Sleep Quality After Modified Uvulopalatopharyngoplasty: Results From the SKUP3 Randomized Controlled Trial

Sleep Quality After Modified Uvulopalatopharyngoplasty: Results From the SKUP3 Randomized... Abstract Study Objectives To investigate whether uvulopalatopharyngoplasty (UPPP) improves sleep quality in patients with obstructive sleep apnea (OSA) using the Functional Outcomes of Sleep Questionnaire (FOSQ) and the Karolinska Sleep Questionnaire (KSQ). Methods Randomized controlled trial used to compare modified UPPP with controls at baseline and after 6 months. The controls received delayed surgery and a 6-month postoperative follow-up. All operated patients were offered a 24-month follow-up. At each follow-up, patients underwent polysomnography and vigilance testing and completed questionnaires. Nine scales were evaluated: five subscales and the total score in the FOSQ and three subscales in the KSQ. Results Sixty-five patients, mean 42.3 years (SD 11.5), Friedman stage I and II, body mass index < 36 kg/m2, and moderate-to-severe OSA, were randomized to intervention (n = 32) or control (n = 33). In the FOSQ and in the KSQ, the mean rate of missing values was 6.2% (range 0%–19%) and 20.5% (3%–38%), respectively. In 8 of 9 scales, significant differences were observed between the groups in favor of UPPP. There were significant correlations between results from the questionnaires and objective measures from polysomnography and the vigilance test. At the 6- and 24-month postoperative follow-ups, eight of nine scales were significantly improved compared with baseline. Conclusions In selected patients with OSA, subjective sleep quality was significantly improved 6 months after UPPP compared with controls, with stable improvements 24 months postoperatively. The correlations between subjective and objective outcomes and the long-term stability suggest a beneficial effect from surgery, although a placebo effect cannot be excluded. Trial registration number NCT01659671. http://www.clinicaltrials.gov subjective outcome, sleep quality, surgical treatment, apnea, randomized controlled Statement of Significance Using a randomized controlled trial (RCT), our group has previously shown that surgical treatment with modified uvulopalatopharyngoplasty (UPPP) improves polysomnographic results and vigilance in selected obstructive sleep apnea (OSA) patients. However, less is known about the effect surgery has on subjective symptoms. Using the same cohort, we investigated whether surgery also improves sleep quality using pre- and post- surgery questionnaires and data gathered at two-year postoperative follow-ups. Compared to controls, the UPPP patients were significantly improved after surgery, and the improvement remained stable at the two-year follow-up. These improvements also correlated with outcomes from polysomnography and the vigilance test. We recommend UPPP for OSA patients who cannot adhere to nonsurgical treatments, although these results need verification with more RCTs of other cohorts. INTRODUCTION Obstructive sleep apnea (OSA) is a common cause of poor sleep, with a prevalence of 15% among men and 4% among women.1 OSA is caused by repetitive upper airway obstruction during sleep, which reduces airflow despite continued respiratory effort, resulting in fragmented sleep and oxygen desaturations. As such, patients with OSA present excessive daytime sleepiness. OSA can be treated with noninvasive methods, most commonly continuous positive airway pressure (CPAP) or a mandibular retaining device (MRD). Although these treatments are effective, they have a high nonadherence rate—34.1%.2 For these nonadherent patients, surgical treatment with uvulopalatopharyngoplasty (UPPP) may be an option. Although some studies have questioned the effectiveness of UPPP because of its potential side effects and lack of efficacy,3 our group has published several studies, including a randomized controlled trial (RCT) (SKUP3),4 which shows that modified UPPP may be safe and effective in improving nocturnal respiration in selected patients with OSA. The in-lab polysomnography (PSG) showed a reduced apnea–hypopnea index (AHI) by 60% from 53.3 (19.7) to 21.1 (16.7) events/hour.4,5 Another RCT of UPPP (42 patients; 23 operated and 19 controls) showed similar findings.6 Moreover, SKUP3 found significant improvements after UPPP in blood pressure,7 quality of life (QoL) (SF36), daytime sleepiness (Epworth Sleepiness scale [ESS]), and sleep latency (evaluated using vigilance testing).8 Sleep can be measured both objectively and subjectively. When objective sleep is measured, PSG is considered the golden standard. When subjective sleep is measured in clinical practice, different standardized questionnaires are often used. Examples of self-administered validated sleep questionnaires are the ESS,9 the Functional Outcome of Sleep Questionnaire (FOSQ),10 and the Karolinska Sleep Questionnaire (KSQ).11 This study assesses subjective sleep quality using the FOSQ and the KSQ before and after modified UPPP compared with controls. In addition, because the controls subsequently received UPPP, we evaluated all patients at a 6-month and a 24-month follow-up and correlated the subjective and objective outcomes using previous reports of the same cohort. METHODS This study consists of two parts: an analysis of secondary outcomes from a prospective randomized controlled trial performed by Browaldh et al. in 2013 (SKUP3)4 with two study groups (intervention and controls) and an analysis of secondary outcomes from the 6- and 24-month follow-ups of both the intervention group and the controls who received delayed surgery. All OSA patients referred to the Ear, Nose, and Throat Department of the Karolinska University Hospital, Stockholm, Sweden from June 2007 to May 2011 for UPPP were eligible for this single-center study (Figure 1). All patients underwent clinical investigations by ENT specialists, with fiber-endoscopy of the upper airways. Patients who were considered suitable (no other obvious anatomical abnormality), and willing to undergo pharyngeal surgery, were asked to participate. The following inclusion criteria were used: males and females > 18 years of age; AHI ≥ 15 events/hour of sleep (from PSG); ESS score ≥ 8; excessive daytime sleepiness three times a week or more; body mass index (BMI) < 36 kg/m2; Friedman stage I or II6; and nonadherence with CPAP and MRD treatments, with the exception of patients with Friedman stage I and BMI < 30 kg/m2. The following exclusion criteria were used: serious psychiatric, cardiopulmonary, or neurological disease or an American Society of Anesthesiologists (ASA) classification of >3; patients who decline surgery; insufficient knowledge of Swedish language to complete questionnaires; nightshift workers; patients who could be dangerous in traffic according to responses in our nonstandardized questionnaire; severe nasal congestion (could be included after topical nasal treatment); previous tonsillectomy (as such patients were considered partially treated); Friedman stage III; and severe clinical worsening of OSA during the study. Figure 1 View largeDownload slide Flowchart of participants. n = number of patients completing at least one subscale. FOSQ = Functional Outcome of Sleep Questionnaire, UPPP = Uvulopalatopharyngoplasty, KSQ = Karolinska Sleep Questionnaire, ITT = Intention-to-treat. Figure 1 View largeDownload slide Flowchart of participants. n = number of patients completing at least one subscale. FOSQ = Functional Outcome of Sleep Questionnaire, UPPP = Uvulopalatopharyngoplasty, KSQ = Karolinska Sleep Questionnaire, ITT = Intention-to-treat. The patients underwent all-night in-lab PSG procedures and completed questionnaires at baseline (both groups), preoperatively (control group), and at 6- and 24-month postoperative follow-ups (both groups). The morning after PSG, the patients underwent vigilance testing, a modified OSLER (only performed once at each evaluation point), to measure sleep latency as described by Browaldh et al.8 Intervention The patients were randomized to receive either modified UPPP, including tonsillectomy, within 1 month (intervention group, n = 32) or no treatment for 6 months (control group, n = 33) before surgery. Our method, a modification of the method initially described by Fujita,12 requires only minor resections of the soft palate and uvula using the cold steel technique4 and suturing of the tonsillar pillars including the palatopharyngeal muscle. All surgeons used the same technique. The Functional Outcomes of Sleep Questionnaire The FOSQ, developed by Weaver et al. in 1997,10 is a 30-item questionnaire that includes a total score and five subscales. The FOSQ measures functions of everyday life and not only the patients’ perception of their sleep. The subscales are General productivity (eight questions), social outcome (two questions), activity level (nine questions), vigilance (seven questions), and intimate relationship (four questions). Responses range from 1 to 4, where 1 is extreme difficulty and 4 is no difficulty. A zero means there was no engagement in the activity for reasons other than excessive sleepiness. Those responses and missing answers are excluded and the remaining answers are averaged. Every subscale therefore ranges from 1 to 4 with a total range of 5 to 20 (total score). If a subscale is entirely without answers, that subscale as well as the total score is excluded. Higher scores indicate that sleepiness had a less impact on everyday life. A normal total score is considered to be higher than 18.10,13 This study uses a validated Swedish version of the FOSQ.13 The Karolinska Sleep Questionnaire The KSQ measures subjective sleep and sleepiness and the