Salivary cortisol levels in women with muscular facial pain


Objectives: the present study was conducted to identify specific biological, clinical and psychological variables in women with chronic muscular facial pain. Materials and Methods: Twenty-three women with myofascial facial pain were included into the case group, whereas thirty healthy women constituted the control group. Basal salivary cortisol levels at 08.00, 16:00 and 24:00 h were the main biological variables considered; the clinical variables detected were Pericranial Tension Score (PTS), Cervical Tension Score (CTS) and Visual Analogue Scale (VAS); the psychological variables considered were depression, anxiety, stress and alexithymia. Results: No significant cortisol level difference was detected between the two groups. Psychometric scales showed a significant difference between the two groups (P=0.01). A VAS cut off value (60) separated patients with higher cortisol value from those with a normal cortisol (fair discriminant ability 0.7). The VAS cut off value was not able in discriminating patients on the basis of the PTS and CTS scores. Conclusions: considering the relationship among cortisol and stress-related diseases, the VAS score revealed a discriminant ability in detecting patients with higher cortisol value. Clinical Relevance: VAS could represent a potential clinical indicator for stress-related temporomandibular disorders.


Comba Benedetta, DDS, Resident, Department of Surgical Sciences, Specialization School of Orthodontics, Dental School, University of Torino, Via Nizza 230 Torino, Italy

De Giorgi Ilaria, DDS, Visiting Professor, Department of Surgical Sciences, Gnathology Unit, Dental School, University of Torino, Via Nizza 230 Torino, Italy

Castroflorio Tommaso, DDS, PhD, Visiting Professor, Department of Surgical Sciences, Specialization School of Orthodontics, Dental School, University of Torino, Via Nizza 230 Torino, Italy

Deregibus Andrea, MD, PhD, Adjunct Professor, Department of Surgical Sciences, Gnathology Unit, Dental School, University of Torino, Via Nizza 230 Torino, Italy


Temporomandibular disorders (TMDs) are a heterogeneous group of disorders involving masticatory muscles, temporomandibular joint and related structures [1]. Muscle disorders are most frequently seen in community samples. There is also general support that biopsychosocial models and methods [2], which reflect integration of biomedical diagnostic and treatment methods as well as assessment of psychological status and psychosocial level of function best support their assessment and management [3-4]. On the basis of the biopsychosocial model stresses are important factors in TMD insurgence and perpetuation. The “stress system” is a complex and sensitive system, composed by the Sympathetic Nervous System (SNS), which acts trough adrenaline or noradrenaline secretion, and the Hypothalamus-Pituitary-Adrenal Gland axis (HPA), which acts trough slow secretion of glucocorticoid hormones [5].  Both catecholamines and glucocorticoids show protective and harmful effects: in the short period they are fundamental for adaptation, homeostasis maintenance and survival, whereas in the long period they could become negative and accelerate the pathological process [6-7]. An excessive and prolonged glucocorticoid response is considered a risk factor for stress-related diseases and, consequently, a prolonged stress exposition could lead to psychophysical diseases [8-10] and to HPA axis alterations [11-14]. Equally, HPA axis dysfunctions may contribute to chronic pain growth [13]. Recent works support the hypothesis that HPA axis is implicated in the physiopathology of muscular facial pain [15]. Higher cortisol levels have been associated to TMDs [16-20]. Thus women affected by muscular facial pain could present HPA axis alterations and reduced management capacities under high levels of stress [21-23].

Considering the potential role of HPA axis alterations in the aetiology of muscle related TMDs, the aims of this study were :1) to compare basal salivary cortisol levels variations between women with muscular facial pain and healthy women and 2) to evaluate the eventual relationship between salivary cortisol (biological variable), pain perception and muscle tension (clinic variables) and psychometric test (psychological variables). The study was conducted to answer the following 2 clinical-research questions:

  1. “Is the myofascial facial pain associated to cortisol level?”.
  2. “Is it possible to identify diagnostic clinical indicators of cortisol level?”

Positive answers need to be sought in the light of clarifying the role of HPA axis in the pathophysiology of muscular related TMDs and in the light of improving the clinical diagnostic ability in the daily routine.

Material and Methods: 


Twenty-three women consecutively referring to the Gnathological Unit of the Dental School of “Città della Salute e della Scienza Hospital (University of Turin)” and fulfilling the RDC-TMD diagnostic criteria for Group Ia and Ib Axis I (Myofascial Pain without and with limitation of the mouth opening) were enrolled (mean age 37,9) [24]. All patients suffered from muscular chronic pain from at least six months (P Group). Exclusion criteria were (a) systemic diseases (endocrine, nervous, cardiovascular, liver, and kidney diseases, acute or chronic inflammatory disorders and headache) (b) diagnosis of Group II and III according to RDC-TMD, (c) history of craniofacial trauma and (d) facial pain medication overuse.

