Vasoconstriction agents are a pillar in the acute therapeutic management of patients suffering from septic shock. ^,, Nowadays, septic-shock treatment guidelines recommend norepinephrine as the first-line choice followed by epinephrine or vasopressin, even before fluid resuscitation is fully completed. ^,, However, there is no clear evidence that any single drug consistently improves microvascular flow in septic-shock patients. Vasopressors such as noradrenalin and vasopressin are commonly used to restore hemodynamic stability in the face of very low cardiac output and vascular tone in severe septic patients. ^,,

The literature contains numerous case studies of patients of all ages, who suffered from vasopressor-induced acute limb ischemia (VIALI), a condition that endangered patients’ limbs with variable measures taken to minimize the damage, e.g., application of local and systemic vasodilator agents alongside reduction of vasopressor dosages. ^,,, Oh and Song found that 0.8 % of septic shock survivors in a South Korean cohort developed peripheral limb ischemia, ultimately leading to limb amputations. Certain comorbidities and treatments were associated with an increased risk of peripheral gangrene and limb loss.

Several authors have examined the underlying mechanisms of vasoplegia in sepsis and defined risk factors associated with VIALI, such as prolonged exposure to vasopressors and the need for combination therapy with multiple vasopressor agents. ^, A meta-analysis assessing the effects and safety of vasopressin receptor agonists compared to catecholamines in septic shock patients across 20 randomized controlled trials, suggests that vasopressin use significantly increased the risk of digital ischemia.

The current study aimed to evaluate the adverse clinical outcomes associated with vasopressor use in patients with septic shock. Our main aim was the assessment of the potential risk of resulting amputations.

All adult patients aged 18 years or older who were admitted to the Sheba Medical Center between 2010 and 2023 with a primary diagnosis of septic shock were initially included in the study. Patients electronic medical records were approached after an Institutional Review Board affirmed this study according to the Helsinki Committee principles (# SMC-23-0348) and waived the need for informed consent due to the retrospective nature of the study. Patients who developed sepsis as a complication of treatment for another primary medical condition were excluded from the study. Additionally, those with insufficient data or those who were admitted to Intensive Care Units (ICUs) were excluded from the analysis in order to focus on patients treated in the settings of general-internal medicine departments. The remaining eligible patients were categorized into two groups: those who received vasopressor therapy and those who did not. Each group was further subdivided based on whether the patient underwent amputation during their hospitalization.

Categorical variables were compared between the study groups using a Chi-squared test. We then described each categorical variable by showing its prevalence, shown as a percentage (%) out of the entire cohort, and of each of the study groups. Continuous numeric variables were tested for normality using the Shapiro-Wilk normality test. A Student’s t-test was utilized to compare normally distributed variables between the study groups, and when normality was rejected, we conducted a Mann–Whitney U test instead. Normally distributed numeric variables are described as the mean ± standard deviation (SD), while non-normal variables are described as the median with interquartile range (IQR).

A logistic regression analysis was performed to evaluate the effect of the use of vasopressors on the likelihood of limb amputation. Results of the logistic regression analysis are presented as odds ratios (ORs) with 95 % confidence intervals (CIs) for both univariate and multivariate models, the latter adjusting for potential confounders. Statistical significance was determined as a P-value < 0.05. All statistical analyses were conducted using R (version 4.3.0, R Foundation for Statistical Computing).