The Effect of Cleft Palate Repair on Polysomnography Results

Abstract

Study Objectives:

In view of the risk that surgical repair of cleft palate may induce or worsen obstructive sleep apnea (OSA), the goal of this study was to assess presurgical and postsurgical polysomnography (PSG) results for children who underwent primary palatoplasty.

Methods:

Retrospective case-control series for children with cleft palate repair performed between January 2008 and December 2016 at a tertiary pediatric center. Children underwent PSG before and after surgery.

Results:

Sixty-four children (53.1% female) with a mean age of 2.0 ± 2.8 years (range 0.6–16.4) were included in the study. Pierre-Robin sequence was the most common comorbidity (67%). Before palatal repair, the mean obstructive apnea-hypopnea index (oAHI) was 3.4 ± 3.9 (range 0–17.9) events/h; this did not significantly change, with 5.9 ± 14.5 (range 0–105.7) events/h after surgery (P = 0.30). However, 34.4% of patients had a worsening of more than 1 obstructive event/h and 18.9% had a worsening of 5 or more obstructive events/h. The presence of a concomitant syndrome (eg, Treacher Collins) was a risk factor for postoperative OSA (odds ratio 4.2, 95% confidence interval 1.1–15.8, P = .03)

Conclusions:

OSA did not develop or worsen following primary palatoplasty. However, the oAHI increased by 5 or more events/h in approximately 20% of study participants. The presence of a syndrome was the only factor predictive of worsening OSA after palatoplasty. These findings suggest that palatoplasty does not worsen or cause OSA in most patients, and that nonsyndromic children are at low risk for the development or worsening of OSA.

Citation:

Bergeron M, Cohen AP, Maby A, Babiker HE, Pan BS, Ishman SL. The effect of cleft palate repair on polysomnography results. J Clin Sleep Med. 2019;15(11):1581–1586.

BRIEF SUMMARY

Current Knowledge/Study Rationale: Although primary cleft palate repair is routinely performed in children, there is a hypothetical risk that this procedure will worsen or induce obstructive sleep apnea (OSA). However, there are no definitive data pertaining to this topic.

Study Impact: Overall, OSA did not develop or worsen following primary palatoplasty. However, the oAHI increased by 5 or more events/h in approximately 20% of study participants. The presence of a syndrome was the only factor predictive of worsening OSA after palatoplasty.

INTRODUCTION

Cleft palate, with or without cleft lip, is a common disorder that occurs in as many as 1 in 680 live births in the United States annually.¹ Children with a cleft palate may have both functional and aesthetic defects even after cleft palate repair (palatoplasty), resulting in difficulty with feeding,² pronunciation,² social integration,³ and hearing.⁴ Given that attention is generally focused on the treatment of these conditions, baseline or resultant obstructive sleep apnea (OSA) may be overlooked.

OSA and sleep-disordered breathing (SDB) are frequently associated with the presence of congenital midface defects. Previous studies suggest that the presence of craniofacial malformations such as cleft palate increase the risk of OSA and SDB and may affect 70% of this patient population.^1,5 If left untreated, the long-term consequences of OSA may include cardiovascular disease, pulmonary hypertension, daytime sleepiness, reduced neurocognitive function, and a negative effect on quality of life.^6–12

Children with cleft palate have both structural and functional anomalies in the upper airway.¹³ Several primary palatoplasty techniques exist, such as the two-flap palatoplasty, a straight line repair, and the double-opposing Z-plasty. Although these changes may worsen upper airway obstruction, thereby inducing or worsening OSA, definitive data pertaining to the true impact of primary palatoplasty on OSA are limited.¹⁴ Furthermore, despite the fact that overnight polysomnography (PSG) is the gold standard for diagnosing OSA, few studies of cleft repair have reported PSG results but rather have reported their findings based solely on clinical symptoms.^1,5

To our knowledge, no studies have evaluated PSG data before and after primary cleft palate repair. Published reports are limited to case series, with PSGs performed either before or after palatoplasty. In addition, some of these patients were treated for correction of velopharyngeal insufficiency and not primary palate repair; thus, there is significant heterogeneity in the patients previously assessed.¹⁵ In view of this gap in knowledge, our objective was to assess changes in PSG results in these children before and after primary cleft palate repair.

METHODS

Study Design

Participants in this study included consecutive patients with unrepaired cleft palates evaluated for cleft palate repair at Cincinnati Children’s Hospital Medical Center (CCHMC) from January 2008 to December 2016 who underwent overnight in-laboratory PSGs both before and after primary cleft repair. All patients underwent primary palatoplasty under general anesthesia, with surgery performed by one of two plastic surgeons. We excluded children who had undergone adenoidectomy and/or tonsillectomy concurrent with palatoplasty or before the postsurgical PSG and those who had undergone revision surgery.

