Bacteriology of Spontaneous Intra-Abdominal Abscess in Patients with Crohn Disease in China: Risk of Extended-Spectrum Beta-Lactamase-Producing Bacteria

Abstract

Background:

Crohn disease (CD) has its own features and the etiology is believed to be related to the gastrointestinal flora, which are the cause of most intra-abdominal infections. However, few experiences have been reported regarding the specific bacteriology of spontaneous intra-abdominal abscess (IAA) in CD patients and related risk factors. In the present study, we aim to investigate the bacteriology of IAA in CD and the associated risk factors.

Methods:

Retrospective study on 97 CD patients suffering from IAA was performed from January 2007 to December 2012. Microbiologic information of pus cultures was pooled, and logistic regression was applied to examine the possible factors affecting the incidence of common organisms and the bacteria producing extended-spectrum beta-lactamase (ESBL).

Results:

Of the 97 patients recruited, 151 aerobic bacteria were cultured from the pus of IAA. The leading bacteria were Escherichia coli (57.7%), Klebsiella pneumoniae (19.6%) and Enterococcus faecium (18.6%). Extended-spectrum beta-lactamase-producing bacteria were more commonly found among CD patients receiving immunosuppressant.

Conclusions:

Broad-spectrum antibiotics are recommended for treating IAA in CD. Empirical antibiotic coverage against ESBL-producing bacteria in patients receiving immunosuppressive therapy is advocated. Considering the predominance of Enterococcus faecium, it may be advisable to use more effective antibiotics when gram-positive cocci are suspected.

Crohn disease (CD), a subtype of inflammatory bowel disease, is characterized by transmural inflammation of the bowel, which can lead to serious complications including intra-abdominal abscess (IAA). Approximately 10% of CD patients develop IAA during the course of their illness [1]. It has been proposed in the European Crohn's and Colitis Organization (ECCO) recommendations that IAA should be managed with antibiotics, percutaneous drainage, or surgery [2]. However, there is considerable variation in the clinical approach among physicians because of the lack of practice guidelines [3]. Notwithstanding, one consensus that has been accepted commonly is that antimicrobial therapy should be initiated promptly once the diagnosis of IAA is made [4], and a subset of patients of IAA can improve with antibiotic therapy alone [5,6].

Little research have been conducted to evaluate the specific bacteriology of spontaneous IAA with regard to CD and the factors affecting the incidence of related organisms, which influences the curative effect of antibiotics administered. To make an effective recommendation for empiric therapy for CD-related IAA, especially for patients whose IAA could not be drained effective, we reviewed our institution's information about the pathogenic bacteria distribution and identified the factors that influence their occurrence.

Patients and Methods

Patient selection

Data for patients with CD who received a diagnosis of IAA between January 2007 and December 2012 in the Department of General Surgery, Jinling Hospital, was retrieved retrospectively. The diagnosis of IAA was confirmed by computed tomography (CT) or ultrasonography after admission. Once the diagnosis was made, the patients would receive percutaneous puncture drainage within one or two days or surgical drainage within one week. Antibiotics, such as cephalosporin and ornidazole, were routinely administered to confirmed cases and adjusted subsequently according to culture results. Patients were excluded in the present study if the following criteria were met: Post-operative IAA, which occurred less than 1 mo after abdominal surgery; the abscess wasn't appropriate for either percutaneous or surgical drainage; pus samples obtained from the IAA were probably contaminated.

Microbiological examination

Specimens of pus were obtained by either the puncture of IAA with guidance of CT or ultrasound using a standardized protocol or the surgical drainage. After collected by needle aspiration, the samples were then transported immediately, at room temperature, to the laboratory for aerobic microbial cultures. All specimens were cultured on blood agar, MacConkey agar, and chocolate agar, then incubated at 37°C for 24 to 48 h. Isolates from cultures were identified by standard aerobic microbial identification methods. Mycobacterial and fungal examinations were also conducted. Anaerobic pathogen was not studied in this present study. The isolates were also screened for extended-spectrum beta-lactamase (ESBL) production by disk agar diffusion (DAD) using third-generation cephalosporins and double disk synergy Test (DDST) with and without clavulanic acid (CA), as recommended by the Clinical Laboratory Standards Institute's guidelines.

Data collection

The following clinical data of enrolled patients were retrospectively collected: age, gender, and disease location of CD, age at diagnosis of CD, duration of CD, and age of diagnosis of spontaneous IAA, prior bowel resections, and types of medication. Microbiological information of pus cultures was pooled for the analysis of the distribution of the pathogenic organisms.

