philpan1c

PhilSPEN Online Journal of Parenteral and Enteral Nutrition

(Article 37 | POJ_0031.html) Issue January 2017 - June 2017: )

Original Clinical Investigation

An analysis of the use of the 1960 Lubchenco growth curves on Filipino newborn infants in a tertiary care hospital in Quezon City, Metro-Manila, Philippines

Abstract | Introduction | Methodology | Results | Discussion | Conclusion | References | PDF (138 KB) |Back to Articles Page

Submitted: May 3, 2017| Posted: June 1, 2017

AUTHORS:

Melissa T. Sy MD and Luisito Llido MD

INSTITUTIONS WHERE RESEARCH WAS CONDUCTED:

Clinical Nutrition Service, St. Luke’s Medical Center, E. Rodriguez Avenue, Quezon City, Philippines, 1102

ABSTRACT | Back

Background and Aims: The 1960 Lubchenco growth curves have been used to evaluate newborns using their birth weight to classify them as small, appropriate or large infants in our institution thus the need to validate it. The goal of this study is to analyze and compare the birthweight of infants using the Lubchenco chart and determine if there is close agreement in values or growth patterns thus allowing us to confidently use the tool for reference until such a time when growth curves based on the local population are available.

Methodology: A cross-sectional study involving 2,387 newborns in the hospital nursery from year 2012 to 2014 was done. All encoded data were processed using the NCSS v10 statistics software to generate the percentile values and create a graph to be plotted against and compared to the Lubchenco chart.

Results: 1) the 50th percentile curve was in agreement with the Lubchenco curve from the 31st week to the 41st week but it falls at 31st week up to the 22nd week. (2) the 90th percentile values were lower compared to the Lubchenco values but these agree with the 75th percentile of the Lubchenco curve; this agreement stops at the 20th week and below. (3) the 10th percentile values were in agreement with the Lubchenco curves only from the 37th week to the 41stweek then it falls from the 36th week to the 22nd week.

Conclusion: The Lubchenco curves can be used in our Filipino newborn population from the 31st week to term. However, it cannot give a good assessment for the premature occupying the 10th percentile of the weight category.

KEYWORDS: Lubchenco, birthweight, newborn, growth curves, Filipino infants

INTRODUCTION | Back

In the evaluation of newborn and infants It is routine to obtain their anthropometric data especially their birth weight to classify them as small, appropriate or large infants. There is a need, therefore, for newborn or intrauterine reference growth curves based on the local population to achieve this goal. There is, however, no reference growth curves developed in the Philippines thus the pediatric department in this institution (St. Luke’s Medical Center) adopted the 1960 Lubchenco infant growth curves for this purpose. (Reference 1) This is one of the older but still widely used reference growth charts in NICU settings outside and inside the United States (2) and have also been most commonly used Asia as in Korea (3). The 1960 Lubchenco curves is used in the neonatal care unit and delivery room of SLMC to determine if the weight of infant is small, average or large for gestational age. it is also used as reference to monitor the premature newborn’s growth.

There are issues raised on this practice however. The absence of an international standard is one of the limitation to estimate prevalence and mortality of small-for-gestational-age infants because references differs from each countries at a particular time points.  (4) An international study showed that the 1960 Lubchenco growth curves underestimated the percentage of infants who were small-for-gestational age except for younger girls (</= 36 weeks) whom it was overestimated.  It also underestimated percentage of infants (</= 36weeks) who are large for gestational age and overestimated percentage of infants (>36 weeks) who were large-for-gestational age. (5) In the Philippines no studies have been done to verify if the Filipino newborn values fit the standards determined by the 1960 Lubchenco chart. The data of this chart were based on US data and not on Filipino data, thus there is still a need for reference charts based on local data to assess the intrauterine growth and birth weight of Filipino infants. Today new intrauterine growth curves based on ultrasound data are developed for the US population because some studies reported that the Lubchenco weight curves may not accurately reflect the current US population.

There is therefore a need in this institution to validate the 1960 Lubchenco chart using our own Filipino data in order to know if objective, accurate and better assessment of the Filipino infant can be done using this tool. The goal of this study is to analyze and compare the birth weight of infants born in St. Luke’s Medical Center-Quezon City using the 1960 Lubchenco Chart and determine if there is close agreement in values or growth patterns. Once good agreement is established then the confidence of the pediatric department in using the Lubchenco growth chart will improve knowing that it has a good tool for infant growth assessment until such a time when intrauterine ultrasound based Filipino data are available. The result of the study will be of help in the process and protocol development for assessing growth of all newborns.

METHODOLOGY | Back

This is a descriptive study involving review of all medical records of newborns in St. Luke’s Medical Center – Quezon City from the years 2012 to 2014. In the delivery room, all newborns were weighed without clothing using the calibrated newborn scales in the delivery room at less than 12 hours of birth to determine the birthweight then recorded in the newborn chart. The weight recorded was in grams. The gestational age of the newborns was based on the Ballard Score done by the pediatric resident on duty at the time the baby was born. Age of gestation is estimated from the first day of last menstrual period or calculated based on early prenatal ultrasound then rounded off to the nearest whole number. To obtain a more accurate gestational age, Ballard scoring can be used. There were studies that have shown that the Ballard score is most consistent with the prenatal ultrasound and last menstrual period in estimating the gestational age (6). These are the terms used to delineate the different age groupings based on weight.

