Objective: The objective of this study was to determine whether persistent lymphopenia on the fourth day following the diagnosis of sepsis predicts mortality. Methods: This was a single-center, retrospective cohort study of 335 adult patients with bacteremia and sepsis admitted to a large university-affiliated tertiary care hospital between January 1, 2010, and July 31, 2012. All complete blood cell count profiles during the first 4 days following the diagnosis of sepsis were recorded. The primary outcome was 28-day mortality. Secondary outcomes included development of secondary infections, 1-year mortality, and hospital and intensive care unit lengths of stay. Results: Seventy-six patients (22.7%) died within 28 days. Lymphopenia was present in 28-day survivors (median, 0.7× 10 3 cells/μL; interquartile range [IQR], 0.4–1.1× 10 3 cells/μL) and nonsurvivors (median, 0.6× 10 3 cells/μL; IQR, 0.4–1.1× 10 3 cells/μL) at the onset of sepsis and was not significantly different between the groups (P= 0.35). By day 4, the median absolute lymphocyte count was significantly higher in survivors compared with nonsurvivors (1.1× 10 3 cells/μL [IQR, 0.7–1.5× 10 3 cells/μL] vs. 0.7× 10 3 cells/μL [IQR, 0.5–1.0× 10 3 cells/μL]; P< 0.0001). Using logistic regression to account for potentially confounding factors (including age, Acute Physiology and Chronic Health Evaluation II score, comorbidities, surgical procedure during the study period, and time until appropriate antibiotic administration), day 4 absolute lymphocyte count was found to be independently associated with 28-day survival (adjusted odds ratio, 0.68 [95% confidence interval, 0.51–0.91]) and 1-year survival (adjusted odds ratio, 0.74 [95% confidence interval, 0.59–0.93]). Severe persistent lymphopenia (defined as an absolute lymphocyte count of 0.6× 10 3 cells/μL or less on the fourth day after sepsis diagnosis) was associated with increased development of secondary infections (P= 0.04). Conclusions: Persistent lymphopenia on the fourth day following the diagnosis of sepsis predicts early and late mortality and may serve as a biomarker for sepsis-induced immunosuppression.