Abstract
Background
In times of modern surgery, transplantation and percutaneous techniques, pyogenic liver abscess (PLA) has essentially become a problem of biliary or iatrogenic origin. In the current scenario, diagnostic approach, clinical behavior and therapeutic outcomes have not been profoundly studied. This study analyzes the clinical and microbiological features, diagnostic methods, therapeutic management and predictive factors for recurrence and mortality of first episodes of PLA.
Methods
A retrospective single-center study was conducted including 142 patients admitted to the Hospital Italiano de Buenos Aires, between 2005 and 2015 with first episodes of PLA.
Results
Prevailing identifiable causes were biliary diseases (47.9%) followed by non-biliary percutaneous procedures (NBIPLA, 15.5%). Seventeen patients (12%) were liver recipients. Eleven patients (7.8%) died and 18 patients (13.7%) had recurrence in the first year of follow up. The isolation of multiresistant organisms (p = 0.041) and a history of cholangitis (p < 0.001) were independent risk factors for recurrence. Mortality was associated with serum bilirubin >5 mg/dL (p = 0.022) and bilateral involvement (p = 0.014) in the multivariate analysis.
Conclusion
NBPLA and PLA after transplantation may be increasing among the population of PLA in referral centers. History of cholangitis is a strong predictor for recurrence. Mortality is associated to hiperbilirrubinemia and anatomical distribution of the lesions.
Introduction
Pyogenic liver abscess (PLA) represents the most common visceral abscess with an incidence of 5–20 in 100,000 hospitalizations in the western population.1 From the first series published by Oschner in 1938,2 PLA has been subjected to significant changes in terms of etiology, diagnosis, management and prognosis. In times of modern surgery, transplantation and percutaneous techniques, PLA has essentially become a problem of biliary or iatrogenic origin, particularly in referral centers where these procedures are performed regularly.3, 4 Therefore, patients with PLA frequently exhibit: reconstructed biliary anatomy, recurrent hospitalizations, regular performance status and history of immunosuppression. In the current scenario, diagnostic approach, clinical behavior and therapeutic outcomes of PLA have not been profoundly studied.
The purpose of this study is to evaluate the clinical and microbiological features, diagnostic and therapeutic management, outcome and predictor factors for recurrence and mortality of first episodes of PLA in a referral institution in Argentina.
Methods
A retrospective study was conducted on a prospectively filled database of patients admitted to the Hospital Italiano de Buenos Aires (HIBA), between January the 1st 2005 and July 31st 2015 with diagnosis of first episode of PLA. We define an abscess with the identification of one or more discrete cavities in the liver identified by imaging methods associated with fever, leukocytosis or bacterial isolates in blood or culture of the lesion material.5 Patients with parasitic, fungal or amoebic abscesses were not included in the study. Patients with sub-diaphragmatic or sub-hepatic abscess, with gallbladder empyema or abscessed tumors were also excluded from the analysis.
Data collected included demographic characteristics, etiopathologic factors, clinical features, laboratory data, number, size and location of lesions, microbiological findings, diagnostic and therapeutic methods, treatment response, mortality and recurrence. The origin of the abscess was based upon imaging studies as well as clinical, pathological and surgical findings.
Definitions
Abscesses were considered from biliary origin when an obstruction, instrumentation and/or anatomic alteration of the biliary tree were present. Non-biliary iatrogenic pyogenic liver abscess (NBIPLA) included abscesses that occur within 60 days after performing liver radiofrequency ablation (RFA), transarterial chemoembolization (TACE) or liver biopsies and occurred in the location where the procedure was performed (Figure 1, Figure 2). Abscesses of portal origin were those that occurred after portal bacteremia secondary to abdominal infections and peritonitis, supra-infected gastrointestinal tumors and chronic inflammatory bowel disease.
Figure 1.
72-year-old female patient with diagnosis of metastatic neuroendocrine liver tumor treated with RFA. a. Axial section of MDCT showed a nodular formation of 14 × 18 mm (white arrow) in segment IV–V. b. The MDCT 4 weeks after the procedure showed a hypodense area with air bubbles inside in segments IV–V (white arrow) suggesting a PLA
Figure 2.
74-year-old patient with diagnosis of melanoma liver metastasis. a. MRCP showed an extensive mass (95 × 83 mm) located in segments VI–VII (White arrow). b. The lesion was treated with super-selective TACE (white arrow). c. The MDCT 6 weeks after the procedure showed a fluid level collection of 112 × 90 mm (white arrow). d. MDCT guided percutaneous drainage (white arrow)
Cure was defined as the absence of signs and symptoms together with negative blood cultures and image studies without findings compatible with PLA. Recurrence was defined as a new episode PLA after cure has been reached.
Failure of nonoperative therapy was defined as a patient having persistent signs of sepsis after 5 days of intravenous antibiotics and percutaneous drainage (PD). Case fatality was defined as death during hospitalization or before cure was reached.
