Transoral Robotic Surgery versus Plasma Ablation for Tongue Base Reduction in Obstructive Sleep Apnea: Meta-analysis

Abstract

Objectives

To compare the effectiveness of transoral robotic surgery (TORS) versus plasma ablation (PA) in tongue base reduction surgery for obstructive sleep apnea (OSA).

Data Sources

PubMed, Scopus, Cochrane Library, OVID.

Review Methods

Keywords searched included OSA, tongue base surgery, TORS, and coblation. Outcomes included pre- to postoperative apnea-hypopnea index, Epworth Sleepiness Scale (ESS), and lowest oxygen saturation. Additional outcomes included surgical success rate, postoperative bleeding, operative time, and length of stay.

Results

A total of 690 unique articles were identified, of which 60 underwent full-text review. Twenty-six articles were included in final analysis, comprising 18 studies on TORS (834 patients) and 11 studies on PA (294 patients). Mean differences of apnea-hypopnea index, ESS, and lowest oxygen saturation for TORS were –23.92, –7.6, and 5.83% (all P < .01). Corresponding values for PA were –22.07, –4.14, and 5.48% (all P < .00001). TORS had greater ESS reduction than PA (P = .02). Follow-up duration was shorter in TORS than PA (mean ± SD: 4.2 ± 2.6 vs 4.6 ± 1.4 months, P = .0482). Surgical success rates in TORS and PA were similar (57.6% vs 60.3%, P = .4474). Postoperative bleeding occurred less frequently in TORS versus PA (3.3% vs 7.5%, P = .0103). Operative time was longer for TORS than PA (77.9 ± 16.4 minutes vs 44.0 ± 12.9 minutes, P < .0001). Length of stay was similar between TORS and PA (3.9 ± 1.6 days vs 3.9 ± 2.5 days, P = .9047).

Conclusion

Tongue base reduction with TORS or PA each effectively treats OSA and provides comparable results. The choice between techniques might depend on patient factors, availability of technology, and associated costs.

Keywords

transoral robotic surgery coblation plasma ablation base of tongue lingual tonsillectomy glossectomy obstructive sleep apnea sleep disordered breathing

The prevalence of obstructive sleep apnea (OSA) has increased in the setting of a growing incidence of obesity worldwide.^1,2 Surgical treatments for OSA include uvulopalatopharyngoplasty (UPPP), maxillary-mandibular advancement, genioglossus advancement, and hyoid advancement. Tracheostomy, hypoglossal nerve stimulation, and tongue base reduction (TBR) are also surgical options for select patients with OSA. Although surgical management typically involves a multilevel approach, the base of tongue represents an anatomic target of interest, as oropharyngeal and retroglossal airway obstructions are commonly identified in patients with OSA.³ Since its description by Woodson and Laohasiriwong,⁴ TBR with plasma ablation technology has offered a viable technique in tongue base surgery. However, the introduction of transoral robotic surgery (TORS) to OSA surgery provided surgeons superior visualization and tissue manipulation in the targeted area.⁵ Due to unique advantages of each technique, comparing the outcomes of TORS and plasma ablation in TBR for OSA is a topic of great interest.

Until now, only a few institutional studies have directly compared TBR with TORS versus plasma ablation.^6-8 Prior systematic reviews have either examined TORS alone^9,10 or broadly summarized differences between techniques.¹¹ In the present study, we aim to conduct a meta-analysis of both surgical approaches, comparing differences in operative factors, outcomes, and complications.

Methods

Search Strategy

This study was conducted according to the PRISMA guidelines.¹² To identify studies for inclusion in this review, a detailed query of the following 4 databases was performed: PubMed (National Library of Medicine, National Institutes of Health), Scopus (Elsevier), Cochrane Library (Wiley), and OVID (Wolters Kluwer). The query used subject headings (eg, MeSH in PubMed [Medical Subject Headings]) and keywords for the following concepts: obstructive sleep apnea, tongue base surgery, lingual tonsillectomy, glossectomy, transoral robotic surgery, TORS, robotic surgical procedures, and coblation. The PubMed search strategy was modified for the other 3 databases, with similar keywords. The databases were searched from inception through July 23, 2019. To identify additional articles, the reference lists of relevant articles were hand-searched, as well as citing articles. References were uploaded to EndNote (Clarivate Analytics, Philadelphia, Pennsylvania) and screened for relevance.

Selection Criteria

Only studies reporting on apnea-hypopnea index (AHI) with means and standard deviations (or with derivable individual patient data) were included. Inclusion criteria were (1) double- or single-blinded randomized controlled trials, (2) double- or single-blinded randomized comparison trials, (3) prospective or retrospective observational studies, and (4) case series with >5 eligible patients. Abstracts were first independently assessed by 2 reviewers (J.A.L. and Y.J.B.) to identify all articles satisfying the inclusion criteria. Non-English studies, nonhuman studies, case reports, and review articles were excluded. Finally, articles duplicating patient data from another study were represented once, with the largest study included for review. Articles were critically appraised to assess level of evidence by using criteria of the Oxford Center for Evidence-Based Medicine. The risk of bias was assessed according to the Cochrane Handbook for Systematic Reviews of Interventions. The latest version of this tool was updated in March 2011 (version 5.1.0).¹³ Two authors (Y.J.B. and J.A.L.) performed a pilot assessment on 3 studies to check for consistency of assessment. Both then performed independent risk assessment on the remaining studies. All disagreements were resolved by way of discussion with a third author (S.A.N.). Risk-of-bias items included the following: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other bias. The risk of bias for each aspect is graded “low,”“unclear,” or “high.”

Data extracted from studies included author, publication year, study design, study characteristics, and patient demographics. Patient data included preoperative body mass index (BMI), study selection criteria, procedure, and operative time. Outcomes included pre- and postoperative AHI, Epworth Sleepiness Scale (ESS), nadir oxygen saturation on polysomnogram (lowest oxygen saturation [LSAT]), as well as surgical complications. Additional outcome measures were extracted but not considered for analysis.

