Prevalence of Temporomandibular Disorder Symptoms After Whiplash Trauma—A Systematic Review and Meta‐Analysis

ABSTRACT

Background and Objectives

Temporomandibular disorders (TMDs) is an umbrella term for pain and dysfunction involving jaw muscles and/or the temporomandibular joint, with whiplash trauma suggested to be one risk factor. The aim was to evaluate prevalence and relative risk of TMDs in the acute and chronic stages after whiplash trauma.

Databases and Data Treatment

This review was registered in Prospero (CRD42023407091) and followed the PRISMA guidelines. A literature search in PubMed, Scopus and Web of Science on 10 March 2023 and updated 29 April 2024 identified studies reporting prevalence of TMD after whiplash trauma. Risk of bias was assessed with Joanna Briggs Institute Prevalence Critical Appraisal Tool. A random effect meta‐analysis was performed for prevalence of TMD pain.

Results

After screening of 671 identified studies, 96 articles were assessed in full text. Fourteen studies, with 840 cases in the acute and 8293 cases in the chronic stage (i.e., > 3 months post‐trauma) were included in a qualitative analysis together with 1591 controls. Nine studies, including 449 cases in the acute and 7912 individuals in the chronic stage after trauma, together with 515 controls, were included in the meta‐analysis. Mean prevalence for TMD pain was 18.9% (95% CI 9.71–29.98) in the acute case group, 26.8% (95% CI 15.07–38.79) in the chronic case group, and 5.7% (95% CI 3.08–8.96) in the control group.

Conclusions

The higher prevalence of TMD pain already in the early stage after whiplash trauma, emphasises the need for early comprehensive clinical assessment as well as targeted research to understand underlying mechanisms.

Significance

The prevalence of Temporomandibular disorder pain was high already in the acute stage after whiplash trauma, and there was no evidence of any decrease from the acute to the chronic stage. This finding suggests that early assessment and management rather than a 'wait‐and see' approach should be recommended when patients present with orofacial pain related to whiplash trauma.

1. Introduction

Orofacial pain is the third most common chronic pain condition, after knee and neck/back pain and affects more than 10% of the adult population worldwide (Breivik et al. 2006; Lövgren et al. 2017). As with many other pain conditions, orofacial pain is more frequent in women and in the 20‐to 50‐year age span (Häggman‐Henrikson et al. 2020). Orofacial pain is also the most common reason for emergency dental visits globally (John et al. 2020). Whereas acute orofacial pain is most often dentally related, chronic orofacial pain is most often related to temporomandibular disorders (TMDs). TMDs is an umbrella term for pain and dysfunction involving jaw muscles and/or the temporomandibular joint (TMJ) (Dworkin and LeResche 1992; Schiffman et al. 2014). The aetiologies of TMDs are multifactorial (Slade et al. 2013) and indirect trauma related to whiplash injury has been suggested to be a contributing factor (Häggman‐Henrikson, Rezvani, and List 2014).

The term whiplash trauma describes an acceleration–deceleration trauma mechanism to the head–neck, most often related to motor vehicle accidents, but also to other types of external trauma, with considerable burden on affected individuals and society (Spitzer et al. 1995). In 2013, the annual incidence of whiplash trauma in the western world was about 0.2%, with most affected individuals having transient symptoms and recovering within a year of the trauma (Walton et al. 2013). However, over a third of affected individuals will develop long‐term symptoms, chronic post‐traumatic neck pain, often labelled as whiplash‐associated disorders (WAD). In addition to the most commonly reported WAD symptoms, neck pain, neck stiffness and headache (Al‐Khazali et al. 2020), psychological comorbidities are common, as is the case with other chronic pain conditions. Moreover, patients with WAD often report pain in other regions, with an especially high prevalence in the upper body regions (Westergren et al. 2018).

In line with this, we previously reported (Häggman‐Henrikson et al. 2013) that the prevalence of TMD in whiplash populations is approximately twice that of the general population (Lövgren et al. 2016). Those results, however, were based only on a qualitative synthesis of the literature, as a meta‐analysis was not performed. Furthermore, in the last decade several new studies have been published, which may allow for a meta‐analysis that provides a more definite evidence base. Pain sensitivity and spread of pain have been shown to be present early after whiplash trauma (Andersen et al. 2022). This is especially relevant since the earlier notion that pain in the orofacial region after whiplash trauma develop over time as part of chronification was refuted by a prospective study reporting that orofacial pain was present already in the early stage after a whiplash trauma, but often persisted into the chronic stage (Lampa et al. 2020). If this is the case, early assessment and management rather than a ‘wait‐and see’ approach should be recommended. Thus, the aim of this systematic review was to evaluate the prevalence of TMDs, together with the relative risks, both in the acute and chronic stages after whiplash trauma.

2. Methods

2.1. Protocol

This review was carried out in accordance with the PRISMA guidelines (Moher et al. 2009) with a protocol registered in Prospero (CRD42023407091). All identified primary studies were imported into Covidence for overall management of the different stages of the review.

2.2. Literature Search

The main search strategy was developed for PubMed and then adopted for the other databases. This search strategy was developed in collaboration with an information specialist at Malmo University and included a combination of free text and Mesh‐terms arranged in two search blocks on whiplash and TMD, respectively. The electronic search was performed in three separate databases; PubMed, Scopus, Web of Science from the inception of each database until 10 March 2023 and updated 29 April 2024. The full search strategy for all databases is provided in Table 1. The electronic search was done without language restrictions and complemented with a hand search of reference lists and relevant literature. Grey literature was not searched and authors were not contacted for additional information. All articles identified in the literature searches were imported from EndNote into Covidence for abstract screening, full‐text assessment, data extraction and quality assessment.