Management and treatment
All patients had one set of blood culture taken prior to the initiation of empirical broad spectrum intravenous antibiotics that were suggested by the Infectious Diseases Department. Antibiotic treatments were tailored once culture and susceptibility results were available. Total duration of antibiotic therapy was at least 21 days, varying according to the clinical and radiological response. Antibiotic therapy was discontinued when there was a clinical and radiological resolution of the abscess. Patients with positive blood culture were invariably treated with IV antibiotics the first 14 days.
For abscesses with a diameter >5 cm, PD was performed. For abscesses between 3 and 5 cm, PD was performed in cases in which patients exhibited poor medical condition, did not respond to medical treatment and lesions were accessible for a percutaneous approach. Abscesses <3 cm were not drained. PD was performed placing a variety of 8–12 Fr catheters (Cook®) under Seldinger technique using an ultrasound or Multidetector compute tomography (MDCT) guide (Figure 2, Figure 3). Drains were removed upon resolution of sepsis, imaging evidence of abscess disappearance, and minimal drain aspirates. Percutaneous or endoscopic biliary drainage was performed in patients in whom biliary obstruction was evidenced. All patients underwent a repeat radiologic imaging at 2–3 week intervals.
Figure 3.
55-year-old man with history of liver transplantation secondary to primary sclerosing cholangitis. a. MRCP showed an heterogeneous hyper intense mass (42 × 30 mm, white arrow). b. MDCT coronal section showed hepaticojejunostomy (arrow head) and aerobilia (white arrow). c. MDCT guided percutaneous drainage (white arrow)
Follow up
Patients included in this study had a follow up of at least 2 years or until death. Follow up included a clinical control with a blood test 7–10 days after discharge together with cross-sectional imaging and blood tests at 3 months, 6 months, 1 year, and 2 years after discharge until complete radiological resolution and normalization of laboratory values.
Statistical analysis
Quantitative variables were presented, according to a parametrical or non-parametrical distribution, as mean ± standard deviation or median and interquartile range, respectively. Qualitative variables were presented as absolute value and percentage. For statistical comparison, the Student's T test was used for quantitative variables, while Chi square and Fisher exact probability tests for qualitative variables. The statistically significant independent factors obtained by univariate analyses were entered into a multiple stepwise logistic regression model to identify independent risk factors for PLA recurrence and mortality. Statistical significance was considered to have been achieved when p < 0.05.
Results
Demographic, clinical and imaging characteristics were summarized according to the liver abscess etiology in Table 1. Prevailing identifiable PLA were those caused by biliary diseases, with a total 68 cases (47.9%), followed by NBIPLA (15.5%; RFA, 7 patients [4.9%]; TACE, 13 [9.2%]; liver biopsies, 2 [1.4%]). Among PLA caused by biliary disease, there were 17 liver transplant recipients (12%). PLA of different etiologies showed no statistically significant differences in terms of patient demographic data, clinical manifestation, laboratory findings or abscesses features (Table 1).
Table 1.
Demographic, clinical, radiological and laboratory findings of pyogenic liver abscess by etiology
| Variable/Etiology | Biliary n = 68 (47.9%) |
Cryptogenic n = 33 (23.2%) |
NBIPLA n = 22 (15.5%) |
Portal n = 14 (9.9%) |
p |
|---|---|---|---|---|---|
| Age ± SD | 63.7 (±16.3) | 61.3 (±17.2) | 61.3 (±11.6) | 58.9 (±21.4) | 0.61 |
| Sex M:F | 42:25 | 25:8 | 15:7 | 6:8 | 0.71 |
| CVD | 28 (41) | 11 (33) | 8 (36) | 6 (42) | 0.26 |
| Diabetes | 18 (26) | 7 (21) | 4 (18) | 1 (7) | 0.53 |
| Immunosuppression | 18 (26) | 4 (12) | 7 (32) | 4 (29) | 0.45 |
| Fever | 59 (87) | 30 (91) | 18 (82) | 14 (100) | 0.51 |
| Jaundice | 18 (26) | – | 4 (18) | – | 0.11 |
| Shock | 9 (13) | 1 (3) | 1 (5) | – | 0.33 |
| Leucocytes >15.000 | 38 (56) | 23 (70) | 13 (59) | 7 (50) | 0.55 |
| ALT >120 | 11 (16) | 1 | 6 (27) | 2 (1) | 0.11 |
| ALP >400 | 14 (21) | 1 (3) | 5 | – | 0.13 |
| TSB > 5 mg% | 6 (9) | 1 (3) | – | – | 0.29 |
| Serum albumin<3 g% | 49 (72) | 20 (61) | 16 (73) | 8 (57) | 0.63 |
| Bacteriemia | 23 (34) | 5 (15) | 7 (32) | 4 (29) | 0.30 |
| ML:SL | 42 (62):26 (38) | 14 (42):19 (58) | 10 (45):12 (55) | 8 (57):6 (43) | 0.25 |
| Bilobular involvement | 19 (28) | 5 (15) | 2 (9) | 2 (14) | 0.19 |
| Abscess Size>5 cm | 34 (50) | 24 (73) | 15 (68) | 8 (57) | 0.75 |
NBIPLA, non-biliary iatrogenic pyogenic liver abscess; M:F, male:female; CVD, cardiovascular disease; TSB, total serum bilirubin; ALT, alanine aminotransferase; ALP, Alkaline Phosphatase; ML:SL, Multiple lesion:Single lesion.
