Infectious Disease Updates
By Carol A. Kemper, MD, FACP
El Feghaly RE, etc. Markers of intestinal inflammation, not bacterial burden, correlate with clinical outcomes in Clostridium difficile infection. CID 2013: 56:1713-21.
While certain factors, such as age, high white blood count, and comorbidities may be used in identifying patients at risk for more severe C. difficile enterocolitis, they are also used to justify the initial choice of antibacterial treatment. For example, current guidelines suggest that patients with mild to moderate disease receive metronidazole, but those patients with higher white blood counts (≥ 15,000 cells/ microL) receive, as initial therapy, orally administered vancomycin. But how valid are these predictors of disease severity?
These authors examined the correlation between 3 different disease severity scores (HINES VA, UPMCI, and CSI), fecal cytokine levels, quantitative fecal C. difficile burden, based on real-time PCR measurements of tcdB DNA, and clinical outcomes, including duration of diarrhea and length of hospital stay. Fecal CXCL-5 mRNA, IL-8 MRNA and quantitative IL-8 protein levels, and fecal lactoferrin concentrations were measured at baseline and then daily (when stools were available). CXCL-5 is a member of the CXC cytokine group, which is 22% homologous to IL-8, binds to the IL-8 receptor, recruits and activates neutrophils, and has been reportedly elevated in patients with inflammatory bowel disease.
A total of 131 patients were enrolled in the study; stools were available in 121. One patient had consistently negative tcdBDNA levels and was excluded from the analysis. Of the remaining 120 patients, 19 (16%) were initially started on oral vancomycin and the rest were treated with metronidazole. Of the latter group, 33 (33%) were switched to oral vancomycin at some point during the treatment course, presumably for lack of response. Diarrhea persisted for 5 or more days in slightly more than half of the patients (52%). Eleven patients (9%) were considered to have severe disease, based on the presence of toxic megacolon, colectomy, admission to ICU or death. Slightly more than half (52%) of the patients were considered immunosuppressed based on previous use of corticosteroids or other immunosuppressants within 90 days. There did not appear to be a difference in presentation between immunosuppressed and immune competent participants.
HINES VA and CSI scores, and leukocytosis (≥ 15,000 cells/mcl) were each significantly associated with disease severity. A more severe HINES VA score was also associated with higher levels of fecal IL-8 mRNA and lactoferrin protein concentrations. A higher UPMC1 score was also associated with higher IL-8 protein concentrations in stool.
Interestingly, fecal C. difficile bacterial burden at presentation did not appear to correlate with any of the severity scores, the frequency of stools, or pain level. But CXCL-5 mRNA and IL-8 mRNA levels at study entry were statistically significantly associated with time to resolution of diarrhea (p = .009 and P = .03, respectively), and elevated levels correlated with diarrhea symptoms throughout the hospital course. As predicted, IL-8 protein levels decreased after starting therapy. Bacterial burden quickly decreased with initiation of therapy, but remained independent of diarrhea symptoms.
It appears that these biological markers are better correlates of disease severity and clinical outcomes (severity and duration of diarrhea) than C. difficile bacterial load. C. difficile toxins cause direct injury to the intestinal endothelium, resulting in a host inflammatory response with cytokine release. Fecal cytokines are therefore more likely reflecting the extent of tissue inflammation and damage a better predictor of disease severity and outcome, and potential treatment failure.
Manhart LE, et al. Standard treatment regimens for nongonococcal urethritis have similar but declining cure rates: A randomized controlled trial. CID 2013:56(7): 934-942.
Declining rates of response to standard treatment regimens for non-gonococcal urethritis (NGU) may prompt changes in accepted treatment strategies. Both single dose azithromycin 1 gram or doxycycline 100 mg twice daily x 7 days are currently accepted first line treatment regimens for patients with NGU, as defined by mucupurulent or purulent urethral discharge or a urethral swab with ≥ 5 white blood cells per HPF in the absence of gonorrhea; or a positive nucleic acid amplification test (NAAT) for Chlamydia trachomatis (CT), Mycoplasma genitalium (MG), Ureaplasma urealyticum biovar 2(UU) or Trichomanas vaginalis (TV). Evidence suggests that rates of response to standard regimens may be declining, especially those infections due to M. genitalium. Previously, azithromycin was consistently more effective than doxycycline against M. genitalium. However, rates of microbiologic cure to azithromycin single dose treatment of M. genitalium in 3 recent U.S. studies ranged from a high of 77% in New Orleans to as low as 40% in Seattle.
These Seattle-based investigators examined the rates of response to standard therapy in a group of men ≥ 16 yrs of age, with NGU, defined as visible urethral discharge or ≥ 5 WBC per HPF in a randomized, double-blind clinical trial. Men were recruited between 2007 and 2011. Urine was tested using NAATs, which are highly sensitive methods for detecting specific pathogens. Clinical cure was assessed at 3 weeks (< 5 WBCs per HPC regardless of the presence of symptoms or absence of discharge) plus microbiologic cure (negative NAAT). Men with either M. genitalium or Ureaplasma spp. who failed to respond to their initial treatment received the alternate treatment, and were re-evaluated at 6 weeks of study.
A total of 606 men were randomized to either azithromycin 1 gram once vs placebo doxycycline (n 304) vs azithromycin placebo plus doxycycline 100 mg bid x 7 days (n = 302). The mean age of participants was 33.7 yrs, slightly more than half had urethral discharge (53%), dysuria(51%) or other urethral complaints (26%). Fifteen of the men were HIV+ (2.5%). NAAT results were positive for CT (23%), UU (24%), and MG ( 13%).
In the intent to treat analyses, 80% of the men receiving azithromycin achieved clinical cure vs 76% for the doxycycline group (p = NS). Interestingly, clinical cure occurred less often in men returning for follow-up "early" within 3 weeks (67%) vs after 3-5 weeks (81%) (p = .004). While clinical cures for all causes were similar between the two treatment groups (83% for azithromycin and 82% for doxycycline), clinical cures in men with MG were significantly lower (63% for azithromycin and 48% for doxycycline). The intent-to-treat analyses for microbiologic cure rates were also similar for both treatment groups. No difference in response to treatment was observed for those few men who were HIV+.
These results suggest that current treatment regimens for NGU in men may not be adequate. Identification of the specific microbiologic cause is useful in guiding therapy clinical testing should be expanded beyond the common protocol testing for CT and GC. Men with TV should be treated with metronidazole, in addition to azithromcyin. In addition, 30% of the men in this survey were diagnosed with MG, for which neither azithromycin nor doxycycline was effective in 37%-52% of individuals. Should they fail, current guidelines recommend switching to the other agent which does not appear to be a very effective treatment strategy. Presently, the authors treat all treatment failures with moxifloxacin 400 mg po x 7 days. (Levofloxacin 500 mg per day for 7 days would be a reasonable alternative). And it may be appropriate to counsel men they should not expect rapid resolution of their clinical symptoms within a week or two of treatment. Based on this data, most men do not have symptomatic resolution for at least 3-5 weeks.
