Introduction

Gynoid lypodistrophy also known as cellulite is a morphological alteration of the tegument tissue, directly interfering in self-esteem, with important psychological and esthetic consequences. According to the literature, cellulite seems to affect 80 to 98% of young and adult women [1].

Since the classic study by Nürnberger and Muller [2], cellulite has been a controversial issue in the health sciences. It is certainly known that important changes in the subcutaneous layer occur, such as edema, lipodystrophy, and fibrous deviation of connective tissue, involving hormonal changes and inflammatory mediators [3].

De La Casa Almeida et al. [4,5,6,7,8,9,10,11,12,13] summarized four existing theories for the etiology of cellulite that include anatomical, vascular, inflammatory, and genetic factors.

Studies using laser Doppler flowmetry found that in some patients with chronic cellulite, the skin blood flow was reduced up to 35%. Vascular disorders such as impaired or stagnant blood flow perhaps due to vasodilatation could cause hyperthermia, while fibrotic areas present lowered temperature (hypothermia) [14]. These contrasts create a very heterogeneous environment characterized by a dalmatian look.

The practical difficulties to study in vivo local vascular activity in cellulite restrict the diagnosis of the actual health status of the subcutaneous tissue. Infrared thermography has been suggested as an alternative for a more accurate diagnostic. It represents an important new tool as a non-invasive method capable of in vivo real-time evaluations of local inflammatory sites. Infrared thermography is based on the fact that human skin has a thermal homeostasis indicating normality. Any thermal change in the skin indicates that something is wrong with the body. Normally, an increase in temperature means that there is a greater local circulation that may be due to an algic or inflammatory process.

In 2013, Nkengne et al. [15] demonstrated that the thermal camera imaging is a repeatable and reproducible method that can be used to assess the severity of cellulite. In 2017, Wilczyński et al. [16] demonstrated that infrared thermography can be used as a consistent tool for cellulite analysis. The main advantages of thermography are as follows: the speed of data collection, the interpretation of images takes place in real time, radiation is not lethal in nature, and it does not require contact with the inspected part during collection.

Here, we investigate a possible correlation between the dalmatian pattern of the infrared images with the visual cellulite scale. The aim of the study was to investigate the hypothesis that the more heterogeneous tissue pattern analyzed by infrared thermography, the more severe is the cellulite grade.

Ethical procedures

This study was approved by the Ethical Committee of the University of Vale do Paraíba Register number 3.617.104 from October 2nd of 2019. After authorization by the Ethics and Research Committee and upon signing the Free and Informed Consent Form (ICF), the proposed protocol was initiated.

The study was performed at the Sensory Motor Rehabilitation Engineering Laboratory located at the Faculty of Health Sciences of the University of Vale do Paraíba (UNIVAP) in São José dos Campos. Forty female participants were selected and 60 legs were analyzed. The female participants who presented gynoid lipodystrophy in the flanks and meet the inclusion criteria of the study were included in the study. Five male participants were taken as negative controls for cellulite.

All studies involved Caucasian women volunteers, phototypes I to V. Subsequently, a cellulite Assessment Protocol (anamnesis form) was applied, which consists of addressing aspects such as clinical history, physical examination, and the classification of the degree of lipodystrophy. This instrument is easy to apply and allows to classify, in an appropriate and objective way, the gynoid lipodystrophy degree, as well as the levels of sensory changes when present.

Inclusion criteria

Body mass index (BMI = Weight / Height2) up to 29.9 kg/m2; presenting “orange peel” foci on the thighs (LDG grades 2 or 3 according to the Curri scale); having regular menstrual cycles (between 26 and 30 days) and having used the same oral contraception in the last 3 months, stable weight for at least 3 months (less than 2.0 kg variation); sedentary and without the use of cosmetics that work in the local circulation. Participants were instructed to refrain from using any cosmetic, product with retinoids, dimethylaminoethanol (DMAE), alpha-hydroxy acid (AHA), or beta-hydroxy acid (BHA) for 1 month before the start of the study.

Exclusion criteria

The following factors were considered exclusion criteria: (a) pathology or injury at the site in the lateral region of the hip; (b) possible or confirmed pregnancy; (c) metallic prosthesis close to the site of irradiation; (d) menopause; (e) any recent surgery on the spot of laser irradiation; (f) use of medications such as anti-inflammatories, corticosteroids, and antibiotics during the study period.

