Transoral Robotic Surgery Versus Traditional Lingual Tonsillectomy for Persistent Pediatric Sleep Apnea

ABSTRACT

Objectives

This study compares postoperative outcomes of transoral robotic surgery (TORS) and traditional lingual tonsillectomy in pediatric patients with persistent obstructive sleep apnea (OSA) following tonsillectomy and adenoidectomy (T&A).

Methods

A systematic search of Ovid MEDLINE, Embase, Cochrane CENTRAL, and Web of Science was conducted through January 2026. Inclusion criteria: pediatric patients (< 18 years) with persistent OSA following T&A, TORS, or traditional lingual tonsillectomy. Outcomes assessed included postoperative apnea–hypopnea index (AHI) reduction, bleeding, and non‐bleeding complications.

Results

A total of 17 studies were analyzed (n = 470; 25 TORS and 445 traditional procedures). Postoperative dysphagia was significantly higher among TORS (32%) versus traditional (1.3%; p < 0.001). Postoperative infection rates were greater after TORS (16% vs. 1.3%; χ ² = 14.88, p < 0.001). Meta‐analysis demonstrated significantly greater postoperative bleeding with TORS (12%) versus traditional (3%, χ ² = 3.98, p = 0.046). AHI reduction was greater in TORS (mean: 16.20 events/h, 95% CI: 9.01–23.39) than in traditional (mean: 3.64 events/h, 95% CI: 2.03–5.25), with a significant subgroup difference (χ ² = 11.41, p < 0.001) indicating greater symptom resolution following TORS. No significant differences were observed in postoperative airway obstruction or bleeding requiring surgical intervention.

Conclusion

Although TORS offers greater improvements in persistent pediatric OSA following T&A, its association with higher rates of postoperative dysphagia, infection, and bleeding warrants careful decision‐making and further investigation in larger cohorts to refine patient selection and optimize outcomes.

1. Introduction

Obstructive sleep apnea (OSA) is a common disorder in the pediatric population, defined by the American Thoracic Society and the American Academy of Pediatrics as intermittent, complete, or partial obstruction of the upper airway during sleep, resulting in disrupted sleep architecture and gas exchange abnormalities [1]. OSA affects approximately 1%–6% of children and adolescents, with peak prevalence occurring between the ages of 2 and 8 years, a period that coincides with the highest incidence of adenotonsillar hypertrophy [1, 2]. Additional factors that lead to upper airway collapse include obesity‐related fat deposition and craniofacial abnormalities such as midface hypoplasia or micrognathia, which are more prevalent in syndromic patients [1, 2, 3]. Accurate diagnosis is essential, and polysomnography (PSG) remains the gold standard for identifying and grading the severity of OSA, with apnea–hypopnea index (AHI) as a measure of OSA severity [4]. An AHI of 1–4.9 events/h is considered mild, 5–9.9 is moderate, and ≥ 10 indicates severe OSA [1].

For children in whom tonsillar hypertrophy is the dominant anatomical contributor to their OSA, the first‐line treatment is tonsillectomy and adenoidectomy (T&A), which often results in substantial clinical improvement [5]. However, 20%–30% of children experience persistent OSA following initial T&A [6]. In many of these cases, multilevel airway obstruction, characterized by collapse at multiple anatomical sites, is a contributing factor. T&A has shown limited efficacy in certain high‐risk subgroups, including children with obesity, craniofacial anomalies, and syndromic conditions such as Down syndrome [1]. For patients with persistent OSA, second‐line interventions such as continuous positive airway pressure (CPAP), oral appliances, hypoglossal nerve stimulation, medical management, or additional surgical procedures may be required [7].

In particular, lingual tonsil hypertrophy is a recognized cause of persistent OSA following T&A, and lingual tonsillectomy has emerged as an effective treatment option in select cases. Several surgical approaches have been described, including sharp dissection, laser excision, endoscopic‐assisted coblation, and unipolar diathermy [8]. Recently, transoral robotic surgery (TORS), initially developed for the resection of oropharyngeal tumors, has been adapted for use in pediatric lingual tonsillectomy [9]. TORS offers superior visualization and surgical precision, contributing to its increasing application in this patient population [10].

