To construct new Korean reference curves for birth weight by sex and gestational age using contemporary Korean birth weight data and to compare them with the Lubchenco and the 2010 United States (US) intrauterine growth curves.
Data of 2,336,727 newborns by the Korean Statistical Information Service (2008-2012) were used. Smoothed percentile curves were created by the Lambda Mu Sigma method using subsample of singleton. The new Korean reference curves were compared with the Lubchenco and the 2010 US intrauterine growth curves.
Reference of the 3rd, 10th, 25th, 50th, 75th, 90th, and 97th percentiles birth weight by gestational age were made using 2,249,804 (male, 1,159,070) singleton newborns with gestational age 23-43 weeks. Separate birth weight curves were constructed for male and female. The Korean reference curves are similar to the 2010 US intrauterine growth curves. However, the cutoff values for small for gestational age (<10th percentile) of the new Korean curves differed from those of the Lubchenco curves for each gestational age. The Lubchenco curves underestimated the percentage of infants who were born small for gestational age.
The new Korean reference curves for birth weight show a different pattern from the Lubchenco curves, which were made from white neonates more than 60 years ago. Further research on short-term and long-term health outcomes of small for gestational age babies based on the new Korean reference data is needed.
Small for gestational age (SGA) means a developing fetus in the uterus or an infant is smaller in size than normal babie's adjusted for sex and gestational age (GA). It is commonly defined as a weight below the 10th percentile for the GA
Some infants born SGA, particularly with IUGR, suffer from acute and chronic consequences. They can suffer from perinatal event such as hypoglycemia, gastro-esophageal reflux, and hypothermia in neonatal period
For a proper assessment and management of SGA, up-to-date ethnicity specific birth weight (or length) references by GA are needed
The objectives of this study were: (1) to make the sex and GA specific birth weight reference data of Korean population and to develop new birth weight curves; (2) to compare these Korean specific birth weight curves with old but commonly used curves (the Lubchenco curves) and 2010 US intrauterine growth curves; and (3) to estimate the prevalence of SGA using new curves and Lubchenco curves.
This study was performed using population data released by the Korean Statistical Information Service (KOSIS). The Korean National Statistic Office opened the KOSIS in July 2007 after establishing a statistical information database and building an integrated service system. It started under the national information strategy scheme in 2005. The KOSIS currently provides 343 types of statistics from domestic to international statistics produced by 87 organizations in Korea. In addition, KOSIS releases 100 major indicators of Korea, which provides information on current economic, social situations and health indicators, including birth and death rate data
For the new Korean reference for birth weight by GA, we used data of singleton newborns. We made birth weight reference of male and female separately as birth size is different by sex. The GA was determined by rounding off. Thus, GA 36 weeks represents 36 weeks plus 0 to 6 days. We excluded data with missing weight (n=5,728); unknown sex; GA under 22 weeks (n=1,244), and over 44 weeks (n=173). Multiple births (n=67,396) were also excluded as it is well known to have negative impact on intrauterine growth. We further excluded "extreme outliers". Extreme outliers were defined as values >2 times the interquartile range (25th to 75th percentiles) below the first quartile and above the third quartile for each GA
The percentile growth curves were created by using the Cole's Lambda Mu Sigma (LMS) method
By definition, 10% of a population should be <10th percentile (SGA), 80% between the 10th and 90th percentiles (appropriate for GA), and 10% >90th percentile (large for GA, LGA)
All data were analyzed using SAS 9.2 (SAS Institute Inc., Cary, NC, USA). The data are presented as the mean±SD, percentile values for birth weight according to GA by sex. The new Korean reference curves for birth weight were generated by using LMS methods. The LMS parameters for the curves are also presented.
All drawings are made using GraphPad Prism ver. 5 (GraphPad Software Inc., La Jolla, CA, USA). As GA 36 weeks represent 36 weeks +0 day to +6 days in our data, we make reference curve to shift right by 0.5 weeks for comparison with other data.
