Polypoid lesions of the gallbladder may be defined as elevations on the mucosa of the gall bladder (GB) which affect 4 to 6% of the normal population [1-3]. As the radiologic tools such as ultrasonography (USG) and computed tomography (CT) develop, the frequency of detecting many diseases, such as GB polyps, has increased [1-4]. To distinguish between a benign polyp and a malignant polyp is very important because untreated malignant GB polyps have a poor prognosis. And because of the poor prognosis, early diagnosis and early treatment are very important [5-8].

Currently, there are predicting factors that are useful to distinguish benign from malignant. Size, number of polyps, and age of the patient are the clinical data that are used. But these factors are not sufficient in deciding which surgical procedures are appropriate, especially in polyps around 10 mm. Improved diagnostic methods are needed to differentiate between benign and malignant GB polyps, and to determine which polyps require operation [9-12].

If we can differentiate cholesterol polyps from non-cholesterol polyps, it may be the first step in detecting whether the polyp has the likelihood of being a malignant one. Therefore, we evaluated clinical data which would be helpful in distinguishing cholesterol polyps from non-cholesterol polyps.

This is a retrospective study on analyzing preoperative USG and CT findings compared with their postoperative gross and microscopic findings associated with the patients' clinical data.

Between January 2000 and December 2009, 359 patients underwent cholecystectomy under the diagnosis of GB polyp of around 10 mm. Only 187 patients (182 laparoscopic, 5 open) were diagnosed with GB polyp of any kind postoperatively and the 187 patients were enrolled in this study. The patients were divided into two groups (146 in cholesterol polyp group and 41 in non-cholesterol polyp group) based on the postoperative pathological findings, retrospectively.

Clinical features such as gender, age, bodyweight, body mass index (BMI), symptoms, laboratory findings, size, number of polyps, presence of GB stone, maximum diameter measured by preoperative radiologic findings of the USG, CT, and the pathological diameter of the postoperative findings were subjected to comparative analysis.

For comparison, the GB polyps were measured on every CT view and the longest was accepted as the polyp length. The specimen was fixed in formalin within 20 minutes and was sent to the pathology department and entirely embedded for pathological comparison. Results were reported as the mean ± standard deviation. For statistical analysis, a chi-square, t-test and Mann-Whitney U test were used PASW ver. 18 (IBM, New York, NY, USA). A P-value < 0.05 was considered statistically significant.

Though a laparoscopic cholecystectomy can be considered a less complicated operation, an operation that is unnecessary is truly a burden to the patient and a waste of cost. To prevent these operations, the diagnosis of GB polyps has to be made correctly. This study was intended to characterize the clinical features of the cholesterol polyp and to determine the accurate radiological predictive factors.

Age is a known risk factor associated with malignancy [13-15]. This study also showed a higher mean age in the non-cholesterol polyp group. Many studies show the relationship between metabolic syndrome and the development of cholesterol polyps [2,16,17]. Our study showed a significant data of higher BMI in cholesterol polyps compared to non-cholesterol polyps [11]. And the cholesterol level was higher in the cholesterol polyp group without significant difference.

It is known that a single polyp is more likely to be a malignant polyp [13,18]. If a single polyp is identified it needs more aggressive work up or more aggressive interventions than multiple polyps. Among our study populations, over 90% of the non-cholesterol polyps was solitary and about 60% of cholesterol polyps was multiple polyps. In other words our study also shows a need for aggressive work up in solitary polyps as the ratio is higher in non-cholesterol polyps. Many studies have reported that polyps over 10 mm have a high risk of being a malignancy, and this is surgical criteria for treating GB polyps [11,19-21]. The pathologic size of the polyps in our study also had a similar tendency. But there were many cholesterol polyps larger than 10 mm in the group. Therefore, it is hard to state size as a definite factor to distinguish benign from malignant polyps [22-25].

We observed discrepancies between preoperative radiological measurements and postoperative pathologic measurements in the cholesterol polyp group. And the cholesterol polyp group also showed low correlation coefficients between the three measured sizes. The damage of cholesterol polyps during operation or handling after obtaining the GB may explain the results. As a result, the preoperative radiologic studies are limited in obtaining the correct measurements for cholesterol polyps. On the contrary, non-cholesterol polyps have a high correlation coefficient between the preoperative diameters measured by USG and CT. And the correlation coefficiency compared with the USG and CT of postoperative pathologic diameters are also significantly high.

The predictive preoperative radiologic signs for a non-cholesterol polyp include a single polyp, observable on both the USG and CT, and a discrepancy of the maximum diameter less than 1 mm between the USG and CT.

In conclusion, young patients with high BMI and high cholesterol levels who have multiple GB polyps near 10 mm in size measured by USG can be recommended a CT scan [26-28]. And if the diameter has a discrepancy or only seen in the USG, the GB polyp can be considered as a cholesterol polyp and require planning for another follow-up after 6 months [29,30] thereby delaying the operation. We suggest that it would be more efficient to make flexible treatment plans for the mentioned cases rather than the fixed guideline.