Oral Sessions

Oral Sessions

Brain & Development 26 (2004) S21–S37 www.elsevier.com/locate/braindev Oral Sessions Oral Session I Child neurology neuroimaging findings was negati...

183KB Sizes 10 Downloads 79 Views

Brain & Development 26 (2004) S21–S37 www.elsevier.com/locate/braindev

Oral Sessions Oral Session I Child neurology

neuroimaging findings was negatively related to the occurrence of epilepsy.

O1 Epilepsy in children with congenital hemiplegia: correlation with clinical characteristics

O2 Autistic features in children with congenital cytomegalovirus infection

E. Pavlidoua,p, M. Tzitiridoua, E. Karatzaa, A. Kovanisb, C. Panteliadisa a 3rd Dept. of Paediatrics, Aristotle University of Thessaloniki, Greece; bDepart. of Paediatric Neurology, Agia Sophia Children Hospital, Athens

V. Mejasˇki-Bosˇnjak, L. Lujic´, J. Bosˇnjak, N. Jakusˇic´, I. Boric´ Children’s Hospital Zagreb, Dpt. Pediatrics, University Medical School, Zagreb, Croatia

Objectives: the aim of this retrospective, multicenter study was to investigate the relationship between epilepsy and clinical characteristics in children with congenital hemiplegia (CH). Patients and methods: two hundred and three children with CH were assessed by history, neurological and developmental examination. Electroencephalogram (EEG) and CT/MRI brain imaging were performed in 150 of them (82/150 had an MRI and 68/150 had a CT scan). Patients were re-evaluated every six months for, at least, a two-year follow-up period (range 2– 14 yrs). Results: The prevalence of epilepsy in 12/62 patients (19.4%) with mild hemiplegia was significantly lower as compared to 67/141 (47.5%) of patients with moderate or severe hemiplegia. 36.7% of the children had their first seizure between the 1st and the 5th year of life, and 26.5% during the first year of life. Complex partial seizures are the most frequent type of seizure (32.9%) followed by generalized tonic – clonic (20.3%), myoclonic – astatic (15.2%), simple partial seizures (12.7%) and other less frequent types (19.0%). EEG was abnormal in 76% of patients; epileptic seizures developed in 38.9% of them. The frequency of epilepsy paralleled the degree of EEG abnormality, approaching 85% in patients with severe EEG abnormalities and was also closely related to the extent of neuroimaging findings. Conclusions: epileptic seizures developed in more than one third of patients with CH, although EEG abnormalities were evident in the majority of them. The prevalence of epilepsy is closely related to the severity of hemiparesis. The absence of EEG abnormalities and/or normal (or minor) doi:10.1016/j.braindev.2004.03.003

Congenital cytomegalovirus (CMV) infection is the most common infective cause of neurodevelopmental disorders in childhood, the most prominent features being microcephaly, mental retardation, cerebral palsy, epilepsy, sensorineural hearing and visual impairment. However, in a few studies autistic problems of children suffering congenital CMV infection were additionally reported. The aim of this report is to present clinical features including autistic behaviour and neuroimaging in two children with congenital CMV infection. Patient 1, preterm neonate presenting at neonatal age as symptomatic, sepsis-like, CMV infection and microcephaly. Subsequent follow up revealed hearing loss, psychomotor delay and autistic behaviour i.e. lacking interest in people, interaction with peers, eye-to-eye gaze, responsivness and speech development and sterotypia. Patient 2 was microcephalic neonate with intrauterine growth retardation. Evaluation for psychomotor delay at age of 8 months, detected additionally hearing loss, poor interest for people, with avoidance of eye-to-eye contact and stereotipia. In both children CMV infection was proven by serologic specific IgG/IgM investigation and presence of CMV-DNA in serum using PCR. MRI of the brain showed in both children widespread leukoencephalopathy and cysts on the tips of the temporal lobes but no migration/gyral abnormalities. These findings suggest that the timing of injury to the developing brain by CMV may be in the third trimester in our patients with autistic disorder. O3 Poistron emission tomography in the assessment of learning disabilities

S22

Abstracts

L. Hsiu-Fena,p, C. Ching-Shianga, M. Suk-Chuna, L. RenShyanb a Department of Pediatrics, Taichung Veterans General Hospital; bDepartment of Nuclear medicine, Taipei Veterans General Hospital Introduction: To date, a number of challenges persist with neuroimaging and metabolic imaging techniques in the assessment of learning disabilities (LD). The utilization of poistron emission tomography (PET) scan for the evaluation of children with LD was studied during Apr. 1994 to May. 2000. Methods: Seventy patients, 2 – 18 years old, were recognized to have LD based on their parents, teachers interviews and DSM-IV. All patients performed static brain PET imaging at 45 minutes after intravenous injection of 3.3 mCi of [F-18] fluorodeoxyglucose. Intelligence quotient (IQ) testing [Wechsler Intelligence Scale for ChildrenRevised (WISC-R)] was also done. Results: Among 70 PET scan results, excluding 2 with inadequate data acquisition, overall abnormal rate of PET findings was 68/68 (100%). Forty (58.8%) performed IQ tests, 13 (32.5%) with mild to profound mental retardation, and 27 (67.5%) with borderline to superior intelligence. Both of PET findings of mental retardation and borderline to superior intelligence showed decreased radioactivity in temporal area and in the cerebellum. Among borderline to superior intelligence group, 4 (14.8%) had pure attention deficit and/or hyperactive disorders (ADHD) and 23 (85.2%) had special learning disability (SLD). PET scan findings were some differences between ADHD and SLD. Both of them revealed decreased radioactivity in the cerebellum. SLD had decreased radioactivity in left, right or bilateral mesial temporal area, while ADHD had this finding in bilateral temporal area. In the other hand, we found that 12 (52.2%) patients combined with ADHD among SLD group. Conclusion: PET scan is helpful in further evaluating the neurobiologic dysfunction in LD. O4 A randomized controlled trial to compare efficacy of oral clobazam with oral diazepam for prophylaxis of febrile seizures S. Gulatia,p, D. Sainia, R.M. Pandeyb, V. Kalraa a Departments of Pediatrics, All India Institute of Medical Sciences, New Delhi-110029, India; bDepartments of Biostatistics, All India Institute of Medical Sciences, New Delhi-110029, India Background: Febrile seizures are the most common convulsive events in childhood. Objectives: To compare the efficacy of intermittent prophylaxis with oral clobazam versus oral diazepam in

the reduction of seizure recurrence risk during subsequent febrile episodes in children with febrile seizures. Design, settings, and methods: Patients attending Pediatric OPD (October 2000 – December 2003) were studied. Inclusion Criteria: Children aged 6 –60 months with one or more episodes of febrile seizure. Exclusion Criteria: Children with neurological abnormalities, progressive neurological diseases, afebrile seizures, symptomatic seizures of other nature and seizures during a CNS infection. 150 children were randomized into two groups: Group 1: Oral diazepam 0.3 mg/kg/dose every 8 hrs for 3 days Group 2: Oral clobazam 1 mg/kg/day for 3 days. Drugs were to be administered during each febrile episode. Use of antipyretics and hydrotherapy was advised to both groups. Outcome: Seizure recurrence following a febrile episode, side effects of the therapy as well as afebrile seizures were appropriately recorded. Statistical analysis: Child was unit of randomization and febrile episode was unit of analysis. Bivariate analysis (Chi square and binary logistic regression) was done to determine the distribution of potential confounding factors at the child level. Factors found statistically significant were used in the Generalized Estimating Equation (GEE). As a child may have more than one febrile episode, therefore, the outcome (occurrence of seizure) of the febrile episodes for a child was correlated. Therefore, multivariate logistics regression analysis using Generalized Estimating Equation approach was used to compare treatment efficacy. STATA 8.0 statistical software was used for data analysis and p value , 0.05 was considered to be statistically significant. Results: 150 children (75 in diazepam group, and 75 in clobazam group) were studied. There were significantly more males in the diazepam group. The other basal characteristics of the two groups were similar which included mean age at presentation, median age at onset, gestational age at birth, mode of delivery, history of birth asphyxia, family history of febrile seizures, family history of epilepsy, number of febrile episodes/febrile seizures in last one year and number of complex febrile seizures. The median follow-up duration was 10 months in both groups during which 135 children (90%) had febrile episodes, 65 of them (86.7%) in diazepam group, and 70 (93.3%) in clobazam group. A total of 559 febrile episodes (254 and 305, respectively) were recorded of these, non-compliance was recorded for 13% and 16% febrile episodes respectively. It was due to fever being noticed just after onset of seizure: 2 febrile episodes in diazepam group and 3 febrile episodes in clobazam group. Sedation was noticed to be more in the clobazam group ðp ¼ 0:007Þ: The odds ratio of seizure recurrence in the diazepam group was 2.3 as compared to clobazam group. The febrile seizure recurrence during fever episodes adequately treated was 11.3% in the diazepam group, and 3.1% in clobazam group

Abstracts

ðp ¼ 0:0004Þ: Three patients in each group developed afebrile seizures. Conclusion: Oral clobazam was found to be more effective in preventing febrile seizure recurrence as compared to diazepam in children with history of at least one episode of febrile seizure. O5 Cerebral palsy and epilepsy Z. Modrusan-Mozetic, E. Paucic-Kirincic, I. Prpic University of Rijeka, Department of pediatric “Kantrida”, Rijeka, Croatia Convulsive phenomena are quite frequent in children with cerebral palsy. According to some authors, the prevalence ranges up to 50%. The diagnosis of epilepsy is reported in 30% to 40% of children with cerebral palsy. The analysis included children with cerebral palsy born during the 1982 –1992 period. Data were obtained from Registry of Risk and Disabled Children. There were a total of 84 children, which is consistent with literature data on the cerebral palsy prevalence of 2– 3/1000 children. There were 44 (52.3%) male and 40 (47.6) female children. In 31 (36.9%) of 84 children, epilepsy was diagnosed according to ILAE criteria. Results are in accordance with those reported elsewhere, and point to the need of systematic follow-up of children at a neurological risk.