Statistics

Logistic regression was applied to compare the incidence of common organisms and ESBL-producing bacteria in various conditions such as age, gender, disease locations, and other well-established factors. The statistical analysis was performed using PASW for Windows version 18.0 (IBM, Inc., Armonk, NY).

Results

Patient characteristics

A total of 97 CD patients with IAA were enrolled in this study. The clinical characteristics are indicated in Table 1. Of the 97 patients, 69 patients (71.1%) were male and 28 (28.9%) were female. The age of onset and diagnosis of CD were 27.8±10.0 and 31.4±10.5 y respectively, and the diagnosis of IAA was made at the age of 34.6±11.1 y. The most common disease location was small intestine (41 patients, 42.3%), followed by small bowel and colon (30 patients, 30.9%), colon (26 patients, 26.8%), and upper gastrointestinal tract (5 patients, 5.2%). For the treatment of CD, the drug used mostly was a corticosteroid.

Table 1.

Clinical Characteristics of CD Patients with IAA

Characteristic	Value (n=97)
Gender
Male/Female	69/28
Age at onset of CD, Mean±SD	27.8±10.0
Age at diagnosis of CD, (y), Mean±SD	31.4±10.5
Age at diagnosis of IAA, (y), Mean±SD	34.6±11.1
CD locations^*, (%)
Small intestine	41 (42.3)
Colon	26 (26.8)
Small bowel and colon	30 (30.9)
Upper gastrointestinal tract	5 (5.2)
Previous bowel resections (%)
Appendectomy	17 (17.5)
Small intestinal resection	16 (16.5)
Colectomy	23 (23.7)
Ileocecoectomy	16 (16.5)
Number of previous bowel resections
0	45 (46.4)
1	25 (25.8)
≥2	27 (27.8)
Medical treatment of CD^†, (%)
Sulfasalazine	20 (20.6)
Mesalazine	28 (28.9)
Corticosteroid	36 (37.1)
Azathioprine	10 (10.3)
TWP^‡	9 (9.3)

As L4 may coexist with L1, L2, and L3, the sum was more than 100%.

†

As a patient may receive a combination of several treatments, the sum was more than 100%.

‡

TWP, Tripterygium wilfordii Hook F, is a Chinese herbal medicine with proved immunosuppressive activity.

CD=Crohn disease; IAA=intra-abdominal abscess; SD=standard deviation.

Microbiological information of pus cultures from IAA in CD

In total, 151 aerobic bacteria and five fungal pathogens were cultured form 97 patients (Table 2). 53 (54.6%) of the 97 cases were polymicrobial. Among the 97 patients, Escherichia coli (57.7%) was the most common isolated organism, followed by Klebsiella pneumoniae (19.6%) and Enterococcus faecium (18.6%). A total of 59 isolates (37.8%) were identified to produce ESBL. Escherichia coli was the most common ESBL-producing enterobacterium (38 isolates). Other organisms were K. pneumoniae (nine isolates), Pseudomonas aeruginosa (four isolates), Acinetobacter baumannii (three isolates), Enterobacter cloacae (three isolates), and Proteus mirabilis (two isolates).

Table 2.

Results of Pus Cultures in 97 Patients with IAA

Organisms	n (%)	ESBL+ (%)
Aerobic gram-positive bacteria
Enterococcus faecium	18 (18.6)	NA
Staphylococcus aureus	8 (8.2)	NA
S. epidermidis	5 (5.2)	NA
Streptococcus pyogenes	3 (3.1)	NA
S. haemolyticus	2 (2.1)	NA
E. faecalis	2 (2.1)	NA
Group D streptococcus	2 (2.1)	NA
Others	3 (3.1)	NA
Aerobic Gram-negative bacteria
Escherichia coli	56 (57.7)	38 (67.9)
Klebsiella pneumoniae	19 (19.6)	9 (47.4)
Pseudomonas aeruginosa	9 (9.3)	4 (44.4)
Acinetobacter baumannii	7 (7.2)	3 (42.9)
Enterobacter cloacae	5 (5.2)	3 (60.0)
Proteus mirabilis	4 (4.1)	2 (50.0)
Stenotrophomnas maltophilia	2 (2.1)	0 (0)
Klebsiella oxytoca	2 (2.1)	0 (0)
Others	4 (4.1)	0 (0)
Fungal
Candida albicans	5 (5.2)	NA
Total	156 (160.8)^*	59 (37.8)

Values in parentheses are percentages. ^*As some patients were polymicrobial, the sum is more than 100%.