If left untreated, pediatric OSA can lead to significant long‐term consequences, including neurocognitive impairment, behavior problems, and cardiovascular and metabolic comorbidities [5]. Children with persistent OSA have shown decreased cognitive function, as evidenced by reduced performance efficiency and decreased capability across verbal and nonverbal commands [11]. In addition, those who sleep less than 5 h per night have shown increased risk patterns for hypertension and obesity. More specifically, an AHI score ≥ 5 events/h alone increases the risk of elevated systolic and diastolic blood pressure, emphasizing the significant cardiovascular burden of OSA [1]. Together, these findings highlight the importance of effective management of pediatric OSA to avoid long‐term consequences. This systematic review compares postoperative AHI score improvements and other complications following TORS‐assisted versus traditional transoral lingual tonsillectomy in pediatric patients with persistent OSA following T&A, providing physicians with the evidence to guide surgical decision‐making.

2. Materials and Methods

This study did not require Institutional Review Board approval since the data were collected from previously published studies.

2.1. Search Strategy

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines. A literature search was conducted to identify publications in the MEDLINE, Embase, Cochrane CENTRAL (via Ovid), and Web of Science databases, covering the period from inception to January 2026. The search strategy included database‐specific subject headings and keywords related to pediatrics, OSA, lingual tonsillectomy, and TORS.

2.2. Eligibility Criteria

The PICO statement for this study is as follows: participants, pediatric patients (< 18 years of age) with persistent OSA following T&A; intervention, TORS‐assisted lingual tonsillectomy; comparator, traditional transoral lingual tonsillectomy techniques; outcomes, improvement in AHI, length of hospital stay, and incidence of postoperative complications; and study design, systematic review.

Inclusion criteria included peer‐reviewed randomized‐controlled trials, retrospective and prospective cohort studies, and case series and reports written in English that investigated pediatric patients who underwent lingual tonsillectomy for persistent OSA following adenotonsillectomy. Exclusion criteria encompassed review articles, meta‐analyses, editorials, cadaveric or animal studies, abstracts without full‐text availability, and articles lacking English full‐text translations. Studies were also excluded if they involved the incorrect population (e.g., adults).

2.3. Study Selection

The final search results were exported into Covidence for assessment of eligibility criteria. The review began with a title and abstract screen, followed by a full‐text review for articles meeting inclusion criteria. Three independent reviewers (M.D., K.M., and S.S.) conducted both stages, with discrepancies in study eligibility resolved by two additional independent reviewers (E.C. and C.Z.).

2.4. Risk of Bias Assessment

Each study was assessed for risk of bias using a Methodological Index for Nonrandomized Studies (MINORS) score detailed in Table 1 [12].

TABLE 1.

Characteristics of included studies and patient cohorts.

Author (year)	Country	Study design	# Children with persistent OSA who underwent lingual tonsillectomy	Type of lingual tonsillectomy	MINORS score
Abdel‐Aziz et al. (2011) [8]	Egypt	Retrospective	16	Traditional	13/16
Barakate et al. (2008) [13]	Australia	Prospective	13	Traditional	12/16
Chan et al. (2012) [14]	USA	Retrospective	68	Traditional	10/16
DeMarcantonio et al. (2016) [15]	USA	Retrospective	57	Traditional	10/16
Leonardis et al. (2013) [16]	USA	Retrospective	11	TORS	11/16
Lin et al. (2009) [17]	USA	Retrospective	26	Traditional	11/16
Maksimoski et al. (2022) [18]	USA	Retrospective	20	Traditional	13/16
Mikulski et al. (2023) [19]	USA	Retrospective	2	TORS	10/16
Montevecchi et al. (2017) [20]	Italy	Retrospective	3	TORS	10/16
Propst et al. (2017) [21]	Canada	Retrospective	13	Traditional	10/16
Prosser et al. (2017) [22]	USA	Retrospective	21	Traditional	11/16
Skirko et al. (2018) [23]	USA	Retrospective	39	Traditional	10/16
Thottam et al. (2015) [24]	USA	Retrospective	9	TORS	10/16
Trandafir et al. (2025) [25]	France	Retrospective	50	Traditional	13/16
Ulualp et al. (2019) [26]	USA	Retrospective	10	Traditional	9/16
Williamson et al. (2023) [27]	USA	Retrospective	101	Traditional	12/16
Williamson et al. (2024) [28]	USA	Retrospective	11	Traditional	12/16

