Abstract
Tinnitus is a prevalent medical disorder which frequently becomes chronic and severe. Furthermore, quality of life can become compromised with many experiencing comorbid insomnia. We hypothesize that insomnia is a highly prevalent symptom and diagnostic category accompanying tinnitus. Our article reviews the tinnitus literature examining the prevalence of insomnia, the sleep disturbances found, and any methodological issues. Our literature search included a number of databases such as PubMed, Cochrane, and Embase. We found that 16 prior studies had sufficient data presented that allowed for an assessment of the prevalence rate of insomnia in tinnitus; the prevalence rate ranged from 10% to 80% (most rates were over 40%). The overwhelming majority of these studies inadequately defined insomnia as a diagnosis but described it only as a symptom. They focused predominantly on questionnaires (sent via the mail) asking only 1 to 4 questions on whether tinnitus disturbs sleep. Frequently, the only question asked was whether tinnitus disturbed a patient’s sleep without clarifying whether there were problems with daytime functioning. Thus, a valid insomnia diagnosis could not be established. Even in the few studies that asked the necessary information to establish a diagnosis of insomnia, only 1 study provided it. The presence of insomnia in tinnitus was associated with a more severe form of tinnitus. Alarmingly, insomnia was mainly untreated despite evidence supporting that successful treatment of insomnia might also help comorbid tinnitus. Because insomnia is significantly prevalent in tinnitus patients and appears to potentially further impact negatively on one’s quality of life, clinicians should address this possibility with a detailed clinical evaluation; incorporating self-rating questionnaires on sleep could be clinically helpful. If insomnia is present, therapy should be considered.
Introduction
Tinnitus is a highly prevalent medical disorder occurring in approximately 10% to 30% of people.1–3 Typically, it consists of the perception of a ringing noise in the absence of an external sound; less commonly, the sound may be perceived as a whining, buzzing, hissing, humming, or a whistling noise.1,4,5 Unfortunately, tinnitus frequently (25%-35%) becomes a chronic and severe illness.6–9 In a large study of more than 13 000 patients with tinnitus conducted by the American Tinnitus Association, tinnitus patients reported a compromise in daily life functioning consisting of impaired social interaction (69%), impaired work functioning (57%), and reduced enjoyment in life (86%). 10 In addition, such patients frequently experience insomnia, diminished concentration, and an increased incidence of anxiety and depression.2,11–13 The idea that a chronic medical condition such as tinnitus can effect daily functioning and mental health is supported in the literature for various other conditions such as rheumatoid arthritis14,15 and cardiovascular disorders.16,17
Of the many associated factors and/or consequences of tinnitus, insomnia is a highly prevalent and major factor that can adversely affect tinnitus itself and lead to stress and potential psychiatric complications.3,10 A large literature has demonstrated that the presence of insomnia, even in a general population, is highly associated with negative consequences of daily functioning and the subsequent development of anxiety, depression, and other psychiatric complications.18–22 Thus, it is likely that patients who have tinnitus plus insomnia will have a particularly significant compromise in daily functioning and quality of life.
The association of tinnitus and insomnia has been described for over 30 years. 3 This relationship is not surprising because a patient having tinnitus is struggling and being distracted with noises at bedtime. A further explanation tying tinnitus to insomnia is that there is a lack of a masking effect of environmental noise just prior to or during sleep allowing tinnitus sounds to interfere with sleep initiation or maintenance.23,24 This appears to be particularly noteworthy for tinnitus patients as they try to fall sleep.25,26 Recently, it has been suggested that both tinnitus and insomnia share a similar underlying physiological mechanism, ie, hyperarousal of the sympathetic nervous system which may be another explanation for the prevalent comorbidity of these disorders. 27
Our article will review the tinnitus literature examining the prevalence of insomnia (either as a symptom or as a diagnostic category) with its wide reported range, the particular sleep disturbances found, the questionnaires used, and the possible methodological issues that were present. We hypothesized that insomnia as a symptom and diagnosis will be widely prevalent in a tinnitus population and will be associated with a more severe form of tinnitus. It is noteworthy that our article is one of the first to critically compare and review prior studies that predominantly focus on the interrelationship of insomnia to tinnitus. We felt this attention would be potentially important and helpful for health professionals and patients as insomnia appears to complicate and probably worsen tinnitus.
