Population-based study of the epidemiology of and the risk factors for pyogenic liver abscess

Background & Aims: Pyogenic liver abscess (PLA) is relatively uncommon in North America but is associated with significant morbidity and mortality. Our objective was to characterize the incidence of, risk factors for, and outcomes of PLA in a nonselected population. Methods: Population-based surveillance was conducted in the Calgary Health Region (CHR) between April 1, 1999 and March 31, 2003. All adult CHR residents with PLA were identified, and charts were reviewed. Results: Seventy-one CHR residents developed a PLA for an annual incidence of 2.3 per 100,000 population. There was an increasing incidence of PLA with advancing age. Men were at much higher risk of acquiring a PLA as compared to women (3.3 vs 1.3 per 100,000; relative risk [RR], 2.6; 95% confidence interval [CI], 1.5–4.6; P < .001), and this was observed across all age groups. A number of comorbid conditions were associated with significantly higher risk for developing a PLA including liver transplantation patients (RR, 444.8; 95% CI, 89.5–1356.0; P < .0001), diabetics (RR, 11.1; 95% CI, 6.3–19; P < .0001), and patients with a history of malignancy (RR, 13.3; 95% CI, 6.9–24.4; P < .0001). No other solid organ transplantation patient was at increased risk. All patients required admission to hospital (median length of stay, 16 days), and 7 (10%) patients died in hospital, corresponding to a mortality rate of 0.22 per 100,000 population. Conclusions: This study provides important data on the burden of PLA and identifies risk groups that might potentially benefit from preventive efforts. Background & Aims: Pyogenic liver abscess (PLA) is relatively uncommon in North America but is associated with significant morbidity and mortality. Our objective was to characterize the incidence of, risk factors for, and outcomes of PLA in a nonselected population. Methods: Population-based surveillance was conducted in the Calgary Health Region (CHR) between April 1, 1999 and March 31, 2003. All adult CHR residents with PLA were identified, and charts were reviewed. Results: Seventy-one CHR residents developed a PLA for an annual incidence of 2.3 per 100,000 population. There was an increasing incidence of PLA with advancing age. Men were at much higher risk of acquiring a PLA as compared to women (3.3 vs 1.3 per 100,000; relative risk [RR], 2.6; 95% confidence interval [CI], 1.5–4.6; P < .001), and this was observed across all age groups. A number of comorbid conditions were associated with significantly higher risk for developing a PLA including liver transplantation patients (RR, 444.8; 95% CI, 89.5–1356.0; P < .0001), diabetics (RR, 11.1; 95% CI, 6.3–19; P < .0001), and patients with a history of malignancy (RR, 13.3; 95% CI, 6.9–24.4; P < .0001). No other solid organ transplantation patient was at increased risk. All patients required admission to hospital (median length of stay, 16 days), and 7 (10%) patients died in hospital, corresponding to a mortality rate of 0.22 per 100,000 population. Conclusions: This study provides important data on the burden of PLA and identifies risk groups that might potentially benefit from preventive efforts. Pyogenic liver abscess (PLA) is a relatively uncommon illness affecting North Americans, but it has been associated with significant morbidity, mortality, and consumption of healthcare resources. 1Shrara A.I. Rockey D.C. Pyogenic liver abscess.Curr Treat Options Gastroenterol. 2002; 5: 437-442Crossref PubMed Google Scholar However, the epidemiology of PLA has not been well-defined. Only one study has defined the incidence of PLA. 2Hansen P.S. Schonheyder H.C. Pyogenic hepatic abscess a ten year population-based retrospective study.APMIS. 