Long term outcome of very low birth weight babies born at Sri Jayawardenepura General Hospital from 1990-1995

Method VLBW babies (<1500g), born at SJGH from 1990 January to 1995 December were compared with age and sex matched children of birthweights >2500g. These children were 6-11 years old at time of study (September-December 2001). Of 161 survivors, 92 were assessed. Their weights, heights, OFC and BMI were noted. Wechsler Intelligence Scale for Children-Revised (WISC-R) and Draw-aman test were used to assess cognitive functions. Educational status was assessed using age appropriate reading, writing, arithmetic tests and grade 5 scholarship marks. Behavioural status was assessed using a check list.


Introduction
Several studies in the West 1,2 have shown that very low birth weight (VLBW) children have impairment of growth, cognitive function, educational status and behaviour.Neurological abnormalities were also reported in these studies.Study done in Southampton Hospital, UK 3 has shown that VLBW babies are liable to be overweight after 7 years and are prone to get cardiovascular complications at an early age.Main aim of present study was to compare long-term outcome of VLBW babies born at Sri Jayewardenepura General Hospital (SJGH), during a period of six years, in terms of their growth parameters (weight, height, occipitofrontal circumference [OFC], body mass index [BMI]), cognitive functions, educational status (reading, writing, arithmetic) and behavioural status with age and sex matched children of normal birth weight.

Method
Study population consisted of a cohort of VLBW (<1500g) survivors born at SJGH from 1990 January to 1995 December.A normal birth weight (>2500g) control population was recruited, individually matched for age and sex, from paediatric ward and clinic.Controls, taken from paediatric ward, were examined after allowing sufficient time for convalescence.Of 161 survivors, with whom we corresponded, 92 responded.Informed consent was obtained from parents and an intervieweradministered questionnaire was used to gather data.
A complete neurological assessment was carried out to identify motor defects, cranial nerve palsies, visual and hearing defects.We used Snellen chart, crude hearing tests, Rinne and Weber tests as tools in above assessments.
Growth assessment included measurement of weight, height and OFC.BMI was calculated using formula weight in Kg/ (height in metres) 2 .Growth measurements were taken by a trained medical officer.
Cognitive function tests were assessed using Weschler Intelligence Scale for Children -Revised (WISC-R) 4,5 and Good enough-Harris.Draw-a-man test 6 .WISC-R consists of 6 verbal and 4 performance tests.Verbal tests were on general information, general comprehension, arithmetic reasoning, remembering series of digits forwards and backwards, similarities in pairs of words and vocabulary.Performance tests were used to assess children's ability on digit symbol coding, picture completion, block designs and picture arrangement.We had to modify test material to suit Sri Lankan children and 6 different question series were used.
Verbal and performance scores were calculated up to 100 separately so that total score was given out of 200.Perceptuomotor function was assessed by Good enough-Harris Draw-a-man test.Children were asked to draw a picture of a man.20 characteristics of the picture were considered in giving a score.One mark was given to each criterion and a developmental quotient (DQ) 6  between the two cognitive function tests was noted for cases and controls.

Results
Table 1 shows perinatal characteristics of the cohort of VLBW children.Inability to match a classmate to each VLBW child regarding educational variables was one of the limitations of this study.This was not feasible as our sample had children who were transferred in utero from various parts of the country.A significant number of VLBW children were lost to follow up.This was the other important limitation.

Conclusions
• VLBW children mainly consisted of preterm babies who were SGA.
• PIH is an important predisposing factor.
• Their catch up growth was not satisfactory.
• VLBW children had significant reading and writing difficulties.
was calculated for each child.Writing was marked out of a score of 4 whereas arithmetic tests were marked out of 8. Before finalizing methods for educational tests, advice of the primary school teacher was sought.In addition, year five scholarship marks were noted and compared.