- Email: [email protected]
Unexplained elevated maternal serum α-fetoprotein levels and perinatal outcome in an urban clinic population
October 1994
Brazerol, Grover, and Donnenfeld
9. 10.
11. 12.
aspiration in labor using amnioinfusion. Obstet Gynecol 1989;73:647-51. Sadovsky Y, Amon E, Bade ME, Petrie RH. Prophylactic amnioinfusion during labor complicated by meconium: a preliminary report. AM J OBSTET GYNECOL 1989; 161 :613-7. Macri CJ, Schrimmer DB, Leung A, Greenspoon JS, Paul RH. Prophylactic amnioinfusion improves outcome of pregnancy complicated by thick meconium and oligohydramnios. AM J OBSTET GYNECOL 1992;167:117-21. Uhing MR, Bhat R, Philobos M, Raju TN. Value of amnioinfusion in reducing meconium aspiration syndrome. Am J Perinatol 1993;10:43-5. Wen TS, Eriksen NL, Blanco JD, Graham JM, Oshiro BT,
Am J Ohstet Gynecol
13. 14. 15. 16.
Prieto JA. Association of clinical intra-amniotic infection and meconium. Am J Perinatol 1993; 10:438-40. Miyazaki FS, Nevarez F. Saline amnioinfusion for relief of repetitive variable decelerations: a prospective randomized study. AM J OBSTET GYNECOL 1985;153:301-6. Meis PJ, Hall M, Marshall JR, Hobel CJ. Meconium passage: a new classification for risk assessment during labor. AM J OBSTET GYNECOL 1978;131:509-13. Katz VL, Bowes WA. Meconium aspiration syndrome: reflections on a murky subject. AM J OBSTET GVNECOL 1992;166:171-83. Jovanovic R, Nguyen HT. Experimental meconium aspiration in guinea pigs. Obstet Gynecol 1989;73:652-6.
Unexplained elevated maternal serum a-fetoprotein levels and perinatal outcome in an urban clinic population William F. Brazerol, MD, Steven Grover, MD, and Alan E. Donnenfeld, MD Philadelphia, Pennsylvania OBJECTIVE: Our purpose was to determine whether obstetric patients with unexplained elevated maternal serum a-fetoprotein levels from an indigent clinic population are at increased risk for adverse perinatal outcome compared with similar patients with normal values. STUDY DESIGN: Perinatal outcomes from inner-city obstetric patients with unexplained elevated maternal serum a-fetoprotein levels (> 2.0 multiples of the median) were compared with patients from the same clinic with normal values. The frequency of adverse outcomes in the two groups was subjected to x2 analysis. RESULTS: Adverse perinatal outcomes occurred in 33 of 57 (58%) of the subjects with unexplained elevated maternal serum a-fetoprotein levels compared with 163 of 719 (23%) patients with normal values (p < 0.001). Statistically significant differences were observed for abruptio placentae (p < 0.025), intrauterine growth retardation (p < 0.025), stillbirth at >20 weeks (p < 0.001), and pregnancy-induced hypertension (p < 0.01). Differences in the frequencies of preterm premature rupture of membranes, preterm delivery, pregnancy loss < 20 weeks, and congenital malformations were not statistically significant. CONCLUSION: In contrast to a previous report, we found that unexplained elevated maternal serum a-fetoprotein levels confer an increased risk of adverse perinatal outcome in an urban clinic population over and above the already increased risk related to socioeconomic status. (AM J OaSlET GYNECOL 1994;171 :1030-5.)
