Hospital Based Gestational Age Specific Birth Charts in Urban Bangladesh

Birth weight is not enough to evaluate growth of the newborn. Around the globe, growth charts at birth as percentiles for various gestational ages are available both in developed and developing countries. This study aimed to develop such growth chart in Bangladesh, to exactly evaluate an individual newborn’s growth pattern by comparing growth of its different parts with that of a standard one. A cross sectional observational study conducted at fifteen non-government maternity centers located in different parts of Dhaka, Narayanganj and Narsingdi. Birth weight, birth length, head and chest circumferences of five thousand one hundred and five singleton newborns’ with gestational ages ranging from 35 to 42 weeks were recorded during June 2014 to July 2015. Trained Physicians and paramedics of respective centers recorded data. Gestational age specific percentile charts were constructed for birth weight, birth length, head circumference and chest circumference. Mean birth weight, birth length, head and chest circumferences were found to be respectively 2966g, 47.1, 33.2 and 32.3 cm which were more or less comparable to some other studies in the country and also higher than some of the studies. Most strikingly, prevalence of low birth weight was found to be 11 percent which is significantly lower than any other study conducted in the country. The 10, 50 and 90 percentile values of birth weight at 40 weeks of gestation were respectively found to be 2600g, 3000g and 3600g. The linear and circumference growth at 40 weeks of gestation match with the percentile values of birth weight. Constructed growth charts are more or less comparable to some South Asian countries but also significantly differ from some other studies. Country specific standard growth chart is required for clinical assessment and identifying high risk newborns for providing special care. The growth charts constructed in this study, will guide policy makers and programme personnel to take necessary interventions for ensuring potential growth of every future newborn. Furthermore, this study has made the opportunity to compare the charts with other South Asian countries and developed world.


Introduction
In Bangladesh, like most other countries, anthropometric studies concerning newborns were primarily focused on birth weight particularly on low birth weight (LBW <2.5kg).Only individual's birth weight is not sufficient to evaluate the growth of newborns.For example, many of the growth-retarded babies have birth weights either equal to or greater than 2.5 kg. 1 Again, though preterm babies (gestational age <37 completed weeks) usually weigh < 2.5 kg, their growth pattern is different from that of intrauterine growth retarded LBW babies.Therefore, taking only weight of the babies, whether they have grown normally or not in the uterus, could not always be able to assess appropriately.
For proper interpretation of growth, Lubchenco first described anthropo-metry at birth as percentiles for various gestations. 2 Thereafter, to interpret the growth of newborns at birth developed countries as well as many developing countries have constructed growth charts at birth as percentiles for various gestations.Usually, growth charts are constructed for birth weight, birth length and head circumference.To recognize the individual baby's risk, a useful supplement to other observations during the early hours after birth is the recording of the baby's weight, length, and head circumference on the intrauterine growth charts.However, in Bangladesh, so far no such charts have been developed.Therefore, due to absence of such chart, no one can evaluate exactly an individual newborn's growth pattern by comparing growth of its different parts (birth weight, length, head circumference) with that of a standard chart.The position of the infant's measurements on the charts may indicate that s/he is within the usual borders of growth for his/her gestational age, s/he may be near or outside of the extremes of normal growth, or there may be inconsistencies between the percentile positions of weight, length, and head circumference.
For liveborn babies, both short and long term consequences of abnormally grown foetus are critical.4] Those who survive adulthood are more likely to suffer from chronic diseases like hypertension, coronary heart disease, diabetes mellitus and obstructive lung disease. 5To prevent these adverse effects, growth charts can be used at primary health care level to screening the vulnerable babies for proper attention by the appropriate health care providers.The incidence of growth retarded infants is high in the country. 6evelopment of growth charts for different parameters and their use by health care providers obviously will play vital role in identifying the growth retarded newborns.Furthermore, as the study was conducted with reasonably large sample size, with certain limitations, the constructed growth charts can be used as reference or even test charts in the country.Once it is considered the magnitude of the problem of growth retardation, the public health professionals and clinicians will feel encourage to use the growth charts for screening and necessary intervention for newborns at risk.Essential instruction can also be provided to mothers and other care givers for taking proper care to those babies at home.
The objectives of the study were to construct gestational age specific centile charts for weight, length, weight length ratio, head circumference and chest circumference at birth and also to assess the growth of newborn and to compare the constructed charts with similar charts in other countries.

