Additive Effect of Multilevel Radiofrequency Ablation in Elderly Patients with Sleep Apnea

Abstract

Elderly patients with Obstructive sleep apnea (OSA) often struggle with positive airway therapy due to low adherence. This study explores Radiofrequency ablation (RFA) as a treatment alongside weight loss and tongue-retaining device for OSA in older adults. Through a randomized trial, we compare RFA effectiveness to a control group, aiming to establish its potential role in managing OSA in this vulnerable population. The control group received only weight loss and tongue-retaining device recommendations. In the intervention group, these measures were supplemented with RFA of the soft palate and inferior turbinates. After three months, both groups were assessed for changes in sleep study parameters, BMI, and tongue-retaining device adherence. Twenty patients in the control group and 23 in the intervention group completed the trial. The average age was 60.25 years old (SD = 5.88) in the control group and 61.83 years old (SD = 5.21) in the RFA group. Both groups experienced significant BMI reductions, with the control group averaging a decrease of 1.5 kg/m² (SD = 2.4 kg/m²) and the RFA group averaging a decrease of 1.3 kg/m² (SD = 1.1 kg/m²). Tongue-retaining device adherence was 50% in the control group and 56.5% in the RFA group. The RFA group achieved a significantly greater proportionate reduction in AHI compared to the control group (p = 0.04, effect size=-0.38). This randomized controlled trial showed that RFA on the soft palate and inferior turbinate is a well-tolerated and potentially effective treatment option for elderly patients with OSA. Further research with larger samples and longer follow-ups is needed to confirm these findings and assess long-term durability.

Introduction

Obstructive sleep apnea (OSA) is a sleep disorder characterized by recurrent pauses in breathing during sleep. It is associated with various clinical consequences, including daytime hypersomnolence, neurocognitive dysfunction, cardiovascular disease, metabolic dysfunction, and pulmonary hypertension [1]. Obstructive sleep apnea (OSA) is highly prevalent in the elderly population, affecting at least 20% of individuals over the age of 65 [2]. The pathophysiology of sleep-disordered breathing in older patients differs from that in younger populations due to age-related factors such as loss of neuromuscular tone and tissue laxity [3].

A range of treatment options are available for obstructive sleep apnea (OSA), including continuous positive airway pressure (CPAP) therapy, oral appliances, surgery, and conservative approaches such as weight loss and positional therapy [4]. CPAP is considered the “gold standard” treatment, but its side effects and low adherence rates make it less tolerable for some patients [5].

Invasive sleep surgery for older adults with obstructive sleep apnea carries potential complications that warrant careful consideration. While generally effective, these procedures involve risks due to the delicate nature of the structures involved and the increased vulnerability of elderly patients [6]. Common complications include bleeding, infection, respiratory complications, cardiovascular issues, neurological complications, and wound healing issues [7].

Radiofrequency ablation (RFA) has been investigated as an alternative therapy for patients with obstructive sleep apnea (OSA) [8]. RFA boasts a significantly reduced morbidity profile. It has been shown that compared to invasive surgeries like uvulopalatopharyngoplasty (UPPP). RFA is associated with fewer incisions, minimal tissue disruption, and consequently, less bleeding and postoperative pain. This translates to faster recovery times, and potentially lower risk of infection [9].

RFA can be effectively performed under local anesthesia at the soft palate and inferior turbinate sites, eliminating the need for general anesthesia and its associated risks in the elderly population [10, 11]. This is particularly beneficial for patients with co-morbidities or cardiovascular concerns who may not be suitable candidates for general anesthesia [12].

This study aims to rigorously assess the efficacy of multilevel radiofrequency ablation RFA as a treatment modality for obstructive sleep apnea in elderly individuals in addition to tongue-retaining device and weight loss. Utilizing a randomized clinical trial design, we objectively measure and compare the effectiveness of RFA against a control group, providing evidence for its potential role in managing OSA within this vulnerable population.

