Microbiologic Data in Acute Cholecystitis: Ten Years' Experience from Bile Cultures Obtained during Percutaneous Cholecystostomy

Abstract

Background:

The aim of the study was to describe the microbiology and susceptibility patterns in acute cholecystitis by examining bile culture results from patients who underwent percutaneous cholecystostomy and examine concordance with empiric treatment.

Patients and Methods:

A total of 124 patients with acute cholecystitis underwent percutaneous cholecystostomy between 2003 and 2012 at Emek Medical Center, Israel. Data on bile and blood culture results, isolate susceptibility, and clinical outcomes were retrieved from patient files.

Results:

Bile cultures obtained from 116 patients were positive in 70 (60.3%) patients. Blood cultures obtained from 77 patients were positive in 23 (31.1%). Escherichia coli was the most common isolate in 28.6% of bile cultures and 43.5% of blood cultures. The concordance between empiric treatment coverage and culture isolate susceptibility was 67.6%. In most discordant cases, the isolates were Enterobacter spp. (40.9%) and Enterococcus spp. (31.8%). Overall, the in-hospital mortality rate was 7%: 2% in patients with concordant treatment compared with 14% in patients with discordant treatment (p = 0.09). Empiric antibiotic regimens were adequate in only two-thirds of patients.

Conclusions:

There might be a trend for poorer outcome in patients treated with inadequate antibiotic agents, emphasizing the importance of tailoring antibiotic treatment.

Acute cholecystitis most commonly results from obstruction of the cystic duct by gallstones [1,2]. Although bile is usually sterile, it may become infected by bacteria originating from the duodenum in the presence of gallstone obstruction and biliary stasis. According to different studies, the rate of bile culture positivity in patients with acute cholecystitis ranged from 35% to 60% [3 –5]. Most bacterial isolates are gram-negative Enterobacteriaceae, such as Escherichia coli, Klebsiella spp., Enterobacter spp., and less frequently gram-positive cocci such as Enterococcus spp. or anaerobic bacteria including Bacteroides and Clostridium spp.

The initial treatment for acute cholecystitis is surgical removal of the gallbladder, in conjunction with antibiotic therapy. Different guidelines recommend various antibiotic regimens that cover gram-negative Enterobacteriaceae with an option of coverage of gram-positive cocci and anaerobes as well, while stating that in any specific case, one should take into consideration disease severity, medical history, previous antibiotic exposure, and local antibiotic susceptibility patterns [6 –8]. During the past few years, there has been a worldwide increase in antibiotic resistance among Enterobacteriaceae, mediated mostly by production of extended spectrum beta-lactamases [9]. This might cause current empiric antibiotic regimens to be inadequate.

In high-risk surgical patients, specifically elderly persons and patients with co-existing illnesses, percutaneous cholecystostomy (PC) may be required to improve clinical outcomes in patients who do not improve sufficiently with antibiotic therapy [10,11]. Although this procedure carries a risk of complications, recent studies advocate its use as a relatively safe and effective procedure [3 –5,10 –13]. One study found higher morbidity rates compared with cholecystectomy [14]. A recent Cochrane review concluded that there is insufficient evidence in the literature to recommend cholecystostomy as an optional treatment for high-risk surgical patients with cholecystitis [15]. An American study from 2011 found that after cholecystostomy was performed in patients with acute cholecystitis, 68% improved clinically; while there was bacterial growth in 52% of bile cultures, positive cultures did not affect patient prognosis [4]. To the best of our knowledge, however, there are no recent data in the literature on susceptibility patterns of bacterial isolates from patients with acute cholecystitis.

The aims of our study were to describe the pathogens isolated and their susceptibility patterns in acute cholecystitis by examining bile cultures obtained from patients who underwent PC. In addition, we aimed to describe complication types and rates from this procedure, and to examine the concordance between empiric treatment coverage and culture isolate susceptibility.

Patients and Methods

The study population included all patients who underwent PC for the treatment of acute cholecystitis between January 2003 and December 2012 in the department of surgery “A” at Emek Medical Center, Afula, Israel. In cases in which several cholecystostomies were performed on the same patient, we only included the first procedure. Data that were extracted from the medical charts include: Demographic characteristics, existing co-morbidities, regular medications, and clinical parameters at hospital admission including vital signs, laboratory and imaging tests, and timing and types of antibiotic preparations that were administered.