NA=not applicable; IAA=intra-abdominal abscess; ESBL=extended-spectrum beta-lactamase.

Risk factors for the occurrence of pathogens in IAA of CD

To reveal the possible relation between clinical features of CD and the growth of pathogenic bacteria in IAA, a logistic regression analysis was performed (Table 3). Among many factors selected, only immunosuppressive medication was found to significantly impact the culture rate of ESBL-producing bacteria with a p value of 0.03.

Table 3.

Logistic Regression Analysis of the Pathogens of IAA in CD

	E. coli		K. pneumoniae		E. faecium		ESBL+ bacteria
Parameter	OR	p	OR	p	OR	p	OR	p
Gender (Male/Female)	0.68	0.29	1.16	0.78	0.79	0.67	1.31	0.46
Smoking history	0.86	0.70	1.08	0.89	0.84	0.77	1.04	0.92
Age at onset of CD	0.48	0.14	1.39	0.59	1.51	0.50	0.85	0.71
≥40
<40
Age at diagnosis of CD	0.56	0.16	1.51	0.44	0.81	0.82	0.81	0.60
≥40
<40
Age at diagnosis of IAA	0.60	0.16	1.05	0.93	2.50	0.07	0.71	0.45
≥40
<40
CD locations
Small intestine	1.00	0.99	0.82	0.69	0.90	0.85	1.09	0.79
Colon	1.59	0.22	0.49	0.28	1.47	0.47	1.62	0.19
Small bowel and colon	0.65	0.24	2.04	0.15	0.77	0.64	0.58	0.13
Upper gastrointestinal tract	0.89	0.87	0.00	0.99	2.34	0.31	0.48	0.33
Immunosuppressive medication	1.58	0.17	0.97	0.95	0.85	0.74	2.08	0.03^*
Number of previous bowel resections
0	1.42	0.29	0.65	0.39	0.72	0.51	1.51	0.21
1	0.67	0.30	1.00	0.99	1.47	0.47	0.60	0.19
≥2	0.94	0.87	1.64	0.34	1.01	0.98	0.97	0.92
Previous hospital admission	1.13	0.89	0.68	0.73	2.28	0.99	0.29	0.16
Previous antibiotic therapy	1.12	0.92	2.38	0.99	2.28	0.99	0.30	0.33
Recent surgery	0.45	0.09	1.09	0.89	1.18	0.79	0.51	0.13
Nasogastric tube	0.42	0.10	0.62	0.54	1.69	0.40	0.63	0.35

Logistic regression was performed using PASW for Windows version 18.0. ^*p <0.05.

IAA=intra-abdominal abscess; CD=Crohn disease; OR=odds ratio.

Discussion

The present study was designed to investigate the specific bacteriology of spontaneous IAA and to discover whether certain possible factors existed affecting the incidence of pathogenic organisms. The leading aerobic bacteria cultured were E. coli (35.9%), K. pneumoniae (12.2%) and E. faecium (11.5%). Additionally, our study showed that ESBL-producing gram-negative bacteria were more commonly found among IAA patients who received immunosuppressive therapy currently.

As shown in our study, a wide variety of pathogens are involved in spontaneous IAA in CD patients. Therefore it is advisable to administer broad-spectrum antibiotics. Moreover, it is quite generally accepted that E. coli is the most common organism isolated from patients who have intra-abdominal infections, followed by K. pneumoniae [7 –10]. The result of our study ultimately reinforced this view. Thus, empiric antibiotic coverage against enteric gram-negative bacilli is recommended for treating IAA in CD. Logistic regression analysis conducted in this study revealed that the CD patients who used immunosuppressants were more likely to have an intra-abdominal infection caused by ESBL-producing bacteria. Hence, more effective antibiotics such as carbapenems should be considered when CD patients suffering from IAA had immunosuppressive medication in the meantime. Though numerous studies assessing risk factors for colonization and infection with ESBL-producing organisms yielded a plethora of conflicting results, some generalizations can be made. Patients had prior antibiotic exposure and previous hospitalizations were at high risk for the colonization [11 –13] and infection [14 –16] of EBSL-producing Enterobacteriaceae. CD patients were reasonably prone to colonize EBSL-positive bacteria, and we showed that immunosuppressive medication was a contributing factor further inducing the intra-abdominal infection of EBSL-producing organisms. Whereas some studies also manifested the role of immunosuppressive therapy in the infection of EBSL-positive infection [17,18], we still have to be cautious about this finding because of the relatively small amount of samples in our study.