Thirty healthy women (mean age 35,5) without history of psychosocial disorders and/or temporomandibular disorders, were also recruited (H Group).

All subjects signed an informed consent. The study was conducted in accordance to the Helsinki declaration and all the subjects were free to withdraw from the experiment at any time.

The study was approved by the Lingotto Dental School Ethical Committee.



All patients and controls underwent a complete clinical evaluation performed by the same expert operator (BC), including palpation of pericranial and cervical muscles. Tenderness was scored for each muscle on a 0 to 3 scale, where 0 indicated no tenderness, 1 mild, 2 moderate and 3 severe tenderness. Pericranial Tenderness Score (PTS) is the mean tenderness score obtained from the palpation of: (1) masseter, (2) lateral pterygoid, (3) medial pterygoid and temporalis (4, mandibular and 5, cranial insertion) muscles, while Cervical Muscle Tenderness Score (CTS) is the mean tenderness score obtained from the palpation of: sternocleidomastoid (1, belly; 2, cranial insertion), (3) trapezius and (4) nuke muscles [25-28]

All the subjects of P Group were asked to indicate their facial pain level using a Visual Analogue Scale (VAS). VAS was a horizontal line, 100 mm in length, anchored by word descriptors at each end. On the left end the descriptor was “no pain”, while on the right end the descriptor was “worst imaginable pain”. The patients marked on the line the point they felt representing their perception of their current state. The VAS score was determined by measuring in millimetres from the left end of the line to the point marked by the patient.

 Furthermore, during the same day, both groups answered the psychological questionnaires, assessing depressive (Beck Depression Inventory (BDI-SF) [29]), anxiety (Hospital Anxiety and Depression Scale (HADS) [30]), distress symptoms (Distress Thermometer (DT) [31], and alexithymia (Toronto Alexithymia Scale (TAS-20) [32-34]. All patients and controls were interviewed by the same expert psychologist (DF) to exclude the presence of major depressive and anxiety disorders (according to DSM-IV TR criteria) .

All subjects received two Salivette kit (SARSTEDT A.G. & Co Numbrecht, Germany) for sampling cortisol, after a training period performed by an expert operator (DI). Each kit was composed by three cotton swabs in three different tubes. A first day of sampling, not used for the present study, was used to assess the compliance of the subjects. Each subject used the kit during the day performing the sampling at 08.00, 16.00 and 24.00 h. The second kit was used to perform the sampling used in the data analysis. Every participant was identified by a code reported on a label, together with the day and the hour of sampling.

Each subject was instructed to insert a pad in the mouth avoiding contact with fingers, chew the pad for 2 minutes to stimulate salivation and put it back in its container, which was finally firmly closed.

According to the Salivette producer guidelines no food, drinks (water excluded) and smoking were allowed 2 h before the salivary sampling; it was forbidden to carry on the process in presence of bleeding; it was prohibited to assume any drugs during the two days preceding the sampling and the Salivettes were preserved at a temperature between –2° and +8° C and were delivered to the analysis laboratory within 48 hours from sampling.

All salivary samples were analysed by the same expert biologist.  The samples of salivary cortisol were analysed with an immunologic test (immunoassay in electrochemiluminescence, ECLIA). The serum cortisol assay on the Elecsys analyser (Elecsys Roche Diagnostics, Laval, Quebec, Canada) is a competitive polyclonal antibody immunoassay that employs magnetic separation step followed by electrochemiluminescence quantitation.


Statistical analyses

Statistical analysis was performed using the software Statistical Package for the Social Sciences (SPSS) 15.0 (SPSS Inc., Chicago, IL, USA).

The statistical analysis was focused on the distribution and variability of the biological (cortisol), clinical (VAS, PTS and CTS) and psychological parameters (depression, anxiety, stress and alexithymia).

All the variables (biological, clinical and psychological) of Group P (patients) and group H (healthy) were compared using the Mann-Whitney test (non parametric test).

An analysis of  the VAS distribution was performed and the Group P was divided into two subgroups (P1 £ 0.18 mg/dL, P2 > 0.18 mg/dL) according to cortisol cut-off value of 0.18 mg/dL at 24.00 h (value from the laboratory of SARSTEDT A.G. & Co).