Oral Session II Pediatric orthopedics O6 Gait analysis before and after lengthening of gastrocnemius fascia in children with cerebral palsy G. Albertinia, M. Gallib, V. Cimolinb, N. Tenorea, G. La Rosac, F. Costicic, M. Crivellinib a “San Raffaele-Pisana” Hospital-Tosinvest Sanita` , Rome, Italy; bBioeng. Dept., Polytechnic of Milan, Milan, Italy; c“Ospedale Pediatrico Bambin Gesu” Rome, Italy Introduction: Equinus deformity is a common problem of children with spastic Cerebral Palsy (CP). Treatment for static or dynamics equinus deformity has traditionally consisted of stretching the contracted tissue and persistent deformity that interferes with function is treated by surgical lengthening of the triceps surae. The aim of this study was the analysis of kinematic and kinetic effects of gastrocnemius fascia lengthening on gait pattern of children with CP and in particular the evaluation of push-off ability before and after treatment. Methods: 20 children (mean age: 8 years) with CP were evaluated by clinical examination and three-dimensional Gait Analysis (GA) before and 1 year after gastrocnemius

S23

fascia lengthening surgery (modified Vulpius’ technique). 10 healthy children (mean age: 9 years) was included as Control Group (CG). A 12-camera optoelectronic system with passive markers (ELITE, BTS S.p.A., Milan, Italy) working at a sampling rate of 100 Hz, to measure the kinematics of movement, two force platforms (Kistler, CH), to obtain the kinetic of movement and a Video system synchronic with optoelectronic and force platform systems.Clinical evaluation and GA data were collected before surgery and about 1 year after surgery. In this study the patterns of ankle and knee kinematic and of ankle kinetic were analysed and some parameters were defined in order to evaluate the effect of gastrocnemius fascia lengthening. Statistical analysis (Student’s t-test, p , 0:05) was used to compare pre- and post-operative states. Results: Before surgery children with CP were characterized by an evident equinus deformity; after gastrocnemius fascia lengthening, the reduction of equinus foot and a significant increase of ankle dorsiflexion during gait were noted. In most cases pre-operatively there was a pattern of abnormal generation of energy in midstance rather than a normal absorption; after surgery there was a reduction in this energy generation in midstance. This surgical approach didn’t result in weakening of triceps surae for push-off. The energy generated for push-off was, in fact, improved post-operatively. The analysis showed an improvement of knee extension during the stance phase, meaning that gastrocnemius fascia lengthening did not lead to production of crouch gait. The value of knee peak in swing improved post-operatively, showing that surgery didn’t give rise to excessive stiffness of knee in swing. Conclusions: The analysis demonstrated positive results from the modified version of Vulpius lengthening technique, concerning not only ankle joint but also knee joint. The improved prepositioning of the foot at initial contact and the incresed knee extension at the same istant of gait cycle produced a better weight acceptance through ankle and knee. This surgical approach didn’t result in weakening of triceps surae for push-off. The energy generated for push-off was, in fact, improved postoperatively. The analysis showed an improvement also in knee kinematic data and no crouch gait cases were found. O7 Relationship between ankle joint angle and foot-ground angle at initial contact for the classification of foot loading type N. Darrasa,p, D. Pasparakis b , M. Tziomaki a, C. Nestoridisa, G. Pistevosb a ELEPAP Gait Lab, Hellenic Society for Disabled Children, Kononos 16, Pagrati 11634, Athens, Greece; b2nd Orth. Dpt. “Aglaia Kyriakou”, Children Hospital, Thivon and Levadias, Goudi 11527, Athens, Greece

S24

Abstracts

Introduction: Initial contact (IC) influences the quality of the subsequent phases [1] in gait cycle. IC is characterized by the Angle of the Ankle Joint and the Angle between Foot and Ground. Usually IC is evaluated using only one of these two angles. This can lead to misinterpretations especially in cases with underlying pathology [2,3] (ie equinus foot). A more accurate classification is very important. The purpose of the study was to describe and classify the relationship between different Ankle Joint Angles and Foot-Ground Angles at IC, based on Gait Analysis data. Methods: 6 Subjects with various pathologies and 2 normal walking subjects have been selected for the study. 3D BIOKIN Motion Analysis System was used to capture a 19 points model, which included: Trunk, Pelvis, Thigh, Shank and Foot. Ankle Joint Angle was defined by the angle between the Shank axis and the Heel-3rd metatarsal axis. Positive values indicated Plantar Flexion, while negative values indicated Dorsiflexion. The Foot-Ground Angle was measured by calculating the angle of the Heel3rd Metatarsal axis to the ground level. Positive angles indicated Toe on the ground-Heel off the ground. Negative angles indicated, Heel on the Ground-Toe off the Ground. Results: 9 different types of Foot Loading at IC have been identified. (Class I) Ankle Joint at Neutral Position – – –

Foot Ground angle Negative (Normal) Foot Ground angle Neutral (Foot Flat) Foot Ground angle Positive (‘Apparent’ equinus) (Class II) Ankle Joint at Plantar Flexion

– – –

Foot Ground angle Negative (Normal – Fast Walking) Foot Ground angle Neutral (Foot Flat, Equinus?) Foot Ground angle Positive (True Equinus, Drop Foot) (Class III) Ankle Joint at Dorsiflexion

– – –

Foot Ground angle Negative (Calcaneus Gait) Foot Ground angle Neutral (Crouch) Foot Ground angle Positive (Apparent Equinus)

Conclusions: Considering the above classification, a better understanding of the relationship between shank-foot and foot-ground angles is achieved This classification, helps to reveal deceptive instances of equinus and describe better the underlying pathology. References [1] Gage, J.R. (1991) Gait Analysis in Cerebral Palsy. Clinics in Developmental Medicine, No 121. London. Mac Keith Press with Blackwell Scientific Publications Ltd.

[2] Winters TF, Gage JR, Hicks R. Gait patterns in spastic hemiplegia in children and young adults. J. Bone Joint Surg [Am], 1987;Vol 69-A, 437– 41. [3] Rodda J., Graham Hk. Classification of gait patterns in spastic hemiplegia and spastic diplegia: a basis for a management algorithm. Eur J Nurol, 2001;8 (Suppl 5): 98 – 108. O8 The importance of proximal medial hamstring release in the hip adduction deformity of patients with cerebral palsy A. Okan Nedretp, A. Bursali, M. Gorgec, A.N. Kara Haydarpasa Numune Training and Research Hospital and SSK Vakif Gureba Training and Research Hospital; Istanbul-Turkey Introduction: Adduction deformity of the hip is one of the most common pathologies seen in patients with Cerebral Palsy (CP). Adductor Brevis and Adductor Longus, together with Gracilis have been considered as the most responsible muscles for this deformity. Recognising the fact of unstable gait after anterior Obturator neurectomy and the release of Adductor Brevis, most surgeons have begun to preserve this muscle. But Adductor Longus tenotomy has still been widely accepted as a safe and useful procedure without side effects. However, Matsuo in his treatment approach for CP, in Orthopaedic Selective Spasticity-control Surgery (OSSCS), strictly recommended the preservation of Adductor Longus together with Adductor Brevis for gaining and preservation of gait stability [1]. He advocated the proximal release of medial Hamstrings (Semimembranosus and Gracilis) and distal section of Adductor Magnus to correct the hip adduction deformity in CP while increasing hip stability by preservation of Adductor Longus. In his approach, Adductor Longus is accepted as an important antigravity muscle. On the other hand, Gracilis, which adducts the hip and Semimembranosus, having a long tendon proximally which extends, internally rotates and adducts the hip are two multiarticular muscles without having antigravity activities around hip region. After releasing these two muscles proximally together with the distal cut of hamstring portion of Adductor Magnus according to the principles of OSSCS, we have seen that adduction deformity could be well corrected in our patients without losing power around hip region. Materials and methods: Between October 2001 to July 2003, we operated 40 CP patients according to the principles of OSSCS. Excluding 1 upper extremity, 1 scoliosis and 6 foot cases, 62 hips of 32 spastic CP patients (3 quadriplegic, 27 diplegic and 2 hemiplegic) underwent surgery. All these hips had adduction deformities with positive Phelps tests. Half of the cases had additional flexion contractures at the same region needing selective flexor releases and 5 of these hips needed open

Abstracts

reduction and bony procedures for severe subluxation and/or dislocation at the same stage of operation. Mean age at the time of surgery was 11,2 years and mean follow-up was 14 months. Adductor Longus muscles were preserved in all cases but only in 8 hips having scissors posture with flexion deformity and dislocation or severe to moderate subluxation, we added intramuscular lengthening of this muscle to the other procedures. In all hips, we sectioned Gracilis proximally, Adductor Magnus distally and lengthened or sectioned Semimembranosus tendon proximally. Results: In all patients, adduction deformities were well corrected with a mean increase of 28 degrees in abduction both with hip extended and flexed. Hygienic care could be given more easily by caregivers. Among 7 non-ambulatory cases, 4 diplegic ones have begun to walk with support. All the patients have gained at least one motor functional level. No complication and/or iatrogenic deformity was seen. Overall satisfaction among the patients and families were complete. Conclusion: Proximal medial Hamstring release combined with the release of hamstring portion of Adductor Magnus is a satisfactorily effective surgical procedure for the adduction deformity of the hip in CP patients without side effects. By this method, we could be able to preserve Adductor Longus together with Adductor Brevis and also preserve hip stability. Adductor Longus is a monoarticular muscle having antigravity activities so it should not be sacrificed easily. Reference [1] Takashi Matsuo: Cerebral Palsy: Spasticity-control and Orthopaedics—An Introduction to Orthopaedic Selective Spasticity-control Surgery (OSSCS)-, 1st ed. Soufusha, Tokyo; 2002. O9 The effect of ankle-foot orthoses on gait parameters in children with spastic cerebral palsy S. Hayeka, M. Kfirb, S. Khamisa, R. Batta, Z. Yizhara,b a Gait and Motion Analysis Laboratory, Pediatric Orthopedic department, Dana Children’s Hospital, TASMC; b Department of Physical Therapy, Sackler Faculty of Medicine, Tel-Aviv University Background: Ankle-foot orthosis (AFO) are frequently recommended to children with CP to improve their functional ambulatory ability, namely to increase walking velocity and stride length, and decrease cadence. Several studies examined the effectiveness of different types of AFOs based on gait analysis, however AFOs are usually prescribed upon clinical examination alone. Based on our first year experience at the Dana Gait Lab not all AFOs significantly improved function.