MDCT was performed in 132 cases (92.9%), ultrasonography (US) in 116 (81.7%) and magnetic resonance imaging (MRI) in 23 patients (16.2%). Abscesses were multiple in 77 cases (54.2%), compromising the right lobe in 90 cases (63.4%), the left in 21 (14.8%) and both in 31 patients (21.8%). The size was >5 cm in 81 cases (57.1%). Seven cases (6%) could not be diagnosed with US, all with size <5 cm. The correlation in the location and number of abscesses between MDCT and MRI was complete in all cases.
Microbiological characteristics
Revenue from blood cultures and cultures from abscess material was 30.3% and 85.4%, respectively (Table 2). Escherichia coli was the most frequently found microorganism in 38 cases (26.7%), followed by Enterococcus spp. in 24 (16.9%) and Klebsiella pneumoniae in 23 (16.2%). In 16 patients (11.3%) multidrug resistant bacteria (MDRB) were isolated. Isolates were polymicrobial in 23 patients (16.2%). In 29 cases (20.4%) it was not possible to isolate any microorganism (Table 2).
Table 2.
Microbiological results in patients with pyogenic liver abscesses
| Total (%) n = 142 | Biliary (%) n = 68 | Cryptogenic (%) n = 33 | NBIPLA (%) n = 22 | Portal (%) n = 14 | Other (%) n = 5 | |
|---|---|---|---|---|---|---|
| Positive blood culture | 43 (30.3) | 23 (33.8) | 6 (18.2) | 7 (31.8) | 4 (28.6) | 3 (60) |
| Abscesses material culture | 123 (86.6) | 56 | 29 | 19 | 13 | 6 |
| Positive culture | 105 (85.4) | 53 (94.6) | 22 (75.7) | 17 (89.5) | 9 (69.2) | 4 (66.7) |
| Gram negative bacilli | 55 (38.5) | 30 (44.1) | 11 (33.3) | 9 (40.9) | 4 (28.6) | 1 (20) |
| E. coli | 38 (26.7) | 21 (30.9) | 6 (18.2) | 8 (36.4) | 2 (14.3) | 1 (20) |
| K. pneumoniae | 19 (13.9) | 6 (8.8) | 7 (21.2) | 4 (18.2) | 1 (7.1) | 1 (20) |
| Pseudomona spp. | 3 (2.1) | 2 (2.9) | 1 (3) | – | – | – |
| Morganella spp. | 3 (2.1) | 3 (4.4) | – | – | – | – |
| Proteus | 2 (1.4) | 2 (2.9) | – | – | – | – |
| Gram positive cocci | 42 (29.5) | 16 (23.5) | 4 (12.1) | 6 (27.3) | 6 (42.9) | 2 (40) |
| Staphylococcus | 14 (12.7) | 5 (7.4) | 2 (6.1) | 3 (13.6) | 1 (7.1) | 2 (40) |
| Streptococcus | 16 (14.5) | 5 (7.4) | 5 (15.2) | 3 (13.6) | 3 (21.4) | – |
| Enterococcus | 24 (21.8) | 14 (20.6) | 1 (3) | 4 (18.2) | – | 1 (20) |
| Multibacterian | 23 (16.2) | 10 (14.7) | 5 (15.2) | 6 (27.3) | – | 2 (40) |
| Negative cultures | 29 (20.4) | 10 (14.7) | 9 (27.3) | 4 (18.2) | 4 (28.6) | 2 (40) |
| Multidrug resistant bacteria (MDRB) | 16 (11.3) | 14 (20.6) | – | 2 (9.1) | – |
NBIPLA, non-biliary iatrogenic pyogenic liver abscess.
Results expressed as absolute value and percentage.
Treatment & outcome
In 122 patients (85.9%) treatment included PD together with antibiotic therapy. A group of 20 patients (14.1%) with abscesses <5 cm received antibiotic treatment exclusively. Only 37 patients (26.1%) received antibiotic therapy more than 4 weeks. Drains were placed under tomographic guidance in 96 cases (78.7%) and US guidance in 26 (21.3%). In 18 cases (14.8%) drain replacements or additional drainages were required. The need for these procedures was found statistically associated with the placement of the initial drainage under US guidance (OR 4.83 [1.68–13.87]; p = 0.031). In 49 of the 68 patients with biliary PLA (72%) biliary instrumentation was also performed, either endoscopically (n = 31, 46%) or percutaneously (n = 18, 26%). Surgical treatment was required in 4 patients (2.8%) in whom the percutaneous treatment had failed.