A receiver operating characteristics (ROC) curve analysis was performed to detect diagnostic accuracy (area under the curve), true-positive rate (TPR, sensitivity) and false positive rate (FPR, 1-specificity) of the VAS cutoff (≥60) to discriminate between high cortisol level (P2) and low cortisol level patients (P1). ROC curve analysis was based on the assumption that an area of 1 represents a perfect test, while an area of 0.5 represents a worthless test. Statistically, a larger area under the curve means that it is identifying more true positives while minimizing the percentage of false positives. Furthermore these subgroups were compared using the Mann-Whitney test

The level of significance was set at p<0.05 (alpha = 0.05).


Salivary cortisol

In both groups a normal circadian rhythm of cortisol secretion was revealed: higher in the morning, lower in the afternoon, and minimum in the evening (Fig.1).

The analysis of cortisol levels between groups P (patients) and H (healthy) at 8.00 (p=0.9), 16.00 (p=0.48) and 24.00 (p=0.11) h, showed no significant differences between the two groups. The cortisol level at 24.00 h was higher in patients than controls without a significant difference because of the low power of the test (P=49%, b=51%) and because of the high variability observed among the subjects of the Group P (0.22±0.29 mg/dL, range 0.07-1.2 mg/dL, s/x>1) (Fig.1).

Figure 1 - Circadian rhythm of cortisol secretion. Group P: 0,58 ±0,45 µg/dL (h8.00), 0,20±0,13 µg/dL (h 16.00) and 0,22±0,29 µg/dL (h 24.00). Group H: 0,45±0,15 µg/dL (h8.00), 0,16±0,11 µg/dL (h 16.00) and 0,10±0,07 µg/dL (h 24.00).


Psychometric scales

DT (p=0.001), BACK (p=0.002), depression HADS-D (p=0.001) and anxiety HADS-A (p=0.002) tests were significantly different between patients and healthy subjects.

Tests about alexithymia showed statistically significant results in TAS20 (a 20-item instrument that is one of the most commonly used measures to identify and describe emotions)

 (p=0.011) and in TASF1 (Difficulty Identifying Feelings subscale) (p=0.003).

The obtained results are shown into Tab. 1.


Cortisol vs VAS

Group P salivary cortisol values at 24.00 were compared to the cut-off value 0.18 mg/dL provided by the laboratory. Group P was then split into 2 subgroups P1 and P2: 17 patients with cortisol level < 0.18 mg/dL at 24.00 formed group P1; the remaining 6 patients with values > 0.18 mg/dL were part of P2. For each of the two subgroups the VAS frequency was calculated considering 6 ranges of value (<10, 11-30, 31-50, 51-70, 71-90, 90-110). A frequency distribution curve for the VAS of the two subgroups P1 and P2 was drawn (Fig.2).

Figure 2 - Frequency distribution curve for the VAS

A VAS value ≥60 was related to higher frequency of patients with higher cortisol levels.

The area below the ROC curve was 0.7, indicating for the VAS test a fair discriminant ability (Fig. 3).

Figure 3 - ROC Curve

The parameters of measurement accuracy of the ROC analysis identified the VAS value ≥60  as the cut-off value.



No significant differences related to PTS and CTS were observed between P2 and P1 patients.


Cortisol vs Psychometric Scales

No significant correlations between cortisol levels and psychometric scales were found.

Discussion and Conclusions: 

According to previous studies a circadian rhythmic expression of cortisol was revealed. A correlation between VAS and cortisol level was demonstrated, suggesting the possible use of the VAS scale as clinical diagnostic indicator of the cortisol level in patients suffering from muscle-related temporomandibular disorders. However no significant differences in daily cortisol level between muscular pain patients and healthy subjects were found, accordingly to Korszun et al., Galli et al., Sjors et al., Nilsson et al. and Jasim et al. [15;18;20;22-23]. Reviewing the existing literature on the topic only Da Silva et al. and Nadendla et al. asserted the importance of cortisol as indicator of stress and its connection with disability, depression and somatization [16;19].

All analysed studies confirmed our results related to the differences of psychological profiles between women with muscular pain and controls. Anxiety, stress, depression and alexithymia levels were significantly higher in patients than controls [18-20].

Nilsson A. et al. [23] investigating the possible relationship between the VAS scale and the cortisol level didn’t find any correlation. With respect to the present study, this difference could be related to the only waking cortisol sampling performed by Nilsson et al. [23]: shortly after awakening, a sharp 38–75% (average 50%) increase occurs in the blood level of cortisol in about 77%[34] of healthy adults, and it occurs in people of all ages.[35]. Thus a cortisol distribution curve was not possible. In our study the classification of patients on the basis of the pain perception demonstrated that VAS was a good instrument to separate patients into two categories: 1) Patients with VAS<60-70, which had normal cortisol; 2) Patients with VAS>60-70 which had hypercortisolism. It was interesting to observe that the two groups seemed similar to clinician on the basis of PTS and CTS analyses, while they had a completely different endocrinological profile. Therefore a closed link between HPA axis and pain perception (VAS) could be hypothesized.

