Abstract
Adverse effects on the oral mucosa after the use of dermal fillers have been increasingly reported due to their increased use for facial aesthetics. The objective of this study was to evaluate the clinical and histologic effects of two types of product, 10% polymethylmethacrylate and 20 mg/ml hyaluronic acid, locally and at long distance, examining initial and late reactions. Each substance was randomly and separately injected in rats’ tongues (polymethylmethacrylate, n = 16; hyaluronic acid, n = 18). They were compared with the control group ( n = 16) at 3 observation times (7, 60 and 90 days) for clinical analysis, intensity of local inflammatory response (haematoxylin and eosin staining), amount of newly formed blood vessels and macrophages (immunohistochemical assays), density of collagen fibres (picrosirius staining) and systemic migration of the product to the liver and kidney (haematoxylin and eosin staining). The results showed inflammation triggered by the injection of the material, suggesting that both substances cause responses in local tissue, although there was biocompatibility with hyaluronic acid. This research highlights the importance of experimental studies on this subject, since adverse reactions have been observed routinely in dental practice.
Recently, dental surgeons have seen patients with nodular lesions particularly in the labial or buccal submucosa, resembling inflammatory or neoplastic diseases of the minor salivary glands. Biopsy and microscopic evaluation of specimens has revealed that these lesions show an exogenous material associated with a foreign body reaction. These products are dermal fillers used by dermatologists and plastic surgeons to restore the volume of subcutaneous tissue lost during the ageing process or to enhance soft tissue, such as the lips. There are reports of numerous cases of foreign body reaction in the orofacial region after injection of filling materials. It has become increasingly common to use these cosmetic procedures to maintain a youthful facial appearance. This technique is chosen mainly because it is minimally invasive and more affordable when compared to traditional facelift surgery.
The reaction caused by the presence of an exogenous substance in tissues takes the form of an inflammatory response that begins with the influx of neutrophils, causing pain and exudation. Later, lymphocytes, plasma cells and macrophages are detected at the site, with the possible formation of giant cells. In this case, a foreign body reaction takes place. On the periphery of the inflammation area there are intense signs of fibroplasia and neovascularization.
In addition to the reactions that occur at the site of injection of these substances, there is also migration. This is a controversial issue in the literature. Most papers on the migration of a foreign body over a distance do not specify its precise location. A study by Rosa and de Macedo noted the presence of hepatic and renal inflammatory infiltrates in rats that received different filling materials in the ear. The authors suggest that these substances could act at a distance in organs of metabolism and excretion.
The most widely used dermal fillers are polymethylmethacrylate (PMMA) and hyaluronic acid (HA). PMMA is long-lasting in tissues due to the body’s difficulty in degrading its constituents. HA is resorbable, with limited permanence, ranging from 6 to 9 months. Both are used for filling wrinkles and correcting nasolabial folds, as well as soft tissue augmentation. PMMA is also applied in the treatment of facial lipodystrophy in HIV patients treated with antiretrovirals.
Most studies have mentioned the reactions resulting from the use of facial filling materials based almost exclusively on case reports. Recent research suggests the involvement of inflammatory cells such as lymphocytes, macrophages and giant cells in the immune response to dermal fillers. The aim of this study was to evaluate the clinical and histopathologic reactions to two different filling materials at early and late study times, using a rat model. The intention was to investigate the association between the type of dermal filler (PMMA and HA) and the pattern of inflammatory response.
Materials and methods
A longitudinal randomized study was conducted using 50 Wistar female rats ( Rattus norvegicus ) from the same breeder. All procedures were performed according to institutional standards for the care and use of experimental animals after approval by local Ethics Committees.
The rats were kept in cages placed on ventilated shelves with a controlled temperature of 22 ± 1 °C and light–dark cycles of 12 h. They were fed with Nuvilab-CR1 and given filtered water ad libitum .
At the beginning of the experiment, each animal weighed about 200 g and had an average age of 2 months. They were randomly divided into 3 groups according to the indicated treatment: PMMA ( n = 16); HA ( n = 18) and control (NaCl 0.9%) ( n = 16). Each group was subdivided into 3 experimental times of 7, 60 and 90 days, according to the interval between treatment and death.
The animals were weighed prior to subsequent anaesthesia using an intraperitoneal injection of 20 mg/ml xylazine hydrochloride (0.05 ml/100 g) and 50 mg/ml (ketamine hydrochloride 0.1 ml/100 g). Once anaesthetized, the rats were placed in a supine position on the operating table for the injection of the test substance. The needle (26G ½, 13 × 4.5) was cautiously introduced in the tissues with its bevel facing up and the long axis as parallel as possible to the mucosa. The submucosa was dissected at 7 mm and standardized with the use of an endodontic cursor. Thus, 0.07 mL of each substance was introduced in the middle third of the ventral tongue, 2 mm to the left of the midline and 7 mm to the front of the frenum.