S25

Objective: To investigate the effect of prescribed AFOs on gait performance of children with cerebral palsy. Methods: Twenty two children diagnosed with cerebral palsy (age 5–17 y; gender: 11m, 11f; diagnosis: 8 hemiplegic, 11 diplegic, 3 quadiplegic) were referred for full 3D instrumented gait analysis for different purposes. Using the Vicon 612 system they were studied walking barefoot and with their prescribed AFOs. The two modes were compared in terms of spatio-temporal parameters and ankle–knee kinematics. Statistical analysis included paired t-test, Pearson correlation coefficient; level of significance was set to 0.05. Results: Using the prescribed AFOs, stride length was significantly increased (on average, 9.95 ^ 0.11 cm, p ¼ 0:000) while no significant changes were found in walking velocity and cadence (p ¼ 0:111; p ¼ 0:420; respectively). Split by diagnosis revealed significant reduction in cadence (12.7 step/min, p ¼ 0:034) in the hemiplegics children. There was no significant improvement in the symmetry index of the stride length and step time due to the use of AFOs. Ankle and knee kinematics at initial contact and at loading response revealed non-significant changes. In the hemiplegic group knee flexion increased significantly ðp ¼ 0:002Þ while ankle dorsiflexion at initial contact was almost significant ðp ¼ 0:3Þ: The consistency of the results within subjects and between modes was highly correlated ðr ¼ 0:858 2 0:928Þ and statistically very significant ðp , 0:000Þ: Discussion: In the current study, the only benefits of AFOs were increasing stride length and some improvement in ankle and knee kinematics. Our findings show that the use of an AFO, by itself, dose not dramatically change other walking parameters. The high correlation between barefoot and AFO modes suggests that the child’s basic capability is the main factor that affects the gait pattern. Inappropriate AFO may be partly the reason for the non-significant results in this study. We conclude that AFOs should be given only after optimization of the child’s physical capabilities. Prescription should be made after careful evaluation using gait analysis whenever possible. O10 Changes in coronal plane knee moments after surgical correction of external tibial torsion in patients with cerebral palsy M. Aiona, R. Pierce, R. Dorociak Shriners Hospitals for Children, Portland Introduction: External tibial rotation in patients with cerebral palsy can cause an awkward gait with compensatory gait motions. The normal knee valgus moment is often reversed to varus with an associated medial thrust of the knee. The hypothesis of this study is that correction of external tibial torsion in patients with cerebral palsy will correct the kinematic and kinetic gait abnormalities,

S26

Abstracts

specifically reduce the abnormal coronal plane force at the knee. Methods: A retrospective study of 20 patients (26 limbs) (mean age 15.4 ^ 3.7) with cerebral palsy who had an internal tibial rotational osteotomy and 3-D computerized gait analysis pre and at least one year postop. were reviewed. Statistical analysis by paired t-test was performed for each stance phase peak coronal plane knee moment with significance set at p , 0:01: Results: Tibial rotation significantly improved 19 degrees, with improvements in the first and second peak valgus moments in stance ðp , 0:0001Þ: All limbs had improved kinematics with 23 of 26 (88%) improved in one or both abnormal peak moments. Discussion: Correcting external tibial torsion improves gait kinematic and kinetic parameters with improved symptoms as the torque across the knee is reduced. Relating clinical symptoms to abnormal joint forces will refine treatment guidelines. O11 Variability of pressures and forces in normal children and children without hemiplegia E. Arvanitakisp, N. Panorios, M. Vlaxou, D. Tsirakos, A. Papavasiliou, D. Dimitriadis Orthopaedic and Neurology Departments of Penteli Children’s Hospital, Athens Greece The forces and pressures created during the gait cycle in normal individuals and in patients with kinetic disabilities have attracted the attention of many researchers over the years (Shaw et al. 1998, Cavanagh, et al. 1997). It has been observed that pressures and forces are differentiated between the patients and normal individuals. The Objective, of the present work is to compare the pressures and forces which are generated during the gait in hemiplegic children and normal individuals. More specifically, the study focuses on the variability of the values and pattern of the pressure as well as, the vertical force during the gait cycle between these two groups. Fourteen children with hemiplegia (six boys and eight girls aged from five year old to twelve years old. Mean age 9) and fourteen children without kinetic problems (seven boys and seven girls aged from six years old to eleven years old. Mean age 8) participated in the study. Forces and pressures were recorded using a pressure platform system (Lorain Engineering, Italy), with sampling frequency 1000 Hz. All children were asked to walk freely at normal speed along the platform for at least five strides. The preliminary results of this study indicate that: † The pattern of force and the values of forces and pressure presented higher variability in the group with hemiplegic children.

† Children without kinetic problems presented smaller variability in the force and the pattern values of the force and pressure. † The difference in the mean force value among hemiplegic children exceeds the 10% of the maximum force value at almost every phase of the gait, whereas the corresponding values were lower than 10% in the group with normal children. † Comparison between, the two groups revealed higher ariability in the force pattern in the experimental group. † Force derivatives (change of force over time) in hemiplegic children appear to be higher than in normal individuals. O12 The split tibialis posterior tendon transfer for correction of spastic varus hind foot M. Vlachou, N. Panorios, E. Arvanitakis, D. Dimitriadis Orthopaedic Department of the Pendeli Children’s Hospital, Athens This is a retrospective study of 38 feet (24 right and 14 left) in 33 ambulant patients (23 females and 10 males) with flexible spastic varus hind foot deformity. 28 patients presented unilateral and 5 bilateral involvement. The mean age at operation was 10.8 ys [range 4– 17], and the mean follow-up was 7.9 ys [range 2 – 14]. There were 20 hemiplegic feet, 11 feet in paraplegic, and 7 feet in quadriplegic patients. 18 feet had also equinus position, requiring concomitant Achilles cord lengthening. The surgical technique applied, was similar to the one described by Green and Griffin, following four skin incisions (two on either side of the foot and ankle). The evaluation of the results was carried out, using Kling and Kaufer’s clinical criteria. According to them, excellent results are graded when the child manages to walk with a plantigrade foot, without fixed or postural deformity, in a regular shoe, and have normal callocities. Patient and parents are pleased with the result, and no brace is required post-operatively. Results are graded good in children, who walk with less than 58 varus, valgus, or equinus posture of the hind foot, wear regular shoes without abnormal shoe wear, have normal callocities and were satisfied with the outcome. Feet with recurrent equinovarus deformity, or overcorrected into a valgus or calcaneovalgus deformity, are considered as poor results. In this group of patients there were 20 excellent, 14 good, and 4 poor feet. Discussion: Split posterior tibial tendon transfer is performed to balance the flexible spastic varus foot and is preferable to the tibialis posterior lengthening, as the muscle does not loose its power and therefore the possibility of a valgus or calcaneovalgus deformity is diminished. The results in our cases were in general satisfactory, as 34 out 38 feet (89,5%) were graded excellent and good. The feet

Abstracts

with poor results had a residual varus deformity, due to technical intraoperative errors.

Oral session III Developmental Medicine/Child Psychiatry O13 Successful prevention of hip dislocation and contractures in children with cerebral palsy first ten years experience of a population based prevention programme G. Ha¨ gglund, S. Andersson, H. Lauge Pedersen, E. Nordmark, L. Westbomp In 1994 a cerebral palsy register and a health-care programme, with the aim of preventing hip dislocation and severe contractures in children with cerebral palsy (CP) was initiated in southern Sweden. The programme includes regular inventories to reach all children with CP in the total population, a standardized screening by physio- and occupational therapists and X-ray screening. New techniques to reduce spasticity and dystonia were introduced: selective dorsal rhizotomy, continuous intrathecal baclofen infusion and botulinumtoxin treatment. A multidisciplinary approach to choose the best treatment for each child in each developmental period is essential. In Lund physicians, physio- and occupational therapists from different disciplines such as child orthopaedic and hand surgery, neurosurgery and neuropaediatrics form a central spasticity team which co-operates with the local child habilitation services. We have found that the common structured and standardized follow-up programme with early intervention against muscle hypertonia and imbalance can prevent hip luxation and contractures in children with CP. The total population of children with CP born $ 1992 are included in the complete screening programme (2,5/1000 children living in the area; . 400 children with CP). No child followed from the time of CP-diagnosis has developed a hip dislocation. This is a significant decrease compared to a reference group of whom eight per cent dislocated a hip before six years of age. The reference group comprised all children with CP born 1990 –1991 and were screened by physiotherapists only. The range of motion in hip, knee and ankle in children without walking capacity improved significantly from the first to the later age groups followed. Children with walking capacity had almost no severe contractures. The proportion of children treated with orthopaedic surgery for contracture or skeletal torsion deformity decreased significantly. One fifth of the children with spastic diplegia had been treated with selective dorsal rhizotomy. One third of all children with CP born 1994 –1995 had been treated with botulinumtoxin before eight years of age and four children had intratecal baklofen before that age.