Statistical Analysis

The meta-analysis utilized pre- to posttreatment measures, with all patients serving as their own controls. Meta-analysis of selected studies with a continuous measure (comparison of means and standard deviations between pre- and posttreatment groups) was performed with Cochrane Review Manager (RevMan, version 5.3; Nordic Cochrane Center, Cochrane Collaboration, Copenhagen, Denmark). Both the fixed effects model and the random effects model were used. Under the fixed effects model, the assumption is that all studies come from a common population and that the effect size (standardized mean difference) is not significantly different among the studies; this assumption is tested by the “heterogeneity test.” If this test yielded a low probability value (P < .05), then the fixed effects model may be invalid. In this case, the random effects model, in which the random variation within the studies and the variation among the studies are incorporated, may be more appropriate. Under the random effects model, the true effects in the studies are assumed to vary among studies, and the summary effect is the weighted average of the effects reported in the different studies.¹⁴ The random effects model provides a more conservative estimate (ie, with a wider confidence interval [CI]), but the results from the 2 models usually agree when there is no heterogeneity. When heterogeneity was present, the random effects model was preferred. In this study, the null hypothesis was that there was no difference between pre- and posttreatment with respect to AHI, ESS, and LSAT. Data are presented as mean ± SD (95% CI) in this text and as mean difference in the figure.

In addition, a meta-analysis of proportions was performed with MedCalc 18.10.2 (MedCalc Software bvba, Ostend, Belgium). Primary outcomes included surgical success, defined as a postoperative reduction in AHI >50% and AHI <20, as well as postoperative bleeding. Each technique was weighted according to the number of patients affected. The program MedCalc lists the proportions (expressed as a percentage) with their 95% CIs as found in the individual studies included in the meta-analysis. The weighted summary proportion is calculated by the Freeman-Tukey transformation.¹⁵ This pooled proportion is reported with 95% CI for the fixed effects model and the random effects model.¹⁶ Both the fixed effects model and the random effects model were used in this study. If there is high heterogeneity (I² > 50%), then a random effects model is used; if low heterogeneity, then a fixed effects model is allowable. Using random effect modeling is more conservative; thus, it is preferable to assume random effects modeling unless I² is small. A P value <.05 was considered to indicate a statistically significant difference for all statistical tests.

Finally, the Sterne and Egger tests were performed for further assessment of risk of publication bias.^17,18 Potential publication bias was evaluated by visual inspection of the funnel plot and Egger’s regression test, which statistically examines the asymmetry of the funnel plot. In a funnel plot, treatment effect is plotted on the horizontal axis and MedCalc plots the standard error on the vertical axis.¹⁹ The vertical line represents the summary estimate derived with fixed effect meta-analysis. Two diagonal lines represent (pseudo) 95% confidence limits (effect ± 1.96 SE) around the summary effect for each standard error on the vertical axis. These show the expected distribution of studies in the absence of heterogeneity or selection bias. In the absence of heterogeneity, 95% of the studies should lie within the funnel defined by these diagonal lines. Publication bias results in asymmetry of the funnel plot.

Results

Search Results

The literature search identified 690 unique articles after removal of duplicates. Initial screening by title and abstract eliminated 630 articles, leaving 60 for full-text review. Among these, 26 articles were included in the final analysis. A PRISMA diagram outlining the summary of the search process is shown in Figure 1 .

Figure 1.

PRISMA flow diagram of search results. TORS, transoral robotic surgery.

Overall Study Characteristics

A summary of study characteristics is detailed in Tables 1 and 2 . Articles were published between 2009 to 2019. Five studies were prospective studies,^20-24 while the remainder were retrospective. Two studies represented level 2 evidence, while the remaining studies were level 3 and 4 evidence according to the 2011 criteria of the Oxford Center for Evidence-Based Medicine.²⁵ The risk of bias was assessed for each study ( Figure 2 ). All studies were considered to have low risk of bias. A funnel plot depicting study bias is presented in Figure 3 .

Table 1.

Study Characteristics: Transoral Robotic Surgery.