Clinical alterations
At each observation time, the animals were sedated for clinical evaluation and the presence or absence of fundamental lesions such as plaques, papules and ulcerations was determined.
After clinical evaluation, the animals were killed with isoflurane, and subsequently, the tongue, right medial lobe of the liver and right kidney were removed. These tissue fragments were fixed in 10% neutral buffered formalin and processed for haematoxylin and eosin (H–E) staining. Tongue samples were also submitted to immunohistochemical analysis and picrosirius staining.
Training was conducted with an experienced pathologist aiming to standardize the criteria for analysis. Before reading the slides, the examiner was calibrated and blinded to the experimental groups.
For microscopic analysis of the tongue in the PMMA and HA groups, slides with identifiable exogenous material were considered viable, and evaluation of the tissue near the implant was made. In the control group, the area being evaluated corresponded to the anatomic region where the material was applied.
Intensity of inflammatory response
A biological microscope (Zeiss ® Axioskop 40, Carl Zeiss, Jena, Germany), at magnifications of 100 times, 200 times and 400 times was used to assess the presence of inflammatory cells (neutrophils, eosinophils, lymphocytes, plasma cells, macrophages and giant cells) qualitatively in the tongue sections stained with H–E. Based on criteria used in other studies, the observation of certain characteristics defined the inflammation score, considering the area where there was the most intense tissue response: 0 absent, absence of inflammation; 1 mild, presence of sparse mononuclear cells; 2 moderate, presence of lymphoplasmacytic infiltrate and/or sparse neutrophils and eosinophils; and 3 intense, presence of neutrophil and eosinophil infiltrate.
Blood vessel neoformation and number of macrophages
Immunohistochemical reactions in the tongue sections were carried out using a streptavidin–biotin complex with the primary antibodies: anti-CD34 (Novocastra, Newcastle, UK, 1:250 dilution) and anti-CD68 (Novocastra, Newcastle, UK, 1:200 dilution). Positive controls were prepared using histological sections of the rat intestine (CD34) and lung (CD68). In the negative control, there was omission of the primary antibody.
The slides were viewed with a microscope (Zeiss ® Axioskop 40, Carl Zeiss, Jena, Germany) coupled to a camera (Cool Snap-Pro cf, Media Cybernetics, Bethesda, USA) connected to a computer (Dell ® Optiplex GX620 model, Round Rock, USA). For each rat, 10 microscopic fields were captured at the highest magnification (400 times) for each marker. A manual count of the structures was performed in each field of 166 μm 2 considering the following criteria of positivity: wall thickness and size of blood vessel in slides stained for CD34 and cellular morphology of macrophage in slides stained for CD68.
Density of collagen fibres
It was possible to capture 3–5 fields of slides with tongues stained with picrosirius using a polarized light microscope (Zeiss ® Axioskop 40, Carl Zeiss, Jena, Germany) coupled to a camera (Cool Snap-Pro cf, Media Cybernetics, Bethesda, USA) connected to a computer (Dell ® Optiplex GX620 model, Round Rock, USA). The chosen area covered the largest amount of collagen fibres by using the 100 times magnification. Images were transported to Image-Pro Plus ® , version 4.5.1 (Media Cybernetics, Inc.; 2005), where the collagen percentage, represented by red birefringence, was calculated for the total slide area (673 μm 2 ).
Migration
Migration was evaluated in the liver and kidney slides stained with H–E using a biological microscope (Zeiss ® Axioskop 40, Carl Zeiss, Jena, Germany) at 100 times, 200 times and 400 times magnifications, based on the presence or absence of inflammatory response or traces of injected material.
Statistical analysis
All data were tabulated and analyzed using the SPSS 17 software (SPSS Inc., USA), with the Kruskal–Wallis non-parametric test, complemented by its multiple comparisons test, at a significance level of 5%, since the data did not fit a normal distribution.
Results
During the experiment, 4 animals died: 2 in the PMMA group (1 at 60 and another at 90 days) and 2 in the control group at the last observation time. Therefore, the final sample consisted of 50 rats.
