Accuracy of gallbladder polyp size in predicting and detecting gallbladder cancer-a single-center study on gallbladder polyp

DOI: https://doi.org/10.21203/rs.3.rs-1600177/v1

Abstract

Background: Despite the fact that most gallbladder polyps are benign, malignant lesions are not so rare as thought. We performed this study to evaluate size distribution of gallbladder polyps and assess efficiency of reported risk factors in predicting malignancy.

Methods: In this study, medical records of patients having undergone laparoscopic cholecystectomy for gallbladder polyps at Department of Hepatobiliary and Pancreatic Surgery, Shenzhen People,s Hospital were reviewed. Chi-square test and binary logistic regression analysis were adopted to determine risk factors for malignant lesions. Additionally, receiver operating curve (ROC) analysis was accomplished to identify the optimal cutoff size.

Results: A total of 1012 patients were included in this study. Of the 1012 patients, 977 ones were diagnosed with benign gallbladder polyps while 35 ones with malignant polyps. Diameters of the 977 benign gallbladder polyps were between 1 and 50 mm while the 35 malignant polyps were between 4 and 54 mm. Incidence of gallbladder cancer among lesions ≥10mm was 15.5%, 31.5% for lesions ≥15mm and 49.0% for lesions ≥20mm. Among polyps in the following four size ranges (≤5mm, 6-9mm, 10-14mm, and 15-19mm), cholesterol polyps were the most common. While among polyps between 20 and 24mm, adenomas were the most common and among polyps ≥25mm, malignant tumors were the most common. Sessile morphology was significantly more common in malignant lesions (91.4%) than in benign lesions (3.7%, P<0.001). Multiple polyps were frequently diagnosed not only as cholesterol polyps (82%) but also as adenomas (4.3%) and gallbladder cancer (0.3%). We demonstrated that sessile morphology, blood flow signal on ultrasonography and single polyps were independent predictive factors for gallbladder malignancy.

Conclusions: Detection rate of malignant lesions increased significantly with the increase in diameter. Risk factors such as sessile morphology, blood flow signal on ultrasonography and single polyps were effective factors predicting malignant lesions. It was difficult for us to accurately confirm the pathological diagnoses of gallbladder polyps solely based on preoperative examinations given the fact that cholesterol polyps and adenomas were quite common, even among large gallbladder polyps. 

Introduction

As projections of the gallbladder wall into the lumen, gallbladder polyps are commonly detected in routine clinical practice[1]. Pathologically, most of these lesions are pseudopolyps that are consisted of cholesterol polyps without the potential to become malignant. However, other types of polyps consisted of malignant lesions (such as gallbladder cancer) or benign lesions (such as adenoma) also exist[2, 3]. Usually, risk of malignancy is evaluated by measuring the size of polyp. Given the fact that polyps larger than 10mm have a significantly higher risk of becoming malignant, several guidelines recommended that these polyps should be resected as soon as possible [46]. However, not all polyps larger than 1cm are malignant. Preoperative evaluation of a gallbladder polyp,s being malignant or benign bears huge significance since range of surgery for malignant lesion is remarkably different that for benign lesions. Preoperative assessment is mainly based on imaging examinations such as ultrasonography, computed tomography and magnetic resonance imaging. Some studies suggested that gallbladder polyps ≥ 2cm are much more likely to be malignant[24, 7, 8]. However, in clinical practice, cases where a benign gallbladder polyp is demonstrated to have a diameter ≥ 2cm are quite common.

Conversely, we have also encountered cases that gallbladder cancer presented as a lesion smaller than 1cm and some studies have reported the existence of gallbladder cancer smaller than 1cm[710]. On the contrary, according to other studies, most of gallbladder polyps smaller than 1cm were benign and almost 23.5% of polyps smaller than 1cm would shrink or vanish with the time gone by[11, 12]. Furthermore, in a few studies, algorithms assessing risk of being malignant and determining the most appropriate treatment choice for gallbladder polyps smaller than 1cm have been proposed by some previously published studies[2, 4, 12, 13]. However, by far no consensus has reached. In 2017, European Society of Gastrointestinal and Abdominal Radiology (ESGAR), European Association for Endoscopic Surgery and other Interventional Techniques (EAES), International Society of Digestive

Surgery-European Federation (EFISDS), and European Society of Gastrointestinal Endoscopy (ESGE) put forward a joint guideline reporting an algorithm stating that gallbladder ployps smaller than 1cm should be divided into those smaller than 6mm and those between 6 and 9mm considering those two groups should adopt different follow-up strategies[6]. Using this algorithm, we may select the optimal treatment choice for gallbladder polyps smaller than 1cm.