S27

With early treatment of spasticity/dystonia, early nonoperative treatment of contracture and prevention of hip dislocation, the need for orthopaedic surgery for contracture or torsion deformity is reduced, and the need for multilevel procedures seems to be eliminated. O14 Gabriela—a case study. Management and, re-education in paediatric feeding and swallowing difficulties I. Marinopulosp, E. Ziudru University of Manchester, MENCAP, Royal School for the Deaf Introduction: A thorough review of literature about swallowing and feeding disorders, coupled by observations made during the surgical and re-educational approach that helped a young neurologically impaired patient to gain oral feeding ability. Methods: An up to date article review, the step by step report of investigative (videofluoroscopy), surgical (Stamm gastrostomy) and communication techniques performed in order to tackle the feeding difficulties of an extreme premature child, diagnosed as suffering from Mobius syndrome. Results: The initial unpleasant nasogastric feeding pattern which forced the child to food denial disappeared following the temporary gastrostomy and appropriate behavioural management which led to significant weight gain and establishment of normal feeding patterns. Conclusion: Observations made during this case study confirmed many studies, which proved that only positive reinforcement and, contingency management treatments could be successful on the management of the feeding problems of neurological impaired children. O15 Etiological yield of comprehensive evaluation of global developmental delay in young children S. Gulatia,p, A. Tikariaa, M. Kabraa, S. Sapraa, R.M. Pandeyb, A.K. Guptac a Department of Pediatrics, All India Institute of Medical Sciences, New Delhi-110029, India; bDepartment of Biostatistics, All India Institute of Medical Sciences, New Delhi-110029, India; cDepartment of Radiodiagnosis, All India Institute of Medical Sciences, New Delhi-110029, India Background: Developmental delay is one of the commonest conditions encountered by pediatricians in clinical practice. Objectives: To study the etiologic profile of young children with developmental delay and to study the role of diagnostic evaluation of these children.

S28

Abstracts

Design, settings, and methods: A descriptive cohort study in children aged six months to five years, with global developmental delay (May 2001 – May 2003). Global developmental delay was defined as significant delay in two or more developmental domains. Children with delay in only one developmental domain, acute postnatal event with initial normal development, and regression of milestones were excluded. Initial assessment included history, physical examination and developmental assessment. Investigations included brainstem evoked responses, thyroid functions, Computed tomography/Magnetic resonance imaging, electroencephalography when indicated, karyotyping, molecular studies for Fragile X syndrome when suspected clinically and metabolic work up [arterial blood gas, arterial lactate, urine aminoacidogram, plasma ammonia, plasma aminoacidogram, and urine organic acid] when suspected clinically. Search for recognizable congenital malformation syndromes was carried out using London Dysmorphology Database [LDDB] when indicated. Statistical analysis: Association between presence of features at initial contact and etiology was analyzed by 2-tailed Fisher exact test and chi-square test. Results: 100 patients were studied. The mean age at the time of assessment was 23.6 months [^ 15.0 months], mean initial age at which parents suspected delay was 5.4 months [^ 2.3 months] and 67 were male, while 33 were female. History of birth asphyxia was present in 20. 67 children had some positive features in history which will be discussed. Developmental delay was mild in 25, moderate in 42 and severe in 33. Positive examination findings were present in 78 which will be discussed. Investigations: Hearing loss was mild in 8, moderate in 7, severe in 5. Three children had hypothyroidism. Abnormal neuroimaging findings included cerebral atrophy in 16, hypoxic insult in 12, cerebral malformations in 10, ventriculomegaly in 6 and intracranial calcifications in 5. Eighteen of 93 children had abnormal karyotypes, of which 13 had Trisomy 21. One child had Fragile X syndrome. Four of 24 children in whom metabolic workup was performed had abnormal metabolic screen. An etiologic diagnosis was made in 73: cerebral dysgenesis, chromosomal disorders including Down syndrome, Hypoxic ischemic encephalopathy (HIE) sequelae, and multiple malformation syndromes accounted for 55/73 [75%]. Associations between features evident from initial history and physical examination and determination of etiologic diagnosis were studied. History of sepsis, seizures or birth asphyxia increased the etiologic yield 1.4 –2.4 fold. Abnormal physical findings on examination increased the etiologic yield 2.3 fold. History and physical examination pointed to a diagnosis in 19/73 children. In 54/73 cases, laboratory investigations contributed to the diagnosis, and in 17/73, laboratory investigations were the sole means of etiologic determination. Conclusions: An etiologic diagnosis was made in 73% children with global developmental delay. Cerebral dysgenesis, chromosomal disorders, HIE sequelae, and

multiple malformation syndromes accounted for 75% of diagnoses. Hypothyroidism is an important treatable cause of developmental delay in 3 percent. Laboratory investigations make a big contribution to successful determination of etiology. O16 Quality of life in adolescents with cerebral palsy I. Prpic´p, Z. Modrusˇan-Mozeticˇ, Z. Korotaj, I. Vlasˇic´Cicvaric´ and Paucic-Kirincic University Hospital Centre Rijeka, Department of Pediatrics “Kantrida”, Rijeka, Croatia Purpose: Evaluation of Quality of life in group of adolescents with cerebral palsy with newly constructed HRQOL questionnaire which was developed regarding Croatian national customs and cultural identity. Method: The newly constructed HRQOL questionnaire covers seven domains; emotional well being, physical activities, family functioning, sexuality, education and treatment satisfaction adapted in regard to age, national customs and life style. The questionnaire consists of 42 items and answers are given on Likert scale (5 degree from never to always), self-assessment. The test was applied on 32 adolescents (10 female, 22 male), age 11– 21 (M ¼ 16; 8 age). All subjects had normal intellectual functioning, good reading ability and no comorbidity. Results: On each scale of seven measured domains adolescents have given good satisfaction level of functioning. No statistical differences have been found between gender accept in the domain of sexuality, were female adolescents are more unsatisfied with quality of their sexual activities then male adolescent with cerebral palsy. Conclusion: Application of this Questionnaire in clinical practice will help identify domains of life, which they found most restricted and dysfunctional because of disease. In this way we will be able to approve psychological support and medical treatment and help to approve Quality of life to chronically ill. O17 Learning from the child with cerebral palsy-a parent’s view D. Economou Cerebral Palsy Greece, Athens, Greece The purpose of this paper is essentially to consider what we may learn from our child with cerebral palsy. The following main areas of experience are taken into account: 1. Speaking of learning 2. The quest of information

Abstracts

3. An exceptional human being 4. A life of relationships 5. Fragmented treatment and care 6. Functioning or living 7. Society, friend or foe 8. A world of barriers 9. Getting to know and like you 10. The future. What does it hold? 11. Love with no conditions 12. The ultimate experience The conclusion is that we have a great deal to receive and learn from our child, as long as we travel with him all the way. It is the journey of a lifetime. O18 The prevalance of malocclusion in children with cerebral palsy I. Tanboga, P. Karataban, S. Altintas Marmara University, School of Dentistry and Metin Sabanci Spastic Children Center Malocclusion has a greater incidence in persons with disabilities. The objective of this investigation was to assess the prevalance of malocclusion (Class II, Class III Malocclusion), cross-bite and open-bite as well as bruxism as a parafunctional habit in children with different types of Cerebral Palsy. Materials and methods: 100 patients (53 boys, 47 girls) with different types of Cerebral Palsy (Spastic, Ataxic, Athetoid and Mixed) who were admitted to Metin Sabanci Rehabilitation Center were evaluated. The mean age of the study group was ^ 6.19 (2.5 – 15.5 years). A questionaire form was used for investigating the oral health status and treatment needs, including orthodontic anomalies, home care, intra oral and extra oral pathologies, parafunctional habits and pacifier usage. Parents were also inquired about how they fed their children whether they were using nursing bottle or not. Results: Data obtained indicated that, 84% Spastic, 8% Athetoid, 1% Ataxic and 7% of the patients had Mixed type of Cerebral Palsy. 22% of the patients also had an accompanying epilepsy. 17% of the patients had Class II, 7% had Class III malocclusion, also 3% of the patients had anterior and 5% had posterior cross-bite, 23% of the children had anterior open-bite and 16% of children were bruxists. According to our survey 11% of the children were fed only by nursing bottle and semi-solid or liquid food, 14% were fed with semi-solid or solid food with spoon and taking liquids with nursing bottle and 75% were fed with solid food. Also 12% of the children were still using pacifier. Conclusion: The results of this study have shown that the prevalance of openbite as a malocclusion in patients with CP was significantly high. This might be due to the usage of

S29

nursing bottle and pacifier accompanying the muscular weakness. Thus, parents should be educated about giving up nursing-bottle feeding at an earlier age to prevent the risk of malocclusions. O19 Efficacy of methylphenidate in patients with cerebral palsy and attention deficit disorder V. Gross-Tsurp, R.S. Shalev, N. Badihi, O. Manor Neuropediatric Unit, Shaare Zedek Medical Center, Jerusalem, Israel Objective: To study the short-term efficacy and safety of methylphenidate (MPH) in patients with the dual diagnosis of cerebral palsy (CP) and attention deficit hyperactivity disorder (ADHD). Patients and methods: Twenty-nine patients (8.0 ^ 4.0 years old) with CP and ADHD completed the study. In a prospective, cross-over, double-blind paradigm, patients were treated with MPH or placebo, each for 4 weeks. Parents and teachers completed a modified Abbreviated Conners’ Rating Scale at the beginning of the study and the end of the first and second month of MPH or placebo. Parents were contacted weekly to report side effects. Results: MPH treatment, as assessed by teachers’ ratings, resulted in a significant improvement in ADHD symptomatology (t ¼ 2:29 df ¼ 27 p , 0:05): however, only a trend for improvement was noted on parents’ ratings. After conclusion of the study, 12 patients continued MPH for 20 ^ 10 months. Side effects were minimal, with the exception of transient hallucinations in 1 patients. Conclusion: In children with CP and ADHD, MPH is effective, rendering it a valuable adjunct to their overall medical therapy.