		No.		Mean ± SD
Study (Year)	OLE	Patients	Male	Age, y	BMI	Selection Criteria	BoT Procedure	Associated Procedure
Friedman⁷ (2012)	3	27	24	43.8 ± 9.2	32.3 ± 3.3	AHI >15	TBR	ZPP
Lee²⁰ (2012)	4	20	14	45.7 ± 10.6	32.6 ± 7.4	OSA with TBO (LTH)	LT	UPPP
Vicini⁵ (2012)	4	20	18	50.2 ± 10.9	27.9 ± 3.3	AHI >20, ESS >11, CPAP intolerance, BoT hypertrophy, adequate BoT exposure	TBR	SGP, UPPP, SNS
Toh²⁶ (2014)	4	20	16	47.1 ± 11.4	26.5 ± 2.9	CPAP intolerance	TBR, LT	TL with uvulopalatal flap, partial ET
Vicini²⁷ (2014)	4	243	197	50.0 ± 12.0	28.5 ± 3.9	OSA with TBO (LTH)	TBR	SGP, TL, UPPP, ZPP, SNS
Vicini²⁸ (2014): 1	3	12	12	49.6 ± 11.3	28.0 ± 2.9	OSA	TBR, SGP	UPPP
Vicini²⁸ (2014): 2	3	12	9	54.2 ± 10.8	27.3 ± 2.0	OSA	TBR, SGP	ESP
Chiffer²¹ (2015)	4	18	16	46.9 ± 9.3	34.2 ± 6.6	AHI >5; Exclusion: active infection, pregnancy, previous head/neck surgery precluding TORS, c/i medical condition	B/L PHG	Limited pharyngectomy + UPPP
Lin²⁹ (2015)	4	39	24	46.5 ± 13.2	32.9 ± 7.0	OSA, including prior surgery (21), tracheostomy (4)	TBR	None (11), UPPP (2), ET (7), UPPP + ET (19)
Muderris³⁰ (2015)	4	6	0	48.7 ± 4.8	27.3 ± 2.2	OSA with TBO (LTH)	TBR	EP (1)
Arora²² (2016)	4	14	13	54.3 ± 14.6	28.7 ± 2.8	AHI >15, BMI <35, TBO, prior OSA treatment intolerance	TBR	Wedge EP (10)
Kayhan³¹ (2016)	4	25	19	50.1 ± 8.5	30.7 ± 5.5	OSA with TBO	TBR	EP (24), UP (22), anterior PaP (22), PhP (18), TL (18)
Folk⁶ (2017)	3	45	33	48.2 ± 11.6	32.3 ± 4.8	OSA with TBO, includes prior surgery	TBR	UPPP, TL, SNS
Meraj³² (2017)	4	101	78	54.7 ± 12.2	28.5 ± 4.0	BMI <40, AHI ≥15. Exclusion: prior TORS LT, no DISE or postoperative PSG	LT	PMG, EP
Aynaci³³ (2018)	3	20	16	45.0 ± 7.1	NR	OSA, CPAP intolerance	TBR	TL, anterior PaP, uvulectomy
Cammaroto³⁴ (2018)	4	51	38	55.5 ± 11.6	29.1 ± 8.9	AHI ≥20, CPAP intolerance. Exclusion: prior OSA surgery	TBR	Tracheostomy, TL, ESP, SNS
Hwang⁸ (2018)	4	16	14	45.1 ± 13.4	25.8 ± 3.4	OSA related to BoT. Exclusion: prior OSA surgery	TBR	Lateral PaP, TL
Lai³⁵ (2018)	4	121	89	43.4 ± NR	23.0 ± 1.7	OSA with BMI <25	TBR	Palate surgery
Huntley³⁶ (2019)	3	24	20	46.4 ± 14.0	29.6 ± 3.8	Moderate to severe OSA with TBO, CPAP intolerance, included prior OSA surgery	TBR	None

Abbreviations: AHI, apnea-hypopnea index; B/L, bilateral; BMI, body mass index; BoT, base of tongue; c/i, contraindicated; CPAP, continuous positive airway pressure; DISE, drug-induced sleep endoscopy; EP, epiglottoplasty; ESS, Epworth Sleepiness Scale; ESP, expansion sphincter palatoplasty; ET, epiglottidectomy; LT, lingual tonsillectomy; LTH, lingual tonsil hypertrophy; NR, not reportable; OLE, Oxford level of evidence; OSA, obstructive sleep apnea; PaP, palatoplasty; PHG, posterior hemiglossectomy; PhP, pharyngoplasty; PMG, partial midline glossectomy; PSG, polysomnography; SGP, supraglottoplasty; SNS, sinonasal surgery; TBO, tongue base obstruction; TBR, tongue base reduction/resection; TL, tonsillectomy; TORS, transoral robotic surgery; UPPP, uvulopalatoplasty; UP, uvuloplasty; ZPP, Z-palatoplasty.

Table 2.

Study Characteristics: Plasma Ablation.

		No.		Mean ± SD
Study (Year)	OLE	Patients	Male	Age, y	BMI	Selection Criteria	BoT Procedure	Associated Procedure
Babademez³⁷ (2010)	4	16	16	41.3 ± 10.5	29.6 ± 2.5	AHI >10. Exclusion: maxillomandibular retrognathia, prior oropharyngeal surgery, weight change > 10% during 6 mo of therapy	TBR	UPPP
Babademez²³ (2011): 1	2	15	NR	NR	NR	Supine OSA, CPAP intolerance. Exclusion: interfering systemic disease, prior oropharyngeal surgery or craniofacial insufficiency	LTBR	UPPP
Babademez²³ (2011): 2	2	15	NR	NR	NR	Supine OSA, CPAP intolerance. Exclusion: interfering systemic disease, prior oropharyngeal surgery or craniofacial insufficiency	SMILE	UPPP
Friedman⁷ (2012)	3	22	20	41.7 ± 8.6	31.5 ± 4.6	AHI >15	SMILE	ZPP
Suh³⁸ (2013)	4	50	42	43.3 ± 8.4	30.9 ± 4.0	AHI >15, CPAP intolerance. Exclusion: FTP II, mild OSA, BMI >40	MG + LT	UPPP
Lin³⁹ (2014)	4	35	32	42.0 ± NR	27.2 ± 2.7	AHI >30, BMI <32, primary OSA surgery only, CPAP intolerance, TBO	TBR	ZPP
Wee⁴⁰ (2015)	4	27	NR	42.4 ± 10.7	27.5 ± 3.6	AHI ≥5, TBO	LT	TL (24), UPPP (24), SNS (8), GGA (4)
Li⁴¹ (2016)	4	30	27	41.5 ± 9.4	26.4 ± 3.0	AHI >20, TBO, CPAP intolerance. Exclusion: no follow-up, prior OSA surgery	CELL	Relocation PhP
Folk⁶ (2017)	3	16	12	46.3 ± 8.4	32.5 ± 3.5	OSA with TBO	PMG, LT	TL, SNS, UPPP
Hwang⁸ (2018)	4	29	26	39.8 ± 10.8	26.8 ± 2.8	AHI >15, TBO. Exclusion: Prior OSA surgery	TBR	Lateral PhP
Park⁴² (2018)	3	17	14	40.2 ± 10.7	25.5 ± 3.6	CPAP intolerance, TBO	TBR	Lateral PhP
Chen²³ (2019)	2	22	20	NR	29.1 ± 3.5	AHI >30, CPAP intolerance, primary OSA surgery only, retropalatal plane concentrically collapsed, TBO	PMG	ZPP

Abbreviations: AHI, apnea-hypopnea index; BMI, body mass index; BoT, base of tongue; CELL, coblation endoscopic lingual lightening; CPAP, continuous positive airway pressure; FTP II, Friedman tongue position; GGA, genioglossus advancement; LT, lingual tonsillectomy; LTBR, low temperature bipolar radiofrequency; MG, midline glossectomy; NR, not reportable; OLE, Oxford level of evidence; OSA, obstructive sleep apnea; PhP, pharyngoplasty; PMG, partial midline glossectomy; SMILE, submucosal minimally invasive lingual excision; SNS, sinonasal surgery; TBO, tongue base obstruction; TBR, tongue base reduction/resection; TL, tonsillectomy; UPPP, uvulopalatoplasty; ZPP, Z-palatoplasty.

