Timing of Cleft Palate Closure Should Be Based on the Ratio of the Area of the Cleft to That of the Palatal Segments and Not on the Age Alone

Transcription

1 17 Timing of Cleft Palate Closure Should Be Based on the Ratio of the Area of the Cleft to That of the Palatal Segments and Not on the Age Alone Samuel Berkowitz Diagnosis and surgical treatment planning in both medicine and dentistry are most frequently dependent on the patient s age, and nature and extent of the tissue s defect. In cleft lip and palate, the timing for surgically closing a cleft palate has been traditionally based solely on the age of the patient and the onset of speech (usually between 6 to 8 months) irrespective of the physical assets and defects of the affected tissue and not on the relative size of the palatal cleft defect to that of the surrounding palatal tissue. The inability to develop quantitative diagnostic criteria to facilitate a differential diagnosis for proper treatment planning is partially due to cleft palate research programs having an insufficient number of investigative data obtained from serial maxillary and mandibular dental palatal casts of patients starting at birth and extending into adolescence as well as a proper instrument to measure the palatal cast s surface. The premise of this investigation is that data extrapolated from serial cast records will establish the relationship of the cleft defect to the palatal size and shape under the influence of corrective surgery. Clinicians who do not have serial cast records have failed to appreciate the importance of cleft size/shape variations that exist within each cleft type at various ages, which may be crucial for making the proper decision as to when to surgically close the cleft space to avoid growth-inhibiting scarring. With the advent of advanced technology to perform 3D spatial-temporal measurements and Cadcam computer software for in-depth analyses, extrapolated surface data from palatal casts can now be subjected to highly sophisticated quantitative analyses to perform differential diagnosis and treatment planning. Hypothesis to be tested: The hypothesis that a relationship exists between cleft size and palatal size to achieve good facial-palatal growth and speech will be tested, also, there is more than one physiological surgical procedure and the Hotz presurgical orthopedic protocol does not increase palatal size in velocity Method and Material (Tables 17.1, 17.2) Employing the palatal casts of 242 male and female individuals from eight institutions in the USA and western Europe, separate serial analyses were conducted of well-growing cases with excellent aesthetics, dental occlusion,and speech and a palatal control series of 17 nonpalatal cleft cases, to access the growth changes in size and velocity from birth through adolescence. These control cases consisted of various clefts of the lip and alveolus and/or soft palate but with no clefts in the hard palate. The various complete cleft lip and palate series were compared to this series. Malcom Johnson (personal communication) has confirmed that this control sample is an appropriate one for the palatal growth comparisons. In this group the midpalatal suture, extending anteriorly to the incisal papilla at the anterior alveolar ridge, served as the medial border dividing the palate into right and left segments. With the exception of the excellent sample of cases from Goteborg, the vomer series had poor occlusion, facial aesthetics, and poor speech. This series was included in order to determine what timing and type of surgical procedure had produced favorable or unfavorable outcomes, and also whether the outcomes varied with the relationship of the size of the cleft defect to the size of the palatal segments medial to the alveolar ridges at time of surgery. The participating institutions were selected on the basis of their excellent records and varied treatment protocols, and different racial, and mixed gender populations. The Miami sample of nonorthopedically treated cases was selected at random from a larger number of similarly treated cases.

