Complication Rate of Prepectoral Implant-based Breast Reconstruction Using Human Acellular Dermal Matrices.

Implant-based breast reconstruction (IBR) is currently the most frequently performed reconstructive technique post mastectomy. Even though submuscular IBR continues to be the most commonly used technique, mastectomy technique optimization, the possibility to check skin viability with indocyanine green angiography, the enhanced propensity of patients undergoing prophylactic mastectomies, and the introduction of acellular dermal matrices (ADMs) have paved the way to the rediscovery of the subcutaneous reconstruction technique. The aim of this article is to update the complication rate of immediate and delayed prepectoral IBR using human ADMs (hADMs).
METHODS: A literature search, using PubMed, Medline, Cochrane, and Google Scholar database according to Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines, was conducted to evaluate complication rates of prepectoral implant-based reconstructions using hADMs. The following MeSH terms were used: "prepectoral breast reconstruction acellular dermal matrix," "prepectoral breast reconstruction ADM," "human ADM breast reconstruction," and "human acellular dermal matrix breast reconstruction" (period: 2005-2020; the last search took place on April 2, 2020).
RESULTS: This meta-analysis includes 1425 patients (2270 breasts) who had undergone immediate or delayed prepectoral IBR using different types of hADMs. The overall complication rate amounted to 19%. The most frequent complication was represented by infection (7.9%), followed by seroma (4.8%), mastectomy flap necrosis (3.4%), and implant loss (2.8%).
CONCLUSIONS: The overall complication rate was 19%. The most frequent complications were infection, seroma, and mastectomy flap necrosis, while capsular contracture was rare.

INTRODUCTION

Breast reconstruction constantly advances, increasing the expectations of patients seeking natural aesthetic results, ideal breast size and shape, with a fast recovery time and low rate of complication.[1] Implant-based breast reconstruction (IBR) is currently the most frequently performed reconstructive technique post mastectomy.[2-5] Even though submuscular implant placement continues to be the most commonly performed technique,[2-4] the new generation of prepectoral IBR has shown encouraging results and has become more widespread in the last 15 years.[6-8] The optimization of mastectomy techniques, together with the possibility to check skin viability with indocyanine green angiography, the enhanced propensity of patients to undergo prophylactic mastectomies, and the introduction of acellular dermal matrices (ADMs), have paved the way to the rediscovery of the subcutaneous reconstruction technique.[8] The main advantages of the prepectoral IBR compared with the submuscular IBR are: the possibility to achieve a more natural breast shape, the complete elimination of animation deformity, and the disappearance of functional deficits and pain caused by the pectoral muscle elevation.[6-9] Similar to the submuscular IBR, the prepectoral IBR can be performed in 1 (direct to implant: DTI) or 2 stages following skin-sparing or nipple-sparing mastectomies. Moreover, patients affected by animation deformity or dysfunctional chronic chest pain can be selected for delayed IBR by way of a pocket conversion from a submuscular to a prepectoral plane.[10-13]

The introduction of ADMs has certainly represented a key point in the modern prepectoral IBR. In fact, the ADM brings an additional interface between the implant and the mastectomy flap as well as additional support to keep the implant in the desired position, leading to the improvement of functional and aesthetic outcomes.[14,15] ADMs consist of biodegradable soft connective tissue grafts created from a decellularization process that presents a regenerative capacity. Upon implantation, this structure serves as a scaffold for donor-side cells to facilitate subsequent incorporation and revascularization.[16] ADMs are derived from human (AlloDerm, LifeCell, Branchburg, N.J.), porcine (Strattice—LifeCell Corporation; Branchburg, N.J. and Permacol—Covidien, Mansfield, Mass., USA), or bovine (SurgiMend—Integra Life Sciences, Plainsboro, N.J. and Braxon—Decomed S.r.l., Venezia, Italy) tissues.[17]

In 2005, Breuing and Warren[18] described the first submuscular IBR using a human-derived ADM (AlloDerm, LifeCell, Branchburg, N.J.) positioned between the inferior border of the pectoralis major muscle and the chest wall, with the aim to expand the lower pole of the breast. Since then, several series of IBR that adopted ADMs have been brought forth. Indeed, a recent study reports that more than 50% of IBRs are performed using an ADM in the United States.[19] Nonetheless, the universal approval of ADMs for IBR has been restrained by worries concerning surgical complications such as seroma, infection, and mastectomy flap necrosis. In the last 10 years, many reports concerning the incidence of complications related to ADMs have been published.[20-23] The aim of this article is to provide an update of the complication rates of immediate and delayed prepectoral IBR using human ADMs (hADMs).