Gram-positive cocci frequently cause intra-abdominal infections as well. Streptococci are supposed to be the most common gram-positive organisms isolated [9,19]. As for Enterococcus spp., it is isolated much less frequently and reported to be found in 10% to 20% of patients [8,9,19,20]. Enterococcus faecalis has been the predominant enterococcal species, accounting for 80% to 90% of all clinical isolates [21], and occupied a large proportion of the enterococcal strains isolated from patients who had intra-abdominal infections [22]. However, our study showed a different bacteriologic pattern, with E. faecium (18.5%) being the gram-positive cocci cultured most commonly, much more frequently than streptococci and E. faecalis. Actually, other research had already mentioned the increasing proportion of E. faecium in clinical isolates [23,24]. Sood et al. proposed that this microbiologic shift is likely to be explained in part by the emergence of vancomycin-resistant enterococci (VRE) in consideration of E. faecium being the dominant species identified among VRE [25]. The risk factors for VRE included longer duration of hospitalization and previous antimicrobial therapy [25], hence the susceptibility for VRE in CD populations may be the possible reason for the predominance of E. faecium. However, we failed to reveal the further and deeper causes inducing the infection of E. faecium in the logistic regression analysis. Considering that levels of resistance among E. faecium isolates are high whereas E. faecalis resistance is notably low against vancomycin and ampicillin, it may be advisable to use more effective antibiotics when gram-positive cocci intra-abdominal infections are suspected.

However, we have to admit the limitations in the present study. Given that this research was conducted from a single center with a relatively small sample size, the results of the epidemiology of bacterial infection may suffer from a certain degree of bias. Notwithstanding, because our center is the biggest specialized treatment center for CD in China, the microbiologic pattern revealed in this study was similar to the general distribution of pathogenic bacteria of intra-abdominal infection reported by other epidemiological studies, with E. coli and K. pneumoniae being the most common organisms isolated. Secondly, the microbiologic information reviewed in this study was retrieved from the clinical laboratory, reporting mostly one or two aerobic organisms per sample. Actually, whereas in other perspective research studies, each sample may isolated five to ten bacterial isolates [7,19], most clinical laboratories typically identify only a few microorganisms [26,27]. However, the pathogenic role of other rare species of bacteria in intra-abdominal infection has not been emphasized.

In conclusion, we ultimately manifested that the pathogenic bacterial distribution of IAA in CD patients was mainly the same as that of the general abdominal infection, except for the predominance of E. faecium in CD patients. Therefore, broad-spectrum antibiotics against gram-negative bacilli are recommended. Empirical antibiotic coverage against the infection of ESBL-producing bacteria in CD patients using immunosuppressive therapy is advisable, such as carbapenems. CD has its own special clinical features and the etiology is believed to be related to the gastrointestinal flora [28], which are the cause of most intra-abdominal infections [8]. Hence we conducted this pilot study. However, in the light of the limitations above, a larger and multicenter study is needed to draw accurate and reliable conclusions.

Footnotes

Acknowledgments

This study was supported by grants from Research Talents of Jiangsu Province, China (BRA2011232) and National Natural Science Foundation of China (81270478).

Author Disclosure Statement

The authors declare that they have no conflicts of interest.

References

Hurst

, Molinari

, Chung

, et al. Prospective study of the features, indications, and surgical treatment in 513 consecutive patients affected by Crohn's disease. Surgery, 1997; 122:661–667.

Dignass

, Van Assche

, Lindsay

, et al. The second European evidence-based consensus on the diagnosis and management of Crohn's disease: Current management. J Crohns Colitis, 2010; 4:28–62.

Dotson

, Nwomeh

, Andridge

, et al. Variation in management of intra-abdominal abscesses in children with Crohn's disease. Inflamm Bowel Dis, 2013; 19:818–825.

Feagins

, Holubar

, Kane

, Spechler

. Current strategies in the management of intra-abdominal abscesses in Crohn's disease. Clin Gastroenterol and Hepatol, 2011; 9:842–850.

Kumar

, Kim

, Haukoos

, et al. Factors affecting the successful management of intra-abdominal abscesses with antibiotics and the need for percutaneous drainage. Dis Colon Rectum, 2006; 49:183–189.

Bermejo

, Garrido

, Chaparro

, et al. Efficacy of different therapeutic options for spontaneous abdominal abscesses in Crohn's disease: Are antibiotics enough?. Inflamm Bowel Dis, 2012; 18:1509–1514.

