Gestational Age Specific Postnatal Growth Curves for Singleton Babies in A Tertiary Hospital of Western Nepal

Introduction: Measurement of birth weight, crown-heel length, head circumference and chest circumference used to assess the intrauterine growth of a baby vary with altitude, race, gender, socioeconomic status, maternal size, and maternal diseases. The study aimed to construct centile charts for birth weight, crown-heel length and head circumference for new born at different gestational ages in Western Nepal. Methods: This was a descriptive cross-sectional study done over a period of 15 months in a tertiary care hospital of Western Nepal. Birth weight, length, head circumference and chest circumference were measured within 12-24 hours of birth. Gestational age was estimated from the first day of last menstrual period and New Ballard’s scoring system. Microsoft 2007 Excel and SPSS-16 was used for data analysis. Cole’s Lambda Mu Sigma method was used for constructing centile curves. Results: Out of 2000 babies analysed, 1910 samples were used to construct smoothed intrauterine growth curve of birth weight, crown-heel length, and head circumference from 33-42 weeks of gestation. Among all, 1147 (57.35%) were male and 853 (42.65%) were female, mean gestational age was 38.13±2.44 weeks. The means of birth weight, crown-heel length, head and chest circumference were 2744.78 gm, 47.80 cm, 33.18 cm, and 30.20 cm with standard deviations of 528.29, 3.124, 1.78, and 2.35 respectively. Conclusions: This necessitates the update in the existing growth charts and development in different geographical regions of a country. _______________________________________________________________________________________


INTRODUCTION:
Birth weight (BW), crown-heel length (CHL), head circumference (HC) and chest circumference (CC) are used to assess the intrauterine growth of a baby. BW is one of the most important characteristics in evaluation of the well-being of a child at birth which is not only a reflection of the intrauterine development, but also is determined by the duration of pregnancy. 1 Since then many studies have been conducted showing that intrauterine growth curves vary with altitude, race, gender, socio-economic status, maternal size, and medical conditions during pregnancy. 5 Although charts from Kathmandu valley are available, 6,7 they do not represent the whole country population due to climate, altitude and population variances.

METHODS
This is a descriptive cross-sectional study done over a period of 15 months (1 st Jan 2014 to 1 st April 2015) in the labour ward, operation theatre, neonatal intensive care unit and post natal ward of Universal College of Medical Sciences, Bhairahawa, Nepal. The hospital is a tertiary care referral hospital. Ethical approval was obtained from the Institutional Research Committee. A written consent was taken from the parent(s).
All live born babies (both vaginal delivery and caesarean section) between 28-44 weeks were included. Their BW, CHL, HC and CC were measured within 12-24 hours of birth. Gestational age was estimated from first day of last menstrual period. In cases where LMP was unknown or in clinically discrepant cases, it was confirmed by clinical assessment using New Ballard's scoring system. Maternal age, weight, height and hemoglobin levels were also recorded. Babies less than 28 weeks, still births, twin pregnancies, with gestational diabetic mothers and with gross congenital anomalies were excluded.
Birth weight was measured within 12-24 hours of birth on the electronic weighing machine named Goldtech Digital Baby Scale, to the nearest ±5 gm. Crownheel length was recorded to the nearest fraction of cm using an infantometer. Head circumference (largest occipitofrontal diameter) and chest circumference (at the level of the nipples) were measured with a fiber glass tape in cm.
Microsoft 2007 Excel and SPSS 16 was used for data analysis. The mean, standard deviation, and 3rd, 10th, 25th, 50th, 75th, 90th, 97th centiles of each variable at each gestation were computed for all neonates. Cole's Lambda Mu Sigma (LMS) method 8 was used for constructing centile curves. This method estimates three age specific parameters: a Box-cox power transformation of skewness (L), median (M), and coefficient of variation (S) that correspond to the relationships in the following formulas: Z= {(x/M L -1)/ LS}, where X is the measured value of weight, length, or HC; and Centile=M (1+LSZ) 1/L , where Z is the z-score that corresponds to a given percentile. A smoothed percentile curve or an individualized score was obtained from the smoothed values of L, M, and S.

Length (cm)
Length for Gestational Age

Chest Circumference for Gestational Age
After the GA=40 weeks, there is decrease in birth weight which may be due to a small number of babies (Table 1). Smoothed curves of 3 rd , 10 th , 25 th , 50 th , 75 th , 90 th , 97 th centiles for weight, CHL, HC and CC were also prepared (Figures 2-5).
The birth weight from 33 to 42 weeks in 50 th centile ranged from 1954.50 gm to 3114.30 gm. The 50 th centile of CHL in the same gestation ranged from 43.30 cm to 49.80 cm. The HC and CC from 33 to 42 weeks in the 50 th centile ranged from 30-34 cm and 26-32 cm respectively.

DISCUSSION
The present study was aimed to develop growth charts for BW, CHL, HC and CC of babies born at different gestational ages in mixed group of population not only from Bhairahawa but also from the neighbouring areas. Further the number of preterm babies were almost three times more in Rupandehi as compared to Kathmandu. 6,7 In the present study, majority (46.2%) of the mothers were in the age group 21-25 years with mean age 24.8±4.56 years, more than half (62.7%) of the mothers were educated below lower secondary and only few (2.4%) had bachelor level degrees. Recent study done in South India depicted 41.7% of the women in the age group 25-29 years followed by 40.1% in the age group 20-24 years demonstrated the impact of maternal education "on growth of babies" (28.6% of the mothers were graduates and 13.6% had secondary education). This suggests that better education in the South Indian mothers may have been an important factor for their heavier babies in the study conducted by Kumar et al. 9 Another study conducted in rural Varanasi, India in 2002 showed 72.6% of normal BW neonates which is almost similar to our study. 10 One other previous Indian study in 1971 conducted by Ghosh et al. also showed smaller babies compared to our study. 11 Therefore, the cut off values would differ from one country to another country and different regions of any country and time to time for which the growth curves should be revised.
For Rupandehi's data, the centiles are obtained after doing the necessary corrections for skewness by LMS method as suggested by Cole et al. 8 Also in the Rupandehi data, sample size is small at the lower and higher gestational ages ( The main limitation of the present the study chart is the absence of the intrauterine growth percentiles from 28 to 32 weeks and after 42 weeks due to the insufficient number of newborns to calculate the percentile for these groups. This study has shown only the unisex intrauterine growth curve because of small sample size which might be different from the gender specific intrauterine growth curves.
The gestational age is calculated only from LMP and New Ballard Scoring System. This study has not used early ultrasound estimation which is found as one of the accurate methods of gestational age estimation. This study also lacks data regarding the maternal nutrition, maternal smoking and weight gain during pregnancy which influences the intrauterine growth pattern. This study has not excluded the maternal diseases like gestational hypertension, gestational diabetes mellitus, anemia which affect the intrauterine growth curve.
Another drawback for these types of charts is their cross-sectional nature of data collection. Larger studies comprising longitudinal data on intrauterine fetuses would be more accurate for assessing fetal growth velocity, though costly to collect.

CONCLUSIONS
The data of the present study represent intrauterine growth curves for weight, length, head and chest circumference for the Western Nepal (Rupandehi district). These weight data were similar to Kathmandu district's data at 50 th centile but differed at 10 th and 90 th centiles. For length at Kathmandu and Rupandehi's data were similar at 10 th and 50 th centiles while the former were significantly higher at the 90th centile. These differences suggest that for intrauterine growth assessment local data be used and updated time to time.
This data of the present study can be merged with the data of other countries to make a single intrauterine growth curve which represents the whole country.