MATERIALS AND METHODS

A literature search via PubMed, Cochrane, and Google Scholar databases according to Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA)[24] guidelines was conducted to evaluate the complication rates of prepectoral implant–based reconstructions using hADMs. The following MeSH terms were used: “prepectoral breast reconstruction acellular dermal matrix,” “prepectoral breast reconstruction ADM,” “human ADM breast reconstruction,” and “human acellular dermal matrix breast reconstruction” (period: 2005–2020; the last search was performed on April 2, 2020). Two independent reviewers performed two-stage screening and data extractions. Abstracts were screened to identify eligible papers. Reference lists of selected studies were screened for additional articles. The search strategy is shown in the flow chart below.

Inclusion and Exclusion Criteria

Studies were selected based on the following inclusion criteria: (i) studies selectively investigating prepectoral implant–based reconstructions using human acellular dermal matrix; (ii) registration of complication after surgical treatment; (iii) minimal sample of 10 breasts; (iv) full text available in English. Studies were excluded due to any of the following criteria: (i) review articles; (ii) case reports; (iii) articles only reporting surgical techniques and not outcomes or complication rates; (iv) articles describing studies that included fewer than 10 breasts; (vi) non-referenced articles; (vii) expert opinion (Level V).

Data Collection

Extracted data included:

1) Preoperative data: number of patients included, number of breasts included, age (mean and range), body mass index (BMI, mean and range), inclusion and number of patients who had undergone previous or adjuvant radiation therapy, inclusion and number of patients with BMI > 30 kg/m2, eligibility criteria of prepectoral IBR (inclusion and exclusion criteria).

2) Intraoperative data: type of mastectomies (skin-sparing mastectomy SSM; nipple-sparing mastectomy NSM), implant or expander characteristics, type of HADM, size of HADM, ADM placement, presence of contemporary dermal flap.

3) Postoperative complications: follow-up (mean and range), overall incidence of complication, incidence of seroma, infection, implant loss, mastectomy flap necrosis, capsular contracture.

4) Outcomes: patient satisfaction, aesthetic outcomes, rippling, revision surgery (fat graft, implant exchange).

Moreover, patients who had undergone preoperative or postoperative radiation therapy (RT) and patients affected by obesity (presenting a BMI > 30 kg/m2) were analyzed separately to calculate the odds ratio relative to the overall complication rate.

Bias Assessment

The Q and the I2 of each study were assessed using fixed-effect models.

Statistical Analysis

Statistical analysis was performed using SPSS statistical software (version 24.0; IBM Corporation, Somers, N.Y.). Pooled random effect estimates for each postoperative surgical complication and 95% confidence interval (95% CI) were calculated using Microsoft Excel (Microsoft Corp, Redmond, Wash.). Univariate logistic regression models were performed to identify two risk factors (radiation therapy and obesity) associated with overall complication rate, including seroma, infection, flap necrosis, implant loss, and capsular contracture.

RESULTS

After exclusion of duplicates, 359 articles were identified. Two different reviewers analyzed all the records by titles and abstracts. Sixty-five full-text articles were examined for eligibility. Eighteen studies[12,13,25-40] published between 2015 and 2020 were considered eligible based on appropriateness, relevance, and recency (Fig. 1). A total of 1425 patients and 2270 breasts were included in the meta-analysis, and the sample size of each study ranged from 23 to 353 breasts. Figure 2 shows the Q and the I2 of each study assessed using fixed-effect models.

Fig. 1.

Flowchart according to PRISMA guidelines.

Fig. 2.

Q and the I2 (fixed-effect models).

Patient age ranged from 19 to 83 years, bringing the average age to 42.2 years. Patient BMI ranged from 17.8 to 54 kg/m2. Twelve studies included obese patients and 4 studies reported the exact number of obese patients included (104). Nine studies included patients who had undergone prior radiation therapy (133 patients). Sixteen studies reported the number of patients who had undergone postoperative radiation therapy (171 patients).

Exclusion criteria resulted to be heterogeneous. All the authors performed an intraoperative mastectomy skin flap evaluation using perfusion assessment devices. Active smokers and previous radiation therapy represented exclusion criteria respectively in 3 and 5 articles. Patients with a BMI greater than 30 or 40 kg/m2 were excluded in 4 studies. Eleven studies included patients with a BMI greater than 30 kg/m2 and 5 studies openly mentioned the number of obese women included in the sample, for a total of 126 patients. Fourteen articles included patients who had undergone previous radiotherapy or adjuvant radiotherapy (235 cases).

