Outcome of gall bladder polypoidal lesions detected by transabdominal ultrasound scanning: A nine year experience

Abstract

AIM: To determine the outcome of polypoidal lesions within the gall bladder (PLG) diagnosed by trans-abdominal scanning.

METHODS: A nine-year (1993-2002) retrospective case-note review of all patients who underwent ultrasound scanning after referral to a single Upper GI Surgeon at a District General Hospital was conducted. Patients who were diagnosed with a PLG were included in our study. A database was constructed and patient details, investigations including ultrasound scan (USS) findings, treatment and histology and final diagnosis were recorded.

RESULTS: Twenty-three (out of 651) patients were diagnosed pre-operatively by USS to have a polyp-like gall bladder lesion (PLG). Post cholecystectomy histological examination revealed 12 gallstones, 7 cholesterol polyps, 3 adenocarcinomas within polyps and 1 normal gall bladder. The specificity of USS in the diagnosis of PLG was 92.3%. All the true polyps were malignant. Overall USS had 66.66% sensitivity and 100% specificity in the pre-operative suspicion of malignancy. Using size greater than 10 mm as measured on USS as a cut-off, we find 100% sensitivity and 86.95% specificity with a positive predictive value of 50% in the diagnosis of malignancy in PLG.

CONCLUSION: A large number of PLG are in fact calculi within diseased gall bladder. In cases of gall bladder polyps more then 10 mm in size on USS further imaging (cross-sectional and/or EUS) is indicated prior to surgery. This will help in the optimal management of patients and avoid histological surprises.

Key Words: Polypoidal lesion in gall bladder; Malignant polyps

INTRODUCTION

A polypoidal lesion may be defined as an elevation on the gall bladder mucosa. Such polypoid lesions affect 5% of the adult population[1]. The majority of gall bladder polyps are benign, most commonly being cholesterol polyps, however it is their malignant transformation that is the cause for concern[1].

Initial imaging for hepatobiliary disease is usually ultrasonography, hence both surgeons and physicians are often presented with an ultrasound report suggestive of a polypoid lesion within the gall bladder (PLG)[2]. Generally polyps in the gall bladder are demonstrable on ultrasonography, only when they are over 5 mm in diameter. Sonographic differentiation between benign and malignant polyps (and calculous disease) relies greatly on the size of a single non-mobile lesion within the gall bladder. Simple cholecystectomy is curative in early gall bladder carcinoma[3]. Such curative resection is limited to cases in which the carcinoma was diagnosed in gall bladders resected for benign disease[4,5]. Their malignant potential increases with size, as demonstrated in a Japanese study where majority of malignant tumors were over 10 mm[6].

Our study was a retrospective case note review of patients referred to a consultant surgeon between 1993 and 2002. The inclusion criterion into the study was the suggestion of PLG on ultrasonography. The aim of this study was to determine the efficacy of ultrasound scan in diagnosing polyps, by correlating post cholecystectomy histopathological findings with the preoperative sonological results. The literature on existing controversy in imaging and management is thereafter reviewed and discussed briefly.

MATERIALS AND METHODS

In the nine years between April 1993 and March 2002, 751 cholecystectomies were performed at a district general hospital. Six hundred and fifty-one case notes were individually reviewed and ultrasound findings were correlated with final gall bladder histopathology report. The Ultrasound probe used was a 3.5/5.0 MHz transducer. The sonological study was performed by consultant radiologists as per hospital protocol. In cases where sonography was suggestive of PLG, the scan was repeated by the same radiologist, to check for the mobility of the lesion in an attempt to exclude calculous etiology.

The only inclusion criterion was the presence of a solitary lesion within the gall bladder on pre-operative USS. This inclusion into the study did not differentiate between the method of definitive surgical management, i.e., laparoscopic or open procedure. The histopathological reporting was done by consultant pathologists, and only formal reports were considered.

