- Email: [email protected]
Clostridium difficile recurrences in Stockholm
Accepted Manuscript Clostridium difficile recurrences in Stockholm Staffan Sandell, Mamun-Ur Rashid, Christina Jorup-Rönström, Kristina Ellström, Carl Erik Nord, Andrej Weintraub PII:
S1075-9964(16)30003-8
DOI:
10.1016/j.anaerobe.2016.01.005
Reference:
YANAE 1531
To appear in:
Anaerobe
Received Date: 8 September 2015 Revised Date:
11 January 2016
Accepted Date: 19 January 2016
Please cite this article as: Sandell S, Rashid M-U, Jorup-Rönström C, Ellström K, Nord CE, Weintraub A, Clostridium difficile recurrences in Stockholm, Anaerobe (2016), doi: 10.1016/j.anaerobe.2016.01.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Original article
RI PT
Clostridium difficile recurrences in Stockholm
Staffan Sandella, Mamun-Ur Rashidb, Christina Jorup-Rönströma, Kristina
Karolinska Institutet, Department of Medicine, Division of Infectious Diseases,
M AN U
a
SC
Ellströma, Carl Erik Nordb*, Andrej Weintraubb
Södersjukhuset, SE-118 83 Stockholm, Sweden b
Karolinska Institutet, Department of Laboratory Medicine, Division of Clinical
TE D
Microbiology, Karolinska University Hospital Huddinge, SE-141 86 Stockholm, Sweden
EP
* Corresponding author. Tel.: +46 8 585 87838; fax: +46 8 585 87933.
AC C
E-mail address: [email protected] (C.E. Nord)
ACCEPTED MANUSCRIPT ABSTRACT Sixty-eight hospital-admitted patients with a first episode of Clostridium difficile infection (CDI) were included and followed up during 1 year. Faeces samples were collected at 1, 2, 6 and 12 months after inclusion and analyzed for the presence of C. difficile toxin B, genes for toxin A, toxin B, binary toxin
RI PT
and TcdC deletion by PCR. All strains were also PCR-ribotyped and the MICs of the isolates were determined against eight antimicrobial agents. In 68 patients initially included, antibiotics, clinical signs and co-morbidities were analyzed and 56 were evaluable for recurrences. The mean number of different antibiotics given during 3 months prior to inclusion was 2.6 (range 0-6). Six patients had not
SC
received any antibiotics and three of them had diagnosed inflammatory bowel disease. Thirty-two patients (57%) had either a microbiological or clinical recurrence, 16 of whom had clinical recurrences
M AN U
that were confirmed microbiologically (13, 23%) or unconfirmed by culture (3, 5%). Twenty-nine patients were positive in at least one of the follow-up tests, 16 had the same ribotype in follow-up tests, i.e. relapse, and 13 a different ribotype, i.e., reinfection. Most common ribotypes were 078/126, 020, 023, 026, 014/077, 001 and 005. No strain of ribotype 027 was found. Strains ribotype 078/126
TE D
and 023 were positive for binary toxin and were the strains most prone to cause recurrence. All strains were sensitive to vancomycin and metronidazole. Patients with recurrences were significantly older (p=0.02) and all patients had a high burden of comorbidities, which could explain the high fatality
AC C
Keywords:
EP
rate, 26 (38%) patients died during the 1-year follow-up.
Antibiotic treatment
Clostridium difficile infection Clinical recurrence
Microbiological recurrence Ribotyping Minimum inhibitory concentration
2
ACCEPTED MANUSCRIPT 1. Introduction Clostridium difficile infection (CDI) is a most common nosocomial diarrheal disease. A high frequency of recurrences (20-30%), either as relapses or reinfections, plays a major role in CDI [1-3]. Many studies have been performed to find out which factors are associated with
RI PT
one or more recurrences. Treatment with broad-spectrum antibiotics, proton pump inhibitors, old age, severe underlying illness and hospitalization are well-known factors associated with recurrences [4-6]. Leukocytosis, hypoalbuminemia and raised creatinine levels are also
SC
associated with an increased frequency of recurrent infection [4-6].
The aims of the prospective study were to compare relapses (same C. difficile ribotype)
M AN U
with reinfections (different C. difficile ribotypes) in relation to clinical signs, prior antibiotic use, and to find out whether certain ribotypes resulted in more recurrent infections during 1 year. The in-vitro activity of cadazolid, fidaxomicin, metronidazole, vancomycin, linezolid, tigecycline, moxifloxacin and clindamycin by determining the minimum inhibitory
TE D
concentrations (MICs) against the C. difficile isolates was also determined.
2. Materials and methods
EP
2.1 Patients and survey methods
The study was conducted at Södersjukhuset, Stockholm, Sweden. In-patients during the
AC C
period 1st May 2009 to 31st July 2011 with a primary CDI were consecutively recruited. The patients were followed during 1 year with stool sampling for C. difficile laboratory diagnoses at 1, 2, 6 and 12 months after the inclusion. C. difficile infection (CDI) was defined as a positive toxin B test and/or growth of toxigenic C. difficile combined with symptoms of gastrointestinal infection. Microbiological recurrence was defined as a C. difficile positive sample during the follow-up period. A clinical recurrence was defined either as clinical symptoms or prescribed antibiotics for treatment of CDI during the follow-up period.
3
ACCEPTED MANUSCRIPT Inclusion criteria were: (i) age over 18 years, (ii) no known CDI during the preceding 3year period, (iii) positive laboratory test, (iv) diarrhea (at least 3 loose stools/day or one bloody loose stool), and (v) life expectancy estimated to be at least 3 months after the primary CDI episode.
RI PT
At inclusion, medical histories, laboratory results and prescription data of the patients were recorded. Medical records before the inclusion as well as during the follow-up period were requested from different hospitals, nursing homes and primary care centers for the
SC
analysis.
A nurse at the hospital had a telephone contact with each patient at the time for collecting
M AN U
fecal sample and send the stool sample container. The samples were submitted to an outpatient clinic and sent to the laboratory for diagnosis. In addition, the nurse asked the patient for any episodes of loose stool incidences between the sampling occasions. The study was approved by the Ethics Committee at Karolinska Institutet, Stockholm
TE D
(EPN 2009/161-31) and all the patients signed an informed consent.
2.2 Collection and typing of strains
EP
All fecal samples were cultured for C. difficile under anaerobic conditions. The isolates were identified by characteristic colony morphology, typical smell and Gram-staining. The
AC C
isolates were then frozen at -70°C until further analyses [7].
2.3 Determination of toxins, detection of toxin gene and ribotyping The production of toxin B was determined by the cell cytotoxicity neutralization assay. The gene for toxin A was detected by conventional PCR [8]. The genes for toxin B, the binary toxin and TcdC deletion were detected by real time PCR using the GeneXpert® System
4
ACCEPTED MANUSCRIPT (Cepheid, Sunnyvale, CA, USA) assay [9]. A sample was considered positive if the CCNA assay was positive and/or the sample was positive for toxigenic culture. PCR ribotyping was performed as described previously [7, 10].
RI PT
2.4 Antimicrobial susceptibility
The antimicrobial susceptibility of the C. difficile strains was determined according to the Clinical and Laboratory Standards Institute (CLSI) guidelines by the agar dilution method
SC
using Bacteroides fragilis ATCC 25285 and Clostridium difficile ATCC 700057 as reference strains [11]. Vancomycin and metronidazole were purchased from Sigma-Aldrich Co. (Saint
M AN U
Louis, MO, USA), clindamycin was obtained from Applichem GmbH (Darmstadt, Germany), moxifloxacin was supplied by Bayer Health Care Pharmaceuticals (Berlin, Germany), fidaxomicin was provided by Optimer Pharmaceuticals, Inc. (San Diego, CA, USA), linezolid and tigecycline were provided by Pfizer Inc. (New York, NY, USA), and cadazolid was from
TE D
Actelion Pharmaceuticals Ltd. (Allschwil, Switzerland). All antimicrobial agents were dissolved and diluted according to the companies' instructions. MIC was defined as the lowest concentration of the drug that inhibited growth
EP
completely. MIC50/90 corresponds to the concentrations that inhibit the growth of 50% and 90% of the strains tested, respectively. For the calculation of resistant isolates against
AC C
vancomycin, metronidazole, tigecycline, and moxifloxacin, breakpoints established by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) were used [12]. For clindamycin, the CLSI breakpoint was used [11]. For ciprofloxacin and linezolid, no breakpoints are available in the EUCAST and CLSI documents and therefore ciprofloxacin and linezolid breakpoints (≥8 and >4 mg/l respectively) were used based on literature recommendations [13, 14].
5
ACCEPTED MANUSCRIPT 2.5 Statistics PCR ribotype profiles from clinical isolates were compared to those profiles which define the database by maximum matching with Pearson correlation. IBM SPSS Statistics 22 (Armonk, NY, USA) software was used to calculate the percentiles 50 and 90 of the MIC
in comparative calculations by Microsoft Word 2010.
