Sepsis is a common cause of mortality in hospitalized patients in the USA and worldwide [1]. Among patients with end-stage kidney disease (ESKD), sepsis is the second most common known cause of mortality, trailing only mortality from cardiac causes [2]. Sepsis is presumed to arise from the dysregulation of the immune response to infection [3]. Although standard treatment relies on maintaining tissue perfusion and initiation of empiric antibiotics, statins have been implicated as having a potential mortality benefit in sepsis [4].
Widely known for their lipid-lowering properties, statins are also thought to have inherent anti-inflammatory properties [5]. Although chronic kidney disease is a risk-enhancing factor for atherosclerotic cardiovascular disease, the most recent guidelines provided by the American Heart Association do not recommend the initiation of statins for adults requiring dialysis treatment based on lack of cardiovascular benefit [6]. Supporting this guideline, rosuvastatin was shown not to have a significant benefit in composite cardiovascular death, nonfatal myocardial infarction and stroke in patients receiving hemodialysis [7]. However, to date, studies have not investigated whether statins affect sepsis mortality in the ESKD population. Thus, we sought to determine whether chronic statin use was associated with mortality benefits among ESKD patients hospitalized for sepsis within the Kaiser Permanente Southern California (KPSC) healthcare system.
A retrospective cohort study was performed among patients admitted to the hospital with a diagnosis of both sepsis and ESKD [obtained through the International Classification of Diseases, 9th revision, Clinical Modification (ICD-9-CM) and ICD-10-CM] between 1 January 2008 and 30 September 2018. The KPSC ESKD population is racially/ethnically diverse and generally reflective of the KPSC membership population [8, 9]. Statin users were defined as those having filled two prescriptions for a statin within 6 months prior to admission, with the second prescription filled within 30 days prior to admission based on the KPSC pharmacy and analytic database. The primary outcome of 30-day all-cause mortality was analyzed unweighted and weighted using the Inverse Probability Treatment Weighting (IPTW) method.
A total of 8858 patients were included in our study, with 2820 on statins and 6038 not on statins at the time of admission for sepsis. At baseline, there were differences between statin users and non-users in age, sex, race and comorbidities such as heart disease, hypertension, cerebrovascular disease and diabetes (Table 1). Thus, we used the IPTW method to create synthetic groups to more adequately compare the effect of statin exposure. After IPTW, statin users and non-users were similar across many of the covariates analyzed such as sex, race, heart failure, ischemic heart disease, cerebrovascular disease and renin angiotensin system blocker use. However, significant differences between statin users and non-users remained for hypertension, diabetes, hyperlipidemia and use of beta-blockers (Table 1).
Table 1.
Patient characteristics before and after IPTW
| Unweighted |
Weighted |
|||||||
|---|---|---|---|---|---|---|---|---|
| Total | Statin user |
P-value | Total | Statin user |
P-value | |||
| Yes | No | Yes | No | |||||
| N | 8858 | 2820 | 6038 | 8858 | 2820 | 6038 | ||
| Age, years | ||||||||
| <45 | 703 (7.9) | 101 (3.6) | 602 (10.0) | <0.0001 | 671 (7.6) | 193 (6.8) | 478 (7.9) | 0.0543 |
| 45–64 | 2999 (33.9) | 774 (27.4) | 2225 (36.8) | 2951 (33.3) | 914 (32.4) | 2038 (33.7) | ||
| ≥65 | 5156 (58.2) | 1945 (69.0) | 3211 (53.2) | 5236 (59.1) | 1713 (60.8) | 3523 (58.3) | ||
| Sex | 0.0249 | 0.4299 | ||||||
