Sleep Breath DOI 10.1007/s11325-013-0920-6

ORIGINAL ARTICLE

Effect of musculoskeletal pain on sleep architecture in patients with obstructive sleep apnea Rashid Nadeem & Hasnain Bawaadam & Asma Asif & Irfan Waheed & Ahmed Ghadai & Adnan Khan & Sara Hamon

Received: 19 September 2013 / Revised: 4 November 2013 / Accepted: 21 November 2013 # Springer-Verlag (outside the USA) 2014

Abstract Rationale Obstructive sleep apnea and chronic musculoskeletal pain both affect sleep. Sleep architecture of patients suffering from both is largely unknown. Objectives This study seeks to define the sleep architecture of patients with chronic musculoskeletal pain and obstructive sleep apnea. Methods Patients with obstructive sleep apnea diagnosed by sleep study during the past 3 years were included. Patients with clinical documentation of chronic musculoskeletal pain constituted cases, while others were classified as controls. Measurements Demographics, clinical factors affecting sleep, medications affecting sleep, Epworth sleepiness scores, and polysomnographic parameters; total sleep time, sleep efficiency, sleep stages, rapid eye movement (REM) sleep onset, apnea– hypopnea index, arousal index, and periodic leg movements were recorded. Results There were 393 subjects: 200 cases (obstructive sleep apnea and chronic musculoskeletal pain) and 193 controls (obstructive sleep apnea alone). There was significant difference in total sleep time (274.5±62.5 vs. 302.2±60.1 min, p =0.0001), sleep efficiency (73.54±15.8 vs. 78.76±14.3 %, p =0.0003), and REM sleep onset (148.18±80.5 vs. 124.8± 70.9 min, p =0.006). Subgroup analysis within the obstructive sleep apnea with chronic musculoskeletal pain group revealed that subjects had better total sleep time and sleep efficiency if they were on REM sleep affecting medications (suppressants and stimulants). Those on REM sleep suppressants slept 25.7 min longer and had 6.4 % more efficient sleep than those not on REM suppressants (p =0.0034 and p =0.0037). Conclusion Patients with obstructive sleep apnea and chronic musculoskeletal pain sleep not only significantly less but also R. Nadeem (*) : H. Bawaadam : A. Asif : I. Waheed : A. Ghadai : A. Khan : S. Hamon RFUMS, North Chicago, IL, USA e-mail: [email protected]

with inferior sleep quality. Their REM sleep is also less in duration and its onset is delayed. Despite low TST and SE, these patients may not exhibit sleepiness. Keywords Sleep apnea . Chronic pain . Sleep architecture

Introduction Chronic pain is among the most disabling disorder in North America [1]. Chronic pain is defined as “pain without apparent biological value that has persisted beyond the normal tissue healing time usually taken to be 3 months” [2]. Fifty to 88 % of patients with chronic pain demonstrate sleep disturbances [3]. Pain disturbs sleep and in turn disturbed sleep further exacerbates pain [4] . Patients experiencing high levels of pain tend to report less sleep time, more delayed sleep onset, and increased nighttime awakenings, which in turn can further increase pain intensity and sleep disturbances [5]. Obstructive sleep apnea (OSA) is a common disorder that affects approximately 2 % of middle-aged women, 4 % of middle-aged men [1] and ≥11 % of the elderly [6]. OSA is also associated with poor sleep efficiency; fragmented sleep with less SWS [7]. Patients with chronic pain and OSA suffer from inefficient sleep pattern [8]; a decrease in total sleep duration with frequent and multiple nighttime awakenings in addition to daytime sleepiness [8–10]. Decreased total sleep, REM sleep, or SWS may adversely impact pain tolerance [11]. Pharmacologic treatment options for pain management have been shown to cause significant effects on sleep architecture, decreased slow wave sleep (SWS), increased stage 2 sleep, and increased REM latency with reduced REM sleep [12]. Patients frequently complain about inadequate medication prescription for pain while clinicians are overcautious in

Sleep Breath

prescribing them since little is known about the magnitude of derangement this combination of disorder can cause on sleep. There is no study to our knowledge that objectively describes the sleep architecture of these patients by utilizing polysomnography (PSG).

sleep stages (percent of time spent in stage N1, stage N2, stage N3, and stage REM sleep), arousal index (AI), sleep and REM onset, periodic limb movement (PLM) index (PLMI), and periodic limb movement arousal index (PLMAI). Study was approved by IRB Hines VA medical center (protocol # 11-065).

