Scholarly article on topic 'Reappraisal of Fetal Abdominal Circumference in an Asian Population: Analysis of 50,131 Records'

Reappraisal of Fetal Abdominal Circumference in an Asian Population: Analysis of 50,131 Records Academic research paper on "Clinical medicine"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{"abdominal circumference" / "Asian population" / fetus / pregnancy}

Abstract of research paper on Clinical medicine, author of scientific article — Szu-Ching Lu, Chiung-Hsin Chang, Chen-Hsiang Yu, Lin Kang, Pei-Ying Tsai, et al.

Summary Objective Fetuses from different populations may show different growth patterns. In obstetrics, fetal abdominal circumference (AC) is a very useful index for assessing fetal growth. In this study, we attempted to establish the normal fetal growth curves of AC in an Asian population in South Taiwan. Materials and Methods We reviewed our computer ultrasound database of fetal AC records from January 1991 to December 2006. During the study period of 1 6 years, only the fetuses examined by ultrasonography with gestational age between 1 4 and 41 weeks were included. We excluded extreme bilateral records after initial analysis. Eventually, 50,131 records of AC were included for final analysis. Results The observed gestation-specific AC values and the predicted AC values were calculated. The best-fit regression equation of AC versus gestational age is a second-order polynomial equation. In general, fetal AC values in our population showed similar patterns to those in Western populations. Besides, we established a table of the predicted AC values based on specific gestational age, including the 5th, 1 0th, 50th, 90th and 95th centiles, for clinical reference. Conclusion To the best of our knowledge, our series is the largest sample of AC reported in the medical literature. We believe that the gestational age-specific nomogram of fetal AC is important for further clinical assessment of fetal growth.

Academic research paper on topic "Reappraisal of Fetal Abdominal Circumference in an Asian Population: Analysis of 50,131 Records"

-■ ORIGINAL ARTICLE ■-

Reappraisal of Fetal Abdominal Circumference in an Asian Population: Analysis of 50,131 Records

Szu-Ching Lu, Chiung-Hsin Chang*, Chen-Hsiang Yu, Lin Kang, Pei-Ying Tsai, Fong-Ming Chang Department of Obstetrics and Gynecology, National Cheng Kung University Medical College and Hospital, Tainan, Taiwan.

SUMMARY

Objective: Fetuses from different populations may show different growth patterns. In obstetrics, fetal abdominal circumference (AC) is a very useful index for assessing fetal growth. In this study, we attempted to establish the normal fetal growth curves of AC in an Asian population in South Taiwan.

Materials and Methods: We reviewed our computer ultrasound database of fetal AC records from January 1991 to December 2006. During the study period of 16 years, only the fetuses examined by ultrasonography with gestational age between 14 and 41 weeks were included. We excluded extreme bilateral records after initial analysis. Eventually, 50,131 records of AC were included for final analysis.

Results: The observed gestation-specific AC values and the predicted AC values were calculated. The best-fit regression equation of AC versus gestational age is a second-order polynomial equation. In general, fetal AC values in our population showed similar patterns to those in Western populations. Besides, we established a table of the predicted AC values based on specific gestational age, including the 5th, 10th, 50th, 90th and 95th centiles, for clinical reference.

Conclusion: To the best of our knowledge, our series is the largest sample ofAC reported in the medical literature. We believe that the gestational age-specific nomogram of fetal AC is important for further clinical assessment of fetal growth. [TaiwanJ Obstet Gynecol 2008;47(1):49-56]

Key Words: abdominal circumference, Asian population, fetus, pregnancy

Introduction

Among the common fetal growth parameters, such as abdominal circumference (AC), biparietal diameter and femur length, AC is the most sensitive single indicator of restricted or accelerated fetal growth [1-3]. To date, many reports have presented reference ranges for AC versus gestational age (GA) from various populations, especially Caucasian populations in Western countries [1,4-10]. Fetuses from different populations may show different growth patterns, and fetal AC values are

different in previous reports [1,4-10]. Moreover, no fetal AC values in Taiwanese have been officially reported in the medical literature. In this series, we attempted to investigate whether the fetal AC values in Taiwanese are different from those in the Western populations [1,4-7,9]. In addition, we also established a GA-specific nomogram of Taiwanese fetal AC for clinical reference.

