RESUMO
BACKGROUND: With a growing obesity epidemic, the approach to care of this patient remains controversial and in many circumstances different than the general population. Appropriate hemodynamic support, although still controversial, remains a cornerstone of septic shock therapy. Catecholamines are currently recommended by guidelines without a preferred dosing strategy. However, the use of weight-based (µg kg-1 min-1) or nonweight-based (µg/min) vasopressor drip rates may impact patient care in these populations. METHODS: A multicenter retrospective chart review was conducted. Patients receiving nonweight-based catecholamine infusions for septic shock were grouped into nonobese (n = 112) or obese (n = 196), and evaluated based on hemodynamic resuscitation. For the primary outcome, groups were analyzed for the requirement of a secondary hemodynamic support agent to obtain a goal mean arterial pressure of greater than or equal to 65 mm Hg. Secondary outcomes included an evaluation of time to a secondary hemodynamic support agent, time to hemodynamic stability (HDS), ability to obtain HDS at 24 hours, and death due to cardiovascular collapse. RESULTS: With the exception of weight and sex, baseline characteristics were similar among groups. Early resuscitative fluids were given at a lower weight based, but not total volume dose in the obese group (nonobese, 34.8 mL/kg vs obese, 22.4 mL/kg; P < .0001). The primary end point of addition of any secondary hemodynamic support agent was significantly greater in obese patients when adjusted for institution (nonobese, 19% vs obese, 27%; adjusted odds ratio, 0.42; 95% confidence interval, 0.23-0.77). Time to HDS was also prolonged (nonobese, 3.5 hours vs obese, 5.3 hours; P = .006). CONCLUSION: This study calls into question the adequacy of a nonweight-based approach to hemodynamic support of critically ill obese patients. This strategy seems to result in less aggressive, lower weight-based vasopressor and fluid doses, and more diverse approach than their nonobese counterparts.