Methods
During 2011 to 2012, we conducted an extensive review of the subject of the comorbidity of insomnia and tinnitus for a series of scientific presentations. An updated review was conducted in 2017 for this current manuscript. A literature search included PubMed as well as other databases such as the Cochrane and Embase databases on all articles from 1960 to the present with cross-references being tinnitus, insomnia, and sleep disturbances. Sixteen studies were reviewed that addressed in some way the prevalence of insomnia in patients with tinnitus.7,9,25,28–41 The study by Meikle and Meikle and colleagues presented the same data in the 2 separate publications9,35 (see Table 1). A focus of the review was to identify what criteria of insomnia were used (a symptom or diagnosis), the questionnaires or questions used to assess insomnia, the setting of the population studied, use of control groups, the major findings, etc. If a diagnosis of insomnia was employed, we sought to establish what sleep nomenclature was used. Insomnia symptoms were defined as any sleep disturbance caused by tinnitus such as disturbances in sleep onset or sleep maintenance or a deterioration in the quality of sleep. A diagnosis of insomnia, depending on the specific sleep nomenclature used, was defined as at least including the presence of a sleep disturbance in combination with a compromise in daytime functioning.
A review of studies evaluating tinnitus comorbid with insomnia.
Abbreviations: BDI, Beck Depression Inventory; DFA, difficulty falling asleep; EMA, early morning awakening; HAMA, Hamilton Anxiety Scale; HAMD, Hamilton Depression Scale; ICD-9, International Classification of Diseases, Ninth Revision; ISI, Insomnia Severity Index; PSQI, Pittsburgh Sleep Quality Index; RIS, Regensburg Insomnia Scale; STAI, State Trait Anxiety Inventory; TEQ, Tinnitus Effects Questionnaire.
Statistics
For the review of the 16 studies examining tinnitus comorbidity with insomnia, the data are presented in Table 1. No attempt at a meta-analysis was performed given that the methodology and outcomes were so varied and that there were a small number of studies; furthermore, the information was predominantly descriptive in nature.
Results
The review of previous studies evaluating the relationship of tinnitus to insomnia consisted of 16 publications with a prevalence of comorbid insomnia and/or sleep disturbances ranging from 10.1% to 79.5% (Table 1).7,9
These prevalence rates were based almost exclusively on the presence (versus absence) of insomnia symptoms and/or sleep disturbances determined to be secondary to tinnitus (see Table 1).
Although 9 (56.2%) studies elicited information about what aspects of sleep were disturbed (difficulty falling asleep, intermittent awakenings, and early morning awakening) in patients with tinnitus,9,25,30,34,35,37–40 only 2 (12.5%) studies provided these data in their articles. Lasisi and Gureje 38 suggested that tinnitus patients had a similar rate of sleep disturbances across the sleep cycle. In contrast, Axelsson and Ringdahl 25 found that although all aspects of sleep were significantly disturbed, difficulty falling asleep was the most prevalent; the latter findings were similarly reported in the past by House. 23
Only 5 (31.2%) studies addressed negative daytime consequences of which are necessary for a diagnosis of insomnia.9,35,38,40,41 Surprisingly, only 1 (6.2%) of the studies used a sleep nomenclature 40 ; thus, a formal comorbid diagnosis of insomnia could not be appropriately given except for one study. The data defining insomnia were obtained predominantly with the use of questionnaires alone,7,25,28,29,31–34,39 a retrospective chart review, 36 rarely by clinical interviews alone, 38 or the combination of questionnaires, retrospective chart reviews, and/or clinical interviews.9,30,35,37,40,41 The questionnaires used varied widely including open-ended ones asking patients to list all their complaints in the order of importance.28,31 Many contained self-rating assessments which included 1 to 4 questions on sleep7,9,25,29,32–35; those who used only 1 question in their questionnaire to elicit information on sleep simply asked whether tinnitus interfered with one’s sleep.7,9,25,32,33,35 Three (18.8%) studies used well-validated self-rating scales for insomnia, such as the Pittsburgh Sleep Quality Index (7 sleep-related items), 39 the Insomnia Severity Index (7 sleep-related items), 40 and the Regensburg Insomnia Scale (10 sleep-related items). 41
Control groups (nontinnitus patients) were only occasionally used.25,34,38,39 In general, even in the nontinnitus controls, insomnia was frequently present. Nonetheless, insomnia rates were greater in the tinnitus group. This was highlighted in the study by Lasisi and Gureje 38 where insomnia was found in 54% of the tinnitus population and 34% of the nontinnitus population (P= .002). The findings of Asplund 34 are of particular interest in that they suggest that even sex might be an important factor regarding the association of tinnitus to insomnia. This study revealed that difficulty getting back to sleep after waking up at night was similar in males with tinnitus compared with males without tinnitus (20% versus 18%); in contrast, females in the tinnitus group had a greater rate of difficulty getting back to sleep after waking up than females in the nontinnitus group (45% versus 35%).