patients’ perception of their sleep and sleepiness.14 The 1992 KSQ version11 included several sections of questions; the first part consisted of 17 questions—seven referred to sleep quality (insomnia symptoms), three to snoring and cessation of breathing (symptoms of sleep apnea), and five to sleepiness and fatigue during the daytime (symptoms of sleepiness). The responses to the two other questions in the KSQ about (number 7) “nightmares” and about (number 11) “sleeping too few hours” were excluded in the present study according to Nordin et al.10 Thus, this study evaluated responses from 15 questions. Responses range from 0 to 5, where 0 is never and 5 is always. The three subscales of the KSQ were named insomnia, apnea, and sleepiness. Low scores mean better sleep. The scores for each dimension were calculated as means across items. In contrast to the FOSQ, a single missing response in any subscale of the KSQ will generate the exclusion of the whole subscale, thus a missing value. Statistical Analysis As the data are ordinal, they are presented with medians and interquartile ranges. However, because the different subscales were originally calculated as means and to facilitate comparisons with other studies, means and standard deviations are shown in the tables. Changes within groups are calculated by subtracting the follow-up from the baseline values. Wilcoxon-matched pair tests were used to compare the results within groups, and Mann–Whitney U tests were used to compare differences in changes between groups. For correlation analyses in the RCT among the vigilance test (sleep latency), polysomnography (AHI), the ESS, the FOSQ, and the KSQ, a nonparametric Spearman’s rank correlation (SRC) test was used. Follow-up data were used since there were nine of 65 (14%) missing values at baseline for the vigilance test. Intention-to-treat (ITT) analyses were performed for all 71 randomized patients, including the six excluded patients at the start of the study. The missing variables were imputed to “no change from baseline” or “no change from follow-up.” Among the patients with missing values at both baseline and follow-up, the median values for the subscales at baseline were used. A “nonresponse analysis” was performed to evaluate participation bias. The baseline values for the group with values at the 24-month follow-up were compared with the group without values at the 24-month follow-up. A Mann–Whitney U test was used. All statistical calculations were performed using Statistica 10.0 and Stata 13.1. The significance level was determined to be p < .05. Ethics The study was first rejected by the Swedish Regional Ethics Committee, 2007/449-31/3, but after an appeal, it was approved by the Central Ethics Committee, Ö21–2007. RESULTS The study included 65 patients (59 men and 6 women); 32 patients were randomized to the intervention group and 33 patients to the control group. All 65 filled in the questionnaire at baseline, although not all questions for all the subscales were completed. Baseline characteristics (Table 1) showed no significant differences between groups, except for the subscales of Intimate relationships in the FOSQ (p = .034). Table 1 Baseline Characteristics of the Two Study Groups in the RCT.   n  Intervention group (n = 32)  n  Control group (n=33)  p  Age  32  41.5 (11.5)  33  42.9 (11.7)  .662  Body mass index (kg/m2)  32  28.2 (2.9)  33  27.7 (3.3)  .519  Epworth Sleepiness Scale  32  12.5 (3.2)  33  12.9 (3.1)  .519  Apnoea/hypopnoea index (events/h sleep)  32  53.3 (19.7)  33  52.6 (21.7)  .901  Sleep latency (min)  27  31 (11.1)  29  34 (9.0)  .279  Functional Outcome of Sleep questionnaire   General productivity  32  3.5 (0.5)  33  3.5 (0.4)  .571   Social outcome  32  3.6 (0.6)  33  3.6 (0.7)  .295   Activity level  32  3.0 (0.7)  33  3.0 (0.6)  .871   Vigilance  32  3.1 (0.6)  33  3.2 (0.7)  .453   Intimate relations  29  3.4 (0.8)  32  3.0 (0.8)  .034   Total score  29  16.8 (2.8)  32  16.4 (2.4)  .198  Karolinska Sleep Questionnaire subscales   Insomnia  27  16.2 (6.2)  31  17.2 (6.3)  .631   Apnea  28  10.8 (4.4)  26  11.0 (3.8)  .817   Sleepiness  32  12.4 (5.0)  33  13.0 (5.2)  .701    n  Intervention group (n = 32)  n  Control group (n=33)  p  Age  32  41.5 (11.5)  33  42.9 (11.7)  .662  Body mass index (kg/m2)  32  28.2 (2.9)  33  27.7 (3.3)  .519  Epworth Sleepiness Scale  32  12.5 (3.2)  33  12.9 (3.1)  .519  Apnoea/hypopnoea index (events/h sleep)  32  53.3 (19.7)  33  52.6 (21.7)  .901  Sleep latency (min)  27  31 (11.1)  29  34 (9.0)  .279  Functional Outcome of Sleep questionnaire   General productivity  32  3.5 (0.5)  33  3.5 (0.4)  .571   Social outcome  32  3.6 (0.6)  33  3.6 (0.7)  .295   Activity level  32  3.0 (0.7)  33  3.0 (0.6)  .871   Vigilance  32  3.1 (0.6)  33  3.2 (0.7)  .453   Intimate relations  29  3.4 (0.8)  32  3.0 (0.8)  .034   Total score  29  16.8 (2.8)  32  16.4 (2.4)  .198  Karolinska Sleep Questionnaire subscales   Insomnia  27  16.2 (6.2)  31  17.2 (6.3)  .631   Apnea  28  10.8 (4.4)  26  11.0 (3.8)  .817   Sleepiness  32  12.4 (5.0)  33  13.0 (5.2)  .701  Data are mean (SD). p-Values calculated with Mann–Whitney U Test except for age, BMI, AHI, and sleep latency, which was calculated with t-test. Significant p-values (p < .050) are marked in bold. View Large The nonresponse analysis in all operated patients did not find any significant group difference at baseline in the group with values at the 24-month follow-up compared with the group without values at the 24-month follow-up, except for social outcome in the FOSQ (p = .032). Randomized Controlled Trial Functional Outcomes of Sleep Questionnaire The mean rate of missing values was 6% (range 0%–19%) (Table 2). The improvements were significantly larger in the intervention group compared with the control group in all subscales but the subscale social outcome. Thus, there were significant group differences between changes in five of six scales. In the intervention group, the median change of the total score was 1.03 (interquartile range from 0.37 to 3.47); in the control group, this change was −0.22 (−0.85 to 0.33). Results including mean values for all subscales are presented in Table 2 and as a boxplot for total score in Figure 2. Table 2 Results From the RCT From the FOSQ (Five Subscales and Total) and the KSQ (Three Subscales), With Mean Difference Within Groups and Comparisons Between Groups.   Intervention group (n = 32)  Within group difference  Control group (n = 33)  Within group difference  Between group comparisons  n (mv %)    n (mv%)    p  FOSQ   General productivity  30 (6%)  0.25 (0.49)  33 (0%)  −0.04 (0.32)  <.010   Social outcome  30 (6%)  0.13 (0.47)  33 (0%)  −0.02 (0.61)  .627   Activity level  30 (6%)  0.47 (0.55)  33 (0%)  −0.08 (0.36)  <.001   Vigilance  30 (6%)  0.48 (0.65)  33 (0%)  −0.04 (0.52)  <.001   Intimate relationship  26 (19%)  0.38 (0.54)  31 (6%)  −0.06 (0.5)  <.050   Total FOSQ  26 (19%)  1.53 (2.64)  31 (6%)  −0.20 (1.22)  <.001  KSQ   Insomnia  25 (22%)  −6.1 (6.1)  25 (24%)  −2.0 (4.3)  <.050   Apnea  20 (38%)  −7.5 (4.9)  24 (27%)  0.7 (2.7)  <.001   Sleepiness  29 (9%)  −7.45 (5.3)  32 (3%)  2.0 (3.5)  <.001    Intervention group (n = 32)  Within group difference  Control group (n = 33)  Within group difference  Between group comparisons  n (mv %)    n (mv%)    p  FOSQ   General productivity  30 (6%)  0.25 (0.49)  33 (0%)  −0.04 (0.32)  <.010   Social outcome  30 (6%)  0.13 (0.47)  33 (0%)  −0.02 (0.61)  .627   Activity level  30 (6%)  0.47 (0.55)  33 (0%)  −0.08 (0.36)  <.001   Vigilance  30 (6%)  0.48 (0.65)  33 (0%)  −0.04 (0.52)  <.001   Intimate relationship  26 (19%)  0.38 (0.54)  31 (6%)  −0.06 (0.5)  <.050   Total FOSQ  26 (19%)  1.53 (2.64)  31 (6%)  −0.20 (1.22)  <.001  KSQ   Insomnia  25 (22%)  −6.1 (6.1)  25 (24%)  −2.0 (4.3)  <.050   Apnea  20 (38%)  −7.5 (4.9)  24 (27%)  0.7 (2.7)  <.001   Sleepiness  29 (9%)  −7.45 (5.3)  32 (3%)  2.0 (3.5)  <.001  n = number of patients for complete data at baseline and 6-month follow-up, mv% is the rate of missing values for each group and subscale. Data are mean (standard deviation). p-Values from independent samples between-group comparisons, Mann–Whitney U tests. Significant differences, p < .050, are shown in bold. View Large Figure 2 View largeDownload slide Showing the results from the three subscales of the Karolinska Sleep Questionnaire (insomnia, apnea, and sleepiness) with box plots (median, 25% and 75%) and bars (10% and 90%) and outliers; at baseline (all), at 6 months for controls, and for all operated on, aKer 6 and aKer 24 months. n = number of patients in each group, total number and percentage. Figure 2 View largeDownload slide Showing the results from the three subscales of the Karolinska Sleep Questionnaire (insomnia, apnea, and sleepiness) with box plots (median, 25% and 75%) and bars (10% and 90%) and outliers; at baseline (all), at 6 months for controls, and for all operated on, aKer 6 and aKer 24 months. n = number of patients in each group, total number and percentage. Karolinska Sleep Questionnaire The mean rate of missing values was 20.5% (range 3%–38%) (Table 2). There were significant group differences between changes in all three subscales. The median reduction of insomnia was 6 (interquartile range from 3 to 11) in the intervention group and 2 (−2 to 5) in the control group. The subscale of apnea had a median reduction of 9.5 (3.5 to 12) in the intervention group, but did not change in the control group. The median reduction of sleepiness was 8 (2 to 11) in the intervention group and 2 (−0.5 to 4) in the control group. Results with mean are presented in Table 2 and as boxplots in Figure 3. Figure 3 View largeDownload slide Results from all operated with total