C. difficile Enterocolitis in Critically Ill and Non-Critically Ill Burn and Non-Burn Populations

Author(s):
Mathilda Nicot-Cartsonis; George Golovko; Juquan Song; Steven Wolf

Background:

C. difficile colitis (C diff colitis) is identified in up to 500k infections/year in the US, with up to 30,000 people per year dying with the diagnosis. Previous studies established risk factors for this condition including long-term antibiotic use. Burn patients are typically exposed to many antibiotics over long periods of time, and may have lengthy hospitalizations similar to other critically ill patients. Neither of these populations have been studied in terms of C diff colitis, risk and outcomes alike. Our aim is to compare C diff colitis rates among critically ill and non-critically ill burn populations, and to investigate the use of antibiotics prior to diagnosis.

Hypothesis:

We suspect that C diff colitis diagnosis would be elevated in critically ill burn patients as compared to critically ill non-burn and non-critically ill burn patients, as they are known to receive generous amounts of antibiotics and are at increased risk of infection at baseline.

Methods:

All analyses were generated with TriNetX platform software (TriNetX, Cambridge, MA) in 2023. We compared incidence of C diff colitis in non-critically ill and critically ill burn populations, and non-burn injury control groups. Cohorts were propensity matched for age at diagnosis, sex, and race/ethnicity.  The difference between groups was tested using the log-rank test and quantified with hazard ratios (95% CI), calculated with TriNetX Analytics features. We examined length of hospital stay of cohorts, antibiotic use prior to C diff colitis diagnosis, death within 1 year, and colectomy/toxic megacolon diagnosis.

Results:

Critically ill patients are diagnosed with C diff colitis at similar rates (OR = 1.065, 95% CI = 0.89, 1.276), but higher rates compared to non-critically ill burn (OR = 1.391, 95% CI = 1.148, 1.686), Non-critically ill burn patients had higher rates compared to non-burn injury counterparts (OR = 1.713, 95% CI = 1.471, 1.995). Antibiotics use 2 weeks prior to C diff colitis diagnosis was higher in C diff colitis patients in all four cohorts. Specifically for critically ill burn C diff colitis patients, penicillin (p=0.0072), tetracyclines (p=0.0037), trimethoprim (p=0.0011), vancomycin (p<0.0001), sulfonamides (p=0.0005) to name a few were given more frequently compared to those without C diff colitis. All C diff cohorts were more likely to die within 1 year of diagnosis (Burn+critical OR=2.196, 95%CI=1.795, 2.732; Trauma+critical OR=1.196, 95%CI=1.747, 2.1; Burn+no critical OR=4.011, 95%CI=3.357, 4.792; non-burn injury+no critical OR=2.958, 95%CI=2.89, 3.027).

Conclusions:

Critically ill burn patients are at increased risk of C diff colitis compared to non-critically ill burn. C diff colitis is associated with higher mortality, antibiotics, and length of stay in both burn and non-burn cohorts.