High Levels of Resistance to Fluoroquinolones among Clostridium difficile Isolates in a Spanish Hospital
- Mar Sánchez-Somolinos,
- Luis Alcalá,
- Teresa Peláez,
- Mercedes Marín,
- Adoración Martín,
- Pilar Catalán, and
- Emilio Bouza
- Clinical Microbiology and Infectious Diseases Department, Hospital General Universitario Gregorio Marañón, Universidad Complutense, Madrid, Spain
- Reprints or correspondence: Dr. Mar Sánchez-Somolinos, Servicio de Microbiología, Hospital General Universitario Gregorio Marañón, Avda. Dr. Esquerdo 46, 28007 Madrid, Spain (msanchezs35{at}yahoo.es).
The appearance of Clostridium difficile NAP1/027 (the epidemic strain) has been blamed for an increase in the incidence and severity of C. difficile infection. Some authors consider quinolone resistance to be a good marker of the epidemic strain. We performed a prospective study to assess C. difficile infection at our institution in Madrid, Spain (Hospital General Universitario Gregorio Marañón, Universidad Complutense). The incidence of C. difficile infection remains stable in our institution, and the epidemic strain has not yet become a problem. However, resistance to quinolones is frequent for several other ribotypes and, therefore, is a poor marker for the NAP1/027 strain.
The incidence and severity of Clostridium difficile–associated diarrhea have been increasing in recent years [1, 2], mainly in hospitals in Canada and the United States, because of the appearance and spread of a characteristic toxin-hyperproducing strain, NAP1/027 [3–5]. Morbidity and mortality resulting from C. difficile–associated diarrhea caused by this epidemic strain are higher than they were in the 1990s, and the disease may even be acquired in the community by previously healthy people, with severe consequences [6]. The ribotype 027 strain is resistant to fluoroquinolones, and some authors consider quinolone resistance to be a good marker of its presence [7]. This strain has been reported in Europe and has caused outbreaks of C. difficile–associated diarrhea in the United Kingdom, The Netherlands, Belgium, Ireland, Germany, and France [8]. Because the presence of the ribotype 027 strain has not yet been reported in Spain, we studied the incidence of C. difficile–associated diarrhea in our institution (Hospital General Universitario Gregorio Marañón, Universidad Complutense; Madrid, Spain) during the past 7 years to assess the clinical features and spectrum of the disease in our hospital and to question the value of quinolone resistance as a surrogate marker for the epidemic strain.
Methods.We determined the evolution of the incidence of C. difficile–associated diarrhea in our hospital from January 2000 through December 2006. To assess the most recent clinical presentation and course of C. difficile–associated diarrhea, we chose a 4-month period (April 2005 through July 2005) in which 1 of the authors (M.S.S.) prospectively attended and followed up patients who developed a first episode of C. difficile–associated diarrhea. All of our patients had a hospital onset of diarrhea. We recorded clinical data using a preestablished protocol. We also evaluated patient hospitalizations that had occurred during the previous year and antibiotics and chemotherapy administered during the previous month, including outpatient medication. Patients were followed up weekly via telephone for up to 2 months after the episode.
Toxin was routinely detected using the cytotoxin test in either stool samples or isolated colonies if the direct test result was negative [9–11]. We routinely cultured unformed stool samples on C. difficile agar (Agar-Clo; bioMérieux) [12–14]. Antimicrobial susceptibility was tested with use of the E-test (AB Biodisk) according to the manufacturer's instructions. In the case of susceptibility to metronidazole, MICs were determined before and after isolates were stored at −70°C, to compare the susceptibility of fresh strains with the susceptibility of frozen strains.
The MIC was defined as the point at which the ellipse intersected the scale at 48 h, although we searched for colonies growing inside the inhibition zone after 5 days and considered these isolates to be heteroresistant. Heteroresistance was defined according to communication C2–2174 from the 2003 Interscience Conference on Antimicrobial Agents and Chemotherapy [15] and communication E–1361 from the 2006 Interscience Conference on Antimicrobial Agents and Chemotherapy [16].