Figure 2.

Risk of bias.

Figure 3.

Funnel plot depicting bias. MD, mean difference.

A total of 834 patients across 18 studies underwent TORS TBR.^{5,6-8,20-22,26-36} The mean age of this group was 49.1 ± 3.84 years with a male:female ratio of 3.5:1. All studies except one³³ reported preoperative BMI, which averaged 28.46 ± 2.89 kg/m². Six studies reported the robot operative time and robot setup time separately, averaging 51.3 and 22.8 minutes, respectively.^{5,22,27,29-31} Eight studies reported the total operative time (encompassing true time in operation and robot setup), with a mean of 77.9 ± 16.38 minutes. The mean preoperative values for AHI, ESS, and LSAT were 40.66 ± 6.41, 12.41 ± 2.41, and 79.44 ± 1.73, respectively. TBR was the primary procedure in all but 2 studies, where the authors performed lingual tonsillectomy alone.^20,32 Procedures performed in conjunction with TBR varied widely but most commonly included UPPP, z-palatoplasty, and epiglottoplasty. The follow-up polysomnography was performed between 3 and 12 months postoperatively.

A total of 294 patients across 11 studies underwent plasma ablation TBR.^{6-8,23,24,37-42} The reported mean age in this group was 42.10 ± 1.57 years, which was significantly younger than that of the TORS group (P < .0001). More male patients underwent plasma ablation (88.20% ± 0.06%) than TORS (78.00% ± 0.09%; P < .0001). All but 1 study²³ reported the preoperative BMI in the plasma ablation group, with an average of 28.70 ± 2.15 kg/m²; this was not significantly different than that of the TORS group (P = .2697). The reported mean operative time was 44.00 ± 12.91 minutes across 3 studies,^8,23,37 and this was significantly shorter than that of TORS (P < .0001). The preoperative values for AHI (45.36 ± 12.95), ESS (10.77 ± 2.07), and LSAT (76.04 ± 5.72) were significantly different than those of the TORS cohort (P < .0001 for each comparison). The most common primary procedures included TBR,^8,37,39,42 lingual tonsillectomy,^6,38,40 and partial midline glossectomy.^6,24,39 Other primary procedures included low temperature bipolar radiofrequency,²³ submucosal minimally invasive lingual excision (SMILE),^7,23 and coblation endoscopic lingual lightening.⁴¹ Associated procedures ranged widely but most commonly included UPPP, z-palatoplasty, and other types of pharyngoplasty. The follow-up polysomnography was performed between 3 and 8 months postoperatively.

Treatment Outcomes

A summary of the treatment outcome for each group is presented in Tables 3 and 4 . The comparison of the study cohort is shown in Table 5 . Mean postoperative follow-up duration was 4.2 ± 2.6 months in 10 TORS studies and 4.6 ± 1.4 months in 7 plasma ablation studies (P = .0482). The postoperative weighted mean differences of AHI in the TORS group and plasma ablation group were –23.92 (95% CI, –27.06 to −20.79) and –22.07 (–27.62 to −16.52), with each demonstrating significant overall effect (P < .00001). Both TORS and plasma ablation achieved a similar magnitude of reduction in AHI (P = .58; Figure 4 ).

Table 3.

Outcomes of Transoral Robotic Surgery.

	Time, min, Mean ± SD
Study (Year)	Total Operative	True Operative	Setup	Follow-up Interval, mo	Success Rate, %	LOS, d, Mean ± SD	Postoperative Bleeding, %	Other Complications (No.)
Friedman⁷ (2012)	NR	NR	NR	2.9	66.7	1.6 ± 0.7	0.0	No major complications
Lee²⁰ (2012)	NR	NR	NR	8	45.0	NR	5.0	Transient dysphagia (5), transient dysgeusia (3), transient globus sensation (2)
Vicini⁵ (2012)	73.7	42.5 ± 15.2	31.2 ± 18.2	NR	60.0	NR	15.0	Pharyngeal edema (1), subcutaneous emphysema (2)
Toh²⁶ (2014)	NR	NR	NR	8.2	55.0	NR	5.0	Transient dysgeusia (7), transient dysphagia (1)
Vicini²⁷ (2014)	75.2	55.7 ±17.1	19.5 ± 10.5	3	66.9	3.5 ± 3.18	2.1	Transient globus sensation (0.4%), transient pharyngeal edema (0.4%), transient hypogeusia (14.2%), pharyngeal stenosis (0.4%)
Vicini²⁸ (2014): 1	125.9 ± 42.4	125.9 ± 42.4	NR	NR	33.3	7.3 ± 1.42	0.0	Moderate pharyngeal edema (1), subcutaneous emphysema (1)
Vicini²⁸ (2014): 2	140.5 ± 35.5	140.5 ± 35.5	NR	NR	83.3	8.3 ± 2.32	0.0	Pneumonia (1)
Chiffer²¹ (2015)	NR	NR	NR	6 to 24	61.1	NR	NR	NR
Lin²⁹ (2015)	83.7	59.5 ± 20.6	24.2 ± 7.7	4 to 6	53.9	3.5 ± 1.5	0.0	Dysphagia (3), transient dysgeusia, prolonged dysgeusia >1 y (3), airway concerns (0)
Muderris³⁰ (2015)	87.4	44.2 ± 6.8	43.2 ± 7.3	6	83.3	5.3 ± 1.7	0.0	Lingual edema (6)
Arora²² (2016)	66	21 ± 9	45 ± 21	6	63.6	NR	7.1	Dysgeusia (1), odynophagia (2)
Kayhan³¹ (2016)	50.4	22.2 ± 6.5	28.2 ± 7.2	3	8.0	NR	0.0	Transient odynophagia, transient pharyngeal edema
Folk⁶ (2017)	NR	NR	NR	NR	75.6	<1	NR	TORS and coblation: dehydration (3), gag sensation (1)
Meraj³² (2017)	NR	NR	NR	3	50.5	NR	NR	NR
Aynaci³³ (2018)	13 ± 4.7	13 ± 4.7	NR	6	NR	NR	15.0	No other complications
Cammaroto³⁴ (2018)	NR	NR	NR	12	74.5	NR	NR	NR
Hwang⁸ (2018)	26.4 ± 26.3	26.4 ± 26.3	NR	4	56.3	8.2 ± 5.1	12.5	Foreign body sensation (8), dysgeusia (4)
Lai³⁵ (2018)	NR	NR	NR	3	NR	NR	NR	NR
Huntley³⁶ (2019)	NR	NR	NR	NR	54.2	NR	0.0	No major complications

