Fangfei Nie,Hongbin Xie

Department of Plastic Surgery,Peking University Third Hospital,Beijing 100191,China

Keywords:Cosmetic filler Polymethyl methacrylate Local massage Intra-arterial injection Embolism Necrosis

ABSTRACT Background: Massage is generally believed to be a simple and effective method for preventing necrosis when intravascular injection emboli are suspected.Due to its good dispersive properties,polymethyl methacrylate(PMMA) was used as the test filler.The main purpose of this study was to observe whether local massage as a simple remedial measure can promote the diffusion of filler and reduce the necrosis rate for intra-arterial embolism.Methods: Rabbit ears with the central ear artery (CEA) main trunk totally or segmentally obstructed 5 min after PMMA injection were studied.In order to simulate the massage effect of the fingers,the massage group was treated with a small beauty massage bar with a fixed vibration frequency along the direction of the blood flow.CEA blood flow and skin lesions were also analyzed.Results:The baseline data were similar between the control and massage groups.Compared to the control group,the recanalization rate of the CEA trunk in the treatment group increased significantly after massage.However,there was no significant difference in the visible rate of transparent embolus on the 1st day after treatment,nor in the necrosis degree or area of soft tissue damage on the 7th day after treatment.Further analysis showed that massage tended to have a positive effect on segmentally obstructed cases at 5 min after injection but a negative effect on totally obstructed ears at 5 min.Conclusion: Local massage cannot reduce the complications of tissue necrosis after intra-arterial PMMA injection.Prevention is key to reducing complications.

1.Introduction

In recent years,there has been a rapid development of modern and minimally invasive injections of cosmetic fillers,with a corresponding increase in complications.The most serious complication is tissue necrosis after injection.Existing research data show that the main cause of tissue necrosis is unintentional injection into the artery,resulting in distal arteriolar embolism and microcirculation disturbance.Once vascular embolism is suspected,there is no clear method for managing different types of fillers.1

There are many expert consensus recommendations for the treatment of soft tissue filler complications.1,2Arterial compromise is typically accompanied by immediate blanching and severe pain.If recognized during the treatment session,the injection should be stopped immediately,and an attempt should be made to aspirate the product.The recommended therapies include immediate administration of a warm gauze to tap the area,local massage,topical nitroglycerin paste,and nitroglycerin sublingual tablets to facilitate vasodilatation.1,3Although local massage has been reported to be informally effective in treating impending artecoll necrosis,these are only personal experience reports.Meanwhile,some scholars believe that hot or cold compression is effective.To verify the effectiveness of these empirical methods further,an animal experiment was designed.This study focused on the effects of massage.

Polymethyl methacrylate (PMMA,representative product:Artecoll)microspheres are widely used as fillers.They are composed of 80%collagen and 20% PMMA and have a diameter of 30–40 μm.Compared with the most widely used dermal filler hyaluronic acid gel (HA,representative product:Restylane),Artecoll is considered to have good dispersion characteristics in blood flow.4Therefore,we speculate that local massage will significantly promote its dissemination in blood vessels to alleviate embolism and reduce tissue damage,and if PMMA-induced arterial embolisms cannot benefit from local massage,it is even less likely to affect HA-induced arterial embolisms.To verify the effectiveness of massage on vascular recanalization and to decrease the risk of soft tissue necrosis,an animal model was used in this study.These results provide experimental support for this empirical application in clinical practice.

2.Materials and methods

2.1.Animal and tested products

Healthy male white rabbits (2.6–3.5 kg,aged approximately 1 year)were purchased from the experimental animal center of Peking University Health Science Center.The rabbits were fed typical pellets,with food and water providedad libitum,and sub-cages were reared at 22 ± 2°C and at a humidity of 60% ± 10% in a clean animal laboratory.All procedures used in this study were approved by the Experimental Animal Ethics Committee of the regional hospital(approval no.A2019010).All applicable international and national guidelines for the care and use of animals were followed.Artecoll (Hafod B.V.,Rotterdam,the Netherlands)was supplied by the manufacturer.

2.2.Animal experiments

Animal models of intravascular embolization were constructed as previously reported.4Briefly,after shaving the local hair,compound lidocaine cream(Tongfang Pharmaceutical Group Co.,Ltd.,China)was applied to the local injection site 30–60 min before injection.Next,0.2 mL of Artecoll was then injected into the main trunk of the central ear artery (CEA) from the proximal to distal direction with a 27-G needle (0.4 mm × 20 mm,manufactured by Terumo Europe N.V.,Belgium) at a speed of 0.2 mL/50 s.