2 364 S. Berkowitz Table Number of Clinics and Patients Involved in Study Clinics Timing of surgery Palatal closure procedure Lip Adhesion Second 2nd Alveolar Hard and Lip-Nose Bone Graft Soft Palate Miami 3 mos 6 8 mos 8 10 yrs mos Von Langenbeck + Vomer Flap Illinois 2 4 mos 6 7 mos Seldome mos Various procedure Wardil-Kilner, Von Langenbeck Nijmegen 1 2 mos udp 6 mos 10 yrs SP mos Von Langenbeck bdp 9 mos HP 6 9 yrs Goteborg udp 1 2 mos 6 9 mos 9 11 yrs SP 6 8 mos Vomer Flap Delayed bdp 2 4 mos HP 9 11 yrs Goteborg 2 mos mos average 14 yrs 2 mos arterior vomer flap, Vomer 9 mos posterior push back Northwestern 6 8 wks primary 6 mos 12 mos intravelar veloplasty bone graft Twelve different groupings of cases were established depending on their institutional location and type of cleft (complete unilateral cleft lip and palate, CUCLP, and complete bilateral cleft lip and palate, CB- CLP), with each group consisting of both males and females. The number and type of cases were as follows: University of Miami: 26 CUCLP, and 17 CBCLP; University of Illinois: 12 CUCLP; Northwestern University: 22 CUCLP: Nijmegen: 10 CBCLP; Goteborg- Delayed Closure: 24 CUCLP, 8 CBCLP; Goteborg- Vomer: 23 CUCLP, 10 CBCLP; Amsterdam: 26 CUCLP, 4 CBCLP; Rotterdam: 60 CUCLP. The statistical analyses of the CUCLP and CBCLP cases were carried out separately. The Amsterdam and Rotterdam cases, which had been treated with the Hotz presurgical protocol (PP), were followed from birth for 48 months, permitting a determination of the PP on palatal growth effects differed due to presurgical orthopedics. Analysis to be made: Comparative effects of treatment of (1) the palates surface area s rate of change (velocity) and growth (size in 2 mm). (2) Size of the posterior cleft space and the velocity of its change. (3) Ratio of cleft size to the palate s size before and after surgery (total surface area). Each of the subsamples of cleft children will be compared to each other and to the age-appropriate control samples. Data Extrapolation from the surface of palatal casts (Fig. 17.1): A highly accurate, 3D, electromechanical palatal cast measuring instrument which gave a measurement error of less than 5% made possible a spatial-temporal (4D) analysis of palatal form and size changes, permitting an in-depth study of how the cleft space and palatal segments of each clinic s cases changed in relationship with each other over time. (Figs. 17.2a, 17.3) Features to be measured and analyzed (Fig ): The size of the palatal segments are measured serially starting at birth and divided into two treatment periods. The first period ends at surgery to close the palatal cleft. The second period is after palatal cleft closure: it includes adding the remaining cleft space with the changing size of the palatal segments (total surface area). In CUCLP: (Fig. 17.2a, b) The sizes of the palatal segments are limited laterally by the alveolar ridges, and anteriorly by a line connecting the most anterior point on the alveolar ridge (AC and AC1). Posteriorly a line is drawn from point gingival (P and P1) which are equivalent to the pterygomaxillary fissure seen on cephaloradiographs which marks the posterior limits of hard palate. When this transpalatal line makes contact with the cleft, points PC and PC1 are created. PC to PC1 measures posterior cleft space width. Medial border of the palatal segments (AC-PC, AC1-PC1) is limited by the cleft space. In CBCLP (Fig. 17.3) the premaxilla s surface area is limited anteriorly by its alveolar ridge (PML-PMR) Method Used for Analyses Cleft Subjects: The subjects had unequal numbers of observations and were observed at different ages. In order to compare the responses of the different groups, the data was recoded into new age intervals of 2, 8, 18, 36, 60, 84, 120, 168, and 192 months. The groups from Amsterdam-Rotterdam were observed only for 4 years and were coded as 2, 8, 18, and 48 months. Infrequently, especially in the early months, subjects would have two observations in a recoded age interval. When this happened, the mean values of the observations were used. The within-group variances

3 Chapter 17 Timing of Cleft Palate Closure Should Be Based 365 a b c d e Fig a e. Various palatal forms being measured. a Normal palate. b Posterior view of normal palate showing position of teeth landmark points: middle of incisal edge and central fossa of molars. c Isolated cleft palate. d Complete unilateral cleft lip and palate. e Complete bilateral cleft lip and palate were compared during subsequent analyses to ensure that the equal-variance requirement of the Analysis of Variance was maintained and not biased by the use of mean values in some instances. Controls: Serial casts from 17 patients who had only clefts of the lip and/or alveolus (CLA), with or without clefts of the soft palate (SP), or clefts of the lip and soft palate (CLA and SP) alone. All of these patients did not have a cleft in the hard palate. One case series had no cleft at all. The data from the control series was handled exactly as that from the cleft series. Statistical Analysis: A preliminary analysis of palatal area was performed using a random effects model in the SAS Proc Mixed software. This preliminary analysis showed not only group differences, but, more importantly, heterogeneity in group responses over time (i.e., a significant Group x Time interaction). Because of the significant interaction, overall differences among groups or times are not directly interpretable. Consequently, the simple effects of differences between groups at each time point were investigated. The SAS Proc GLM software was used to test the fixed effects model among groups at each time, followed by a pair-wise comparison among mean values. At each time point it was sometimes possible to compare among eight groups, which would lead to 28 comparisons (some time points had fewer groups because not all groups were observed at every time point). A strict control for multiple comparisons, say by the Bonferroni procedure, would severely decrease the power to detect differences. Therefore, the alpha level for each pair-wise test was set at No statistical comparisons were made across time points but the consistency of the significant differences found in the pair-wise tests was noted. In this way, biologically consistent differences could be elucidated with sufficient power Treatment Protocols at Each of the Centers (Table 17.2) Miami Craniofacial Anomalies Foundation, South Florida Cleft Palate Clinic (Figs ) All of the lip and palate surgery was performed by D. Ralph Millard Jr., M.D., secondary alveolar bone graft by S.A. Wolfe, M.D., and all staged orthodontics procedures were performed by Samuel Berkowitz, D.D.S. All cases studied from the participating institutions had similar graphs created.

