Does Palatoplasty Technique Impact Resolution of Eustachian Tube Dysfunction?

Abstract

Objective

To determine whether palatoplasty technique affects the resolution of eustachian tube dysfunction and postoperative speech outcomes in children with cleft palate (CP).

Design

Retrospective cohort.

Setting

Multidisciplinary cleft and craniofacial clinic at a tertiary care center.

Patients

Seventy-three patients with nonsyndromic CP who underwent palatoplasty between 2005 and 2019. Inclusion criteria included soft palate repair with Furlow technique or intravelar veloplasty (IVV) and Veau classifications I-III.

Interventions

Either Furlow or IVV repair was performed based on the surgeon's clinical judgment. All patients had bilateral ear tubes placed prior to or at the time of palate repair, with postpalatoplasty ear tubes placed at the ENT surgeon's discretion. Patients received routine follow-up care for over 5 years. Data were analyzed with Wilcoxon tests, χ² tests, and negative binomial regression.

Main Outcome Measures

Number of postpalatoplasty ear tubes placed, rates of velopharyngeal insufficiency, and speech surgery after palatoplasty in each group.

Results

Furlow repair patients required a similar number of postpalatoplasty ear tubes (P = .321) and underwent additional sets at similar rates compared to those who underwent IVV repair. Negative binomial regression found no covariates (age, race, Veau, repair type, speech surgery, fistula repair) that predicted additional ear tube requirements.

Conclusion

Furlow repair patients required postpalatoplasty ear tubes at a similar rate compared to IVV repair patients. While the palatoplasty techniques differ, patients may still need the same amount of time for resolution of their eustachian tube dysfunction.

Introduction

Children with cleft palate (CP) have an increased risk of otitis media with effusion (OME), with an incidence of up to 90% to 100%. ¹ Cleft palate is characterized by a failure of palatal fusion during development that leads to abnormal insertion sites of the tensor and levator veli palatini muscles, which are synergistically responsible for opening the eustachian tube (ET). ¹ In cases of CP, the tensor veli palatini inserts into the posterior aspect of the hard palate instead of the soft palate aponeurosis and is unable to course around the pterygoid hamulus and function as a muscular sling. The levator veli palatini, which also typically inserts on the palatine aponeurosis, inserts into the cleft margin and the posterior aspect of the hard palate. ² These aberrant attachments lead to abnormal contraction of these structures, resulting in hindered ET dilation or causing ET obstruction. ¹

As a result of these anatomical differences, children with CP are more likely to have ET dysfunction (ETD) and are subsequently at risk of frequent episodes of OME due to disturbed ventilation of the middle ear. Otitis media with effusion can lead to severe sequelae, such as tympanic membrane retraction, ossicular defects, and middle ear cholesteatoma. ³ Persistent OME can result in conductive hearing loss in the long term and, as a result, delays in speech and behavioral development ⁴ with the potential to lead to psychosocial issues.^5,6 Cleft palate repair alone is not always sufficient to restore the ventilation function of the middle ear, as evidenced by persistent rates of OME in patients who have undergone palatoplasty.^7,8 Tympanic tube placement performed early or concurrently with palate repair is effective in preventing the development of OME in patients with CP.^9,10 The utilization of ear tubes to correct ETD in CP patients thus provides a means for this intervention to serve as a measure of ETD severity in the case of subsequent required placements.

The extent of the cleft defect is critical to define when considering a patient's comorbidities and risk of future sequelae. The Veau classification system allows for a standardized characterization of CP anatomy: Veau I CP involves only the soft palate; Veau II involves both soft and secondary hard palate but spares the primary hard palate; Veau III involves the soft and entire hard palate, extending unilaterally to the alveolar ridge; and Veau IV involves the soft and entire hard palate, extending bilaterally anterior to the incisive foramen to include the alveolus and lip. Additional patient factors, such as the etiology of CP, play an important role in assessing individualized risk for ETD. Cleft palate can present in an isolated manner or in the setting of a broader genetic syndrome. Syndromic cases of CP as in cases in Pierre Robin sequence (PRS) or Stickler syndrome present with additional craniofacial abnormalities such as micrognathia that may contribute to a patient's worsened ETD.^11,12

In addition to ETD, children with CP are at risk of developing velopharyngeal insufficiency (VPI).^13–15 In CP, abnormal attachments of the velar muscles result in dysfunction of the palatal muscle sling that is responsible for soft palate elevation during speaking and swallowing. Velopharyngeal insufficiency can also occur after palate repair due to surgical scarring of the soft palate, misalignment of the levator veli palatini muscle fibers, or the development of an oronasal fistula. ¹⁶ Mitigating the risks of VPI is critical to reduce potential sequelae of significant speech impairments.

