Introduction

Gynecomastia is a common problem in male, characterized by enlargement of the breast, and can present bilaterally, and rarely unilateral [1,2,3]. The enlargement of breast tissue is caused by an increase in amount of ductal tissue, fat, and/or both [2, 4]. Diagnosis can be determined by patient’s history and physical examination. There are three phases in life where gynecomastia can occur. The first is right after birth, due to stimulation of estradiol and progesterone, but usually regresses after several weeks. The second phase occurs on pubertal stage, usually detected at peak age of 14. Pubertal gynecomastia usually regresses spontaneously in 3 years within onset. It is caused by an imbalance in androgen to estrogen ratio. Elderly stage is the third phase in which gynecomastia can occur. It is hypothesized to be caused by increased amount of fat that will produce excessive estrogen and a decrease in testosterone [5, 6].

Gynecomastia causes social anxiety and sometimes pain or tenderness [5]. When gynecomastia does not resolve spontaneously, it can be treated with either medical therapy or in adjunct with surgery [7]. Surgery is only indicated as last resort for patients with persistent gynecomastia and suspected malignancy. Generally, the aim is to reduce the volume of the breast with minimal scar possible [5, 6].

Many techniques are available and can be used according to the grade of gynecomastia. Minimally invasive surgery such as endoscopic approach, or liposuction assisted minimal incision surgery are the choices as it has been associated with less complications, less recovery time, and better aesthetic outcome [6, 8]. Complications of gynecomastia surgery can be divided into early and late complications. Some examples of early complications include hematoma, seroma, infection, and bleeding. Late complications include residual breast tissue, hypertrophic scar/keloid, numbness, asymmetry, nipple necrosis, and contour irregularity [5]. This study aims to analyse the quality of the studies on liposuction-assisted minimal incisional surgery and revisit the techniques by summarizing the satisfaction and complication rate for the treatment of gynecomastia.

Methods

Search Strategies and Screening Procedure

A review using the PRISMA guideline [10] was conducted to studies in online databases of PubMed, ScienceDirect, Cochrane Library, and Scopus using keywords: “gynecomastia” AND “liposuction.” All searches were limited to time range of January 2011-November 2020, published in English. Access to all of the database was conducted on 28th of October 2020. Inclusion criteria includes studies on: (1) male patients with gynecomastia that underwent minimal incision liposuction assisted surgery with or without pharmacological intervention; (2) specified Simon’s grade of gynecomastia grade I and II; (3) with minimum follow up of 6 months; and (4) there is either complication, satisfaction rate or both in the result. Minimal periareolar incision is defined as periareolar incision that is made across ½ to < ¾ periareolar line. Exclusion criteria were studies on transgender, surgery with skin excision, and patient treated with endoscopic surgery. Letter to Editor, Discussion, Comments, and Expert Opinion were excluded in the screening process. The data were then screened by two authors (AGB and IA) independently.

Data Extraction and Quality Assessment

Data were extracted by two authors independently (AGB and IA) and then combined into a single Microsoft Excel spreadsheet. Discrepancies were justified by further discussion by all authors to get conclusion agreement. Corresponding authors of each journal were contacted via email when additional data are needed. A timeframe of 1 month was given to get response from each author.

Average age particularly was obtained from multiplying the average age of each study times number of subjects per study; all studies added, then divided by total number of subjects. Due to the inconsistency between mean as median vice versa, the obtained age data that comes as mean and median in all included studies are treated as average.

The patient satisfaction score was obtained from generalizing different scores such as Visual Analogue Score (VAS), Likert score, and general satisfaction score; into gross standardized groups defined by the authors. Knowing VAS and general satisfaction score are defined as interval data, while Likert score is defined as ordinal data; the authors acknowledged that this method is not ideal. Nevertheless, some studies and reports do use this method and treat Likert scale as interval data [9].

Quality assessment of the studies was performed using Methodological Index for Non-Randomized Studies (MINORS) [11] by two authors independently, then combined. The MINORS assessment shows ideal if the score is 16 for non-comparative studies and 24 for comparative studies [11].

Results

Records identified through PubMed, ScienceDirect, Cochrane Library, and Scopus showed 98, 67, 2, and 248 relevant studies, respectively. Ninety-five studies were duplicates, and 227 studies were excluded after abstract and title screening. Full text of 9 studies were not available. Eighty-four articles were assessed for inclusion and exclusion criteria. Finally, there were 18 studies eligible for review.

The patients’ average age is 23.13 years, calculating from 244 available patients. The remaining 1384 patients were not included in the calculation due to incomplete data. The total patients treated by liposuction-assisted surgery that have gynecomastia grade I are 618, and 1036 for grade II, excluding 37 patients from the study by Cigna et al. [12] due to the incomplete detail of data. (Table 1)

Table 1. Characteristics of the studies
Full size table

Table 2 shows the detail of quality assessment using MINORS. The mean score of the non-comparative studies concerning satisfaction rate of improved quality of life after treated by liposuction with or without small incision surgery is 13.69. Two non-comparative studies are deemed ideal. Mean of comparative study included in this study is 22±1.73.

