The Laryngoscope VC 2013 The American Laryngological, Rhinological and Otological Society, Inc. Does the Type of Cleft Palate Contribute to the Need for Secondary Surgery? A National Perspective James A. Owusu, MD; Meixia Liu, MS; James D. Sidman, MD; Andrew R. Scott, MD Objectives/Hypothesis: To determine whether the type of cleft palate is associated with a need for secondary surgery (oronasal fistula repair, speech surgery) after primary cleft palate repair. Study Design: Retrospective analysis of a national pediatric database (2009 Kids Inpatient Database). Methods: We identified two distinct cohorts of children who underwent primary and secondary (revision) cleft palate repairs, respectively, from a national, pediatric database (2009 Kids Inpatient Database). Revision ratios for each cleft palate diagnosis were calculated to identify diagnoses with higher rates of revision. Revision ratio was calculated by dividing the relative frequency of each diagnosis in the secondary repair cohort by the corresponding relative frequency in the primary repair cohort. Results: In 2009, there were 1942 cases of primary cleft palate repair in the 44 states participating in the KID s inpatient database. Fifty-two percent (n51018) were male. The average age at the time of surgery was 13.36 months. In the same year secondary cleft palate procedures were performed on 724 different patients, 54% (n5388) were males. The average age for secondary procedures was 59 months. Cleft lip and palate diagnoses had higher revision rate ratios (1.92) compared to cleft palate only (0.54) P <0.05. Conclusion: Children with an initial diagnosis of cleft lip and palate, which is more severe than cleft palate only, have comparatively higher rates of secondary cleft palate procedures than children with cleft palate only. Key Words: Cleft palate, secondary surgery, revision surgery, HCUP KID database. Level of Evidence: N/A. Laryngoscope, 123:2387 2391, 2013 INTRODUCTION Cleft palate presents over a wide spectrum of severity ranging from simple asymptomatic bifid uvula to wide bilateral complete cleft involving the hard and soft palates. The hard palate is mostly static, providing a partition between the nasal and oral cavities. In contrast, the soft palate is a dynamic structure that plays a crucial role in normal speech. The soft palate is composed primarily of muscle fibers covered by mucous membrane with minor salivary glands. The major muscles of the soft palate are the tensor veli palatini and levator veli palatini. Normally these two muscles insert into the palatine aponeurosis at the midline of the palate. In the case of palatal cleft, however, the muscles From the Department of Otolaryngology Head and Neck Surgery (J.A.O, J.D.S.), University of Minnesota; the Children s Hospitals and Clinics of Minnesota (M.L., J.D.S.), Minneapolis, Minnesota; and the Department of Otolaryngology Head and Neck Surgery (A.R.S.), Floating Hospital for Children Tufts Medical Center, Boston, Massachusetts, U.S.A Editor s Note: This Manuscript was accepted for publication January 2, 2013. Presented at the Society for Ears Nose and Throat Advancement in Children Annual Meeting, Charleston, SC, November 30 December 2, 2012. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to James A. Owusu, MD, Department of Otolaryngology, University of Minnesota Medical School, Mayo Mail Code 396, 420 Delaware Street SE, Minneapolis, MN 55455. E-mail: owusuja@gmail.com DOI: 10.1002/lary.24008 are oriented along the margin of the cleft and insert onto the posterior edge of the hard palate, making them ineffective. Additionally clefts may extend into the hard palate, creating an opening between the nasal and oral cavities. The goals of cleft palate repair are: to reorient the palatine muscles, to lengthen the soft palate to restore its dynamic sphincter function, and to close the opening between the oral and nasal cavities. These goals are not always achieved through a single surgical procedure. In some cases, additional surgeries may be required to achieve optimal results. 1 Several factors have been shown to affect the outcome of primary repair of cleft palate, including surgical technique, surgeons experience, and timing of the repair. The severity of the cleft may also influence the outcome of primary repair. Conventional wisdom suggests that a wide cleft of the hard and soft palate will more likely require a secondary procedure than an isolated cleft of the soft palate. There are several studies on the rates of secondary procedures in cleft palate repair; however, most of these studies are based on single institutional experience. 2 4 In 1997, the Healthcare Cost and Utilization Project (HCUP) sponsored by the Agency for Healthcare Quality and Research (AHRQ) provided the first version of the Kids Inpatient Database (KID). The KID is a nationwide all-payer encounter-level database sampling of pediatric inpatient admissions. The KID is produced every 3 years, with 2009 being the most recent year. Compiling data from 44 states on patients aged 20 and younger, 2387
