Russian Medical Review
ISSN 2311-7729 (Print), 2619-1571 (Online)

Russian Medical Review

E-library РГБ
VAK russian citation index

Impact factor - 0,378*

* Impact factor according to the SCIENCE INDEX 2017

Паттерн красивого лица. Не описанные ранее пропорции


Orthognathic surgery is aimed to correct a bad bite and to deal with respiratory dysfunctions but also to make a more beautiful face. The latter gains more importance as the means of achieving the first develop and evolve. As pre-surgical planning techniques improve and surgical procedures become widespread, we are close to resolving a problem of restoring lost functions while aesthetic goal comes to the fore. However, unpredictable aesthetic outcome remains an important issue [1, 2].

It increasingly applies to orthognathic procedures designed to treat severe snoring and sleep apnea that require significant maxillomandibular advancement. Patient dissatisfaction with aesthetic outcomes in this group is close to 30% [3].

A formalized description of a beauty was a concern throughout the human existence [4]. Ideal facial proportions were described as early as in the Ancient Greece and specified during the Renaissance. The appearance of orthodontia was the next step of the development of the doctrine of the beauty as doctors were able to interfere with facial proportions and the general perception of face. E.H. Angle, the father of orthodontia, focused on the classical ancient parameters using the Apollo Belvedere as a benchmark [5]. In 1937, E.H. Wuerpel proposed the classification of facial types (Greek, Roman, and Mongoloid) and argued that every facial type had its own laws of beauty [6].

The next step of the development of facial beauty doctrine was the addition of the analysis of bony structures to the analysis of soft tissues. R.M. Ricketts described optimal lip-chin proportions and their relation to the facial angle [7]. However, these proportions vary greatly in different nationalities and races [8-10]. The patterns found failed to address a practical challenge of developing the rules of the construction of facial beauty. As a result, there was a doubt regarding the achievement of this goal. In 1964, R.S. Hambleton suggested that it is impossible to find out a formula that satisfies each orthodontist. He agreed with J.D. Subtelny (1961) who believed that just a doctor’s “internal opinion” exists [11-13].

In 2004, T.A. Auger claimed that the perception of facial beauty changes depending on the society and time [14]. The allegation of the individual perception of facial beauty resulted in the attempts to refrain fr om the term “beauty” in favor of such terms as “attractiveness”, “charm” or “well-balanced face”. G.W. Arnett formulated that facial beauty combines two components, i.e., the quality of specific parts (eyes, eyebrows, lips etc.) and the balance of face parts. Unsuccessful outcomes are accounted for by a surgeon’s attempt to interpret facial proportions based on his/her subjective perceptions but not careful measurements [15].

Preliminary phase of the study

An error of the searching for the patterns of facial beauty should lie in the very beginning of the reasonings, in the first step. Meanwhile, the first step of the analysis of any 3D shape is to define a coordinate system. There are no problems in facial height or width. However, the position of the horizontal line produces a great controversy. Different analyses use the Frankfort horizontal plane that passes through the inferior borders of the bony orbits and the upper margins of the auditory meatus, the sella-nasion plane that passes through the sella turcica and the bridge of the nose, Camper's plane that passes through the nasal alar and bilateral tragus etc.

All these planes were constructed in a similar fashion, i.e., two points were selected on a photo or X-ray image and connected with a line. Meanwhile, there are no actual facial horizontal structures which comply with these planes. The Fox plane that is perpendicular to the true vertical line (TVL) and is independent of any benchmark is somewhat different. Moreover, it was demonstrated that none of the horizontal structures anchored to the facial surface were described earlier. The term “proportionality” implies the need of a benchmark to which the proportionality is evaluated. It should be a linear structure that remains visible irrespective of head position.

Tele X-ray images were excluded from the search at once. Their analysis was performed since 1887 when the horizontal line introduced by Virchow was accepted in Frankfurt as a major benchmark for skull analysis. Since then, a variety of methodologies of working with X-ray images emerged. Therefore, it’s highly unlikely to find something new [2].