Various palatoplasty techniques have been described in the literature, including 2 of the most commonly used approaches: intravelar veloplasty (IVV) and Furlow double-opposing Z-palatoplasty. ¹⁷ While each technique addresses the primary goal of soft palate repair, they differ in their approaches and outcomes. Intravelar veloplasty encompasses 2 methods: 2-flap palatoplasty and the V-Y pushback palatoplasty. The 2-flap technique involves raising bilateral mucoperiosteal flaps, allowing for primary closure without tension, while V-Y pushback employs a V-shaped incision followed by palatal tissue rearrangement to elongate the palate. ¹⁸ Intravelar veloplasty in this study refers to levator reconstruction where the abnormally attached levator veli palatini muscles are dissected off the hard palate and realigned to restore the typical anatomy of the muscular sling. In contrast to IVV, Furlow Z-palatoplasty is characterized by a distinctive Z-shaped incision in the soft palate which allows for reorientation of the palatal musculature, facilitating soft palate expansion and improving velopharyngeal function.^17,18

There is no consensus in the literature regarding the impact of different palatoplasty techniques on improving OME outcomes in children with CP.^3,19,20 This study aims to assess the effect of palatoplasty technique on postoperative ETD improvement. We hypothesize that patients who undergo Furlow repair will require fewer sets of postpalatoplasty ear tubes compared to those who undergo IVV repair.

Methods

Study Design and Patient Selection

A retrospective cohort study was conducted using data from our institutional CP database registry under Institutional Review Board approval. Screening for eligibility of patients diagnosed with CP who underwent palatoplasty from 2005 to 2019 at our institution was conducted. Inclusion criteria for our study included children with Veau classifications I, II, and III only who underwent either Furlow or IVV repair techniques. In this study, we define IVV as any patient who underwent 2-flap palatoplasty, V-Y pushback, or a combination of the 2 techniques, which is consistent with other literature definitions of soft palate repair.^21,22 All patients with IVV had the musculature dissected off the hard palate as well as dissection of the musculature from the oral and nasal layers. Inclusion in the study also required ear tube placement either prior to or at the time of primary palate repair and patients seen in follow-up at least 5 years out from their initial repair date. Ear tube placement was determined by the consulted otolaryngologist in accordance with the American Academy of Otolaryngology—Head and Neck Surgery (AAO-HNS) clinical practice guidelines for tympanostomy tube placement in children. Patients were excluded if their CP repair was performed outside of our institution or performed with adjunct repair (eg, buccal flaps). All cases of syndromic CP were also excluded from the study. Any patients with documented diagnoses of a syndrome, confirmed by either genetic testing or strong clinical suspicion given the patient's constellation of features, were excluded. These syndromes include, but are not limited to: PRS, Stickler syndrome, Pallister-Killian syndrome, Apert syndrome, Dandy-Walker syndrome, Turner syndrome, and 22q11.2 deletion syndrome.