SC
3. Results
RI PT
results. Descriptive statistics were used to summarize the results and student t-test was applied
3.1 Demographic data
M AN U
During the study period, stool samples from 2150 patients at Södersjukhuset were submitted to the laboratory for C. difficile analysis, and 157 samples were positive. Eightynine (57%) patients were excluded since they did not meet the inclusion criteria. Sixty-eight (43%) in-patients between the ages of 19 and 98 years, 35 men and 33 women, with a first
TE D
episode of CDI were included. Twelve patients (18%) did not submit any follow-up sample because of death or terminal disease (7 patients), severe but not deadly disease (1 patient) and lost to follow-up (4 patients). Fifty-six patients (82%) (29 men, 27 women, median age 74
EP
years, range 48-98 years) submitted at least one follow-up sample and were evaluable for recurrences. Figure 1 shows the demographic data of the patients. The patients were treated at
AC C
departments of surgery (10), internal medicine (27), infectious diseases (15), intensive care (3), and orthopedic surgery (1). As shown in Table 1, recurrent patients were significantly older than cured (p=0.02). Among recurrent patients, women were significantly older than men (p=0.05).
3.2 Clinical data
6
ACCEPTED MANUSCRIPT Sixty-three (93%) of 68 patients suffered from at least one chronic comorbidity, and cardiovascular diseases (ischemic heart disease, heart failure, hypertension, cerebrovascular disease) were the most frequent clinical diagnosis in 31 patients. Other conditions among the patients were: diabetes mellitus (12 patients), active malignancy (11 patients), chronic
RI PT
infections (9 patients), chronic liver disease (8 patients), renal disease (10 patients), and 3 patients with end-stage renal failure. Alcoholism (6 patients), chronic lung disease (6
patients), inflammatory bowel disease (5 patients), cognitive impairment (4 patients), and
SC
rheumatic/autoimmune disease (4 patients).
Nineteen (28%) patients had ongoing acute non-CDI infections at the time of primary
M AN U
CDI: septicemia (10 patients), endocarditis (2 patients), pneumonia (1 patient), appendicitis (1 patient), mammary abscess (1 patient), abdominal infection (1 patient), cholecystitis (1 patient), pancreatitis (1 patient), and intestinal tuberculosis (1 patient). Of the remaining 49 (72%) patients without concomitant acute infection, 21 had fever at
TE D
the time of primary infection. Nineteen had abdominal pain. Nine patients had vomiting and 5 had stools containing blood. Median duration of symptoms with diarrhea at the time of inclusion was 4 days (range 1-31).
EP
Twenty-four (35%) of 68 patients died during the 1-year follow-up, median age of 81 years at the time of death (range 59-97 years), 13 were men (median age 74 years), and 11
AC C
women (median age 86 years). Nine of the originally included patients died within one month after inclusion. Of the 15 patients who died within the 1-year follow-up and submitted more than one sample, 9 had at least one positive sample. In addition, 4 of the 9 patients with microbiological recurrence were positive for a binary toxin strain. The mean time from inclusion until death was 3.3 months. Causes of death were cardiovascular events (6 patients), infections other than CDI (3 patients), malignancy (2 patients) and pancreatitis (1 patient). In 14 patients, the direct cause of death was unknown, and of these, 3 patients had symptoms
7
ACCEPTED MANUSCRIPT implying ongoing CDI, and 1 died while having colitis. All patients that died within 1 year after inclusion had at least one comorbidity.
3.3 Antibiotics
RI PT
Antibiotics had been prescribed to 62 (91%) of 68 patients during 3 months prior to
inclusion. Six (9%) patients had not been exposed to antibiotics during these 3 months, and of them, 3 had a diagnosis of inflammatory bowel disease (IBD). Mean number of antimicrobial
SC
agents prescribed were 2.6 agents/patient during the 3 months prior to inclusion. Five patients had received at least 6 different antibiotics. Most common antibiotics were cephalosporins,
M AN U
isoxazolyl-penicillins and clindamycin (Table 2).
In 67 (99%) patients, treatment for CDI was given, metronidazole in 53 patients, vancomycin in 1 patient, and a combination of metronidazole and vancomycin in 13 patients. Mean duration of treatment was 13.7 days. Severe colitis occurred in 4 patients, 2 of which
TE D
underwent colectomy.
3.4 Chemical laboratory findings at inclusion in relation to recurrences
EP
Values of C-reactive protein, leucocytes and albumin are shown in Table 3. The denominator (displayed within parenthesis) varies between the categories depending on the
AC C
number of available results.
3.5 Microbiological findings Two hundred and thirteen (average 3.8 samples/patient) fecal samples were collected from the 56 patients. One hundred and twelve samples (53%) were positive. One hundred and fifteen C. difficile strains were isolated from the samples and 107 were toxigenic.
8
ACCEPTED MANUSCRIPT 3.6 Toxin gene detection and PCR-ribotyping All toxigenic isolates were positive for Toxin B by the CCNA assay and for the Toxin A and Toxin B genes by PCR. In addition, 24 isolates were positive for the binary toxin gene (15 isolates of PCR-ribotype 023 and 9 isolates of PCR-ribotype 078/126). Of the 107
RI PT
toxigenic C. difficile isolates, 99 were PCR-ribotyped (8 isolates could not be recovered from the frozen stock). Twenty-six different PCR-ribotypes were identified and the most common ribotypes were: 078/126 and 020 (15 strains each); 023 (9 strains); 026 and 014/077 (7 strains
SC
each); 001, 005 (5 strains each). C. difficile 027 was not detected in any of the samples during
3.7 Antimicrobial susceptibility
M AN U
the study period.
The antimicrobial susceptibility pattern of 81 of the 107 (26 isolates could not be recovered from the frozen stock) toxigenic C. difficile strains is shown in Table 4. All strains
TE D
were sensitive to cadazolid, fidaxomicin, vancomycin, metronidazole and tigecycline with MIC90 values of 0.5 mg/l, 0.125 mg/l, 1 mg/l, 1 mg/l, and 0.125 mg/l, respectively. Two strains were resistant (8 mg/l) to linezolid, 10 strains were resistant (8-32 mg/l) to
EP
moxifloxacin and 59 strains were resistant (8-256 mg/l) to clindamycin.
AC C
3.8 Recurrences and ribotypes Thirteen (23%) of 56 patients had both a clinical and laboratory recurrence. The median age was 85 years. Nine of the patients had the recurrence within 1 month after the primary CDI. Four of these patients had a second recurrence 4-8 weeks after the first recurrence. Sixteen (29%) of 56 patients had laboratory verified recurrence only, median age 75 years.
9
ACCEPTED MANUSCRIPT Three patients had clinical recurrence only (median age 76 years) and did not provide a sample for laboratory diagnosis at the time of clinical symptoms. However, they were successfully treated for CDI. In addition, one patient with a clinical recurrence was also positive for the toxin in the CCNA assay but culture negative.
RI PT
Twenty-four (43%) of 56 patients did not have any clinical or laboratory recurrence (median age 72 years).
A toxigenic C. difficile strain was isolated from 53 (78%) of 68 patients at inclusion.
SC
None of the toxigenic C. difficile isolates belonged to the 027 ribotype. Thus, of the 29
patients positive for toxigenic C. difficile in any of the follow-up samples, 16 had the same
M AN U
ribotype, i.e., relapsed, and 13 patients had changed ribotype, i.e., were re-infected. Twentysix patients with microbiological recurrence (with or without clinical recurrence) had continuous positive laboratory diagnosis whereas 7 had one initial negative test before recurring.
TE D
In total, 32 (57%) of the evaluable patients had either a clinical or laboratory recurrence or both of CDI (Figure 1). Of the 29 patients with at least one C. difficile positive stool sample during the follow-up period, 13 (45%) patients had also a clinical manifestation of CDI.
EP
Proton pump inhibitors were used by 23 (41%) of the 56 patients. Among the patients with PPI, 15 (65%) had relapsed, whereas among patients without PPI, 18 (54%) patients had
AC C
relapsed.
4. Discussion
In this prospective study of 68 included patients, 56 patients were followed with a first CDI during 1 year concerning new clinical symptoms and presence of toxigenic C. difficile in the fecal samples. Twelve patients were excluded concerning recurrences due to death or severe disease. The median age of the 68 patients was 74 years and the burden of
10
ACCEPTED MANUSCRIPT comorbidities was high, two factors that well reflect the CDI patient group in general and partly explain the high proportion of recurrences [6, 15-19]. Younger patients were observed in the cured group compared to recurring patients (p=0.02). Recurring women were significantly older than recurring men (p=0.05). A high number of the included patients died
RI PT
during the 1-year follow-up period, 26 patients of the originally included 68 patients. Since autopsy is rarely performed on elderly patients in Sweden, a definite cause of death could not be made in all but one patient. However, the high fatality rate among the included patients
patients, diarrhea was ongoing at the time of death.
