Background β-Blockers are used widely in outpatient care for chronic disease, but little is known about therapy changes with septic shock. We sought to characterize the epidemiology of β-blocker therapy during and after septic shock among patients administered chronic β-blocker therapy in the year prior to hospitalization.

Methods We studied electronic health records of patients who received outpatient b-blocker therapy in the 12 months prior to hospitalization at 12 UPMC hospitals from 2010 to 2014. Eligible patients were adults (age ≥ 18 years) with septic shock, defined using modified Sepsis-3 criteria. Demographics, biomarkers, β-blocker and vasopressor administration were abstracted for 14 admission days. Patients were stratified as those who, during the hospital stay, were administered β-blockers daily (“continued”), had chronic therapy stopped for more than 24 hours with or without restart (“held”), or never received β-blockers (“discontinued”). Descriptive data were compared using the Kruskal-Wallis test or c2 test, as appropriate.

Results Of 22,208 patients, 3,748 were hospitalized with septic shock and received chronic β-blockers (mean age 67 ± 14 years, 56% male, 87% White), of which 405 (11%) continued therapy, 2,025 (54%) held therapy, and 1,317 (35%) discontinued therapy during their hospital stay. Patients in whom β-blockers were discontinued had greater SOFA score (continued, median SOFA 8.0 (IQR: 6.0-11.0); held, median SOFA 8.0 (IQR: 6.0-11.0); discontinued, median SOFA 10.0 (IQR: 7.0-12.0); p<0.001), while first-measured biomarkers, such as serum lactate, troponin, and platelets, were similar across groups. Among patients with β-blockers held, 50% of patient-days were on β-blockers alone, 22% on vasopressors alone, 7% on both, and 22% on neither. 92% of patients restarted β-blocker therapy with a median time of 4 days from admission (IQR: 2-6 days). In-hospital mortality was 17% if β-blockers were continued, 16% if held, and 41% if discontinued (p<0.001). Among 1,908 patients who survived 6 months after septic shock, 89% received outpatient β-blockers--most often those with inpatient β-blockers continued (95%) or held (94%) versus those discontinued (74%; p<0.001).

Conclusion In a multicenter cohort study of septic shock patients who received chronic β-blocker therapy, most patients had therapy discontinued or held during hospitalization. Among those with β-blockers held, inpatient prescribing was variable, overlapped with vasopressor use, and resumed, on average, four days after admission; nearly all patients restarted β-blocker therapy in the 6 months post-hospitalization. With increasing use of β-blockade for chronic disease, variable inpatient prescribing for septic shock necessitates optimizing adrenergy to improve resuscitation outcomes.