Introduction

In addition to pain and stiffness, fatigue, and sleep problems are important concerns in patients with ankylosing spondylitis (AS) [1]. Various sleep problems including poor quality sleep, sleep onset insomnia, difficulty awakening, and obstructive sleep apnoea syndrome (OSAS) has been reported in AS [2]. The importance of fatigue in AS was already highlighted [3]. Fatigue and quality of life in AS is closely related to sleep disturbances (SD) [4].

Tumor necrosis factor-α (TNF-α) is major therapeutic target in AS [5] and there are several anti-TNF agents which significantly improve the signs and symptoms of AS [6, 7]. But much is not known in a quantitative way about the effect of treatments on sleep problems in AS.

This study is aimed first, to investigate sleep disturbances in AS and secondly, to evaluate the effects of anti-TNF treatment on SD in AS.

Patients and methods

Between January 2008 and 2009, consecutive AS patients fulfilling modified New York criteria [8] and giving informed consent from our university hospital rheumatology outpatient clinic were included into the study. Demographic data and disease characteristics [disease duration, acute phase reactants (erythrocyte sedimentation rate, C-reactive protein, hemogram)] and treatments were recorded from hospital files. Disease activity in AS was assessed with The Bath Ankylosing Spondylitis Functional Index (BASFI) [9] and The Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) [10].

As a control, 86 age- and gender-matched consecutive controls giving informed consent without inflammatory diseases were recruited. For both AS and control group, The Medical Outcomes Study (MOS) Sleep Questionnaire was used for evaluating sleep and problems of sleep [11].

Measurements

The Bath Ankylosing Spondylitis Functional Index (BASFI) consists of eight questions on daily activities and two additional questions that assess patients’ ability to cope with everyday life. Each question is answered on 10-cm horizontal visual analogue scale (VAS). The VAS have no distinguishing marks except the words “easy” and impossible “impossible” at either end of the line to indicate the direction of the severity. The mean of ten scales gives in the BASFI score (0–10), with higher scores indicating more severe impairment.

The Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) is a self-administrated questionnaire consisting of six questions relating to the five major symptoms fatigue, spinal pain, joint pain/swelling, areas of localized tenderness, and morning stiffness, measured in terms of both severity and duration. The patients were asked to mark the degree to which they had experienced aforementioned symptoms over the previous week. Each of the first five questions are answered on 10-cm, unmarked, horizontal VAS, except fort the words “none” and “very severe” on the left and right ends, respectively, while the scale for duration of morning stiffness is graded every 15 min between 0 and 2 h. After converting the question on length of morning stiffness to 0–10, BASDAI is calculated with the formulae [1 + 2 + 3 + 4 + ([5 + 6]/2)]/5.

The Medical Outcomes Study (MOS) Sleep Questionnaire (Table 1) is a 12-item self-report questionnaire that yields 2 sleep problem indexes (SPI-I, 6 items and SPI-II, 9 items) and 6 scale scores: the sleep disturbance scale (SDS, 4 items), daytime somnolence (3 items), snoring (1 item), awaken short of breath or with headache (1 item), and quantity of sleep (1 item). Answers refer to a retrospective assessment over the past 4 weeks. The quantity of sleep is scored as the average number of hours slept per night. Except for the quantity of sleep measure, the MOS sleep scales and indexes are scored on a 0–100 possible range, higher scores indicating more sleep problems.

Table 1 Question concepts and associated sleep scales of the MOS scale
Full size table

Statistical analysis

Statistical analysis was performed with SPSS 11.0 and statistical significance was set at p less than 0.05. Data are presented as mean ± SD or median (range). Student’s t-test was used for comparison of parametric data and Mann–Whitney U test for nonparametric data. Correlation was tested using Pearson’s correlation coefficient for parametric values and Sperman’s correlation test for nonparametric values.

Results

One hundred seventy-one (Female/male: 90/81) AS patients and 86 controls (F/M: 56/30) were included into the study. Demographic characteristics, disease activity parameters, and treatments of AS patients were shown in Table 2.