Table 1 shows each study’s patient preoperative characteristics in detail. Ten studies described cases of immediate IBR, 4 studies described cases of delayed IBR, and 4 studies reported both. In summary, 1015 patients underwent immediate breast reconstruction (DTI or two-stage), while 410 underwent delayed breast reconstruction. Among them, 274 patients (426 breast) underwent pocket conversion from a subpectoral to a prepectoral plane. Implant size ranged from 125 to 770 ml.

Table 1.

Sample Characteristics

Study	Sample	Age (Mean-range in Years)	BMI	Radiation Therapy*	BMI > 30	Inclusion Criteria	Exclusion Criteria
Becker[25]	26 pts 56 b	51	—	Included 5 pts	—	DTI prepectoral BRVertical mastectomy incision	None
Woo[26]	75 pts 135 b	50,2 (26–76)	26,9 (17, 9–44,6)	—	Included	Prepectoral ADM BR	Poor mastectomy flap perfusion (SPY Elite System)Active smokers
Downs[27]	45 pts 79 b	46,8 ± 9,6	24,3 ± 5,4	Included 34 pts	—	DTI prepectoral BR in NSM	Poor mastectomy flap perfusion (SPY Elite System)
Zhu[28]	29 pts 50 p	50,48 ± 8,8	27,8 ± 5,35	Included 21 pts	Included	Two-stage prepectoral BR	Poor mastectomy flap perfusion (SPY Elite System)Active smokers
Schnarrs[29]	188 b	54 (29–76)	27,3 (range 17, 8–51)32 pts > 30	—	Included 32 pts	Prepectoral hADMprimary or revision BR	Cosmetic proceduresXenogenic or syntheticADMFenestrated human ADM
Bohac[30]	22 pts 26 b	46 (30–69)	—	Included 6 pts	—	Two-stage prepectoral BR	None
Sigalove[31]	207 pts 353 b	2370	22–39	Included 27 pts	Included	Prepectoral ADM BR	BMI > 40Previous RTActive smokersImmunocompromissionHbA1c > 7.5%Oncologic: deep tumors, >5 cm, chest wall involvement, grossly positive axillary involvement, high risk of recurrencePoor mastectomy flap perfusion
Jones[32]	50 pts/73 b	47 (26–67)	27 (18–44)	Included 11 pts	Included	Prepectoral ADM BR	Poor mastectomy flap perfusion (SPY Elite System)
Jones[33]	90 pts 142 b	55 (29–77)	28 (19–42)	Included 21 pts	Included	Animation deformity after subpectoral BR	No exclusion criteria in terms of skin flap thickness, RT, BMI, smoke, or diabetes
Gabriel[12]	57 pts 102 b	53,2 (34–77)	27,3 (19–47)14 pts BMI > 30	Included 9 pts	Included 14 pts	Animation deformity or chronic pain ± implant malposition after subpectoral BR	Poor mastectomy flap perfusion (PDE, Hamamatsu)Previous RTActive smokersDiabetes
Lentz[13]	31 pts 55 b(46/55 ADM)	49.8 (34–72)	26.06 (20–39)	Included 7 pts	Included	Animation deformity or chronic pain ± implant malposition after subpectoral BR
Jones[34]	140 pts 194 b+ 94 pts 111b	—	—	—	—	Immediate implant-based BR	Poor mastectomy flap perfusion (SPY Elite System)
Sigalove[35]	33 pts 52 b	50.6 (23–75)	27,7 (16–42)BMI > 30 12 pts	Included 33 pts	Included 12 pts	Immediate prepectoral BR following SSM or NSM + postoperative RT	Previous RTActive smokersBMI > 40HbA1c > 7.5%Lack of fat donor sitesOncologic: deep tumors, >5 cm, chest wall involvement, grossly positive axillary involvement, high risk of recurrencePoor mastectomy flap perfusion (SPY Elite System)
Lee[36]	23 pts 23 b	45 (30–53)	22,1 (19, 30, 9)	Included 9 pts	—	Prepectoral BRSmall-to-moderate breast sizePinch test ≥ 1.5 cmMastectomy flap ≥ 0.5 cm	Severely ptotic breastBMI > 30Diabetes mellitusActive smokersPrevious RT
Safran[37]	214 pts 313 b	48.6 (19–83)	25.9 (18–54)>30 22 pts	Included 38 pts	Included 22 pts	Prepectoral DTI BR	Previous failed implant-based BRExtensive skin envelope radiation damageLocally advanced cancer, Extensive skin excisionDelayed or autologous BR
Manrique[38]	19 pt 36 b	45 (39–47)	27,6 (20–38)	Included 3 pts	Included	Prepectoral 2-stage BR	Poor mastectomy flap perfusion (SPY Elite System)
Bilezikian[39]	131 pts 230 b	29–82	19–48 >3046 pts 79 b	19–48 >3046 pts 79 b	Included 46 pts	Prepectoral BR	—
Holland[40]	45 pts 80 b	50.3	26,2	26,2	—	Animation deformity after subpectoral BR	—

*Previous or adjuvant.