Materials and Methods

* Correspondence to: Dr Chiung-Hsin Chang, Department of Obstetrics and Gynecology, National Cheng Kung University Hospital, 138, Victory Road, Tainan 70428, Taiwan. E-mail: ahsin@mail.ncku.edu.tw Accepted: July 5, 2007

ELSEVIER

We retrospectively reviewed our computer ultrasound database of fetal AC records from January 1991 to December 2006. The analysis data included the examination date, GA, and fetal AC assessed by ultrasonography. During the study period of16 years, only fetuses with GA ranging from 14 to 41 weeks were included.

Initially, 53,723 records were included for calculation of mean and standard deviation (SD) values for each specific gestational week (Figure 1). Then, we excluded the extreme bilateral records, which were larger or smaller

Figure 1. Flow chart of data analysis.

than the mean ± 1.66 SD. Finally, 50,131 records of AC were included for final analysis.

All examinations were performed using conventional ultrasound scanners with a 3.5-MHz real-time abdominal probe, including equipments manufactured by Aloka (Tokyo, Japan), Medison-Kretz (Zipf, Austria), and Toshiba (Tokyo, Japan). The ultrasound measurements followed a previous method reported by Tamura et al [1]. The anteroposterior (AD1) and transverse (AD2) diameters were measured, and AC was calculated from the formula: 3.14x (AD1 + AD2)/2. The mean and SD values of AC for each gestational week were calculated. Using GA as the independent variable and AC as the dependent variable, the linear regression analysis and the polynomial regression analysis (up to the fourth order) were undertaken to find the best-fit regression equations for predicted values and age-related variances [4]. In addition, we used the best-fit polynomial regression equations of GA versus AC to establish a GA-specific nomogram for clinical use.

Table 1. The gestational age (GA)-specific mean and standard deviation (SD) for abdominal circumference

GA (wk) Original records Observed values

Mean (mm) SD (mm) Numbers Mean (mm) SD (mm) Numbers

14 88.6 14.3 133 89.2 13.9 130

15 97.5 13.0 385 96.9 9.3 361

16 106.0 14.2 869 106.0 9.0 830

17 117.0 13.4 948 117.0 9.4 877

18 127.0 15.2 881 127.0 10.6 818

19 139.0 34.9 849 138.0 14.0 839

20 152.0 13.7 2,049 153.0 9.9 1,933

21 162.0 22.8 5,592 162.0 10.6 5,545

22 170.0 12.6 5,568 171.0 9.0 5,152

23 178.0 13.9 2,783 178.0 10.1 2,587

24 188.0 16.1 1,782 188.0 11.9 1,640

25 198.0 16.9 1,688 200.0 12.3 1,543

26 208.0 17.8 1,349 209.0 12.5 1,239

27 219.0 20.3 1,257 220.0 13.9 1,162

28 231.0 18.7 1,373 232.0 13.9 1,269

29 242.0 20.1 1,567 242.0 14.4 1,462

30 251.0 21.3 1,809 252.0 14.9 1,703

31 262.0 21.7 1,692 263.0 15.4 1,577

32 271.0 20.2 1,992 271.0 15.0 1,819

33 280.0 21.9 1,974 281.0 15.7 1,827

34 290.0 22.9 2,120 290.0 16.4 1,952

35 300.0 22.4 2,218 300.0 16.4 2,027

36 309.0 23.0 2,546 309.0 17.3 2,309

37 319.0 23.6 2,983 319.0 17.1 2,752

38 327.0 21.1 3,015 327.0 16.0 2,739

39 332.0 25.2 2,474 333.0 17.7 2,369

40 337.0 20.8 1,433 337.0 15.8 1,321

41 335.0 20.7 393 336.0 15.8 356

Total 53,723 50,131

Fetal AC in an Asian Population

Results

Table 1 presents the sample size and the observed values of the mean and SD of AC for each GA in the initial 53,723 data and in the final 50,131 eligible data.