We performed this study with the aim of assessing the distribution of malignant lesions among polyps of different sizes and evaluating the predictive capabilities of some previously reported risk factors. Particularly, we furthermore investigated whether the algorithm proposed by the joint ESGAR, EAES, EFISDS and ESGE guidelines could also be adopted to efficiently detect small gallbladder cancer among patients with polyps undergoing cholecystectomy at our center.

Materials And Methods

Data retrieval and management

Data of patients who had undergone laparoscopic cholecystectomy or conversion to open surgery at Department of Hepatobiliary and Pancreatic Surgery, Shenzhen People,s Hospital between January 2018 and December 2021 were retrieved from the medical records. Then those with the diagnosis of gallbladder polyp or gallbladder cancer were screened out for further analysis. Sizes of gallbladder polyps or cancer were identified by measuring resected fresh specimens. At our institution, laparoscopic cholecystectomy was initially prepared for all the patients (even for patients with suspected gallbladder cancer), and conversion to open cholecystectomy was performed when evidence of gallbladder cancer was confirmed, such as invasion of serosal layer, invasion of adjacent liver tissue or apparent lymph node metastases. For patients whose malignant lesions were confirmed after surgery, radical resection would be determined by depth of cancer invasion. And adjuvant chemotherapy using gemcitabine or tegafur/oteracil/gimeracil would be recommended if metastatic lymph nodes were identified. 

Medical records of patients included in this study were carefully analyzed to determine size of gallbladder lesions treated at our department, considering the fact that size of gallbladder polyp was the most crucial factor determining whether surgery should be applied. According to joint guidelines, for gallbladder polyps ≥10 mm and those between 6 and 9mm with other risk factors, surgery should be applied as prompt as possible[1]. According to the maximum diameter, gallbladder lesions detected between 2003 and 2019 were categorized according to the maximum diameter and numbers of each kind of lesions of different size ranges were counted. In this study, superficial or fat cancers were also excluded since these lesions could be easily differentiated from benign gallbladder lesions[14]. And for lesions suspected to be malignant, intraoperative frozen section examination would be routinely performed. 

Ethical approval

This study was approved by the Ethics Committee, Shenzhen People,s Hospital and each patient involved in this study had given his or her informed consent in written form. Declaration of Helsinki was adhered to during the whole process of this study[15]. 

Statistical analysis

Continuous variables were presented as medians while those categorical ones as frequencies and percentages. Differences in proportions were evaluated by Chi-square test while significance of differences for continuous variables were assessed by Mann-Whitney u test. Additionally, independent risk factors for gallbladder malignancy were identified by binary logistic regression analysis. All the statistical analyses in this study were accomplished by the SPSS 22 (Armonk, New York, IBM Corp). Comparisons with P values <0.05 were considered as statistically significant. 

Results

Size distribution of malignant and benign gallbladder polys

After reviewing the medical system of our hospital, we identified 1012 patients diagnosed with gallbladder polyps. Sizes of benign gallbladder polyps ranged from 1 mm and 50 mm while diameters of malignant gallbladder polyps were between 4 mm and 54 mm. Representative figures demonstrating different pathological diagnoses were presented in Figure 1. Distribution of malignant polyps and benign polyps was demonstrated in Table 1 and Figure 2, from which we could reveal that significantly more gallbladder cancers were detected among larger polyps. Sensitivities and specificities of different cutoff values were calculated by receiver operating curve analysis (Figure 3) and specific values were presented in Table 2. Area under curve (AUC) was 0.949 (P<0.001, 95%CI: 0.916-0.982) and 13.5mm was the optimal cutoff value. Incidence of gallbladder cancer among lesions ≥10mm was 15.5%, 31.5% for lesions ≥15mm and 49.0% for lesions ≥20mm. 

