By Deborah J. DeWaay, MD, FACP
Assistant Professor, Medical University of South Carolina, Charleston, SC
Dr. DeWaay reports no financial relationships in this field of study
SOURCE: Brown AT, Seifert CF. Effect of Treatment Variation on Outcomes in Patients with Clostridium difficile. Am J Med. 2014; 127(9):865-870
SYNOPSIS: Patients have decreased disease recurrence and mortality when physicians follow the IDSA/Society for Healthcare Epidemiology of America guidelines for the treatment of C difficile infection.
There is an increasing focus on Clostridium difficile (C. diff) given its increasing prevalence, incidence and treatment failures. This gram-positive, anaerobic, cytotoxic bacterium is the most common cause of nosocomial diarrhea in industrialized countries. In 2010, the Infectious Disease Society of America (IDSA) and the Society of Healthcare Epidemiology of America (SHEA) published guidelines for the treatment of Clostridium difficile. Their recommendations for C. diff colitis were as follows:
-
A mild to moderate case, whether initial or first recurrence, should be treated with 10-14 days of Metronidazole 500 mg PO TID.
-
An initial severe case as defined by leukocytosis (WBC ≤ 15,000 cells/¦ÌL) or a serum creatinine ≤ 1.5 times normal for the patient or the second recurrence should be treated with 10-14 days of vancomycin 125 mg PO 4x per day.
-
An initial severe/complicated case, defined by the presence of shock, megacolon or ileus, should be treated with vancomycin 500 mg 4x per day by mouth (or feeding tube) AND metronidazole 500 Q8 hours intravenously.
Complications from a C. diff infection include, but are not limited to, recurrence, toxic megacolon requiring surgery, and death. Authors of this study hypothesized that adherence to the treatment guidelines would decrease the known complications of C. diff.
The authors conducted a retrospective case-control study of patients treated for C. diff infection during a five-month period in 2011. The patients included in the study were at least 18 years old, identified from the International Classification of Diseases-9th Revision with a diagnosis code of “intestinal infection due to C.diff (008.45) and received treatment for their infection. The electronic health records used were from a single, tertiary care teaching hospital in Texas. The patient information was collected via chart review and placed in a standardized form. Patients were independently classified as mild/moderate, severe, or severe/complicated based on IDSA/SHEA guidelines by two authors, who subsequently concurred on the classifications. After patients were classified, the authors reviewed the charts and divided the patients into two groups: guideline-concordant and guideline-discordant. The authors then compared the complication rate of recurrence, surgery, toxic megacolon or death between the two groups. Any patient who received treatment that was more aggressive than the guidelines was put in the “guideline-concordant” group. The Shapiro-Wilk test for normality, Mann-Whitney U and Kruskal-Wallis tests were used to analyze the continuous data. Nominal data was analyzed using Pearson chi-squared and Fisher’s exact tests. Complications, recurrence and mortality were analyzed using multiple logistic regression analysis.
180 out of 207 patients were included after the exclusion criteria were applied.
64.4% of all patients had antibiotic exposure within 8 weeks of developing C. diff.
55.6% of patients had taken a proton-pump inhibitor within 8 weeks of developing C.diff. The guideline-concordant and guideline-discordant groups had similar demographics except the guideline-concordant group was older (67 ¡À 22.7 years vs. 58 ¡À 24 years; P = .008). The guideline-concordant group was less likely to have a recurrence when compared to the discordant group (14% vs. 35.6%; P=.0007). The mortality rate was also less in the concordant group versus the discordant group (5.4% vs. 21.8%; P=.0012) leading to a decreased overall complication rate in the concordant group versus the discordant group (17.2% vs. 53.6%; P < .0001). The clinical cure rate was increased in the guideline-concordant group compared to the discordant group (93.5% vs. 71.3%; P<.0001). Patients with mild/moderate C. diff infection were more likely to have received guideline-concordant therapy as compared to patients who had a severe or severe/complicated infection (82.1% vs. 35.3%; P<.0001). The patients with severely complicated infections were older (67 vs. 59.5 vs 61; P = .0083) and had higher complication (54.1% vs. 38.2% vs. 22.4%; P = .0004) and mortality rates (26.2% vs. 20.6% vs. 1.2%; P<.0001) compared to the severe and mild/moderate groups.
Recurrence of disease was 72.5% less likely in patients who received guideline-based therapy (OR 0.2747; 95% CI 0.1308-0.5769; P = .0004). The most common pattern of discordant therapy among those with recurrent mild-moderate C. diff was treating with metronidazole monotherapy rather than vancomycin in a tapered or pulsed schedule. Only 19.7% of severe/complicated patients were treated per the guidelines. Most often there was a failure to add IV metronidazole to PO vancomycin or the patient was treated with PO metronidazole alone.
COMMENTARY
This study demonstrates that there may be clinical benefit to guideline-concordant therapy. The patients who received guideline-based therapy had less recurrence and decreased mortality. Unfortunately, in this study, only slightly over half of the patients received treatment per the guidelines. The authors make the point that only one-half of Americans receive guideline-based therapy when they see a doctor in general. Oral vancomycin was available at the hospital this study was performed, so resources were not limited. They suggest that the main problem is a lack of familiarity by physicians with the guidelines. With the rising infection rates of C. diff, preventing hospitalizations will be challenging. Preventing recurrence of disease when possible will be essential to minimizing hospitalizations, and following the guidelines for treatment will be an important way to prevent recurrence and also mortality.
There are several limitations to this study. First, it is a retrospective, case-control study so causality cannot be assessed. Second, authors could not always be sure that the markers for severity of disease were secondary to C. diff. Third, this was a single-center study so generalizability is questionable. Fourth, there was no follow-up as an outpatient, so patients could potentially have had uncounted outpatient treatment failures.
C. diff infection is a common illness that hospitalists must understand how to diagnose and treat in their patients. I believe this study offers several key lessons that can be applied to a practice in hospital medicine. First, as C. diff causes increasing illness in our patients, physicians must be up to date on how to treat it according to the guidelines. It is not surprising that the mild/moderate group was more likely to have guideline-concordant therapy, since the treatment for this group is the traditional treatment of C. diff. In patients with more severe disease, guidelines were not followed as often and outcomes were worse. Second, it is a reminder of the importance of being up to date in the literature in general and the severe consequences that can occur if the patients are not given the most evidence-based treatments possible. Although these are very simple and obvious points, I believe their simplicity is what makes them profound. Physicians, in general, are trying to give the best care possible, yet we are not successful in doing so a significant portion of the time. The quality improvement movement will hopefully help change the course of this reality.
