Crystalloid Liberal or Vasopressors Early Resuscitation in Sepsis-Study of Treatment's Echocardiographic Mechanisms (CLOVERS-STEM)

Abstract

Importance: Receipt of fluid and vasopressors, common treatments in septic shock, may affect cardiac function.

Objectives: We sought to determine whether a liberal or restrictive fluid resuscitation strategy was associated with changes in cardiac function.

Design: We prospectively studied a subset of patients enrolled in the Crystalloid Liberal or Vasopressors Early Resuscitation in Sepsis (CLOVERS) trial, performing echocardiography at baseline and at 24 hours after randomization. Among patients who had an echocardiogram performed at 24 hours, we measured left ventricular global longitudinal strain (LV GLS) and right ventricular free-wall longitudinal strain (RVFWLS). We performed linear regressions with dependent variables of LV GLS, change in LV GLS (ΔLV GLS), and RVFWLS using treatment assignment as an independent variable. We adjusted for ratio of early diastolic mitral inflow velocity to early diastolic mitral annulus velocity, mean arterial pressure, and history of congestive heart failure and myocardial infarction.

Setting: Emergency department and ICUs.

Patients: Adults with sepsis enrolled in the CLOVERS trial.

Main outcomes and measures: We enrolled 180 patients. Our analytic cohort comprised 131 patients with an echocardiogram performed at 24 hours. We observed no differences between treatment arms with respect to demographic, clinical, or echocardiographic data at baseline. We observed no association between restrictive fluid assignment and LV GLS (coefficient, 1.22; p = 0.23), ΔLV GLS (-1.97; p = 0.27), or RVFWLS (2.33; p = 0.19).

Conclusions and relevance: In a subset of patients enrolled in CLOVERS, we observed no association between receipt of fluid and vasopressors and short-term changes in cardiac function. Decreased enrollment may limit inferences.