the KID has become a valuable resource for research. Sampling weights provided in the database are used to generate national estimates. 5 The goal of this study is to determine whether the initial cleft palate diagnosis is associated with the need for secondary surgery, using the latest KID data. In this study, secondary surgery includes oronasal fistula closure, as well as palatopharyngeal procedures, to improve speech (such as pharyngeal flap or other pharyngoplasties). Enhancing the knowledge base on factors that contribute to the need for secondary cleft palate surgery will be beneficial for healthcare providers in counseling parents of children with cleft palate. MATERIALS AND METHODS We retrospectively analyzed the 2009 KID for all cases of primary and revision cleft palate repairs. As the KID is a public, national database made up of de-identified data, this study was exempt from institutional review board approval. The database was queried using International Classification of Disease Ninth Revision (ICD-9) codes to identify records with a primary admitting diagnosis of cleft palate (ICD-9: 749.00-749.04 and 749.20-749.25). The records were divided into two groups using the primary procedure code for cleft palate repair (2762) to select a primary repair group. A revision repair group was selected using the procedure code for revision cleft palate repair (2763). An inclusion criteria of age 3 years was used for the primary repair group. Patients who underwent alveolar bone grafting (procedure codes: 7691, 8452) were excluded from the revision group. Alveolar bone grafting was considered to be an expected procedure. Relative frequencies were calculated for each cleft palate diagnosis in the primary repair and revision repair groups. Relative frequency allows a proportional comparison of each cleft palate diagnosis between the two groups. In both groups, the following formula was used: Frequency of diagnosis ðicd 9Þ in group Relative FrequencyðRFÞ5 Total number cases in group Thereafter, revision ratios (RRs) were calculated by dividing the relative frequency determined for each diagnosis in the revision group by the corresponding relative frequency in the primary group. This is shown in the formula below: Relative frequency in revision group Revision Ratio ðrrþ5 Relative frequency in primary group A high revision ratio indicates a higher rate of requiring secondary surgery for a particular cleft palate diagnosis. For example, bilateral complete cleft lip and palate (ICD-9: 749.23) represented 5.3% of all cleft palate cases in the primary group. However, it accounted for 10.8 % of cases in the revision group. Per our formula above, this generates a RR of 2.08, which suggests that patients with this diagnosis disproportionally undergo revision procedures compared to certain peers with isolated cleft palate. Taking the analysis a step further, a revision index (RI) was determined, using the RR of unilateral complete cleft lip and palate (ICD-9: 749.21). Unilateral complete cleft lip and palate is the most common specified cleft palate diagnosis in the dataset, and consequently its RR was used as a reference for calculating the revision index. The revision index (see formula below) allows for comparisons between the revision rates of the different types of cleft palate diagnoses in the sample. 2388 TABLE I. Characteristics of Patients Undergoing Primary and Revision Cleft Palate Repair in 2009. Primary Repair Revision Ratio for diagnosis Revision Index ðriþ5 Revision Ratio ICD 9ð749:21Þ Secondary Procedures Age in months (SD) 13.4 (5.8) 58.9 (35.1) Gender Females, n (%) 924 (47.6) 319 (44.1) Males, n (%) 1018 (52.4) 388 (53.6) Diagnosis Cleft palate 1294 (66) 259 (35.8) only, n (%) Cleft lip and palate, n (%) 648 (33.4) 464 (64.1) SD 5 standard deviation; n 5 number of patients. Student t test was performed to compare mean RR and RI between cleft lip and palate diagnoses and cleft palate-only diagnoses. All statistical analysis was performed using SPSS version 19 (IBM). The numbers of cases reported are weighted estimates from the KID. RESULTS Our analysis shows that in 2009, 1942 patients under age 3 underwent primary repair of cleft palate, of which 52% (n51018) were males. Mean age at primary repair was 13.3 months. The most frequently recorded primary diagnosis was unspecified cleft palate (749.00) 38.86 % (n5755), followed by unilateral complete cleft lip and palate (749.21) in 14% (n5272). In the same year, secondary cleft palate procedures were performed in a cohort of 724 patients. The average age for secondary procedures was 59 months. Males accounted for 53.59% (n5388) of secondary procedures. The most common diagnosis for secondary procedures was unspecified cleft palate 24.6% (n5178), followed by unilateral incomplete cleft lip and palate (749.22) 14.60% (n5106) (Tables I and II). A revision ratio of 1 indicates that the diagnosis represents an equal proportion of primary repair and secondary cleft palate procedures. A revision ratio >1 indicates a higher chance of requiring secondary procedures compared to a revision ratio of 1. The revision index compares the RR of each cleft palate diagnosis to that of the most commonly named cleft palate diagnosis in the dataset: unilateral complete cleft lip and palate. Figure 1 compares the relative frequencies of cleft palate diagnoses in the primary and revision groups. Higher RR and RI are seen in more extensive forms of cleft palate (cleft lip and palate) compared to clefts isolated to the palate only (Table II). It is important to note that there were no cases of bilateral incomplete cleft lip and palate in the primary repair group; thus, RR and RI could not be determined for this diagnosis. Student t test comparison of mean RR and RI between cleft lip and palate diagnoses and cleft palate-
