed]
- Tuegel, C.; Bansal, N. Heart failure in patients with kidney disease. Heart 2017, 103, 1848–1853. [Google Scholar] [CrossRef] [PubMed]
- Matsushita, K.; Ballew, S.H.; Wang, A.Y.; Kalyesubula, R.; Schaeffner, E.; Agarwal, R. Epidemiology and risk of cardiovascular disease in populations with chronic kidney disease. Nat. Rev. Nephrol. 2022, 18, 696–707. [Google Scholar] [CrossRef] [PubMed]
- Grande, D.; Gioia, M.I.; Terlizzese, P.; Iacoviello, M. Heart Failure and Kidney Disease. Adv. Exp. Med. Biol. 2018, 1067, 219–238. [Google Scholar] [CrossRef] [PubMed]
- Ronco, C.; Haapio, M.; House, A.A.; Anavekar, N.; Bellomo, R. Cardiorenal syndrome. J. Am. Coll. Cardiol. 2008, 52, 1527–1539. [Google Scholar] [CrossRef] [PubMed]
- Schefold, J.C.; Filippatos, G.; Hasenfuss, G.; Anker, S.D.; von Haehling, S. Heart failure and kidney dysfunction: Epidemiology, mechanisms and management. Nature reviews. Nephrology 2016, 12, 610–623. [Google Scholar] [CrossRef] [PubMed]
- Ronco, C.; Bellasi, A.; Di Lullo, L. Cardiorenal Syndrome: An Overview. Adv. Chronic Kidney Dis. 2018, 25, 382–390. [Google Scholar] [CrossRef] [PubMed]
- Roehm, B.; McAdams, M.; Hedayati, S.S. Novel Biomarkers of Kidney Disease in Advanced Heart Failure: Beyond GFR and Proteinuria. Curr. Heart Fail. Rep. 2022, 19, 223–235. [Google Scholar] [CrossRef] [PubMed]
- Nowak, C.; Ärnlöv, J. Kidney Disease Biomarkers Improve Heart Failure Risk Prediction in the General Population. Circ. Heart Fail. 2020, 13, e006904. [Google Scholar] [CrossRef] [PubMed]
- Tomasoni, D.; Adamo, M.; Lombardi, C.M.; Metra, M. Highlights in heart failure. ESC Heart Fail. 2019, 6, 1105–1127. [Google Scholar] [CrossRef] [PubMed]
- Levey, A.S.; Titan, S.M.; Powe, N.R.; Coresh, J.; Inker, L.A. Kidney Disease, Race, and GFR Estimation. Clin. J. Am. Soc. Nephrol. CJASN 2020, 15, 1203–1212. [Google Scholar] [CrossRef] [PubMed]
- Fonarow, G.C.; Abraham, W.T.; Albert, N.M.; Stough, W.G.; Gheorghiade, M.; Greenberg, B.H.; O’Connor, C.M.; Nunez, E.; Yancy, C.W.; Young, J.B. A smoker’s paradox in patients hospitalized for heart failure: Findings from OPTIMIZE-HF. Eur. Heart J. 2008, 29, 1983–1991. [Google Scholar] [CrossRef] [PubMed]
- Han, X.; Zhang, S.; Chen, Z.; Adhikari, B.K.; Zhang, Y.; Zhang, J.; Sun, J.; Wang, Y. Cardiac biomarkers of heart failure in chronic kidney disease. Clin. Chim. Acta; Int. J. Clin. Chem. 2020, 510, 298–310. [Google Scholar] [CrossRef] [PubMed]
Variable, n (%) | Normal GFR mL/min/1.73 m2 (≥60) (n = 373 **) | Low GFR mL/min/1.73 m2 (<60) (n = 277 **) | p-Value | |