Brook

. Microbiology of polymicrobial abscesses and implications for therapy. J Antimicrob Chemother, 2002; 50:805–810.

Mazuski

, Solomkin

. Intra-abdominal infections. Surg Clin North Am, 2009; 89:421–437.

Marshall

. Intra-abdominal infections. Microbes Infect, 2004; 6:1015–1025.

10.

Goldstein

EJC

. Intra-abdominal anaerobic infections: Bacteriology and therapeutic potential of newer antimicrobial carbapenem, fluoroquinolone, and desfluoroquinolone therapeutic agents. Clin Infect Dis, 2002; 35:S106–S11.

11.

Lonchel

, Melin

, Gangoue-Pieboji

, et al. Extended-spectrum beta-lactamase-producing Enterobacteriaceae in Cameroonian hospitals. Eur J Clin Microbiol Infect Dis, 2013; 32:79–87.

12.

Azap

, Arslan

, Karaman

, Togan

. Risk factors for fecal carriage of extended-spectrum beta-lactamase producing Escherichia coli and Klebsiella spp. in the community. Turk J Med Sci, 2007; 37:31–38.

13.

Luvsansharav

U-O

, Hirai

, Nakata

, et al. Prevalence of and risk factors associated with faecal carriage of CTX-M beta-lactamase-producing Enterobacteriaceae in rural Thai communities. J Antimicrob Chemother, 2012; 67:1769–1774.

14.

Lautenbach

, Patel

, Bilker

, et al. Extended-spectrum beta-lactamase-producing Escherichia coli and Klebsiella pneumoniae: Risk factors for infection and impact of resistance on outcomes. Clin Infect Dis, 2001; 32:1162–1171.

15.

Colodner

, Rock

, Chazan

, et al. Risk factors for the development of extended-spectrum beta-lactamase-producing bacteria in nonhospitalized patients. Eur J Clin Microbiol Infect Dis, 2004; 23:163–167.

16.

Kuster

, Hasse

, Huebner

, et al. Risks factors for infections with extended-spectrum beta-lactamase-producing Escherichia coli and Klebsiella pneumoniae at a tertiary care university hospital in switzerland. Infection., 2010; 38:33–40.

17.

Johnson

, Anderson

, May

, Drew

. Utility of a clinical risk factor scoring model in predicting infection with extended-spectrum beta-lactamase-producing Enterobacteriaceae on hospital admission. Infect Control Hosp Epidemiol, 2013; 34:385–392.

18.

Oguz Mizrakci

, Arda

, Erdem

, et al. Risk factors for gastrointestinal colonization by ESBL-producing Klebsiella pneumoniae and Escherichia coli in anaesthesiology and reanimation intensive care unit. Mikrobiyol Bul, 2013; 47:223–229.

19.

Goldstein

EJC

, Snydman

. Intra-abdominal infections: Review of the bacteriology, antimicrobial susceptibility and the role of ertapenem in their therapy. J Antimicrob Chemother, 2004; 53:29–36.

20.

Waites

, Duffy

, Dowzicky

. Antimicrobial susceptibility among pathogens collected from hospitalized patients in the United States and in vitro activity of tigecycline, a new glycylcycline antimicrobial. Antimicrob Agents and Chemother, 2006; 50:3479–3484.

21.

Bhat

, Paul

, Ananthakrishna

. Drug resistant enterococci in a South Indian hospital. Trop Doct, 1998; 28:106–107.

22.

Teppler

, McCarroll

, Gesser

, Woods

. Surgical infections with enterococcus: Outcome in patients treated with ertapenem versus piperacillin-tazobactam. Surg Infect, 2002; 3:337–349.

23.

Edwards

. Enterococci attract attention of concerned microbiologists. ASM News, 2000; 66:540–545.

24.

Huycke

, Sahm

, Gilmore

. Multiple-drug resistant Enterococci: The nature of the problem and an agenda for the future. Emerg Infect Dis, 1998; 4:239–249.

25.

Sood

, Malhotra

, Das

, Kapil

. Enterococcal infections & antimicrobial resistance. Indian J Medical Res, 2008; 128:111–121.

26.

Roehrborn

, Thomas

, Potreck

, et al. The microbiology of postoperative peritonitis. Clin Infect Dis, 2001; 33:1513–1519.

27.

Blot

, De Waele

. Critical issues in the clinical management of complicated intra-abdominal infections. Drugs, 2005; 65:1611–1620.

28.

Nagalingam

, Lynch

. Role of the microbiota in inflammatory bowel diseases. Inflamm Bowel Dis, 2012; 18:968–980.