2.5. Data Extraction

Basic patient demographics were collected, including age, sex, and race. In addition, preoperative characteristics were collected, including obesity, underlying syndromes, and prior surgeries, to ensure that every patient analyzed met our strict inclusion criteria. Study design and procedural details were collected for descriptive analysis. Primary outcomes included symptom amelioration as determined by improvement in AHI, length of hospital stay, and incidence of postoperative complications, including bleeding, dysphagia, airway obstruction, and infection.

2.6. Statistical Analysis

For each categorical preoperative and postoperative outcome, we compared the number of affected children in the TORS and traditional lingual tonsillectomy cohorts with Fisher's exact test, which is appropriate given the small cell counts in several tables.

All meta‐analyses were performed in R (packages: meta and metafor). Continuous outcomes (postoperative AHI and ΔAHI) were pooled using metamean, while binary outcomes were analyzed using metaprop. A restricted‐maximum‐likelihood random‐effects model with Hartung–Knapp confidence limits was applied whenever heterogeneity exceeded 40% (I ² > 40%); otherwise, a common‐effect model was reported.

For outcomes with fewer than 20 total events, a generalized linear mixed model was fitted, and a 0.5 continuity correction was applied to zero cells. Forest plots display study‐level estimates, subgroup diamonds for each surgical technique, and the χ ² test that corresponds to the model shown.

3. Results

3.1. Study Selection

The literature search initially identified 5391 studies, with 4175 unique citations remaining after the removal of 1216 duplicates. After title and abstract screening, 32 potential full texts were reviewed for inclusion. Of these, 17 articles were selected for data abstraction, 13 of which involved traditional lingual tonsillectomy, while the remaining four utilized TORS tonsillectomy (Figure 1).

FIGURE 1.

PRISMA flow diagram illustrating the study selection process for inclusion in the systematic review.

3.2. Patient and Study Characteristics

The final dataset involved 470 pediatric patients who either underwent TORS (n = 25) or traditional lingual tonsillectomy (n = 445) for persistent OSA following T&A. From the 17 included studies, 11 of them included stratified data on this specific cohort (n = 220); 137 were male, and 83 were female, with a mean age of 11.5 years. Table 1 provides a brief overview of the included studies, including study design, risk of bias, and procedural details.

3.3. Postoperative Non‐Bleeding Complications

Fisher's exact test was used to assess associations between postoperative complications and surgical technique. The resulting p value (< 0.001) indicated a statistically significant association, suggesting complication rates differed by intervention type (Table 2).

TABLE 2.

Postoperative non‐bleeding complications by intervention.

Postoperative complications	TORS lingual tonsillectomy, no. of children (%) (n = 25)	Traditional lingual tonsillectomy, no. of children (%) (n = 445)	p
Dysphagia	8 (32%)	6 (1.3%)	< 0.001 a
Obstructive airway	0 (0%)	21 (4.7%)
Infection	4 (16%)	6 (1.3%)

^a Fisher's exact test was used to assess associations between postoperative complications and surgical technique.

3.3.1. Dysphagia

Pooled data from 17 studies demonstrated a higher incidence of postoperative dysphagia in the TORS group compared to the traditional lingual tonsillectomy group. Among the 13 traditional lingual tonsillectomy studies (n = 445), the pooled incidence was 0.01 (95% CI: 0.01–0.03) (Figure 2). In contrast, four TORS tonsillectomy studies (n = 25) had a pooled incidence of 0.32 (95% CI: 0.17–0.52). Heterogeneity was minimal overall (I ² = 15%) and absent within each subgroup (I ² = 0%) (Figure 2). The χ ² test for subgroup difference was significant (χ ² = 35.5, df = 1, p < 0.001), indicating a significantly higher dysphagia rate following TORS (Figure 2).