Image acquisition

Image acquisition was performed with the S65 camera (Flir system, Sweden). The FLIR S65 camera has a measurement range of − 20 to 120° C, accuracy of 1%, sensitivity of 0.05 °C, infrared spectral band of 7.5 and 13 μm, refresh rate of 60 Hz, autofocus and a resolution of 320 × 240 pixels. The camera was mounted on a tripod and aligned perpendicular to the surface of interest. The distance to the patient was adjusted to 50 cm, allowing you to see a wide field of view. Male subjects were evaluated as negative controls in this study. Figure 1 illustrates both thermogragraphic images of male and female members in basal conditions.

Fig. 1
figure 1

Infrared thermography of thighs of a man (left) and a woman (right) as representative image of male and female patterns for cellulite. Five men were evaluated as negative controls for thermographic images

Full size image

Environmental conditions

All measurements were carried out in a controlled environment with a temperature adjusted to 20 ± 2 °C and a relative humidity of 50 ± 5%. The images of posterior lateral views of the lower limbs (breeches) were taken in the morning period with the objective of physiological standardization.

The thermal images were analyzed one by one, using the equipment’s software. The average temperature and the maximum amplitude, corresponding to the difference between the extremes, were calculated from the original images.

Infrared thermography analysis × cellulite grade

After image acquisition and clinical evaluation by an experienced biomedical scientist specialized in esthetics, a second blinded researcher performed image evaluations. A grid of 30 points was established in each image covering the whole latero-posterior area. Each point generated one temperature value, consisting of 30 temperature values from each leg. All images were unidentified and temperature measurements were blinded. All temperature values were recorded in an Excel file for statistical analysis. The standard deviations of temperature analysis for each thigh were categorized in 8 intervals and plotted as a frequency histogram—Fig. 2 (intervals 0–0.24; 0.25–0.49; 0.5–0.74; 0.75–0.99; 1–1.24; 1.25–1.49; 1.5–1.74; 1.75–1.99).

Fig. 2
figure 2

Frequency histogram of standard deviation intervals calculated from a sample size of 60 thighs. Each thigh presented 30 points of affeered temperatures by infrared thermography. Eight equal intervals were generated and frequency distribution is represented

Full size image

A third researcher, the coordinator of the study, performed the statistical tests without access to the images and participant IDs, only the temperature values. For each leg side, we calculated the average of 30 measurements and the standard deviation. As a reflection of data dispersion, standard deviation of 30 values for each leg was tested against the cellulite grade established by the first researcher. Correlation analyses were performed, generating the Spearman value for correlation.

Phototype × cellulite grade

A second analysis was performed. The phototype of the 40 participants was evaluated by the first biomedical researcher as part of the clinical evaluation. Besides, visual cellulite grade was also recorded in the clinical anamnesis. All values were recorded in the Excel spreadsheet. At the end of the studies, skin phototype was tested against visual cellulite grade by the third researcher. Correlation analyses were performed, generating the Spearman value for correlation.

Results

Table 1 summarizes the anthropometric data of female participants. For thermographic measurements, 60 thighs with 30 temperature points each were taken into account. Figure 2 demonstrates the frequency histogram of standard deviations of thigh temperature.

Table 1 Anthropometric data of 40 female participants of the study
Full size table

Infrared thermography analysis × cellulite grade

As described in the methods, a grid of 30 points was established in each image covering the whole latero-posterior area. Each point generated one temperature value, consisting of 30 temperature values from each leg. The standard deviation of 30 points of each thigh was tested against visual cellulite grade of 60 thighs. Figure 3 demonstrates the linear regression and correlation (r Spearman). As we could observe, the present data reveals a correlation of 0.92 between standard deviation (temperature dispersion of 30 temperature points/thigh) and classical visual cellulite evaluation.

Fig. 3
figure 3

Linear regression and correlation of standard deviation of thigh temperatures × visual cellulite grade. N = 60 thighs; correlations with Spearman r were considered significant

Full size image

Phototype × cellulite grade

The skin phototypes of 40 participants were correlated to the standard deviation of 30 temperature points of each thigh. Figure 4 demonstrates the linear regression and correlation (r Spearman). As we could observe, the present data reveals a correlation of − 0.67 between standard deviation (temperature dispersion of 30 temperature points/thigh) and classical skin phototype evaluation.

