Safety profile of autologous adipose-derived stromal vascular fraction in clinical use: an exhaustive literature review

Abstract

Adipose tissue–derived stromal vascular fraction (SVF) has rapidly emerged as a promising tool in regenerative medicine due to its accessibility, autologous origin, and potent reparative potential. However, despite encouraging preclinical outcomes, the clinical safety of SVF across organ systems remains insufficiently characterized. This comprehensive review synthesizes current evidence on the safety and tolerability of autologous SVF therapy in humans, organized by target organ systems including cardiovascular, pulmonary, hepatic, renal, musculoskeletal, cutaneous, neurological, and inflammatory applications. Across studies, SVF therapy appears technically feasible and generally well tolerated, with adverse events predominantly mild and procedure-related (e.g., transient pain, swelling, or local inflammation). Serious complications — including embolism, infection, fibrosis, or tumor formation — have not been reported to date in clinical settings. However, available studies suffer from major limitations: small and heterogeneous cohorts, non-randomized designs, variability in cell preparation and dosing, and short follow-up durations. The lack of standardized isolation protocols and absence of mechanistic endpoints further restrict interpretation. Overall, current data support the short-term safety of autologous SVF therapy, while underscoring the need for large-scale randomized controlled trials, harmonized methodologies, and long-term surveillance to fully establish its risk–benefit profile.

Graphical abstract

Introduction

Autologous stromal vascular fraction (SVF) isolated from adipose tissue has attracted growing attention in regenerative medicine. This heterogeneous cell population (Fig. 1)—comprising adipose-derived stromal/stem cells (ADSCs), endothelial cells and progenitor, pericytes, immune cells, and other vascular niche components—offers a minimally invasive and readily accessible autologous source with notable reparative potential [1]. ADSCs represent the multipotent stem cell component of SVF, capable of differentiating into adipogenic, myogenic, chondrogenic, osteogenic, and neurogenic lineages. SVF is therefore considered a primary, non-expanded source of mesenchymal-like stem cells used in point-of-care regenerative applications.

Fig. 1.

A schematic diagram of the cellular composition of SVF and its relationship to adipose-derived stromal/stem cells

A major advantage is its use immediately after isolation without ex vivo expansion, enabling a streamlined, same-day, point-of-care therapeutic approach that reduces manipulation, operational complexity, cost, and risks associated with cell culture such as phenotypic drift or contamination.

SVF has been evaluated in a wide array of clinical indications, including osteoarthritis, cartilage defects, fat grafting enrichment, ischemic cardiomyopathy, chronic obstructive pulmonary disease, pulmonary fibrosis, post-COVID complications, liver cirrhosis, renal insufficiency, perianal Crohn’s fistulas, chronic wounds, systemic sclerosis, stroke, and other neurological diseases [2].

A key challenge, however, lies in clarifying the global safety and tolerability profile of autologous SVF. Safety considerations include tissue harvesting, isolation techniques, administration routes, short- and long-term monitoring, and identification of adverse events. The current literature is limited by small sample sizes, insufficient long-term follow-up, heterogeneity in cell processing, and regulatory divergences. Notably, the European Medicines Agency (EMA) has ruled that minimally manipulated SVF may fall outside the ATMP classification under certain conditions—an important distinction for clinical translation. Emerging concerns also include the possibility that SVF-derived cells may adopt pro-fibrotic phenotypes in damaged tissues rather than regenerative ones [3].

Given these uncertainties, a comprehensive synthesis of human clinical data is necessary. The purpose of this review is to evaluate the safety profile of autologous SVF across organ systems, summarize reported adverse events, identify potential risks, and highlight regulatory and methodological considerations essential for safe clinical implementation.

General mechanistic and regulatory considerations

Autologous SVF represents a highly heterogeneous and biologically dynamic cell mixture, comprising adipose-derived stromal/stem cells, endothelial and progenitor cells, pericytes, immune populations, and extracellular matrix fragments. This intrinsic heterogeneity is further influenced by inter-donor variability, anatomical harvesting sites, and technical differences in isolation methods—enzymatic versus mechanical processing, centrifugation parameters, and device-dependent protocols. As a consequence, both cellular composition and functional activity can differ substantially across preparations, creating uncertainty regarding the reproducibility and predictability of therapeutic effects.

