Safety and efficacy of USG-guided catheter drainage in liver abscesses.

Background: This study aimed to evaluate the safety and efficacy of USG-guided percutaneous drainage in liver abscesses of >5 cm. A lot of literature is available on the minimally invasive treatment of liver abscesses since its introduction in the early 1980s. This study focuses on the eastern Indian population and the outcome of treatment of liver abscess of >5 cm by means of catheter drainage and the use of antibiotics. Patients and
Methods: This is a retrospective study conducted on a total of fifty patients over a period of 1 year, 1 month (from June 2017 to June 2018). Only patients with liver abscess with size >5 cm were included in the study. The demographic characteristics; comorbidities; and clinical, radiological, and bacteriological characteristics of liver abscesses in the eastern Indian population and the safety and efficacy of catheter drainage were evaluated.
Results: It was found that because of preprocedural empirical antibiotic intake, 70% of the patients had no growth in the pus, whereas 12% had Entamoeba histolytica, 8% had Escherichia coli, and 6% had Klebsiella pneumoniae as the causative agent. The total duration of hospital stay ranged from 3 to 22 days, and the duration of intravenous antibiotics ranged from 1 to 9 days with a clinical success rate of 96% without any drainage-related complications.
Conclusion: In contradiction to the earlier belief, percutaneous drainage is a safe and effective means of treatment in liver abscesses of >5 cm with high clinical success rate and reduced duration of intravenous antibiotic requirement as well as hospital stay.

INTRODUCTION

Liver abscess is a common health problem in tropical countries including India.[12] If not treated properly, it carries a high mortality.[34] With the advancements of imaging and the development of ultrasound-guided drainage procedures by the 1980s, antibiotic therapy coupled with percutaneous drainage had been accepted as the treatment modality for liver abscesses.[56] There are many studies on percutaneous drainage of liver abscesses worldwide, but we could find only very few studies on that of the eastern Indian population. Moreover, majority of the studies have found that catheter drainage is efficacious for the treatment of large abscesses (>10 cm). However, in this study, we evaluate the efficacy of catheter drainage in abscesses >5 cm. Hence, this study included patients undergoing sonographic diagnosis of liver abscess >5 cm. Moreover, the aim of this study was to find the demographic characteristics; comorbidities; and clinical, radiological, and bacteriological characteristics of liver abscesses as well as the efficacy of catheter drainage in patients with liver abscess >5 cm in eastern India.

PATIENTS AND METHODS

In this retrospective study, patients with liver abscesses of size >5 cm who had undergone percutaneous pigtail catheter drainage procedure in the department of radiology of our institution were included. The study period was from January 2018 to January 2020. Small-sized abscesses were excluded from the study. A total of fifty patients were included in this study. Past medical and surgical history; suspected immediate cause of the abscess; immunologic status; demographic data; laboratory findings; location, number, and size of the abscess; type of drainage; pre- and postprocedural duration of antibiotic therapy; and duration of hospital stay were recorded and analyzed. Follow-up for complete resolution or recurrence was done.

Preprocedural preparation

Written consent of the patient or guardian was taken

Total platelet count and prothrombin time (PT)/international normalized ratio (INR) of the patients were checked. The total platelet count had to be above 1 lakh and INR had to be below 1.5

USG screening was done to rule out ascites because presence of moderate-to-gross ascites is considered a contraindication for the procedure [Figure 1].

Figure 1

Ultrasound image shows a liver abscess ( arrow ) in the left lobe measuring approximately 8X10 cms

Materials

18G 10/15-cm Chiba needle

Injection xylocaine (1% or 2%)

No. 11 blade

10 F pigtail or Malecot catheter set (plastic dilator and 0.035” stiff guidewires)

2-0 Mersilk suture

Uro bag

10-mL syringes

Drape sheet, gauze pieces.

Procedure

A pigtail catheter was inserted into the liver abscesses using Seldinger technique. The patient was laid in the supine position. The site of percutaneous drainage was cleaned with betadine and draped. Under USG guidance, 5–7 mL of 2% xylocaine was infiltrated in the skin, deep up to the capsule of the liver, and then a small nick was given on the skin surface. 18G 10/15-cm Chiba needle was inserted percutaneously into the abscess cavity. Then, a stiff guidewire was inserted through the needle and the tract was dilated by placing 8F and 10F dilators over the guidewire. Then, 10 F pigtail or Malecot catheter was inserted into the abscess cavity and connected to an Uro bag to allow gravity drainage of the pus [Figures 2 and 3]. About 5 mL of the pus was put into a container for Gram stain, culture, and sensitivity.