Among the patients who survived the first episode, 18 patients (13.7%) exhibited recurrence in the first year of follow up (Table 3). All the recurrences were ipsilateral to the lesions of the first event. In univariate analysis, there was an association between recurrence and hypoalbuminemia (OR 9.6 [1.24–75.41]; p = 0.031), the presence of multiple abscesses (OR 3.96 [1.23–12.78]; p = 0.022) or lesions in both liver lobes (OR 7.07 [2.4–20.83]; p < 0.001), polymicrobial (OR 3.26 [1.07–10.08]; p = 0.038) or MDRB isolation (OR 3.21 [1.42–17.58]; p = 0.012), and a history of cholangitis previous to the development of the abscess (OR 89.08 [18.29–565.98]; p < 0.001). Of these, MDRB isolation (OR 8.61 [1.08–69.2]; p = 0.04) and a history of cholangitis (OR 201.57 [15.48–2624.2]; p < 0.001) were also found as independent risk factors for recurrence in the first year.
Table 3.
Risk factors of mortality and recurrence of pyogenic liver abscess
| No recurrence n = 113 (86.3%) |
Recurrence n = 18 (13.7%) |
OR | Univariate |
OR | Multivariate |
|||
|---|---|---|---|---|---|---|---|---|
| CI 95% | p | CI 95% | p | |||||
| Albuminemia<3 g% | 62 | 18 | 9.6 | (1.24–75.41) | 0.03 | 10.44 | (0.65–168.42) | 0.09 |
| Multiple abscesses | 43 | 14 | 3.26 | (1.23–12.78) | 0.02 | 2.84 | (0.38–21.01) | 0.3 |
| Bilobular involvement | 14 | 9 | 7.07 | (2.4–20.83) | <0.001 | 2.81 | (0.36–21.82) | 0.32 |
| Multibacterian infection | 15 | 6 | 3.26 | (1.07–10.08) | 0.038 | 3.68 | (0.51–26.78) | 0.19 |
| MDRB isolation | 7 | 6 | 3.21 | (1.42–17.58) | 0.012 | 8.61 | (1.08–69.2) | 0.04 |
| Previous episodes of cholangitis | 2 | 11 | 89.08 | (18.29–565.98) | <0.001 | 201.57 | (15.48–2624.2) | <0.001 |
| Age >65 y/o | 55 | 6 | 0.55 | (0.19–1.56) | 0.21 | – | – | – |
| Diabetes | 34 | 3 | 0.64 | (0.17–2.37) | 0.53 | – | – | – |
| LTR | 11 | 3 | 1.85 | (0.46–7.42) | 0.38 | – | – | – |
| Serum ALP >400 | 13 | 5 | 2.95 | (0.91–9.64) | 0.07 | – | – | – |
| Biliary etiology | 48 | 11 | 2.81 | (0.98–8.02) | 0.054 | – | – | – |
| Abscess >5 cm | 76 | 9 | 0.68 | (0.25–1.86) | 0.46 | – | – | – |
| Bacteremia | 29 | 7 | 1.84 | (0.65–5.20) | 0.25 | – | – | – |
| Antibiotic therapy <4 weeks | 74 | 11 | 0.54 | (0.19–1.53) | 0.25 | – | – | – |
| No mortality n = 131 (92.3%) | Mortality n = 11 (7.7%) | OR | Univariate |
OR | Multivariate |
|||
|---|---|---|---|---|---|---|---|---|
| CI 95% | p | CI 95% | p | |||||
| TSB >5 mg% | 3 | 4 | 24.38 | (4.54–130.1) | <0.001 | 60.11 | 4.49–804.68 | 0.02 |
| Multiple abscesses | 67 | 10 | 9.55 | (1.18–76.76) | 0.03 | 2.49 | (0.21–30.11) | 0.47 |
| Bilobular involvement | 23 | 8 | 12.5 | (3.08–50.83) | <0.001 | 9.95 | (1.61–62.1) | 0.014 |
| Biliary etiology | 59 | 9 | 5.49 | (1.14–26.4) | 0.03 | 2.97 | (0.55–16.09) | 0.2 |
| Age >65 y/o | 61 | 6 | 1.42 | (0.41–4.88) | 0.57 | – | – | – |
| Diabetes | 37 | 2 | 0.74 | (0.15–3.75) | 0.72 | – | – | – |
| LTR | 14 | 3 | 3.13 | (0.74–13.2) | 0.12 | – | – | – |
| Serum ALP>400 | 18 | 4 | 3.58 | (0.95–13.5) | 0.06 | – | – | – |
| Abscess >5 cm | 76 | 3 | 0.27 | (0.07–1.06) | 0.68 | – | – | – |
| Bacteremia | 36 | 6 | 3.1 | (0.91–11.02) | 0.08 | – | – | – |
| Multibacterian infection | 21 | 2 | 1.16 | (0.23–5.77) | 0.85 | – | – | – |
| MDRB | 13 | 3 | 3.38 | (0.79–14.32) | 0.99 | – | – | – |
| Antibiotic therapy <4 weeks | 95 | 10 | 3.79 | (0.47–30.67) | 0.21 | – | – | – |
MDRB, multidrug resistant bacteria; TSB, total serum bilirubin; LTR, liver transplant recipient; ALP, alkaline phosphatase; OR, odd ratio; CI, confidence interval.
