Mechanisms of fat and soft tissue filler embolism following aesthetic injections: A cadaveric systematic review

Abstract

Background

Fat embolism (FE), fat embolism syndrome (FES), and soft tissue filler embolisms have emerged as significant complications following aesthetic procedures involving fat grafting and soft tissue fillers. Despite increasing reports of embolic events, the anatomical pathways and mechanisms remain poorly understood. We hypothesize that specific anatomical features and injection techniques are associated with increased embolism risk in aesthetic procedures. This systematic review synthesizes cadaveric studies to map vascular pathways, identify high-risk zones, identify high-risk injection instruments and techniques, and improve procedural safety.

Methods

A systematic review following PRISMA guidelines was conducted. Cadaveric studies examining fat or filler embolization pathways were included. Animal studies and those lacking full-text availability were excluded. A comprehensive search of PubMed and MEDLINE (2000–2025) was performed without language restrictions.

Results

Of 1014 identified studies, 31 met the inclusion criteria. High-risk anatomical zones included the temporal, periorbital, nasal, and gluteal regions. Key vascular structures, such as the middle temporal vein, ophthalmic artery, and deep gluteal vessels, were identified as primary conduits for embolization. Small filler volumes were sufficient to cause severe complications, including blindness, stroke, and pulmonary embolism.

Discussion

Anatomical variability significantly influences embolization risk. Study limitations include differences from in vivo conditions and variations in injection techniques. Safer approaches, such as ultrasound guidance, low-pressure injections, and blunt cannulas, are recommended to minimize vascular complications.

Other

This study received no funding.

Introduction

Fat embolism (FE) and fat embolism syndrome (FES) have traditionally been linked to orthopedic trauma and surgery. However, in recent years, fat and soft tissue filler embolisms have been increasingly recognized as severe complications following aesthetic procedures. The popularity of cosmetic interventions such as autologous fat grafting and dermal filler injections has led to a rise in reports of embolic events affecting both arterial and venous circulations, resulting in potentially devastating consequences, including skin necrosis, blindness, cerebral infarction, and pulmonary embolism.1, 2, 3 Despite the growing awareness of these complications, significant gaps remain in understanding the precise anatomical pathways and mechanisms underlying fat and filler embolization. To address these gaps, this study is guided by the hypothesis that specific anatomical features and injection techniques are associated with an increased risk of embolism in aesthetic procedures. By systematically reviewing cadaveric studies, we aim to:

• •Identify vascular pathways implicated in embolization following filler injections.
• •Identify high-risk anatomical zones for filler-induced embolism.
• •Identify high-risk injection instruments and techniques for filler-induced embolism.
• •Provide evidence-based recommendations for safer injection techniques.

Materials and methods

A systematic review was conducted to identify cadaveric studies on fat and soft tissue filler embolism, focusing on anatomical pathways and mechanisms in cosmetic procedures. The study follows the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses)⁴ Guideline.

Eligibility criteria

This study included cadaveric studies that met the following criteria:

• •Cadaveric models of fat or soft tissue embolism, addressing anatomical pathways and mechanisms of embolism following fat grafting or soft tissue filler injection.
• •Cadaveric models of artery and vein anatomical studies addressing the mechanism of embolism following fat or soft tissue filler injection.
• •Articles that lacked full-text availability online were excluded.

Information sources and search strategy

A comprehensive literature search was performed in PubMed and MEDLINE for articles published between 2000 and 2025, with no language restrictions. Studies without English-language abstracts were excluded. Search terms included ``Fat Embolism,'' ``Filler Embolism,'' ``Soft Tissue Filler Embolism,'' ``Hyaluronic Acid Embolism,'' AND ``Cadaver,'' and ``Anatomical.'' The reference lists of selected studies and reviews were also checked to identify additional relevant studies. Non-English studies with English abstracts were translated using Google Translate (Mountain View, California, USA).

Selection and data collection process

The search results were extracted and documented using EndNote® X21 software. Three authors independently screened the articles to remove duplicates. The lead author reviewed the extracted data. All discrepancies were resolved by consensus discussion in face-to-face meetings. The final set of studies represents comprehensive experimental cadaveric research on fat or soft tissue filler embolism.

Results

The search strategy identified a total of 1014 results. After deduplication, 876 articles remained. Title and abstract screening further narrowed the selection to 145 articles, and a detailed full-text assessment yielded 31 studies (Table 1) that met the inclusion criteria (Figure 1). These studies comprised cadaveric investigations, providing crucial insights into the anatomical pathways and mechanisms of fat and soft tissue filler embolism (Figure 2).