Figure 2

Post Procedure image of the same patient showing a pigtail catheter ( arrow ) inserted into the abscess

Figure 3

External site of drain insertion in a case of left lobe liver abscess

All patients received preprocedural intravenous (IV) piperacillin, tazobactam 4.5 g TDS, and IV metronidazole 800 mg TDS, which were continued post procedure for 2–3 days, and then the patients were discharged with oral antibiotics, mostly oral cefixime 200 mg BD or ofloxacin 400 mg BD and metronidazole 800 mg TDS for 7–14 days.

Follow-up

Patients were monitored clinically. USG was done when drain output was <10 mL/day. Duration to attain clinical recovery, duration of hospital stay, duration of antibiotic use, complications, and death were recorded. After discharge, the patients were followed in the outpatient department, clinically as well as by USG on a monthly basis for 6 months.

RESULTS

In this study, a total of fifty cases of liver abscesses were included. Among these, 39 (78%) were male and 11 (22%) cases were female. The age group affected was 21–66 years, with the mean age being 45 years. Fifty percent of the patients had underlying comorbidity, the most common of which was diabetes in 32% of the cases. The most common symptom was fever which was found in 70% (n = 35) of the cases. Abdominal pain was the next most common, seen in 52% (n = 26) of cases. The most common sign was upper abdominal tenderness in 44% (n = 22) of cases.

The demographic data, the associated habits and comorbidities, and the signs and symptoms are summarized in Table 1. Right lobe was the most common site of occurrence of abscess (n = 28) in 56% of cases. It has been found that isolated left lobe affection (n = 7 equivalent to 14%) is less common than both lobe affection (n = 15 equivalent to 30%). Solitary abscess cavity was most frequently encountered in 70% (n = 35) of cases. Double abscess cavities and multiple abscess cavities were less commonly seen in 18% and 12% of the cases, respectively. The size of abscess cavity varies from 7.5 cm × 6.5 cm to 15 cm × 13 cm. No growth was seen in 70% of patients who had been empirically treated with abscess drainage and broad-spectrum antibiotics that included piperacillin-tazobactam and metronidazole. While Entamoeba histolytica was the causative agent in 12% of the patients, 8% of the patients were infected by Escherichia coli, 6% of the patients were infected by Klebsiella pneumoniae, and Acinetobacter and Burkholderia made 2% each.

Table 1

Demographic data, associated habits, comorbidities, signs and symptoms

Parameters	Number of patients	Mean/percentage
Age (years)	21-66	45
Gender
Male	39	78
Female	11	22
Habit/comorbidities
Diabetes	16	32
Alcoholic/smoker	7	14
GI surgery	1	2
CKD	1	2
Symptoms
Fever	35	70
Pain abdomen	26	52
Nausea/vomiting	12	24
Signs
Tenderness of upper abdomen	22	44
Hepatomegaly	15	30
Jaundice	6	12

GI: Gastrointestinal, CKD: Chronic kidney disease

Among the various laboratory parameters, total leukocyte count (TLC) had increased (>11 × 103/μl) in 35 patients equivalent to 70%, serum bilirubin had raised (>1.2 mg/dL) in 13 patients equivalent to 26%, serum total protein was low (<6 g/dL) in 16 patients equivalent to 32%, and INR had raised (>1.5) in nine patients equivalent to 18%. The laboratory findings are summarized in Table 2.

Table 2

Abnormal laboratory parameters in liver abscess patients

Parameters	Range and mean (M)	Percentage of patients with abnormal parameters
Total leukocyte count (1000/µl)	5.63-36.79 (17.81)	(>11) 70%
Total bilirubin (mg/dL)	0.44-14.9 (1.98)	(>1.2) 26%
Total protein (g/dL)	3.2-8.4 (6.2)	(<6) 32%
INR	1.04-2.52 (1.39)	(>1.5) 18%

INR: International normalized ratio

Clinical recovery was seen in 96% of the patients (n = 48), whereas death was seen in two patients, both were above 70 years of age. One patient was with pan drug-resistant K. pneumoniae and the other was a diabetic with multiple liver abscesses. The total duration of hospital stay varied from 3 to 24 days, with a mean duration of 12.8 days, while after abscess aspiration, the duration of hospital stay ranged from 1 to 22 days with a mean duration of 7.6 days. Young age with amebic liver abscess was seen to have early clinical recovery, while patients with advanced age, diabetes, jaundice, and pneumonia had delayed recovery with prolonged hospital stay. The duration of intravenous antibiotics post procedure varied from 1 to 9 days with a mean of 4 days and that of oral antibiotics varied from 7 to 14 days.