SC
could be attributed to the high age and high burden of comorbidities. In only 3 of 26 deceased
M AN U
PCR-ribotyping showed that of the 29 patients positive for toxigenic C. difficile in the follow-up samples, 16 patients had the same ribotype i.e. relapsed and 13 patients changed the ribotype, i.e. were re-infected. None of the toxigenic C. difficile isolates belonged to the 027 ribotype. However, in the present study among relapsing patients, ribotypes 023 (n=15) and
TE D
078/126 (n=9) were dominating, and were found to be binary toxin producing. The role of the binary toxin as a virulence factor has until now not been thoroughly investigated [20]. In addition, the role of binary toxin in recurrences is still unclear. Stewart et al showed that the
EP
binary toxin significantly associated with a recurrent CDI [21]. However, in a recent study from Spain it has been shown that in a non-027 setting, the presence of binary toxin had no
AC C
role in recurrent CDI [22]. The findings in our, although limited, study show that the presence of binary toxin seems to be linked to an increased risk of recurrence. Decreased sensitivity against metronidazole and vancomycin, the drugs generally used for treatment, was not seen in any of the strains. Resistance should thus not have had any impact on the recurrences. Only 6 of 68 patients had not received any antibiotics during 3 months prior to inclusion, i.e., the risk for community-acquired infection seems low, contradictory to other authors. In
11
ACCEPTED MANUSCRIPT addition, 3 of the patients had an IBD diagnosis and it is well known that these patients are more prone to acquire CDI [23, 24]. The antibiotics given prior to the CDI are well known, i.e., cephalosporins, clindamycin and ciprofloxacin, to cause CDI. Isoxazolyl-penicillins were also common, and patients treated with these antibiotics often received other antibiotics as
RI PT
well, and long duration of therapy against different infections might have played a role.
In the present study, antibiotic use after the primary CDI did not increase the risk of
recurrence, which is contradictory to previous studies [19, 25, 26]. This may be due to a long-
to exceptionally high age and frequent comorbidities.
SC
lasting ecological change which may have been more pronounced than in previous studies due
have been published worldwide.
M AN U
During the last years (2012-2015), several papers covering recurrent CDI and risk factors
In a Japanese investigation [27], 20 patients with CDI recurrences were analyzed by PCR ribotypes, cytotoxicity and antimicrobial susceptibility. Sixteen patients were relapse cases
TE D
and four patients were reinfection cases. There was no association between PCR ribotypes and incidence of recurrences. All C. difficile strains were susceptible to vancomycin and metronidazole but resistance against clindamycin was found in 87.7% of the isolates.
EP
Taori et al. [28] studied 56 patients with recurrent CDI in Scotland for one year. Ribotype 001 was the most common type and no ribotype 027 was detected. High leucocyte count and
AC C
low albumin were associated with mortality. No association with proton-pump inhibitors and with vancomycin-resistant enterococci were found. Recently Hikone et al. [29] published another report on recurrent CDI in 14 patients at a Japanese university hospital. The recurrent patients had more underlying malignant diseases and more intensive care unit hospitalization. A systemic review of risk factors for recurrence of CDI has been published by Chakra et al. [30]. Older age, use of antibiotics after diagnosis, use of proton-pump inhibitors,
12
ACCEPTED MANUSCRIPT hypoalbuminemia, leukocytosis, renal failure and infection with ribotype 027 strains were risk factors for recurrent CDI. Deshpande et al. [31] confirmed the findings of Chakra et al [30] concerning risk factors for recurrent CDI. The most frequent risk factors were age > 65 years, additional antibiotics
RI PT
during follow up, proton-pump inhibitors, renal insufficiency and quinolone therapy.
Abdelfatah et al. [32] also studied risk factors for developing recurrence of CDI. They concluded that most recurrences occurred in patients with comorbidities such as chronic
SC
kidney disease. Use of proton-pump inhibitors was another significant risk factor.
When comparing the laboratory and clinical data on CDI recurrences and risk factors,
M AN U
there are similar findings in most publications, but some differences are found. These findings
AC C
EP
TE D
may be explained by the epidemiological situations in Europe, America and Asia.
13
ACCEPTED MANUSCRIPT References [] [1]
Vardakas KZ, Polyzos KA, Patouni K, Rafailidis PI, Samonis G, Falagas ME.
Treatment failure and recurrence of Clostridium difficile infection following treatment with
RI PT
vancomycin or metronidazole: a systematic review of the evidence. Int J Antimicrob Agents 2012;40:1-8. [2]
Eyre DW, Walker AS, Wyllie D, Dingle KE, Griffiths D, Finney J, et al. Predictors of
SC
first recurrence of Clostridium difficile infection: implications for initial management. Clin
[3]
M AN U
Infect Dis 2012;55 Suppl 2:S77-87.
Louie TJ. Treatment of first recurrences of Clostridium difficile-associated disease:
waiting for new treatment options. Clin Infect Dis 2006;42:765-7. [4]
DuPont HL, Garey K, Caeiro JP, Jiang ZD. New advances in Clostridium difficile
2008;21:500-7.
Kelly CP, Kyne L. The host immune response to Clostridium difficile. J Med
EP
[5]
TE D
infection: changing epidemiology, diagnosis, treatment and control. Curr Opin Infect Dis
[6]
AC C
Microbiol 2011;60:1070-9.
Louie TJ, Miller MA, Crook DW, Lentnek A, Bernard L, High KP, et al. Effect of age
on treatment outcomes in Clostridium difficile infection. J Am Geriatr Soc 2013;61:222-30. [7]
Rashid MU, Lozano HM, Weintraub A, Nord CE. In vitro activity of cadazolid against
Clostridium difficile strains isolated from primary and recurrent infections in Stockholm, Sweden. Anaerobe 2013;20:32-5.
14
ACCEPTED MANUSCRIPT [8]
Kato H, Kato N, Watanabe K, Iwai N, Nakamura H, Yamamoto T, et al. Identification
of toxin A-negative, toxin B-positive Clostridium difficile by PCR. J Clin Microbiol 1998;36:2178-82. Huang H, Weintraub A, Fang H, Nord CE. Comparison of a commercial multiplex
RI PT
[9]
real-time PCR to the cell cytotoxicity neutralization assay for diagnosis of Clostridium difficile infections. J Clin Microbiol 2009;47:3729-31.
Stubbs SL, Brazier JS, O'Neill GL, Duerden BI. PCR targeted to the 16S-23S rRNA
SC
[10]
gene intergenic spacer region of Clostridium difficile and construction of a library consisting
[11]
M AN U
of 116 different PCR ribotypes. J Clin Microbiol 1999;37:461-3.
Clinical and Laboratory Standards Institute. Methods for antimicrobial susceptibility
testing of anaerobic bacteria; Approved standard. 8th ed. Document M11-A8. Wayne, PA:
[12]
TE D
CLSI; 2012.
The European Committee on Antimicrobial Susceptibility Testing. Breakpoint tables
for interpretation of MICs and zone diameters. Version 3.1. Keessen EC, Hensgens MP, Spigaglia P, Barbanti F, Sanders IM, Kuijper EJ, et al.
EP
[13]
AC C
Antimicrobial susceptibility profiles of human and piglet Clostridium difficile PCR-ribotype 078. Antimicrob Resist Infect Control 2013;2:14. [14]
Pelaez T, Alcala L, Blanco JL, Alvarez-Perez S, Marin M, Martin-Lopez A, et al.
Characterization of swine isolates of Clostridium difficile in Spain: a potential source of epidemic multidrug resistant strains? Anaerobe 2013;22:45-9. [15]
Stanley JD, Bartlett JG, Dart BWt, Ashcraft JH. Clostridium difficile infection. Curr
Probl Surg 2013;50:302-37.
15
ACCEPTED MANUSCRIPT [16]
Samie AA, Traub M, Bachmann K, Kopischke K, Theilmann L. Risk factors for
recurrence of Clostridium difficile-associated diarrhoea. Hepatogastroenterology 2013;60:1351-4. Jones AM, Kuijper EJ, Wilcox MH. Clostridium difficile: a European perspective. J
RI PT
[17]
Infect 2013;66:115-28. [18]
Wilcox MH, Shetty N, Fawley WN, Shemko M, Coen P, Birtles A, et al. Changing
SC
epidemiology of Clostridium difficile infection following the introduction of a national
[19]
M AN U
ribotyping-based surveillance scheme in England. Clin Infect Dis 2012;55:1056-63. Kim J, Shaklee JF, Smathers S, Prasad P, Asti L, Zoltanski J, et al. Risk factors and
outcomes associated with severe Clostridium difficile infection in children. Pediatr Infect Dis J 2012;31:134-8.