FIGURE 2.

Forest plot comparing the incidence of postoperative dysphagia between TORS and traditional lingual tonsillectomy in pediatric patients with persistent OSA.

3.3.2. Obstructive Airway Events

Obstructive airway events were infrequent overall, but 21 cases were reported in the 13 traditional lingual tonsillectomy studies. This resulted in moderate heterogeneity (I ² = 60%) and a pooled incidence of 0.06 (95% CI: 0.03–0.12) under a random‐effects model. No airway obstruction events were reported in the four TORS lingual tonsillectomy studies. Despite this, the pooled estimate for TORS was 0.08 (95% CI: 0.03–0.22). The χ ² test for subgroup difference was not significant (χ ² = 0.34, df = 1, p = 0.56), indicating no statistical difference between techniques.

3.3.3. Infection

Postoperative infection was uncommon among traditional lingual tonsillectomy studies but was more prevalent in TORS lingual tonsillectomy. Across the 13 traditional studies, the pooled infection rate was 0.01 (95% CI: 0.01‐ 0.03) with no heterogeneity (I ² = 0%) (Figure 3). The χ ² test for subgroup difference was statistically significant (χ ² = 14.88, df = 1, p < 0.001), suggesting a higher infection risk associated with TORS. However, this result should be interpreted cautiously due to the small TORS sample size.

FIGURE 3.

Forest plot comparing the incidence of postoperative infection between TORS and traditional lingual tonsillectomy in pediatric patients with persistent OSA.

3.4. Postoperative Bleeding Complications

Using Fisher's exact test, the p value was 0.025, indicating a statistically significant difference in rates of postoperative bleeding between TORS and traditional lingual tonsillectomy groups (Table 3).

TABLE 3.

Postoperative bleeding complications by intervention.

Bleeding complications	TORS lingual tonsillectomy, no. of children (%) (n = 25)	Traditional lingual tonsillectomy, no. of children (%) (n = 445)	p
Postoperative bleeding without surgical management	3 (12%)	15 (3.4%)	0.025 a
Postoperative bleeding with surgical management	1 (4.0%)	10 (2.2%)

^a Fisher's exact test was used to assess associations between postoperative complications and surgical technique.

3.4.1. Postoperative Bleeding Complications Without Surgical Intervention

Although postoperative bleeding was rare, the incidence was higher in TORS lingual tonsillectomy cohorts. In 13 traditional lingual tonsillectomy studies, the pooled incidence was 0.03 (95% CI: 0.02–0.06). In contrast, the pooled incidence from four TORS lingual tonsillectomy studies was 0.12 (95% CI: 0.04–0.31). There was no heterogeneity within subgroups (I ² = 0%). The χ ² test for subgroup difference reached statistical significance (χ ² = 3.98, df = 1, p = 0.046), indicating a higher bleeding risk following TORS (Figure 4).

FIGURE 4.

Forest plot comparing postoperative bleeding without surgical intervention between TORS and traditional lingual tonsillectomy in pediatric patients with persistent OSA.

3.4.2. Postoperative Bleeding Complications Requiring Surgical Intervention

Surgical management for postoperative bleeding was extremely rare in both groups. Ten cases occurred across 13 traditional lingual tonsillectomy studies, yielding a pooled incidence of 0.02 (95% CI: 0.01–0.04). One event occurred in the TORS cohorts, yielding a pooled incidence of 0.04 (95% CI: 0.01–0.24). No heterogeneity was observed (I ² = 0%). The χ ² test for subgroup difference was not significant (χ ² = 0.47, df = 1, p = 0.49).

3.5. Postoperative AHI Reduction

Limited evidence was available to compare postoperative AHI outcomes. Two traditional lingual tonsillectomy studies (n = 140) yielded a pooled postoperative mean AHI reduction of 3.64 events/h (95% CI: 2.03–5.25; I ² = 0%). A single TORS study (n = 9) reported a higher postoperative mean AHI reduction of 16.20 events/h (95% CI: 9.01–23.39). Pooled analysis using a random‐effects model (τ ² = 36.38; I ² = 83%) gave a combined postoperative mean AHI reduction of 8.40 events/h (95% CI: −8.07 to 24.88) (Figure 5). The χ ² test for subgroup difference was statistically significant (χ ² = 11.41, df = 1, p < 0.001), indicating a notable discrepancy between techniques, though this result is based on very limited data and should be interpreted cautiously.