TABLE 1.

Electronic search strategy and identified records on April 29 2024.

	Databases and search strategy	Records
	Pubmed
#1	temporomandibular joint OR temporomandibular joints OR temporomandibular disorder OR temporomandibular disorders OR temporomandibular joint disorder OR temporomandibular joint disorders OR tmd OR tmj disorder OR tmj disorders OR craniomandibular disorder OR craniomandibular disorders OR “Craniomandibular Disorders”[Mesh] OR “temporomandibular joint disorders”[MeSH]	43,830
#2	“Whiplash Injuries”[Mesh] OR whiplash[Title/Abstract] OR ((cervical[Title/Abstract] OR neck[Title/Abstract] OR “Neck”[Mesh]) AND (injur*[Title/Abstract] OR traum*[Title/Abstract] OR traffic[Title/Abstract] OR “Accidents”[Mesh]))	51,677
#1 AND #2		392
	Scopus
#1	TITLE‐ABS‐KEY (“craniomandibular disorder” OR “craniomandibular disorders” OR “temporomandibular joint disorder” OR “temporomandibular joint disorders” OR “temporomandibular disorder” OR “temporomandibular disorders” OR tmjd OR tmd OR “tmj disorder” OR “tmj disorders”)	34,065
#2	TITLE‐ABS‐KEY (whiplash OR ((cervical OR neck) AND (injur* OR traum* OR traffic OR accident*)))	108,862
#1 AND #2		450
	Web of Science
#1	TS = (“craniomandibular disorder” OR “craniomandibular disorders” OR “temporomandibular joint disorder” OR “temporomandibular joint disorders” OR “temporomandibular disorder” OR “temporomandibular disorders” OR tmjd OR tmd OR “tmj disorder” OR “tmj disorders”)	19,496
#2	TS = (whiplash OR ((cervical OR neck) AND (injur* OR traum* OR traffic OR accident*)))	55,084
#1 AND #2		169

2.3. Inclusion and Exclusion Criteria

The inclusion criteria based on PECO were as follows:

Population: Adults (≥ 18 years) with a history of whiplash trauma and symptoms and/or signs related to TMD, and/or a TMD diagnosis.

Exposure: Whiplash trauma.

Comparison: Control groups without a history of whiplash trauma.

Outcomes: Prevalence of TMD symptoms including pain and functional limitations, other self reported symptoms or signs related to the masticatory muscles and the temporomandibular joint.

Peer‐reviewed studies in English, Swedish or German were included. Systematic reviews, narrative reviews, case reports, letters to editor, editorials, commentaries, proceedings papers and conference abstracts were excluded.

2.4. Study Selection

Screening and full‐text assessment were conducted in Covidence. Two reviewers (AL, BHH) independently assessed all identified abstracts to determine possible eligibility. Articles deemed as possibly eligibly by at least one of the reviewers, or with abstract missing, were included for full text assessment. Full‐text assessment in relation to inclusion and exclusion criteria was carried out by two independent reviewers (AL) and (BHH, TL or CP) considering any possible conflict of interest related to co‐authorship. Disagreements at full‐text level was resolved by discussion or by consulting a third reviewer (TL/CP).

2.5. Assessment of Risk of Bias in Included Studies

Risk of bias was assessed with the Joanna Briggs Institute Prevalence Critical Appraisal Tool (Munn et al. 2014) by two independent reviewers; (AL) and (BHH/TL/CP), again considering any possible conflict of interest related to co‐authorship. Disagreements were resolved by discussion or by consulting a third reviewer (TL/CP).

2.6. Data Extraction

Data extraction was conducted in a form adapted from the Covidence template. The following data were extracted: author, publication year, country, aim, study design, study setting, study population, methods for assessment, TMD outcomes, results and authors' conclusions. Data were extracted by two independent reviewers (AL) and (BHH/TL/CP). Disagreements were resolved by discussion or by consulting a third reviewer (TL/CP).

2.7. Data Synthesis

A qualitative synthesis was carried out together with a random‐effect meta‐analysis. Prevalence from individual studies were transformed using the double‐arcsine transformation and pooled using a random‐effects model to account for between‐study heterogeneity (Wang 2023). The restricted maximum‐likelihood estimator was used for between‐study variance estimation. The prevalence of TMD in the acute and chronic phases was analysed separately. Additionally, meta‐analyses of relative risks in acute and chronic phases were performed in relation to the control groups as defined by our PECO. All statistical analyses were performed using R version 4.3.3 with meta and metafor packages and presented in forest plots.

3. Results

In total, 1021 articles were identified in the literature search. After removal of duplicates and abstract screening of 671 primary studies, 96 studies (including three studies identified in the hand search) were assessed in full text. A total of 82 studies were excluded at full‐text level, with the most common reason for exclusion being that the study population was not representative of a true whiplash population (Table 2).

TABLE 2.

Main reasons for exclusion at full text level (n = 82).

