Introduction Pyogenic liver abscess (PLA) is an uncommon condition associated with considerable morbidity and mortality despite modern treatment. Patients have prolonged lengths of stay, require repeated imaging and procedures, and frequently develop significant complications. We audited the burden of PLA and the outcomes of treatment in Chelsea and Westminster Hospital, London. We used our data to develop local guidelines for PLA, incorporating information from previous case series and a review of the literature.
Methods Between January 2000 and September 2011, 41 patients had a correctly coded diagnosis of PLA. Electronic or paper medical records were available for 40 patients. We collected anonymised information regarding the medical history, investigations and clinical course.
Results 40 patients (78% male, mean age 54 years, 70% caucasian) were treated for acute PLA, with no in-hospital deaths. Mean length of stay was 19.3 days (CI 15.1 to 23.4 days). Presentations were non-specific, and the diagnosis rarely considered on admission. Blood results in most patients on admission demonstrated a characteristic pattern of hypoalbuminaemia, high CRP, and non-specific liver dysfunction. 61.1% of patients were septic on admission. In general, appropriate imaging was arranged early in the admission. Diagnosis of PLA was made with CT (54%) or ultrasound (46%). 75% of abscesses were found in the right lobe, with a mean maximum diameter of 6.5 cm (SD 2.4 cm). 48% of patients had multiple abscesses, and 40% had signs of loculation on imaging. 78% of abscesses were larger than 5 cm. 68% of abscesses were aspirated or drained, with three patients requiring repeated intervention, and one referred to a hepatobiliary unit. Length of stay was longer in older patients (p<0.05), those requiring drainage (23.8 vs 11.3 days, p<0.001) and those with complications (28 vs 11.8 days, p<0.001). Abscess size was not related to length of stay or need for drainage. Decisions regarding antibiotics and drainage varied between clinicians. Severe complications developed in 50% of patients and included venous thrombosis (n=4) and need for ITU support (n=4). Resolution was typical, with no in-hospital mortality. Underlying causes for PLA included abdominal malignancy (n=4), appendicitis (n=2) and diabetes mellitus (n=5). Often no cause was found, and screening for causes with MRCP and colonoscopy was not universal.
Conclusion Our study describes a large UK cohort of patients with PLA, with a lower mortality than described in other case series, and good outcomes without drainage in selected patients with large abscesses. Our cohort showed considerable variation in PLA treatment, and in response we have developed PLA management guidelines.
Competing interests None declared.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.