[Longitudinal study on the age at childhood onset in children from Shanghai and its effects on child growth]

OBJECTIVE

The infancy-childhood-puberty (ICP) growth model divides human growth into three additive and partly superimposed phases: infancy, childhood, and puberty reflecting the endocrinology of the growth process. The childhood phase of the ICP model has been assumed to start during the second half-year of life and slowly decelerates, although it continues into puberty until growth ceases. This phase has been taken to mirror the effect of growth hormone (GH). The age at onset of the phase of growth, normally occurring between 6 and 12 months of age is recognized to be an important time event at postnatal human growth. The aims of this study were to study the age at childhood onset in Shanghai children and its effect on child growth, and to compare these results with the results obtained from Swedish longitudinal studies.

METHODS

The study population consisted of 1 720 children born between January 1, 1980, and December 31, 1990 in Fenglin Community, Shanghai, who got a continual monitoring at the clinic from birth to six years of age. The age at onset of the 'childhood phase' of growth was determined individually, in units of one month, by visual inspection of the data contained within the individual, computer-generated, ICP-based growth chart for both length and length velocity. T tests, chi(2) tests and multiple linear regression analyses were used in data analysis.

RESULTS

There was a significant (P < 0.05) difference in the distribution of the age at onset of the childhood phase of growth between Shanghai and Swedish populations. The mean age of the infants at onset of the 'childhood phase' of growth was 11.2 months in boys and 10.7 months in girls. Compared to their Swedish counterparts, these means occurred 1.3 months later in boys and 1.4 months later in girls. Both age at onset of the 'childhood phase' of growth and length at six months of age significantly (P < 0.05) contributed to the attained height from 12 months of age onward; one month delay in the onset of the 'childhood phase' of growth reduced height, at 5 years of age, by 0.4 cm in boys and 0.56 cm in girls. The mean values of length/height among the four groups of the age of the childhood onset, i.e. 6 m-, 9 m-, 12 m-, >/= 15 m, were found to be significantly (P < 0.05) different from 12 month of age onward. However, a significant (P < 0.05) difference in the mean length/height velocities among the four groups was in principle only found during the following intervals: 3 - 6, 6 - 9, 9 - 12, and 12 - 18 months of age. The mean values of BMI among the four groups of the age of the childhood onset showed significantly (P < 0.05) different only in 9 and 12 months of age. The mean values of BMI velocity among the four groups of the age of the childhood onset were found to be significantly (P < 0.05) different in four age intervals: 3 - 6, 6 - 9, 9 - 12, and 12 - 18 months of age. The age at onset of the 'childhood phase' of growth was negatively associated (P < 0.05) with mid-parental height, though positively related (P < 0.05) to height at six months of age. A simple linear regression was applied to the age at onset of the 'childhood phase' of growth and mid-parental height. It was found that the age at onset of the childhood phase' of growth was 1.5 months later in boys and 1.3 months later in girls, for children with a mid-parent height being -2SD below the mean in comparison to the children of tall parents, i.e. a mid-parental height equals to mean +2SD.

CONCLUSIONS

The age at childhood onset is equally important when studying children from Shanghai, as it is with their Swedish counterparts.

Birthweight and psychological distress in adult twins: a longitudinal study

AIM

To assess the relation between birthweight and psychological distress as measured by the Malaise Inventory in adult twins.

METHODS

Data were drawn from the 1958 British birth cohort study, which included twins followed from birth to age 42 y. We examined the relation between birthweight and psychological distress at ages 23, 33 and 42 y measured by the psychological scale of the Malaise Inventory. Analyses were performed both between subjects (n = 282) and within twin pairs (n = 112). The generalized estimating equations approach was used to handle the repeated measurements.

RESULTS

Between the 282 twins, the difference in psychological distress score was -0.45 (95% confidence interval -0.74 to -0.15) per Z-score increase in birthweight-for-gestational age. Within twin pairs, the heavier co-twins tended to have a psychological distress score lower than that of their lighter co-twins, the mean difference being -0.35 (-0.78 to 0.09).

CONCLUSION

Results from the between-subject analysis agreed with previous findings from adult singletons that psychological health is related to birthweight. The within-pair analysis suggested a similar relation but did not attain statistical significance.

Maternal smoking: an increasing unique risk factor for sudden infant death syndrome in Sweden

AIM

To assess the change of risk factors that are specific to sudden infant death syndrome (SIDS) after the initialization of a campaign to reduce the risk (RTR) of SIDS compared to non-SIDS postneonatal deaths.

METHODS

Data were extracted from the Swedish Medical Birth Registry, 1982-1991 and 1993-1998. 1105 infants died from SIDS during the postneonatal period. 2115 postneonatal deaths were from other causes and 11,050 live birth controls were selected. Risk factors previously identified to be related to SIDS were defined as high parity, prematurity, young maternal age, low Apgar score, birth during the night, single motherhood, multiple births, maternal smoking, male gender, short length standard deviation score (SDS) and small weight-to-length SDS.

RESULTS

Non-SIDS deaths were more significantly related to a low 5-min Apgar score, smaller weight-to-length SDS, and/or short length SDS values; while SIDS deaths were more closely related to mothers with higher parity or multiple births, mothers who smoked during pregnancy and single-parent (mother) families. Maternal smoking was even more prominent among SIDS deaths in the post-campaign period. The adjusted odds ratios, compared with non-SIDS deaths, increased from 1.84 (95% CI: 1.48, 2.28) in the pre-campaign period to 4.11 (95% CI: 2.72, 6.21) in the post-campaign period.

CONCLUSIONS

Maternal smoking during pregnancy remains the most important modifiable risk factor for SIDS in the post-campaign period in comparison with non-SIDS postneonatal deaths. Other than putting babies in a supine sleeping position, maternal smoking should be the next most important issue to be considered, if there is to be a second campaign.

Do men have a higher case fatality rate of severe acute respiratory syndrome than women do?

Severe acute respiratory syndrome (SARS) has been reported in 30 countries and regions, with a cumulative total of 8,099 probable cases and 774 deaths as of July 31, 2003, according to the World Health Organization. In Hong Kong, People's Republic of China, 1,755 SARS cases and 299 deaths had occurred as of September 22, 2003. The authors analyzed data from the Department of Health, Hong Kong SAR. The data series includes details regarding sex, age, and chronic disease history. Using data from early March to September 22, 2003, the authors found that males had a significantly (p < 0.0001) higher case fatality rate than females did, 21.9% versus 13.2%; the relative risk was 1.66 (95% confidence interval (CI): 1.35, 2.05), and it was 1.62 (95% CI: 1.21, 2.16) after adjustment for age. Subgroup analysis was conducted by excluding health care workers (n = 386) from the analysis. The overall crude relative risk of mortality was 1.41 (95% CI: 1.15, 1.74), and the adjusted relative risk was 1.48 (95% CI: 1.10, 2.00). Thus, among SARS patients, males may be more severely affected by the disease than females are. This finding could be related to a nonuniform case definition of SARS disease, a different treatment regimen, a past smoking history, work-environment factors, or gender-specific immune-defense factors, for instance.

Refining the Apgar score cut-off point for newborns at risk

AIM

To evaluate the Apgar score predictive power for mortality during different periods in the first year of life in a population with a very low mortality rate.