Table 2 Demographic characteristics and disease activity parameters of 171 AS patients
Full size table

Although there was no difference between AS patients and controls regarding ages and sex, AS patients had higher SDS and SPI-II scores (Table 3). There was no difference between the groups regarding snoring, awakening, sleep adequacy, and somnolence.

Table 3 Sleep abnormalities in patients with AS and controls
Full size table

In subgroup analysis of AS patients according to the treatment, Group A (Patients using NSAID and/or DMARD treatment) had higher BASDAI and BASFI compared with Group B (Patients using anti-TNF treatments) (4.29 ± 2.38 vs. 2.46 ± 2.32, P < 0.001; 1.95 ± 2.15 vs. 0.93 ± 1.31, P < 0.001, respectively). Group A had higher scores of SDS, awaken short of breath or headache, somnolence, and SPI-II than controls. In the comparison of Group B and controls, none of the sleep parameters were statistically different between patients on anti-TNF treatments and controls (Table 4).

Table 4 Comparison of AS subgroups and controls
Full size table

Correlation analysis of AS disease activity with sleep parameters

BASDAI was positively correlated with SPI-I, SPI-II, SDS, and somnolence scale (P < 0.001 and r = 0.480; P < 0.001 and r = 0.548, P < 0.001 and r = 0.481, P < 0.001 and r = 0.481, respectively). BASFI had positive correlation with SDS and somnolence scale (P = 0.044 and r = 0.255, P = 0.001, r = 0.41) ESR was positively correlated with SDS (P = 0.04 and r = 0.277).

Discussion

This is the first study assessing sleep from several perspectives as disturbance, snoring, somnolence, adequate sleep, and awakening in AS. Additionally, effects of treatments on these parameters were investigated. Our results showed that even though AS patients had increased sleep disturbances and sleep problems compared with controls, no significant difference was found between patients using anti-TNF agents and controls.

Although sleep problems are common in the general population, pain conditions, chronic illness, and mood disorders are associated with increased sleep disturbance [12]. In parallel to previous studies [4, 13], our AS patients had increased sleep problems and disturbances than controls. Sleep disturbance was clearly correlated with disease activity measured with BASDAI and BASFI. Moreover, it was correlated with ESR. Patients on NSAID and/or DMARD treatment had higher disease activity so, sleep problems, disturbances, and awakening.

As it was to be expected, treatments that improve AS clinical status also reduce sleep disturbance. For none of the sleep parameters that was a statistically significant difference between patients on anti-TNF treatments and controls. There are observations that etanercept and infliximab could reduce daytime sleepiness in RA [12]. In an efficacy and safety study of golimumab in AS patients, a decrease in sleep disturbance by 3 unit on a 0–20 units scale assessed with Jenkins sleep disturbance scale was shown [14]. We found that using anti-TNF therapy could decrease not only sleep disturbances but also awakening and other sleep problems.

OSAS was found more prevalent in AS than in the general population (11.7–22.5% vs. 1–4%) in previous studies [15, 16]. In our study, there was no difference between AS and controls regarding snoring, somnolence, and awakening. But patients on NSAID and/or DMARD treatment had higher somnolence and awakening. It was shown that 41% of AS patients complaining fatigue, awakened more than 3 times each night [2]. Increased awakening by short of breath could be a result of costochondral and back pain due to disease activity of AS. Anti-TNF treatment could improve somnolence and awakening. Effect of anti-TNF treatment on OSAS in AS patients should be investigated with further studies using polysomnography.

In our study, we had used MOS sleep indexes and scales. But administration and scoring of MOS questionnaires are very difficult [11, 12]. On the other hand, BASDAI is moderately correlated with sleep problem index II and sleep disturbance scale. It is not surprising that patients with increased disease activity (measured with BASDAI) would have more sleep problems and disturbances.

Conclusion

Our data indicate that disease activity plays an important role in sleep disturbance in AS. Anti-TNF agents improve sleep problems in addition to other symptoms of AS.