Clinical alterations
In clinical analysis at 7 days, all animals showed lesions in the form of ulcers close to the PMMA injection site, while no change was found in the HA group. At 60 days, 1 of 5 rats in the PMMA group showed papules and 50% of those in the HA group developed white plaques in the region. When assessed 90 days after injection of the material, both test groups showed white plaques: 2 animals from the PMMA and 1 from the HA group. The control group did not show any clinical alteration at the 3 observation times.
For histological analysis, the intra-examiner calibration was performed using the intraclass correlation by re-analysis of 30 slides or fields with an interval of 7 days. There was excellent correlation between the readings related to the intensity of inflammatory response ( P = 0.904), blood vessel neoformation and number of macrophages ( P = 0.967) and density of collagen fibres ( P = 1.000).
Intensity of inflammatory response
At 7 days, the inflammatory response at the site of injection of PMMA was intense ( Fig. 1 a) , differing significantly from the moderate response at other times ( P = 0.004) ( Fig. 1 b). Animals treated with HA displayed a distinct behaviour: in the first week, they showed a response that ranged from moderate to intense ( Fig. 2 a) , dropping significantly by 90 days, when a mild to moderate inflammatory process was observed ( P = 0.018) ( Fig. 2 b). No changes were observed in the control group ( Table 1 ).
| Material | Score | Time | ||
|---|---|---|---|---|
| Day 7 | Day 60 | Day 90 | ||
| PMMA | 0 – Absent | 0 | 0 | 0 |
| 1 – Mild | 0 | 0 | 0 | |
| 2 – Moderate | 2 | 5 | 5 | |
| 3 – Intense | 4 | 0 | 0 | |
| Total ( n ) | 6 | 5 | 5 | |
| Mean rank | 11.83 A | 6.50 B | 6.50 B | |
| HA | 0 – Absent | 0 | 0 | 0 |
| 1 – Mild | 0 | 1 | 3 | |
| 2 – Moderate | 3 | 5 | 3 | |
| 3 – Intense | 3 | 0 | 0 | |
| Total ( n ) | 6 | 6 | 6 | |
| Mean rank | 13.50 A | 8.75 AB | 6.25 B | |
| Control | 0 – Absent | 6 | 6 | 4 |
| 1 – Mild | 0 | 0 | 0 | |
| 2 – Moderate | 0 | 0 | 0 | |
| 3 – Intense | 0 | 0 | 0 | |
| Total ( n ) | 6 | 6 | 4 | |
| Mean rank | 8.50 A | 8.50 A | 8.50 A | |
The test groups showed a significantly higher inflammatory response than the control group at 7, 60 and 90 days. PMMA tended to have more intense inflammatory scores than HA, though no statistically significant difference could be seen ( P = 0.750) ( Table 2 ).
| Time | Score | Material | ||
|---|---|---|---|---|
| PMMA | HA | Control | ||
| Day 7 | 0 – Absent | 0 | 0 | 6 |
| 1 – Mild | 0 | 0 | 0 | |
| 2 – Moderate | 2 | 3 | 0 | |
| 3 – Intense | 4 | 3 | 0 | |
| Total ( n ) | 6 | 6 | 6 | |
| Mean rank | 13.00 A | 12.00 A | 3.50 B | |
| Day 60 | 0 – Absent | 0 | 0 | 6 |
| 1 – Mild | 0 | 1 | 0 | |
| 2 – Moderate | 5 | 5 | 0 | |
| 3 – Intense | 0 | 0 | 0 | |
| Total ( n ) | 5 | 6 | 6 | |
| Mean rank | 12.50 A | 11.58 A | 3.50 B | |
| Day 90 | 0 – Absent | 0 | 0 | 4 |
| 1 – Mild | 0 | 3 | 0 | |
| 2 – Moderate | 5 | 3 | 0 | |
| 3 – Intense | 0 | 0 | 0 | |
| Total ( n ) | 5 | 6 | 4 | |
| Mean rank | 11.50 A | 8.75 A | 2.50 B | |
The evaluation of the presence of giant cells showed that this cell type was present in the PMMA group at all observation times ( Table 3 ).
| Material | Time | |||
|---|---|---|---|---|
| Day 7 | Day 60 | Day 90 | ||
| PMMA | Presence of giant cells (# of rats) | 5 | 3 | 4 |
| Total ( n ) | 6 | 5 | 5 | |
| HA | Presence of giant cells (# of rats) | 1 | 0 | 0 |
| Total ( n ) | 6 | 6 | 6 | |
| Control | Presence of giant cells (# of rats) | 0 | 0 | 0 |
| Total ( n ) | 6 | 6 | 4 | |
Stay updated, free dental videos. Join our Telegram channel
VIDEdental - Online dental courses
