Secondary Logo

Ultrastructural Analysis of 3 Hyaluronic Acid Soft-Tissue Fillers Using Scanning Electron Microscopy

Flynn, Timothy Corcoran MD*,†; Thompson, David H. PhD‡,§; Hyun, Seok-Hee PhD‡,§; Howell, David J. PhD, RRT

doi: 10.1097/01.DSS.0000452647.14389.a7
Original Article
Free

BACKGROUND Although hyaluronic acid (HA) specifications such as molecular weight and particle size are fairly well characterized, little information about HA ultrastructural and morphologic characteristics has been reported in clinical literature.

OBJECTIVE To examine uniformity of HA structure, the effects of extrusion, and lidocaine dilution of 3 commercially available HA soft-tissue fillers.

MATERIALS AND METHODS Using scanning electron microscopy and energy-dispersive x-ray analysis, investigators examined the soft-tissue fillers at various magnifications for ultrastructural detail and elemental distributions.

RESULTS All HAs contained oxygen, carbon, and sodium, but with uneven distributions. Irregular particulate matter was present in RES but BEL and JUV were largely particle free. Spacing was more uniform in BEL than JUV and JUV was more uniform than RES. Lidocaine had no apparent effect on morphology; extrusion through a 30-G needle had no effect on ultrastructure.

CONCLUSION Descriptions of the ultrastructural compositions and nature of BEL, JUV, and RES are helpful for matching the areas to be treated with the HA soft-tissue filler architecture. Lidocaine and extrusion through a 30-G needle exerted no influence on HA structure. Belotero Balance shows consistency throughout the syringe and across manufactured lots.

*Cary Skin Center, Cary, North Carolina;

Department of Dermatology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina;

Department of Chemistry, Purdue University, West Lafayette, Indiana;

§Pandion Laboratories LLC, West Lafayette, Indiana;

San Francisco, California

Address correspondence and reprint requests to: Timothy Corcoran Flynn, MD, Cary Skin Center, 200 Wellesley Trade Lane, Cary, NC 27519, or e-mail: flynn@caryskincenter.com

T. C. Flynn is a consultant for Merz. S.-H. Hyun is a Senior Research Scientist for Pandion Laboratories LLC. Pandion Laboratories has received research support from Merz. D. J. Howell is a self-employed medical writer, compensated by Merz for his editorial assistance. The other authors has indicated no significant interest with commercial supporters. This study was funded by Merz.

The popularity of hyaluronic acid (HA) soft-tissue fillers for nonsurgical aesthetic treatments is hard to contest. According to the American Society of Dermatologic Surgery, in 2012, almost 1 million soft-tissue procedures (916,000) were performed.1 Indeed, soft-tissue fillers are second only to neurotoxin injections in frequency of delivery in the clinical setting.1,2

Concomitant with the rise in use of soft-tissue fillers has been the emergence—in this journal and in other peer-reviewed journals—not only of cross-HA comparisons of efficacy and complications3 but also of product-specific variables such as rheology,46 degree of cross-linking,7 and molecular weight.8 We examined at length in late 2013 the molecular weight aspect of 3 HA soft-tissue fillers—Belotero Balance (BEL; Merz USA, Greensboro, NC), Restylane (RES; Valeant Pharmaceuticals International Inc., Bridgewater, NJ), and Juvéderm Ultra (JUV; Allergan, Irvine, CA).8 Differences in the molecular weight distributions of the HA fillers were demonstrated in our earlier research. The different HA fillers also had individualized responses to hyaluronidase.

In this study, we wanted to investigate whether there were differences in the ultrastructure of these materials. Using cryogenic scanning electron microscopic (SEM) techniques, combined with energy-dispersive x-ray (EDX) detection, we examined cross-product HA filler structural details such as extrusion effects, gel consistency within the syringes, and whether addition of lidocaine had any effect on 3 commercially available HAs. Variation between individual lots of BEL was also investigated.