Main reason for exclusion	Study
Not a representative whiplash population	Kaye, Moran, and Fritz (1979); Harkins and Marteney (1985); Weinberg and Lapointe (1987); Pullinger and Monteiro (1988); Goss and Bosanquet (1990); Wiens (1990); Burgess (1991); Pullinger and Seligman (1991); Pressman et al. (1992); Probert, Wiesenfeld, and Reade (1994); Burgess et al. (1996); De Boever and Keersmaekers (1996); De Wijer et al. (1996); Garcia Jr. and Arrington (1996); Seligman and Pullinger (1996); Kolbinson et al. (1997, 1998); Krogstad et al. (1998); Ciancaglini, Testa, and Radaelli (1999); De Boever, Van Den Berghe, and Vanwormhoudt (1999); Plesh et al. (1999); Friedman and Weisberg (2000); Kamisaka et al. (2000); Rauhala et al. (2000); Dalkiz, Pakdemirli, and Beydemir (2001); Steed and Wexler (2001); Abd‐Ul‐Salam, Kryshtalskyj, and Weinberg (2002); Choi et al. (2002); Häggman‐Henrikson, Zafar, and Eriksson (2002); Velly, Gornitsky, and Philippe (2002); Zafar, Nordh, and Eriksson (2006); Eriksson, Häggman‐Henrikson, and Zafar (2007); Grushka et al. (2007); Grönqvist, Häggman‐Henrikson, and Eriksson (2008); Vita (2010); Ohrbach et al. (2011); Marini et al. (2013); Ohrbach et al. (2013); Gualniera et al. (2016); Lee et al. (2018, 2019); Sharma et al. (2019); Bal et al. (2020); Sharma et al. (2020); Singh, Prabakar, and Duraisamy (2020); Figueiredo et al. (2021); Lee, Lee, and Auh (2021); Corsalini et al. (2022); Rangel et al. (2024)
Not peer‐reviewed full paper	Frankel (1965); Roydhouse (1973); Moses and Skoog (1986); Kupperman (1988); Mandel (1988); Kirk Jr. (1992); Kerwin (1999); Niamtu (1999); Brooke and Merskey (2002); Cohen (2005); Bucholtz (2007); Ernst (2008); Miller (2020)
Review	Stenger (1977); Lader (1983); Levandoski (1993); O'Shaughnessy (1995a, 1995b); Brown (1997); Goldberg (1999); Gremillion (2000); John (2011)
Overlapping study population	Schrader, Obelieniene, and Ferrari (2000); Bunketorp et al. (2005); Häggman‐Henrikson et al. (2016); Lampa et al. (2019); Eklund et al. (2020); Böthun et al. (2023); Böthun et al. (2024)
Wrong age study population	Martín Berrocal et al. (2018)
Diagnostic guideline	Headache Classification Committee of the International Headache Society (2004)
No reported TMD outcome	Hulse and Losert‐Bruggner (2008)
Experimental Study	Howard et al. (1998)

Fourteen studies based on a total of 11 study populations were included (Figure 1). Most of these primary studies were conducted in hospital or specialist clinic settings in northern Europe. Seven studies were longitudinal (Heise, Laskin, and Gervin 1992; Kasch et al. 2002; Carroll, Ferrari, and Cassidy 2007; Sale and Isberg 2007; Severinsson, Bunketorp, and Wenneberg 2010; Sale, Bryndahl, and Isberg 2014; Lampa et al. 2020) and thus included data from both the acute and chronic (i.e., more than 3 months) stages after trauma. The individual studies included between 19 and 7127 cases with whiplash trauma and 8 of the studies also included control groups with 20–186 individuals without a history of whiplash trauma. In total, 840 individuals in the acute stage, 8293 individuals in the chronic stage, and 1591 controls without a history of whiplash trauma were included in the qualitative synthesis (Table 3).

FIGURE 1.

PRISMA flow diagram of included and excluded studies at different stages.

TABLE 3.

Extracted data for all included studies (n = 14).

