Treatment of First Recurrences of Clostridium difficile-Associated Disease: Waiting for New Treatment Options

In this issue of the journal, Pepin et al. [1] have reported on the outcomes of treatment of 463 first recurrences of Clostridium difficile-associated disease (CDAD) during the period of 1991 through mid-2005, spanning the period prior to and during the outbreak of infection with the C. difficile ribotype 027/NAP-1 pulsotype strain in Sherbrooke, Quebec. Almost two-thirds of the recurrences occurred in the outbreak period 2003–2005. A major question arising in the infectious diseases community is whether and when to abandon the use of metronidazole as first-line therapy for C. difficile diarrhea, both for initial and recurrent episodes in the face of the changing epidemiology of this reemerging infectious disease [2, 3]. Earlier this past year, Pepin et al. [4] and Musher et al. [5] reported their recent findings of higher relapse rates and delayed or suboptimal responses for metronidazole treatment of initial episodes of CDAD. The implication by extrapolation was that metronidazole should be abandoned in favor of vancomycin. The accompanying editorial by Gerding [6] cautioned that these clinical experiences are still subject to uncontrolled biases, that carefully conducted randomized trials are required to resolve the relative merits or deficiencies of treatments, and that metronidazole should remain the agent of choice for the majority of cases of initial treatment. In this current report, Pepin et al. [1] reexamine the issue of treatment choices for first recurrences. Outcomes focused on (1) the risk of an additional recurrence of CDAD and (2) the development of complicated CDAD (defined as megacolon, perforation, colectomy, shock, and all-cause death within 30 days). The findings demonstrate that the risks of recurrence for metronidazole and vancomycin therapy are similarly high (36.5%–40%) and are unaffected by the treatment of the prior episode, and the risk of recurrence before the outbreak period does not differ from the risk during the outbreak period. The implication is that neither therapy is helpful in restoration of the microbial flora that is severely impaired in relapsing cases of CDAD [7]. Age and duration of hospitalization were found to be risk factors favoring a second recurrence of disease. In the analysis of recurrence, it was concluded that reinfection rather than relapse of CDAD was more likely on the basis of longer duration of hospitalization. In the presence of an outbreak of CDAD in which spores of the outbreak strain become dominant in the hospital, the traditional definitions of endogenous relapse of infection with the same strain or exogenous reinfection with a different strain (or strains) become much less incisive, because this disease perpetuates itself via intestinal and environmental persistence of spores. A simple designation of recurrence seems appropriate in these circumstances. With respect to the equally important (and perhaps more important) end point of development of complicated disease, it was observed that advancing age, a leukocyte count of >20 × 109 cells/L, and renal insufficiency were risk factors favoring a poorer outcome. A complicated course supervened in 12 (7%) of 171 vancomycin-treated episodes, compared with 15 (13%) of 115 metronidazole-treated episodes (P = .12). (According to table 2 in the article by Pepin and colleagues, combined therapy with both vancomycin and metronidazole appeared to result in a higher likelihood of a complicated course; this was also shown in the initial infection study [8], and the complicated course was much more likely associated with the initiation of combination therapy for patients who were severely ill, rather than the development of severe disease as a result of combined therapy.) It was concluded that there was a tendency for vancomycin to be less likely to be accompanied by a complicated course, but the trend did not reach statistical significance. In these same investigators' first report on outcomes of initial treatment [8], using the same definitions of complicated CDAD, 125 (13.1%) of 951 metronidazole recipients developed complicated CDAD, compared with 5 (5.9%) of 85 vancomycin recipients (adjusted OR, 0.2; 95% CI, 0.05–0.8; P < .05). Taken together, combining observations on the treatment of initial and first recurrences, it could be concluded that differences in the rates of less-complicated outcome exist for the 2 therapies. However, several caveats apply: the observations are not based on randomized, controlled studies; it is a single health care center's experience; the health care center had experienced an outbreak of infection with ribotype 027/NAP-1 strain; and the definition of a