Prevalence and distribution of musculoskeletal pain in patients with dizziness—A systematic review

Abstract Background and purpose Musculoskeletal disorders are among the leading causes of disability globally, but their role in patients with dizziness and imbalance is not well understood or explored. Such knowledge may be important as musculoskeletal pain and dizziness can mutually influence each other, leading to a complex condition requiring more comprehensive approaches to promote successful recovery. We conducted a systematic review to examine the extent and characteristic of reported musculoskeletal pain in patients with dizziness. Methods A comprehensive literature search in Medline, Embase, Cochrane, Scopus, Amed, Google Scholar, SveMed+, and Web of Science was conducted in March 2021. Inclusion criteria were studies examining patients with a vestibular diagnosis, patients with cervicogenic dizziness and patients included based on having dizziness as a symptom; and reported musculoskeletal pain. Data regarding age, sex, sample size, diagnosis and musculoskeletal pain was extracted. The Crowe Critical Appraisal Tool was used for assessing methodical quality of the included studies. Results Out of 1507 screened studies, 16 studies met the inclusion criteria. The total sample consisted of 1144 individuals with dizziness. The frequency of patients reporting pain ranged between 43% and 100% in the included studies. Pain intensity were scored between 5 and 7 on a 0–10 scale. Pain in the neck and shoulder girdle was most often reported, but musculoskeletal pain in other parts of the body was also evident. Discussion In the included studies, musculoskeletal pain was highly prevalent in patients with dizziness, with pain intensity that may have a moderate to severe interference with daily functioning. Pain in the neck and shoulder is well documented, but there are few studies addressing musculoskeletal pain in additional parts of the body. More research is needed to understand the relations between dizziness and musculoskeletal pain.

Previous studies have found associations between dizziness and neck pain Malmström et al., 2007), where a possible explanation is that neck pain may cause dizziness via connections between the cervical proprioceptive system and the vestibular nuclei (Kristjansson & Treleaven, 2009;Peng, 2018). It could be advocated that the same rationale would apply for reduced somatosensory input from other parts of the body due to pain. Another explanation could be that patients "lock" the head to restrict head-neck movements, to avoid triggering dizziness. This could lead to increased muscular tension in the neck/upper trunk area and pain may thus develop over time when dizziness persists (Kvåle et al., 2008;Wilhelmsen & Kvåle, 2014).
Some studies have reported that patients with cervicogenic dizziness (CD) (Malmström et al., 2007) and patients with peripheral vestibular dysfunction (Iglebekk et al., 2013;Kvåle et al., 2008;Malmström et al., 2021) also have pain in other parts of the body. The prevalence and distribution of musculoskeletal pain in patients with dizziness is uncertain. To further explore this is interesting since neck pain, which is frequently reported as a local phenomenon in this population , is often part of a wider pain pattern in the general population (Natvig et al., 2010).
The clinical picture in musculoskeletal disorders shows a striking overlap with what features can be found in disorders associated with dizziness, such as predominance in women (Neuhauser et al., 2005;Smith et al., 2014;Treaster & Burr, 2004;Yardley et al., 1998), increasing prevalence with increasing age (Sloane et al., 2001;Smith et al., 2014) and the common comorbidity of anxiety and depression (Blair et al., 2003;Cousins et al., 2017;Neuhauser et al., 2005). As there is a high prevalence of both musculoskeletal pain and dizziness in the general population, the chance of concomitant occurrence is high. We have argued that dizziness and pain can mutually influence each other negatively and lead to a complex condition that requires more comprehensive assessments and treatments to promote successful recovery. Thus, our aim was to conduct a systematic review to examine the extent and characteristic of reported musculoskeletal pain in patients with dizziness in the literature.

| Study design
We performed a systematic review employing the guidelines in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement (Moher et al., 2009;Page et al., 2021). The review study was registered in the Prospero database in advance of the data inclusion (CRD42020183285).