ADM, acellular dermal matrix; BMI, body mass index; pts, patients; b, breasts; BR, breast reconstruction; NSM, nipple-sparing mastectomy; SSM, skin-sparing mastectomy.

Alloderm (LifeCell, Branchburg, N.J.) was the most common hADM (12 studies, 1062 patients), and size 16 × 20 cm was chosen in all cases. Four studies (201 patients) described prepectoral IBR using AllodermRTU (LifeCell, Branchburg, N.J.) 16 × 16 cm or 16 × 20 cm. Three studies (58 patients) reported the adoption of FlexHD (Ethicon, Inc., Somerville, N.J. and Musculoskeletal Transplant Foundation, Edison, N.J.), with a size of 16 × 16 cm or 16 × 20 cm. Dermacell (LifeNet Health, Virginia Beach, Va.) and Cryoderm (CGBio Corp., Seongnam, Korea) were used in 3 articles. Figure 3 shows the characteristics of hADM included in the review. Twelve of 18 studies reported author disclosures (affiliation with the company providing the breast implant or the ADM); no funding was received for any of the included studies.

Fig. 3.

HADM included in the metanalysis.

In 7 articles the hADM was manually fenestrated to facilitate the drainage. Concerning the hADM placement, 10 articles described complete anterior implant coverage, and 7 studies described complete anterior and posterior implant coverage using 2 ADM sheets. Lastly, 1 study described coverage limited to the lower pole of the implant. In 2 studies, the hADM was combined with a local autodermal-adipous flap harvested from the inferior and lateral pole of the breast.

Table 2 shows the intraoperative data of each study in detail. Mean follow-up was 17.6 months (range, 1–103). Pooled random effect estimates of overall complication rates (95% CI) was 19% (CI 8.9%–29.1%; 193 patients). Pooled random effect estimates of each postoperative surgical complication (95% CI) resulted in 4.8% (CI 2.25%–7.49%; 59 patients) for seroma, 7.9% (CI 2.74%–7.9%; 107 patients) for infection, 2.8% (CI 0.34%–5.26%; 40 patients) for implant loss, 3.4% (CI 0.24%–7.03%; 51 patients) for mastectomy flap necrosis, and 1.2% (0%–2.4%; 8 patients) for capsular contracture (Table 3). With regard to infections (the most common complication found), only 1 study[39] stated the grade of infection when reporting 10 cases of superficial cellulitis and 7 cases of pocket infection. Moreover, in 69 of 107 cases, the type of treatment put in place was specified: 27 patients had undergone oral antibiotic therapy, 18 patients had undergone intravenous antibiotic therapy, and 24 patients had undergone revision surgery. Among patients undergoing revision surgery, 11 had undergone an “explantation-washout-debridement-reimplantation” procedure. Lastly, 13 cases of implant loss due to infection were reported. One study[36] reported 9 cases of implant loss without stating how many of them were due to infection.

Table 2.