Pathological diagnoses of benign gallbladder polyps

Numbers of different pathological diagnoses of the 977 benign gallbladder polyps were demonstarted in Table 3. Of the 977 polyps, cholesterol polyps (643 cases; 65.8%) were the most common and adenomyomatosis were the second most common (159 cases; 16.3%). A total of 104 benign adenomas (10.6%) were detected and all of these 104 adenomas were tubular adenomas. Cholesterol polyps were the most common in the following size ranges: ≤5mm, 6-9mm,10-14mm, and 15-19mm. While among benign polyps ≥20mm, adenomas were the most common. 

Risk factors predicting malignant lesions of gallbladder

It has been suggested by joint guidelines that risk factors predicting malignant lesions of gallbladder are senior age (older than 50 years), a broad-base morphology (or sessile), and focal thickening of gallbladder wall. We assessed these risk factors for gallbladder malignancy. Sessile morphology was more frequently identified among malignant polyps (91.4%, 32 of 35 cases) than among benign lesions (3.8%, 37 of 977 cases, P<0.001). 15 adenomas were demonstrated to have a sessile morphology (14.4%, 15 of 104 cases). Percentage of older patients (>50 years) was significantly lower for benign polyps (36.3%, 355 of 977 cases) than gallbladder malignant lesions (82.9%, 29 of 35 cases) (P<0.001). Appearance of multiple lesions is usually considered more likely to signify the presence of cholesterol polyps while a single lesion is much more likely to be an adenoma. However, in the present study, many patients with multiple polyps were not only diagnosed with cholesterol polyps (82.0%, 486 of 592 cases) but also with adenomas (6.9%, 41 of 592 cases) and gallbladder cancer (0.33%, 2 of 592). Adenomas or gallbladder cancer presented as multiple lesions usually consisted of a single adenoma or cancer and multiple cholesterol polyps. Then binary logistic regression analysis was performed to identify independent predictive factors for gallbladder cancer, revealing that age (P=0.006, OR=1.094, 95%CI: 1.026-1.165), size (P=0.01, OR=1.683, 95%CI: 1.131-2.505), sessile morphology (P=0.001, OR=16.239, 95%CI: 3.047-86.542), and flow signal on ultrasonography (P<0.001, OR=48.505, 95%CI: 7.125-330.198) were independent predictive factors for gallbladder cancer (Table 4 and Table 5). 

Four cases of gallbladder cancer smaller than 10mm and nine cases of benign lesions of gallbladder with a diameter  30mm

Four gallbladder polyps smaller than 10mm were confirmed as gallbladder cancer. The first one was a 60-year-old female patient with a 4mm lesion. The second one was a 48-year-old female patient with a 8mm lesion. The third one was a 57-year-old female patient with a 8mm lesion. The fourth one was a 84-year-old male patient with a 8mm lesion. There were nine cases of benign lesions of gallbladder with a diameter ≥ 30mm, six ones of which were adenoma, one adenomyomatosis, and two mixed polyps (cholesterol polyp and adenoma). 

Discussion

In this study, 88.6% of all the malignant polyps were larger than 10mm (31 of 35 cases). Although four malignant polyps smaller than 10mm were identified, all the four malignant polyps own other characteristics predicting gallbladder cancer. Three ones were with sessile morphology, three ones with blood flow signal on ultrasonography and three ones were single polyps. According to the joint guideline, sessile morphology, blood flow signal on ultrasonography and single polyps were risk factors for gallbladder cancer[1]. And in this study we also confirmed these three factors as independent predictive factors for gallbladder malignant lesions. In routine clinical practice, it is not realistic for surgeons to resect all the polyps smaller than 10mm. And gallbladder polyps smaller than 10mm were frequently encountered and the detection rate of gallbladder polyps smaller than 10mm was about 0.3% to 9.5%[1]. Results of our study demonstrated that criteria set by the joint guideline were efficient in selecting gallbladder lesions with the possibility of being malignant. However, according to Tsuchiya et al, about 4.6% of gallbladder polyps between 1 and 5mm were malignant[16]. These small gallbladder lesions (with diameters less than 10mm) were reported not to invade the subserosa[16]. Therefore, in our opinion, it is reasonable to recommend 10mm as the threshold and gallbladder polyps between 6 and 9mm with other risk factors such as sessile morphology, blood flow signal on ultrasonography and single polyps should be resected as soon as possible. However, with the increase of polyp size, the likelihood of malignancy remarkably increased. Despite the fact that we could encounter benign lesions larger than 2cm, patients with polyps larger than 2cm should undergo laparoscopic cholecystectomy since the remaining benign lesions are adenomas which still have the potential of becoming malignant. As a matter of fact, in this study, 8 polyps larger than 30mm were identified as adenomas. Therefore, combining the recommendations from the joint guideline and results of our study, polyps larger than 10mm should be resected and polyps smaller than 10mm but with other risk factor such as sessile morphology, blood flow signal on ultrasonography and single polyps should also be surgically managed. 

