By Brenda L. Mooney

Unadjusted Risks of Recurrence and All-Cause 30-Day Mortality for Patients With Vancomycin Hydrochloride and Metronidazole Hydrochloride Stratified by Disease Severity CohortA, Patients treated with vancomycin and metronidazole had similar risks of recurrence across disease severity cohorts. B, Among patients with severe Clostridium difficile Infection, those treated with vancomycin were less likely to die compared with patients treated with metronidazole.

SALT LAKE CITY — Severe illness caused by the bacteria Clostridium difficile (C. diff) is now the most common hospital-acquired infection in the United States. A new VA-led study suggests, however, that it is not always being treated appropriately.

In a study published online by JAMA Internal Medicine, researchers from the Veterans’ Salt Lake City Health Care System and University of Utah reported that patients with the infection were less likely to die when treated with the antibiotic vancomycin compared to the standard treatment of metronidazole.1

One of the challenges in treating C. diff is the high mortality rate, as high as 26%. In 2011, almost a half-million Americans, mostly 65 or older, developed CDI, and 83,000 of them experienced recurrence of infection within 30 days of completing the standard course of antibiotics, according to the national Centers for Disease Control and Prevention (CDC).

“This is a very real problem that impacts the patients’ quality of life,” explained lead author Vanessa Stevens, PhD, a research assistant professor in the department of internal medicine and an investigator at the IDEAS 2.0 Center at the VA.

While the antibiotics metronidazole or vancomycin are recommended in practice guidelines to treat the condition, metronidazole has tended to be favored over the past few decades because of its lower cost and because of concerns about vancomycin resistance in other hospital-acquired infections.

“For many years, the two antibiotics were considered to be equivalent in their ability to cure C. diff and prevent recurrent disease,” noted Stevens, whose report pointed out that the study guidelines are based on small clinical trials from about three decades ago. “Our work and several other studies show that this isn’t always the case.”

For the current study, researchers sought to determine effectiveness of the two drugs by focusing on the risk of mortality after treatment.

Eligible patients included those with CDI as measured by the CDC laboratory—identified based on a laboratory test result that indicated the presence of C difficile toxin or toxin gene in a stool sample.

For purposes of the research, a severe case of Clostridium difficile infection (CDI) was defined as elevated white blood cell count or serum creatinine within four days of the CDI diagnosis, while a mild to moderate case of CDI was defined as normal white blood cell counts and creatinine levels.

From Jan. 1, 2005, to Dec. 31, 2012, 47,471 patients, mean age of 68.8 and about 96% male, developed CDI at VA facilities, were treated with vancomycin or metronidazole and met criteria for entry into the study

Researchers noted that, during the study period, the number of first treated cases classified as severe decreased from 7,432 (40.3%) in 2005 to 5,138 (30.5%) in 2012. At the same time, the number of patients treated with vancomycin increased from 147 (2.0%) in 2005 to 428 (8.3%) in 2012.

Of the nearly 50,000 patients with first eligible treatment episodes, about 4.4% received vancomycin as initial therapy, more often with mild to moderate infections or CDI of unknown severity and slightly less often among those diagnosed as severe. Those 2,068 patients were matched to 8,069 patients in the metronidazole group for a total of 10,137 patients for the analysis.

With about 35% of those considered severe, lower mortality rates occurred when patients were treated with vancomycin compared to metronidazole—15.3% vs. 9.8%. In fact, researchers calculated that only 25 patients with severe CDI would need to be treated with vancomycin instead of metronidazole to prevent one death.

This difference was largely driven by a 4.5%—15.3% vs. 19.8%—reduction in absolute risk of mortality among patients with severe CDI being treated with vancomycin compared with those treated with metronidazole. No statistically significant difference in the risk of mortality by treatment group was documented among patients with mild to moderate CDI (5.9% for the vancomycin group vs the metronidazole group, (6.9%).

“That is a powerful, positive outcome for our patients’ well-being,” Stevens emphasized, although she added that the study team does not fully understand why the differences in the antibiotic affects mortality rates.

Study results did not show a difference in the rate of the illness recurrence, no matter the antibiotic treatment or illness severity.

Unchanging Treatment

Background information in the study suggested that treatment of infections due to C difficile has undergone minimal changes since the pathogen was first identified as the main cause of pseudomembranous colitis in the late 1970s. While vancomycin was the first antibiotic was found to be effective for the treatment of CDI, metronidazole speedily became the drug of choice for two reasons: concerns about the cost of branded vancomycin tablets and the possible emergence of vancomycin resistance among Enterococcus.

Initially, clinical trial data indicated that metronidazole was noninferior to vancomycin for the treatment of CDI, but a 2007 study from the University of Chicago found that, while metronidazole and vancomycin are equally effective for the treatment of mild C. Diff, vancomycin is superior for treating patients with severe cases.2

At the same time, study authors cited evidence that primary cure rates with metronidazole and vancomycin are lower than previously expected based on early trial data, suggesting that “the consequences of these treatment failures are often overlooked and underestimated.”

The VA study indicated mortality rates of 6.8% for patients with mild to moderate CDI and 19.3% for patients with severe CDI. Study authors suggested that was lower than the 26% reported in a systematic review of 20 published studies because their data mixed low-mortality outpatients (2.6%) with higher-mortality inpatients (14.3%).

“Despite strong evidence and clinical practice guidelines to support vancomycin treatment for severe CDI, it remains an underused treatment option,” study authors emphasized, adding, “Despite the increase in vancomycin use during the study period, half of the patients with severe CDI did not receive vancomycin in 2012. Our results are in accordance with previous findings, which indicate that vancomycin is used in 15% or less of all patients with CDI and that patients with severe CDI are no more likely to receive vancomycin than patients with mild to moderate CDI unless an active antimicrobial stewardship protocol is in place to give oral vancomycin to patients with severe CDI.”

Research limitations included the study’s observational nature and that subjects were primarily male, according to Stevens, who called for more research. “The optimal way to move forward is to do decision analysis that allows us to weigh the pros and cons of the various treatment strategies,” she said.

Study authors also urged that future studies go beyond the question of clinical cure and looked at “important downstream outcomes, such as recurrence and mortality” in selecting an initial therapy. About 16% of patients in the recent study developed recurrent infection


  1. Stevens VW, Nelson RE, Schwab-Daugherty EM, Khader K, et. al. Comparative Effectiveness of Vancomycin and Metronidazole for the Prevention of Recurrence and Death in Patients With Clostridium difficile Infection. JAMA Intern Med. 2017 Feb 6. doi: 10.1001/jamainternmed.2016.9045. [Epub ahead ofprint] PubMed PMID: 28166328.
  2. Zar FA, Bakkanagari SR, Moorthi KM, Davis MB. A comparison of vancomycin and metronidazole for the treatment of Clostridium difficile-associated diarrhea, stratified by disease severity. Clin Infect Dis. 2007 Aug 1;45(3):302-7. PubMed PMID: 17599306.