From a mechanistic perspective, SVF is believed to act primarily through paracrine and immunomodulatory mechanisms rather than direct tissue engraftment. Nonetheless, important questions remain unresolved. Context-dependent cellular plasticity may lead SVF-derived cells to adopt fibroblastic or even pro-fibrotic phenotypes in chronically inflamed or severely damaged tissues—an effect described in preclinical cardiac models and raising theoretical safety concerns in high-risk organs. The absence of standardized mechanistic endpoints, such as in vivo biodistribution, immunologic consequences, or lineage tracking, further limits understanding of SVF’s behavior once administered.

Regulatory classifications add another layer of complexity. Jurisdictions diverge substantially in their interpretation of “minimal manipulation” and “homologous use.” The European Medicines Agency (EMA) has concluded that mechanically isolated autologous SVF, when used for certain homologous applications, may fall outside the ATMP classification, whereas enzymatic processing or non-homologous indications typically require full medicinal product regulation. In contrast, U.S. FDA guidance generally considers enzymatically isolated SVF as more than minimally manipulated, thereby necessitating an Investigational New Drug (IND) pathway. These discrepancies influence trial design, required quality controls, and post-treatment monitoring, which in turn affect the consistency and completeness of safety data across studies.

Finally, standardization remains a central challenge linking mechanistic uncertainty and regulatory diversity. Current SVF studies vary widely in processing techniques, sterility controls, quantification of viable cells, administration routes (intravenous, intra-arterial, intra-articular, or intralesional), and dosing strategies. This heterogeneity complicates comparisons across clinical trials and may obscure rare or delayed adverse events. Harmonized protocols—covering tissue harvesting, processing parameters, release criteria, dosing ranges, and standardized adverse event reporting—are essential to improve reproducibility, enable regulatory convergence, and allow a more accurate evaluation of SVF’s true safety profile across organ systems.

Organ-system based review of clinical safety data

In order to provide a consolidated overview of the current safety evidence, Table 1; Fig. 2 summarize all available clinical studies using autologous SVF across organ systems, indications, and highlighting study design, patient characteristics, and reported adverse events.

Table 1.

Summary of the main adverse events in applications involving the cardiovascular, pulmonary, hepatic, renal, musculoskeletal, cutaneous, maxillofacial, neurological and inflammatory systems

