SU Ya-jing, WU Huan-liang, CHEN Yun-jing, HE Gui-sheng, WU Huang-fu?

1. Hainan Medical University, Haikou 570100, China

2. Hainan Cancer Hospital, Haikou 570100, China

3. The Second Affiliated Hospital of Hainan Medical University, Haikou 570100, China

Keywords:Intraductal papilloma of breast Vacuum-assisted excision Meta analysis Effect Prognosis

ABSTRACT Objective: To evaluate the safety and effectiveness of vacuum-assisted excision for benign intraductal papilloma of breast. Methods: The databases including PubMed, Embase, Cochrane Library, WangFang, CNIK, VIP and Web of Science were searched by computer, according the inclusion and exclusion criteria after screening of literature. The Cochrane handbook and Newcastle-Ottawa scale was used to evaluate the studies. Then the Review Manager 5.3 software was to used analyze the data.Results: A total of 1 016 patients was included in 9 articles. The amount of blood loss in vacuum-assisted surgery was less than that in open surgery [MD=-6.38,95%CI(-9.90,-2.86),P＜0.05]. The drainage[MD=-2.56,95%CI(-4.97,-0.15 ,P＜0.05] and drainage time[MD=-0.25,95%CI(-0.40,-0.09),P＜0.05] in vacuum-assisted surgery was less than that in open surgery. There were also few postoperative complications than that in open surgery [RR=0.43,95%CI(0.29,0.64),P＜0.05]. Compared with open surgery, the postoperative recurrence rate was lower in vacuum-assisted surgery [RR=0.26,95%CI (0.14, 0.49), P＜0.05]. Last,there was no statistically significant difference in the operative time between vacuum-assisted surgery and open surgery,[SD=-12.82,95%CI (-25.70, 0.06),P=0.05].Conclusion: Compared with open surgery, vacuum-assisted excision has the advantages of less blood loss and lower postoperative complications, but the operative time is not statistically significant compared with open surgery. In addition, compared with open surgery in this study, vacuum-assisted excision has the advantages of lower recurrence rate, but it still needs long-term dynamic observation.

1. Introduction

Intraductal papilloma of breast is a benign tumor with cancerous potential originating in breast ducts, which is caused by abnormal proliferation of ductal epithelial cells. The common onset age is between 33 and 55 years old, and the onset is related to estrogen.Risk factors include oral contraceptives, estrogen replacement therapy, lifetime exposure to estrogen, and a family history of breast cancer [1]. Intraductal papilloma of the breast is usually divided into central type and peripheral type. The lesion location is determined according to ultrasound examination results. Lesions less than 2cm from the nipple are often defined as central type, and solitary lesions are more common. Any further location is defined as a peripheral papilloma, which is often multiple [2] [3]. Patients often seek medical treatment due to spontaneous hemorrhage or discharge from the nipple. A few people can touch the mass around the areola, but most of them are asymptomatic [4] and seek medical treatment based on physical examination findings. For the treatment of intraductal papilloma of the breast, surgical resection is the standard method, but surgical resection has the disadvantages of inaccurate target, large resection range, large injury, and many postoperative complications, while vacuum-assisted excision (VAE)is an emerging surgical method. Is a minimally invasive technique in the field of breast surgery and a reflected, VAE has the advantages of small trauma, fewer complications and faster recovery, in the treatment of benign breast intraductal papilloma has get more and more attention, but at the moment about VAE research literature for the treatment of breast intraductal papilloma were insufficient,lack of major cases reported and overall assessment of the longterm prognosis, As a supplement and integration of the previous literature, this study analyzed and evaluated the efficacy and prognosis of VAE in the treatment of benign intraductal papilloma of the breast, so as to provide more options for the clinical treatment of intraductal papilloma of the breast.

2. Data and Methods

2.1 Inclusion and exclusion criteria

2.1.1 Study types

Randomized controlled trials or retrospective controlled trials,whether or not blind. The language is limited to English and Chinese.

2.1.2 Study objectPatients with intraductal papilloma were enrolled. The diagnostic criteria were in accordance with WHO standards.

2.1.3 InterventionThe experimental group was treated with vacuum-assisted excision(VAE). In the control group, breast ductoscopy surgery, traditional open surgery, methylene blue or guidewire assisted open surgery were used.

