- Email: [email protected]
Assessment of fetal risk in postdate pregnancies
Assessment of fetal risk in postdate pregnancies Robert H. Usher, MD, Mark E. Boyd, MD, Frances H. McLean, BScN, and Michael S. Kramer, MD Montreal, Quebec, Canada To assess postdate fetal risk, pregnancies in which menstrual history was confirmed by early ultrasound examination were reviewed; 5915 pregnancies within 1 week of term, 1408 1 to 2 weeks postdate, and 340 at least 2 weeks postdate. Fetal distress and meconium release were twice as frequent and meconium aspiration eight times as frequent postterm. Birth asphyxia was unrelated to gestational age. Fractures and palsies were more frequent because of primiparity and macrosomia. Only one antepartum fetal death occurred in 1748 postdate pregnancies. Review of 674 perinatal deaths at 37 plus weeks in Quebec showed no increase in deaths postterm. The increase in fetal distress and meconium aspiration postterm without an increase in birth asphyxia or fetal death may reflect greater responsiveness of the more mature fetus to mild asphyxic insults. Findings of this study could not justify increased fetal monitoring in postdate pregnancies. (AM J OBSTET GvNECOL 1988;158:259·64.)
Key words: Postterm, fetal distress, birth asphyxia, meconium, perinatal mortality
There is a historic concern for fetal well-being in the postterm period. 1-6 This concern is based on studies in which fetal maturity was established by clinical means. With ultrasound examination done early in pregnancy, there is reason to believe that a diagnosis of postterm birth was mistaken in some 70% of cases. 7 Fetal hazard must now be reevaluated in this light. In this article fetal outcomes in the postterm period are compared with those of term infants. Fetal age was determined in all fetuses by last normal menstrual period and confirmed by early ultrasound examination. It was hoped that an answer could be given to the question posed in the title of a recent article, "Postmaturity-much ado about nothing?". 8 This study was one of several conducted on the same population base; the others dealt with obstetric considerations and fetal growth in postterm pregnancy. Methods This study was carried out on patients cared for and delivered of infants at the Royal Victoria Hospital, Montreal, and on the Obstetrical Teaching Service of McGill University between jan. 1, 1978, and March 31, 1986. With few exceptions, patients received antenatal care from their obstetrician starting in the first trimester and had a dating ultrasound measurement of hi-
From the Departments of Obstetrics and Gynecology, Pediatrics, and Epidemiology and Biostatistics, McGill University Faculty of Medicine, and the Royal Victoria Hospital. Presented at the Forty-third Annual Meeting of The Society of Obstetricians and Gynaecologists of Canada, Ottawa, Ontario, Canada, june 24-27, 1987. Reprint requests: Robert H. Vsher, MD, Neonatal Unit, Women's Pavilion, Royal Victoria Hospital, 687 Pine Avenue West, Montreal, Quebec HJA JAJ, Canada.
parietal diameter performed, usually between 16 and 18 weeks. All births meeting the following criteria were included in the study: (1) The date of the last normal menstrual period was recorded; (2) there was a record of an early ultrasound dating examination; (3) gestational age calculated from early ultrasound examination was concordant within 7 days with that calculated from menstrual history; and (4) delivery occurred at or after 273 days (39 completed weeks) from the last normal menstrual period. Of the 15,226 births during the period for which both ultrasound and menstrual data were recorded, 10930 births were concordant within 7 days. The study population included the 7663 infants delivered at or after 273 days and was divided according to gestational age based on last normal menstrual period into three groups: (I) 5915 infants delivered at 39 or 40 completed weeks (273 to 286 days); (2) 1408 infants delivered at 41 completed weeks (287 to 293 days); and (3) 340 infants delivered at 42 completed weeks (294 days) or more. Although high-risk pregnancies at this hospital receive frequent nonstress tests and ultrasound evaluations in the hospital's antenatal center, there is no routine testing of postterm patients, and those reported here were, in fact, seldom monitored with either technique. Inductions of labor were performed in 17% of patients delivered at 39 and 40 weeks, 24% of patients delivered at 41 weeks, and 44% of patients delivered at 42 weeks or later. Deliveries were performed by or in the presence of an attending obstetrician with residents, interns, and students participating. A diagnosis of fetal distress from abnormal fetal heart tracings during labor (88% of labors were electronically monitored) was made when late or variable
259
260 Usher et al.
February 1988 Am J Obstet Gynecol
Table I. Incidence of fetal distress and meconium in amniotic fluid (per 100 births) Gestational age (wk)
39
No. of births Fetal distress Meconium Combinations No fetal distress or meconium Fetal distress alone Meconium alone Fetal distress + meconium Total
+ 40
RR* (95% Cl)
5915 15.9 15.3
1408 21.9 27.0
73.3 11.4 10.9 4.5 100.1
59.4 13.6 18.8 8.2 100.0
1.38 (1.23-1.55) 1.76 (1.58-1.96) 0.81 1.20 1.73 1.84
(0.78-0.84) ( 1.03-1.39) (1.51-1.97) ( 1.49-2.27)
340 29.7 31.5 54.1 14.4 16.2 15.3 100.0
RR (95% CI)
Probabilityf
1.87 (1.56-2.25) 2.05 (1.72-2.46)
<0.001 <0.001
0.74 1.27 1.49 3.41
<0.001 <0.01 <0.001 <0.001
(0.68-0.80) (0.96-1.66) (1.15-1.93) (2.60-4.48)
RR = Relative risks; CI = confidence interval. *Relative risks are in comparison with deliveries at 39 and 40 weeks. tProbability is based on x' for linear trend analysis.
