Factors Affecting Postoperative Complications of Suction-Curettage by Arthroscopic Shaver for Bromhidrosis

Abstract

Background 

Suction-curettage by arthroscopic shaver is the most effective treatment for bromhidrosis; however, postoperative complications require wound management and exhibit a high risk of hypertrophic scarring. We investigated factors affecting postoperative complications.

Methods 

We retrospectively evaluated data for 215 patients (430 axillae) with bromhidrosis treated with suction-curettage by arthroscopic shaver between 2011 and 2019. Cases followed for less than 1 year were excluded. Complications of hematoma or seroma, epidermis decortication, skin necrosis, and infection were recorded. Multinomial logistic analysis was used to calculate odds ratios and corresponding 95% confidence intervals for the complication of the surgery, adjusting for relevant statistically significant variables.

Results 

Complications occurred in 52 axillae (12.1%). Epidermis decortication occurred in 24 axillae (5.6%), with a significant difference for age (P < 0.001). Hematoma occurred in 10 axillae (2.3%) with a significant difference in tumescent infiltration use (P = 0.039). Skin necrosis occurred in 16 axillae (3.7%) with a significant difference for age (P = 0.001). Infection occurred in 2 axillae (0.5%). Severe scarring occurred in 15 axillae (3.5%), with complications related to more severe skin scarring (P < 0.05).

Conclusion 

Older age was a risk factor for complications. Use of tumescent infiltration resulted in good postoperative pain control and less hematoma. Patients with complications presented with more severe skin scarring, but none experienced limited range of motion after massage.

Bromhidrosis is a clinical disorder characterized by excessive or abnormally foul-smelling axillary odor. The condition is further aggravated by poor hygiene or underlying disorders that promote bacterial overgrowth, including diabetes, intertrigo, erythrasma, and a high body mass index (≥24 kg/m²) or being overweight.^1,2

Several treatments are available for axillary bromhidrosis, including topical drying or antibacterial agents, injected botulinum toxin A,³ iontophoresis, surgical approaches, laser treatment,⁴ and noninvasive microwave technology (miraDry) treatment.⁵ However, the effects of noninvasive treatments are temporary, and invasive treatments are limited by their complications. In the last 10 years, suction-curettage offered a safe procedure with few adverse effects, low complication rate, the shortest possible time away from daily activities, and reduced postoperative pain.^6,7 High efficiency is associated with a high incidence of complications of subcutaneous hematoma, wound necrosis, and delay of wound healing, which are associated with a high risk of hypertrophic scar formation. No studies exist in the literature that discusses the factors affecting complications in suction-curettage treatment of bromhidrosis. Our study looked at factors related to complications and how to reduce complications.

MATERIALS AND METHODS

Patients

We retrospectively evaluated 215 patients and 430 axillae with bromhidrosis that were treated with suction-curettage by arthroscopic shaver between 2011 and 2019 at Ditmanson Medical Foundation Chia-Yi Christian Hospital, Taiwan. All patients were under local anesthesia, and all procedures were performed by 9 surgeons in our outpatient Medical Cosmetic Center. Cases followed for less than 1 year were excluded.

Surgical Procedures

For all patients, axillary hair was shaved before the operation. The patient was placed in supine position with the upper arms abducted to 100°. The operative fields were draped for asepsis, and bilateral axillary hair line and 1 cm beyond the outline were marked. Diluted local anesthesia was induced either with 40 mL of 1% lidocaine or with a tumescent solution. When a tumescent solution was used, 300 to 600 mL of solution (according to the size of axilla) was injected into both axillae with a 22-gauge spinal needle. One tiny stab incision of approximately 1 cm was made at the 12-o'clock position of the outer circle of marking in each axilla, and tunneling of subcutaneous tissue was performed. An arthroscopic shaver was inserted through the incision, and suction-curettage (at 1500 rpm at a pressure of 250 mm Hg) of apocrine glands in subdermal plane was performed for 7 to 15 minutes (Table 1).

