Overview: Sepsis, biomarkers, and context in Rwanda
Sepsis remains a leading cause of death globally, with resource-limited settings bearing a disproportionate burden. In Rwanda, where ICU capacity is constrained, identifying simple, affordable prognostic tools is critical. This study examines whether low-cost white blood cell (WBC) indices and procalcitonin (PCT) levels can predict mortality among adults with suspected sepsis admitted to intensive care units (ICUs) at two national referral hospitals.
Study design and population
The researchers conducted a prospective cohort study in the ICUs of Butare University Teaching Hospital (CHUB) and Rwanda Military Referral and Teaching Hospital (RMRTH). Adults (≥18 years) with suspected sepsis based on Sepsis-2 criteria were enrolled, and a 40-day follow-up tracked survival. A total of 125 participants were included, with data captured within 24 hours of ICU admission and analyzed using survival methods such as Kaplan–Meier curves and Cox proportional hazards models.
Biomarkers and laboratory methods
Blood samples were divided for procalcitonin measurement and complete blood count (CBC). PCT was quantified using an electrochemiluminescence immunoassay, while CBC parameters included neutrophils, lymphocytes, monocytes, and the neutrophil-to-lymphocyte ratio (NLR). These low-cost, routine tests are widely available in many sub-Saharan African hospitals, offering practical tools for early risk stratification in sepsis care.
Key findings: neutrophils, NLR, and PCT
The study found that elevations in neutrophil counts and NLR were strongly associated with higher mortality risk in univariate and multivariable analyses. Specifically, a threefold increase in neutrophil count nearly doubled the hazard of death (adjusted hazard ratio ~2.0). Procalcitonin was higher among non-survivors and particularly elevated in patients who died within 24 hours, underscoring its role as a marker of sepsis severity, though it did not retain independent significance in all adjusted models when neutrophil counts were considered.
When PCT was combined with neutrophil count or total WBC, model discrimination improved, suggesting a synergistic value of biomarker panels. Time-dependent analyses showed that models integrating neutrophils or total WBC with clinical data achieved higher predictive accuracy in the early ICU period, with AUCs around 0.68–0.71 within the first 10–15 days of follow-up.
Clinical implications for resource-limited settings
In Rwanda’s context—where ICU beds are limited and advanced diagnostics may be scarce—the practical takeaway is clear: using neutrophil counts and NLR, alongside PCT, can support timely triage and guide early therapeutic decisions. These low-cost biomarkers may help identify patients needing escalated care, prompt antibiotic therapy, and prioritization of supportive measures, thereby potentially reducing sepsis mortality.
Limitations and future directions
The study focused on Sepsis-2 criteria and excluded advanced HIV/AIDS and other immunocompromised states, which may influence generalizability. Lymphocyte preservation observed in non-survivors contrasts with findings in other populations, suggesting regional differences in immune response. Future research should validate these results in broader African cohorts and explore context-specific scoring systems that incorporate readily available biomarkers like neutrophil counts and NLR.
Conclusion
Low-cost WBC indices, especially neutrophil counts and NLR, alongside PCT, show meaningful prognostic value for mortality in Rwandan ICU patients with suspected sepsis. The findings advocate for integrating these markers into bedside triage protocols to optimize outcomes where resources are constrained and rapid decision-making is essential.
