Collagen injectables are useful primarily for patients with limited, predominantly superficial contour defects. The choice of agent and collagen preparation is largely dependent upon the depth of the contour defect. Superficial small rhytides are best treated with Zyderm I or II (formerly McGhan Medical, Santa Barbara, CA, now Inamed Aesthetics, Santa Barbara, CA), which can be placed very superficial in the superficial papillary dermis. Injection should be performed with visualization of the needle beneath the skin. If the needle cannot be visualized during injection, the augmentation will be much less effective. Zyplast (McGhan Medical) differs from Zyderm I and II in that it contains glutaraldehyde cross-linked collagen, which is less susceptible to collagenase degradation (1). Zyplast cannot be injected into the superficial papillary dermis or in areas of thin skin because it forms palpable beads if placed superficial. Moderate sized depressions can best be addressed with collagen preparations, such as Zyplast injected deep into the midreticular or deep reticular dermis at the dermal subcutaneous interface (1).

Bovine collagen preparations contain lidocaine, but topical anesthesia (such as Betacaine or 4% lidocaine) increases patient comfort and tolerance (2). Both Zyderm and Zyplast are injected with a 30-gauge needle with the bevel up and tunneling to the proper depth. The needle will be visible with Zyderm I and II, but Zyplast must be injected into deeper layers to avoid beading (1). The needle is advanced along the length of the rhytid and the material injected while the needle is withdrawn. Blanching of the overlying skin is observed with Zyderm. Overcorrection of 50% to 100% of the augmented area is needed because the water and lidocaine are resorbed in the first 24 hours.

Bovine collagen injections are specifically contraindicated in persons undergoing treatment with steroids or who have a history of allergy to other bovine products or meat (3). In addition, any person with a history of an autoimmune disorder and especially a collagen vascular disorder should not be treated.

One of the main risks associated with the use of bovine collagen is severe allergy. The allergy skin test is performed on all patients to determine the risk of a severe allergic response to bovine collagen. The skin test entails placing a 0.1-mL aliquot of collagen injectable beneath the skin, often in the hairline or forearm. The test site is observed closely during the first 3 days after injection and then for a total of 4 weeks before a treatment series would be administered. Approximately 3% to 10% of persons tested will demonstrate a significant response to the test dose and should not be treated (4). In addition, 1% to 2% of persons will develop an acquired allergy to bovine collagen despite initial nonreactivity to the test dose (5). These reactions can be severe and often are difficult and sometimes impossible to treat. Treatment reactions may consist of prolonged redness, swelling, itching, and firmness at some or all of the injection sites. However, severe reactions with superficial skin necrosis and extensive scarring are possible. Some authors have suggested that the immunologic reaction to collagen is universal but varies in severity. They suggest the augmentation due to collagen is more a function of the intradermal edema generated during immunologic reactivity than volume augmentation (6). Rare systemic complications have been reported with bovine collagen. Approximately 5 in 1,000 patients may experience systemic symptoms with flu-like symptoms, paresthesias, or difficulty breathing. Total yearly injections should not exceed 30 mL.

The rate of collagen removal can be quite variable, based upon facial anatomic location of injection, depth of injection, and injectable agent used. The biomechanical
forces of facial movement tend to increase resorption in areas such as around the mouth. Injections in these sites tend to be of shorter duration. Overall, injections last for 2 to 4 months. In regions such as the glabella, concomitant use of chemodenervation can slow resorption of collagen by decreasing dynamic facial movement. Of note, Zyplast is specifically contraindicated for treatment of glabellar rhytides due to the possibility of central retinal artery occlusion (5).

Autologous human tissue matrix (Autologen, Collagenesis, Beverly, MA) is an injectable human material derived from intact collage fibrils (6,7). It is processed from the patient’s own skin, which is harvested from elective surgery. After excision, the skin is sent to Collagenesis and processed for use. Two square inches of donor material is required to formulate a 1-mL syringe of injectable material, which can be stored for 6 months.

The processed material is diluted to a 5% suspension; thus, overcorrection of approximately 30% is needed. In addition, repeat injections, usually three over several weeks, are needed for adequate augmentation due to the low percentage of collagen material. The injectable material is not prepared with lidocaine; thus, the injections are more painful and require nerve blocks and topical anesthesia. Autologen augmentation can last for 3 to 6 months. No difference in observed persistence was noted compared to injection of bovine collagen over a 12-week period (7).