Table 1.

Studies included in this systematic review.

Number	Author	Year	Country	Sample size
1	Huang et al.5	2017	China	8 cadavers
2	Tansatit et al.6	2015	Thailand	16 cadavers
3	Yang et al.7	2015	Korea	9 Cadaver
4	Jung et al.8	2014	Korea	9 Cadaver
5	Jiang et al.9	2014	China	10 cadavers
6	Taylor et al.10	2020	Australia	27 cadavers
7	Tansatit et al.11	2015	Thailand	6 cadavers
8	Zhao et al.12	2019	China	21 cadavers
9	Cho et al.13	2019	USA	6 cadavers
10	Lee et al.14	2019	Korea	41 cadavers
11	Kim et al.15	2014	Korea/Thailand	30 cadavers
12	Wu et al.16	2017	China	12 cadavers
13	Tansatit et al.17	2017	Thailand	30 cadavers
14	Hufschmidt et al.18	2019	France	18 cadavers
15	Zhang et al.19	2019	China	18 cadavers
16	Khan et al.20	2017	USA	6 cadavers
17	Tansatit et al.21	2020	Thailand	31 cadavers
18	Tansatit et al.22	2019	Thailand	46 cadavers
19	Lee et al.23	2015	Korea/Thailand	36 cadavers
20	Tansatit et al.24	2014	Thailand	26 cadavers
21	Phumyoo et al.25	2014	Thailand	14 cadavers
22	Kelly et al.26	2006	USA	9 cadavers
23	Hong et al.27	2024	Korea	10 cadavers
24	Cong et al.28	2016	Thailand	11 cadavers
25	Ordenana et al.29	2019	USA	10 cadavers
26	Álvarez et al.30	2018	Mexico	10 cadavers
27	Ramos-Gallardo et al.31	2017	Mexico	5 cadavers
28	Wall et al.32	2018	USA	2 cadavers
29	Del Vecchio et al.33	2018	USA	8 cadavers
30	Gascoigne et al.34	2020	Australia	10 cadavers
31	Tansatit et al.35	2016	Thailand	10 cadavers

USA, United States of America.

Figure 1.

Flowchart illustrating the study selection process.

Figure 2.

A mind map visualizing the vascular pathways implicated in embolization following fat or soft tissue filler injection.

Temporal area

Temporal fat compartments and “zones of caution”

Through the dissection of 16 temples from 8 cadavers, Huang et al.⁵ identified four distinct fat compartments—the lateral temporal-cheek fat compartment (LTFC), lateral orbital fat compartment (LOFC), upper temporal compartment (UTC), and lower temporal compartment (LTC). Notably, the anterior half of the LTC was designated a high-risk “zones of caution” because of the dense distribution of sentinel veins and perforators of the middle temporal vein (MTV). These anatomical details underscore the importance of selecting a safe cannula entry point (the medial border of the hairline and temporal line) and employing a multi-plane approach to achieve aesthetic longevity and minimal risk of vascular complications.

Venous pathways and risk of embolization

Multiple studies highlight the venous drainage patterns in the temporal region that predispose patients to non-thrombotic pulmonary embolism (NTPE) or ocular complications. Tansatit et al.⁶ demonstrated, via anterograde and retrograde injections in 16 cadavers, that the MTV drains directly into the internal jugular vein, creating a conduit for embolic material to travel to the lungs. The findings suggest that accidental injection of the MTV rarely leads to ocular symptoms due to blockage of the superior or inferior ophthalmic vein. They also introduced a preventive maneuver—manual compression at the pretragal venous confluence—significantly reduced venous flow, suggesting a practical technique for minimizing venous embolization risk during temporal augmentation.

The contribution of the medial zygomaticotemporal vein (MZTV), or “sentinel vein,” was elucidated by Yang et al.⁷ in 18 hemifaces. They discovered that the MZTV frequently anastomoses with the periorbital venous system (27.8 % of cases), thus creating a potential route for filler embolism into the intracranial circulation (cavernous venous sinus). This finding, combined with the difficulty of palpating the vein (83.3 % pierce the orbicularis oculi muscle), highlights the need for slow injections and aspiration checks to reduce the risk of inadvertent complications.