Photos are equally significant in history and get an attention from the orthodontists. Soft tissue points are used in the majority of tele X-ray image analyses. A photo is the key method of determining the true facial vertical to perform all measurements during presurgical planning. The analysis of both photos and tele X-ray images is virtually perfect. The only question is what a photo is missing. It is worth recalling that we are talking about science photos. Background choice and the position of additional light aim to eliminate the shadows [16]. Therefore, the lack of the shadows is an important hallmark of science photos. Meanwhile, shadow analysis is a correct direction for further search.

3D photography provides an opportunity to study the position of the shadows depending on the light source. The analysis of 3D photos from this viewpoint allowed for identifying previously undescribed facial structure, i.e., the belt of the game of light and shade that passes along the line connecting the following regions (see Fig. 1):

·               a prominent area in the posterior portion of the zygomatic arch corresponding to the superior border of the zygomatic process of the temporal bone with the superimposed superior pole of the parotid gland, anterior auricular muscle, temporal extension of Bichat fat pad, and temporal fat pad;

·               the area of the maximum projection of the zygomatic bone formed by the mutual imposition of the prominent area of the zygomatic bone, zygomatic mimic muscles, and zygomatic fat package;

·               soft tissue thickening in the lateral surface of the nose formed by the mutual imposition of the elevator muscle of upper lip and wing of nose and the superior portion of the nasolabial fat package.

Рис. 1. «Пояс» светотени при косом освещении Fig. 1. Chiaroscuro in oblique illumination

The belt is characterized by a volatile width and some deviations from the line. Moreover, it cannot be seen on science photos. As a result, a scientific description was never afforded to the belt of the game of light and shade.

The belt divides the face into lower and upper portions thus determining its actual shape. Facial projection in the projection of the belt has a maximum area decreasing upward and downward. At side face photos, the belt is projected to the line running from the most lateral point on the alar contour to the base of the inferior crus of the antihelix.

In jewelry, girdle is a thin perimeter dividing a diamond into upper and lower parts and determining its size. It is generally accepted in Russia to use Dutch jewelry terminology. We introduced a term “rondist” (Ro) to denote the above-described belt and a term “rondist plane” to denote a plane that passes through rondist.

The resulting plane meets our requirements as it is applied directly to face. It is visible irrespective of head posture and is automatically fixated by the gaze as a line. Finally, this plane divides the head into upper and lower parts. Rondist plane can be used as a horizontal plane in the coordinate system for describing facial shape.

Since the in-depth analysis of tele X-rat images was already introduced, there are no chances to find any novel patterns in this field. However, all methods share some properties, i.e., they are limited by the level of glabella. As a result, nasion becomes a reference point to determine the anteroposterior position of teeth and mandilble. Therefore, the shape of neurocranium is completely excluded from the facial analysis. Meanwhile, forehead is a part of the face. To prevent the subjectivity when selecting benchmark location, tuber frontale (Tf) point was introduced. This point is created by the contact of the frontal contour with the line oriented at an angle of 45° to rondist plane. Further, three planes being perpendicular to rondist were selected:

·               subnasale (Sn) plane going through a point wh ere the nasal septum and the upper lip meet in the midsagittal plane;

·               glabella (Gb) plane going through the prominence between the eyebrows;

·               tuber frontale (Tf) plane going through the upper margin of the frontal bone.

The ratio of these planes in the anteroposterior direction defines the proportions of the upper portion of face, i.e., its shape. The distance between the planes determines whether a face appears flat or convex. A profile depth (PD) is the ratio of Sn-Gb plane to Gb-Tf plane (see Fig. 2).

Рис. 2. Анализ профиля лица с использованием плоско- сти рундиста (Ro). Плоскость рундиста (Ro) проециру- ется на линию, соединяющую верхний край крыла носа и нижнюю ножку противозавитка. Перпендикулярные ей плоскости проходят через точки основания носа S

Inclination that determines the profile (distal or mesial) is of crucial importance for the lower one-third of face. Considering that this line is natural for the perception by an ordinary person, it should connect the chin and the upper lip. The upper lip is seen as a single entity, therefore, a point corresponding to the middle was chosen. This levels the effect of the protrusion of the maxillary incisors. The angle between this line and Ro was referred to as a lower face inclination (LFI).

The primary analysis of photos has revealed that in beautiful female face the profile depth and the distance between Sn, Gb, and Tf planes are almost equal while the angle between Ro and LFI is about 90°. This rule has provided the basis for our study (see Fig. 2).