Back to Top | Article Outline

Scanning Electron Microscope and Energy-Dispersive X-Ray Detection

Scanning electron microscope produces images through a beam of electrons9 instead of light as in conventional microscopy. The electron beam, generated at the superior part of the SEM instrument, travels vertically downward through a series of magnetic fields and lenses in a vacuum and is focused on the sample to be investigated. When the beam strikes the sample, secondary electrons and electromagnetic rays are emitted by the sample. These electrons and electromagnetic rays are then captured and converted into a signal that is sent to a monitor, to be displayed as an image. The samples are prepared by freezing them and coating the sample with a thin layer of platinum to make them more electron conductive. Samples are kept at cryogenic temperatures to limit electron beam damage to the samples during imaging.

An Oxford Instruments INCA Energy 250 system with a 30-mm window (Abingdon, United Kingdom) EDX detector was used in some of the experiments described below to produce elemental maps of the 3 HA samples. Elemental mapping of the sample involves creation of a pixel grid, commonly called a bitmap, from each of the samples by applying an electron beam raster across the surface and collecting the stimulated emission of secondary electrons and x-rays as a function of position along the sample surface. The energies of the stimulated x-rays are analyzed at each position within the bitmap with regard to known characteristics of atomic emissions of all elements. In the study described here, investigators analyzed the samples for the presence of oxygen, carbon, and sodium.

To collect SEM data, the investigators used a NOVA NanoSEM field emission SEM device (FEI, Hillsboro, OR), coupled with a Gatan Alto 2500 Cryo Unit (Pleasanton, CA). The NOVA NanoSEM (Figure 1) is a high-resolution device that provides for a wide range of magnifications during imaging. The GATAN Alto 2500 Cryo Unit uses a liquid nitrogen bath for flash freezing of the sample. The frozen sample is then cleaved to expose its ultrastructural morphology before brief high-vacuum sublimation to remove surface water, and application of a thin platinum layer before transfer onto a cryogenically cooled stage in the microscope. Scanning electron microscope images obtained using the NOVA NanoSEM equipped with an EDX detector are reported in the Results section.

Figure 1

Figure 1

Back to Top | Article Outline

Study Objective

In this study, investigators sought to answer several questions using SEM methods. The questions concerned ultrastructural morphology, lot-to-lot consistency, effect of filler extrusion through a 30-G needle, and the effects—if any—of 1% lidocaine solution addition to 3 commercially available HA soft-tissue fillers: BEL, RES, and JUV.

Questions were as follows:

  • (1) What is the elemental distribution in the BEL, RES, and JUV gels?
  • (2) What is the morphologic appearance of the gels in the syringes by high-resolution cryo-SEM?
  • (3) What is the effect of lidocaine on the HA gel morphology, if any?
  • (4) What is the effect of extrusion of the HA gels through a 30-G needle versus gentle release through the Luer-Lok opening?
  • (5) Does the ultrastructure of the filler remain consistent throughout the BEL syringe?
  • (6) Is the gel morphology consistent across multiple lots of BEL?
Back to Top | Article Outline

Materials and Methods

Sample specimens were prepared as shown in Figure 2. Samples from each HA gel (BEL, RES, and JUV) were cryoimmobilized by plunge freezing into a liquid nitrogen bath at −196°C. Frozen samples were then transferred in vacuo to a cryopreparation chamber where they were fractured and sublimated for 3 minutes before sputter coating with a platinum layer for 2 minutes. After transfer to the SEM chamber, samples were examined at 5 kV under high vacuum on the cryo-SEM stage (Figure 3). The EDX detector mapped the samples for the presence of the elements of carbon (green), oxygen (red), and sodium (blue). These colors are used consistently in all the images reported in the Results section. For lot analysis, 3 different lots of a single product (BEL) were evaluated. The lidocaine effect experiments were performed by preparing lidocaine-containing samples in the following manner: 0.1 mL of a 1% lidocaine clinical solution was loaded into a 1-mL syringe and 0.9 mL of the HA filler was loaded into a second 1-mL syringe; the 2 syringes were then connected by a female-to-female Luer-Lok adapter free of air bubbles, and the contents of the syringes were gently pushed back and forth through the Luer-Lok fitting 30 times before removal of a small portion of the filler–lidocaine mixture for cryo-SEM analysis. The samples for extrusion analysis were performed by pushing a small sample onto the SEM sample holder with gentle pressure, either through the unmodified Luer-Lok fitting on the manufacturer's syringe or through a 30-G 0.5 inch BD needle that was attached to the Luer-Lok fitting on the syringe. In all cases, the samples were then flash frozen and processed for cryo-SEM analysis as described above.

