Introduction
The cleft palate is the most common craniofacial malformation. The treatment of this condition is surgical. The aim of the surgical procedure to repair a palatal cleft is to restore the function of the velopharyngeal sphincter during speech. That is, correcting the inherent velopharyngeal insufficiency (VPI) which occurs because of the cleft palate1–3.
Cleft palate repair provides a functional velopharyngeal sphincter in about 70-80% of the cases. However, 20-30% of patients persist with VPI following the surgical procedure1,4,5. These patients require a secondary surgical procedure to correct persistent or residual VPI6,7.
The American Cleft Palate Association guidelines for performing surgery for correcting persisting VPI following palatal repair recommends the performance of imaging before surgical treatment of VPI8. The guidelines describe only videonasofaringoscopia (VNP) as the imaging procedure. However, other reports support the use of a combination of VNP and videofluoroscopy (VF)1,9,10.
The combination of VNP and VF is the best approach for the evaluation of the velopharyngeal sphincter (VPS) during speech11. Multiple reports describe that the best success rates for correcting residual VPI are achieved when the surgical procedure is customized according to imaging findings1,12,13.
VNP provides an in vivo examination of the (VPS) during speech. However, young children do not always tolerate this procedure well enough to complete all the necessary speech tasks with the best articulation placement possible13. In contrast, most young children comply with the performance of VF without difficulty. The craniofacial team should not consider these two procedures as options but as a combination for an optimal assessment of the VPS during speech. The rationale is that each procedure provides useful but different data. As mentioned herein, VNP provides direct in vivo examination of the VPS, but there are limitations. VNP provides only a 2-dimensional view. Furthermore, it cannot provide actual-size measurements of the structures of the sphincter and their movements6,13. VF provides imaging of the sphincter from different views allowing a 3-dimensional assessment of the structures and their movements. Furthermore, VF does provide actual-size measurements1,12–14.
Although young children tolerate VF better than VNP, it does not provide an accurate assessment of adenoid and palatine tonsils. The assessments of these structures are essential for the planning of surgical procedures for correcting VPI. Furthermore, VNP allows a complete examination of the entire vocal tract, not only the VPS12,13.
VF involves the use of X-rays which has created controversy in medical literature12,15,16. However, following specifically designed protocols can decrease the risks of radiation effectively. Moreover, young children can undergo VF as outpatients (OPs) without any sedation12,16.
Young children can also undergo VNP as an OP procedure, but they do not necessarily comply with allowing a complete examination of the whole vocal tract during speech12,17,18.
The purpose of this paper is to report the preliminary results of customizing the surgical procedure for correcting VPI in patients under 10 years of age by performing VF as an OP and performing VNP in the operating room (OR), immediately before the surgery, under light sedation.
Material and methods
The Internal Review Board of the Corewell William Beaumont University Hospital in Royal Oak, Michigan, USA, approved this research protocol.
The protocol indicated recruiting all patients with residual VPI following palatal repair for the study group. Patients included in the group had to meet the following inclusion criteria: (a) unilateral or bilateral cleft lip and palate or isolated cleft palate; (b) primary palate and lip repair during the 1st month of life; (c) secondary palate repair before 6 months of age per the protocol established in the hospital11,12,17,18; (d) residual or persistent VPI following palatal repair as demonstrated by speech pathology evaluation and nasometry19; (e) patients undergoing pharyngeal flap (PF) surgery for correcting VPI; (f) PFs customized individually according to imaging findings as described previously11,12,17,18; (g) the patients included in the study group should be ≤10 years of age.
The protocol also indicated the following exclusion criteria: (a) patients with post-operative fistula; (b) patients with previous secondary surgical procedures for VPI; (c) syndromic patients; (d) patients with concomitant neurologic disorders or severe language impairments.
VF was performed according to the previously reported protocol of the craniofacial clinic as an OP procedure12. For the adequate performance of VF, the patient had to be able to repeat a speech sample with adequate articulation placement at least during the articulation of short words. All patients underwent a speech and language pathology (SLP) evaluation before the procedure. Patients with consistent compensatory articulation patterns (e.g., glottal stops, nasal fricatives, etc.) underwent a SLP intervention before the VF12.
VNP was performed in the OR under light sedation. It should be noted that the patient should be able to repeat the speech sample used in the craniofacial clinic with adequate articulation placement. The protocol for the analysis of velopharyngeal function during speech by VNP at the craniofacial clinic was published previously11,17.
From January 2013 to July 2024, patients meeting the inclusion criteria were included in the study group.
A total of 74 patients underwent PF surgery according to imaging findings during this period. Only 59 patients completed pre-operative and post-operative evaluations.
The sample size was calculated according to success rates for correcting residual VPI at the craniofacial clinic in the 5 years before the start of the recruiting process for the protocol of PF surgery according to OP VF and VNP in the OR. A 95% interval and a power of 80% were selected for the calculation. A minimum of seventy patients should have complete pre-operative and post-operative evaluations to provide reliable results. The project is still ongoing.