METHODS

The records of all singleton live births without severe congenital malformations and length of gestation >25 wk (n = 976635) were collected from the Swedish Medical Birth Registry, 1990 to 1998. Receiver operating characteristic (ROC) analysis was utilized.

RESULTS

Both the 1-min and the 5-min Apgar scores were shown to be good discriminators for early mortality, with the area under the ROC curve >0.85. For babies at risk of early death, the selected cut-off values for the 1-min Apgar score was <8 for preterm (true-positive (TP) rate: 83.9%; false-positive (FP) rate: 17.7%) and term babies (TP rate: 69.4%; FP rate: 6.7%). At 5 min, the analysis revealed that newborns with an Apgar score <9 were at risk for early death (preterm babies: TP rate: 79.8%; FP rate: 13.3%; term babies: TP rate: 73.8%; FP rate: 3.4%).

CONCLUSIONS

Our analysis did not support the common practice in the clinic or in research of grouping infants at risk in Apgar score groups, i.e. a score below 4 or a score below 7. However, the data presented here allow the clinicians and researchers to identify and define a suitable cut-off point in relation to the quality of neonatal care and resources available, rather than adhering to a historical cut-off value that has not been studied in depth.

Reference values for change in body mass index from birth to 18 years of age

Body mass index (BMI) has become the measure of choice for determination of nutritional status during the paediatric years, as in adults. Recently, several cross-sectional BMI childhood reference values standards have been published. In order precisely to evaluate childhood nutritional interventions, reference values allowing for the evaluation of changes in BMI values are also needed. For the first time, such reference values can be presented based on 3650 longitudinally followed healthy Swedish children born full term. The reference values for the change in BMI are given as the change in BMI standard deviation scores. The reference values are given as means of mathematical functions adjusting for gender, age of the child and the length of the interval between two measurements for interval lengths of 0.25 to 1.0 y before 2 y of age and of 1 to 5 y between birth and 18 y. The usefulness of the reference values is proved by a graph that forms a part of a clinical computer program; the -2 to +2 standard deviation range of the predicted change in BMI can be computed for an individual child and drawn in the graph as an extended support for clinical decision-making.

CONCLUSION

For the first time this communication gives access to BMI growth rate values that can be used both in research and in the clinic to evaluate various interventions, be they nutritional, surgical or therapeutic.

Birthweight distribution in southern Chinese infants with symptomatic congenital heart disease

OBJECTIVE

Western infants with congenital heart disease have frequently been reported to have a low birthweight for gestational age. Studies in Asian infants seem to be lacking in this area. This is the first extensive study presenting the birthweight distribution of Chinese newborns with symptomatic congenital heart disease.

METHODS

The birth data of 454, mainly southern-Chinese infants with symptomatic cardiovascular defects, born between 1990 and 1995 and admitted to Grantham Hospital, Hong Kong in 1994 and 1995, were analysed retrospectively. Infants with (non)-genetic syndromes or other major extracardiac malformations were excluded.

RESULTS

Fifteen per cent of all newborns had a birthweight below the reference mean of - 2 Z-score. After correction for length of gestation, no significant difference could be detected in birthweight between the cyanotic and acyanotic groups, nor between the different haemodynamic disturbances. Infants with atrial or ventricular septal defects, tetralogy of Fallot, pulmonary atresia with ventricular septal defect, heart with univentricular atrio-ventricular connection or double outlet right ventricle showed significant birthweight deficits. Transposition of the great arteries was not related to being small for gestational age.

CONCLUSIONS

Similar to Western infants, prenatal growth impairment was a common feature in Chinese infants with symptomatic congenital heart disease. The birthweight distribution in Chinese might be comparable to that in Western populations. Exceptions are possibly the high frequency of low birthweight in Chinese newborns with atrial septal defect or a single-ventricle abnormality. Further studies on fetal anthropometry and haemodynamics are necessary to provide insight into the relation between cardiovascular malformations and being small for gestational age.

Swedish population-based longitudinal reference values from birth to 18 years of age for height, weight and head circumference

This study aimed to update growth reference values for height, weight and head circumference in order to reflect the changes in body size in the Swedish population during the past two decades. The data came from a large longitudinal growth study on 3650 full-term healthy Swedish children who were born between 1973 and 1975. All of these 1801 girls and 1849 boys had longitudinal data for height and weight from birth to final height. Comparison with previous Swedish growth reference values based on children born between 1955 and 1958 revealed that there have been secular changes in body size. For instance, at 18 y of age, the updated height and weight reference values are 180.4 cm for males and 167.7 cm for females, i.e. 1.9 cm taller and 5.7 kg heavier for males and 2.3 cm taller and 3.4 kg heavier for females compared with the previous reference values.

CONCLUSION

These new growth reference values provide current national standards for growth monitoring and evaluation since the year 2000.

Association between psychological symptoms in adults and growth in early life: longitudinal follow up study

OBJECTIVES

To test the hypothesis that birth weight for gestational age and weight gain in early childhood have a long term association with psychological distress in adults.

DESIGN

Longitudinal study of 1958 birth cohort followed to age 42 years.

SETTING

Population based birth cohort study.

PARTICIPANTS

9731 cohort members with valid perinatal, postnatal, and adult data.

MAIN OUTCOME MEASURES

Malaise inventory scores measured at ages 23, 33, and 42 years. Generalised estimating equations approach used to analyse repeated measures.

RESULTS

Psychological distress score was inversely related to birthweight z score and weight gain from birth to the age of 7 years. A unit increase in birthweight z score or childhood weight gain was associated with a mean reduction in psychological distress score of 0.10 (95% confidence interval 0.05 to 0.15) and 0.06 (0.02 to 0.10), respectively. Birth weight and weight gain were also inversely related to the odds of having a high level of psychological distress, with odds ratios being 0.90 (0.85 to 0.95) and 0.93 (0.89 to 0.98), respectively.

CONCLUSIONS

Psychological health in adults is related to fetal growth and growth in early childhood.

Benn Index at birth is associated with postnatal linear growth

While previous research has suggested that body thinness is related to subsequent linear growth in children, it is unclear whether thinness at birth is related to linear growth in newborns and catch-up growth in small-forgestational age newborns. Drawing on data from a longitudinal growth study of 3,650 full-term Swedish babies, this study examines linear growth from birth to 6 months of age in three groups of newborns with short (< -2 SDS), appropriate (-2 to 2 SDS) and long (> 2 SDS) body length for gestational age. Among infants short at birth, the Benn Index (kg/m2.69) at birth was not related to the odds of short stature (< -2 SDS) at age 6 months (odds ratio = 1.03; p > 0.10). Nonetheless, the Benn Index was positively related to growth velocity in the first 6 months of life in the short (p = 0.060), appropriate (p < 0.05), and tall (p < 0.05) for gestational age newborns. Use of the Ponderal Index (kg/m3) would give similar results. The findings suggest that nutritional status at birth is related to linear growth velocity in newborns.

Timing of birth and infant and early neonatal mortality in Sweden 1973-95: longitudinal birth register study

OBJECTIVE

To assess the impact of time of birth on infant mortality and early neonatal mortality in full term and preterm births.

DESIGN

Analysis of data from the Swedish birth register, 1973-95.

PARTICIPANTS

2 102 324 spontaneous live births of infants without congenital malformation.