1998; 106: 396-402Crossref PubMed Scopus (69) Google Scholar Hansen and Schonheyder 2Hansen P.S. Schonheyder H.C. Pyogenic hepatic abscess a ten year population-based retrospective study.APMIS. 1998; 106: 396-402Crossref PubMed Scopus (69) Google Scholar conducted a population-based study in the County of Northern Jutland, Denmark and identified 52 cases of PLA during a 10-year period for an annual incidence of 1.1 per 100,000. Although several case series have suggested that a number of factors might increase the risk of acquiring a PLA including biliary disease, diabetes, malignancy, and hepatobiliary surgery, no studies to date have been adequately designed to establish actual risk factors for acquiring PLA in a nonselected population. 3Huang C.J. Pitt H.A. Lipsett P.A. Osterman Jr, F.A. Liliemoe K.D. Cameron J.L. Zuidema G.D. Pyogenic hepatic abscess changing trends over 42 years.Ann Surg. 1996; 223: 600-607Crossref PubMed Scopus (403) Google Scholar, 4Chou F.F. Sheen-Chen S.M. Chen Y.S. Chen M.C. Chen F.C. Tai D.I. Prognostic factors for pyogenic abscess of the liver.J Am Coll Surg. 1994; 179: 727-732PubMed Google Scholar, 5Lee K.T. Wong S.R. Sheen P.C. Pyogenic liver abscess an audit of 10 years’ experience and analysis of risk factors.Dig Surg. 2001; 18: 459-466Crossref PubMed Scopus (133) Google Scholar, 6Chu K.M. Fan S.T. Lai E.C. Lo C.M. Wong I. Pyogenic liver abscess an audit of experience over the past decade.Arch Surg. 1996; 131: 148-152Crossref PubMed Scopus (128) Google Scholar, 7Barakate M.S. Stephen M.S. Waugh R.C. Gallagher P.J. Solomon M.J. Storey D.W. Sheldon D.M. Pyogenic liver abscess a review of 10 years’ experience in management.Aust NZ J Surg. 1999; 69: 205-209Crossref PubMed Scopus (109) Google Scholar, 8Mischinger H.I. Hauser H. Rabl H. Quehenberger F. Werkgartner G. Rubin R. et al.Pyogenic liver abscess study of therapy and analysis of risk factors.World J Surg. 1994; 18: 852-857Crossref PubMed Scopus (82) Google Scholar, 9Perez J.A.A. Gonzalez J.J. Baldonedo R.F. Sanz L. Carreno G. Junco A. Rodriguez J.I. Martinez M.D. Jorge J.I. Clinical course, treatment and multivariate analysis of risk factors for pyogenic liver abscess.Am J Surg. 2001; 181: 177-186Abstract Full Text Full Text PDF PubMed Scopus (152) Google Scholar, 10Wong W.M. Wong B.C.Y.W. Hui C.K. Ng M. Lai K.C. Tso W.K. Lam S.K. Lai C.L. Pyogenic liver abscess retrospective analysis of 80 cases over a 10-year period.J Gastroenterol Hepatol. 2002; 17: 1001-1007Crossref PubMed Scopus (146) Google Scholar, 11Mohsen A.H. Green S.T. Read R.C. McKendrick M.W. Liver abscess in adults ten years experience in a UK centre.QJM. 2002; 95: 797-802Crossref PubMed Scopus (179) Google Scholar, 12Furugaki K. Chijiiwa K. Ogawa T. Tanaka M. Pyogenic liver abscess after hepatobiliary and pancreatic surgery.Int Surg. 2001; 86: 67-71PubMed Google Scholar In addition, the mortality rate associated with PLA has not been well-established. Hospital-based case series that are highly subject to selection bias have reported case-fatality rates that vary between 6%–32%. 3Huang C.J. Pitt H.A. Lipsett P.A. Osterman Jr, F.A. Liliemoe K.D. Cameron J.L. Zuidema G.D. Pyogenic hepatic abscess changing trends over 42 years.Ann Surg. 1996; 223: 600-607Crossref PubMed Scopus (403) Google Scholar, 4Chou F.F. Sheen-Chen S.M. Chen Y.S. Chen M.C. Chen F.C. Tai D.I. Prognostic factors for pyogenic abscess of the liver.J Am Coll Surg. 1994; 179: 727-732PubMed Google Scholar, 5Lee K.T. Wong S.R. Sheen P.C. Pyogenic liver abscess an audit of 10 years’ experience and analysis of risk factors.Dig Surg. 2001; 18: 459-466Crossref PubMed Scopus (133) Google Scholar, 