Introduction

The use of tissue fillers in facial corrective surgery has become a reality over the last 6 to 8 years. Several techniques have been described for augmenting soft tissues such as fluorinated polymers [1], hydroxyapatite [1], purified autologous collagen [2,3,4,5,6], polymethylmethacrylate microspheres [7], aminocaproic acids [8,9], expanded polytetrafluoroethylene (E-PTFE) [10] and other microparticles as well as autologous fat injections [11,12,13,14]. Some of these techniques have been employed with more or less good results because of problems such as migration, reabsorption, scar tissue formation and toxicity.

We have conducted a retrospective study of 104 patients treated with polyacrylamide hydrogel (PAAG) injections for facial correction. The good results we found combined with the convenient mode of administering this product warrants its publication in anticipation of prospective clinical trials, which are being initiated.

Description of Material

Polyacrylamide hydrogel, now marketed as (Aquamid®), is an atoxic, stable, non-resorbable, sterile watery gel for injection into the soft tissues. It consists of approximately 2.5% cross-linked polyacrylamide and non-pyrogenic water. Due to its unique characteristics the gel is highly bio-compatible.

PAAG has been used in plastic and aesthetic surgery in Russia for more than 10 years. Approximately 30,000 patients have been treated with it. Comprehensive data regarding the safety of the gel includes pre-clinical as well as clinical studies.

PAAG is widely used in biomedical research as well as in industry. For decades it has been used in the preparation of soft contact lenses. In biomedical research the product serves as tissue implant material, and it is used in tissue models, body fluid models, detectors of penicillin antibodies, as well as carriers of hormones and drugs in animal studies. In the United States most of the polyacrylamides are used in water and wastewater treatment as a thickening and suspending agent. In lesser amounts, polyacrylamide is used in sugar clarification and drug production [14]. Another branch extensively using polyacrylamides is the cosmetic industry [15].

The interest in polyacrylamides originated because of their use in the groundwater clearance. Obviously, chemicals used for the groundwater treatment have to be non-toxic to avoid possible hazards to humans or animals. Comprehensive toxicology studies of polyacrylamides has shown that polyacrylamide is non-toxic to humans as well as animals [16]. Thus, a long-term feeding study (2 years) in dogs and rats with polyacrylamides revealed that there were no specific toxic effects on animals. D. J. King and R. R. Noss from the University of Massachusetts summarize in their review numerous articles reporting toxicology studies with polyacrylamides [17]. They concluded that polyacrylamides are non-toxic when used at a limited range [17].

It is generally believed that the main toxicological concern with polyacrylamide is its acrylamide monomer content [18] which is below 0.0064 µg/ml. Thus, by implantation of PAAG the single dose of monomer is lower than the exposure to drinking water in 1 or 2 days [19].

The stability as well as toxicity of PAAG has been extensively investigated in Russia (the previous Soviet Union) during the past 10 years and for the Aquamid® formula it has now been confirmed in Denmark in several toxicology studies according to ISO standards (data on file with CONTURA SA). Some investigations were performed also in Norway and USA. The stability of the gel was assessed by various methods, e.g., exposure of the gel to enzymes, bacteria, and oxidizing agents [20,21]. All studies revealed that PAAG is a stable, practically nondegradable product.

In vitro experiments concerning cytotoxicity were performed on cell cultures as well as on human blood cells. No cytotoxic effects were detected after incubation for 24, 48 hours and 10–12 days with PAAG [20,22]. Additionally, no impact on human blood cells was observed after incubation with 5% PAAG for 24 hours [23]. In vivo studies were performed on various species including rats, mice, rabbits and dogs. In BALB/c mice, PAAG had no mutagenic effects, as assessed by micronuclei testing [24]. Subchronic toxicology studies in rats revealed no effects on blood and liver biochemistry, and no inflammatory processes were observed [25]. After 6–8 weeks of PAAG injection (s.c.) in rabbits, no physiological or histological changes were observed, and liver and spleen were not affected [26]. No carcinogenic effects have been seen 18 month after implantation of PAAG (i.m.), assayed by immunodetection of organospecific tumor-associated antigens [27].

