Predictors of Intravenous Immunoglobulin Treatment Success in Severe Clostridium difficile–AssociatedDisease

Background:

Clostridium difficile–associated disease (CDAD) is the most common cause of health care–associated infectious diarrhea. The past decade has witnessed an unprecedented increase in CDAD incidence, fatality rate, and resistance to metronidazole. catalyzing the search for new therapies. To date, 5 small studies have reported the use of intravenous immunoglobulin (IVIG) for severe CDAD with contradictory results, In our report, the largest so far, we investigated the usefulness of different variables in predicting IVIG treatment success for severe CDAD.

Methods:

We conducted a chart review of all episodes of severe CDAD treated with IVIG at a single center between July 1, 2002, and April 30, 2006. Severe CDAD was defined as positive stool toxin test and diarrhea with either radiologic or colonoscopic evidence of pancolitis or severe abdominal distension and ileus requiring total parenteral nutrition and lCU or lCU step‐down unit transfer. The primary outcome was the determination of independent predictors of hospitalization survival using single and multiple Cox regression analyses, These variables were: APACHE II score and serum lactate level on day 1 of IVIG infusion, age, sex, history of CDAD, number of days before IVIG use, peak white blood cell (WBC) count and creatinine level, lowest serum albumin and potassium levels, and number of antibiotics used that are not active against CDAD.

Results:

Twenty‐one of 1230 patients with CDAD were treated with IVIG (8 men, 13 women). The mean age was 68 years (35–98 years), with a mean hospital stay of 23 days (9–64 days). Conventional treatment was used for a mean of 8 days (1–25 days) before IVIG infusion. The mean APACHE II score was 25 (6–39) on day 1 of IVIG infusion. Nine patients (43%) survived their hospitalization with clinical symptoms resolution, whereas 12 patients (57%) died. One patient developed pulmonary edema after IVIG infusion. Symptoms resolved after an average of 10 days (range 2‐20 days) for survivors. Two patients underwent urgent colectomy. Both APACHE II score (P = 0.006) and lactate level (P = 0.004) on day 1 of IVIG infusion were found to be independent predictors of survival. However, only APACHE II score remained significant (P = 0.04) after adjusting for sex, previous history of CDAD, number of days before IVIG use, lactate level on day 1 of IVIG infusion, lowest albumin level, and number of antibiotics used that are not active against CDAD. No adjustments were made for age. WBC count, or potassium or creatinine levels, as those are included within the APACHE II score.

Conclusions:

IVIG treatment benefit in severe CDAD seems to depend on the extent of systemic involvement, with low APACHE II score on the first day of IVIG infusion being an independent predictor of survival. Thus, IVIG may only be beneficial early in the course of severe CDAD, before significant systemic involvement. Larger, prospective controlled trials are needed to verify the present findings.

Author Disclosure:

M Abougergi, none; A. Broor, none; W. Cui, none; B. Jaar, none.