Sepsis is a leading cause of death in the hospital setting and the prevalence of sepsis is increasing in developing countries (1). Researchers have previously speculated that the aging population (2) and other factors such as unhealthy lifestyle trends, as reflected by higher rates of obesity and complications of the metabolic syndrome (e.g., diabetes mellitus), were contributing factors. The latter seemed to be supported by observations linking certain diets to higher sepsis risk (3), and morbid obesity to increased sepsis mortality (4, 5). However, several studies reported that obesity is associated with lower sepsis mortality risk (6, 7). Despite recent meta-analyses that seem to support the “obesity paradox,” the controversy persists because many of the studies included in these analyses failed to adjust for confounding patient factors (inaccurate body mass index [BMI], smoking history, treatment bias) and/or selection bias (e.g., inclusion of subjects with less severe sepsis, analysis restricted to elderly populations) (8).
In this issue of Critical Care Medicine, Pepper et al (9) report the results of a large retrospective cohort analysis designed to confirm if obesity confers protection against sepsis mortality. The study by Pepper et al (9) was conducted using the CERNER Healthfacts electronic health record database (Cerner, North Kansas City, MO) derived from 139 U.S. hospitals from 2009 to 2015, which included 55,038 sepsis cases. This dataset includes specific clinical data that is routinely recorded, such as BMI, and the participating hospitals are representative of academic and nonacademic, large and small, urban and rural status. Although retrospective, the study by Pepper et al (9) is the largest study conducted to date to determine the relationship between obesity and sepsis mortality. The BMI data were collected just before or at the time of initial sepsis diagnosis prior to fluid resuscitation, which tends to inflate the BMI, and statistical adjustments were made for known confounders such as comorbidities, site of infection, and geographical location. After adjusting for potential confounding factors (e.g., chronic obstructive pulmonary disease, cancer, heart failure), lower the short-term (in-hospital or discharge to hospice) sepsis mortality correlated with higher BMI, with the highest mortality observed in underweight (31%) and lowest mortality (14%) in obese class-III (BMI > 40 kg/m2). Overall the relative risk of sepsis mortality among obese patients was 0.73 (0.70–0.77; p < 0.0001).
As with any retrospective study, there were study limitations. Historically, the diagnosis of sepsis based upon coding criteria is inconsistent among healthcare providers. Thus, the investigators did not rely solely on diagnosis codes and instead relied upon objective documentation of sepsis-related orders (blood cultures, antibiotics prescribed for at least 4 d) and documentation of organ failures based on the Sequential Organ Failure Assessment score, which is in keeping with the Sepsis-3 definition. However, data were often missing, resulting in the exclusion of greater than 75,000 sepsis encounters due to inadequate data. Another limitation was the reliance solely on sepsis mortality, without knowing the sepsis denominator within the obese and nonobese cohorts. As such, the prevalence of sepsis among obese patients compared with nonobese patients was unknown. Thus, it remains possible that sepsis may be more common in the obese population but is less severe, as reflected by a lower mortality rate.
Assuming that the findings of the study by Pepper et al (9) are valid, what explains the protective effects of obesity during sepsis? Are obese patients more likely to receive optimal care, as per established bundles of sepsis care? A study by O’Brien et al (10) indicates otherwise, showing that obese critically ill patients, many of which were septic, were less likely to receive recommended low tidal volume ventilation (i.e., 6 mL/kg based on ideal body weight), had higher plateau airway pressures, and received more opioid infusions. Despite more sedation and noncompliance with the ventilator bundle, mortality was lower among obese patients in the study by Pepper et al (9).
What other patterns of ICU care could confer protection in obese patients? As might be expected, a larger proportion of obese patients received insulin infusions (10), which has interesting implications in terms of promoting cell viability and tissue repair (11). As for the administration of noninvasive ventilation for acute respiratory failure, a large study by Lindenauer et al (12) conducted on patients presenting with pneumonia, the most common cause of sepsis, showed that obese patients, particularly those with obstructive sleep apnea, more often require invasive mechanical ventilation and paradoxically had lower mortality. Another possible explanation for the obesity paradox could be the tendency to perform fewer unessential procedures or diagnostic tests due logistical concerns (e.g., transport, imaging platform weight limits) and lower diagnostic yield (13). The contribution of variables relating to the patterns of ICU care that are unique to obese patients needs to be further studied.