The disadvantages of Autologen include difficulty in isolating material, limited supply, painful injections, and time required for processing. The advantage is that infectious or allergic complications are virtually eliminated.

Dermalogen (Collagenesis) is a suspension of human collagen prepared from human donor tissue processed from cadavers (6). The donors undergo extensive screening for infectious diseases, and the material is irradiated before use. Skin testing is not required by the United States Food and Drug Administration (FDA), and the risk of allergic complications is very rare. Dermalogen is not formulated with lidocaine; thus, the injections tend to be painful and require nerve blocks and topical anesthesia.

Overcorrection of 20% to 30 % is recommended, and serial injections are necessary due to initial rapid resorption. Two- to threefold of injection material is needed for augmentation results similar to that achieved with bovine collagen. This increased amount of material is due to the initial high resorption, but expense is also increased two- to threefold. The material is especially useful for patients with contraindications, allergies, or unwillingness to use bovine collagen.

Fine wrinkles, particularly those around the lips and mouth, can be especially problematic to patients. The fine vertical lines of the lips or “lipstick lines” can be treated both directly and indirectly with collagen. Injection of these lines along their length using a preparation designed for superficial use (Zyderm) reduces their appearance. In addition, these fine lines can be softened by deeper injections along
the vermilion. Injection can be performed continuously or in a beading fashion to create beads of material. With either method, massage can redistribute and mold the material to some degree. A slightly deeper injection plane can rotate the lips to provide a youthful upper tooth show and full lips. Injection at one or all three of these planes provides many options for lip contouring.

Many patients become concerned with small problem areas that they view as significant. Collagen injections can provide a reproducible and predictable treatment for these problem areas. For example, many patients are concerned with the brow furrows produced by the corrugator and/or procerus muscle. The problem in these cases is twofold: dynamic and static. We often use chemodenervation, in addition to collagen treatments, to relax the inciting muscle groups. The chemodenervation reduces the dynamic component, while collagen augmentation can effectively treat the static component. Collagen injections along these furrows provide improved contouring of these focal points. Chemodenervation and collagen injections can provide synergistic effects, especially when implemented with a good skin care regimen. Although beyond the focus of this article, skin care regimens with consistent use of retinoic acid or chemical or laser peels will greatly augment the overall appearance and satisfaction with collagen injections because fine lines will be diminished.

In our practice, collagen injections have a variety of roles as discussed. Collagen is especially useful for focal points as determined by the patient. In addition, we occasionally use collagen to simulate longer-term augmentation methods such as fat injections. It often is challenging to counsel a patient regarding potential changes in appearance resulting from facial augmentation. Collagen injections can provide a temporary trial that allows the patient and physician to envision the same goals for facial recontouring. Again, deep facial defects are best treated with deep augmentation methods such as fat. However, collagen occasionally can simulate these results in regions such as the nasolabial fold and lips. For example, many patients will be interested in augmenting their lips or reducing creases such as the nasolabial fold, but they cannot envision the results of such augmentation. Often before considering more permanent augmentation such as fat injections, we use collagen injections to simulate future interventions. Collagen augmentation can provide the groundwork for discussions and emphasize realistic expectations for both the physician and patient.

Collagen injections represent a useful method in the armamentarium of facial augmentation. Consideration of the use of any agent first begins with the definition of the facial deformity. Once the problem is defined, a rational treatment approach can be formulated. Each of the agents offers distinct advantages and limitations. By understanding the advantages and disadvantages, the physician and patient can precisely communicate and agree upon a rational treatment plan.

In his preliminary assessment, Piacquadio (14) originally reported in 1994 the use of cross-linked HA (hylan B gel) in 150 patients enrolled in a multicenter clinical study initiated in 1991. At week 12, 80% of patients reported moderate or higher satisfaction. Adverse reactions, including persistent erythema and ecchymosis, were rare. In 1997, Piacquadio et al. (15) reported an initial study in which they used a 12-month guinea pig model to investigate the tissue effects of hylan B gel versus “collagen” controls. The study demonstrated significantly higher histologic demonstrability of the hylan B gel (87 %) compared to that of collagen (25%) at 26 weeks. At 1 year, only hylan B gel implants were evident (12/16 test sites). Thus, hylan B gel was found to be biologically compatible and more stable in dermal tissues than collagen.