OUTCOME MEASUREMENTS

Absolute and relative risk of infant mortality, early neonatal mortality, and early neonatal mortality related to asphyxia.

RESULTS

Infant mortality, early neonatal mortality, and early neonatal mortality related to asphyxia were higher in infants who were born during the night (9 pm to 9 am) compared with those born during the day for 1973-9, 1980-9, and 1990-5. The difference was more dramatic for preterm infants. The largest difference was observed during 1990-5, when there was a 30% increase in early neonatal mortality (relative risk 1.31, 95% confidence interval 1.10 to 1.57) and a 70% increase in early neonatal mortality related to asphyxia (1.70, 1.22 to 2.38) in preterm infants born during the night compared with rates for preterm infants born during the day. A detailed analysis over 24 hours revealed two "high risk" periods: between 5 pm and 1 am and around 9 am.

CONCLUSIONS

Infants born during the night have a greater risk of infant and early neonatal mortality and early neonatal mortality related to asphyxia than those born during the day. There has been no improvement over the past two decades. The problem is more serious for preterm births and was even worse in the 1990s. Shift changes and the hours immediately after such changes are high risk periods for neonatal care.

Body mass index reference values (mean and SD) for Swedish children

Body mass index (BMI) is an important indicator of nutritional status. Many studies have been done to present BMI reference values in centile values rather than mean and SD values since its statistical distribution is positively skewed. Both height and weight growth charts are usually available in terms of mean and 1, 2 and 3 SD around the means; it would be of clinical value to produce BMI reference charts in a similar way. The aim of this work was to derive the mean and +/- 1, 2 and 3 SD BMI reference ranges as a supplement to the BMI centile reference values published previously for the same group of Swedish children. The method was based on an age-dependent Box transformation, and the beta-value was given as a third-degree polynomial function over the paediatric age. The BMI reference values can be given from mathematical functions in addition to values for specific ages.

CONCLUSION

The BMI reference values and charts derived as described effectively reflect the nature of the variant age-dependent positive skewed statistical distribution of BMI values in the population, and can serve as a valid supplementary tool in the evaluation of growth and nutrition during paediatric years.

Onset of significant GH dependence of serum IGF-I and IGF-binding protein 3 concentrations in early life

Forty-eight normal full-term Chinese babies (25 boys and 23 girls) were followed up every 2 mo in the first year and every 3 mo during the second year of life for anthropometric measurements. Blood samples were taken at birth and at 6, 10, 12, and 18 mo after birth for serum GH-binding protein, IGF-I, and IGF-binding protein 3 analysis. Onset of the childhood phase of growth in the infants was determined from the growth data plotted on Infancy-Childhood-Puberty growth charts. The serum GH-binding protein concentrations were low in cord blood but rose significantly at 6 mo, with slower rises in late infancy and early childhood. However, a significant rise in serum IGF-I and IGF-binding protein 3 levels was only observed from 10 mo of life onward. The change in IGF-I between birth and 6 mo was significantly correlated with length gain (r(2) = 0.35, p < 0.05) and body mass index gain (r(2) = 0.41, p < 0.01) during the same period. The 34 infants with onset of childhood phase of growth between 6 and 10 mo had a higher mean serum IGF-I value at 10 mo (8.8 +/- 5.8 nM versus 4.9 +/- 3.1 nM; p < 0.05) and higher length velocity between 10 and 12 mo (16.3 +/- 4.7 cm/y versus 8.8 +/- 4.3 cm/y; p < 0.001) compared with the 14 infants with a later onset after 10 mo of age. A significant correlation between a change in serum IGF-I and IGF-binding protein 3 levels was observed during the three 6-mo periods between birth and 18 mo, but a significant correlation between a change in serum GH-binding protein and a change in serum IGF-I or IGF-binding protein 3 levels was only seen between 12 and 18 mo of age. The multiple regression analysis (r(2) = 0.43, p = 0.0002) revealed that the change in serum GH-binding protein and IGF-I concentrations between 6 and 12 mo of age and the age of onset of childhood phase of growth could explain 43% of the length gain between 6 and 12 mo of age in our babies. The results of our study support the hypothesis that the onset of the childhood phase of growth is associated with the onset of significant GH action on growth.

Modelling variation of lower leg length growth in early life

We consider the estimation of sources of variation for panel data with repeated measurements. With no repeated measurements and known measurement error, models for variation decomposition have been proposed when there are one or more types of measurements. Estimation was performed using the EM algorithm accompanied by model augmentation that demands more computational efforts. In this article we extend previous variation models and modify the estimation methods in order to estimate various variation components after eliminating the unknown effects of measurement error. Specifically, methods that dispense with model augmentation and estimation of time-dependent covariates are considered. A set of lower leg length data from Chinese infants is analysed by using the proposed model. Interestingly, our results are consistent with the well-accepted three-phase (infancy-childhood-puberty) growth transition proposition for human growth. Moreover, gender effect is found to be time-varying.

Seasonality of growth in Shanghai infants (n=4128) born in 11 consecutive years

OBJECTIVE

To describe the seasonal growth patterns in Shanghai infants, to explore seasonal time lag between weight gain and length gain, and to investigate the long-term effect of birth season on early postnatal growth.

DESIGN

Community-based longitudinal study.

SETTING

Shanghai, People's Republic of China.

METHOD

Children were followed up monthly from 1 to 6 months, 3 monthly from 6 to 12 months, and 6 monthly from 12 to 24 months.

SUBJECTS

A total of 6018 children born between 1 January 1980 and 31 December 1990.

MAIN OUTCOME MEASURES

Weight gain, length gain and change in body mass index (BMI) over the seasons of the year.

RESULTS

The infants tended to grow faster in height in spring and summer, and faster in weight and BMI in autumn and winter. The seasonal effect on weight gain and length gain is largely independent. The mean length value at 1 month of age was about 2.0 cm higher in infants born in May to July than in those born in November to February. At 24 months of age this difference was reduced to about 0.7 cm.

CONCLUSIONS

There is a clear and consistent seasonality in growth in Shanghai infants. The seasonality seems to act independently on weight and length. Birth month has some association with attained size, but this is reduced during the first 2 y of life.

Critical growth phases for adult shortness in Hong Kong Chinese

OBJECTIVES

This study was designed to explore whether the influence of subnormal growth in fetal, infancy, childhood and pubertal phases on adult short stature was the same when comparing privileged and underprivileged populations.

METHODS

Data came from two longitudinal growth studies: 1) Hong Kong Chinese children who were born in 1967 (n = 132), and 2) the comparatively more privileged Swedish children who were born in 1973-1975 (n = 2,850).

RESULTS

68% of Hong Kong Chinese children had two or more growth phases subnormal, much higher than the 12.4% for Swedish children. 42.4% of Hong Kong Chinese were short at final height, much higher than the 2.2% for Swedish children. Subnormal growth in any growth phase was associated with an increased risk of adult short stature in both series (p<0.01). After adjustment for mid-parental height, the place of residence (Hong Kong/Sweden) was not significant for adult shortness (p>0.05) in the pooled data.

CONCLUSIONS

The impact of subnormal growth in any phase on adult shortness is similar in privileged and underprivileged populations. The much higher prevalence of subnormal growth and consequently adult short stature in developing countries is likely mainly attributable to adverse extrinsic or environmental influences.

