Abstract & Commentary

Does Aspirin Reduce Mortality in the Systemic Inflammatory Response Syndrome?

By David J. Pierson, MD, Editor, Professor Emeritus, Pulmonary and Critical Care Medicine, University of Washington, Seattle, is Editor for Critical Care Alert.

Synopsis: In this retrospective study, patients with systemic inflammatory response syndrome who had received aspirin had a lower mortality rate than those who had not. The groups were different in numerous ways, although propensity analysis to account for confounding variables and make the groups more comparable failed to completely eliminate the mortality difference.

Source: Eisen DP, et al. Acetyl salicylic acid usage and mortality in critically ill patients with the systemic inflammatory response syndrome and sepsis. Crit Care Med 2012;40:1761-1767.

In this retrospective cohort study of patients with the systemic inflammatory response syndrome (SIRS) in an Australian ICU, those with SIRS who received acetyl salicylic acid (ASA; aspirin) were compared to those who did not receive aspirin. Using the hospital's electronic medical records, the authors identified all patients admitted to the ICU during a 9.5-year period who met two or more of the four SIRS criteria proposed by the 1992 American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference:1 temperature > 38° C, heart rate > 90 beats/min, respiratory rate > 20 breaths/min (or PaCO2 < 32 mmHg), and white blood cell count > 12,000/mL (or < 4000/mL or > 10% band forms). For all such patients, they determined whether aspirin or another non-steroidal anti-inflammatory drug (NSAID), such as ibuprofen, had been administered during the same 24-hour period in which the SIRS criteria were met, along with patient demographics, diagnoses, selected laboratory data, and in-hospital mortality. The investigators then performed covariable and propensity analyses to decrease the effect of confounding variables in comparing aspirin and non-aspirin patients with respect to mortality, documented bleeding complications (not further specified), and a decrease in renal function (defined as a doubling of initial serum creatinine level).

During the interval reviewed, there were 7945 admissions to the study ICU. Of these, 6131 met the authors' SIRS inclusion criteria. Excluding repeat admissions, 2082 patients with SIRS received aspirin and 3441 did not. Overall in-hospital mortality in the two groups was 9% and 20%, respectively. Seven percent of all patients with SIRS received an NSAID other than aspirin, and these patients were about evenly distributed between the aspirin and non-aspirin groups. The patients in the aspirin-receiving and non-aspirin groups were quite different: the former were older by a mean of 8 years, more likely to be male (71% vs 59%), and had cardiac-related comorbidities, 48% of them having been admitted for coronary artery bypass grafting vs 15% of the non-aspirin-receiving patients. Propensity analysis, in an attempt to eliminate these differences, matched 1445 of the aspirin-receiving patients with an equal number of non-aspirin patients.

The difference in mortality among aspirin and non-aspirin patients with SIRS was less pronounced after the propensity analysis, but remained statistically significant (10.9% mortality vs 17.2%), with an absolute difference of -6.2% (95% confidence interval [CI], -9.5% to -3.5%). After propensity matching, the investigators also found a higher risk of developing renal impairment (6.2% incidence vs 2.9%; absolute risk difference, 3.3%; 95% CI, 2.5% to 5.0%) in patients who received aspirin, but a lower risk for bleeding (8.8% incidence vs 17.6%; absolute risk reduction, -8.8%; 95% CI, -11.1% to -7.3%).

Commentary

In the discussion, the authors point out that because it is retrospective and thus quasi-experimental, "this study cannot establish a causal association between ASA and death from SIRS or sepsis. Despite efforts to adjust for confounding factors, the difference in mortality between ASA users and nonusers could potentially be attributable to residual confounding." This is the crux of the matter for all readers of this article, and especially for clinicians wishing to determine whether the findings are relevant to their practices.

This study prompts consideration of the hypothesis that aspirin might have a mortality-reducing effect on SIRS. However, two issues make me reluctant to embrace this hypothesis. The first has to do with how different the patient groups turned out to be, as illustrated by likely explanations for the renal and bleeding effects. Renal toxicity is not a common adverse effect of aspirin in therapeutic doses, and the authors state in the discussion that the patients in this study "had been using ASA before admission to the ICU." Thus, it is difficult to generate an aspirin-related hypothesis as to the decline in renal function after ICU admission. In addition, bleeding is one of the common adverse effects of aspirin use and the study found the rate of bleeding to be twice as high in the non-aspirin-receiving patients. These things suggest to me that the renal and bleeding rates were most likely the results of factors other than the use of aspirin, and they weaken the assertion that the mortality difference was due to this agent.

The second issue is that ibuprofen, an NSAID like aspirin whose effects in SIRS and sepsis would be expected to be similar, has already been subjected to a multicenter, randomized, controlled trial in patients with sepsis,2 as mentioned in the introduction. That study was negative with respect to an effect of ibuprofen on mortality.

Aspirin is cheap and very widely used, and administering it to patients with SIRS who do not have contraindications would likely be safe. Nonetheless, I agree with the authors' call for caution in interpreting their findings, quoted above, and also with their statement in the abstract that "the effect of acetyl salicylic acid on mortality of patients with systemic inflammatory response syndrome and sepsis needs to be evaluated with prospective randomized intervention studies."

References

  1. Bone RC, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992; 101:1644-1655.
  2. Bernard GR, et al. The effects of ibuprofen on the physiology and survival of patients with sepsis. The Ibuprofen in Sepsis Study Group. N Engl J Med 1997;336:912-918.