Gerding DN, Johnson S, Rupnik M, Aktories K. Clostridium difficile binary toxin
TE D
[20]
CDT: mechanism, epidemiology, and potential clinical importance. Gut Microbes 2014;5:1527.
Stewart DB, Berg A, Hegarty J. Predicting recurrence of C. difficile colitis using
EP
[21]
[22]
AC C
bacterial virulence factors: binary toxin is the key. J Gastrointest Surg 2013;17:118-24. Reigadas E, Alcala L, Marin M, Martin A, Iglesias C, Bouza E. Role of binary toxin in
the outcome of Clostridium difficile infection in a non-027 ribotype setting. Epidemiol Infect 2015;1-6. [23]
Nitzan O, Elias M, Chazan B, Raz R, Saliba W. Clostridium difficile and
inflammatory bowel disease: role in pathogenesis and implications in treatment. World J Gastroenterol 2013;19:7577-85.
16
ACCEPTED MANUSCRIPT [24]
Sha S, Xu B, Wang X, Zhang Y, Wang H, Kong X, et al. The biodiversity and
composition of the dominant fecal microbiota in patients with inflammatory bowel disease. Diagn Microbiol Infect Dis 2013;75:245-51. Johnson S. Recurrent Clostridium difficile infection: a review of risk factors,
treatments, and outcomes. J Infect 2009;58:403-10. [26]
RI PT
[25]
Garey KW, Dao-Tran TK, Jiang ZD, Price MP, Gentry LO, Dupont HL. A clinical
[27]
M AN U
antibiotics. J Hosp Infect 2008;70:142-7.
SC
risk index for Clostridium difficile infection in hospitalised patients receiving broad-spectrum
Oka K, Osaki T, Hanawa T, Kurata S, Okazaki M, Manzoku T, et al. Molecular and
microbiological characterization of Clostridium difficile isolates from single, relapse, and reinfection cases. J Clin Microbiol 2012;50:915-21.
Taori SK, Wroe A, Poxton IR. Clostridium difficile infections in South East Scotland:
TE D
[28]
mortality and recurrence in a region without PCR ribotype 027. J Med Microbiol 2013;62:1468-77.
Hikone M, Ainoda Y, Tago S, Fujita T, Hirai Y, Takeuchi K, et al. Risk factors for
EP
[29]
AC C
recurrent hospital-acquired Clostridium difficile infection in a Japanese university hospital. Clin Exp Gastroenterol 2015;8:191-6. [30]
Abou Chakra CN, Pepin J, Sirard S, Valiquette L. Risk factors for recurrence,
complications and mortality in Clostridium difficile infection: a systematic review. PloS one 2014;9:e98400.
17
ACCEPTED MANUSCRIPT [31]
Deshpande A, Pasupuleti V, Thota P, Pant C, Rolston DD, Hernandez AV, et al. Risk
factors for recurrent Clostridium difficile infection: a systematic review and meta-analysis. Infect Control Hosp Epidemiol 2015;36:452-60. Abdelfatah M, Nayfe R, Nijim A, Enriquez K, Ali E, Watkins RR, et al. Factors
RI PT
[32]
Predicting recurrence of Clostridium difficile infection (CDI) in hospitalized patients:
AC C
EP
TE D
M AN U
SC
Retrospective study of more than 2000 patients. J Investig Med 2015;63:747-51
18
ACCEPTED MANUSCRIPT
Table 1
All patients
Women No. of Range
Age
Range patients
All
74
56
48-98
81
Any recurrence
77
32
48-98
85b
83a
16
63-98
86
75
16
48-91
77
72a
24
49-89
Clinical
recurrence only
EP
Microbiological
72
Range
48-98
71
29
54-91
48-98
70b
16
54-91
7
66-98
75
9
63-91
9
48-91
63
7
54-88
11
49-89
71
13
55-87
a
Clinically recurrent patients were significantly older than cured (p= 0.02).
b
Women were significantly older than men in any recurrence (p=0.05).
AC C
No. of patients
16
TE D
recurrence
27
Age
M AN U
patients
Cured
Men
SC
No. of Age
RI PT
Age (median) and gender distribution among the 56 evaluable patients.
ACCEPTED MANUSCRIPT Table 2 Number of patients prescribed different antibiotics during 3 months before inclusion. Men (35)
Women (33)
Cephalosporina
34
23
11
Isoxazolyl penicillina
21
11
10
Clindamycina
17
7
Ciprofloxacin
12
10
Trimethoprim-Sulfametoxazola
12
4
Piperacillin-Tazobactam
10
Phenoxymethyl- and benzylpenicillina
10
10 2 8 6
4
6
10
5
5
7
4
3
7
5
2
5
3
2
4
0
4
3
2
1
2
0
2
2
1
1
Rifampicin
2
0
2
Trimethoprim
2
1
1
Vancomycin i.v
1
0
1
Carbapenems Gentamicin
Pivmecillinam
Nitrofurantoin
AC C
Clarithromycin
EP
Amoxicillin-clavulanate
TE D
Metronidazole
Oral and/or intravenous.
M AN U
4
Ampicillin, Amoxicillina
a
RI PT
Total (68)
SC
Antibiotic
ACCEPTED MANUSCRIPT
Table 3
All included
Evaluable for recurrences (56)
patients (68)
All recurrences Survived
Deceased
(68)
(23/56)
(33/56)
(42/68)
(26/68)
(49)
(15)
(24)
(28)
(21)
mean
121
109
147
126
115
(range)
(5 - 323)
(12 - 322)
(9 – 323)
(9 – 323)
(5 – 290)
(49)
(15)
(24)
(28)
(21)
mean
16.6
15.8
18.6
15.8
17.5
(range)
(4.1 – 47.7)
(5.9 – 36.3)
(4.1 – 47.7)
(5.9 – 41.4)
(4.1 – 47.7)
(9)
(16)
(15)
(15)
23
27
27
23
(14 – 32)
(15 – 36)
(15 – 36)
(14 – 34)
Albumin (g/L)
(31)
mean
25
(range)
(14 – 36)
M AN U
TE D
Leucocytes (x 109/L)
EP
CRP (mg/L)
SC
Cured
(No. of patients)
1
All included (68)
All
AC C
Biochemical parameter
RI PT
Chemical laboratory findings at enrolment in relation to recurrences. The results of the patients with other concomitant infections are excluded.
68 patients were included in the study and 56 were evaluable for recurrences.
ACCEPTED MANUSCRIPT Table 4 Minimum inhibitory concentration of 81 C. difficile isolates against 8 antimicrobial agents. Antimicrobial
MIC50
MIC90
Breakpoint
agent
mg/l
mg/l
Cadazolid
0.125
0.5
0.064-0.5
Fidaxomicin
0.064
0.125
0.008-0.125
Vancomycin
0.5
1
0.125-1
Metronidazole
0.5
1
0.125-2
>2b
0.064
0.125
0.032-0.25
>0.25b
Linezolid
1
4
Moxifloxacin
2
Clindamycin
8
Range mg/l
RI PT
−a
−a
SC
>2b
M AN U
Tigecycline
(mg/l)
0.125-8
>4c
16
0.25-32
>4b
64
0.25-256
≥8d
No breakpoint established.
b
Breakpoints established by the EUCAST (The European Committee on Antimicrobial
TE D
a
Susceptibility Testing. Breakpoint tables for interpretation of MICs and zone diameters.
EP
Version 3.1, 2013. http://www.eucast.org.). Breakpoints based on literature [13, 14].
d
Breakpoints established by the CLSI [11].
AC C
c
22
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1. Demographic characteristics of the included 68 CDI patients.