complicated course including all-cause mortality within 30 days. In elderly patients with multiple underlying comorbidities, all-cause mortality may lead one overestimate the impact of the CDAD infection. In favor of the utility of these single-center observations, it has been shown that the outbreak strain is more widespread than originally appreciated, appearing in multiple states in the United States [2] and in provinces outside of Quebec, Canada (author's unpublished data). Although the results of current double-blind, randomized trials comparing new agents with standard therapy are awaited (initial outcomes are expected in early 2007), how is the clinician to respond to the results of the reports above? Of the 2 end points for therapy, prevention of severe morbidity/mortality should rank higher than recurrence as an outcome variable, with the knowledge that each case of relapse/recurrence runs the risk of developing complications. The first priority is the prompt initiation of therapy. Detection of CDAD is still to be optimized, and although EIAs for detection of C. difficile toxins A/B offer a rapid turnaround time and high specificity, sensitivity is in the order of 55%–89%. Tissue culture detection of cytotoxin B requires a minimum 48 h of turnaround time. Additional laboratory tests, including strain recovery, toxigenic culture, and PCR-based methods, would improve diagnostic precision, but these tests are not widely available to clinical services. Therefore, clinicians must still rely on clinical judgment (including endoscopy findings) to determine whether to and when to initiate therapy. Because no therapeutic modality predictably averts complications, all patients treated for CDAD should be closely monitored daily for adverse outcomes, especially during the first week of therapy. In the absence of severe leukocytosis, renal insufficiency, and advanced age, for patients who appear to have disease of mild-to-moderate severity, metronidazole remains first-line therapy. This likely applies to the majority of patients. For those with risk factors for severe disease or who are deemed to be suboptimal responders to metronidazole, vancomycin is the current practical alternative. Treatment with vancomycin is targeted at attaining a clinical response, with no added benefit over metronidazole in terms of recurrence. At present, there are few attractive choices or strategies for the treatment of relapsing disease. It is not certain whether tapering the dose of vancomycin, which is commonly practiced, is effective for prevention of recurrences. Although it has been demonstrated that vancomycin suppresses the gram-negative anaerobic fecal flora (as well as the expected suppression of infection with most gram-positive organisms) [9], a detailed sequential sampling of microfloral shifts during metronidazole therapy of primary CDAD is not available. Nevertheless, the presumption is that both first-line therapies suppress the normal flora, accounting for or contributing to relapse/recurrences. Limited data on bacitracin and fusidic acid for treatment of initial infection reveal that vancomycin and metronidazole have comparable response and relapse rates [10–12], and neither appears to be a better alternative for recurrent infections. The mechanisms of relapse have not been investigated and do not necessarily involve the same process between agents. In a limited study, teicoplanin was shown to be associated with a lower relapse rate, but this agent is not used in the United States [13]. Of the new therapies that might offer selective suppression of C. difficile while preserving components of the normal flora, tolevamer (a liquid polystyrene preparation that binds and neutralizes C. difficile toxins A/B) [14], tiacumicin B complex (PAR-101; formerly OPT-80) [15], rifaximin [16, 17], and nitazoxanide [18] appear to have potential normal flora sparing effects. Ingestion of oligofructose to enhance the growth of bifedobacteria has recently been shown to reduce recurrent disease [19]. Confirmatory studies should be available to verify these capabilities in the next 1–2 years. This study verifies that metronidazole and vancomycin treatment of first recurrences of CDAD are associated with the same potential for subsequent (second) recurrences, and it revealed a trend toward a lower frequency of complicated outcomes for vancomycin treatment. These findings are of temporary utility while the results of new therapies are awaited. Potential conflicts of interest. T.J.L. has received recent research funding from Optima Pharma (for a clinical trial investigation), Genzyme (for a clinical investigation), ActivBiotics (for a clinical trial), and Bristol-Myers Squibb.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.