2.1.4 Outcomes

Operation time (min), the amount of blood loss (ml), tissue resection volume (m2), postoperative drainage (ml) and drainage time (d), length of hospital stay (d); postoperative complications indicators: hematoma, ecchymosis, intraoperative effusion,intraoperative infection, breast deformity; postoperative recurrence rate.

2.1.5 Exclusion criteria(1) Intraductal papilloma of the breast with the opposite sex was confirmed preoperatively;(2) Studies in which data cannot be accurately extracted or data is missing;(3) Reviews, case reports, no control group, and republished studies were excluded.

2.2 Document retrieval

By searching CNKI, Wanfang, VIP, Web of Science, PubMed,Embase, Cochrane Library, etc, the retrieval date was from the database establishment to 2021.4.25. The English search terms are:Papilloma*, Papillary*, Intraductal*, Minimally Invasive Surgical Procedures, Minimal Surgical Procedure*, Minimally Invasive Surgical Procedures Surgery*, Minimal Access Surgical Procedure*,MIS, VACUUM*, Mammotome, Encon. A total of 2651 literatures were collected.

2.3 Literature screening and data extraction

The literature and data extraction is carried out by two independent researchers back to back. If two researchers have different opinions on the literature or the extracted data, they can consult a superior expert or the whole group to discuss and resolve. If the information is incomplete, you can contact the original author by email to add. The extracted data include: ①Basic information of the included research:including the author's name, year of publication, and country;②Basic characteristics of the research object: including sample size, age, and symptoms; ③The specific situation of intervention measures; ④Key elements of bias risk evaluation; ⑤The main data of the outcome indicators.

2.4 Literature quality evaluation

The risk of bias of the included studies was evaluated using the risk of bias evaluation tool for randomized controlled trials recommended by Cochrane Manual Version 5.1.0 and the Newcastle-Ottawa scale for case-control studies to evaluate the quality of the literature.

2.5 Statistical method

RevMan 5.3 software was used for statistical analysis. Continuous variables, using mean difference (SD) as the effect indicator,binary variables using odds ratio (RR) as the effect indicator, each effect size is given a 95% confidence interval (CI), P＜0.05 means the difference is statistical Learn meaning. The heterogeneity of the included studies was analyzed by the χ2test (the test level is α=0.1), and the I2 statistic was combined for evaluation. If the heterogeneity test result is I2＜50%, the fixed effects model is used for meta-analysis; if I2＞50%, the source of heterogeneity is further analyzed, and the random effects model is used for meta-analysis.

3. Result

3.1 Literature search and screening

The literature screening of this study includes two parts:preliminary screening and re-screening. 1) Preliminary screening:Import the documents retrieved from the database into the Endnote document management software, use the software's duplicate check function to check the duplicates, and exclude 172 duplicate articles;then read the titles and abstracts to exclude 2448 articles, and finally 34 articles are performed by reading the full text Re-screening.2) According to PICOS and inclusion and exclusion criteria, 3 conference abstracts, 8 uncontrolled trials, and 3 incomplete data were excluded, 9 articles were confirmed to be included .See Figure 1 for details.

Figure 1 Flow chart of literature search and screening

3.2 Basic characteristics of the included literature

This study included a total of 9 studies [6,7,8,9,10,11,12,13,14], 8[6,7,8,9,10,11,13,14] from China, 1 Article [12] from South Korea,published from 2011 to 2020. A total of 1016 patients, including 544 in the experimental group and 472 in the control group. 7[6,7,8,9,10,13,14] randomized controlled trials, 2 [11,12] retrospective analysis; 9 [6,7,8,9,10,11,12,13 ,14] The experimental group of the study all underwent vacuum minimally invasive atherectomy under the guidance of B-ultrasound. Two articles [6,11,14] used ductoscopy for positioning before surgery, and ductoscopy were used for positioning. Gel, Nacl solution and air as media enhance the visualization of the tumor under ultrasound. See Table 1 for details.Note: RCT: randomized controlled trials; VAE: vacuum-assisted excision; OB: open surgery. Clinical manifestations: ①nipple bloody discharge, ②nipple discharge, ③single mass, ④multiple mass, ⑤cystic solid mass. Outcome indicators: ① operation time; ② intraoperative blood loss; ③ postoperative drainage,④ drainage time, ⑤ hospitalization time, ⑥ tissue volume removed, ⑦ hematoma, ecchymosis, ⑨ incision infection, ⑩ recurrence rate, ? breast deformity, ? treatment effect, ? recovery time, ?, upgrade rate.