Middle temporal vein (MTV) variations and safe injection zones

Further emphasizing the variability and vulnerability of the MTV, Jung et al.⁸ analyzed 18 hemifaces and found that the vein follows a path averaging 23.5 mm above the zygomatic arch at the jugale and 18.5 mm at the zygion, with a mean diameter of 5.1 mm. In 28 % of specimens, the MTV exhibited a splitting and reuniting pattern, increasing its caliber and rendering it more susceptible to injury. Because of potential connections to the cavernous sinus, inadvertent intravascular injection may lead to serious complications such as cavernous sinus thrombosis. Jung et al.⁸ recommended targeting a zone at least one finger breadth above the zygomatic arch, where no major MTV trunk was observed, and employing supraperiosteal injections with microcannulas to minimize venous perforation and subsequent embolic events.

Jiang et al.⁹ highlighted the potential for life-threatening pulmonary embolism when the material is introduced into the MTV during temporal augmentation procedures. Their analysis of 10 formalin-fixed cadavers and three clinical cases (one fatal) revealed that the MTV—originating from multiple tributaries at the lateral orbital angle—maintains a sizable caliber (2.06 to 3.02 mm) and is supported by the surrounding temporal fat, preventing its collapse during injection. This anatomical configuration increases the likelihood of direct venous entry and subsequent migration of filler or fat to the lungs. Jiang et al.⁹ thus stressed the importance of opting for blunt-tip cannulas, slowly retrograde injection, and multiple injecting sites.

Orbital and periorbital area

Evidence of arterial and venous pathways for retrograde embolization

Using lead oxide injections in 27 fresh cadavers, Taylor et al.¹⁰ revealed true anastomoses between the ophthalmic and facial arteries, as well as arteriovenous connections within the orbit and inner canthus, permitting both arterial and venous routes of retrograde filler migration. Based on these findings, three clinical patterns of blindness were identified: immediate onset with ocular pain, delayed onset (hours to a day), and late-onset (days to weeks), underscoring variable embolic mechanisms. The glabella, nasal dorsum, and inner canthi were confirmed as high-risk zones due to anchored ophthalmic arterial branches and dense venous pathways.

Tansatit et al.¹¹ used six soft-embalmed cadaver injections to confirm dye migration from the superficial temporal and facial arteries into the ophthalmic circulation, affecting both ipsilateral and contralateral globes in over half of the cases. Manual compression at the superior nasal orbital rim was shown to reduce ocular dye migration, suggesting a practical technique to lessen the risk of retrograde embolization during injections. Similarly, Zhao et al.¹² using three-dimensional computed tomography techniques assessed 21 cadavers, emphasized the variability of ophthalmic artery (OA) branching patterns and documented anastomoses between the OA, facial artery (FA), and superficial temporal artery (STA). These vascular interconnections increase the risk of retrograde embolization, posing a potential threat of blindness and stroke if filler material gains access to the OA. They advised caution when injecting within 2 cm above the superior orbital rim. Notably, Cho et al.¹³ used a fresh cadaver head perfusion model in six cadavers to confirm that retrograde embolization occurs when injection pressures surpass 166.7 mmHg, higher than typical systolic values, further stressing the importance of injecting slowly and at minimal pressures to mitigate retrograde flow into the ophthalmic circulation.

Anatomical variations in the FA and its branches further add complexity. Lee et al.¹⁴ identified the midface, particularly the nasolabial fold, marionette lines, and perinasal regions, as the most susceptible to occlusion due to smaller arterial diameters and thinner vessel walls. Their findings underscore that high-pressure injections or prominent bolus placements in these narrow arterial territories may precipitate retrograde embolization into the ophthalmic circulation, causing severe complications. Kim et al.¹⁵ detailed morphological variations of the angular artery (AA) across four distinct patterns, including instances where the AA arises directly from the ophthalmic artery or is absent, compensated by alternative branches. Because of frequent AA anastomoses with the dorsal nasal and ophthalmic arteries, injections near the medial canthal region, nasolabial fold, or tear trough can inadvertently lead to filler migration into the OA.