Abbreviations: LOS, length of stay; NR, not reportable; TORS, transoral robotic surgery.

Table 4.

Outcomes of Plasma Ablation.

Study (year)	Total Operative Time, min^a	Success Rate, %	Follow-up Interval, mo	LOS, d^b	Postoperative Bleeding, %	Other Complications (No.)
Babademez³⁷ (2010)	25-30	62.5	6	NR	0.0	Pain requiring analgesic >5 d (5), dysphagia >10 d (2), dysphagia >30 d (2), halitosis >30 d (1)
Babademez²³ (2011): 1	60.4 ± 13.1	NR	3	NR	0.0	No major complications
Babademez²³ (2011): 2	58.8 ± 11.9	NR	3	NR	0.0	No major complications
Friedman⁷ (2012)	NR	45.5	6	1.5 ± 0.9	0.0	No major complications
Suh³⁸ (2013)	NR	56.0	3-6	<1	4.0	Temporary dysgeusia (2), prolonged dysphagia (4), oropharyngeal stenosis (1), HGN injury (0)
Lin³⁹ (2014)	NR	62.9	3	NR	2.9	Airway concerns (0), pain requiring meds (mean, 8.4 d), transient VPI (19), transient ear fullness (12), transient dysgeusia (15), persistent dysgeusia >3 mo (8), hypopharyngeal stenosis (0), HGN injury (0)
Wee⁴⁰ (2015)	NR	55.6	6	2-3	3.7	Respiratory distress (1), dental injury (0), tongue mobility (0), taste (0), swallow (0), stenosis (0)
Li⁴¹ (2016)	NR	73.0	6-8	NR	NR	NR
Folk⁶ (2017)	NR	56.3	NR	<1	NR	TORS and coblation: dehydration (3), gag sensation (1). TORS/coblation unspecified
Hwang⁸ (2018)	35.6 ± 23.9	62.1	4	4.4 ± 2.4	27.6	Dysgeusia (9), foreign body sensation (18)
Park⁴² (2018)	NR	70.6	NR	NR	29.4	Transient FBS (9), transient dysgeusia (5), VPI (0)
Chen²⁴ (2019)	NR	63.6	6	7.6 ± 1.3	18.2	Submental swelling (4), prolonged dysphagia (5), HGN injury (0)

Abbreviations: FBS, foreign body sensation; HGN, hypoglossal nerve; LOS, length of stay; NR, not reportable; TORS, transoral robotic surgery; VPI, velopharyngeal insufficiency.

Range or mean ± SD.

Mean ± SD, range, or No.

Table 5.

Comparisons of TORS vs Plasma Ablation Cohorts.

	TORS		Plasma Ablation
Measure	No.	Mean ± SD^a	No.	Mean ± SD^a	Mean Difference (95% CI)	P Value
Age, y	834	49.1 ± 3.8	242	42.1 ± 1.6	7.0 (6.5 to 7.5)	<.0001^b
Male, %	834	78.0 ± 0.1	237	88.2 ± 0.1	−10.2 (–11.4 to −8.9)	<.0001^b
BMI, kg/m²	814	28.5 ± 2.9	264	28.7 ± 2.1	−0.2 (–0.6 to 0.2)	.2697
Total operative time, min	362	77.9 ± 16.4	75	44.0 ± 12.9	33.9 (30.0 to 37.9)	<.0001^b
LOS, d	355	3.9 ± 1.6	89	3.9 ± 2.5	0.7 (–0.4 to 0.4)	.9047
Pre-AHI	834	40.7 ± 6.4	294	45.4 ±13.0	−4.7 (–5.8 to −3.6)	<.0001^b
Pre-ESS	610	12.4 ± 2.1	150	10.8 ± 2.1	1.6 (1.3 to 2.0)	<.0001^b
Pre-LSAT, % (O²)	620	79.4 ± 1.7	215	76.0 ± 5.7	3.4 (2.9 to 3.9)	<.0001^b
Follow-up interval, mo	563	4.2 ± 2.6	181	4.6 ± 1.4	0.4 (0.0 to 0.8)	.0482^b
Success rate, %	693	57.6 (49.4-65.6)	264	60.3 (54.3-66.2)	−2.7 (–4.3 to 9.5)	.4474
Postoperative bleeding, %	498	3.3 (1.9-5.2)	248	7.5 (2.7-14.3)	−4.2 (0.9 to 8.4)	.0103^b

Abbreviations: AHI, apnea-hypopnea index; BMI, body mass index; ESS, Epworth Sleepiness Scale; LSAT, lowest oxygen saturation; LOS, length of stay; TORS, transoral robotic surgery.