The breakpoints of the different antimicrobial agents used against C. difficile were those recommended by the Clinical and Laboratory Standards Institute for anaerobes, if available. Otherwise, we used the Clinical and Laboratory Standards Institute interpretive categories for Staphylococcus aureus. The breakpoints were as follows: vancomycin, ⩾2 µg/mL; metronidazole, ⩾32 µg/mL; clindamycin, ⩾4 µg/mL; ciprofloxacin, ⩾2 µg/mL; moxifloxacin, ⩾1 µg/mL; levofloxacin, ⩾2 µg/mL; bacitracin, ⩾8 µg/mL; fusidic acid, ⩾4 µg/mL; rifampicin, ⩾2 µg/mL; teicoplanin, ⩾16 µg/mL; and linezolid, ⩾4 µg/mL.
The genetic diversity of the strains was analyzed by PCR ribotyping, with amplification of the 16S and 23S intergenic regions [17]. The primer sequences were 59-GTG CGG CTG GAT CAC CTC CT-39 (16S primer) and 59-CCC TGC ACC CTT AAT AAC TTG ACC-39 (23S primer). Amplification involved 35 cycles of 1 min for denaturation at 94°C, 1 min for primer annealing at 57°C, and 1 min for extension at 72°C. Ribotyping patterns were separated by electrophoresis on MS-8 agarose (3.0%) at 100 V for 3 h and analyzed with ultraviolet light after ethidium bromide staining. Strain patterns were compared with that of the epidemic strain PCR ribotype 027 (provided by Dr. Edward J. Kuijper of the Department of Medical Microbiology, Leids Universitair Medisch Centrum, Leiden, The Netherlands).
The following definitions were used: diarrhea was defined as the passage of ⩾3 unformed stools per day for at least 2 consecutive days; nosocomial diarrhea was defined as diarrhea that developed >48 h after hospitalization; C. difficile infection was defined as diarrhea that was not attributed to any other cause and that was associated with a stool cytotoxin test result positive for C. difficile; C. difficile infection onset was defined as onset of symptoms. Severity was classified as follows: mild (self-limiting C. difficile infection that required no specific therapy), moderate (an episode of C. difficile infection that required treatment with antimicrobial therapy and did not fulfill the criteria of moderate or severe), severe (refractory C. difficile infection or C. difficile infection with sepsis, hypotension, leukocytosis [leukocyte count, >20,000 cells/µL], colonic dilation, or perforation), and fulminant (C. difficile infection associated with paralytic ileum, toxic megacolon, or need for colectomy [18]). First episode was defined as a C. difficile infection episode that occurred during the study period, excluding patients who had experienced an episode in the previous 2 months. Persistent diarrhea was defined as diarrhea that did not disappear after 10 days of treatment. Recurrence was defined as the reappearance of symptoms after treatment of and recovery from diarrhea caused by C. difficile toxins. Recurrence was confirmed by direct cytotoxin test and stool culture. We studied a 2-month period after the first episode to look for relapses.
All data were collected and entered into a Microsoft Access database. Qualitative variables were described as proportions, and quantitative variables were described as means when they were normally distributed or as medians when they were not normally distributed. Proportions were compared using Fisher's exact test with 2 tails. P<.05 was considered to be statistically significant. Data were analyzed using SPSS statistical software, version 15.0, for Windows (SPSS).
Results.The mean incidence of C. difficile infection during the period from 2000 through 2006 was 8.7 cases per 10,000 patient-days. The highest incidence was in the year 2002 (10.6 cases per 10,000 patient-days). From 2004 through 2006, we observed a decrease in incidence. During the last year of the study, the incidence was 7.8 cases per 10,000 patient-days.
We analyzed a series of 111 patients who received consecutive diagnoses of first episodes of C. difficile infection in our institution during a period of 4 months. The departments with the greatest number of patients with C. difficile infection were internal medicine and gastroenterology. Less than 25% of the patients with C. difficile infection were in surgical departments at the time of the infection. The demographic and epidemiologic features of the patients are summarized in table 1.
Table 1
Demographic and epidemiologic features of 111 patients with Clostridium difficile–associated diarrhea.
The main clinical parameters, diagnosis procedures, and treatment strategies are summarized in table 2. Antibacterial agents were administered during the month before the onset of diarrhea in at least 91% of the patients. Quinolones were the most frequently used antibacterial agent (in 49.5% of cases), followed by cephalosporins (29.7%). Overall, 16% of patients had ⩾1 recurrence, and all of them had previously received metronidazole as their first treatment. Colonoscopy was performed for 4 patients, and pseudomembranes were only observed in 1 patient. The overall mortality in this series of patients was 14.4%. Mortality was clearly related to an episode of C. difficile infection in only 1 patient.