Specific arterial branches in the periorbital region have been repeatedly implicated in filler-induced blindness. Wu et al.¹⁶ mapped the OA and FA in 12 cadaver heads, identifying four key arteries—supratrochlear, supraorbital, dorsal nasal, and angular—that directly connect to the OA. Injection experiments in the glabella and nasal dorsum demonstrated that deep, periosteal-layer placements place fillers in immediate proximity to these vessels, increasing the likelihood of embolic events. Similarly, Tansatit et al.¹⁷ described the superior medial orbital rim (SMOR), particularly near the trochlea of the superior oblique muscle, as a critical “danger zone” where the supratrochlear and subtrochlear arteries form major external–internal carotid anastomoses. Trochlear compression proved a potentially useful maneuver to impede arterial flow during filler injections in this region. Hufschmidt et al.¹⁸ dissected the infraorbital artery (IOA). They demonstrated two high-risk areas: 1-The periosteal layer of the infraorbital hollow (tear trough) near the nasal branch of the IOA, which communicates with internal carotid branches 2-The lateral zygomatic region, where the zygomaticomalar branch’s perforators may lead to skin necrosis upon occlusion. Figure 3 demonstrates the retrograde embolism mechanism.

Figure 3.

This schematic illustrates the vascular pathways implicated in retrograde filler embolism, emphasizing high-risk anatomical zones (glabella, nasal dorsum, inner canthi, tear trough, and nasolabial fold) and their connections to the ophthalmic circulation. CRA, central retinal artery; ICA, internal carotid artery.

Minimal injection volumes required for ophthalmic artery occlusion

Multiple studies demonstrate that small filler volumes can precipitate central retinal artery occlusion (CRAO). Zhang et al.¹⁹ dissected 36 fresh cadaver hemifaces. They analyzed three clinical cases of filler-induced blindness, concluding that as little as 0.08–0.09 mL of hyaluronic acid (HA) can enter the ophthalmic artery to cause irreversible visual impairment. Similarly, Khan et al.²⁰ reported a mean supratrochlear artery volume of 0.085 mL (radius 0.72 mm, length 51.75 mm), indicating that 0.1 mL of injected filler is sufficient to provoke an embolus.

Chin, labia, and nasal areas

Chin region and the ascending mental artery (AMA)

Tansatit et al.²¹ dissected 31 cadaveric faces to investigate the ascending mental artery (AMA) location, a key vessel in chin augmentation with fillers. Notably, the dominant AMA entered the chin approximately 6 mm from the midline at a mean depth of 4.15 mm in the muscular plane. Occasional anastomoses with the sublingual artery underscore a potential risk of lingual infarction if a filler embolus travels retrograde. The authors recommended favoring midline injections, deemed safer due to fewer major vascular structures. They advocated using cannulas, pull-down tissue maneuvers, and bolus avoidance for paramedian injections to minimize arterial occlusion risk.

Expanding on this, Tansatit et al.²² employed a larger sample of 42 formaldehyde-embalmed cadavers and four soft-embalmed cadavers, coupled with ultrasonographic assessment in 10 volunteers. Most AMAs arose from the submental artery; however, 7.1 % originated from the sublingual artery, increasing the likelihood of lingual ischemia in embolization. With an average diameter of 0.7 mm, the AMA was consistently located at the mental protuberance in the muscular layer. The study reinforced that even low filler volumes may pose a threat if inadvertently injected into this vessel. To reduce complications, it advocated midline, slow, intermittent, retrograde injections and chin pulling techniques.

Lip anatomy, superior and inferior labial arteries

Lee et al.²³ examined 60 hemifaces from 36 cadavers to map the superior labial artery (SLA). They identified four major branching patterns, with the most common (Type I, 56.7 %) featuring separate SLA and alar branches. In 25 % of cases, the SLA coursed superficial to the orbicularis oris muscle, increasing susceptibility to vascular compromise if an injection occurs near the midline. On average, the artery's origin was 12.1 mm lateral to the oral commissure at a depth of 3 mm. Due to these variations, the authors advised avoiding injections within 1.5 cm of the mouth corner.

In a complementary study, Tansatit et al.²⁴ dissected 26 soft-embalmed cadavers. They demonstrated that the SLA lies an average of 4.5 mm beneath the skin, running deep between the oral mucosa and orbicularis oris muscle. In contrast, the inferior labial artery (ILA) commonly arises from a shared trunk with the labiomental artery. Superficial filler injection is considered safe along the vermilion borders of both the upper and lower lips. The entire lower lip is suitable for filler placement due to the presence of only small ascending arteries. For upper lip augmentation, filler should be placed centrally and superficially, avoiding deep injections between the muscle and mucosal layers to reduce the risk of damaging the anastomotic arch of the superior labial arteries.