Organ / System	Indication	Route / Treatment	Study Type	Patients & Follow-up	Safety / Adverse Events	Reference
Heart	Chronic ischemic cardiomyopathy	Intramyocardial SVF	Phase I	28 patients	No AEs	Comella et al., 2016 [4]
Brain	Stroke	IV SVF	Case report	1 patient	No AEs	Michalek et al., 2019 [5]
Lungs	Idiopathic pulmonary fibrosis	Endobronchial SVF ×3	Prospective pilot	14 patients, 12 mo	No AEs	Tzouvelekis et al., 2013 [7]
	Advanced COPD	IV SVF	Prospective pilot	12 patients, 12 mo	One liposuction hematoma; no pulmonary AEs	Comella et al., 2017 [8]
	Post-COVID pulmonary fibrosis	IV SVF	Prospective case series	5 patients, 12 mo	No AEs	Carstens et al., 2024 [9]
Liver	Cirrhosis (HBV, HCV, NASH)	Intra-arterial SVF	Phase I	4 patients, 1–12 mo	No AEs	Sakai et al., 2017; 2020 [10, 11]
Kidney	CKDu	Renal artery SVF	Prospective cohort	18 patients, 36 mo	No treatment-related AEs	Carstens et al., 2023 [12]
Inflammatory / GI	Crohn’s fistulas	Local microfat + SVF	Case series / pilot	1–10 patients, up to 3 yrs	Only minor AEs	Philandrianos et al., 2018 [13]; Guillo et al., 2022 [14]
	Inflammatory disorders	Local	Narrative review	Variable	Minimal AEs	Emile et al., 2025 [15]
Systemic sclerosis	Digital ischemia	SC SVF	Pilot trial	12 patients, 6 mo	Minor AEs	Granel et al., 2015 [17]; Daumas et al., 2017 [18]
	Scleroderma (various protocols)	SC SVF ± fat graft	Systematic reviews / case series	Multiple	Minor AEs	Cao et al., 2023 [19]; Suliman et al., 2023 [20]; Perrier et al., 2025 [21]; Orlandi et al., 2025 [22]
Bone & joints	Bone regeneration	SVF vs. ADSCs	Translational	NA	No AEs	Nyberg et al., 2019 [23]
	Orthopaedic procedures	Various SVF uses	Narrative review	NA	Well tolerated	Rodriguez-Merchan, 2022 [24]
	Knee osteoarthritis	Intra-articular SVF	Phase I-II trials & cohorts	40–200 patients	Mild effusion, minor donor-site pain; no serious AEs	Hong et al., 2019 [25]; Yokota et al., 2019 [26]; Nakamura et al., 2020 [27]; Boada-Pladellorens et al., 2020 [28]; Yokota et al., 2022 [29]; Shanmugasundaram et al., 2021 [30]; Riggle et al., 2024 [31]
Cutaneous / Reconstruction	Acne	Fat graft + PRP + SVF	Observational & meta-analysis	Multiple	No major AEs	Han et al., 2023 [32]; Pensato et al., 2024 [33]; Alessandri Bonetti et al., 2024 [34]
	Breast reconstruction (CAL)	SVF-enriched fat grafting	Meta-analysis	353 patients	Minor AEs only	Li & Chen, 2021 [35]
	CAL / fat grafting	SVF or ADSCs	Reviews / meta-analyses	Variable	Minor AEs	Asimakopoulos & Anastasatos, 2022 [36]; Shimizu et al., 2025 [37]
	Facial rejuvenation	SVF-enriched lipografts	Observational	784 patients	Minor AEs	Van Dongen et al., 2019 [38]
	SVF gel	Microdroplet dermal injections	Prospective cohort	58 patients, 12 mo	Mild transient AEs	Jiang et al., 2020 [39]
	Tear trough deformity	SVF vs. fat grafting	Retrospective pooled	2,372 patients	1.5% mild AEs	Li X et al., 2025 [40]
	Chronic wounds	SVF-enriched fat	Prospective + retrospective	453 patients	Rare mild AEs	Malik et al., 2020 [41]
	Pure SVF wound injections	Intradermal/perilesional SVF	Prospective	267 patients	Minimal AEs	Lee MH et al., 2022 [42]
Maxillofacial	Craniofacial defects	SVF-enriched fat	Pilot	12 patients	Minor AEs	Bourne et al., 2021 [43]
Neurology	ALS, MS, neuropathies	IV / IT / IM SVF	Large observational	676 patients, 36 mo	4.3% mild transient AEs	Comella et al., 2017 [44]
	Neurodegenerative diseases	Intracerebroventricular SVF	Phase I	9 patients	Transient moderate AEs	Dumas et al., 2019 [45]
	Neurodegenerative diseases	Intracerebroventricular SVF	Prospective cohort	31 patients, 3 yrs	Mild transient AEs	Farmer et al., 2025 [46]

Abbreviations: IV = intravenous, SC = subcutaneous, AEs = adverse events, ADSCs = adipocyte derived stromal/stem cells

Fig. 2.

A graphical summary of organ systems studied to date and types of adverse events reported

Cardiac and cerebral Ischemia

Preliminary studies suggest that autologous SVF administration is feasible and well tolerated in ischemic disorders. Comella et al. reported no serious complications following intramyocardial SVF injections in 28 patients with ischemic cardiomyopathy [4]. In cerebral ischemia, Michalek et al. described a single intravenous SVF case with no adverse effects [5]. Despite encouraging findings, evidence remains extremely limited. Long-term risks—such as delayed vascular events, aberrant proliferation, or fibrosis—cannot be excluded. Packer et al. further highlight that mesenchymal stromal cells exposed to injured myocardium may adopt fibroblast-like phenotypes, potentially worsening myocardial fibrosis rather than contributing to regeneration [6].