Oral Session IV Cerebral palsy management O20 Is there a role for selective dorsal rhizotomy in children with cerebral palsy? Paper 1-selection criteria of patients G.F. Colep, J.H. Patrick, C. Stewart, A.P. Roberts, S.E. Farmer, U.K. Oswestry Selective dorsal rhizotomy (SDR) remains a controversial option for children with cerebral palsy. Several randomised controlled trials [1 – 3] have failed to agree whether the procedure leads to long term functional improvement. Disparate results from different centres may be related to lack of standardisation in (a) patient selection (b) intra-operative technique (c) post-operative

S30

Abstracts

physiotherapy regimes (d) outcome measures. Cerebral palsy is a heterogeneous condition and we have attempted to establish clinical criteria for selecting children who will most benefit. Defining and refining selection criteria is an evolving process and we report our experience here. In the last 5 years 53 children have been referred for consideration of SDR. Each child was reviewed and examined by a multidisciplinary team including an Orthopaedic Surgeon, Paediatric Neurologist and Physiotherapist. Gait analysis including dynamic EMG and energy cost assessment was carried out where appropriate. Selection criteria employed were as follows:-

J.H. Patrickp, A.P. Roberts, C. Stewart, S.E. Farmer, G.F. Cole, U.K. Oswestry There is a broad consensus that the surgical procedure of Selective Dorsal Rhizotomy (SDR) leads to a reduction in spasticity in cerebral palsy. Whether the procedure improves function is an on-going debate and evidence of improvement in post operative gait patterns is not readily available. Outcome measures lack standardisation and scales such as PEDI, Ashworth and GMFM have all been employed to evaluate SDR [1 –3]. In Oswestry we have performed SDR on 19 children (13 boys, 6 girls average age 8.6 years) using rigorous

History

Examination

Investigation

Age range 5– 10

Diagnosis spastic diplegia, severe hemiplegia or HSP Spasticity moderate to severe Mean lower limb power . 3 on MRC scale Movement control at least moderate

Moderate to low energy cost (No PCI . 3). No hip dysplasia No basal ganglia change on MRI

Absence of chronic conditions Cognitive ability-IQ 70 or above Well motivated, emotionally robust child No previous multilevel surgery Good family/social support

Weight not disproportionately . height

Balance-at least moderately good Absence of severe fixed joint deformity No involuntary movements or dystonia

19 children were chosen for SDR and our guidelines will be expanded upon. 34 children were not selected for SDR (including 14 considered to have insufficient spasticity). This and other reasons for exclusion of these 34 children, as well as suggested alternative treatment options will be discussed. The children selected for SDR are considered in the accompanying paper. We believe our encouraging results are related to our stringent selection criteria and that establishing clearly defined, widely accepted criteria will lead to improved outcomes for SDR. Further work is required to identify the relative importance of individual criteria. References [1] McLaughlin JF, et al. Dev. Med. Child Neurol. 1998;40:220–232. [2] Steinbok P, et al. Dev. Med. Child Neurol. 1997;39:178–184. [3] Wright FV, et al. Dev. Med. Child Neurol. 1996;40:239-247. O21 Is there a role for selective dorsal rhizotomy in children with cerebral palsy? Paper 2-instrumented gait analysis as an outcome measure

selection criteria and employing gait analysis as an outcome measure. All children were assessed pre operatively and this included neurological and orthopaedic examination, video filming, 3D gait analysis with marker tracking, dynamic EMG of critical muscles and energy cost assessment. Only 13 children completed a full gait analysis, although all were able to comply with some aspects of it. Post operative gait analysis was performed on average 1.5 years after surgery. We compared pre and post operative data using paired t-tests where valid, otherwise Wilcoxon signed-rank tests were applied. All the changes reported below are significant ðp , 0:05Þ: Average changes in the clinical examination were as follows: Hip extensor power increased by 0.6 grades, hip abduction increased by 6.68, hidden flexion (Thomas) decreased by 8.18 and anteversion decreased by 7.78. Knee extensor power increased by 0.3 grades. At the ankle dorsiflexion range increased by 8.88 with knee flexed and by 7.38 with knee extended. The plantarflexion range decreased by 5.58 and dorsiflexor power increased by 0.5 grades. In the hips beneficial changes were achieved by increasing the range of joint movement, whereas at the ankle the available range was shifted towards dorsiflexion, although there was an increase in the overall range of ankle movement.

Abstracts

The kinematic and kinetic data from gait analysis are entirely consistent with these clinical changes and will be demonstrated graphically and by video. At the knee joint there was a marked increase in knee extension. This was observed by an 9.88 decrease in knee flexion at initial contact and a 12.88 increase in the maximum extension angle during stance phase. The transition into swing phase became much freer with an increased rate of knee flexion. There was also a significant increase in post operative step length (increased by 0.1 metre) and walking speed (increased by 0.15 metres per second) indicating that the gait changes described had a functional and not purely cosmetic impact. Our primary goal was to improve the children’s gait and we feel that our promising results are a reflection of careful patient selection. References [1] Grubert C, et al. J Paediatr Orthop. 2000;20:496 –500. [2] McLaughlin JF, et al. Dev. Med. Child Neurol. 1998;40:220-232. [3] Steinbok P, et al. Dev. Med. Child Neurol. 1997;39:178-184. O22 Functionaland societal limitation and disabilityoutcomes in a randomized control trial of btox-a for children with spastic diplegic cerebral palsy K. Bjornsonp, R. Hays, C. Graubert, F. Won, R. Price, J. Mclaughlin, V. Pinedo Children’s Hospital and Regional Medical Center, 4800 Sand Point Way NE, Seattle, WA, USA Introduction: The purpose of this project was to document the functional and societal limitation and disability level effects of gastrocnemius BTX-A injections in children with spastic diplegic cerebral palsy. Methods: We carried out a randomized double masked placebo control trial in a tertiary-care children’s hospital. A sample of 33 children, with spastic diplegic cerebral palsy participated. Mean age was 5.5 years (range 3.0– 11.9) and 19 were male. GMFCS Levels were: I ¼ 12, II ¼ 15 and III ¼ 6. All participants were randomized to receive either 12 units/kg BTX-A or placebo saline injections evenly distributed to bilateral gastrocnemius muscles. Outcomes were collected at baseline, 3, 8, 12 and 24 weeks after injections. Results: Functional limitation and disability outcomes: the GMFM-88, GMFM-66, energy cost index (ECI) and COPM performance scores were collected at 3, 12, and 24 weeks post injection. GMFM-88 total score and the GMFM-66 significantly increased for the treatment group compared to placebo at 24 weeks (p ¼ 0:001 and 0.03, respectively). No significant group differences were found for the ECI at any follow-up point. Societal limitation

S31

outcomes: the Canadian Occupational Performance Measure (COPM) satisfaction scores and Goal Attainment Scaling (GAS) were obtained at 12 and 24 weeks. The COPM performance scores were significantly greater for the treatment group at 12 weeks ðp ¼ 0:04Þ and approached a significant difference at 24 weeks ðp ¼ 0:06Þ: COPM satisfaction scores and GAS were not significantly different by group at any follow-up point. Conclusions: The GMFM-88 and 66 scores and the performance scores of the COPM demonstrated a significant difference at 12 weeks with a diminishing effect noted at 24 weeks. In contrast, satisfaction with performance as measured by the COPM satisfaction scores and performance/importance of goals set as measured by the GAS were not significantly different between treatment groups. These data documented consistent changes on the functional limitation and disability levels up to six months post treatment, but not on the societal outcome level. Acknowledgments: National Center for Medical Rehabilitation Research, National Institutes of Health, (RO1 HD35750). O23 Pathophysiologic and impairment outcomes in a randomized control trial of BTX-a for children with spstic diplegic cerebral palsy K. Bjornson, R. Hays, C. Graubert, R. Price, F. Won, J. Mclaughlin, V. Pinedo Children’s Hospital And Regional Medical Center, 4800 Sand Point Way NE, Seattle, WA 98105, USA Introduction: The purpose of this project was to document the impairment and pathophysiologic level effects (NCMRR framework) of gastrocnemius BTX-A injections in children with spastic diplegic cerebral palsy. Methods: We carried out a randomized double masked placebo control trial in a tertiary-care children’s hospital. A sample of 33 children, with spastic diplegic cerebral palsy participated. Mean ages were 5.5 years (range 3.0 – 11.9) and19 were male. GMFCS Levels were: I ¼ 12, II ¼ 15 and III ¼ 6. All participants were randomized to receive either 12 units/kg BTX-A or placebo saline injections evenly distributed to bilateral gastrocnemius muscles. Outcomes were collected at baseline, 3, 8, 12 and 24 weeks after injections. Results: Pathophysiologic outcomes were EMG electrical responses to maximal voluntary contraction (QEK), which documented significant effect ðp ¼ 0:05Þ of the treatment versus placebo group at 3 weeks post injection. Impairment level outcomes were an electromechanical torque measure of spasticity (SMS), Ashworth scores, achilles deep tendon reflexes (DTR), clonus, ROM and maximum torque of the gastroc soleus muscle. SMS total and elastic path-lengths were significantly lower for the treatment group at 8 weeks ðp , 0:05Þ: Ashworth scores were not significantly different at any time point. Achilles DTR