All children underwent an overnight 16-channel diagnostic PSG before and after palatoplasty in an accredited sleep center at CCHMC. Standard PSG parameters were measured, including pulse oximetry, chest movement, nasal airflow, electrocardiogram, end-tidal carbon dioxide (ETCO₂) electroencephalogram, electromyogram, and electrooculogram. All variables were recorded with a digital acquisition system. The studies were scored according to the pediatric scoring criteria of the American Academy of Sleep Medicine, using a 3% desaturation criteria¹⁴ and were read by board-certified sleep physicians. The changes in PSG parameters before and after surgery were analyzed, including the apnea-hypopnea index (AHI), obstructive AHI (oAHI), oxygen saturation nadir, total sleep time, sleep efficiency, and elevated ETCO₂. The AHI is the number of apnea and hypopnea events/h of total sleep time. The oAHI is the number of obstructive apnea and hypopnea events/h of total sleep time. For patients who underwent PSG more than once prior to surgery, the test performed closest to the date of surgery was used. The severity of OSA was defined by the oAHI: mild OSA was defined as one to fewer than 5 events/h, moderate OSA was defined as 5 to fewer than 10 events/h, and severe OSA was defined as 10 or more events/h. The oxygen saturation nadir was defined as the lowest oxygen saturation reading during an obstructive respiratory event.

Body mass index (BMI) was calculated for each patient and compared using BMI z-scores as derived from the growth standards of the Centers for Disease Control and Prevention published in 2000.¹⁶ This study was reviewed and approved by the CCHMC institutional review board.

Statistical Analysis

For demographic and PSG variables, the one- and two-sided Fisher exact tests and Wilcoxon rank-sum test were used. The chi-square test was used for categorical variables (race, sex, comorbidities). Because the results were not normally distributed, significance results were obtained with the two-sample Wilcoxon rank-sum test.

The Spearman correlation was used to evaluate the relationships between comorbidities and demographic data and PSG results. Univariable regression analysis was used to assess the relationship between baseline sleep study parameters and worsening of the oAHI, AHI, OSA severity, oxygen nadir saturation, sleep efficiency, and elevated carbon dioxide (CO₂) overnight (> 50 mmHg). Comorbidities that were still significant in backward and forward multivariate model selection were included in the multivariate model. Values of P < .05 were used to determine significance. Statistical analysis was performed using XLSTAT 9.4 (Addinsoft, Paris, France).

RESULTS

As shown in Table 1, 64 patients (34 females, 53.1%) were included in the study; 52 patients (81.3%) were white. The mean age of participants was 2.0 ± 2.8 years (range 0.6–16.1) at the time of surgery. The single most frequent comorbidity was Pierre Robin sequence (67.2%). Sixteen patients (25%) had a syndrome or craniofacial disorder other than Pierre Robin sequence (eg, Treacher Collins, CHARGE syndrome). The mean BMI percentile was 52.5 ± 33.1 (0.5–99.9). Baseline PSG revealed that 31% of patients had no OSA, 33% had mild OSA, 23% had moderate OSA, and 11% had severe OSA. Overall, there was no significant difference between sleep parameters measured before and after primary palatoplasty (Figure 1, Table 2). The median time between the pre- and post-PSG was 315.5 days [interquartile range 212.5–612.5].

Table 1.

Baseline demographic data for children before primary cleft palate repair.

Figure 1. Box plot of the obstructive apnea-hypopnea index (oAHI) before and after primary palatoplasty for children with a cleft palate (n = 64).

Table 2.

Sleep study parameters for all children (overall) undergoing primary palatoplasty and a comparison of the change in these parameters before and after surgery for those who had an increase of 5 or more events in their obstructive apnea-hypopnea index and those who did not.

Patients had an oAHI of 3.4 ± 3.9 (0–17.9) events/h before surgery and 5.9 ± 14.5 (0–106.7) events/h after surgery (P = .30). Twenty-two children (34.4%) had an oAHI increase of one or more obstructive events/h, whereas 12 (19%) had an increase of 5 or more obstructive events/h. For these 12 patients, the preoperative oAHI was 3.2 ± 3.7 events/h (0–9.4) and the postoperative oAHI was 22.5 ± 28.3 events/h (6.1–105.7, P < .001). The oxygen saturation nadir changed from 91.3 ± 3.7% prior to surgery to 79.9 ± 17.8% following surgery (P = .008). The total sleep time with CO₂ > 50 mm Hg increased from 0 ± 0.0 to 2.9 ± 6.8% (0–24.8, P = .19). This can be seen in Figure 2. Of the 26 patients with no OSA prior to surgery, 20 (76.9%) had no OSA after surgery.

Figure 2. Plot of the obstructive apnea-hypopnea index (oAHI) for the 22 children who had an increase in the oAHI of 1 or more events after primary palatoplasty.