To perform the VNP in the OR, the patients were sedated by Midazolam administered orally at a dose of 0.05 mg/kg in the pre-operative unit. The patient was taken to the OR and nitrous oxide was administered by an anesthesia mask.
The patient was in a semi seated position on the surgical bed, already strapped and held by 2 OR surgical nurses.
One of the nurses held the arms and the other one held the head tightly to avoid any movement.
The patient had to be able to repeat the speech sample with adequate articulation placement of most plosive or fricative phonemes. Inconsistent but not frequent compensatory articulation patterns were acceptable during the repetition of the speech sample. A pediatric flexible scope with 2.4 mm at the tip was used in all cases.
The VNP was recorded with sound.
After the VNP had been completed the general anesthesia induction was started by the anesthesiologist with sevoflurane through an anesthesia mask. Simultaneously, an IV line was placed, and IV anesthesia induction was performed. The patient was intubated and the Dingman mouth gag was placed.
The previous VF and the VNP performed in the OR were analyzed by the attending surgeon and the coordinator of SLP of the craniofacial clinic. The protocol for assessing the function of the VPS during the speech was described in detail in previous reports11,12,17.
All PFs were performed by the four craniofacial surgeons of the craniofacial clinic. The surgical techniques for performing the PFs were slightly different depending on the surgeon. Three surgeons split the palate following the technique. One did not. The surgical techniques have been described in detail in previous reports11,17.
All patients underwent a post-operative evaluation 2 months following the surgical procedure including a SLP evaluation and nasometry. Only cases with persistent data of VPI following PF surgery underwent further VNP. VF was performed only in cases undergoing another surgical procedure to correct persistent VPI.
Results
The ages of the 59 patients included in the study group ranged from 4 years and 5 months to 9 years and 11 months. The median age was 5 years and 4 months. Thirty-nine patients were males, and 20 patients were females.
VF was appropriately completed as OP in all 59 cases. The mean total radiation time was 2.80 mSieverts (mSv). The standard deviation (SD) was 1.25 mSv. The range was 1.4-9.45 mSV.
When the adenoid pad was covering ≥ 50% of the airway (≥ 2 +/ 4) as observed by the lateral view of the VF, adenoidectomy was scheduled. Adenoidectomy was performed using Coblator II as reported previously11,12. It should be emphasized that all adenoid lymphoid tissue had to be completely removed. Fifty patients (84%) underwent adenoidectomy in preparation for PF surgery.
Palatine tonsils were assessed by direct intraoral vision and by the frontal view of the VF. When tonsils were classified ≥ 3 +/ 4, tonsillectomy was performed in the same surgical stage as adenoidectomy (tonsillectomy + adenoidectomy = T & A). Forty out of the 50 patients with adenoid hyperplasia underwent T & A (80%).
PF surgery was scheduled at least 4 months but not more than 8 months following adenoidectomy or T & A. VNP was appropriately performed in the OR in all cases. No cases of lymphoid tissue remnants or regrowth were demonstrated by VNP.
Table 1 displays the number of patients operated on by each of the surgeons participating in this study (Table 1). There were no complications during the surgical procedure in any of the cases.
Table 1. Cases operated on by the surgeons participating in this study
| Surgeon | No. cases |
|---|---|
| 1 | 12 |
| 2 | 15 |
| 3 | 22 |
| 4 | 10 |
| Total | 59 |
Pre-operative nasometry demonstrated mean nasalance (MN) above mean + 2 SD per the normative study of the craniofacial clinic in all cases (Table 2). MN ranged from 55% to 90%. The mean MN was 59%; SD = 14%. All patients demonstrated inconsistent nasal emission by perceptual assessment. Nasal emission was not phoneme-specific in any of the cases. Fifty-seven cases presented with moderate hypernasality. Two of the cases were classified as severe hypernasality by SLP evaluation. The scale reported by the interdisciplinary group for the evaluation of universal parameters for reporting speech outcomes in individuals with cleft palate20 was used for the resonance assessment.
Table 2. Nasometry normative data
| *Normative data (per a normative study performed 2023) at the Speech and Voice Laboratory. Department of Speech & Language Pathology. Corewell Health William Beaumont University Hospital | |
| Males | |
| X = 26.71 | |
| SD = 4.15 | |
| Ref = 18-35% | |
| Females | |
| X = 29.50 | |
| X = 29.50 | |
| Ref = 23-35% | |
|
Sample: Rainbow Passage |
|
Two months postoperatively, 54 patients (91.5%) demonstrated normal nasal resonance, absence of nasal emission, and MN within normal limits (Table 2). In contrast, five patients were persistent with signs of VPI including MN above normal limits, inconsistent nasal emission, and moderate hypernasality. It should be noted that the two cases with severe hypernasality did not demonstrate signs of VPI postoperatively. The distribution of persistent VPI postoperatively as stratified by a surgeon is displayed in table 3. A Chi-square test demonstrated that there was no significant dependence of post-operative VPI between surgeons (p < 0.05). None of the patients reported data suggestive of sleep-disordered breathing 2 months following PF surgery.