tbl01_lubc

These are included in the study: all newborns, post term, term and premature, born in the delivery room of St. Luke’s Medical Center- Quezon City in the year 2012-2014. These were excluded: Infants with major congenital anomalies or chromosomal anomaly, still birth and Infants born outside the delivery room of St. Luke’s Medical Center-Quezon City.
Data of birth weight and age of gestation were gathered from the medical chart review of all newborns of the year 2012-2014. The data was encoded in an Excel file and was processed using the NCSS v10 statistics software to generate the percentile values in the different ages of gestation in weeks (8). The newborn reference weight percentile (both sexes) curves (graph) was then created and plotted against and compared to the Lubchenco growth curves (Figure 1)

fig01_lubch

RESULTS | Back

A total of 2,387 newborns were included in the study. 60% were delivered normally while 38.5% were by Cesarean section. There was equal distribution between males and females. Table 2 shows the profile of the population while Table 3 shows the values that comprise the different mean weights in grams in each age of gestation.

tbl02_lubch

tbl03_lubch

The ages covered were from 22 weeks to 41 weeks and table 3 shows the profile of the newborns. The data however was mixed in order to fit the data of the Lubchenco graph which was based on a mixed male and female population. Figure 2 shows the final newborn percentile curves of the St. Luke’s Medical Center Nursery and Figure 3 shows the comparison of the 1960 Lubchenco graph with the 2013-2014 St. Luke’s Nursery graph.

fig02_lubch

fig03_lubch


The compared graphs in Figure 3 showed the following characteristicss:

  1. The 50th percentile values of the St. Luke’s newborns were in agreement with the Lubchenco curves from the 31st week to the 41st week. But it falls below the 50th percentile values at 31st week up to the 22nd week
  2. The 90th percentile values of the St. Luke’s newborns were lower compared to the 1960 Lubchenco values; these are in agreement with the 75th percentile of the Lubchenco curve however, this agreement stops at the 20th week and below.
  3. The 10th percentile values of the St. Luke’s newborns were in agreement with the Lubchenco curves only from the 37th week to the 41stweek; the curves then falls below the 10th percentile value from the 36th week to the 22nd week.

Observations:

  1. The Lubchenco growth curves can be used as reference values fairly well for mature births and for premature births from 31st week to 36th weeks of gestation.
  2. It cannot give a good assessment for the premature occupying the 10th percentile of the weight category however.
  3. The 90th percentile of the St. Luke’s newborns or the “big babies” is only at the 75th percentile of the Lubchenco indicating that the “big US babies” were larger than the “big Filipino babies”.

DISCUSSION: | Back

This analysis of the 1960 Lubchenco growth curves shows that it can be used in our Filipino newborn population from the 31st week to term. Premature newborns from 31st week to 38th week are also “represented well”, meaning the value and classification of these babies using the Lubchenco growth curve can be relied on. Figure 4 shows the suggested cut-offs in the Lubchenco growth curve.

fig04_lubch

The findings for the large gestational age newborns show that Filipinos babies at the 90th percentile fit the Lubchenco 75th percentile indicating that American babies may be larger in this category. It may reflect the difference in racial body composition, Asians compared to Caucasians, which was also reflected in a similar study done in Korea which showed that their 90th percentile values were lower than Lubchenco’s 90th percentile values for those less than 30 weeks. The 90th percentile curve was drawn near Lubchenco’s 75th percentile area (3).

For the small for gestational age newborns (10th percentile), these are only represented well when they are term. The 10th percentile St. Luke’s newborn curve tend to be lower compared to the 10th percentile Lubchenco graph at 37th week and younger. The opposite was observed with the Koreans - their 10thpercentile values were higher than the Lubchenco’s 10th percentile values for 35th week and younger. (Figure 5)

fig05_lubch

CONCLUSION: | Back

The Lubchenco curves can be used in our Filipino newborn population from the 31st week to term. However, it cannot give a good assessment for the premature occupying the 10th percentile of the weight category.

REFERENCES: | Back

  1. Lubchenco LO, Hansman C, Dressler M, Boyd E. Intrauterine growth as estimated from liveborn birth-weight data at 24 to 42 weeks of gestation. Pediatrics 1963;32: 793–800.
  2. Olsen IE et al. New Intrauterine Growth Curves Based on United States Data. Pediatrics 2010;125;e214.
  3. Aum JA, Jung HJ, Huh JW, Son SH. Analysis of anthropometric data for premature infants of 26 to 35 weeks of gestation comparison with the data of 1960’s. Korean J Pediatr 2007; 50(6): 543-548.
  4. Villar et al. International standards for newborn weight, length, and head circumference by gestational age and sex: the Newborn Cross-Sectional Study of the INTERGROWTH-21st Project. Lancet 2014: 384: 857-68.
  5. de Onis M, Onyango A, Borghi E, et al. Worldwide implementation of the WHO Child Growth Standards. Public Health Nutr 2012; 15: 1603–10.
  6. Ballard JL, Khoury JC, Wedig K et al. New Ballard Score, expanded to include extremely premature infants. J Pediatr 1191; 119:417.
  7. Appendix F, Standard terminology for reporting of Reproductive Health Statistics in the United States. AAP Guidelines for Perinatal Care, 7th Ed. 2012
  8. NCSS statistical software. Available at https://www.ncss.com

 

Abstract | Introduction | Methodology | Results | Discussion | References | Back to Articles Page