Abstract
Abstract
Background:
The role of bactibilia as a risk factor for an unfavorable outcome during biliary disease or surgical procedures remains obscure. Our aim was to identify possible risk factors for bactibilia and their possible relations to complications after elective cholecystectomy. As secondary aims, composition and antibiotic resistance patterns were studied.
Methods:
Bile and gallbladder mucosa samples from 358 elective cholecystectomies were collected prospectively between June 2009 and June 2012. Ordinary microbiologic cultures and antibiograms were performed. All pre-operative factors associated with bactibilia were studied by stepwise logistic regression multivariable analysis.
Results:
The bacteria isolated most frequently from 103 positive cultures were Escherichia coli (21.3%), Enterococcus spp. (14.7%), and Enterobacter spp. (14.7%) with a global amoxicillin-clavulanic acid resistance rate of 53.7%. Age >65 y, male gender, previous instrumentation or disease of the biliary tract, and high American Society of Anesthesiologists score were independent risk factors. No correlation was found between bactibilia and surgical complications.
Conclusions:
Although the influence of bactibilia in developing surgical complications is limited, its composition and the high rate of resistance can be influential enough to modify antibiotic treatment in biliary tract infections, especially in high-risk patients.
B
The frequency and the microbial composition of bactibilia are not well known. Although gram-negative and anaerobic bacteria commonly are involved in biliary tract infections, recent results from several surveys after hepatobiliary surgery show a significant increase in previously unrelated organisms such as gram-positive or multi-drug-resistant enteric bacteria [1–4,7]. As a consequence, the presence of bactibilia should be related to a higher rate of post-operative infectious complications. However, this relation remains inconsistent because the incidence of surgery-related infections after elective cholecystectomy, even in the presence of complications, remains low. Moreover, the need for preventive measures such as antibiotic prophylaxis during elective cholecystectomy in the absence of risk factors is still under discussion [14,15].
Our aim was to study the presence of bactibilia and its causes in a prospective cohort of 368 patients who underwent elective cholecystectomy. Additionally, antibiotic resistance was recorded, and a multivariable analysis was performed to identify patient- or procedure-related risk factors for bactibilia as a post-operative complication.
Patients and Methods
Study design, clinical features, and exclusion criteria
Data from 368 patients who underwent elective cholecystectomy between June 2009 and June 2012 at our institution were included in a prospective observational and multidisciplinary study. The main inclusion criteria for the study were an indication for elective cholecystectomy and absence of any clinical or radiographic signs or laboratory data that suggested acute biliary disease. Cholecystectomy was performed, not only as an isolated procedure, but also during other elective major operations. Diagnosis of acute biliary disease was based on clinical criteria (right upper-quadrant pain; tenderness or fever 48 h before surgery), laboratory data (leukocytosis; abnormal liver function tests), or positive radiographic studies (ultrasonography or magnetic resonance imaging). Patients with hydrops, gallbladder empyema, or pathologically confirmed cholecystitis after surgery were excluded.
Endpoints
For the primary endpoint, we considered the presence of bactibilia and its causes. Bactibilia was defined as a positive bile or gallbladder mucosa culture. Secondary endpoints were patient characteristics (age, gender, body mass index [BMI], presence of diabetes mellitus, American Society of Anesthesiologists [ASA] score, previous biliary tract disease and fever or right upper quadrant pain during the last 2 mos but not the immediate 48 h before surgery), or procedure (indication, surgical approach, previous biliary tract manipulation, pre-operative antibiotic administration) as risk factors for bactibilia relative to the post-operative outcome (surgical site infection according to the U.S. Centers for Disease Control and Prevention [CDC] definitions) [16].