4 366 S. Berkowitz Fig a Twelve serial digital images of a complete unilateral cleft lip and palate from 13 days of age to 14 years and 2 months of age. The 2D images generated from 3D casts, are accurate in form but are not related in size to each other. The image at birth (13 days) is followed by (0.3 and 1.6) that had undergone molding as a result of muscular contractive forces created by lip adhesion at 14 months of age. Palatal cleft closure using a von Langenbeck with modified vomer flap was performed at 14 months of age. A secondary alveolar bone graph was placed at 8 years and 3 months of age after some minor orthodontics to align the anterior teeth. No presurgical orthopedics was performed a b Fig b Graph depicting serial palatal 3D bony growth changes and the size of the cleft space of case 2A. The combined surface areas of the large and small palatal segments changed from 679 mm 2 at 13 days to 1121 mm 2 at 1 year 6 months (an increase of 39.5%) at the time of palatal surgery. The cleft space reduced from 365 mm 2 to 202 mm 2 at 3 months due to molding. Additional molding and palatal growth further reduced the cleft space prior to palatal surgical closure. Palatal growth occurred gradually up to 5 years and 6 months. The continued marked increase in palatal sizes up to 10 years, 7 months reflects posterior palatal growth to accommodate the developing molars. This posterior palatal area appears not to be affected by the palatal surgery

5 Chapter 17 Timing of Cleft Palate Closure Should Be Based 367 Table Treatment Protocol and Sequence Controls = Soft Palate + Lip/Alveous + Soft Palate (Miami) 5 5 Lip Alveous 7 Clinics CBCLP CUCLP Presurgical Orthopedics Miami None Illinois 0 12 None Nijmegen 10 0 Hotz 3 4 wk for 2 3 yr Goteborg Delayed wk to yr Goteborg Vomer wk Northwestern 0 23 Posterior 1/2 palate Fig Digitized serial 2D cast images of a one side incomplete and the opposite side complete bilateral cleft lip and palate. The serial form cast s serial changes are accurate but not so as it relates to their relative size.at 0-1 (1 month) palatal segments are laterally displaced. At 0-4 (4 months), this image shows medial movement (molding) of the palatal segments, as a result of lip adhesion. The premaxilla is positioned forward of the lateral palatal segments with marked closure of the cleft space. At 0-8 and 1-1, these images show further spontaneous closure of the palatal cleft space as a result of palatal growth.at 1-4 palatal closure was performed 1 month earlier using a von Langenbeck with a modified vomer flap. Note that the following cast images show the gradual incorporation of the premaxilla within the palatal arch with orthodontic treatment 17.3 Results Comparison of Total Surface Area in Unilateral Cases The mean values, standard errors, and sample sizes for the unilateral data are shown in Table 17.3 and Fig. 17.2a,b. From Table 17.3 it appears that all other groups are below the Controls and that the Goteborg- Vomer group and the Northwestern group behave somewhat differently from the rest. Both are lower. The results of the statistical comparisons are shown in various Tables which are set up as a matrix to be read as follows: In the cell defined by the intersection of a row for one group and the column for another group the times where the two groups are significantly different are indicated. As an example, following the Miami row and the Goteborg-V column,we see that these two groups had significantly different total areas at 36, 60, and 84 months. That this is reasonable can be determined by inspecting Fig. 17.6, Unilateral Cleft Lip and Palate Data, or looking at the mean values in Table The most striking feature in Tables 17.3 and 17.4 is that all of the groups differ from Controls over time. (*see note) The differences decrease over time due to growth changes in the posterior molar area. This growth area is associated with the developing perma-