Only 3 (18.8%) studies focused on a geriatric population7,34,38 despite the fact that tinnitus and insomnia are more often seen in the elderly. 3 Nonetheless, many of the studies had a wide age range of participants. Most of the sites recruiting patients were from ENT and tinnitus clinics,7,9,29–33,35,37,40,41 one (6.2%) study used a tinnitus self-help group, 28 4 (25%) studies used a community sample.25,34,37,39 Questionnaires were sent to patients prior to or at their initial visit for a tinnitus evaluation7,9,29–33,35,37,40,41 or as part of a general survey.28,34,38,39
Three (18.8%) studies reported rates of insomnia symptoms as infrequent (sometimes or few days a week) versus frequent (often or nearly every day of the week). As expected, the insomnia rates decreased from the more frequent symptom group to the infrequent symptom group by the following: 44.5% versus 24.6%,9,35 40.3% versus 35.6%, 32 and 46.3% versus 33.2%. 33 Regarding listing any exclusionary criteria, this was rarely addressed; 2 (12.5%) studies excluded tinnitus patients if they had depressive symptoms warranting a diagnosis of a psychiatric disorder 30 or if they had “unstable” psychiatric or medical disorders 41 ; the evaluation process to establish these exclusions was generally unclear.
A detailed characterization of tinnitus including common associated features was not consistently reported. Although by definition, patients with tinnitus had ringing in the ears (or “ear noises” as described by Axelsson and Ringdahl) 25 without any external origin, other defining aspects were described less clearly. A clarification of whether patients had chronic tinnitus (greater than 6 months) versus acute tinnitus (6 months or less) or pulsatile versus nonpulsatile tinnitus was sketchy; only the studies by Hallam, 30 Folmer and Greist, 32 Folmer, 33 and Schecklmann et al 41 exclusively studied patients with chronic tinnitus. Thus, one can only presume that most of the studies (12 of 16) (75%) reported on a mixture of tinnitus patients (acute plus chronic) because it was unstated.7,9,25,28,29,31,34–40 Regarding pulsatile tinnitus, 1 study clearly excluded patients with pulsatile tinnitus 36 and only 1 study identified that 24% of their tinnitus sample had pulsatile tinnitus 7 ; interestingly, most of the studies (14 of 16) (88%) did not state the number of patients with pulsatile tinnitus. Hearing loss and/or impairment is well known to be highly associated with tinnitus.7,9,25,28–35,37–41 As highlighted by Sanchez et al, 36 approximately 85% to 96% of patients with tinnitus demonstrate some form of hearing loss. Similar rates of hearing loss were reported by the studies in our review that conducted audiometric testing.7,9,29,31–33,35,36
Discussion
Our review found that the overwhelming majority of the studies evaluating comorbid insomnia in patients with tinnitus (15 [93.8%] of the 16 studies) reported on the presence of insomnia as a symptom.7,9,25,28–39,41 The latter was defined as having sleep disturbances secondary to tinnitus in 14 (93.3%) of 15 studies7,9,25,28–37,39,41 or having a sleep disturbance, nonrestorative sleep, or daytime sleepiness in 1 (6.7%) of the 15 studies. 38 The actual prevalence rate of insomnia symptoms ranged from 10% to 80% with most studies reporting a rate above 40% (see Table 1). Only 1 (8.8%) of the 16 studies listed in Table 1 defined insomnia as a diagnosis using the International Classification of Diseases, Ninth Revision (ICD-9) sleep nomenclature; they reported a prevalence rate of 27%. 40
Most striking in this review was a marked diversity of criteria and assessments used to define insomnia. Many of the studies relied solely on questionnaires to assess sleep, asking as few as 1 to 4 specific questions,7,9,25,32–35 whereas others used validated self-rating insomnia scales39–41; a small number of studies used open-ended questionnaires, asking about accompanying disturbances to a patients’ tinnitus in general.28,31 The variability of the criteria to define insomnia and the methodology assessing it likely contributed to the wide prevalence rate. Highlighting this issue is the study that counted tinnitus patients as also having insomnia criteria if they only had “daytime sleepiness.” 38 “Daytime sleepiness” may be a consequence of insomnia but is only one of the necessary requirements defining insomnia—having an impairment of daytime functioning. The rationale to define insomnia solely by this one criterion is aberrant and not explained by the authors.
The fact that only 1 (6.2%) of the 16 studies listed in Table 1 defined insomnia as a diagnosis 40 is worthy of highlighting and is not totally surprising; these studies were predominantly conducted by specialists involved in the otolaryngology field and not involved in the sleep, psychiatry, or behavioral medicine fields.7,9,25,28–41 This may have contributed to unfamiliarity with sleep nomenclatures and therefore lack of their use. Interestingly, the information necessary to have made a diagnosis of insomnia—a sleep disturbance plus daytime dysfunction, plus a minimum duration of symptoms—was obtained in 4 of these studies.9,35,38,40,41 Nonetheless, the actual data were not presented in these articles so that a reader could not retroactively make a diagnosis of insomnia.
The difference between a symptom versus a diagnosis of insomnia has significant epidemiologic and severity of illness implications. As reviewed by Neubauer and Flaherty, episodic sleep disturbances are common occurring in about 30% of adults. In contrast, when one assesses the general population for more sustained symptoms lasting at least a month, the prevalence of insomnia is reduced to only a 10% rate. 42 We suspect that the diagnosis of insomnia will be reserved for only a minority of tinnitus patients who have insomnia. As our review points out, the literature has not critically answered how frequently tinnitus patients develop a diagnosis of insomnia resulting from their physical disorder. An insomnia diagnosis requires that a tinnitus patient who has sleep disturbances must also have impairment of daytime functioning. Therefore, tinnitus patients who only have insomnia symptoms likely do not have this more severe complication. Because the severity of tinnitus appears to be highly associated with the severity of sleep disturbances and vice versa,7,9,25,30,32,35,39,40,43–45 we predict that tinnitus patients with a diagnosis of insomnia will have more severe tinnitus presentations and in greatest need of hypnotherapy.
No matter what the exact prevalence rate of insomnia is in patients with tinnitus (either as a symptom or as a diagnostic entity), it is a significant comorbidity. Furthermore, it is likely that the relationship of tinnitus to insomnia is highly correlated and may be bidirectional. The presence of insomnia in tinnitus patients has been reported to predict poor tolerance and increased discomfort to tinnitus.43,46,47 Tinnitus patients with sleep difficulties had more severe forms of tinnitus than tinnitus patients without insomnia7,9,25,30,32,35,39,40,43–45,48 Supporting a bidirectionality and important linkage between tinnitus and insomnia is the finding by Folmer that after 22 months of following up a tinnitus population who underwent various treatments for comorbid conditions (psycho-education, masking, psychotherapy), both tinnitus and insomnia improved. 33 In addition, treatments for tinnitus using sound stimulation are associated with decreased tinnitus intensity as well as improvement of sleep quality.49,50 Previously, Folmer 32 suggested that effective treatment of insomnia should help the severity of tinnitus. Patients with tinnitus have themselves rated that improved sleep was the most important condition that lessened the severity of tinnitus. 47 Supporting this are the findings that melatonin51,52 and nortriptyline 53 (both non-FDA [Food and Drug Administration]–approved hypnotics) not only improved insomnia but also alleviated symptoms of tinnitus.
Tinnitus patients who have a comorbid diagnosis of insomnia should be carefully identified and appropriately treated. Furthermore, because tinnitus is so highly correlated and bidirectional, insomnia, irrespective of whether insomnia satisfies the definition of a diagnosis, should be treated. Unfortunately, insomnia in tinnitus populations is seldom treated25,32,39,40,46 and possibly not well identified. This is not surprising in that the National Sleep Foundation found that insomnia was rarely diagnosed and treated.54–56 Early treatment of insomnia should ease the burden of having tinnitus, lessen its severity, and possibly prevent or reduce depression and anxiety frequently associated with it. Thus, health specialists should make every effort to assess every tinnitus patient for comorbid insomnia. The use of self-rating scales for insomnia such as the Pittsburgh Sleep Quality Index to assist the clinician in identifying insomnia/sleep disturbances should be considered when evaluating patients with tinnitus. 57 Furthermore, to increase the likelihood of identifying insomnia in tinnitus patients, the latter should be made aware of the possibility of having insomnia (eg, via pamphlets) and encouraged to discuss it with their physician.
One major limitation in most of the studies reviewed in this article is that other causes of insomnia besides tinnitus may exist which were not investigated. When researching tinnitus with insomnia, one should always include an assessment for psychiatric disorders, particularly generalized anxiety disorder (GAD) and major depressive disorder (MDD) as well as psychiatric symptoms. These psychiatric disorders and symptoms are frequently comorbid with tinnitus and correlate to tinnitus severity3,12,58–60; they are also independently associated with insomnia. 61 In a recent review of the relationship of mood disorders to tinnitus, the authors found that the presence of depression was highly correlated with tinnitus severity in 18 of 20 studies. 59 Thus, it is important to include a psychiatric examination. Of the 16 studies reviewed here (see Table 1), only 2 studies acknowledged a prior psychiatric history. The study by Meikle et al9,35 excluded patients with a diagnosis of depression, and the study by Schecklmann et al 41 excluded patients who had any “unstable” psychiatric disorder. Nonetheless, the aforementioned studies did not define how the psychiatric diagnoses were determined other than via a chart review.9,35,41 If a psychiatric disorder is the main cause of the sleep disorder, appropriate treatment of the mental disorder might be the most appropriate action with or without the addition of hypnotherapy. To help clarify the relationship of tinnitus to insomnia, future studies must include a psychiatric examination perhaps with the aid of a structured rating scale conducted by mental health professionals.
Another cause of insomnia, particularly in the elderly, is organic sleep disorders (OSDs) such as a breathing-related sleep disorder (eg, sleep apnea), narcolepsy, or periodic limb movement disorder (PLMD).11,61–63 Organic sleep disorders might be sizeable in a tinnitus population because these patients tend to be older increasing the likelihood of a sleep-related breathing disorder. 64 Two recent investigators evaluating polysomnography in tinnitus patients with insomnia found that 34% in one study 65 and 50% in another study 47 had sleep obstructive apnea or restless leg syndrome/PLMD. These findings suggest that a significant subgroup of tinnitus patents with insomnia have sleep disorders that are not necessarily secondary to tinnitus per se and might require special treatment for their specific sleep-related disorder. The 16 studies reviewed did not screen for this possibility nor perform polysomnography. They could have inquired about the presence of snoring, whether ones sleeping partner witnessed episodes of apnea during sleep or observed leg and arm movements going into or during sleep. If the responses to these questions suggested an OSD, a definitive diagnosis would be possible in the sleep laboratory. If a patient had one of these disorders, the diagnosis of insomnia or presence of insomnia symptoms secondary to tinnitus might not have been justified.61,62 Therefore, the prevalence of insomnia secondary to tinnitus would have been somewhat lower as these other insomnia conditions preclude the former diagnosis. The correct insomnia diagnosis has bearing on treatment. For example, the use of hypnotic medications appropriate for some insomnia may be contraindicated in certain breathing-related sleep disorders such as sleep apnea where hypnotics could suppress respiration and increase apnea episodes.66,67
In conclusion, insomnia as a symptom is highly prevalent in patients with tinnitus. A prospective study should be undertaken to better understand the relationship of insomnia to tinnitus, including the prevalence rate of the diagnosis of insomnia as well as insomnia symptoms. First, such a study should evaluate patients directly for tinnitus to establish the correct diagnosis. Langguth and colleagues proposed a series of recommendations for future studies of tinnitus focusing on procedures to insure appropriate diagnosis and assessments including physical examinations by a specialist and accompanying otological and audiological examinations such as psychophysic measures of tinnitus (eg, maskability and loudness match). 68 As Meikle and her colleagues have stressed, there is a need for detailed quantitative information about tinnitus.9,35 Therefore, routine use of self-rating questionnaires to supplement clinical interviews, focusing on the specifics of tinnitus such as the duration of illness, severity, and its effect on quality of life, should be considered. Highly recommended is the Tinnitus Functional Index I, a well-validated questionnaire including assessments for the presence of insomnia symptoms resulting from tinnitus. 69 Detailing features of tinnitus, such as whether tinnitus is constant versus intermittent, pulsatile versus nonpulsatile, etc, may help in understanding the complicated interrelationship of tinnitus to insomnia.54,70 A thorough medical workup must be emphasized here. Many comorbid medical illnesses and medications are associated with insomnia. They must be evaluated, detected, and possible treatment interventions made. 54 Second, self-rating scales on sleeps should be administered to establish whether insomnia symptoms are present and whether an insomnia diagnosis is warranted. 68 Clinical interviews should then validate any positive findings from self-ratings as did the studies by Meikle and colleagues.9,35 The database should allow for the use of various sleep nomenclatures to detect insomnia secondary to tinnitus as well as insomnia of different origins. Third, a nontinnitus control group matched for age and sex should be included because insomnia and tinnitus are significantly related to age and other factors.3,67 As noted in our review, only 5 studies used a control group25,34,38–40 of which 3 of them presented their data34,38,39; these controls had high rates of insomnia. Furthermore, in the study by Asplund, 34 certain subgroups of controls did not differ from the tinnitus population—the male tinnitus group did not differ from the male nontinnitus control group regarding being able to go back to sleep after nocturnal awakenings. Finally, because the comorbid psychiatric diagnoses, particularly MDD and GAD, are frequently present,3,58–60,68 which may explain the presence of insomnia, 61 a comprehensive psychiatric examination should also be performed; a structured interview such as the Mini-International Neuropsychiatric Interview would be helpful. 71 Ideally, self-rating scales for depression and anxiety should complement psychiatric clinical interviews. 58 Besides performing a mental status evaluation, a future study should perform polysomnography where insomnia and possible OSDs could be assessed. As mentioned earlier, 2 recent studies suggest that OSDs are seen in one-third to one half of tinnitus patients with insomnia.45,65 Detecting insomnia may have significant treatment implications. The relationship of insomnia to tinnitus appears to be bidirectional; thus, effective treatment of insomnia may also be beneficial in the treatment of tinnitus.
Footnotes
Acknowledgements
The authors wish to thank Ying Chen, MD, for translation assistance with the article by Shao and colleagues that was referenced in this manuscript. This study was presented at 28th CINP World Congress, June, 2012 in Stockholm, Sweden and the Annual Meeting of the American Psychiatric Association, May, 2013 in San Francisco, USA.
Funding:
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests:
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Author Contributions
All authors contributed equally to the conception of the idea for this manuscript, data extraction from references, and writing and editing of the manuscript.