Table 2. MINORS quality assessment of the studies
Full size table

We categorized the satisfaction rate of the quality of life into three broad range as a gross descriptive index, converting the results of each study into ‘Highly satisfied’, ‘Satisfied’, and ‘Unsatisfied’. We converted the 5-point Likert scale, the 10-point scoring system, and the 4-point categorization used by each study to each category as described in Table 3. There is no universally accepted cut-off point for these scores; thus, the cut-off is determined arbitrarily, modified from studies that used similar method [28]. The interval of each score was determined by the formula for tertile; the highest possible average minus the lowest possible average, then divided by three. VAS is an interval data; thus, cut-off can be done with the method described in Table 3.

Table 3. Gross descriptive categories to interpret mean satisfaction score
Full size table

As already mentioned in the methods, the limitations are to be considered when interpreting the data. A study by Voutilainen et al. shows that the usage of VAS results in lower satisfaction rate compared to Likert scale [29]. As a result, the findings are fair for VAS studies. All studies included reported a high satisfaction rate, having only two studies reporting ‘satisfied’ patients: using power-assisted/ultrasound [14] and suction-assisted liposuction [17] methods; the former was caused by residual breast tissue, and latter not explained The overall reoperation percentage ranges from 0.6 to 25%. Table 4 shows the detail of each study’s complication rate, satisfaction score, and the reoperation rate.

Table 4. Extracted data from reviewed studies
Full size table

The studies reveal 4 different techniques of liposuction. If no technique was specified, we consider it as vacuum-assisted liposuction. The complication and the rate of each technique are described in Table 5. The most common complications were bleeding and scar which occurred to 31 patients each, and revisions were done to 35 patients out of 1628 patients analyzed in this paper. Power-assisted liposuction has the highest complication rate as high as 9.07%. [12,13,14, 16, 20, 23, 24] With the use of tumescent technique, excessive intraoperative-bleeding was reduced throughout the years. The reduced amount of bleeding referred was accounted as the amount of blood in the lipoaspirate, while also comparing preoperative and postoperative hemoglobin. [30, 31] The estimated blood loss on tumescent technique liposuction is 1% of volume aspirated, while 20-45% of blood in the lipoaspirate were estimated in liposuction without any fluid infiltration [32] (Fig. 1).

Table 5. Liposuction techniques with its complications
Full size table

Discussion

Biased assessment of the study endpoint is the most common cause of MINOR low score. However, when assessed further, the cause of getting 0 for the point is because the point was inapplicable for the study. All the studies’ endpoint were the patient’s own judgement; thus, bias is unavoidable. Since there is a high risk of bias, the determination of which surgical technique is better cannot be obtained. The lack of statistical result is also one of the most common contributors of low score. A study by Fagerlund et al. [1] shows similar result. The low score of methodological review of each study does not allow us to draw any conclusion, as any conclusion will be biased.

The lower satisfaction rate on power-assisted liposuction may be due to high percentage of residual breast disc (19.2%) and resulting in high rate of reoperation. Under-resection is a common complication in liposuction only cases [33]. Cause of reoperation found in this study is mostly due to residual breast tissues. Other causes are hematoma and scar. This wide range of reoperation percentage can be caused by: (1) the discrepancy of number of subjects; (2) the surgeon’s experience; and (3) the surgical technique that is used, such as liposuction only technique, compared to combined technique.

The high rate of bleeding and scar complication shown in this review was mostly contributed by Lee et al. [16]. But the conclusion of bleeding as the highest complication rate could not be drawn, because the total patient in that study were 1011, which means only less than 1% have bleeding and/or scar as its complication.

Our study has several limitations. First, there were limitations on detail of each patient while extracting the data, such as no explanation about complications or satisfaction level. This results in small number of subjects that can be analyzed. Secondly, the lack of subjects concerning laser-assisted and suction-assisted liposuction studies also contributes to higher risk of bias. Thirdly, we found that the studies available for gynecomastia has low quality of methodological technique. This results in the inability to draw firm conclusions without high risk of bias.

Conclusions

Small incisional design for breast parenchymal removal in gynecomastia assisted by liposuction showed a good technical approach for consistent improvement in quality of life. The studies included in this review mostly does not comply as an ideal non-comparative study, mostly due to high risk of bias. This review can be used as a reference for further studies to have a better methodological quality, less bias, and have a standardized method of measurement in regards to satisfaction rate for gynecomastia patients.

Fig. 1
figure 1

Identification of new studies via database and registers

Full size image