RESULTS

Preoperative ultrasonography suspected PLG (polyp like gall bladder lesion) in 23 out of 651 patients. In the remaining 628 patients the USS diagnosis was multiple gallstones with or without acute/chronic inflammation.

Among our study group (n = 23) the female to male ratio was 16:7 with a mean age of 56.8 years. Twenty-one patients had symptomatic disease while the remaining two were symptom free.

A diagnosis of polyp like gall bladder lesion was made on the basis of: (1) An immobile gall bladder lesion with no post-acoustic shadows; (2) Normal common bile duct diameter. The range of the sizes of PLG was 2-22 mm (mean 7.91, standard deviation 4.26). Histological examination revealed multiple lesions in three patients all of which turned out to be gall bladder calculi. These results are detailed in Table 1. The sensitivity of USS in making a diagnosis of PLG was 100% (this is obvious given the inclusion criteria of our study), the proportion of patients correctly diagnosed as having a gall bladder abnormality (specificity) was 92.3%, since one patient had a histologically-normal gall bladder.

Table 1 Summary of USS and pathological features.

	Symptom	Post acoustic shadow	Lesion –number/mobility	Size mm	USS diagnosis	Histologic diagnosis
1	Bil Colic	No	1/immobile	2-3	PLG	Cholesterosis
2	Bil Colic	No	1/immobile	5	PLG	Calculi
3	Bil Colic	No	1/immobile	5	PLG	Cholesterosis
4	Bil Colic	No	1/immobile	5	PLG	Cholesterosis
5	Bil Colic	No	1/immobile	5	PLG	Calculi
6	Bil Colic	No	1/immobile	6	PLG	Cholesterosis
7	Bil Colic	No	1/immobile	6	PLG	Calculi
8	Bil Colic	No	1/immobile	7	PLG	Calculi
9	Bil Colic	No	1/immobile	8	PLG	Calculi
10	Bil Colic	No	1/immobile	8	PLG	Calculi
11	Bil Colic	No	1/immobile	9	PLG	Calculi
12	Bil Colic	No	1/immobile	9	PLG	Calculi
13	Bil Colic	No	1/immobile	10	PLG	Calculi
14	Bil Colic	No	1/immobile	12	PLG	Adenocarcinoma
15	Bil Colic	No	1/immobile	18	? malignant	Adenocarcinoma
16	Atyp pain	No	1/immobile	2-3	PLG	Cholesterosis
17	Atyp pain	No	1/immobile	5	PLG	Calculi
18	Atyp pain	No	1/immobile	5	PLG	Cholesterosis
19	Atyp pain	No	1/immobile	6	PLG	Calculi
20	Atyp pain	No	1/immobile	6	PLG	Normal
21	Atyp pain	No	1/immobile	20	? malignant	Adenocarcinoma
22	Asymp	No	1/immobile	10	PLG	Calculi
23	Asymp	No	1/immobile	12	PLG	Cholesterosis

Bil Colic – biliary colic.

Ultrasound scan could not exclude gall bladder malignancy in two PLG cases. The suspicion about the malignant nature of these two PLG was the size of the polyp within the gall bladder (18 mm and 20 mm). In spite of the suspicion of cancer of gall bladder, no further imaging was performed in both cases because: (1) Cross sectional imaging was not widely available in the earlier years of our study; (2) Solitary lesions within the gall bladder were assumed to be due to calculous disease necessitating cholecystectomy.

In all the 23 patients laparoscopic cholecystectomy was performed. Median hospital stay was 2 d with a range of 2-5. There were no peri- or post-operative complications.

Histopathological examination of these 23 resected gall bladders revealed calculous cholecystitis in 12 (52%) cases, cholesterosis in eight patients and a focus of adenocarcinoma in three specimens. No histopathological abnormality was found in one gall bladder. No true benign polyps were seen, all the three true polyps contained a focus of adenocarcinoma.