Eleven of the 142 patients included in the series died during hospitalization (7.8%) (Table 3). Cause of death was multiorgan failure due to severe sepsis in all but one patient, who suffered a massive pulmonary embolism. In univariate analysis, patient fatality was associated with biliary etiology (OR 5.49 [1.14–26,4]; p = 0.031), the presence of multiple abscesses (OR 9.55 [1.18–78.76]; p < 0.03) or lesions in both lobes (OR 12.5 [3.08–50.83]; p < 0.001) and bilirrubinemia >5 mg/dL (OR 24.38 [4.54–130.1]; p < 0.001). In the multivariate analysis, only hyperbilirubinemia >5 mg/dL (OR 60.11 [4.49–804.68]; p = 0.022) and bilateral involvement (OR 9.95 [1.61–62.1]; p = 0.014) were statistically significant.
Discussion
In our study, biliary disease represented the most common cause of PLA (47.9%). This group of patients included a notable number of liver transplant recipients, whose bile ducts have been reconstructed and had a history of immunosuppression. The scientific evidence in PLA in transplanted livers is limited to case reports or small series of patients.6, 7 Hepatico-jejunal anastomosis and stenosis of the bile ducts have been widely reported as risk factors for PLA. Tachopoulou et al. showed that a history of hepatic artery thrombosis is a major risk factor for the development of PLA. Coincident with this, all transplant recipients included in our series reported either hepatico-jejunal anastomosis or biliary stenosis at the time of diagnosis and 23% of the patients had a history of thrombosis of the hepatic artery. Moreover, it is remarkable that NBIPLA represented the second most common identifiable etiology in our study. Both RFA and TACE generate a localized destruction of tissue that can favor the formation of an abscess. PLA has been reported as a rare complication that occurs in less than 5% of TACE and 1% of the RFA.8, 9, 10, 11 Although the rate of this complication at our center was similar to the literature (4.1% in TACE and 1.2% in RFA), NBIPLA represented 15.5% of the overall population with first episodes of PLA.
In our study, E. coli was the most frequently isolated agent, while K. pneumoniae resulted surprisingly scarce. This distribution has been respected even in the diabetic population, in which E. coli was isolated in 28% of the cases. The results contrast with numerous series that reported progressive increase of K. pneumoniae over E. coli as primary pathogen causing PLA, especially in patients with diabetes mellitus.1, 5, 12, 13, 14, 15, 16, 17 The isolation of MDRB was slightly higher as compared to other reports17 and could be explained by the notable presence of patients with chronic disease and history of prolonged hospitalizations. A 75% of MDRB were isolated in transplant recipients or patients with NBIPLA with active oncologic disease.
US demonstrated limitations for detecting small PLA. Lin et al. highlighted the size and location of the abscess as the most influential factors in revenue from US.18 Our study showed a sensitivity of 95%, with 7 false negative cases that occurred in patients with PLA <5 cm including 5 cases with PLA <3 cm. At the same time, 33% of the drains placed by US guidance required relocation or placement of an additional drain for the same abscess. More sophisticated cross-sectional modalities such as MDCT or MRI showed a higher value for both the initial evaluation of patients with high suspicion of PLA and the eventual guidance for PD.
The advent of interventional radiology has turned surgical drainage into a rare indication. PD have been shown to be as effective as open surgical therapy and has become the standard of care in patients with abscesses >3 cm19, 20, 21with case fatality rates of up 8%.22, 23 Still, there is a role for surgical drainage if PD fails, especially in cases with large multilocular abscesses.12 Even so, these situations can be frequently manage by ‘‘re-drainage’’ or improvement of existing drainage systems. In our series, only four of the 122 patients (3.3%) treated with PD required surgery, but 18 cases (14.8%) required a drain replacement or additional drainage. In total, 21 patients (17.2%) had more than one catheter placed. Coincident to the literature, our overall mortality in patients that were subjected to PD was 2.5% (3 patients out of 118).23, 24 We did not performed needle aspiration in our series in any case. A recent study published by Yu et al., compared repeated needle aspiration versus catheter aspiration in 64 patients with PLA larger than 3 cm in diameter.25 Even though the authors concluded that the two techniques were equivalent in terms of success morbidity and mortality, 59% of patients required more than one procedure. We believe that subjecting the patient to multiple procedures should be avoided, especially considering the technical difficulty that may represent puncturing an abscess that has been already drained.