In a study by Dilek ON et al, it was reported that cholesterol polyps were usually smaller than 10mm and sizes of neoplastic polyps (mainly adenomas) were usually between 2 and 20mm [4]. However, in this study, cholesterol polyps could be detected in all size ranges even among polyps larger than 20mm. As a matter of fact, the largest cholesterol polyp had a diameter of 25 mm. This may be explained by the high prevalence of cholesterol polyps (63.5% in this study and 60% to 68% from another report [4]). 104 of the 977 benign polyps in this study were identified as adenomas. Neoplastic polyps such as adenomas have been reported to be single ones while non-neoplastic polyps tend to be multiple[4]. In fact, in this study, a certain number of multiple polyps were diagnosed with adenomas and interestingly, these multiple lesions usually consisted of a single adenoma and multiple hyperplastic polyps or cholesterol polyps, suggesting that single or multiple appearance was not an efficient factor distinguishing cholesterol polyps from adenomas. Thus, it is not so easy for us to accurately predict the pathological diagnoses of gallbladder polyps. 

However, some inherent shortcomings of this study exist. Firstly, this study was a retrospective one in nature and more prospective studies are needed to verify findings of this study. Secondly, a relative small number of patients were included in this study and studies including more patients were warranted. Thirdly, among polyps larger than 30mm, gallbladder cancer was not so common as in other studies. In a study by Fujiwara K et al, it was reported that among polyps larger than 25mm, incidence of gallbladder cancer was as high as 100% [14]. This disparity might be caused by different diagnostic criteria adopted and some adenomas in this study were actually early gallbladder cancer. However, our study could provide some useful guidance for future clinical practice given the fact that our study included 1012 patients with gallbladder polyps and conclusions of this study were consistent with the joint guideline [1]. 

Conclusions

In conclusion, the larger gallbladder polyps are, the higher incidence of malignancy is. Consistent with the joint guideline, a sessile morphology, blood flow signal on ultrasonography and single polyps were risk factors for gallbladder malignancy. It is difficult for us to accurately evaluate the pathological diagnoses of gallbladder polyps preoperatively since some benign polyps could be rather large and some malignant ones are very small. Gallbladder lesions with a diameter of 10mm or larger should be surgically removed as soon as possible. However, the final pathological diagnosis might be different from preoperative evaluation.

Abbreviations

ROC: receiver operating curve; ESGAR: European Society of Gastrointestinal and Abdominal Radiology; EAES: European Association for Endoscopic Surgery and other Interventional Techniques; EFISDS: International Society of Digestive Surgery-European Federation; ESGE: European Society of Gastrointestinal Endoscopy; AUC: area under curve; OR: odds ratio; CI: confidential interval

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee, Shenzhen People,s Hospital (NO. LL-KY-2021864) and each patient in this study had given his or her informed consent in the written form. 

Consent for publication

Not applicable. 

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare that they have no competing interests.

Funding

This study was supported by Science and Technology Innovation Foundation of Shenzhen (No.JCYJ20180228164603659, No.JCYJ20180507182437217).

Authors' contributions

Shuwang Liu, Yan Wang, Tailai An and Jiling Jiang designed this study. Shuwang Liu, Lifen Zhang, Jing Xu, and Tianchong Wu collected the data. Tailai An and Jiling Jiang supervised this study. Shuwang Liu, Tailai An and Yan Wang performed statistical analysis. Jing Xu, Xiaofang Lu and Lingna Deng assessed the pathological diagnosis. Shuwang Liu and Yan Wang wrote the manuscript. Tailai An revised the manuscript. Jiling Jiang and Tailai An submitted this study. All the authors read and approved the final manuscript.

Acknowledgements

The authors would thank Lyujia Cheng from Jinan University (Guangzhou) for his help with proofreading the manuscript.