Out of the three histologically proven gall bladder malignancies, pre-operative USS raised the suspicion of malignancy in two cases. With regard to a diagnosis of malignancy, USS had 66.66% sensitivity and 100% specificity (true positives: 2, false positives: 0, true negatives: 20, false negative: 1 [USS diagnosis was a solitary calculus]). One other gall bladder specimen was reported to have no abnormality on histopathology despite a preoperative suspicion of PLG on ultrasonography.

DISCUSSION

Majority of polypoidal lesions within the gall bladder are due to calculous biliary disease. Cholesterolosis is defined as the result of accumulation of triglycerides and esterified sterols in macrophages of the lamina propria and can give ultrasonographic appearance of small polyps[7]. Such polyps are reportedly pedunculated and are attached to the gall bladder by a delicate pedicle that becomes easily detached before or during a cholecystectomy[8]. A gallstone impacted within the gall bladder wall may be easily mistaken for a polyp on ultrasound scanning[2]. Thus, several factors could contribute to the discrepancies of the USS findings and the final diagnosis after surgical excision.

In 12 of our 23 cases the PLG turned out be true gall bladder calculi. These findings are in keeping with a retrospective analysis of 41 patients with PLG who underwent cholecystectomy where the definitive diagnosis was cholesterolosis (17 of 41) or cholelithiasis (15 of 41)[2].

In our study ultrasonography was not accurate in differentiating polyps (true or pseudo that is; cholestrerol polyps) from calculi (66.7% sensitivity). This differentiation assumes significant importance given that true polyps that is, adenomas/papillomas have the potential for malignancy[9]. A level of 15 mm as a cut-off, which appears to have been used as a USS marker for malignancy has missed suspecting a 12 mm PLG as a possible malignant polyp. All the three malignant gall-bladder polyps have been above 10 mm in size. Using size greater than 10 mm as measured on USS as a marker, we find 100% sensitivity and 86.95% specificity with a positive predictive value of 50% (three lesions greater that 10 mm were non-malignant).

The efficacy of ultrasound in diagnosing gall bladder polyps is controversial as evidenced by the following two studies. In a study reported in 1993, out of 23 patients with polyps on USS only 13 (57%) true polyps were found on histology[10]. Yang et al[11] (1992), have reported an ultrasound scan sensitivity of 90% in detecting gall bladder polyps after reviewing post-cholecystectomy specimens with a preoperative diagnosis of polyps of the gall bladder. Sugiyama et al (1995), suggested routine use of endoscopic ultrasound (EUS) in the patients with a USS suggestive of a polyp. They suggested that aggregation of echogenic spots seem to be pathognomonic of cholesterol polyps. Another group[12] suggest that the nonenhanced and enhanced CT scan differentiate neoplastic from non-neoplastic PLGs. CT biliary cystoscopy has recently been suggested as a non-invasive and accurate means of assessing gall bladder polyps[13]. However, in most of the units it may not be logistically practical and financially feasible to perform EUS or CT purely to differentiate PLG.

The management of polyps found on USS is conflicting with some offering selective surgery while others offer routine laparoscopic cholecystectomy[12,14]. In our study, USS was not very specific for the diagnosis of gall bladder polyps as 52% (12 out of 23) were found to be calculi within the gall bladder. However as 22 of the 23 resected gall bladders were pathological, we suggest that gall bladder with polypoidal lesions on USS have high probability of having intrinsic gall bladder disease and hence should undergo cholecystectomy. Further a size criteria of greater than 10 mm as a marker for further investigation of the gall bladder abnormality seems reasonable. This will aid optimal management planning and avoid histological surprises.

In conclusion, ultrasound evidence of a polypoidal lesion within the gall bladder is an indication for a cholecystectomy since nearly all such gall bladders are diseased. Most of such lesions turn out to be having calculous etiology. Lesions which, measure more than 10 mm on ultrasonography need to be imaged further.