Authors (year)	Study design	Total n (mean age, SD/range)	TMD assessment	Results	Authors' conclusions
Country	Whiplash assessment	Men/women (%)	Outcome measures
Setting	Criteria/time after trauma
Bergman, Andersson, and Isberg (1998) Sweden Specialist clinic	Cohort with control group Clinical assessment + MRI Cases: Quebec Grade 1–3 > 72 h < 3 months after MVA + visit emerg dept. Controls: General population	Cases: 60 (34, 16–55 years) 22♂/37♀(62%) Controls: 53 (36/35, 15–63 years) 22♂/31♀(58%)	Interview TMJ Imaging Jaw catching/locking TMJ sounds MRI	Cases: 53% DD; 6% TMJs effusion Controls: 45% DD; 8% TMJs effusion Cases versus controls: NS	This prospective study does not show any significantly increased incidence of disk displacement, joint effusion, or any other injury to the TMJ after whiplash trauma that could be revealed by MR imaging
Carroll, Ferrari, and Cassidy (2007) a Canada Insurance register	Cohort study Self reported: questionnaire after 6 weeks, 4 months, 8 months, 12 months Cases: MVA Exclusion: hospitalised > 2 days	Cases: 7127 Age ≥ 18 years 2855♂/4269♀(60%) No control group	Questionnaire Reduced/painful jaw movements	Cases: 17.4% TMD pain 95% CI: 16.5–18.3 (♀20.0%, ♂13.2%) Women higher incidence (♀18.4% versus ♂12.4%) > 50% of incidence cases recovered within first year	Reduced or painful jaw movement was more common in people with WADs than in those with other collision‐related injuries. Among those with WADs, reduced or painful jaw movement was more common in women and younger people
Ferrari, Schrader, and Obelieniene (1999) a Lithuania Traffic accident register	Cohort with control group Self reported: questionnaire Cases: MVA 27 months (14–41 months) Controls: General population	Cases: 165 139♂ (40 years, SD 10) 26♀(16%) (36 years, SD 8) Controls: 186 156♂ (40 years, SD 10) 24♀(13%) (34 years, SD 7)	Questionnaire (symptoms once a month or more) Jaw pain Jaw locking TMJ sounds	Cases: 4.8% jaw pain 6.6% jaw locking; 9.1% TMJ sounds Controls: 9.3% jaw pain 8.3% jaw locking; 18% TMJ sounds Cases versus controls: NS	Unlike whiplash claimants in many Western societies, Lithuanian accident victims do not appear to report the chronic symptoms of temporomandibular disorders despite their acute whiplash injuries
Heise, Laskin, and Gervin (1992) a United States Hospital clinic	Cohort study Clinical assessment Baseline (Within 72 h), 1, 12 months Cases: MVA + visit emerg dept	Cases: 155 (38, 20‐71 years) 59♂/96♀(62%) No control group	Clinical examination at baseline. Phone interview follow‐ups Jaw pain TMJ clicking	Cases baseline: 14% jaw pain TMJ clicking: 1 (1%) Symptoms generally diminished over time No new symptoms reported	These data indicate that the incidence of TMJ pain and clicking following whiplash injury is extremely low, and that patients who do not have clicking on resolution of their initial pain/dysfunction subsequently do not develop this problem
Kasch et al. (2002) a Denmark Hospital clinic	Case control study Self reported Cases: Quebec Grade 1, MVA, visit emerg dept. ≤ 48 h Age 20–35 years Examined within 4 weeks and after 6 months Controls: ankle injury, visit emerg dept. ≤ 48 h	Cases: 19 (26.3 years, SD 4.5) 9♂/10♀(53%) Controls: 20 (25.4 years, SD 5.7) 9♂/11♀(55%)	Clinical examination RDC/TMD Questionnaire TMD pain Max jaw opening TMJ clicking TMJ crepitus	Cases: 5% TMD pain Jaw op 56.8 mm (SD 7.0) 20% clicking; 10% crepitus Controls: 5% TMD pain Jaw op 52.6 mm (SD 14.0) 21.1% clicking; 0% crepitus Follow‐up cases: Jaw op 56.4 mm (SD 7.2) 20.0% clicking; 0% crepitus Follow‐up controls: Jaw op 56.4 mm (SD 7.6) 10.5% clicking; 0% crepitus Cases versus controls: All NS Baseline versus follow‐up: All NS	TMD pain after whiplash injury and ankle injury is rare, suggesting that whiplash injury is not a major risk factor for the development of TMD problems. Further studies are needed to identify which other factors may contribute to TMD pain
Klobas, Tegelberg, and Axelsson (2004) a Sweden Specialist clinic	Case control study Clinical assessment 8–200 months (mean 34 months) Cases: Refered for 4‐week rehabilitation programme Controls: Recall patients dental clinic	Cases: 54 (37.4, 23‐64 years) 22♂/32♀(59%) Controls: 66 (38.4, 19‐62 years) 26♂/40♀(61%)	Questionnaire Clinical examination Helkimo's index Pain Pain on palp Pain on function Reduced jaw op TMJ clicking	Cases: 61.1% palp pain TMJ Muscle palp score: 3.3 30.2% pain on function Jaw op 48.2 mm (CI 46.1–50.3) 16.7% < 40 mm jaw op; 22.2% clicking Controls: 7.6% palp pain TMJ Muscle palp score: 3.3, p < 0.001 3.0% pain function, p < 0.001 54.2 mm (52.5–56.0), p < 0.05 1.5% < 40 mm jaw op 18.2% clicking, NS	In conclusion, the prevalence of TMD was higher among individuals with chronic WAD compared with an age‐ and sex‐stratified cohort of patients in a general dental practice. The results indicate that trauma to the neck also affects temporomandibular function
Kronn (1993) a Ireland Hospital clinic	Case control study Clinical assessment emerg dept. 6–38 days Cases: Traffic accident Controls: Physiotherapy patients	Cases: 40 (32.2, 17‐53 years) 26♂/14♀(35%) Controls: 40 (33.4, 16‐35 years) 26♂/14♀(35%)	Questionnaire Clinical examination Pain Pain on palp Reduced jaw op TMJ sounds	Cases: 30% TMJ pain (♂23%, ♀43%) 30% Pain on palp (♂15%, ♀57%) Reduced jaw op: 37.5% (♂27%, ♀57%) 37.5% TMJ sounds (♂27%, ♀57%) Controls: 2.5% TMJ pain (♂0%, ♀21%) 7.5% pain on palp (♂5%, ♀3%) Reduced jaw op: 7.5% (♂ 4%/♀21%) 42.5% TMJ sounds (♂ 42%, ♀43%) Cases versus controls: TMJ pain p < 0.001 Pain on palpation p < 0.01 Jaw opening p < 0.01 TMJ sounds: NS	These findings warrant further study and suggest the benefit of a routine examination of the temporomandibular joint and masticatory system in patients with cervical whiplash injury