Our results showed that recurrence was associated with albuminemia <3 g/dL, polymicrobial or MDRB isolations, presence of multiple abscesses and compromise of both hepatic lobes. Multivariate analysis confirmed that MDRB isolation and the presence of lesions in both lobes represent independent risk factors for PLA recurrence. Cheng et al. reported biliary etiology as a related factor for recurrence.26 In our series, 11 of 13 patients with PLA and history of cholangitis recurred in the first year. This association was confirmed in both univariate and multivariate analysis. These previous cholangitis episodes included 7 patients who had ascending cholangitis secondary to hepaticojejunostomy and 6 patients who had obstructive cholangitis secondary to biliary strictures. Although the source of infection was addressed in these patients at that moment, the original risk factor for biliary infection – diseased biliary tree or biliodigestive anastomosis-persisted after the cholangitis episode. The mechanism by which PLA and cholangitis develop would be identical; microorganisms could migrate to the liver parenchyma from the digestive tract through diseased or obstructed bile ducts. Cholangitis and biliary PLA may be therefore different phases of the same process. It is remarkable that the history of cholangitis might not only revealed biliary disease but also a particular predisposition to biliary infections within patients with altered biliary anatomy. This is consistent with the fact that every recurrence was ipsilateral to the first event.
Mortality in our series was 7.8%, similar to that reported in the16, 17, 22, 27, 28 studies have analyzed risk factors for mortality, including the presence of active oncologic17, 22, 26, 29, 30 renal failure,17, 18, 22, 26, 29 hiperbilirrubinemia,17, 29, 30 hypoalbuminemia,17, 29 MDRB isolation17, 29 and multiple abscesses17, 22, 26, 29, 30 with inconsistent results. In our series, factors associated with mortality in univariate analysis included biliary disease, multiple abscesses, the bilateral liver involvement, and bilirrubinemia >5 mg/dL. The presence of lesions in both liver lobes and hyperbilirubinemia were also independent risk factors for mortality in the multivariate analysis. As opposed to other reports,17, 29 we found that hepatobiliary factors such as the presence of multiple abscesses, lesions in both lobes and hyperbilirubinemia were associated to mortality. On the contrary, parameters of inflammatory systemic response to PLA, such as high leukocytosis, hemodynamic instability or kidney failure, failed to predict the prognosis of the disease in our center.
Our study represents a reference of the actual situation of PLA in Argentina, providing valuable epidemiological, clinical and microbiological information regarding first cases of PLA in the region. Most previous reports of PLA prognostic factors were based on univariate analysis or included only a limited number of cases.23, 29, 30, 31, 32, 33 We applied a multivariate analysis to a large cohort, unique in the region. Study limitations include its retrospective, single-center cross design.
To conclude, our results suggest that NBIPLA and PLA after liver transplantation are increasing among the general population of PLA in referral centers. The majority of the patients resolve with antibiotics and PD. History of cholangitis prior to PLA is a strong predictor for recurrence in the first year. Mortality was associated to hyperbilirubinemia, etiological origin, number and anatomical distribution of the lesions.
Author contributions
-
•
Study conception and design: Matías Czerwonko, Martín de Santibañes, Pablo Huespe, Santiago Bertone, Pablo Pellegrini, Sung Ho Hyon
-
•
Acquisition of data: Matias Czerwonko, Martin de Santibañes, Santiago Bertone, Pablo Pellegrini, Oscar Mazza, Sung Ho Hyon
-
•
Analysis and interpretation of data: Matias Czerwonko, Martin de Santibañes, Pablo Huespe, Oscar Mazza, Eduardo De Santibañes, Juan Pekolj, Sung Ho Hyon
-
•
Drafting of manuscript: Matias Czerwonko, Martin de Santibañes, Pablo Huespe, Pablo Pellegrini, Eduardo De Santibañes, Juan Pekolj
-
•
Critical revision: Pablo Huespe, Santiago Bertone, Oscar Mazza, Eduardo De Santibañes, Juan Pekolj
Conflicts of interest
None declared.