Table 1 The basic characteristics of the study

3.3 Literature quality evaluation

The included 9 [6,7,8,9,10,11,12,13,14] literatures used Cochrane Handbook Version 5.1.0 risk bias criteria for randomized controlled trials from random methods, allocation concealment, and patients Or the interventionists were evaluated in 7 aspects, including blinding,loss to follow-up, selective reporting, and other biases, and the comprehensive evaluation level was B level. The retrospective study was evaluated using the Newcastle-Ottawa scale (NOS scale) [15],and the comprehensive evaluation level was A level.

3.3.1 The quality evaluation of 7 [6,7,8,9,10,13,14] RCT documents is shown in Figure 2.

Figure 2 The basic characteristics of the RCTS

3.3.2 Two of the [11, 12] retrospective research quality evaluation tables are shown in Tab 2

3.4 Comparison of operation time

There was 6[7,8,9,10,13,14] studies compared the operation time between the VAE and OB among the nine included[6,7,8,9,10,11,12,13,14] studies. The test results of heterogeneity between the two groups showed significant statistical heterogeneity(I2=100%). Random effects model was used for analysis. The analysis showed that there was no statistical difference in the comparison of operation time. [SD=-12.82, 95%CI (-25.70, 0.06),P=0.05]. See Figure 3.

3.5 Intraoperative blood loss

Five of the included studies [7,8,9,13,14] compared the bleeding volume of the two groups of operations. The heterogeneity test results showed significant statistical heterogeneity (I2=96%). The random effects model was used for analysis. The analysis showed that the two groups had statistical significance in the comparison of intraoperative blood loss. [MD=-6.38,95%CI(-9.90,-2.86), P＜0.05].Further analysis of the heterogeneity of the intraoperative blood loss of the two groups, such as the proficiency of the surgical operator,the location of the tumor, the amount of surgical tissue removed,and the patient's coagulation function, etc., will all affect the intraoperative blood loss, but among them, four [7,8,9,13] studies did not give a specific description of the above situation. See Figure4.

Table 2 The basic characteristics of the retrospective studies

Figure 3 Forest plot of operation time in vacuum-assisted excisim versus open surgery

3.6 Postoperative drainage

Three of the included studies [7,8,13] compared the postoperative drainage of the two groups. The heterogeneity test results showed significant statistical heterogeneity (I2=88%). The random effects model was used for analysis. The analysis showed that the two groups had statistical significance in the comparison of postoperative drainage. [MD=-2.56, 95%CI (-4.97, -0.15), P＜0.05]. See Figure 5.

3.7 Postoperative drainage time

Three of the included studies [7, 8, 13] compared the postoperative drainage time of the VAE and OB. There was no statistical heterogeneity between the two groups (P=0.28, I2=22%). The fixedeffect model was used for analysis. The analysis showed that the two groups had statistical significance in the comparison of postoperative breast drainage. [MD=-0.25, 95%CI (-0.40, -0.09), P＜0.05]. See Fig ure6.

3.8 Surgical complications

Five of the included studies [6,7,8,9,13,14] compared the complications of the VAE and OB. There was no statistical heterogeneity between the two groups (P=0.39, I2=30%). The fixed effects model was used for analysis. The two groups had statistical significance in the occurrence of complications. [RR=0.43,95%CI(0.29,0.64), P＜0.05].See Figure 7.

3.9 Recurrence rate

Four of the included studies [8,9,12,13] compared the postoperative recurrence rate of the experimental group and the observation group during the postoperative follow-up. The heterogeneity test results between the two groups showed no statistical heterogeneity (P=0.13,I2=46%), using a fixed-effect model for analysis, the two groups had statistical significance in the comparison of recurrence rates.[RR=0.26,95%CI(0.14,0.49), P＜0.05]. See Figure 8.

3.10 Evaluation of publication bias

Due to the inclusion of the document is less than 10, so there is no publication bias were evaluated.