In addition, we collected data concerning the procedure including date of performance, method (under ultrasonography [US] or computed tomography [CT] guidance), complications, and duration of drainage. Culture results from blood and bile were noted, including the susceptibility profile of each pathogen. Patient outcomes, as manifested by length of stay and death during hospitalization, were assessed. Concordance of empiric antibiotic therapy with culture results was assessed. Antibiotic treatment was considered concordant when it covered all pathogens that grew from bile and blood cultures, and discordant when there was an isolate in bile or blood cultures that was resistant to treatment. A comparison of outcomes between the group of patients with concordant versus discordant therapy was performed. The study was approved by the hospital's Helsinki Committee.

Statistical analysis

Categoric variables were presented using frequency and percentages; continuous variables are summarized using means and standard deviation or medians and interquartile range as appropriate. Categoric risk factors for in-hospital death were assessed using the chi-square test or Fisher exact test as appropriate, and continuous risk factors were assessed with the t-test or the Wilcoxon two sample test as appropriate. The p value of less than 0.05 for two-tailed tests was considered statistically significant. Statistical analysis was performed using SAS 9.2 software.

Results

During the 10-year study period, 721 patients were hospitalized with acute cholecystitis in the department of surgery at the Emek Medical Center. Of these patients, 124 (17.2%) underwent PC. The demographic and clinical characteristics of these patients, as well as vital signs and blood test results on admission, are summarized in Table 1.

Table 1.

Demographic Characteristics, Medical History, Vital Signs, and Blood Test Results on Admission of All Patients with Cholecystitis who Underwent Cholecystostomy (n = 124)

Age	Years, Mean (range)
	68 (33–96)
Gender	Number (%)
Male	63 (50.8%)
Female	61 (48.2%)
Vital signs on admission	Mean (range)
Temperature (^oC)	37.2 (36–40.2)
Pulse (beats per minute)	87 (50–129)
Systolic blood pressure (mm Hg)	136 (113–214)
Diastolic blood pressure (mm Hg)	77 (48–85)
Temperature before cholecystostomy (^oC)	38 (36.7–40)
Co-morbidity	Number (%)
Hypertension	78 (62.9%)
Diabetes mellitus	57 (46%)^**
Ischemic heart disease	38 (30.6%)
Atrial fibrillation	10 (12.4%)
Congestive heart failure	12 (9.7%)
Chronic lung disease	11 (8.9%)
Peripheral vascular disease	6 (4.8%)
Malignancy	5 (4%)
Medication use	Number (%)
Aspirin	54 (43.5%)
Insulin	9 (7.4%)
Warfarin	8 (4.8%)
Clopidogrel	6 (4.8%)
Corticosteroids	3 (2.4%)
Laboratory tests on admission	Mean (range)
White blood cells (/μL)	15,000 (3,400–31,500)
Hemoglobin (g/dL)	12.9 (9.3–1.18)
Platelets (/μL)	252,000 (83,000–688,000)
Serum creatinine (mg/dL)	1.12 (0.39–3.97)
Blood urea nitrogen (mg/dL)	44.6 (0.91–154)
Aspartate aminotransferase (IU/L)	102 (10–938)
Alanine transaminase (IU/L)	73 (7–482)
Alkaline phosphatase (IU/L)	131 (32–718)
Total bilirubin (mg/dL)	1.61 (0.3–7.78)
Direct bilirubin (mg/dL)	0.71 (0.02–5)
Glucose (mg/dL)	173 (54–539)

In all cases, imaging of the gallbladder and biliary tract was performed during hospitalization. In 121 cases (97.6%), US was performed, and in 72 cases (58.5%), an abdominal CT was performed. The pathologic radiographic findings included cholelithiasis in 103 (85.2%) cases in which US was performed, peri-cholecystic fluid in 20 (27.8%) of the CT studies, and a peri-cholecystic infiltrate in 10 (13.9%) CT studies.

All patients received empiric antibiotic therapy: 109 (87.9%) received cefuroxime and metronidazole, 10 (8.1%) received ceftriaxone and metronidazole, three (2.4%) received ciprofloxacin and metronidazole, and one (0.8%) each received ofloxacin and metronidazole and pipracillin-tazobactam (Table 2).

Table 2.