References
- 1.Huang C.J., Pitt H.A., Lipsett P.A., Osterman F.A., Jr., Lillemoe K.D., Cameron J.L. Pyogenic hepatic abscess. Changing trends over 42 years. Ann Surg. 1996;223:600–607. doi: 10.1097/00000658-199605000-00016. 607–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ochsner A., De Bakey M., Murrey S. Pyogenic liver abscess II: an analysis of forty-seven cases with review of the literature. Am J Surg. 1938;40:292–319. [Google Scholar]
- 3.Siu L.K., Yeh K.-M., Lin J.-C., Fung C.-P., Chang F.-Y. Klebsiella pneumoniae liver abscess: a new invasive syndrome. Lancet Infect Dis. 2012;12:881–887. doi: 10.1016/S1473-3099(12)70205-0. [DOI] [PubMed] [Google Scholar]
- 4.McDonald M.I., Corey G.R., Gallis H.A., Durack D.T. Single and multiple pyogenic liver abscesses. Natural history, diagnosis and treatment, with emphasis on percutaneous drainage. Medicine (Baltimore) 1984;63:291–302. doi: 10.1097/00005792-198409000-00003. [DOI] [PubMed] [Google Scholar]
- 5.Branum G.D., Tyson G.S., Branum M.A., Meyers W.C. Hepatic abscess. Changes in etiology, diagnosis, and management. Ann Surg. 1990;212:655–662. doi: 10.1097/00000658-199012000-00002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Farges O., Leese T., Bismuth H. Pyogenic liver abscess: an improvement in prognosis. Br J Surg. 1988;75:862–865. doi: 10.1002/bjs.1800750910. [DOI] [PubMed] [Google Scholar]
- 7.Tachopoulou O.A., Vogt D.P., Henderson J.M., Baker M., Keys T.F. Hepatic abscess after liver transplantation: 1990–2000. Transplantation. 2003;75:79–83. doi: 10.1097/00007890-200301150-00014. [DOI] [PubMed] [Google Scholar]
- 8.Guiu B., Deschamps F., Aho S., Munck F., Dromain C., Boige V. Liver/biliary injuries following chemoembolisation of endocrine tumours and hepatocellular carcinoma: lipiodol vs. drug-eluting beads. J Hepatol. 2012;56:609–617. doi: 10.1016/j.jhep.2011.09.012. [DOI] [PubMed] [Google Scholar]
- 9.Woo S., Chung J.W., Hur S., Joo S.M., Kim H.C., Jae H.J. Liver abscess after transarterial chemoembolization in patients with bilioenteric anastomosis: frequency and risk factors. AJR Am J Roentgenol. 2013;200:1370–1377. doi: 10.2214/AJR.12.9630. [DOI] [PubMed] [Google Scholar]
- 10.Goldberg S.N., Charboneau J.W., Dodd G.D., 3rd, Dupuy D.E., Gervais D.A., Gillams A.R. Image-guided tumor ablation: proposal for standardization of terms and reporting criteria. Radiology. 2003;228:335–345. doi: 10.1148/radiol.2282021787. [DOI] [PubMed] [Google Scholar]
- 11.Livraghi T., Solbiati L., Meloni M.F., Gazelle G.S., Halpern E.F., Goldberg S.N. Treatment of focal liver tumors with percutaneous radio-frequency ablation: complications encountered in a multicenter study. Radiology. 2003;226:441–451. doi: 10.1148/radiol.2262012198. [DOI] [PubMed] [Google Scholar]
- 12.Tan Y.M., Chung A.Y., Chow P.K., Cheow P.C., Wong W.K., Ooi L.L. An appraisal of surgical and percutaneous drainage for pyogenic liver abscesses larger than 5 cm. Ann Surg. 2005;241:485–490. doi: 10.1097/01.sla.0000154265.14006.47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Shelat V.G., Chia C.L., Yeo C.S., Qiao W., Woon W., Junnarkar S.P. Pyogenic liver abscess: does Escherichia coli cause more adverse outcomes than Klebsiella pneumoniae? World J Surg. 2015 doi: 10.1007/s00268-015-3126-1. 268-015-3126-1. [DOI] [PubMed] [Google Scholar]
- 14.Chan D.S.G., Archuleta S., Llorin R.M., Lye D.C., Fisher D. Standardized outpatient management of Klebsiella pneumoniae liver abscesses. Int J Infect Dis. 2013;17:e185–e188. doi: 10.1016/j.ijid.2012.10.002. [DOI] [PubMed] [Google Scholar]
- 15.Lederman E.R., Crum N.F. Pyogenic liver abscess with a focus on Klebsiella pneumoniae as a primary pathogen: an emerging disease with unique clinical characteristics. Am J Gastroenterol. 2005;100:322–331. doi: 10.1111/j.1572-0241.2005.40310.x. [DOI] [PubMed] [Google Scholar]
- 16.Rahimian J., Wilson T., Oram V., Holzman R.S. Pyogenic liver abscess: recent trends in etiology and mortality. Clin Infect Dis. 2004;39:1654–1659. doi: 10.1086/425616. [DOI] [PubMed] [Google Scholar]