Figure 2

Figure 2

Figure 3

Figure 3

Back to Top | Article Outline

Results

Elemental distribution confirms visualization of the HA network by SEM. Scanning electron microscope–EDX images of each HA filler are shown in Figure 4. The top row of images are unmapped cryo-SEM micrographs of the 3 different HA fillers. The second, third, and fourth rows are images of carbon (C), oxygen (O), and sodium (Na) elemental distribution maps, respectively. The column of images on the left are of BEL, the middle are of RES, and the right are of JUV. The unmapped images of the fractured HA gel material reveal the presence of a dense 3-dimensional structure with irregularly spaced fibrous or slab-like structures having an interconnected network appearance consistent with interconnected HA molecules. Elemental mapping experiments indicate that the irregular structures are rich in C and Na. Oxygen mapping shows that the dark gray zones in the unmapped images that are present between the lighter gray HA network structures are rich in oxygen. We infer from the elemental maps, and the overlays with the unmapped cryo-SEM images from the same sample area (Figure 4, bottom row), that the walls in the network are comprised of HA that is suspended in aqueous media. We attribute the sodium signal enrichment within the HA network structure to electrostatic attraction of Na counterions to the glucuronic acid carboxylate (CO2) units within the HA polymer.

Figure 4

Figure 4

Differences in morphologic architecture are evident among BEL, RES, and JUV. As can be seen in Figure 5, a more uniform network having a finer consistency (i.e., smaller average pore dimensions—pore being the term used to describe the small spaces in between the HA polymeric chains) was found in BEL when compared with JUV and RES. Juvéderm Ultra exhibits a network structure that is more heterogeneous, with some regions of finer porosity and other zones that are of larger average pore size. Restylane images show multiple pore sizes, some of which are quite dense and others that are of the lowest density observed in any of the samples. There was no evidence of discrete particles in BEL or JUV. However, the RES samples contained many highly irregular particles with very small pores that were surrounded by a more diffuse network of HA material with larger pores.

Figure 5

Figure 5

Lidocaine addition shows no effect on HA gel morphology. In clinical practice, small amounts of anesthetic solution are frequently mixed with HA gels. We sought to determine the effect of lidocaine addition on the ultrastructure of the HA gels. As shown in Figure 6, little change is noted in the network structure appearance after addition of lidocaine to each of the 3 soft-tissue fillers. As noted in the unmodified samples (Figure 5), the appearance of BEL and JUV is more uniform before and after lidocaine addition than the RES samples that still possessed particulate matter after lidocaine dilution.

Figure 6

Figure 6

Extrusion of the HA gels through a 30-G needle has no effect on gel architecture. Each of the 3 HAs was extruded in 2 ways: (1) through a 30-G needle and (2) simply releasing it through the Luer-Lok opening of the syringe. The images in the top row of Figure 7 show that the effect of extrusion through the needle is negligible compared with the network structure apparent when the sample was extruded through the much larger diameter Luer-Lok port of the manufacturer's syringe. This suggests that the architecture of the HA networks in each soft-tissue filler remains fundamentally intact during an operation that mimics the deposition of filler product in the patient through a 30-G needle.

Figure 7

Figure 7

BEL HA gel is consistent throughout the BEL syringe and across multiple lots. Using a single syringe of each BEL gel, HA filler samples were applied by extrusion through the Luer-Lok syringe tip onto cryo-SEM stubs at 3 different stages of sample extrusion. Specifically, we applied material from the first 0.3 mL, the middle 0.3 mL, and the final 0.3 mL of the syringe onto the SEM sample holders in an effort to determine whether the sample ultrastructure changes as one proceeds from the front, mid-section, and back-section of the syringe. Comparison of each row of images in Figure 8 suggests that there is no appreciable change in sample ultrastructure as one draws material from the front, mid-section, or back of the syringe. Regarding lot consistency, note the relatively uniform network structure of BEL, when the columns of data are compared, showing a substantial uniformity of the HA gel across 3 separate lots of the soft-tissue filler. The findings show consistency in a narrow range of network porosities ranging from 1 to 10 μm both across the different BEL lots and within different positions in the syringe.