FIGURE 5.

Forest plot comparing postoperative AHI reduction between TORS and traditional lingual tonsillectomy in pediatric patients with persistent OSA.

4. Discussion

Adenotonsillectomy remains the first‐line surgical treatment for pediatric OSA and is effective in the majority of uncomplicated cases. Previous studies have reported an average AHI reduction of 13.92 events/h and normalization of PSG parameters in 82.9% of pediatric OSA patients [29]. Yet, up to 30% of children can experience persistent OSA, defined as an AHI greater than 1, following T&A due to residual obstruction at other airway levels, such as the lingual tonsils [24]. Choosing the most effective surgical or medical option requires an intricate examination and understanding of the patient's unique structural anatomy, OSA severity, and medical history.

TORS was introduced in 2005 as a minimally invasive approach, specifically for the treatment of oropharyngeal carcinomas, offering three‐dimensional high‐definition imaging, magnification, tremor filtration, and wristed instrument articulation [30]. These advantages allowed for more precise resection while preserving surrounding anatomical structures [31]. While TORS has been widely adopted in adult head and neck surgery, including treatment of oropharyngeal carcinoma and adult OSA, its role in the pediatric population has only recently been explored [16, 32, 33]. Initial pediatric studies have demonstrated that TORS lingual tonsillectomy is feasible in selected patients, with operative times comparable to traditional approaches and adequate exposure despite challenging anatomy. Reported complications were generally self‐limited and without lasting sequelae [16, 24].

This systematic review and meta‐analysis found that TORS was associated with a significantly greater mean reduction in AHI (16.20 vs. 3.64 events/h), suggesting superior symptom resolution. This finding aligns with prior pediatric TORS studies, which observed that children undergoing TORS base of tongue (BOT) reduction and lingual tonsillectomy experienced a mean obstructive AHI decrease from 27.1 to 10.9, an average reduction of 16.1 events/h, with all patients achieving at least 50% improvement [24]. Similarly, a case report of children with trisomy 21 and lingual tonsillar hypertrophy describes marked improvements in airway patency after TORS, supporting its utility in higher‐risk and anatomically challenging pediatric populations [19].

Several mechanisms likely explain why TORS achieves greater AHI reductions than traditional lingual tonsillectomy. First, its superior visualization and precision facilitate more complete resection of obstructive tissue, even in deep and narrow anatomical regions [16, 30]. Second, the stability and dexterity of robotic arms allow for safe dissection at multiple depths, which is particularly advantageous in multilevel obstructions common in persistent OSA. Third, by improving surgical access, TORS enables more thorough clearance of the BOT region, which may otherwise remain partially untreated with conventional approaches [34, 35].

Despite promising surgical advances offered by the TORS approach, consistent with other studies, our analysis also revealed higher rates of postoperative complications, including dysphagia, infection, and bleeding. Postoperative bleeding is the most common postoperative complication following TORS with neck dissection for treatment of oropharyngeal carcinoma, often requiring readmission and, in some cases, even leading to death [36, 37]. A retrospective survey of 45 respondent TORS‐trained physicians in the United States reported 6 postoperative bleeding‐related deaths among 2015 procedures within 30 days of TORS surgery [38]. The evidence of a greater bleeding rate in TORS cases in our analysis is consistent with individual pediatric reports describing postoperative hemorrhage requiring intervention following TORS [16, 24]. Importantly, no statistically significant differences were observed between TORS and traditional techniques for postoperative airway obstruction or bleeding requiring surgical intervention. This suggests that while TORS may increase some complication risks, these events are generally manageable and not associated with worse airway safety outcomes.

