Despite many studies linking compliance with sepsis bundles to improved survival, some authors have questioned their value (1–3). In this issue of Critical Care Medicine, Pepper et al (4) have published the results of an important meta-analysis evaluating the effect of sepsis bundles on mortality in patients with sepsis and septic shock. Clinicians invested in providing best care should hope the current study by Pepper et al (4) will put an end to the controversy about the importance of implementing sepsis bundles widely. Based on published data, it is already known that bundles save lives compared with “maximum flexibility,” as noted by the authors in their review of the evidence. By way of background, the original Surviving Sepsis Campaign (SSC) bundles were introduced in 2005 in partnership with the Institute for Healthcare Improvement. They were later revised in 2012 and then again in 2018 based on data from several studies that demonstrated increased compliance with the bundles to be associated with improved survival in patients with sepsis (5–8). The SSC sepsis bundles formed the basis for the New York State sepsis initiative (9) and the Centers for Medicare and Medicaid Services (CMS) SEP-1 (the sepsis reporting initiative from the Centers for Medicare and Medicaid Services) program (10).
The authors are to be commended for generating the current study (4), which demonstrates that implementation of sepsis bundles, based on the 17 papers included in the meta-analysis, is associated with improved survival. The authors were also able to show that the implementation of sepsis bundles alters clinician behavior, as evidenced by the increased number of times antibiotics and fluids were given within the bundle’s targeted time. This meta-analysis confirms the results of a previous meta-analysis of published studies of sepsis bundles by Damiani et al (11). In the prior meta-analysis, 50 published reports were evaluated and found to be associated with a statistically significant reduction in mortality with an odds ratio of 0.66.
To put the value of the study by Pepper et al (4) in context, several points are important as their study supports several statewide regulatory requirements for measuring compliance with sepsis bundles and the CMS SEP-1 initiative.
First, it is important to examine the inclusion criteria for these two meta-analyses of sepsis bundles. Because no published randomized controlled trials of sepsis bundles exist, these meta-analyses must examine prospective cohort studies. The study by Damiani et al (11) incorporated all prospective cohorts, including studies that use interrupted time-series methodology, a methodology that often provides better adjustment for temporal changes and confounding than either before-and-after or concurrent control groups. The meta-analysis by Pepper et al (4) included only studies that used before-and-after or concurrent control group methodologies. Unfortunately, this approach did not include several of the largest and most recently published prospective cohort trials of sepsis bundles, thereby excluding three studies that had more than 30,000 patients in each study (6–8). The relatively small numbers of patients in many of the studies examined by Pepper et al (4) contributed to an underpowered meta-analysis for examining the secondary outcomes reported by the authors: timing of antibiotics and fluid administration.
Second, acknowledging the limitations of both the relatively small sample size and the heterogeneity in design and quality for the included studies is important. Based on their inclusion criteria, the authors examined studies with fewer than 250 patients for 10 of the 17 studies. Of the 17 studies included: five used adjusted analyses while five did not and 10 examined hospital mortality while six used different endpoints. In addition, some studies used various bundle times, variable time zeros, and variable control groups. There was also heterogeneity across settings of care (emergency department, wards, or ICU) and different geographical regions. In reporting the secondary outcomes of the meta-analysis—namely timing of antibiotics and volume of fluid administration—the studies included in the analysis that recommended antibiotic administration in less than 1 hour had a combined population of 1,200 patients compared with the more than 28,000 in the less than 3-hour group. Given the size and heterogeneity of the included studies in the meta-analysis, it should not be surprising that the authors did not demonstrate statistical significance for either timing of antibiotics or volume of fluid. Importantly, the three excluded studies with much larger numbers of patients provide strong support for the delivery of antibiotics within 1 hour. Further, because many of the included studies did not report the exact amount of fluids administered to patients receiving the sepsis bundles, assessing the volume of fluid is nearly impossible. Given these limitations, drawing conclusions about these secondary outcomes from this meta-analysis is difficult.
Finally, the authors correctly highlight the importance of adhering to the National Quality Forum (NQF) criteria for quality measure endorsement. The sepsis quality metrics that are the basis for SEP-1 and most of the published studies on sepsis bundles are based on the 0500 NQF sepsis bundle that has been endorsed by NQF (12).
The authors also raise some important questions that should be examined in future studies with sepsis bundles. First, is 1 hour more beneficial than 3 hours as a target for antibiotic administration? Although a large volume of published studies demonstrates significantly increased mortality associated with hour-by-hour delay in antibiotic administration (13), this remains an extremely important question.
