ISOLATED CLEFT PALATE IS AMONG

Transcription

1 ORIGINAL ARTICLE JOURNAL CLUB Impact of Cleft Width in Clefts of Secondary Palate on the Risk of Velopharyngeal Insufficiency Scan for Author Audio Interview Derek J. Lam, MD, MPH; Lynn L. Chiu, MD; Kathleen C. Y. Sie, MD; Jonathan A. Perkins, DO Objective: To investigate the association between velopharyngeal insufficiency (VPI), a common finding after cleft palate repair, and various risk factors, including cleft width. Methods: We performed a retrospective cohort study of patients with isolated cleft palates repaired from 2003 to 2008 at a tertiary children s hospital. Patients were observed postoperatively for development of VPI and palatal fistula. The primary risk factor was cleft width. Covariates included cleft length, age at surgery, and presence of associated syndrome. Logistic regression analysis was used to calculate adjusted and unadjusted odds ratios (ORs). Results: The cohort comprised 61 patients. Mean (SD) patient age at the time of cleft repair was 13 (3) months. Fistula rate was 3%. Overall rate of postoperative VPI was 32%. We found significant associations between VPI diagnosis and increasing age in months at the time of palate repair (OR, 1.4 [95% CI, ]) and between VPI and cleft width greater than 10 mm (OR, 5.3 [95% CI, ]). The association between VPI and cleft width was similar after adjustment for cleft length, patient age, and syndrome presence (OR, 4.5 [95% CI, ]). Conclusions: Our results suggest that increased palatal cleft width is associated with a greater risk of postoperative VPI. Clinicians should consider this when counseling patient families for cleft palate repair. Arch Facial Plast Surg. 2012;14(5): Published online April 16, doi: /archfacial Author Affiliations: Departments of Otolaryngology Head and Neck Surgery, Cincinnati Children s Hospital Medical Center, Cincinnati Ohio (Dr Lam), University of Washington, Seattle (Drs Lam, Chiu, Sie, and Perkins), and University of California, Irvine (Dr Chiu); Craniofacial Center, Seattle Children s Hospital, Seattle, Washington (Drs Sie and Perkins). ISOLATED CLEFT PALATE IS AMONG the most common birth defects, with an estimated incidence of 1 in 2000 live births. The different types of cleft palate can be grouped according to the Veau classification 1 : incomplete cleft of the soft palate only (type 1), cleft of both hard and soft palate but limited to the secondary palate (type 2), complete unilateral cleft of primary and secondary palate as well as the lip (type 3), and complete bilateral cleft of primary and secondary palate (type 4). 2 The primary goals of palatoplasty are to restore the structural competence of the hard and soft palates and allow closure of the velopharyngeal port during speech and swallowing. Inadequate closure of the velopharynx during speech production results in velopharyngeal insufficiency (VPI), which is characterized by hypernasal speech and nasal air escape during speech production and may be associated with nasopharyngeal reflux with swallowing. Velopharyngeal insufficiency after cleft palate repair is typically reported in 2% to 30% of patients with cleft palates There have been many studies investigating risk factors for VPI among patients with cleft palates reported in the literature, but most focus on surgical technique, demographic factors, or type of palatal cleft. 3,6,7,9-12 We sought to examine the impact of cleft width on the development of VPI among patients with isolated cleft palate. CME available online at and questions on page 309 Journal Club slides available at METHODS SUBJECTS This was a retrospective cohort study of consecutive patients with isolated cleft palates (Veau type 2) who underwent palate repair by a senior pediatric otolaryngologist ( J.A.P.) at Seattle Children s Hospital between 2003 and 2009 with a minimum of 1 year of follow-up Author Affil Departments Otolaryngolo Surgery, Cinc Hospital Med Cincinnati O University o Seattle (Drs Perkins), and California, Ir Craniofacial Children s H Washington Perkins). 360