The characteristics of the study subjects are presented in
The mean and percentile birth weight of male newborns by GA are presented in
The new Korean reference curves for birth weight were plotted with the 2010 US intrauterine growth curves for comparison (
The new Korean reference curves for birth weight were also plotted with the Lubchenco curves (
In this study, we developed the new Korean reference curves for birth weight by GA and sex based on the data of contemporary Korean infants. This new Korean birth weight data set and growth curves might be useful not only in assessing short-term health risks of SGA infants in neonatal intensivecare units but also in assessing long-term health risks such as short stature in childhood and metabolic risk in adulthood among children born SGA or LGA. Furthermore, possible misclassification of SGA infants using the Lubchenco curves, which was based on the old data of the US neonates and has commonly been used in Korea, supports the need for updated and Korean specific reference curves for birth weight.
Contemporary ethnicity specific birth weight references are needed to identify SGA infants who might suffer from acute and chronic consequences
A controlled trial reported the results of using metformin in girls born SGA with premature pubarche
It is well known that there is racial and ethnic difference in birth weight
When compared with New Japanese neonatal anthropometric charts for GA
When compared with the Lubchenco's data, the 90th percentile birth weight cutoff of Korean curves was lower than that of Lubchenco's and the 10th birth weight cutoff of Korean curves was higher after 32 weeks than the Lubchenco curves. Lubchenco curves were made from White neonates in 1948 to 1963
In conclusion, we calculated the means and percentiles of birth weight in Korean neonates by GA on the basis of KOSIS data. We created new sex specific reference curves for birth weight by GA using the data of almost all contemporary Korean neonates. We also created new cutoffs of SGA and LGA. As some SGA and LGA infants had short-term and long-term health risks, these data and curves provide useful information not only for endocrinologists but also for other specialists, including neonatologists in Korea for planning research or targeting to prevent metabolic diseases, such as obesity, diabetes, and CVD.
No potential conflict of interest relevant to this article was reported.
The new Korean birth weight curves for male (A) and female (B) newborns.
The new Korean birth weight curves (black solid line) compared with the 2010 United States intrauterine growth curves (red dashed line) for male (A) and female (B) newborns/Lubchenco curves (blue dashed line) for male (C) and female (D) newborns. The line represents 10th, 50th, and 90th birth percentile, respectively.
The characteristics of study subjects