- 17.Chen S.C., Tsai S.J., Chen C.H., Huang C.C., Lin D.B., Wang P.H. Predictors of mortality in patients with pyogenic liver abscess. Neth J Med. 2008;66:196–203. [PubMed] [Google Scholar]
- 18.Lin A.C., Yeh D.Y., Hsu Y.H., Wu C.C., Chang H., Jang T.N. Diagnosis of pyogenic liver abscess by abdominal ultrasonography in the emergency department. Emerg Med J. 2009;26:273–275. doi: 10.1136/emj.2007.049254. [DOI] [PubMed] [Google Scholar]
- 19.Ch Yu S., Hg Lo R., Kan P.S., Metreweli C. Pyogenic liver abscess: treatment with needle aspiration. Clin Radiol. 1997;52:912–916. doi: 10.1016/s0009-9260(97)80223-1. [DOI] [PubMed] [Google Scholar]
- 20.Gerzof S.G., Johnson W.C., Robbins A.H., Nabseth D.C. Intrahepatic pyogenic abscesses: treatment by percutaneous drainage. Am J Surg. 1985;149:487–494. doi: 10.1016/s0002-9610(85)80045-3. [DOI] [PubMed] [Google Scholar]
- 21.Johnson R.D., Mueller P.R., Ferrucci J.T., Jr., Dawson S.L., Butch R.J., Papanicolaou N. Percutaneous drainage of pyogenic liver abscesses. AJR Am J Roentgenol. 1985;144:463–467. doi: 10.2214/ajr.144.3.463. [DOI] [PubMed] [Google Scholar]
- 22.Chen C.-H., Wu S.-S., Chang H.-C., Chang Y.-J. Initial presentations and final outcomes of primary pyogenic liver abscess: a cross-sectional study. BMC Gastroenterol. 2014;14:133. doi: 10.1186/1471-230X-14-133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mohsen A.H., Green S.T., Read R.C., McKendrick M.W. Liver abscess in adults: ten years experience in a UK centre. QJM. 2002;95:797–802. doi: 10.1093/qjmed/95.12.797. [DOI] [PubMed] [Google Scholar]
- 24.Lardière-Deguelte S., Ragot E., Amroun K., Piardi T., Dokmak S., Bruno O. Hepatic abscess: diagnosis and management. J Visc Surg. 2015;152:231–243. doi: 10.1016/j.jviscsurg.2015.01.013. [DOI] [PubMed] [Google Scholar]
- 25.Yu S.C., Ho S.S., Lau W.Y., Yeung D.T., Yuen E.H., Lee P.S., Metreweli C. Treatment of pyogenic liver abscess: prospective randomized comparison of catheter drainage and needle aspiration. Hepatology. 2004;39:932–938. doi: 10.1002/hep.20133. [DOI] [PubMed] [Google Scholar]
- 26.Cheng H.-C., Chang W.L., Chen W.Y., Kao A.W., Chuang C.H., Sheu B.S. Long-term outcome of pyogenic liver abscess: factors related with abscess recurrence. J Clin Gastroenterol. 2008;42:1110–1115. doi: 10.1097/MCG.0b013e318157e4c1. [DOI] [PubMed] [Google Scholar]
- 27.Alvarez J.A., González J.J., Baldonedo R.F., Sanz L., Carreño G., Jorge J.I. Single and multiple pyogenic liver abscesses: etiology, clinical course, and outcome. Dig Surg. 2001;18:283–288. doi: 10.1159/000050153. [DOI] [PubMed] [Google Scholar]
- 28.Koneru S., Peskin G.W., Sreenivas V. Pyogenic hepatic abscess in a community hospital. Am Surg. 1994;60:278–281. [PubMed] [Google Scholar]
- 29.Chou 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;179:727–732. [PubMed] [Google Scholar]
- 30.Mischinger H.J., Hauser H., Rabl H., Quehenberger F., Werkgartner G., Rubin R. Pyogenic liver abscess: studies of therapy and analysis of risk factors. World J Surg. 1994;18:852–857. doi: 10.1007/BF00299085. discussion 858. [DOI] [PubMed] [Google Scholar]
- 31.Lee K.T., Sheen P.C., Chen J.S., Ker C.G. Pyogenic liver abscess: multivariate analysis of risk factors. World J Surg. 1991;15:372–376. doi: 10.1007/BF01658732. discussion 376–7. [DOI] [PubMed] [Google Scholar]
- 32.Wong W.M., Wong B.C., Hui C.K., Ng M., Lai K.C., Tso W.K. Pyogenic liver abscess: retrospective analysis of 80 cases over a 10-year period. J Gastroenterol Hepatol. 2002;17:1001–1007. doi: 10.1046/j.1440-1746.2002.02787.x. [DOI] [PubMed] [Google Scholar]
- 33.Alvarez Pérez J.A., González J.J., Baldonedo R.F., Sanz L., Carreño G., Junco A. Clinical course, treatment, and multivariate analysis of risk factors for pyogenic liver abscess. Am J Surg. 2001;181:177–186. doi: 10.1016/s0002-9610(00)00564-x. [DOI] [PubMed] [Google Scholar]