2 Table 1. Patient Characteristics by Dichotomous Cleft Width a Figure. Measurement of cleft width and length. postoperatively. While most patients had nonsyndromic isolated clefts, patients with clefts associated with other syndromes or chromosomal abnormalities were included in the study. Exclusion criteria included other Veau cleft types (1, 3, or 4), missing cleft measurements, or patients who were nonverbal owing to developmental delay at the time of follow-up. Risk factors including age at the time of repair, syndrome presence, type of palate repair, and the primary outcome of VPI diagnosis were gleaned from electronic medical record review. This study was approved by the Seattle Children s Hospital institutional review board. SURGICAL TECHNIQUE All clefts were repaired with either a Furlow double-opposing Z-plasty or a combined Furlow palatoplasty and V to Y (V-Y) pushback. Intraoperative measurements of cleft width and length were taken at the time of palate repair prior to local anesthetic infiltration. Cleft width was measured at the junction of the hard and soft palate, and cleft length was measured from the anterior limit of the cleft to the midpoint of the transverse line connecting the 2 uvulae (Figure). In the Furlow palatoplasty technique, double-opposing, posteriorly based musculomucosal flaps were raised on either side of the cleft and repositioned transversely to simultaneously close the cleft and lengthen the soft palate while realigning the palatal levator sling. When the anterior extent of the cleft did not allow favorable closure using just the Furlow technique, a V-Y pushback was also used. In this technique, symmetric posteriorly based mucoperiosteal flaps were raised and reapproximated in the midline to close the palatal cleft. Nasal mucosal flaps were also elevated and closed to recreate the nasal floor. This repair resulted in anterolateral mucosal gaps healed by secondary intention. DIAGNOSIS OF VPI Patients were observed postoperatively for a minimum of 1 year for the development of VPI. This was diagnosed by 1 of 2 speech language pathologists using standard diagnostic criteria, which included ratings of hypernasal resonance, nasal air escape, speech intelligibility, and compensatory misarticulations. 13 STATISTICAL ANALYSIS Logistic regression analysis was used to assess unadjusted and adjusted associations between the primary risk factor of cleft width (continuous variable, reported in millimeters) and the primary outcome of diagnosis with VPI (dichotomous variable, reported as presence or absence). Secondary analysis included cleft width as a dichotomous variable (either 10 mm or 10 mm). Covariates included cleft length as both a continuous (in millimeters) and dichotomous ( 30 mm or 30 mm) variable, patient age at time of palate repair (continuous variable, measured in months), presence of syndrome (dichotomous variable), and type of palate repair (dichotomous variable, Furlow vs Furlow plus V-Y pushback). Results are presented as odds ratios (ORs) of association. RESULTS Cleft Width, mm Characteristic Total (n = 73) 10 (n = 45) 10 (n = 28) P Value b Sex Male 24 (33) 14 (31) 10 (36) Female 49 (67) 31 (69) 18 (64).80 Age at surgery, 12.5 (2.9) 12.0 (3.3) 13.3 (1.9).07 mean (SD), mo Duration of 21.7 (15.8) 22.4 (17.8) 20.5 (11.9).62 follow-up, mean (SD), mo Palate length, mm c Mean (SD) 28.5 (10.6) 23.7 (5.3) 34.8 (12.4) (58) 31 (86) 6 (21) (42) 5 (14) 22 (79) Type of repair Furlow 31 (42) 28 (62) 3 (11) Furlow plus 42 (58) 17 (38) 25 (89).001 V-Y pushback Syndrome presence d Yes 10 (14) 6 (14) 4 (14) No 60 (86) 36 (86) 24 (86).99 Fistula presence Yes 2 (3) 2 (7) No 65 (97) 40 (100) 25 (93).16 VPI presence Yes 23 (32) 8 (18) 15 (54) No 50 (68) 37 (82) 13 (46).002 Abbreviation: VPI, velopharyngeal insufficiency. a Unless otherwise indicated, data are reported as number (percentage) of patients. b P values calculated using the Student t test for continuous variables and Fisher exact test for dichotomous variables. c Palate length data available for only 64 patients. d Syndromic diagnoses included Stickler syndrome, Kabuki syndrome, velocardiofacial