| Eligibility criteria
This review was restricted to articles written in English or Scandinavian languages with no limitation to publication date. Published peer-reviewed studies were included; unpublished studies, books, reviews, case reports, conference papers, editorials and papers not available in full text were not included. The inclusion criteria were studies examining participants with a vestibular diagnosis, patients with CD and patients included based on having dizziness as a symptom; and reported musculoskeletal pain. Regarding musculoskeletal pain, there were no restrictions concerning the type of outcome measures. To eliminate other types of pain, only studies that specified pain as musculoskeletal pain or referred to pain as pain in muscles, joints, bones or tendons were included. Studies concerning exclusively elderly >65 years were excluded to avoid bias as a result of natural changes due to aging. However, studies that included an adult population with a lifespan perspective were retained, even some in the sample were over 65 years of age. Baseline results originally published in the first study (Karlberg et al., 1996) on the same study population.

| Study selection
8 of 15reviewers for inclusion. Any disagreement was resolved through discussion between the reviewers and the co-authors. The reviewer process was facilitated using Rayyan systematic review web application, which allowed blinding in each step of the process (Ouzzani et al., 2016). The selection process is documented in the PRISMA flowchart ( Figure 1).

| Data extraction
Bibliographic data (author, title, year, and study design), diagnosis, inclusion-and exclusion criteria, population (age and gender distribution), sample size, outcome measures regarding musculoskeletal pain were extracted, compared and compiled in a spreadsheet by both reviewers (Table 1). Only baseline information was extracted from intervention-or prospective studies to avoid bias due to treatment effect.

| Data analysis
Due to the heterogeneous nature of the included studies concerning design and outcome measures, meta-analyses of results were not feasible for this review. Thus, only a qualitative assessment of the data was conducted.

| Quality assessment
The quality of the studies was assessed using the Crowe Critical

| Search result and study selection
The initial search identified 1507 articles. After duplicates were Two of the included studies reported from the same study participants (Malmström et al., 2021;Malmström et al., 2019). This is highlighted in Table 1, and no information was duplicated in the analyses. The reason for exclusion is documented in Appendix S2.

| Demographics and diagnosis
The and 10 (100 mm) as "the worst pain ever possible". GCPS consists of 8 items questioning pain intensity, duration of pain and interference of daily activities, social life and work, where 0 points equal "no pain"/ "no interference" and 10 points equal pain as "bad as could be"/ "unable to carry on any activities".
However, the diagnostic criteria for CD varied between the different studies.
In Bracher et al. (2000), 14 out of 15 (93%) patients reported pain in the neck and/or shoulder girdle, 47% had pain in both areas. reported head/neck/eye pain. Krabak et al. (2000) found the Sternocleidomastoid muscle to be the most common site of pain in 73% of the patients (n = 15), followed by the Trapezius, Levator scapulae, Occipitalis, Masseter and Supraspinatus muscles. The same muscles were reported as painful in more than 50% of the individuals in the study by Malmström et al. (2007). In Morinaka (2009)

| Oto-vestibular diagnoses
Among 24 patients with Ménière's disease (Bjorne & Agerberg, 2003), 75% reported neck and shoulder pain, and 50% reported pain in the face or jaws. Similar results were found in two studies on BPPV (Iglebekk et al., 2013;Tjell et al., 2019) where 82%-87% reported neck pain, and 13% reported widespread/generalized pain. In the same studies, peri-/retroorbital pain was reported by 67%-74%, and temporomandibular pain by 59%. In a study addressing clinical features and triggers of vestibular migraine (Beh et al., 2019), neck pain was reported as a trigger of attacks by 3 out of 131 patients. Twelve 3.5.1 | Dizziness from various causes  found that 58% of 238 patients referred to a specialized ear-nose-throat department due to dizziness and balance issues, reported neck pain. Malmström et al. (2021) found that 65.3% (n = 49) of patients referred to a vestibular unit suffering from dizziness and balance dysfunction, reported pain with severity that influenced daily life. In another study with the same participants, Malmström et al. (2019) reported that 61.2% had pain in neck, shoulder or back. Of these, pain was most often reported in the neck and shoulders (57%), followed by back pain (42%), legs (29%), arms (28%), feet (15%) and upper/lower torso (11%). In the patients with dizziness from both groups (n = 85), pain in the head, neck/shoulders and feet was reported significantly more often (p < 0.001), compared with those who did not report dizziness. Women suffered more often from pain than men (p = 0.013), and the patients reporting pain scored significantly higher in the Dizziness Handicap Inventory (DHI) total (p = 0.004) (Malmström et al., 2021).