Nasal and nasolabial region, facial artery, and associated anastomoses

Focusing on the nasolabial fold, Phumyoo et al.²⁵ investigated 14 soft-embalmed Thai cadavers to determine the exact course, depth, and diameter of the facial artery near the oral commissure and nasal ala. The artery was found approximately 15.3 ± 3.7 mm lateral to the oral commissure at a depth of 11.1 ± 3.1 mm and 6.7 ± 4.4 mm lateral to the nasal ala at a depth of 11.6 ± 3.7 mm. This study identifies a high-risk zone, termed the ``danger line,'' where facial artery injury is most likely during filler injections near the oral commissure and nasal ala. The most significant risk occurs 15–20 mm lateral to the oral commissure and 0–5 mm from the nasal ala, both at a depth of 10–15 mm beneath the skin.

Kelly et al.²⁶ explored the functional anastomotic relationship between the supratrochlear and facial arteries, particularly around the paranasal and medial canthal region, in nine fresh frozen cadaver heads. They confirmed a robust vascular network encompassing the supratrochlear, infraorbital, and angular arteries, thus allowing perfusion of forehead flaps even when the supratrochlear artery is ligated. From our study’s perspective, these inter-arterial connections provide a potential pathway for retrograde filler embolization, which may impact ocular and cerebral circulation through the dorsal nasal artery. Therefore, injections in the paranasal, glabellar, and medial canthal regions should be performed with extreme caution to avoid serious ocular or neurological complications.

Additional findings

Hong et al.²⁷ delineated sub-superficial musculoaponeurotic system (subSMAS) spaces—such as the subgalea-frontalis, interfascial-temporalis, prezygomatic, premaxillary, and premental compartments—that offer protective layers against vascular and nerve injuries. These compartments accommodate filler material while minimizing the risk of arterial compromise and nerve impingement.

In the forehead region, Cong et al.²⁸ classified supratrochlear and supraorbital arterial distributions into two main types, underscoring that both superficial and deep branches of these arteries supply the frontalis muscle and overlying skin. Their proposed safest entry point for forehead augmentation is below the deep branch of the supraorbital artery and lateral to the supraorbital foramen. A 23–25-gauge cannula should be used with a slow retrograde technique, injecting medially and deep to the frontalis muscle. Manual molding can be used for lateral augmentation. The metopion is a safer entry for glabellar injections due to fewer vessels. Local anesthesia is recommended, and post-injection massage helps ensure even filler distribution (Tables 2 and 3).

Table 2.

Danger zones of the face for embolism following filler injection.

Region	High-risk structures	Potential complications
Glabella	Supratrochlear & supraorbital arteries	Blindness, skin necrosis
Nasal Dorsum	Dorsal nasal artery, connections to ophthalmic circulation	Blindness, skin necrosis
Infraorbital	Infraorbital artery, risk of retrograde embolization	Ocular embolism, necrosis
Temporal	Middle temporal vein, risk of venous embolization	Venous embolism, pulmonary complications
Zygomatic	Zygomaticotemporal vein, risk of ocular migration	Retrograde ocular embolism
Chin	Ascending mental artery, risk of lingual infarction	Lingual ischemia, necrosis
Periorbital	Ophthalmic artery, direct connection to the central retinal artery	Central retinal artery occlusion, blindness

Table 3.

Procedural and other suggestions for embolism prevention in the face filler injection.

Category	Recommendations
Injection Technique	Inject slowly. Perform retrograde injection. Manual compression at the pretragal venous confluence during temporal injections. Manual trochlear compression during periorbital injections. Perform chin pulling techniques for chin injections.
Cannula vs. Needle	Blunt cannula reduces arterial perforation risk.
Aspiration	Always aspirate before injection, though not entirely reliable.
Volume and Pressure	Small aliquots, low injection pressure
Anatomical Guidance	Avoid high-risk zones.

Gluteal area

Vascular mapping and the importance of strictly subcutaneous injections

Ordenana et al.²⁹ used latex injections in 20 cadaveric gluteal regions to confirm that the deeper, medial gluteal planes contain large vessels (e.g., the inferior gluteal vein up to 13.65 mm in diameter), while the subcutaneous layer harbors smaller vessels (arteries averaging 0.9 mm). They emphasize a subcutaneous-only approach to minimize inadvertent vascular entry and strongly recommend avoiding downward angulation of the cannula, which promotes deep muscle penetration.