S32

Abstracts

ðp , 0:03Þ and clonus ðp ¼ 0:05Þ were significantly decreased in the BTX-A group at 3 weeks. At 12 weeks, ankle dorsiflexion ROM was significantly greater for the treatment group ðp ¼ 0:05Þ: Maximum voluntary torque was significantly greater in the treatment group at 24 weeks ðp ¼ 0:03Þ: Conclusions: The significant decrease in EMG activity at 3 weeks is consistent with the known pathophysiological effect of the intervention. Achilles DTR and clonus were significantly decreased in the treatment group also at 3 weeks. The significant difference in SMS at 8 weeks and subsequently the dorsiflexion ROM at 12 weeks may be due to changes in the visco-elastic properties of the muscle. The significant change in maximum voluntary torque at 24 weeks may be a direct response to the improved ROM. Acknowledgments: National Center for Medical Rehabilitation Research, National Institutes of Health, (RO1 HD35750). O24 The evaluation of botulinum toxin injection for the treatment of equinus foot in cerebral palsy G. Albertinia, M. Gallib, V. Cimolinb, M. Romeib, N. Tenorea, E.M. Valentec, M. Crivellinib a “San Raffaele-Pisana” Hospital-Tosinvest Sanita`, Rome, Italy; bBioeng. Dept, Polytechnic of Milan, Milan, Italy; c Istituto CSS Mendel, Rome, Italy Introduction: Intramuscular botulinum toxin A (BTA) injection in the calf muscle has an important role in the conservative management of dynamic equinus deformity in children with Cerebral Palsy (CP). It can be used to reduce, temporarily, their muscle spasticity and to obtain a better placement of the foot on the ground during walking; in this way it is possible to delay the need for more invasive treatments, such as surgery, too. Methods: 15 children (mean age: 10.1 years) with dynamic contracture of the calf muscles were selected for BTA treatment to correct equinus foot. Three selection criteria were applied for inclusion of subjects in the study: diagnosis of CP, presence of equinus foot caused by calf muscle spasticity and no previous orthopaedic surgery at lower limb level. 20 healthy children (mean age 10.6 years) formed the group of healthy controls (CG). A 12-camera optoelectronic system with passive markers (ELITE, BTS S.p.A., Milan, Italy) working at a sampling rate of 100 Hz, for measuring the kinematics of movement, and two force plates (Kistler, CH), for measuring the kinetik of movement, and a Video system synchronic with optoelectronic and force platform systems were used. The patients received injections of BTA into the triceps surae: two injection sites were selected (medial and lateral gastrocnemius) and an initial dose of 2 – 6 U/kg per muscle was administered ([email protected], Allergan Inc.).

The effects of the BTA were monitored around one month after treatment and in some cases through several postinjection GA acquisitions. The patterns of ankle and knee kinematic and of ankle kinetic were analysed and some parameters were defined in order to evaluate the effectiveness of the treatment. Results: Before treatment, all the patients presented severe equinus foot in stance and excessive flexion of the knee upon initial contact with the ground: this pattern is indicative of the presence of equinus foot. The posttreatment GA data revealed two different effects: the 80% of the subjects showed an significant improvement at ankle, but not at knee level. In the remaining 20% a significant improvement at knee level was observed, but equinus foot persisted; effects are never observed contemporaneously at both joints. About ankle kinetic, peak of power absorbed was found to be very similar to the value recorded in the healthy controls ðp , 0:05Þ; while no significant increases in peak of power generated were observed. Conclusions: The results of this study show that injecting BTA into the gastrocnemius to correct equinus foot gives satisfactory results: placement of the foot on the ground improved in 80% of the subjects studied. In the other cases the effect of the BTA manifested itself only at the knee and not at ankle level. Furthermore, BTA injection produces temporary paralysis of the gastrocnemius: for this reason, no improvement in power generated at the ankle joint was observed in any patient. One particular feature of this study was that, unlike previous studies in the literature, it focused not only on the ankle joint, but also on the knee. Given that the gastrocnemius is, in fact, a biarticular muscle, it was deemed important to evaluate the effects of BTA on both these joints. O25 The lateral distal femur: an alternative dxa site for kids with cerebral palsy H.H. Kecskemethyp,a, H.T. Harckeb,d, S.J. Bachrachc,d a Departments of Research, Nemours Children’s ClinicWilmington, Alfred I. duPont Hospital for Children, Wilmington, Delaware, USA; bDepartments of Medical Imaging, Nemours Children’s Clinic-Wilmington, Alfred I. duPont Hospital for Children, Wilmington, Delaware, USA; c Departments of Pediatrics, Nemours Children’s ClinicWilmington, Alfred I. duPont Hospital for Children, Wilmington, Delaware, USA; dThomas Jefferson University, Philadelphia, Pennsylvania, USA The distal femur is a site that is accessible in most children with CP and at times it is the only site than can be meaningfully measured by dual energy X-ray absorptiometry (DXA). The presence of metallic implants, positioning

Abstracts

difficulties, and excessive involuntary movement often preclude the use of typical DXA measurement sites. It is our intent to discuss the use of the lateral distal femur (LDF) in children with CP. Methods: An overview of the LDF technique will be provided including how it is obtained, what the positioning looks like, and how it is used clinically. Case examples will be shown. Results: Explanation of the technique will be provided including visual images of the positioning for the LDF DXA in children with CP and of analyzed scans. Normative data have been published [1] and the clinical utility of the scan results will be shown on normal curves for the three regions of interest in the LDF. Clinical application of the LDF over time will be demonstrated. Conclusion: The lateral distal femur is an accessible and meaningful site to measure by DXA in children with CP. Pediatric centers in the US are starting to use the LDF site for DXA on their children with CP. Reference [1] Henderson RC, Lark RK, Newman JE, Kecskemethy H. Fung EB, Renner JB, Harcke HT: Normal Pediatric Reference Data for DXA Measures of Bone Density in the Distal Femur. Am. J. Radiol. 2002;178:439– 443. O26 Treatment of osteopenia with pamidronate in children with spastic quadriplegic cerebral palsy and recurrent fractures S.J. Bachrachp,a,e, H.H. Kecskemethyb, H.T. Harckec,e, R.K. Larkf, F. Millerd,e, R.C. Hendersonf,g a Department of Pediatrics, Nemours Children’s ClinicWilmington, Alfred I. duPont Hospital for Children, Wilmington, Delaware, USA; bDepartment of Research, Nemours Children’s Clinic-Wilmington, Alfred I. duPont Hospital for Children, Wilmington, Delaware, USA; c Department of Medical Imaging, Nemours Children’s Clinic-Wilmington, Alfred I. duPont Hospital for Children, Wilmington, Delaware, USA; dDepartment of Orthopaedics, Nemours Children’s Clinic-Wilmington, Alfred I. duPont Hospital for Children, Wilmington, Delaware, USA; e Thomas Jefferson University, Philadelphia, Pennsylvania, USA; fDepartment of Orthopedics, University of North Carolina, Chapel Hill, North Carolina, USA; gDepartment of Pediatrics, University of North Carolina, Chapel Hill, North Carolina, USA The safety and efficacy of IV pamidronate to treat low bone mineral density (BMD) in children with CP has been evaluated and reported by our group [1]. The purpose of this study is to report on our expanded experience with IV pamidronate in children with severe CP, including

S33

longitudinal bone density evaluations for up to 4 years after treatment. Methods: Seventeen non-ambulatory skeletally immature patients with spastic quadriplegic cerebral palsy, osteopenia, and a history of previous fracture(s) were treated with a 3-day dosing cycle of intravenous pamidronate at 3 to 4 month intervals for 1 year (five dosing cycles). BMD, radiographs and blood studies were obtained serially. Results: At the end of the treatment period bone mineral density (BMD) in the metaphyseal region of the distal femur was increased compared to pre-treatment BMD by 66 ^ 56% (mean ^ SE), and increased 47 ^ 38% in the lumbar spine. Post-treatment BMD was measured in 13 subjects a minimum of 3 months after treatment ended (mean 21 months, maximum 49 months). Subsequent changes in BMD were variable. Both late significant decreases and continued increases in BMD were observed after treatment. During the treatment period only one fracture occurred, and it was within one week of the first dosing cycle. One subject sustained a non-traumatic fracture during prolonged followup, 51 months after starting treatment. Conclusion: All subjects tolerated the treatment regimen without significant untoward effects. Rate of fracture was reduced. Long term effects on BMD were variable. Reference [1] Henderson RC, Lark RK, Kecskemethy H, Miller F, Harcke HT, Bachrach SJ. Bisphosphonates to treat osteoporosis in children with quadriplegic cerebral palsy: A randomized, placebo-controlled clinical trial. J. Pediatri. 2002;141:644 –651. O27 The influence of an adapted training program on the development of the gross motor function and performance in adolescents with spastic hemiplegia I. Kandralia, C. Evaggelinoub, G. Katsimanisb, N. Aggelopoulouc a Public Special High School, Eleones Pilaias, 55535, Thessaloniki; bAristotle University of Thessaloniki, Agios Ioannis, 62100, Serres; cAristotle University of Thessaloniki, 54006, Thessaloniki Introduction: The purpose of this study was to evaluate the influence of a training program on the improvement of the gross motor function and performance in adolescents with spastic hemiplegia (SH) as a result of cerebral palsy (CP). Methods: Ten adolescents with SH, 13 – 17 years of age were assigned to two groups of 5 individuals (experimental and control group). Both groups attended the physiotherapeutic program of the public special school based on the Neurodevelopmental Treatment (NDT), Bobath Approach. An adapted training program was

