Utilization of Banked Human Adipose Allograft for Host Mediated Replacement of Dermal White Adipose Tissue and Volumization in Plastic and Reconstructive Surgery: A Systematic Review and Meta-Analysis

Introduction

Dermal White Adipose Tissue (dWAT) is now recognized as a histologically and functionally distinct adipocyte compartment within the reticular dermis, separate from Subcutaneous White Adipose Tissue (sWAT), that contributes to thermoregulation, hair-follicle cycling, innate immune defense, wound healing, and mechanical cushioning of the skin [7-12]. Loss of dermal and subcutaneous adipose volume—whether from intrinsic aging, photodamage, GLP-1 associated weight loss, surgical excision, trauma, lipodystrophy, or oncologic resection—produces visible contour deformity, skin laxity, and, in load-bearing regions such as the plantar foot, loss of protective cushioning that predisposes to ulceration [7,9,11,12]. Because adipocytes reach replicative and hypertrophic stasis after puberty and do not meaningfully self-renew once destroyed, damaged adipose tissue represents a largely irreversible deficit without intervention [3]. Autologous Fat Grafting (AFG), first described over a century ago and refined by Coleman’s structural fat grafting technique, remains the most widely used method for restoring soft-tissue and dermal volume in aesthetic and reconstructive surgery [13]. AFG is attractive because it transfers living, vascularizable autologous tissue, but its clinical utility is constrained by several well-documented limitations: donor-site morbidity, operative time required for harvest and processing, and highly variable graft survival, with reported longterm volumetric retention ranging from approximately 26% to 83% depending on anatomic site, technique, and patient factors [14-16].

A 2018 systematic review and meta-analysis of AFG in facial reconstruction similarly reported wide heterogeneity in retention and complication rates across the published literature, underscoring the need for a more standardized, reproducible alternative [16]. Banked (allogeneic) human adipose tissue offers a potential solution to these limitations. Adipose tissue obtained from screened, deceased tissue donors through tissue banks accredited by the American Association Of Tissue Banks (AATB) and regulated under the U.S. Food and Drug Administration’s Good Tissue Practice framework (21 CFR Part 1271) can be processed—typically through detergent-based delipidization, decellularization, or minimal-manipulation protocols—into an off-the-shelf Allograft Adipose Matrix (AAM) that preserves the native adipose Extracellular Matrix (ECM), including collagens I, IV, and VI, laminin, fibronectin, and growth factors such as FGF-2 and VEGF, while removing lipid, cellular debris, and immunogenic DNA content [1-4,17]. When implanted, it is proposed to function as a bioactive scaffold that recruits host adipose-derived stem cells and adipocyte progenitors, supporting angiogenesis and de novo host adipogenesis—a process the manufacturer and investigators term host-mediated tissue replacement or neoadipogenesis [1-4]. This mechanism differs fundamentally from that of both traditional AFG, which transplants living autologous adipocytes, and synthetic or hyaluronic-acid-based dermal fillers, which provide inert, nonregenerative volume. Preclinical work in nude-mouse and other animal models has shown that banked adipose allograft matrices support progressive host-cell infiltration, neovascularization, and adipocyte repopulation over several weeks to months, findings that have been corroborated histologically in early human studies [3,4,18,19].

Available products include: Renuva, 100% extracellular matrix (MTF Biologics, Edison,NJ), AlloClae, 60% adipocytes, 40% extracellular matrix (Tiger Aesthetics, Conshohocken, PA) and most recently, Lipoderma 90% adipocytes, 10% extracellular matrix. (Frederick, MD)On the basis of this preclinical and early clinical evidence, banked adipose allografts and related investigational matrices have been applied clinically to a range of dermal and subcutaneous adipose deficits, including facial volume loss (temple, malar, and prejowl regions), dorsal hand and foot rejuvenation, plantar fat-pad restoration in diabetic neuropathy, breast and genital augmentation, vocal-fold augmentation, and post-surgical or post-traumatic contour deformities [11,12,20-25]. This Clinical Study included injections to the temples, earlobes, malar/submalar region, tear troughs, nasolabial/melolabial folds, lower one third of the face, breasts, dorsal hands, and the buttock. Two recent narrative/systematic reviews—one in Plastic and Reconstructive Surgery (Morel et al., 2025) and one in Advances in Wound Care (Mehta et al., 2025)—have summarized the emerging preclinical and clinical literature on acellular/ allograft adipose matrices and concluded that the evidence, while promising, remains limited by small sample sizes, absence of randomized comparative trials, and wide heterogeneity in volumeretention reporting (21.5-100% across studies) [26,27] Neither prior review, however, performed a formal quantitative metaanalysis restricted to human clinical outcome data using PRISMAcompliant methodology. This study introduces a human adipose allograft with a novel patented wash method which effectively removes the hematopoieticand immunogenic components while leaving the graft structure most closely resembling native adipose tissue at (90% adipocytes and 10 % extracellular matrix) [28]. This systematic review and meta-analysis subsequently undertaken to [14] comprehensively and reproducibly identify the human clinical literature on banked human adipose tissue allograft for dermal white adipose tissue replacement and volumization comparing those results to this study, [15] quantitatively pool comparable clinical outcome data across studies where possible, and [1] critically appraise the strength, consistency, and limitations of the current evidence base to inform clinical decision-making and future research priorities.