6Chu K.M. Fan S.T. Lai E.C. Lo C.M. Wong I. Pyogenic liver abscess an audit of experience over the past decade.Arch Surg. 1996; 131: 148-152Crossref PubMed Scopus (128) Google Scholar, 7Barakate M.S. Stephen M.S. Waugh R.C. Gallagher P.J. Solomon M.J. Storey D.W. Sheldon D.M. Pyogenic liver abscess a review of 10 years’ experience in management.Aust NZ J Surg. 1999; 69: 205-209Crossref PubMed Scopus (109) Google Scholar, 8Mischinger H.I. Hauser H. Rabl H. Quehenberger F. Werkgartner G. Rubin R. et al.Pyogenic liver abscess study of therapy and analysis of risk factors.World J Surg. 1994; 18: 852-857Crossref PubMed Scopus (82) Google Scholar, 9Perez J.A.A. Gonzalez J.J. Baldonedo R.F. Sanz L. Carreno G. Junco A. Rodriguez J.I. Martinez M.D. Jorge J.I. Clinical course, treatment and multivariate analysis of risk factors for pyogenic liver abscess.Am J Surg. 2001; 181: 177-186Abstract Full Text Full Text PDF PubMed Scopus (152) Google Scholar, 10Wong W.M. Wong B.C.Y.W. Hui C.K. Ng M. Lai K.C. Tso W.K. Lam S.K. Lai C.L. Pyogenic liver abscess retrospective analysis of 80 cases over a 10-year period.J Gastroenterol Hepatol. 2002; 17: 1001-1007Crossref PubMed Scopus (146) Google Scholar, 11Mohsen A.H. Green S.T. Read R.C. McKendrick M.W. Liver abscess in adults ten years experience in a UK centre.QJM. 2002; 95: 797-802Crossref PubMed Scopus (179) Google Scholar The only population-based study reported a case-fatality rate of only 6%. 2Hansen P.S. Schonheyder H.C. Pyogenic hepatic abscess a ten year population-based retrospective study.APMIS. 1998; 106: 396-402Crossref PubMed Scopus (69) Google Scholar The incidence of and mortality associated with PLA in a North American population have not been defined. Furthermore, no studies to date have evaluated actual risk factors for acquiring PLA in a nonselected population. We conducted a population-based surveillance cohort study in the largest fully integrated health region in Canada to define the incidence of, quantify the risk factors for, and determine the outcomes of PLA. The Calgary Health Region (CHR) provides virtually all medical and surgical care to more than 1 million residents of the cities of Calgary and Airdrie and several surrounding communities. 13Calgary Health Region. Available at: http://www.ozcan.ca/chr_map/.Google Scholar Only patients requiring liver, heart, and/or lung transplantation surgery are routinely referred out of the region, although much of their subsequent care is provided in the CHR. The CHR is composed of approximately 80% white, 12% Asian, 3% First Nations/Metis, and 5% other people, and its age and gender distribution is comparable to many other North American centers. 14Calgary Health Region Web site. Available at: http://www.calgaryhealthregion.ca//hocr/influ/demo/demo.htm.Google Scholar All adult (18 years of age or older) residents of the CHR diagnosed with a PLA between April 1, 1999 and March 31, 2003 were included in this study. The study utilized a population-based active-surveillance design. Study subjects were identified through 2 complementary mechanisms. First, all International Classification of Diseases 9 and 10 codes for PLA were searched by using the CHR Data Warehouse, which contains administrative data on all emergency department and hospital admission encounters in the region. Second, all clinical samples taken from liver submitted for microbiology testing were identified by the Calgary Laboratory Services, a regional-based microbiology laboratory that handles all routine bacterial specimens from CHR patients in hospitals and the community. 