Using GA as the independent variable and AC as the dependent variable, we underwent the polynomial regression analysis up to the fourth order. The result showed that the second-order polynomial regression was the best-fit polynomial regression equation: AC = -0.0653 GA2 + 13.368 GA - 90.932 (R2 = 0.9983). The best-fit equation for SD was: SD = -0.0028 GA2 + 0.5326 GA +1.9693. We used the second-order polynomial regression to calculate the predicted values and established a table and a figure of predicted values for further clinical reference (Table 2 and Figure 2).

To compare our AC data with other previous series in the medical literature, Tables 3-5 and Figures 3-5 are listed to show the comparisons and depict the trends

of 5th, 50th and 95th centiles in our population and in the other populations from previous reports.

Discussion

Several cross-sectional and longitudinal studies have documented racial variations in fetal growth [9,11]. Previous studies involving one Korean population have found that biometric parameters, including fetal AC, in the Korean population during pregnancy were smaller than those in Caucasian populations [10]. When compared with the 5th percentile measurements, our predictive values were similar to other references (Table 4 and Figure 3). For the 50th percentile measurements, our predictive values were similar to other series from 14 to 32 weeks and smaller than the Western series after 33 weeks (Table 3 and Figure 4). However, for the 95th percentile measurements, the difference appeared

Table 2. The gestational age (GA)-specific predicted values of abdominal circumference

Abdominal circumference (mm) GA -

(wk) 5th 10th 25th 50th 75th 90th 95th SD

percentile percentile percentile percentile percentile percentile percentile

14 69.0 72.4 77.7 83.4 89.2 94.4 97.9 8.7

15 79.8 83.4 88.9 94.9 100.9 106.4 110.0 9.1

16 90.5 94.3 100.0 106.2 112.5 118.2 122.0 9.5

17 101.1 105.0 110.9 117.5 124.0 129.9 133.8 9.9

18 111.5 115.6 121.8 128.5 135.3 141.4 145.5 10.2

19 121.9 126.1 132.5 139.5 146.5 152.9 157.1 10.6

20 132.1 136.5 143.1 150.3 157.5 164.1 168.5 11.0

21 142.2 146.7 153.5 161.0 168.5 175.3 179.8 11.3

22 152.2 156.8 163.9 171.6 179.3 186.3 190.9 11.7

23 162.0 166.8 174.1 182.0 189.9 197.1 201.9 12.0

24 171.8 176.7 184.1 192.3 200.4 207.9 212.8 12.4

25 181.4 186.5 194.1 202.5 210.8 218.5 223.5 12.7

26 190.9 196.1 203.9 212.5 221.1 228.9 234.1 13.0

27 200.2 205.6 213.6 222.4 231.2 239.2 244.6 13.4

28 209.5 215.0 223.2 232.2 241.2 249.4 254.9 13.7

29 218.6 224.2 232.6 241.8 251.1 259.4 265.0 14.0

30 227.6 233.3 241.9 251.3 260.8 269.3 275.1 14.3

31 236.5 242.3 251.1 260.7 270.4 279.1 284.9 14.6

32 245.3 251.2 260.2 270.0 279.8 288.7 294.7 14.9

33 253.9 260.0 269.1 279.1 289.1 298.2 304.3 15.2

34 262.4 268.6 277.9 288.1 298.3 307.6 313.8 15.5

35 270.8 277.1 286.6 297.0 307.3 316.8 323.1 15.7

36 279.1 285.5 295.1 305.7 316.3 325.9 332.3 16.0

37 287.3 293.8 303.5 314.3 325.0 334.8 341.3 16.3

38 295.3 301.9 311.8 322.8 333.7 343.6 350.2 16.5

39 303.2 309.9 320.0 331.1 342.2 352.3 359.0 16.8

40 311.0 317.8 328.1 339.3 350.6 360.8 367.6 17.0

41 318.7 325.6 336.0 347.4 358.8 369.2 376.1 17.3

SD = standard deviation.

Gestational age (wk)

Figure 2. The gestational age-specific nomogram of fetal abdominal circumference.