Histology done on different tissues from various animals as well as from humans after longer time of incubation (6 month - 10 years) have shown that the gel inter-grows with thin inter-layers of connective tissue and elastic fibres [28]. No thick fibrous capsules, dystrophic nor necrotic changes, local allergic reactions, haemodynamic disturbances, carcinogenic effects or calcium salt deposits have been observed [28], and gel has not been demonstrated in lymphatic nodes [29,30,31].

Method of Administration

The gel is injected under local anesthesia, but for correction of wrinkles and folds, local anesthesia is not necessarily required. For lip augmentation, anesthesia through a nerve block is recommended. The procedure must be conducted under sterile conditions. The gel is delivered in pre-filled sterile syringes of 1 ml with Luer lock and should be injected subcutaneously with a thin-gauge needle, e.g., 27 G. The necessary amount of gel is injected subcutaneously in a retrograde manner by injecting the gel while withdrawing the needle. After the injection a light manipulation helps to obtain an even distribution of the gel. The injected gel will form a stable, soft part in the connective tissue. Secondary injections can be given and gel can be aspirated within a year if deemed excessive.

Clinical Experience

Since the 80s a hydrogel consisting of 5% polyacrylamide and 95% water has been used in aesthetic/plastic surgery. Most of the patients treated with PAAG have undergone breast augmentation and/or facial corrections. In 1996 a retrospective clinical study was performed by a group of Danish investigators on 175 patients who were randomly selected from a list of 390 patients who had undergone mammoplasty before 1/7/94 at the clinic in Kiev [32]. The tolerability of the injections, assessed by side effects and the results of laboratory tests, showed that the injection sessions were well tolerated and only a few side effects (2%) were observed. These occurred mostly due to incorrect technique of injection which could be altered, or they were judged not to be related to the gel injection. The overall cosmetic results assessed by surgeons and patients themselves were very good [32].

Material and Methods

The investigation was carried out according to a protocol adapted to Good Clinical Practice (GCP) requirements. The investigational site in Kiev identified from their hospital records all the patients who had had injections of PAAG in the face and invited them by letter to participate. Patients were invited to call the center and arrange for an appointment during weeks 39 and 40, 2000. The clinical investigation was conducted by one or more of four independent specialist surgeons in the presence of an interpreter, and all findings were noted in a pre-printed case report form (CRF). The patients filled in a questionnaire to assess their evaluation of both the cosmetic result and safety of the procedure. Both CRF and patient questionnaire assessed the health status of the patient regarding local (facial) as well as systemic effects/events. Normal aesthetic criteria were used by both surgeons and patients. None of the patients had been examined systematically prior to the injection(s), and no pre-injection photos existed. The photos attached to this study originate from Danish and Swedish patients having received facial injections of PAAG, produced in Denmark under the trade name (Aquamid®) (Figs. 4, 5, 6, 7, 8). For all patients the volume of polyacrylamide hydrogel previously injected was recorded. If more than one injection had been administered, the volume of each injection was recorded separately in the CRF. If the patient had had injection of polyacrylamide for breast augmentation, this was recorded as well along with the date and volume injected in order to evaluate the total patient exposure to the product.

The cosmetic result was judged by both the investigator(s) and the patient and rated on an interval scale with four options ranging from very good to very unsuccessful. The safety assessment included the following signs and symptoms, which could possibly be related to the injections: skin color, skin thickness, development of subcutaneous fat, teleangiectasia, depigmentation, wrinkles, hyperpigmentation, peau d’orange, scratch marks, hyperemia, rash, polish-like skin, scars, hypertricosis and sebaceous glands hyperplasia. All of these signs and symptoms were recorded as present or absent. In addition, the skin sensitivity at the injection site was assessed.