---|---|---|---|---|
Gender | Male | 251 (67.7) | 152 (55.7) | 0.002 * |
Female | 120 (32.3) | 121 (44.3) | ||
Age (years) | <40 | 22 (6.3) | 3 (1.2) | <0.001 * |
40–49 | 30 (8.5) | 2 (0.8) | ||
50–59 | 76 (21.7) | 28 (11.2) | ||
60–69 | 106 (30.2) | 72 (28.9) | ||
≥70 | 117 (33.3) | 144 (57.8) | ||
Hypertension | Yes | 263 (77.1) | 234 (89.3) | <0.001 * |
No | 78 (22.9) | 28 (10.7) | ||
Diabetes | No | 118 (34.6) | 61 (23.3) | 0.003 * |
Yes | 223 (65.4) | 201 (76.7) | ||
Smoking | No | 199 (58.4) | 211 (80.5) | <0.001 * |
Yes | 142 (41.6) | 51 (19.5) | ||
Alcohol | No | 334 (97.9) | 259 (98.9) | 0.387 |
Yes | 7 (2.1) | 3 (1.1) | ||
Type of Heart Failure | Chronic | 276 (75.2) | 201 (73.1) | 0.544 |
Acute | 91 (24.8) | 74 (26.9) | ||
Dyslipidemia | No | 121 (35.5) | 97 (37.0) | 0.697 |
Yes | 220 (34.5) | 165 (63.0) | ||
Obesity | No | 303 (88.9) | 235 (89.7) | 0.742 |
Yes | 38 (11.1) | 27 (10.3) | ||
Family History of Premature Death | No | 306 (89.7) | 220 (84.0) | 0.035 * |
Yes | 35 (10.3) | 42 (16.0) | ||
Family History of ASCVD | No | 326 (95.6) | 156 (59.5) | <0.001 * |
Yes | 15 (4.4) | 106 (40.5) | ||
History of ASCVD | No | 47 (16.3) | 40 (17.2) | 0.783 |
Yes | 242 (83.7) | 193 (82.8) | ||
History of Arrhythmias | No | 261 (76.1) | 157 (61.4) | <0.001 * |
Yes | 109 (23.9) | 118 (38.6) | ||
History of Implanted Device | No | 283 (97.9) | 220 (94.4) | 0.034 * |
Yes | 6 (2.1) | 13 (5.6) | ||
History of Structural Heart Disease | No | 271 (93.8) | 213 (91.4) | 0.303 |
Yes | 18 (6.2) | 20 (8.6) | ||
History of HF | No | 106 (28.6) | 65 (23.7) | 0.162 |
Yes | 264 (71.4) | 209 (76.3) | ||
Admissions in Past 6 Months | 0 | 206 (57.4) | 141 (51.6) | 0.474 |
1 | 67 (18.7) | 53 (19.4) | ||
2 | 22 (6.1) | 15 (5.5) | ||
>2 | 64 (17.8) | 64 (23.4) | ||
Mechanical Ventilation | No | 255 (97.3) | 207 (93.7) | 0.049 * |
Yes | 7 (2.7) | 14 (6.3) | ||
Death | No | 351 (94.1) | 224 (80.9) | <0.001 * |
Yes | 22 (5.9) | 53 (19.1) |
Variable, n (%) | Normal BUN mg/dL (n = 437) | High BUN mg/dL (>20) (n = 1362) | p-Value | |
---|---|---|---|---|
Gender | Male | 174 (39.8) | 594 (43.6) | 0.163 |
Female | 263 (60.2) | 768 (56.4) | ||
Age (years) | <40 | 27 (6.3) | 38 (3.0) | <0.001 * |
40–49 | 45 (10.5) | 72 (5.8) | ||
50–59 | 83 (19.3) | 178 (14.3) | ||
60–69 | 116 (27.0) | 326 (26.1) | ||
≥70 | 158(36.8) | 633 (50.8) | ||
Hypertension | No | 86 (20.6) | 244 (19.0) | 0.459 |
Yes | 331 (79.4) | 1042 (81.0) | ||
Diabetes | No | 135 (32.4) | 379 (29.4) | 0.243 |
Yes | 282 (67.6) | 912 (70.6) | ||
