Stress and salivary cortisol levels among temporomandibular disorders: a case-control study

2.1 Study design

This case-control study was conducted as part of a multidisciplinary investigation of TMDs patients within the framework of a PhD project at the Dental University Hospital, King Saud University, Riyadh, Saudi Arabia. The study protocol was approved by the Institutional Review Board (IRB) at King Saud University under project number E-22-7168, issued on 11 September 2022, in accordance with the ethical principles outlined in the Declaration of Helsinki. The study followed the STROBE guidelines [25]. Written informed consent was obtained from all participants. This study applied the PECO framework, with TMDs patients and healthy controls (population), lifetime stress (exposure) and salivary cortisol levels (outcome), to examine the relationship between stress and cortisol in both groups.

2.2 Selection criteria

A total of 110 participants were prospectively recruited between November 2022 and April 2023. The study included a TMDs group (n = 55) and a gender- and age-matched control group (n = 55). All participants were informed regarding the study, and written consent was obtained.

The inclusion criteria required participants to be adults aged between 18–40 years. The TMDs group consisted of individuals diagnosed with symptomatic disc displacements (DDs) and/or myalgia according to the Diagnostic Criteria for Temporomandibular Disorders (DC/TMD). The control group comprised individuals visiting the Dental University Hospital for routine dental check-ups. Participants were excluded if they (1) had conditions that affect pain sensitivity (e.g., fibromyalgia, rheumatoid arthritis, autoimmune diseases, migraines, neurological or neuropsychiatric disorders); (2) had used anti-inflammatory drugs, opioids, analgesics or steroids within the past 30 days (unless they had discontinued use); (3) were taking medications that affect saliva secretion (e.g., calcium channel blockers, antidepressants, antihistamines); (4) were pregnant or lactating, obese, or smokers; (5) had salivary gland diseases (e.g., tumours, stones, hyposalivation); (6) had complaints of dry mouth, edentulism or prosthodontic rehabilitation (complete/partial dentures); (7) had poor oral hygiene (Plaque or Gingival Index >2.0) or severe periodontal disease (clinical attachment loss ≥5 mm, probing depth ≥6 mm, significant bone loss); (8) had untreated or actively treated mental health disorders (e.g., depression, anxiety, post-traumatic stress disorder (PTSD); (9) had obstructive sleep apnoea (OSA); (10) had malignancies with ongoing radiotherapy/chemotherapy; or (11) had adrenal hyperfunction or Cushing’s disease. Participants were from the same cohort that was used in a prior study [26].

2.3 Clinical evaluation

The Diagnostic Criteria for Temporomandibular Disorders (DC/TMD) tool was published in 2014. The tool establishes standardized and valid diagnostic criteria for clinical and research settings and provides a globally recognized framework to be followed for both clinical and research purposes [21].

A thorough assessment of the participants’ jaw function and related structures was conducted. The clinician measured maximum unassisted and assisted mouth opening, along with lateral and forward movements, to evaluate range of motion and detect any restrictions. While the patient performed these movements, the temporomandibular joints (TMJs) were examined for clicking, popping or crepitus sounds. The masticatory muscles were palpated to check for tenderness or referred pain. The TMJs were also palpated, both at rest and during movement.

2.4 STRAIN questionnaire

The Stress and Adversity Inventory for Adults (STRAIN) questionnaire was used to assess stressors throughout the lifespan. This is a validated, structured interview-based scale designed to assess an individual’s lifetime stressors across multiple life domains (https://is.gd/i3VlNT). The questionnaire categorizes the stressors into the following domains: housing, financial, work, relationship, humiliation, marital/partner, health/treatment, death, physical danger. This scale has demonstrated excellent test-retest reliability, discriminant validity, and good concurrent and predictive utility for several stress-related health outcomes, including anxiety and depression [22]. As per the recommendations of the Institutional Review Board (IRB) and in consideration of cultural sensitivities, two questions, (19) “Were you assaulted or attacked (e.g., someone tried to hurt, molest, or rape you)” and (20) “Have you experienced ongoing sexual abuse (e.g., rape, molestation or unwanted sexual contact)”, were removed from the questionnaire. Any discussion of such topics is highly sensitive in this study population, where participants may be unwilling to disclose such experiences. This exclusion was specifically requested by the IRB to protect participant’s well-being and respect their autonomy and comfort.

The interviews were conducted in a controlled clinical setting at the Dental University Hospital. Each interview was conducted in a quiet and private consultation clinic, with only the researcher and the participant present. The seating arrangement provided adequate space to ensure a relaxed setting. Before the interview commenced, the participants received a detailed explanation of the study’s purpose and procedures. Each session lasted approximately 20 minutes, and participants were provided with time to reflect on their responses, if necessary.

2.5 Saliva collection

The saliva sample collection process was meticulously organized to ensure accuracy and consistency from the clinic to the laboratory. The participants were asked to not brush their teeth, eat or perform activities that could cause blood contamination before collection. After one minute of rest, participants were requested to rinse their mouths thoroughly with water to remove any debris. Two millilitres of unstimulated saliva were collected in pre-graduated polypropylene vials with conical-bottom centrifuge tubes and immediately stored at −20 °C in ultra-low temperature freezers (TSX60086A, ThermoFisher Scientific, Waltham, MA, USA). Upon arrival at the laboratory, the saliva samples were kept at room temperature for 10 minutes. The samples were centrifuged in an Andreas Hettich GmbH & Co. KG centrifuge (EBA 270, Tuttlingen, BW, Germany) at 3000 Revolutions Per Minute (RPM) for 5 minutes. For cortisol conjugation, the salivary samples were reconstituted using 16 μL of sterile dH₂O.