S34

Abstracts

applied on the experimental group, three times per week, for about 50 minutes, with a duration of 3 months. The adapted training program was based on the Neurodevelopmental Treatment (NDT) combined with an athletic program based on the principles of basketball. Pre- and post-exercise tests were performed to assess the gross motor function and performance of the adolescents with SH of both groups. The instruments for the evaluation were the Gross Motor Function Measure (GMFM) and Gross Motor Performance Measure (GMPM). Results: A two-way ANOVA analysis was used and the results revealed that the adapted training program had a positive effect on the experimental group and improved both their quantitative and qualitative gross motor skills ðp , 0:001Þ: There was no change between the initial and final measurement on the control group. Discussion/Conclusion: Conclusively, adolescents with SH are able to show improvement both in their quantitative and qualitative gross motor skills. At the same time, the adapted training program is a particularly effective method of intervention, since it can help them deal with their movement problems better as well as to conduct various everyday activities in a more willing and lively way. O28 Effect of isotonic strengthening exercises on children with cerebral palsy L. Movsisyanp, L. Michaud, A. Mirzoyan Republican Pediatric Rehabilitation Center, Yerevan, Armenia Introduction: The purpose of this study was find out how the strengthening of different muscle groups in children with cerebral palsy (CP), who underwent the Progressive Resistive Strengthening Exercises (PRSE) course, will impact on their functional skills, their general strength and energy expenditure. All patients didn’t have any surgical intervention prior to inclusion in this study. Material and methods: This study includes 14 children with CP, ages from 3 to 19 years (7 males and 7 females) who didn’t have any kind of orthopedic surgeries or other kind of manipulations and who underwent a short-term, heavy progressive resistive exercise program. The majority of children (12) had spastic diplegia, one was double hemiplegic patient and another one had dyskenitic form of CP. The investigated group of patients was evaluated for Gross Motor Function by GMFM 88 and Gross Motor Function Classification System for Cerebral Palsy before and after training. Spasticity of quadriceps and hamstring muscles were measured using Modified Ashword Scale. Energy Expenditure Index was calculated before and after training by measuring the resting heart rate, walking heart rate and walking velocity (walking distance for 3 minutes), with or without assistive devices (braces, walker, etc.). After manual testing of the different groups of muscles

the isometric muscle strength was determined using hand hold dynamometer to the proximal muscle groups of lower extremities (hip flexors, extensors, abductors, adductors, quadriceps femoris and hamstrings). Testing was done before and after six days of training, as well as at the end of training course and three months after. Calculation of the load for each muscle was determinated by measuring of muscle strength by dynamometer. This measurement was repeated three times and the average of results was used as bases for the future calculation. This amount was duplicated and 65% was used as a load for isotonic strengthening exercises. The standardized Muscle Test Positions was used for each muscle groups. The frequency of exercises was once a day during first six days, then the strength was reassessed and new load was calculated for the next six days. After that, final assessment was done. Each exercise was repeated 20 times, in sets of 5 repetitions. Between each repetition children took 30 seconds rest. The training duration was 12 days. Results: The isotonic muscle strength training program showed significant increase of muscle strength during training from beginning to end: significant increase was evident during the first section of exercises. During the second six days it increased slightly. Further investigations showed general decrease of muscle strength, but it finally stayed slightly above beginning values. All children showed improvement in gross motor function, but the changes were not too impressive. Gains made in gross motor ability as a result of the training program were maintained by all subjects over the three months after the end of the training course. No changes were seen in Gross Motor Function level. No significant changes were demonstrated in walking velocity and energy efficiency. Conclusions: Short-term strength training program demonstrates positive changes for children with CP: increase of muscle strength, improvement of gross motor ability. The positive response to exercise program indicates that the heavy resistance exercise program can be used as one of the methods in the treatment of children with CP. For the improvement of gross motor function we are planning to alternate the strengthening exercises with gross motor function training in the future.

Oral Session V Neurodevelopmental Disorders O29 Early development of infants with cerebral palsy as measured by the test of infant motor performance S.K. Campbellp, V.M. Barbosa University of Illinois at Chicago, USA Introduction: Both information on the natural history of motor development in infants with cerebral palsy (CP) and

Abstracts

tests which identify children with CP as early as possible are needed in order to properly plan and deliver intervention. The Test of Infant Motor Performance (TIMP) is a test of functional movement specifically designed to assess the postural and selective control of movement needed for functional activities in early infancy. The TIMP can be used for infants from 32 weeks postconceptional age through 4 months adjusted age (AA). Purpose: The purpose of this paper is to present results on the Test of Infant Motor Performance (TIMP) through 4 months of age in a group of infants later diagnosed as having CP (Barbosa et al., 2003). Methods: Subjects were 96 infants with varying risk for poor developmental outcome of whom 10 were later diagnosed as having CP. Infants were tested weekly from 0 –4 months AA on the TIMP and at 12 months on the Alberta Infant Motor Scale. Physicians provided an independent opinion of neurologic condition at 3, 6, 9, and 12 months AA, and infants were followed for developmental outcome up to 4 –5 years of age. Results: Six infants with CP had TIMP measures 2 0.5 SD below the mean already at 7 days AA, one from 30 days on, and one beginning at 60 days. Poor ability to lift the head during pull-to-sit was a particularly important early sign of CP, and other items requiring anti-gravity control and midline orientation of the head also showed delayed development. Unlike children with delayed development but no CP, children with CP showed areas of regression in development, i.e. loss of previously demonstrated skills in the lower extremities. Conclusion: The TIMP is a useful test for early diagnosis of delayed motor development in children with CP. Items requiring anti-gravity and midline control of the head were particularly problematic for infants with CP. Repeated testing revealed regression in lower extremity skills. Reference [1] Barbosa, V. M., Campbell, S. K., Sheftel, D., Singh, J., Beligere, N. Longitudinal performance of infants with cerebral palsy on the Test of Infant Motor Performance and on the Alberta Infant Motor Scale. Physical and Occupational Therapy in Pediatrics, 2003;23(3):7-29. O30 Neurodevelopmental assessment of infants with neonatal clinical subtle seizures D. El Metwally, A. Abd El Rahman Pediatrics Department. Faculty of Medicine. Suez Canal University Objectives: (1)-Neurodevelopmental outcome assessment among preterms with subtle seizures (SS). (2)-Detection of Factors associated with poor neurodevelopmental outcome among infants with SS.

S35

Methods: A prospective analytic study of 44 preterm: 22 with history of subtle SS in their neonatal period, compared to 22 control preterm infants with no seizures. Neurodevelopmental assessment was conducted at 9 – 12 months corrected age, using neurological exam and CAT-CLAM scale, abnormal developmental quotient (DQ) was defined as DQ , 70. Multiple regression analysis was run to detect factors associated with unfavorable outcome. Results: Abnormal/suspect neurological examination occurred among 81.8% of the SS infants compared to 50% among the control preterm infants, p , 0:026: Mean CAT-DQ among the SS infants was 67.7 ^ 9.8 compared to 76.1 ^ 8.4 among the control infants, p , 0:004: Mean CLAM-DQ among the SS infants was 64.4 ^ 11.2 compared to 75 ^ 9.2 among the control infants, p , 0:001: Multiple regression analysis revealed a significant inverse relationship between birth weight and CAT-CLAM , 70 with an OR: 0.03, CI: 0 –0.8, p , 0:041: CAT , 70 was associated with the cause of Seizures, with an OR: 7.1, CI: 1.5 – 33 p , 0:013:Conclusion: Subtle seizures are associated with remarkable adverse neurodevelopment in the first year. The etiology of seizures and birth weight are the determinants of poor outcome. Further follow-up of this cohort is recommended. O31 Neurodevelopmental outcome of high risk neonates: the importance of a detailled assessment at 4 years of age H. Bouzap, I. Antoniadou, H. Antonopoulou, M. Morozini, J. Sarafidou, M. Anagnostakou, F. Anatolitou, M. Xanthou B’ Neonatal Intensive Care Unit, “Aghia Sophia” Children’s Hospital, Athens; Institute of Child Health, Athens Introduction: Very Low Birth Weight(VLBW) Neonates and neonates with Hypoxic/Ishaemic Encephalopathy (HIE),are at increased risk of developing major neurodevelopmental disabilities as well as more subtle developmental deficits. The purpose of our study was to assess VLBW infants and infants with HIE at a preschool age in order to detect those subtle difficulties which may affect different areas of development and may lead to learning difficulties. Methods: 47 children, which were graduates of our Neonatal Intensive Care Unit (NICU),were assessed at 4 years of age.40 children were VLBW neonates (Birth weight , 1500 g) and 7 children had moderate or severe HIE. The children were born between May 1997 and December 1998 and were assessed from May 2001 to December 2002.After discharge from the NICU, the children were followed up longitudinally and neurodevelopmental disabilities were identified. At the age of 4 years, the children had a detailed assessment with a combination of structured tests in order to identify subtle deficits which might have been undiagnosed. We used as a tool the Towen’s examination for minor neurological dysfunctions, the Griffiths mental, development scales and the Miller