Intraoperative Data

Study	MastectomyType	Time of BR	Type of BR	Implant	ADM	Treatment	Size	Location	DermalFlap
Becker[25]	NSM (vertical incision)	Immediate	DTI	Smooth round saline (Spectrum, Mentor Corp)	FlexHD Pliable(84% of pts)	Trimming	16 × 16 cm	Complete anterior coverage	Lateral flap
Woo[26]	SSMNSM	Immediate	DTI 10%Two-stage 90%	—	AllodermRTU	FenestrationTrimming	16 × 20 cm	Complete anterior and posterior coverage	No
Downs[27]	NSM	Immediate	DTI	Silicone gel anatomical (Mentor)	AlloDerm 69%/FlexHD 31%	—	16 × 20 cm	Complete anterior or anterior and posterior coverage	No
Zhu[28]	NSM 21 (42%)SSM 29 (58%)	Immediate	Two-stage	Expander (Natrelle 133 Mx or MV)	AlloDerm [15/29 pts]		Not reported	Lower pole coverage	Inferior flap3/29 pts
Schnarrs[29]	SSMNSM	Immediate 53 ptsDelayed 135 pts	DTI 1Two-stage 5236 PocketConversion	—	AlloDerm 143AlloDerm RTU 19FlexHD 18hmatrix 32	Trimming	16 × 16 or16 × 20 cm	Complete anterior coverage	No
Bohac[30]	—	Delayed 23 ptsImmediate 3 pts	Two-stage	Expander (Mentor)	ADM developed in CTB	Decellularization	273 cm2(155–655)	Complete anterior coverage	No
Sigalove[31]	SSMNSM	ImmediateDelayed	DTI 46Two- stage 293Two-stage 14	Natrelle MX/Natrelle Inspira/-	AlloDerm	—	—	—	No
Jones[32]	SSM 47%NSM 53%	Immediate	DTI	Anatomic cohesive gel full height and profile-textured surface implants (Allergan)	AlloDermRTU	WashingFenestrationTrimming	16 × 20 cm	Complete anterior coverage	No
Jones[33]	SSMNSM	Delayed	Pocket Conversion	FX or FF implant profile	AlloDerm	Trimming	16 × 20 cm	Complete anterior coverage	No
Gabriel[12]	SSM 72%NSM 28%	Delayed	Pocket Conversion	Round silicone implantSize: 603ml (400–800)	AlloDerm RTULifeCell	Fenestration	16 × 20 cm	Complete anterior coverage	No
Lentz[13]	—	Delayed	Pocket Conversion	Smooth round cohesive or responsive silicone gel (Allergan)Upsize 90ml	AlloDerm (46/55)	WhashingFenestration	One or two sheets16 × 20 cm (±6 × 16, 8 × 10 or 16 × 20)	Two sheetsComplete anterior and posterior coverage	No
Jones[34]	—	Immediate	DTI	—	AlloDerm	Trimming	16 × 20 cm	Complete anterior coverage	No
Sigalove[35]	SSM 49 (94,2%)NSM 3 (5,8%)	Immediate	DTI 19 (36,5%)Two-stage(implant 6 weeks PO)33 (63,5%)	—	AlloDerm	Fenestration	16 × 20 cm	Two sheetsComplete anterior and posterior coverage	No
Lee[38]	SSM 7NSM 16	Immediate	DTI	Textured, anatomic (Mentor or Allergan)Size: 252, 4 ml (125–440)	CryoDerm	Washing (Betadine)Fenestration	212 cm	Two crossed ADMsComplete anterior and posterior coverage (PICO)	No
Safran[37]	SSM 45 (14,3%)NSM 267 (85,3%)	Immediate	DTI	Size: 392.9 ml (220–700)	AlloDerm243 b (77,6%)	Fenestration	—	Complete anterior coverage	[35 pts 11.2%]Autoderm Wise pattern
Manrique[39]	NSM	Immediate	Two-stage	Round, smooth, silicone, cohesive gel (Allergan)Size: 435ml (125–750)	AlloDerm	—	16 × 20 cm	Complete anterior and posterior coverage	No
Bilezikian[39]	SSMNSM—	ImmediateDelayed	DTIPocket Conversion15 pts/20 b	Round, smooth, SCX (Allergan Inspira)Size: 240–800 ml	DermaCell	—	16 × 20 cm	Complete anterior coverage	No
Holland[40]	—	Delayed	Pocket Conversion	Cohesive gel implantsSize 588 ml (220–770)	AlloDerm65 pts (81,3%)	—	16 × 20 cm	Complete anterior, Partial posterior and inferior coverage	No

ADM, acellular dermal matrix; BR, breast reconstruction; DTI, direct to implant; NSM, nipple-sparing mastectomy; PO, postoperative; pts, patients; SSM, skin-sparing mastectomy.

Table 3.

Pooled Random Effect Estimates for Each Postoperative Surgical Complication and 95% Confidence Interval

	Complication	Rate
Global	19%	IC 95% 8,9–29,1%
Seroma	4,8%	IC 95% 2,25–7,49%
Infection	7,9%	IC 95% 2,74–11,9%
Implant Loss	2,8%	IC 95% 0,34–5,26%
Mastectomy flap necrosis	3,4%	IC 95% 0,24–7,03%
Capsular Contracture	1,2%	IC 95% 0–2,4%

Table 4 reports in detail the complication rate by study. A forest plot is provided to better visualize the pooled data on complications in Figure 4. Aesthetic outcomes were mentioned in 5 studies, and 2 studies used the “Aesthetic Items scale” and the “Likert scale.” Only 5 studies reported the number of patients showing postoperative rippling, the incidence of which ranged from 0.8% to 35%. Patient satisfaction was reported in 3 studies: 1 study adopted the “KNUH Breast Reconstruction Satisfaction Questionnaire” and another the “Breast Q” reporting satisfactory scores. Nine of 16 studies mentioned cosmetic revision procedures. The most common intervention consisted of one or more session of fat grafting. Implant exchange (cohesive implant or different size) was also performed in some cases.

Table 4.