score at baseline (all), at 6 months (controls), and for all 6 months and 24 months postoperatively. n (%) = the number of patients who responded to the questionnaire, divided with the total number in each group, and the percentage. Boxes represent the median and quartile values, bars are the 10% and 90% values, and dots the outliers. Figure 3 View largeDownload slide Results from all operated with total score at baseline (all), at 6 months (controls), and for all 6 months and 24 months postoperatively. n (%) = the number of patients who responded to the questionnaire, divided with the total number in each group, and the percentage. Boxes represent the median and quartile values, bars are the 10% and 90% values, and dots the outliers. There were significant correlations between follow-up values from the RCT between the questionnaires and the objective measures sleep latency and AHI. Sleep latency correlated on all subscales except social outcome. AHI correlated on all subscales except social outcome, intimate relationship, and total FOSQ. There were significant correlations between the results from the FOSQ and the KSQ and their subscales, as well as with the ESS. The r-values are shown in Table 3. ITT analyses in the RCT were performed for all 71 randomized patients in both questionnaires and this did not change the results. Table 3 Results From the RCT at 6-Month Follow-up With Correlation Tests Between Sleep Latency, AHI, ESS, and Questionnaires.   Sleep latency  AHI  ESS  Insomnia  Apnea  Sleepiness  FOSQ   General prod.  0.61  −0.34  −0.70  −0.70  −0.63  −0.72   Social outc.  0.19  −0.01  −0.40  −0.43  −0.35  −0.42   Activity level  0.54  −0.30  −0.73  −0.76  −0.67  −0.73   Vigilance  0.49  −0.26  −0.75  −0.67  −0.48  −0.71   Intimate rel.  0.46  −0.25  −0.71  −0.72  −0.49  −0.65   Total FOSQ  0.50  −0.22  −0.73  −0.78  −0.64  −0.79  KSQ   Insomnia  −0.36  0.28  0.75    0.66  0.75   Apnea  −0.36  0.65  0.64  0.66    0.76   Sleepiness  −0.32  0.42  0.81  0.75  0.76      Sleep latency  AHI  ESS  Insomnia  Apnea  Sleepiness  FOSQ   General prod.  0.61  −0.34  −0.70  −0.70  −0.63  −0.72   Social outc.  0.19  −0.01  −0.40  −0.43  −0.35  −0.42   Activity level  0.54  −0.30  −0.73  −0.76  −0.67  −0.73   Vigilance  0.49  −0.26  −0.75  −0.67  −0.48  −0.71   Intimate rel.  0.46  −0.25  −0.71  −0.72  −0.49  −0.65   Total FOSQ  0.50  −0.22  −0.73  −0.78  −0.64  −0.79  KSQ   Insomnia  −0.36  0.28  0.75    0.66  0.75   Apnea  −0.36  0.65  0.64  0.66    0.76   Sleepiness  −0.32  0.42  0.81  0.75  0.76    Data are showing r-values. Significant correlations are marked in bold, p < .050. Spearman Rank Order Correlation test. View Large All Operated Patients at the 6- and 24-month Postoperative Follow-up Functional Outcomes of Sleep Questionnaire The per protocol analysis of all operated patients showed a mean rate of missing values of 13% (11%–20%) after 6 months and 25% (23%–29%) after 24 months (Table 4). The median value for general productivity was significantly improved after 6 months—from 3.63 (interquartile range from 3.25 to 3.88) to 4.00 (3.84 to 4.00)—and stable after 24 months. The median value for social outcome was unchanged after both follow-ups—from 4.00 (3.00 to 4.00) at baseline to 4.00 (4.00 to 4.00). The median value for activity level showed a significant improvement after 6 months—from median 3.11 (2.75 to 3.56) to 3.78 (3.36 to 3.89)—and stable after 24 months. The median value for vigilance improved significantly—from 3.29 (2.86 to 3.57) to 3.71 (3.43 to 4.00) after 6 months to 4.00 (3.71 to 4.00) after 24 months (a significant increase between 6 and 24 months). The median value for intimate relationship was significantly improved—from median 3.25 (3.00 to 4.00) to 4.00 (3.25 to 4.00) at 6 months to 4.00 (4.00 to 4.00) after 24 months. The median total FOSQ was significantly improved after 6 months—from 17.28 (15.00 to 18.38) to 19.47 (18.01 to 19.89)—and stable after 24 months. For results with mean values, see Table 4. Table 4 Results for All Operated Patients From the Functional Outcome of Sleep Questionnaire and the Karolinska Sleep Questionnaire at Baseline, 6 Months, and 24 Months.   Baseline  6 months  24 months  Comparison between baseline and 6 months  Comparison between baseline and 24 months  Comparison between 6 and 24 months  n  Mean (SD)  n  Mean (SD)  n  Mean (SD)  p  p  p  FOSQ   General prod.  65  3.5 (0.4)  58  3.8 (0.5)  50  3.8 (0.3)  <.001  <.001  .931   Social outc.  65  3.6 (0.6)  57  3.8 (0.5)  49  3.8 (0.4)  .098  .071  .933   Activity level  65  3.0 (0.6)  58  3.6 (0.5)  50  3.6 (0.5)  <.001  <.001  .812   Vigilance  65  3.2 (0.6)  58  3.6 (0.5)  50  3.7 (0.5)  <.001  <.001  <.050   Intimate rel.  61  3.2 (0.8)  52  3.7 (0.6)  46  3.7 (0.6)  <.001  <.001  .754   Total FOSQ  61  16.6 (2.57)  52  18.5 (2.3)  45  18.8 (1.6)  <.001  <.001  .658  KSQ   Insomnia  58  16.8 (6.3)  55  9.5 (6.8)  47  10.1 (7.1)  <.001  <.001  .327   Apnea  54  10.9 (4.1)  48  3.0 (3.5)  42  2.8 (3.0)  <.001  <.001  .200   Sleepiness  65  12.7 (5.1)  56  5.1 (4.8)  50  5.2 (4.9)  <.001  <.001  .748    Baseline  6 months  24 months  Comparison between baseline and 6 months  Comparison between baseline and 24 months  Comparison between 6 and 24 months  n  Mean (SD)  n  Mean (SD)  n  Mean (SD)  p  p  p  FOSQ   General prod.  65  3.5 (0.4)  58  3.8 (0.5)  50  3.8 (0.3)  <.001  <.001  .931   Social outc.  65  3.6 (0.6)  57  3.8 (0.5)  49  3.8 (0.4)  .098  .071  .933   Activity level  65  3.0 (0.6)  58  3.6 (0.5)  50  3.6 (0.5)  <.001  <.001  .812   Vigilance  65  3.2 (0.6)  58  3.6 (0.5)  50  3.7 (0.5)  <.001  <.001  <.050   Intimate rel.  61  3.2 (0.8)  52  3.7 (0.6)  46  3.7 (0.6)  <.001  <.001  .754   Total FOSQ  61  16.6 (2.57)  52  18.5 (2.3)  45  18.8 (1.6)  <.001  <.001  .658  KSQ   Insomnia  58  16.8 (6.3)  55  9.5 (6.8)  47  10.1 (7.1)  <.001  <.001  .327   Apnea  54  10.9 (4.1)  48  3.0 (3.5)  42  2.8 (3.0)  <.001  <.001  .200   Sleepiness  65  12.7 (5.1)  56  5.1 (4.8)  50  5.2 (4.9)  <.001  <.001  .748  Data are mean (standard deviation). p-Values from the Wilcoxon matched pairs test for within-group comparisons. Significant differences, p < .050, are shown in bold, n = number of patients. View Large Karolinska Sleep Questionnaire In the per protocol analysis of all operated patients, the mean drop-out rate was 18% (range 14%–26%) after 6 months and 26% (23%–35%) after 24 months (Table 4). The median value for insomnia was significantly reduced—from 16.5 (interquartile range 12 to 20) at baseline to 8 (4 to 12) after 6 months—and stable after 24 months. The median value for apnea was significantly reduced—from 12 (9 to 14) to 2 (0 to 4) after 6 months—and stable after 24 months. The median value for sleepiness was reduced highly significantly—from 13 (9 to 15) to 4 (2 to 6.5) after 6 months—and stable after 24 months. See Table 4 for results with mean values. DISCUSSION This RCT showed significant group differences in the changes of the FOSQ and the KSQ. Altogether, eight of nine scales improved in the intervention group. There were also several significant, although moderate, correlations between the results from questionnaires and objective parameters (AHI and sleep latency). Compared with baseline, all operated patients exhibited significant improvements in the FOSQ (except social outcome) and the KSQ at the 6- and 24-month follow-up. BMI, a possible confounder, remained unchanged both in the RCT and in all operated patients during the 6- and 24-month follow-up.15 The subscale vigilance in the FOSQ had the lowest values of all subscales at baseline and had the best improvement in the RCT. This change is of interest since this subscale includes questions on the ability to operate motor vehicles at long distances (question 7), and patients with OSA have a well-established increased risk of traffic accidents.16 In addition, this subscale also correlated with our published results from the RCT in the objective vigilance test (sleep latency, r = 0.49), which showed a significant improvement in the operated group with a mean increase of 9.2 minutes compared with controls,8 providing further indications of a beneficial effect. The subscale of intimate relationships is difficult to interpret due to group differences at baseline and a drop-out rate of 19% in the intervention group. Nonetheless, there was a significant improvement in this group compared with the control group, which showed no improvement. Interestingly, social outcome was the only subscale without significant group differences in the RCT, and no significant improvement for social outcome was evident for all operated patients. However, the baseline value was the highest among the subscales, indicating that the study population did not consider these activities to be much of a problem even before surgery, so there was little room for improvement. The FOSQ has previously been used by Billing et al. in a randomized trial to evaluate treatment of OSA with CPAP after 3 months of treatment.17 As with our results, they found that CPAP-treatment achieved a normalized total FOSQ score above 18. Furthermore, in a cohort study of hypoglossal nerve stimulation of 126 patients with moderate-to-severe OSA, the total FOSQ improved from mean 14.3 (median 14.6) to 17.3 (median 18.2).18 In contrast, a RCT conducted by Marklund et al. evaluating MRD treatment versus placebo in patients with mild-to-moderate OSA showed no significant group differences at follow-up on any subscale of the FOSQ or the KSQ after 4 months with MRD even if the polysomnographic results were significantly improved.19 Although CPAP and/or MRD should remain the first line treatment for OSA, these treatments should be interpreted in the context of low adherence. In