To our knowledge, this is the first study with reasonable number of patients evaluating the long-term effect of gastric bypass surgery on OSA. Our study demonstrates that bariatric surgery is an effective treatment of OSA even 5 years after the surgery. Three of five patients were considered to have been cured (i.e., they had an AHI < 5), and one out of every four (26.1%) patients were observed to have only mild OSA postoperatively. Furthermore, marked subjective improvements in the QoL were demonstrated.

According to a recent meta-analysis, the short-term effect of bariatric surgery on OSA was associated with an overall rate of remission of 65% [16]. In our study, the corresponding rate was 55% which can be regarded being in line with earlier studies taken into account the long follow-up time of the present study. There are number of studies evaluating the effect of bariatric surgery on OSA, but in most of them, the follow-up time is short, and the number of patients included is small [11,12,13,14,15]. Previous long-term studies including 1 to 36 patients have reported 66–100% improvement of OSA after sleeve gastrectomy [17,18,19,20]. Probably, the most comparable study with 65 patients comparing sleeve gastrectomy and LRGYP reported 42% and 45%, respectively, improvement of OSA 10 years after surgery [21]. Our study is well in line with these previous reports.

Not all patients with OSA were cured from the disease, but considering the rate of patients with mild OSA at 5-year follow-up, we can assume that roughly 80% of patients benefit from bariatric surgery. This is essential, since it is well-documented that the severity of OSA is associated with an increased risk for cardiovascular morbidity and mortality [22, 23]. OSA is a chronic disease with a tendency to worsen with time. Based on current knowledge about the evolution of OSA, we believe that weight gain represents a high risk for future progression towards more severe disease. This study provides long-term evidence that sustained weight reduction by LRYGB can not only significantly improve OSA in obese patients but also prevent the progression of OSA, thus reducing the risk for cardiovascular morbidity and mortality.

Since in our study there was a fifth of OSA patients still suffering moderate or severe symptoms of OSA after surgery, it is important to have a postoperative assessment to identify these patients. A postoperative cardiorespiratory recording is recommended for all OSA patients, and it is important to counsel the patients not to discontinue the treatment, e.g., with CPAP only based on symptoms.

To support our belief of reduced risk of cardiovascular morbidity, the present study also showed marked improvements in desaturation times under 90% and 80% and in mean heart rate. In a recent study, the severity of OSA as measured by degree of nocturnal oxygen desaturation is associated with cardiovascular consequences especially in women, who may be more susceptible to the impact of nocturnal hypoxemia [24]. Previous studies have also demonstrated the importance of desaturation levels on OSA-related symptoms and consequences. The severity of sleep-related desaturations has been shown to be more significant contributor to daytime sleepiness compared to AHI [25]. Moreover, night-time desaturation has also been shown to be associated with memory impairment in adults and increased incidence of diabetes [26, 27]. It could be assumed that surgery induced marked decrease in desaturation time is one reason for beneficial overall effects on obesity-related comorbidities.

Obesity and OSA are associated with decreased QoL, and bariatric surgery has been shown to increase it both short- and long-term regardless of indications for surgery [28,29,30,31,32]. Studies comparing the effect of bariatric surgery on QoL in patients with OSA and without OSA are still scarce. In a retrospective study by de Raaff et al. [33], patients with OSA had lower postoperative scores on public distress and work after laparoscopic gastric bypass compared to patients without OSA. This finding is in contrast with our results since we noticed distinct improvements in four dimensions and total score in patients with OSA. All physiological parameters measured in our study, such as AHI, weight, heart rate etc., improved supporting beneficial changes in subjective well-being. Besides the study design, another explanatory factor compared to the earlier study may be the difference in follow-up time. In our study, the follow-up time was longer (5 years vs. 15 months). On the other hand, it has been reported that QoL after surgery improves in the short term but declines slightly after 2 years [30].

Some limitations of the present study have to be addressed. Only 60.7% ( 91/150) of the initial 150 OSA patients at baseline had polysomnography at the 5-year follow-up visit which weakens the statistical power of the study. In OSA patients, the number was a little higher 63.1% (70/111). However, in Finland, bariatric surgery is centralized in larger hospitals, and patients coming to bariatric surgery may live hundreds of kilometers away from the hospital providing a challenge for additional visits. Instead of using in-laboratory polysomnography, portable recording devices (Embletta®) were used in the present study. However, clinical guidelines for the use of portable monitors have been introduced, and there is now a recommendation that portable monitoring may be used as an alternative to polysomnography for the diagnosis and treatment follow-up of OSA [34]. In addition, being a multicenter study, we aimed to standardize all methods used and choose them accordingly. Although these data are encouraging, they need to be replicated in a larger study.