15Church D. Hall P. Centralization of a regional clinical microbiology service the Calgary experience.Can J Infect Dis. 1999; 10: 393-402PubMed Google Scholar Once potential study candidates were identified, medical records were then reviewed by using a standardized form. Information abstracted included confirmation of the diagnosis of PLA and assessment of demographic, clinical, radiologic, and laboratory data. The presence of potential comorbid illnesses including the presence of a solid organ transplantation, human immunodeficiency virus seropositivity, diabetes mellitus, current or past history of malignancy, chronic dialysis dependence (hemodialysis or peritoneal), alcoholism, cirrhosis, hepatobiliary surgery, rheumatoid arthritis, and systemic lupus erythematosus were recorded by using a priori–defined criteria. Patient management including therapies and requirement and duration for hospital and intensive care unit admission were recorded. Survival was assessed to the point of hospital discharge. The study was approved by the Conjoint Health Research Ethics Board before its commencement. PLA was diagnosed if at least one of the following criteria were met: (1) liver aspirate or surgical tissue culture positive from a discrete liver abscess, (2) diagnostic imaging consistent with liver abscess with a concurrent positive blood culture, or (3) diagnostic imaging consistent with liver abscess and a defined radiologic and clinical response to antimicrobial management even if cultures were negative or not performed. Patients with amoebic hepatic lesions solely were excluded. Community-acquired infections were those that were either present or incubating at the time of hospital admission or associated with the first positive culture result obtained within 48 hours of admission. 16Garner J. Jarvis W. Emori T. Horan T.C. Hughes J.M. CDC definitions for nosocomial infections.Am J Infect Control. 1988; 16: 128-140Abstract Full Text PDF PubMed Scopus (5194) Google Scholar The time to diagnosis was defined as the date of symptom onset to date of tissue diagnosis (or date of radiologic identification when culture not obtained). The time to presentation was defined as the date of symptom onset to date of first physician assessment. Fever was defined as a temperature greater than 38.3°C within 24 hours of presenting to medical attention. Shock was defined as systolic blood pressure less than 90 mm Hg, unresponsive to fluid resuscitation, or requirement for vasopressor infusion. Antimicrobial-resistant organisms were defined as methicillin-resistant Staphylococcus aureus, vancomycin-resistant Enterococcus faecalis or faecium, penicillin-resistant Streptococcus pneumoniae, or any gram-negative organisms resistant to 1 or more of ciprofloxacin, tobramycin, gentamicin, ceftazidime, piperacillin, or imipenem. Local data sources that account for all patients in the CHR were used to either precisely ascertain or estimate the yearly prevalence of a priori selected potential risk factors for PLA. The size and demographic profile of the CHR adult population at risk were established by using 2001 national census data from Statistics Canada with yearly adjustment for population growth based on data from the Alberta Health Population Registry. 17Statistics Canada Web site. Available at: http://www.statcan.ca/english/concepts/health.Google Scholar The number of patients residing in the CHR who were solid organ transplantation recipients was ascertained from the Provincial Transplant Program Research Office at the University of Alberta. The number of human immunodeficiency virus–seropositive residents in the CHR was estimated by using statistics from the Southern Alberta Clinic. 18Southern Alberta Clinic Web site. Available at: http://www.crha-health.ab.ca/clin/sac/epidprov.htm.Google Scholar The number of patients requiring chronic dialysis was obtained from a regional database. 