In view of the data provided by Pepper et al (9) and by others (6, 7), it is reasonable to speculate that the biological and immunological consequences of obesity, and not the processes of care, are the primary determinants of improved sepsis outcomes. For instance, leptin is an adipokine that is typically elevated in obese humans that possesses pleomorphic functions, including modulation of the immune response. Prior studies show that obese compared with nonobese humans have elevated peripheral and CNS leptin levels and higher leptin levels correlate with higher BMI (14). Based on evidence that leptin functionally activates monocytes in vitro (15), a well-conceived study by Tschöp et al (16) evaluated supplementation of leptin via minipumps placed peripherally, or in intraperitoneal or intracerebroventricular locations. They demonstrated that increased CNS leptin levels promoted a more potent immune response, as reflected by bacterial clearance at the source of infection, reduced organ failure and improved survival in a murine sepsis model (16). The mechanisms of immune leptin sensitivity and CNS leptin regulation of the systemic immune response to promote sepsis survival in animal models are unclear, but merits further investigation and validation in humans.
Another variable not considered in the study of Pepper et al (9) is the obesity phenotype. Specifically, in comparison to peripheral obesity truncal obesity is associated with higher risk for the metabolic syndrome (e.g., insulin dependence) and related cardiovascular risks. The protective effects of peripherally located brown fat, which burns calories to produce heat, is thought to be linked to suppression inflammatory pathways (17). In support of this mechanism, it is shown that increased visceral adipose tissue distribution, compared with subcutaneous, is associated with increased sepsis mortality (18). Perhaps brown fat is the key to the obesity paradox?
Regardless of the mechanisms explaining the “obesity sepsis paradox,” the large, well-designed retrospective study by Pepper et al (9) provides the strongest evidence to date linking lower sepsis mortality to obesity. Future prospective, mechanistic studies are needed to determine which obesity phenotypes are most protected and to consider the mechanisms by which obesity confers protection.
1. De La Rica AS, Gilsanz F, Maseda E. Epidemiologic trends of sepsis
in western countries. Ann Transl Med 2016; 4:325
2. Martin GS, Mannino DM, Moss M. The effect of age on the development and outcome of adult sepsis
. Crit Care Med 2006; 34:15–21
3. Gutiérrez OM, Judd SE, Voeks JH, et al. Diet patterns and risk of sepsis
in community-dwelling adults: A cohort study. BMC Infect Dis 2015; 15:231
4. El-Solh A, Sikka P, Bozkanat E, et al. Morbid obesity in the medical ICU. Chest 2001; 120:1989–1997
5. Wang HE, Griffin R, Judd S, et al. Obesity and risk of sepsis
: A population-based cohort study. Obesity (Silver Spring) 2013; 21:E762–E769
6. Martino JL, Stapleton RD, Wang M, et al. Extreme obesity and outcomes in critically ill patients. Chest 2011; 140:1198–1206
7. Hutagalung R, Marques J, Kobylka K, et al. The obesity paradox in surgical intensive care unit patients. Intensive Care Med 2011; 37:1793–1799
8. Ng PY, Eikermann M. The obesity conundrum in sepsis
. BMC Anesthesiol 2017; 17:147
9. Pepper DJ, Demirkale CY, Sun J, et al. Does Obesity Protect Against Death in Sepsis
? A Retrospective Cohort Study of 55,038 Adult Patients. Crit Care Med 2019; 47:643–650
10. O’Brien JM Jr, Philips GS, Ali NA, et al. The association between body mass index
, processes of care, and outcomes from mechanical ventilation: A prospective cohort study. Crit Care Med 2012; 40:1456–1463
11. Yao H, Han X, Han X. The cardioprotection of the insulin-mediated PI3K/Akt/mTOR signaling pathway. Am J Cardiovasc Drugs 2014; 14:433–442
12. Lindenauer PK, Stefan MS, Johnson KG, et al. Prevalence, treatment, and outcomes associated with OSA among patients hospitalized with pneumonia. Chest 2014; 145:1032–1038
13. Modica MJ, Kanal KM, Gunn ML. The obese emergency patient: Imaging challenges and solutions. Radiographics 2011; 25:27–36
14. Schwartz MW, Peskind E, Raskind M, et al. Cerebrospinal fluid leptin
levels: Relationship to plasma levels and to adiposity in humans. Nat Med 1996; 2:589–593
15. Santos-Alvarez J, Goberna R, Sánchez-Margalet V. Human leptin
stimulates proliferation and activation of human circulating monocytes. Cell Immunol 1999; 194:6–11
16. Tschöp J, Nogueiras R, Haas-Lockie S, et al. CNS leptin
action modulates immune response and survival in sepsis
. J Neurosci 2010; 30:6036–6047
17. Villarroya F, Cereijo R, Gavaldà-Navarro A, et al. Inflammation of brown/beige adipose tissues in obesity and metabolic disease. J Intern Med 2018; 284:492–504
18. Pisitsak C, Lee JG, Boyd JH, et al. Increased ratio of visceral to subcutaneous adipose tissue in septic patients is associated with adverse outcome. Crit Care Med 2016; 44:1966–1973