To develop a universal cephalometric analysis of a beautiful female face and a simple method of the planning of orthognathic surgery for correcting malocclusions, improving airway permeability, and achieving a good stable esthetic outcome.

Patients and Methods

Traditionally, a beauty is assessed by presenting photos to respondents [17]. They make up a group of beautiful faces or, alternatively, the investigator determines beautiful and ugly faces. Meanwhile, the effect of the physiological factors on facial beauty is not eliminated. As to our study, we assume that the most beautiful women have similar health features and the levels of estrogens. Moreover, they have such facial proportions that allow to choose them from millions of faces. By using international beauty ratings, we eliminated the need to assess the individual beauty whereby receiving a large group of faces similar in their attractiveness and harmony.

Meta-analysis of open access photos was performed. Internet ratings emerged from the survey of multiple respondents were used to sel ect the faces to be analyzed. In other words, we performed the analysis of the faces considered as beautiful ones by the majority of people.

The following groups were enrolled in the study.

Group 1 (study group) included the members of the list of the most beautiful Caucasian women in the modern world (hereinafter – models) according to (top 50 most beautiful women in the world 2020) and (top 18 most beautiful Russian women).

To assess the universality of identified facial proportions, the photos of the most beautiful women of distinct races and epochs were analyzed.

Group 2 included the 20th century's most beautiful women according to (Elle’s the most beautiful women of all time, 27 faces).

Group 3 included 19 Asian women from the list of the most beautiful Asian models according to

Group 4 included 18 Negroid women from the list of the most beautiful Negroid models according to

Side portrait image for each face was found on the Web. If no side portrait image with appropriate head posture and good resolution was found, the model was excluded from the study. The photos were printed on A4 paper. The next step was to overlay Ro plane, frontal planes (Tf, Gb, and Sn), and LFI plane. The distances between the frontal planes, their ratio (PD), and the angle between Ro and LFI were measured. The data obtained were summarized in an Excel table. Statistical analysis was performed using StatFi software (AnalystSoft Inc., USA). The mean, standard deviation and standard error, kurtosis, and dispersion were calculated for each group. The Lilliefors (Kolmogorov-Smirnov) test and Shapiro-Wilk test were used for evaluating normality. Considering deviations from a Gaussian distribution, the values obtained were compared between the groups for their belonging to the general population using Mann-Whitney U-test.


As Table 1 illustrates, in group 1, PD was, on average, 0.97 (95% CI: 0.94-0.99). The asymmetry of -0.12 indicates symmetrical distribution while the kurtosis of 5.4 (the kurtosis of normal distribution is 3) means that the distribution produces rare extreme outliers. The normal distribution was confirmed by the Kolmogorov-Smirnov test. However, the Shapito-Wilk test rejected the hypothesis of normality due to high kurtosis. Considering this, nonparametric tests were used to compare the groups. LFI was 93.3° (95% CI: 92.0-94.5). The distribution was symmetrical, the kurtosis was 2.54. The normal distribution was confirmed. Data for other groups are listed in table 1. The normal distribution of PD and LFI was confirmed by the Lilliefors (Kolmogorov-Smirnov) test and Shapiro-Wilk test.

Таблица 1. Анализ значений глубины профиля (PD) и на- клона нижней трети лица (LFI) по группам Table 1. The analysis of profile depth (PD) values and lower facial inclination (LFI) in the groups

No significant differences in PDI and LFI between all four groups were revealed using Mann-Whitney U-test (p>0.05). Therefore, all investigated parameters belong to the same general population (see Table 2).

Таблица 2. Проверка принадлежности значений глубины профиля (PD) и наклона нижней трети лица (LFI) в груп- пах женщин различных эпох и рас к одной генеральной совокупности Table 2. The test of belonging of profile depth (PD) values and lower facial inclin


The concept of beauty

The perception of beauty is diverse. Beauty is not easy to lay a finger on. There are many definitions of beauty, however, none of them is a universal one. Currently, three concepts of facial beauty are discussed in scientific papers. The first one, physiological, postulates that facial beauty is determined by a woman’s health and estrogen levels. The second concept, intermediate, claims that people like average faces. And the third concept hypothesizes that beautiful faces follow the golden ratio. The possible explanation is the mixing of two independent processes occurring in the brain. The first process is older and utilizes the hypothalamus and limbic system. This is an emotional process related to the function of reproduction. The brain responds to the signs of health and high levels of estrogens (attractive, in the Russian language). The second process is new and utilizes the cerebral (mainly prefrontal) cortex. When the objects with complex internal proportions are demonstrated, this pathway is activated to compare the observed object with an ideal image available in the human brain (harmonic, in the Russian language) [18]. This dialectic approach allows for reconciling conflicting opinions and further studying of each of the processes without mixing them.