19Manns B. Mortis G. Taub K. McLaughlin K. Donaldson C. Ghali W.A. The Southern Alberta Renal Program database a prototype for patient management and research initiatives.Clin Invest Med. 2001; 24: 164-170PubMed Google Scholar The prevalence of chronic medical conditions including alcoholism, diabetes mellitus, and history of a present or past malignancy was estimated by using Calgary-specific data from the 2002 Canadian Community Health Survey. 17Statistics Canada Web site. Available at: http://www.statcan.ca/english/concepts/health.Google Scholar The prevalence of systemic lupus erythematosus and rheumatoid arthritis was estimated by using American data. 20Hochberg M. Perlmutter D. Medsger T. Steen V. Weisman M.H. White B. Wigley F.M. Prevalence of self-reported physician-diagnosed systemic lupus erythematosus in the USA.Lupus. 1995; 4: 454-456Crossref PubMed Scopus (78) Google Scholar, 21Gabriel S. Crowson C. O’Fallon W. The epidemiology of rheumatoid arthritis in Rochester, Minnesota, 1955-1985.Arthritis Rheum. 1999; 42: 415-420Crossref PubMed Scopus (303) Google Scholar Analysis was performed with Stata version 8.0 (Stata Corp, College Station, TX) software program. Normally distributed variables were reported as means ± standard deviations (SDs) and non-normally distributed variables as medians with interquartile ranges (IQRs). Category-specific risks for developing PLA associated with a predisposing condition were calculated as described elsewhere 21Gabriel S. Crowson C. O’Fallon W. The epidemiology of rheumatoid arthritis in Rochester, Minnesota, 1955-1985.Arthritis Rheum. 1999; 42: 415-420Crossref PubMed Scopus (303) Google Scholar and reported as relative risks (RRs) with exact 95% confidence intervals (CIs). Two-sided P values less than .05 were considered to be significant for all comparisons. During the 4-year study 71 residents of the CHR were identified with a PLA for an annual incidence of 2.3 per 100,000. All patients were admitted to hospital, and medical records were available for review in all cases. The yearly incidence was stable, and no seasonal trend in occurrence was observed. Fifty-one (72%) patients were male, and the mean ± SD age was 62.4 ± 14 years. There was an increasing population incidence of PLA observed in association with advancing age as shown in Figure 1. Patients aged 65 years and older were 10 times more likely to acquire a PLA as compared to younger individuals. Men were at an increased risk of acquiring a PLA compared to women (3.3 vs 1.3 per 100,000; RR, 2.6; 95% CI, 1.5–4.6; P < .001), and this was observed across all age groups (Figure 1). Liver transplantation, diabetes, alcoholism, and history of a malignancy were identified as significant risk factors for developing a PLA as shown in Table 1. No cases of PLA were observed among patients with human immunodeficiency virus, on chronic dialysis, or with solid organ transplantation other than liver. Rheumatoid arthritis (1 patient) and systemic lupus erythematosus (1 patient) were not significantly associated with the development of a PLA.Table 1Factors Assessed for an Increased Risk of Acquiring a Pyogenic Liver Abscess Among Residents of the CHR, CanadaUnderlying conditionNo. of patients (n = 71)Estimated no. with underlying condition at riskaEstimated 4-year prevalence of the underlying condition in the CHR.Incidence per 100,000bIncidence of PLA among those with the underlying condition.RR (95% CI)cRR for developing PLA among those with as compared to those without the underlying condition.P valueLiver transplant3309970.9444.8 (89.5–1356.0)<.0001Cancer1456,08123.113.3 (6.9–24.4)<.0001Diabetes20105,93118.911.1 (6.3–19)<.0001Alcoholism786,0018.64.1 (1.6–8.9).003Rheumatoid arthritis131,1563.21.4 (0.04–8.1).3Systemic lupus erythematosus111,5908.63.8 (0.1–22).1a Estimated 4-year prevalence of the underlying condition in the CHR.b Incidence of PLA among those with the underlying condition.c RR for developing PLA among those with as compared to those without the underlying condition. Open table in a new tab Of the 14 patients who had a prior diagnosis of malignancy, 5 (36%) had developed a PLA within 1 year of their cancer diagnosis. Half of the patients had been diagnosed with a malignancy within the gastrointestinal tract (2 colorectal, esophageal, hepatocellular, pancreatic, cholangiocarcinoma, and gastric lymphoma), 6 (43%) had a hematologic malignancy (2 acute myelogenous leukemia, and 1 each of non-Hodgkin’s lymphoma, gastric lymphoma, chronic lymphocytic leukemia, and multiple myeloma), and the final 2 patients had prostate cancer. A number of other potential risk factors for developing a PLA were identified; however, RRs were not calculated because adequate denominator data were not available. These potential risk factors included 4 patients (6%) who had a prior choledochojejunostomy, 10 (14%) who had a prior cholecystectomy, 4 (6%) who had a prior splenectomy, and 4 (6%) with underlying cirrhosis. The median time to presentation was 7 (IQR, 4–14) days, whereas the median time to diagnosis was 11 (IQR, 7–18) days. The most common presenting symptoms included fever, chills, and right upper quadrant pain as shown in Table 2. The presenting laboratory findings are displayed in Table 3, with the most notable being hypoalbuminemia, increased liver enzyme levels, and leukocytosis.Table 2Frequency of the Presenting Symptoms Associated With a Pyogenic Liver AbscessSymptomNumber with symptoms (%) N = 71Fever54 (73)Chills32 (45)Right upper quadrant pain27 (38)Nausea21 (30)Fatigue17 (24)Vomiting14 (20)Shortness of breath12 (17)Diarrhea12 (17)Cough10 (14)Weight loss10 (14)Jaundice7 (10)Altered mental status6 (9) Open table in a new tab Table 3Laboratory Abnormalities Among Patients With a Pyogenic Liver AbscessLaboratory abnormalityNo. with characteristic (%)Mean ± SD or Median (IQR)Hypoalbuminemia (<33 g/L)47/49 (96)24 ± 5Elevated γ-glutamyltransferase (>63 U/L)30/37 (81)150 (80–214)Elevated alkaline phosphatase (>130 U/L)50/70 (71)208 (141–315)Leukocytosis (>11 × 109/L)49/71 (69)15 (10–19)Hyponatremia (<133 mmol/L)38/70 (54)134 ± 4Elevated ALT (>60 U/L)36/70 (51)62 (36–116)Anemia (hemoglobin < 120 g/L)35/71 (49)118 ± 18Hyperbilirubinemia (>24 μmol/L)26/69 (38)19 (11–33)Bandemia (>0)26/71 (37)0 (0–1.2)Azotemia (>120 μmol/L)23/70 (33)93 (70–128)Thrombocytopenia (<150 × 109/L)17/71 (24)250 (153–392)Hyperglycemia (>11 mmol/L)13/61 (21)7.5 (6.2–9.5)Hypokalemia (<3.5 mmol/L)14/70 (20)3.9 ± 0.6 Open table in a new tab The initial admitting diagnosis was a confirmed or suspected PLA in only 19 (27%) cases. In nearly one half (34, 48%) of patients the admitting diagnosis was either an extra-abdominal or nonapparent focus. These admitting diagnoses included 17 fever/sepsis of unknown source, 8 pneumonia/empyema, 5 diabetic ketoacidosis or hyperosmolar non-ketotic states, 2 urosepsis, 1 upper gastrointestinal bleed, and 1 acute coronary syndrome. The diagnosis of a PLA was based on a positive liver culture in 45 (63%) patients and a positive blood culture with compatible imaging findings in 14 (20%) patients. Twelve (17%) cases had negative cultures but were diagnosed on the basis of diagnostic imaging findings and a clear radiologic and clinical response to antimicrobial therapy. All culture-negative cases had preceding antimicrobial therapy. Among the cohort of 