Method

Study Design and Population

This single-center, randomized controlled trial (RCT) enrolled 50 participants diagnosed with OSA between January and November 2023. Patients aged 55 years or older were included. The study was registered at the Iranian Registry of Clinical Trials (IRCT: IRCT20190602043791N4) and received ethical approval from the Tehran University of Medical Sciences ethics committee (IR.TUMS.MEDICINE.REC.1401.356). All participants provided informed consent, and the study adhered to the Declaration of Helsinki principles.

All patients underwent a pre-operative type 1 sleep study, with AHI, mean oxygen saturation, time spent below 90% oxygen saturation (T90%), and lowest oxygen saturation recorded. On the same night, a type 4 sleep study was conducted using the Berry-BM2000A wearable device, and this device was used for measuring and comparing follow-up sleep parameters. Patients were included if the type 1 polysomnography confirmed OSA and AHI > 5 measured by the wearable sleep study, and were unable to use positive airway pressure (PAP) therapy. Exclusion criteria included Body Mass Index (BMI) > 40, previous OSA surgery or head and neck surgery, and uncontrolled metabolic, cardiovascular, or respiratory disorders. Age, sex, BMI, Friedman tongue position, and tonsil grade were documented at admission.

Soft Palate RFA

Each patient in the RFA group underwent one session for the soft palate RFA. Following topical application of 10% lidocaine spray and bilateral injection of 1 mL of 2% lidocaine to each soft palatal side, an additional 1 mL injection was administered to each side five minutes later. Immediately, the curved soft palate handpiece (AvanTeb, Smart Relax Sleep handpiece) was introduced, and radiofrequency ablation was performed at 12 watts with four applications per side (eight shoots for each patient), entering the soft palate at two separate entry points on each side. All patients were monitored for 30 min post-procedure and discharged with amoxicillin-clavulanate 625 mg three times daily for five days. Ibuprofen was provided for pain as needed. Patients were instructed to maintain a semi-sitting position for one night, consume soft and cold foods for one day, and avoid aspirin for one day. They were advised to seek immediate medical attention in case of any breathing difficulties.

Inferior Turbinate Reduction

One week following soft palate RFA, the inferior turbinate reduction was performed in one session. One week following soft palate RFA, the inferior turbinate reduction was performed. Two drops each of tetracaine and phenylephrine were applied to both nostrils, followed by the insertion of a ribbon gauze soaked in equal parts tetracaine and phenylephrine into each nasal cavity for 5 min. Bilateral 1 mL injections of 2% lidocaine were administered to each inferior turbinate, repeated after 5 min. Using a straight inferior turbinate handpiece (AvanTeb, Smart Relax Breath handpiece), each turbinate was penetrated 4 cm and ablated at 5 mm intervals, progressing posteriorly to the anterior tip and power set at 15 watts. This ablation was repeated three additional times for the anterior 1 cm. Patients were instructed to use saline nasal irrigation and mupirocin nasal ointment for one week, with follow-up in-office necrotic crust debridement scheduled.

Outcome Measures

The Sleep study parameters (AHI, Mean oxygen saturation, Lowest oxygen desaturation, T90% measured by Berry-BM2000A wearable device), BMI, Tongue retaining device usage, Epworth Sleepiness Scale (ESS), and Stanford Subjective Snoring Scale (SSSS) were measured at baseline and 3 months after the intervention in both the intervention and control groups.

Stanford Subjective Snoring Scale (SSSS)

The Stanford Subjective Snoring Scale (SSSS) is a self-administered questionnaire used to assess the impact of snoring on the patient’s sleep partner and overall sleep environment. It assigns a score from 0 to 10 based on the severity and impact of snoring, ranging from “no snoring” (0) to “partner leaves the room due to snoring” (10) [13].

Epworth Sleepiness Scale (ESS)

The Epworth Sleepiness Scale (ESS) is a validated, self-administered questionnaire used to quantify daytime sleepiness in patients. Developed in 1991, the ESS asks participants to rate their likelihood of dozing in eight different everyday situations on a scale of 0 (“would never doze”) to 3 (“high chance of dozing”). The total score ranges from 0 to 24, with higher scores indicating greater daytime sleepiness [14].