Lampa et al. (2017) Sweden Specialist clinic	Case control study Clinical assessment emerg dept. Cases: MVA + visit emerg dept. within 72 h, Quebec Grade I‐III, age 18–70 years Examined 2–4 weeks after trauma Controls: General population	Cases: 80 (34.1,18–67 years) 33♂ (34.3,18–67 years) 47♀ (59%) (33.9,18–66 years) Controls: 80 (35.0, 18–66 years) 33 ♂ (23.5, 18–66 years) 47♀ (59%) (36.7, 20–64 years)	Questionnaires JDC Chewing task Pain on function	Cases: 30% pain on function JDC 0–6 Controls: 10% pain on function JDC 0–2 Cases versus controls: Pain on function p = 0.003 JDC p < 0.001	Taken together, the results indicate that jaw‐neck sensory‐motor function is impaired already within 1 month after a whiplash trauma. The association between neck disability and jaw impairment underlines the close functional relationship between the regions, and stresses the importance of multidisciplinary assessment
Lampa et al. (2020) a Sweden Specialist clinic Same cohort as Lampa 2017	Cohort study w control group Clinical assessment emerg dept. Cases: MVA + visit emerg dept. within 72 h, Quebec Grade I–III, age 18–70 years Examined 2–4 weeks after trauma Controls: General population, age 18–70 years	Cases: 176 (35.2 years, SD 14.4) 72♂ (37.6 years, SD 14.3) 104♀ (59%) (33.5 years SD 14.2) Controls: 116 (32.8 years, SD 12.8) 48♂ (30.8 years, SD 12.1) 68♀ (58%) (34.2 years, SD 13.1)	Questionnaires Pain intensity (NRS) 3Q/TMD: Q1: Pain Q1 Q2: Pain on function Q3: Catching locking JDC	Baseline Cases: NRS 0–8 33.5% Q1; 19.3% Q2; 14.8% Q3 JDC: 0–8; DS:0–90 Baseline Controls: NRS 0–3 6.0% Q1; 4.3% Q2; 7.8% Q3 JDC: 0–2; DS: 0–90 Baseline cases versus controls: NRS: p = 0.002 Q1 0.002; Q2 0.004; Q3 NS JDC and DS: p = 0.002 Follow‐up Cases: NRS: 0–7 35.3% Q1; 27.7% Q2; 18.5% Q3 JDC: 0–8; DS: median 0–80 Follow‐up Controls: NRS: 0–5 12.5% Q1; 6.7% Q2; 7.7% Q3 JDC: 0–4; DS: 0–23 Follow‐up cases versus controls: NRS: p = 0.004 Q1 0.002; Q2 0.002; Q3 NS JDC and DS: p = 0.002	Orofacial pain and jaw disability related to neck pain are often present already at the acute stage after whiplash trauma and persist into the chronic stage for most individuals. Assessment following whiplash trauma should therefore include both the neck and the orofacial regions. More studies are needed to further evaluate risk factors for development of orofacial pain after whiplash trauma
Magnusson (1994) a Iceland Specialist clinic	Case‐series Clinical assessment 6–44 months (mean 17 months) Late whiplash syndrom	Cases: 38 (33/31.5 years, 17–52) 13♂/25♀ (66%) No control group	Clinical examination TMJ pain TMJ clicking Jaw dysfunction	Cases: 28.9% TMJ pain Clicking: 0% Jaw dysfunction 10.5	It is concluded that many of the symptoms of patients that suffer from late whiplash syndrome conform with the criteria of other specific diagnoses
Sale and Isberg (2007) a Sweden Hospital clinic Same cohort as Bergman 1998	Cohort with control group Clinical assessment Cases: Quebec Grade 1–3 > 72 h < 3 months after MVA + visit emerg dept. Baseline: 3–15 days after trauma 1‐year follow‐up: whiplash mean 16 months (13–21), controls mean 14 months (12–17) Controls: General population	Cases: 59 (33/35, 16–55 years) 22♂/37♀(62%) Controls: 53 (36/35, 15–63 years) 22♂/31♀(58%)	Questionnaires Interview Pain Functional limitation	Baseline cases: TMJ pain 12% 34% clicking;2% locking 15% Other TMJ symptoms Baseline controls: TMJ pain 2% 34% clicking; 2% locking 15% Other TMJ symptoms Cases versus controls: TMJ pain p = 0.048 Follow‐up cases: TMJ pain 32% 42% clicking; 15% locking 34% Other TMJ symptoms Follow‐up controls: TMJ pain 6% 19% clicking; 8% locking 7% Other TMJ symptoms Cases versus controls: TMJ pain p = 0.004	Our results suggest that one in three people who are exposed to whiplash trauma is at risk of developing delayed TMJ symptoms that may require clinical management
Sale, Bryndahl, and Isberg (2014) Sweden Specialist clinic Same cohort as Bergman 1998	Cohort with control group Clinical assessment + MRI Cases: Quebec Grade 1–3 > 72 h < 3 months after MVA + visit emerg dept. Baseline: 3–15 days after trauma 1‐year follow‐up +15‐year follow‐up Controls: General population	Cases: 57 (49, 31–71 years) 21♂/36♀(63%) Controls: 50 (50, 29–78 years) 20♂/30♀(60%)	Questionnaires Interview MRI TMJ pain Catching/locking DD	Baseline cases: TMJ pain 10.5% 33% clicking; 2% locking; 63% DD Baseline controls: TMJ pain 2% 34% clicking; 2% locking; 53% DD Cases versus controls: DD: NS Follow‐up cases Year 1/Year 15: TMJ pain 33%/21% clicking 42%/42%; locking 16%/19% DD 63% Follow‐up controls Year 1/Year 15: TMJ pain 6%/4%; DD −/55% Baseline versus follow‐up: DD: NS	This prospective 15‐year follow‐up suggests that the development of TMJ symptoms, both immediate and delayed, is common in whiplash patients
Severinsson, Bunketorp, and Wenneberg (2010) Sweden Mixed settings	Cohort study Clinical assessment Cases: Quebec Grade 1–3 0–35 days after MVA (mean 16 days) + 1 year. follow‐up (questionnaires only)	Cases baseline: 146 65♂ (38, 20–75 years) 81♀ (55%) (40, 19–66 years) Cases baseline: 120 52♂/68♀ (57%) No control group	Clinical examination + questionnaires Jaw symptoms Jaw signs	Cases baseline: Jaw symptoms 4% (♂4%/♀5%) Jaw signs: 36% (♂35%/♀37%) Cases follow‐up: Jaw symptoms 8% (♂0%/♀15%) Jaw signs: 36% (♂23%/♀24%)	Jaw symptoms are seldom reported during the acute phase after a whiplash trauma. Women more often than men develop jaw symptoms during the first year. Jaw symptoms and signs may develop also after low‐speed impacts, especially after rear‐end collisions. Jaw symptoms and signs should be observed after whiplash trauma, especially in those with headache, pronounced neck problems, cranial neck symptoms and post‐traumatic stress
Visscher et al. (2005) a Netherlands Specialist clinic	Case–control study Self‐report, neck pain > 6 months after whiplash trauma Cases: Quebec Grade 1–3 Controls 1: Neck pain Controls 2: No neck pain	Total: 65 29♂/36♀ (55%) (39.8 years, SD 11.7, 19–61) Cases: 25 Controls 1:15 Controls 2:25	Clinical examination + questionnaires RDC/TMD TMD pain	Cases: TMD pain 52% Controls 1: TMD pain 27% Controls 2: TMD pain 8% Cases versus Controls 1: p < 0.028 Cases versus Controls 2: p = 0.001	The higher prevalence of widespread pain and psychologic distress in patients with chronic whiplash‐associated disorder suggests that the higher prevalence of temporomandibular disorder pain in these patients is part of a more widespread chronic pain disorder