| Female | 3765 (42.5) | 1150 (40.8) | 2615 (43.3) | 3799 (42.9) | 1227 (43.5) | 2572 (42.6) | ||
| Male | 5093 (57.5) | 1670 (59.2) | 3423 (56.7) | 5059 (57.1) | 1593 (56.5) | 3466 (57.4) | ||
| Race | <0.0001 | 0.9226 | ||||||
| White | 2762 (31.2) | 1011 (35.9) | 1751 (29.0) | 2759 (31.2) | 874 (31.0) | 1885 (31.2) | ||
| Black | 1885 (21.3) | 520 (18.4) | 1365 (22.6) | 1883 (21.3) | 600 (21.3) | 1283 (21.2) | ||
| Hispanic | 3070 (34.7) | 858 (30.4) | 2212 (36.6) | 3054 (34.5) | 963 (34.2) | 2091 (34.6) | ||
| Asian | 1048 (11.8) | 408 (14.5) | 640 (10.6) | 1070 (12.1) | 354 (12.5) | 716 (11.9) | ||
| Othera | 93 (1.1) | 23 (0.8) | 70 (1.2) | 92 (1.0) | 29 (1.0) | 63 (1.1) | ||
| Comorbidity | ||||||||
| Liver disease | 1406 (15.9) | 321 (11.4) | 1085 (18.0) | <0.0001 | 1353 (15.3) | 399 (14.2) | 954 (15.8) | 0.0439 |
| Heart failure | 4771 (53.9) | 1649 (58.5) | 3121 (51.7) | <0.0001 | 4837 (54.6) | 1580 (56) | 3256 (53.9) | 0.0636 |
| Chronic pulmonary disease | 2051 (23.2) | 678 (24.0) | 1373 (22.7) | 0.1756 | 2068 (23.3) | 666 (23.6) | 1403 (23.2) | 0.7016 |
| Ischemic heart disease | 4709 (53.2) | 1731 (61.4) | 2978 (49.3) | <0.0001 | 4781 (54) | 1562 (55.4) | 3218 (53.3) | 0.0638 |
| Rheumatological disease | 148 (1.7) | 47 (1.7) | 101 (1.7) | 0.9834 | 147 (1.7) | 45 (1.6) | 102 (1.7) | 0.8007 |
| Hypertension | 8460 (95.5) | 2788 (98.9) | 5672 (93.9) | <0.0001 | 8510 (96.1) | 2742 (97.2) | 5768 (95.5) | 0.0001 |
| Cerebrovascular disease | 1169 (13.2) | 404 (14.3) | 765 (12.7) | 0.0319 | 1190 (13.4) | 395 (14) | 795 (13.2) | 0.2798 |
| Diabetes mellitus | 6745 (76.2) | 2328 (82.6) | 4417 (73.2) | <0.0001 | 6868 (77.5) | 2259 (80.1) | 4609 (76.3) | 0.0001 |
| Dementia | 833 (9.4) | 268 (9.5) | 565 (9.4) | 0.8262 | 847 (9.6) | 278 (9.9) | 569 (9.4) | 0.5062 |
| Malnutrition | 4458 (50.3) | 1382 (49.0) | 3076 (50.9) | 0.0894 | 4475 (50.5) | 1438 (51) | 3037 (50.3) | 0.5458 |
| Peptic ulcer disease | 529 (6.0) | 173 (6.1) | 356 (5.9) | 0.6587 | 522 (5.9) | 164 (5.8) | 358 (5.9) | 0.8545 |
| Any malignancy | 4585 (51.8) | 1616 (57.3) | 2969 (49.2) | <0.0001 | 4569 (51.6) | 1447 (51.3) | 3121 (51.7) | 0.740 |
| Prior transplantation | 1508 (17.0) | 522 (18.5) | 986 (16.3) | 0.011 | 1490 (16.8) | 470 (16.7) | 1020 (16.9) | 0.795 |
| History of GI hemorrhage | 2224 (25.1) | 685 (24.3) | 1539 (25.5) | 0.2259 | 2214 (25) | 697 (24.7) | 1517 (25.1) | 0.688 |
| Alcohol/drug use | 987 (11.1) | 221 (7.8) | 766 (12.7) | <0.0001 | 944 (10.7) | 276 (9.8) | 668 (11.1) | 0.0647 |
| AIDS/HIV | 54 (0.6) | 16 (0.6) | 38 (0.6) | 0.727 | 52 (0.6) | 16 (0.6) | 36 (0.6) | 0.7771 |
| Hyperlipidemia | 7602 (85.8) | 2734 (97.0) | 4868 (80.6) | <0.0001 | 7688 (86.8) | 2503 (88.8) | 5184 (85.9) | 0.0002 |
| Cardiovascular medications | ||||||||
| Digoxin | 392 (4.4) | 147 (5.2) | 245 (4.1) | 0.0138 | 398 (4.5) | 127 (4.5) | 271 (4.5) | 0.9815 |
| Beta-blockers | 6032 (68.1) | 2176 (77.2) | 3856 (63.9) | <0.0001 | 6121 (69.1) | 2000 (70.9) | 4121 (68.2) | 0.0107 |
| ACE inhibitors | 3090 (34.9) | 1110 (39.4) | 1980 (32.8) | <0.0001 | 3122 (35.3) | 1009 (35.8) | 2114 (35) | 0.477 |
| Angiotensin receptor blockers | 1937 (21.9) | 716 (25.4) | 1221 (20.2) | <0.0001 | 1979 (22.3) | 653 (23.1) | 1326 (22) | 0.2123 |
| Immunosuppressants | ||||||||
| Biologics | 1106 (12.5) | 417 (14.8) | 689 (11.4) | <0.0001 | 1105 (12.5) | 353 (12.5) | 753 (12.5) | 0.9664 |
| Oral corticosteroids | 2499 (28.2) | 894 (31.7) | 1605 (26.6) | <0.0001 | 2521 (28.5) | 811 (28.8) | 1710 (28.3) | 0.6779 |
| DMARDs | 251 (2.8) | 83 (2.9) | 168 (2.8) | 0.6708 | 253 (2.9) | 79 (2.8) | 173 (2.9) | 0.8696 |
Including multiple, Native American, Alaskan, other or unknown race. P-value was generated by Chi-square test for categorical variables and Kruskal–Wallis test.