Mean (95% CI) where indicated.

P < .05.

Figure 4.

Apnea-hypopnea index of patients receiving transoral robotic surgery (TORS) vs plasma ablation.

Twelve studies comprising 610 patients in the TORS group showed a weighted mean difference in ESS of –7.6 (95% CI, –9.94 to −5.26). A total of 150 patients in the plasma ablation group across 6 studies demonstrated a weighted mean difference in ESS of –4.14 (–5.88 to −2.40; Figure 5 ). Both achieved significant overall reduction in ESS (P < .0001), but ESS reduction was greater in TORS than in plasma ablation (P = .02).

Figure 5.

Epworth Sleepiness Scale of patients receiving transoral robotic surgery (TORS) vs plasma ablation.

Finally, 12 studies comprising 620 patients in the TORS group showed improvements in LSATs, with a weighted mean difference of 5.83% (95% CI, 1.95-9.71). For 215 patients in the 8 studies in the plasma ablation group, the weighted mean difference of LSAT was 5.48% (3.47%-7.50%). Each intervention demonstrated significant overall improvement of LSAT (P < .0001) with comparable results (P = .88; Figure 6 ).

Figure 6.

Lowest oxygen saturation of patients receiving transoral robotic surgery (TORS) vs plasma ablation.

Surgical success was defined as a 50% reduction in AHI with a postoperative AHI <20 in 16 TORS studies and 9 plasma ablation studies. Corresponding surgical success rates were 57.6% (95% CI, 49.4%-65.6%) in TORS and 60.3% (54.3%-66.2%) in the plasma ablation group ( Figure 7 ). The difference in success rates was not statistically significant (P = .4474). Seven studies in the TORS group reported the length of stay (LOS), with an average of 3.9 ± 1.56 days; in 3 plasma ablation studies that reported LOS, the average was 3.9 ± 2.53 days. The difference in LOS was not statistically significant (P = .9047; Table 5 ).

Figure 7.

Surgical success rates in transoral robotic surgery (TORS; left) vs plasma ablation (right).

Surgical Complications

Thirteen TORS studies and 9 plasma ablation studies provided adequate data for evaluation of surgical complications. Postoperative bleeding occurred at a significantly lower rate at 3.3% (95% CI, 1.9%-5.2%) in the TORS group as compared with 7.5% (2.7%-14.3%) in the plasma ablation group (P = .0103; Table 5 ). No studies reported cases of iatrogenic hypoglossal nerve injury. Other common side effects included dysgeusia, dysphagia, and globus sensation, but these complaints were often transient and infrequent; thus, they were not included in statistical analysis.

Discussion

TORS and plasma ablation are the most commonly published techniques in TBR for OSA.¹¹ Using plasma ablation in TBR offers a reasonable means of treating OSA related to the base of tongue by allowing one to target a wide and superficial field without harming deeper neurovascular structures. TORS employs a similar technique but using a robot additionally offers superior magnification, 3-dimensional perception, and precise tissue handling.⁴³ Despite these advantages, decreased tactile feedback, access to technology, and added costs have prohibited its wide adoption.⁷ Therefore, data concerning efficacy and complications of either technique are of particular interest to the sleep medicine community.

Treatment Efficacy

In the present study, the effectiveness of TORS and plasma ablation surgery in TBR were compared with AHI, ESS, and LSAT, with both techniques demonstrating significant improvement in these outcome variables. Notably, our analysis revealed comparable effectiveness between the modalities as well as equivocal rates of surgical success (TORS, 57.6%; plasma ablation, 60.3%). Although significant differences in preoperative parameters existed between the cohorts, patients receiving either intervention on average far exceeded the threshold for severe OSA (AHI of 40.66 and 45.36 for TORS and plasma ablation, respectively), suggesting that the preoperative difference for this parameter might be clinically insignificant. TORS also demonstrated significantly greater reduction in ESS as compared with plasma ablation, but further investigation is needed to evaluate the relevance of difference seen in patient-perceived outcomes. In this study, the mean preoperative ESS of the TORS cohort was higher than the plasma ablation cohort, thereby providing a wider range for potential improvement. Altogether, our results indicate that TORS and plasma ablation represent effective treatments for OSA and offer comparable results.

Consistent with our report, 3 institutional studies that directly compared TORS with plasma ablation found that TORS was no different than plasma ablation in achieving surgical success.^6-8 The study by Friedman et al,⁷ however, found significantly greater percentage reduction in AHI with TORS (60.5% ± 24.9%) as compared with plasma ablation (37% ± 51.6%). In that study, the authors used plasma ablation in the SMILE technique, while Hwang et al⁸ and Folk and D’Agostino⁶ used plasma ablation in tongue base resection and posterior midline glossectomy, respectively. It is possible that the plasma ablation partial glossectomy technique results in improved visualization to allow more tissue removal as compared with the SMILE technique, which is largely performed through blind submucosal tunnels. Although the present study did not compare the efficacy between specific plasma ablation techniques, such comparison in a future study may shed light on the advantages of different approaches.

Cammaroto et al also compared TORS with plasma ablation for OSA.¹¹ In their systematic review, the authors described nuanced differences among various surgical techniques and approaches. Unlike the present study however, that study failed to meta-analyze their primary outcomes and statistically compare them between TORS and plasma ablation. In addition, that study included 15 articles, while the present study gathered data from 26 articles.

TORS vs Plasma Ablation: Other Factors to Consider in Surgical Decision

Given comparable efficacy in TORS and plasma ablation, other aspects of surgical management might ultimately guide the technique of choice. For instance, postoperative bleeding occurred at a significantly lower rate in patients receiving TORS (3.26%) than those receiving plasma ablation (7.46%). Decreased bleeding risk in TORS might indeed be due to the superior visualization of the tongue base. Of note, bleeding more often required hospital admission in patients who had received plasma ablation, although we did not quantify these data.

