Predicting Vasopressor Requirement in Sepsis: A Comparison Between Perfusion Index and Blood Lactate

Abstract

Background: Sepsis continues as a devastating global burden, largely due to microvascular failure that progresses rapidly towards hypotension. Timely vasopressor therapy is essential along with fluid resuscitation. Metabolic markers like lactate often fail to reflect early circulatory compromise, need invasive intervention and time consuming. So a more reliable predictor is crucial for guiding rapid intervention. This study was undertaken to determine whether the perfusion index, an indirect, noninvasive, real time indicator of peripheral microcirculatory failure offers superior predictive accuracy for vasopressor requirement compared to blood lactate during early sepsis management. Methods: This prospective observational study enrolled 96 adult patients with sepsis in the ICU of Dhaka Medical College Hospital, applying strict inclusion and exclusion criteria to ensure diagnostic accuracy and clinical consistency. After informed consent, each patient underwent detailed baseline assessment, including measurement of perfusion index and serum lactate, alongside comprehensive hemodynamic and laboratory evaluation. Patients were closely monitored during early resuscitation, with vasopressors initiated when shock or persistent hypotension occurred despite adequate fluid therapy. All collected data were systematically recorded, verified, and analyzed using SPSS, maintaining rigorous statistical standards throughout the study. Result: The study population demonstrated a balanced age and sex distribution with no significant differences in mean age between males and females (48.6±8.1 vs. 47.8±8.6; p=0.262), and comorbidities such as hypertension (42.7%) and diabetes (27.1%) were similarly represented across groups (all p>0.05), ensuring a comparable baseline for evaluating perfusion markers. Patients presented with diverse ICU diagnoses, including trauma, respiratory, neurological, and infectious conditions, reflecting a realistic spectrum of sepsis etiologies. Nearly half of the cohort exhibited low mean arterial pressure (<65 mmHg in 47.9%) and elevated lactate (mean 3.03 mmol/L), yet the perfusion index more clearly differentiated hemodynamic compromise, with 42 of 46 hypotensive patients having Perfusion Index ≤0.3. Its diagnostic accuracy was striking, achieving 91.3% sensitivity, 94.0% specificity, and a Positive Predictive Value of 93.3%, outperforming lactate, which showed lower sensitivity (87%), specificity (90%), and Positive Predictive Value (88.9%). This superiority was reinforced by a higher Area Under the Receiver Operating Characteristic Curve for Perfusion Index (0.900 vs. 0.870), establishing the perfusion index as a more dependable predictor of vasopressor requirement in sepsis. Conclusion: The findings demonstrate that the perfusion index provides a more accurate and responsive predictor of vasopressor requirement in sepsis than blood lactate.

Bangladesh Crit Care J March 2026; 14 (1): 28-33