Data are presented as n (%). GI, gastrointestinal; ACE, angiotensin-converting enzyme; DMARDs, disease-modifying anti-rheumatic drugs. Bolded values were considered significant as P-value <0.05.
Analysis of the IPTW 30-day mortality hazard ratio for ESKD patients hospitalized for sepsis demonstrated a mortality hazard ratio (HR) of 0.82 [95% confidence interval (CI) 0.74–0.91] for statin users compared with non-users. A similar effect was observed using an unweighted comparison, with a HR of 0.85 (95% CI 0.77–0.91). Furthermore, the effect persisted when comparing mortality between statin users and non-users at 90 days with a HR of 0.82 (95% CI 0.76–0.90) and 0.86 (95% CI 0.76–0.90) for weighted and unweighted comparisons, respectively.
Our findings from a real-world population suggest that statin use is associated with lower mortality among ESKD patients hospitalized for sepsis. A strength of our study includes the large, diverse patient population obtained through the comprehensive electronic health records. Potential limitations of our study include the fact that we were not able to fully propensity match certain covariates such as hypertension, diabetes and hyperlipidemia between statin users and non-users, which may confound the interpretation of our findings. In addition, we did not have information on type of dialysis access (fistulas, grafts and catheters) for the entire study population. Nonetheless, our study sheds light on the potential benefits of statins in the ESKD population, a population that is highly prone to infections and infection-related mortality. Although previous studies have not demonstrated a cardiovascular mortality benefit from statin use in ESKD patients, statins have been shown to provide a mortality benefit in a subset of the ESKD population, such as those with acute myocardial infarction [10]. While additional studies are warranted, our findings suggest a potential benefit of statins for ESKD patients hospitalized for sepsis, a population that is highly susceptible to infections.
FUNDING
This work was supported and funded by the Kaiser Permanente Regional Research Committee and the KPSC Clinician Investigator Award (to J.J.S.).
CONFLICT OF INTEREST STATEMENT
None of the authors has any conflicts of interest relevant to this manuscript. The results presented in this paper have not been published previously in whole or part, except in abstract format.
ACKNOWLEDGEMENTS
The authors would like to thank the KPSC Renal Business Group for their support and providing information on the KPSC ESKD population.
REFERENCES
- 1.Cohen J, Vincent JL, Adhikari NK et al. Sepsis: a roadmap for future research. Lancet Infect Dis 2015; 15: 581–614 [DOI] [PubMed] [Google Scholar]
- 2.United States Renal Data System. 2020 USRDS Annual Data Report: Epidemiology of kidney disease in the United States. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 2020
- 3.van der Poll T, van de Veerdonk FL, Scicluna BP et al. The immunopathology of sepsis and potential therapeutic targets. Nat Rev Immunol 2017; 17: 407–420 [DOI] [PubMed] [Google Scholar]
- 4.Lee CC, Lee MG, Hsu TC et al. A population-based cohort study on the drug-specific effect of statins on sepsis outcome. Chest 2018; 153: 805–815 [DOI] [PubMed] [Google Scholar]
- 5.Golia E, Limongelli G, Natale F et al. Inflammation and cardiovascular disease: from pathogenesis to therapeutic target. Curr Atheroscler Rep 2014; 16: 435. [DOI] [PubMed] [Google Scholar]
- 6.Grundy SM, Stone NJ, Bailey AL et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of blood cholesterol: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol 2019; 73: e285–e350 [DOI] [PubMed] [Google Scholar]
- 7.Fellstrom BC, Jardine AG, Schmieder RE et al. Rosuvastatin and cardiovascular events in patients undergoing hemodialysis. N Engl J Med 2009; 360: 1395–1407 [DOI] [PubMed] [Google Scholar]
- 8.Sim JJ, Zhou H, Shi J et al. Disparities in early mortality among chronic kidney disease patients who transition to peritoneal dialysis and hemodialysis with and without catheters. Int Urol Nephrol 2018; 50: 963–971 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sim JJ, Huang CW, Selevan DC et al. COVID-19 and survival in maintenance dialysis. Kidney Med 2020. (Epub ahead of print 1 December 2020); doi: 10.1016/j.xkme.2020.11.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ercan E. Statin treatment in dialysis patients after acute myocardial infarction improves overall mortality. Atherosclerosis 2017; 267: 156–157 [DOI] [PubMed] [Google Scholar]