Safety of infraorbital hyaluronic acid injections: Outcomes of a meta‐analysis on prospective clinical trials

Hollowing of the infraorbital region represents a common concern among aesthetic patients. In the past decade, an increasing number of patients have resorted to noninvasive aesthetic procedures to treat these concerns. The objective of this study was to evaluate the safety profile of infraorbital hyaluronic acid injections for aesthetic rejuvenation.


| INTRODUC TI ON
The infraorbital region can be defined as the under-eye area which originates at the medial canthus and extends laterally to the zygomatic arch. 1 The lower orbital rim acts as its superior border and the malar prominence as its inferior border. 1 The infraorbital hollow, also known as the infraorbital sulcus, is divided into medial (tear trough) and lateral (palpebromalar) portions. 1,2 Hollowing of the infraorbital region represents a common concern among aesthetic patients, as a sunken appearance of the under-eye area can contribute to a fatigued appearance. Structural changes associated with the aging process contribute to infraorbital hollowing, including increased skin laxity, reduced soft tissue volume, changes in the orientation of retaining ligaments, and resorption of bone. 3 In the past decade, an increasing number of patients have resorted to non-invasive aesthetic procedures to treat these concerns. 4 The American Society for Aesthetic Plastic Surgery reported a 30% increase in hyaluronic acid (HA) injections from 2012 to 2021; and has consistently ranked these treatments as the second most popular non-surgical procedure. 5 Correspondingly, the use of HA injections has become a central strategy for the rejuvenation of the infraorbital area. 4,6 The treatment indication is generally based on a careful assessment of bone, local intra, and extra-orbital soft tissue, as well as muscle and overlying skin, to restore volume loss, camouflage volume transposition, lift, tighten, and/or improve skin quality. 3 Many factors render the infraorbital region a complex site for injecting HA. For example, irregularities (e.g., bumps, lumps) resulting from erroneously placed HA can be highlighted by the thin periorbital skin, the region's fat distribution, and the dynamic musculature of the area. 3 Furthermore, the compartmentalization of the area by various ligaments increases the risk of edema, and the presence of large critical vessels (angular artery and vein, infraorbital artery, zygomaticofacial artery, etc…) increases the risks of vascular incidents related to the procedure. 2,3 It has been proposed that the selection of the appropriate injection technique can measurably limit these adverse events (AEs). For example, using a blunt cannula or a sharp needle may be associated with different distributions and/or incidence rates of AEs. 2,7 Therefore, the objective of this study was to evaluate the safety profile of infraorbital HA for aesthetic rejuvenation, through a systematic review and meta-analysis of prospective clinical trials.

| Research question
Does the use of needle versus cannula during infraorbital HA injections result in the same incidence rate (%) of specific AEs, notably ecchymosis, and edema?

| ME THODS
The protocol of this systematic review and meta-analysis was registered in the PROSPERO database (ID number: 328232). The systematic review was performed in accordance with the recommendations listed in the Cochrane Handbook for Systematic Reviews of Interventions. The findings are reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. 8 The review process (study selection, data extraction, risk of bias, and quality of evidence assessments) was undertaken using Excel (Version 2203, Microsoft Corporation). 9

| Eligibility criteria
Preliminary scoping of the literature revealed an insufficient number of randomized controlled trials (RCTs) available to perform a traditional meta-analysis. This low incidence of RCT is in and of itself concerning. All prospective clinical trials which assessed AEs associated with infraorbital HA used for aesthetic indications were eligible for inclusion (e.g., RCTs, cohort studies, and case-controlled studies). Given that retrospective trials typically capture data less rigorously than prospective trials, authors often report no rare events in these studies. 10 As such, retrospective studies were excluded from the analysis. Furthermore, manuscripts including cadaveric studies, animal interventions, retrospective chart reviews, literature reviews, consensus papers, guidelines, letters, and comments or editorials were excluded. However, the bibliographies of these manuscripts (e.g., reviews, consensus papers, and guidelines) were searched manually to identify further eligible studies. Studies that allowed concomitant treatments (e.g., facials, chemical peels, lasers, injectables other than HA [e.g., polyl-lactic acid and calcium hydroxylapatite]) during the trial, other than rescue medication, such as hyaluronidase or aspirin, were also excluded.
There were no limitations applied to the demographic profiles of study participants in terms of gender or ethnicity. However, the patient population(s) had to include healthy adults (aged ≥18 years).
Studies including subject populations with major morbidities (e.g., HIV) were excluded. Only studies investigating HA use for agerelated soft tissue changes in the infraorbital region were eligible (e.g., studies investigating other indications, such as anophthalmic volume deficient sockets, phthisic eyes, cicatricial ectropion, and lower eyelid retraction, were excluded). There were no limitations to the reporting source of AE information (clinician, nurse, research assistant, and subject) or study setting (e.g., hospital and private clinic). Each database was searched from inception to December 2021.