Figure 8

Figure 8

Back to Top | Article Outline

Discussion

In the clinical setting, physicians have long acknowledged that 1 HA is not simply interchangeable with another. One HA has applicability and usefulness for certain areas of the face, and for one purpose, whereas another has usefulness in other areas. Examples would be the use in the tear trough or lips versus the use for deeper subdermal volume replacement. The professional literature is useful for highlighting distinctions between and among soft-tissue fillers but, ultimately, the decisions rest with the treating health care provider.

The HA fillers have different qualities with respect to tissue integration when placed intradermally.8,10 Our first study, in 2011, noted differences in diffusion throughout the skin in both biphasic and monophasic soft-tissue fillers. The HA with the cohesive polydensified filler (BEL) appeared “evenly placed in and between the collagen fibers throughout the reticular dermis,” whereas JUV showed “material throughout the dermis but in large clumps and channels.” RES showed “large pools of HA distributed as clumps or beads of material located at the lower part of the dermis.” Another study was published in 2013 that has helped better understand integration of HA fillers into the dermis, in this case, into the superficial dermis.11 Using ultrasound and histologic techniques, investigators found that the HA with the cohesive polydensified matrix, that is, BEL, was useful for the superficial dermis “because of its high degree of integration into the dermis.” In comparison, JUV showed a heterogenous appearance with a cone of shadow at the edge of the papule on intradermal injection, and RES showed a granular appearance, with a cone of shadow underneath.

In 2013, our research group published a follow-on study about the high–molecular weight and low–molecular weight components of BEL, RES, and JUV.8 Belotero Balance had the largest high–molecular weight species, followed by JUV and RES (p < .0001). Differences among the low–molecular weight species in the 3 HAs were not significant. In that study, the 3 HAs were exposed to ovine testicular hyaluronidase at 6 time points, to determine degradation rates of the 3 HAs. Degradation rates of the 3 HAs were similar, with full degradation at 24 hours. However, the proportion of high–molecular weight components in BEL remained high at early time points.

In the study reported here, we used SEM techniques to further understand the nature of these HA polymers. All of the polymers were rich in oxygen, carbon, and sodium; these elements were used to identify the localization and ultrastructure of the HA polymers in the images. The HA polymers seem to form a fibrous network structure that is suspended in an aqueous medium. In terms of appearances under SEM imaging, BEL presented as the HA with the most uniform distribution, followed by JUV. The appearance of Restylane was the least uniform of the 3, presenting with highly irregularly sized particles in a diffuse network. The findings of BEL and RES support 2 earlier findings about evenness of the 2 products after dermal injection.12,13

Much research has been conducted to investigate the rheological properties of HA and other fillers, in an effort to correlate measurements such as elastic modulus (G′) and viscosity with how the filler acts within tissue after injection.4,14,15 For example, it has been postulated that HA gels demonstrating a high G′ provide more structural support and tissue “lifting,” whereas those with a low G′ are best suited for areas of thin skin and/or superficial placement.

Our work here correlates rheologic and histologic findings with the overall ultrastructural morphology shown in the SEM images. Belotero Balance exhibits the lowest G′ and a high degree of histologic dermal integration, followed by JUV, and then RES. Our findings here show that BEL has the highest degree of uniform interlocking HA strands compared with JUV. Restylane is shown to be formed of denser particles suspended in HA strands. The findings of this study, combined with previous work, may explain why BEL is more ideally suited for superficial use and intradermal placement, whereas JUV is used subdermally and deeper, with RES also suitable as a deeper HA filler.

We also investigated how the addition of lidocaine and extrusion force might affect overall morphology. We did not find noticeable differences in the gel's structure with the addition of lidocaine in any of the 3 HAs examined. The distinctions noted above in the nonlidocaine images remained present in the ones with the lidocaine added. Similarly, no marked changes were observed in the 3 HA gels when delivered through a 30-G needle compared with simple delivery through the Luer-Lok opening of the syringe. In addition, the consistency of the HAs remained intact in both the needle and syringe aperture sampling. This is an interesting finding when one considers speculation among physicians that injection of HA through small-bore needles (30–32 G) would have a shearing force effect, altering the overall architecture of the gel. Our results show no effect on gel morphology when passed through a 30-G needle.