| Pain intensity
Reported VAS/NRS intensity varied from 5.0 to 7.2, an average score of 6.1 in total.

| Cervicogenic dizziness
The mean VAS score for neck pain in the five studies with CD was 6.0 (range: 5.0-6.9). Grande-Alonso et al. (2018)  to those in the lower end, and the total mean score was 3.6.

| Vestibular diagnoses
Using the VAS scale, 24 patients with Ménière (Bjorne & Agerberg, 2003), scored 7.2 in the neck/shoulder and 6.9 for the face/ jaw. Malmström et al. (2019) found that the mean VAS score of neckshoulder-back pain intensity was 5.4, and the pain severity was 6.0 in patients with dizziness/balance disorders.

| Duration of pain
Duration of musculoskeletal pain was described in four studies (Bracher et al., 2000;Cuenca-Martinez et al., 2018;Malmström et al., 2007Malmström et al., , 2019Malmström et al., , 2021, all of them reported long-lasting pain for at least 3 months. 3.8.1 | Cervicogenic dizziness Bracher et al. (2000) reported 3.8.2 | Dizziness from various causes Malmström et al. (2019) found that the mean duration of pain in patients with symptoms of dizziness and balance dysfunction (n = 49) was 7.8 years. Mean duration of dizziness was 5.8 years.

| Quality of included studies
CCAT scores range from 21 to 37 indicating moderate to good quality.

| DISCUSSION
Approximately 70% of the individuals in the included studies reported musculoskeletal pain. Overall, pain was most often reported in the neck and shoulder areas, but this review showed that pain in other parts of the body was also evident. Muscles located along the nuchal line and shoulder girdle were the most prominent muscles related to pain, such as the Sternocleidomastoid, Trapezius, Levator scapulae, the suboccipital area, and the paraspinal-and interscapular muscles.
Only four studies systematically examined pain in all parts of the body, thus, we still know little about the prevalence of musculoskeletal pain in other parts of the body. A high prevalence of neck pain in CD is not surprising as pain or cervical dysfunction is a central part of the diagnostic criteria. However, pain in the neck and shoulder girdle also seems to be common in patients with dizziness regardless of diagnosis (Bjorne & Agerberg, 2003;Iglebekk et al., 2013;Kvåle et al., 2008;Malmström et al., 2021;Tjell et al., 2019). Neck pain is a common musculoskeletal complaint in general, ranked as the fourth greatest contributor to global disability measured in years lived with disability ), yet the prevalence among the patients included in this review was higher (>50%) than in the general population (Hoy et al., 2010) (approximately 30%). Iglebekk et al. (2013) found that pain in general was the second highest ranked symptom in chronic BPPV, and suggested a likely connection between neck pain, widespread pain and BPPV. The was a retrospective chart review, pain was not a main outcome in the study, and only symptoms accompanying most (>50%) of the migraine attacks were reported. Thus, it is possible that pain was inadequately documented in the medical record.
The pain intensity/severity stated in the respective studies are consistent, reporting pain levels between 5 and 6 on a scale 0-10.
According to the NRS cut-off values for patients with chronic musculoskeletal pain (Boonstra et al., 2016), this reveals that pain overall has moderate to severe interference with functioning in daily life. An exception was seen in the study on patients with Ménière's disease who reported a slightly higher pain intensity in neck and shoulder (7.2), and face or jaw (6.9; Bjorne & Agerberg, 2003). A possible explanation may be the close relation between Ménière's and migraine (Radtke et al., 2002). An important note is that one study (Krabak et al., 2000) stated pain intensity after a 1-year period of non-standardized treatment for the cervical pain. The fact that the pain score in these patients remined high (6.8), emphasizes the substantiality of pain is in this population.
In the studies reporting the duration of pain, nearly all patients report pain for at least 3 months, which is classified as chronic pain (Treede et al., 2015). Previous studies have shown that chronic pain is a significant challenge in the general population, ranging between 12% and 34% in Europe (Breivik et al., 2006), but still not as common as reported among the individuals with dizziness in this review. Bracher et al. (2000) suggested that dizziness arises as an aggravating factor of chronic musculoskeletal dysfunction in the cervical spine and shoulder girdle, based on the marked difference between the average duration of vertigo (52 days) and the average duration of musculoskeletal symptoms (7.5 years).