contrast, surgery, even if less effective compared with CPAP, is beneficial on both subjective and objective sleep quality and the issue of adherence is nonrelevant. As previously presented, there were only minor-to-moderate postoperative complications, which had resolved at the 6-month follow-up4 and there were no serious complications or mortality after 2 years.15 The primary symptom of OSA—excessive daytime sleepiness—is probably caused by frequent sleep apnea episodes, with fragmented sleep and oxygen desaturations. As expected, there was a marked reduction in the KSQ of the subjective apnea and sleepiness subscale scores in the follow-ups of the operated patients. There was also a slight and significant decrease in the insomnia subscale, which is not surprising since insomnia has been shown to be a common symptom of OSA.20 Mendes and Santos found that CPAP effectively treated both insomnia and OSA symptoms,21 and Glidewell et al.22 had similar results in a study that assessed the relationships among insomnia symptoms, OSA variables, and CPAP use. These studies support the results of the present study, and that also modified UPPP may improve subjective sleep quality. Furthermore, other studies have shown that sleepy patients have improved physiologic response, e.g., decreased blood pressure after CPAP use, compared with nonsleepy patients.23 The patients in the present study can be defined as a sleepy cohort, with a mean ESS at inclusion of 13. As would have been expected with CPAP-treatment, their systolic and diastolic blood pressures were significantly reduced after our modified UPPP, also after 2 years.7 For all interventions, there is the possibility of the placebo effect. In the present study, there were correlations between subjective and objective data in that the questionnaires correlated not only to each other but also to objective measures such as sleep latency and AHI. These results suggest a real positive effect of UPPP, although a placebo effect cannot be excluded. In addition, the 2-year follow-up did not show any sign of a diminishing effect and a placebo effect would presumably lessen over time. In our previous SKUP3 studies of the same cohort, we reported that AHI as well as blood pressure was significantly reduced in the intervention group compared with controls and that the QoL (SF-36) and daytime sleepiness measured by the ESS and sleep latency (vigilance test) had a similar pattern. Reductions of AHI also correlated significantly with improved QoL and reduced ESS.8 These results strengthen the evidence that modified UPPP should be offered to selected patients with OSA. The main strength of the present study is the randomized controlled design and the prospective long-term follow-ups (6 and 24 months) as well as the consistency of subjective and objective results from our previous studies of the same cohort.4 There are several limitations in this study. First, the power analysis was calculated from our original study and based on improvement in AHI and not on the secondary outcomes in this study. Second, with the use of questionnaires, there is always the possibility of recall bias and that the mere context of participating in a study could influence answers. Another weakness is the mean rate for missing values (20%) in the KSQ, with the highest rate in the 24-month follow-up, probably due to single missing responses, leading to exclusion of the subscale. The ITT analyses in the RCT, however, did not change the results. Regardless, the long-term results should be interpreted with caution. Third, there is a problem with generalization of the results, and compared with the general population with OSA,22 our subjects had a lower mean age, a lower mean BMI, and a higher AHI. It should also be noted that patients who had undergone previous tonsillectomy were not included in the study. In selected patients with OSA, subjective sleep quality was significantly improved 6 months after UPPP compared with controls, with stable improvements 24 months postoperatively. The correlations between subjective and objective outcomes and the long-term stability suggest a beneficial effect from surgery, although a placebo effect cannot be excluded. FUNDING This work was partly supported by grants from the Stockholm County Council (ALF) and the Acta Otolaryngologica Foundation. DISCLOSURE STATEMENT None declared. REFERENCES 1. Peppard PE, Young T, Barnet JH, Palta M, Hagen EW, Hla KM. Increased prevalence of sleep-disordered breathing in adults. Am J Epidemiol . 2013; 177( 9): 1006– 1014. Google Scholar CrossRef Search ADS PubMed  2. Rotenberg BW, Murariu D, Pang KP. Trends in CPAP adherence over twenty years of data collection: a flattened curve. J Otolaryngol Head Neck Surg . 2016; 45( 1): 43. Google Scholar CrossRef Search ADS PubMed  3. Franklin KA, Anttila H, Axelsson Set al.   Effects and side-effects of surgery for snoring and obstructive sleep apnea – a systematic review. Sleep . 2009; 32( 1): 27– 36. Google Scholar PubMed  4. Browaldh N, Nerfeldt P, Lysdahl M, Bring J, Friberg D. SKUP3 randomised controlled trial: polysomnographic results after uvulopalatopharyngoplasty in selected patients with obstructive sleep apnoea. Thorax . 2013; 68( 9): 846– 853. Google Scholar CrossRef Search ADS PubMed  5. Lundkvist K, Januszkiewicz A, Friberg D. Uvulopalatopharyngoplasty in 158 OSAS patients failing non-surgical treatment. Acta Otolaryngol . 2009; 129( 11): 1280– 1286. Google Scholar CrossRef Search ADS PubMed  6. Sommer UJ, Heiser C, Gahleitner Cet al.   Tonsillectomy with uvulopalatopharyngoplasty in obstructive sleep apnea. Dtsch Arztebl Int . 2016; 113( 1–02): 1– 8. Google Scholar PubMed  7. Fehrm J, Friberg D, Bring J, Browaldh N. Blood pressure after modified uvulopalatopharyngoplasty: results from the SKUP(3) randomized controlled trial. Sleep Med . 2017; 34: 156– 161. doi: 10.1016/j.sleep.2017.02.030 Google Scholar CrossRef Search ADS PubMed  8. Browaldh N, Bring J, Friberg D. SKUP(3) RCT; continuous study: changes in sleepiness and quality of life after modified UPPP. Laryngoscope . 2016; 126( 6): 1484– 1491. Google Scholar CrossRef Search ADS PubMed  9. Johns MW. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep . 1991; 14( 6): 540– 545. Google Scholar CrossRef Search ADS PubMed  10. Weaver TE, Laizner AM, Evans LKet al.   An instrument to measure functional status outcomes for disorders of excessive sleepiness. Sleep . 1997; 20( 10): 835– 843. Google Scholar PubMed  11. Kecklund G, Åkerstedt T. The psychometric properties of the Karolinska Sleep Questionnaire. J Sleep Res . 1992; 1 (Suppl 1): 113. 12. Fujita S, Conway W, Zorick F, Roth T. Surgical correction of anatomic azbnormalities in obstructive sleep apnea syndrome: uvulopalatopharyngoplasty. Otolaryngol Head Neck Surg . 1981; 89( 6): 923– 934. Google Scholar CrossRef Search ADS PubMed  13. Korpe L, Lundgren J, Dahlström L. Psychometric evaluation of a Swedish version of the Functional Outcomes of Sleep Questionnaire, FOSQ. Acta Odontol Scand . 2013; 71( 5): 1077– 1084. Google Scholar CrossRef Search ADS PubMed  14. Nordin M, Åkerstedt T, Nordin S. Psychometric evaluation and normative data for the karolinska sleep questionnaire. Sleep Biol Rhythms . 2013; 11( 4): 216– 226. Google Scholar CrossRef Search ADS   15. Browaldh N, Bring J, Friberg D. SKUP3: 6 and 24 months follow-up of changes in respiration and sleepiness after modified UPPP. Laryngoscope . 2017. doi: 10.1002/lary.26835. [Epub ahead of print] 16. Tregear S, Reston J, Schoelles K, Phillips B. Obstructive sleep apnea and risk of motor vehicle crash: systematic review and meta-analysis. J Clin Sleep Med . 2009; 5( 6): 573– 581. Google Scholar PubMed  17. Billings ME, Rosen CL, Auckley Det al.   Psychometric performance and responsiveness of the functional outcomes of sleep questionnaire and sleep apnea quality of life instrument in a randomized trial: the HomePAP study. Sleep . 2014; 37( 12): 2017– 2024. Google Scholar CrossRef Search ADS PubMed  18. Strollo PJJr, Soose RJ, Maurer JTet al.  ; STAR Trial Group. Upper-airway stimulation for obstructive sleep apnea. N Engl J Med . 2014; 370( 2): 139– 149. Google Scholar CrossRef Search ADS PubMed  19. Marklund M, Carlberg B, Forsgren L, Olsson T, Stenlund H, Franklin KA. Oral appliance therapy in patients with daytime sleepiness and snoring or mild to moderate sleep apnea: a randomized clinical trial. JAMA Intern Med . 2015; 175( 8): 1278– 1285. Google Scholar CrossRef Search ADS PubMed  20. Krakow B, Melendrez D, Ferreira Eet al.   Prevalence of insomnia symptoms in patients with sleep-disordered breathing. Chest . 2001; 120( 6): 1923– 1929. Google Scholar CrossRef Search ADS PubMed  21. Mendes M, JM dos S. Insomnia as an expression of obstructive sleep apnea syndrome – the effect of treatment with nocturnal ventilatory support. Rev Port Pneumol . 2015; 21( 4): 203– 208. Google Scholar PubMed  22. Glidewell RN, Renn BN, Roby E, Orr WC. Predictors and patterns of insomnia symptoms in OSA before and after PAP therapy. Sleep Med . 2014; 15( 8): 899– 905. Google Scholar CrossRef Search ADS PubMed  23. Zhang D, Luo J, Qiao Y, Xiao Y. Continuous positive airway pressure therapy in non-sleepy patients with obstructive sleep apnea: results of a meta-analysis. J Thorac Dis . 2016; 8( 10): 2738– 2747. Google Scholar CrossRef Search ADS PubMed  © Sleep Research Society 2017. Published by Oxford University Press on behalf of the Sleep Research Society. All rights reserved. For permissions, please e-mail journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png SLEEP Oxford University Press

Sleep Quality After Modified Uvulopalatopharyngoplasty: Results From the SKUP3 Randomized Controlled Trial

Loading next page...