The rabbits were numbered according to their cage numbers.Sixty rabbit ears with their CEA main trunks totally or segmentally obstructed at 5 min postoperatively were randomly divided into two groups:massage and control groups.The massage group was treated with a small beauty massage bar with a fixed vibration frequency of 100 Hz for 5 min from the proximal to the distal end of the CEA main trunk.At the same time,a certain degree of pressure was applied to simulate the massage effect of the fingers to promote the diffusion of transparent emboli in blood vessels to the distal end of the circulation.Because rabbit ear skin is thin,a water bag was used as a pad to simulate human skin tissue.The ventral side of the ear was padded,whereas the dorsal side was massaged(Fig.1).The injections and massages were performed by the same senior plastic surgeon.

2.3.Observation of CEA blood flow and skin necrosis

Blood flow and skin necrosis were observed and recorded using a camera(Canon 70D with a 24–105 lens,Japan)with a strong flashlight.Recordings were taken before injection,5 min,10 min (5 min after massage/no massage),1 day,and 7 days after injection.Countercurrent rates were evaluated in each group according to if the filler flowed reversely from distal to proximal direction during injection.Obstruction rates of CEA were calculated according to the rate of cases when the main trunks of the CEA were totally or segmentally obstructed in comparison to the total cases.The blood flow status was judged mainly according to the visible transparency or blood embolus in the vessels under strong light.On the 7th day,different skin lesion grades were classified,and the severity of necrosis was evaluated by the proportion of these grades in each group.The hematoma area at 10 min and the necrotic area on the 7th day were measured using ImageJ software(NIH,Bethesda,MD,USA)(Fig.2).

Fig.1.The obstructed CEA was treated with a small beauty massage bar from the proximal to the distal end of the CEA main trunk.A padded water bag was placed on the ventral side of the ear while the dorsal side was massaged.CEA,central ear artery.

2.4.Histological examination

The rabbits were euthanized by carbon dioxide euthanasia on the 7th day after intra-arterial injection of PMMA.Rabbit ear skin specimens were fixed with 4% neutral formaldehyde,dehydrated with a conventional gradient of ethanol,and embedded in paraffin.Routine sections were prepared for hematoxylin and eosin staining to observe the vascular emboli and verify the degree of tissue damage.

2.5.Statistical analysis

SPSS 16.0 software package(IBM Corp.Armonk,NY,USA)was used for all the statistical analyses.The median (quartile) [M (Q1,Q3)] was used to measure skewed distributions,and the rate (%) was used for counting data.The measurements of each group were compared and analyzed using independent sample nonparametric and chi-square tests.The significance level for comparisons was set at 0.05.

Fig.2.The main observation time points of CEA blood flow in rabbit ears.Initiation of injection was considered to be day 0.

3.Results

3.1.No significant benefits were found in the massage group

A total of 78 ears from 39 rabbits were used in this experiment.However,the injections failed in 13 ears,and in 5 ears,the blood flow of the CEA main trunk was completely restored within 5 min after injection.A total of 60 ears meeting the inclusion criteria(ears with their CEA main trunk totally or segmentally obstructed at 5 min postoperatively) were randomly divided into two groups according to a random number table.

On day 0,the two groups had the same average body weight(3.06 ± 0.24 kg) and the same obstruction rates in the CEA main trunks within the first 5 min after injection (21/30 cases of total obstruction and 9/30 cases of segmental obstruction,Fig.3B).There were also no significant differences in the countercurrent rates during the injection (P=0.554,Fig.3A).These data suggested that the baseline situation was similar between the two groups.

Fig.3.Comparisons between baseline figures of the control and massage groups.(A) Countercurrent rate during the injection.(B) Main CEA trunk obstruction rate within 5 min after filler injection.N=30/group.

At 10 min,there were three main blood flow states:no occlusion,segmental occlusion,and total occlusion.In the latter two cases,an obvious hematoma was observed at the injection point (Fig.4A).The CEA trunk obstruction rate significantly decreased (P=0.000,Fig.4B),and the hematoma area correspondingly increased in the massage group(P=0.001,Fig.4C).At the same time,we noticed that in the control group,there was also a decrease in the obstruction rate at 10 min compared to that at 5 min(Figs.3B and 4B),but there was no significant difference(P=0.300).

Fig.4.Comparisons of the control and massage groups at 10 min.(A) CEA trunk recanalization grades at 10 min.The arrow indicates the CEA main trunk.No:No obstruction in blood flow in the main trunk of the CEA;segmental:segmentally obstructed;total:totally obstructed.(B)CEA main trunk obstruction rate at 10 min;the obstruction rate was significantly decreased in the massage group.(C) The hematoma area increased significantly in the massage group at 10 min.＊＊P＜0.01.

On day 1 post-injection,the blood flow in most of the ears remained unchanged.There was no significant difference between the two groups in the rate of visible transparent embolus at 1 day (20% in the control group and 26.67%in the treatment group,P=0.784).