Abbreviations: DD, disk displacement; DS Jaw‐pain related disability; JDC, Jaw disability checklist; MRI, Magnetic Resonance Imaging; NRS, Numerical Rating Scale; TMD, Temporomandibular disorders; TMJ, temporomandibular joint.

^a Studies included in the meta‐analysis (n = 9).

In most studies, the whiplash trauma was related to traffic injuries, specifically car accidents. There was a range of definitions of the whiplash populations in the individual studies, from insurance registers and self‐report to clinical assessments in hospital settings. The most commonly reported outcome variables related to TMDs were TMD pain and TMJ sounds. Three studies reported joint crepitation and clicking grouped as TMJ sounds (Kronn 1993; Bergman, Andersson, and Isberg 1998; Ferrari, Schrader, and Obelieniene 1999). Only one study specifically reported joint crepitation, present in 10% of cases in the acute stage, and in no cases in a follow‐up after 6 months (Kasch et al. 2002). Six studies reported TMJ clicking, ranging from 0% to 42% in cases and 10%–34% in controls (Heise, Laskin, and Gervin 1992; Magnusson 1994; Kasch et al. 2002; Klobas, Tegelberg, and Axelsson 2004; Sale and Isberg 2007; Sale, Bryndahl, and Isberg 2014).

Prevalence of TMD pain was the only outcome suitable for inclusion in a meta‐analysis. Different studies reported a range of different outcomes, but it was possible to extract data on prevalence of TMD pain for nine studies that reported this outcome measure, Thus, data from 449 individuals in the acute stage after whiplash trauma, 7912 individuals in the chronic stage after whiplash trauma and 515 controls were included in the meta‐analysis for the outcome of TMD pain. The mean prevalence for TMD pain was 18.9% (95% CI 9.71, 29.98) in the acute case group, 26.1% (95% CI 15.07, 38.79) in the chronic case group, and 5.7% (95% CI 3.08, 8.96) in the control group (Figure 2). High heterogeneity was observed in the acute case group (Q = 26.98 and I ² = 85%) and the chronic case group (Q = 77.1 and I ² = 92%) but not in the control group (Q = 9.71 and I ² = 38%). To further investigate the sources of heterogeneity, we used the Baujat plot and diagnostic tests, which identified the Lampa et al.'s (2020) study in the acute case group and the Ferrari, Schrader, and Obelieniene's (1999) study in the chronic case group as potential outliers and influential studies. After excluding these studies, the mean prevalence of TMD pain in the acute case group decreased to 15.0% (95% CI 7.69, 23.93) with notable reduction in heterogeneity (Q = 7.19, I ² = 58%). In the chronic case group, the mean prevalence of TMD increased slightly to 30.35% (95% CI 21.64–39.80), but the heterogeneity remained high (Q = 48.58, I ² = 90%), indicating substantial variation between studies.

FIGURE 2.

Forest plots for the prevalence of TMD pain in (A) cases in the acute stage after whiplash trauma, (B) cases in the chronic stage after whiplash trauma, and (C) controls with no history of whiplash trauma.

Based on the studies that included control groups, the Risk Ratio for TMD pain was 5.54 (95% CI 2.91, 10.55) in the acute case group (Q = 2.05, I ² = 0%) and 2.85 (95% CI 1.09, 7.43) in the chronic case group (Q = 20.75, I ² = 81%) (Figure 3). After excluding the Ferrari, Schrader, and Obelieniene (1999) study, the Risk Ratio for TMD pain in the chronic case group increased to 3.61% (95% CI 2.38, 5.47) with substantially reduced heterogeneity (Q = 3.21 and I ² = 6%).

FIGURE 3.

Forest plots for the Risk Ratio for TMD pain in (A) cases in the acute stage after whiplash trauma, (B) cases in the chronic stage after whiplash trauma, based on the studies that included control groups with no history of whiplash trauma.

