With the advances in the understanding and treatment of cleft palate patients, our center currently achieves velar competence in 93.3% of cases, with 6.7% requiring a secondary procedure to treat velopharyngeal insufficiency (VPI).1 Where speech therapy, obturators, and posterior pharyngeal augmentation have been credited with some improvement in VPI, secondary pharyngeal procedures, pharyngeal flap or sphincterpharyngoplasty remain the mainstay treatments for VPI.
As we gained experience with nasoendoscopy as our dynamic study of choice for patients with VPI, we identified a group of patients with good pharyngeal and soft palate mobility with an isolated small (1–3 mm) closure defect resulting in an audible nasal air escape (ANAE) and turbulence during speech. We considered the risks of a sphincterpharyngoplasty too high for the potential benefit. Stimulated by the resurgence of fat grafting resulting from Coleman’s technique2 and based on our new understanding of these patients velopharyngeal (VP) anatomy, we hypothesized that fat grafting the palate alone in patients with an isolated small gap on active nasopharyngeal closure could improve speech outcomes without the risks of a potentially obstructive pharyngeal procedure.
After Institutional Review Board approval, a retrospective chart review from June 2010 to August 2013 identified 46 patients with VPI that were treated with autologous fat grafting to the soft palate at Children’s of Alabama, University of Alabama in Birmingham. Inclusion criteria were a history of optimal cleft palate repair with intravelar veloplasty,1 subjective and nasoendoscopic evidence of VPI, soft palate elevation and pharyngeal wall contraction on speech leaving only a small persistent VP closure gap.
Our standard procedure for cleft palate repair has been previously described.1
Cases referred to us after the initial cleft repair elsewhere had the soft palate re-repair optimized by an intravelar veloplasty. When speech evaluation after paltoplasty detected ANAE or hypernasal speech despite adequate speech therapy, awake fiberoptic nasoendoscopy was performed to document the anatomic site of the nasal air escape. In patients with otherwise good palatal motion, verified transverse levator veli palatini orientation and a small VP closure defect, autologous fat transfer was offered as a biologic obturator.
Surgical Technique: All cases of fat grafting were performed by a single surgeon (J.H.G.) under general anesthesia and after 1 dose of antibiotic prophylaxis. Autologous fat was harvested in the supine position either from the periumbilical or gluteal region. A 3.7-mm Tulip cannula and a 60-mL Luer-Lok syringe was used to obtain the fat under manual suction (Fig. 1). The graft was processed by sequential washes with injectable saline using hand centrifugation to separate the aspirate into the 3 phases as described by Coleman,2 a supernatant oil phase, the central layer composed of fat, and decanting the denser layer composed of blood and water. After 2 to 3 wash cycles, the fat was considered ready for grafting based on its color, washing until there is no more blood in the collection and only yellow fat, then injected through the syringe into a container for further processing. The oil phase was also discarded from the fat to be injected. The fat was then passed through an 18-gauge needle and any tissue, such as dermis, obtained with the harvesting was discarded. When the processed fat would easily pass through the 18-gauge needle, the fat was aspirated into 1-mL syringes.
The volume of fat to be harvested and location to be injected were previously decided based on awake nasoendoscopy findings. We estimated 1 mL of fat per mm of closure deficit.
A Dingman mouth gag retractor was placed providing direct visualization of the soft palate. Simultaneous fiber optic nasoendoscopy allowed visualization of the nasal surface of the soft palate.
The injections were performed trans-orally with a 17-gauge Touhey needle entering the oral mucosa of the soft palate, and the fat was injected between the muscle layer and the nasal mucosa of the soft palate at the predetermined VP defect location (Fig. 2). The needle puncture site was closed with a purse-string stitch with absorbable suture and the fat final location recorded (Fig. 3). The first few cases were observed overnight for airway control. Because no events occurred in the first 6 months, the latter part of the study was treated as outpatients.
Preoperative and postoperative speech evaluation was performed by a multidisciplinary team, including a plastic surgeon, an ENT surgeon, and a speech pathologist. Measured outcomes were patients’ demographic data, speech improvement based both subjectively and objectively with Pittsburgh Weighted Speech Scale (PWSS), number of patients requiring a second graft or a secondary procedure, time interval between the initial graft and the decision for a second graft, closure deficit on nasoendoscopy, volume of fat injected, local and donor site complications, operative times, bleeding, infection, and airway obstruction.
Nominal data were analyzed with descriptive statistics.
A total of 61 autologous fat grafting procedures were performed in 46 patients. The average age of the study population is 5.59 ± 2.05 years. Eight cases (17.39%) had the cleft palate associated with a syndrome or sequence (Stickler in 1, DiGeorge in 3, Pierre Robin in 2, caudal regression syndrome in 1, and suspected but not confirmed genetic mutation in 1), the remaining 38 (82.61%) were considered by our team geneticist to have isolated clefts. The majority of cases had the initial palate repair performed at our institution (36/46 or 73.9%), the remaining 12 cases (26.1%) had the primary repair performed elsewhere and a secondary palatoplasty with intravelar veloplasty performed by our senior author prior to consideration for fat grafting.