Table II. Fetal morbidity (rates per 1000 live births) Gestational age (wk)
39
Number of live births Depression at birth Moderate Severe Postasphyxic encephalopathy Meconium aspiration Fractures/palsies
+ 40
5906 23.4 3.7 0.7 2.2 10.7
RR* (95% Cl) 1407 31.3 3.6 1.4 5.7 14.2
RR (95% Cl)
Probabilityf
340 1.34 0.95 2.10 2.58 1.33
(0.96-1.87) (0.36-2.51) (0.40-11.02) (1.11-6.03) (0.81-2.91)
26.5 8.8 2.9 17.6 29.4
1.13 2.37 4.34 8.02 2.76
(0.58-2.20) (0.74-7.61) (0.58-32.26) (3.56-18.04) (1.46-5.21)
NS NS NS <0.001 <0.01
RR = Relative risks; CI = confidence interval. *Relative risks are in comparison with deliveries at 39 and 40 weeks. tProbability is based on x' for linear trend analysis.
decelerations or other abnormal fetal heart patterns were noted by the obstetric staff. Fetal heart pattern and meconium staining of amniotic fluid were recorded on a checklist on the obstetric chart after each delivery by physicians attending the birth and have not been reviewed further. All neonatal records were reviewed after discharge by R. U. for standardization and confirmation of diagnoses. Depression at birth was defined as the need for positive pressure bag and mask ventilation to initiate respirations, moderate if for 3 minutes or less, and severe if for more than 3 minutes. Postasphyxic encephalopathy included abnormalities in tone or neurologic state, including convulsions in the most severe infants, which occurred during the first days of life after perinatal asphyxic insult. Meconium aspiration syndrome was diagnosed when any degree of respiratory distress and abnormal radiographic findings followed delivery and when meconium was present in amniotic fluid. Birth trauma included fractures (clavicular) or paralyses (brachial or facial), the latter of which often resolved quickly. The McGill Obstetrical and Neonatal Data Base9 was used to computerize the data and perform the analyses.
Proportions were compared with the x' test. Confidence intervals around the reported relative risks are based on the test-based method ofMiettinen. 10 Tests for linear trends in proportion used the x' test described by Armitage.11
Results Fetal distress and meconium release occurred more commonly postterm (Table 1). The rates of both approximately doubled among deliveries at or after 42 weeks compared with deliveries occurring at term. Relative risks were intermediate at 41 weeks. Fetal distress developed in combination with meconium release in 4.5% of term deliveries and in 15.3% of deliveries at or after 42 weeks. There was no significant increase in depression at birth postterm either of moderate or severe degree or of postasphyxic encephalopathy (Table II). Signs of the latter developed in only three of 1747 infants delivered at or after 41 weeks. There was a progressive (eightfold) increase in meconium aspiration, which occurred in 2.2/1000 births or one infant in every 455 delivered at term, one in
Fetal risks in postdate pregnancies 261
Volume 158 Number 2
Fetal 01stress
Mecon1um
40 i
30 1-
z
u.J
~
20
u.J
0...
10
~· .................. .. ··'·''''''·······• ,
8/8
o~---r-----,,-----~
Oepress1on at B1rth
Fracture or Palsy
8
•
6
•-•
Pr~tliiP'
............
Pr11111p .1[)[)[)q ·
o"'"'"''o
M1111q1 .1[)[1(1;1
~(J()(lq
5 1-
z
u.J
'--'
a:
·~······0 .. :: ......
4
u.J
~-
~,. ..... o
a
0...
•
3
...
~0 2
, 0:,.,,,
··········· f
0
.... ,,,,,,,,o
e•"''"'""'""•
o~---r-----,,-----~
39 40
42
41
39 40
41
42
GESTATIONAL AGE (weeksl
Fig. 1. Rates of fetal distress, meconium release, depression at birth, and birth trauma for infants at 39 and 40 weeks, 41 weeks, and 42 + completed weeks of gestation. Effects of parity and fetal size are demonstrated.
175 at 41 weeks, and one in 57 by 42 weeks (Table II). These cases were, however, relatively mild with no deaths, no need for mechanical ventilation, and only 56% of the affected infants requiring oxygen therapy. Meconium, when released, was more often thick postterm: 37.4% at 39 and 40 weeks, 44.5% at 41 weeks, and 51.4% at 42 weeks among those with meconiumstained amniotic fluid. This 1.4-fold increase in the thickness of meconium postterm was insufficient to account for the observed fourfold increase in meconium aspiration syndrome among postterm-exposed infants. Fractures and paralyses were also more frequent postterm; the risk increased from one infant with birth trauma in every 93 at term to one in 34 at 42 weeks (Table II). In this study postterm pregnancies were more often primiparous: 4 7% of deliveries at 39 and 40 weeks, 60% at 41 weeks, and 66% at 42 weeks or later. Postterm
fetuses were also more often macrosomic, with birth weight of 4000 gm or greater in 23% of primiparas and 40% of multiparas at 42 weeks or later compared with 9% and 14%, respectively, at term. Because both primiparity and macrosomia are associated with an increased risk of prolonged labor and difficult delivery, it was necessary to reassess the role of postterm labor and delivery on fetal outcome after controlling for parity and fetal size. With these variables controlled, fetal distress and meconium release continued to show an increased risk postterm, whereas depression at birth and birth trauma showed no relationship to gestational age (Fig. 1). Fetal distress and meconium release increased postterm, yet without the expected parallel increase in depression at birth. The relationship between these signs of fetal jeopardy during labor and the condition of the infant at birth was therefore analyzed (Table III).
262
Usher et al.