Figure 4 Forest plot of bleeding volume in vacuum-assisted excisim versus open surgeryB

Figure 5 Forest plot of postoperative drainage in vacuum-assisted excisim versus open surgery

Figure 6 Forest plot of the time of postoperative drainage in vacuum-assisted excisim versus open surgery

Figure 7 Forest plot of postoperative complications in vacuum-assisted excisim versus open surgery

Figure 8 Forest plot of recurrence rates in vacuum-assisted excisim versus open surgery

4. Discussion

Intraductal papilloma of the breast accounts for about 5.3% of all benign breast lesions [16], and about 3% of solid breast lesions [17],but there is a 6.3% risk of malignant transformation [18]. With people’s health awareness and self-examination. The detection rate of breast intraductal papilloma is increasing year by year.Middle-aged women are the high-risk group of this disease. Early detection, early diagnosis and precise treatment are very important.Vacuum-assisted excision is an application of minimally invasive technology in the field of breast surgery, adding a new way to the surgical selection of breast diseases; because breast intraductal papilloma have small lesions, negative palpation, and sometimes only show nipple discharge, traditional open surgery usually uses syringe needles, epidural anesthesia tubes, guide wires, etc. to insert the breast duct retrogradely from the nipple along the discharge catheter port, or retrograde injection of methylene blue; then choose areola incision or radial incision for breast. Segmental or quadrant resection; followed by scars and postoperative breast deformities that accompany lifelong. Obviously, traditional surgery has the shortcomings of inaccurate target, large resection area, large damage,and many postoperative complications, which have caused a great degree of damage to women's physiology and psychology. VAE can avoid these problems well, and has the unparalleled advantages of traditional surgery.First of all, VAE has the advantages of accurate positioning, less damage, and low postoperative complications. The application of VAE in breast surgery is realized under the guidance of B-ultrasound,mammography X-ray, and MRI. Compared with the other two,B-ultrasound has simple operation, convenient and fast, accurate positioning, no radiation and real-time guidance. It is the best and preferred guidance technique for vacuum-assisted minimally invasive atherectomy [19, 20]. According to the clinical characteristics of breast intraductal papilloma, Sun [21] believes that in the diagnosis of breast intraductal papilloma, high-frequency ultrasound due to its high resolution is better than ductal X-ray angiography, ductoscopy, etc.Able to make a preparatory judgment on the relationship between dilated breast ducts and masses. The 9 articles [6,7,8,9,10,11,12,13,14]included in this study all chose B-ultrasound as the guidance technology, Xia [6], Wu [11] and others also used breast ductoscopy for positioning before surgery, which further achieved precise positioning of the lesion. Wei [19] believes that real-time and accurate ultrasound guidance plays a key role in the success of surgery. The combined guidance of longitudinal and transverse sections can more accurately locate the lesion, reduce the frequency of atherectomy,reduce the cutting of normal breast tissue, and shorten the operation time, reduce the occurrence of complications after surgery. This study also further verified that compared with open surgery, VAE has the characteristics of less postoperative complications. Unfortunately,only Xia [6] and Wu [11] studied the amount of tissue resected in the two groups. In comparison, the amount of data is small, and it cannot be further confirmed that VAE cuts normal breast tissue less than OB in the treatment of breast ductal papilloma; theoretically,accurate positioning and fewer resections will shorten the operation time, but the comparison of time between the two groups is not statistically significant. Considering that the included studies have large clinical heterogeneity, this may be related to the surgical skills of the surgeon, the cooperation with the sonographer, the size and size of the tumor, and the postoperative suture technique.Secondly, a large number of studies have proved that VAE surgery is safe to remove benign breast tumors. Although VAE is used to obtain pathological tissue through multiple rotary-cuts, it destroys the integrity of the tumor and makes it difficult for pathologists to determine the tumor boundary and resection margins [22]. There is a risk of residual lesions or insufficient resection of the lesion, so it has a risk of tumor recurrence or pathological underestimation .However, a meta-analysis by Ding[23] et al. compared the residual rate of tumors after open surgery and Mammotome minimally invasive surgical resection of benign breast tumors, and the study indicated that there was no statistical difference between the two. In terms of pathological underestimation rate, a prospective study by Chang[24] et al. showed that 60 patients with breast papillary lesions diagnosed by VAE (11G) were subsequently surgically removed,and the malignant escalation rate was 0%, so they believed that the vacuum Minimally invasive atherectomy is safe for breast intraductal papilloma without heterogeneity. But they also pointed out that the 6.1% (3 of 49; 95% CI 1.3-16.9) atypical papilloma escalation rate is worthy of attention. Choi [25] et al. conducted a retrospective study on 500 benign breast intraductal papilloma with no atypical lesions diagnosed preoperatively. They were divided into surgery group (206), VAE group (233) and In the non-intervention followup group (61), the malignant pathological escalation rates of the surgical group and the VAE group were 1.9% and 2.1%, respectively.The escalation rates of malignant tumors in the two groups were similar, and there was no statistical difference. Dayoung Ko [12]also conducted statistics on the malignant pathological escalation rate of 405 patients with benign breast intraductal papilloma who were clearly diagnosed by thick needle puncture before surgery.The rate of malignant pathological escalation was 5.2% and 0.5%in the surgery group and VAE group, respectively. The upgrade rate of the group was significantly higher than that of the VAE group,but Chang [24], Dayoung Ko [12] and others subsequently conducted further studies to show that the size of the mass is a predictor of pathological upgrade. VAE is feasible for small, non-specific breast intraductal papilloma, but surgical resection is recommended for IDP with dysplasia.