6 368 S. Berkowitz Table Total Palate Surface Area (Mean ± Std. Error, Sample Size) of Unilateral Clefts by Data Source and Age in Months Months Source Amsterdam 817 ± ± ± ± 29 a (33) (33) (31) (33) Miami 811 ± ± ± ± ± ± ± ± 75 (28) (22) (24) (30) (30) (28) (28) (18) Controls 930 ± ± ± ± ± ± ± ± 123 (11) (13) (12) (14) (15) (17) (12) (7) Goteberg (D) 882 ± ± ± ± ± ± ± ± 46 (24) (22) (18) (21) (20) (20) (24) (23) Goteberg (V) 842 ± ± ± ± ± ± ± ± 57 (23) (22) (19) (20) (20) (22) (23) (21) Illinois 892 ± ± ± ± ± ± ± ± 61 (11) (11) (19) (17) (11) (9) (10) (9) Northwestern 730 ± ± ± ± ± ± ± ± 95 (28) (12) (11) (9) (7) (6) (21) (6) Rotterdam 779 ± ± ± ± 25 a (60) (59) (59) (59) a Measurements made at 48 months. Table Statistically Significant Differences in Unilateral Total Surface Area (p < 0.01) at Specific Time Points by Sample Source Amster- Miami Controls Gote- Gote- Illinois North- Rotterdam berg D berg V western dam Amsterdam 2 2 Miami 2, 18, 36, 36, 60, , 84, 120 Controls 2 2, 18, 36, 18, 36, 60, 18, 36, 60, 18, 36, 60 2, 36, 60, 2, 8, 18, 36 60, 84, , , 120 Goteberg D 18, 36, 60, 36, Goteberg V 36, 60, 84 18, 36, 60, 36, 60 36, , 120 Illinois 18, 36, , North , 36, 60, western 84, 120 Rotterdam 2, 8, 18, nent molars. The other feature is that the Goteborg- Vomer group seems to be consistently low over the period of months Growth Velocity in the Unilateral Cases (Fig. 17.5) In order to compute the rate of growth (velocity) of Total Surface Area, the slope of the growth between time points was computed for each subject by dividing difference in area by the difference in time. The slope was then assigned the midpoint of the two time observations for analysis and plotting. The Northwestern and Illinois groups show the highest initial velocities, with Amsterdam, Miami, Goteborg-Vomer, and Rotterdam showing very similar intermediate velocities. The Goteborg-Delayed group and the Controls show very similar but lower initial velocities. Interestingly, the Controls show a slight increase in velocity between week 8 and week 18. Between 18 and 60 months the Vomer group drops in velocity and stays constant. The other groups level off and stay

7 Chapter 17 Timing of Cleft Palate Closure Should Be Based 369 Fig Unilateral cleft palate data Fig Unilateral cleft palate data Fig Unilateral cleft palate data Fig Unilateral cleft palate data approximately constant after 36 months. The plot of the mean velocities is shown in Fig Some of the purposes of the Amsterdam and Rotterdam studies were to investigate growth velocity before surgery. The data on the velocity changes are plotted in an expanded graph in Figs and This plot substantiates the observations made above. These data show that the Controls have a substantially higher growth velocity than the other groups after the very early period. The Vomer group shows consistently low velocities. Interestingly, the Amsterdam and Rotterdam groups do not differ from Miami, Goteborg-Delayed,Illinois,and Northwestern.The velocity data showed rapid growth early except for the Controls and the Goteborg-Delayed groups. During the period 8 18 months, the Amsterdam, Miami, Goteborg-Delayed, Illinois, and Rotterdam groups came together and showed constant growth thereafter. The Goteborg-Vomer group and the Northwestern group both showed a lower growth velocity than did the other groups.after the period of 8 18 months, the controls showed a consistently higher growth velocity.

8 370 S. Berkowitz Fig Unilateral cleft palate data Fig Unilateral cleft palate data Comparison of Unilateral Posterior Cleft Areas (Figs. 17.8, 17.9) The outstanding feature of these data is that Illinois started out lower than the others and Northwestern started out higher than the others, but neither showed decreasing posterior cleft areas after 8 months. They actually increased in size to 18 months then slowly decreased. The other groups showed patterns similar to each other, becoming smaller in size at different rates and to different degrees Comparisons of the Ratio of Posterior Cleft Area to Total Surface Area in Unilateral Cases (Fig. 17.9) At month 18 there were no significant differences between groups. These data have almost exactly the same pattern as the Posterior Cleft Space plots shown in Fig Tracking of the Large and Small Segments in Unilateral Cases (Fig a h) The large and small segments tracked each other closely in all groups. Plots of the two segments for each unilateral group are found in the full-page plots. In early adulthood, although the data are sparse in this region, it appears that all four groups achieve similar total palate surface areas. The clinical decision to effect a relative early (18 24 months) closure of the unilateral cleft based on the Miami model appears to have had the same results as delaying the closure in the Goteborg-Delayed series until approximately 5 9 years of age. Both groups show a tendency for growth to start increasing at 5 6 years, coinciding with permanent molar development, and continue through adolescence and early adulthood.