Based on the risk of bias assessment a majority of the primary studies had an overall low or medium risk of bias as assessed by the Joanna Briggs Institute Prevalence Critical Appraisal Tool (Figure 4). Most included studies, 10 out of 14, also provided details about received funding (Magnusson 1994; Bergman, Andersson, and Isberg 1998; Kasch et al. 2002; Klobas, Tegelberg, and Axelsson 2004; Visscher et al. 2005; Sale and Isberg 2007; Severinsson, Bunketorp, and Wenneberg 2010; Sale, Bryndahl, and Isberg 2014; Lampa et al. 2017, 2020), whereas a declaration of no conflict of interest was provided only in one study (Lampa et al. 2017).

FIGURE 4.

Risk of bias assessment with the Joanna Briggs Institute Prevalence Critical Appraisal Tool for included studies (n = 14).

4. Discussion

The main finding in the present systematic review was a higher prevalence of TMD pain following whiplash trauma, compared to individuals without whiplash trauma. There was no clear evidence of any change (increase or decrease) from the acute to the chronic stage. This finding is important, especially in a clinical context, where knowledge on the natural course of pain conditions is crucial for appropriate patient management.

The meta‐analysis revealed an overall significantly higher prevalence of TMD pain in study populations exposed to whiplash trauma, compared to nonwhiplash groups, both in the acute and chronic stages after trauma. This finding extends the results from a previous review based solely on a qualitative synthesis (Häggman‐Henrikson et al. 2013). There was a marked heterogeneity between the primary studies, with I ² values exceeding 80% in the whiplash groups. The heterogeneity in the acute whiplash group was reduced when one outlier (Lampa et al. 2020) was removed, however in the chronic case group the heterogeneity remined high even after removal of the identified outlier (Ferrari, Schrader, and Obelieniene 1999). This is probably due to a generally higher degree of selection bias in the chronic case group. The explanation for the variance is likely multifaceted, with one main reason being that WAD itself is a condition with considerable heterogeneity in symptoms and a range of manifestations (Al‐Khazali et al. 2020). Thus, the umbrella term WAD encompasses a spectrum of features including pain related symptoms, cognitive symptoms, and non‐specific physical symptoms, such as fatigue and dizziness (Sterner and Gerdle 2004). In addition to this, differences in study designs, definitions, diagnostic criteria, traumatic events (which may or may not have included also direct head trauma), and principles for recruitment, where some studies recruited patients in primary or secondary care, whereas other studies recruited patients at emergency departments or through traffic injury registers, will have affected the overall composition of the study populations. For primary studies that included control groups without a history of whiplash trauma, these were included in the meta‐analysis to provide a reference for the case groups. The prevalence of TMD pain for the control groups, was slightly lower than the prevalence reported in the general population. This could partly be due to the applied exclusion criteria, such as direct head–neck trauma and systemic disease, factors that are related to higher prevalence of TMDs. Together with the finding of a lower heterogeneity, this indicates that the control groups are representative for general population samples without history of whiplash trauma.

For the acute stage after whiplash trauma, Kasch et al. (2002) reported in a case group of almost 50% men, the lowest prevalence of TMD pain among the primary studies (5.3%), which is comparable to the prevalence reported in the general population (Lövgren et al. 2016; Häggman‐Henrikson et al. 2020). One reason for this relatively low prevalence could be related to the fact that the study only included cases with WAD grade 1 according to the Quebec classification (Spitzer et al. 1995). A higher WAD grade, grade 2 or 3, has been shown to be related to a slightly poorer prognosis compared to grade 0 or 1 in a meta‐analysis focusing on predictive factors (Walton et al. 2013). High initial neck pain intensity was the strongest predictive factor of a poor long‐term outcome and female sex was confirmed as a risk factor (Walton et al. 2013). This is noteworthy in relation to the study that reported the lowest prevalence of TMD pain, 4.9%, in the present review (Ferrari, Schrader, and Obelieniene 1999). This study had several methodological issues, including a non‐representative sample with predominately men (84%) recruited based on police reports. This is in line with the finding that this study was identified as an outlier in the meta‐analysis. Furthermore, the control group reported a higher prevalence of TMD pain, 9.4%, than the case group (Ferrari, Schrader, and Obelieniene 1999). It should also be noted that although the mean TMD prevalence generally was higher in the chronic case groups compared to the acute case groups, this may be partly related to selection bias, as several of the studies included cases with persistent pain recruited at specialist clinics (Magnusson 1994; Klobas, Tegelberg, and Axelsson 2004; Visscher et al. 2005) and therefore reporting data from study populations more likely to present with more comorbidities and more severe pain conditions.

TMDs following whiplash trauma may relate to several possible mechanisms. An early hypothesis of a ‘mandibular whiplash’ causing intra‐articular structural damage in the TMJ, has been replaced by the notion of overlap, spread and referral of pain, as demonstrated in experimental studies (Hellström et al. 2000; Svensson et al. 2004; Schmidt‐Hansen et al. 2006). There is also increasing evidence for central sensitisation being prominent in whiplash populations (Van Oosterwijck et al. 2013) with clinical features including such pain spread and referral. There is also a functional integration between the jaw and neck sensorimotor systems (Eriksson et al. 2000) and it has been shown that whiplash trauma can impair jaw function (Häggman‐Henrikson, Zafar, and Eriksson 2002) and that experimental pain can alter jaw–neck motor strategy (Wiesinger et al. 2013). In line with the biopsychosocial model, also psychosocial factors may contribute and among these, kinesiophobia have emerged as a strong predictor (Anarte‐Lazo et al. 2024). With recent pathophysiological models for development of chronic pain, this pain could well be partly or completely nociplastic (Kaplan et al. 2024).

