Ultrasonography in the diagnosis of true gallbladder polyps: the contradiction in the literature

Abstract

Polypoid lesions of the gallbladder (PLGs) are often incidentally identified during ultrasonographic examination of abdominal pain. The present study was designed to determine the reliability of ultrasonography (US) in the diagnosis of PLGs. The records of 853 patients who underwent laparoscopic cholecystectomy (LC) for PLGs in Gazi Medical School from January 2000 to January 2004 were reviewed. Data were collected regarding the patients’ gender, age, symptoms, serum lipid levels, the size and the number of polyps on US, surgical indications for PLGs and histopathological diagnosis. In all, 56 of 853 patients had PLGs and underwent LC. Right upper quadrant pain (59%) was the most common presenting symptom that led to gallbladder US. Nearly 75% of the lesions were smaller than 10 mm. At histopathologic examination cholesterolosis was found in 17 of 56 (30%) patients, and 12 of 56 (21%) demonstrated only cholelithiasis; 17 (30%) patients had both cholesterolosis and stones. Only 10 (18%) patients had adenomatous polyp and 8 of these polyps were larger than 1 cm. Overall US-based diagnosis of gallbladder polyp was inaccurate in 82%. The sensitivity and specificity of US for polyps <1 cm was 20% and 95.1%, respectively, whereas the sensitivity and specificity of US for polyps >1 cm was 80% and 99.3%, respectively. The accuracy of US in diagnosing PLGs was poor, especially in polyps <1 cm.

Introduction

Polypoid lesions of the gallbladder (PLGs) are defined as masses protruding from the mucosal surface of the gallbladder. These lesions, as reported in the literature, range from 1.3% to 6.9% in both genders ¹,²,³. PLGs can be benign or malignant. In 1958, Carrera and Oschsner ⁴ evaluated polyps taken from more than 1300 cholecystectomies and described five types of mucosal polypoid lesions: inflammatory, cholesterol, adenoma, adenomyoma, and carcinoma. In 1970, Christensen and Ihsak ⁵ evaluated 180 gallbladder lesions at the Armed Forces Institute of Pathology. In addition to the categories described by Carrera and Oschsner, they also noted two granular cell tumors of the cystic duct and seven polyps of gastric heterotopic tissue. PLGs are usually diagnosed with ultrasonography (US) as nonshadowing, fixed, solid masses. Symptoms are unusual unless other abnormalities, such as stones, are present. The widespread use of US in patients with suspected gallstones and screening for other abdominal diseases has increased the detection rate of PLGs. Once identified, they pose a dilemma with respect to their proper long-term management ⁶. Physicians interpreting US reports describing PLGs must understand the accuracy and significance of this US finding, as it could affect the determination of the necessity for cholecystectomy. The present retrospective study was designed to determine the reliability of US in the diagnosis of PLGs. The literature was also reviewed for current diagnostic modalities for PLGs.

Patients and methods

The records of patients who underwent laparoscopic cholecystectomy (LC) with an US report showing PLGs in our institution from January 2000 to January 2004 were reviewed and data were collected regarding patients’ gender, age, symptoms, serum cholesterol levels, serum low-density lipoprotein (LDL) cholesterol levels, serum high-density lipoprotein (HDL) cholesterol levels, serum triglyceride levels, the size and the number of polyps on US, surgical indications for PLGs and pathologic evaluation of the gallbladder.

A 3.5-MHz linear-array transducer attached to an EUB-420 scanner (Hitachi, Tokyo, Japan) was used for the ultrasonographic examinations. The patients fasted overnight for 8 h, and then multiple scans were done with the patients supine and in left lateral decubitus position. To ensure that the lesions were not mobile, additional scans were made with the patients in different positions such as sitting or standing. PLG was defined on US as a lesion projecting into the lumen of the gallbladder which did not cast an acoustic shadow and did not move with the position of the patient (Figure 1). When two or more polyps were identified, the size of the larger polyp was used for analysis. All scans were performed by radiologists with special training in abdominal US.

Figure 1. .

Cholesterol polyp of the gallbladder. Ultrasonography shows a 12-mm, granular-surfaced pedunculated mass with an internal echo pattern characterized as an aggregation of echogenic spots with scattered echopenic areas.

Three months after undergoing an uneventful LC, a face to face interview was carried out to determine whether the symptoms had resolved in patients with gallbladder polyp on preoperative US.

Statistical analysis

The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of US for gallbladder polyps were calculated. Mean and standard deviations (SD) of the data were also calculated when necessary.