Bmi in childhood and its association with height gain, timing of puberty, and final height

No large population-based study has addressed the question of how overnutrition is related to subsequent height gain in childhood, timing of puberty, and final height. The present data represent a large Swedish population-based longitudinal growth study. Height gain in childhood, timing of reaching peak height velocity and height gain during adolescence, and final height were regarded as the short-term, interim, and long-term outcomes of childhood nutritional status, i.e. body mass index (BMI) change between 2 and 8 y. Midparental height was adjusted as the genetic influence on linear growth of the child. Childhood BMI gain was related to an increased height gain during the same period, i.e. an increase of 1 BMI unit was associated with an increase in height of 0.23 cm in boys and 0.29 cm in girls. A higher BMI gain in childhood was related to an earlier onset of puberty; the impact on the timing of puberty was 0.6 y in boys and 0.7 y in girls. Each increased unit of BMI gain in childhood also reduced the height gain in adolescence, 0.88 cm for boys and 0.51 cm for girls. No direct correlation was shown between childhood BMI gain and final height. We conclude that overnutrition between 2 and 8 y of age will not be beneficial from a final height point of view, as the temporary increase in height gain in childhood will be compensated by an earlier pubertal maturity and a subnormal height gain in adolescence.

Mortality of twins and singletons by gestational age: a varying-coefficient approach

This study used data from the Swedish Medical Birth Registry between 1982 and 1995 to address the question of whether there is higher mortality in twins in relation to singletons of the same gestational age and to examine the optimal gestational age range for twins. A "varying-coefficient approach" was adopted to estimate the gestational age-specific relative and absolute risks of mortality in twins and singletons, adjusting for size at birth and risk factors of short gestational duration. The models showed that twins born between 29 and 37 weeks of gestation had lower mortality than did singletons of the same gestational age. Twins born at older gestational age had higher mortality than did their singleton counterparts, because longer gestational duration was more advantageous to singletons than to twins. Without adjustment for size at birth, there was an upturn of mortality in twins born after 38 weeks. It is postulated that twins have better health than singletons initially, but they could not enjoy the benefit of a longer gestational duration as much as singletons could. The optimal gestational age for twins appeared to be 37-39 weeks according to neonatal and infant mortality.

Risk factors of obesity in preschool children in China: a population-based case--control study

OBJECTIVE

To identify the risk factors of obesity in preschool children in China.

DESIGN

A nationwide case-control study in eight cities in China.

SUBJECTS

A total of 748 boys and 574 girls (age 0.1 - 6.9 y), including the obese and non-obese. One obese child was matched with one non-obese child by sex and age.

MEASUREMENTS

Weight, height and other measurements were taken for all the subjects. The information on child activity, feeding pattern and family background was collected by our study team from the parents and the kindergarten teacher.

RESULTS

Birth weight > or =4.0 kg, high eating speed, obesity among the child's relatives > or =25%, mother's body mass index (BMI) > 25 kg/m2 and father's BMI > 25 kg/m2 were identified as the major significant (P < 0.05) risk factors of obesity in preschool children in China.

CONCLUSION

Family history of obesity, high birth weight and eating speed were identified as the risk factors of obesity in preschool children in China. Our results show the need for family therapy as part of an intervention program for childhood obesity, which includes behavior modification.

Fetal growth and early postnatal growth are related to blood pressure in adults

It is commonly agreed that birth weight is associated with blood pressure in adults. However, not much is known about birth length, ponderal index, and early postnatal growth, whose effects on adult blood pressure, if any, can affect the interpretation of the birth weight-blood pressure association. This study examined the association between fetal growth, early postnatal growth, and blood pressure in Chinese adults. One hundred twenty-two subjects born in Hong Kong in 1967 were followed from birth to age 30 years. Multiple linear regression was used to analyze the association between size at birth, postnatal changes in body size, and systolic and diastolic blood pressure at age 30 years. Having adjusted for potential confounders and each other explanatory variable, it is found that birth length standard deviation score (regression coefficient or beta=-3.2), ponderal index at birth (beta=-1.8), and postnatal changes in ponderal index from age 6 months to 18 months (beta=-2.2) were inversely associated with systolic blood pressure (each P<0.05). Postnatal changes in length standard deviation score were not significantly associated with systolic blood pressure. Birth length standard deviation score was inversely associated with diastolic blood pressure at age 30 years (beta=-2.6; P<0.05). Other anthropometric variables were not associated with diastolic blood pressure. The results support the hypotheses that both fetal growth and early postnatal growth may have a long-term impact on blood pressure in adults. It also highlights the importance of differentiating length and weight for length.

Children born small-for-gestational age: postnatal growth and hormonal status

It is generally recognized that children born small-for-gestational age (SGA) have a 5-7 times higher risk of short stature than children born at normal size. It has been suggested that the programming of the endocrine axes occurs during critical phases of fetal development and is affected by intrauterine growth retardation. This study was undertaken to characterize the postnatal growth pattern and the final height of children born SGA, as part of a population- based study (n = 3,650), from birth to final height, and to evaluate the hormonal status in another group of prepubertal children born SGA (n = 134) without postnatal catch-up growth. The majority (88%) of 'healthy' full-term singleton SGA infants achieved catch-up growth during the first 2 years of life, and most of the increase in height occurred by 2 months of age. The SGA children who remained short at 2 years of age had a higher risk of short stature later in life. The risk of having a short final height (<-2 SDS) was five times higher for children with a low birth weight and seven times higher for those with a low birth length in comparison with children with a normal birth size. Moreover, about 20% of all children of short stature were born SGA. As a group, children born SGA will have a final height, expressed in SDS, as they had during the prepubertal years. This is in contrast to children, who became short postnatally. During puberty, these short children will have a mean height gain of 0.6 SDS for girls and 0.7 SDS for boys. The mean estimated secretion rate for growth hormone (GH) was lower in the short children born SGA compared with the reference groups born at an appropriate size for gestational age, of either short (p < 0.05) or normal stature (p < 0.001). Moreover, in the youngest children born SGA (2-6 years of age) a different pattern of GH secretion was found, with a high basal GH level, low peak amplitude, and high peak frequency. The majority of the children born SGA had levels of GH-binding protein within the range previously reported for normal children. However, the levels of insulin-like growth factor I (IGF-I), IGF-binding protein-3 (IGFBP-3) and leptin were significantly reduced compared with the reference values (p < 0.001, p < 0.01 and p < 0.001, respectively). In conclusion, the low spontaneous GH secretion rate and a disturbed GH secretion pattern, together with low serum levels of IGF-I, IGFBP-3 and leptin, might contribute to the reduced postnatal growth in some of the subgroup of children born SGA who remained short during childhood.