Methods

Protocol and Reporting Standard

This systematic review and meta-analysis was conducted and is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement.7 A review protocol specifying the research question, eligibility criteria, search strategy, and planned analyses was defined prospectively before data extraction.

Eligibility Criteria

Studies were eligible for qualitative synthesis if they [14] enrolled human subjects; [15] evaluated a banked, allogeneic, humantissue-bank-derived adipose tissue allograft (processed as an acellular, decellularized, minimally manipulated, or cryopreserved adipose matrix) administered by injection or surgical implantation for replacement, augmentation, or volumization of dermal or subcutaneous adipose tissue; and [1] reported at least one clinical, histologic, imaging, or patient-reported outcome. Studies were eligible for quantitative meta-analysis if they additionally reported extractable numerator/denominator or mean ± standard deviation data for a comparable outcome (responder/satisfaction rate or percentage volume retention) in a prospective cohort of five or more subjects. Exclusion criteria were: animal-only or in-vitro-only studies; studies of non-adipose allografts (e.g., acellular dermal matrix used for indications other than adipose replacement); narrative reviews, editorials, or expert-panel commentaries without original patient data; conference abstracts without sufficient methodological detail; and duplicate reports of the same patient cohort (the report with the longest follow-up or most complete dataset was retained). No restriction was placed on anatomic site, publication date, or manufacturer, provided the tissue product met the banked human adipose allograft definition above.

Search Strategy

A systematic electronic search of PubMed/MEDLINE, Embase, Cochrane CENTRAL, and Scopus was performed for records published through 30 June 2026, using combinations of the terms “allograft adipose matrix,” “decellularized adipose matrix,” “acellular adipose matrix,” “banked adipose tissue,” “human adipose allograft,” “dermal white adipose tissue,” “soft tissue volumization.” ClinicalTrials.gov was searched for registered and completed trials, and the reference lists of all included studies and of the two prior related reviews were hand-searched for additional eligible records [26,27]

Study Selection and Data Extraction

Titles and abstracts were screened against the eligibility criteria, followed by full-text review of potentially eligible records. Data extracted from each included study comprised: first author and year; study design; tissue product and processing method; anatomic site(s) treated; sample size; patient demographics; injected/ implanted volume; follow-up duration; effectiveness outcomes (volume retention, Global Aesthetic Improvement Scale [GAIS] scores, patient/investigator satisfaction, histologic adipogenesis and angiogenesis); and safety outcomes (adverse event type, frequency, and resolution). Where a study reported outcomes for more than one anatomic subsite (e.g., malar and prejowl regions in the same cohort), each subsite was treated as a separate dataset for the purposes of quantitative synthesis, as prespecified.

Risk-of-Bias Assessment

Because all identified human studies were single-arm, nonrandomized case series or pilot cohort studies, risk of bias was assessed using a modified Methodological Index For NonRandomized Studies (MINORS) framework, evaluating clarity of study aim, consecutive/prospective enrollment, adequacy of follow-up, appropriateness of outcome measurement, and prospective calculation of sample size. Given the near-universal absence of a comparator arm and the involvement of device manufacturers in study funding or authorship for the majority of included reports, the overall risk of bias across the evidence base was judged high, and this is reflected in the interpretation of pooled estimates.

Statistical Analysis

For outcomes reported as proportions (e.g., percentage of patients rated as clinical responders or satisfied/very satisfied with treatment), a random-effects meta-analysis using the DerSimonianLaird method was performed, with between-study variance weighting and inverse-variance pooling on the raw proportion scale.8 For outcomes reported as continuous means with standard deviations (percentage volumetric retention), a random-effects mean pooling model with inverse-variance weighting was similarly applied. Heterogeneity was quantified using Cochran’s Q and the I² statistic; given the small number of studies contributing to each pooled estimate, I² and 95% confidence intervals should be interpreted cautiously. All calculations were performed in Python 3.11 using standard meta-analytic formulae; a fixed-effect estimate was reported when between-study variance (τ²) was estimated at zero. Because only two to three comparable datasets were available for any single outcome, formal publication-bias testing (e.g., funnel plot asymmetry, Egger’s test) was not performed, consistent with current guidance that such tests are underpowered and potentially misleading with fewer than ten studies.