Fig. 4
figure 4

Linear regression and correlation of visual cellulite grade with skin phototype. Standard deviation of thigh temperatures × visual cellulite grade. N = 60 thighs; correlations with Spearman r was considered significant

Full size image

Discussion

The female body undergoes numerous physiological changes during life. Such changes can cause biological responses, sometimes of an inflammatory nature.The concept of aggression and inflammatory response today goes much further the original ones and may include tissue ischemia, physiological imbalances like glycemic problems, emotional stress, obesity, hormonal imbalances (puberty, pregnancy, menopause, menstrual cycle), and aging. [17,18,19].

Cellulitis presents a series of pathological changes that have been compared to those that occur in the inflammatory process [5]. Edema, painful hypersensitivity, pain, temperature changes, and infiltration of inflammatory cells, among others, have been reported.. The crucial point is whether or not there is an inflammatory process in the affected region. Although cellulite is generally asymptomatic, the most severe stages can be accompanied by the appearance of painful nodules and an increase in local temperature in the dermis and subcutaneous adipose tissue [13]. A third point concerns on temperature changes in the affected area. Cellulite can cause a blood flow reduction of up to 40%, resulting in a normally cooler region accompanied by edema and fibrosis. However, very frequently, we can observe important areas of temperature increases and pain/hypersensitivity. In the present study, we presented a simple method based on infrared thermography that can be adopted in any esthetics office with a correlation of 0.92 with the visual classic evaluation, but, besides, may be very helpful to the clinician to decide which treatment will be adopted, i.e., an aggressive and inflammatory approach such as the radiofrequency of shockwave therapy or an anti-inflammatory approach such as photobiomodulation, depending on the inflammatory status of cellulite. Here, we demonstrate that together with an inaesthetic condition, a subcutaneous inflammatory problem is taking place and should be treated. Infrared thermography is a viable technology for both diagnostic and analysis of treatment effectivity.

In 2014, Yoo et al. [20] tried to establish optimized methods to evaluate cellulite grades. The authors analyzed volume of skin (cavities), dermo-subcutaneous interface length and subcutaneous thickness and concluded that such assessments could be used to evaluate the pathology, specially to evaluate the effects of cosmetics. However, the inflammatory symptoms were not taken in account. More recently, Bauer et al. [21] reported a complex method based on artificial intelligence for prediction and identification of cellulite grade. The reported method was developed using a combination of histogram of oriented gradients and artificial neural network.

At the time of this study, the authors analyzed 40 patients who met the inclusion criteria, totaling 60 limbs submitted to clinical evaluation and infrared thermography. Observing the following panel, of images obtained in the laboratory by the technique of infrared thermography, we can notice numerous points of temperature changes, that is, abnormal foci of temperature increase in the regions affected by cellulite. The heterogeneity of the images performed in the breeches of the participants, translated by the standard deviation of the temperature averages of 30 points marked in the latero-posterior region, was remarkable, especially in more advanced degrees of cellulite. The standard deviation of the mean temperatures ranged from 0.1 to 1.9, demonstrating that the regions, as well as the sides of the body, may present differences in inflammatory involvement that should be considered before the treatment election. Interestingly, the heterogeneity of skin temperature int the affected tissue, as demonstrated in our study, is strongly correlated to the classical visual cellulite scale, which made us to think about what really is happening in the subcutaneous tissue. Perhaps we should consider that temperature may be a valid biomarker of inflammation as stated at the cardinal inflammatory signs for centuries. The present study is the first to quantify and categorize de temperature heterogeneity as a valuable and consistent marker to evaluate cellulite. The previous studies [20, 21] have not attempted to the dalmatian pattern. Here, we provide additional evidences for an antique and undesirable esthetic problem that deserves to be carefully analyzed.

In conclusion, this study proposes an easy, fast, and low-cost method to aid in the diagnosis and monitoring of cellulite, based on the actual pathophysiological conditions of the affected region, helping to make decisions regarding the best treatment methodology for each patient. Thus, the diagnosis of cellulite with the aid of infrared thermography should contribute to greater assertiveness in cellulite treatments.