Table 3. The comparison of 50th percentile for gestational age (GA)-specific abdominal circumference

50th percentile for GA-specific abdominal circumference

GA - Predicted values

(wk) Tamura Snijders Chitty Kurmanavicius Smulian Salomon of this report

et al [1] et al [5] et al [4] et al [6] et al [7] et al [9]

14 90.0 78.9 79.8 79.0 93.8 83.4

15 99.0 90.3 91.3 94.0 102.5 94.9

16 108.0 101.6 102.7 107.0 111.7 106.2

17 118.0 112.9 114.0 116.0 121.2 117.5

18 131.0 128.0 124.1 125.3 128.0 131.1 128.5

19 144.0 139.0 135.2 136.4 139.0 141.2 139.5

20 154.0 149.0 146.2 147.5 150.0 151.6 150.3

21 170.0 161.0 157.1 158.5 159.0 162.2 161.0

22 180.0 172.0 168.0 169.3 172.0 172.9 171.6

23 193.0 183.0 178.7 180.0 182.0 183.6 182.0

24 205.0 195.0 189.3 190.6 193.0 194.5 192.3

25 213.0 207.0 199.8 201.1 204.0 205.3 202.5

26 221.0 219.0 210.2 211.5 215.0 216.0 212.5

27 237.0 231.0 220.4 221.7 226.0 226.7 222.4

28 253.0 243.0 230.6 231.8 236.0 237.2 232.2

29 269.0 254.0 240.5 241.7 246.0 247.4 241.8

30 274.0 266.0 250.4 251.5 258.0 257.5 251.3

31 280.0 277.0 260.1 261.1 268.0 267.2 260.7

32 287.0 287.0 269.7 270.6 278.0 276.6 270.0

33 290.0 297.0 279.1 279.9 289.0 285.6 279.1

34 301.0 307.0 288.4 289.0 299.0 294.2 288.1

35 322.0 316.0 297.5 298.0 309.0 302.3 297.0

36 333.0 324.0 306.4 306.8 317.0 309.8 305.7

37 344.0 332.0 315.1 315.4 329.0 316.8 314.3

38 357.0 339.0 323.7 323.8 340.0 323.1 322.8

39 359.0 345.0 332.1 332.0 352.0 328.7 331.1

40 361.0 340.4 340.0 359.0 333.7 339.3

41 371.0 348.4 347.8 366.0 337.8 347.4

as early as 30 weeks (Table 5 and Figure 5). Given the above comparison, there seems to be a trend that our fetuses had smaller AC in the third trimester, especially for the upper limits at the 95th centile. Although there

were some differences in AC centiles among these references, there is, in fact, no appropriate statistical method to examine whether there is a significant difference between our report and other reports. Future

Table 4. Th e comparison of 5th percentile for gestational age (GA)-specific abdominal circumference

GA (wk) 5th percentile for GA-specific abdominal circumference Predicted values of this report (50,131 cases)

Tamura et al [1 ] (197 cases) Snijders et al [5] (1,040 cases) Chitty et al [4] (425 cases) Kurmanavicius et al [6] (5,807 cases) Smulian et al [7] (10,070 cases) Salomon et al [9] (19,647 cases)