On day 7,different degrees of skin and soft-tissue lesions were observed.“None”refers to no skin damage or blood flow obstruction,and“mild”refers to when there was only darkening of the tissue or abnormal blood flow.Any skin lesions with unilateral or bilateral ulcer/necrosis were referred to as“moderate”or“severe”,respectively (Fig.5A).Corresponding histopathological changes were observed(Fig.5B).

Fig.5.Different degrees of skin and soft tissue lesion on day 7.(A) Gross observations under strong light (the arrow points to the skin lesions).Mild:No ulcer was observed,and only blood flow was obscured;moderate:lesions with unilateral ulcers,which is on the dorsal side of this case;severe:bilateral skin obvious necrosis,appears transparent under strong light.(B)Histopathological changes of different degrees of skin and soft tissue lesions.In a typical mild necrosis case,dilated blood vessels with microspheres inside could be seen,as indicated by the arrow,and skin damage was not obvious.In moderate necrosis cases,unilateral epidermal necrosis was visible(the arrow points to the necrotic part of the skin).In the severe necrosis lesions,skin full-thickness necrosis without structure could be seen.Bar,500 μm.

Statistically,there was no significant difference in the necrosis severity ratio (P=0.839,Fig.6A) or the skin lesion area at 7 days(P=0.527,Fig.6B).

Fig.6.Comparisons of the control and massage groups on day 7.(A) Incidence of different skin necrosis grades.(B) Skin lesion area.

3.2.There was a negative benefit of massage on totally obstructed ears,while minor benefits were found in the segmentally obstructed cases

Based on the blood flow status of the CEA main trunk at 5 min,the ears were divided into four subgroups for statistical analysis(Ct,control:totally obstructed;Cs,control:segmentally obstructed;Mt,massage:totally obstructed;and Ms,massage:segmentally obstructed).There were 21 cases in the Ct and Mt groups and 9 cases in the Cs and Ms groups(Fig.3B).There were no significant differences in body weight or countercurrent rate during injection among the four subgroups.

At 10 min,the CEA main trunk obstruction rates were still significantly different between the Ct and Mt subgroups (the Mt group had a lower obstruction rate than the Ct group,P=0.001)and between the Cs and Ms subgroups(the Ms group had a lower obstruction rate than the Cs group,P=0.000)(Fig.7A).There was no significant difference in the rate of clear embolus formation among the four groups on day 1.The difference between the Mt (38.10%) and Ms (0%) groups was the largest(P=0.067,Fig.7B).According to the necrosis rate and the median value of the necrotic area,the severity of necrosis decreased in the order of Mt,Ct,Cs,and Ms.Compared with the Mt group,the necrosis degree of the Ms subgroup on the 7th day was significantly reduced (P=0.019,Fig.7C).In terms of the necrosis area,the differences between the Ct and Cs (P=0.035) and Mt and Ms (P=0.007) groups were both significant(Fig.7D).

Fig.7.Subgroup analysis between totally and segmentally obstructed cases at 5 min after injection.(A) Obstruction rate at 10 min.(B) Transparent embolus formation rates on day 1.(C) Necrosis severity ratio on day 7.(D) Skin lesion area on day 7.The four subgroups were:Ct,control:totally obstructed;Cs,control:segmentally obstructed;Mt,massage:totally obstructed;Ms,massage:segmentally obstructed.＊P＜0.05,＊＊P＜0.01.

4.Discussion

Arterial occlusion is typified by immediate,severe,and disproportionate pain and color changes.5Because intravascular occlusions are rare,present recommendations for prevention and management are based almost exclusively on expert opinion and consensus reports.6–10Treatment strategies include hyaluronidase if HA is injected,warm compression,massaging,or tapping the area,and applying 2% nitroglycerin paste to promote vasodilatation and facilitate blood flow in the affected area.1,3,5Digital massage11is recommended to treat occlusion of the central retinal artery(CRA).It is important to note that ocular massage is not currently supported by high-quality evidence for treating CRA occlusion,with only a few successful case reports published.In addition,the method of ocular massage may vary greatly among ophthalmologists and requires evaluation in larger studies.Other reports in the literature suggest that acetylsalicylic acid is useful for emboli,but its efficacy and that of other antiaggregants have not been proven for filler-related retinal vascular complications.12Therefore,it is still a difficult clinical problem to manage embolism complications caused by filler entering the blood vessel by mistake.

Some researchers report that rare complications such as arterial occlusion and retinal vascular occlusion were not associated with a particular type of dermal filler.13Similar to HA-induced arterial embolism,PMMA-induced arterial embolism is rare;however,the consequences are serious.2Because of its good dispersive properties,we used Artecoll as the research product to explore whether simple local massage can promote dispersion and reduce skin necrosis after the filler enters the artery.If massage cannot benefit PMMA-induced arterial embolism,it is easy to speculate that massage may not play a positive role in HA-induced arterial embolism.