Blood pressure is associated with body mass index in both normal and obese children

Obesity is associated with elevated blood pressure (BP) both in adults and children. Childhood obesity has become a severe health problem, especially during the last few decades. So far there has not been any large-scale study specifically focusing on the association between obesity and BP in early life. The aim of this study is to examine systematically the association between obesity and BP in preschool Chinese children in mainland China. In 1996, measurements of weight, height, and BP values were collected in a nationwide, case-control study of 748 boys and 574 girls who ranged in age from 0.1 to 6.9 years in 8 cities in mainland China. One obese child and 1 nonobese child were matched for gender and age. The BP differences of the mean-matched pair were approximately 5 mm Hg for systolic blood pressure (SBP) and approximately 4 mm Hg for diastolic blood pressure (DBP) (P<0.05); a higher value was noted in obese children. The BP value of 19.4% children in the obese group and 7.0% children in the nonobese group was higher than the 95th percentile value (P<0.0001), which is defined as high BP by the Task Force on Blood Pressure Control in Children. Both SBP and DBP were significantly (P<0.05) positively related to body mass index (BMI) values (P<0.05) for children in obese and nonobese groups after adjustment for age, gender, and height. To be specific, an increase of 1 BMI unit was associated with, on average, an increase of 0.56 mm Hg and 0.54 mm Hg in SBP and DBP, respectively, for obese children. In nonobese children, the increase in SBP and DBP was 1.22 mm Hg and 1.20 mm Hg, respectively. An increase in the BMI is conclusively associated with elevated SBP and DBP in nonobese children. Furthermore, an increase in the adjusted BMI was associated with an increase in SBP and DBP in obese and nonobese children.

Critical growth phases for adult shortness

Previous growth studies have not explored how different growth phases-the fetal, infancy, childhood, and puberty phases-interact with each other in the development of adult shortness. In this paper, the authors attempt to describe the importance of each growth phase for adult shortness and the effect of growth in one phase on other, subsequent phases. The authors analyzed data from a longitudinal population-based growth study of 2,850 healthy, full term Swedish children born between 1973 and 1975. The height values were transformed into a centimeter score (CMS) by subtracting the raw values from the reference mean values for a particular age and sex. Subnormal growth in any growth phase, as defined by a decrease of 3 CMS or more during a growth phase, was associated with significant increased risk for final heights below 0, -6, and -12 CMS. For children with subnormal growth during one, two, and three phases, the percentages for final height below -12 CMS (a standard deviation score of approximately -2) were 0.5%, 9.4%, and 75%, respectively. Most children (57/62) with a final height below -12 CMS had subnormal growth in two or three phases. Height gains during the four growth phases were interdependent. The infancy phase was negatively associated with fetal growth (r = -0.33, p < 0.01); the childhood phase was positively associated with infancy growth (r = 0.21, p < 0.01); and the puberty phase was negatively associated with childhood growth (r = -0.10, p < 0.01).

New reference for the age at childhood onset of growth and secular trend in the timing of puberty in Swedish

The objectives of the present work were to present a new reference for the age at childhood onset of growth and to investigate the secular trend in the timing of puberty in a community-based normal population in Sweden. A total of 2,432 children with longitudinal length/height data from birth to adulthood were used to determine the two measures by visual inspection of the measured attained length/height and the change in growth velocity displayed on a computer-generated infancy-childhood-puberty (ICP) based growth chart. The series represents a sample of normal full-term children born around 1974 in Göteborg, Sweden. We found about 10% of children were delayed (>12 mo of age) in the childhood onset of growth based on the previous reported normal range, i.e. 14% in boys and 8% in girls. Distribution of the age at childhood onset of growth was skewed. The medians were 10 and 9 mo for boys and girls, respectively. After natural logarithmic transformation, the mean and standard deviation (SD) were 2.29 (anti-log 9.9 mo) and 0.226 for boys, 2.23 (anti-log 9.3 mo) and 0.220 for girls, respectively. The 95% normal ranges were 6.3-15.4 and 6.0-14.3 for boys and girls, respectively. The distribution of the timing of PHV was close to the normal distribution. The mean values were 13.5 y for boys and 11.6 y for girls with 1 y SD for both sexes.

CONCLUSION

A downward secular trend in the onset of puberty was clearly shown in the population. The age at childhood onset of growth did not correlate with the timing of puberty (r = -0.01 and 0.05, p > 0.7 and 0.1 in boys and girls, respectively). Normal ranges of the age at childhood onset of growth are in need of revise, as this study indicates. The new reference presented here could be a reliable indicator in further studies.

Foetal size to final height

It is well known that some adult diseases, such as cardiovascular diseases, may be programmed during foetal life. It is not clear, however, whether final height may be predicted from foetal growth. A longitudinal cohort of full-term healthy Swedish babies (n = 3,650) was followed up from birth to maturity in a population-based growth study. Length or height and its changes were analysed from birth to 18 y of age; 2,807 children, with data available on birth length, final height and parental height, were included in this analysis. The result clearly shows that length at birth relates to final height. In terms of standard deviation scores (SDS), the mean difference in length at birth from the mean was greatly decreased in final height, but retained the same order as was seen at birth. In terms of centimeter difference from the reference mean values, the difference in length at birth remained roughly stable into final height. For instance, babies 5 cm above or below the mean birth length will end up approximately 5 cm above or below the mean in final height. Parental height-a surrogate value of the genetic final height potential of an individual-is shown to influence postnatal growth in height strongly. However, the difference from the mean in length at birth remained into adulthood within the same midparental height group.

CONCLUSION

This study reveals that trends in foetal linear growth continue into maturity. Foetal growth is a significant predictor of postnatal growth. Final height is dependent on both the magnitude of foetal growth and the genetic potential in stature, and appears to some extent to be programmed from foetal growth.

Population-based body mass index reference values from Göteborg, sweden: birth to 18 years of age

The body mass index or BMI (weight/height2) is a somewhat crude estimate of nutritional status. However, due to its simplicity and high correlation with total body fat, it has been the method of choice in both paediatric clinics and research over the years. Since BMI is not an equivalent measure of the percentage of body fat in different ethnic groups and in the two sexes, population-specific BMI reference data is needed. Several BMI reference values have been published for French, American, British and Hong Kong children in recent years. In Sweden, weight-for-age and height-for-age reference values, which were published in 1976, are still used as the current national growth reference values. Updated growth reference values are needed for assessing nutritional status due to the secular trend toward and increasing prevalence of childhood obesity. The aim of this study was to produce BMI reference values for Swedish children of paediatric age. The series came from a large Swedish population-based longitudinal growth study of 3650 full-term babies followed from birth to 18 y of age. The children in this data set were born in the early 1970s. The pattern and level of 50th centile BMI values presented here are quite similar to those of the Swedish cohort study in the 1950s. In comparison with the US BMI reference values, the Swedish values are much lower, especially for the higher centile values.

CONCLUSION

The new Swedish BMI chart from our study may provide a useful tool for paediatricians to assess body fat, and consequently nutritional status, in Swedish children.

Long-term consequences of early linear growth retardation (stunting) in Swedish children

Linear growth retardation (stunting) is prevalent (10-80%) in developing countries. It takes place between 6 and 18 mo of age and can be characterized by the delayed onset of the childhood phase of growth. The objective of this work was to study the existence of stunting in early life and, when present, its lifelong effects on final height in a normal healthy population in a developed country. A total of 2432 children with longitudinal height data from birth to adulthood were analyzed. The series represents a normal healthy population-based sample of full-term children born around 1974 in Göteborg, Sweden. We found the incidence of stunting to be about 10% as indicated by delayed onset of the childhood phase of growth (>12 mo of age): 14% in boys and 8% in girls. In a multiple linear regression model, final height was described by length at 6 mo of age, age at onset of the childhood phase of growth, and age at peak height velocity during puberty. All three measures contributed significantly (p < 0.05) to final height with a total R2 of 0.33. A final height difference of 5.0 cm could be explained by abnormal onset of the childhood phase of growth in this normal population. The timing of the childhood onset of growth did not correlate with the timing of puberty. For the first time, we can conclude that stunting in early life is not only a problem in developing countries. Its importance on final height achieved is clearly shown, even after controlling for the timing of puberty.