14 78.0 70.1 66.8 62.4 82.7 69.0

15 86.0 80.5 77.5 75.7 90.4 79.8

16 94.0 90.8 88.1 88.7 98.5 90.5

17 103.0 100.9 98.6 99.4 106.9 101.1

18 103.0 112.0 111.2 109.1 111.4 115.7 111.5

19 116.0 122.0 121.3 119.5 122.4 124.8 121.9

20 126.0 130.0 131.3 129.8 131.7 134.1 132.1

21 142.0 141.0 141.2 140.0 140.7 143.6 142.2

22 152.0 151.0 151.1 150.0 153.7 153.2 152.2

23 165.0 160.0 160.8 159.9 162.1 162.9 162.0

24 177.0 171.0 170.4 169.7 174.7 172.6 171.8

25 185.0 181.0 179.9 179.4 185.7 182.4 181.4

26 193.0 192.0 189.3 189.0 193.4 192.1 190.9

27 209.0 203.0 198.5 198.4 204.4 201.6 200.2

28 225.0 214.0 207.7 207.7 214.4 211.1 209.5

29 241.0 223.0 216.6 216.9 224.4 220.3 218.6

30 246.0 234.0 225.5 225.9 236.4 229.2 227.6

31 252.0 244.0 234.2 234.7 244.8 237.9 236.5

32 259.0 252.0 242.8 243.4 256.4 246.2 245.3

33 262.0 260.0 251.2 251.9 264.1 254.2 253.9

34 273.0 269.0 259.5 260.2 274.1 261.7 262.4

35 294.0 277.0 267.6 268.4 282.4 268.7 270.8

36 305.0 284.0 275.5 276.4 288.8 275.1 279.1

37 316.0 292.0 283.2 284.2 300.8 281.0 287.3

38 329.0 298.0 290.8 291.9 310.1 286.3 295.3

39 331.0 303.0 298.2 299.3 318.8 290.9 303.2

40 333.0 305.5 306.5 330.8 294.7 311.0

41 343.0 312.5 313.5 339.4 297.8 318.7

Table 5. The comparison of 95th percentile for gestational age (GA)-specific abdominal circumference

95th percentile for GA-specific abdominal circumference

GA - Predicted values

(wk) Tamura Snijders Chitty Kurmanavicius Smulian Salomon of this report

et al [1] et al [5] et al [4] et al [6] et al [7] et al [9]

14 102.0 87.7 92.8 95.6 104.9 97.9

15 112.0 100.1 105.1 112.3 114.7 110.0

16 122.0 112.4 117.3 125.3 124.9 122.0

17 133.0 124.9 129.4 132.6 135.5 133.8

18 159.0 144.0 137.0 141.5 144.6 146.5 145.5

19 172.0 156.0 149.1 153.3 155.6 157.7 157.1

20 182.0 168.0 161.1 165.2 168.3 169.1 168.5

21 198.0 181.0 173.0 177.0 177.3 180.8 179.8

22 208.0 193.0 184.9 188.6 190.3 192.5 190.9

23 221.0 206.0 196.6 200.1 201.9 204.4 201.9

24 233.0 219.0 208.2 211.5 211.3 216.3 212.8

25 241.0 233.0 219.7 222.8 222.3 228.1 223.5

26 249.0 246.0 231.1 234.0 236.6 240.0 234.1

27 265.0 259.0 242.3 245.0 247.6 251.7 244.6

28 281.0 272.0 253.5 255.9 257.6 263.2 254.9

29 297.0 285.0 264.4 266.5 267.6 274.6 265.0

30 302.0 298.0 275.3 277.1 279.6 285.7 275.1

31 308.0 310.0 286.0 287.5 291.2 296.5 284.9

32 315.0 322.0 296.6 297.8 299.6 307.0 294.7

33 318.0 334.0 307.0 307.9 313.9 317.1 304.3

34 329.0 345.0 317.3 317.8 323.9 326.7 313.8

35 350.0 355.0 327.4 327.6 335.6 335.9 323.1

36 361.0 364.0 337.3 337.2 345.2 344.5 332.3

37 372.0 372.0 347.0 346.6 357.2 352.5 341.3

38 385.0 380.0 356.6 355.7 369.9 359.9 350.2

39 387.0 387.0 366.0 364.7 385.2 366.6 359.0

40 389.0 375.3 373.5 387.2 372.6 367.6

41 399.0 384.3 382.1 392.6 377.8 376.1

) 350 -

cir200

Predictive value of this report

....... Tamura et al [1]

---Snijders et al [5]

- Chitty et al [4]

-- - Kurmanavicius et al [6]

--Smulian et al [7]

— - Salomon et al [9]

~~1-1-1-1-T~

24 26 28 30 32 Gestational age (wk)

34 36 38

Figure 3. The comparison of 5th percentile for gestational age-specific abdominal circumference.

international collaborative studies are warranted to confirm whether there is a significant difference in AC between our population and those of other reports.