A range of internal and external factors interact in the transition from acute to chronic pain, or in the transition from acute pain to recovery following trauma. High initial neck pain intensity and neck disability in the early stage after whiplash trauma have emerged as strong predictive factors (Walton et al. 2013). Factors that relate to individual susceptibility are, among others, female sex, pain catastrophising, fear of movement and coping strategies (Walton et al. 2013; Anarte‐Lazo et al. 2024). Thus, a lack of coping ability, increased vulnerability, or decreased neurophysiological and/or psychosocial adaptation may play a crucial role in the development of TMD pain following whiplash trauma, particularly in the transition from acute to chronic TMD pain. Also post‐traumatic stress symptoms are associated with poor recovery and related to pain sensitisation (Andersen et al. 2022) and generalised hypersensitivity after whiplash trauma may increase pain sensitivity also in the orofacial region (Westergren et al. 2018). Taken together, many of these identified predictive factors are not markedly different from other chronic pain's risk factors and suggest that other factors not related to the whiplash trauma per se predict chronification (Glare, Aubrey, and Myles 2019; Sharma et al. 2020; Böthun et al. 2024).

4.1. Methodological Considerations

The protocol for the present review was registered and a broad literature search conducted in three data bases in line with present guidelines. The most common reason for exclusion at the full‐text assessment stage was that the study population was not representative for a whiplash population. A majority of the primary studies were performed before the publication of the ICOP criteria and ICOP categories could not be confirmed as most studies were based only on self‐report of individual symptoms and not on a comprehensive assessment or clinical examination. Furthermore, there was a significant uneven geographical distribution, both for studies identified in our literature search, and for the primary studies included in the analysis. This may be related to a high volume of research conducted in Northern Europe on TMD in general and in particular on the topic of the present review, when studies in whiplash populations are combined with the presence of TMD. The heterogeneity of the included studies was expected, and partly related to the condition under study, but also to variations in study design, recruitment and the definition of whiplash. The random effect meta‐analysis allows for some of this variability, and the cases were also subgrouped into the acute and chronic stage to reduce heterogeneity in terms of follow‐up times in relation to the trauma. In accordance with guidelines, a funnel plot to illustrate possible publication bias was not performed as the number of studies included in the meta‐analysis was < 10. The risk of bias assessment was carried out with a widely used instrument especially developed for evaluating prevalence studies (Munn et al. 2014) that has been recommended for use in systematic reviews on prevalence (Kolaski, Logan, and Ioannidis 2023).

4.2. Clinical Implications and Future Research

It was previously suggested that TMDs following whiplash trauma develop in the first year after trauma (Severinsson, Bunketorp, and Wenneberg 2010). However, a recent longitudinal study (Böthun et al. 2024), reported that compared to a baseline examination one month after trauma, whiplash trauma was not a predictor for development of additional orofacial pain over a 2‐year period. Based on this, together with the results from the present review, it seems reasonable to assume that for individuals that develop TMD pain following whiplash trauma, a majority, but not all, do so within the first 3 months.

In the clinical assessment of patients with whiplash trauma, three validated screening questions for TMD (Lövgren et al. 2016) can be utilised (Table S1) to determine whether a more thorough examination of the jaw system is required. These screening questions can easily be incorporated in health questionnaires in any setting. In the management of patients with both TMD pain and post‐traumatic neck pain, a multidisciplinary and multimodal approach, tailored to the individual patient, is recommended. In this context, it should also be noted that exercise therapy is recommended both for patients with WAD (Chrcanovic et al. 2022) and for patients with TMD (List and Axelsson 2010). It is therefore reasonable to explore the possibility of evaluating individually tailored exercise therapy to patients with combined WAD and TMD utilising appropriate patient‐reported outcome measures (PROMs; Sterling et al. 2023). Finally, it should be noted the Global Burden of Disease project predict that neck pain will continue to increase and the importance of collecting global neck pain data (Collaborators 2024). To our knowledge, global data on orofacial pain are however still scarce.

5. Conclusion

TMD pain is more common in individuals with whiplash trauma compared to individuals without a history of neck trauma. The finding of higher prevalence of TMD pain in the chronic stage after trauma compared to the acute stage is probably related to selection bias rather than an actual increase over time, and further studies could discern if, for example, nervous system sensitisation was contributing. Nevertheless, the results suggest an overlap and spread between pain in the jaw and neck regions, triggered by post‐traumatic neck pain. The higher prevalence of TMD pain already in the early acute stage after whiplash trauma, emphasises the need for early comprehensive clinical assessment as well as targeted research for understanding underlying mechanisms.

Author Contributions

All authors reviewed and commented on the manuscript. B.H.H.: conceptualisation, data extraction and qualitative synthesis, draft preparation, critically reviewing manuscript, approval of final version; A.L.: conceptualisation, data extraction, critically reviewing manuscript, approval of final version; C.P.: data extraction, critically reviewing manuscript, approval of final version; H.W.: conceptualisation, critically reviewing manuscript, approval of final version; W.W.: data extraction, meta‐analysis, critically reviewing manuscript, approval of final version; T.L.: conceptualisation, critically reviewing manuscript, approval of final version.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1.

Data Availability Statement

Presented data from the primary studies are available by an electronic bibliographic search. The full search strategy for all data bases is presented in Table 1.