Prognostic Nutritional Index (PNI) in Patients With Diabetes and Sepsis: A Cross-Sectional Study

BACKGROUND AND OBJECTIVES: Sepsis and multiorgan dysfunction syndrome (MODS) are serious global health concerns. Patients with critical illnesses are assessed using the sequential organ failure assessment (SOFA) score for morbidity. The prognostic nutritional index (PNI) is a novel composite biomarker for sepsis. It is based on two parameters: absolute lymphocyte count and serum albumin. The purpose of this study was to examine PNI in patients with diabetes and sepsis. Additionally, we correlated the SOFA score and PNI of the study participants at admission. METHODS: This cross-sectional study was carried out from August 2025 to December 2025 at the Kalinga Institute of Medical Sciences (KIMS), Bhubaneswar, India. Adult diabetes patients of both sexes fulfilling Sepsis-3 criteria were included in this study. We recorded their absolute lymphocyte count and serum albumin for PNI calculation. SOFA scores were noted at day 1, 3, and 7. The categorical and continuous variables were assessed with the Chi-square and the Wilcoxon test, respectively. We used the Spearman correlation to assess the association between the SOFA score at day 1 and the subjects' PNI values. R software (version 4.3.2) was used for data analysis. RESULTS: /L, respectively. The median PNI of the subjects was 34.64 (interquartile range (IQR), 30.32-39.39). The median SOFA score during admission was 5.0 (IQR, 3.0-8.0). There was a negative correlation between SOFA score at day 1 and PNI (-0.153, 95% CI: -0.233 to -0.071, p < 0.001). Male and female subjects demonstrated similar negative correlations. CONCLUSION: Male and female subjects with sepsis had comparable PNI values. However, serum albumin, absolute lymphocyte count, and SOFA scores showed statistically significant differences between genders. The PNI and SOFA scores were negatively correlated. The majority of the subgroup analyses also showed weakly negative associations. However, the study findings cannot be generalized due to a single-centric design, a small sample size, and missing data on hemodynamic, renal, hepatic, and glycemic parameters, antibiotics used, daily fluid intake, urine output, comorbidities, and concomitant medications.