Pulmonary diseases

SVF therapy has shown a favorable short-term safety profile in chronic lung diseases. Tzouvelekis et al. conducted a Phase Ib trial administering endobronchial SVF to idiopathic pulmonary fibrosis patients, reporting no serious adverse events and stabilization of pulmonary function over 12 months [7]. Similarly, intravenous SVF infusions for end-stage COPD were well tolerated without pulmonary or cardiac complications [8]. More recently, SVF infusion for post-COVID pulmonary sequelae showed no infusion-related adverse events and trends toward functional improvement [9]. Although these results indicate feasibility and short-term safety, larger randomized controlled trials with standardized protocols are needed.

Liver diseases

Sakai et al. evaluated intrahepatic arterial infusion of freshly isolated autologous SVF/ADSCs in cirrhosis patients, reporting no serious adverse events, embolic complications, or infections [10, 11]. Biomarkers remained stable or improved. While these results suggest procedural safety, invasive arterial infusion theoretically carries embolic and hemorrhagic risks. Long-term outcomes remain insufficiently documented.

Kidney pathologies

A first-in-human trial by Carstens et al. assessed intra-arterial renal infusion of autologous SVF in chronic kidney disease, reporting no serious adverse events in 18 patients [12]. Renal perfusion improved in most cases, and in patients with preserved baseline kidney function (eGFR ≥ 30), renal function stabilized over 12 months. Although encouraging, these findings require confirmation in controlled, larger-scale studies.

Inflammatory and systemic diseases

SVF-based therapies have shown promise in Crohn’s perianal fistulas, with early case reports and long-term follow-up demonstrating durable healing and a strong safety profile [13, 14]. Meta-analyses confirm these findings [15]. In systemic sclerosis, SVF injections into the fingers resulted in good tolerability and functional improvement, with long-term follow-up showing sustained benefits [17, 18]. Several systematic reviews support the safety of adipose-derived therapies in this context [19–21]. These applications highlight the potential of SVF in localized manifestations of systemic diseases.

Bone and joint disorders

Orthopedic applications have shown consistently favorable safety outcomes. SVF injections for bone regeneration and osteoarthritis rarely produce more than transient, mild adverse events such as swelling or pain [23–31]. No serious complications—including embolism, neoplasia, or severe infection—have been reported. Systematic reviews confirm a low complication rate and good overall tolerability.

Skin and soft tissue regeneration

Across dermatologic and reconstructive applications—acne scars, tear trough deformities, breast augmentation, facial rejuvenation, and chronic wounds—SVF consistently demonstrates mild, transient adverse effects (e.g., swelling, bruising) and no serious systemic complications [32–42]. These results reinforce its safety in superficial and soft tissue applications.

Maxillofacial surgery

SVF-enriched fat grafting for craniofacial reconstruction shows favorable safety, with only expected donor-site discomfort reported and no serious adverse events [43]. Limitations include small sample sizes and short follow-up.

Neurological disorders

Large cohorts treated with intravenous, intrathecal, or intramuscular SVF show a low incidence of mild, short-lived adverse events and no treatment-related serious complications [44]. Intracerebroventricular injections in Phase I trials were also well tolerated, with only transient symptoms such as low-grade fever or mild meningism [45, 46]. Medium-term follow-up demonstrates an absence of delayed complications.

Route-specific mechanistic safety considerations

By explicitly linking anticipated adverse events to route-specific physiological constraints, this section strengthens the mechanistic foundation of the safety analysis and helps contextualize the variation in observed tolerability across indications. Intra-articular injections are delivered into a non-distensible compartment with limited dispersal capacity, which may transiently increase intra-articular pressure and contribute to short-term inflammatory flares. Intravenous administration carries distinct mechanism-based risks, including the potential for micro-embolism related to cell size, pro-coagulant activity, and the pulmonary first-pass effect. Intralesional injections may induce localized inflammatory responses shaped by tissue architecture and remodeling dynamics. Finally, subcutaneous or soft-tissue delivery may be associated with fibro-adipogenic reactions or the formation of palpable nodules. Together, these route-specific considerations highlight the need for tailored administration strategies and structured monitoring frameworks to ensure the safe and consistent clinical translation of autologous SVF therapies.