 
/lp/ou_press/sleep-quality-after-modified-uvulopalatopharyngoplasty-results-from-Fi2h3cnCtU
Publisher
Sleep Research Society
Copyright
© Sleep Research Society 2017. Published by Oxford University Press on behalf of the Sleep Research Society. All rights reserved. For permissions, please e-mail journals.permissions@oup.com.
ISSN
0161-8105
eISSN
1550-9109
D.O.I.
10.1093/sleep/zsx180
Publisher site
See Article on Publisher Site

Abstract

Abstract Study Objectives To investigate whether uvulopalatopharyngoplasty (UPPP) improves sleep quality in patients with obstructive sleep apnea (OSA) using the Functional Outcomes of Sleep Questionnaire (FOSQ) and the Karolinska Sleep Questionnaire (KSQ). Methods Randomized controlled trial used to compare modified UPPP with controls at baseline and after 6 months. The controls received delayed surgery and a 6-month postoperative follow-up. All operated patients were offered a 24-month follow-up. At each follow-up, patients underwent polysomnography and vigilance testing and completed questionnaires. Nine scales were evaluated: five subscales and the total score in the FOSQ and three subscales in the KSQ. Results Sixty-five patients, mean 42.3 years (SD 11.5), Friedman stage I and II, body mass index < 36 kg/m2, and moderate-to-severe OSA, were randomized to intervention (n = 32) or control (n = 33). In the FOSQ and in the KSQ, the mean rate of missing values was 6.2% (range 0%–19%) and 20.5% (3%–38%), respectively. In 8 of 9 scales, significant differences were observed between the groups in favor of UPPP. There were significant correlations between results from the questionnaires and objective measures from polysomnography and the vigilance test. At the 6- and 24-month postoperative follow-ups, eight of nine scales were significantly improved compared with baseline. Conclusions In selected patients with OSA, subjective sleep quality was significantly improved 6 months after UPPP compared with controls, with stable improvements 24 months postoperatively. The correlations between subjective and objective outcomes and the long-term stability suggest a beneficial effect from surgery, although a placebo effect cannot be excluded. Trial registration number NCT01659671. http://www.clinicaltrials.gov subjective outcome, sleep quality, surgical treatment, apnea, randomized controlled Statement of Significance Using a randomized controlled trial (RCT), our group has previously shown that surgical treatment with modified uvulopalatopharyngoplasty (UPPP) improves polysomnographic results and vigilance in selected obstructive sleep apnea (OSA) patients. However, less is known about the effect surgery has on subjective symptoms. Using the same cohort, we investigated whether surgery also improves sleep quality using pre- and post- surgery questionnaires and data gathered at two-year postoperative follow-ups. Compared to controls, the UPPP patients were significantly improved after surgery, and the improvement remained stable at the two-year follow-up. These improvements also correlated with outcomes from polysomnography and the vigilance test. We recommend UPPP for OSA patients who cannot adhere to nonsurgical treatments, although these results need verification with more RCTs of other cohorts. INTRODUCTION Obstructive sleep apnea (OSA) is a common cause of poor sleep, with a prevalence of 15% among men and 4% among women.1 OSA is caused by repetitive upper airway obstruction during sleep, which reduces airflow despite continued respiratory effort, resulting in fragmented sleep and oxygen desaturations. As such, patients with OSA present excessive daytime sleepiness. OSA can be treated with noninvasive methods, most commonly continuous positive airway pressure (CPAP) or a mandibular retaining device (MRD). Although these treatments are effective, they have a high nonadherence rate—34.1%.2 For these nonadherent patients, surgical treatment with uvulopalatopharyngoplasty (UPPP) may be an option. Although some studies have questioned the effectiveness of UPPP because of its potential side effects and lack of efficacy,3 our group has published several studies, including a randomized controlled trial (RCT) (SKUP3),4 which shows that modified UPPP may be safe and effective in improving nocturnal respiration in selected patients with OSA. The in-lab polysomnography (PSG) showed a reduced apnea–hypopnea index (AHI) by 60% from 53.3 (19.7) to 21.1 (16.7) events/hour.4,5 Another RCT of UPPP (42 patients; 23 operated and 19 controls) showed similar findings.6 Moreover, SKUP3 found significant improvements after UPPP in blood pressure,7 quality of life (QoL) (SF36), daytime sleepiness (Epworth Sleepiness scale [ESS]), and sleep latency (evaluated using vigilance testing).8 Sleep can be measured both objectively and subjectively. When objective sleep is measured, PSG is considered the golden standard. When subjective sleep is measured in clinical practice, different standardized questionnaires are often used. Examples of self-administered validated sleep questionnaires are the ESS,9 the Functional Outcome of Sleep Questionnaire (FOSQ),10 and the Karolinska Sleep Questionnaire (KSQ).11 This study assesses subjective sleep quality using the FOSQ and the KSQ before and after modified UPPP compared with controls. In addition, because the controls subsequently received UPPP, we evaluated all patients at a 6-month and a 24-month follow-up and correlated the subjective and objective outcomes using previous reports of the same cohort. METHODS This study consists of two parts: an analysis of secondary outcomes from a prospective randomized controlled trial performed by Browaldh et al. in 2013 (SKUP3)4 with two study groups (intervention and controls) and an analysis of secondary outcomes from the 6- and 24-month follow-ups of both the intervention group and the controls who received delayed surgery. All OSA patients referred to the Ear, Nose, and Throat Department of the Karolinska University Hospital, Stockholm, Sweden from June 2007 to May 2011 for UPPP were eligible for this single-center study (Figure 1). All patients underwent clinical investigations by ENT specialists, with fiber-endoscopy of the upper airways. Patients who were considered suitable (no other obvious anatomical abnormality), and willing to undergo pharyngeal surgery, were asked to participate. The following inclusion criteria were used: males and females > 18 years of age; AHI ≥ 15 events/hour of sleep (from PSG); ESS score ≥ 8; excessive daytime sleepiness three times a week or more; body mass index (BMI) < 36 kg/m2; Friedman stage I or II6; and nonadherence with CPAP and MRD treatments, with the exception of patients with Friedman stage I and BMI < 30 kg/m2. The following exclusion criteria were used: serious psychiatric, cardiopulmonary, or neurological disease or an American Society of Anesthesiologists (ASA) classification of >3; patients who decline surgery; insufficient knowledge of Swedish language to complete questionnaires; nightshift workers; patients who could be dangerous in traffic according to responses in our nonstandardized questionnaire; severe nasal congestion (could be included after topical nasal treatment); previous tonsillectomy (as such patients were considered partially treated); Friedman stage III; and severe clinical worsening of OSA during the study. Figure 1 View largeDownload slide Flowchart of participants. n = number of patients completing at least one subscale. FOSQ = Functional Outcome of Sleep Questionnaire, UPPP = Uvulopalatopharyngoplasty, KSQ = Karolinska Sleep Questionnaire, ITT = Intention-to-treat. Figure 1 View largeDownload slide Flowchart of participants. n = number of patients completing at least one subscale. FOSQ = Functional Outcome of Sleep Questionnaire, UPPP = Uvulopalatopharyngoplasty, KSQ = Karolinska Sleep Questionnaire, ITT = Intention-to-treat. The patients underwent all-night in-lab PSG procedures and completed questionnaires at baseline (both groups), preoperatively (control group), and at 6- and 24-month postoperative follow-ups (both groups). The morning after PSG, the patients underwent vigilance testing, a modified OSLER (only performed once at each evaluation point), to measure sleep latency as described by Browaldh et al.8 Intervention The patients were randomized to receive either modified UPPP, including tonsillectomy, within 1 month (intervention group, n = 32) or no treatment for 6 months (control group, n = 33) before surgery. Our method, a modification of the method initially described by Fujita,12 requires only minor resections of the soft palate and uvula using the cold steel technique4 and suturing of the tonsillar pillars including the palatopharyngeal muscle. All surgeons used the same technique. The Functional Outcomes of Sleep Questionnaire The FOSQ, developed by Weaver et al. in 1997,10 is a 30-item questionnaire that includes a total score and five subscales. The FOSQ measures functions of everyday life and not only the patients’ perception of their sleep. The subscales are General productivity (eight questions), social outcome (two questions), activity level (nine questions), vigilance (seven questions), and intimate relationship (four questions). Responses range from 1 to 4, where 1 is extreme difficulty and 4 is no difficulty. A zero means there was no engagement in the activity for reasons other than excessive sleepiness. Those responses and missing answers are excluded and the remaining answers are averaged. Every subscale therefore ranges from 1 to 4 with a total range of 5 to 20 (total score). If a subscale is entirely without answers, that subscale as well as the total score is excluded. Higher scores indicate that sleepiness had a less impact on everyday life. A normal