DISCUSSION

In recent years, minimally invasive percutaneous radiological interventions along with antibiotics usage have been successful in the treatment of liver abscesses, reducing the morbidity and mortality.[678] Several studies are available on the effectiveness of these minimally invasive procedures, however there are very few studies from eastern India. Again, this study focuses on the safety and efficacy of catheter drainage of abscesses >5 cm. Moreover, <5-cm abscesses are managed conservatively by suitable antibiotics after diagnostic aspiration.

In our study, the mean age of those affected was 45 years, which is similar to that of the study conducted by Mangukiya et al.[9] We found that males are more commonly affected. It is in accordance to the literature.[10] The most common comorbidity associated was diabetes mellitus. Yu et al.[11] and Jha et al.[12] also had similar findings in their studies, while Zerem and Hadzic had found cholecystitis as the most common associated comorbidity.[13]

The most common symptom the patients had in our study was fever in 72% of cases followed by pain in the abdomen in 52% of cases, which is consistent with the study of Yu et al.[11] However, Zerem and Hadzic found pain abdomen as the most common symptom.[13] Oschner et al.[14] and Norman et al.[15] showed that clinical features of liver abscess may be nonspecific. Hence, ultrasound scan can be very helpful in the diagnosis of this entity, which has reduced the mortality from 40% to 10%–25% in the last two decades.[161718]

A single liver abscess was our most common observation, seen in 70% of the cases studied and its most common location was right lobe of the liver, which is consistent with the study done by Mangukiya et al.[9] The mean size of abscess in their study was 6.87 cm (range 2–15 cm), while in our study, the mean abscess size was 8.5 cm (range 6.5–15 cm).

No microorganisms could be isolated by either Gram staining or culture in about 70% of cases which is because of the empirical use of antibiotics and antiamebic drugs before the percutaneous drainage procedure. Among the rest of the 30% positive culture or Gram stain, the most common organism found was E. histolytica (12%), followed by E. coli (8%) and K. pneumoniae (6%). In many previous studies, the rate of negative microbial study of the pus was very less, which is discordant with that of the present study. The most common organism attributed as the causative agent in previous studies was E. histolytica followed by E. coli and Klebsiella species, which is similar to that of our study.[1920]

The most common laboratory abnormality in patients with liver abscess in the present study was leukocytosis (TLC >11,000/mm3) in 70% of cases. Many other studies, including that of Rajak et al.[21] and Yu et al.,[11] had also found similar results. Furthermore, in the present study, serum bilirubin was raised in 26% of patients, INR was prolonged in 18% of patients, and serum protein was low in 32% of patients.

The average duration of IV antibiotics given in our study was 4 days (range: 1–9 days). In the study by Simon et al.,[11] the average duration was 12 days, and that in the study by Khan et al.[22] was 8.4 days. The duration of oral antibiotic administration in our study varied from 7 to 14 days. This reduced duration of intravenous antibiotics in our study may be because of catheter drainage in all the patients instead of needle aspiration. The success rate in our study was 96%. Similar success rate was seen in the study by Khan et al.[22] and Haider et al.[23] The duration of hospital stay varied from 3 to 24 days. Patients with diabetes and renal disorders require longer duration of injectable antibiotics and prolonged hospital stay.

One of the studies on percutaneous drainage of liver abscess from eastern India by Jha et al.[12] showed that the success rate for percutaneous management in patients with pyogenic liver abscess of size >300mL was 100%. In our study with the concurrent usage of catheter drainage and medical management for abscesses >5 cm, we were able to achieve a success rate of 96%.

Complications such as bleeding, pleural effusion or empyema, catheter displacement, and sepsis which were reported in the study by Lambiase et al.[3] were not seen in our study. Hence, percutaneous drainage of liver abscess is safe if performed properly in experienced hands.

CONCLUSION

This study revealed that for abscesses >5 cm, percutaneous catheter drainage is an effective and safe treatment of choice. As opposed to the earlier belief, it is not associated with longer hospital stay, more nursing care, or increased risk of complications. There was very good success rate with good clinical recovery.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.