The hypothesis of an average face was corroborated by many studies. A composite facial photo created by combining 32 photos is perceived as a more attractive than any original photo. However, this is contrary to the “domestic experience” since there are plenty beautiful faces differed from each other. One of the versions of the rightness of these hypotheses is accounted for by the fact that the photos of one type are initially taken. As a result, the deviations from the proportions that are optimal for brain’s perception are minimized.

Face as a three-dimensional shape

All proportions are traditionally measured on the photos and images. Meanwhile, the projections on the frontal plane, not real distances, are measured; the projection of a planar and 3D object is equal. Hence, M.J. Gunson has recently refused from the practice to demonstrate postoperative en face and side portrait photos to assess postoperative results claiming that only a ¾ view provides the insight into the true esthetic outcome [19]. Searching for the natural horizontal plane and the third axis is useful fr om this viewpoint. We have demonstrated that the majority of beautiful faces follow the same template. Mean PD is 0.97, variations are rare. LFI is more variable but still demonstrates the normal distribution and is, on average, 93°. The distribution is asymmetrical, there are large variations of values. Therefore, the perception of a beautiful face allows for the mesialization of a profile.

The study on modern beautiful female faces of three races and female faces considered beautiful before 21stcentury demonstrates similar PD and LFI distributions comparable to group 1 (see Fig. 3).

Рис. 3. Паттерн красивого лица. Объемные характеристи- ки идентичны у красивых лиц, принадлежащих женщи- нам разных рас и эпох Fig. 3. The pattern of facial beauty. Beautiful faces of distinct races and historical epochs have equal volumetric characterist

A method of orthognathic surgery planning was developed based on these findings [21].


The introduction of rondist plane regarded as a null horizontal plane for facial analysis has revealed previously undefined regularities explaining the fact that the perception of a female facial beauty by people of different nationalities are similar. Cephalometric analysis has also identified the criteria related to 3D characteristics that are intuitively defined by the majority of people. It was demonstrated that beautiful faces follow the same template with PD closer to 1.0 and LFI closer to 90°. Moreover, these parameters are independent of the race and historical epoch.

PD and LFI values can be easily changed by orthognathic surgery thus providing novel perspectives for its planning to achieve stable functional and esthetic outcomes.

About the authors:

Mikhail M. Solov’ev — Cand. of Sci. (Med.), Associate Professor of the Department of Maxillofacial Surgery and Surgical Stomatology, I.P. Pavlov First St. Petersburg State Medical University, 6–8, Lev Tolstoy str., St. Petersburg, 197022, Russian Federation; Head of the Department of Maxillofacial Surgery, City Multidisciplinary Hospital No. 2, 5, Uchebnyy lane, St. Petersburg, 194354, Russian Federation; ORCID iD 0000-0003-2305-3466.

Elena B. Katinas — Cand. of Sci. (Med.), Associate Professor of the Department of Otorhinolaryngology with Clinics, I.P. Pavlov First St. Petersburg State Medical University, 6–8, Lev Tolstoy str., St. Petersburg, 197022, Russian Federation; ORCID iD 0000-0002-7898-0219.

Irina A. An — maxillofacial surgeon, CJSC “Medical Center RAMI”, 13, Kirochnaya str., St. Petersburg, 191014, Russian Federation; ORCID iD 0000-0001-5102-9901.

Contact information: Mikhail M. Solov’ev, e-mail: Financial Disclosure: no authors have a financial or property interest in any material or method mentioned. There is no conflict of interests. Received 31.07.2020, revised 14.08.2020, accepted 25.08.2020.

