Monte Minute‎ > ‎Monte Minute‎ > ‎

Severe Clostridium Difficile Infection - 10/16/2012

posted Oct 16, 2012, 6:38 AM by Rohit Das   [ updated Dec 27, 2012, 6:42 AM by Purnema Madahar ]
Severe CDI at the Weiler Division, by Rohit Das

At autopsy conference yesterday (which was awesome), we went through a case of an unfortunate woman who had recently had been sent to a long-term facility on antibiotics for chronic osteomyelitis, who subsequently developed diarrhea and sepsis, all likely due to severe Clostridium difficile infection (CDI). On gross pathology, there were diffuse pseudomembranes throughout her colon, consistent with this diagnosis. SO – I’d thought I’d talk about CDI a bit, focusing on some epidemiology, how’s that’s been changing and why, and finally, the definition and management of severe infection.

What is the epidemiology of CDI, and how has it been changing since the early 2000s?

·         The attached NEJM article provides an excellent overview of CDI and the change in its behavior over the last decade or so. As the authors summarize, in the 1990s, the incidence of CDI remained pretty stable at 30-40 cases per 100,000 population. By 2001, that number rose to 50, and by 2005, rose to around 85 (around 3-times the rate of the mid-90s).

·         Of particular interest, in 2003, an epidemic in Quebec was reported by Pepin et. al., reporting an almost 4-fold increase in incidence (from 25 per 100,000 population, to 95 in 2003).

·         Perhaps more importantly, this increase in incidence has been accompanied by increased severity of disease, and worse clinical outcomes. In one study in England, CDI was listed as the primary cause of death in 500 patients in 1999, which rose to nearly 2000 in 2005, and ~3400 in 2006. In Quebec, the increased incidence was also associated with a substantial increase in severity/mortality, with CDI being listed as a contributing factor in 8% of case-mortalities, and the leading cause of death in 7% of cases (Around 1-3% previously).

These are some crazy numbers…why is this happening?

·         BECAUSE WE DON’T WASH OUR HANDS! Actually, it’s a bit more than that. First, let’s go over some pathophysiology. The central pathogenic process in CDI is production of toxin, and movement of that toxin into the cytosol of enterocytes. This ultimately leads to, via cellular pathways, apoptosis and badness.

·         CDI typically produces two toxins – Toxin A and B. As depicted in Figure 2 of the attached NEJM article, both toxins have binding domains that allow the enzymatic portion to enter cells. Toxin B is particularly virulent, being 10-times more potent than Toxin A. In vivo, toxin levels correlate with disease severity. Importantly, there are CDI strains that are NON-toxigenic and colonize the colon without clinical ramifications.

·         In association with the epidemic described in Canada in 2003, investigators identified a strain of CDI labeled NAP1/027 as the predominant strain responsible for the numbers described above (Lancet article attached). This strain has also been shown to be responsible for increased incidence/severity in U.S. studies. Pepin et al. showed that this strain produces significantly more toxin than previous strains, thus explaining both the recently increased incidence and severity of CDI.

·         There are a few possible pathophysiologic reasons why NAP1/027 is hypervirulent. It’s been shown to produce “binary toxin” more than previous strains, though the role of this toxin in pathogenesis is not clear (it causes enterolysis in vivo, though). Secondly, NAP1/027 is also characterized by an 18 base-pair deletion of the tcdC locus of the toxin gene (see the NEJM review) – the normal role of this gene is to inhibit toxin transcription, and the deletion may impair that ability. FINALLY, NAP1/027 has high-level resistance to flouroquinolones, suggesting that their use has selected for this hypervirulent strain.

Given all this…how do we define “severe CDI”? Is there data to support certain management options over others?

·         There is really no consensus definition for what defines severe infection. Many studies have defined certain clinical factors associated with worse clinical outcomes, such as a WBC > 20k, elevated creatinine, immunosuppression (e.g., chemotherapy, chronic steroid use), age > 60, and signs of sepsis. Scoring systems have been developed in small retrospective studies, but never validated in prospective cohorts. Ultimately, judging whether CDI is severe or not is based on clinician’s judgment.

·         There has been much debate as to whether CDI, initially, should be treated with oral Metronidazole or Vancomycin. Consensus agreement is that Vancomycin should be the initial treatment for severe infection, and this is somewhat supported by randomized trials. In the 2007 paper attached by Zar et. al., patients were stratified by severity of infection, and randomized to Metronidazole or Vancomycin, along with a matched placebo. Severe CDI was defined by 2 of the following characteristics – age >60, temp > 38.3oC, albumin <2.5mg/dL, WBC > 15k – within 48 hours of enrollment.

·         With a primary outcome of cure (resolution of diarrhea within 6d of treatment initiation), there was no difference between Metronidazole and Vancomycin in cure rate for non-severe infection  (90 vs. 98%), but a significant difference for severe infection (76% versus 97%). Interesting results, though many gripes with this paper, including a severity scoring system that hasn’t really been verified, lack of confidence intervals for the therapeutic benefits they report, and primary outcomes that were rather subjective and not “hard” clinically (i.e., mortality, surgical intervention, etc…).

·         So, the general agreement is to use Vancomycin for severe infection, with some limited supportive data. In addition, in patients with significant ileus or toxic megacolon (colonic dilatation >6-7cm with signs of systemic toxicity), IV Metronidazole and/or intracolonic Vancomycin are recommended (though only with case reports backing them up).

·         Regarding surgical intervention, previous recommendations were to wait for 48 hours to deem CDI unresponsive to medical therapy and require surgical intervention. With the emergence of the aforementioned hypoervirulent strain, earlier surgery is now favored, though optimal timing to surgery is still unclear. In the setting of peritoneal signs, ileus or toxic megacolon surgical intervention is definitely advisable.

·         Other potential therapeutic options, like probiotics, Fidaxomicin, IV immunoglobulin, and fecal transplant have been used, but not substantiated by randomized controlled trials (to clarify, Fidaxomicin was approved on the basis of a non-inferiority study, and in a secondary analysis, found to be better that Vancomycin for recurrent CDI).

Evil, evil bug…Oh and those alcohol-based hand rubs don’t work, so PLEASE WASH YOUR HANDS – otherwise you’re responsible for spreading evil…out tomorrow, shall be back Thursday…till then – KEEP READING!!!

Comments