23
ACCEPTED MANUSCRIPT
2150 patients tested for C. dificile at Södersjukhuset, Stockholm, Sweden May 1, 2009 - July 31, 2011
157 patients C. dificile positive
68 Included Median age 74 35 men 33 women
M AN U
89 excluded
SC
RI PT
1993 patients C. dificile negative
12 no follow-up
EP
TE D
56 evaluable (at least one follow-up sample) Median age 74 29 men (median age 71) 27 women (median age 81)
13 with clinical and microbiological recurrence Median age 85 7 men (median age 81) 6 women (median age 86)
16 with microbiological recurrence without clinical recurrence Median age 75 7 men (median age 63) 9 women (median age 77)
19 Received antibiotics after initial episode
5 Received antibiotics after initial CDI preceding recurrence
5 Received antibiotics after initial CDI preceding recurrence
AC C
24 “cured” no microbiological or clinical recurrence Median age 72 13 men (median age 71) 11 women (median age 72)
3 with clinical recurrence only 74, 75, 86 years 2 men 1 woman
0 received antibiotics after initial CDI preceding recurrence
ACCEPTED MANUSCRIPT Highlights
EP
TE D
M AN U
SC
RI PT
Clostridium difficile recurrences 56 patients in Stockholm 16 patients had relapses, i.e., same ribotype 13 patients had reinfection, i.e., new ribotype High fatality rate: 26 patients died within 1 year
AC C
• • • • •
S1075-9964(16)30003-8
DOI:
10.1016/j.anaerobe.2016.01.005
Reference:
YANAE 1531
To appear in:
Anaerobe
Received Date: 8 September 2015 Revised Date:
11 January 2016
Accepted Date: 19 January 2016
Please cite this article as: Sandell S, Rashid M-U, Jorup-Rönström C, Ellström K, Nord CE, Weintraub A, Clostridium difficile recurrences in Stockholm, Anaerobe (2016), doi: 10.1016/j.anaerobe.2016.01.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Original article
RI PT
Clostridium difficile recurrences in Stockholm
Staffan Sandella, Mamun-Ur Rashidb, Christina Jorup-Rönströma, Kristina
Karolinska Institutet, Department of Medicine, Division of Infectious Diseases,
M AN U
a
SC
Ellströma, Carl Erik Nordb*, Andrej Weintraubb
Södersjukhuset, SE-118 83 Stockholm, Sweden b
Karolinska Institutet, Department of Laboratory Medicine, Division of Clinical
TE D
Microbiology, Karolinska University Hospital Huddinge, SE-141 86 Stockholm, Sweden
EP
* Corresponding author. Tel.: +46 8 585 87838; fax: +46 8 585 87933.
AC C
E-mail address: [email protected] (C.E. Nord)
ACCEPTED MANUSCRIPT ABSTRACT Sixty-eight hospital-admitted patients with a first episode of Clostridium difficile infection (CDI) were included and followed up during 1 year. Faeces samples were collected at 1, 2, 6 and 12 months after inclusion and analyzed for the presence of C. difficile toxin B, genes for toxin A, toxin B, binary toxin
RI PT
and TcdC deletion by PCR. All strains were also PCR-ribotyped and the MICs of the isolates were determined against eight antimicrobial agents. In 68 patients initially included, antibiotics, clinical signs and co-morbidities were analyzed and 56 were evaluable for recurrences. The mean number of different antibiotics given during 3 months prior to inclusion was 2.6 (range 0-6). Six patients had not
SC
received any antibiotics and three of them had diagnosed inflammatory bowel disease. Thirty-two patients (57%) had either a microbiological or clinical recurrence, 16 of whom had clinical recurrences
M AN U
that were confirmed microbiologically (13, 23%) or unconfirmed by culture (3, 5%). Twenty-nine patients were positive in at least one of the follow-up tests, 16 had the same ribotype in follow-up tests, i.e. relapse, and 13 a different ribotype, i.e., reinfection. Most common ribotypes were 078/126, 020, 023, 026, 014/077, 001 and 005. No strain of ribotype 027 was found. Strains ribotype 078/126
TE D
and 023 were positive for binary toxin and were the strains most prone to cause recurrence. All strains were sensitive to vancomycin and metronidazole. Patients with recurrences were significantly older (p=0.02) and all patients had a high burden of comorbidities, which could explain the high fatality
AC C
Keywords:
EP
rate, 26 (38%) patients died during the 1-year follow-up.
Antibiotic treatment
Clostridium difficile infection Clinical recurrence
Microbiological recurrence Ribotyping Minimum inhibitory concentration
2
ACCEPTED MANUSCRIPT 1. Introduction Clostridium difficile infection (CDI) is a most common nosocomial diarrheal disease. A high frequency of recurrences (20-30%), either as relapses or reinfections, plays a major role in CDI [1-3]. Many studies have been performed to find out which factors are associated with
RI PT
one or more recurrences. Treatment with broad-spectrum antibiotics, proton pump inhibitors, old age, severe underlying illness and hospitalization are well-known factors associated with recurrences [4-6]. Leukocytosis, hypoalbuminemia and raised creatinine levels are also
SC
associated with an increased frequency of recurrent infection [4-6].
The aims of the prospective study were to compare relapses (same C. difficile ribotype)
M AN U
with reinfections (different C. difficile ribotypes) in relation to clinical signs, prior antibiotic use, and to find out whether certain ribotypes resulted in more recurrent infections during 1 year. The in-vitro activity of cadazolid, fidaxomicin, metronidazole, vancomycin, linezolid, tigecycline, moxifloxacin and clindamycin by determining the minimum inhibitory
TE D
concentrations (MICs) against the C. difficile isolates was also determined.
2. Materials and methods
EP
2.1 Patients and survey methods
The study was conducted at Södersjukhuset, Stockholm, Sweden. In-patients during the
AC C
period 1st May 2009 to 31st July 2011 with a primary CDI were consecutively recruited. The patients were followed during 1 year with stool sampling for C. difficile laboratory diagnoses at 1, 2, 6 and 12 months after the inclusion. C. difficile infection (CDI) was defined as a positive toxin B test and/or growth of toxigenic C. difficile combined with symptoms of gastrointestinal infection. Microbiological recurrence was defined as a C. difficile positive sample during the follow-up period. A clinical recurrence was defined either as clinical symptoms or prescribed antibiotics for treatment of CDI during the follow-up period.
3
ACCEPTED MANUSCRIPT Inclusion criteria were: (i) age over 18 years, (ii) no known CDI during the preceding 3year period, (iii) positive laboratory test, (iv) diarrhea (at least 3 loose stools/day or one bloody loose stool), and (v) life expectancy estimated to be at least 3 months after the primary CDI episode.
RI PT
At inclusion, medical histories, laboratory results and prescription data of the patients were recorded. Medical records before the inclusion as well as during the follow-up period were requested from different hospitals, nursing homes and primary care centers for the
SC
analysis.
A nurse at the hospital had a telephone contact with each patient at the time for collecting
M AN U
fecal sample and send the stool sample container. The samples were submitted to an outpatient clinic and sent to the laboratory for diagnosis. In addition, the nurse asked the patient for any episodes of loose stool incidences between the sampling occasions. The study was approved by the Ethics Committee at Karolinska Institutet, Stockholm
TE D
(EPN 2009/161-31) and all the patients signed an informed consent.
2.2 Collection and typing of strains
EP
All fecal samples were cultured for C. difficile under anaerobic conditions. The isolates were identified by characteristic colony morphology, typical smell and Gram-staining. The
AC C
isolates were then frozen at -70°C until further analyses [7].
2.3 Determination of toxins, detection of toxin gene and ribotyping The production of toxin B was determined by the cell cytotoxicity neutralization assay. The gene for toxin A was detected by conventional PCR [8]. The genes for toxin B, the binary toxin and TcdC deletion were detected by real time PCR using the GeneXpert® System
4
ACCEPTED MANUSCRIPT (Cepheid, Sunnyvale, CA, USA) assay [9]. A sample was considered positive if the CCNA assay was positive and/or the sample was positive for toxigenic culture. PCR ribotyping was performed as described previously [7, 10].
RI PT
2.4 Antimicrobial susceptibility
The antimicrobial susceptibility of the C. difficile strains was determined according to the Clinical and Laboratory Standards Institute (CLSI) guidelines by the agar dilution method
SC
using Bacteroides fragilis ATCC 25285 and Clostridium difficile ATCC 700057 as reference strains [11]. Vancomycin and metronidazole were purchased from Sigma-Aldrich Co. (Saint
M AN U
Louis, MO, USA), clindamycin was obtained from Applichem GmbH (Darmstadt, Germany), moxifloxacin was supplied by Bayer Health Care Pharmaceuticals (Berlin, Germany), fidaxomicin was provided by Optimer Pharmaceuticals, Inc. (San Diego, CA, USA), linezolid and tigecycline were provided by Pfizer Inc. (New York, NY, USA), and cadazolid was from
TE D
Actelion Pharmaceuticals Ltd. (Allschwil, Switzerland). All antimicrobial agents were dissolved and diluted according to the companies' instructions. MIC was defined as the lowest concentration of the drug that inhibited growth
EP
completely. MIC50/90 corresponds to the concentrations that inhibit the growth of 50% and 90% of the strains tested, respectively. For the calculation of resistant isolates against
AC C
vancomycin, metronidazole, tigecycline, and moxifloxacin, breakpoints established by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) were used [12]. For clindamycin, the CLSI breakpoint was used [11]. For ciprofloxacin and linezolid, no breakpoints are available in the EUCAST and CLSI documents and therefore ciprofloxacin and linezolid breakpoints (≥8 and >4 mg/l respectively) were used based on literature recommendations [13, 14].
5
ACCEPTED MANUSCRIPT 2.5 Statistics PCR ribotype profiles from clinical isolates were compared to those profiles which define the database by maximum matching with Pearson correlation. IBM SPSS Statistics 22 (Armonk, NY, USA) software was used to calculate the percentiles 50 and 90 of the MIC
in comparative calculations by Microsoft Word 2010.