Data Collection

Data were extracted from provider notes and operative reports via manual chart review in an electronic medical record. Patient demographics, diagnostic characteristics, CP repair details, and postoperative outcomes were extracted. Patient demographics included race, ethnicity, sex, insurance type, history of prematurity, and prenatal risk factors. Diagnostic characteristics included location and timing of the CP diagnosis, newborn hearing screening data, and pertinent family history. Palatoplasty data included the patient's Veau classification, their age and weight at the time of their repair, preoperative hearing exams, repair technique, and the presence of middle ear fluid at the time of their initial ear tube placement. Multiple postoperative outcomes were extracted, including the number of postpalatoplasty ear tube sets required and sequential surgeries undergone including secondary velopharyngeal surgery, palate revision, palatal fistula repair, and adenoidectomy. Patients that had secondary velopharyngeal surgery were determined based on the assessment at our institution's monthly VPI conference. To assess candidacy for secondary velopharyngeal surgery, patients have a thorough assessment with our speech language pathologists, which includes an articulation assessment. This speech assessment is coupled with a video nasoendoscope evaluation and reviewed together as a team with multiple surgeons and multiple speech language pathologists to diagnose the patient's VPI and differentiate it from articulation disorder. Over this study's 14-year period, there were 13 speech language pathologists involved in diagnosing and managing the patients that required secondary velopharyngeal surgery. The diagnosis of VPI is not reliant on 1 speech language pathologist for each patient given our multidisciplinary VPI team conference, but given the variability in the documentation and skill level for the speech language pathologists involved there was not further analysis on the assessment of VPI and focus was just on secondary velopharyngeal surgery rates. All data were stored in a secure REDCap electronic database. ²³ The database was then used to compile the retrospective CP registry based on the aforementioned criteria.

Primary Outcomes

Two primary outcomes were considered: (1) rates of ear tube sets required following palate repair; and (2) rates of secondary velopharyngeal surgery. Postpalatoplasty ear tube requirements were defined as the number of myringotomies with tympanostomy tube placements performed after the primary palatoplasty. During this study's time period, there was not an audiologic protocol for the patient cohort, so there was inconsistent documentation and assessment of hearing tests. Since the audiologic information was inconsistent for the patients included and as it was not a primary outcome or evaluation for the study, further details on this topic were not analyzed. Our institution has since developed and authored an audiologic protocol for our patients with CP. ²⁴

Statistical Analysis

Patients were stratified by palatoplasty technique into 2 repair groups: Furlow and IVV. For descriptive statistics, categorical variables were reported as frequencies and proportions while continuous variables were reported as medians with interquartile ranges. Pearson χ² test was performed to compare categorical variables between groups, while continuous variables were compared between groups using the Wilcoxon test. A negative binomial generalized linear regression model was used to analyze the association between the primary outcome and other variables. Independent variables were selected based on variables of interest to this study (repair type, Veau classification, age at repair, fistula repair, and speech surgery) and demographics with statistical significance on our univariate models (race). All statistical analyses were performed using R Studio. Confidence intervals and P values for Veau classification in the regression model were generated by R package multicomp with adjustment for multiple comparisons. Statistical significance was determined at a level of P < .05.

Results

Patient Demographics and Diagnostics

A total of 73 patients were included in our study. Patients were stratified into those who underwent Furlow palatoplasty (n = 19) and IVV (n = 54). The patients across these 2 groups shared similar demographic characteristics regarding sex, ethnicity, history of prematurity, prenatal risk factors, and insurance status (Table 1). Approximately half of the patients were male (n = 36, 49%), and the majority were non-Hispanic (n = 62, 85%) without a history of prematurity (n = 63, 86%). Prenatal risk factors including parental drug use, parental smoking, parental alcohol consumption, maternal obesity, and gestational diabetes were present in a minority of patients in both groups. Most patients had either public insurance, including Medicaid/Medicare, (n = 34, 47%), or private insurance (n = 35, 48%). The 2 repair groups were significantly different in terms of race (P = .03), with more IVV patients identifying as Asian (n = 12, 22%) than Furlow patients (n = 1, 5%).

Table 1.

Patient Demographics and Preoperative Diagnostics Comparing Those That Had Furlow Palatoplasty Versus IVV.

