TABLE IX.D.9.

Summary of evidence for durability of response to surgery

Study		Year	LOE	Study design		Study groups	Clinical end-point	Conclusion
He et al.812		2019	2a	SR and meta-analysis (two RCTs and nine cohort studies)		Adults who underwent UPPP or its modification for OSA and had short-term (n = 285) and/or long-term (n = 368) follow-up (at least 34 months) with objective sleep study results	AHI	1) Despite surgical efficacy decreasing over time (67.3% short-term and 44.35% long-term), UPPP surgeries are effective. 2) BMI, lowest O2 sat, and proportion of sleep time with O2 sat <90% were potentially predictive of long-term response.
Costantino et al.1890		2019	2a	SR and meta-analysis of cohort studies		Adult patients (n = 350) who underwent hypoglossal nerve stimulation (HNS) for treatment of moderate-to-severe OSA	AHI, ESS, adverse events	1) At 5-year follow up: surgical success rate was 75%, mean AHI reduction was 18.00, and ESS mean reduction was 5.27 (for Inspire HNS). 2) 6% of patients with serious device-related adverse events.
Janson et al.1958		1997	2b	Cohort study		Patients (n = 34) with OSA who underwent UPPP with (n = 25) or without (n = 9) long-term follow-up	Symptoms, AHI (responders with >50% AHI reduction and AHI ≤10)	1) 48% were responders at long-term (4–8 year) follow up. 2) Responders had lower preoperative AHI. 3) Some initially successfully treated patients (n = 4, 16%) relapsed in the long term.
Riley et al.1959		2000	2b	Cohort study		Patients (n = 40) with severe OSA who underwent phase II (skeletal and soft tissue) surgical reconstruction for sleep apnea	PSG variables (RDI, LSAT), QOL, complications	90% of patients showed persistent clinical success (improved quality of life assessments and polysomnography equivalent to CPAP patients; or postoperative RDI < 20 with at least 50% reduction and LSAT levels equivalent to nasal CPAP patients) at time of long-term follow-up (mean follow-up 50.7 ± 31.9 months).
Boot et al.1665		2000	2b	Cohort study		Patients (n = 58) with OSA who underwent UPPP with long-term follow up data (11–74 months, median 34 months)	Snoring, excessive daytime sleepiness, ODI	1) Response to UPPP for OSA decreases progressively over years after surgery. 2) UPPP in combination with tonsillectomy more effective than UPPP alone.
Lee et al.1964		2018	3b	Retrospective case–control study		1) Adults with OSA who underwent UPPP (n = 22,213) 2) Adults with OSA who did not undergo surgical intervention (n = 170,103) 3) Matched control group of adults without OSA (n = 961,590)	Newly diagnosed MI, CHF, and AF	UPPP, regardless of its effects on AHI, can significantly reduce risk of CHF (hazard ratio [HR]: 1.17 [1.10–1.24] without surgery to 0.76 [0.60–0.96]) and AF (HR: 1.39 [1.32–1.46] without surgery to 1.12 [0.94–1.32]) in patients with OSA for up to 8 years after surgery.
Walker-Engström et al.1507	2002		1a	Randomized controlled trial	Patients with mild-to-moderate OSA treated with oral appliance (OA, n = 45) vs. UPPP (n = 43)		AI, AHI (success defined as >50% reduction of AI/AHI), symptoms, compliance, need for complementary treatment, adverse effects	1) Success rate at 4-year follow-up was higher in OA group vs. UPPP (81% vs. 53%), but effectiveness of OA partly invalidated by its long-term compliance of 62%. 2) Success rate of UPPP decreased from 1 to 4 year follow-up: 70% to 53% (AI), and from 60% to 35% (AHI).

Abbreviations: ESS, Epworth Sleepiness Scale; lateral PP, lateral pharyngoplasty; LAUP, laser-assisted uvulopalatoplasty; LSAT, low oxyhemoglobin desaturation; OA, oral appliance; ODI, nocturnal oxygen desaturation index; TA, tongue advancement (mandibular osteotomy); TCRTA, temperature-controlled radiofrequency tissue ablation; TS, tongue suspension; RDI, respiratory disturbance index; UPPP, uvulopalatopharyngoplasty.