Fig. 1. Distribution of cleft palate subtypes in patients who underwent primary repair and revision cleft palate surgeries in 2009. The cleft palate types are based on ICD- 9 codes in HCUP KID. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] only diagnoses reveal significantly higher RR (1.39 vs. 0.44) and RI (1.45 vs. 0.45) in the cleft lip and palate group (P<0.05). DISCUSSION The goals of cleft palate repair include reorienting the palatine muscles and lengthening the soft palate to restore its dynamic sphincter function, as well as closing the opening between the oral and nasal cavities. In some cases, these goals are achieved with one procedure, while in others secondary procedures may be needed. Although several surgical techniques for successful cleft palate repair have been described in the literature, the ideal technique remains a subject of contention. There is general agreement, however, that the management of cleft palate is challenging and requires precise surgical technique and experience for optimal results. The predominant indications for secondary cleft palate procedures are velopharyngeal insufficiency (VPI) and oronasal fistula. Secondary procedures in this study include oronasal fistula repair, palatal lengthening procedures, and pharyngeal procedures to improve speech. Secondary cleft palate procedures in this database are collectively represented by a single procedure code 2763 (revision cleft palate repair). The lack of individual procedure codes for each secondary procedure limits the depth of analysis in this study. Reported rates of VPI and oronasal fistula vary between 0% and 60%. 1 4,6,7 Several studies have reviewed factors affecting outcomes of primary repair of cleft palate. Most agree that early age at the time of cleft palate repair is associated with lower rates of VPI. In contrast, there appears to be no significant association between age and the formation of oronasal fistula. 6,8 Studies by Witt et al. and Salyer et al. have reported decreasing rates of VPI, oronasal fistula, and secondary surgeries with increasing surgeons experience. 9,10 Studies on the association between cleft severity and surgical outcomes have produced conflicting results. In most studies, clefts are classified according to the Veau classification. The Veau classification divides clefts into four groups depending on the extent of involvement: Veau group I is limited to the soft palate only; Veau group II involves the soft and hard palates; Veau group III involves the soft and hard palate as well as the lip; and Veau group IV is bilateral complete cleft. Some studies report no correlation between the Veau classification and surgical outcomes; others report higher rates of VPI and oronasal fistula in groups III and IV clefts. 6,8,10,11 The Veau classification was not used in this study because the KID provides information based solely on the ICD-9 diagnosis. Unspecified complete cleft lip and palate had the highest revision ratio of 2.12, as well as the highest revision index of 2.21. Unspecified cleft lip and palate could most likely represent complete cleft lip and palate as this is the most common presentation of cleft lip with 2389
TABLE II. Distribution of Cleft Palate Subtypes in Patients Undergoing Primary and Revision Palatoplasty. ICD-9 Diagnosis Primary Relative Frequency Secondary Relative Frequency Revision Ratio Revision Index 749.00 Unspecified Cleft Palate 38.9% 24.6% 0.63 0.66 749.01 Unilateral Complete Cleft Palate 5.1% 2.2% 0.43 0.45 749.02 Unilateral Incomplete Cleft Palate 9.8% 6.6% 0.67 0.70 749.03 Bilateral Complete Cleft Palate 3.3% 1.0% 0.29 0.30 749.04 Bilateral Incomplete Cleft Palate 9.4% 1.4% 0.15 0.15 749.20 Unspecified Cleft Lip and Palate 5.1% 10.9% 2.12 2.21 * 749.21 Unilateral Complete Cleft Lip and palate 14.0% 13.4% 0.96 1.00 749.22 Unilateral Incomplete Cleft Lip and Palate 7.3% 14.6% 2.00 2.09 749.23 Bilateral Complete Cleft Lip and Palate 5.3% 10.8% 2.03 2.12 749.24 Bilateral Incomplete Cleft Lip and Palate 0.0% 12.4% N/A N/A 749.25 Other Cleft Lip and Palate 1.5% 1.9% 1.25 1.30 749.00-04 Cleft Palate Only 66.6% 35.8% 0.54 0.56 749.20-25 Cleft Lip and Palate 33.4% 64.1% 1.92 2.00 Note: Revision ratio represents the likelihood of needing revision surgery for the corresponding cleft type. The revision index compares the rate of revision of each cleft type to the