Seth Bauer, PharmD, FCCM
Clinical Pharmacist
Cleveland Clinic
Lyndhurst, Ohio
Disclosure information not submitted.
Gretchen Sacha, BCCCP, PharmD
Critical Care Clinical Specialist
Cleveland Clinic
Cleveland, Ohio
Disclosure information not submitted.
Ajit Moghekar, MD, MBA
Staff Physician
Cleveland Clinic, United States
Disclosure information not submitted.
.jpg "Ryota Sato, MD photo")
Ryota Sato, MD
Staff Intensivist
The Queens Medical Center
Honolulu, Hawaii
Disclosure information not submitted.
Patrick Collier, MD, PhD
Staff Physician
Cleveland Clinic, United States
Disclosure information not submitted.
Richard Grimm, DO
Staff Physician
Cleveland Clinic, United States
Disclosure information not submitted.
Vidula Vachharajani, MD, FCCP, FCCM
Professor of Medicine
Cleveland Clinic Lerner College of Medicine
Westlake, Ohio
Disclosure information not submitted.
Siddharth Dugar, MD, FCCM
Associate Staff
Cleveland Clinic Foundation
Cleveland, Ohio
Disclosure information not submitted.
Title: Association of echocardiographic profiles and response to vasopressin in patients with septic shock
Introduction:
Septic shock patients have heterogeneous and dynamic cardiovascular dysfunction. Diverse hemodynamics leading to differential response to vasopressors may explain the lack of patient-centered outcome differences in clinical trials. Vasopressin, used as a catecholamine adjunct and a pure vasoconstrictor, may be detrimental in some hemodynamic profiles. We hypothesized that pre-drug echocardiographic (“echo”) profile is predictive of vasopressin response.
Methods:
This retrospective study included patients with septic shock receiving vasopressin as a catecholamine adjunct who had an echo performed after shock onset and prior to vasopressin initiation. Vasopressin response was defined as attainment of MAP ≥65 mm Hg with a decrease from baseline in catecholamine dose at 6 hours after vasopressin start. Left ventricular (LV) systolic dysfunction was defined as LVEF < 45%. LV diastolic dysfunction was defined as septal e’ < 8 cm/sec. A multimodal technique was used to define right ventricular (RV) dysfunction. Vasopressin response groups were compared.
Conclusions:
A total of 129 patients were included with 72 (56%) vasopressin responders. Most patients (99 [77%]) were medical admissions; echo was performed median 6.2 [IQR 2.2-13.8) hours after shock onset and 7.6 (3.2-19.5) hours before vasopressin start. Vasopressin non-responsive (NR), compared with responsive (R), groups were similar at baseline except the need for mechanical ventilation during echo (NR: 34 [60%] vs. R: 55 [76%], p=0.04). Importantly, we found between-group differences in LV systolic dysfunction (NR: 17 [30%] vs. R: 10 [14%], p=0.03) and LV diastolic dysfunction (NR: 33 [58%] vs. R: 29 [40%], p=0.05); no significant difference was detected in RV dysfunction (NR: 48 [84%] vs. R: 52 [72%], p=0.11). Patients with LV systolic dysfunction were less frequently vasopressin responsive (with LV dysfunction: 10/27 [37%] vs. without LV dysfunction: 62/102 [61%], absolute difference -24%; 95% CI -44% to -3%). After controlling for lactate, catecholamine dose, and APACHE III with multivariable logistic regression, higher LVEF was associated with vasopressin response (per 10% LVEF, OR 1.28; 95% CI 1.01-1.62). In conclusion, patients with septic shock receiving adjunctive vasopressin had differential response based on pre-drug echo profile.