SC
3. Results
RI PT
results. Descriptive statistics were used to summarize the results and student t-test was applied
3.1 Demographic data
M AN U
During the study period, stool samples from 2150 patients at Södersjukhuset were submitted to the laboratory for C. difficile analysis, and 157 samples were positive. Eightynine (57%) patients were excluded since they did not meet the inclusion criteria. Sixty-eight (43%) in-patients between the ages of 19 and 98 years, 35 men and 33 women, with a first
TE D
episode of CDI were included. Twelve patients (18%) did not submit any follow-up sample because of death or terminal disease (7 patients), severe but not deadly disease (1 patient) and lost to follow-up (4 patients). Fifty-six patients (82%) (29 men, 27 women, median age 74
EP
years, range 48-98 years) submitted at least one follow-up sample and were evaluable for recurrences. Figure 1 shows the demographic data of the patients. The patients were treated at
AC C
departments of surgery (10), internal medicine (27), infectious diseases (15), intensive care (3), and orthopedic surgery (1). As shown in Table 1, recurrent patients were significantly older than cured (p=0.02). Among recurrent patients, women were significantly older than men (p=0.05).
3.2 Clinical data
6
ACCEPTED MANUSCRIPT Sixty-three (93%) of 68 patients suffered from at least one chronic comorbidity, and cardiovascular diseases (ischemic heart disease, heart failure, hypertension, cerebrovascular disease) were the most frequent clinical diagnosis in 31 patients. Other conditions among the patients were: diabetes mellitus (12 patients), active malignancy (11 patients), chronic
RI PT
infections (9 patients), chronic liver disease (8 patients), renal disease (10 patients), and 3 patients with end-stage renal failure. Alcoholism (6 patients), chronic lung disease (6
patients), inflammatory bowel disease (5 patients), cognitive impairment (4 patients), and
SC
rheumatic/autoimmune disease (4 patients).
Nineteen (28%) patients had ongoing acute non-CDI infections at the time of primary
M AN U
CDI: septicemia (10 patients), endocarditis (2 patients), pneumonia (1 patient), appendicitis (1 patient), mammary abscess (1 patient), abdominal infection (1 patient), cholecystitis (1 patient), pancreatitis (1 patient), and intestinal tuberculosis (1 patient). Of the remaining 49 (72%) patients without concomitant acute infection, 21 had fever at
TE D
the time of primary infection. Nineteen had abdominal pain. Nine patients had vomiting and 5 had stools containing blood. Median duration of symptoms with diarrhea at the time of inclusion was 4 days (range 1-31).
EP
Twenty-four (35%) of 68 patients died during the 1-year follow-up, median age of 81 years at the time of death (range 59-97 years), 13 were men (median age 74 years), and 11
AC C
women (median age 86 years). Nine of the originally included patients died within one month after inclusion. Of the 15 patients who died within the 1-year follow-up and submitted more than one sample, 9 had at least one positive sample. In addition, 4 of the 9 patients with microbiological recurrence were positive for a binary toxin strain. The mean time from inclusion until death was 3.3 months. Causes of death were cardiovascular events (6 patients), infections other than CDI (3 patients), malignancy (2 patients) and pancreatitis (1 patient). In 14 patients, the direct cause of death was unknown, and of these, 3 patients had symptoms
7
ACCEPTED MANUSCRIPT implying ongoing CDI, and 1 died while having colitis. All patients that died within 1 year after inclusion had at least one comorbidity.
3.3 Antibiotics
RI PT
Antibiotics had been prescribed to 62 (91%) of 68 patients during 3 months prior to
inclusion. Six (9%) patients had not been exposed to antibiotics during these 3 months, and of them, 3 had a diagnosis of inflammatory bowel disease (IBD). Mean number of antimicrobial
SC
agents prescribed were 2.6 agents/patient during the 3 months prior to inclusion. Five patients had received at least 6 different antibiotics. Most common antibiotics were cephalosporins,
M AN U
isoxazolyl-penicillins and clindamycin (Table 2).
In 67 (99%) patients, treatment for CDI was given, metronidazole in 53 patients, vancomycin in 1 patient, and a combination of metronidazole and vancomycin in 13 patients. Mean duration of treatment was 13.7 days. Severe colitis occurred in 4 patients, 2 of which
TE D
underwent colectomy.
3.4 Chemical laboratory findings at inclusion in relation to recurrences
EP
Values of C-reactive protein, leucocytes and albumin are shown in Table 3. The denominator (displayed within parenthesis) varies between the categories depending on the
AC C
number of available results.
3.5 Microbiological findings Two hundred and thirteen (average 3.8 samples/patient) fecal samples were collected from the 56 patients. One hundred and twelve samples (53%) were positive. One hundred and fifteen C. difficile strains were isolated from the samples and 107 were toxigenic.
8
ACCEPTED MANUSCRIPT 3.6 Toxin gene detection and PCR-ribotyping All toxigenic isolates were positive for Toxin B by the CCNA assay and for the Toxin A and Toxin B genes by PCR. In addition, 24 isolates were positive for the binary toxin gene (15 isolates of PCR-ribotype 023 and 9 isolates of PCR-ribotype 078/126). Of the 107
RI PT
toxigenic C. difficile isolates, 99 were PCR-ribotyped (8 isolates could not be recovered from the frozen stock). Twenty-six different PCR-ribotypes were identified and the most common ribotypes were: 078/126 and 020 (15 strains each); 023 (9 strains); 026 and 014/077 (7 strains
SC
each); 001, 005 (5 strains each). C. difficile 027 was not detected in any of the samples during
3.7 Antimicrobial susceptibility
M AN U
the study period.
The antimicrobial susceptibility pattern of 81 of the 107 (26 isolates could not be recovered from the frozen stock) toxigenic C. difficile strains is shown in Table 4. All strains
TE D
were sensitive to cadazolid, fidaxomicin, vancomycin, metronidazole and tigecycline with MIC90 values of 0.5 mg/l, 0.125 mg/l, 1 mg/l, 1 mg/l, and 0.125 mg/l, respectively. Two strains were resistant (8 mg/l) to linezolid, 10 strains were resistant (8-32 mg/l) to
EP
moxifloxacin and 59 strains were resistant (8-256 mg/l) to clindamycin.
AC C
3.8 Recurrences and ribotypes Thirteen (23%) of 56 patients had both a clinical and laboratory recurrence. The median age was 85 years. Nine of the patients had the recurrence within 1 month after the primary CDI. Four of these patients had a second recurrence 4-8 weeks after the first recurrence. Sixteen (29%) of 56 patients had laboratory verified recurrence only, median age 75 years.
9
ACCEPTED MANUSCRIPT Three patients had clinical recurrence only (median age 76 years) and did not provide a sample for laboratory diagnosis at the time of clinical symptoms. However, they were successfully treated for CDI. In addition, one patient with a clinical recurrence was also positive for the toxin in the CCNA assay but culture negative.
RI PT
Twenty-four (43%) of 56 patients did not have any clinical or laboratory recurrence (median age 72 years).
A toxigenic C. difficile strain was isolated from 53 (78%) of 68 patients at inclusion.
SC
None of the toxigenic C. difficile isolates belonged to the 027 ribotype. Thus, of the 29
patients positive for toxigenic C. difficile in any of the follow-up samples, 16 had the same
M AN U
ribotype, i.e., relapsed, and 13 patients had changed ribotype, i.e., were re-infected. Twentysix patients with microbiological recurrence (with or without clinical recurrence) had continuous positive laboratory diagnosis whereas 7 had one initial negative test before recurring.
TE D
In total, 32 (57%) of the evaluable patients had either a clinical or laboratory recurrence or both of CDI (Figure 1). Of the 29 patients with at least one C. difficile positive stool sample during the follow-up period, 13 (45%) patients had also a clinical manifestation of CDI.
EP
Proton pump inhibitors were used by 23 (41%) of the 56 patients. Among the patients with PPI, 15 (65%) had relapsed, whereas among patients without PPI, 18 (54%) patients had
AC C
relapsed.
4. Discussion
In this prospective study of 68 included patients, 56 patients were followed with a first CDI during 1 year concerning new clinical symptoms and presence of toxigenic C. difficile in the fecal samples. Twelve patients were excluded concerning recurrences due to death or severe disease. The median age of the 68 patients was 74 years and the burden of
10
ACCEPTED MANUSCRIPT comorbidities was high, two factors that well reflect the CDI patient group in general and partly explain the high proportion of recurrences [6, 15-19]. Younger patients were observed in the cured group compared to recurring patients (p=0.02). Recurring women were significantly older than recurring men (p=0.05). A high number of the included patients died
RI PT
during the 1-year follow-up period, 26 patients of the originally included 68 patients. Since autopsy is rarely performed on elderly patients in Sweden, a definite cause of death could not be made in all but one patient. However, the high fatality rate among the included patients
patients, diarrhea was ongoing at the time of death.