total score is considered to be higher than 18.10,13 This study uses a validated Swedish version of the FOSQ.13 The Karolinska Sleep Questionnaire The KSQ measures subjective sleep and sleepiness and the patients’ perception of their sleep and sleepiness.14 The 1992 KSQ version11 included several sections of questions; the first part consisted of 17 questions—seven referred to sleep quality (insomnia symptoms), three to snoring and cessation of breathing (symptoms of sleep apnea), and five to sleepiness and fatigue during the daytime (symptoms of sleepiness). The responses to the two other questions in the KSQ about (number 7) “nightmares” and about (number 11) “sleeping too few hours” were excluded in the present study according to Nordin et al.10 Thus, this study evaluated responses from 15 questions. Responses range from 0 to 5, where 0 is never and 5 is always. The three subscales of the KSQ were named insomnia, apnea, and sleepiness. Low scores mean better sleep. The scores for each dimension were calculated as means across items. In contrast to the FOSQ, a single missing response in any subscale of the KSQ will generate the exclusion of the whole subscale, thus a missing value. Statistical Analysis As the data are ordinal, they are presented with medians and interquartile ranges. However, because the different subscales were originally calculated as means and to facilitate comparisons with other studies, means and standard deviations are shown in the tables. Changes within groups are calculated by subtracting the follow-up from the baseline values. Wilcoxon-matched pair tests were used to compare the results within groups, and Mann–Whitney U tests were used to compare differences in changes between groups. For correlation analyses in the RCT among the vigilance test (sleep latency), polysomnography (AHI), the ESS, the FOSQ, and the KSQ, a nonparametric Spearman’s rank correlation (SRC) test was used. Follow-up data were used since there were nine of 65 (14%) missing values at baseline for the vigilance test. Intention-to-treat (ITT) analyses were performed for all 71 randomized patients, including the six excluded patients at the start of the study. The missing variables were imputed to “no change from baseline” or “no change from follow-up.” Among the patients with missing values at both baseline and follow-up, the median values for the subscales at baseline were used. A “nonresponse analysis” was performed to evaluate participation bias. The baseline values for the group with values at the 24-month follow-up were compared with the group without values at the 24-month follow-up. A Mann–Whitney U test was used. All statistical calculations were performed using Statistica 10.0 and Stata 13.1. The significance level was determined to be p < .05. Ethics The study was first rejected by the Swedish Regional Ethics Committee, 2007/449-31/3, but after an appeal, it was approved by the Central Ethics Committee, Ö21–2007. RESULTS The study included 65 patients (59 men and 6 women); 32 patients were randomized to the intervention group and 33 patients to the control group. All 65 filled in the questionnaire at baseline, although not all questions for all the subscales were completed. Baseline characteristics (Table 1) showed no significant differences between groups, except for the subscales of Intimate relationships in the FOSQ (p = .034). Table 1 Baseline Characteristics of the Two Study Groups in the RCT.   n  Intervention group (n = 32)  n  Control group (n=33)  p  Age  32  41.5 (11.5)  33  42.9 (11.7)  .662  Body mass index (kg/m2)  32  28.2 (2.9)  33  27.7 (3.3)  .519  Epworth Sleepiness Scale  32  12.5 (3.2)  33  12.9 (3.1)  .519  Apnoea/hypopnoea index (events/h sleep)  32  53.3 (19.7)  33  52.6 (21.7)  .901  Sleep latency (min)  27  31 (11.1)  29  34 (9.0)  .279  Functional Outcome of Sleep questionnaire   General productivity  32  3.5 (0.5)  33  3.5 (0.4)  .571   Social outcome  32  3.6 (0.6)  33  3.6 (0.7)  .295   Activity level  32  3.0 (0.7)  33  3.0 (0.6)  .871   Vigilance  32  3.1 (0.6)  33  3.2 (0.7)  .453   Intimate relations  29  3.4 (0.8)  32  3.0 (0.8)  .034   Total score  29  16.8 (2.8)  32  16.4 (2.4)  .198  Karolinska Sleep Questionnaire subscales   Insomnia  27  16.2 (6.2)  31  17.2 (6.3)  .631   Apnea  28  10.8 (4.4)  26  11.0 (3.8)  .817   Sleepiness  32  12.4 (5.0)  33  13.0 (5.2)  .701    n  Intervention group (n = 32)  n  Control group (n=33)  p  Age  32  41.5 (11.5)  33  42.9 (11.7)  .662  Body mass index (kg/m2)  32  28.2 (2.9)  33  27.7 (3.3)  .519  Epworth Sleepiness Scale  32  12.5 (3.2)  33  12.9 (3.1)  .519  Apnoea/hypopnoea index (events/h sleep)  32  53.3 (19.7)  33  52.6 (21.7)  .901  Sleep latency (min)  27  31 (11.1)  29  34 (9.0)  .279  Functional Outcome of Sleep questionnaire   General productivity  32  3.5 (0.5)  33  3.5 (0.4)  .571   Social outcome  32  3.6 (0.6)  33  3.6 (0.7)  .295   Activity level  32  3.0 (0.7)  33  3.0 (0.6)  .871   Vigilance  32  3.1 (0.6)  33  3.2 (0.7)  .453   Intimate relations  29  3.4 (0.8)  32  3.0 (0.8)  .034   Total score  29  16.8 (2.8)  32  16.4 (2.4)  .198  Karolinska Sleep Questionnaire subscales   Insomnia  27  16.2 (6.2)  31  17.2 (6.3)  .631   Apnea  28  10.8 (4.4)  26  11.0 (3.8)  .817   Sleepiness  32  12.4 (5.0)  33  13.0 (5.2)  .701  Data are mean (SD). p-Values calculated with Mann–Whitney U Test except for age, BMI, AHI, and sleep latency, which was calculated with t-test. Significant p-values (p < .050) are marked in bold. View Large The nonresponse analysis in all operated patients did not find any significant group difference at baseline in the group with values at the 24-month follow-up compared with the group without values at the 24-month follow-up, except for social outcome in the FOSQ (p = .032). Randomized Controlled Trial Functional Outcomes of Sleep Questionnaire The mean rate of missing values was 6% (range 0%–19%) (Table 2). The improvements were significantly larger in the intervention group compared with the control group in all subscales but the subscale social outcome. Thus, there were significant group differences between changes in five of six scales. In the intervention group, the median change of the total score was 1.03 (interquartile range from 0.37 to 3.47); in the control group, this change was −0.22 (−0.85 to 0.33). Results including mean values for all subscales are presented in Table 2 and as a boxplot for total score in Figure 2. Table 2 Results From the RCT From the FOSQ (Five Subscales and Total) and the KSQ (Three Subscales), With Mean Difference Within Groups and Comparisons Between Groups.   Intervention group (n = 32)  Within group difference  Control group (n = 33)  Within group difference  Between group comparisons  n (mv %)    n (mv%)    p  FOSQ   General productivity  30 (6%)  0.25 (0.49)  33 (0%)  −0.04 (0.32)  <.010   Social outcome  30 (6%)  0.13 (0.47)  33 (0%)  −0.02 (0.61)  .627   Activity level  30 (6%)  0.47 (0.55)  33 (0%)  −0.08 (0.36)  <.001   Vigilance  30 (6%)  0.48 (0.65)  33 (0%)  −0.04 (0.52)  <.001   Intimate relationship  26 (19%)  0.38 (0.54)  31 (6%)  −0.06 (0.5)  <.050   Total FOSQ  26 (19%)  1.53 (2.64)  31 (6%)  −0.20 (1.22)  <.001  KSQ   Insomnia  25 (22%)  −6.1 (6.1)  25 (24%)  −2.0 (4.3)  <.050   Apnea  20 (38%)  −7.5 (4.9)  24 (27%)  0.7 (2.7)  <.001   Sleepiness  29 (9%)  −7.45 (5.3)  32 (3%)  2.0 (3.5)  <.001    Intervention group (n = 32)  Within group difference  Control group (n = 33)  Within group difference  Between group comparisons  n (mv %)    n (mv%)    p  FOSQ   General productivity  30 (6%)  0.25 (0.49)  33 (0%)  −0.04 (0.32)  <.010   Social outcome  30 (6%)  0.13 (0.47)  33 (0%)  −0.02 (0.61)  .627   Activity level  30 (6%)  0.47 (0.55)  33 (0%)  −0.08 (0.36)  <.001   Vigilance  30 (6%)  0.48 (0.65)  33 (0%)  −0.04 (0.52)  <.001   Intimate relationship  26 (19%)  0.38 (0.54)  31 (6%)  −0.06 (0.5)  <.050   Total FOSQ  26 (19%)  1.53 (2.64)  31 (6%)  −0.20 (1.22)  <.001  KSQ   Insomnia  25 (22%)  −6.1 (6.1)  25 (24%)  −2.0 (4.3)  <.050   Apnea  20 (38%)  −7.5 (4.9)  24 (27%)  0.7 (2.7)  <.001   Sleepiness  29 (9%)  −7.45 (5.3)  32 (3%)  2.0 (3.5)  <.001  n = number of patients for complete data at baseline and 6-month follow-up, mv% is the rate of missing values for each group and subscale. Data are mean (standard deviation). p-Values from independent samples between-group comparisons, Mann–Whitney U tests. Significant differences, p < .050, are shown in bold. View Large Figure 2 View largeDownload slide Showing the results from the three subscales of the Karolinska Sleep Questionnaire (insomnia, apnea, and sleepiness) with box plots (median, 25% and 75%) and bars (10% and 90%) and outliers; at baseline (all), at 6 months for controls, and for all operated on, aKer 6 and aKer 24 months. n = number of patients in each group, total number and percentage. Figure 2 View largeDownload slide Showing the results from the three subscales of the Karolinska Sleep Questionnaire (insomnia, apnea, and sleepiness) with box plots (median, 25% and 75%) and bars (10% and 90%) and outliers; at baseline (all), at 6 months for controls, and for all operated on, aKer 6 and aKer 24 months. n = number of patients in each group, total number and percentage. Karolinska Sleep Questionnaire The mean rate of missing values was 20.5% (range 3%–38%) (Table 2). There were significant group differences between changes in all three subscales. The median reduction of insomnia was 6 (interquartile range from 3 to 11) in the intervention group and 2 (−2 to 5) in the control group. The subscale of apnea had a median reduction of 9.5 (3.5 to 12) in the intervention group, but did not change in the control group. The median reduction of sleepiness was 8 (2 to 11) in the intervention group and 2 (−0.5 to 4) in