Optimal Time for Vasopressin Initiation to Decrease Mortality Risk in Septic Shock

The risk of mortality was found to decrease among patients with septic shock when vasopressin was initiated during low-dose norepinephrine administration. These findings, from a retrospective observational study, were published in BMC Infectious Diseases.

Researchers in China sourced patient data for this study from the database of Beth Israel Deaconess Medical Center in the United States. Adults (N=1817) with septic shock who were admitted to an intensive care unit were evaluated on the basis of whether they received low- (<0.25 μg/kg/min) or high-dose (≥0.25 μg/kg/min) norepinephrine at the time of vasopressin initiation. The primary endpoint was mortality 28 days following septic shock diagnosis. Outcomes were also compared among a subset of 535 patients from both groups who were balanced via 1:1 propensity score matching.

Among unmatched patients in the low- (n=613) and high-dose (n=1204) norepinephrine groups, the mean (SD) ages were 65.3 (15.9) and 65.5 (15.7) years, 54.5% and 54.4% were men, 61.8% and 62.7% were White, respectively. During the course of septic shock, 22.5% vs 32.3% of patients in the low- vs high-dose groups were receiving hydrocortisone (P <.001), 7.5% and 18.3% were receiving epinephrine (P <.001), 8.0% and 13.0% were receiving dopamine (P =.002), and 25.3% and 34.6% required mechanical ventilation (P <.001).

Among propensity score matched patients (both groups n=535), the mean norepinephrine dose administered among those in the low- and high-dose groups was 0.14 (0.07) and 0.41 (0.28) μg/kg/min, respectively.

At 28 days, the mortality rate was 49.2% among patients in the low-dose group and 59.4% among those in the high-dose group. The risk of 28-day mortality was also decreased among patients in the low- vs high-dose groups when the groups were matched via other modalities. Results were as follows:

• Propensity score matching (odds ratio [OR], 0.660; P <.001);
• Propensity-scored based inverse probability weighting (OR, 0.622; P <.001);
• Doubly robust with unbalanced covariates (OR, 0.637; P <.001);
• Double robust with all covariates (OR, 0.635; P <.001); and
• Multivariable logistic regression (OR, 0.604; P <.001).

Patients in the low- vs high-dose groups also had significantly increased mechanical ventilation-free days (mean, 11.7 vs 8.8 days; P <.001) and continuous kidney replacement therapy-free days (mean, 13.9 vs 11.0 days; P =.001). In addition, significantly greater urine volumes at day 2 (mean, 1087 mL vs 899 mL; P =.012) and a significantly shorter duration of norepinephrine therapy (mean, 63.9 vs 76.0 hours; P =.014) were observed among patients in the low- vs high-dose groups.

Study limitations include the single-center retrospective design and the lack of adjustment for changes in septic shock management during the study period. In addition, increased disease severity potentially contributed to the comparatively higher mortality rate observed among patients who received high-dose norepinephrine.

According to the researchers, “[O]ur study proves that the optimal timing of vasopressin initiated can be earlier than recommended by guidelines, which provides a rational for prospective studies.”