Results

The records of 853 patients who underwent elective LC in our institution were reviewed. This revealed that 56 of 853 patients had PLGs. There were 40 women and 16 men; the median age was 48.0 years (range 24–66). All the patients in this study had a preoperative ultrasonographic examination. The interval between the US study and LC ranged from 1 to 23 months (median 7 months). Eight (14%) of the 56 patients were asymptomatic and their PLGs were detected on US during a check-up. The remaining 48 (86%) patients had symptoms, which led to ultrasonographic examination. Right upper quadrant pain (59%) was the most common presenting symptom that led to gallbladder US. The other symptoms were nausea (41%), epigastric discomfort (32%), flatulence (22%), and food intolerance (15%). In the US study, the diameter of the polyp measured 0–5 mm in 22 patients, 6–10 mm in 20 patients, and >10 mm in 14 patients. Hepatosteatosis was seen only in one patient and only that patient had high serum cholesterol and LDL levels. Serum LDL, HDL, triglyceride and cholesterol levels were in normal ranges in patients with PLGs. On preoperative ultrasonographic evaluation 10 patients had multiple polyps and all of them were 4–7 mm in diameter. Indications for cholecystectomy in patients with PLGs were symptomatic gallbladder disease (n=49), enlarging size of polyp between serial ultrasonograms (n=6), and concomitant cholelithiasis (n=1). The characteristics of the patients are listed in Table I.

Table I. Clinical and demographic data of patients with gallbladder polyp on preoperative US.

Characteristic	Patients (n=56)
Age (years), median and range	48.0 (24–66)
Sex (male/female)	40/16
Laboratory data
Serum cholesterol (U/L) mean±SD	160.2±15.3
Serum HDL (U/L) mean±SD	42.1±10.2
Serum LDL (U/L) mean± SD	65.3±14.1
Serum triglyceride (U/L) mean±SD	70.4±22.2
Symptoms
Asymptomatic (%)	6(11)
Right upper quadrant pain (%)	33 (59)
Nausea (%)	23 (41)
Epigastric discomfort (%)	18 (32)
Others (%)	10 (18)
Indications
Any symptomatic patients (%)	49 (87)
Enlarging size of polyp (%)	6(11)
Concomitant cholelithiasis (%)	1(2)

At histopathologic examination cholesterolosis was seen in 17 of 56 (30%) patients, and 12 of 56 (21%) demonstrated only cholelithiasis; 17 (30%) patients had both cholesterolosis and stones. Only 10 patients had an adenomatous polyp and 8 of these were >1 cm in diameter. No malignancies were identified (Table II). The remaining 797 patients had only cholelithiasis on preoperative US. On the other hand, pathological examinations revealed that 10 of these patients also had various types of polyps (Table III). No malignancy was seen in these patients.

Table II. Histopathological diagnosis and maximal diameter of polypoid lesions.

	Polyp diameter on US
Histopathological diagnosis	<6 mm	6–10 mm	>10 mm	Total%
Cholesterolosis	10	6	1	17 (30)
Cholelithiasis	5	5	2	12 (22)
Cholesterolosis and cholelithiasis	6	8	3	17 (30)
Adenomatous polyp	1	1	8	10 (18)
Total	22	20	14	56 (100)

Table III. Histopathological findings of PLGs that were missed on preoperative US.

Age (years)	Gender	Permanent histopathologic examination	Size (mm)
44	Male		4
52	Female		3
55	Female	Inflammatory polyp	4
48	Female		5
51	Female		5
56	Male	Gastric heterotopic tissue	5
43	Female		4
35	Female		12
46	Female	Adenomatous polyp	11
54	Female		4

During the follow-up period, a face to face interview was carried out with 41 of 49 patients who underwent LC because of symptomatic gallbladder polyp. Eight patients did not reply to the invitation. The symptoms resolved in 38 (92.6%) of these 41 patients. The histopathological examination of enlarging polyps between serial ultrasonograms revealed that five of six patients had adenomatous polyps and one had cholethiasis with cholesterolosis. The sensitivity and specificity of US for polyps <1 cm was 20% and 95.1%, respectively. On the other hand, the sensitivity and specificity of US for polyps >1 cm was 80% and 99.3%, respectively (Table IV).

Table IV. The sensitivity and specificity of US for patients with PLGs.

Parameter	Polyps <1cm	Polyps >1 cm
Sensitivity (%)	20	80
Specificity (%)	95.1	99.3
PPV (%)	4.76	57.14
NPV (%)	98.99	99.74

PPV, Positive predictive value; NPV, negative predictive value.