9 Chapter 17 Timing of Cleft Palate Closure Should Be Based 371 a b c d e f Fig a h. Unilateral cleft palate data. a Goteborg Vomer Sample. b Goteborg Delayed Sample. c Northwestern Sample. d Amsterdam Sample. e Illinois Sample. f Miami Sample

10 372 S. Berkowitz g h Fig a h. (continued) g Rotterdam Sample. h Control Sample Table Surface Area Growth Velocity (Mean 1 Std. Error, Sample Size) of Unilateral Clefts by Data Source and Age in Months Months Source Amsterdam 31.0 ± ± ± 0.7 a (33) (34) (31) Miami 30.9 ± ± ± ± ± ± ± 0.6 (22) (19) (23) (28) (26) (26) (16) Controls 19.2 ± ± ± ± ± ± ± 0.5 (10) (12) (11) (13) (15) (12) (7) Goteberg (D) 18.9 ± ± ± ± ± ± ± 0.6 (22) (17) (16) (17) (18) (20) (23) Goteberg (V) 29.4 ± ± ± ± ± ± ± 0.6 (21) (18) (16) (17) (18) (22) (20) Illinois 37.2 ± ± ± ± ± ± ± 1.5 (7) (10) (9) (6) (4) (5) (4) Northwestern 48.1 ± ± ± ± ± ± ± 0.4 (7) (5) (4) (5) (3) (4) (4) Rotterdam 29.2 ± ± ± 0.4 a (50) (58) (58) a Estimate made at 33 month Comparisons of Surface Area in the Bilateral Cases (Table 17.5) The statistical methodology for the bilateral comparisons was the same as that for the unilateral comparisons. The mean values for the bilateral cases are shown in Table Clearly, the sample sizes are smaller than in the unilateral cases, but the standard errors reflect statistically stable observations. A plot of these values is shown in Fig The most striking feature of these data is that the Control group and the Miami group are behaving similarly with respect to the other groups, although the Miami group shows lower growth until approximately 84 months. To show early differences more clearly, an expanded plot is shown in Figs and These plots show clearly that the two Goteborg groups start off slowly.

11 Chapter 17 Timing of Cleft Palate Closure Should Be Based 373 Fig Bilateral cleft palate data Fig Bilateral cleft palate data Fig Bilateral cleft palate data Fig Early growth velocity by group and time Growth Velocity in the Bilateral Cases (Fig ) The Miami and Nijmegen groups show the highest initial velocities, with Goteborg-Delayed and Goteborg-Vomer showing very similar intermediate velocities. The Amsterdam group and the Controls show very similar but lower initial velocities. Interesting, the Controls show a slight increase in velocity between week 8 and week 18. Overall growth velocities are shown in Fig This plot substantiates the observations made above. Figure shows that there is very little difference in growth velocities among the bilateral cases Comparison of Bilateral Posterior Cleft Areas (Fig ) The posterior cleft areas up to 36 months are shown in Fig All the groups had similar patterns for the closure of the posterior cleft space Comparisons of the Ratio of Posterior to Total Surface Area in Bilateral Cases (Fig ) At month 8 Nijmegen is significantly different from each of the other groups. There are no further statistically significant differences. These data have almost exactly the same pattern as the Posterior Cleft Space plots shown in Fig

12 374 S. Berkowitz Conclusions for the Bilateral Series The different practices showed similar results in palatal growth in that growth curves paralleled controls, but never entirely caught up. The only exception was that the Miami bilateral group did overtake the Controls. Early (18 24 months) or later closure had similar results Clinical Significance of the Results Fig Bilateral cleft palate data Fig Bilateral cleft palate data Bilateral Cases: The Bilateral total surface growth curves revealed that both Goteborg groups grew slowly throughout the complete time period while the Nijmegen group showed slowed growth after 60 months. The Miami group showed a pattern similar to the Controls but with somewhat lower total growth between 18 and 84 months. The velocity profiles for the bilateral cases were very similar to those for the unilateral cases. The closure of the posterior cleft spaces and the ratio of posterior cleft space to palatal surface area before surgery and to total surface area after surgery was also similar to that in the unilaterals. The only noteworthy feature was that the Nijmegen group had larger posterior cleft spaces and consequently higher ratios from month 3 through month 36. This may be due to the longer and more constant use of the presurgical orthopedic appliance which prevented the palatal segments from moving together (molding). This study highlights that differences in palatal osteogenesis is reflected in differences in cleft space size at the same age at birth and for the next 12 months. The vomer series, where the palatal cleft was closed before 1 year followed by a velar pushback, showed that extensive pushback procedures interfered with palatal growth. This finding was reported by the clinic staff and was the reason for discontinuing this surgical protocol and changing to delayed closure without vomerine flaps and palatal pushbacks. The small modified vomer flap with a von Langenbeck procedure used in the Miami series did not cause palatal growth retardation or cause an excessive number of posterior crossbites. CBCLP cases appear to have more osteogenic deficiency than CUCLP cases at birth. All cleft palates are smaller than noncleft hard palates (controls). There is more than one physiological surgical procedure which can be used to close palatal clefts. In all cases the most common age for the smallest velocity of growth was between 18 to 24 months of age. The best time to close the palatal cleft is when palatal growth change has significantly slowed down so that cellular growth activity can proceed without interference. Physiological surgery (the procedure that interferes least with normal cellular activity) allows for catch-up palatal growth. It is now possible to mathematically determine the best time to close the palatal cleft, other than have it based solely on the age of the patient. It appears that the effect of cleft closure surgery involves only that part of the palate anterior to the first permanent molars. Subsequent palatal growth is necessary to accommodate the second and third molars and this area seems to grow independently of the effect from early palatal surgery. The lateral palatal segments in the CBCLP and CUCLP cases grow at a similar rate.