TABLE 1 - Differences in Our Surgeons, Including Tumescent or Not, Shaving Time, and Dressing Methods

Surgeon	Tumescent	Suction-Curettage Time	Dressing
A	Y	10	Artiss + C1
A	Y	10	C3
B	N	8	Anchor + C3
B	Y	15	Drain + Anchor + C3
C	N	8	Drain + Anchor + C3
D	Y	10	Anchor + C3
E	N	10	Tie + Anchor + C3
F	N	8	Anchor + C3
G	N	7	Anchor + C3
H	N	10	Anchor + C3

Anchor, multiple anchor sutures; Artiss, Artiss using; C1, compressive bandage for 1 day; C3, compressive bandage for 3 days; Drain, 1 Penrose drain leaving; N, no; Tie, tie-over dressing; Y, yes.

The wound was closed after water irrigation and ensuring hemostasis. The dressing type varied by surgeons, including the use of fibrin sealant (ARTISS; Baxter, 2 mL), anchoring sutures (five 4-0 nylon stitches go through the full-thickness of the skin to nail the deep fascia), drain, and tie-over or compression dressing for 1 or 3 days (Table 1). The patient returned to our clinic for dressing changes on postoperative days 2 and 3, with removal of the stitches at 1 week. Complications were evaluated at 1, 3, 6, and 12 months, unless the patient experienced wound problems or poor results requiring secondary management. The patient was advised to avoid full abduction and elevation of the arms for 1 week and to massage the scars approximately 2 weeks after the operation.

Assessment

Complications of hematoma or seroma, epidermis decortication, skin necrosis, and infection were recorded. A hematoma was an accumulation of fluid under the skin that was drawn as blood, whereas a seroma was drawn as a serous discharge. Epidermal decoration referred to postoperative epidermis exfoliation. Skin necrosis referred to full-thickness skin necrosis that required further debridement. Infection was defined as local inflammation and positive cultures. Scar was graded by medical staff as “minimal,” “mild,” or “severe” based on total score on the Vancouver Scar Scale for vascularity, pigmentation, pliability, and height. “Minimal” indicated a total score of 0 to 1, “mild” indicated a total score of 2 to 4, and “severe” indicated a total score of 5 and greater. All complications and scars were evaluated and recorded by 2 attending physician.

Statistical Analysis

Surgeons' experience was a dependent variable for statistical analysis and a well-known factor associated with clinical outcomes. Statistical analyses were performed (SPSS for Windows version 21.0; IBM Corp, Armonk, NY). Continuous variables were expressed as mean ± SD, and categorical data were expressed as numbers and percentages. Comparisons of continuous data between groups were evaluated with Student t test or the Mann–Whitney U test, and comparisons of categorical data were evaluated with the χ² test or Fisher exact test, as appropriate. Multiple comparisons were evaluated by 1-way analysis of variance with Tukey post hoc testing. Multinomial logistic analysis was used to calculate odds ratios (ORs) and corresponding 95% confidence intervals (CIs) for the complication of the surgery, adjusting for relevant statistically significant variables. P < 0.05 was considered statistically significant.

RESULTS

Patient Characteristics and Efficacy

Table 2 summarizes the characteristics of the 215 patients (160 women and 55 men) who underwent bilateral suction-curettage for bromhidrosis. Mean age was 25.6 ± 22.0 years. Efficacy results were “excellent” for 382 axillae (88.84%), “good” for 36 axillae (8.37%), and “failed” for 12 axillae (2.79%). Of the 12 “failed” axillae, 11 underwent secondary suction-curettage and achieved excellent results after secondary suction-curettage.

