Motivation: A sick, critically ill - possibly infected - patient remains somnolent despite many bags of antibiotics. A question that often occurs in this setting: is the patient immunocompromised? The answer determines whether to broaden antibiotic coverage. Occasionally, one of the clinical tests used to assess immune function is to measure the CD4 T cell subset count. But, is this is a valid tool in the setting of critical illness? Does the CD4 count stay normal in immunocompetent patients even when critically ill?
Paper: Aldrich, J., Gross, R., Adler, M., et. al. "The Effect of Acute Severe Illness on CD4+ Lymphocyte Counts in Nonimmunocompromised Patients." Arch. Intern. Med. (2000); 160: 715-716.
Methods: Prospective observational study in which patients admitted to the medical intensive care unit of Hospital of the University of Pennsylvania from 2/1996 to 4/1997 had CD4 cell count measured. Exclusion criteria were known HIV diagnosis, primary immunosuppressing disease, treatment with glucocorticosteroids, radiation therapy, or chemotherapy. Mortality outcomes were recorded for all patients.
Results:
Cohort: Of 353 admissions, 144 met inclusion criteria (exclusion criteria most frequently met were recent steroid use, immunosuppressive therapy, or refusal to be tested for HIV). Of the 144 meeting inclusion criteria, only 53 consented to participate in the study. In the final cohort, median age was 56 with 47% male. The four top admitting diagnoses were infection, gastrointestinal bleeding, end stage liver failure, and respiratory failure.
CD4+ Cell Count: The median CD4 cell count was 510 (normal range of 560 to 1840). Nine patients (17%) had counts lower than 200 while 29 (55%) had counts lower than 500. The median CD4/CD8 ratio was 2.2 (normal range: 0.9-3.4)
Mortality: Of the nine patients with CD4 counts less than 200, five (56%) died while only ten (23%) died with higher CD4 counts. The relative risk of death was 2.4 (95% CI: 1.1-5.3). The median CD4 count of those who died was 401 compared to the CD4 count of 510 for survivors (p = 0.05).
Discussion:
This observational study shows that critical illness alone is associated with depleted blood CD4 counts. In HIV, the blood CD4 count is a reflection of systemic T cell depletion in the tissue and lymph nodes. In critical illness, the CD4 count is likely changed by redistribution from the circulation to activated tissue sites. In this setting, the blood CD4 count may not be an accurate assessment of immune function and perhaps reflects severity of illness (increased mortality in lower CD4 group). Consequently, CD4 count in critical illness should be interpreted with caution.
Paper: Aldrich, J., Gross, R., Adler, M., et. al. "The Effect of Acute Severe Illness on CD4+ Lymphocyte Counts in Nonimmunocompromised Patients." Arch. Intern. Med. (2000); 160: 715-716.
Methods: Prospective observational study in which patients admitted to the medical intensive care unit of Hospital of the University of Pennsylvania from 2/1996 to 4/1997 had CD4 cell count measured. Exclusion criteria were known HIV diagnosis, primary immunosuppressing disease, treatment with glucocorticosteroids, radiation therapy, or chemotherapy. Mortality outcomes were recorded for all patients.
Results:
Cohort: Of 353 admissions, 144 met inclusion criteria (exclusion criteria most frequently met were recent steroid use, immunosuppressive therapy, or refusal to be tested for HIV). Of the 144 meeting inclusion criteria, only 53 consented to participate in the study. In the final cohort, median age was 56 with 47% male. The four top admitting diagnoses were infection, gastrointestinal bleeding, end stage liver failure, and respiratory failure.
CD4+ Cell Count: The median CD4 cell count was 510 (normal range of 560 to 1840). Nine patients (17%) had counts lower than 200 while 29 (55%) had counts lower than 500. The median CD4/CD8 ratio was 2.2 (normal range: 0.9-3.4)
Mortality: Of the nine patients with CD4 counts less than 200, five (56%) died while only ten (23%) died with higher CD4 counts. The relative risk of death was 2.4 (95% CI: 1.1-5.3). The median CD4 count of those who died was 401 compared to the CD4 count of 510 for survivors (p = 0.05).
Discussion:
This observational study shows that critical illness alone is associated with depleted blood CD4 counts. In HIV, the blood CD4 count is a reflection of systemic T cell depletion in the tissue and lymph nodes. In critical illness, the CD4 count is likely changed by redistribution from the circulation to activated tissue sites. In this setting, the blood CD4 count may not be an accurate assessment of immune function and perhaps reflects severity of illness (increased mortality in lower CD4 group). Consequently, CD4 count in critical illness should be interpreted with caution.