Injection angles and anatomical zones

Álvarez et al.³⁰ conducted a cadaveric study of 10 fresh specimens. This study evaluated the safety of different fat graft injection techniques for gluteal augmentation. Safe and effective access was achieved via the upper medial intergluteal sulcus (UMIGS) at 0° and −10° angles, and the middle buttock (MB) at a −30° lateral angle, as these approaches avoided contact with neurovascular structures and reached the intended grafting planes. In contrast, injections through the middle lower gluteal sulcus (MLGS) at all tested angles and the peritrochanteric (PT) access at 0° resulted in complications due to proximity to critical anatomical structures. Clinicians should avoid MLGS and PT approaches and instead use UMIGS (excluding −30°) or MB (−30°) techniques to minimize risk and ensure safe fat placement.

In another investigation, Ramos-Gallardo et al.³¹ corroborated that deep gluteal fat injections significantly increase the risk of intravascular contact. Their cadaver dissections identified a 45° injection angle as more likely to reach the superior gluteal artery and sciatic nerve than a 30° angle that largely confined fat within safer zones. These authors pinpoint quadrants 1 and 3 of the gluteal region as preferable sites, advising practitioners to avoid deep, steeply angled trajectories.

Role of the gluteus maximus fascia and avoidance of intramuscular placement

Wall et al.³² investigated cadaveric hemibuttocks under varied injection conditions, including different fascial states (intact, perforated, fenestrated) and expansion vibration lipofilling (EVL). Fat remained in the subcutaneous space despite high pressures when the fascia was intact or only perforated. However, when larger fascia defects were present, fat migrated into the submuscular space. These results confirm that the gluteus maximus fascia is a substantial barrier preventing deep fat migration unless compromised. The study concludes that all fat injections should be performed exclusively in the subcutaneous tissue to avoid dangerous intramuscular fat migration, recommending a new standard of care against subfascial or intramuscular injections.

Del Vecchio et al.³³ further emphasized that no portion of the gluteus maximus muscle is safe for fat deposition. Their dissections demonstrated a phenomenon termed ``deep intramuscular migration'' (DIM), where fat introduced into the muscle could track along intermuscular planes into deeper neurovascular structures. These results confirm that even superficial intramuscular injections carry a risk of venous entry.

Additional insights from other cadaveric studies

Gascoigne et al.³⁴ used a cadaveric perfusion model in bilateral upper limbs to examine whether increasing perfusion pressure would dilate perforator arteries or transform choke anastomoses into larger, more direct anastomoses. These results have broader implications for understanding flap perfusion, suggesting that acute increases in perfusion pressure alone cannot remodel vascular architecture. Extrapolated to aesthetic fat grafting, it indicates that simply increasing injection pressure does not necessarily create safer ``potential channels'' but can instead encourage undesired retrograde flow or venous backfill, mechanisms also implicated in embolic events.

Tansatit et al.³⁵ examined the capacity of cannulas to perforate arterial walls under various angles and conditions, challenging the notion that blunt cannulas are uniformly safe. They discovered that arterial perforation can occur when the cannula tip aligns perpendicularly to a fixed artery, especially at bifurcations or in tortuous vascular segments. Conversely, parallel cannula orientation minimized the risk of arterial trauma.

In summary, cadaveric studies confirm that the gluteal subcutaneous plane contains smaller vessels and is safer for fat grafting. In contrast, deeper planes house larger veins like the inferior gluteal vein (up to 13.65 mm), increasing embolism risk. Angles of 0° to 30° and upper medial or mid-buttock access avoid critical structures and intramuscular fat migration. Intramuscular or steep-angle injections (e.g., 45°) risk contacting the superior gluteal artery and sciatic nerve. Intact gluteus fascia prevents fat from reaching deeper planes, even under high pressure. Blunt cannulas reduce but do not eliminate perforation risk—gentle, parallel orientation and small aliquots are essential (Tables 4 and 5).

Table 4.

Danger zones of the gluteal for embolism following filler injection.

Region	High-risk structures	Potential complications
Upper Medial	Superior gluteal artery, sciatic nerve proximity	Gluteal necrosis, fat embolism, nerve damage
Lower Medial	Inferior gluteal vein, high venous caliber	Massive venous embolism, pulmonary fat embolism
Upper Lateral	Deep intramuscular vessels, risk of venous entry	Fat embolism, deep tissue necrosis
Lower Lateral	Gluteal perforators, risk of migration	Gluteal ischemia, vascular occlusion
Intergluteal Sulcus	Risk of deep penetration and vascular compromise	Deep penetration leading to systemic embolization

Table 5.