Average follow-up was 12.6 ± 9 months observed in 42 patients, the remaining 4 patients in the study were lost to follow-up after the first postoperative visit.
The majority of patients underwent a single procedure (32/46 or 69.6%), 13 of the 46 patients (28.2%) had 2 fat grafting procedures and only 1 patient (2.2%) had 3 fat grafting procedures. The average time interval after each procedure before offering a second procedure was 13.45 ± 4.34 months. This time was to allow patients to undergo appropriate time for speech therapy before evaluation for regrafting.
Early in the study, the periumbilical region (22/46 or 47.8%) was the donor site of preference. We noted a prominent oil phase we feared could result in increased inflammation and converted to harvest from the gluteal region. Therefore, the later part of the study had the fat harvested from the gluteal region (20/46 or 43.5%). Another 4 patients had the fat harvested from the periumbilical region on the first procedure but subsequentially obtained from the gluteal region (4/46 or 8.7%).
Nasoendoscopic evaluations were described as a numeric or nominal estimate of millimeter of closure deficit. Of all patient undergoing fat grafting for the first time, closure defects between 1 and 5 mm were described in 26 of 46 patients (56.52%), averaging 2.96 ± 1.03 mm. The remaining patients had a non-numeric description of the closure defect. The volume of fat grafted on the first procedure group was recorded in 34 of 46 (73.91%) patients averaging 2.4 ± 1.1 mL. On the group of patients undergoing a second round of fat grafting, a closure defect between 2 and 3 mm was described in 8 of 14 patients (57.14%), with an average of 2.12 ± 0.35. Fat graft volume on this group was recorded in 7 of 14 procedures (50%) and averaged 2.15 ± 0.66 mL.
The fat was injected primarily in the soft palate. Of the 46 patients, in 40 patients (86.95%), the graft remained in the soft palate, in 5 patients (10.86%), the fat extruded into the palatoglossal pillar and in 1 particular case, fat was injected in the soft palate and posterior pharynx at Passavant ridge. This case represents a cleft palate patient that only presented with VPI after adenoidectomy was performed. Grafting of the posterior pharynx was undertaken to replace some of the volume resected by the surgical procedure.
A subjective assessment of the procedure results was recorded in 41 patients (89.13%). Of those, 28 of 41 patients (68.29%) had a significant improvement in hypernasality, 9 of 41 patients (21.95%) had no benefit from the operation, and 4 of 41cases (9.75%) have had a single postoperative visit 1 month after the procedure. These 4 cases were considered subjectively improved on this initial postoperative visit, but then lost to follow-up.
Of the 46 patients in the study, 38 (82.6%) had a preoperative and postoperative PWSS assessments. The average preoperative score was 8.21 ± 4.38 compared to 5.21 ± 3.19 postoperative score, moving them from inadequate/unintelligible speech to Marginal speech/intelligible to lay listeners.
In 15 (15/38 or 39.5%) patients, the magnitude of speech improvement resulted in a PWSS category change, 22 of 38 patients (57.9%) had a subjective improvement but remained on the same category, and 1 of 38 patients (2.6%) deteriorated. This particular case was assigned a PWSS score of 6 preoperatively and a score of 7 postoperatively, moving from a moderate incompetent to an incompetent speech category.
Further break down of the patient groups according to the PWSS categories, 25(65.79%) of these patients were preoperatively considered incompetent and 13 (34.21) borderline incompentent. Of the 25 patients initially designated as incompetent, 3 (12%) improved to borderline competent, 7 (28%) improved to borderline incompetent, and 16 (60%) remained in the incompetent category, with a lower PWSS score. Of the13 patients initially categorized as borderline incompetent speech, 2 (15.4%) improved to competent, 3 (25%) improved to borderline competent, 7 (53.9%) remained borderline incompetent, and 1 (7.7%) deteriorated into incompetent category.
Of the 61 procedures performed, 31 (50.82%) were performed in combination with another procedure, including transverse wedge excision, minor revision of cleft lip, tip rhinoplasty. In the 30 cases (49.18%) where fat graft was the only procedure performed, the average operative time was 39 ± 12.55 minutes.
There were no local or donor site complications.
In our experience, since the introduction of radical intravelar veloplasty to our primary cleft palate repair, 6.7% of the patients have an incompetent speech requiring a secondary pharyngeal procedures. Another 10 % of patients have a borderline incompetent speech that does not merit the risks of a pharyngeal obstructive procedure, but still lacks normal speech.1 Our growing experience with nasoendoscopy has shown that that a number of patients in these groups have good palatopharyngeal mobility with an isolated small VP closure defect that results in ANAE. We suspected that a tailored approach, augmenting the soft tissues in a targeted fashion to close the gap would result in an improvement in speech, potentially eliminating the nasal turbulence observed during subjective assessment.