Complication Rates

Study	Follow-up(mo)	ComplicationRate	Seroma	Infection	Implant Loss (total)	Implant Loss due to Infection	Mastectomy Flap Necrosis	Capsular Contracture
Becker[25]	23,1 (1–55)	29% 6 pts	3,2% 1 pt	3,2% 1 pt	6,4% 2 pts	3,2% 1 pt	6,4% 2 pts	0
Woo[26]	10 (2–36)	11,4% 9 pts	3,8% 3 pts	3,8% 3 pts/4b	3,8% 3 pt/2b	3,8% 3pts/4b	0	0
Downs[27]	22 (12, 7–33, 5)	81% 36b	15,2% 6 pts	10% 4 pts	17,7% 7pts	7,5% 3 pts	27,8 11 pts	10% 4 pts
Zhu[28]	—	16% 5 pts2% major14% minor	10% 5 pts	4% 2 pts	4% 2 pts (4 b)	4% 2 pts (4 b)	4% 2 pts	0
Schnarrs[29]	—	19,7% 18 ptsNo significant difference between 4 types of HADMs	3,7% 4 pts	18% 17 pts	0	0	3,1% 3 pts	0
Bohac[30]	12,6 (4–21)	3,8% 1 pt	0	3,8% 1 pt (RT)	0	3,8% 1 pt (RT)	0	0
Sigalove[31]	6–26	9% 19 pts	2% 7 pts	4,5% 16 pts	0	0	2,5% 9 pts	0
Jones[32]	48 (13–103)	32% 16 pts	12,2% 9 pts	5,4% 4 pts	2.7% 2 pts	2.7% 2 pts	0	1,3% 1 pt
Jones[33]	19,5	12% 11 pts	2,1% 3 pts	4,2% 6 pts	0,7% 1 pt	0	0,7% 1 pt	0
Gabriel[12]	16,7 (4–65)	3,9% 4 pts	2% 2 pts	0	3.9% 4 pts	0	3.9% 4 pts	0
Lentz[13]	8.3 (1–27)	19,6% 9 pts	0	17,4% 8 pts	2,2% 1 pt	2,2% 1 pt	0	0 pts ADM
Jones[34]	Group 1: <3,8 yGroup 2: 15,1 m (max 3,6 y)	27,8% 39 pts1% 1 pt	5,2% 7 pts0	5,2% 7 pts0	6,7% 9 pts0	—	4,1% 6 pts(superficial)0	00,9% 1 pt
Sigalove[35]	25,1 (15–37)	5,9% 2 pts	2,9% 1 pt	0	2,9% 1 pt	0	0	0
Lee[36]	12	17,3% 4 pts	13% 3pts	0	0	0	0	4,3% 1 pt
Safran[37]	>12	18,5% 44 pts (among ADM group)	2,2% 7 pts (among all groups)	5,4% 17 pts (among all groups)	0	0	4,4% Superficial 3,8% Full thickness0,6% 12 pts (among all groups)	—
Manrique[38]	17	0	0	0	0	0	0	0
Bilezikian[39]	24	4% 5 pts	0	4% 10b	4% 10b	0	0	0
Holland[40]	15.2 (1.5–45)	17,5% 8 pts	2,5% 2 pt	13,8% 11 pts	1,5% 1	1,5% 1	1,25% 1 pt	1,5% 1

pts, patients; b, breasts.

Fig. 4.

Forest plot (pooled data on complications).

Tables 3, and 5 report the postoperative outcomes in detail. Among the group of patients who had undergone prior (133) or postoperative (171) RT (total: 304), 22 showed postoperative complications (6 prior RT and 16 postoperative RT). In detail, 11 patients showed infection, 5 showed mastectomy flap necrosis, 3 showed seroma, and 1 showed a grade III capsular contracture. Overall, 3 patients who had undergone prior or postoperative radiation therapy underwent implant removal. With respect to obesity, neither of the studies reported complications in the obese patient subgroup. Univariate logistic regression models considered radiation therapy (RT) and obesity separately. Neither of the 2 subgroups showed an increased risk of complications compared with the control group (Table 6). Pooled complications stratified by RT exposure analyzing rate of implant loss and capsular contracture did not show an increased risk of complication when comparing the RT group with the control group (Table 7). Table 8 provides a descriptive overview on complications in patients who had undergone radiation therapy in the analyzed studies.

Table 5.