Statistics Analysis

All data analyses were conducted using JASP software version 0.18.01. The proportion of change was calculated by dividing the difference between final and baseline values by the baseline value. Chi-square tests were used for categorical variables. Normality of the numeric data assed by Shapiro-Wilk. Normally distributed paired data were analyzed using paired t-tests, while non-normally distributed data were analyzed using Wilcoxon signed-rank tests. Independent samples with normally distributed data were analyzed using independent t-test and non-normally distributed data were analyzed using Mann-Whitney U test. A significance level of p < 0.05 was used for all analyses.

Results

The study enrolled a total of 43 patients, with 23 patients randomized to the RFA group and 20 patients to the control group. The second test was not completed by two participants in the intervention group and five participants in the control group. Two participants in the intervention group did not experience any complications. The RFA group had a higher proportion of males (78.2%) compared to the control group (55.0%), but this difference did not reach statistical significance (p = 0.104). In the control group, the mean age was 60.25, with a median age of 57.5 years (SD = 5.88, IQR = 6.25). The RFA group had a mean age of 61.83 years, and a median age of 61 years (SD = 5.21, IQR = 5.5). The two groups were comparable with a p-value of 0.141.

Comorbidities

Smoking prevalence was slightly higher in the RFA group (17.4%) compared to the control group (5.0%), but this difference was also not statistically significant (p = 0.206). The RFA group had slightly more patients with hypertension (7 patients vs. 5 patients, p = 0.723) and ischemic heart disease (5 patients vs. 2 patients, p = 0.298), while the control group had slightly more patients with diabetes (6 patients vs. 5 patients, p = 0.489). Additionally, two patients in the RFA group had a history of coronary artery bypass grafting (CABG), one had asthma, and one had Chronic Obstructive Pulmonary Disease (COPD).

Baseline BMI

The control group had a mean BMI of 30.03, median of 29.4, SD of 4.82, and IQR of 4.38. The RFA group had a mean BMI of 29.14, a median of 28.7, an SD of 3.01, and an IQR of 3.1. These differences were statistically insignificant (p = 0.626).

Baseline Polysomnograhy Evaluation

The control group had a mean AHI of 43.73 (SD = 24.87), while the RFA group had a mean AHI of 43.11 (SD = 20.53), p value = 0.93. The detailed polysomnography information is presented in Table 1.

Table 1.

Summary statistics of polysomnografic measures before interventions

	AHI		Lowest O2%		Mean O2%		T90%
	RFA	Control	RFA	Control	RFA	Control	RF	Control
Median	41.2	39.3	76	74.5	92	92	8.9	8.15
Mean	43.11	43.72	74.913	72.4	91.56	91.75	21.11	18.92
Std. Deviation	20.54	24.87	11.361	15.19	2.59	1.83	27.77	22.53
IQR	23.25	31.28	10.5	13.25	1.5	1.25	17.35	20.65
P value	0.93		0.71		0.4		1

AHI: Apnea-Hypopnea index, T90%: Time spent below 90% oxygen saturation, SD: standard deviation, IQR: interquartile range, RFA: Radiofrequency ablation

Baseline Wearable Sleep Study

The control group exhibited a mean AHI of 19.4 (SD = 13.03), while the RFA group had a mean AHI of 25.43 (SD = 21.04), p value = 0.52. Further details on the baseline wearable sleep study parameters are presented in Table 2.

Table 2.

Summary statistics of wearable sleep study measures before interventions

	Initial AHI		Lowest O2%		Mean O2%		T90%
	RFA	Control	RFA	Control	RFA	Control	RFA	Control
Median	17.41	16.86	77	77	92.5	93.4	10.6	4.2
Mean	25.43	19.4	75.9	76.08	91.62	92.73	22.53	16.9
SD	21.036	13.037	6.959	8.875	3.2	2.71	30.51	24.92
IQR	21.98	14.5	9.5	17.75	2.15	4.32	19.85	13.82
P value	0.52		0.94		0.27		0.25

AHI: Apnea-Hypopnea index, T90%: Time spent below 90% oxygen saturation, SD: standard deviation, IQR: interquartile range, RFA: Radiofrequency ablation

Friedman Tonsil Grade

In the control group, the average tonsil grade was 1.7 (SD = 1.17) with a median of 1 (IQR = 2). The RFA group had an average tonsil grade of 1.3 (SD = 0.56) and a median of 1 (IQR = 1). Statistically, there was no significant difference between the groups (p = 0.58).