Table 3. Post-operative outcomes
| Surgeon | Corrected VPI | Persistent VPI |
|---|---|---|
| 1 | 11 | 1 |
| 2 | 14 | 1 |
| 3 | 20 | 2 |
| 4 | 9 | 1 |
| Total | 54 | 5 |
|
VPI: velopharyngeal insufficiency. |
||
Discussion
Previous reports demonstrated that performing PF surgery according to findings of VNP and VF provides the best data for planning surgery for VPI1,6–8,14. Furthermore, customizing PFs according to imaging findings yields a success rate of over 90%11,14,17.
The present report of preliminary results of performing VF as an outpatient, and performing VNP in the OR, suggests that this is a safe and reliable protocol for planning and performing PF surgery in children below 10 years of age. The success rate following this protocol (91.5%) is not significantly different as compared to the best success rates previously reported in the literature (81-97%)14,17.
VF has been reported as well tolerated in children below 10 years of age. The compliance for this procedure including adequate repetition of a standardized speech sample is 100%12. The controversy concerning the use of radiation seems to have been resolved when adequate standardized protocols are being followed, reducing the total radiation dosage to non-risk values12,16.
VNP is not always tolerated well in children below 10 years of age21. The procedure is uncomfortable despite the use of a nasal vasoconstrictor and topical anesthetic. All the patients including those in the study group for this preliminary report were able to repeat the speech sample in the OR under light sedation. In addition, a post-operative interview demonstrated that none of the patients remembered the procedure.
There are reports describing the use of magnetic resonance imaging (MRI) as an imaging procedure for analyzing the velopharyngeal sphincter in cases of VPI. MRI offers the advantage of not using radiation. Furthermore, MRI has a higher rate of patient compliance as compared to VNP. However, currently, there are multiple challenges associated with the clinical use of MRI for the assessment of the velopharyngeal sphincter. It can take significant time to implement standardized protocols, streamline the MRI ordering process, and secure scanner time. Only sustained phonation is elicited during the MRI, consequently, repetition of an adequate phonetically standardized speech sample cannot be used. In addition, a sustained phonation which is usually a sustained/e/phonation, does not accurately reflect velopharyngeal function during speech. Further research is needed before MRI can become the diagnostic tool of choice for assessing velopharyngeal function in cases of VPI21. Planning the surgical procedure and customizing the surgical technique to each individual imaging finding is the optimal approach for correcting VPI.
Four different surgeons with slightly different surgical techniques for performing PF participated in this study. Despite these differences, there was a non-significant dependence of the post-operative success rate between surgeons and techniques.
In the five patients who persisted with signs of VPI postoperatively, the most common findings were inferior migration of the flap, shrinking, lateral dehiscence, or total detachment. Only one case demonstrated total detachment of the flap. This patient presented with a severe viral infection of the upper respiratory tract 24 h postoperatively. The patient had to stay in the hospital for 10 more days with IV fluids and respiratory therapy. The patient presented with intense coughing fits during the hospital stay. Besides, the total detachment of the flap and a small dehiscence of the repaired palatal split was also detected. Inferior migration or shrinking of the flap can occur by technical error, but these post-operative complications can be secondary to the healing process of the donor area.
There are limitations to this study. First, this is a preliminary report as a minimal number of cases per speech sample calculation (70 cases) have not been reached yet. Because of the strict inclusion criteria, recruitment of patients for this protocol takes time. The project is still ongoing, and results will be available to be submitted soon.
Only patients operated on using a customized PF have been included in the study group. Other surgical techniques for correcting residual VPI were not studied including sphincter pharyngoplasty, secondary Furlow’s “Z” palatoplasty, fat grafting, or synthetic filler injection. Although different options should be considered in each case, the most crucial factor is to select and perform the surgical technique according to a complete clinical history, and imaging findings as provided by VF and VNP17,22.
Conclusion
From the preliminary results of this study it can be assumed that performing VF as OP, and VNP in the OR under light sedation are safe and reliable procedures for planning and performing pharyngeal flap surgery with optimal success rate.
VF and VNP provide complementary information for evaluating VPI and planning corrective PF. VF must be performed preoperatively as is well tolerated in the outpatient setting. On the other hand, VNP can be uncomfortable and is best performed under light sedation in the OR. VF allows the surgeon to measure the needed dimensions-size of the PF. VNP allows the surgeon to determine the optimal position of the flap in the posterior wall of the nasopharynx which can be factored into the operative plan just before beginning the surgical procedure. The findings here suggest that these standardized protocols for measurement and placement of pharyngeal flaps obviate minor differences in surgical techniques, allowing for a high rate of success.
Funding
The authors declare that they have not received funding.
Conflicts of interest
The authors declare no conflicts of interest.
Ethical considerations
Protection of humans and animals. The authors declare that no experiments involving humans or animals were conducted for this research.
Confidentiality, informed consent, and ethical approval. The study does not involve patient personal data nor requires ethical approval. The SAGER guidelines do not apply.
Declaration on the use of artificial intelligence. The authors declare that no generative artificial intelligence was used in the writing of this manuscript.