Prevalence of childhood asthma, allergic rhinitis and eczema in Urumqi and Beijing

OBJECTIVE

To assess the prevalence of asthma and other allergies in children in Urumqi and Beijing, compared with that in Hong Kong.

METHODS

A total of 7754 primary school students were randomly selected to participate in the study. Data were collected in 1995-96 using the International Study of Asthma and Allergies in Childhood (ISAAC) protocol for 6- and 7-year-olds. The study design and data quality assurance in all aspects followed the protocol, including the double entry of data.

RESULTS

The questionnaire response rate was high (98.9%). Beijing children reported significantly more asthma than those living in Urumqi in three categories: wheezing or whistling in the chest in the past year (6.0%, 95% confidence interval (CI): 5.1-6.9% vs 2.9%, 2.3-3.5%, P < 0.001), sleeping disturbed due to wheezing (1.3%, 0.9-1.7% vs 0.6%, 0.3-0.9%, P < 0. 03) and having experienced asthma ever (10.7%, 9.6-11.8% vs 7.6%, 6. 6-8.6%, P < 0.001). The prevalence of allergic rhinitis (30.0% vs 31. 1% estimated as 12-month nasal symptoms in Beijing and Urumqi, respectively) and eczema (2.8% vs 2.0% recorded as 12-month itchy rash) in the two cities were not significantly different. A similar study was also performed in Hong Kong in 1995 using the same ISAAC protocol. The children in Urumqi and Beijing had fewer (P < 0.05) allergic symptoms compared to those living in Hong Kong (n = 3618). The 12-month prevalence of wheezing, nasal symptoms and itchy rash found in Hong Kong were 9.2% (95% CI: 8.2-10.2%), 35.1% (33.5-36.7%) and 4.2% (3.5-4.9%), respectively.

CONCLUSION

: Urumqi, Beijing and Hong Kong represent communities at different stages of westernization and the results from these three cities reflect a worldwide trend for an increasing prevalence of allergies along with westernization. These three cities could assist in identifying risk factors involved in the increase in asthma, allergic rhinitis and eczema.

Distribution of symptomatic congenital heart disease in Hong Kong

Racial group studies have identified differences in the occurrence of congenital heart disease (CHD) among ethnic populations. The aim of this study was to characterize the proportionate frequency and clinical profile of children with symptomatic cardiac abnormalities in Hong Kong. The hospital records of 666, mainly Southern Chinese children with symptomatic CHD, who were 4 years of age or younger and who were admitted to Grantham Hospital, Hong Kong, in 1994 and 1995 were analyzed retrospectively. Left-to-right shunting (45.0%) and pulmonary outflow obstruction (34.4%) were the most frequently diagnosed categories, followed by left ventricular outflow obstruction (8.3%), transposition of the great arteries (4.2%), conditions with intracardiac mixing (3.9%), and other cardiac lesions (4.2%). Compared with Western studies, pulmonary outflow obstruction (p<0.0001), particularly tetralogy of Fallot and critical pulmonary stenosis, were more frequent in Chinese children. In contrast with previous reports, coarctation of the aorta (5%) does not seem to be uncommon in Chinese patients. Conversely, aortic stenosis and hypoplastic left ventricle may be rare in these children (1% vs 3% and 3-7%). Other cardiac lesions showed no consistent racial difference in the frequency of occurrence. Chinese patients with Down's syndrome had ventricular septal defect (38%) as the predominant lesion followed by atrioventricular septal defect (25%). Western studies usually report a reverse pattern for these two lesions. The mortality rate for the total cohort was 7.5%. However, of those with conditions with intracardiac mixing and left ventricular outflow tract obstruction many did not survive childhood (20% and 21%, respectively).

Fetal size to final height in Hong Kong Chinese children

OBJECTIVES

It has been known that size at birth is important for postnatal growth and final height. However, there are few data in the literature on the difference in height growth patterns from fetal size to final height between less privileged and more privileged populations. The aim of this study was to describe the important features in height growth from birth to maturity in an underprivileged Hong Kong Chinese cohort in comparison to a more privileged Swedish cohort.

METHODS

The longitudinal height growth data from birth to maturity in full-term healthy Hong Kong Chinese children (n=132) who were born in 1967 were analyzed, and compared with those for Swedish children who were born in 1973-75 (n=3650).

RESULTS

Children with longer birth length achieved taller adult stature with respect to their target height. The mean final height retained the same order as that of the mean length at birth for various birth length groups. All children in the Hong Kong Chinese series showed catch-down height growth during the first 2 years of life, in contrast to the catch-up in smaller babies and catch-down in larger babies for the Swedish series. The growth deficit for the Hong Kong Chinese was -0.9 SDS at birth, -1.8 SDS at 2.0 years of age, -2.1 SDS at 8 years of age, and -1.7 SDS at final height.

CONCLUSIONS

Fetal size is important for postnatal growth and attained final height with respect to a child's familial genetic potential in stature, not only for privileged populations, but also for underprivileged populations. However, children in underprivileged populations experience a persistent increasing growth deficit during infancy and childhood. Special attention should be given to monitor their growth status in early years and to institute appropriate intervention programs.

Prediction of adult overweight during the pediatric years

Obesity in adults is associated with an increased mortality rate from various diseases. Childhood overweight or obesity may persist into adulthood, and for this reason it is important to identify such children at risk. The data were taken from a larger longitudinal growth study of 3650 full-term and healthy Swedish babies followed from birth to 18 y of age. Body mass index (BMI) was used to estimate (during the pediatric years) the risk of obesity at 18 y of age. A probability chart for becoming overweight (>25 kg/m2) by 18 y of age was constructed, For example, in girls, a BMI of 16 kg/m2 at 4 y of age is associated with a 20% risk of attaining a BMI value over 25 kg/m2 at 18 y, and in boys at 4 y of age, a BMI of 19 kg/m2 leads to a 60% risk that they will have a BMI value over 25 kg/m2 at 18 y. The probability risk charts for adult overweight developed here is the first one presented. They provide an easy and novel instrument to use to identify those children who are at high risk of becoming obese adults, so that they may have clinical intervention at younger ages.

Early linear growth retardation in Chongqing, China

OBJECTIVES

The objectives of this study were to investigate early linear growth retardation and to identify potential risk factors for it.

METHODOLOGY

A community-based prospective study was performed in Chongqing, China, with infants being examined at 12 and 15 months of age. A total of 101 infants were examined twice. Supine length, bodyweight, lower leg length, head circumference, middle upper arm circumference and biceps skinfold thickness were measured.

RESULTS

The prevalence of short stature (length-for-age standard deviation score, < -2) was 22% and 23% at 12 and 15 months of age, respectively, while the incidence of growth stunting (low growth velocity) between 12 and 15 months of age was 24%, using the stunting screening method. A mother having a history of abortion and infants having current episodes of diarrhoea were identified as risk factors for growth stunting.

CONCLUSIONS

We conclude that both short stature and growth stunting are common in the population studied. The stunting screening method used is applicable in developing communities.