Results

Study Selection

In this Study, under the International Cell Surgical Society IRB, (ICSS-2021-011) 35 consecutive patients were treated from

08/2025-08 2026 with a hATA. There were 29 females and 6 males. Ages ranged from 21 to 78 years. Exclusion criteria included any injectable filler or bioregenerative substance within 6 months. Injection sites included: temples, earlobes, malar/ submalar region, tear troughs, nasolabial/melolabial, lower 1/3 of face, breasts, hands and buttock. Buttock patients had to consent to a randomized, ssplit-buttock protocol with CaHA injected on one side to act as a bioregenerative control, and the hATA injected on the contralateral side. The electronic search identified 412 records from databases and 14 additional records from ClinicalTrials.gov and hand-searching, for 426 records before de-duplication. After removal of 137 duplicates, 289 unique records were screened by title and abstract, of which 231 were excluded (animal-only or in-vitro-only, n = 138; non-adipose allograft, n = 41; unrelated indication, n = 33; not peer-reviewed or no extractable data, n = 19). Fifty-eight full-text articles were assessed for eligibility; 44 were excluded (animal or in-vitro only, n = 19; overlapping/ duplicate cohort, n = 6; no original human outcome data such as expert-panel commentary, n = 11; insufficient extractable data, n = 5; non-adipose allograft product, n = 3). Fourteen human studies, comprising approximately 96 patients in aggregate, were included in qualitative synthesis; four datasets derived from three prospective studies met criteria for quantitative pooling. The study-selection process is summarized in the PRISMA 2020 flow diagram (Figure 1).

Figure 1: PRISMA 2020 flow diagram of study identification, screening, eligibility assessment, and inclusion.

Characteristics of Included Studies

Included studies spanned facial aesthetic volumization (temple, malar, and prejowl regions), [21-23] abdominal softtissue histologic evaluation in the pre-abdominoplasty pannus model,6 tissue-engineered reconstruction using mechanical preconditioning, [20] vocal-fold augmentation for glottic insufficiency, [24] diabetic plantar fat-pad restoration, [11,12] aesthetic and functional contour-deformity correction, [25] and combined preclinical/human histologic characterization of the injectable matrix itself [3]. All studies used a processed, acellular or minimally cellular banked human adipose tissue allograft derived from AATB-accredited tissue banks; most (8 of 14) evaluated the commercially available allograft adipose matrix marketed as Renuva or Leneva (MTF Biologics), reflecting the concentration of the current clinical evidence base around a small number of tissue-bank-affiliated products. Study designs were uniformly single-arm, open-label, prospective pilot studies or retrospective case series; no randomized controlled trial comparing banked adipose allograft with autologous fat grafting or dermal filler was identified. (Table 1) summarizes the characteristics of the studies contributing quantitative outcome data.

Table 1: Characteristics and key quantitative outcomes of prospective human studies contributing data to quantitative synthesis. AE, adverse event; GAIS, Global Aesthetic Improvement Scale.

Quantitative Synthesis: Responder / Satisfaction Rate

Two prospective studies reporting a comparable binary responder/satisfaction outcome were pooled: the temple pilot study (7 of 10 patients satisfied/very satisfied) [21] and the malar/prejowl pilot study (investigator-assessed GAIS responders, 9 of 11 patients) [22] The random-effects (fixed-effect-equivalent, since Q < degrees of freedom) pooled responder rate was 77.2% (95% CI, 59.4-95.0%; Cochran’s Q = 0.40, df = 1, p = 0.53; I² = 0%), indicating that a majority of patients across these small pilot cohorts experienced a clinically meaningful, investigator- or patient-rated improvement following banked adipose allograft treatment (Figure 2). The absence of statistical heterogeneity should be interpreted cautiously given that only two studies contributed to this estimate. Quantificare qualitative results revealed improvements in the tone, quality and clarity of overlying skin in injection areas, Volume increase progressing from 6 and 9 months, on average with 2 sessions spaced 3 months apart. (Figures 3-5). Linear analog patient satisfaction scores at 1 year ranged from 24 %(temples) to 82% (hands)

Figure 2: Forest plot of pooled clinical responder/satisfaction rate following banked adipose allograft treatment (random-effects, DerSimonian-Laird).

Figure 3: 78-year female 10 months post 2 sessions Lipoderma spaced 3 months apart. 10 ml injected each session into temples, malar, lower 1/3, pre-jowl, and lips. Quantificare Analysis shows improvement of wrinkles, pores, evenness, and volume.