Demographics	Participants (n = 73)
	Furlow	IVV	P
	(n = 19, 26%)	(n = 54, 74%)
Sex			.737
Female	47% (9)	52% (28)
Male	53% (10)	48% (26)
Race			.034
Asian	5% (1)	22% (12)
Pacific Islander	0% (0)	2% (1)
Black or African American	11% (2)	2% (1)
White	63% (12)	70% (38)
Unknown/not reported	21% (4)	4% (2)
Ethnicity			.221
Hispanic or Latino	16% (3)	13% (7)
Non-Hispanic/Latino	79% (15)	87% (47)
Unknown/not reported	5% (1)	0% (0)
History of prematurity	5% (1)	17% (9)	.214
Prenatal risk factors
Parental drug use	0% (0)	4% (2)	.395
Parental smoking	5% (1)	17% (9)	.214
Parental alcohol consumption	0% (0)	2% (1)	.551
Maternal obesity	0% (0)	2% (1)	.551
Gestational diabetes	11% (2)	2% (1)	.101
Insurance
Public (Medicare/Medicaid)	47% (9)	46% (25)	.710
Private	42% (8)	50% (27)
Other (eg, military)	5% (1)	2% (1)
None	5% (1)	2% (1)
Diagnostics
Prenatal diagnosis	5% (1)	15% (8)	.276
Newborn hearing screen performed/documented	53% (10)	48% (26)	.737
Failed newborn hearing screen	10% (1)	38% (10)	.097
Family history of cleft palate	0% (0)	6% (3)	.294

Abbreviation: IVV, intravelar veloplasty.

Approximately half of the patients in each repair group had a documented newborn hearing screen in their charts (n = 36, 49%), with 10% failing this initial hearing screen in the Furlow group and 38% failing in the IVV group (P = .097). There were no differences between the 2 repair groups regarding a family history of CP.

Cleft Palate Repair Characteristics

Patients in both the Furlow and IVV groups were similar in age and weight at the time of their CP repair, with a median age of 12 months in both groups and median weights of 9.4 kg in the Furlow group and 9.5 kg in the IVV group (Table 2). They additionally had similar preoperative hearing results, with most patients having either conductive hearing loss (n = 10, 45%) or mixed hearing loss (n = 7, 32%) among those with documented screening results available (n = 22, 30%). Out of the 7 mixed hearing loss patients, 3 of them had genetic testing that was negative and the remaining 4 did not have any concerning exam findings suggesting an underlying syndrome. Both sets of patients shared similar types of ear tubes inserted at the initial placement, with Armstrong tubes being the most common type placed at our institution among those with the tube type documented in their operative reports (n = 55, 92%). The majority of patients in both groups also had middle ear fluid present in at least 1 ear at the time of their initial tube placement (n = 71, 97%). Patients who underwent Furlow palatoplasty were more likely to have a Veau I classification of their CP (n = 15, 79%), while patients who underwent IVV were more likely to have a Veau II (n = 26, 48%) or Veau III (n = 27, 50%) classification (P < .001).

Table 2.

Cleft Palate Comparison of Preoperative and Postoperative Details.

Cleft Palate Repair	Furlow	IVV
Age at time of palate repair, months (median [IQR])	12.0 (3.5)	12.0 (3.0)	.752
Veau classification			< . 001
Veau I	79% (15)	2% (1)
Veau II	21% (4)	48% (26)
Veau III	0% (0)	50% (27)
Preop hearing results a			.874
Normal hearing	25% (2)	21% (3)
Conductive hearing loss	50% (4)	43% (6)
Mixed hearing loss	25% (2)	36% (5)
Ear tube type at first placement a
Armstrong	63% (12)	80% (43)	.152
T-tubes	11% (2)	4% (2)	.261
Paparellas	0% (0)	2% (1)	.551
Fluid in ear(s) at time of first ear tube placement	95% (18)	98% (53)	.433
Weight at time of palate repair, kilograms
(median [IQR])	9.4 (1.5)	9.5 (2.2)	.472
Number of postpalatoplasty ear tube sets			.321
0	37% (7)	30% (16)
1	37% (7)	43% (23)
2	11% (2)	22% (12)
3	16% (3)	4% (2)
4	0% (0)	2% (1)

Data are presented as incidence (frequency) or median (interquartile range). Significant P values are in bold.

^a Characteristics with a smaller sample size due to missing data.

Postpalatoplasty Ear Tube Sets

The number of postpalatoplasty ear tube sets was not statistically significant (P = .321) between the 2 repair groups (Figure 1). For patients who underwent Furlow repair, 37% required no ear tube sets after their initial placement by the time of palate repair while 37% required 1 additional set, 11% required 2 sets, and 16% required 3 sets. Among those who underwent IVV repair, 30% required 0 subsequent ear tube sets while 43% required 1 additional set, 22% required 2 sets, 4% required 3 sets, and 2% required 4 additional sets (Table 2). Table 2 also shows that there were no statistically significant differences between the groups with regard to the type of ear tube used with Armstrong (P = .152), T-tubes (P = .261), or Paparellas (P = .551). Both groups had a majority of Armstrong tubes used (63% for Furlow, 80% for IVV) at primary repair.