TABLE 2 - Demography of Patients of Bromhidrosis

	Female	Male	Total
n (%)	160 (74.42)	55 (25.58)	215
Age, y	12–67 (26.28 ± 22.00)	13–63 (23.84 ± 21.00)	12–67 (25.65 ± 22.00)

Risk Factors for Epidermis Decortication

Epidermis decortication was observed in 24 axillae (5.58%), and a significant difference was found with respect to age (adjusted OR, 1.11; 95% CI, 1.07–1.16; P < 0.001). When compressive bandage for 3 days was used as the reference group, fibrin sealant plus compressive bandage for 1 day seemed to be associated with a high risk for epidermis decortication (adjusted OR, 66.22; 95% CI, 2.41–1821.97; P = 0.013). No significant differences were found in shaving time, gender, and use of tumescent infiltration between patients with epidermis decortication and patients without complications (Table 3).

TABLE 3 - Crude and Adjusted OR and 95% CI of Suction-Curettage–Related Epidermis Decortication Associated With Predictors by Logistical Regression Model

	Epidermis Decortication
	No (n = 406)	Yes (n = 24)	P
Age, y	24.97 ± 10.62	37.25 ± 14.00	<0.001
Shaving time, min	9.20 ± 2.20	9.42 ± 2.41	0.464
Gender			0.131
Female	299 (73.65%)	21 (87.50%)
Male	107 (26.35%)	3 (12.50%)
Tumescent			0.634
No	286 (70.44%)	18 (75.00%)
Yes	120 (29.56%)	6 (25.00%)
Dressing			0.258
C3	49 (12.07%)	1 (4.17%)
Artiss + C1	9 (2.22%)	1 (4.17%)
Anchor + C3	222 (54.68%)	14 (58.33%)
Tie + Anchor + C3	39 (9.61%)	5 (20.83%)
Drain + Anchor + C3	87 (21.43%)	3 (12.50%)
Outcome = epidermis decortication
	Crude OR (95% CI)	P	Adjusted OR (95% CI)	P
Age, y	1.07 (1.04–1.11)	<0.001	1.11 (1.07–1.16)	<0.001
Shaving time, min	1.04 (0.88–1.24)	0.643	1.49 (0.70–3.19)	0.299
Gender		0.143		0.373
Female	Reference		Reference
Male	0.40 (0.12–1.37)		0.55 (0.15–2.06)
Tumescent		0.634		0.805
No	Reference		Reference
Yes	0.79 (0.31–2.05)		0.70 (0.04–12.30)
Dressing
C3	Reference		Reference
Artiss + C1	5.44 (0.31–95.21)	0.246	66.22 (2.41–1821.97)	0.013
Anchor + C3	3.09 (0.40–24.06)	0.281	18.26 (0.91–366.91)	0.058
Tie + Anchor + C3	6.28 (0.71–56.01)	0.100	11.68 (0.25–537.36)	0.208
Drain + Anchor + C3	1.69 (0.17–16.69)	0.653	0.38 (0.01–27.94)	0.659

Anchor, multiple anchor sutures; Artiss, Artiss using; C1, compressive bandage for 1 day; C3, compressive bandage for 3 days; Drain, 1 Penrose drain leaving; Tie, tie-over dressing.

Risk Factors for Hematoma

Hematoma was observed in 10 axillae (2.33%), and a significant difference was found in whether tumescent infiltration was used (P = 0.039). No patient in the tumescent infiltration group experienced hematoma. Hematoma showed a significant difference in shaving time (P = 0.013) in χ² test but not in adjusted OR (P = 0.101). No significant differences were observed in gender, age, and dressing type between patients with and without hematoma (Table 4).

TABLE 4 - Crude and Adjusted OR and 95% CI of Suction-Curettage–Related Hematoma Associated With Predictors by Logistical Regression Model

