Background: Obesity is an important risk factor for common chronic conditions, such as diabetes and heart disease. However, obesity has been associated with better outcomes for patients with sepsis. The underlying pathophysiologic basis for this obesity paradox is unclear. We sought to assess patient characteristics, quality of care, and clinical outcomes of patients with obesity who are hospitalized with community-onset sepsis.
Methods: From November 2020 to May 2023, professional abstractors at 67 hospitals participating in the Michigan Hospital Medicine Safety Consortium (HMS) retrospectively collected patient-level data on admissions for community-onset sepsis. Patients were included if their principal discharge diagnosis was sepsis, influenza, COVID-19, acute respiratory failure, or pneumonia, and they had objective evidence of suspected infection and acute organ dysfunction by the 2nd hospital day. We compared patient characteristics and quality of sepsis care (i.e., initial sepsis resuscitation bundle compliance) of patients with obesity (body mass index [BMI] >30 kg/m2) versus normal BMI (18.5-24.9 kg/m2). Primary outcomes were in-hospital mortality or discharge to hospice, and 30-day mortality. Associations between obesity and primary outcomes were assessed using multivariable logistic regression while adjusting for age, gender, admitted from LTAC/SNF/SAR/AR, hospitalization in prior 90 days, moderate/severe kidney disease, moderate/severe liver disease, congestive heart failure, malignancy, predicted mortality score, max lactate within 6 hours of hospital arrival, max creatinine on day 1, minimum PaO2:FiO2 ratio within 3 hours of hospital arrival, mechanical ventilation within 6 hours, altered mental status, receipt of vasopressor within 6 hours, and random effect for hospital.
Results: Of 33,649 patients hospitalized with community-onset sepsis, 13,151 (39.1%) patients were obese, 8,496 (25.2%) were overweight, 9,873 (29.3%) had normal BMI, and 2,129 (6.3%) were underweight. Patients with obesity were younger (median age 68 years [IQR 57-76 years] vs 74 years [IQR 63-83 years], p<.001), and had lower frequency of dementia (7.9% vs 18.1%, p<.001) and cancer (19.6% vs 28.5%, p<.001) compared to patients with normal BMI. Skin and soft tissue infection was more common in patients with obesity vs normal BMI (17.9% vs 8.6%, p<.001). The median predicted 30-day mortality was lower with obesity compared to normal BMI (7.7% [IQR 3.6%-17.0%] vs 17.1% [IQR 8.3%-32.5%], p<.001). Receipt of initial sepsis resuscitation bundle was similar in both groups (37.0% vs 37.4%, p=.76). However, lung protective ventilation strategy was used less frequently in patients with obesity (69.2% vs 80.0%, p=.001). In-hospital mortality or discharge to hospice care occurred in 13.6% of patients with obesity versus 23.7% with normal BMI (p<.001), while 30-day mortality was 13.9% and 24.2% respectively (p<.001). After adjusting for prespecified covariates, obesity was associated with lower odds of in-hospital mortality or hospice discharge (aOR, 0.82 [95% CI, 0.76-0.89]) and 30-day mortality (aOR, 0.81 [95% CI, 0.74-0.87]).
Conclusions: Among patients hospitalized with community-onset sepsis, obesity was associated with lower predicted mortality, similar initial resuscitation practices, and lower mortality, even after adjusting for illness severity.