13 Chapter 17 Timing of Cleft Palate Closure Should Be Based Discussion Recent trends in the study of human biology generally and cleft palate research in particular exhibit an increasingly emphatic recognition of an important fact: that mass cross-sectional studies of large groups can be significantly less valuable than studies made of groups of single individuals over lengthy periods of time. Mass studies tend to smooth out significant individual differences and to obscure them; while consistent and prolonged serial studies focusing on individual areas tend to emphasize such variations by bringing them out in high relief. To date, investigating the growth of the palate has been studied only so far as it is displayed in the superficial soft structures overlying the bony palate. All measurements in this paper, therefore, unless otherwise stated, refer to the soft structures of the palate and not to the skeletal palate. For purposes of convenience, whenever necessary, we shall hereafter in our series speak of the fleshy palate as distinguished from the bony palate. The design of this study would have been better if it included test cases treated with the same palatal surgery but at 6 12 months as well as at months of age. This comparative growth study would demonstrate the importance of using surgery based on the size of the cleft space relative to total palatal size and eliminate any resultant growth differences due to the type of surgery used. Unfortunately, we only had a small sample of these early closure cases. Since they developed excessive scarring with poor dental occlusion and midfacial growth, we did not accrue enough cases to perform valid statistical comparisons. Knowing that many organ systems in neonates grow very rapidly within the first 24 months, it was decided to use the same range of 12 to 24 months as the suspected ideal period for palatal cleft closure. A modified vomer flap was added to the von Langenbeck procedure to create a normal vault space. Since some cases had extremely large cleft palate spaces that reflected an excessively high degree of osteogenic deficiency, we focused on this factor rather than the age of the patient to avoid creating excessive scarring. The palatal growth velocity measurements showed that our reasoning as to when best to perform palatal surgery was supported by the subsequent analyses. Surgical Goals: In 1938 Kilner listed the primary objectives of cleft palate treatment in order of importance to be speech, followed by chewing and aesthetics [1]. Unfortunately, this priority of goals still seems to be preferred by many plastic surgeons due to the influence of speech-language pathologists who fear the consequences of an open palate. Thus, they favor palatal cleft closure before 1 year of age.as a result,the surgical history of cleft palate repair is replete with varied attempts to close the cleft space as if it was a stagnant hole, with minimal concern as to the surgical effects on palatal and facial growth [2]. Clinicians in all specialties have criticized the poor long-term aesthetic and dental occlusion results created by nonphysiological surgical procedures (as to type and timing) that did too much mucoperiosteal undermining too soon, creating excessive scarring. Unfortunately, the speech argument for early palatal cleft closure before 12 months still prevails and with it the hunt for the magical cut that will answer all aesthetic problems at birth and in the future. Slaughter and Brodie [3], commenting on poor midfacial growth, stressed that reduction in blood supply and constriction by scars would jeopardize palatal growth, yet this message was disregarded. In addition, they stated that unwarranted trauma to hard and soft tissue, due to the fracturing of bone, and stripping of mucoperiosteum, would cause permanent damage to growth sites that were active until 5 years of age. There was no criticism of the timing of surgery, only the procedures being performed. Not having appropriate records, they could not relate the surgical outcome to the disproportionately small palatal segments to cleft size that existed before 1 year of age [4]. What surgical procedures to use and when to close the palatal cleft were questions that had no universally acceptable answers.while surgeons such as Veau [5] and Brophy [6] believed that early closure of the cleft palate improved speech development, there were many who took a contrary position. Koberg and Koblin [7] favored palatal cleft closure between 2 and 3 years of age to achieve both good midfacial growth and speech. Delaying Palatal Closure: There were many European cleft palate clinics which emphasized in the 1960s to the present time the achievement of good midfacial, palatal development and speech. They recommended that cleft palate closure be postponed until the eruption of either the deciduous or as late as the permanent dentition [8 13]. Why Presurgical Orthopedics: The rationale of the four European centers in this study for using presurgical orthopedics within 2 weeks after birth is based on their speculation that this treatment may aid speech development and feeding. The Goteborg clinic uses the PSO appliance solely as an obturator to aid feeding. The Goteborg, Amsterdam, Rotterdam, and Nijmegen centers usually performed delayed, staged palatal closure between 5 and 9 years of age.children s Memorial Medical Center, Northwestern University Cleft Palate Institute, employs presurgical orthope-