To the best of our knowledge, the data presented in our AC nomogram is based on the largest sample size

reported in the medical literature. There is a trend that recent reports have larger and larger sample sizes than previous reports (Table 4). The larger and larger sample sizes indicate the attempt of investigators to avoid type II error in statistics, i.e. insufficient sample size. However,

Fetal AC in an Asian Population

I 200-

> Predictive value of this report ■ Tamura et al [1] Snijders et al [5] • Chitty et al [4] Kurmanavicius et al [6] Smulian et al [7] Salomon et al [9]

24 26 28 30 32 Gestational age (wk)

34 36 38 40

Figure 4. The comparison of 50th percentile for gestational age-specific abdominal circumference.

cir 200

Predictive value of this report Tamura et al [1] Snijders et al [5] Chitty et al [4] Kurmanavicius et al [6] Smulian et al [7] Salomon et al [9]

24 26 28 30 32 Gestational age (wk)

Figure 5. The comparison of 95th percentile for gestational age-specific abdominal circumference.

it is not an easy task to undertake a clinical study with a sample size ofmore than 50,000 records. In this series, we endeavored to establish our computer database of fetal AC records between 1991 and 2006, i.e. a study period of 16 years. With standardized operation procedure and computerized digital data management, we have taken great effort during the 16 years to collect the data. Eventually, 50,131 records of AC which fit the criteria were included for final analysis. With our sample size which is believed to be the largest in the medical literature, our AC data should be a good representation of fetal AC in Taiwanese.

In conclusion, our study provides new reference equations, tables and figures of fetal AC in an Asian population in South Taiwan. These novel reference equations, tables and figures of fetal AC should be very useful in the screening, diagnosis and management of fetal abnormal growth, such as fetal growth restriction, large-for-gestational-age, macrosomia, in prenatal care. We are now undergoing a clinical study to test the efficacy of fetal AC in screening fetal abnormal growth using the fetal AC references from this report. We will report our results in the near future.

Acknowledgments

Special thanks to all the staff of Laboratory of Prenatal

Ultrasound, Department of Obstetrics and Gynecology, National Cheng Kung University Medical Center,

for their assistance in completing this manuscript.

References

1. Tamura RK, Sabbagha RE. Percentile ranks of sonar fetal abdominal circumference measurements. Am J Obstet Gynecol 1980;138:475-9.

2. Smith GC, Smith MF, McNay MB, Fleming JE. The relation between fetal abdominal circumference and birth weight: findings in 3,512 pregnancies. Br J Obstet Gynaecol 1997; 104:186-90.

3. Hedriana HL, Moore TR. A comparison of single versus multiple growth ultrasonographic examinations in predicting birth weight. Am J Obstet Gynecol 1994;170: 1600-4.

4. Chitty LS, Altman DG, Henderson A, Campbell S. Charts of fetal size: 3. abdominal measurements. Br J Obstet Gynaecol 1994;101:125-31.

5. Snijders RJ, Nicolaides KH. Fetal biometry at 14-40 weeks' gestation. Ultrasound Obstet Gynecol 1994;4:34-48.

6. Kurmanavicius J, Wright EM, Royston P, Zimmermann R, Huch R, Huch A, Wisser J. Fetal ultrasound biometry: 2. Abdomen and femur length reference values. Br J Obstet Gynaecol 1999;106:136-43.

7. Smulian JC, Ananth CV, Vintzileos AM, Guzman ER. Revisiting sonographic abdominal circumference measurements: a comparison of outer centiles with established nomograms. Ultrasound Obstet Gynecol 2001;18:237-43.

8. Paladini D, Rustico M, Viora E, Giani U, Bruzzese D, Campogrande M, Martinelli P. Fetal size charts for the

Italian population: normative curves of head, abdomen and long bones. Prenat Diagn 2005;25:456-64.

9. Salomon LJ, Duyme M, Crequat J, Brodaty G, Talmant C, Fries N, Althuser M. French fetal biometry: reference equations and comparison with other charts. Ultrasound Obstet Gynecol 2006;28:193-8.

10. Jung SI, Lee YH, Moon MH, et al. Reference charts and equations of Korean fetal biometry. Prenat Diagn 2007;27:545-51.

11. Drooger JC, Troe JW, Borsboom GJ, et al. Ethnic differences in prenatal growth and the association with maternal and fetal characteristics. Ultrasound Obstet Gynecol 2005; 26:115-22.