Second, is there unintended harm associated with sepsis bundles? The authors state that the potential harm of the sepsis bundles has not been evaluated; this is not entirely true as the studies included in the meta-analysis demonstrate the bundles to be associated with improved all-cause mortality. Therefore, if there are some unevaluated harms, those harms must be counter-balanced by benefit. In addition, a study evaluating the positive effect of sepsis bundles on patient populations thought to be at considerable risk for unintended consequences, namely those with congestive heart failure and end-stage renal disease, has been published (14). It demonstrated improved survival specifically in this patient population. Nonetheless, future studies should assess the effect of sepsis bundles on nonseptic patients and the downstream complications associated with antibiotic utilization in patients receiving sepsis bundles.
Third, an opportunity to develop bundles that are personalized to individual patients is apparent. It is important to note that, although the concept of individualized bundles may be attractive as we enter the era of precision medicine, there is compelling data from across different fields of medicine, as well as extensive experience in quality improvement, that key components of effective quality improvement include standardization of care, implementation of change, and measurement of improvement. In fact, this is precisely what was demonstrated by Pepper et al (4) in this important meta-analysis—standardizing clinical behavior through the use of bundles can save lives. Pepper et al (4) make a case for flexibility, which is always part of every responsible bundle. Individual clinicians caring for individual patients are always responsible for assessing whether a particular bundle is in the best interests of this particular patient and responsible bundles must allow clinicians to opt-out of the bundle requirements with documentation of their rationale.
Given the strong consistency in the overall results of these two important meta-analyses, it is clear that sepsis bundles are associated with improved survival. Unfortunately, the quality of the existing evidence makes it difficult to draw definitive conclusions about the single best choice for the timing of antibiotics and the volume of fluid. Without question, additional data are necessary to better understand individual bundle elements, and specific recommendations for bundles will always need to be updated with new data. Nonetheless, given that bundles save lives, clinicians have an obligation to act now on the best available data. Confirmed by the work of Pepper et al (4) and the recently published study by Kahn et al (15), there is an urgent need to implement sepsis bundles widely for the benefit of the severely ill septic patients.
1. Klompas M, Calandra T, Singer M. Antibiotics for sepsis
-finding the equilibrium. JAMA 2018; 320:1433–1434
2. Marik PE, Farkas JD, Spiegel R, et al.; Collaborating Authors: POINT: Should the Surviving Sepsis
Campaign Guidelines be retired? Yes. Chest 2019; 155:12–14
3. Jaswal DS, Natanson C, Eichacker PQ. Endorsing performance measures is a matter of trust. BMJ 2018; 360:k703
4. Pepper DJ, Sun J, Cui X, et al. Antibiotic- and Fluid-Focused Bundles Potentially Improve Sepsis
Management, but High-Quality Evidence Is Lacking for the Specificity Required in the Centers for Medicare and Medicaid Service’s Sepsis
). Crit Care Med 2019; 47:1290–1300
5. Levy MM, Dellinger RP, Townsend SR, et al.; Surviving Sepsis
Campaign: The Surviving Sepsis
Campaign: Results of an international guideline-based performance improvement program targeting severe sepsis
. Crit Care Med 2010; 38:367–374
6. Levy MM, Rhodes A, Phillips GS, et al. Surviving Sepsis
Campaign: Association between performance metrics and outcomes in a 7.5-year study. Crit Care Med 2015; 43:3–12
7. Seymour CW, Gesten F, Prescott HC, et al. Time to treatment and mortality during mandated emergency care for sepsis
. N Engl J Med 2017; 376:2235–2244
8. Levy MM, Gesten FC, Phillips GS, et al. Mortality changes associated with mandated public reporting for sepsis
. The results of the New York state initiative. Am J Respir Crit Care Med 2018; 198:1406–1412
11. Damiani E, Donati A, Serafini G, et al. Effect of performance improvement programs on compliance with sepsis bundles
and mortality: A systematic review and meta-analysis of observational studies. PLoS One 2015; 10:e0125827
13. Peltan ID, Brown SM, Bledsoe JR, et al. ED door-to-antibiotic time and long-term mortality in sepsis
. Chest 2019; 155:938–946
14. Liu VX, Morehouse JW, Marelich GP, et al. Multicenter implementation of a treatment bundle for patients with sepsis
and intermediate lactate values. Am J Respir Crit Care Med 2016; 193:1264–1270
15. Kahn JM, Davis BS, Yabes JG, et al. Association between state-mandated protocolized sepsis
care and in-hospital mortality among adults with sepsis
. JAMA 2019; 322:240–250