Friedman Tongue Position Grade

The control group had an average tongue position grade of 2.7 (SD = 0.64) and a median of 3 (IQR = 0.625). The RFA group had an average tongue position grade of 2.83 (SD = 0.67) and a median of 3 (IQR = 0.5). The difference between groups was not statistically significant (p = 0.89).

Baseline Snoring Severity

Both groups had a mean and median score of 7.1 and 8.0 in SSSS. While the RFA group displayed a slightly lower standard deviation (2.6 compared to 2.8 in the control group), the distributions were similar, as evidenced by the IQR of 5, p value = 1.0.

Baseline Sleepiness

The control group had a mean ESS of 9.8 (SD = 6.4), with a median of 9.0 and IQR of 10.75. The RFA group had a slightly higher mean ESS of 11.6 (SD = 5.7), a median of 11.5, and an IQR of 7.0, p value = 0.97.

BMI Changes

Both groups experienced significant reductions in BMI.

• Control group: average BMI decrease of 1.5 kg/m² (standard deviation 2.4 kg/m², p < 0.01).

• RFA group: average BMI decrease of 1.3 kg/m² (standard deviation 1.1 kg/m², p < 0.01).

The final BMI and proportional BMI change were not statistically different between the control and RFA groups (p value > 0.05).

Tongue-Retaining Device Usage

The prevalence of tongue-retaining device use was similar between the control and RFA groups. Roughly half of the participants in both groups (50% in control, 56.5% in RFA) reported using this device. This difference was not statistically significant (p = 0.669).

Wearable Sleep Study Follow-up

AHI Changes

• Control group: Average AHI decrease of 5.5 events/hour (standard deviation 7.8, effect size 0.87, p < 0.01).

• RFA group: Average AHI decrease of 14.63 events/hour (standard deviation 17.5, effect size 0.99, p < 0.01).

However, the final AHI values after treatment were not statistically different between the groups (p = 0.304). While the RFA group achieved a larger AHI proportionate change from baseline, this difference was statistically significant (p = 0.04, effect size=-0.38). Refer to Table 3 for detailed AHI data.

Table 3.

Summary statistics of wearable sleep study measures after interventions

	Final AHI		AHI change		AHI proportionate change
	RFA	Control	RFA	Control	RFA	Control
Median	6.65	10.86	-8.66	-4.66	-0.52	-0.36
Mean	11.66	14.15	-14.63	-5.50	-0.55	-0.29
SD	10.89	10.56	17.48	7.82	0.33	0.44
IQR	16.88	17.92	11.86	5.15	0.54	0.44
Effect size	-0.19		-0.31		-0.38
P value	0.3		0.11		0.04
	Final Lowest O2%		Lowest O2% change		Lowest O2% proportionate change
	RFA	Control	RFA	Control	RFA	Control
Median	84	81.5	4.0	0.5	0.05	0.007
Mean	80.3	79.35	4.10	3.14	0.06	0.048
SD	7.71	7.29	7.48	7.84	0.11	0.109
IQR	12.5	6.75	8.5	11.75	0.12	0.158
Effect size	0.13		0.12		0.1
P value	0.68		0.7		0.77
	Final Mean O2		Mean O2 change		Mean O2 proportionate change
	RFA	Control	RFA	Control	RFA	Control
Median	93.4	92.3	1.1	-0.05	0.012	-0.0005
Mean	92.81	92.34	1.16	-0.38	0.013	-0.004
SD	2.5	2.08	1.48	2.19	0.016	0.023
IQR	2.6	1.88	1.82	1.48	0.021	0.016
Effect size	0.2		0.49		0.49
P value	0.51		0.01		0.01
	Final T90%		T90% change		T90% proportionate change
	RFA	Control	RFA	Control	RFA	Control
Median	1.40	6.35	-2.50	-0.45	-0.63	-0.11
Mean	15.37	13.51	-5.72	-3.32	0.67	0.81
SD	26.14	19.98	28.84	18.03	5.04	2.38
IQR	17.85	10.48	9.75	4.25	0.79	1.24
Effect size	-0.17		0.3		0.44
P value	0.34		0.13		0.02