Figure 1.

Average postpalatoplasty ear tube sets per palatoplasty repair type.

Out of 16 patients with Veau I, 5 (31%) of them required no additional ear tube sets in the postoperative period, 6 (38%) needed 1 set of postpalatoplasty ear tubes, 3 (19%) required 2 sets, 2 (12%) needed 3 sets, and none required 4 postpalatoplasty ear tube sets. Among the 57 patients with Veau II (n = 30) and Veau III (n = 27), 18 (32%) of them required no postpalatoplasty ear tube sets, 24 (42%) needed 1 set, 11 (19%) needed 2 sets, 3 (5%) needed 3 sets, and 1 (2%) needed 4 postpalatoplasty ear tube sets.

The negative binomial regression did not reveal repair type as a significant predictor for the number of postpalatoplasty ear tube sets (95% CI = −1.028, 1.602, P = .540). Similarly, all other covariates in the regression model (repair age, Veau classification, subsequent palate fistula repair, subsequent speech surgery, White race, Asian race) were not statistically significant predictors of additional ear tube requirements (Table 3).

Table 3.

Regression Coefficients for Predicting Number of Postpalatoplasty Ear Tube Sets.

Variable			95% CI
	B	SE	Lower	Upper	Z-Value	P
Repair age	−0.010	0.017	−0.056	0.036	−0.613	.540
Veau classification II versus I	−0.238	0.489	−1.368	0.892	−0.486	.872
Veau classification III versus I	−0.452	0.554	−1.732	0.827	−0.817	.680
Veau classification II versus III	0.214	0.283	−0.439	0.868	0.758	.717
Repair type—IVV	0.287	0.485	−1.028	1.602	0.592	.554
Palatal fistula repair	−0.116	0.394	−1.840	0.951	−0.295	.768
Speech surgery	−0.114	0.326	−0.998	0.770	−0.349	.727
Asian race	0.127	0.490	−1.201	1.455	0.259	.796
White race	0.057	0.365	−0.931	1.046	0.158	.875

Abbreviation: IVV, intravelar veloplasty.

Postpalatoplasty Outcomes

The 2 surgical repair groups differed significantly in terms of rates of secondary velopharyngeal surgery by the age of 6. Patients in the IVV repair group were also more likely to require secondary velopharyngeal surgery by the age of 6 (n = 17, 31%) compared to those in the Furlow group (n = 0, 0%) (P = .005), with pharyngeal flap and Furlow being the most commonly performed secondary velopharyngeal surgeries in this group. Rate of palate fistula repair (n = 9, 17%) in the IVV group was greater than in the Furlow group (n = 0, 0%) but not statistically significant (P = .057). Lastly, rates of adenoidectomy were not significantly different between the 2 repair groups, with 26% (n = 5) of patients who underwent Furlow repair and 17% (n = 9) of those who underwent IVV requiring adenoidectomy later in childhood. All patients who had adenoidectomy had conservative removal leaving the inferior one-half of the adenoids and strictly removing adenoids covering the choana and ET orifice. No secondary speech surgeries were required after this surgery due to the adenoidectomy causing or unveiling VPI. The Furlow group's indications for adenoidectomy were for obstructive sleep apnea in combination with tonsillectomy (4/5), and for ETD (1/5) coupled with repeat ear tubes. The IVV group's indications for adenoidectomy were prophylactically prior to pharyngeal flap repair (6/9) out of precaution for obstructive sleep apnea (in combination with tonsillectomy), and for obstructive sleep apnea (3/9).

Discussion

Otitis media with effusion in children with CP can significantly impact audiologic health, speech, and social development, and appropriate middle ear ventilation is critical in mitigating the increased risks these patients face. The current study assessed the relationship between palatoplasty technique and ETD with the number of postpalatoplasty ear tube sets serving as a proxy as well as the effect of the surgical repair on speech development and other postoperative outcomes. We found no significant difference between the number of postpalatoplasty ear tube sets required in patients who underwent either Furlow or IVV palatoplasty with at least 5 years of clinical follow-up. However, rates of secondary velopharyngeal surgery were higher in patients who underwent IVV compared to Furlow repair.