| Implications for clinical practice and research
Whether symptoms of dizziness and pain co-exist independently of each other, or whether dizziness leads to muscle tenderness or vice versa, cannot be concluded based on our review. Our results do, however, indicate that it could be beneficial to assess and treat symptoms of pain as well as dizziness when pain is present in persistent dizziness.
Musculoskeletal pain seems to be highly prevalent in dizzy patients as indicated in our study, and it may be reasonable to ask whether assessment and treatment of pain need to be addressed routinely and more systematically than what is undertaken in clinical practice today. Increased knowledge of the extent, characteristics and distribution of musculoskeletal pain in patients with dizziness can broaden the perspective of treatment and understanding of longstanding dizziness. The total burden of complaints could have a predictive value and inform expectations for recovery after treatment. Focusing on dizziness in isolation may overlook other important contributors to recovery and the patient's overall function.

| Limitations
As dizziness occurs for many different reasons, a clear delimitation of inclusion criteria was challenging. We wanted to include patients suffering from dizziness where dizziness was the primary problem, excluding studies where dizziness may have been a symptom accompanying psychological, neurological or cardiovascular disorders, or along with natural aging or medications. CD and dizziness caused by whiplash or neck trauma may be overlapping etiologies, indicating potential bias in the inclusion process. However, as CD is viewed as a dizziness condition, we chose to include it in this review.
The same argument was used for two of the papers including patients with chronic BPPV. One of the studies only included patients with a history of head trauma (Iglebekk et al., 2013) which would also likely be accompanied with pain from the musculoskeletal system. In addition to the fact that "chronic" BPPV is disputed, the patients were not diagnosed with BPPV according to international guidelines (von Brevern et al., 2015). However, we included the studies that primarily examined patients with dizziness.
The data on musculoskeletal pain was extracted regardless of the purpose of the study. The authors were aware of the risk of misrepresentation, and the result should therefore be interpreted with caution. Pain can originate from other sources than from the musculoskeletal system, but this was not discussed in any of the included studies. However, based on the context and the inclusion criteria in the separate studies, it is reasonable to assume that there was no other type of pain included in the studies. Studies with unclear definitions of pain, followed by missing or inadequate outcome measures of pain, were excluded. Further, the review reflects the prevalence of musculoskeletal pain in the included studies and does not necessarily reflect the prevalence of pain in all patients with dizziness.

| CONCLUSION
Results from this review indicate that musculoskeletal pain affects a large proportion of individuals with dizziness disorders, with pain intensity that may have a moderate to severe interference with daily functioning. Duration and intensity of pain were comparable across the studies. Neck pain was prominent, not only in patients with CD but for dizziness disorders in general. Pain in other parts of the body was also evident, but the literature emphasizing this is scarce. The prevalence and intensity of musculoskeletal pain are valid for the included studies but cannot necessarily be generalized to all patients with dizziness. Further research to evaluate the association between musculoskeletal pain and dizziness is needed, also considering the characteristics of pain within the different diagnoses.