SC
could be attributed to the high age and high burden of comorbidities. In only 3 of 26 deceased
M AN U
PCR-ribotyping showed that of the 29 patients positive for toxigenic C. difficile in the follow-up samples, 16 patients had the same ribotype i.e. relapsed and 13 patients changed the ribotype, i.e. were re-infected. None of the toxigenic C. difficile isolates belonged to the 027 ribotype. However, in the present study among relapsing patients, ribotypes 023 (n=15) and
TE D
078/126 (n=9) were dominating, and were found to be binary toxin producing. The role of the binary toxin as a virulence factor has until now not been thoroughly investigated [20]. In addition, the role of binary toxin in recurrences is still unclear. Stewart et al showed that the
EP
binary toxin significantly associated with a recurrent CDI [21]. However, in a recent study from Spain it has been shown that in a non-027 setting, the presence of binary toxin had no
AC C
role in recurrent CDI [22]. The findings in our, although limited, study show that the presence of binary toxin seems to be linked to an increased risk of recurrence. Decreased sensitivity against metronidazole and vancomycin, the drugs generally used for treatment, was not seen in any of the strains. Resistance should thus not have had any impact on the recurrences. Only 6 of 68 patients had not received any antibiotics during 3 months prior to inclusion, i.e., the risk for community-acquired infection seems low, contradictory to other authors. In
11
ACCEPTED MANUSCRIPT addition, 3 of the patients had an IBD diagnosis and it is well known that these patients are more prone to acquire CDI [23, 24]. The antibiotics given prior to the CDI are well known, i.e., cephalosporins, clindamycin and ciprofloxacin, to cause CDI. Isoxazolyl-penicillins were also common, and patients treated with these antibiotics often received other antibiotics as
RI PT
well, and long duration of therapy against different infections might have played a role.
In the present study, antibiotic use after the primary CDI did not increase the risk of
recurrence, which is contradictory to previous studies [19, 25, 26]. This may be due to a long-
to exceptionally high age and frequent comorbidities.
SC
lasting ecological change which may have been more pronounced than in previous studies due
have been published worldwide.
M AN U
During the last years (2012-2015), several papers covering recurrent CDI and risk factors
In a Japanese investigation [27], 20 patients with CDI recurrences were analyzed by PCR ribotypes, cytotoxicity and antimicrobial susceptibility. Sixteen patients were relapse cases
TE D
and four patients were reinfection cases. There was no association between PCR ribotypes and incidence of recurrences. All C. difficile strains were susceptible to vancomycin and metronidazole but resistance against clindamycin was found in 87.7% of the isolates.
EP
Taori et al. [28] studied 56 patients with recurrent CDI in Scotland for one year. Ribotype 001 was the most common type and no ribotype 027 was detected. High leucocyte count and
AC C
low albumin were associated with mortality. No association with proton-pump inhibitors and with vancomycin-resistant enterococci were found. Recently Hikone et al. [29] published another report on recurrent CDI in 14 patients at a Japanese university hospital. The recurrent patients had more underlying malignant diseases and more intensive care unit hospitalization. A systemic review of risk factors for recurrence of CDI has been published by Chakra et al. [30]. Older age, use of antibiotics after diagnosis, use of proton-pump inhibitors,
12
ACCEPTED MANUSCRIPT hypoalbuminemia, leukocytosis, renal failure and infection with ribotype 027 strains were risk factors for recurrent CDI. Deshpande et al. [31] confirmed the findings of Chakra et al [30] concerning risk factors for recurrent CDI. The most frequent risk factors were age > 65 years, additional antibiotics
RI PT
during follow up, proton-pump inhibitors, renal insufficiency and quinolone therapy.
Abdelfatah et al. [32] also studied risk factors for developing recurrence of CDI. They concluded that most recurrences occurred in patients with comorbidities such as chronic
SC
kidney disease. Use of proton-pump inhibitors was another significant risk factor.
When comparing the laboratory and clinical data on CDI recurrences and risk factors,
M AN U
there are similar findings in most publications, but some differences are found. These findings
AC C
EP
TE D
may be explained by the epidemiological situations in Europe, America and Asia.
13
ACCEPTED MANUSCRIPT References [] [1]
Vardakas KZ, Polyzos KA, Patouni K, Rafailidis PI, Samonis G, Falagas ME.
Treatment failure and recurrence of Clostridium difficile infection following treatment with
RI PT
vancomycin or metronidazole: a systematic review of the evidence. Int J Antimicrob Agents 2012;40:1-8. [2]
Eyre DW, Walker AS, Wyllie D, Dingle KE, Griffiths D, Finney J, et al. Predictors of
SC
first recurrence of Clostridium difficile infection: implications for initial management. Clin
[3]
M AN U
Infect Dis 2012;55 Suppl 2:S77-87.
Louie TJ. Treatment of first recurrences of Clostridium difficile-associated disease:
waiting for new treatment options. Clin Infect Dis 2006;42:765-7. [4]
DuPont HL, Garey K, Caeiro JP, Jiang ZD. New advances in Clostridium difficile
2008;21:500-7.
Kelly CP, Kyne L. The host immune response to Clostridium difficile. J Med
EP
[5]
TE D
infection: changing epidemiology, diagnosis, treatment and control. Curr Opin Infect Dis
[6]
AC C
Microbiol 2011;60:1070-9.
Louie TJ, Miller MA, Crook DW, Lentnek A, Bernard L, High KP, et al. Effect of age
on treatment outcomes in Clostridium difficile infection. J Am Geriatr Soc 2013;61:222-30. [7]
Rashid MU, Lozano HM, Weintraub A, Nord CE. In vitro activity of cadazolid against
Clostridium difficile strains isolated from primary and recurrent infections in Stockholm, Sweden. Anaerobe 2013;20:32-5.
14
ACCEPTED MANUSCRIPT [8]
Kato H, Kato N, Watanabe K, Iwai N, Nakamura H, Yamamoto T, et al. Identification
of toxin A-negative, toxin B-positive Clostridium difficile by PCR. J Clin Microbiol 1998;36:2178-82. Huang H, Weintraub A, Fang H, Nord CE. Comparison of a commercial multiplex
RI PT
[9]
real-time PCR to the cell cytotoxicity neutralization assay for diagnosis of Clostridium difficile infections. J Clin Microbiol 2009;47:3729-31.
Stubbs SL, Brazier JS, O'Neill GL, Duerden BI. PCR targeted to the 16S-23S rRNA
SC
[10]
gene intergenic spacer region of Clostridium difficile and construction of a library consisting
[11]
M AN U
of 116 different PCR ribotypes. J Clin Microbiol 1999;37:461-3.
Clinical and Laboratory Standards Institute. Methods for antimicrobial susceptibility
testing of anaerobic bacteria; Approved standard. 8th ed. Document M11-A8. Wayne, PA:
[12]
TE D
CLSI; 2012.
The European Committee on Antimicrobial Susceptibility Testing. Breakpoint tables
for interpretation of MICs and zone diameters. Version 3.1. Keessen EC, Hensgens MP, Spigaglia P, Barbanti F, Sanders IM, Kuijper EJ, et al.
EP
[13]
AC C
Antimicrobial susceptibility profiles of human and piglet Clostridium difficile PCR-ribotype 078. Antimicrob Resist Infect Control 2013;2:14. [14]
Pelaez T, Alcala L, Blanco JL, Alvarez-Perez S, Marin M, Martin-Lopez A, et al.
Characterization of swine isolates of Clostridium difficile in Spain: a potential source of epidemic multidrug resistant strains? Anaerobe 2013;22:45-9. [15]
Stanley JD, Bartlett JG, Dart BWt, Ashcraft JH. Clostridium difficile infection. Curr
Probl Surg 2013;50:302-37.
15
ACCEPTED MANUSCRIPT [16]
Samie AA, Traub M, Bachmann K, Kopischke K, Theilmann L. Risk factors for
recurrence of Clostridium difficile-associated diarrhoea. Hepatogastroenterology 2013;60:1351-4. Jones AM, Kuijper EJ, Wilcox MH. Clostridium difficile: a European perspective. J
RI PT
[17]
Infect 2013;66:115-28. [18]
Wilcox MH, Shetty N, Fawley WN, Shemko M, Coen P, Birtles A, et al. Changing
SC
epidemiology of Clostridium difficile infection following the introduction of a national
[19]
M AN U
ribotyping-based surveillance scheme in England. Clin Infect Dis 2012;55:1056-63. Kim J, Shaklee JF, Smathers S, Prasad P, Asti L, Zoltanski J, et al. Risk factors and
outcomes associated with severe Clostridium difficile infection in children. Pediatr Infect Dis J 2012;31:134-8.
Gerding DN, Johnson S, Rupnik M, Aktories K. Clostridium difficile binary toxin
TE D
[20]
CDT: mechanism, epidemiology, and potential clinical importance. Gut Microbes 2014;5:1527.
