Abstract
This study determined the intelligibility (words, sentences and story telling) of 43 children (mean age 4.9 years) with unilateral cleft lip and palate who received a Wardill–Kilner palatoplasty, as judged by their parents and determined the influence of age and gender. A comparison with normative data for intelligibility of 163 Dutch speaking children was made. Each child and the parents completed The Dutch intelligibility test. Measures reported include group mean intelligibility percentages for words, sentences and story telling for children aged 2.5–5.0 and 5.0–7.6 years. A significant correlation between age of the children and intelligibility was measured. No significant gender-related differences were found. An interesting finding is the absence of significant difference in intelligibility percentage between the children with cleft palate and the normative data for story telling. The reported intelligibility percentages provide important prognostic reference information for surgeons who perform palatoplasty and speech pathologists who assess the speech production of children with cleft palate.
Evaluation of the efficacy of palatal surgery in patients with cleft palate and a determination of the long-term speech outcome is of interest to surgeons who treat cleft palates. The success of cleft palate surgery is determined by the subsequent speech intelligibility. The speech of children with cleft palate is often characterized by resonance, articulation and phonatory disorders. These three aspects can adversely affect overall speech intelligibility. Speech intelligibility can be defined as what is understood by the listener. It is the product of a series of interactive processes including articulation, phonation, resonance and prosody . Intelligibility has been described as the most important measure of speech disorder and increasing intelligibility as the primary goal of surgical and therapeutic intervention . The importance of assessing the interactive components of speech intelligibility is well recognized in the cleft palate population but only a few authors recommended the inclusion of intelligibility in a standard speech assessment. The success of cleft palate surgery in children is determined by the overall intelligibility and the subsequent speech characteristics.
Several types of listeners (surgeons, speech language pathologists, parents of children with or without cleft palate, cleft palate children) can be used to judge overall intelligibility. In a critical review of intelligibility studies of speakers with cleft palate W hitehill summarized the impact of different types of listeners. The majority of studies used listeners who were qualified speech language pathologists or graduate students. To the best of the authors’ knowledge only three studies used parent groups to judge the overall speech intelligibility of children with cleft palate. In the study of S tarr et al. 12 parents of children with cleft palate, 12 parents of children without cleft palate and several clinic/school clinicians rated the speech samples of 15 children (aged 8–21 years) with cleft palate using an equal-appearing interval and category scales. Five children were judged as having normal nasality, five had mild and five moderate hypernasality. Ratings for speech intelligibility were not mentioned. S traus et al. questioned 102 patients with cleft palate (aged 13–19 years) and their parents about satisfaction with a number of variables including speech intelligibility. Most patients (69%) were very pleased with their current speech status, though several patients rated themselves as only moderately understandable (19%) or not understandable (9%). Parents and children expressed increased satisfaction with speech as the children’s age increased. In the study of W itt et al. a parent group (32 parents of cleft palate subjects) rated the speech intelligibility of 20 children with cleft palate (mean age 10.6 years) using a 12-item rating scale with speech descriptors ranked on a binary (yes/no) scale. The results of this study show that parents of children with cleft palate may effectively screen for impaired speech in the cleft palate population. 69% found the speech of the children clear, 78% found the speech was not hard to understand and 71% reported normal speech. More details of these studies are provided in Table 1 .
| Authors | Subjects | Method | Listener group | Results |
|---|---|---|---|---|
| S tarr et al. | 15 children with CP (age range: 8–21 years) | Equal-appearing interval and category scales | 12 parents with and without CP children | Results of the parents of children with CP |
| Speech sample: 250 recorded word reading passage | Several clinic/school clinicians | Five children: normal nasality | ||
| Children with cleft palate and nasal speech | Five children: mild nasality | |||
| Five children: moderate nasality | ||||
| S trauss et al. | 102 children with CP (age range: 13–19 years) | Questionnaire | Results of the parents of their own children with CP | |
| The parents of the children with CP | 72% of the parents thought that their child did not have a speech problem | |||
| 19% thought that their child had a moderate problem | ||||
| 9% thought that their child had a severe speech problem | ||||
| W itt et al. | 20 children with CP (m.a.: 10.6 years) (age range: 8–12 years) (10 boys, 10 girls subjects excluded) | 12-item rating scale with speech descriptors ranked on a binary (yes/no) scale | Parents | Results of the parents of children with cleft palate |
| Isolated cleft lip | Speech sample: realistic variety of consonant phonemes of sufficient linguistic complexity to allow a comprehensive speech evaluation | Teachers | Speech is clear: 69% yes | |
| Major craniofacial syndrome | Subjects | Nose is stuffy: 70% no | ||
| Intellectual impairment | Speech pathologists | Not hard to understand: 78% no | ||
| Sensorineural hearing loss | Speech is hoarse: 76% no | |||
| Psychiatric disorder | Sounds too quiet: 82% no | |||
| Sounds too loud: 98% no | ||||
| Needs speech therapy: 66% no | ||||
| Speech is excellent: 67% no | ||||
| Speech is normal: 71% yes | ||||
| Speaks too fast: 78% no | ||||
| Speech is squeaky: 88% no | ||||
Comparing the results of these studies is difficult because different age and possibly gender groups, different methods and different listener groups were used. To the best of the authors’ knowledge only S traus et al. used the children’s own parents as listeners. Parental involvement in the assessment of intelligibility, especially in children with reduced speech intelligibility (such as cleft palate children) has several advantages. Parents are more familiar with their child’s phonetic repertoire and behavior, involvement may improve the parent’s knowledge of the child’s development and subsequent participation in intervention programs, parental assessment may save time and early identification of speech abnormalities may be encouraged as an attempt to prevent or minimize disabilities through appropriate interventions .