Aesthetic Outcomes and Revision Surgery

Study	Rippling	Revision	Fat Graft %	Implant Exchange	Pt Satisfaction	Aesthetic Scale
Becker[25]	—	Reported	15,3%	14% implant exchanged for silicone implants	—	—
Woo[26]	—	—	Second stage	—	—	-2,95/4 overall aesthetic score-95% of patients has similar or improved aesthetic outcome.
Downs[27]	35% 14 pts	Reported	60%–70%	No	—	—
Zhu[28]	—	—	0	0	—	—
Schnarrs[29]	—	—	—	—	—	—
Bohac[30]	—		0	0	—	—
Sigalove[31]	—	—	Second or third stage	0	—	—
Jones[32]	12% 9 pts	Reported	37%	Two patients requested upsizing to a larger, higher profile implant at 12–15 months	High	—
Jones[33]	4,9% 7 pts	Reported	18,3% 26	0,7% 1 pt requested a smaller implant size	—	Overall breast aesthetics were improved
Gabriel[12]	—	—	—	—	—	Animation deformity resolution
Lentz[13]	—	Reported	21.4% (group without prior fat graft)	—	—	Animation deformity resolution
Jones[34]	15% 18 pts	Reported	38% 13,5%	7,2% 1,8%	—	Successful outcome was achieved in 93.3% of cases
Sigalove[35]	—	Reported	—	1 pt TRAM	—	—
Lee[36]	—	Reported	—	1 pt latissimus dorsi flap	KNUHBreastReconstruction Satisfaction Questionnaire:high	Successful outcome
Safran[37]	—	—	—	—	—	—
Manrique[38]	—	Reported	21% 4 pts	0	Breast Q 84.8	Aesthetic Items Scale mean 20.9Likert Scale
Bilezikian[39]	0,8% 2b	3% 7 breasts cosmetic revisions	2 pts/4b	1 larger implant1 cohesive gel because of visible rippling	—	—
Holland[40]	—	6.2% 4 cosmetic revision	—	—	—	7 asymmetric

Table 6.

Binary Logistic and Linear Regression Models for Radiotherapy and Obesity Variables on Surgical Outcomes

	Complication Rate
	OR	IC	P
RT	1,2163	0,82–1,79	0,32
BMI > 30 kg/m2	OR	IC	P
	1,03	0,75–1,43	0,8

Table 7.

Binary Logistic and Linear Regression Models for Radiotherapy Variable on Seroma, Infection, Mastectomy Flap Necrosis, Implant Loss, and Capsular Contracture

	OR	IC	P
Seroma	0,1236	0,03–0,50	0,0038
Infection	0,279	0,134–0,580	0,0006
Mastectomy flap necrosis	0,1449	0,035–0,599	0,0077
Implant loss	0,4205	0,128–1,3766	0,152
Capsular contracture	0,5252	0,064–4,285	0,54

Table 8.

Overview on Complications by RT Exposure

	Radiation Therapy (no. pts.)	No Radiation Therapy (no. pts.)
Overall patients	304	1121
Overall complications	22	193
Seroma	2	57
Infection	8	99
Implant loss	3	48
Mastectomy flap necrosis	2	49
Capsular contracture	1	7

DISCUSSION

One of the most meaningful advancements in prepectoral IBR was the introduction of ADMs. In 2015, Reitsamer[41] described the first series of 13 patients who had undergone a prepectoral IBR using complete anterior implant coverage with a porcine ADM. Since then, several authors have published their series of 2-stage or DTI prepectoral IBRs using bovine, porcine, or human ADMs.[15,20,42] In truth, prepectoral IBR can be performed with or without ADMs but different authors have assessed that complication rate and, in particular, the capsular contracture rates are significantly lower when prepectoral IBR is performed with an ADM support.[12,43,44]

However, there is a lack of long-term results regarding prepectoral IBR because the technique is relatively new. Our study has the aim to provide an update regarding complication rates related to prepectoral breast reconstruction using the subgroup of ADMS derived from human tissue. The present meta-analysis includes 1425 patients (2270 breasts) who have undergone immediate or delayed IBR employing different types of hADMs summarized in Figure 2. No comparative analysis between the different products has been performed in this study. Previous meta-analysis reported similar risks of complication using different types of hADMs.[26] The most frequently described hADM was AlloDerm (LifeCell, Branchburg, N.J.), followed by AllodermRTU (LifeCell, Branchburg, N.J.), FlexHD (Ethicon, Inc., Somerville, N.J. and Musculoskeletal Transplant Foundation, Edison, N.J.), Dermacell (LifeNet Health, Virginia Beach, Va.), and Cryoderm (CGBio Corp., Seongnam, Korea).

ADM assemblage can be executed by positioning the matrix into the subcutaneous pocket due to soft tissue support (on-label technique) creating complete anterior coverage or, by wrapping the matrix around the tissue expander or the implant before insertion into the subcutaneous pocket, creating complete anterior and posterior coverage (off-label technique). In our review, 10 articles described complete anterior coverage of the implant (180 degree wrapping), 7 studies described complete anterior and posterior implant coverage using 2 ADM sheets (360 degree wrapping). Moreover, in 2 studies, hADM support was combined with the use of a local autodermal-adipous flap harvested from the inferior and lateral pole of the breast.