Antibiotic Regimens Used for Treatment of Patients with Cholecystitis Who Underwent Cholecystostomy (n = 124)

Antibiotic regimen	Patients treated N (percentage)
Cefuroxime and metronidazole	109 (87.9%)
Ceftriaxone and metronidazole	10 (8.1%)
Ciprofloxacin and metronidazole	3 (2.4%)
Ofloxacin and metronidazole	1 (0.8%)
Pipracillin-tazobactam	1 (0.8%)

Cholecystostomy was performed by a specialized interventional radiologist in all patients. In 87 (70.2%) patients, drainage was performed under US guidance, and in 37 (29.8%) patients, under CT guidance. Complications of this procedure were noted in 15 (12.1%) patients: Two (1.6%) patients had hemobilia, one of whom also had a perforation of the gallbladder (0.8%), and in 13 (10.5%) patients, peri-procedural signs of sepsis developed that improved on administration of fluids and continuation of antibiotic therapy. The PC was removed after a mean of 10.3 days (range 1–29 d).

In 122 patients, bile and/or blood was sent for culture. Bile cultures were obtained at the time of cholecystostomy insertion in 116 (93.5%) patients and were positive in 70 (60.34%); blood cultures were obtained during hospitalization in 77 (62.1%) patients and were positive in 23 of 74 (31.1%) patients (data were missing concerning three samples). The bacteria that were isolated from bile and blood cultures are depicted in Table 3, with Escherichia coli being the most prominent isolate in bile (28.6%) and blood cultures (43.5%). The same pathogen was isolated from bile and blood in 13 (19.12%) cases.

Table 3.

Results of Positive Bile (n = 70) and Blood (n = 23) Cultures of Patients with Cholecystitis who Underwent Cholecystostomy ^*

Pathogen	Isolates from bile no. (%)	Isolates from blood no. (%)	Same pathogen in blood and bile no. (% of isolate in bile or blood)
Escherichia coli	20 (28.57)	10 (43.47)	6 (23.08)
Klebsiella spp.	12 (17.14)	4 (17.39)	1 (7.69)
Enterobacter spp.	12 (17.14)	4 (17.49)	3 (25)
Enterococcus spp.	8 (11.42)	1 (4.3)	1 (12.5)
Other Streptococcus spp.	5 (7.14)	3 (13.04)	1 (14.28)
Bacteroides fragilis	1 (1.43)	0	0
Candida spp.	2 (2.85)	1 (4.3)	1 (50)
Other pathogen	12 (17.14)	0	0
Mixed flora^a	3 (4.28)	2 (8.6)	NA

Some cultures were positive for more than one pathogen.

Mixed flora–three or more pathogens isolated from one specimen.

To assess concordance of antibiotic treatment with culture susceptibility results, we excluded positive cultures with mixed flora (growth of three or more different pathogens in culture; in such cases, antibiotic susceptibility is not examined), leaving 68 patients who had positive bile or blood cultures. Concordance was found in 46 of 68 (67.6%) cases. In 22 (32.3%) patients, at least one bacterial isolate did not match empiric antibiotic therapy, and thus treatment was regarded as discordant. The discordant cases were: Nine (40.9%) cultures of resistant Enterobacter spp., seven (31.8%) enterococcal isolates, two (9.1%) E. coli isolates, two (9.1%) cases with Candida spp., one case (4.55%) of Klebsiella pneumonia, and one isolate (4.55%) of Serratia spp. In 21 cases (16.9% of all patients), antibiotic treatment was changed after receipt of culture results.

Examination of patient outcomes revealed that the mean length of hospital stay was 14.3 (1–37) days; in nine (7.3%) patients, an urgent cholecystectomy was performed, 10 (8.1%) patients were transferred to the intensive care unit, and nine (7.3%) patients died during hospitalization. Culture results of patients who died are summarized in Table 4.

Table 4.

Bile and Blood Culture Results of Patients with Cholecystitis Who Underwent Cholecystostomy and Died during Hospitalization (n = 9)

Patient	Empiric antibiotic treatment	Bile culture	Blood culture	Concordant/ discordant treatment
1	Cefuroxime and metronidazole	Enterococcus spp.	No growth	Discordant
2	Cefuroxime and metronidazole	Enterococcus spp.	No growth	Discordant
3	Ciprofloxacin and metronidazole	Candida spp.	Candida glabrata	Discordant
4	Cefuroxime and metronidazole	Mixed growth	Not obtained	NA
5	Cefuroxime and metronidazole	Not obtained	Streptococcus viridans	Concordant
6	Cefuroxime and metronidazole	Not obtained	Not obtained	NA
7	Cefuroxime and metronidazole	Not obtained	No growth	NA
8	Cefuroxime and metronidazole	No growth	No growth	NA
9	Cefuroxime and metronidazole	No growth	No growth	NA

In-hospital death was 6.7% in patients with positive cultures (bile or blood, n = 75) versus 4.2% (two of 47) in patients with negative bile and blood cultures (p = 0.7). All deaths were related to sepsis from underlying cholecystitis. Death was not found to be statistically significantly associated with diabetes mellitus (p = 0.3370) or more than three chronic diseases on admission (p = 0.2748). When we assessed whether concordant empiric antibiotic treatment matching bacterial susceptibility affected death, we found that in patients who received concordant therapy, death was 2.2% (one patient of 46) versus 13.6% (three patients of 22) in those who received discordant therapy, indicating a relative risk (RR) for in-hospital death of 6.28 with discordant therapy (p = 0.09).