A comparison of target height estimated and final height attained between Swedish and Hong Kong Chinese children

Target height, the genetic potential in stature, is commonly estimated by the corrected midparental height (CMH) method. A new model for estimating target height has recently been introduced based on a large, Swedish, population-based study. The aim of this study was to compare the validity of the two methods for estimating target height in Hong Kong Chinese children. The Hong Kong Chinese were more than 10 cm shorter than the Swedes in stature. The secular increase in height over the two generations, however, was 4.2-4.8 cm for the Hong Kong Chinese, much larger than that of the Swedes (0.7-1.0 cm). The two populations are thus at different stages in the secular trend. The new model derived from Swedish population for estimating target height was shown to be applicable to Hong Kong Chinese children; the mean of residual final height values was close to zero (-0.15 cm, p = 0.74). However, the mean of residual final height was significantly above the expected value of zero (4.5 cm, p<0.0001) when the CMH method was applied to the data, which implies an underestimation bias of 4.5 cm. Consequently, if the CMH method is used to estimate target height and evaluate growth hormone treatment responses in short children, it may inflate the treatment response by 4.5 cm. In conclusion, the recently proposed model for target height estimation offers a better alternative for estimating target height in Hong Kong Chinese children and for assessing growth-promoting treatments.

Length and body mass index at birth and target height influences on patterns of postnatal growth in children born small for gestational age

OBJECTIVE

Previous growth studies on children born small for gestational age (SGA) indicate that birth length, weight, and target height are important predictors for postnatal catch-up growth in SGA. Their influences on different phases of catch-up growth are still not described. The aim of this study was to clarify the influences of target height, length, and nutritional status at birth on different phases of postnatal catch-up growth (infancy, childhood, puberty) in SGA and the long-term consequences.

METHODS

Data were obtained from a longitudinal population-based growth study on Swedish children (N = 2815). Primary outcome measurements include heights, the changes in height standard deviation scores (SDS) during various phases of growth and relative risk for adult shortness.

RESULTS

The difference in final height in children born SGA was attributable to their difference in target height and the magnitude of catch-up growth during the first 6 months of life, rather than the difference in length or body mass index (BMI) at birth. Length at birth showed negative influence on catch-up growth during infancy (0 to 2 years of age), but no significant influence thereafter. The BMI or weight for length SDS at birth showed no significant influence on catch-up growth during any growth phase. Target height showed positive influence on catch-up growth from the onset of childhood. Neither target height nor length and BMI at birth showed any significant influence on catch-up growth during puberty. The magnitude of catch-up growth during infancy, especially the first 6 months of life, is most critical in decreasing risk at adult shortness. We confirmed that the SGA group had a sevenfold greater risk for adult shortness than the non-SGA group (relative risk = 7.31; 95% confidence interval: 3.96-13.52). However, approximately 40% of children who were below -2 in height SDS at 2 years of age remained short at final height in both SGA and non-SGA groups. The mean height SDS of children born SGA increased by 1.65 from birth to final height, but the length deficit in centimeters at birth (-5.4 cm) persisted into adulthood (-5.9 cm).

CONCLUSIONS

BMI at birth is not related to postnatal catch-up growth in infants born SGA, but birth length and target height are important. The genetic influence on catch-up growth appears to start from the onset of childhood. Being born short or becoming short during the first 2 years of life is similar in terms of risk for adult short stature.

Allergen sensitization of asthmatic and nonasthmatic schoolchildren in Costa Rica

The prevalence of asthma among schoolchildren in Costa Rica is very high -- at the level of 20-30% -- and the reason is still unknown. A group of children from our previous epidemiologic study was randomly selected in order to establish the relation between asthma symptoms and allergy sensitization to common allergens. Serum samples from children with and without asthma were analyzed for the presence of IgE antibodies to 36 different allergens, for the presence of IgE antibodies to a pool of 10 common allergens, and for total serum IgE. The most prevalent IgE antibodies were those to mite, cockroach, dog, and house-dust allergens with MAST pipettes for the serologic measurements. Positive reactions to house dust, mite, cat, and the two molds (Alternaria and Cladosporium), and food allergens such as egg white, peanut, and shellfish were significantly more prevalent among the asthmatics than the nonasthmatics. Sensitization was equally prevalent at different ages, but the house-dust, mite, cat, dog, cockroach, Alternaria, and egg-white allergens had sensitized boys more often than girls (P < 0.01). The result of the analysis of IgE antibodies to a pool of 10 common allergens by Phadiatop supported the MAST pipette results, showing allergen sensitization in 57.7% of the asthmatic children and 42.3% in the nonasthmatic group. The concentration of IgE was significantly higher among the asthmatic children (372.2 kU/l) than among the nonasthmatic children (249.1 kU/l) (P < 0.00001). Parasitic infestations were not examined in this study, but in most of Costa Rica these have largely been eliminated and could not explain the high total IgE levels. Our data indicate that the very high prevalence of bronchial asthma in Costa Rican schoolchildren can be related to sensitization, especially to airborne indoor allergens such as those of mites, cockroaches, and dogs.

High seroprevalence of Helicobacter pylori in active bronchiectasis

Helicobacter pylori causes chronic inflammation of the gastric mucosa and has been identified in tracheobronchial secretions. Serum IgG against H. pylori was therefore measured prospectively in consecutive subjects with bronchiectasis (n = 100; mean age +/- SD 55.1 +/- 16.7 yr), active pulmonary tuberculosis (n = 87; age, 57.3 +/- 19.1 yr), and healthy volunteers (n = 94; age, 54.6 +/- 7.6 yr). Seropositivity was found in 76.0% of bronchiectatic subjects, which was significantly higher than that of the control (54.3%, p = 0.001) and tuberculous (52.9%, p = 0.0001) groups. Multiple logistic regression, adjusted for age, sex, occupational social class, and number of persons living in the household, showed that H. pylori IgG levels of the bronchiectatic group were still significantly higher than that of the control (p = 0.0014) and tuberculous (p = 0.0154) groups. Multiple regression analysis revealed associations between H. pylori serology and sputum volume (p = 0.03) and age (p = 0.001) in the bronchiectatic patients, but not lung function indices or causes of bronchiectasis. The H. pylori seroprevalence in bronchiectasis was significantly (p = 0.0002) higher in patients who produced more (83.1%) than those who produced less than 5 ml sputum/24 h (58.6%). This is the first report of a high H. pylori seroprevalence in bronchiectasis which appears to be specific. Further studies are indicated to evaluate the possible pathogenic role of H. pylori in bronchiectasis.

Target height as predicted by parental heights in a population-based study

The corrected midparental height method was introduced by Tanner in 1970 (Tanner method) and is commonly used to estimate target height in children to evaluate the effectiveness of growth-promoting therapies. It has not been established if the equation used to compute target height should be the same for children with short, normal, or tall parents. In this study, we examined the predicted target height values by parental heights in a large population-based study (n = 2402). A simple linear function of midparental height (x) was proposed to estimate target height (y): y = 45.99 + 0.78x (boys), y = 37.85+0.75x (girls), with a 95% predicted interval of about +/-10 cm. The prediction model was similar for boys and girls in SD scores (SDS), and was not affected by assortative mating or difference in parental heights. The model may underestimate the potential stature by about 2 cm for children with midparental height below -2 SDS, or 163 cm. In comparison, the Tanner method may lead to a 6-cm error in underestimating target height for these children. The function would be a better choice than the Tanner method for estimating target height in the clinical evaluation of growth promotion treatments because it is common that short children also have short parents. Children with very short parents will usually be much taller than their parents in adult stature, and we believe that a different function should be developed. The results support the proposed nondominant, non-sex-linked, polygenic inheritance in stature. The estimated heritability values were 0.75-0.78 in cm or 0.55-0.60 in SDS.