Eligibility for prepectoral breast reconstruction represents a very relevant issue. Our meta-analysis exclusion and inclusion criteria resulted quite heterogeneous. All authors agreed on the need to perform an intraoperative mastectomy skin flap evaluation using perfusion assessment devices in cases of both immediate and delayed breast reconstruction. Active smokers, previous RT, and obesity represented exclusion criteria in 5, 3, and 4 articles, respectively. Most studies included obese patients and patients who had undergone previous or adjuvant radiation therapy.

Even if there is a lack of long-term outcomes in our meta-analysis (mean follow-up, 17.6 months), results with more than 2 years of follow-up are encouraging. The overall complication rate amounted to 19%, including 4.8% cases of seroma, 7.9% cases of infection, 2.8% cases of implant loss, and 3.4% cases of mastectomy flap necrosis. No authors mentioned the onset of red breast syndrome.

Capsular contracture was reported in 1.2% of patients, which was derived from local inflammation related to an exaggerated production of collagen by fibroblastic cells in contact with the breast implant.[45] Our data are in line with previous literature that reported a significantly lower rate of capsular contracture in cases of ADMs compared with conventional IBR.[12,43-47] In fact, clinical and experimental studies demonstrated that hADMs induce a reduction of chronic inflammation and fibroblasts’ proliferation compared with native breast capsules.[48] Lately, prepectoral reconstruction in the setting of radiotherapy has become a crucial topic of interest.[15,34,45] The present univariate logistic regression analysis suggests that previous or adjuvant radiation therapy should not be considered as an independent risk factor (Table 6).

Nahabedian hypothesized that the side effects of radiotherapy are more marked toward the pectoralis muscles mostly when the inferior origin has been interrupted.[15] In fact, while radiotherapy, in the setting of a subpectoral device, commonly caused skin tightening and an inframammary fold elevation with a cephalad displacement of the implant, when radiation is delivered in the setting of a prepectoral device, elevation of the inframammary fold is not observed.[15] Similarly, univariate logistic regression analysis investigating the subgroup of patients with a BMI greater than 30 kg/m2 did not show a significant increase in overall complication rates (Table 6).

Implant rippling still remains an unwanted side effect of prepectoral breast reconstruction, with an incidence reported between 0.8% and 35%.[48] Preoperative or postoperative fat graft, especially in patients with low BMI, represents the most frequently used technique to repair this cosmetic complication. In 2017, Saliban recommended creating a prepectoral pocket with a smaller base than that of the prosthesis and closing the inframammary fold incision under a moderate amount of tension in leaner patients.[45] In our review, only 5 studies mentioned the onset of rippling (ranging from 0.8% to 35%), and all cases were corrected by fat graft or implant exchange (upsizing or cohesive gel implant). Astoundingly, only 4 studies reported patient satisfaction scores or objective aesthetic evaluation scales. These data point out that there is still a lack of homogeneity concerning breast reconstruction outcome evaluation.

Our meta-analysis incorporated different types of breast reconstructions, including direct-to-implant, two-stage reconstruction and pocket conversion from a subpectoral to a prepectoral plane. The latter technique was described in 6 studies and requires the dissection of the plane between the pectoralis major muscle from the overlying mastectomy skin flap, the removal of the indwelling implants, the anchorage of the pectoralis muscle to the anterior chest wall, and the placement of the implant in the new prepectoral pocket. All patients who had undergone this operation reached a complete resolution of animation deformity and chronic pain. Gabriel[12] compared the results of two groups of patients who had undergone a pocket conversion with or without the use of hADM, showing a significant decrease in the complication rate in the first group.

There are several limitations to the meta-analysis, such as the possible conflict of interest related to the use of ADMs and possible publication bias. Six of 18 studies had no conflicts of interest to declare.[28,30,33,35,36,38] However, no funding or financial support was received from any company for any of the included studies. The relative small sample size (<100 breasts) in 9 studies, the fact that many studies were set as retrospective analysis, and the lack of correction of the confounding factors limited the power of statistical analysis.

CONCLUSIONS

This meta-analysis includes 1425 patients who had undergone immediate or delayed IBR adopting hADMs, with the aim to provide updated information regarding complication rates related to this procedure. The overall complication rate amounted to 19%. The most frequent complication was infection (7.9%), followed by seroma (4.8%), mastectomy flap necrosis (3.4%), and implant loss (2.8%). Capsular contracture was reported in fewer than 1% of patients.