syndrome, skeletal dysplasia, and AEC syndrome (ankyloblepharon-ectodermal defects cleft lip/palate). Cleft width and length measurements were available for 101 of 128 patients who underwent cleft palate repair during the study period. Of these, speech pathology follow-up data were available for 73 patients. The mean age of patients at the time of surgery was 12.5 months (age range, 8-30 months), and mean follow-up for the development of VPI was 22 months postoperatively (Table 1). Mean (SD) palate width was 8.2 (3.2) mm, and mean palate length was 28.5 (10.6) mm. Most patients had nonsyndromic isolated clefts that were repaired with a combination of Furlow and V-Y pushback palatoplasty. Only 2 patients developed a persistent oronasal fistula postoperatively (3%). Velopharyngeal insufficiency was diagnosed in 23 of the patients (32%). 361

3 Table 2. Unadjusted Associations Between Patient Characteristics and VPI Diagnosis Characteristic OR (95% CI) P Value Covariates Female sex 1.2 ( ).77 Age at surgery, mo Continuous 1.5 ( ).01 Categorical 12 1 [Referent] ( ) (1.2-20).03 Type of repair.002 Furlow 1 [Referent] Furlow plus V-Y pushback 8.5 (2.2-33) Syndrome presence 2.5 (0.6-10).19 Palate Cleft Size Cleft width, mm Continuous 1.4 ( ).001 Dichotomous, ( ).002 Cleft length, mm Continuous 1.0 ( ).10 Dichotomous, (1.5-14).01 Cleft area, mm 2 Continuous 1.01 ( ).01 Ordinal, 100-mm 2 intervals 1.8 ( ).003 Dichotomous, 150 mm ( ).02 Abbreviations: OR, odds ratio; VPI, velopharyngeal insufficiency. When compared with patients with cleft widths narrower than 10 mm, patients with clefts of 10 mm or wider had significantly longer clefts and were far more likely to have a Furlow with V-Y pushback palatoplasty rather than a Furlow palatoplasty alone. Patients with wider clefts also had a significantly higher rate of VPI diagnosis (54% vs 18%) (P=.002) (Table 1). When we examined the cohort stratified by surgery type comparing those who underwent only Furlow palatoplasty with those who underwent Furlow plus V-Y pushback, we found significant differences in mean (SD) cleft width (6.0 [1.9] mm and 9.8 [3.0] mm, respectively) (P.001), cleft length (21.6 [4.6] mm and 34.0 [10.8] mm, respectively) (P.001), and subsequent diagnosis with VPI (10% and 48%, respectively) (P.001). In unadjusted logistic regression analysis, significant associations were found between presence of VPI and greater age at time of surgery (OR, 1.5 [95% CI, ]) and increasing cleft width (OR, 1.4 [95% CI, ]) (Table 2). In addition, cleft width greater than 10 mm appeared to be strongly associated with VPI (OR, 5.3 [95% CI, ]). Cleft length greater than 30 mm also demonstrated a large association (OR, 4.5 [95% CI, ]). After adjusting for cleft length, patient age at the time of surgery, age at follow-up, and presence of syndrome, we found that the association between VPI and increasing cleft width remained significant (OR, 1.4 [95% CI, ]) for the continuous variable. Cleft width greater than 10 mm also remained strongly associated with VPI (OR, 4.5 [95% CI, ]) in adjusted analysis (Table 3). In the final multivariable model, the associations between VPI and the other covariates, including cleft length, did not achieve statistical significance. Table 3. Adjusted Associations Between Cleft Width and VPI Diagnosis Cleft Width OR (95% CI) P Value Overall cohort Continuous 1.4 ( ) a.01 Dichotomous, 10 mm 4.5 (1.1-19) a.04 Stratifying by surgery type Furlow Continuous 1.8 ( ) b.049 Dichotomous, 10 mm 6.3 ( ) b.24 Furlow plus V-Y pushback Continuous 1.1 ( ) b.50 Dichotomous, 10 mm 1.0 ( ) b.96 Abbreviations: OR, odds ratio; VPI, velopharyngeal insufficiency. a Adjusted for cleft length, patient age at time of surgery, age at follow-up, and presence of syndrome. b Adjusted for patient age at time of surgery and cleft length. To account for the impact of type of surgical repair, the logistic regression analysis was also carried out stratifying the cohort by surgery type. This demonstrated continued significant adjusted associations between VPI and larger cleft width (OR, 1.8, [95% CI, ]) among patients who