Hong Kong clinical trials published in Medline between 1987 and 1996

The aims of this study were to determine the number of randomised clinical trials where the correspondence address included the words 'Hong Kong' for the years of publication between January 1987 and December 1996, and to study their characteristics: year of publication, disease area, sample size, and correspondence address of the department and institute; MEDLINE EXPRESS was used for the search. A total of 5605 publications were identified, of which only 170 (3.0%) were found to be randomised clinical trials. No significant increase in the proportion of randomised clinical trials could be seen during the decade of publication (P>0.05). Approximatley 50% of the trials had a small sample size (fewer than 75 subjects) and most randomised clinical trials were performed in the field of internal medicine, followed by surgery, and obstetrics and gynaecology (total, 69%). The predominant research area was gastrointestinal disease (34%). Despite a relatively high academic output from Hong Kong, the number of randomised clinical trials has not increased much during the decade.

The effect of platyspondyly and pubertal growth spurt on the stature of patients with beta-thalassaemia major

OBJECTIVE

To study the effect of the body proportion and pubertal growth spurt on the stature of children with beta-thalassaemia major.

METHODS

The height, sitting height, upper to lower segment (U:L) ratio and pubertal development were determined in 71 Chinese children (38 girls and 33 boys) with beta-thalassaemia. The growth patterns of 20 patients with complete growth data between 3 years and final height, were analyzed according to whether they underwent a pubertal growth spurt or not.

RESULTS

27% of the boys and 32% of the girls had a height below the 3rd percentile. About 60% of all the children had a U:L ratio below the 10th percentile for age. Abnormal body proportion was found in patients with or without growth retardation. 34% of the 41 children over the age of 14 years underwent spontaneous puberty. In 28 patients over the age of 16 years, a growth spurt was observed in 46% of the children during spontaneous or induced puberty. The retrospective analysis showed that the height deviation from the mean in adulthood was significantly higher in patients without pubertal growth acceleration than in those with a growth spurt (x = -11.8 cm, s = 7.6 cm vs x = -4.4 cm, s = 4.4 cm; P = 0.02).

CONCLUSIONS

An abnormal U:L ratio was commonly observed in patients with beta-thalassaemia major and may be one factor contributing to the short stature of these patients. Abnormal puberty was present in a significant proportion of children and the lack of a pubertal growth spurt was found to be detrimental to adult height.

Modification of the MTT method for the study of bilirubin cytotoxicity

BACKGROUND

We propose a modification of the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] method to study the cytotoxicity of bilirubin. The original method involves reading the intensity of a purplish blue color resulting from the conversion of MTT to formazan crystals by the mitochondria of viable cells. We have found that when the method is applied to study the effect of bilirubin on growing cells, precipitation of the yellow bilirubin pigment interferes with the colorimetric reading.

METHODS

A human liver cell line was used. The interference of bilirubin deposition on the MTT assay was investigated by comparing the value of optical density of the MTT solution in the presence and absence of bilirubin. The effect of 0.04 mol/L HCL-isopropanol on the bilirubin precipitate was tested by recovering the amount of bilirubin from the wells after the isopropanol treatment.

RESULTS

Bilirubin deposition increases MTT reading by 10-24%. Hydrochloride-isopropanol (0.04 mol/L) dissolves MTT formazan only without disturbing the bilirubin precipitates. The bilirubin extracted into the supernatant was less than 5% of the total bilirubin deposited.

DISCUSSION

This indirect MTT assay, as developed in this study, could eliminate the interference of bilirubin deposits and serve as a good method for the study of bilirubin cytotoxicity.

Evidence-based medical practice: as viewed by a clinical epidemiologist

Evidence-based medicine shifts the basis for clinical decision making from intuition and unsystematic clinical experience to the examination of evidence resulting from clinical research. Evidence-based medicine is the same as good clinical research. The selection of a proper study design for a specific study objective is the most important cornerstone of good clinical research. Evidence-based medicine places great importance on the design of a study, with optimum evidence being obtained from the randomised, controlled clinical trial. However, various study designs are equally important-if properly used-in the process of searching for solid and important evidence for use in clinical practice. There should be an emphasis on a quality improvement shift in research design from retrospective to prospective, cross-sectional to longitudinal, uncontrolled to controlled, and non-randomised to randomised. The reasons for using suboptimal study designs in clinical research are the lack of formal research training and the pressure to obtain academic output without being motivated. Research design is one of the most important aspects to study and practice, and there are four simple rules: (1) avoid retrospective studies; (2) focus on prospective studies; (3) use controls, randomisation, and blinding; and (4) always discuss the research design with an experienced researcher or statistician before commencing the study.

Risk factors for impaired length growth in early life viewed in terms of the infancy-childhood-puberty (ICP) growth model

A difference in length or height growth is observed between populations in developing and developed countries and, within a country, between well-off and poor subpopulations. The aim of this community-based study was to identify possible risk factors for growth stunting in early life using the infancy-childhood-puberty (ICP) growth model. A periurban slum, a village and an urban slum, plus a local control group in Lahore, Pakistan, were investigated. In total, there were 425 infants born from September 1984 to March 1987 in the study. Children were followed from birth to 24 months and anthropometric data were recorded at each examination. A well-designed questionnaire was used to collect background information to identify risk factors for growth faltering over this period. The age at onset of the childhood component was determined individually using the ICP growth model. A late onset was defined as onset after 12 months. Univariate odds ratio analysis identified 20 significant risk factors for a delayed childhood onset in the total sample. Backward logistical analysis showed that only six variables were significant, i.e. small number of rooms in a household, episode(s) of diarrhoeal diseases, poor house structure, low body mass index, early end of breastfeeding and birth in warm season. Stratified analysis showed that episode(s) of diarrhoeal diseases and birth in warm season were the only significant risk factors in the two poor areas. This study also found that the impact of diarrhoeal diseases on length growth persisted after allowing for birth season. Seasonal variation in length growth was seen in the age at onset of the childhood component. General socioeconomic status was found to be a risk factor for growth faltering in early life. Diarrhoeal diseases and being born in a warm season are particular risk factors for growth faltering in early life.

The need for a clinical trials research methodology training programme in Hong Kong

Training courses in the concepts of clinical trials research methodology that include rules in good clinical practice have not yet been extensively implemented in Hong Kong. This study aims to define the current knowledge of rules in good clinical practice and identify any need for such training programmes. Between May and August 1996, 161 clinical research staff were asked non-randomly to fill in a questionnaire about their knowledge of research methodology and their interest in specific courses. The median number of correctly answered questions (maximum score, 20) was 5 and the mode was 4, which was the expected score if questions had been answered randomly. Only minor differences in score were detected between doctors, research staff, and industry employees. Many researchers were keen, however, to further their knowledge by attending future courses; on average, each person showed an interest in three of the eight proposed courses. The study shows that the knowledge of rules in good clinical practice among clinical research staff in Hong Kong is poor, but there is nevertheless a demand for training programmes.