Eustachian tube dysfunction in children with CP leads to conductive hearing loss even after passing a newborn hearing screen, ²⁴ necessitating earlier audiologic testing and timely ventilation of the middle ear space. While middle ear effusions occur bilaterally in unilateral cleft lip and palate supporting the practice of bilateral tube placements, conductive hearing loss, and middle ear complications have been shown to arise more on the cleft side in unilateral cleft lip and palate, ²⁵ suggesting worse ET function on the cleft side. Given the association between middle ear function and palate integrity, palatoplasty technique reasonably may impact CP repair outcomes. In this study, children who underwent IVV had a similar rate of postpalatoplasty ear tube placements compared to those who received Furlow double-opposing Z-plasty with 30% and 37%, respectively, avoiding a second set of ear tubes. Previous studies have found that children who underwent 2-flap palatoplasty had higher rates of ear tube insertions compared to those who underwent Furlow repair.^26,27 However, both of these previous studies included syndromic cases of CP, with Kitaya et al ²⁷ finding that the incidence of PRS was significantly higher in their 2-flap palatoplasty group. Patients with PRS have significantly wider clefts at the soft palate level ¹¹ with a “U-shaped” cleft which tends to more closely involve the oral and nasal cavities in contrast to the “V-shaped” cleft in patients without PRS, ²⁸ resulting in worse hearing and middle ear conditions. ²⁹ Our study's exclusion of all syndromic cases of CP attempted to limit the potential confounding effects of patient comorbidities to further elucidate the relationship between repair technique and ETD.

We aimed to assess the long-term effect of palatoplasty type on ETD resolution through focusing on the number of postpalatoplasty ear tube placements required in addition to necessitating patient follow-up of at least 5 years. Previous studies conducted on this topic did not require ear tube placement by or at the time of palate repair in their inclusion criteria. A recent meta-analysis found significant benefit of ear tube insertion when performed concurrently with palatal repair, with insufficient evidence for nonconcurrent placement for OME prevention. ⁹ Our study's focus on the number of ear tube sets needed only after this initial placement removes variables of differing initial placement times following palate repair that exists in these previous studies and may contribute to alternate findings. Additionally, follow-up for the previously referenced studies was limited to 2 or 3 years.^26,27 Other studies that found no difference between IVV and Furlow repair types were similarly limited to 2 years of follow-up data ²⁰ or included patients with short-term follow-up in aggregate longer-term follow-up data. ³⁰ This study elected to only include patients who were seen with at least 5 years of follow-up given literature has shown that it can take up to 5 to 6 years for most children with CPs to recover normal ET function after palatoplasty.^31,32

While the Furlow repair technique allows for more expansion of the soft palate and thus theoretically better nasopharyngeal closure^33,34 in addition to reorienting palatal musculature in a more natural “sling” orientiation, ³⁵ we did not observe a significantly lower rate of ear tube placements in this group compared to the IVV repair group. When we further analyzed other factors that may contribute to the number of postoperative ear tubes required, we found no predictors among the variables of age, Veau classification, repair type, subsequent palate fistula repair, subsequent speech surgery, and race. One explanation for a lack of observed difference between these repair groups may be that the muscle insertions on the ET are not altered during palatoplasty despite recreating the more natural orientation of soft palate musculature with Furlow technique, such that timing for ET maturity may remain constant regardless of the surgical approach. Kane et al ³⁶ studied the effects of pterygoid hamulus fracturing, which dislocates the tensor veli palatini muscle to minimize tension on the repair while not disrupting its ET attachments, and observed no changes in speech and otologic outcomes with this surgical maneuver. This may be further corroborated by studies that showed minimal improvement of ETD with palatoplasty repairs alone ⁷ and potentially explained by the efficacy of the initial ear tube placed at the time of repair in restoring ET function. ⁹