February 1988 Am J Obstet Gynecol
Table III. Incidence of depression at birth as related to fetal distress and meconium release(%)
Table IV. Perinatal mortality (per 1000 births) Gestational age (wk)
39 + 40
Gestational age (wk)
42
39 and 40
No fetal distress or meconium Fetal distress and/or meconium
1.9% (82/4335) 5.3% (84/1580)
2.2% (18/836) 5.6% (32/572)
0% (0/184) 8.3% (13/156)
Fetal distress and meconium release were associated with an increased risk of depression at birth at all gestational ages. Perinatal death was not increased in postterm pregnancy compared with patients delivered of infants at term, although the overall number of deaths (19, eight of which were from malformations) was too .few for meaningful analysis (Table IV). Of the four infants who died at least I week postterm, two had lethal malformations, a third weighing 2150 gm died at 41 weeks before labor of unsuspected fetal malnutrition in an obese woman, and the fourth weighing 5640 gm died of birth asphyxia sustained during a difficult delivery complicated by severe shoulder dystocia at 42 weeks. These latter two cases were the only nonanomaly deaths among the 1748 pregnancies that had continued at least 7 days postterm.
Comment The purpose of this retrospective study was to ascertain whether there was an increased risk to the fetus carried postterm. If so, what was the nature of this risk? To what degree could long-term deprivation of oxygen supply to the fetus be demonstrated? What was the significance and role played by meconium release and aspiration? Was dystocia still a concern in the modern postterm delivery? Most of the emphasis in the study was placed on fetal and neonatal morbidity related to fetal distress, birth asphyxia, trauma, and meconium aspiration. These complications represent the concerns specifically related to postdate pregnancy. Perinatal death at and after term is now so rare that very large populations are needed to reflect an increased risk postterm, and many of the deaths that occur postterm are from causes such as malformations that cannot be attributed to the duration of pregnancy. This study shows that fetal distress and meconium release develop more commonly postterm than at term, with intermediate rates at 41 completed weeks. Although meconium release has regularly been reported
No. of births Fetal deaths Total Nonanomaly Neonatal deaths Total Nonanomaly Perinatal deaths Total Nonanomaly
5915
42+
1408
340
1.5 1.5
0.7 0.7
0 0
1.0 0.0
1.4 0
3.0 3.0
2.5 1.5
2.1 0.7
3.0 3.0
as much more frequent postterm, 12 " 13 fetal distress has been less clearly recognized as a risk. 14 Meconium aspiration syndrome is a problem to which the postterm infant is particularly predisposed. 15 This study shows a twofold increase in meconium release and an eightfold increase in frequency of meconium aspiration syndrome postterm. Expressing the frequency of meconium aspiration syndrome as a proportion of infants at risk (those with meconium in amniotic fluid), this rate rose from 14% (13/90) at term to 21% (eight/38) at 41 weeks, and 55% (six/11) postterm (Tables I and II). That is to say that not only is meconium present more often in amniotic fluid postterm, but when present, is more often aspirated. There was no concomitant increase in frequency of birth asphyxia, at least as reflected by depression at birth requiring manual ventilation. Although the number of depressed infants at 42 weeks or more is too small to speak with confidence about the lack of risk postterm, the number at 41 weeks is large enough and should represent some degree of postterm asphyxic risk if such were present. We are left with an apparent inconsistency: more fetal jeopardy in utero without a concomitant increase in depression at birth. One explanation might be that fetal distress and meconium may have less of a predictive value for birth asphyxia at term, but this could not be demonstrated (Table III). Rather, the reason for the inconsistency is that a comparatively large number of infants (82/166) were depressed after delivery at term but did not show fetal distress or meconium release during labor, whereas this did not occur postterm (0113). When present, fetal distress and meconium have similar ominous significance at term as postterm, but their absence is much more reassuring postterm. A hypothesis can be developed to explain these findings. The more mature postterm fetus may be more responsive and react more readily to levels of asphyxic stimuli that may not initiate a response at term. These
Fetal risks in postdate pregnancies
Volume 158 Number 2
263
Table V. Causes of perinatal death at term and postterm, Province of Quebec, April I, 1980 to March 31, 1982 21 Gestational age (complete wk*)
Estimated No. of birthst No. of fetal deaths No. of neonatal deaths No. of deaths by cause Malformation Fetal malnutrition Infection Rh hemolytic disease Maternal diseases Abruptio, placenta previa Dystocia, malpresentation Birth asphyxia, other Other known causes Unexplained fetal death antepartum Unexplained fetal death intrapatrum Unexplained neonatal death Total
37-41
>42
Total
% postterm
131,085 352 271
20646 27 24
151,731 379 295
13.6 7.1 8.1
219 66 13 2 20 36 21 24 100 103 15 4
22 2 2 0 0 2 4 2 8 7 2 0
241 68 15 2 20 38 25 26 108 110 17 4
~
---sT
-----m4
9.1 2.9 13.3 0.0 0.0 5.3 16.0 7.7 7.4 6.4 11.8 0.0 7.6
*At time of delivery or at time of fetal death if different. tBased on a previous analysis of gestational age distribution of births in the Province of Quebec.