the control group. Results with mean are presented in Table 2 and as boxplots in Figure 3. Figure 3 View largeDownload slide Results from all operated with total score at baseline (all), at 6 months (controls), and for all 6 months and 24 months postoperatively. n (%) = the number of patients who responded to the questionnaire, divided with the total number in each group, and the percentage. Boxes represent the median and quartile values, bars are the 10% and 90% values, and dots the outliers. Figure 3 View largeDownload slide Results from all operated with total score at baseline (all), at 6 months (controls), and for all 6 months and 24 months postoperatively. n (%) = the number of patients who responded to the questionnaire, divided with the total number in each group, and the percentage. Boxes represent the median and quartile values, bars are the 10% and 90% values, and dots the outliers. There were significant correlations between follow-up values from the RCT between the questionnaires and the objective measures sleep latency and AHI. Sleep latency correlated on all subscales except social outcome. AHI correlated on all subscales except social outcome, intimate relationship, and total FOSQ. There were significant correlations between the results from the FOSQ and the KSQ and their subscales, as well as with the ESS. The r-values are shown in Table 3. ITT analyses in the RCT were performed for all 71 randomized patients in both questionnaires and this did not change the results. Table 3 Results From the RCT at 6-Month Follow-up With Correlation Tests Between Sleep Latency, AHI, ESS, and Questionnaires.   Sleep latency  AHI  ESS  Insomnia  Apnea  Sleepiness  FOSQ   General prod.  0.61  −0.34  −0.70  −0.70  −0.63  −0.72   Social outc.  0.19  −0.01  −0.40  −0.43  −0.35  −0.42   Activity level  0.54  −0.30  −0.73  −0.76  −0.67  −0.73   Vigilance  0.49  −0.26  −0.75  −0.67  −0.48  −0.71   Intimate rel.  0.46  −0.25  −0.71  −0.72  −0.49  −0.65   Total FOSQ  0.50  −0.22  −0.73  −0.78  −0.64  −0.79  KSQ   Insomnia  −0.36  0.28  0.75    0.66  0.75   Apnea  −0.36  0.65  0.64  0.66    0.76   Sleepiness  −0.32  0.42  0.81  0.75  0.76      Sleep latency  AHI  ESS  Insomnia  Apnea  Sleepiness  FOSQ   General prod.  0.61  −0.34  −0.70  −0.70  −0.63  −0.72   Social outc.  0.19  −0.01  −0.40  −0.43  −0.35  −0.42   Activity level  0.54  −0.30  −0.73  −0.76  −0.67  −0.73   Vigilance  0.49  −0.26  −0.75  −0.67  −0.48  −0.71   Intimate rel.  0.46  −0.25  −0.71  −0.72  −0.49  −0.65   Total FOSQ  0.50  −0.22  −0.73  −0.78  −0.64  −0.79  KSQ   Insomnia  −0.36  0.28  0.75    0.66  0.75   Apnea  −0.36  0.65  0.64  0.66    0.76   Sleepiness  −0.32  0.42  0.81  0.75  0.76    Data are showing r-values. Significant correlations are marked in bold, p < .050. Spearman Rank Order Correlation test. View Large All Operated Patients at the 6- and 24-month Postoperative Follow-up Functional Outcomes of Sleep Questionnaire The per protocol analysis of all operated patients showed a mean rate of missing values of 13% (11%–20%) after 6 months and 25% (23%–29%) after 24 months (Table 4). The median value for general productivity was significantly improved after 6 months—from 3.63 (interquartile range from 3.25 to 3.88) to 4.00 (3.84 to 4.00)—and stable after 24 months. The median value for social outcome was unchanged after both follow-ups—from 4.00 (3.00 to 4.00) at baseline to 4.00 (4.00 to 4.00). The median value for activity level showed a significant improvement after 6 months—from median 3.11 (2.75 to 3.56) to 3.78 (3.36 to 3.89)—and stable after 24 months. The median value for vigilance improved significantly—from 3.29 (2.86 to 3.57) to 3.71 (3.43 to 4.00) after 6 months to 4.00 (3.71 to 4.00) after 24 months (a significant increase between 6 and 24 months). The median value for intimate relationship was significantly improved—from median 3.25 (3.00 to 4.00) to 4.00 (3.25 to 4.00) at 6 months to 4.00 (4.00 to 4.00) after 24 months. The median total FOSQ was significantly improved after 6 months—from 17.28 (15.00 to 18.38) to 19.47 (18.01 to 19.89)—and stable after 24 months. For results with mean values, see Table 4. Table 4 Results for All Operated Patients From the Functional Outcome of Sleep Questionnaire and the Karolinska Sleep Questionnaire at Baseline, 6 Months, and 24 Months.   Baseline  6 months  24 months  Comparison between baseline and 6 months  Comparison between baseline and 24 months  Comparison between 6 and 24 months  n  Mean (SD)  n  Mean (SD)  n  Mean (SD)  p  p  p  FOSQ   General prod.  65  3.5 (0.4)  58  3.8 (0.5)  50  3.8 (0.3)  <.001  <.001  .931   Social outc.  65  3.6 (0.6)  57  3.8 (0.5)  49  3.8 (0.4)  .098  .071  .933   Activity level  65  3.0 (0.6)  58  3.6 (0.5)  50  3.6 (0.5)  <.001  <.001  .812   Vigilance  65  3.2 (0.6)  58  3.6 (0.5)  50  3.7 (0.5)  <.001  <.001  <.050   Intimate rel.  61  3.2 (0.8)  52  3.7 (0.6)  46  3.7 (0.6)  <.001  <.001  .754   Total FOSQ  61  16.6 (2.57)  52  18.5 (2.3)  45  18.8 (1.6)  <.001  <.001  .658  KSQ   Insomnia  58  16.8 (6.3)  55  9.5 (6.8)  47  10.1 (7.1)  <.001  <.001  .327   Apnea  54  10.9 (4.1)  48  3.0 (3.5)  42  2.8 (3.0)  <.001  <.001  .200   Sleepiness  65  12.7 (5.1)  56  5.1 (4.8)  50  5.2 (4.9)  <.001  <.001  .748    Baseline  6 months  24 months  Comparison between baseline and 6 months  Comparison between baseline and 24 months  Comparison between 6 and 24 months  n  Mean (SD)  n  Mean (SD)  n  Mean (SD)  p  p  p  FOSQ   General prod.  65  3.5 (0.4)  58  3.8 (0.5)  50  3.8 (0.3)  <.001  <.001  .931   Social outc.  65  3.6 (0.6)  57  3.8 (0.5)  49  3.8 (0.4)  .098  .071  .933   Activity level  65  3.0 (0.6)  58  3.6 (0.5)  50  3.6 (0.5)  <.001  <.001  .812   Vigilance  65  3.2 (0.6)  58  3.6 (0.5)  50  3.7 (0.5)  <.001  <.001  <.050   Intimate rel.  61  3.2 (0.8)  52  3.7 (0.6)  46  3.7 (0.6)  <.001  <.001  .754   Total FOSQ  61  16.6 (2.57)  52  18.5 (2.3)  45  18.8 (1.6)  <.001  <.001  .658  KSQ   Insomnia  58  16.8 (6.3)  55  9.5 (6.8)  47  10.1 (7.1)  <.001  <.001  .327   Apnea  54  10.9 (4.1)  48  3.0 (3.5)  42  2.8 (3.0)  <.001  <.001  .200   Sleepiness  65  12.7 (5.1)  56  5.1 (4.8)  50  5.2 (4.9)  <.001  <.001  .748  Data are mean (standard deviation). p-Values from the Wilcoxon matched pairs test for within-group comparisons. Significant differences, p < .050, are shown in bold, n = number of patients. View Large Karolinska Sleep Questionnaire In the per protocol analysis of all operated patients, the mean drop-out rate was 18% (range 14%–26%) after 6 months and 26% (23%–35%) after 24 months (Table 4). The median value for insomnia was significantly reduced—from 16.5 (interquartile range 12 to 20) at baseline to 8 (4 to 12) after 6 months—and stable after 24 months. The median value for apnea was significantly reduced—from 12 (9 to 14) to 2 (0 to 4) after 6 months—and stable after 24 months. The median value for sleepiness was reduced highly significantly—from 13 (9 to 15) to 4 (2 to 6.5) after 6 months—and stable after 24 months. See Table 4 for results with mean values. DISCUSSION This RCT showed significant group differences in the changes of the FOSQ and the KSQ. Altogether, eight of nine scales improved in the intervention group. There were also several significant, although moderate, correlations between the results from questionnaires and objective parameters (AHI and sleep latency). Compared with baseline, all operated patients exhibited significant improvements in the FOSQ (except social outcome) and the KSQ at the 6- and 24-month follow-up. BMI, a possible confounder, remained unchanged both in the RCT and in all operated patients during the 6- and 24-month follow-up.15 The subscale vigilance in the FOSQ had the lowest values of all subscales at baseline and had the best improvement in the RCT. This change is of interest since this subscale includes questions on the ability to operate motor vehicles at long distances (question 7), and patients with OSA have a well-established increased risk of traffic accidents.16 In addition, this subscale also correlated with our published results from the RCT in the objective vigilance test (sleep latency, r = 0.49), which showed a significant improvement in the operated group with a mean increase of 9.2 minutes compared with controls,8 providing further indications of a beneficial effect. The subscale of intimate relationships is difficult to interpret due to group differences at baseline and a drop-out rate of 19% in the intervention group. Nonetheless, there was a significant improvement in this group compared with the control group, which showed no improvement. Interestingly, social outcome was the only subscale without significant group differences in the RCT, and no significant improvement for social outcome was evident for all operated patients. However, the baseline value was the highest among the subscales, indicating that the study population did not consider these activities to be much of a problem even before surgery, so there was little room for improvement. The FOSQ has previously been used by Billing et al. in a randomized trial to evaluate treatment of OSA with CPAP after 3 months of treatment.17 As with our results, they found that CPAP-treatment achieved a normalized total FOSQ score above 18. Furthermore, in a cohort study of hypoglossal nerve stimulation of 126 patients with moderate-to-severe OSA, the total FOSQ improved from mean 14.3 (median 14.6) to 17.3 (median 18.2).18 In contrast, a RCT conducted by Marklund et al. evaluating MRD treatment versus placebo in patients with mild-to-moderate OSA showed no significant group differences at follow-up on any subscale of the FOSQ or the KSQ after 4 months with MRD even if the polysomnographic results were significantly improved.19 Although CPAP and/or MRD