References

  1. Foley KG, Lahaye MJ, Thoeni RF, Soltes M, Dewhurst C, Barbu ST, et al. Management and follow-up of gallbladder polyps: updated joint guidelines between the ESGAR, EAES, EFISDS and ESGE. Eur Radiol. 2022;32(5):3358–68.
  2. McCain RS, Diamond A, Jones C, Coleman HG. Current practices and future prospects for the management of gallbladder polyps: A topical review. World J Gastroenterol. 2018; 24(26):2844–52.
  3. Donald G, Sunjaya D, Donahue T, Hines OJ. Polyp on ultrasound: now what? The association between gallbladder polyps and cancer. Am Surg. 2013; 79(10):1005–8.
  4. Dilek ON, Karasu S, Dilek FH. Diagnosis and Treatment of Gallbladder Polyps: Current Perspectives. Euroasian J Hepatogastroenterol. 2019; 9(1):40–8.
  5. Owen CC, Bilhartz LE. Gallbladder polyps, cholesterolosis, adenomyomatosis, and acute acalculous cholecystitis. Semin Gastrointest Dis. 2003; 14(4):178–88.
  6. Wiles R, Thoeni RF, Barbu ST, Vashist YK, Rafaelsen SR, Dewhurst C, et al. Management and follow-up of gallbladder polyps: Joint guidelines between the European Society of Gastrointestinal and Abdominal Radiology (ESGAR), European Association for Endoscopic Surgery and other Interventional Techniques (EAES), International Society of Digestive Surgery - European Federation (EFISDS) and European Society of Gastrointestinal Endoscopy (ESGE). Eur Radiol. 2017; 27(9):3856–66.
  7. Sugiyama M, Xie XY, Atomi Y, Saito M. Differential diagnosis of small polypoid lesions of the gallbladder: the value of endoscopic ultrasonography. Ann Surg. 1999; 229(4):498–504.
  8. Sadamoto Y, Oda S, Tanaka M, Harada N, Kubo H, Eguchi T, et al. A useful approach to the differential diagnosis of small polypoid lesions of the gallbladder, utilizing an endoscopic ultrasound scoring system. Endoscopy. 2002; 34(12):959–65.
  9. Kalbi DP, Bapatla A, Chaudhary AJ, Bashar S, Iqbal S. Surveillance of Gallbladder Polyps: A Literature Review. Cureus. 2021; 13(7):e16113.
  10. Spaziani E, Di Cristofano C, Di Filippo AR, Caruso G, Orelli S, Spaziani M, et al. Polypoid lesions of the gallbladder in a consecutive series of 2631 patients. A single-center experience. Ann Ital Chir. 2019; 90:305–10.
  11. Csendes A, Burgos AM, Csendes P, Smok G, Rojas J. Late follow-up of polypoid lesions of the gallbladder smaller than 10 mm. Ann Surg. 2001; 234(5):657–60.
  12. Babu BI, Dennison AR, Garcea G. Management and diagnosis of gallbladder polyps: a systematic review. Langenbecks Arch Surg. 2015; 400(4):455–62.
  13. Chavan S, Rathi P. Gallbladder Polyp: Review and Proposed Algorithm for Management. J Assoc Physicians India. 2022;70(1):11–2.
  14. Fujiwara K, Abe A, Masatsugu T, Hirano T, Sada M. Effect of gallbladder polyp size on the prediction and detection of gallbladder cancer. Surg Endosc. 2021; 35(9):5179–85.
  15. World Medical A: World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA. 2013; 310(20):2191–4.
  16. [Collective review of 503 cases of small polypoid lesions (less than 20mm in maximum diameter) of the gallbladder: size distribution in various diseases and the depth of carcinomatous invasion]. Nihon Shokakibyo Gakkai Zasshi. 1986; 83(9):2086–7.