S36

Abstracts

assessment for preschoolers. The Mann – Whitney test has been used to compare the scores in the different subscales of the different tests in the premature children with a birth weight , 1000 g compared to those with a birth weight of 1000 – 1500 g and in the premature children with a gestational age , 29 weeks compared to those with a gestational age of 30– 32 weeks. Results: The Towen neurological examination revealed that the percentage of the children with suboptimal scores was high mainly at the subscale of coordination, gross motor function and associated movements. The Griffiths developmental assessment revealed low on the average scores in the subscale of eye-hand coordination and performance. The Miller assessment for preschoolers revealed a high percentage of children needing intervention at the subscale of foundation, coordination and complex tasks. The combination of these tests overts that the most common problem of this population is coordination disorders. More immature neonates had significantly lower scores in various subscales. Conclusion: The combination of the three tests used for the evaluation of high risk neonates at a preschool age is believed to be a useful tool that could identify the subtle disabilities of that population that could be associated with learning difficulties. The early identification of these disabilities is important in order to suggest intervention before school age. O32 Variability of early abnormal neurologic signs in a longitudinal study of high risk neonates H. Bouzap, G. Nyktari, M. Anagnostakou, M. Koumenidou, M. Xanthou B’ Neonatal Intensive Care Unit, “Aghia Sophia” children’s Hospital, Athens Introduction: Very Low Birth Weight neonates and neonates with Hypoxic/Ichaemic Encephalopathy (HIE) admitted in Neonatal Intensive Care Units (NICU) are at high risk for developing cerebral palsy. The purpose of the study was to follow longitudinally high risk neonates in order to detect early abnormal neurologic signs, to study their variability within the first two years of life and to compare the variability within the different groups studied. Methods: 196 neonates(53 preterm with a birth weight , 1000 g (group A), 116 preterm with a birth weight 1000 – 1500 g (group B), and 27 full term with HIE (group C),admitted and survived in our unit between January 1999 and December 2001, were followed up longitudinally from discharge till the age of 2 years. A structured neurological examination was performed focusing on passive tone, motor activity, reflexes, postural reactions and gross motor millestones.The neurological examination was performed before discharge, at 3 months, 6 months,12 months and 24 months of age. For the preterm infants the corrected age was used for evaluation. The neurological examination was

considered as normal or borderline if no abnormal or subtle signs were depicted, or abnormal if deviations from normal of muscle tone of the trunk or the limbs, lack of head control, presence of abnormal movements, brisk reflexes, delay in gross motor millestones or asymmetry were found. Results: In our study, out of the newborn infants with HIE, 1/3 had already an abnormal neurological examination on discharge, which remained unchanged until 24 months. At the age of 3 or 6 months some infants became abnormal and remained abnormal thereafter. No infant considered abnormal at 6 months improved at the following examinations. In premature infants, the neurological examination showed a greater variability with most abnormal signs appearing at 3 months or 6 months. If the abnormal signs appeared at 3 months, the outcome was poor in most cases. If the abnormal signs appeared at 6 months, a great percentage of these signs disappeared by 12 months. In almost all infants, major abnormal neurologic signs were found up to the age of 1 year. Conclusion: The neurologic examination of the high risk infant shows a great variability in the first year of life with abnormal signs appearing at any stage of development. In some infants abnormal signs are transient, in some others they are persistent as early markers of cerebral palsy. Structured follow up examination is needed for the early diagnosis of those signs in order to initiate early intervention programs. O33 Evaluation of peri-intraventricular hemorrhages in newborn infants taking into account laterality A. Kapiki, A. Triantafyllidou, E. Stratiki, E. Papathoma, A. Koutroumba, S. Sevastiadou, C. Costalos Neonatal Unit, Alexandra General Hospital, Athens, Greece The IVH in the neonatal period could be bilateral(asymmetrical or not) or unilateral. The aim of this study was to investigate whether the IVH in the neonatal period occurs more frequently on the left side of the brain. Study group: We randomly selected the records of 125 neonates born in our Unit in the period of 1999– 2003 regardless of gestation age and other problems and having had an U/S brain for whatever reason. The median gestational age was 31,4wks (^ 3,7), the median gestational weight was 1602 gr (^ 782) and 72 were boys and 53 girls. The incidence of IVH was as follows: IVH1E degree: 2/125, IVH2E degree: 6/125, IVH3E degree: 29/125, IVH4E degree: 7/125, Hydrocephalus: 1/125. For most of them there were two U/S brain performed during their hospitalization. For analysis reasons the newborn were subdivided in fo\ur categories based on the IVH degree and the following were recorded: mothers’ age, IVF, prenatal corticosteroids and doses, premature rupture of membranes, chorioamnionitis, diabetes, preeclampsia, other problems,

Abstracts

mode of delivery, gestational age, gestational weight, SGA, gender, origin, Apgar score at 10 and 50 , need and mode of resuscitation, RDS 1 and RDS 2, need of mechanical ventilation, surfactant type and doses, patent ductus arteriosus, treatment of PDA, infections,antibiotic treatment, positive blood cultures, infections type (meningitis, sepsis etc),NEC, postnatal corticosteroids, eyes examination, ROP and outcome (death,survival). The location of the IVH regardless of degree was statistically analyzed. Results: In the first U/S brain the IVH regardless of the degree was more frequently left-sided compared to the right-sided or symmetrical bilateral ðp , 0:001Þ: In the second brain U/S no difference was found to the location of the IVH. In the third brain U/S left sided IVH was more frequent than the right sided IVH in a statistically significant manner ðp , 0:001Þ but not compared to the bilateral IVH. Conclusion: We are looking forward to recording all the newborns having had an U/S brain during this period in order to establish our observation for the left side location of the IVH. The reason of this laterality remains to be elucidated. O34 Intraventricular hemorrhage: incidence, risk factors and outcome in a population of very low birth weight infants E. Papathoma, J. Kaleyiasp, A. Kapiki, Z. Hatzistamatiou, A. Koutrouba, A. Anastasiadou, A. Andrikopoulou, C. Kostalos NICU, General Hospital of Athens “Alexandra”, Athens, Greece Aim: Estimation of the incidence of intraventricular hemorrhage (IVH), in premature VLBW infants, identification of possible perinatal variables that affect the risk of IVH, and assessment of the outcome of very low birth weight (VLBW) infants with IVH. Materials and methods: The cohort of infants used in this analysis was selected from a five-year period (1999 – 2003). In the study were included, all VLBW infants (birth weight #1500 g) born in our hospital and admitted to the NICU during this period. Criteria for exclusion were death within 48 hours after birth, transfer to another hospital, and/or lack of cranial ultrasound. The diagnosis of IVH was based on the ultrasound scan, and when present, the grade of IVH was determined according to Papile et al. The relative risk for each factor and the significance ðp , 0:05Þ were determined

S37

by the chi-square method. Comparison of mean rates was carried out using the Student’s t-test. Statistical analysis was performed using the SPSS. Results: During the study period, 395 VLBW infants were nursed in the NICU, and of these 345 (87%), were included in the study group, with mean BW ¼ 1110 (^ 270) g and mean GA ¼ 29.2 (^ 2.7) weeks. The overall incidence of IVH was 26% (90/345): 11.5% grade I/II (40/345), 10% (34/345) grade III and 4.6% (16/345) grade IV. The incidence of major IVH was 17% during the period 1999 – 2001 and declined to 11.4% during the period 2002 – 2003 ðp ¼ 0:14Þ: Risk factors (OR . 1) for IVH (as a total) were: GA # 28 wks (OR ¼ 1.8, 95% CI 1.4– 2.3), BW#1000 g (OR ¼ 1.4, 95% CI 1.16– 1.73), Apgar score at 1 min#4 (OR ¼ 1.3, 95% CI 1.1– 1.53), resuscitation at the delivery room (OR ¼ 2.5, 95% CI 1.5 – 4.2), RDS (OR ¼ 5.2, 95% CI 2.9– 9.3), PDA (OR ¼ 3.5, 95% CI 2 –6.2), sepsis (OR ¼ 3.5, 95% CI 2.1 – 6), NEC (OR ¼ 1.7, 95% CI 1.02 –3). Regarding major IVH (grade III/IV), additional risk factors were: chorioamnionitis (OR ¼ 1.6, 95% CI 1.2 –3) and air leaks (OR ¼ 2.5, 95% CI 1.1– 5.8). Protective factors (OR , 1) against IVH were: toxenemia (OR ¼ 0.48, 95% CI 0.24 –0.96) and delivery by caesarean section (OR ¼ 0.65, 95% CI 0.46 –0.95). Regarding major IVH, additional protective factor was prenatal steroid administration (OR ¼ 0.53, 95% CI 0.28– 0.9). No correlation was found between IVH and premature rupture of membranes ðp ¼ 0:5Þ; IVF ðp ¼ 0:3Þ and multiple pregnancy ðp ¼ 0:34Þ: Neonates with IVH spent more days on ventilator ðp ¼ 0:05Þ; need higher FiO2 ðp ¼ 0:02Þ and length of hospitalization was similar ðp ¼ 0:4Þ to that of neonates with no IVH. There was a strong association between major IVH and BPD (50% of neonates with major IVH developed BPD vs 14% of neonates without IVH, p , 0:0001). The overall mortality rate was 27% (103/376), while the mortality rate of the neonates with IVH was 43% (37/86). Taking into account the grade of IVH, the mortality rate was 27.7% (10/36) for grade I/II and 55% (26/47) for grade III/IV ðp , 0:0001Þ: Conclusions: Our results confirm that the development of IVH is associated with extreme prematurity (BW # 1000 g, GA # 28 weeks), low Apgar score at 1 min, need for resuscitation in the delivery room, RDS, PDA, NEC, sepsis and chorioamnionitis (for major IVH). Protective factors against IVH were the prenatal corticosteroid administration (for major IVH), delivery by caesarean section and toxenemia.