14 376 S. Berkowitz dics with primary bone grafting while closing the palatal cleft at approximately 1 year of age. In some instances, PSO is also used to manipulate the lateral palatal segments to aid surgical closure of the lips and establish an alveolar butt relationship prior to primary bone grafting. Advocates of delayed palatal closure (5+ years) wanted to avoid secondary malformations of the palate and severe deformities of the maxilla caused by scarring created by extensive mucoperiosteal undermining with transpositioning of the tissue. This occurred when surgery created exposed lateral areas of denuded bone due to stripping off of the overlying mucoperiosteum. Similar malformations were created in animals by Kremenak et al. [14]. To avoid the consequences of early surgery, Hotz [9] and Weil [12] advocated the use of an obturator in the interim until additional palatal growth occurred which reduced cleft space width. They believed that postponing palatal cleft closure would not jeopardize speech development. Bzoch [20] stated that early speech therapy, between 1 and 3 years of age, corrected early speech problems. Speech language-pathologists and surgeons have mistakenly disregarded the possibility that significant speech problems can be corrected with therapy. Many benefits were claimed for the use of presurgical orthopedics, such as the stimulation of palatal growth, aiding speech development, and the reduction of middle ear disease. However, in a state-of-theart report on oro-facial growth, no supporting literature was reported [15]. The same failure of having any supporting literature is in evidence even today. The findings by the Amsterdam, Nijmegen, and Rotterdam prospective studies, which used presurgical orthopedics (PSOT) for 30-plus years, state that PSOT has a very limited effect on feeding, and have recently concluded that it has no lasting effect on palatal arch form. Therefore, the cost/benefit ratio may not warrant its further use [16]. They are beginning to question whether earlier palatal surgery may be warranted. Berkowitz [17] believes the timing pendulums had swung too far to the opposite extremes: from early to very late closure, and is now swinging back again to early closure, between 6 to 12 months of age. Berkowitz [18] states that those who favor either extreme timing periods are still not focusing, as Berkowitz and Millard did, on the size of the cleft defect but continue to be fixated on the patient s age alone. Some surgeons have found a middle of the road treatment plan, around months of age, but not on the size of the cleft space Good Speech Is Dependent on a Normal Palate Sally Peterson-Falzone et al. [19] have written that malocclusion needs to be considered during the early speech learning years. She points out that the dental and orthodontic literature contains fairly consistent information regarding the effects of dental problems and malocclusions on speech. A Need for Differential Diagnosis and Treatment Planning: A careful review of cleft palate surgical history makes it clear that a single mode of surgery based on age alone for all cases frequently results in severe palatal and midfacial deformities as well as poor speech development. In general, this literature tells us that dental and occlusal problems are more likely to be causative factors in speech problems (1) when they occur in combination rather than singly, (2) when they are present during the speech-learning years as opposed to later years, and (3) when they influence the spatial relationship between the tip of the tongue and the incisors [18]. The literature also indicates that speech problems are fairly common when there is a restriction in the size of the palatal vault which is more apt to be found in Class III occlusions compared with Class II [18]. Children with clefts are obviously vulnerable to restriction in size of the palatal vault, and the possibility of Class III occlusions due to the presence of dental or occlusal problems (possibly several at one time) during the speech learning years. The question is: Will the speech problems diminish as the dentition or occlusion improves? This statement convincingly acknowledges that good speech development is contingent on good tongue teeth relationships within a normal vault space of proper volume. This study has demonstrated that a scientific basis for selecting the best time to close the palatal cleft, in both CUCLP and CBCLP, is when the cleft space surface area is 10% or less of the surrounding palatal surface area bounded by the alveolar ridges. Early palatal surgery (before 1 year of age in most instances) may not always jeopardize palatal and facial development provided conservative surgical methods are employed when the cleft space is sufficiently small. The overarching thesis of this report favors consideration of the total emotional and physical health of the child with a cleft, based on the desired attainment of a cosmetically attractive face, and adequate dental function and respiration, as well as speech. Many sur-