Salivary cortisol levels were analysed for all samples using an ELISA kit (RE52071, IBL International Corporation, Lucerne, LU, Switzerland) under consistent conditions for both the control and TMDs groups [26]. All collected information was systematically entered into SPSS Version 23.0 (IBM, Armonk, NY, USA).

2.6 Statistical analysis

A power analysis (α = 0.05, β = 0.80) determined a minimum required sample size of 50 per group for the detection of medium effect sizes. Descriptive statistics mean ± Standard deviation (SD) were used for demographic and clinical data. Normality was confirmed using the Kolmogorov-Smirnov test.

The demographic and clinical data obtained from the participants were recorded and transferred into the digital database (Microsoft Excel 2020). Frequencies, mean values and standard deviations (SDs) were obtained for the included variables. Comparisons between groups were conducted using the Chi-square test. An independent t-test was utilized to calculate the mean STRAIN score.

The Z-score transformation was applied for cross-comparisons between stress and cortisol levels. The results were considered significant if the p-value was ≤ 0.05. The data were analysed using IBM SPSS Statistics (ver. 23.0 for Windows; IBM Corporation, Armonk, NY, USA). One-way ANOVA with Tukey’s post hoc test was employed to compare stress scores among different TMDs subtypes. Cronbach’s alpha was used to assess the reliability of the scale by determining how well its items correlate with one another. A higher alpha value indicates strong internal consistency.

To account for multiple comparisons in the multivariable regression analysis, the Bonferroni correction was applied, with the adjusted significance threshold set at α = 0.05/10 = 0.005. p-values were adjusted accordingly, and only variables with adjusted p-values below this threshold were considered statistically significant.

3.1 Sociodemographic characteristics

This study included 110 participants, all of whom underwent clinical examination for TMDs according to the Diagnostic Criteria for Temporomandibular Disorders (DC/TMD). The participants were aged between 18–40 years, with a male-to-female ratio of 1:2. In the study group, the majority of the participants were aged between 18–30 years (53.1%), whereas the majority of the control group were aged between 31–40 years (52.5%). The majority in both groups had college degrees (study: 47.5%, control: 52.5%) and were married (study: 44.4%, control: 55.6%).

3.2 Stress scale

A comparison of STRAIN scores between the groups demonstrated a statistically significant difference (p = 0.001), with the study group exhibiting higher stress levels than the controls (11.1091 ± 3.26). The mean score was higher among males (6.37 ± 6.10; p = 0.91) than females. Significant variations in stress levels were observed among the subgroups.

The myalgia group had the highest mean STRAIN score (11.69 ± 3.72), while the myofascial pain group had the lowest mean score (8.80 ± 1.14). The disc displacement without reduction with limited mouth opening (DDWoR with LO) group was positioned in between, with a mean score of (11.10 ± 1.52), which reflected a distinct stress pattern (p = 0.043).

Higher mean scores were noted among participants aged between 31–40 years (6.52 ± 5.46), those with college degrees (6.51 ± 5.43) and those who were married (6.59 ± 5.27). However, none of the demographic variables (gender, age, education or marital status) showed statistically significant differences in STRAIN scores (p > 0.05) (Table 1). Stressors, including acute life events and chronic difficulties, were significantly associated with TMDs (p < 0.001). Specific stressors related to housing, treatment/health, marital/partner relationships, humiliation and other interpersonal relationships were significantly associated with TMDs (p < 0.001).

3.3 Salivary cortisol levels and TMDs subgroups

Table 2 (Ref. [26]) presents the mean salivary cortisol levels of the TMDs subgroups and the control group. These values represent the average of saliva samples collected at both intervals (early and late morning). Disc displacement without reduction with limited mouth opening showed the highest cortisol levels, with a value of 82.49 ± 124.34 (8.32–398.64). Patients with myofascial pain had a mean cortisol level of 13.03 ± 8.46 (5.69–34.68), while those with myalgia had a mean of 4.69 ± 3.90 (0–15.53) [26].

3.4 Stressor domains

Before applying multiple comparison correction, six predictors were statistically significant (p < 0.05). However, after applying a Bonferroni correction, only four predictors remained significant: relationship stress (adjusted p = 0.04), humiliation (adjusted p = 0.02), marital/partner stress (adjusted p < 0.001) and death-related stress (adjusted p = 0.01). Housing and health/treatment stress were initially significant (p = 0.005 and p = 0.039, respectively) but did not meet the adjusted significance threshold (0.005).

The regression analysis, with TMDs as a dependent variable, showed several significant predictors. Significant domains included housing (B = −1.644, p = 0.005), relationship (B = −1.372, p = 0.004), humiliation (B = 1.860, p = 0.002), marital/partner, (B = 2.337, p < 0.001), health/treatment (B = 1.736, p = 0.039) and death (B = −1.582, p = 0.001). These findings indicate that the humiliation and partner domains had the strongest positive correlations with TMD. Other variables, including financial, work, crime and physical danger domains, were not significant predictors (p ≥ 0.05). Overall, the model explained 61.8% of the variance in the TMD, with an adjusted R² of 42.7% (p = 0.001) (Table 3).

4.1 Limitation

Sample size limitations pose a challenge, particularly in the TMDs subgroups, which may limit the generalizability of the findings, which should be interpreted with caution. The exclusion of two questions may limit cross-cultural comparability. Since cortisol fluctuates throughout the day, the inclusion of additional afternoon and evening measurements in future studies could provide more comprehensive insight.

4.2 Recommendations for future studies

Future studies with larger cohorts are needed to validate these results.

These studies should collect cortisol samples at multiple time points (morning, afternoon, evening) for more robust cortisol analysis. The sample should be expanded to include different age groups and socioeconomic backgrounds. Different scales that capture a wide range of stressors could be used in future studies.