71 patients, 67 (94%) had a computed tomography scan, and 54 (76%) had an ultrasound performed. Computed tomography detected the PLA in 66 of 67 (99%) patients (the 1 negative case was a nonenhanced study), whereas ultrasound was diagnostic in 49 of 54 (91%) patients. Imaging studies revealed solitary abscesses in 43 (61%) cases, 7 (10%) had 2, and the rest had 3 or more lesions. The PLA was isolated to the right lobe in 51 (72%) cases, an isolated left lobe abscess was seen in 11 (15%) cases, and both lobes were involved in 9 (13%) patients. A computed tomography or ultrasound guided aspirate was performed on 62 (87%) patients, with 44 (71%) returning a positive culture. Fifty-nine (83%) patients had either liver (63%) or blood culture (42%) positive results; liver aspirate and blood cultures were concomitantly positive in 16 (23%) cases. Seventeen (24%) patients had a biliary source identified as the underlying etiology of their PLA. In 40 (56%) patients no primary cause for the development of the PLA could be attributed, as shown in Table 4.Table 4Causes of Pyogenic Liver AbscessEtiologyNumber (%)Cryptogenic40 (56)Biliary origin17 (24) Choledocholithiasis 5 (7) Cholecystitis 5 (7) Ascending cholangitis 4 (6) Common bile duct stricture 3 (4) Pyelophlebitis 1 (1) Cholangiocarcinoma 1 (1) Bile leak post-cholecystectomy 1 (1) Choledochojejunostomy 1 (1)Nonbiliary origin14 (20) Neutropenia 3 (4) Perforation 2 (3) Pneumonia/empyema 2 (3) Diverticulitis 1 (1) Perinephric abscess 1 (1) Endocarditis 1 (1) Dental abscess 1 (1) Appendicitis 1 (1) Grade 5 traumatic liver injury 1 (1) Open table in a new tab Thirty-five (49%) of the PLA infections were polymicrobial. The most common organisms identified were Streptococcus milleri group (44%), Klebsiella species (27%), and anaerobes (20%) as outlined in Table 5. There were 6 (9%) nosocomial PLA infections. All of these patients had a monomicrobial infection including Citrobacter freundii, Clostridium perfringens, Streptococcus milleri, 2 fungal infections, and 1 patient with negative culture. Only 1 patient had a PLA caused by an antimicrobial-resistant organism. This patient was a 77-year-old man with chronic lymphocytic leukemia who developed a PLA with vancomycin-resistant Enterococcus faecalis and Pseudomonas aeruginosa.Table 5Frequency of Organisms, Both Monomicrobial and Polymicrobial, Cultured From the Liver or Blood in Patients With Pyogenic Liver AbscessMicroorganismNo. of patients (%)Number of monomicrobial (%)Streptococcus milleri group31 (44)15 (48)Klebsiella species19 (27)13 (68)Anaerobes14 (20)2 (14) Clostridium perfringens3 (4)1 (33) Bacteroides species3 (4)0 Peptostreptococcus species2 (3)0 Fusobacterium necrophorum2 (3)1 (50) Actinomyces species1 (1)0 Prevotella species1 (1)0 Lactobacillus1 (1)0 Veillonella1 (1)0 Escherichia coli11 (16)5 (45)Enterococcus speciesaOne of the 5 organisms was a vancomycin-resistant enterococcus.5 (7)1 (20)Staphyloccus aureus4 (6)3 (75)Other bacteriabBacteria included Group B streptococcus, Eikenella, Citrobacter, coagulase negative staphylococcus, Pseudomonas aeruginosa, Plesiomonas.11 (16)4 (36)Fungus2 (3)2 (100)Culture negative12 (17)a One of the 5 organisms was a vancomycin-resistant enterococcus.b Bacteria included Group B streptococcus, Eikenella, Citrobacter, coagulase negative staphylococcus, Pseudomonas aeruginosa, Plesiomonas. Open table in a new tab The most commonly used antimicrobial therapies included ampicillin, gentamicin/ciprofloxacin, and metronidazole (n = 28, 40%), piperacillin/tazobactam (n = 21, 30%), and ceftriaxone and metronidazole (n = 8, 11%). The remaining patients were treated with narrow spectrum agents, and 1 patient did not receive empirical