Materials and Methods
This was a cross-sectional observational study carried out at fifteen non-government maternity centers located in Dhaka North and South City Corporations, Narayanganj, Narsingdi and Keraniganj.Birth weight, birth length, head and chest circumferences of 5,105 singleton newborns with gestational age ranging from 35 to 42 weeks were recorded over a period of 14 months from June 2014 to July 2015.
Trained physicians and paramedics of respective maternity centers recorded the data in the questionnaire which was pretested and modified earlier.The centers included 8 maternity clinics of Urban Primary Health Care Project (UPHCP-2) in Dhaka; five clinics of Marie-Stopes Society (Two centers at Dhaka and three at Narayanganj, Narsingdi and Keraniganj); Ad-din Hospital, Dhaka; and maternity center of Reproductive Health Services Training and Education Program (RHSTEP) Dhaka.These are secondary level maternity clinics in the country, where pregnant women mostly from middle and lower class attend these centers for antenatal care, delivery and postnatal care.The gestational age was calculated in completed weeks either from the date of last menstrual period or from the early dating ultrasound report.Cases with uncertain gestational age were excluded from the study.The weights of newborns were measured on a digital weighing machine with a variability of 50g.The weighing machines were checked frequently with known standard weight and calibrated during each session.The head and chest circumferences were measured with a fibre-glass measuring (nonstretchable) tape to the nearest of 0.1 cm.Maximum occipito-frontal circumference of head was recorded and chest circumference was measured at the level of nipple at the end phase of expiration.The length of the infant was measured in centimeter on an infantometer with the baby supine, knees fully extended and soles of the feet held firmly against the foot board and head touching the fixed board.These centres usually do not deal with any form of risk pregnancy.Thus, it was obtained anthropometric data of newborns of normal pregnancy.A very few number of cases was obtained before 35 weeks of gestation and therefore was not included in the study.Post mature babies (after 42 weeks of gestation) were also excluded.
SPSS software (version 20) was used for data analysis.Percentile tables for weight, length, weight-length ratio, head and chest circumferences and growth curves are presented and comparison is made with other related studies.Percentiles charts were constructed using the LMS Chart Marker Software. 7LMS chart maker is a program to fit smooth centile curves to reference data using the LMS method.The LMS method summarizes the changing distribution by three curves representing the skewness (L), median (M) and coefficient of variation (S).Using penalized likelihood the three curves can be fitted as cubic splines by non-linear regression, and the extent of smoothing required can be expressed in terms of smoothing parameters or equivalent degrees of freedom.Percentiles drawn for this study were 10 th , 25 th , 50 th, 75 th and 90 th to construct the charts.