1. Kosyreva T.F. Esthetics of the face and its analysis. Educational and methodical manual of the Department of orthodontics and children’s prosthetics MMSI. М.; 1996.
2. Khoroshilkina F. Ya. Elimination of functional, morphological and aesthetic disorders in the treatment of dental-facial anomalies edgeways-technique. Pumpa; 1995.
3. Islam S., Aleem F., Ormiston I.W. Subjective assessment of facial aesthetics after maxillofacial orthognathic surgery for obstructive sleep apnoea. Br J Oral Maxillofac Surg. 2014;53:235–238. DOI: 10.1016/j.bjoms.2014.11.018.
4. Fadeev R.A., Izpravnikova A.N. Results of expert evaluation of violations of facial aesthetics in various forms of dental anomalies. Institut stomatologii. 2009;45(4):21–25.
5. Angle E.H. Treatment of malocclusion of the teeth and fractures of the Maxillae, Angle’s system. White Dental Manufacturing Company, Philadeiphia; 1900.
6. Wuerpel E.H. On facial balance and harmony. Angle Orthod. 1937;7:81–89.
7. Ricketts R.M. Esthetics, environment, and the law of lip relation. Am J Orthod. 1968;54(4):272–289. DOI: 10.1016/s0002-9416 (68) 90278-9.
8. Erbay E.F., Caniklioğlu C.M., Erbay S.K. Soft tissue profile in Anatolian Turkish adults: Part I. Evaluation of horizontal lip position using different soft tissue analyses. Am J Orthod Dentofacial Orthop. 2002;121(1):57–64. DOI: 10.1067/mod.2002.119780.
9. Alam M.K., Mohd Noor N.F., Basri R. et al. Multiracial Facial Golden Ratio and Evaluation of Facial Appearance. PLoS One. 2015;10(11): e0142914. (Published 2015 Nov 12). DOI: 10.1371/ journal.pone.0142914.
10. Saurabh R., Piyush B., Sourabh B. et al. Assessment of facial golden proportions among central Indian population. J Int Soc Prev Community Dent. 2016;6(Suppl 3):182–186. DOI: 10.4103/2231-0762.197188.
11. Case C.S. A practical treatise on the technics and principles of dental orthopedia. DOI: 10.1016/s0099-6963 (23) 80019-9.
12. Hambleton R.S. The soft-tissue covering of the skeletal face as related to orthodontic problems. Am J Orthod. 1964;50(6):405–420. DOI: 10.1016/0002-9416 (64) 90204-0.
13. Subtelny J.D. A longitudinal study of soft tissue facial structures and their profile characteristics, defined in relation to underlying skeletal structures. Am J Orthod. 1959;45(7) 481–507. DOI:10.1016/0002–9416 (59) 90014–4.
14. Auger T.A., Turley P.K. The female soft tissue profile as presented in fashion magazines during the 1900s: a photographic analysis. Int J Adult Orthodon Orthognath Surg. 1999;14(1):7–18.
15. Arnett G.W., Bergman R.T. Facial keys to orthodontic diagnosis and treatment planning. Part I. Am J Orthod Dentofacial Orthop. 1993;103(4):299–312. DOI: 10.1016/0889-5406 (93) 70010-L.
16. Schaaf H., Streckbein P., Ettorre G. et al. Standards for digital photography in cranio-maxillo-facial surgery-Part II: Additional picture sets and avoiding common mistakes. J Craniomaxillofac Surg. 2006;34(7):444–455. DOI: 10.1016/j.jcms.2006.08.002.
17. Solov’ev M.M., Aleksina L.A. Assessment of the emotional and aesthetic perception of the face from photographs and the influence of photogrammetry indicators on it. Institut stomatologii. 2011;4(53):88–89.
18. Zeki S., Romaya J.P. The Brain Reaction to Viewing Faces of Opposite and Same-Sex Romantic Partners. PLoS ONE. 2010;5(12): e15802. DOI: 10.1371/journal.pone.0015802.
19. Gunson M.J., Arnett G.W. Orthognathic virtual treatment planning for functional esthetic results. Seminars in Orthodontics. 2019;3(25):230–247. DOI: 10.1053/j.sodo.2019.08.008.
20. Soloviev M.M. Method for planning the position of the upper jaw during orthognathic operations RF Patent: RU 2 709 832 C1. Saint-Petersburg. 29.03.2019.

License Creative Commons
This work is licensed under a Creative Commons «Attribution» 4.0 License.

Previous article
Next article