The main limitation of this study is that we have not compared the home cortisol sampling to a gold standard (plasma levels). Thus the gold standard evaluation should be needed to fully confirm the excellent results achieved with this study. Notwithstanding that, the VAS scale proved to be suitable for measuring the cortisol level in patients suffering from muscle-related tempormandibular disorders.

Considering that only some patients showed high cortisol levels (HPA axis hyperactivation), the execution of a preventive cortisol test in patients with muscular pain in the daily clinical practice, could be expensive and excessive. Thus the answer to the first clinical-research question on which the present study was based we can conclude that the myofascial facial pain seems not related to high cortisol levels. The use of VAS as an instrument to discriminate between a dangerous cortisol level (HPA axis imbalance) or a normal cortisol level can be considered answers to the second clinical research question of the present study.


Table 1

Test Group P Group H P value
BECK 6.9±3.8 2.4±3.1 0.0002
HADS-D 6±3.1 3.1±2.8 0.001
HADS-A 7.4±3.6 4.5±3.3 0.002
TAS 20 49.5±13 39±11.2 0.011
TAS F1 17.6±6.2 12.1±5.2 0.003
TAS F2 13.9±5.2 11.6±6.0 0.23
TAS F3 18.1±5.5 15.8±4.8 0.27
DT 5±2.5 2.4±2.5 0.001