The investigators filled in the case report from which included all fields. No patient identifiable data were recorded, just name, date of birth, CRF number, date of investigation and investigators signature which were kept in a log at the study site. After completion of all CRFs, the complete sets were reviewed by a monitor from a Contract Research Organization (CRO) for completeness and legibility and all corrections were entered by the investigators prior to double data entry into the database.

Results

Demographics

The data contained information on 104 patients, however, many records contained no information on one or several of the pre-planned variables. Seven patients did not wish to give day of birth, but their birth year and most other variables were known. These patients, who are not included in the comparative analyses, will be described separately under each heading. One woman failed to report on breast augmentation. As this variable may potentially influence the analyses of tolerability to PAAG, it was decided to exclude her from further analyses.

A total of 96 patients were left to be included in the overall analyses. Their age distribution is illustrated in the frequency histogram presented in Figure 1.

Figure 1
figure 1

Patient demographics: frequency distribution of patient age at time of investigation. N = 96, mean ± SD = 37.4 ± 9.7.

Full size image

The majority of women belonged to younger age groups. Only a minority (6) were 50 years or older. The 7 women with unknown birth date showed an age span of 30–61 years, with a mean age of 39 years.

Examination of the patients took place from a few months to 9 years after their facial correction, with the majority examined 2–6 years after PAAG injection. The distribution of time since facial correction can be seen in Figure 2.

Figure 2
figure 2

Demographics: frequency distribution of time (years) from time of last facial correction to date of examination. N = 93, mean time ± SD = 3.9 ± 2.

Full size image

A majority of women had a follow-up time ranging from 1 1/2 to years. The follow-up time of the 7 women with unknown birth date was from 9 months to 9 years, with a mean time of 3 1/2 years.

The patients received from 1 to 3 facial corrections so that the presented material consists of a total of 174 interventions in 96 patients. The distribution according to number of corrections in each patient and the facial site is presented in Table 1.

Table 1 emographics: number of corrections at each facial area examined in 96 patients.
Full size table

A total of 49 patients (51%) had had breast augmentation as well, most of these belonging to the group with only 1 facial correction. In the group of 7 women with unknown birth date, 2 also had had breast augmentation. They did not present with any symptoms.

The amount of gel injected for facial correction was recorded and data were available from 93 patients. In Figure 3 it can be seen that the majority received up to 5 ml PAAG. The group of 7 women with unknown birth date did not differ from the main group in number of facial corrections or amount of injected gel.

Figure 3
figure 3

Demographics: frequency distribution of total ml gel injected for facial correction. N = 93, mean ± SD = 5.7 ± 5.5 ml.

Full size image
Figure 4
figure 4

Upper lip augmentation with PAAG in a 41-year-old woman. A: front view, before; B: front view, after, C: side view, before; D: side view, after augmentation.

Full size image
Figure 5
figure 5

Upper lip augmentation with PAAG in a 55-year-old woman. A: front view, before; B: front view, after; C: side view, before; D: side view, after augmentation.

Full size image
Figure 6
figure 6

Glabella augmentation with PAAG in a 42-year old woman. A: front view, before; B: after augmentation.

Full size image
Figure 7
figure 7

Chin augmentation with PAAG in a 50-year-old male; A: side view, before; B: after augmentation.

Full size image
Figure 8
figure 8

Nose-ridge augmentation with PAAG in a 33-year-old oriental woman, side view. A: before; B: after augmentation.

Full size image

A special sub-analysis of women with versus women without concurrent breast augmentation with PAAG was performed in order to identify any treatment-related effects of the amount of gel injected, as the volume employed in the breast augmentation procedure is considerably higher than in the facial correction procedure. No correlation was found between women with a large quantity of PAAG injected (simultaneous breast augmentation) and women with only facial gel injections regarding adverse effects or complications (data not shown).

Tolerability

Most variables investigated in the analyses of tolerability showed very small effects due to facial corrections. Though age, previous breast augmentation with PAAG (large amount of gel injected) and amount of gel injected at facial corrections potentially might influence treatment results, this was generally undetected in the present study. Thus, the very few cases of unwanted treatment effects in most cases render analyses of the above mentioned co-variables meaningless. The findings related to tolerability at physical examination of the injection site in the face are presented in Tables 2, 3, 4 and 5. The Case Record Form required specific examination for 13 other findings. The results are presented in Table 3.