antimicrobials. Sixteen (23%) patients were treated with antibiotics alone. Fifty-two (73%) patients had their abscess drained either by an ultrasound (21) or computed tomography scan (31) guided percutaneous catheter. In the patients treated with a catheter, 6 (12%) required a second catheter to successfully drain their abscess. Three patients required surgical drainage of their abscess, although in 1 of these patients the PLA was incidentally discovered during surgery. The median intravenous antibiotic duration was 17 days (IQR, 10–29), whereas the median duration of all antibiotics (oral and intravenous) was 31 days (IQR, 18–45). Fifteen (21%) endoscopic retrograde cholangiopancreatographies (ERCPs) were conducted, of which 11 identified a biliary etiology (9 choledocholithiasis and 2 common bile duct strictures). Ten of these patients were treated with a sphincterotomy, 5 had a stone removed, and 5 had a stent inserted. There were no complications attributable to ERCP. All patients were admitted to hospital for at least initial management of their PLA, and the median length of stay was 16 (IQR, 11–32) days. Eight (11%) patients developed septic shock as a result of their PLA and required an admission to an intensive care unit. Seven (10%) patients died of a PLA in hospital, corresponding to a mortality rate of 0.22 per 100,000 per year. All patients who died were older than the age of 49 years and had a community-acquired bacteremic PLA. All but 1 patient had a polymicrobial PLA. Four of the 7 were diabetic, and 3 had a prior diagnosis of a malignancy. Two presented with shock. The etiology of the PLA was biliary (2 ascending cholangitis, cholecystitis) in 3, cryptogenic in another 3, and 1 had endocarditis. This is the first population-based study of PLA conducted in a North American region and identifies a significant burden of disease associated with this diagnosis. We used a population-based cohort design that is ideal for defining the impact of a disease. In these designs all cases of a disease in a defined population are studied, and because sampling is not performed, the selection bias that commonly influences other designs is minimized. 3Huang C.J. Pitt H.A. Lipsett P.A. Osterman Jr, F.A. Liliemoe K.D. Cameron J.L. Zuidema G.D. Pyogenic hepatic abscess changing trends over 42 years.Ann Surg. 1996; 223: 600-607Crossref PubMed Scopus (403) Google Scholar, 4Chou F.F. Sheen-Chen S.M. Chen Y.S. Chen M.C. Chen F.C. Tai D.I. Prognostic factors for pyogenic abscess of the liver.J Am Coll Surg. 1994; 179: 727-732PubMed Google Scholar, 5Lee K.T. Wong S.R. Sheen P.C. Pyogenic liver abscess an audit of 10 years’ experience and analysis of risk factors.Dig Surg. 2001; 18: 459-466Crossref PubMed Scopus (133) Google Scholar, 6Chu K.M. Fan S.T. Lai E.C. Lo C.M. Wong I. Pyogenic liver abscess an audit of experience over the past decade.Arch Surg. 1996; 131: 148-152Crossref PubMed Scopus (128) Google Scholar, 7Barakate M.S. Stephen M.S. Waugh R.C. Gallagher P.J. Solomon M.J. Storey D.W. Sheldon D.M. Pyogenic liver abscess a review of 10 years’ experience in management.Aust NZ J Surg. 1999; 69: 205-209Crossref PubMed Scopus (109) Google Scholar, 8Mischinger H.I. Hauser H. Rabl H. Quehenberger F. Werkgartner G. Rubin R. et al.Pyogenic liver abscess study of therapy and analysis of risk factors.World J Surg. 1994; 18: 852-857Crossref PubMed Scopus (82) Google Scholar, 9Perez J.A.A. Gonzalez J.J. Baldonedo R.F. Sanz L. Carreno G. Junco A. Rodriguez J.I. Martinez M.D. Jorge J.I. Clinical course, treatment and multivariate analysis of risk factors for pyogenic liver abscess.Am J Surg. 2001; 181: 177-186Abstract Full Text Full Text PDF PubMed Scopus (152) Google Scholar, 10W