Table 2
Clinical manifestations, diagnosis, and outcome of patients with Clostridium difficile–associated diarrhea.
We determined the antimicrobial susceptibility of 101 strains (obtained from 111 patients). Resistance to ciprofloxacin was found in 92.3% of the strains (MIC, 0.5 to >32 µg/mL; geometric mean, 49.7 µg/mL), resistance to levofloxacin was found in 90% (MIC range, 0.5 to >32 µg/mL; geometric mean, 41.2 µg/mL), and resistance to moxifloxacin was found in 73.3% (MIC range, 0.125 to >32 µg/mL; geometric mean, 14.3 µg/mL). Resistance to clindamycin was found in 65% of the strains (MIC range, 0.125 to >256 µg/mL; geometric mean, 49 µg/mL), resistance to rifampicin was found in 55% (MIC range, 0.002 to >32 µg/mL; geometric mean, 1.4 µg/mL), and resistance to bacitracin was found in 100% (MIC range, 8 to >256 µg/mL; geometric mean, 473.5 µg/mL). All isolates were susceptible to vancomycin (MIC range, 0.012–1 µg/mL; geometric mean, 0.75 µg/mL), teicoplanin (MIC range, 0.016–1 µg/mL; geometric mean, 0.15 µg/mL), fusidic acid (MIC range, 0.03–2 µg/mL; geometric mean, 0.61 µg/mL), and linezolid (MIC range, 0.125–1 µg/mL; geometric mean, 0.63 µg/mL).
Of the 101 patients with a representative strain, 73 were treated with metronidazole. When we determined susceptibility to metronidazole of these fresh isolates of C. difficile, we observed 9 strains with heteroresistant subpopulation colonies growing inside the inhibition zone after 5 days of incubation. Of the 9 patients with heteroresistant strains, 4 patients had ⩾1 recurrent episode. Of the 64 patients with fully susceptible strains, only 10 patients had ⩾1 recurrence. Heteroresistance was lost in frozen strains when they were studied immediately after thawing.
PCR ribotyping was performed on the strains collected during the study period to detect the possible appearance of the epidemic ribotype 027 strain. We found a predominant endemic strain in our hospital that was not related to ribotype 027. Recurrent cases were caused by the same strain.
Discussion.In recent years, the incidence of C. difficile infection has remained stable or decreased in our center. Most of our patients had mild or moderate disease, probably reflecting the absence of the hypervirulent epidemic strain. Resistance to quinolones was common among isolates obtained from our patients and was a poor surrogate marker for the presence of the 027 strain. Quinolone resistance has been suggested as a good surrogate marker for the presence of the 027 strain in certain areas [19], but our work and the reports of other authors show that quinolone resistance cannot be used as a surrogate marker for the presence of the hypervirulent strain [20–22].
Resistance to metronidazole has been described by our group [15–16] and others [24, 25]. Resistance is heterogeneous and, if properly determined, is not very uncommon, because it occurred in 12% of the 73 strains tested. Such heteroresistance can be demonstrated using freshly isolated C. difficile and E-test strips after 5 days of incubation. The proportion of relapses in patients treated with metronidazole who had heteroresistant or susceptible strains was 4 (44.4%) of 9 and 10 (15.6%) of 64, respectively. These differences almost reached statistical significance (P =.06). Despite the limitations of the sample, we consider this finding to be potentially relevant and worthy of further investigation in long-term prospective studies.
Our work shows another aspect of C. difficile infection. Recommendations from institutions and countries in which the hypervirulent ribotype 027 strain is widespread cannot be blindly applied to other centers.
Acknowledgments
We thank Dr. Edward J. Kuijper of the Department of Medical Microbiology, Leids Universitair Medisch Centrum, Leiden, The Netherlands, for kindly providing the epidemic strain PCR ribotype 027.
Financial support.Fondo de Investigación sanitaria (PI070869) and INIA FAU2006-0005-C02-00. Red Española de Investigación en Patología Infecciosa C/03/14 (REIPI).
Potential conflicts of interest.All authors: no conflicts.
- Received February 10, 2008.
- Accepted May 26, 2008.
- © 2008 Infectious Diseases Society of America.
references
This Article
- ExtractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