Smoking | No | 248 (59.5) | 921 (71.6) | <0.001 * |
Yes | 169 (40.5) | 365 (28.4) | ||
Alcohol | No | 412 (98.8) | 1281 (99.6) | 0.060 |
Yes | 5 (1.2) | 5 (0.4) | ||
Type of HF | Chronic | 330 (75.3) | 896 (66.4) | <0.001 * |
Acute | 108 (24.7) | 454 (33.6) | ||
Dyslipidemia | No | 119 (28.5) | 627 (48.8) | <0.001 * |
Yes | 298 (71.5) | 659 (51.2) | ||
Obesity | No | 378 (90.6) | 1192 (92.7) | 0.177 |
Yes | 39 (9.4) | 94 (7.3) | ||
Family History of Premature Death | No | 365 (87.5) | 1259 (97.9) | <0.001 * |
Yes | 52 (12.5) | 27 (2.1) | ||
Family History of ASCVD | No | 373 (89.4) | 903 (70.5) | <0.001 * |
Yes | 44 (10.6) | 380 (29.0) | ||
History of ASCVD | No | 47 (13.5) | 218 (22.2) | <0.001 * |
Yes | 302 (86.5) | 766 (77.8) | ||
History of Arrhythmias | No | 267 (65.3) | 798 (67.4) | 0.329 |
Yes | 170 (34.7) | 567 (32.6) | ||
History of Implanted Device | No | 334 (95.7) | 946 (96.1) | 0.720 |
Yes | 15 (4.3) | 38 (3.9) | ||
History of Structural Heart Disease | No | 329 (94.3) | 920 (39.5) | 0.609 |
Yes | 20 (5.7) | 64 (6.5) | ||
History of HF | No | 69 (15.8) | 265 (19.3) | 0.098 |
Yes | 368 (84.2) | 1107 (80.7) | ||
Admissions in Past 6 Months | 0 | 166 (39.1) | 545 (41.1) | 0.031 * |
1 | 90 (21.2) | 241 (18.2) | ||
2 | 43 (10.1) | 93 (7.0) | ||
>2 | 126 (29.6) | 446 (33.6) | ||
Mechanical Ventilation | No | 227 (93.8) | 912 (95.3) | 0.340 |
Yes | 15 (6.2) | 45 (4.7) | ||
Death | No | 392 (89.3) | 1243 (90.0) | 0.667 |
Yes | 47 (10.7) | 138 (10.0) |
Variable, n (%) | Normal Cr µmol/L (n = 1113) | High Cr µmol/L (>115) (n = 814) | p-Value | |
---|---|---|---|---|
Gender | Male | 463 (41.6) | 345 (42.4) | 0.730 |
Female | 650 (58.4) | 469 (57.6) | ||
Age | <40 | 59 (5.8) | 11 (1.4) | <0.001 * |
40–49 | 95 (9.3) | 32 (4.2) | ||
50–59 | 197 (19.3) | 87 (11.3) | ||
60–69 | 271 (36.5) | 194 (25.3) | ||
≥70 | 401 (39.2) | 443 (57.8) | ||
Hypertension | No | 42 (23.7) | 111 (13.9) | <0.001 * |
Yes | 779 (76.3) | 688 (86.1) | ||
Diabetes | No | 380 (37.1) | 175 (21.9) | <0.001 * |
Yes | 645 (62.9) | 625 (78.1) | ||
Smoking | No | 645 (63.2) | 611 (76.5) | <0.001 * |
Yes | 376 (36.8) | 188 (23.5) | ||
Alcohol | No | 1012 (99.1) | 797 (99.7) | 0.060 |
Yes | 9 (0.9) | 2 (0.3) | ||
Type of HF | Chronic | 792 (72.1) | 538 (65.9) | 0.004 * |
Acute | 307 (27.9) | 278 (34.1) | ||
Dyslipidemia | No | 417 (40.8) | 371 (46.4) | 0.017 * |
Yes | 604 (59.2) | 428 (53.6) | ||
Obesity | No | 935 (91.6) | 738 (92.4) | 0.540 |
Yes | 86 (8.4) | 61 (7.6) | ||
Family History of Premature Death | No | 976 (95.6) | 753 (94.2) | 0.190 |