Discussion

The relatively elderly age of patients undergoing cholecystostomy in our study is in line with previous studies [10,11,14,16,17] as well as the high prevalence of co-morbidities and chronic medication use [5], reflecting the high operative risk in these patients, and thus drainage is preferred. Fifteen percent of patients had acalculous cholecystitis, which is concordant with previous reports from series of patients with acute cholecystitis [5,11].

In our study, 12% of patients had complications of cholecystostomy. Other studies found 4.1%–19% complication rates [5,15,17]. The difference among complication rates might be attributed to different definitions: Whereas some studies included only local complications, others also included sepsis, which was the most common complication in our study. As concluded in other studies, we found this to be a relatively safe procedure. The mean time until tube extraction was 10 days in our study. Other studies reported mean duration of gallbladder drainage of 54–73 days [5,16]. Possibly, these longer drainage periods reflect the approach of leaving the drain in place until performance of cholecystectomy, which is not the customary approach in our center, where during the study period, the policy was removal of the drain 10 days after insertion.

In 60% of cases, the bile cultures obtained during PC were positive. Other studies found positive bile cultures in 35%–60% of cases [3 –5,18,19]. In our study, blood cultures were positive in 31%, with no data found in the literature from other studies. As expected, and similar to other published data [19], the most commonly isolated organisms were Enterobacteriaceae (61%), followed by streptococci, Candida, and anaerobes. On assessment of matching the empiric antibiotic treatment with bacterial susceptibility in blood and bile cultures, we found concordant treatment in 67.6% of cases, which might indicate the need for a broader spectrum empiric regimen. Discordant cases were mostly resistant Enterobacter and enterococcal isolates.

It might be that patients who undergo PC are more prone to resistant bacteria or enterococcal infection than other patients with cholecystitis (because of other risk factors, such as age, underlying illnesses, etc.). In addition, it is possible that after exposure to antibiotic agents, a shift occurs to pathogens that are not covered by the empiric antibiotic regimens (such as Enterococcus and Candida spp.).

Concerning patient outcomes, mean length of hospitalization was 14 days, similar to that reported in other studies (10–20.7 days) [11,14]. Nine patients (7.2%) died during hospitalization. Other studies found a 30-day mortality rate of 9%–15% [11]. Nikfarjam et al. [5] and Sanjay et al. [17] found a 37.7% mortality rate during the first year after cholecystostomy. These high mortality rates might reflect the fact that these patients are usually older and sicker than the average patients with cholecystitis who do not undergo this procedure. For patients who received concordant therapy, mortality was 2.2% versus 13.6% in those who received discordant therapy (RR 6.3, p = 0.09). In addition, we found that bile cultures were positive for enterococci in two of the patients who died, and bile and blood cultures grew Candida in another patient who died, meaning that one-third of in-hospital deaths occurred in patients who were not covered by the hospital's empiric regimen.

This might indicate the need for assessing patient-specific risk factors for infection with enterococci or Candida, such as previous antibiotic exposure and underlying illnesses, and tailoring therapy accordingly. On the other hand, the finding of positive cultures for Enterococcus and Candida spp. may be a marker of those who are ill instead of actually being an infectious cause of death. Both Candida and enterococcal infections are often seen in patients with compromised immunity. In our study, we did not find a correlation between baseline chronic diseases and death. This study emphasizes the importance of administering empiric antibiotic agents that match bacterial susceptibility according to local epidemiology. This is the first study that raises the question of the adequacy of empiric regimens in covering pathogens of cholecystitis and the clinical relevance of this matter.

This study has a several weaknesses. The retrospective nature of the study and the relatively low in-hospital deaths do not allow for clearly concluding whether concordant empiric therapy affects patient outcomes. Moreover, we did not assess other important outcomes, such as long-term death and recurrent episodes of cholecystitis.