Procedural and other suggestions for embolism prevention in gluteal filler injection.

Category	Recommendations
Injection plane	Strictly subcutaneous, avoid intramuscular planes.
Injection angle	Use 0° to −10° angle, minimize deep penetration
Aspiration	Aspirate before injecting, though limitations exist.
Cannula choice	Use a blunt-tip cannula; larger gauge preferred.
Volume management	Inject small aliquots, avoid overfilling.
Anatomical guidance	Consider ultrasound guidance, avoid high-risk zones.

Discussion

The results of this review align with existing evidence on the risks associated with fat and soft tissue filler embolism. The findings reinforce the critical importance of strict adherence to safe injection techniques, particularly in high-risk anatomical regions. Tillo et al.³⁶ and Del Vecchio et al.³⁷ highlight the necessity of subcutaneous-only fat placement in gluteal fat grafting to minimize the risk of pulmonary fat embolism, a key recommendation echoed in our study. Jorge et al.³⁸ employed cone-beam CT to map gluteal vasculature in 32 patients, demonstrating that intramuscular vessels are significantly larger in diameter (median 1.47 mm) and more densely distributed than subcutaneous vessels (median 1.09 mm). Similarly, Turin et al.³⁹ demonstrate that the superior and inferior gluteal veins pose an inherent embolic risk when deep intramuscular injections are performed, further validating our conclusions on the importance of procedural caution. A recent systematic review offers a comprehensive analysis of fat embolism following fat grafting, consolidating reported cases and evaluating patient demographics, procedural characteristics, clinical features, and outcomes. Their findings indicate that ocular manifestations were the most common presentation (50.8 %), followed by neurological (42.2 %) and cardiopulmonary symptoms (32.0 %); these findings are consistent with our review, which underscores the vulnerability of the ophthalmic circulation to embolic events, particularly following facial filler injections in high-risk areas such as the glabella, nasal dorsum, and temporal region. The study further highlights the high mortality rate (34.3 %) associated with fat embolism syndrome, emphasizing the need for improved safety protocols in aesthetic procedures.³

Our study supports the recommendations made by Villanueva et al.⁴⁰ and Cansancao et al.⁴¹ advocating for real-time ultrasound guidance in gluteal fat grafting to ensure precise cannula placement and minimize vascular injury. A retrospective study involving 1815 patients demonstrated that ultrasound-guided gluteal fat grafting significantly reduces complications, with no reported fat embolism or mortality, underscoring its efficacy in enhancing procedural safety.⁴² In facial procedures, Kapoor et al.⁴³ and Beleznay et al.⁴⁴ emphasize that understanding vascular anatomy and modifying injection techniques can reduce the risk of severe complications such as blindness. These findings highlight that ultrasound enables real-time visualization of vascular structures, reducing the risk of intravascular injection; however, its adoption is limited by cost, equipment accessibility, and the steep learning curve requiring formal training.

While this review provides valuable insights, several limitations must be acknowledged. Although rich in anatomical detail, Cadaveric studies do not fully replicate the dynamic physiological conditions of living subjects, limiting generalizability. Variations in study design, sample size, and methodology introduce heterogeneity, and despite an extensive search strategy, some relevant studies may have been missed due to publication bias or database limitations. Additionally, our review did not comprehensively address how demographic factors—such as age, sex, ethnicity, and body habitus—may influence vascular anatomy and embolism risk. These population-specific differences underscore the need for individualized injection planning. While blunt cannulas are recommended to reduce vascular injury, they are not without limitations. High-pressure injections can still cause inadvertent arterial perforation, and navigating dense tissue planes may be challenging. Moreover, the perceived safety of cannulas may lead to operator overconfidence, emphasizing the importance of technique optimization and personalized approaches to enhance procedural safety.

Future research should focus on prospective clinical trials to validate these findings in real-world settings. Further studies using advanced imaging modalities, such as 3D-computed tomography,⁴⁵ may enhance our understanding of vascular variations and their implications for filler embolization. Finally, interdisciplinary collaboration between plastic surgeons, radiologists, and anesthesiologists will be crucial in developing comprehensive safety protocols to minimize embolic risks across various cosmetic procedures.

Ethical approval

Not required.

Conflicts of interest

All authors declare that they have no competing interests.