reactions are manifested by fetal heart abnormalities, meconium release, and gasping movements. The less responsive term fetus may be sufficiently asphyxiated to be depressed at birth without affecting the heart rate during labor or causing the release of meconium. The fetus at term may also pass meconium in utero and yet not be responsive enough to gasp and aspirate it. These interpretations would account for why fetal distress, meconium release, and meconium aspiration occur more often postterm and yet apparently do not indicate that the postterm infant is more often asphyxiated (as evidenced by depression at birth) than those delivered at term. Furthermore, little serious morbidity is associated with perinatal asphyxia postterm. None of the infants with meconium aspiration required ventilatory assistance, and none of the depressed infants demonstrated postasphyxic neurologic sequelae on follow-up examination. The criterion for birth asphyxia-need for ventilation-used in this study has previously been found more specific and meaningful than Apgar scores, which is the index often used. 16 Had the definition of birth asphyxia instead been a !-minute Apgar score <7, the incidence would have been 7.5% at term, 7.9% at 41 weeks, and 9.9% at 42 weeks compared with 2.7%, 3.5%, and 3.5% having "need for ventilation." The major concern in postterm pregnancy remains the state of the fetus before the onset oflabor. Problems that develop during labor can usually be effectively managed. The paucity of postterm antepartum deaths in the present study in the absence of a program of
antepartum fetal monitoring is reassuring and consistent with the findings reported over the years from this institution of a relative absence of increased risk with postterm pregnancy. 17"19 The excess deaths associated with postterm delivery in the literature are not related to unexplained asphyxia in utero but rather to a general increase in deaths from all causes. 20 To ascertain whether a careful analysis of cause of death in a large number of term and postterm births would indicate an asphyxic or other risk of death postterm, perinatal deaths from a recent population of births from a whole province were reviewed by one of the authors (R. U.) (Table V)! 1 The cause of death was established from questionnaires completed for each of the deaths, including autopsy reports. Gestational age from last normal menstrual period was known in 86% of cases. The distribution of gestational ages of the population from which these deaths were derived was estimated from figures previously obtained for gestational age of all births in the province. The analysis shows that of all perinatal deaths at or after 37 weeks, the proportion dying postterm is less than the proportion of total births postterm (7.6% of deaths and 13.6% of births). Perinatal death was estimated at 4.39 at term and 2.47 postterm per 1000 births. There was no tendency for deaths from longterm deprivation of oxygen (i.e., unexplained fetal deaths, deaths from fetal malnutrition, or unexplained birth asphyxia) to occur disproportionately often postterm. Using the estimated number of births by gestational age, fetal malnutrition deaths occurred in I : 2298 pregnancies at term and in I : 10,323 postterm, where-
264 Usher et al.
as unexplained antepartum fetal death occurred in I : 1473 pregnancies at term and I : 2949 postterm. * It is possible that the high risk previously attributed to postterm birth is related not to those who are truly postterm, but to those whose menstrual histories are misleading. The incidence of postterm birth in most of the original studies was 10% to I2% for 42-week births and 3% to 4% at 43 or more weeks of all pregnancies. 3 This contrasts sharply with the findings of the present study in which gestational age was very precisely confirmed, in which only 340 of 7663 or 4.4% of pregnancies reaching 39 weeks exceeded 293 days of gestation. The excess of "postterm" cases in less welldefined populations includes women who have mistaken their dates (often of low socioeconomic class) and women with very irregular cycles. 5 Both categories are likely to be at higher risk of perinatal death, which would have been falsely attributed to postmaturity. 22 The fetus in a postterm pregn'!ncy cannot be found to be at higher risk than at term before the onset of labor. The intensive antepartum fetal monito.ring of postdate pregnancies currently advocated by many authors is unlikely to be of more value than a similar program would be at term. There are, however, psychological and medicolegal reasons why postterm pregnancies will continue to engender anxiety. Even if it is true that the risks of fetal death and birth asphyxia are not increased postterm, as the results reported here would indicate, the loss of a fetus after the due date is more difficult to accept than one occurring earlier. The natural response is always to think of how unfortunate it is that the infant was not delivered earlier ot followed more closely. No amount of scientific fact will change this reaction. *These calculations incorporate postterm births with term births when calculating the population at risk of fetal loss at term, since postterm pregnancies are also at risk of fetal death at term.
REFERENCES 1. Gruenwald P. The fetus in prolonged pregnancy. AM 1 0BSTET GYNECOL 1964;89:503-9.
February 1988 Am J Obstet Gynecol
2. Evans T, Koeff S, Morley G. Fetal effects of prolonged pregnancy. AM1 0BSTET GYNECOL 1963;85:701-12. 3. Nesbitt R. Prolongation of pregnancy. A review. Obstet Gynecol Surv 1955;10:311-62. 4. Ballantyne 1W. The problem of the postmature infant. 1 Obstet Gynecol Br Emp 1902;2:521-54. 5. Zwerdling MA. Factors pertaining to prolonged pregnancy and its outcome. Pediatrics 1967;40:202-12. 6. Gibberd GF. The choice between death from postmaturity and death from induction of labor. Lancet 1958;1:64-6. 7. Boyd ME, Usher RH, McLean FH, Kramer MS. Obstetric consequences of postmaturity. AM 1 0BSTET GYNECOL 1988; 158:334-8. 8. Steer PJ. Commentary: postmaturity-much ado about nothing? Br 1 Obstet Gynaecol1986;93:105-8. 9. Smith LP, deLeon A, Funnell WR1, Lalonde AB, McLean FH, Usher RH. A research-oriented system for McGill obstetrical and neonatal data (MONO). Acta Obstet Gynecol Scand [Suppl]1982;109:49-50. 10. Miettinen OS. Simple interval estimation of risk ratio. Am 1 Epidemiol1974;100:515-6. 11. Armitage P. Statistical methods in medical research. Oxford: Blackwell Scientific Publications, 1971:353-68. 12. Eden RD, Seifert LS, Winnegar A, Spellacy WN. Perinatal characteristics of uncomplicated postdate pregnancies. Obstet Gynecol 1987;69:296-9. 13. Green 1N, Paul RH. The value of amniocentesis in prolonged pregnancy. Obstet Gynecol 1978;51 :293-8. 14. Klapholz H, Friedman E. The incidence of intrapartum fetal distress with advancing gestational age. AM 1 OBSTET GYNECOL 1977;127:405-7. 15. Gregory GA, Goodling CA, Phibbs H, Tooley WH. Meconium aspiration in infants. A prospective study. 1 Pediatr 1974;85:848-52. 16. Cyr RM, Usher RH, McLean FH. Changing patterns of birth asphyxia and trauma over 20 years. AM 1 OBSTET GYNECOL 1984;148:490-8. 17. Usher RH. Clinical implications of perinatal. mortality statistics. Clin Obstet Gynecol 1971; 14:885-925. 18. O'Brien 1R, Usher RH, Maughan GB. Causes of birth asphyxia and trauma. Can Med Assoc 1 1966;94: 1077-85. 19. Scott KE, Usher R. Fetal malnutrition: its incidence, causes, and effects. AM 1 0BSTET GYNECOL i 966;94: 951-63. 20. Naeye RL. Causes of perinatal mortality excess in prolonged gestations. Am 1 Epidemiol 1978; 108:429-33. 21. Brault R. Reports of perinatal mortality 1980-81 and 1981-82, and analysis of data collected on medical certificates of childbirth, 1973, Perinatal Mortality Committee of Quebec (unpublished). 22. Dewhurst CJ, Beazley1M, Campbell S. Assessment of fetal maturity and dysmaturity. AM 1 0BSTET GYNECOL 1972; 113:141-9.