The authors gratefully acknowledge the support from Dental School (University of Turin) and the excellent technical assistance of Baldi-Riberi’s biomedical laboratory (Molinette Hospital, Turin).
Conflict of Interest: 
We certify that there are no known conflicts of interest associated with this publication and there hasn’t been significant financial support for this work that could have influenced its outcome.
  1. American Academy of Orofacial Pain guidelines (McNeil C Editor. Temporomandibular Disorders. Guidelines for Classification, Assessment and Management. The American Academy of Orofacial Pain. Quintessence Publishing Co, Carol Stream, Illinois, 1993
  2. Engel GL. From biomedical to biopsychosocial. Being scientific in the human domain.Psychosomatics. 1997 Nov-Dec;38(6):521-8.
  3. Fricton JR, Dubner R, editors. Advances in pain research and therapy.Orofacial pain and temporomandibular disorders, vol. 21. NewYork: Raven Press; 1995.
  4. Clark G.T. Etiologic theory and the prevention of temporomandibular disorders, Adv. Dent. Res.1991;5: 60–66.
  5. Chrousos G, Gold P: The concept of stress and stress system disorders. JAMA 1992; 267: 1244-52.
  6. McEwen, B.S., 2000. Allostasis and allostatic load: implications for neuropsychopharmacology. Neuropsychopharmacology 22, 108— 124.
  7. Sarzi Puttini PC, Cazzola M. La sindrome fibromialgica. Edi Ermes, Milano, 2001.
  8. Holleman Michiel, Vreeburg Sophie A., Dekker Jack J.M., Penninx Brenda W.J.H.. The relationships of working conditions, recent stressors and childhood trauma with salivary cortisol levels. Psychoneuroendocrinology (2012) 37, 801—809
  9. Plotsky PM: Stress system plasticity: neural adaptations to neonatal pain and stress. J Musculoskel Pain .1998; 3: 57-60.
  10. Rystedt, L.W., Cropley, M., Devereux, J.J., Michalianou, G., 2008. The relationship between long-term job strain and morning and evening saliva cortisol secretion among white-collar workers. J. Occup. Health Psychol. 13, 105—113.
  11. McFarlane, A.C., 2007. Stress-related musculoskeletal pain. Best Pract. Res. Clin. Rheumatol. 21, 549—565
  12. Lariviere WR, Melzack R. The role of corticotropin-releasing factor in pain and analgesia. Pain. 2000 Jan;84(1):1-12.
  13. Maina G., Palmas, A., Bovenzi, M., Filon, F.L., 2009a. Salivary cortisol and psychosocial hazards at work. Am. J. Ind. Med. 52, 251—260.
  14. McLean SA, Williams DA, Harris RE, Kop WJ, Groner KH, Ambrose K et al. Momentary relationship between cortisol secretion and symptoms in patients with fibromyalgia. Arthritis Reum 2005;52:3660-9.
  15. Galli Ursula, Gaab Jens, Ettlin Dominik A., Ruggia Fidel, Ehlert Ulrike, Palla Sandro. Enhanced negative feedback sensitivity of the hypothalamus–pituitary–adrenal axis in chronic myogenous facial pain. European Journal of Pain 2009;13:600–605
  16. Da Silva Andrade, Gamero G.H., Pereira L.J., Junqueira Zanin I.C., Gaviao M.B.. Salivary cortisol levels in young adults with temporomandibular disorders. Minerva Stomatol 2008;57(3):109-116.
  17. Jones David A., Rollman Gary B., Brooke Ralph I.. The cortisol response to psychological stress in temporomandibular dysfunction.Pain 1997; 72:171–182
  18. Korszun A, Young EA, Singer K, Carlson NE, Brown MB, Crofford L. Basal circadian cortisol secretion in women with temporomandibular disorders. J Dent Res 2002;81:279-83.
  19. Nadendla Lakshmi Kavitha, Meduri Venkateswarlu, Paramkusam Geetha, Pachava Rao Koteswara. Evaluation of Salivary Cortisol and Anxiety Levels in Myofascial Pain Dysfunction Syndrome. Korean J Pain (2014) 27(1):30-4.
  20. Jasim H., Louca S., Christidis N., Ernberg M. Salivary cortisol and psychological factors in women with chronic and acute oro-facial pain. J Oral Rehabil. (2014); 41(2):122-32.
  21. Solberg Nes Lise, Carlson Charles R., Crofford Leslie J., De Leeuw Reny, Segerstrom Suzanne C. Self-regulatory deficits in fibromyalgia and temporomandibular disorders
. Pain 151 (2010) 37–44
  22. Sjors Anna, Larsson Britt, Karlson Bjorn, Osterberg Kai, Dahlman Joakim, Gerdle Bjorn. Salivary cortisol response to acute stress and its relation to psychological factors in women with chronic trapezius myalgia–A pilot study.Psychoneuroendocrinology (2010) 35, 674—685
  23. Nilsson Anne-Marie, Dahlstrom Lars. Perceived symptoms of psychological distress and salivary cortisol levels in young women with muscular or disk-related temporomandibular disorders. Acta Odontologica Scandinavica, 2010; 68: 284–288
  24. Dworkin S.F. , LeResche L. , Research diagnostic criteria for temporomandibular disorders: review, criteria, examinations and specifications, critique. J Craniomandib Disord. 1992 Fall;6(4):301-55
  25. Jensen R, Rasmussen BK, Pedersen B, Olesen J (1993) Muscle tenderness and pressure pain thresholds in headache. A population study. Pain 52:193–199
  26. Langemark M, Olesen J (1987) Pericranial tenderness intension headache. A blind, controlled study. Cephalalgia 7:249–255
  27. Simons DG, Travell JG, Simons LS. 1999. Travell and Simon's Myofascial Pain and Dysfunction: the Trigger Point Manual: Volume I, edition 2: The Upper Body. Baltimore: Williams and Wilkins.
  28. Mongini F., Rota E., Deregibus A., Mura F., Francia Germani A., Mongini T.  A comparative analysis of personality profile and muscle tenderness between chronic migraine and chronic tension-type headache. Neurol Sci (2005) 26:203–207)
  29. Beck AT, Beck RW. Screening depressed patients in family practice: a rapid technic. Postgraduate Medicine 1972;52:81-85.
  30. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scandinavica 1983;67:361–70.
  31. Jacobsen PB, Donovan KA, Trask PC, Fleishman SB, Zabora J, Baker F, Holland JC . Screening for psychologic distress in ambulatory cancer patients. Cancer 2005;103:1494-1502
  32. Bagby RM, Parker JDA, Taylor GJ. The twenty-item Toronto Alexithymia Scale-I. Item selection and cross-validation of the factor structure. J Psychosom Res 1994;38:23-32
  33. Castelli LDe Santis FDe Giorgi IDeregibus ATesio VLeombruni PGranieri ADebernardi CTorta R. Alexithymia, anger and psychological distress in patients with myofascial pain: a case-control study. Front Psychol. 2013 Jul 31;4:490
  34. Wüst S, Wolf J, Hellhammer DH, Federenko I, Schommer N, Kirschbaum C. (2000).The cortisol awakening response – normal values and confounds. Noise Health. 2(7):79-88. P
  35. Fries E, Dettenborn L, Kirschbaum C. (2009). The cortisol awakening response (CAR): facts and future directions. Int J Psychophysiol. 72(1):67–73.
Corresponding Author: 

Comba Benedetta

Resident, Department of Surgical Sciences, Specialization School of Orthodontics, Dental School, University of Torino, Via Nizza 230 Torino, Italy

Tel 0039 011 6331550