Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Statins, sepsis, and chronic kidney disease

Bandolier once came across a paper that claimed that at least half of all indications for drug use arose from observations made by perceptive clinicians, rather than from the original intentions for their use by pharmaceutical companies. It is interesting, therefore, to perhaps see one swim into our ken, and perhaps watch it develop. The case of the possible effect of statins in reducing sepsis may be one of these.


A prospective observational study [1] has examined the use of statins and rate of sepsis in dialysis patients. Situated in the USA, the study began in 1995 to examine treatment choices and outcomes. Eligibility included long-term outpatient dialysis in the preceding three months in adults of at least 17 years, and it enrolled 1041 participants up to mid-1998, with observations continuing up to 2005.

Statin use was determined by review of clinic notes and computerised records. Data collected was extensive, including demographics, comorbidity, drug therapy, and laboratory values. The primary outcome was hospital admission for sepsis, where sepsis was defined using ICD codes. A number of different statistical analyses were performed, including multivariate regression and propensity score matching.


The mean age of patients was 57 years, about half men, and about 80% white. Statin users were more likely to be white, and have higher cholesterol levels, cardiovascular disease, and a history of sepsis, but were less likely to have used street drugs, and consumed less alcohol.

In the 1041 patients there were 303 hospital admissions for sepsis over the mean follow up of 3.4 years. The crude incidence rate was 4% per year in statin users and 11% per year in non-users (Figure 1). In the main statistical analysis, the crude incidence rate ratio was 0.37 (95% confidence interval 0.22 to 0.61). Using multivariate analysis with more complex interaction models, or propensity scoring, did not reduce the effect, but if anything made it larger. Various sensitivity analyses did not change the findings.

Figure 1: Crude rate of hospital admission for sepsis with and without statin


This was an extremely detailed study, with a moderate number of events, and with extensive efforts to discover possible sources of confounding, especially confounding by indication. It found none of these, and the result, a 60% reduction in the risk of sepsis with statins in dialysis patients looks strong.

Several other observational studies in bacteraemia or bacterial infection have also found improved outcomes in statin users, and a study of hospital admission for cardiovascular events found a lower incidence of sepsis with statin use. Moreover, there appears to be a biological plausibility, as the first statin was originally identified from a penicillin fungus, where it is theorised that it may have benefited the fungus by preventing replication of microorganisms requiring cholesterol for growth.

All in all an intriguing story based on some good observation. It will be interesting to see where it leads.


  1. R Gupta et al. Statin use and hospitalization for sepsis in patients with chronic kidney disease. JAMA 2007 297: 1455-1464.

previous or next story