Key words: Postterm, fetal distress, birth asphyxia, meconium, perinatal mortality
There is a historic concern for fetal well-being in the postterm period. 1-6 This concern is based on studies in which fetal maturity was established by clinical means. With ultrasound examination done early in pregnancy, there is reason to believe that a diagnosis of postterm birth was mistaken in some 70% of cases. 7 Fetal hazard must now be reevaluated in this light. In this article fetal outcomes in the postterm period are compared with those of term infants. Fetal age was determined in all fetuses by last normal menstrual period and confirmed by early ultrasound examination. It was hoped that an answer could be given to the question posed in the title of a recent article, "Postmaturity-much ado about nothing?". 8 This study was one of several conducted on the same population base; the others dealt with obstetric considerations and fetal growth in postterm pregnancy. Methods This study was carried out on patients cared for and delivered of infants at the Royal Victoria Hospital, Montreal, and on the Obstetrical Teaching Service of McGill University between jan. 1, 1978, and March 31, 1986. With few exceptions, patients received antenatal care from their obstetrician starting in the first trimester and had a dating ultrasound measurement of hi-
From the Departments of Obstetrics and Gynecology, Pediatrics, and Epidemiology and Biostatistics, McGill University Faculty of Medicine, and the Royal Victoria Hospital. Presented at the Forty-third Annual Meeting of The Society of Obstetricians and Gynaecologists of Canada, Ottawa, Ontario, Canada, june 24-27, 1987. Reprint requests: Robert H. Vsher, MD, Neonatal Unit, Women's Pavilion, Royal Victoria Hospital, 687 Pine Avenue West, Montreal, Quebec HJA JAJ, Canada.
parietal diameter performed, usually between 16 and 18 weeks. All births meeting the following criteria were included in the study: (1) The date of the last normal menstrual period was recorded; (2) there was a record of an early ultrasound dating examination; (3) gestational age calculated from early ultrasound examination was concordant within 7 days with that calculated from menstrual history; and (4) delivery occurred at or after 273 days (39 completed weeks) from the last normal menstrual period. Of the 15,226 births during the period for which both ultrasound and menstrual data were recorded, 10930 births were concordant within 7 days. The study population included the 7663 infants delivered at or after 273 days and was divided according to gestational age based on last normal menstrual period into three groups: (I) 5915 infants delivered at 39 or 40 completed weeks (273 to 286 days); (2) 1408 infants delivered at 41 completed weeks (287 to 293 days); and (3) 340 infants delivered at 42 completed weeks (294 days) or more. Although high-risk pregnancies at this hospital receive frequent nonstress tests and ultrasound evaluations in the hospital's antenatal center, there is no routine testing of postterm patients, and those reported here were, in fact, seldom monitored with either technique. Inductions of labor were performed in 17% of patients delivered at 39 and 40 weeks, 24% of patients delivered at 41 weeks, and 44% of patients delivered at 42 weeks or later. Deliveries were performed by or in the presence of an attending obstetrician with residents, interns, and students participating. A diagnosis of fetal distress from abnormal fetal heart tracings during labor (88% of labors were electronically monitored) was made when late or variable
259
260 Usher et al.
February 1988 Am J Obstet Gynecol
Table I. Incidence of fetal distress and meconium in amniotic fluid (per 100 births) Gestational age (wk)
39
No. of births Fetal distress Meconium Combinations No fetal distress or meconium Fetal distress alone Meconium alone Fetal distress + meconium Total
+ 40
RR* (95% Cl)
5915 15.9 15.3
1408 21.9 27.0
73.3 11.4 10.9 4.5 100.1
59.4 13.6 18.8 8.2 100.0
1.38 (1.23-1.55) 1.76 (1.58-1.96) 0.81 1.20 1.73 1.84
(0.78-0.84) ( 1.03-1.39) (1.51-1.97) ( 1.49-2.27)
340 29.7 31.5 54.1 14.4 16.2 15.3 100.0
RR (95% CI)
Probabilityf
1.87 (1.56-2.25) 2.05 (1.72-2.46)
<0.001 <0.001
0.74 1.27 1.49 3.41
<0.001 <0.01 <0.001 <0.001
(0.68-0.80) (0.96-1.66) (1.15-1.93) (2.60-4.48)
RR = Relative risks; CI = confidence interval. *Relative risks are in comparison with deliveries at 39 and 40 weeks. tProbability is based on x' for linear trend analysis.
Table II. Fetal morbidity (rates per 1000 live births) Gestational age (wk)
39
Number of live births Depression at birth Moderate Severe Postasphyxic encephalopathy Meconium aspiration Fractures/palsies
+ 40
5906 23.4 3.7 0.7 2.2 10.7
RR* (95% Cl) 1407 31.3 3.6 1.4 5.7 14.2
RR (95% Cl)
Probabilityf
340 1.34 0.95 2.10 2.58 1.33
(0.96-1.87) (0.36-2.51) (0.40-11.02) (1.11-6.03) (0.81-2.91)
26.5 8.8 2.9 17.6 29.4
1.13 2.37 4.34 8.02 2.76
(0.58-2.20) (0.74-7.61) (0.58-32.26) (3.56-18.04) (1.46-5.21)
NS NS NS <0.001 <0.01
RR = Relative risks; CI = confidence interval. *Relative risks are in comparison with deliveries at 39 and 40 weeks. tProbability is based on x' for linear trend analysis.
decelerations or other abnormal fetal heart patterns were noted by the obstetric staff. Fetal heart pattern and meconium staining of amniotic fluid were recorded on a checklist on the obstetric chart after each delivery by physicians attending the birth and have not been reviewed further. All neonatal records were reviewed after discharge by R. U. for standardization and confirmation of diagnoses. Depression at birth was defined as the need for positive pressure bag and mask ventilation to initiate respirations, moderate if for 3 minutes or less, and severe if for more than 3 minutes. Postasphyxic encephalopathy included abnormalities in tone or neurologic state, including convulsions in the most severe infants, which occurred during the first days of life after perinatal asphyxic insult. Meconium aspiration syndrome was diagnosed when any degree of respiratory distress and abnormal radiographic findings followed delivery and when meconium was present in amniotic fluid. Birth trauma included fractures (clavicular) or paralyses (brachial or facial), the latter of which often resolved quickly. The McGill Obstetrical and Neonatal Data Base9 was used to computerize the data and perform the analyses.
Proportions were compared with the x' test. Confidence intervals around the reported relative risks are based on the test-based method ofMiettinen. 10 Tests for linear trends in proportion used the x' test described by Armitage.11
Results Fetal distress and meconium release occurred more commonly postterm (Table 1). The rates of both approximately doubled among deliveries at or after 42 weeks compared with deliveries occurring at term. Relative risks were intermediate at 41 weeks. Fetal distress developed in combination with meconium release in 4.5% of term deliveries and in 15.3% of deliveries at or after 42 weeks. There was no significant increase in depression at birth postterm either of moderate or severe degree or of postasphyxic encephalopathy (Table II). Signs of the latter developed in only three of 1747 infants delivered at or after 41 weeks. There was a progressive (eightfold) increase in meconium aspiration, which occurred in 2.2/1000 births or one infant in every 455 delivered at term, one in
Fetal risks in postdate pregnancies 261
Volume 158 Number 2
Fetal 01stress
Mecon1um
40 i
30 1-
z
u.J
~
20
u.J
0...
10
~· .................. .. ··'·''''''·······• ,
8/8
o~---r-----,,-----~
Oepress1on at B1rth
Fracture or Palsy
8
•
6
•-•
Pr~tliiP'
............
Pr11111p .1[)[)[)q ·
o"'"'"''o
M1111q1 .1[)[1(1;1
~(J()(lq
5 1-
z
u.J
'--'
a:
·~······0 .. :: ......
4
u.J
~-
~,. ..... o
a
0...
•
3
...
~0 2
, 0:,.,,,
··········· f
0
.... ,,,,,,,,o
e•"''"'""'""•
o~---r-----,,-----~
39 40
42
41
39 40
41
42
GESTATIONAL AGE (weeksl
Fig. 1. Rates of fetal distress, meconium release, depression at birth, and birth trauma for infants at 39 and 40 weeks, 41 weeks, and 42 + completed weeks of gestation. Effects of parity and fetal size are demonstrated.
175 at 41 weeks, and one in 57 by 42 weeks (Table II). These cases were, however, relatively mild with no deaths, no need for mechanical ventilation, and only 56% of the affected infants requiring oxygen therapy. Meconium, when released, was more often thick postterm: 37.4% at 39 and 40 weeks, 44.5% at 41 weeks, and 51.4% at 42 weeks among those with meconiumstained amniotic fluid. This 1.4-fold increase in the thickness of meconium postterm was insufficient to account for the observed fourfold increase in meconium aspiration syndrome among postterm-exposed infants. Fractures and paralyses were also more frequent postterm; the risk increased from one infant with birth trauma in every 93 at term to one in 34 at 42 weeks (Table II). In this study postterm pregnancies were more often primiparous: 4 7% of deliveries at 39 and 40 weeks, 60% at 41 weeks, and 66% at 42 weeks or later. Postterm
fetuses were also more often macrosomic, with birth weight of 4000 gm or greater in 23% of primiparas and 40% of multiparas at 42 weeks or later compared with 9% and 14%, respectively, at term. Because both primiparity and macrosomia are associated with an increased risk of prolonged labor and difficult delivery, it was necessary to reassess the role of postterm labor and delivery on fetal outcome after controlling for parity and fetal size. With these variables controlled, fetal distress and meconium release continued to show an increased risk postterm, whereas depression at birth and birth trauma showed no relationship to gestational age (Fig. 1). Fetal distress and meconium release increased postterm, yet without the expected parallel increase in depression at birth. The relationship between these signs of fetal jeopardy during labor and the condition of the infant at birth was therefore analyzed (Table III).
262
Usher et al.
February 1988 Am J Obstet Gynecol
Table III. Incidence of depression at birth as related to fetal distress and meconium release(%)
Table IV. Perinatal mortality (per 1000 births) Gestational age (wk)
39 + 40
Gestational age (wk)
42
39 and 40
No fetal distress or meconium Fetal distress and/or meconium
1.9% (82/4335) 5.3% (84/1580)
2.2% (18/836) 5.6% (32/572)
0% (0/184) 8.3% (13/156)
Fetal distress and meconium release were associated with an increased risk of depression at birth at all gestational ages. Perinatal death was not increased in postterm pregnancy compared with patients delivered of infants at term, although the overall number of deaths (19, eight of which were from malformations) was too .few for meaningful analysis (Table IV). Of the four infants who died at least I week postterm, two had lethal malformations, a third weighing 2150 gm died at 41 weeks before labor of unsuspected fetal malnutrition in an obese woman, and the fourth weighing 5640 gm died of birth asphyxia sustained during a difficult delivery complicated by severe shoulder dystocia at 42 weeks. These latter two cases were the only nonanomaly deaths among the 1748 pregnancies that had continued at least 7 days postterm.
Comment The purpose of this retrospective study was to ascertain whether there was an increased risk to the fetus carried postterm. If so, what was the nature of this risk? To what degree could long-term deprivation of oxygen supply to the fetus be demonstrated? What was the significance and role played by meconium release and aspiration? Was dystocia still a concern in the modern postterm delivery? Most of the emphasis in the study was placed on fetal and neonatal morbidity related to fetal distress, birth asphyxia, trauma, and meconium aspiration. These complications represent the concerns specifically related to postdate pregnancy. Perinatal death at and after term is now so rare that very large populations are needed to reflect an increased risk postterm, and many of the deaths that occur postterm are from causes such as malformations that cannot be attributed to the duration of pregnancy. This study shows that fetal distress and meconium release develop more commonly postterm than at term, with intermediate rates at 41 completed weeks. Although meconium release has regularly been reported
No. of births Fetal deaths Total Nonanomaly Neonatal deaths Total Nonanomaly Perinatal deaths Total Nonanomaly
5915
42+
1408
340
1.5 1.5
0.7 0.7
0 0
1.0 0.0
1.4 0
3.0 3.0
2.5 1.5
2.1 0.7
3.0 3.0
as much more frequent postterm, 12 " 13 fetal distress has been less clearly recognized as a risk. 14 Meconium aspiration syndrome is a problem to which the postterm infant is particularly predisposed. 15 This study shows a twofold increase in meconium release and an eightfold increase in frequency of meconium aspiration syndrome postterm. Expressing the frequency of meconium aspiration syndrome as a proportion of infants at risk (those with meconium in amniotic fluid), this rate rose from 14% (13/90) at term to 21% (eight/38) at 41 weeks, and 55% (six/11) postterm (Tables I and II). That is to say that not only is meconium present more often in amniotic fluid postterm, but when present, is more often aspirated. There was no concomitant increase in frequency of birth asphyxia, at least as reflected by depression at birth requiring manual ventilation. Although the number of depressed infants at 42 weeks or more is too small to speak with confidence about the lack of risk postterm, the number at 41 weeks is large enough and should represent some degree of postterm asphyxic risk if such were present. We are left with an apparent inconsistency: more fetal jeopardy in utero without a concomitant increase in depression at birth. One explanation might be that fetal distress and meconium may have less of a predictive value for birth asphyxia at term, but this could not be demonstrated (Table III). Rather, the reason for the inconsistency is that a comparatively large number of infants (82/166) were depressed after delivery at term but did not show fetal distress or meconium release during labor, whereas this did not occur postterm (0113). When present, fetal distress and meconium have similar ominous significance at term as postterm, but their absence is much more reassuring postterm. A hypothesis can be developed to explain these findings. The more mature postterm fetus may be more responsive and react more readily to levels of asphyxic stimuli that may not initiate a response at term. These
Fetal risks in postdate pregnancies
Volume 158 Number 2
263
Table V. Causes of perinatal death at term and postterm, Province of Quebec, April I, 1980 to March 31, 1982 21 Gestational age (complete wk*)
Estimated No. of birthst No. of fetal deaths No. of neonatal deaths No. of deaths by cause Malformation Fetal malnutrition Infection Rh hemolytic disease Maternal diseases Abruptio, placenta previa Dystocia, malpresentation Birth asphyxia, other Other known causes Unexplained fetal death antepartum Unexplained fetal death intrapatrum Unexplained neonatal death Total
37-41
>42
Total
% postterm
131,085 352 271
20646 27 24
151,731 379 295
13.6 7.1 8.1
219 66 13 2 20 36 21 24 100 103 15 4
22 2 2 0 0 2 4 2 8 7 2 0
241 68 15 2 20 38 25 26 108 110 17 4
~
---sT
-----m4
9.1 2.9 13.3 0.0 0.0 5.3 16.0 7.7 7.4 6.4 11.8 0.0 7.6
*At time of delivery or at time of fetal death if different. tBased on a previous analysis of gestational age distribution of births in the Province of Quebec.
reactions are manifested by fetal heart abnormalities, meconium release, and gasping movements. The less responsive term fetus may be sufficiently asphyxiated to be depressed at birth without affecting the heart rate during labor or causing the release of meconium. The fetus at term may also pass meconium in utero and yet not be responsive enough to gasp and aspirate it. These interpretations would account for why fetal distress, meconium release, and meconium aspiration occur more often postterm and yet apparently do not indicate that the postterm infant is more often asphyxiated (as evidenced by depression at birth) than those delivered at term. Furthermore, little serious morbidity is associated with perinatal asphyxia postterm. None of the infants with meconium aspiration required ventilatory assistance, and none of the depressed infants demonstrated postasphyxic neurologic sequelae on follow-up examination. The criterion for birth asphyxia-need for ventilation-used in this study has previously been found more specific and meaningful than Apgar scores, which is the index often used. 16 Had the definition of birth asphyxia instead been a !-minute Apgar score <7, the incidence would have been 7.5% at term, 7.9% at 41 weeks, and 9.9% at 42 weeks compared with 2.7%, 3.5%, and 3.5% having "need for ventilation." The major concern in postterm pregnancy remains the state of the fetus before the onset oflabor. Problems that develop during labor can usually be effectively managed. The paucity of postterm antepartum deaths in the present study in the absence of a program of
antepartum fetal monitoring is reassuring and consistent with the findings reported over the years from this institution of a relative absence of increased risk with postterm pregnancy. 17"19 The excess deaths associated with postterm delivery in the literature are not related to unexplained asphyxia in utero but rather to a general increase in deaths from all causes. 20 To ascertain whether a careful analysis of cause of death in a large number of term and postterm births would indicate an asphyxic or other risk of death postterm, perinatal deaths from a recent population of births from a whole province were reviewed by one of the authors (R. U.) (Table V)! 1 The cause of death was established from questionnaires completed for each of the deaths, including autopsy reports. Gestational age from last normal menstrual period was known in 86% of cases. The distribution of gestational ages of the population from which these deaths were derived was estimated from figures previously obtained for gestational age of all births in the province. The analysis shows that of all perinatal deaths at or after 37 weeks, the proportion dying postterm is less than the proportion of total births postterm (7.6% of deaths and 13.6% of births). Perinatal death was estimated at 4.39 at term and 2.47 postterm per 1000 births. There was no tendency for deaths from longterm deprivation of oxygen (i.e., unexplained fetal deaths, deaths from fetal malnutrition, or unexplained birth asphyxia) to occur disproportionately often postterm. Using the estimated number of births by gestational age, fetal malnutrition deaths occurred in I : 2298 pregnancies at term and in I : 10,323 postterm, where-
264 Usher et al.
as unexplained antepartum fetal death occurred in I : 1473 pregnancies at term and I : 2949 postterm. * It is possible that the high risk previously attributed to postterm birth is related not to those who are truly postterm, but to those whose menstrual histories are misleading. The incidence of postterm birth in most of the original studies was 10% to I2% for 42-week births and 3% to 4% at 43 or more weeks of all pregnancies. 3 This contrasts sharply with the findings of the present study in which gestational age was very precisely confirmed, in which only 340 of 7663 or 4.4% of pregnancies reaching 39 weeks exceeded 293 days of gestation. The excess of "postterm" cases in less welldefined populations includes women who have mistaken their dates (often of low socioeconomic class) and women with very irregular cycles. 5 Both categories are likely to be at higher risk of perinatal death, which would have been falsely attributed to postmaturity. 22 The fetus in a postterm pregn'!ncy cannot be found to be at higher risk than at term before the onset of labor. The intensive antepartum fetal monito.ring of postdate pregnancies currently advocated by many authors is unlikely to be of more value than a similar program would be at term. There are, however, psychological and medicolegal reasons why postterm pregnancies will continue to engender anxiety. Even if it is true that the risks of fetal death and birth asphyxia are not increased postterm, as the results reported here would indicate, the loss of a fetus after the due date is more difficult to accept than one occurring earlier. The natural response is always to think of how unfortunate it is that the infant was not delivered earlier ot followed more closely. No amount of scientific fact will change this reaction. *These calculations incorporate postterm births with term births when calculating the population at risk of fetal loss at term, since postterm pregnancies are also at risk of fetal death at term.
REFERENCES 1. Gruenwald P. The fetus in prolonged pregnancy. AM 1 0BSTET GYNECOL 1964;89:503-9.
February 1988 Am J Obstet Gynecol
2. Evans T, Koeff S, Morley G. Fetal effects of prolonged pregnancy. AM1 0BSTET GYNECOL 1963;85:701-12. 3. Nesbitt R. Prolongation of pregnancy. A review. Obstet Gynecol Surv 1955;10:311-62. 4. Ballantyne 1W. The problem of the postmature infant. 1 Obstet Gynecol Br Emp 1902;2:521-54. 5. Zwerdling MA. Factors pertaining to prolonged pregnancy and its outcome. Pediatrics 1967;40:202-12. 6. Gibberd GF. The choice between death from postmaturity and death from induction of labor. Lancet 1958;1:64-6. 7. Boyd ME, Usher RH, McLean FH, Kramer MS. Obstetric consequences of postmaturity. AM 1 0BSTET GYNECOL 1988; 158:334-8. 8. Steer PJ. Commentary: postmaturity-much ado about nothing? Br 1 Obstet Gynaecol1986;93:105-8. 9. Smith LP, deLeon A, Funnell WR1, Lalonde AB, McLean FH, Usher RH. A research-oriented system for McGill obstetrical and neonatal data (MONO). Acta Obstet Gynecol Scand [Suppl]1982;109:49-50. 10. Miettinen OS. Simple interval estimation of risk ratio. Am 1 Epidemiol1974;100:515-6. 11. Armitage P. Statistical methods in medical research. Oxford: Blackwell Scientific Publications, 1971:353-68. 12. Eden RD, Seifert LS, Winnegar A, Spellacy WN. Perinatal characteristics of uncomplicated postdate pregnancies. Obstet Gynecol 1987;69:296-9. 13. Green 1N, Paul RH. The value of amniocentesis in prolonged pregnancy. Obstet Gynecol 1978;51 :293-8. 14. Klapholz H, Friedman E. The incidence of intrapartum fetal distress with advancing gestational age. AM 1 OBSTET GYNECOL 1977;127:405-7. 15. Gregory GA, Goodling CA, Phibbs H, Tooley WH. Meconium aspiration in infants. A prospective study. 1 Pediatr 1974;85:848-52. 16. Cyr RM, Usher RH, McLean FH. Changing patterns of birth asphyxia and trauma over 20 years. AM 1 OBSTET GYNECOL 1984;148:490-8. 17. Usher RH. Clinical implications of perinatal. mortality statistics. Clin Obstet Gynecol 1971; 14:885-925. 18. O'Brien 1R, Usher RH, Maughan GB. Causes of birth asphyxia and trauma. Can Med Assoc 1 1966;94: 1077-85. 19. Scott KE, Usher R. Fetal malnutrition: its incidence, causes, and effects. AM 1 0BSTET GYNECOL i 966;94: 951-63. 20. Naeye RL. Causes of perinatal mortality excess in prolonged gestations. Am 1 Epidemiol 1978; 108:429-33. 21. Brault R. Reports of perinatal mortality 1980-81 and 1981-82, and analysis of data collected on medical certificates of childbirth, 1973, Perinatal Mortality Committee of Quebec (unpublished). 22. Dewhurst CJ, Beazley1M, Campbell S. Assessment of fetal maturity and dysmaturity. AM 1 0BSTET GYNECOL 1972; 113:141-9.