AHI: Apnea-Hypopnea index, T90%: Time spent below 90% oxygen saturation, SD: standard deviation, IQR: interquartile range, RFA: Radiofrequency ablation

Mean O2 Saturation Changes

• Control group: Mean O2 saturation decrease of 0.38% (standard deviation 2.2, p = 0.82).

• RFA group: Mean O2 saturation increase of 1.2% (standard deviation 1.5, effect size 0.78, p = 0.003).

The final O2 saturation values after treatment were not statistically different between the groups (p = 0.51). While the RFA group experienced a larger O2 saturation proportionate change (p = 0.011). Refer to Table 3 for detailed O2 saturation data.

Lowest Oxygen Desaturation Changes

• Control group: Average increase in lowest oxygen desaturation of 3.1% (standard deviation 7.8, effect size 0.49, p = 0.006).

• RFA group: Average increase in lowest oxygen desaturation of 4.1% (standard deviation 7.5, effect size 0.55, p = 0.028).

While the RFA group showed a slightly larger increase than the control group, this difference was not statistically significant (p = 0.13). Additionally, the proportionate change in lowest oxygen desaturation did not differ significantly between the groups (p = 0.77). Further details on the changes in lowest oxygen desaturation can be found in Table 3.

T90% Changes

• Control group: Average decrease in T90% of 3.3% (standard deviation 18.0, p = 0.64).

• RFA group: Average decrease in T90% of 5.7% (standard deviation 28.8, effect size 0.57, p = 0.03).

However, the absolute final T90% values after treatment were not statistically different between the groups (p = 0.34). The RFA group experienced a larger increase in proportionate change of T90%, (p = 0.02). Refer to Table 3 for detailed T90% data.

SSSS Changes

• Control group: Average decrease in SSSS of 1.4 points (standard deviation 2.1, effect size 0.74, p = 0.006).

• RFA group: Average decrease in SSSS of 2.78 points (standard deviation 2.0, effect size 0.95, p < 0.001).

However, the final SSSS scores after treatment were not statistically different between the groups (p = 0.07). The proportionate change in SSSS was significantly higher in the RFA group compared to the control group (p = 0.02). For further details on the changes in SSSS, refer to Table 4.

Table 4.

Summary statistics of snoring and sleepines measures after interventions

	Final SSSS		SSSS change		SSSS proportionate change
	RFA	Control	RFA	Control	RFA	Control
Median	3	6	-3	-1	-0.40	-0.20
Mean	4.35	5.70	-2.78	-1.40	-0.35	-0.20
Std. Deviation	2.72	2.56	2.94	2.01	0.50	0.31
IQR	3.50	3.25	3	1.25	0.41	0.23
Effect size	0.32		0.4		0.42
P value	0.07		0.025		0.02
	Final ESS		ESS change		ESS proportionate change
	RFA	Control	RFA	Control	RFA	Control
Median	9	6	2.5	2	0.24	0.25
Mean	9.22	7.45	2.68	2.4	0.24	0.25
Std. Deviation	4.90	5.71	2.06	3.00	0.30	0.36
IQR	5.5	4.5	1	2.25	0.20	0.43
Effect size	0.26		0.18		0.02
P value	0.14		0.32		0.93

SSSS: Stanford Subjective Snoring Scale, ESS: Epworth Sleepiness Scale. SD: standard deviation, IQR: interquartile range, RFA: Radiofrequency ablation

ESS Changes

• Control group: Average decrease in ESS of 2.4 points (standard deviation 3.0, effect size 0.84, p = 0.002), indicating a significant improvement.

• RFA group: Average decrease in ESS of 2.06 points (standard deviation 2.0, effect size 0.89, p < 0.001).