An additional difference found between techniques was time under anesthesia. As expected, the total operative time in TORS significantly exceeded that of plasma ablation by approximately 34 minutes, owing to the time required for robotic setup. Typical patients with moderate to severe OSA already carry higher operative risk. The operative time therefore may represent a relevant factor in choosing an appropriate surgical technique. Longer operative times may reduce room turnover and therefore hold implications for opportunity cost in addition to the increased direct cost of the operation itself. The LOS did not differ between interventions and, as such, might not contribute to the overall difference in expenditure. However, anecdotal evidence supports a shorter LOS for plasma ablation, and only 3 reports of plasma ablation provided evaluable data on LOS.^14,15,30 Therefore, future studies investigating this factor would potentially highlight differences in associated costs. Nevertheless, as established medical device companies are now developing competing robotic technologies, cost of TORS may soon decline and permit more widespread use.⁴³

Limitations

Our study had several limitations. First, the majority of included studies represented institutional chart reviews and thus carried biases inherent to a retrospective study. Next, inconsistencies in patient selection existed, where some studies excluded revision surgery for OSA. Patients in the plasma ablation cohort received different forms of TBR, ranging from lingual tonsillectomy to larger procedures such as midline glossectomy. While all patients underwent TBR, they also had a variety of associated pharyngeal and nasal procedures that were concurrently performed. As multilevel surgery represents a standard of OSA surgery,⁴⁴ an isolated comparison of TBR surgery was not feasible, though this comparison would provide insightful information in future investigations.

Conclusion

OSA is an increasingly prevalent public health concern and warrants an exploration of a reliable treatment option. TORS and plasma ablation in TBR both provide effective reductions in AHI and adequate rates of surgical success with comparable results. Ultimately, the choice between techniques might depend on individual patient risk factors and knowledge of associated costs of operation.

Author Contributions

Joshua A. Lee, acquired and analyzed data, drafted and revised manuscript, approved final version, agreed to be accountable for work; Young Jae Byun, acquired and analyzed data, drafted and revised manuscript, approved final version, agreed to be accountable for work; Shaun A. Nguyen, contributed to study design, analyzed data, revised manuscript, approved final version, agreed to be accountable for work; Eric J. Lentsch, contributed to study design, interpreted data, revised manuscript, approved final version, agreed to be accountable for work; M. Boyd Gillespie, contributed to study design, interpreted data, revised manuscript, approved final version, agreed to be accountable for work.

Disclosures

Competing interests: M. Boyd Gillespie is a consultant for Medtronic, Cook, and Cryosa; a research and investigator for Cook, ImThera, Inspire, and Olympus; and on the Scientific Advisory Board for Cryosa and Zelegent.

Sponsorships: None.

Funding source: None.

References

Franklin

Lindberg

. Obstructive sleep apnea is a common disorder in the population-a review on the epidemiology of sleep apnea. J Thorac Dis. 2015;7(8):1311-1322.

Senaratna

Perret

Lodge

, et al. Prevalence of obstructive sleep apnea in the general population: a systematic review. Sleep Med Rev. 2017;34:70-81.

Cho

Park

Min

Chung

Lee

Kim

. Endoscope-guided coblator tongue base resection using an endoscope-holding system for obstructive sleep apnea. Head Neck. 2016;38(4):635-639.

Woodson

Laohasiriwong

. Lingual tonsillectomy and midline posterior glossectomy for obstructive sleep apnea. Oper Tech Otolayngol Head Neck Surg. 2012;23(2):155-161.

Vicini

Dallan

Canzi

, et al. Transoral robotic surgery of the tongue base in obstructive sleep apnea-hypopnea syndrome: anatomic considerations and clinical experience. Head Neck. 2012;34(1):15-22.

Folk

D’Agostino

. Transoral robotic surgery vs endoscopic partial midline glossectomy for obstructive sleep apnea. World J Otorhinolaryngol Head Neck Surg. 2017;3(2):101-105.

Friedman

Hamilton

Samuelson

, et al. Transoral robotic glossectomy for the treatment of obstructive sleep apnea-hypopnea syndrome. Otolaryngol Head Neck Surg. 2012;146(5):854-862.

Hwang

Kim

, et al. Comparison of robotic and coblation tongue base resection for obstructive sleep apnoea. Clin Otolaryngol. 2018;43(1):249-255.

Justin

Chang

Camacho

Brietzke

. Transoral robotic surgery for obstructive sleep apnea: a systematic review and meta-analysis. Otolaryngol Head Neck Surg. 2016;154(5):835-846.

10.

Miller

Nguyen

Ong

Gillespie

. Transoral robotic base of tongue reduction for obstructive sleep apnea: a systematic review and meta-analysis. Laryngoscope. 2017;127(1):258-265.

11.

Cammaroto

Montevecchi

D’Agostino

, et al. Tongue reduction for OSAHS: TORSs vs coblations, technologies vs techniques, apples vs oranges. Eur Arch Otorhinolaryngol. 2017;274(2):637-645.

12.

Moher

Liberati

Tetzlaff

Altman

; PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-analyses: the PRISMA statement. Int J Surg. 2010;8(5):336-341.

13.

Higgins

JPT

Green

. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. London, UK: The Cochrane Collaboration; 2011.

14.

Borenstein

. Introduction to Meta-analysis. Chichester, UK: John Wiley & Sons; 2009.

15.

Freeman

Tukey

. Transformations related to the angular and the square root. Annals of Mathematical Statistics. 1950;21(4):607-611.

16.

DerSimonian

Laird

. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177-188.

17.

Egger

Davey Smith

Schneider

Minder

. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629-634.

18.

Sterne

Egger

. Funnel plots for detecting bias in meta-analysis: guidelines on choice of axis. J Clin Epidemiol. 2001;54(10):1046-1055.

19.

Sterne

JAC

Sutton

Ioannidis

JPA

, et al. Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ. 2011;343:d4002.

