Sepsis Management in a Resource-Limited Setting: A Clinical Trial of Quantitative Resuscitation in a National Referral Hospital in Kenya

Document Type

Conference Paper


Anaesthesiology (East Africa)


RATIONALE Fluid resuscitation for sepsis in sub-Saharan Africa (SSA) is controversial. Previous clinical trials of protocol-based sepsis resuscitation for adults in SSA yielded conflicting results, with one showing improved mortality and two others showing harm. A widely used guideline for resuscitation of sepsis in low resource settings is the World Health Organization’s Integrated Management of Adult Illness (IMAI) District Clinician Manual. The efficacy of the IMAI guideline has not been rigorously evaluated. METHODS We examined the efficacy of the IMAI sepsis guideline at Moi Teaching and Referral Hospital in Eldoret, Kenya in a before-and-after clinical trial design. Adult subjects presenting to the Casualty Department with modified sepsis criteria (suspected infection and ≥2 of: temperature >38.0°C or <36.0°C; heart rate >90 beats per minute; respiratory rate >20 breaths per minute; systolic blood pressure <100 mmHg) and lactate ≥2.5 mmol/L were enrolled. In the “before” (control) phase, subjects received usual care and were observed prospectively. In the “after” (intervention) phase, newly enrolled subjects were treated with a resuscitation protocol based on the IMAI guideline (Figure). The primary outcome measure was percent lactate clearance at six hours, defined as the difference between initial lactate and lactate at six hours, divided by initial lactate; lactate clearance has been shown to correlate closely with survival in other settings. Secondary outcomes included 24-hour, in-hospital, and 30- day survival. For the primary outcome of lactate clearance, we used the student’s t-test and linear regression for unadjusted and adjusted analyses, respectively. For the secondary survival outcomes, we used the χ-squared test and logistic regression for unadjusted and adjusted analyses, respectively. RESULTS Of 139 subjects enrolled, 136 completed the study (49.3% female, mean age 42 years, standard deviation [SD] 16 years). Nearly half (49.0%) had HIV. Mean initial lactate was 4.1 mmol/L (SD 2.2 mmol/L). Lactate clearance at six hours was significantly higher in the intervention group than the control group (mean clearance 40.8% versus 14.0%, p<0.001). 24-hour, hospital, and 30-day survival were higher in the intervention group, but the difference was not statistically significant in bivariate or adjusted analysis (98.4% versus 93.2% [p=0.22], 80.7% versus 74.3% [p=0.38], and 82.3% versus 75.7% [p=0.35], respectively). CONCLUSIONS The IMAI algorithm resulted in significantly improved lactate clearance in this population with severe sepsis in Kenya. Survival was not significantly different between the control and intervention groups. A potential correlation between greater lactate clearance and survival warrants further study in this region.


American Journal of Respiratory and Critical Care Medicine