Conclusions

Patients who underwent cholecystostomy for acute cholecystitis at our medical center had few peri-procedural complications; most received antibiotic therapy according to local guidelines. In most cases, cultures were positive, but therapy was concordant with bacterial susceptibility in only two-thirds of cases, with possible effects on in-hospital death. Prospective trials are needed to clarify the importance of concordant empiric therapy, and possibly guidelines of empiric therapy should be updated.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Jackson

, Evans

. Biliary system. In: Townsend

Jr , Beauchamp

, Evers

, Mattox

, eds. Sabiston Textbook of Surgery, 19th ed. Kidlington, UK: Elsevier, 2012: chap 54, 55.

Parmet

, Lynm

, Glass

. JAMA patient page. Acute cholecystitis. JAMA, 2003; 289:124.

Beardsley

, Shlansky-Goldberg

, Patel

, Freiman

. Predicting infected bile among patients undergoing percutaneous cholecystostomy. Cardiovasc Intervent Radiol, 2005; 28:319–325.

Joseph

, Unver

, Hwang

, et al. Percutaneous cholecystostomy for acute cholecystitis: Ten-year experience. J Vasc Interv Radiol, 2012; 23:83–88.

Nikfarjam

, Shen

, Fink

, et al. Percutaneous cholecystostomy for treatment of acute cholecystitis in the era of early laparoscopic cholecystectomy. Surg Laparosc Endosc Percutan Tech, 2013; 23:474–480.

Sartelli

, Viale

, Koike

, et al. WSES consensus conference: Guidelines for first-line management of intra-abdominal infections. World J Emerg Surg, 2011; 6:2.

Solomkin

, Mazuski

, Bradley

, et al. Diagnosis and management of complicated intra-abdominal infection in adults and children: Guidelines by the Surgical Infection Society and the Infectious Diseases Society of America. Clin Infect Dis, 2010; 50:133–164.

Yoshida

, Takada

, Kawarada

, et al. Antimicrobial therapy for acute cholecystitis: Tokyo Guidelines. J Hepatobiliary Pancreat Surg, 2007; 14:83–90.

Lynch

3rd , Clark

, Zhanel

. Evolution of antimicrobial resistance among Enterobacteriaceae (focus on extended spectrum β-lactamases and carbapenemases. Expert Opin Pharmacother, 2013; 14:199–210.

10.

McKay

, Abulfaraj

, Lipschitz

. Short- and long-term outcomes following percutaneous cholecystostomy for acute cholecystitis in high-risk patients. Surg Endosc, 2012; 26:1343–1351.

11.

Nasim

, Khan

, Alvi

, Chaudhary

. Emerging indications for percutaneous cholecystostomy for the management of acute cholecystitis—a retrospective review. Int J Surg, 2011; 9:456–459.

12.

Anderson

, Chang

, Talamini

. A nationwide examination of outcomes of percutaneous cholecystostomy compared with cholecystectomy for acute cholecystitis, 1998–2010. Surg Endosc, 2013; 27:3406–3411.

13.

Bakkaloglu

, Yanar

, Guloglu

, et al. Ultrasound guided percutaneous cholecystostomy in high-risk patients for surgical intervention. World J Gastroenterol, 2006; 12:7179–7182.

14.

Abi-Haidar

, Sanchez

, Williams

, Itani

. Revisiting percutaneous cholecystostomy for acute cholecystitis based on a 10-year experience. Arch Surg, 2012; 147:416–422.

15.

Gurusamy

, Rossi

, Davidson

. Percutaneous cholecystostomy for high-risk surgical patients with acute calculous cholecystitis. Cochrane Database Syst Rev, 2013; 8:CD007088.

16.

Byrne

, Suhocki

, Mitchell

, et al. Percutaneous cholecystostomy in patients with acute cholecystitis: Experience of 45 patients at a US referral center. J Am Coll Surg, 2003; 197:206–211.

17.

Sanjay

, Mittapalli

, Marioud

, et al. Clinical outcomes of a percutaneous cholecystostomy for acute cholecystitis: A multicenter analysis. HPB (Oxford), 2013; 15:511–516.

18.

Sosna

, Kruskal

, Copel

, et al. US-guided percutaneous cholecystostomy: Features predicting culture-positive bile and clinical outcome. Radiology, 2004; 230:785–791.

19.

Tejero

, Riofrío

, Aiquel

, et al. [Bacteriological study of bile from the gallbladder and bile ducts of patients surgically treated for biliary pathology]. (Spa) Enferm Infecc Microbiol Clin, 1990; 8:565–567.