Challenges, knowledge gaps, and future directions

Despite the growing clinical interest in autologous SVF therapies, several important challenges limit the consolidation of a robust safety framework. First, substantial variability persists among isolation technologies—including enzymatic, mechanical, and hybrid methods—leading to significant differences in cell yield, viability, and stromal composition that complicate cross-study comparisons. Second, the absence of validated dose–response relationships and the lack of standardized purity or potency metrics hinder the ability to define therapeutic windows and identify risk thresholds. Reporting of adverse events remains heterogeneous across studies, with inconsistent terminology, incomplete follow-up duration, and limited documentation of route-specific complications, thereby restricting reliable meta-analytic assessment. In parallel, regulatory definitions distinguishing minimally manipulated versus non-minimally manipulated SVF products are not harmonized across jurisdictions, generating uncertainty for clinical translation and study design. Looking ahead, the development of mechanistic biomarkers, standardized potency assays, and consensus-based safety reporting frameworks will be essential to advance the field. Addressing these gaps will enable higher-quality evidence, guide rational trial design, and ultimately support the safe and predictable clinical use of autologous SVF.

General limitations

Despite the overall favorable safety profile observed across clinical applications, the current body of evidence evaluating autologous SVF therapy presents several important limitations that constrain definitive conclusions. Most available studies include small sample sizes, often fewer than 20 participants, and frequently rely on uncontrolled or single-arm designs, limiting the ability to detect rare complications or establish causality. Follow-up durations are typically short—ranging from a few months to one year—insufficient to assess long-term risks such as fibrosis, ectopic tissue formation, or tumorigenicity. Considerable methodological heterogeneity also affects literature: SVF isolation protocols differ in enzymatic versus mechanical processing, centrifugation parameters, stromal cell quantification, and administration routes, making comparisons between studies difficult and potentially masking procedure-specific risks. Additionally, adverse event reporting is inconsistent, with many studies documenting only procedure-related symptoms without standardized grading or systematic surveillance. Mechanistic endpoints—such as cell phenotype characterization, biodistribution analyses, and biomarker-based safety assessments—are largely absent, impeding understanding of SVF behavior in vivo. Collectively, these limitations underscore the need for harmonized protocols, long-term follow-up, and adequately powered randomized trials to fully define the safety profile of autologous SVF in regenerative medicine.

Statement of significance

This review provides the first exhaustive, organ-specific synthesis of clinical safety data on autologous SVF therapy. It serves as a practical reference for clinicians and researchers, highlighting both the favorable short-term safety profile and the substantial knowledge gaps. Given increasing regulatory scrutiny of minimally manipulated autologous cell therapies, accumulating human data may support classification of freshly isolated SVF as a low-risk, point-of-care treatment when prepared according to standardized minimal manipulation protocols. Harmonized guidelines, dose quantification standards, and quality-control frameworks will be essential for future clinical translation.

Conclusions

Across cardiovascular, pulmonary, hepatic, renal, musculoskeletal, cutaneous, maxillofacial, neurological, and inflammatory applications, autologous SVF therapy appears feasible and well tolerated, with adverse events predominantly minor and transient. No serious or systemic complications attributable to SVF have been reported in clinical studies to date. However, most evidence comes from small, heterogeneous cohorts with limited long-term monitoring. Large, methodologically rigorous trials incorporating standardized isolation protocols, systematic adverse event reporting, mechanistic biomarkers, and extended surveillance are critically needed to fully define the long-term safety profile of autologous SVF.

Author contributions

CN & NS contribute equally to this work.

Data availability

Not applicable.

Declarations

Declarations section

The authors declare that they have not use AI-generated work in this manuscript.

Competing interests

The authors declare no competing interests.