This same observation has generated a multitude of posterior pharyngeal augmentation techniques with different fillers and mixed results.3–11 A renewed interest in pharyngeal augmentation stimulated by the resurgence of fat grafting with Coleman technique has caused some centers to study and recently report the speech outcomes of fat grafting to the pharynx. Although effective and with lower risk, pharyngeal augmentation still carries a risk of obstructive sleep apnea.
Based on this experience, and motivated to improve speech outcomes by targeted soft tissue augmentation with minimal risks of airway obstruction, we devised a new approach with fat grafting of the palate alone.
With an average follow-up of 12.6 ± 9 months, this study confirmed our hypothesis that fat grafting to the palate alone is safe and well tolerated. The procedure results in significant speech improvement, with 75.5% showing a subjective improvement. This is comparable to the recent reports of fat grafting to the posterior pharynx, with lower risks of airway obstruction due to a much smaller injected volume. There is the added benefit that such small volume, targeted augmentation can be performed as an outpatient procedure with minimal risks to the patient.
Initial concern about possible acute changes to the airway and potential risk of obstruction secondary to edema led to patients being observed overnight. After a 6-month period of evaluation where no airway events occurred and no symptoms from airway edema were observed on clinical examination, we changed the protocol to outpatient. Currently, parents often report that patients return to their regular activities on the same day and are back to school on the next day.
As with fat grafting of any other part of the body, the final grafting result can be observed within 3 months. We reassessed the patients at 30 days and 6 months based on our clinic flow, but a decision to repeat the grafting can be safely done at the 3-month mark. Fat grafting to the palate also varies with the patient’s body weight changes, and significant weight gain or loss could theoretically compromise results. We do not have enough follow-up on these patients to draw any conclusions, but it seems intuitive that repeated grafting or debridement of excess fat can address any issues that may occur.
Thirty percent of the cases required a second or even a third procedure to achieve the target result. Between each session, a reassessment with nasoendoscopy and discussion with family was done. The options of a repeated fat grafting procedure versus a sphincterpharyngoplasty were offered and invariably the patients and their families were willing to undergo the fat grafting procedure again; based on the subjective improvement, convenience, quick recovery, and low risk of complications. We believe that not only the success rate but also the number of procedures required can be improved by a more selective criteria and experience with the procedure. In this study, patients with a closure deficits ranging from 2 to 5 mm were included. Although the numbers are too small to draw a conclusion, we believe that limiting this procedure to patients with smaller gaps is the preferred approach if a single augmentation is the goal. As we gained experience, we also felt the need to upgrade our nasoendoscopy equipment to permit pictures and videos from the awake nasoendoscopy performed in clinic to be saved and used intraoperatively to more accurately guide the injection.
In conclusion, fat grafting to the palate alone is safe, effective and should be considered an option for patients with good palatopharyngeal motion and small focal VP closure defects.
1. Andrades P, Espinosa-de-los-Monteros A, Shell DH 4th, et al. The importance of radical intravelar veloplasty during two-flap palatoplasty. Plast Reconstr Surg
. 2008; 122: 1121–30.
2. Coleman SR. Structural fat grafting
: more than a permanent filler. Plast Reconstr Surg
. 2006; 118(suppl 3): 108S–120S.
3. Remacle M, Bertrand B, Eloy P, et al. The use of injectable collagen to correct velopharyngeal insufficiency
. 1990; 100: 269–274.
4. Bluestone C, Musgrave RH, McWilliams BJ, et al. Teflon injection pharyngoplasty. Cleft Palate J
. 1968; 5: 19–22
5. Furlow LT Jr, Block AJ, Williams WN. Obstructive sleep apnea following treatment of velopharyngeal incompetence by teflon injection. Cleft Palate J
. 1986; 23: 153.
6. Blocksma R. Correction of velopharyngeal insufficiency
by silastic pharyngeal implant. Plast Reconstr Surg
. 1963; 31: 268.
7. Brauer RO. Retropharyngeal implantation of silicone gel pillows for velopharyngeal incompetence. Plast Reconstr Surg
. 1973; 51: 254.
8. Wolford LM, Oelschlaeger M, Deal R. Proplast as a pharyngeal wall implant to correct velopharyngeal insufficiency
. Cleft Palate J
. 1989; 26: 119.
9. Gersuny R. About a subcutaneous prosthesis. Asch Heilkuns
. 1900; 21: 199.
10. Trigos I, et al. Surgical treatment of borderline velopharyngeal insufficiency
using homologous cartilage implantation with video nasopharyngoscopic monitoring. Cleft Palate J
. 1988; 25: 167.
11. Denny AD, Marks SM, Oliff-Carneol S. Correction of velopharyngeal insufficiency
by pharyngeal augmentation using autologous cartilage: a preliminary report. Cleft Palate Craniofac J
. 1993; 30: 46.