20.

Lee

Weinstein

O’Malley

Jr Thaler

. Transoral robot-assisted lingual tonsillectomy and uvulopalatopharyngoplasty for obstructive sleep apnea. Ann Otol Rhinol Laryngol. 2012;121(10):635-639.

21.

Chiffer

Schwab

Keenan

Borek

Thaler

. Volumetric MRI analysis pre- and post-Transoral robotic surgery for obstructive sleep apnea. Laryngoscope. 2015;125(8):1988-1995.

22.

Arora

Chaidas

Garas

, et al. Outcome of TORS to tongue base and epiglottis in patients with OSA intolerant of conventional treatment. Sleep Breath. 2016;20(2):739-747.

23.

Babademez

Yorubulut

Yurekli

, et al. Comparison of minimally invasive techniques in tongue base surgery in patients with obstructive sleep apnea. Otolaryngol Head Neck Surg. 2011;145(5):858-864.

24.

Chen

Wang

Huang

. Z-palatopharyngoplasty combined with 70-degree endoscopy-assisted coblator partial medial glossectomy on severe obstructive sleep apnea. Acta Otolaryngol. 2019;139(10):902-907.

25.

OCEBM Levels of Evidence Working Group. The Oxford levels of evidence 2. https://www.cebm.net/index.aspx?o=5653. Published 2011. Accessed August 2, 2019.

26.

Toh

Han

Tay

Kiong

KLQ

. Transoral robotic surgery for obstructive sleep apnea in asian patients a Singapore sleep centre experience. JAMA Otolaryngol Head Neck Surg. 2014;140(7):624-629.

27.

Vicini

Montevecchi

Campanini

, et al. Clinical outcomes and complications associated with TORS for OSAHS: a benchmark for evaluating an emerging surgical technology in a targeted application for benign disease. ORL J Otorhinolaryngol Relat Spec. 2014;76(2):63-69.

28.

Vicini

Montevecchi

Pang

, et al. Combined transoral robotic tongue base surgery and palate surgery in obstructive sleep apnea-hypopnea syndrome: expansion sphincter pharyngoplasty versus uvulopalatopharyngoplasty. Head Neck. 2014;36(1):77-83.

29.

Lin

Rowley

Folbe

Yoo

Badr

Chen

. Transoral robotic surgery for treatment of obstructive sleep apnea: factors predicting surgical response. Laryngoscope. 2015;125(4):1013-1020.

30.

Muderris

Sevil

Bercin

Gul

Kiris

. Transoral robotic lingual tonsillectomy in adults: preliminary results. Acta Otolaryngol. 2015;135(1):64-69.

31.

Kayhan

Kaya

Koc

, et al. Multilevel combined surgery with transoral robotic surgery for obstructive sleep apnea syndrome. J Craniofac Surg. 2016;27(4):1044-1048.

32.

Meraj

Muenz

Glazer

Harvey

Spector

Hoff

. Does drug-induced sleep endoscopy predict surgical success in transoral robotic multilevel surgery in obstructive sleep apnea? Laryngoscope. 2017;127(4):971-976.

33.

Aynaci

Karaman

Kerşin

Fındik

. Comparison of radiofrequency and transoral robotic surgery in obstructive sleep apnea syndrome treatment. Acta Otolaryngol. 2018;138(5):502-506.

34.

Cammaroto

Meccariello

Costantini

, et al. Trans-oral robotic tongue reduction for OSA: does lingual anatomy influence the surgical outcome? J Clin Sleep Med. 2018;14(8):1347-1351.

35.

Lai

Lin

, et al. Effects of upper airway surgery on daytime sleepiness in nonobese patients with obstructive sleep apnea/hypopnea syndrome. Ann Otol Rhinol Laryngol. 2018;127(12):912-918.

36.

Huntley

Topf

Christopher

Doghramji

Curry

Boon

. Comparing upper airway stimulation to transoral robotic base of tongue resection for treatment of obstructive sleep apnea. Laryngoscope. 2019;129(4):1010-1013.

37.

Babademez

Ciftci

Acar

, et al. Low-temperature bipolar radiofrequency ablation (Coblation) of the tongue base for supine-position-associated obstructive sleep apnea. ORL J Otorhinolaryngol Relat Spec. 2010;72(1):51-55.

38.

Suh

. Evaluation of open midline glossectomy in the multilevel surgical management of obstructive sleep apnea syndrome. Otolaryngol Head Neck Surg. 2013;148(1):166-171.

39.

Lin

Friedman

Chang

Yalamanchali

. Z-palatopharyngoplasty combined with endoscopic coblator open tongue base resection for severe obstructive sleep apnea/hypopnea syndrome. Otolaryngol Head Neck Surg. 2014;150(6):1078-1085.

40.

Wee

Tan

Lee

Rhee

Kim

. Evaluation of coblation lingual tonsil removal technique for obstructive sleep apnea in Asians: preliminary results of surgical morbidity and prognosticators. Eur Arch Otorhinolaryngol. 2015;272(9):2327-2333.

41.

Lee

Kezirian

. Efficacy of coblation endoscopic lingual lightening in multilevel surgery for obstructive sleep apnea. JAMA Otolaryngol Head Neck Surg. 2016;142(5):438-443.

42.

Park

Chung

Park

, et al. Surgical outcomes of overlapping lateral pharyngoplasty with or without coblator tongue base resection for obstructive sleep apnea. 2018;275(5):1189-1196.

43.

Garas

Kythreotou

Georgalas

, et al. Is transoral robotic surgery a safe and effective multilevel treatment for obstructive sleep apnoea in obese patients following failure of conventional treatment(s)? Ann Med Surg (Lond). 2017;19:55-61.

44.

Kao

Shnayder

Lee

. The efficacy of anatomically based multilevel surgery for obstructive sleep apnea. Otolaryngol Head Neck Surg. 2003;129(4):327-335.