underwent Furlow palatoplasty. However, for patients who underwent Furlow and V-Y pushback palatoplasty, there was no longer a significant association when adjusted for the other covariates in the model. COMMENT It seems intuitively clear that patients with more severe clefting, which is technically more challenging to repair, are more likely to have poor outcomes. This has been demonstrated in multiple studies where palatal fistula was the primary outcome of interest and severity of clefting was categorized according to the Veau classification scheme However, there have been very few authors who have attempted to specifically quantify cleft width as a risk factor for poor postoperative outcomes. Parwaz et al 15 found that a cleft width greater than 15 mm was significantly associated with increased fistula formation, but their study did not address speech outcomes. Velopharyngeal insufficiency is a common finding after cleft palate repair and is typically reported in the range of 2% to 30% of patients with cleft palates following repair with varied surgical techniques Our finding of an overall VPI rate of 32% and fistula rate of 3% was consistent with previous reports. Among the studies investigating speech outcomes after palate repair, several investigators have noted a negative or nonlinear association with cleft severity as defined by the Veau classification. 4,5,7,9,10,18-21 Marrinan et al 5 noted that patients with Veau types 2 and 4 appear to have poorer speech outcomes than those with types 1 and 3 and suggested that the presumed gradation of cleft severity delineated by the Veau classification scheme may not be clinically prognostic, at least with respect to speech outcomes. They further postulated that the relatively normal position of the vomer muscular attachments in Veau types 1 and 3 may 362

4 account for their findings. While this hypothesis may provide a physiologic explanation for a greater risk of postoperative VPI in these patients, it still relies on the Veau classification scheme as a method of prognostication, which we feel may be inadequate. We suggest instead that cleft width should be considered as a prognostic indicator of postoperative VPI, regardless of Veau type. It has been our observation that even within a single Veau category, there can be wide variation in cleft width and that those patients with wider clefts frequently have poor postoperative speech outcomes. Our finding that patients with a cleft width greater than 10 mm have roughly 4.5 times the odds of developing VPI as those with a smaller clefts confirms our clinical observation. In addition, we found that for every millimeter increase in cleft width, the odds of developing VPI increased by 40%. When we stratified by surgery type, we found conflicting associations with increasing width among patients who had either a Furlow palatoplasty compared with those who had a 2-flap pushback in combination with a Furlow palatoplasty. Unadjusted analysis suggested that patients who underwent the Furlow and pushback palatoplasty had approximately 8.5 times the odds of developing VPI as those who underwent a Furlow procedure alone. From this result, it is tempting to conclude that the Furlow plus pushback palatoplasty procedure itself is associated with a greater risk of VPI. However, because the pushback was primarily used in wider clefts to close the hard palate defect and had little impact on the soft palate, we believe that this finding is likely related more to the greater cleft size than the nature of the procedure. Because the type of palatoplasty used was dependent on the cleft width, the independent effect of the procedure cannot be clearly ascertained in the context of the current study. As summarized in Table 1, there was a significant selection bias based on the indication for the 2 different procedures such that patients with wider clefts were far more likely to have the Furlow and pushback procedure. The skewed distribution of the procedures did not allow adequate adjustment for repair type in a multivariate model. To account for this, we performed