Stewart DB, Berg A, Hegarty J. Predicting recurrence of C. difficile colitis using
EP
[21]
[22]
AC C
bacterial virulence factors: binary toxin is the key. J Gastrointest Surg 2013;17:118-24. Reigadas E, Alcala L, Marin M, Martin A, Iglesias C, Bouza E. Role of binary toxin in
the outcome of Clostridium difficile infection in a non-027 ribotype setting. Epidemiol Infect 2015;1-6. [23]
Nitzan O, Elias M, Chazan B, Raz R, Saliba W. Clostridium difficile and
inflammatory bowel disease: role in pathogenesis and implications in treatment. World J Gastroenterol 2013;19:7577-85.
16
ACCEPTED MANUSCRIPT [24]
Sha S, Xu B, Wang X, Zhang Y, Wang H, Kong X, et al. The biodiversity and
composition of the dominant fecal microbiota in patients with inflammatory bowel disease. Diagn Microbiol Infect Dis 2013;75:245-51. Johnson S. Recurrent Clostridium difficile infection: a review of risk factors,
treatments, and outcomes. J Infect 2009;58:403-10. [26]
RI PT
[25]
Garey KW, Dao-Tran TK, Jiang ZD, Price MP, Gentry LO, Dupont HL. A clinical
[27]
M AN U
antibiotics. J Hosp Infect 2008;70:142-7.
SC
risk index for Clostridium difficile infection in hospitalised patients receiving broad-spectrum
Oka K, Osaki T, Hanawa T, Kurata S, Okazaki M, Manzoku T, et al. Molecular and
microbiological characterization of Clostridium difficile isolates from single, relapse, and reinfection cases. J Clin Microbiol 2012;50:915-21.
Taori SK, Wroe A, Poxton IR. Clostridium difficile infections in South East Scotland:
TE D
[28]
mortality and recurrence in a region without PCR ribotype 027. J Med Microbiol 2013;62:1468-77.
Hikone M, Ainoda Y, Tago S, Fujita T, Hirai Y, Takeuchi K, et al. Risk factors for
EP
[29]
AC C
recurrent hospital-acquired Clostridium difficile infection in a Japanese university hospital. Clin Exp Gastroenterol 2015;8:191-6. [30]
Abou Chakra CN, Pepin J, Sirard S, Valiquette L. Risk factors for recurrence,
complications and mortality in Clostridium difficile infection: a systematic review. PloS one 2014;9:e98400.
17
ACCEPTED MANUSCRIPT [31]
Deshpande A, Pasupuleti V, Thota P, Pant C, Rolston DD, Hernandez AV, et al. Risk
factors for recurrent Clostridium difficile infection: a systematic review and meta-analysis. Infect Control Hosp Epidemiol 2015;36:452-60. Abdelfatah M, Nayfe R, Nijim A, Enriquez K, Ali E, Watkins RR, et al. Factors
RI PT
[32]
Predicting recurrence of Clostridium difficile infection (CDI) in hospitalized patients:
AC C
EP
TE D
M AN U
SC
Retrospective study of more than 2000 patients. J Investig Med 2015;63:747-51
18
ACCEPTED MANUSCRIPT
Table 1
All patients
Women No. of Range
Age
Range patients
All
74
56
48-98
81
Any recurrence
77
32
48-98
85b
83a
16
63-98
86
75
16
48-91
77
72a
24
49-89
Clinical
recurrence only
EP
Microbiological
72
Range
48-98
71
29
54-91
48-98
70b
16
54-91
7
66-98
75
9
63-91
9
48-91
63
7
54-88
11
49-89
71
13
55-87
a
Clinically recurrent patients were significantly older than cured (p= 0.02).
b
Women were significantly older than men in any recurrence (p=0.05).
AC C
No. of patients
16
TE D
recurrence
27
Age
M AN U
patients
Cured
Men
SC
No. of Age
RI PT
Age (median) and gender distribution among the 56 evaluable patients.
ACCEPTED MANUSCRIPT Table 2 Number of patients prescribed different antibiotics during 3 months before inclusion. Men (35)
Women (33)
Cephalosporina
34
23
11
Isoxazolyl penicillina
21
11
10
Clindamycina
17
7
Ciprofloxacin
12
10
Trimethoprim-Sulfametoxazola
12
4
Piperacillin-Tazobactam
10
Phenoxymethyl- and benzylpenicillina
10
10 2 8 6
4
6
10
5
5
7
4
3
7
5
2
5
3
2
4
0
4
3
2
1
2
0
2
2
1
1
Rifampicin
2
0
2
Trimethoprim
2
1
1
Vancomycin i.v
1
0
1
Carbapenems Gentamicin
Pivmecillinam
Nitrofurantoin
AC C
Clarithromycin
EP
Amoxicillin-clavulanate
TE D
Metronidazole
Oral and/or intravenous.
M AN U
4
Ampicillin, Amoxicillina
a
RI PT
Total (68)
SC
Antibiotic
ACCEPTED MANUSCRIPT
Table 3
All included
Evaluable for recurrences (56)
patients (68)
All recurrences Survived
Deceased
(68)
(23/56)
(33/56)
(42/68)
(26/68)
(49)
(15)
(24)
(28)
(21)
mean
121
109
147
126
115
(range)
(5 - 323)
(12 - 322)
(9 – 323)
(9 – 323)
(5 – 290)
(49)
(15)
(24)
(28)
(21)
mean
16.6
15.8
18.6
15.8
17.5
(range)
(4.1 – 47.7)
(5.9 – 36.3)
(4.1 – 47.7)
(5.9 – 41.4)
(4.1 – 47.7)
(9)
(16)
(15)
(15)
23
27
27
23
(14 – 32)
(15 – 36)
(15 – 36)
(14 – 34)
Albumin (g/L)
(31)
mean
25
(range)
(14 – 36)
M AN U
TE D
Leucocytes (x 109/L)
EP
CRP (mg/L)
SC
Cured
(No. of patients)
1
All included (68)
All
AC C
Biochemical parameter
RI PT
Chemical laboratory findings at enrolment in relation to recurrences. The results of the patients with other concomitant infections are excluded.
68 patients were included in the study and 56 were evaluable for recurrences.
ACCEPTED MANUSCRIPT Table 4 Minimum inhibitory concentration of 81 C. difficile isolates against 8 antimicrobial agents. Antimicrobial
MIC50
MIC90
Breakpoint
agent
mg/l
mg/l
Cadazolid
0.125
0.5
0.064-0.5
Fidaxomicin
0.064
0.125
0.008-0.125
Vancomycin
0.5
1
0.125-1
Metronidazole
0.5
1
0.125-2
>2b
0.064
0.125
0.032-0.25
>0.25b
Linezolid
1
4
Moxifloxacin
2
Clindamycin
8
Range mg/l
RI PT
−a
−a
SC
>2b
M AN U
Tigecycline
(mg/l)
0.125-8
>4c
16
0.25-32
>4b
64
0.25-256
≥8d
No breakpoint established.
b
Breakpoints established by the EUCAST (The European Committee on Antimicrobial
TE D
a
Susceptibility Testing. Breakpoint tables for interpretation of MICs and zone diameters.
EP
Version 3.1, 2013. http://www.eucast.org.). Breakpoints based on literature [13, 14].
d
Breakpoints established by the CLSI [11].
AC C
c
22
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1. Demographic characteristics of the included 68 CDI patients.
23
ACCEPTED MANUSCRIPT
2150 patients tested for C. dificile at Södersjukhuset, Stockholm, Sweden May 1, 2009 - July 31, 2011
157 patients C. dificile positive
68 Included Median age 74 35 men 33 women
M AN U
89 excluded
SC
RI PT
1993 patients C. dificile negative
12 no follow-up
EP
TE D
56 evaluable (at least one follow-up sample) Median age 74 29 men (median age 71) 27 women (median age 81)
13 with clinical and microbiological recurrence Median age 85 7 men (median age 81) 6 women (median age 86)
16 with microbiological recurrence without clinical recurrence Median age 75 7 men (median age 63) 9 women (median age 77)
19 Received antibiotics after initial episode
5 Received antibiotics after initial CDI preceding recurrence
5 Received antibiotics after initial CDI preceding recurrence
AC C
24 “cured” no microbiological or clinical recurrence Median age 72 13 men (median age 71) 11 women (median age 72)
3 with clinical recurrence only 74, 75, 86 years 2 men 1 woman
0 received antibiotics after initial CDI preceding recurrence
ACCEPTED MANUSCRIPT Highlights
EP
TE D
M AN U
SC
RI PT
Clostridium difficile recurrences 56 patients in Stockholm 16 patients had relapses, i.e., same ribotype 13 patients had reinfection, i.e., new ribotype High fatality rate: 26 patients died within 1 year
AC C
• • • • •