A major functional goal of CP repair is to promote normal speech development and production. Rates of secondary velopharyngeal surgery were lower in patients who underwent Furlow repair compared to IVV in this study. This study was conducted over 14 years with over 13 different speech language pathologists involved in the diagnosis of velopharyngeal dysfunction with varying documentation, skill, and assessments in differentiating velopharyngeal dysfunction from articulation disorders. Because of these variables, we chose to focus on just the rates of secondary velopharyngeal surgery and not delve into velopharyngeal dysfunction diagnosis assessment. Thus, would caution readers to not draw conclusions between patients’ ETD and velopharyngeal dysfunction in this study. We felt it important to include our secondary velopharyngeal surgery rates as our findings are consistent with previous research studies which demonstrate lower rates of secondary pharyngeal flap surgery with primary Furlow repair compared to other techniques.^14,37–39 We did not find a statistical difference in the rates of oronasal fistula in our 2 groups that would drive this difference in secondary velopharyngeal surgery. The differences in muscular rearrangement likely drive these observations of need for secondary velopharyngeal surgery. Intravelar veloplasty does not involve extending the soft palate to the same degree as Furlow repair. ¹⁸ Magnetic resonance imaging (MRI) studies have additionally demonstrated that the position of the levator veli palatini muscle may predict VPI. Patients with VPI following CP repair had a shorter velum and a higher incidence of levator veli palatini discontinuity on MRI, which was also found in noncleft VPI.^40,41 This anatomical anomaly may explain why Furlow repair, which repositions the muscular sling, decreases VPI rates compared to IVV.

Previous studies have found that cleft width is associated with higher VPI rates as well. ⁴² This correlates with our repair findings, given the majority of Veau I patients in our study mostly underwent Furlow repair while the majority of Veau II and Veau III patients underwent IVV, which is consistent with other literature findings. ²⁷ Furlow repair is used in patients with cleft widths that are often more narrow due to the propensity for fistula formation and reduced capacity for velar stretch when the approach is utilized for wider clefts. ³⁴ Thus, Veau I patients with less severe clefts in terms of width can be more likely to undergo Furlow repair. Our study included multiple surgeons performing CP repair and width was not standardized in the operative notes, so was not included for analysis. One other aspect to point out in our patients diagnosed with VPI is that this can also be diagnosed later in the teenage years after adenoid involution and midface growth. Considering that not all patients in this study were followed until 18 years of age, this is a limitation on the comparison between the groups regarding their VPI outcomes.

This study has several limitations. First, the palatoplasties were performed by different experienced cleft surgeons across the Otolaryngology and Plastic Surgery departments at our institution, which may have resulted in interprovider variability in practice that we were unable to capture. Additionally, the decision to pursue additional sets of ear tubes in the postpalatoplasty period may vary among ENT surgeons, despite the presence of the AAO-HNS clinical practice guidelines. The distribution of Veau classification was unequal across our 2 repair groups; however, we performed a negative binomial regression to account for this which showed that the number of postpalatoplasty ear tube sets was similar across our 2 repair groups, regardless of Veau classification. The study focused on nonsyndromic patients and not every patient had genetic testing performed, and thus a clinical judgment was made by the providers involved with their care to determine that the remainder of patients without testing did not have an underlying syndrome. This study also spanned 2 decades, and genetic testing and the array of known syndromes have increased since some of the patients may have been tested, but the authors were not able to account for these possibilities. Lastly, this study was not designed to account for hearing status or hearing outcomes as a proxy for ETD, so this aspect was not thoroughly evaluated but would be an interesting topic for future studies.

Conclusion

In this study, children who underwent Furlow palatoplasty required a similar number of postpalatoplasty ear tube sets and underwent additional sets at similar rates when compared to those who underwent IVV. While more Veau I patients received Furlow repair than Veau II and Veau III patients, these results were independent from cleft classification. Additionally, no other factors, including patient age at time of repair, race, and need for future fistula repair or speech surgery, were predictive of the number of postpalatoplasty ear tubes required over a time period of at least 5 years. The IVV group experienced higher rates of secondary velopharyngeal surgery by the age of 6 compared to the Furlow group. Case-by-case consideration of the benefits and drawbacks of each palatoplasty technique, when reasonable, is crucial prior to determining the most appropriate repair. Regular follow-up is also necessary to ensure appropriate recovery related to middle ear function. Future studies should consider a multi-institutional approach to aggregate more long-term data for further evaluation of these relationships.