	Hematoma
	No (n = 420)	Yes (n = 10)	P
Age, y	25.58 ± 11.19	28.70 ± 10.85	0.384
Shaving time, min	9.24 ± 2.22	8.30 ± 0.95	0.013
Gender			0.073
Female	315 (75.00%)	5 (50.00%)
Male	105 (25.00%)	5 (50.00%)
Tumescent			0.039
No	294 (70.00%)	10 (100.00%)
Yes	126 (30.00%)	0 (0.00)
Dressing			0.758
C3	50 (11.90%)	0 (0.00)
Artiss + C1	10 (2.38%)	0 (0.00)
Anchor + C3	230 (54.76%)	6 (60.00%)
Tie + Anchor + C3	43 (10.24%)	1 (10.00%)
Drain + Anchor + C3	87 (20.71%)	3 (30.00%)
Outcome = hematoma
	Crude OR (95% CI)	P	Adjusted OR (95% CI)	P
Age, y	1.02 (0.97–1.07)	0.386	1.03 (0.97–1.08)	0.319
Shaving time, min	0.72 (0.43–1.19)	0.202	2.68 (0.83–8.70)	0.101
Gender		0.087		0.217
Female	Reference		Reference
Male	3.00 (0.85–10.57)		2.28 (0.62–8.43)
Tumescent
No	—	—	—	—
Yes	—	—	—	—
Dressing				—
C3	—	—	—	—
Artiss + C1	—	—	—	—
Anchor + C3	—	—	—	—
Tie + Anchor + C3	—	—	—	—
Drain + Anchor + C3	—	—	—	—

Anchor, multiple anchor sutures; Artiss, Artiss using; C1, compressive bandage for 1 day; C3, compressive bandage for 3 days; Drain, 1 Penrose drain leaving; Tie, tie-over dressing.

Risk Factors for Skin Necrosis

Skin necrosis was observed in 16 axillae (3.72%), and a significant difference was found with respect to age (adjusted OR, 1.07; 95% CI, 1.03–1.11; P = 0.001). After adjusting for age, gender, shaving time, and dressing type, the risk of skin necrosis was higher in the tumescent infiltration group than in the nontumescent infiltration group (adjusted OR, 18.62; 95% CI, 1.32–263.21; P = 0.030) (Table 5).

TABLE 5 - Crude and Adjusted OR and 95% CI of Suction-Curettage–Related Skin Necrosis Associated With Predictors by Logistical Regression Model

	Skin Necrosis
	No (n = 414)	Yes (n = 16)	P
Age, y	25.22 ± 10.67	36.88 ± 17.44	0.018
Shaving time, min	9.19 ± 2.18	9.81 ± 2.79	0.269
Gender			0.523
Female	307 (74.15%)	13 (81.25%)
Male	107 (25.85%)	3 (18.75%)
Tumescent			0.196
No	295 (71.26%)	9 (56.25%)
Yes	119 (28.74%)	7 (43.75%)
Dressing			0.523
C3	48 (11.59%)	2 (12.50%)
Artiss + C1	10 (2.42%)	0 (0.00)
Anchor + C3	229 (55.31%)	7 (43.75%)
Tie + Anchor + C3	43 (10.39%)	1 (6.25%)
Drain + Anchor + C3	84 (20.29%)	6 (37.50%)
Outcome = skin necrosis
	Crude OR (95% CI)	P	Adjusted OR (95% CI)	P
Age, y	1.07 (1.03–1.11)	<0.001	1.07 (1.03–1.11)	0.001
Shaving time, min	1.11 (0.92–1.35)	0.274	0.63 (0.38–1.04)	0.073
Gender		0.526		0.881
Female	Reference		Reference
Male	0.66 (0.16–2.37)		0.90 (0.22–3.60)
Tumescent		0.203		0.030
No	Reference		Reference
Yes	1.93 (0.70–5.30)		18.62 (1.32–263.21)
Dressing
C3	Reference		Reference
Artiss + C1	—	—	—	—
Anchor + C3	0.73 (0.15–3.64)	0.705	5.64 (0.66–48.23)	0.114
Tie + Anchor + C3	0.56 (0.05–6.38)	0.639	13.62 (0.35–538.16)	0.164
Drain + Anchor + C3	1.71 (0.33–8.83)	0.519	13.56 (0.97–190.45)	0.053

Anchor, multiple anchor sutures; Artiss, Artiss using; C1, compressive bandage for 1 day; C3, compressive bandage for 3 days; Drain, 1 Penrose drain leaving; Tie, tie-over dressing.

