Association of periodic limb movements during sleep and tinnitus in humans

Both of periodic limb movements during sleep (PLMS) and tinnitus were related with dopaminergic system dysfunction. However, it was still unclear whether PLMS, one kind of sleep disturbances, was associated with chronic tinnitus or not. Thus, we aimed to investigate this issue in humans. Clinical and overnight polysomnographic data of 2849 adults from a community hospital during Nov. 2011 to Jun 2017 in Taiwan was collected retrospectively. The association of PLMS and chronic tinnitus was analyzed by Student’s t-test, Pearson’s Chi-Square test, and multivariate logistic regression. The results showed that the mean age was 50.6 years old (standard deviation, SD = 13.3, range = 18~91) for all subjects. There were 1886 subjects without tinnitus and 963 subjects with tinnitus in this study. The PLMS was not significantly different between subjects without tinnitus (mean = 1.0/h, SD = 3.5/h) and subjects with tinnitus mean = 1.1/h, SD = 3.4/h) by Student’s t-test. The severity of PLMS was not significantly between non-tinnitus and tinnitus subjects by Pearson’s Chi-Square test. Multivariate logistic regression also showed that PLMS was not significantly associated with tinnitus after adjusting age, sex, subjective hearing loss, Parkinson’s disease, and insomnia. In conclusion, PLMS was not associated with chronic tinnitus in humans.

www.nature.com/scientificreports www.nature.com/scientificreports/ there were significantly higher co-morbidity rates of coronary artery disease and cardiovascular diseases (CVD), but not acute myocardial infarction, in subjects with PLMS 22 . Meanwhile, these PLMS-related complications and/ or sequelae might also contribute to tinnitus formation 2,3,5-7 .
To our knowledge, the association of PLMS and tinnitus was never reported till now. Therefore, the aim of this study was to examine the association of PLMS and chronic tinnitus in humans.

Methods
From Nov. 2011 to Jun 2017, clinical and overnight polysomnographic data of 2849 adult patients with sleep disturbances at Dalin Tzu Chi Hospital were retrospectively collected. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Research Ethics Committee of Dalin Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation (No. B10604018). Informed written consent was waived because the study was a retrospective data analysis.
Clinical data including age, sex, subjective hearing loss, insomnia, and Parkinson's disease were acquired before overnight PSG examination. Subjective hearing loss was graded as "no" and "yes" in response to a question of "Do you have difficulty hearing other people clearly in daily activities". Insomnia was graded into five grades as "no", "rare", "sometimes", "often", and "usually". PLMS was treated by two methods. First, PLMS was originally a continuous variable (/h). Second, PLMS was graded into four grades as "no", "mild (>=5/h but <25/h)", "moderate (>=25/h but <50/h), and "severe (>=50/h)". And, chronic tinnitus was graded as "no" and "yes", and regarded as non-tinnitus group and tinnitus group, respectively. Statistical analysis. Student's t-test was used to test the difference of continuous variables between non-tinnitus and tinnitus group. Pearson's Chi-Square test was used to test the association of PLMS with different severity and chronic tinnitus. Multivariate logistic regression was used to test the association of PLMS (continuous values) and chronic tinnitus. All analyses were performed using STATA 10.0 software (Stata Corp, College Station, Texas). P < 0.05 was considered to be significant.

Results
There were 989 females and 1860 males in this study. The mean age was 50.6 years old (standard deviation, SD = 13.3, range=18~91) for all subjects. Among that, there were 1886 subjects without chronic tinnitus and 963 subjects with chronic tinnitus in this study. Table 1 showed the basic characteristics of non-tinnitus and tinnitus groups. Compared to non-tinnitus group, tinnitus group had significantly higher age (mean =54.2 years old, SD = 12.2 versus mean=48.8 years old, SD = 13.5, p < 0.0001), proportion of females (39.4% versus 32.3%, p < 0.001), and prevalence of subjective hearing loss (37.7% versus 5.6%, p < 0.001) and insomnia (87.8% versus 76.5%, p < 0.001). But, PLMS (mean =1.1/h, SD = 3.4 versus mean=1.0/h, SD = 3.5, p = 0.4557) and prevalence of Parkinson's disease (0.9% versus 0.8%, p = 0.808) were not significantly different between non-tinnitus and tinnitus groups. Table 2 showed the association of PLMS with different severity and chronic tinnitus. The severity of PLMS was not significantly associated with chronic tinnitus (p = 0.639 by Pearson's Chi-Square test). Table 3 showed the multivariate logistic regression analysis for the association of PLMS on chronic tinnitus with adjustment of other variables. The results showed that PLMS was not significantly associated with chronic tinnitus after adjusting for age, sex, subjective hearing loss, Parkinson's disease, and insomnia.