The first evaluation of use of the material in humans was reported in 1998, when Olenius (16) evaluated 113 subjects after their facial rhytides were injected with partially cross-linked HA produced in bacterial culture. The study demonstrated the highest degree of effectiveness at 2 weeks (98%), with subsequent decline at 3 months, 6 months, and 1 year to 82%, 69%, and 66%, respectively.
Olenius also reported no allergic reactions and infrequent side effects, which mostly included redness and swelling related to overly superficial placement of the material, lumpiness secondary to an uneven injection technique, and darkening secondary to small hematoma formations.

In 1998, Duranti et al. (17) evaluated 158 Caucasian female patients who were treated with facial intradermal implant of HA gel for augmentation therapy. The safety evaluation showed a 12% to 13% rate of postoperative immediate adverse events, which were limited to intermittent swelling, erythema, and slight discomfort, all of which resolved without further sequelae. There was no evidence of major systemic side effects. In a histologic evaluation of five patients from the same study, the product was shown to last as long as 24 weeks with significant resorption by 52 weeks after injection.

In the latest study by Lowe et al. (18), 709 patients treated with Hylaform and Restylane (Q-Med AB, Uppsala, Sweden) were followed up clinically for at least 1 year. The evaluation showed a 0.42% rate of delayed inflammatory skin reactions, which started approximately 8 weeks after injection (18).

Micheels (19) reported eight patients with possible allergic reactions after being treated with HA gel, thus raising a question for the need for routine intradermal skin allergy testing and consensus about its formulary.

There have been several case reports describing isolated reactions to HA. Shafir et al. (20) described sterile abscesslike swellings at the site of injection 2 months after injection of Restylane for nasolabial augmentation requiring incision and drainage. Kavouni and Stanec (21) reported erythematous swelling and mildly tender nodules, which resolved with 2.5% hydrocortisone cream without residual intradermal defects. Another case of a Caucasian patient who developed acute sterile multiple tender red nodules within the treatment area after the third treatment session with HA was reported by Lupton and Alster (22). The lesions resolved after a course of oral steroids and salt compresses to site. Schanz et al. (23) reported a case of arterial embolization of the dorsal nasal artery caused by injection of HA. Symptoms included a reticular pattern affecting the glabellar region and the side of the nose. The patient was treated with low-molecular-weight heparin for 1 week. The patient subsequently developed an ulcer at the injection site, which eventually healed without any residual defects.

From recently reported data from an estimated 144,000 patients treated with Restylane, Perlane, or Restylane Fine Lines, Friedman et al. (24) identified localized hypersensitivity reaction as a major side effect of the injections. According to their review of the worldwide data gathered by the manufacturer on reported adverse events from using nonanimal-stabilized HA gel, the overall proportion of patients experiencing an adverse event was 0.15% in 1999 and 0.06% in 2000.

The lack of immunogenic reaction to the HA preparations is one of the main advantages of the material. Larsen et al. (25) originally evaluated the immunogenic activity of hylan gel in rabbits in 1993. Their work detected no antibody production to hylan gels. Thus, hypersensitivity to HA most likely is secondary to impurities of the bacterial fermentation process resulting in contaminating DNA in HA preparations (24). The significance of trace protein in HA preparations was evaluated by Manna et al. (26), who showed up to four times the quantity of protein in certain lots of Restylane compared to that in the same volume of preparation in certain lots of Hylaform. Filion and Phillips (27) further recognized this when they tested low-molecular-weight fragments obtained from seven different HA preparations. They showed that 2 of the 7 HA preparations stimulated the synthesis of interleukin-12 and tumor necrosis factor-α in human monocytic cells. Treatment of
those two HA preparations with deoxyribonuclease reduced the induction of proinflammatory cytokines by monocytes. These finding are consistent with reports of delayed hypersensitivity reactions in patients treated with hyaluform acid derivatives (17,20). In their study, Friedman et al. (24) also analyzed the differences between side effects in patients treated with newer formulations of HA containing lower levels of trace proteins and those treated with original batches. The side effects occurred 5.9 times more frequently in patients treated with the batch containing the “old” higher levels of trace protein (24).