Risk Factors for Infection

Infection was observed in 2 axillae (0.47%). No significant difference was found in age, shaving time, gender, use of tumescent infiltration, and dressing type between the infection and noninfection groups (Table 6).

TABLE 6 - Crude and Adjusted OR and 95% CI of Suction-Curettage–Related Infection Associated With Predictors by Logistical Regression Model

	Infection
	No (n = 428)	Yes (n = 2)	P
Age, y	25.67 ± 11.20	21.50 ± 6.36	0.599
Shaving time, min	9.20 ± 2.19	11.50 ± 4.95	0.142
Gender			1.000
Female	318 (74.30%)	2 (100.00%)
Male	110 (25.70%)	0 (0.00)
Tumescent			0.501
No	303 (70.79%)	1 (50.00%)
Yes	125 (29.21%)	1 (50.00%)
Dressing			0.860
C3	50 (11.68%)	0 (0.00)
Artiss + C1	10 (2.34%)	0 (0.00)
Anchor + C3	235 (54.91%)	1 (50.00%)
Tie + anchor + C3	44 (10.28%)	0 (0.00)
Drain + anchor + C3	89 (20.79%)	1 (50.00%)

Anchor, multiple anchor sutures; Artiss, Artiss using; C1, compressive bandage for 1 day; C3, compressive bandage for 3 days; Drain, 1 Penrose drain leaving; Tie: tie-over dressing.

Risk Factors for Overall Complications

Overall complications were observed in 52 axillae (12.09%). The cases with complications were older than those without complications (adjusted OR, 1.08; 95% CI, 1.05–1.11; P < 0.001). Ten axillae (2.33%) with skin necrosis or infection needed secondary management of debridement and closure under local anesthesia. No significant differences were found in shaving time, gender, use of tumescent infiltration, and dressing type between patients with complications and patients without complications (Table 7).

TABLE 7 - Crude and Adjusted OR and 95% CI of Suction-Curettage–Related Complication Associated With Predictors by Logistical Regression Model

	Complication
	No (n = 378)	Yes (n = 52)	P
Age, y	24.38 ± 9.97	34.88 ± 14.76	<0.001
Shaving time, min	9.19 ± 2.17	9.40 ± 2.45	0.509
Gender			0.501
Female	279 (73.81%)	41 (78.85%)
Male	99 (26.19%)	11 (21.15%)
Tumescent			0.748
No	266 (70.37%)	38 (73.08%)
Yes	112 (29.63%)	14 (26.92%)
Dressing			0.580
C3	47 (12.43%)	3 (5.77%)
Artiss + C1	9 (2.38%)	1 (1.92%)
Anchor + C3	208 (55.03%)	28 (53.85%)
Tie + anchor + C3	37 (9.79%)	7 (13.46%)
Drain + anchor + C3	77 (20.37%)	13 (25.00%)
Outcome = complication
	Crude OR (95% CI)	P	Adjusted OR (95% CI)	P
Age, y	1.07 (1.05–1.10)	<0.001	1.08 (1.05–1.11)	<0.001
Shaving time, min	1.04 (0.92–1.18)	0.508	0.92 (0.64–1.32)	0.659
Gender				0.797
Female	Reference	0.436	Reference
Male	0.76 (0.37–1.53)		0.90 (0.42–1.95)
Tumescent		0.688		0.538
No	Reference		Reference
Yes	0.88 (0.46–1.68)		1.85 (0.26–13.15)
Dressing
C3	Reference		Reference
Artiss + C1	1.74 (0.16–18.68)	0.647	5.96 (0.49–72.51)	0.162
Anchor + C3	2.11 (0.62–7.23)	0.235	5.68 (0.92–35.00)	0.061
Tie + anchor + C3	2.96 (0.72–12.26)	0.134	8.19 (0.67–100.60)	0.101
Drain + anchor + C3	2.65 (0.72–9.77)	0.145	4.10 (0.52–32.54)	0.182

Anchor, multiple anchor sutures; Artiss, Artiss using; C1, compressive bandage for 1 day; C3, compressive bandage for 3 days; Drain, 1 Penrose drain leaving; Tie, tie-over dressing.