revision rate for bilateral incomplete cleft palate. *Diagnosis 749.21 was used as the reference group to calculate the revision index. There were no cases of bilateral incomplete cleft lip and palate in the primary repair group and thus the revision ratio and revision index could not be determined for this diagnosis. cleft palate. Bilateral incomplete cleft palate had the lowest revision ratio (0.15) and revision index (0.15). The general trend noted in this analysis is relatively lower revision ratios and revision indices for cleft palateonly diagnoses when compared with cleft lip and palate diagnoses, which are more extensive. The disproportionate rate of secondary cleft palate procedure for those with cleft lip and palate is reflected in the increased proportion of males undergoing secondary cleft palate procedures as compared to females (Table I). This follows the overall observation that isolated cleft palate (with a lower need for secondary procedures) is more often seen in females, while cleft lip and palate (which more commonly requires secondary procedures) is disproportionately found among males. Based on our analysis of this national database, it can be concluded that some level of association exists between cleft palate diagnosis and the need for secondary cleft palate procedures. 2390 Limitations The findings in this study are subject to several limitations. Although the KID is a highly representative database, the numbers reported are weighted estimates and may not accurately reflect the actual number of cases performed. Additionally, the KID does not include outpatient data; therefore, minor revision surgeries performed on an outpatient basis are not captured in this study. Since the KID is compiled from individual hospital data, the accuracy of the data is also highly dependent on accurate coding by the provider. This study is not longitudinal and utilizes data from a database collected across 44 states over a single year. Therefore, this 2009 KID database study provides a snapshot of cleft palate repair cases, with the assumption that the rates and distribution of primary and secondary cases do not vary significantly from year to year. Ideally, a prospective study following individual patients over time will provide the most precise rate of secondary surgeries. Due to patient confidentiality, the KID does not include unique patient identifiers to permit such a study. However, the sample size and the national coverage afforded by the KID are strengths that can hardly be matched by single institutional studies. Despite these limitations, this study presents a general assessment of the effect of cleft palate type on the need for secondary procedures after primary repair of cleft palate. CONCLUSION Surgical repair of cleft palate, even in the best of hands, can be challenging. In some cases, results have to be improved upon through secondary procedures. Our study supports the notion that more extensive clefts have higher rates of secondary procedures. The rate of secondary procedures for cleft lip and palate is approximately 4 times the rate of secondary procedures for cleft palate only. Acknowledgements James A. Owusu, MD: Conception and design, analysis and interpretation of data, drafting of this article. Meixia Liu: Data analysis, review of manuscript. James D Sidman, MD: Critical review of article, final approval. Andrew R Scott, MD: Conception, design, critical review of article, and final approval. BIBLIOGRAPHY 1. Cohen M. Residual deformities after repair of cleft of the lip and palate. Clin Plast Surg 2004;31:331 345.
2. Phua YS, de Chalain T. Incidence of oronasal fistulae and velopharyngeal insufficiency after cleft palate repair: an audit of 211 children born between 1990 and 2004. Cleft Palate Craniofac J 2008;45:172 178. 3. Lu Y, Shi B, Zheng Q, Hu Q, Wang Z. Incidence of palatal fistula after palatoplasty with levator veli palatini retropositioning according to Sommerlad. Br J Oral Maxillofac Surg 2010;48:637 640. 4. Andersson EM, Sandvik L, Semb GF, Abyholm F. Palatal fistulas after primary repair of clefts of the secondary palate. Scand J Plast Reconstr Surg Hand Surg 2008;42:296 299. 5. HCUP Kids Inpatient Database (KID). Healthcare Cost and Utilization Project (HCUP). 2006 and 2009. Agency for Healthcare Research and Quality, Rockville, MD. www.hcup-us.ahrq.gov/kidoverview.jsp. 6. 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. JCraniofacSurg2009;20:612 616. 7. Bicknell S, McFadden LR, Curran JB. Frequency of pharyngoplasty after primary repair of cleft palate. J Can Dent Assoc 2002;68: 688 692. 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;35:95 100. 9. Witt PD, Wahlen JC, Marsh JL, et al. The effect of surgeon experience on velopharyngeal functional outcome following palatoplasty: is there a learning curve? Plast Reconstr Surg 1998;102:1375 1384. 10. Salyer KE, Sng KW, Sperry EE. Two-flap palatoplasty: 20-year experience and evolution of surgical technique. Plast Reconstr Surg 2006;118: 193 204. 11. 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;108:1515 1518. 2391