Discussion

Conventional US is routinely used to evaluate the gallbladder. During ultrasonographic examination, the nonshadowing single or multiple fixed masses that project into the lumen of the gallbladder are regarded as PLGs ¹,⁷. The polyps can be neoplastic (adenoma, cancer) or non-neoplastic (cholesterol polyp, inflammatory). Surgeons and other physicians need to be aware of the significance of PLGs noted on the ultrasonogram. Preoperative correct diagnosis is mandatory for decisions regarding surgery. The present study demonstrated that US-based diagnosis of PLGs was inaccurate in 82% of patients. There is a disagreement in the literature concerning the accuracy of US in diagnosing PLGs. Yang et al.⁸ reported a 94% sensitivity of US for detecting PLGs. On the other hand Damore et al.⁹ reported that US-based diagnosis of PLGs was inaccurate in 95% of patients undergoing cholecystectomy. In this study the following ultrasonographic criteria were used for the diagnosis of polyps. (1) An image with similar echogenicity to the gallbladder wall that projects to lumen, (2) fixed image, (3) lacks displacement, and (4) lacks an acoustic shadow. To differentiate an impacted stone that can mimic a polyp on US, scanning was performed after repositioning the patient in the prone or right lateral decubitis position. This makes it technically easier to prove the presence of a stone, because repositioning the patient causes the stone to move. A combination of stones and cholesterolosis may mimic PLGs. Echoes produced by gallstones are usually high in amplitude; however, occasionally they are less echogenic than expected. This appearance occurs most often in patients with soft stones that have mud-like consistency, as seen in patients with cholesterolosis. Additionally, a gallstone impacted in the gallbladder wall that is located in a dependent position can give the appearance of a small PLG.

The differences in the findings between gallbladder US and gallbladder pathology reports are related to several factors. It was very difficult to diagnose PLGs on the basis of the size of lesion. A number of papers have reported that diagnosis of PLGs <10 mm remains difficult in many cases ¹⁰,¹¹,¹². In the present study, histopathological examination revealed only two true polyps in patients who had PLG <10 mm on preoperative US. Strikingly, 14 patients had PLGs >10 mm on preoperative US and at histopathological examination 8 of them were true polyps. In this study, the sensitivity and specificity of US for polyps < 1 cm was 20% and 95.1%, respectively. On the other hand, the sensitivity and specificity of US for polyps > 1 cm was 80% and 99.3%, respectively.

Cholesterolosis of the human gallbladder is a common disease and its reported prevalence has varied from 9% to 24% ¹³,¹⁴. Cholesterolosis is characterized histologically by the accumulation of lipids in the lamina propria of the gallbladder. These lipid deposits consist mostly of cholesterol ester and tri-glycerides ¹⁵. The etiology of cholesterolosis is not yet fully understood. In this study cholesterolosis was seen in 34 of 56 (61%) patients and we did not find any correlation between serum LDL, HDL, triglyceride and cholesterol levels and cholesterolosis. Similar to our findings, Mendez-Sanchez et al.¹⁶ reported that cholesterolosis is not associated with high cholesterol levels in patients with and without gallstone disease. Also, Watanabe et al.¹³ suggested that cholesterol ester synthesis of gallbladder mucosa might play an etiological role in the development of cholesterolosis.

The ultrasonographic technology continues to improve and endoscopic ultrasonography (EUS) seems to be superior to conventional US for imaging the gallbladder. US can demonstrate various polypoid lesions of the gallbladder; however, the differential diagnosis of polypoid lesions <10 mm remains difficult in many cases. US is limited as a diagnostic tool, if the diagnosis depends solely on the shape and surface characteristics of the polypoid lesions. In contrast, EUS can provide information not only about the morphology and surface features, but also the internal echo status of the polyp. The internal echo of a true polyp is usually hypoechoic to isoechoic and almost homogenous, if not slightly heterogeneous. EUS can readily image the layer structure of gallbladder wall and provide high-resolution images of small lesions ¹⁰,¹⁷,¹⁸. Once a polyp of the gallbladder has been identified by standard US, the next question comes to mind:‘Is this benign or malignant?’. Judging from the US diagnostic results in the present study, for patients with polyps <10 mm in size, EUS should be conducted for differential diagnosis. If patients with PLGs are asymptomatic, it is particularly important to differentiate between benign and malignant lesions, because implementation of surgery depends on the differentiation. Therefore, when it is difficult to differentiate between malignant and nonmalignant US-detected gallbladder polyps in asymptomatic patients by US alone, EUS could be best applied. Therefore, EUS is thought to play an important role in determining the treatment strategy for PLGs.

A recent study assessed the use of positron emission tomography (PET) scanning using F-labelled deoxy-glucose (FDG) and the authors reported that FDG-PET may become one of the most useful tools for accurate preoperative diagnosis of small PLGs ¹⁹. Helical computed tomography seems to be another alternative method for differentiating neoplastic and non-neoplastic small PLGs ²⁰.

Taking into account the inaccuracy of US in the diagnosis of PLGs, patients with symptomatic PLG should be offered laparoscopic cholecystectomy. Asymptomatic patients with PLGs, especially lesions < 1 cm, should rescanned. If the lesion increases in size between serial ultrosonograms, they should once again be offered surgery.