Yes | 45 (4.4) | 46 (5.8) | ||
Family History of ASCVD | No | 979 (95.9) | 401 (50.2) | <0.001 * |
Yes | 42 (4.1) | 398 (49.8) | ||
History of ASCVD | No | 164 (20.1) | 118 (18.8) | 0.519 |
Yes | 650 (79.9) | 510 (81.2) | ||
History of Arrhythmias | No | 695 (62.7) | 465 (56.6) | 0.007 * |
Yes | 414 (37.3) | 357 (43.3) | ||
History of Implanted Device | No | 781 (95.9) | 605 (96.3) | 0.703 |
Yes | 33 (4.1) | 23 (3.7) | ||
History of Structural Heart Disease | No | 761 (93.5) | 587 (93.5) | 0.989 |
Yes | 53 (6.5) | 41 (6.5) | ||
History of HF | No | 227 (20.5) | 131 (15.9) | 0.011 * |
Yes | 882 (79.5) | 692 (84.1) | ||
Admissions in Past 6 Months | 0 | 457 (42.8) | 324 (40.0) | 0.368 |
1 | 195 (18.2) | 159 (19.7) | ||
2 | 84 (7.9) | 62 (7.7) | ||
>2 | 333 (31.1) | 264 (32.6) | ||
Mechanical Ventilation | No | 648 (96.6) | 559 (93.3) | 0.008 * |
Yes | 23 (3.4) | 40 (6.7) | ||
Death | No | 1064 (95.0) | 688 (83.0) | <0.001 * |
Yes | 56 (5.0) | 141 (17.0) |
Model | Accuracy | AUC | Sensitivity | Specificity |
---|---|---|---|---|
Random Forest Classifier | 90.02% | 80.51% | 32.56% | 96.39% |
Logistic Regression | 73.78% | 79.15% | 72.09% | 73.97% |
Support Vector Machine | 80.74% | 73.65% | 46.51% | 84.54% |
eXtreme Gradient Boosting | 90.02% | 78.21% | 39.53% | 95.62% |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Izraiq, M.; Alawaisheh, R.; Ibdah, R.; Dabbas, A.; Ahmed, Y.B.; Mughrabi Sabbagh, A.-L.; Zuraik, A.; Ababneh, M.; Toubasi, A.A.; Al-Bkoor, B.; et al. Machine Learning-Based Mortality Prediction in Chronic Kidney Disease among Heart Failure Patients: Insights and Outcomes from the Jordanian Heart Failure Registry. Medicina 2024, 60, 831. https://doi.org/10.3390/medicina60050831
Izraiq M, Alawaisheh R, Ibdah R, Dabbas A, Ahmed YB, Mughrabi Sabbagh A-L, Zuraik A, Ababneh M, Toubasi AA, Al-Bkoor B, et al. Machine Learning-Based Mortality Prediction in Chronic Kidney Disease among Heart Failure Patients: Insights and Outcomes from the Jordanian Heart Failure Registry. Medicina. 2024; 60(5):831. https://doi.org/10.3390/medicina60050831
Chicago/Turabian StyleIzraiq, Mahmoud, Raed Alawaisheh, Rasheed Ibdah, Aya Dabbas, Yaman B. Ahmed, Abdel-Latif Mughrabi Sabbagh, Ahmad Zuraik, Muhannad Ababneh, Ahmad A. Toubasi, Basel Al-Bkoor, and et al. 2024. "Machine Learning-Based Mortality Prediction in Chronic Kidney Disease among Heart Failure Patients: Insights and Outcomes from the Jordanian Heart Failure Registry" Medicina 60, no. 5: 831. https://doi.org/10.3390/medicina60050831