Table 2 Tolerability: frequency of results from examination of skin color, skin thickness and subcutaneous fat at sites of gel injection.
Full size table
Table 3 Tolerability: frequency of symptoms as judged by examination of the treatment site after facial correction.
Full size table
Table 4 Tolerability: frequency of symptoms when palpating the skin area.
Full size table
Table 5 Tolerability: table of other results when palpating the skin.
Full size table

All these examinations demonstrated very good tolerability and low reactivity in the tissues around the sites of injection. No symptoms of the above-mentioned conditions were encountered among the 7 women with unknown birth date.

One category (palpatory regional lymph nodes, Table 5) showed a considerable number of Yes answers (approx. 12%). This was, however, not regarded as abnormal, as all of them were small, smooth, rounded and non-adherent. It was also tested if median age, median amount of gel injected at facial correction, and median time (year) since last facial correction would deviate between the ‘Yes’ and the No’ group with respect to this category. The test results appear in Table 6.

Table 6 Tolerability: results of the Mann-Whitney test for differences between patients detected positive for ‘Reg. Lymph. Nodes’ in age, total amount of gel injected at facial corrections and time since last facial correction.
Full size table

The Mann-Whitney test failed to reveal significant p-values in all the variables tested. It is therefore concluded that age, total amount of gel injected in facial corrections and time since last facial correction has no effect on the probability of finding palpable regional lymph nodes.

The same test performed with the same variables for probability of having had previous breast augmentation with PAAG did not demonstrate any correlation. In total, apart from the palpable cervical lymph nodes (12%), which we consider to be in accordance with normal incidental findings [33], and the palpatory finding in the lips (small “pearls”), the complication rate was small. Migration of the gel was demonstrated in 3 women, all in connection with the naso-labial fold, but the migration was only a few mm.

Evaluation of Cosmetic Outcome

The evaluation of the cosmetic outcome was done by both the physician and the patient on a fixed scale. The results of the evaluations are presented in Table 7 and those of the patients’ evaluations appear in Table 8. The frequencies of responses to the question “If you were going to need another facial correction would you accept a procedure with injection of PAAG again?” are given in Table 9. All 3 types of cosmetic outcome evaluation results from the 7 women were similar to those mentioned in the tables. The cosmetic outcome as judged by both the patient and the physicians thus seems to be satisfactory and in addition, the result is judged identically by patient and physician.

Table 7 reatment results: result of investigator’s global assessment of the result of facial corrections.
Full size table
Table 8 Treatment results: result of patient’s global assessment of facial correction.
Full size table
Table 9 Treatment results: Results when patients were asked “If you were going to need another facial correction would you accept a procedure with injection of PAAG again?”
Full size table

Conclusions

The present retrospective study with its thorough investigation of a large sample of patients in one center which who had undergone facial correction procedures with injections of polyacrylamide hydrogel has demonstrated that the product injected has been very well tolerated. Despite active search and questioning regarding both local side effects or systemic ones, including any possible intercurrent illness, medical treatment or large volume, we failed to demonstrate any such complications with the injection of polyacrylamide hydrogel.

Regional lymph nodes were palpable in approximately 12% of the women. Palpable cervical lymph nodes is a frequent incidental palpatory finding and in this study there was no association with the 1) age of the patient, 2) time since gel injection or 3) amount of gel injection. The finding was therefore considered unrelated to the facial PAAG injection.

Both the physician and patient rated the outcome of the facial correction procedures to be either very good or good. The examination of the patients took place up to 9 years after the facial correction, thus PAAG also seems to be a durable product for these interventions. The present study has demonstrated that polyacrylamide hydrogel seems to be a well tolerated and effective for facial augmentation. Currently, the product Aquamid® is being studied in several prospective clinical trials, one of which is completed and in the process of preparation for publication.