Variable | No. (%) | Birth weight (g) | |
---|---|---|---|
Total | 2,336,727 (100) | 3,218±461 | |
Sex | |||
Male | 1,199,897 (51.3) | 3,266±466 | |
Female | 1,131,003 (48.4) | 3,168±451 | <0.001 |
Unknown | 5,827 | ||
Multiplicity | |||
Singleton | 2,249,804 (96.3) | 3,244±434 | |
Twin | 66,064 (2.8) | 2,394±478 | |
Triplet+ | 1,071 | 1,751±507 | <0.001 |
Unknown | 1,669 |
Mean and percentile birth weight by gestational age in male Korean neonates (n=1,159,070)
GA |
No. | Birth weight (g) |
Percentiles |
|||||||
---|---|---|---|---|---|---|---|---|---|---|
Mean | SD | 3rd | 10th | 25th | 50th | 75th | 90th | 97th | ||
23 | 105 | 620 | 80 | 480 | 510 | 570 | 630 | 680 | 720 | 760 |
24 | 197 | 730 | 100 | 520 | 610 | 680 | 740 | 800 | 860 | 920 |
25 | 298 | 820 | 130 | 520 | 640 | 750 | 830 | 900 | 970 | 1,040 |
26 | 416 | 940 | 140 | 600 | 750 | 860 | 950 | 1,030 | 1,100 | 1,190 |
27 | 548 | 1,060 | 170 | 690 | 830 | 970 | 1,070 | 1,180 | 1,270 | 1,370 |
28 | 745 | 1,180 | 210 | 710 | 900 | 1,060 | 1,210 | 1,320 | 1,420 | 1,530 |
29 | 891 | 1,330 | 250 | 760 | 1,000 | 1,200 | 1,360 | 1,490 | 1,610 | 1,750 |
30 | 1,130 | 1,510 | 280 | 940 | 1,130 | 1,350 | 1,530 | 1,680 | 1,810 | 2,030 |
31 | 1,480 | 1,690 | 290 | 1,080 | 1,280 | 1,530 | 1,715 | 1,870 | 2,030 | 2,210 |
32 | 2,175 | 1,900 | 320 | 1,240 | 1,490 | 1,730 | 1,920 | 2,100 | 2,270 | 2,520 |
33 | 3,205 | 2,120 | 350 | 1,380 | 1,680 | 1,930 | 2,130 | 2,340 | 2,530 | 2,820 |
34 | 6,155 | 2,350 | 360 | 1,600 | 1,910 | 2,140 | 2,360 | 2,570 | 2,800 | 3,100 |
35 | 11,175 | 2,600 | 360 | 1,870 | 2,150 | 2,380 | 2,600 | 2,820 | 3,060 | 3,330 |
36 | 27,092 | 2,830 | 370 | 2,100 | 2,380 | 2,600 | 2,820 | 3,060 | 3,300 | 3,600 |
37 | 96,514 | 3,080 | 370 | 2,380 | 2,620 | 2,840 | 3,080 | 3,320 | 3,560 | 3,820 |
38 | 295,872 | 3,250 | 360 | 2,590 | 2,800 | 3,000 | 3,230 | 3,480 | 3,700 | 3,960 |
39 | 341,418 | 3,340 | 360 | 2,700 | 2,900 | 3,100 | 3,320 | 3,580 | 3,800 | 4,040 |
40 | 294,588 | 3,440 | 360 | 2,800 | 3,000 | 3,200 | 3,420 | 3,680 | 3,900 | 4,140 |
41 | 71,644 | 3,500 | 360 | 2,850 | 3,040 | 3,250 | 3,500 | 3,730 | 3,960 | 4,200 |
42 | 3,212 | 3,510 | 380 | 2,840 | 3,010 | 3,250 | 3,500 | 3,750 | 4,000 | 4,270 |
43 | 210 | 3,400 | 430 | 2,610 | 2,900 | 3,100 | 3,395 | 3,720 | 3,900 | 4,320 |
GA, gestational age; SD, standard deviation.
GA 36 weeks represent 36 weeks+0 day to+6 days.
Mean and percentile birth weight by gestational age in female Korean neonates (n=1,090,734)
GA |
No. | Birth weight (g) |
Percentiles |
|||||||
---|---|---|---|---|---|---|---|---|---|---|
Mean | SD | 3rd | 10th | 25th | 50th | 75th | 90th | 97th | ||
23 | 74 | 580 | 80 | 420 | 490 | 540 | 590 | 640 | 670 | 720 |
24 | 191 | 680 | 90 | 470 | 570 | 620 | 690 | 740 | 800 | 850 |
25 | 311 | 770 | 120 | 490 | 620 | 700 | 780 | 740 | 920 | 990 |