Finally, through long-term follow-up to observe the postoperative recurrence, Dayoung Ko [12] followed up 293 patients after resection for 11 to 110 months (median follow-up time was 20.2 months),and found that the postoperative recurrence rates were 19% and 2% among the OB and VAE, and only 14% recurred in situ, 86%recurred in a new site, and all recurred masses were excised and confirmed to be benign tumors. Choi et al. [25] also followed up 116 IDP patients after VAE for 24 to 115 months (median follow-up time was 45.1 months), and the postoperative recurrence rate was 3.6% (6/166); this study also confirmed that the recurrence rate of the VAE group was lower than that of the traditional surgery group.Huang Wei et al. [26] conducted close follow-up of 104 IDP patients who had undergone resection of MMD for 45 months. The results showed that the number of lesions was the main risk factors for recurrence after Mammotome (OR=1.780, P＜0.05) ,it not related to the patient’s age, tumor size, location, and BI-RADS classification.Therefore, it is feasible to choose VAE to treat breast intraductal papilloma, but For multiple breast intraductal papilloma, there is a certain recurrence.

In addition, vacuum-assisted minimally invasive atherectomy is also used in accessory mastectomy and treatment of gynecomastia;because of its advantages, VAE is playing an increasingly important role in the field of breast surgery.

At present, there is no meta-analysis of vacuum-assisted atherectomy for the treatment of breast intraductal papilloma in China. There is a meta-analysis [27] from abroad, which is about the pathology underestimated of intraductal papilloma by thick needle puncture. This study also has certain shortcomings; 1. The incidence of this disease is low, there are few studies in this area,the total number of samples included in the study is limited, and some indicators are included in the literature, especially the foreign literature included is also small, and Bias cannot be fully eliminated;2. The studies included in this study include randomized controlled trials and non-randomized controlled trials, and there is a risk of selection bias; 3. The quality of the included literature is not high,and only one literature specifies the central random method The principle of distribution and hiding; the effective information that can be provided is not comprehensive enough, and the conclusions are easily affected by the low level of research evidence; 4. Finally,due to the limited amount of literature, it is impossible to analyze all the outcome indicators after surgery, especially postoperative pathological escalation rate. Therefore, we should be cautious about the conclusion of this meta-analysis. At present, VAE has long been used in the treatment of benign breast tumors. In the selective treatment of breast intraductal papilloma, the safety of VAE has been accepted by most scholars, but there is still a lack of large case reports and sufficient clinical randomized controls. The test is corroborated, so the safety and long-term prognosis of the operation still need to be dynamically observed.

Author's contribution degree description:

Su Ya-jing, the first author, design executor, participated in literature screening, data analysis and writing of the first draft of the paper. Wu Huang-fu, corresponding author, design plan and article content review. Wu Huan-liang, participated in literature search and screening. Chen Yun-jing, participated in literature search and screening, and also helped with data extraction and statistical software. He Gu-sheng, participated in the correction of the content of the first draft.