Tables

Table 1. Number of benign lesions and gallbladder cancer within different size ranges and the corresponding cancer detection rates

Size

Total

Benign

lesions 

Gallbladder

cancer

Cancer detection 

rate (%)

≤5mm

661

660

1

0.2

6-9mm

151

148

3

2.0

10-14mm

111

108

3

2.7

15-19mm

40

36

4

10.0

20-24mm

16

12

4

25.0

25-29mm

8

4

4

50.0

≥30mm

25

9

16

64.0

Total

1012

977

35

3.5

Table 2. Different cut-off values and corresponding sensitivities, specificities, positive predictive values and negative predictive values

Cut-off size of the 

lesion (mm)

Sensitivity 

Specificity

Positive predictive 

value 

Negative predictive 

value 

3.5

1.000

0.424

0.059

1.000

6.5

0.971

0.739

0.118

0.999

10.5

0.857

0.901

0.236

0.994

13.5

0.829

0.937

0.319

0.993

15.5

0.686

0.959

0.375

0.988

20.5

0.657

0.982

0.561

0.988

27.5

0.457

0.991

0.640

0.981

Table 3. Numbers of different kinds of benign lesions in various size ranges

Size

Total(%)

Cholesterol polyp (%)

Adenomyomatosis

(%)

Inflammatory 

polyp (%)

Hyperplastic 

polyp (%)

Adenoma

(%)

≤5mm

660(67.6)

487(75.7)

106(66.6)

46(70.8)

4(66.6)

17(16.3)

6-9mm

148(15.1)

88(13.7)

33(20.8)

10(15.4)

1(16.7)

16(15.4)

10-14mm

108(11.0)

49(7.6)

14(8.8)

9(13.8)

1(16.7)

35(33.7)

15-19mm

36(3.7)

17(2.6)

3(1.9)

0(0)

0(0)

16(15.4)

20-24mm

12(1.2)

1(0.2)

2(1.3)

0(0)

0(0)

9(8.6)

25-29mm

4(0.4)

1(0.2)

0(0)

0(0)

0(0)

3(2.9)

≥30mm

9(0.9)

0(0)

1(0.6)

0(0)

0(0)

8(7.7)

Total

977(100)

643(65.8)

159(16.3)

65(6.7)

6(0.6)

104(10.6)

Table 4. Comparisons between benign lesions and gallbladder cancer regarding different clinical characteristics

Characteristics

Total(%)

Benign lesions(%) 

Gallbladder cancer(%)

χ2/t

P value

Gender

 

 

 

2.11

0.147

     Male

439(43.4)

428(43.8)

11(31.4)

 

 

     Female

573(56.6)

549(56.2)

24(68.6)

 

 

Age

46.24±12.96

45.68±12.62

61.91±12.53

-7.48

<0.001

<50

628(62.1)

622(63.7)

6(17.1)

31.057

<0.001

>50

384(37.9)

355(36.3)

29(82.9)

 

 

BMI

23.81±3.41

23.82±3.41

23.40±3.35

0.73

0.469

Size

 

 

 

147.72

<0.001

≤5mm

661(65.3)

660(67.6)

1(2.9)

 

 

6-9mm

151(14.9)

148(15.1)

3(8.6)

 

 

10-14mm

111(11.0)

108(11.1)

3(8.6)

 

 

15-19mm

40(3.9)

36(3.7)

4(11.4)

 

 

20-24mm

16(1.6)

12(1.2)

4(11.4)

 

 

25-29mm

8(0.8)

4(0.4)

4(11.4)

 

 

≥30mm

25(2.5)

9(0.9)

16(45.7)

 

 

Sessile

 

 

 

394.83

<0.001

Yes

69(6.8)

37(3.8)

32(91.4)

 

 

     No

943(93.2)

940(96.2)

3(8.6)

 

 

Flow signal

 

 

 

399.72

<0.001

     Yes

64(6.3)

33(3.4)

31(88.6)

 

 

     No

948(93.7)

944(96.6)

4(11.4)

 

 

Single or Multiple

 

 

 

41.61

<0.001

Single

420(41.5)

387(39.6)

33(94.3)

 

 

Mutiple

592(58.5)

590(60.4)

2(5.7)

 

 

Concurrent stone

 

 

 

0.581

0.446

Yes

612(60.5)

593(60.7)

19(54.3)

 

 

No

400(39.5)

384(39.3)

16(45.7)

 

 

Table 5. Binary logistical regression analysis identifying risk factors for gallbladder cancer

 

B value

Wald χ2

OR value(95%CI)

P value

Age

0.089

7.581

1.094(1.026~1.165)

0.006

Size

0.521

6.585

1.683(1.131~2.505)

0.01

Sessile

2.787

10.661

16.239(3.047~86.542)

0.001

Flow signal

3.882

15.733

48.505(7.125~330.198)

<0.001

Single or Mutiple

-2.035

3.799

0.131(0.017~1.017)

0.052