26 | 397 | 880 | 150 | 510 | 660 | 810 | 900 | 980 | 1030 | 1110 |
27 | 512 | 970 | 180 | 570 | 720 | 865 | 1000 | 1100 | 1180 | 1260 |
28 | 682 | 1,120 | 200 | 720 | 840 | 1000 | 1140 | 1250 | 1360 | 1480 |
29 | 723 | 1,260 | 240 | 820 | 920 | 1,100 | 1,280 | 1,410 | 1,540 | 1,690 |
30 | 946 | 1,410 | 290 | 840 | 1,040 | 1,240 | 1,420 | 1,590 | 1,720 | 1,960 |
31 | 1,176 | 1,590 | 300 | 970 | 1,160 | 1,410 | 1,630 | 1,790 | 1,910 | 2,070 |
32 | 1,693 | 1,810 | 370 | 1,090 | 1,330 | 1,590 | 1,810 | 2,020 | 2,230 | 2,650 |
33 | 2,374 | 2,020 | 350 | 1,320 | 1,580 | 1,810 | 2,040 | 2,230 | 2,440 | 2,700 |
34 | 4,622 | 2,270 | 390 | 1,500 | 1,780 | 2,040 | 2,260 | 2,500 | 2,750 | 3,060 |
35 | 8,367 | 2,500 | 380 | 1,760 | 2,040 | 2,280 | 2,500 | 2,730 | 2,970 | 3,260 |
36 | 20,926 | 2,740 | 380 | 2,010 | 2,280 | 2,500 | 2,740 | 2,980 | 3,220 | 3,520 |
37 | 77,453 | 2,980 | 370 | 2,290 | 2,520 | 2,740 | 2,960 | 3,200 | 3,460 | 3,720 |
38 | 255,375 | 3,130 | 350 | 2,500 | 2,700 | 2,900 | 3,110 | 3,350 | 3,590 | 3,830 |
39 | 318,377 | 3,220 | 350 | 2,600 | 2,800 | 3,000 | 3,200 | 3,450 | 3,680 | 3,910 |
40 | 309,996 | 3,320 | 350 | 2,700 | 2,900 | 3,080 | 3,300 | 3,550 | 3,780 | 4,010 |
41 | 8,296 | 3,380 | 350 | 2,760 | 2,950 | 3,140 | 3,380 | 3,600 | 3,840 | 4,060 |
42 | 3,611 | 3,390 | 370 | 2,720 | 2,920 | 3,140 | 3,380 | 3,620 | 3,860 | 4,100 |
43 | 232 | 3,350 | 350 | 2,700 | 2,850 | 3,105 | 3,340 | 3,600 | 3,800 | 3,950 |
GA, gestational age; SD, standard deviation.
GA 36 weeks represent 36 weeks+0 day to+6 days.
The new Korean birth weight curves L, M, and S parameters
GA |
Male growth curves |
Female growth curves |
||||
---|---|---|---|---|---|---|
L | M | S | L | M | S | |
23 | –0.30 | 630 | 12.86 | –0.41 | 590 | 13.22 |
24 | –0.26 | 740 | 13.26 | –0.06 | 690 | 13.90 |
25 | –0.45 | 830 | 15.77 | –0.34 | 780 | 15.74 |
26 | –0.52 | 950 | 15.09 | –0.72 | 900 | 16.83 |
27 | –0.38 | 1,070 | 16.17 | –0.59 | 1,000 | 18.48 |
28 | –0.44 | 1,210 | 18.03 | –0.17 | 1,140 | 18.22 |
29 | –0.46 | 1,360 | 18.82 | –0.06 | 1,280 | 18.84 |
30 | –0.13 | 1,530 | 18.51 | 0.12 | 1,420 | 20.38 |
31 | –0.23 | 1,720 | 17.09 | –0.27 | 1,630 | 18.74 |
32 | –0.14 | 1,920 | 16.70 | 0.24 | 1,810 | 20.60 |
33 | –0.13 | 2,130 | 16.43 | –0.06 | 2,040 | 17.08 |
34 | –0.01 | 2,360 | 15.40 | 0.07 | 2,260 | 16.99 |
35 | 0.04 | 2,600 | 14.04 | 0.03 | 2,500 | 14.99 |
36 | 0.07 | 2,820 | 13.23 | 0.12 | 2,740 | 13.99 |
37 | 0.08 | 3,080 | 12.11 | 0.15 | 2,960 | 12.45 |
38 | 0.13 | 3,230 | 11.12 | 0.16 | 3,110 | 11.18 |
39 | 0.15 | 3,320 | 10.68 | 0.17 | 3,200 | 10.81 |
40 | 0.14 | 3,420 | 10.47 | 0.17 | 3,300 | 10.56 |
41 | 0.16 | 3,500 | 10.27 | 0.16 | 3,380 | 10.23 |
42 | 0.22 | 3,500 | 10.79 | 0.13 | 3,380 | 10.83 |
43 | 0.05 | 3,400 | 12.73 | –0.14 | 3,340 | 10.52 |
GA, gestational age; LMS, Lambda Mu Sigma.
GA 36 weeks represent 36 weeks+0 day to+6 days.