Our last area of analysis was lot-to-lot consistency of BEL. Our findings in this area mirrored our findings in the across-HA analyses, namely, consistency within the individual syringe and across multiple lots.

Back to Top | Article Outline

Conclusion

Differences were observed in terms of size and uniformity across all 3 HAs investigated. In descending order of uniformity, BEL presented more uniformly than JUV, which presented more uniformly than RES. Restylane has more dense particles of HA suspended in looser HA strands. Neither lidocaine nor extrusion through a 30-G needle appeared to influence inherent properties in the 3 HAs. In terms of consistency both within portions of the syringe and across lots of BEL, little variability was noted. These findings are useful for physicians to help them better understand HAs' structural correlates with patients' needs and appropriate product selection.

Back to Top | Article Outline

Acknowledgments

We sincerely appreciate the contributions of Matt Chansky (Research Triangle Park, NC) for his illustrations and of Dale Murphy (San Francisco, CA) for his graphics.

Back to Top | Article Outline

References

1. ASDS.net. Available from: http://www.asds.net/_Media.aspx?id=6598 Accessed October 16, 2013.
3. Cassuto D, Sundaram H. A problem-oriented approach to nodular complications from hyaluronic acid and calcium hydroxylapatite fillers: classification and recommendations for treatment. Plast Reconstr Surg 2013;132(4S-2):48S–58S.
4. Sundaram H, Voigts B, Beer K, Meland M. Comparison of the rheological properties of viscosity and elasticity in two categories of soft tissue fillers: calcium hydroxylapatite and hyaluronic acid. Dermatol Surg 2010;36(suppl 3):1859–65.
5. Sundaram H, Cassuto D. Biophysical characteristics of hyaluronic acid soft-tissue fillers and their relevance to aesthetic applications. Plast Reconstr Surg 2013;132(4S-2):5S–21S.
6. Stocks D, Sundaram H, Michaels J, Durrani MJ, et al.. Rheological evaluation of the physical properties of hyaluronic acid fillers. J Drugs Dermatol 2011;10:974–80.
7. Kablik J, Monheit GD, Yu L, Chang G, et al.. Comparative physical properties of hyaluronic acid dermal fillers. Dermatol Surg 2009;35(suppl 1):302–12.
8. Flynn TC, Thompson DH, Seok-Hee H. Molecular weight analyses and enzymatic degradation profiles of the soft-tissue fillers Belotero Balance, Restylane, and Juvéderm Ultra. Plast Reconstr Surg 2013;132(4S-2):22S–32S.
9. Scanning Electron Microscope. Available from: http://www.purdue.edu/rem/rs/sem.htm Accessed June 4, 2013.
10. Flynn TC, Sarazin D, Bezzola A, Terrani C, et al.. Comparative histology of intradermal implantation of mono and biphasic hyaluronic acid fillers. Dermatol Surg 2011;37:637–43.
11. Micheels P, Sarazin D, Besse S, Sundaram H, et al.. A blanching technique for injection of the hyaluronic acid Belotero. Plast Reconstr Surg 2013;132(4S-2):59S–68S.
12. Prager W, Steinkraus V. A prospective, rater-blind, randomized comparison of the effectiveness and tolerability of Belotero® Basic versus Restylane® for correction of nasolabial folds. Eur J Dermatol 2010;20:1–5.
13. Prager W, Wissmueller E, Havermann I, Bee EK, et al.. A prospective, split-faced, randomized, comparative study of safety and 12-month longevity of three formulations of hyaluronic acid dermal filler for treatment of nasolabial folds. Dermatol Surg 2012;38:1143–50.
14. Ascher B, Rzany B, Sundaram H. (Chairs). Fillers: from characteristics to recommendations. Consensus panel discussion. Consensus panel discussion at: International Master Course on Aging Skin 2013 Annual Meeting; January 31–February 3, 2013; Paris, France.
15. Lorenc P, Fagien S, Flynn TC, Waldorf HA. Clinical application and assessment of Belotero: a roundtable discussion. Plast Recon Surg 2013;132(4S-2):69S–76S.
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. All rights reserved.