Samples, procurement, and processing
Microbial cultures
Bile and gallbladder mucosa samples were obtained under aseptic conditions after the surgical specimen was removed. Four milliliters of bile were collected via direct puncture of the specimen with a 16F catheter, and a 1.5-cm2 mucosal biopsy was taken; both were placed immediately in an additive-free sterile tube before being processed. Both bile and mucosal samples were cultured in aerobic, enriched, selective, and differential media (blood agar, chocolate agar plus PolyVitex, McConkey agar, colistin–nalidixic acid agar, Schaedler agar, kanamycin–vancomycin agar, Salmonella–Shigella agar, and selenite-enriched solution (bioMérieux, Marcy l'Etoile, France) for at least 48 h under aerobic and anaerobic conditions at 35°–37°C. Microorganisms were identified initially by direct visual examination during incubation; later, genus and species characterization was performed using biochemical methods (API 10; bioMérieux) following the manufacturer's guidelines. Finally, antimicrobial susceptibility testing was performed using disk diffusion methods according to the Clinical and Laboratory Standards Institute (CLSI) guidelines [17]. Isolates of Escherichia coli, Klebsiella spp., and Enterobacter spp. with reduced susceptibilities to cefotaxime or ceftazidime according to the guidelines for laboratory detection of extended-spectrum beta-lactamases (ESBLs) from the CDC were examined by the E-test method (AB Biodisk Inc., Solna, Sweden). Susceptibility breakpoints were determined according to the recommendations of the CLSI.
Data collection and statistical analysis
Demographic, clinical, and microbiologic records from every patient were registered prospectively in an ad hoc database. Files were reviewed monthly, and if any missing item was detected, the electronic medical charts (SAP-SE, Walldorf, Baden-Württemberg, Germany) were reviewed, and the patients were contacted by telephone if necessary to provide it. The study received approval from our Institution's Ethics Committee, and informed consent was obtained from all the patients. Recommendations from the Spanish Clinical Data and Patient's Autonomy Law (41/2002) were strictly observed. The correlation between variables and the identification of possible risk factors for bactibilia was conducted through univariable (χ2 and Fisher exact tests) and multivariable analysis plus forward stepwise logistic regression (SPSS v 13.0 for Microsoft Windows, IBM Inc., Armonk, NY). A 95% confidence level was considered significant.
Results
Descriptive data
During this study, we recruited 368 patients. After applying the exclusion criteria, the final group consisted of 358 patients. Of these, 134 patients were male. The median age was 57 y with a range of 16–88 y. In the entire patient cohort, 103 cultures (28.7%) were positive for bactibilia. From those cultures, we isolated 136 microorganisms (Table 1). In total, there were 27 patients with two microorganisms in their cultures, two patients with three isolates, and 74 patients with a single isolate. The bacteria isolated most frequently were E. coli (29 cases), followed by Enterococcus spp. (20 cases), Enterobacter spp. (20 cases), and Klebsiella spp. (19 cases). Other microorganisms were Pseudomonas (five cases), other Enterobacteriaceae (14 cases), Stenotrophonomas maltophilia (two cases), Bacteroides spp. (five cases), Clostridium spp. (three cases), viridans streptococci (10 cases), Leuconostoc spp. (one case), and yeasts (eight cases). When cultures from laparoscopic cholecystectomies were studied, 43 of 218 (19.62%) were positive. The bacteria isolated were E. coli (seven cases); Enterococcus faecalis (six cases); Enterobacter cloacae (five cases); Klebsiella spp. (five cases); other Enterobacteriaceae (two cases), Pseudomonas (two cases); one case each of ESBL-producing E. cloacae, Enterococcus faecium, and Candida albicans; mixed flora (10), and other gram-positive microorganisms (three cases).