15 Chapter 17 Timing of Cleft Palate Closure Should Be Based 377 gical, medical, and dental therapies may be necessary in the best-treated cases. As long as the surgeon individualizes the treatment plan, taking care to do no harm to growing structures, all goals are obtainable Conclusions 1. When the ratio of posterior cleft space to the total palatal surface area medial to the alveolar ridges is no more than 10%, it is the best time to surgically close the palatal cleft space. Therefore, one need not wait until 5 9 years of age to close the cleft space in order to maximize palatal growth. 2. Presurgical orthopedics does not stimulate palatal growth beyond its normal growth potential. 3. There is more than one physiological surgical procedure to achieve good palatal growth. 4. Extensive velar flaps with or without palatal pushback surgery is detrimental to palatal growth Participating Treatment Programs and Co-Investigators Principal Investigator: Miami Craniofacial Anomalies Foundation, South Florida Cleft Palate Clinic, Samuel Berkowitz, D.D.S., M.S., F.I.C.D. Co investigators: University of Miami, School of Medicine, Robert Duncan, MD. Center for Craniofacial Anomalies, University of Illinois College of Medicine, Carla Evans., D.D.S., D.M.Sc. Childrens s Memorial Medical Center, Northwestern University Cleft Palate Institute, Sheldon Rosensteim, D.D.S., M.S.D. Cleft Palate Center, Sahlgrenska University Hospital, Goteborg Sweden, Hans Friede, D.D.S., O.DR. University Hospital of Nijmegen Cleft Palate Center, Anne Marie Kuijpers-Jagtman, D.D.S., PhD. Free University of Amsterdam Cleft Palate Center, Birte Prahl-Andersen, D.D.S, PhD. Academic Hospital (Dijksigt/Sophia) Rotterdam Cleft Palate Center, M.L.M. Mobers, D.D.S References 1. Millard DR. Cleft craft: the evolution of its surgery, Part I: the unilateral deformity. Boston: Little Brown; Kaplan EN. Cleft palate repair at three months. Ann Plast Surg 1981; 7: Slaughter WB, Brodie AG. Facial clefts and their surgical management in view of recent research. Plast Reconstr Surg 1949; 4: Maisels DO. The timing of various operations required for complete alveolar clefts and their influence on facial growth. B J Plast Surg 1966; 20: Veau V. Le Sequalette due bec-de-lievre. Ann Anat Pathol 1934; 11: Brophy TW. Cleft lip and palate. Philadelphia: Blakiston s; p Koberg W., Koblin I. Speech development and maxillary growth in relation to technique and timing of palatoplasty J: Maxillofac Surg. 1973; 1: Hotz M, Gnoinski W. Comprehensive care of cleft lip and cleft palate children at Zurich University: a preliminary report. Am J Orthod 1976; 70: Hotz MM. 22 years of experience in cleft palate management and its consequences for treatment planning. In: Kherer B, Slongo T, Graf B, et al. (eds.) Long-term treatment in cleft lip and palate with coordinated approach. Bern: Hans Huber; p Schweckendiek W. Two stage closure of cleft palate. Rationale for its use. In: Kherer B, Slongo T, Graf B, et al. (eds.) Long-term treatment in cleft lip and palate with coordinated approach. Bern: Hans Huber; 1979a. p Schweckendiek W: Speech development after two stage closure of cleft palate. In: Kherer B, Slongo T, Graf B, et al. (eds.) Long-term treatment in cleft lip and palate with coordinated approach. Bern: Hans Huber; 1979b. p Weil J. Orthopaedic growth guidance and stimulation for patients with cleft lip and palate. Scand J Plast Reconstr Surg 1987; 21: Friede H. Growth sites and growth mechanisms at risk in cleft lip and palate. Acta Odontol Scand 1998; 56: Kremenak CR, Huffman WC, Olin WM. Maxillary growth inhibition by mucoperiosteal denudation of palatal shelf bone in noncleft beagles. Cleft Palate J 1970; 7: Berkowitz S. State of the art in cleft palate, orofacial growth. Am J Orthodont1978; 74:5, Prahl C, Kuijpers-Jagtman AM,Van t Hof MA, Prahl-Andersen B.A randomized prospective clinical trial into the effect of infant orthopaedics on maxillary arch dimensions in unilateral cleft lip and palate. Eur J Oral Sci 2001; 109: Berkowitz S. Cleft lip and palate and craniofacial anomalies perspectives in management. San Diego: Singular Press; 1996b. 18. Berkowitz S. Timing cleft palate closure-age should not be the sole determinant. J Craniofac Genet Devel Biol 1985; (Suppl.) 1: Peterson-Falzone SJ, Hardin-Jones MA, Karnell MP. Cleft palate speech. 3rd edn. St. Louis: Mosby; p Bzoch KR. Clinical studies of the efficacy of speech appliances compares to pharyngeal flap surgery. Cleft Palate J 1964; 35: