Newborn assessment and long-term adverse outcome: A systematic review

AJOG REVIEWS

Newborn assessment and long-term adverse outcome: A systematic review Jeroen E. van de Riet, MSc,a Frank P.H.A. Vandenbussche, MD,a Saskia Le Cessie, PhD,b and Marc J.N.C. Keirse, MD, DPhilc Leiden, The Netherlands, and Adelaide, South Australia, Australia The medical literature was searched for publications between 1966 and September 1997 for data on the association of Apgar score, umbilical blood pH, or Sarnat grading of encephalopathy with long-term adverse outcome. Odds ratios for these associations were combined to calculate common odds ratios with 95% confidence intervals. Our search identified abstracts of 1312 studies and 81 articles with sufficient numeric data to formulate contingency tables. Forty-two of these qualified for inclusion in our meta-analysis. The strongest associations in the prediction of neonatal death were found by comparing umbilical artery pH <7 with pH ≥7 (common odds ratio 43; 95% confidence interval 15-124) and by comparing Sarnat grade III with grade II (common odds ratio 24; 95% confidence interval 13-45). In the prediction of cerebral palsy, the strongest associations were found for Sarnat grade III versus grade II (common odds ratio 20; 95% confidence interval 6-70) and for 20-minute Apgar score 0 to 3 versus 4 to 6 (common odds ratio 15; 95% confidence interval 5-50). (Am J Obstet Gynecol 1999;180:1024-9.)

Key words: Apgar score, cerebral palsy, encephalopathy, hydrogen ion concentration, meta-analysis, neonatal mortality, umbilical cord blood

Perinatal asphyxia, whether it is defined clinically or biochemically, can occur before, during, or after birth and may be followed by death, complete recovery, or recovery with residual defects. The latter may appear after a period of apparently complete recovery during which normal maturation processes were nevertheless inhibited. Signs that may relate to perinatal asphyxia, such as fetal bradycardia or low Apgar score, occur quite frequently. Uncertainty about the relationship (whether causal or not) between such signs and impaired brain development is a continuing cause of concern for both lay people and clinicians. Several studies have reported on the association between postnatal indicators of perinatal asphyxia and subsequent adverse outcome. The severity of perinatal asphyxia in these studies has been indicated either by clinical scoring systems, such as the Apgar and Sarnat scores,1,2 or by biochemical measurements, such as cord blood pH.3 Most of these studies define adverse outcome as neonatal death or cerebral palsy. In this study we used the principles of classic metaFrom the Department of Obstetrics and Gynaecologya and the Department of Medical Statistics,b Leiden University Medical Centre, and the Department of Obstetrics and Gynaecology, The Flinders University of South Australia and Flinders Medical Centre.c Reprints not available from the authors. Copyright © 1999 by Mosby, Inc. 0002-9378/99 $8.00 + 0 6/1/94204

1024

analysis for analyzing and synthesizing quantitative associations between perinatal asphyxia and serious adverse outcome for 2 purposes: first, to provide more insight in the relative importance of different clinical and biochemical markers for perinatal asphyxia and adverse outcome; and second, to assist clinicians in drawing realistic infant prognoses from such data. Methods We searched the medical literature from 1966 to September 1997 for studies examining the association between newborn assessment scores and long-term adverse outcome. DIMDI (Deutsches Institut für Medizinische Dokumentation und Information) was used as a host to search 20 common medical databases, including MEDLINE, on the association between Apgar score or cord blood pH and long-term adverse outcome. We searched for the following key words, alone or in combination: Apgar score, umbilical cord blood, blood pH, acid-base equilibrium, newborn, neonatal, death, and cerebral palsy. MEDLINE was used to search for the association between neonatal encephalopathy and longterm adverse outcome. In the latter search we used neonatal encephalopathy, Sarnat, and neurologic disability as text words, and infant mortality, cerebral palsy, mental retardation, developmental disabilities, and brain damage as key words. From the abstracts identified in this manner, we studied the eligible articles and their ref-

van de Riet et al 1025

Volume 180, Number 4 Am J Obstet Gynecol

Table I. Meta-analysis of associations between newborn assessment scores and neonatal death in birth cohort and selective follow-up studies Neonatal deaths/live births Newborn assessment 1-min Apgar score2,4,6,7,11,31-36 0-3 versus 4-6 4-6 versus 7-10 5-min Apgar score2,5,6,8-10,30-32 0-3 versus 4-6 4-6 versus 7-10 10-min Apgar scorel0,30,31 0-3 versus 4-6 4-6 versus 7-10 20-min Apgar scorel0 0-3 versus 4-6 4-6 versus 7-10 Sarnat Score2,13,22-29,31,37 III versus II II versus I Umbilical artery pH12,14-21 pH < 7.10 versus pH ≥7.10 pH < 7.20 versus pH ≥7.20 pH < 7.00 versus pH 7.00-7.10 pH < 7.00 versus pH 7.00-7.20

Lower score

Higher score

Common odds ratio

95% CI

833/5,462 379/11,744

379/11,744 600/85,587

5.7 7.1

5.0-6.5 4.0-12.8

1,761/3,354 939/6,510

939/6,510 2,113/370,785

5.7 20.4

4.2-7.9 10.8-38.6

250/368 181/672

181/672 181/5,312

5.7 6.0

4.3-7.6 1.8-20.5

156/178 95/233

95/233 91/1,950

10.3 14.1

6.1-17.3 10.1-19.7

124/184 17/270

19/306 0/366

23.7 7.2

12.5-44.9 2.0-25.5

44/595 36/145 7/87 7/87

130/4,722 42/756 1/385 7/3,412

7.6 4.9 33.6 42.6

3.3-17.5 2.2-10.8 4.1-278.0 14.6-124.2

Table II. Meta-analysis of associations between newborn assessment scores and cerebral palsy in surviving children in birth cohort, selective follow-up and case-control studies Cerebral palsy/survivors Newborn assessment 1-min Apgar score10 0-3 versus 4-6 4-6 versus 7-10 5-min Apgar score10 0-3 versus 4-6 4-6 versus 7-10 10-min Apgar score10 0-3 versus 4-6 4-6 versus 7-10 20-min Apgar score10 0-3 versus 4-6 4-6 versus 7-10 Sarnat score29,31,38 III versus II II versus I Umbilical artery pH*15,16,21,39-41 pH <7.00 versus pH >7.00 pH <7.10 versus pH ≥7.10 pH <7.20 versus pH ≥7.20

Lower score

Higher score

Common odds ratio

95% CI

39/2,285 28/6,421

28/6,421 90/39,036

4.0 1.9

2.4-6.5 1.2 - 2.9

22/434 18/1,356

18/1,356 116/46,537

4.0 5.4

2.1-7.5 3.3-8.9

14/114 11/470

11/470 31/5,012

5.8 3.9

2.6-13.3 1.9-7.7

9/22 6/138

6/138 11/1,859

15.2 7.6

4.7-49.6 2.8-21.0

30/34 13/54

13/54 0/30

20.0 6.7

5.7-70.0 1.1-40.1

3/4 15/51 10/77

39/71 71/482 43/429

2.5 2.4 1.6

0.2-24.8 1.1-5.1 0.6-4.6

*In preterm or low birth weight infants.

erence lists. For inclusion in the analysis, we selected only those articles that studied the appropriate associations and provided sufficient numeric data to allow tabulation of contingency tables. Revman 3.0b (Cochrane Review Manager, Cochrane Collaboration) was used to calculate the common (or typical) odds ratio for each association. Odds ratios and their 95% confidence intervals (95% CI) were calculated using the random effect model because differences in

odds ratios were probably based on coincidence, as well as on differences among study populations. The size of each study was taken into account for assessment of the common odds ratio. Results Our literature search resulted in 1312 abstracts. We read 281 articles, 213 derived from the search and 68 found among the references of these papers. Of these

1026 van de Riet et al

April 1999 Am J Obstet Gynecol

Fig 1. Graphic representation of odds ratios (OR) and their 95% confidence intervals (95% CI) of 12 studies on the association between neonatal death and Sarnat grading of encephalopathy.2,13,22-29,31,37 Their common (or typical) odds ratio was calculated with the Revman 3.0b program (Cochrane Review Manager, Cochrane Collaboration). n, Number of neonatal deaths; N, number of live births.

281 studies, 81 provided sufficient numeric data to construct contingency tables. Thirty-nine of these 81 studies were subsequently excluded because they reported on patients already included in the meta-analysis, or because they used cutoff levels or outcome definitions that differed from those in most other studies. In the remaining 42 studies, Apgar scores were classified into 3 groups (scores of 0-3, 4-6, and 7-10).1 We used the Sarnat grading of encephalopathy (0, I, II, III).2 Umbilical artery pH measurements were classified with intervals of 0.1 between 7.0 and 7.2. The meta-analysis for the prediction of neonatal death was based on different types of observational studies: 8 birth cohort studies,4-11 and 27 follow-up studies of selected infants.2,12-37 The neonatal mortality rate in the 8 birth cohort studies was 1.4%.4-11 The meta-analysis for the prediction of cerebral palsy was based on 1 birth cohort,10 7 selective follow-up studies,15,16,21,29,31,38,39 and 2 case-control studies.40,41 The cerebral palsy rate in survivors in the birth cohort study was 0.3%.10 In the birth cohort studies,4-11 very low (0-3) and low (4-6) Apgar scores were given to 5.2% and 11.3% of the infants after 1 minute,4,6,7,10,11 0.9% and 1.7% after 5 minutes,5,6,8-10 0.7% and 1.2% after 10 minutes,10 and 0.3% and 0.4% after 20 minutes, respectively.10 Ten- and 20-minute Apgar scores were available for 11.6% and 4.4% of newborns, respectively, in our analysis.10 Goldaber et al12 found very low (<7.0) and low (7.0-7.1) umbilical artery pH in 0.3% and 1.3% of the newborns. Thornberg et al13 found neonatal encephalopathy in 0.5% (Sarnat grade III), 0.4% (Sarnat grade II), and 0.9% (Sarnat grade I) of newborns, whereas 98.2% showed no signs of neonatal encephalopathy (Sarnat grade 0). Prediction of neonatal death. Table I lists the common odds ratios for the association between newborn assess-

ment scores and neonatal death.2,4-37 Fig 1 illustrates how a common odds ratio was obtained. In this example, the common odds ratio for neonatal death comparing Sarnat grade III versus II is derived from 12 studies with odds ratios ranging from 5.0 to 130.3, all with large 95% CI.2,13,22-29,31,37 The common odds ratio in this example is 23.7 (95% CI 13-45), implying that the odds of neonatal death are 24-fold greater for an infant with Sarnat grade III encephalopathy than for an infant with Sarnat grade II encephalopathy. From the data in Table I, it can be deduced that the odds of neonatal death for infants with Sarnat grade III are 171 times those for infants with Sarnat grade I (odds ratio Sarnat III vs II times odds ratio Sarnat II versus I, or 23.7 × 7.2 = 171). Although some studies allowed us to stratify groups of infants according to their birth weight, odds ratios derived in this manner revealed no consistent pattern. For example, in the study by Nelson and Ellenberg,10 the odds of death with lower 1- and 5-minute Apgar scores increased with decreasing birth weight, whereas in the studies by Ikonen and Lauslahti these odds decreased with decreasing birth weight.34,35 Prediction of cerebral palsy and other neurologic disabilities. Table II lists the common odds ratios for the association between newborn assessment scores and cerebral palsy.10,15,16,21,29,31,38-41 Common odds ratios for umbilical artery pH and cerebral palsy were derived from 6 studies, all dealing with preterm or low birth weight infants.15,16,21,39-41 As shown in Table II, common odds ratios were arrayed around 2 (1.6-2.5), and only the difference between pH <7.10 versus pH ≥7.10 reached conventional statistical significance. Six studies examined the association between neonatal encephalopathy and neurologic disabilities accompanied by mental retardation.13,22-24,37,42 Table III lists the common odds ratios for these associations. Long-term

van de Riet et al 1027

Volume 180, Number 4 Am J Obstet Gynecol

Table III. Meta-analysis of associations between Sarnat scores and neurologic disabilities with mental retardation in surviving children in 6 selective follow-up studies Disabilities/survivors Newborn assessment Sarnat score13,22-24,37,42 III versus II II versus I I versus 0

Lower score

Higher score

Common odds ratio

95% CI

18/23 45/159 0/36

56/174 1/212 3/149

4.6 27.4 0.6

1.5-13.9 7.8-95.6 0-11.4

neurologic disability was found in 84% of survivors with Sarnat grade III, in 30% of survivors with Sarnat grade II, and in fewer than 0.5% of infants graded Sarnat grade 0 or I.13,22-24,29,31,37,38,42 Two studies investigated the association between umbilical vein pH and neurologic disabilities.43,44 Huisjes et al43 found an odds ratio of 8.2 (95% CI 4-15) when comparing the frequency of neurologic disabilities in infants with low (≤7.19) versus high (>7.19) umbilical vein pH. However, Dweck et al44 found an odds ratio of 0.1 (95% CI 0-3) when comparing the frequency of neurologic disabilities in infants with low (≤7.10) versus high (7.11-7.20) umbilical vein pH. Comment Although numerous studies address newborn assessment and serious adverse outcome, only a limited number provide sufficient data to construct contingency tables. Eight studies provided data on large cohorts of children4-11; the others provided data on more restricted groups.2,12-44 A reasonable criticism of studies such as ours is that data are pooled over an extended period of time during which neonatal outcome has improved considerably. For example, the mean neonatal mortality rate of 1.4% in our pooled data4-11 is 3 to 4 times higher than the neonatal mortality rate observed in current clinical practice. We found no indications, however, that the relative impact of a less favorable result had changed drastically from the earlier to the later studies included in our analysis. This is exemplified in Fig 1, with data obtained over a 20-year period, showing individual odds ratios with a variation that is visibly unrelated to year of publication. On the other hand, the cerebral palsy rate in survivors in our pooled analysis was 0.4%,10 and this is equivalent to the rate in more recent studies.45-51 There have been a number of follow-up studies on infants with low umbilical artery pH or buffer base at birth, but only a few included data on control infants.52-57 It appears from these studies, though, that cerebral palsy is a rare outcome after low umbilical blood pH at birth. Therefore large case-control studies would be necessary to document the strength of an association, if any, between low umbilical blood pH and cerebral palsy. In our meta-analysis, only severe neonatal en-

cephalopathy (Sarnat grade III) was strongly predictive of serious adverse outcome. The other newborn assessments predominantly indicate a relative increase or decrease in the probability of long-term adverse outcome. For example, very low (0-3) 1-minute Apgar scores were associated with a 15% neonatal mortality rate, 6 times higher when compared with low (4-6) Apgar scores and 40 times higher when compared with high (7 to 10) Apgar scores. Very low (<7) umbilical artery pH was associated with 8% neonatal mortality rate, a rate 43 times higher than with an umbilical artery pH ≥7. The most important cutoff points in the prediction of neonatal death were low (4-6) versus high (7-10) Apgar scores after 5 and 20 minutes, encephalopathy grade III versus II, and an umbilical artery pH <7 versus ≥7. Common odds ratios of neonatal death showed a clear statistical difference for all comparisons between low and high scores (Table I). Associations between unfavorable newborn assessment scores and cerebral palsy were much weaker than for neonatal death. Only low 20-minute Apgar scores and severe neonatal encephalopathy (Sarnat grade III) were highly predictive of cerebral palsy in survivors. Very low (0-3) 1-minute Apgar scores were associated with cerebral palsy in 1.7% of survivors, a rate 4 times higher than with low (4-6) and 8 times higher than with high (7-10) 1minute Apgar scores. Common odds ratios for the association between umbilical artery pH and cerebral palsy were approximately 2. However, these odds are based on only 6 studies, all in preterm or low birth weight infants, and most confidence intervals included unity. Sarnat grade II encephalopathy was associated with long-term neurologic disability in 30% of survivors (Table III). In summary, newborn assessment scores are reasonable clinical tools in the selection of infants at risk of serious long-term adverse outcome. The different assessment systems each provide their own information, but it is clear that they cannot be assumed to be independent of each other. Severe neonatal encephalopathy (Sarnat grade III) occurred in 0.5% of neonates and was strongly predictive of long-term adverse outcome. Very low (0-3) 20-minute Apgar scores were obtained in 0.3% of neonates; 88% of these infants died and, of the survivors, 41% had cerebral palsy. Although very low (0-3) and low (4-6) 1- and 5-minute Apgar scores, as well as very low

1028 van de Riet et al

(<7) umbilical artery pH, were associated with an at least 5-fold increase in the odds of death, the association with cerebral palsy was less obvious. Fewer than 5% of survivors with a very low (0-3) Apgar score at 5 minutes and fewer than 2% of those with a very low 1-minute Apgar score eventually had cerebral palsy. Associations do not imply causal relationships. Umbilical blood pH, Apgar scores, and Sarnat grading merely reflect the newborn’s condition, which is affected by genetic, prenatal, and perinatal factors. An unfavorable newborn assessment score does not necessarily indicate an adverse outcome, nor does it indicate the cause of the poor assessment result. Nevertheless, the association between newborn assessment scores and long-term adverse outcomes is important. These scores may play a prominent role when deciding whether to withdraw neonatal medical care. They also help to provide realistic counseling to parents and prevent undue pessimism that may cause emotional distress and interference with parent-child bonding. In addition, the relative strength of the associations adds to the understanding of the prognostic value of different scoring systems. Therefore this systematic compilation of data may broaden the epidemiologic base for further studies on the relationship between perinatal factors and adverse long-term outcome. REFERENCES

1. Apgar V. A proposal for a new method of evaluation of the newborn infant. Curr Res Anesth Analg 1953;32:260-7. 2. Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress, a clinical and electroencephalografic study. Arch Neurol 1976;33:696-705. 3. Saling E. Die Blutgasverhältnisse und der Säure-Basen-Haushalt des Feten bei ungestörtem Geburtsablauf. Geburtshilfe Gynäkol 1964;161:262-92. 4. Apgar V, James LS. Further observations on the newborn scoring system. Am J Dis Child 1962;104:419-27. 5. Atkinson D. An evaluation of Apgar scores as predictors of infant mortality. N C Med J 1983;44:45-54. 6. Jennett RJ, Warford HS, Kreinick C, Waterkotte GW. Apgar index: a statistical tool. Am J Obstet Gynecol 1981;140:206-12. 7. Langrehr D, Dollmann A. Behandlungsergebnisse der akuten, postnatalen Asphyxie. Der Anaesthesist 1971;20:262-5. 8. Miller JM, Levine DH, Michel Y. Low Apgar scores and neonatal mortality. Obstet Gynecol 1984;63:602-3. 9. Morales del Valle Z, Vazquez Calzada JL. Apgar scores and infant mortality in Puerto Rico. Puerto Rico Health Sci J 1994;13:175-81. 10. Nelson KB, Ellenberg JH. Apgar scores as predictors of chronic neurologic disability. Pediatrics 1981;68:36-44. 11. Ahvenianen EK, Veistola T. Evaluation of the newborn infant. Ann Paediatr Fenn 1958;5:27-33. 12. Goldaber KG, Gilstrap LC 3rd, Leveno KJ, Dax JS, McIntire DD. Pathologic fetal acidemia. Obstet Gynecol 1991;78:1103-6. 13. Thornberg E, Thiringer K, Odeback A, Milsom I. Birth asphyxia: incidence, clinical course and outcome in a Swedish population. Acta Paediatr 1995;84:927-32. 14. Bresadola M, Lo Mastro F, Arena V, Bellaveglia L, Scarpellini F. Preterm labour and neonatal parameters. Clin Exp Obstet Gynecol 1995;3:235-9. 15. Kato EH, Yamada H, Matsumoto Y, Hattori S, Makinoda S, Fujimoto S. Relation between perinatal factors and outcome of very low birthweight infants. J Perinat Med 1996;24:677-86. 16. Luthy DA, Shy KK, Strickland D, Wilson J , Bennett RC, Brown

April 1999 Am J Obstet Gynecol

17. 18.

19.

20.

21.

22.

23.

24.

25.

26. 27.

28.

29.

30.

31.

32.

33. 34.

35. 36. 37.

38.

39.

40.

41.

ZA, et al. Status of infants at birth and risk for adverse neonatal events and long-term sequelae: a study in low birth weight infants. Am J Obstet Gynecol 1987;157:676-9. Yu VYH, Hollingsworth E. Improving prognosis for infants weighing 1000 g or less at birth. Arch Dis Child 1979;55:422-6. Yudkin PL, Johnson A, Clover LM, Murphy KW. Clustering of perinatal markers of birth asphyxia and outcome at age five years. Br J Obstet Gynaecol 1994;101:774-81. Vintzileos AM, Antsaklis A, Varvarigos A, Papas C, Sofatzis I, Montgomery JT. A randomized trial of intrapartum electronic fetal heart rate monitoring versus intermittent auscultation. Obstet Gynecol 1993;81:899-907. Tejani N, Verma UL. Correlation of Apgar scores and umbilical artery acid-base status to mortality and morbidity in the low birthweight neonate. Obstet Gynecol 1989;73:597-600. Beeby PJ, Elliott EJ, Henderson-Smart DJ, Rieger ID. Predictive value of umbilical artery pH in preterm infants. Arch Dis Child 1994;71:F93-6. Robertson C, Finer N. Term infants with hypoxic-ischemic encephalopathy: outcome at 3.5 years. Dev Med Child Neurol 1985;27:473-84. Fernandez F, Verdu A, Quero J, Perez-Higuaras A. Serum CPKBB isoenzyme in the assessment of brain damage in asphyctic term infants. Acta Pediatr Scand 1987;76:914-8. Levene MI, Sands C, Grindulis H, Moore JR. Comparison of two methods of predicting outcome in perinatal asphyxia. Lancet 1986;1:67-9. Muttitt SC, Taylor MJ, Kobayashi JS, Macmillan L, Whyte HE. Serial visual evoked potentials and outcome in term birth asphyxia. Pediatr Neurol 1991;7:86-90. Airede AI. Birth asphyxia and hypoxic-ischaemic encephalopathy: incidence and severity. Ann Trop Paediatr 1991;11:331-5. Perlman JM, Risser R. Severe fetal acidemia: neonatal neurologic features and short-term outcome. Pediatr Neurol 1993;9:277-82. Yeo CL, Tudehope DI. Outcome of resuscitated apparently stillborn infants: a ten year review. J Paediatr Child Health 1994;30:129-33. De Vries LS. Somatosensory-evoked potentials in term neonates with postasphyxial encephalopathy. Clin Perinatol 1993;20:46382. Misra PK, Srivastava N, Malik GK, Kapoor RK, Srivastava KL, Rastogi S. Outcome in relation to Apgar score in term neonates. Indian Pediatr 1994;31:1215-8. Lipp-Zwahlen AE, Deonna T, Micheli JL, Calame A. Prognostic value of neonatal CT scans in asphyxiated term babies: low density score compared with neonatal neurologic signs. Neuropediatrics 1985;16:209-17. Manganaro R, Mami C, Gemelli M. The validity of the Apgar scores in the assessment of asphyxia at birth. Eur J Obstet Gynecol Reprod Biol 1994;54:99-102. Colburn DW, Salzman M. The Apgar scoring system in evaluation of the newborn infant. New York State J Med 1960;60:246-9. Ikonen RS, Lauslahti K. Apgar scoring and neonatal morbidity in infants weighing 1000-1990 grams. Ann Clin Res 1974;6:1614. Ikonen RS. Apgar scoring and neonatal morbidity in full-sized newborn infants. Ann Clin Res 1973;5:380-4. Øain P, Skråmm I, Hannisdal E, Bjøro K. Breech delivery: an obstetrical analysis. Acta Obstet Gynecol Scand 1988;67:75-9. Khare MD, Merchant RH. Diagnostic and prognostic value of CT brain scan in term neonates with moderate and severe birth asphyxia. Indian Pediatr 1990;27:267-71. Thompson CM, Puterman AS, Linley LL, Hann FM, van der Elst CW, Molteno CD, et al. The value of a scoring system for hypoxic ischaemic encephalopathy in predicting neurodevelopmental outcome. Acta Paediatr 1997;86:757-61. Gaudier FL, Goldenberg RL, Nelson KG, et al. Acid-base status at birth and subsequent neurosensory impairment in surviving 500 to 1000 gm infants. Am J Obstet Gynecol 1994;170:48-53. Grether JK, Nelson KB, Stanley-Emery E 3rd, Cummins SK. Prenatal and perinatal factors and cerebral palsy in very low birthweight infants. J Pediatr 1996;128:407-14. Murphy DJ, Sellers S, MacKenzie IZ, Yudkin PL, Johnson AM.

van de Riet et al 1029

Volume 180, Number 4 Am J Obstet Gynecol

42.

43.

44.

45.

46.

47.

48.

49.

Case-control study of antenatal and intrapartum risk factors for cerebral palsy in very preterm singleton babies. Lancet 1995;346:1449-54. Groenendaal F, van der Grond J, van Haastert IC, Eken P, Mali WPThM, De Vries LS. Bevindingen bij cerebrale protonkernspinresonantie-spectroscopie bij pasgeborenen met asfyxie, en psychomotore ontwikeling. Ned Tijdschr Geneeskd 1996;140:255-9. Huisjes HJ, Aarnoudse JG. Arterial or venous umbilical pH as a measure of neonatal morbidity. Early Hum Dev 1979;3:15561. Dweck HS, Huggins W, Dorman LP, Saxon SA, Benton JW, Cassady G. Developmental sequelae in infants having suffered perinatal asphyxia. Am J Obstet Gynecol 1974;119:811-5. Cummins SK, Nelson KB, Grether JK, Velie EM. Cerebral palsy in four northern California counties, births 1983 through 1985. J Pediatr 1993;123:230-7. Dowding VM, Barry C. Cerebral palsy: changing patterns of birthweight and gestational age (1976/81). Irish Med J 1988;81: 25-9. Hagberg B, Hagberg G, Zetterstrom R. Decreasing perinatal mortality—increase in cerebral palsy morbidity. Acta Obstet Gynecol Scand 1989;78:664-70. Pharoah PO, Cooke T, Cooke RW, Rosenbloom L. Birthweight specific trends in cerebral palsy. Arch Dis Child 1990;65:602-6. Riikonen R, Raumavirta S, Sinivuori E, Seppala T. Changing pat-

50.

51.

52.

53.

54.

55.

56.

57.

tern of cerebral palsy in the southwest region of Finland. Acta Obstet Gynecol Scand 1989;78:581-7. Stanley FJ. Survival and cerebral palsy in low birthweight infants: implications for perinatal care. Paediatr Perinat Epidemiol 1992;6:298-310. Yeargin-Allsopp M, Murphy CC, Oakly GP, Sikes RK. A multiplesource method for studying the prevalence of developmental disabilities in children: the Metropolitan Atlanta Developmental Disabilities Study. Pediatrics 1992;89:624-30. Fee SC, Malee K, Deddisch R, Minogue JP, Min D, Socol ML. Severe acidosis and subsequent neurologic status. Am J Obstet Gynecol 1990;162:802-6. Dennis J, Johnson A, Mutch L, Yudkin PL, Johnson P. Acid-base status at birth and neurodevelopmental outcome at four and one-half years. Am J Obstet Gynecol 1989;161:213-20. Low JA, Galbraith RS, Muir DW, Killen HL, Pater EA, Karchmar EJ. Intrapartum fetal hypoxia: a study of long-term morbidity. Am J Obstet Gynecol 1983;145:129-34. Goodwin TM, Belai I, Hernandez P, Durand M, Paul RH. Asphyxial complications in the term newborn with severe umbilical acidemia. Am J Obstet Gynecol 1992;162:1506-12. Low JA, Galbraith RS, Muir DW, Killen HL, Pater EA, Karchmar EJ. Motor and cognitive defects after intrapartum asphyxia in the mature fetus. Am J Obstet Gynecol 1988;158:356-61. Nagel HTC, Vandenbussche FPHA, Oepkes D, JennekensSchinkel A, Laan LAEM, Bennebroek Gravenhorst J. Follow-up of children born with an umbilical arterial blood pH <7. Am J Obstet Gynecol 1995;173:1758-64.

Correction In the article by Lam et al, entitled “Prenatal ultrasonographic prediction of homozygous type 1 α-thalassemia at 12 to 13 weeks of gestation” (Am J Obstet Gynecol 1999;180:148-50), Fig 1 on page 149 is incorrect. The correct Fig 1 together with its legend is shown here.

Fig 1. Cardiac 4-chamber view of a healthy fetus at 13 weeks of gestation on abdominal ultrasonographic examination. Plus sign, Transverse cardiac diameter 9.2 mm; x, transverse thoracic diameter 20.6 mm; cardiothoracic ratio 0.45.