Methods Statistical methodology All consecutive patients who were referred for polysomnogram first time for clinically suspected obstructive sleep apnea by sleep clinic physicians at James A Lovell Federal Healthcare center who were found to have OSA were recruited for the study. A total of 393 consecutive individuals were sampled retrospectively from January 1, 2009 to December 31, 2011. One hundred and ninety-three of these individuals had OSA and 200 of these individuals had OSA and pain. Pain was defined as chronic musculoskeletal pain documented in computerized patient records system (CPRS) for a duration of more than 3 months. This chronic pain could be a result from any of these disorders: chronic low back pain, osteoarthritis, rheumatoid arthritis, gouty arthritis, SLE, scleroderma, spinal stenosis, cervical or lumbar spondylosis, ankylosing spondylitis, Sjorgen syndrome, fibromyalgia traumatic, war, work, sports, accidentrelated injury patients, and pain secondary to neoplasm arising from bone, muscle, and joint. From this point onwards, in order to avoid any confusion, the group with both OSA and pain will be addressed as OSA plus Pain group. Data for both groups was collected from CPRS for the demographic variables age, gender, body mass index (BMI); co-morbid conditions like posttraumatic stress disorder (PTSD), benign prostatic hypertrophy (BPH), asthma, and gastroesophageal reflux disease (GERD) as these are the significant confounding factors affecting sleep time and efficiency by causing multiple arousals; medication history that included whether taking REM suppressants [benzodiazepines, barbiturates, tricyclic antidepressants (TCA), monoamine oxidase inhibitors (MAOI), selective serotonin receptor inhibitors (SSRI), clonidine, buspirone, antipsychotics, opioids] or REM stimulants [non-tricyclic and non-SSRI’s; Nefazodone, Trazodone], beta-blockers; metoprolol, atenolol, propranolol, alpha-2 agonist; methyldopa, MAO depletory drugs; Reserpine, alpha1-antagonist; Prazosin, and Terazosin]. We also recorded if subjects were on stimulants: d-amphetamine, l-amphetamine, methyphenidate, nicotine patches/gums, since these drugs tend to suppress REM sleep. Each patient was diagnosed to have obstructive sleep apnea by polysomnography (Somnostar-Pro 7-3A; Carefusion) and sleep stages were scored according to rules described by the Rechtschaffen and Kales (R & K) scoring manual [13] by registered polysomnographic technologist. Epworth sleepiness scale score was also recorded. Data recorded from sleep study included sleep characteristics; total sleep time (TST), sleep efficiency (SE), OSA severity (apnea–hypopnea index (AHI) and rapid eye movement sleep AHI (REM-AHI),

For the sample characteristics, to determine if there was a significant difference between the OSA and OSA plus Pain groups for any continuous variables, a two-sample t test was performed and for any of the categorical variables a Fisher’s exact test was performed. To determine if there was a significant difference between the OSA and OSA plus Pain groups for each of the sleep characteristics, a linear regression was performed. Any sample characteristics found to be significantly different between the OSA and OSA plus Pain groups were included as covariates in the linear model. Finally, to determine if there is a significant difference between sleep characteristics and use of REM suppressants and use of REM stimulants, two-sample t tests were performed. Significance was considered at alpha=0.05. No multiple corrections factor is necessary for Table 1 because it was used to identify any possible confounding variables in the model to use as covariates in Table 2. There is a high level of correlation between the results from Tables 2, 3, 4, 5, and 6 because of the relationships between the sleep characteristics and the same variables are examined on different subsets of the data. The new alpha threshold for the 70 tests performed in these five tables would be 0.00089 would most likely be overly conservative so the p values that meet a point wise significance of 0.05 are reported but any p values that meet this stringent threshold are noted. All statistical analyses were performed using R version 2.14.1.

Results Table 1 presents the sample characteristics. The sample primarily consisted of obese (BMI 33.1±5.7), older (Age 61.8± 11.5), and male (96.7 %) subjects. There was no significant difference for the demographic variables age, BMI or sex between the OSA and OSA plus Pain groups (p >0.05). Approximately 16 % of the sample had PTSD, 15.8 % had BPH, 4.3 % had asthma, with no significant difference between the OSA and OSA plus Pain groups (p >0.05). The OSA plus Pain group was significantly more likely (CI 2.07 (1.26–3.44), p =0.0024) to have GERD, to be taking REM suppressants (CI 1.83 (1.2–2.8), p =0.0043) and finally to be taking REM stimulants (CI 1.67 (1.06–2.63, p =0.0231) than the OSA group.

Sleep Breath Table 1 Sample characteristics

Age BMI (kg/m2) Sex (male) PTSD BPH Asthma GERD Taking REM suppressants Taking REM stimulants

All individuals (N =393)

OSA (N =193)

OSA and pain (N =200)

61.76 (11.52) 33.07 (5.69) 380 (96.7) 63 (16) 62 (15.8) 17 (4.3) 98 (24.9) 170 (43.3) 126 (32.1)

61.17 (11.85) 32.91 (5.79) 189 (96.7) 28 (14.5) 28 (14.5) 10 (5.2) 35 (18.1) 69 (35.7) 51 (26.4)

62.33 (11.2) 33.22 (5.63) 191 (97.9) 35 (17.5) 34 (17) 7 (3.5) 63 (31.5) 101 (50.5) 75 (37.5)

Table 2 presents the summary of the sleep characteristics for the overall group, OSA, and OSA plus Pain groups. There was no significant difference between the groups for OSA severity as measured by AHI and REM.AHI, sleep architecture (percent of time spent in stage N1, stage N2, or stage N3), the Arousal Index, Sleep Onset, PLMI or PLMAI (Table 2, p >0.05). There was a significant influence of pain on both Total Sleep Time (TST) and Sleep efficiency (SE), with the OSA plus Pain group sleeping 27.7 min less and having 5.2 % less efficient sleep than the OSA group (p

Effect of musculoskeletal pain on sleep architecture in patients with obstructive sleep apnea.

Obstructive sleep apnea and chronic musculoskeletal pain both affect sleep. Sleep architecture of patients suffering from both is largely unknown...
183KB Sizes 0 Downloads 0 Views