Despite its technical advantages and demonstrated improvements in AHI, TORS has not yet achieved widespread adoption in pediatric OSA surgery. Pediatric TORS remains technically challenging, especially in patients with trisomy 21, who often have anatomical variations such as macroglossia, lingual hypertrophy, and a more narrowed airway, which increase their risk of persistent OSA [19].

Beyond complication rates, there are substantial practical barriers to TORS adoption in pediatric OSA. The da Vinci robot system has an acquisition cost exceeding $1–$2 million, with annual maintenance fees and per‐procedure instrument costs that add considerably to institutional expenditure [39]. The financial demand places TORS out of reach for many centers, particularly community hospitals and lower‐resource settings. Furthermore, proficiency in pediatric TORS requires specialized training beyond standard otolaryngology residency, including fellowship experience and a robotic surgery learning curve that has been estimated at 20–50 cases before achieving consistent operative efficiency as measured by margin status and operative times [40]. Even at high‐volume academic centers, the pediatric airway presents unique anatomical constraints. Limited oropharyngeal working space, smaller instrument clearance, and more complex airway management compound the technical difficulty relative to adult TORS [16]. However, implementing the use of smaller instruments and a 12‐mm endoscope, combined with retractors, can improve visualization to facilitate successful TORS lingual tonsillectomy [16].

In addition, the pediatric evidence base is limited, with our review identifying only 25 total TORS cases from four studies, most of which were retrospective and often combined with other airway procedures. When these institutional, financial, and training burdens are weighed against the current evidence, which is limited to 25 pediatric TORS cases, higher rates of dysphagia, infection, and bleeding, the cumulative disadvantages do not yet justify routine adoption of TORS over traditional lingual tonsillectomy in pediatric OSA. Until prospective multicenter data establish a more favorable risk–benefit profile and robotic platforms become more widely accessible, traditional lingual tonsillectomy remains the preferred surgical approach for persistent pediatric OSA.

In the context of existing pediatric OSA literature, our results add comparative outcome data for TORS and traditional lingual tonsillectomy, showing greater AHI reduction with TORS but higher short‐term complication rates. Clinically, this highlights the need to weigh potential symptom improvement against the risk of transient morbidity when selecting a surgical approach. Future studies should prioritize prospective, multicenter designs to refine patient selection, evaluate long‐term functional outcomes, and explore strategies to reduce complications.

4.1. Limitations

Several limitations should be considered when interpreting our findings. The number of pediatric TORS cases reported in the literature remains small, with only 25 patients identified across four studies, limiting the statistical power of subgroup comparisons. Most included studies were retrospective and single‐center, introducing potential selection bias and limiting generalizability. Variability in surgical technique, surgeon experience, and institutional resources may also influence reported outcomes. Concurrent airway procedures were performed in several included studies, such as midline posterior glossectomy and tongue base reduction, which could confound the observed differences in AHI reduction and complication rates. Furthermore, insufficient data were available to stratify results by important patient factors, including obesity, craniofacial anomalies, and syndromic diagnoses such as trisomy 21, which significantly affect OSA severity and surgical outcomes, thereby serving as another confounding variable in our study.

5. Conclusion

While adenotonsillectomy remains the cornerstone of surgical management for pediatric OSA, a significant population continues to face persistent disease following the surgery. Traditional lingual tonsillectomy has historically been utilized in this setting despite its limitations in visualization and access. This review introduces a novel comparative evaluation between TORS and traditional lingual tonsillectomy techniques. While our findings demonstrate that TORS is associated with greater reductions in AHI compared to traditional methods, its short‐term complication rates, technical demands, and limited experience in pediatric populations have slowed its widespread adoption. The incorporation of robotic surgery in pediatric patients demonstrates exciting potential; however, it requires continued investigation. Future studies are needed to clarify the specific role of lingual tonsillectomy in pediatric OSA management. Presently, traditional techniques remain the standard treatment approach.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Dabbas M., Zhu C., Clementi E., Sampat S., Mathew K., and Harley E., “Transoral Robotic Surgery Versus Traditional Lingual Tonsillectomy for Persistent Pediatric Sleep Apnea,” Laryngoscope Investigative Otolaryngology 11, no. 4 (2026): e70491, 10.1002/lio2.70491.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.