The final ESS scores after treatment were not statistically different between the groups (p = 0.14). The proportional change in ESS did not differ significantly between the control and RFA groups (p = 0.93). For further details on the changes in ESS, refer to Table 4.

Complications

One patient with severe uvula edema responded well to conservative management with 8 mg of intravenous dexamethasone. Uvula edema developed approximately eight hours after soft palate treatment and subsided following a two-hour course of corticosteroid injection.

Discussion

This randomized controlled trial demonstrates the potential of radiofrequency ablation (RFA) on the soft palate and inferior turbinate as a well-tolerated treatment option for obstructive sleep apnea (OSA) in elderly patients intolerant to continuous positive airway pressure (CPAP).

Our findings offer some key insights. The addition of RFA to standard weight management and tongue retaining devices significantly improved several OSA parameters, including AHI proportionate change from baseline, absolute and proportionate mean oxygen saturation change during sleep, proportionate change of time spent below 90% oxygen saturation from baseline, Absolute and proportionate snoring severity change. Our study is consistent with other studies [15].

RFA can improve obstructive sleep apnea (OSA) through several mechanisms. RFA delivers heat energy, causing controlled thermal injury and subsequent collagen deposition, leading to stiffening of the soft palate tissue. This reduces its collapsibility during sleep, preventing airway obstruction. RFA can also cause some tissue ablation, resulting in a slight decrease in the size of the soft palate and inferior turbinate creating more space for the airway [16].

While several studies have examined the individual efficacy of RFA, weight management, and tongue-retaining devices for OSA, our research is the first to assess their combined impact. Combining RFA’s targeted tissue modification with weight loss and tongue-retaining device mechanical support could lead to a synergistic effect, potentially achieving greater improvement in OSA severity compared to individual interventions alone. Tongue retaining devices address tongue base collapse [17], weight loss decreases pharyngeal lateral wall and tongue volume [18], and RFA improves soft palate collapsibility and nasal airflow.

While tongue base RFA can be performed under local anesthesia [8, 19], we opted to spare this site. The sensitive nature of the tongue base and the potential for discomfort due to the gag reflex make this procedure uncomfortable. Achieving adequate pain control might have required large volumes of lidocaine, which can have its own set of side effects. Tongue base RFA can cause tongue abscess formation, a potential complication in this specific patient population [20].

This study observed a greater effect of RFA on mean oxygen saturation and time spent below 90% compared to AHI. This suggests that RFA may be more effective in reducing the severity of oxygen desaturations during respiratory events even, suggesting a larger impact on oxygen deprivation during events compared to event frequency. The number of respiratory events alone does not provide a comprehensive assessment of OSA severity and decreasing desaturation may have a positive effect on cardiac consequences [21].

The relatively small sample size (23 RFA vs. 20 control) necessitates further studies with larger cohorts to confirm the findings. Long-term follow-up is needed to assess the durability of the treatment effect and potential late complications.

Conclusion

This randomized controlled trial shows that RFA on the soft palate and inferior turbinate is a well-tolerated and potentially effective treatment option for elderly patients with OSA who cannot tolerate CPAP. Our findings indicate that RFA, combined with weight management and tongue-retaining devices, significantly improved various OSA parameters, including AHI, sleep oxygen desaturation, and snoring severity. This study, which is the first to investigate the combined effects of these interventions in the elderly population, suggests a potential synergistic approach for managing OSA. Further research with larger samples and longer follow-ups is needed to confirm our findings and assess long-term durability.

Author Contributions

Conceptualization and Design: Reza Erfnian, Akbar Jafari, and Reihaneh Heidari. Data analysis and Manuscript drafting: Reza Erfanian. and Akbar Jafari. Manuscript revision: Mitra Khorsandi.

Funding

Tehran University of Medical Sciences supported this study.

Data Availability

Upon reasonable request, the data can be made available.

Declarations

Ethical Approval

This study adhered to the Declaration of Helsinki principles. Ethical approval was granted by the Tehran University of Medical Sciences ethics committee (IR.TUMS.MEDICINE.REC.1401.356). The study was registered at the Iranian clinical registry system (IRCT20190602043791N4).

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.