The main purpose of the present study was to determine overall speech intelligibility for words, sentences and story telling of children with unilateral cleft lip and palate who received a one-stage Wardill–Kilner approach as judged by their own parents and to determine the influence of age and gender of the children on the judgements. The possible influence of the gender of the listeners was also studied and comparison with normative data for parental intelligibility judgements of normal Dutch speaking children was made. This study was approved by the human subjects committee of the University Gent.
Methods
53 children were selected to participate in this study. None had a cleft associated with a syndrome, had undergone secondary pharyngeal surgery, had a cognitive deficiency, neuromotor dysfunction or a hearing threshold above 20 dB in the worst ear. Each child was assessed by an otorhinolaryngologist who performed a complete ear, nose and throat examination to exclude nasal and hearing pathologies. The nasopharyngeal and laryngeal examination included indirect nasoendoscopy, laryngoscopy and macroscopic otoscopy. Hearing was assessed for both ears separately at standard audiometric frequencies. Children with a nasal pathology (septum deviation, polyposis and allergic rhinitis, each in 1 child) and those with a common cold on the day of testing (4 children) were excluded from the study. 43 children and their parents responded positively to participate in this intelligibility study. They were 28 boys and 15 girls, ranging from 2.5 to 7.6 years (mean 4.9 years). All children had a unilateral cleft lip and palate. To measure a possible age effect, the children were divided into two age categories (group 1: 2.5–5.0 years, n = 25; group 2: 5.0–7.6 years, n = 18).
All patients consulted the same craniofacial team and had undergone an identical surgical protocol. Surgical closure of the lip was performed at an average age of 5.4 months (range 3–5.11 months). The cleft had been closed using one-stage Wardill–Kilner palatoplasty at an average age of 13 months (range 11–24.12 months). After infiltration with levorinine solution (1 ml/100 cm 3 saline) an incision is made along the margins of the cleft at the junction between oral and nasal mucosa. Lateral incisions are made just behind the teeth and extending up to the area of the canine tooth (von Langenbeck). The lateral incision above and the medial incision are connected (Y-incision). After elevation of the mucoperiosteal flaps entirely, based solely on the greater palatine pedicle posteriorly and the releasing of the muscles of the posterior flaps, the entiral flaps are sutured over the midline. The palatal cleft is closed with pushback of the posterior flaps. Two large triangular areas of exposed bare bone are seen. This heals secondarily. All patients have been operated on by the same surgeon (K.B.) and with the same surgical technique in the same conditions. 18 children had speech therapy for a minimum of 6 months, twice a week. The goals of speech therapy were to establish correct phonetic placement and to eliminate compensatory articulations or developmental errors.
The Dutch intelligibility test ‘Percentage Spraakverstaanbaarheid bij kinderen ’ (meaning Percentage Speech Intelligibility in children) was used to measure the speech intelligibility. This test, with a test–retest reliability of 88% , consists of three subtests: naming words, producing sentences and telling a story ( Fig. 1 ). In the first and second subtests, 25 words and 7 sentences appropriate to a specific picture are elicited, respectively. In the third subtest, the child tells a short story using four illustrations. The speech samples were audio recorded (Sony ICD-SX56) and subsequently presented to the parents (laptop, Toshiba M70 with sound card and acoustic amplifier) who were asked to write down what they understood of the phonetic realisation of their child’s speech production. Afterward an intelligibility percentage for words, sentences and story telling, as described by S hriberg et al. was calculated.