separate analyses stratifying the cohort by surgery type. We found that for patients with smaller clefts ( 10 mm) who primarily underwent Furlow palatoplasty, there was a persistent significant association between increasing cleft width and diagnosis with VPI. However, there was virtually no association between increasing width and VPI among those who had wider clefts ( 10 mm) and underwent a Furlow and pushback palatoplasty. We interpret this to be due to a possible ceiling effect with respect to the impact of increasing cleft width and risk of developing VPI. It may be that in patients with a cleft width of 10 mm or greater, the risk of VPI is so high that further increases in cleft width are not significantly associated with increasing rate of VPI. To our knowledge, this is the first report quantifying the risk of VPI according to an objective measurement of cleft width. The high rate of VPI among patients with clefts at least 10 mm in width suggests that in patients with isolated palatal clefts wider than 10 mm, there should be careful consideration of strategies to minimize postoperative VPI. This would include the use of lengthening procedures such as the Furlow palatoplasty, and in some cases, one might even consider procedures to correct VPI such as a pharyngeal flap or sphincter pharyngoplasty at the time of primary palate repair. 22 This study has several limitations that are worth noting. First, we did not include patients with other types of clefts; therefore, it is difficult to generalize the impact of cleft width beyond Veau type 2 clefts, particularly when one considers the difference in rate of VPI among the different Veau types. Veau type 2 clefts have generally been found to be at greatest risk of developing VPI, and it is possible that the impact of greater cleft width may be less pronounced in other types of clefts. Second, the method of cleft repair was limited to Furlow palatoplasty or Furlow and V-Y pushback palatoplasties, both of which are aimed at lengthening the palate in addition to closing the cleft and have been demonstrated to have lower rates of VPI than other methods of cleft repair. 6,10,23 At least 1 author has argued that preoperative palate length is the most important prognostic factor in the development of VPI, 22 and one could argue that the relative impact of palate length on development of VPI is lessened in this setting compared with cleft width. Regardless, the fact that there is still substantial VPI in spite of the use of the Furlow palatoplasty further highlights the need for delineation of prognostic indicators and careful preoperative planning. Future prospective studies investigating the impact of cleft width on postoperative VPI should include other cleft types and other methods of repair to allow greater generalization. In conclusion, the risk of developing VPI after isolated cleft palate repair appears to increase with increasing width of the palatal cleft, and the rate of postoperative VPI is particularly high in patients with a cleft width greater than 10 mm. Careful consideration should be given to maximizing velopharyngeal closure at the time of primary palate repair in these patients. Accepted for Publication: February 8, Published Online: April 16, doi: /archfacial Correspondence: Derek J. Lam, MD, MPH, Department of Otolaryngology Head and Neck Surgery, MLC 2018, Cincinnati Children s Hospital Medical Center, 3333 Burnet Ave, Cincinnati, OH (derekjlam@gmail.com). Author Contributions: Study concept and design: Lam, Chiu, and Perkins. Acquisition of data: Chiu and Perkins. Analysis and interpretation of data: Lam, Chiu, and Sie. Drafting of the manuscript: Lam and Perkins. Critical revision of the manuscript for important intellectual content: Lam, Chiu, Sie, and Perkins. Statistical analysis: Lam and Perkins. Administrative, technical, and material support: Chiu. Study supervision: Chiu and Perkins. Financial Disclosure: None reported. Online Only Material: Listen to an author interview about this article, and others, at This article is featured in the Archives Journal Club. Go to to download teaching PowerPoint slides. 363