| Data sources and search strategies
Searches were conducted using a combination of study design-, treatment-, and disease-specific keywords and/or Medical Subject Heading (MeSH) terms. While AEs were the outcomes of interest for this review, the use of outcome-specific keywords was avoided in the search strategies due to the possibility that a study investigating the efficacy of an HA filler may not mention terms related to AEs in the title, abstract, or keywords. The search was limited to English language publications and to human subjects. The availability of individual patient data was not anticipated, nor sought. Detailed search strategies for PubMed, MEDLINE, Scopus, and CENTRAL databases are reported in Table 1.
The bibliographies of manuscripts meeting eligibility criteria were searched manually to identify further studies that may have been eligible. The website clini caltr ials.gov was consulted, to identify unpublished studies. Searches were re-run prior to the final analysis.

| Study selection and data extraction
Two members of the review team independently evaluated each title and abstract to exclude irrelevant studies, according to the predefined eligibility criteria. At this step, the criteria for reporting AE information were not considered for selection, as studies focusing on the efficacy of treatment may not report data about AEs in the title or abstract. After this first step, the two reviewers independently reviewed the full texts of the articles not excluded during the initial screening stage to determine whether the studies met all selection criteria. Those who did not meet these criteria were excluded. All differences of opinion regarding the selection of articles were resolved through discussion and consensus between the two reviewers, and any persistent disagreement was solved with the in-

| Outcomes of interest
The primary outcomes of interest were the incidence rates (%) of ecchymosis and edema in subject groups treated with a needle or cannula.

| Statistical analyses
Meta regression analyses using generalized linear models with mixed effects 11 were used to compare the incidence rate of ecchymosis and edema in the needle versus canula groups. The models included variables for RoB and quality of research (i.e., NIH assessment tool) as covariates. A sensitivity analysis was conducted by excluding studies in which zero percent of the outcome was reported or data was unknown. All analyses were performed using SPSS Statistics, Version 28.0 (IBM Corp.). 12

| Ecchymosis
Results of a meta-regression analysis comparing incidence rates of ecchymosis in subjects treated with either needle or cannula, during HA injections in the infraorbital region, revealed a significant group effect (  (Table 3).

| Edema
Results of a meta-regression analysis comparing incidence rates of edema in subjects treated with either needle or cannula, during HA injections in the infraorbital region, revealed a significant group effect ( error: 0.010). The results of the sensitivity analyses supported these findings (Table 4).

| DISCUSS ION
Previously, the comparative safety of injection-related devices (i.e., needle and cannula) has not been convincingly demonstrated in clinical trials. In the present evaluation, a comprehensive literature search was conducted using optimized database combinations, 13 and adequately powered meta-analyses were used to supplement primary data. It was revealed that the incidence rate of ecchymosis following infraorbital injections using HA fillers ranged from 0.016% to 0.228%, depending on whether a cannula or needle was used, respectively. It was also revealed that the incidence rate of edema ranged from 0.217% to 0.547%, depending on whether a needle or cannula was used, respectively. Certain observations can be made from these findings. First, edema is a more common AE compared to ecchymosis in the infraorbital region, regardless of the device selected (needle or cannula). Given that edema is the primary AE associated with infraorbital injections of HA, it can be suggested that any treatment recommendations should focus on reducing its incidence rate. Moreover, given that the infraorbital region can be susceptible to chronic and recurrent edema or "puffiness" prior to HA injections (e.g., genetics, fluid retention, allergies, fatigue, and diet), attempts to avoid aggravating pre-existing conditions by inducing any injection-related edema should be considered in an individual's treatment plan. Post-treatment, mild edema can also not be disguised with the use of camouflaging agents, as can ecchymosis. 14

| Influencing factors
In general terms, a complete evaluation of the infraorbital and sur- The infraorbital region consists of the thinnest skin on the face, and as such, fluid shifts (e.g., due to menstrual cycle or diet) would be prone to worsening temporary AEs. With respect to anatomy, the ability to inject deep into the bone would minimize edema, as the muscle would camouflage the product. However, in comparison to deep injections, placing the product superficially may lead to more edema. Another consideration is whether the patient presents with an expanded venous plexus, as this may increase the risk of bleeding regardless of whether a needle or cannula is used. For example, the use of a needle requires more punctures than a cannula, but a TA B L E 3 Results of a meta-regression analysis comparing incidence rates of ecchymosis in subjects treated with either needle or cannula, during hyaluronic acid injections in the infraorbital region.

TA B L E 4
Results of a meta-regression analysis comparing incidence rates of edema in subjects treated with either needle or cannula, during hyaluronic acid injections in the infraorbital region.

| CON CLUS IONS
The incidence rates of AEs following the administration of HA injections in the infraorbital region vary depending on whether a needle or cannula is used; with needles being associated with a greater risk of ecchymosis and cannulas being associated with a greater risk of edema. These findings should be discussed with patients prior to treatment consultation. Finally, as with most techniques, it is usually prudent to develop expertise with one technique before using a second, especially in cases where both approaches can be used and have different adverse event profiles.

FU N D I N G I N FO R M ATI O N
The authors received no financial support for the research, authorship, and publication of this article.

CO N FLI C T O F I NTER E S T S TATEM ENT
The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.

E TH I C S S TATEM ENT
This article does not contain the results of any primary studies involving human or animal subjects. All analyses were conducted on datasets containing publically available information.