Results
Five thousand one hundred and five singleton live birth newborn babies' birth weight, length, head and chest circumference were recorded from fifteen non-government maternity centres.Gestational age of these newborns ranges from 35 to 42 weeks.Centre specific mean anthropometric values did not vary significantly.Mean values for birth weight, length, head and chest circumference were found to be respectively 2966g, 47.1, 33.2 and 32.3 cm.Anthropometric measurements between male and female newborns were also not found to vary significantly (table I).
The number of newborns with birth weight less than 2500g was found to be only 11 percent.Percentage of low birth weight (<2500g), inadequate birth weight (2500 -<3000g) and adequate birth weight (≥3000g) for male newborns were found to be 10.7, 40.5 and 48.8 percent respectively and that for female newborns were 11.2, 43.9 and 44.8 percent respectively, while for both sexes combined values were found to be 11.0, 42.1 and 46.9 percent respectively (table II).Only 4 percent newborns were born prematurely (before 37 weeks of pregnancy).
Percentile distribution of birth weight, length, weight-length ratio, head and chest circumferences (for both sexes combined) are presented in table III and birth charts are constructed for birth weight, length, head and chest circumferences (figure 1-4).
The 10 th , 50 th and 90 th percentile values of birth weight at 40 th week of gestation were found to be 2600g, 3000g and 3600g.The linear and circumference growth at 40 th weeks of gestation match with the percentile values of birth weight.0][11][12][13] The prevalence of low birth weight in the present study was found to be 11.0 percent which is significantly below in comparison to other studies conducted in the country.In National Low Birth Weight Survey conducted twelve years back found proportion of low birth weight as 36 percent.Whereas the lead author of the present study, in his two previous hospital based studies found prevalence of low birth weight as 15.0 percent and 17.3 percent. 6,13roper antenatal care from the early period of pregnancy of mothers of the newborns, increase of health consciousness of family members and also gradual progression of socioeconomic status of general people might be the reasons for low incidence of low birth weight of the present study.
Though skeletal growth (birth length and head circumference) in intrauterine life is an important determinant for future morbidity and mortality, such records are not documented in most of the studies in the country.National Low Birth Weight Survey recorded birth length and the mean value was found to be 48.5cmwhich does not coincide with the mean birth weight (2632g). 8o other studies in the country documented birth length, head circumference and chest circumference and the respective values were found to be 47.9cm, 33.5cm and 31.7cm for one study, and 48cm, 33cm and 31.6cm for the other study. 6,13an anthropometric values of all the variables of the present study do not differ significantly between male and female babies.Therefore, only combine (male and female together) curves have been constructed.As such growth charts with percentile distribution are not available in the country, comparison cannot be shown in country context.However, such charts are available in developed countries as well as in some developing countries including India.The charts can be considered as normative, because women with any known morbidity (e.g.pre-eclampsia/ eclampsia, diabetes mellitus, hypertension) that might interfere with growth were excluded from the study.
In this study, birth weight data at gestational age of 40 th week the 10 th , 50 th and 90 th percentile values were found to be 2600g, 3000g and 3600g respectively which are lower in comparison to Lubchenco's study. 2 In a Sri Lankan study corresponding values at 10 th , 50 th and 90 th were found to be 2580g, 3037g and 3600 g for male babies and 2800g, 3180g and 3600g for female babies. 14The results of a Nepalese study are comparable to this study findings. 15In one South Indian study corresponding values were found to be lower. 16e 50 th and 90 th percentile values for length of the present study at 40 th week of gestation were also found to be lower than Lubchencho's another study, 17 but 10 th percentile value was found to be same ( 45.8cm).Average length at 40 th weeks was found to be 47.5cm in our study and 49.4 in Lubchenco's study. 17owever, head circumference values at 40 th weeks of gestation were found to be more or less comparable.
The weight-length ratio (ponderal index) reveals how heavy the baby is for its length.Though both birth weight and length were found to be less than that of Lubchenco's studies, deficit of linear growth is more pronounced in this study and therefore weight-length ratio values were found to be more in this study. 2,179][20][21] However, there are substantial differences in the various published standards that are defined by the 10 th percentile alone.For example, in Lubchenco's study, 10 th percentile values for birth weight at 40 th weeks of gestation was found to be 2630g. 2whereas in Blinder's study the corresponding value was found to be 2895g. 22New Dutch reference curves for birth weight has been constructed for male and female infants born from primiparous and multiparous women and a separate reference curve for Hindustani babies. 23Percentile values for Dutch male and female infants at all gestational ages were found to be significantly higher than that of percentile values of this study.Even reference curves for Hindustani babies were found to be higher (2740g, 3238g and 3817g respectively for 10 th , 50 th and 90 th percentils at 40 th week of gestation) than the values in this study.In one Southern Indian study 10 th percentile values at 40 th week of gestation for birth weight for males and females were found to be respectively 2488g and 2412g. 16In another Southern Indian study at same gestational age, 10 th percentile values for birth weight for males and females were found to be respectively 2850g and 27660g, and 50th and 90 th percentile values were found to be 3258g and 3764g for male babies and that for females babies were 3155g and 3652g respectively.In the same study, mean length at 40 th week of gestation were found to be respectively 50.3cm and 49.8 cm for male and female babies and mean head circumference values at 40 th week was found to be 34.3cm and 33.9cm respectively. 24Accordingly, it is clearly revealed that anthropometric values of newborns at birth differ from study to study and also from population to population.Therefore, country specific and even area specific birth charts are necessary.
In spite of the best effort in calculating the percentiles using considerably large cohort, the study has yet certain limitations.There were fifteen centers from where data were collected and inter-center variability in anthropometric measurements is not unlikely.Even in the same center anthropometric measurements were under taken by different doctors or paramedics which might have contributed to inter-observer variability.Moreover, many women might not accurately remember the date of last menstrual period.Furthermore, data were collected from all socioeconomic groups and hence might not reflect the ideal potential for intrauterine growth.However, it is needless to say that, this is an important step in developing growth charts at birth for the first time in the country.

Conclusion
Growth chart at birth for weight, length and head circumference is prerequisite to help in clinical assessment and identifying high risk newborns for special care.The constructed growth charts will also guide policy makers and programme managers to undertake necessary strategies and interventions for ensuring potential growth of every future newborn.The study also creates an opportunity to comparing these charts with that of other South Asian countries as well as developed world.Growth charts should be periodically reviewed to examine the temporal influence of socioeconomic and socio demographic changes on growth of newborns.It should be updated at least in every five year interval.Furthermore, construction of population based growth chart for nationwide use in the country is also recommended.
Acknowledgments: Authors would like to acknowledge deeply the mothers of newborns who allowed us to undertake anthropometric measurements of their newborn babies.Authority of fifteen maternity centres deserves appreciation for letting us in conducting the study.Doctors and paramedics of the respective maternity centres also deserve special thanks for their inexorable effort in collecting data.Finally, we are profoundly grateful to BMRC for necessary funding to completing the study.

Table I :
Mean anthropometric parameters by each sex separately and combined

Table II :
Birth weight status of male and female newborns separately and of both sexes combined

Table III :
Intrauterine growth standard by weight, length, weightlength ratio, head circumference and chest circumference