The only factor predictive of worsening OSA after surgery in univariable regression analysis was the presence of a syndrome, which also remained significant after a multivariate regression analysis (odds ratio 4.2, 95% confidence interval 1.1–15.8, P = .03) (Table 3). There was neither no correlation between PSG improvement and age nor timing of the PSG (r = .21, P = .19). Last, because 67% of our patients had Pierre Robin sequence, we performed a subgroup analysis but found no significant association between Pierre Robin sequence and worsening of OSA after surgery.

Table 3.

Multivariable regression results of the factors predictive of worsening obstructive sleep apnea.

DISCUSSION

Overall, mean PSG sleep parameters did not change following primary palatoplasty. However, the oAHI increased by 5 or more events/h in approximately 20% of study participants after surgery. The only factor predictive of worsening OSA was the presence of a syndrome; these children were four times more likely to experience development of OSA after surgery than were nonsyndromic children. Baseline OSA is often reported as a contraindication to palatoplasty and may result in delay of cleft palate repair. However, these data suggest that palatoplasty does not worsen or cause OSA in most patients and that nonsyndromic children are at low risk for the development or worsening of OSA.

Furthermore, because syndromic children are at increased risk of the development of OSA following palatoplasty, these patients should be closely monitored with postoperative PSG. Moreover, they and their families should be made aware that medical or surgical intervention to manage OSA may be required after palatoplasty.

OSA and SDB are frequently associated with congenital midface defects. Previous studies suggest that the presence of craniofacial malformations such as cleft palate increases the risk of OSA and SDB and could affect as many as 70% of these children.^1,5 Our findings are similar to those reported in the literature,^1,5 with 60% of our patients having at least a mild degree of OSA.

One patient experienced worsening of their oAHI up to 105 events/h (Figure 2). This patient had multiple comorbidities, including Pierre-Robin sequence, velocardiofacial syndrome, and ventricular septal defect, as well as generalized hypotonia and muscle weakness. We were not able to identify a specific factor that resulted in this severe degree of obstruction. However, the generalized hypotonia might have predisposed this patient to significant postoperative obstruction after cleft repair.

Given the retrospective design of this study, it is unclear whether the surgery itself led to an increase in oAHI or whether craniofacial abnormalities associated with cleft palate might be the cause of this change. In 2015, Moraleda-Cibrian et al¹⁷ demonstrated that children with syndromic craniofacial disorders are more likely to have a baseline AHI of 5 or more events/h (33%) versus nonsyndromic children (15%). It has also been reported that patients with cleft palate are already more at risk for SDB than the general population.^13,18 For children with craniofacial deformities, airway obstruction is often multifactorial.¹ The development of OSA has been attributed to morphologic changes that result in a small midface and retrognathic mandible as well as the presence of nasal deformities.¹³

Our study has several limitations. First, this is a retrospective study thus introducing selection bias. In addition, preoperative PSG was not performed for all children with cleft palate prior to palatoplasty, but only for those in whom there was a clinical suspicion of OSA. Thus, we have no information regarding those patients with cleft palate who did not undergo PSG; this likely resulted in selection bias where children with mild or asymptomatic OSA were underrepresented. Moreover, only the last PSG results prior to surgery was assessed and thus we cannot comment on the trends in OSA prior to palate closure. In addition, we only assessed the first PSG results after closure and thus the long-term effect of cleft palate repair on OSA is also unknown. Furthermore, we excluded patients who had a procedure to address their OSA between PSGs or at the time of the palatoplasty. These children likely had severe baseline OSA. In addition, this cohort was predominantly white, thus limiting external validity. Last, PSG was not always performed shortly before or after the procedure. Even if we did not note a correlation between time of the PSG and OSA change, it is possible that “watchful waiting” in some of these patients influenced the results.

CONCLUSION

Overall, OSA did not develop or worsen following primary palatoplasty. However, the oAHI increased by 5 or more events/h in approximately 20% of study participants. The presence of a syndrome was the only factor predictive of worsening OSA after palatoplasty. These data suggest that palatoplasty does not worsen or cause OSA in most patients and that nonsyndromic children are at low risk for the development or worsening of OSA.

DISCLOSURE STATEMENT

All authors have read and approved this manuscript. This study was a podium presentation at the SENTAC 2017 Annual Meeting in Toronto, Ontario, Canada on December 2, 2017. The authors report no conflicts of interest.

ABBREVIATIONS

AHI

apnea-hypopnea index

BMI

body mass index

CO₂

carbon dioxide

CCHMC

Cincinnati Children’s Hospital Medical Center

ETCO₂

end-tidal carbon dioxide

oAHI

obstructive apnea-hypopnea index

OSA

obstructive sleep apnea

PSG

polysomnography

SDB

sleep-disordered breathing