5 Previous Presentations: This research was presented at the American Academy of Facial Plastic and Reconstructive Surgery Annual Meeting; October 1-3, 2009; San Diego, California. REFERENCES 1. Veau V. Bec-de-Liévre, Formes Cliniques Chirurgie: Avec la Collaboration de J. Récamier. Paris, France: Masson et Cie; van Aalst JA, Kolappa KK, Sadove M. MOC-PSSM CME article: Nonsyndromic cleft palate. Plast Reconstr Surg. 2008;121(1)(suppl): Koh KS, Kang BS, Seo DW. Speech evaluation after repair of unilateral complete cleft palate using modified 2-flap palatoplasty. J Craniofac Surg. 2009;20(1): Inman DS, Thomas P, Hodgkinson PD, Reid CA. Oro-nasal fistula development and velopharyngeal insufficiency following primary cleft palate surgery: an audit of 148 children born between 1985 and Br J Plast Surg. 2005;58(8): Marrinan EM, LaBrie RA, Mulliken JB. Velopharyngeal function in nonsyndromic cleft palate: relevance of surgical technique, age at repair, and cleft type. Cleft Palate Craniofac J. 1998;35(2): McWilliams BJ, Randall P, LaRossa D, et al. Speech characteristics associated with the Furlow palatoplasty as compared with other surgical techniques. Plast Reconstr Surg. 1996;98(4): Sullivan SR, Marrinan EM, LaBrie RA, Rogers GF, Mulliken JB. Palatoplasty outcomes in nonsyndromic patients with cleft palate: a 29-year assessment of one surgeon s experience. J Craniofac Surg. 2009;20(suppl 1): Chen PK, Wu J, Hung KF, Chen YR, Noordhoff MS. Surgical correction of submucous cleft palate with Furlow palatoplasty. Plast Reconstr Surg. 1996;97 (6): Härtel J, Gundlach KK, Ruickoldt K. Incidence of velopharyngoplasty following various techniques of palatoplasty. J Craniomaxillofac Surg. 1994;22(5): Kirschner RE, Wang P, Jawad AF, et al. Cleft-palate repair by modified Furlow double-opposing Z-plasty: the Children s Hospital of Philadelphia experience. Plast Reconstr Surg. 1999;104(7): Cheng N, Zhao M, Qi K, Deng H, Fang Z, Song R. A modified procedure for velopharyngeal sphincteroplasty in primary cleft palate repair and secondary velopharyngeal incompetence treatment and its preliminary results. J Plast Reconstr Aesthet Surg. 2006;59(8): LaRossa D, Jackson OH, Kirschner RE, et al. The Children s Hospital of Philadelphia modification of the Furlow double-opposing z-palatoplasty: long-term speech and growth results. Clin Plast Surg. 2004;31(2): Dalston RM, Warren DW. The diagnosis of velopharyngeal inadequacy. Clin Plast Surg. 1985;12(4): Andersson EM, Sandvik L, Semb G, Abyholm F. Palatal fistulas after primary repair of clefts of the secondary palate. Scand J Plast Reconstr Surg Hand Surg. 2008;42(6): Parwaz MA, Sharma RK, Parashar A, Nanda V, Biswas G, Makkar S. Width of cleft palate and postoperative palatal fistula: do they correlate? J Plast Reconstr Aesthet Surg. 2009;62(12): Muzaffar AR, Byrd HS, Rohrich RJ, et al. Incidence of cleft palate fistula: an institutional experience with two-stage palatal repair. Plast Reconstr Surg. 2001; 108(6): Cohen SR, Kalinowski J, LaRossa D, Randall P. Cleft palate fistulas: a multivariate statistical analysis of prevalence, etiology, and surgical management. Plast Reconstr Surg. 1991;87(6): Greene MC. Speech analysis of 263 cleft palate cases. J Speech Hear Disord. 1960;25: Pickrell K, Clifford E, Quinn G, Massengill R. Study of 100 cleft lip-palate patients operated upon 22 to 27 years ago by one surgeon. Plast Reconstr Surg. 1972;49(2): Massengill R Jr, Pickrell K, Robinson M. Results of pushback operations in treatment of submucous cleft palate. Plast Reconstr Surg. 1973;51(4): Spriestersbach DC, Moll KL, Morris HL. Subject classification and articulation of speakers with cleft palates. J Speech Hear Res. 1961;4: Randall P, LaRossa D, McWilliams BJ, Cohen M, Solot C, Jawad AF. Palatal length in cleft palate as a predictor of speech outcome. Plast Reconstr Surg. 2000; 106(6): Perkins JA, Lewis CW, Gruss JS, Eblen LE, Sie KC. Furlow palatoplasty for management of velopharyngeal insufficiency: a prospective study of 148 consecutive patients. Plast Reconstr Surg. 2005;116(1):