Surgical aspects
The most common indication for cholecystectomy was symptomatic cholelithiasis (Table 2), and the most common procedure was laparoscopic cholecystectomy (LC) (Table 3). In order to determine whether there was a relation between surgical site infection (SSI) and bactibilia, we examined the data from the entire group of cholecystectomies. We found 44 SSIs (11.95%), as shown in Table 4. In 15 cases, a site culture was positive, as was true in 88 of the sites that were not infected (odds ratio [OR] 1.32; 95% CI 0.68–2.60; p=0.4) (Table 5). Additionally, we found no relation between these two variables. However, the incidence of SSI after LC was 4.1% (9/218). In this particular group, 190 patients (87.15%) received prophylactic antibiotics according to our institution's protocol. Thirty-nine patients with bactibilia (88%) received prophylactic antibiotics prior to LC, and in 36 of them (92%), the prescription was according to the protocol recommendations. Neither major nor minor complications were observed after LC excepting for nine cases of SSI. No significant correlation between the presence of SSI after LC and the use of prophylactic antibiotic (3.8 vs. 7.6%; p=0.356) or the presence of bactibilia (4.8 vs. 4.16%; p=0.85) was found. Cultures were obtained from every SSI case after LC. Two of them (22%) were positive, one for E. faecalis plus Morganella morganii and one for E. faecalis plus Serratia marcescens (in this case, the same microorganisms that were found in the bile culture).
In obese patients, cholecystectomy was performed during bariatric surgery only if symptomatic cholelithiasis was present.
Associated factors and predictors of bactibilia
All of the pre-operative factors studied were compared with bactibilia in a univariable analysis. Several factors have been associated with bactibilia, such as age (>65 years), male gender, hospital admission (from emergency departments or from another hospital), antibiotic treatment (two mos before surgery), history of biliary pathology, history of instrumentation of the biliary duct, high ASA score (class III or IV), and fever and pain before surgery (Table 6). All pre-operative factors associated with bactibilia in univariate analyses were studied by stepwise logistic regression multivariable analysis. Age >65 y, male gender, history of biliary pathology, history of instrumentation of the biliary duct, and high ASA score were significant by multivariable analysis (Table 7).
ASA=American Society of Anesthesiologists; CI=confidence interval; NS=not significant.
ASA=American Society of Anesthesiologists; CI=confidence interval; ED=emergency department.
Antibiotic resistance patterns
We performed antibiograms on every isolated microorganism. In the four most frequently isolated microorganisms (E. coli, Enterococcus spp., Enterobacter spp., and Klebsiella spp.), we identified 47% resistance to amoxicillin-clavulanic acid with a good rate of susceptibility to the third-generation cephalosporins. In Table 8, we show the most important antibiotic resistance patterns. In summary, we showed a 53.8% rate of resistance to amoxicillin-clavulanic acid. When resistances to these drugs was studied only in those patients who underwent LC, a global 54.05% rate was observed: 36% (12/33) in Enterobacteriaceae and other non-fermenting gram-negative bacteria (one of them with acquired resistance) and 57% (8/14) in Enterococcus, streptococci, yeasts, and other aerobic flora. No relation was found between resistance to amoxicillin-clavulanic acid and the presence of diabetes mellitus (p=0.453), male gender (p=0.864), or ASA score (p=0.545). However, those patients who had received antibiotic treatment before LC showed higher resistance rates (p=0.025).
Discussion
Bactibilia is believed to represent a microbiologic population that colonizes bile during a chronic infection; however, it is also believed to be a population that becomes pathogenic during acute biliary infection. With this consideration in mind, the present study was designed to evaluate the relation between the presence of bacteria in the bile and several aspects of the clinical outcome after elective cholecystectomy. The reported prevalence of bactibilia in elective cholecystectomy is highly variable (between 8% and 42%) [6] and depends on many factors, not only the patient's characteristics but also the clinical management. In our work, the observed overall prevalence of bactibilia was 28.7%. However, when only laparoscopic cholecystectomies were considered, the incidence diminished to 19.6%. In our study, several factors, including an age greater than 65 y, male gender, malignant disease, admission from the emergency department or other institutions (regional community hospital), previous instrumentation or history of biliary tract disease, presence of pain or fever two mos before surgery, high ASA score, and antibiotic treatment in the two mos before surgery were associated with bactibilia. After multivariable analysis, age, previous biliary duct disease or manipulation, and high ASA score remained independent risk factors.
