**5. Postoperative speech results comparing to the previously operated patients and noncleft controls**

Postoperative speech results of 94 patients who underwent palate repair based on our surgical strategy during 2006–2012 (strategy group) and those of 109 patients who previously underwent palate repair without following strategy during 2000–2005 (previous group) were compared. As control group, speech data on 37 Japanese noncleft controls were used. For speech assessment, perceptual rating of hypernasality and nasal emission was classified into four categories: none, slight/mild, moderate, and severe, by one experienced speech language therapist for all participants. Articulation was also evaluated using the articulation test. For objective assessment, Nasometer test was performed for all patients. This study was approved by the Clinical Research Ethical Review Boards of Kagoshima University Hospital.

Comparison of the rate of achieving normal resonance in each cleft type is shown in **Figure 5**. Normal resonance was achieved in 35/37 (94.6%) in Unilateral cleft lip and palate (UCLP), 15/18 (83.3%) in Bilateral cleft lip and palate (BCLP), 24/27 (88.9%) in CP, and 8/12 (66.7%) in Submucous cleft palate (SMCP) in the strategy group. Severe hypernasality was observed in each one patient with BCLP and SMCP. On the other hand, normal resonance was achieved in 40/57 (70.2%) in UCLP, 16/25 (64.0%) in BCLP, and 19/27 (70.3%) in CP in the previous group. Successful achievement of normal resonance was obtained more reliably in all types of CP following palate repair based on our surgical strategy.

 **Figure 5.** Postoperative hypernasality of each cleft type in the strategy and previous group.

The mean and SD of the nasalance scores of the strategy and previous groups and controls are shown in **Table 2**. The mean nasalance scores in the strategy group were less than 20% and were significantly lower than those of the previous group. When comparing the nasalance scores of control groups, those in the previous group were significantly higher on phonating /i:/ and the low-pressure sentence than in controls. On the other hand, there was no significant difference between the strategy and control groups. In other words, the nasalance scores representing hypernasality in the subjects of the strategy group recovered to almost the same levels as those of Japanese children without cleft palate.

Regarding articulation at 4 years of age, normal articulation was obtained in 68.4% in the strategy group, and this was better than that of the previous group (**Figure 6**).


**Table 2.** Mean ± SD of the nasalance score in the strategy, previous, and control groups.

 **Figure 6.** Postoperative articulation in the strategy and previous groups.

#### **6. Discussion**

When considering the postoperative VPI following CP repair, there are several main causes, including a wide cleft, short palate, deep pharynx, and unsatisfactory muscle reconstruction, when syndromic conditions, hearing loss, and mental retardation are excluded (**Figure 7a, b**). The preoperative portion between the velar length and pharyngeal depth bilaterally often differed, especially in subject with UCLP whose major and minor segments dislocated anteroposteriorly. During palatal repair, Z-plasty was usually used for adjusting the velar length; however, complete closure of the mucosal defect by large Z-plasty sometimes moved the uvula forward remaining asymmetry of the uvula position and pharyngeal arches (**Figure 7c**). The authors thought that these asymmetries in the velopharyngeal form may disturb the symmetrical muscular approximation and cause different sizes of the velopharyngeal orifice, resulting in persistent VPI following palatal repair [15, 16]. Therefore, it is thought to be useful to add a mucosal graft on the nasal side to fill the mucosal defect and to avoid an asymmetric VP form that may facilitate symmetrical velar motion in the VP closure mechanism.

The mean and SD of the nasalance scores of the strategy and previous groups and controls are shown in **Table 2**. The mean nasalance scores in the strategy group were less than 20% and were significantly lower than those of the previous group. When comparing the nasalance scores of control groups, those in the previous group were significantly higher on phonating /i:/ and the low-pressure sentence than in controls. On the other hand, there was no significant difference between the strategy and control groups. In other words, the nasalance scores representing hypernasality in the subjects of the strategy group recovered to almost the same

Regarding articulation at 4 years of age, normal articulation was obtained in 68.4% in the

/i/ 20.3 ± 13.5 <0.01 33.6 ± 23.9 <0.01 22.7 ± 14.4 /tsu/ 16.8 ± 13.5 <0.05 22.6 ± 19.3 <0.05 15.2 ± 8.5 /youihaooi/ 19.7 ± 13.6 NS 24.2 ± 17.0 <0.01 13.0 ± 9.7 /kitsutsuki ga kiwo tsutsuku/ 19.2 ± 12.7 NS 23.6 ± 18.3 NS 17.5 ± 9.8

**Strategy (***n* **= 94) Previous (***n* **= 109) Controls (***n* **= 37)**

When considering the postoperative VPI following CP repair, there are several main causes, including a wide cleft, short palate, deep pharynx, and unsatisfactory muscle reconstruction,

strategy group, and this was better than that of the previous group (**Figure 6**).

**Nasalance score (%)**

**Table 2.** Mean ± SD of the nasalance score in the strategy, previous, and control groups.

 **Figure 6.** Postoperative articulation in the strategy and previous groups.

levels as those of Japanese children without cleft palate.

68 Designing Strategies for Cleft Lip and Palate Care

**6. Discussion**

**Figure 7.** The reasons for postoperative VPI following CP repair: (a) the wide cleft, (b) short palate, (c) asymmetry of the uvula position and pharyngeal arch.

Furthermore, in the authors' experience during endoscopic examination of patients with persistent VPI, an asymmetric pharyngeal form or movement of the velopharynx and the midline defect of the velum were often observed, and they might be critical causes of VP closure dysfunction. Regarding the midline defect of the velum, Kuehn and Perry also reported that a midline defect suggested the presence of a deficiency or lack of musculus uvulae tissue or unsatisfactory surgical repair of this muscle (**Figure 8**) [17]. The anatomy and functional significance of the uvular muscle for VP closure was described by Kuehn et al. [18]. The uvular muscle courses posteriorly from its origin along the midline of the velum near the nasal surface of the velum. It is in its most cohesive form in the area overlying, and cradled by, the levator sling. The uvular muscle adds bulk to the dorsal aspect of the velum, thereby helping to fill the area between the velum and posterior pharyngeal wall. Without such bulk, the dorsal region would be concave, rather than convex, demonstrating a midline defect in the velum. In these cases, complete VP closure would not be achieved [17].

Considering the above, to ensure complete VP closure on CP repair, it is important to construct a symmetrical and functional velopharynx. Therefore, the authors have established a surgical strategy for palatal repair focusing on sufficient lengthening of the nasal mucosa, retropositioning the palatal muscles to produce a symmetrical levator sling, and unionizing the palatal muscles with a certain width in the midline of the velum. In the result, the surgical strategy for palatal repair facilitates successful speech outcomes in almost the same levels as those of Japanese children without cleft palate. There was no description about speech results of the CP patients based on the successful achievement of postoperative Velopharyngeal closure function equal to normal children.

 **Figure 8.** The midline defect of the velum of endoscopic examination might be a critical cause of VP closure dysfunction.