Discussion
This retrospective study based on clinical and PSG data showed that PLMS, one kind of sleep disturbances, was not significantly associated with chronic tinnitus before and after adjusting other variables. This clinical finding did not support the expectation based on the dopaminergic system dysfunction for both of PLMS and chronic tinnitus 15,18,19 . Also, although PLMS-related complications and/or sequelae might contribute to tinnitus 2,3,5-7 , current results did not show positive correlation between PLMS and chronic tinnitus in humans at all.
In this study, we showed that older age, male subjects, hearing loss, and insomnia had higher risk for having chronic tinnitus. These findings were very reasonable and expectative. Talking about the relationship of tinnitus and sleep disturbance, loud tinnitus was associated with 2.8-fold and 3.3-fold increases in the odds ratios of  18 . Pramipexole, an agonist of DR2/3, could reduce tinnitus associated with presbycusis 24 . Bupropion, a dopamine reuptake inhibitor, did not enhance the treatment effect of repetitive transcranial magnetic stimulation of the temporal cortex for tinnitus 25 . Salicylate-induced tinnitus might be associated with increasing mRNA expressions of DR1A gene in the cochlear, brainstem and inferior colliculus, hippocampus and parahippocampus, and temporal lobes, but not in the frontal lobes 15 . So, it seemed that DR2/3 activation would suppress tinnitus, whereas DR1 activation would enhance tinnitus.
As for sleep-related movement disorders, although restless legs syndrome (RLS), PLMS, and periodic limb movement disorder (PLMD) were different by definition 26 , their treatments were very similar 27 . RLS might be associated with several conditions, including growing pains, kidney diseases, migraine, diabetes mellitus, epilepsy, rheumatologic disorders, CVD, liver and gastrointestinal disorders, and neuropsychiatric disorders [e.g., attention deficit hyperactivity disorder (ADHD), depression, and conduct disorder 28 . The prevalence of associated diagnoses of PLMS in pediatric population was 46.8% for OSA, 13% for ADHA, 9.1% for migraine, 9.1% for seizure, 6.5% for autism spectrum disorders, 9.1% for narcolepsy, and 96.6% for decreased serum ferritin 29 .
RLS and/or PLMS might also increase cardiovascular risk by increasing heart rate and blood pressure, resulting in sleep fragmentation and sleep deprivation, which produced adverse consequences for neural, metabolic, oxidative, inflammatory, and vascular systems, and iron deficiency, which was also a risk for CVD 30 . Also, PLMS are found in 30% of patients suffering from OSA 31 . Patients with PLMS had higher artery stiffness than their controls and those with moderate/severe OSA did. And, a possible additive effect of OSA and PLMS on arterial stiffness was found 32 . Thus, PLMS might increase the risk of CVD directly and/or indirectly via its comorbid OSA. However, we found that PLMS was not related with tinnitus. Otherwise, it was possible that PLMS might be related with tinnitus indirectly via its comorbid diseases, including OSA as our previous study shown 2 .
This finding did not against the use of DR agonists and/or antagonists on chronic tinnitus clinically. We still supposed that some DR agonists might be helpful for patients with PLMS and/or chronic tinnitus by reducing the chances of PLMS-related sequelae. But, it needs more studies to confirm our hypothesis in the future.

conclusion
This clinical study found that PLMS was not associated with chronic tinnitus in humans, even though possible similar pathopysiology was established in both symptoms.   Table 3. Multivariate logistic regression for tinnitus. *PLMS was treated as a continuous variable, and others were treated as category variables.