The Association Between Postoperative Complications and Scar Formation

Regarding scarring, 15 axillae (3.49%) presented with a severe scar, 48 axillae (11.16%) presented with a mild scar, and 367 axillae (85.35%) presented with a minimal scar. Frequency distributions of postoperative complications among the different levels of scar formation are shown in Table 8. Patients with postoperative complications presented with more severe skin scarring compared with patients without complications (P < 0.05), except with hematoma. No patients experienced limited motion of axilla related from scar formation after following up over 3 months.

TABLE 8 - Crude and Adjusted OR and 95% CI of Complication Related Scar Associated With Predictors by Logistical Regression Model

	Scar
	Good	Mild	Severe	P
Hematoma				0.575
No (420 [97.67%])	359 (97.82%)	46 (95.83%)	15 (100.00%)
Yes (10 [2.33%])	8 (2.18%)	2 (4.17%)	0 (0.00)
Skin necrosis				0.003
No (414 [96.28%])	358 (97.55%)	43 (89.58%)	13 (86.67%)
Yes (16 [3.72%])	9 (2.45%)	5 (10.42%)	2 (13.33%)
Epidermis decortication				<0.001
No (406 [94.42%])	359 (97.82%)	36 (75.00%)	11 (73.33%)
Yes (24 [5.58%])	8 (2.18%)	12 (25.00%)	4 (26.67%)
Infection				0.002
No (428 [99.53%])	366 (99.73%)	48 (100.00%)	14 (93.33%)
Yes (2 [0.47%])	1 (0.27%)	0 (0.00)	1 (6.67%)
Total complications				<0.001
No (378 [87.91%])	341 (92.92%)	29 (60.42%)	8 (53.33%)
Yes (52 [12.09%])	26 (7.08%)	19 (39.58%)	7 (46.67%)

DISCUSSION

The open method of excision of apocrine glands for bromhidrosis treatment is simple and effective, but it often results in severe scarring. Over the past 20 years, many modified methods have been proposed for open or minimally invasive surgery, including liposuction and various shaving techniques.^8,9

Of these methods, open surgery with curettage has been reported to exhibit the lowest incidence of recurrence (1%), followed by open surgery (3%), liposuction with curettage (4.5%), liposuction (5%–7%), and laser treatment (8%).^10,11 However, liposuction exhibited the lowest incidence of complications (hematoma, 1.6%; necrosis, 1.5%), followed by surgery (hematoma, 1.9%; necrosis, 2.1%) and laser treatment (hematoma, 3.1%; necrosis, 4.5%). A meta-analysis by Zhang et al¹² concluded that no statistical difference was found in recurrence or complete response between suction-curettage and the open method but that the incidence of complications was lower with suction-curettage than with the open method. Therefore, a trend has been present toward minimally invasive approaches for bromhidrosis treatment. Because of its efficacy in malodor eradication and concerns about postoperative complications and scarring with the open method, suction-curettage is now more widely used than liposuction and the open method for the treatment of bromhidrosis.

Arthroscopic shavers¹³ that provide curettage and mechanical suction exhibit the advantage of faster surgery than manual curettage. However, operators inexperienced in the use of the technique should be careful to avoid perforation of the skin. If the suction is too strong, the dermis can be strongly sucked against the blade. Overall, however, the use of an arthroscopic shaver for suction-curettage offers favorable cost-effectiveness and excellent results.