should remain the first line treatment for OSA, these treatments should be interpreted in the context of low adherence. In contrast, surgery, even if less effective compared with CPAP, is beneficial on both subjective and objective sleep quality and the issue of adherence is nonrelevant. As previously presented, there were only minor-to-moderate postoperative complications, which had resolved at the 6-month follow-up4 and there were no serious complications or mortality after 2 years.15 The primary symptom of OSA—excessive daytime sleepiness—is probably caused by frequent sleep apnea episodes, with fragmented sleep and oxygen desaturations. As expected, there was a marked reduction in the KSQ of the subjective apnea and sleepiness subscale scores in the follow-ups of the operated patients. There was also a slight and significant decrease in the insomnia subscale, which is not surprising since insomnia has been shown to be a common symptom of OSA.20 Mendes and Santos found that CPAP effectively treated both insomnia and OSA symptoms,21 and Glidewell et al.22 had similar results in a study that assessed the relationships among insomnia symptoms, OSA variables, and CPAP use. These studies support the results of the present study, and that also modified UPPP may improve subjective sleep quality. Furthermore, other studies have shown that sleepy patients have improved physiologic response, e.g., decreased blood pressure after CPAP use, compared with nonsleepy patients.23 The patients in the present study can be defined as a sleepy cohort, with a mean ESS at inclusion of 13. As would have been expected with CPAP-treatment, their systolic and diastolic blood pressures were significantly reduced after our modified UPPP, also after 2 years.7 For all interventions, there is the possibility of the placebo effect. In the present study, there were correlations between subjective and objective data in that the questionnaires correlated not only to each other but also to objective measures such as sleep latency and AHI. These results suggest a real positive effect of UPPP, although a placebo effect cannot be excluded. In addition, the 2-year follow-up did not show any sign of a diminishing effect and a placebo effect would presumably lessen over time. In our previous SKUP3 studies of the same cohort, we reported that AHI as well as blood pressure was significantly reduced in the intervention group compared with controls and that the QoL (SF-36) and daytime sleepiness measured by the ESS and sleep latency (vigilance test) had a similar pattern. Reductions of AHI also correlated significantly with improved QoL and reduced ESS.8 These results strengthen the evidence that modified UPPP should be offered to selected patients with OSA. The main strength of the present study is the randomized controlled design and the prospective long-term follow-ups (6 and 24 months) as well as the consistency of subjective and objective results from our previous studies of the same cohort.4 There are several limitations in this study. First, the power analysis was calculated from our original study and based on improvement in AHI and not on the secondary outcomes in this study. Second, with the use of questionnaires, there is always the possibility of recall bias and that the mere context of participating in a study could influence answers. Another weakness is the mean rate for missing values (20%) in the KSQ, with the highest rate in the 24-month follow-up, probably due to single missing responses, leading to exclusion of the subscale. The ITT analyses in the RCT, however, did not change the results. Regardless, the long-term results should be interpreted with caution. Third, there is a problem with generalization of the results, and compared with the general population with OSA,22 our subjects had a lower mean age, a lower mean BMI, and a higher AHI. It should also be noted that patients who had undergone previous tonsillectomy were not included in the study. In selected patients with OSA, subjective sleep quality was significantly improved 6 months after UPPP compared with controls, with stable improvements 24 months postoperatively. The correlations between subjective and objective outcomes and the long-term stability suggest a beneficial effect from surgery, although a placebo effect cannot be excluded. FUNDING This work was partly supported by grants from the Stockholm County Council (ALF) and the Acta Otolaryngologica Foundation. DISCLOSURE STATEMENT None declared. REFERENCES 1. Peppard PE, Young T, Barnet JH, Palta M, Hagen EW, Hla KM. Increased prevalence of sleep-disordered breathing in adults. Am J Epidemiol . 2013; 177( 9): 1006– 1014. Google Scholar CrossRef Search ADS PubMed  2. Rotenberg BW, Murariu D, Pang KP. Trends in CPAP adherence over twenty years of data collection: a flattened curve. J Otolaryngol Head Neck Surg . 2016; 45( 1): 43. Google Scholar CrossRef Search ADS PubMed  3. Franklin KA, Anttila H, Axelsson Set al.   Effects and side-effects of surgery for snoring and obstructive sleep apnea – a systematic review. Sleep . 2009; 32( 1): 27– 36. Google Scholar PubMed  4. Browaldh N, Nerfeldt P, Lysdahl M, Bring J, Friberg D. SKUP3 randomised controlled trial: polysomnographic results after uvulopalatopharyngoplasty in selected patients with obstructive sleep apnoea. Thorax . 2013; 68( 9): 846– 853. Google Scholar CrossRef Search ADS PubMed  5. Lundkvist K, Januszkiewicz A, Friberg D. Uvulopalatopharyngoplasty in 158 OSAS patients failing non-surgical treatment. Acta Otolaryngol . 2009; 129( 11): 1280– 1286. Google Scholar CrossRef Search ADS PubMed  6. Sommer UJ, Heiser C, Gahleitner Cet al.   Tonsillectomy with uvulopalatopharyngoplasty in obstructive sleep apnea. Dtsch Arztebl Int . 2016; 113( 1–02): 1– 8. Google Scholar PubMed  7. Fehrm J, Friberg D, Bring J, Browaldh N. Blood pressure after modified uvulopalatopharyngoplasty: results from the SKUP(3) randomized controlled trial. Sleep Med . 2017; 34: 156– 161. doi: 10.1016/j.sleep.2017.02.030 Google Scholar CrossRef Search ADS PubMed  8. Browaldh N, Bring J, Friberg D. SKUP(3) RCT; continuous study: changes in sleepiness and quality of life after modified UPPP. Laryngoscope . 2016; 126( 6): 1484– 1491. Google Scholar CrossRef Search ADS PubMed  9. Johns MW. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep . 1991; 14( 6): 540– 545. Google Scholar CrossRef Search ADS PubMed  10. Weaver TE, Laizner AM, Evans LKet al.   An instrument to measure functional status outcomes for disorders of excessive sleepiness. Sleep . 1997; 20( 10): 835– 843. Google Scholar PubMed  11. Kecklund G, Åkerstedt T. The psychometric properties of the Karolinska Sleep Questionnaire. J Sleep Res . 1992; 1 (Suppl 1): 113. 12. Fujita S, Conway W, Zorick F, Roth T. Surgical correction of anatomic azbnormalities in obstructive sleep apnea syndrome: uvulopalatopharyngoplasty. Otolaryngol Head Neck Surg . 1981; 89( 6): 923– 934. Google Scholar CrossRef Search ADS PubMed  13. Korpe L, Lundgren J, Dahlström L. Psychometric evaluation of a Swedish version of the Functional Outcomes of Sleep Questionnaire, FOSQ. Acta Odontol Scand . 2013; 71( 5): 1077– 1084. Google Scholar CrossRef Search ADS PubMed  14. Nordin M, Åkerstedt T, Nordin S. Psychometric evaluation and normative data for the karolinska sleep questionnaire. Sleep Biol Rhythms . 2013; 11( 4): 216– 226. Google Scholar CrossRef Search ADS   15. Browaldh N, Bring J, Friberg D. SKUP3: 6 and 24 months follow-up of changes in respiration and sleepiness after modified UPPP. Laryngoscope . 2017. doi: 10.1002/lary.26835. [Epub ahead of print] 16. Tregear S, Reston J, Schoelles K, Phillips B. Obstructive sleep apnea and risk of motor vehicle crash: systematic review and meta-analysis. J Clin Sleep Med . 2009; 5( 6): 573– 581. Google Scholar PubMed  17. Billings ME, Rosen CL, Auckley Det al.   Psychometric performance and responsiveness of the functional outcomes of sleep questionnaire and sleep apnea quality of life instrument in a randomized trial: the HomePAP study. Sleep . 2014; 37( 12): 2017– 2024. Google Scholar CrossRef Search ADS PubMed  18. Strollo PJJr, Soose RJ, Maurer JTet al.  ; STAR Trial Group. Upper-airway stimulation for obstructive sleep apnea. N Engl J Med . 2014; 370( 2): 139– 149. Google Scholar CrossRef Search ADS PubMed  19. Marklund M, Carlberg B, Forsgren L, Olsson T, Stenlund H, Franklin KA. Oral appliance therapy in patients with daytime sleepiness and snoring or mild to moderate sleep apnea: a randomized clinical trial. JAMA Intern Med . 2015; 175( 8): 1278– 1285. Google Scholar CrossRef Search ADS PubMed  20. Krakow B, Melendrez D, Ferreira Eet al.   Prevalence of insomnia symptoms in patients with sleep-disordered breathing. Chest . 2001; 120( 6): 1923– 1929. Google Scholar CrossRef Search ADS PubMed  21. Mendes M, JM dos S. Insomnia as an expression of obstructive sleep apnea syndrome – the effect of treatment with nocturnal ventilatory support. Rev Port Pneumol . 2015; 21( 4): 203– 208. Google Scholar PubMed  22. Glidewell RN, Renn BN, Roby E, Orr WC. Predictors and patterns of insomnia symptoms in OSA before and after PAP therapy. Sleep Med . 2014; 15( 8): 899– 905. Google Scholar CrossRef Search ADS PubMed  23. Zhang D, Luo J, Qiao Y, Xiao Y. Continuous positive airway pressure therapy in non-sleepy patients with obstructive sleep apnea: results of a meta-analysis. J Thorac Dis . 2016; 8( 10): 2738– 2747. Google Scholar CrossRef Search ADS PubMed  © Sleep Research Society 2017. Published by Oxford University Press on behalf of the Sleep Research Society. All rights reserved. For permissions, please e-mail journals.permissions@oup.com.

Journal

SLEEPOxford University Press

Published: Jan 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial