Objectives: To determine the range of T-lymphocyte subsets (CD4, CD8,
and CD4/CD8 ratios) in acutely ill, hospitalized patients and to deter
mine whether these concentrations correlate with illness severity, sur
vival rate, or immunodepression. Design: Cross-sectional study, compar
ing Acute Physiology and Chronic Health Evaluation II (APACHE II) scor
es and the calculated, disease-specific, predicted mortality rate with
T-lymphocyte subsets. Setting: Urban county hospital intensive care u
nit (ICU), serving as the designated trauma center. Patients: One hund
red two consecutively admitted ICU patients (72 medical and 30 surgica
l). Interventions: None. Measurements and Main Results: Patient clinic
al data, APACHE IP scores, and their associated predicted mortality ra
te were recorded, Blinded human immunodeficiency virus (HIV) and lymph
ocyte testing was performed on samples from all patients on ICU admiss
ion, Despite only three (2.9%) of 102 patients testing positive for HI
V antibodies, 41% (42/102) of patients had CD4 concentrations of <400
cells/mu L, and 29% (29/102) had CD4 concentrations of <300 cells/mu L
. Mean CD8 concentrations were even lower, compared with normal labora
tory values, resulting in a slight increase in CD4/CD8 ratios, althoug
h 16% (16/102) of patients had a CD4/CD8 ratio of <1. CD4 counts were
linearly related to total lymphocyte concentrations (Pearson correlati
on coefficient = 0.948), but no relationship was found between total l
ymphocyte or lymphocyte subset counts and APACHE II score, predicted m
ortality rate, or survival rate. Conclusions: Acute illness alone, in
the absence of HIV infection, can be associated with profound decrease
s of T-lymphocyte populations, This problem is unpredictable and does
not correlate with severity of illness, predicted mortality rate, or a
ctual mortality rate, No conclusions regarding HIV serostatus or survi
val can be made based on single measurements of T-cell concentrations
in acutely ill hospitalized patients.