In this study,the results showed that although local massage at 5 min after injection could provide an immediate increase in the main trunk recanalization rate of occluded blood vessels,it did not significantly reduce the 1-day transparent thrombus and 7-day skin necrosis rates.The reasons for this may include:(1)the massage starts at the root of the ear,and if the intravascular filler travels to the proximal end of the bloodstream,massage may aggravate the reverse countercurrent and obstruct the blood vessels;(2)it only reduces the amount of filler in the trunk and does not reduce diffusion in the smaller blood vessels or capillaries.In the worst case,local massage promotes the concentration of the filler in smaller blood vessels without promoting its dispersion into the blood flow.Therefore,it did not achieve the effect of significantly reducing the necrosis rate.

These results are similar to those of another study on the application of the vasodilator nitroglycerin paste.14In an animal model with fillers injected intra-arterially,the filler caused decreased vascular perfusion.However,no statistically significant improvement in perfusion was noted after the topical application of nitroglycerin paste.In addition,the skin of the rabbit ear post-application appeared to have a more congested appearance than the controls.

It is generally agreed that massage can spread fillers,decrease the pressure on the vessel wall,and improve the microcirculation.15In this study,the blood flow state in the CEA main trunks was improved significantly a short time after massage.However,it did not significantly improve microcirculation on day 1 or reduce the skin necrosis rate on day 7.This is similar to some atypical cases of delayed vascular complications over 24 h observed clinically.16In that study,the authors concluded that the fillers inside the artery may have traveled over time and reached a distal terminal branch,where they generated localized skin damage and pain.In contrast,the hematoma area was apparently larger in the massage group,which indicates that massage pressure can result in blood leakage through the tiny needle hole in the vessel.

It is suggested that the dose of intravascular injection and factors other than the immediate arterial obstruction status after injection are decisive factors in tissue necrosis.Therefore,once fillers are injected into the blood flow,the possible consequences are determined by multiple factors,including the physical characteristics of the products,injection dose,individual hemodynamics,vascular endothelial injury,and endothelial cell response to injury.Admittedly,delivering hyaluronidase to the blocked artery is theoretically feasible for HA-induced arterial embolism.However,its real effectiveness is greatly reduced due to many practical difficulties.17Therefore,prevention is still the best course of action.We must grasp the characteristics of the product before injection,use a slow injection speed,and use shaking injection tips to avoid high-dose intravascular injections.

The subgroup analysis indicated that the initial blood flow state had a significant impact on the effect of massage therapy.The different recanalizations of CEA trunk blood flow within 5 min indicated significant individual differences in the tendency of embolus formation.As shown in Fig.7C,there was no difference in the degree of necrosis between the two subgroups of the control group,whereas the data of the massage group were statistically different.This suggests that massage can reduce necrosis in individuals with easily recanalized blood vessels,whereas tissue necrosis may be aggravated in cases of severe blood flow obstruction.As reported by other researchers,nitroglycerin could worsen the outcomes of dermal filler necrosis.14,18Therefore,empirical treatment with filler endovascular embolization may not be effective.It may need to be analyzed on a case-by-case basis,and the specific application conditions need to be further clarified.

Admittedly,clinical practice often involves a combination of various treatments,and animal models cannot fully represent the human tissue structure.Therefore,there is insufficient evidence to verify the effectiveness of these empirical treatment methods.Another limitation of this study is that we only studied the effects of a fixed massage frequency,intensity,and duration.In clinical practice,a fixed massage intensity cannot be guaranteed.Despite these limitations,this animal study deepens our understanding of the complications of arterial embolization with fillers and further suggests the importance of preventing intraarterial injection of fillers.The results of this study also indicate that local massage should be avoided when intra-arterial embolization is suspected;this is because tissue necrosis may be aggravated,especially in cases of severe blood flow obstruction.

5.Conclusion

In summary,local massage cannot reduce the complications of tissue necrosis when intra-arterial PMMA injection is performed.Prevention is key to reducing complications.

Ethics approval and consent to participate

This study was carried out following ethical approval from the Animal Ethics Committee of Peking University Third Hospital (approval no.A2019010).

Competing interests

The authors declare that they have no competing interests.

Authors’contributions

Xie H:Conceptualization,Methodology,Writing-Review and editing.Nie F:Data curation,Writing-Original draft.

Acknowledgments

This study was funded by Hafod Bioscience B.V.,which also provided Artecoll products for this experiment.Thanks are due to Dr.Guanhuier Wang and Dr.Yimou Sun for their participation in the acquisition,analysis,or interpretation of data.