Establishing the end point for curettage that achieves no malodor with the lowest risk of complications of skin necrosis or scarring is important but technically challenging. Rho et al¹⁴ reported that the skin became very thin and easy to pinch. Seo et al¹⁵ reported that the skin became violet to pale-colored with petechiae. Researchers suggested that the procedure should be interrupted when it was seen that the skin was being sucked through the holes of the cannula. We used these studies as our guides for completion of suction-curettage with the arthroscopic shaver.

A longer shaving time could increase efficacy but exhibited a higher risk of skin complications. No previous reports of the appropriate shaving time for suction-curettage exist. However, in the present study, 7 to 15 minutes of suction-curettage with an arthroscopic shaver was sufficient to achieve good efficacy for bromhidrosis treatment,⁶ although the end point should be guided by skin thickness to prevent complications. The use of tumescent infiltration before suction-curettage did not affect the evaluation of skin thickness after shaving or the efficacy of bromhidrosis treatment. No hematoma was found when tumescent infiltration was used because of vessel contracture, and the same good results were found in the study by Boeni.¹⁶ However, it resulted in significantly better pain control after the operation.

Complications of hematoma or seroma, epidermis decortication, skin necrosis, and infection were experienced in approximately 12% of axillae. Hematoma and seroma were conservatively managed with aspiration, and epidermis slough and skin necrosis were managed with wound care. Debridement and closure were used in 8 axillae with skin necrosis and 2 axillae with infection. Tumescent infiltration and a spatula single-port liposuction cannula made it easier to create subdermal tunneling to prevent hematoma and skin perforation in suction-curettage, as well as causing less postoperative pain.¹⁷

Dermis thickness was thinner in woman, and it continuously decreased with age.^18,19 Complications of epidermis decortication and skin necrosis showed a significant difference with respect to age. They were related to skin envelope circulation after shaving, and thicker dermis of skin envelope exhibited better circulation. Older patients with bromhidrosis exhibited a higher risk of postshaving complication because of thinner dermis, and adjusting the end point of dermis thickness after curettage was more difficult. A prospective cohort study by Kaoutzanis et al²⁰ concluded that age older than 45 years, smoking, and diabetes resulted in slightly higher complication rates in liposuction, and body mass index was an independent risk factor for surgical site infections. No patients with diabetes existed in our series, and all patients abstained from smoking for preoperative 2 weeks and postoperative 2 weeks to lower risk of complications.^21,22

Patients with postoperative complications of epidermis decortication, skin necrosis, and infection presented with more severe skin scarring, except those with hematoma. No patient existed with limited motion of axilla related to scarring after good management of scar massage during 3 months of follow-up. Different degrees of axillary contracture occurred in most cases in the first 1 to 2 months after operation, which was not directly related to complications, and could be recovered by massage.

Different postoperative dressing methods exhibited no significant difference in complications. Compression dressing for 3 days was effective with a lower rate of complications, and it did not need the additional procedures of anchoring sutures,²³ drain, and tie-over dressing. Fibrin sealant with compression dressing for 1 day was less uncomfortable for the patient and achieved the same lower complication incidence as a compression dressing for 3 days, but the cost was higher.

Limitations

This was the result of a retrospective statistical analysis of a single center, and many factors cannot be objectively tested and excluded. The surgeon's experience was a dependent variable in statistical analysis, but it was not easy to analyze, and it also affected the effectiveness and complications of surgery.

CONCLUSION

Post–suction-curettage complications were experienced by 12% of patients who experienced hematoma or seroma, epidermis decortication, skin necrosis, and infection. Tumescent infiltration and a spatula single-port liposuction cannula made creating subdermal tunneling easier to prevent hematoma and skin perforation. Older patients demonstrated a higher risk of epidermis decortication and skin necrosis, and the end point of suction-curettage should be carefully determined. Dressing methods did not affect complications, but fibrin sealant with compression dressing for 1 day was less uncomfortable in the postoperative recovery period. Patients with postoperative complications presented with more severe skin scarring, but no patients demonstrated limited motion of axilla after massage. Suction-curettage with an arthroscopic shaver is the most effective treatment in bromhidrosis, and it is safe with the fewest associated complications.