A relation between aging and bactibilia was reported by Illig et al. [8], who recommended that antibiotic prophylaxis be used for simple LCs in patients older than 65 y. Although aging is related to several pathophysiologic phenomena involved in the development of bactibilia, such as biliary dyskinesia [10], our belief is that the true importance of aging as a risk factor is the result of an additive effect of all of the conditions found in older patients. The presence of bactibilia in patients with higher ASA scores could be explained similarly; for instance, patients with higher ASA scores most often suffer risk factors traditionally associated with SSI, such as diabetes mellitus or liver disease. Therefore, the score may represent a summation of established diseases.
There are several papers that correlate manipulation of the biliary tract and previous biliary disease with bactibilia [6,7]. In both cases, the obstruction of bile flow is constant, establishing the conditions for colonization. Moreover, biliary manipulations can create a direct communication with the duodenum, especially if papillotomy is performed during endoscopic retrograde cholepancreatography (ERCP) [13]. Usually, patients with a history of biliary tract disease or instrumentation suffer pain or fever before surgery. In addition, they often are admitted through the emergency department or are referred from smaller hospitals to tertiary care institutions such as ours. Furthermore, at the time of admission, most patients have received antibiotic treatment, which may explain the fact that only the first two variables have been identified as independent risk factors.
Although the presence of bactibilia is related to the development of post-operative complications, specifically in patients with obstructive jaundice [12], in our series, there was no correlation between bactibilia and the presence of unfavorable outcomes. Despite the fact that severe complications are uncommon after elective cholecystectomy, minor events such as SSI still can occur [18]. Nevertheless, the incidence of SSI was not related to the presence of bactibilia [19,20], which could be explained not only by a high antibiotic prophylaxis protocol success rate (87%) but also by the inclusion of samples from patients who underwent LC. In this particular group, the use of prophylactic antibiotics was not related to lower SSI rates, and the incidence of SSI after LC was 4.1%. It therefore is difficult to assess the influence of bactibilia on SSI. In our series, only two patients with SSI after LC had bactibilia. Nevertheless, in one of them, the microorganisms isolated from bile and surgical site samples were the same. In addition, both SSIs yielded microorganisms with intrinsic high amoxicillin-clavulanic acid resistance, namely, S. marcescens and M. morganii. Our institutional protocol recommends 1 g of amoxicillin-clavulanic acid (with ciprofloxacin plus metronidazole as an alternative) intravenously 30 min before incision. If the procedure is longer than 3 h, another dose should be administered. We have found high rates of resistance to amoxicillin-clavulanic acid even in LC. In addition, the most important risk factor for bactibilia harboring resistant microorganisms was the use of antibiotics prior to the surgical procedure. According to our results, the high incidence of amoxicillin-clavulanic acid-resistant microorganisms in both bile and surgical incision cultures suggests that, in our context, antibiotic prophylaxis recommendations could be reevaluated and antibiotic prescription policies revised. In addition, we are aware that the presence of bactibilia was not the only risk factor involved when SSI was detected, especially in major procedures. Organ-space infections were associated with major resections in which bactibilia could be only a co-factor in the complications.
The clinical goal of our work was to identify the composition and the antibiotic resistance patterns of the microorganisms that colonize the biliary tract and cause biliary sepsis under favorable conditions. The presence of bactibilia is related to several variables that are becoming more frequent in patients who undergo elective cholecystectomy. Moreover, it does not seem to play a pivotal role in the development of post-operative complications, especially SSI [18,21]. However, it provides important information to determine the most probable etiology in cases of biliary sepsis. In addition, although our microbiologic findings, especially the high rate of resistance, are not particularly surprising, our results could influence the choice or empiric treatment of infection or change prophylaxis policy [22].
Author Disclosure Statement
The authors declare that no competing financial interests exist.