Key words: Unexplained elevated maternal serum a-fetoprotein levels, adverse perinatal outcome a-Fetoprotein (AFP) was first identified in 1956,1 and in 1973 elevated amniotic fluid levels were initially used clinically to diagnose anencephaly.2 Subsequently, elFrom the Section of Genetics, Department of Obstetrics and Gynecology, Pennsylvania Hospital. Presented at the Fourteenth Annual Meeting of the Society of Perinatal Obstetricians, Las Vegas, Nevada, January 24-29, 1994. Reprint requests: Alan E. DonnenfeId, MD, Section of Genetics, Department of Obstetrics and" Gynecology, Pennsylvania Hospital, 800 Spruce St., Philadelphia, PA 19107. Copyright © 1994 by Mosby-Year Book, Inc. 0002-9378/94 $3.00 + 0 6/6/57824
1030
evated matenal serum values became the standard of care as a screening test for neural tube defects. In 1984 low maternal serum AFP values were found to be associated with fetal aneuploidy." As a result of two decades of clinical investigation, maternal serum AFP has become an indispensible screening test that aids in the identification of pregnant women from the general obstetric population who are at risk for a number of disorders, including open neural tube defects, abdominal wall defects, and chromosomal abnormalities. For these reasons, offering maternal serum AFP screening
Volume 171, Number 4 Am ] Obstet Gynecol
to all pregnant women has been recommended! Abnormally elevated values are routinely followed by ultrasonographic examination to confirm gestational age, establish viability, rule out multiple gestation, or identity another cause for the elevated level. In approximately 5% of patients ultrasonographic evaluation will identity an open neural tube or ventral wall defect. In those patients whose ultrasonography confirms dates and no explanation for the elevated maternal serum AFP level is found, many patients will opt to undergo amniocentesis, Most will have a normal amniotic fluid AFP in spite of the elevated maternal serum AFP. This group comprises those with unexplained elevated maternal serum AFP levels, representing approximately 2% to 3% of the population undergoing maternal serum AFP screening. Initially this group of patients had been reassured that there was no evidence of an open fetal defect, indicating a favorable pregnancy outlook. In the past several years a number of investigators have demonstrated an increased risk of adverse perinatal outcome associated with unexplained maternal serum AFP levels, including spontaneous abortion, stillbirth, prematurity, intrauterine growth restriction (IUCR), abruptio placentae, premature rupture of membranes, pregnancy-induced hypertension, preeclampsia, and preterm labor. 5-9 The rate of adverse perinatal outcome has been reported to range from 16% to 47% and has been found to be directly correlated with the degree of elevation of maternal serum AFP. 10 Most previous studies have examined the perinatal outcome of patients with unexplained elevated maternal serum AFP levels in private obstetric populations at low risk. A single study from Memphis has evaluated the implications of an unexplained elevated test result in an indigent urban population and compared the perinatal outcome with matched-paired controls from the same clinic with similar obstetric risk factors.ll The authors found that unexplained elevated levels offered little, if any, predictive value for adverse perinatal outcome in populations already at high risk on the basis of obstetric or socioeconomic criteria. The purpose of this study was to perform a similar investigation of indigent urban patients from Philadelphia in a historic cohort analysis with independent groups to determine whether unexplained elevated maternal serum AFP levels identity a subset of clinical patients at further increased risk for adverse perinatal outcome above that related to obstetric or socioeconomic status. Methods
Patient data from women attending the Pennsylvania Hospital prenatal clinic in Philadelphia from January 1988 to June 1992 were analyzed. All patients were offered maternal serum AFP screening between 15 and 20 weeks' gestation, usually in association with simulta-
Brazerol, Grover, and Donnenfeld
1031
neous gestational dating ultrasonography to avoid false-negative and false-positive results because of inaccurate gestational dating. A total of 2764 women had maternal serum AFP screening during this period, with a 3.2% screen-positive rate. An elevated maternal serum AFP level was defined as ;;::: 2.0 multiples of the median adjusted for race, weight, and diabetes mellitus. An abnormal test was repeated in 1 week (unless the patient was past 18 weeks or the multiples of the median value was;;::: 3.0) and a gestational dating ultrasonography was obtained, if not previously performed. If two values were;;::: 2.0 multiples of the median, patients were offered genetic counseling, diagnostic ultrasonography and amniocentesis for amniotic fluid AFP level, chromosome analysis, and, when applicable, acetylcholinesterase level. If amniocentesis was declined, high-resolution ultrasonography was performed. An elevated maternal serum AFP level was considered unexplained if, after erroneous gestational dating (i.e., a ;;::: 10-day discrepancy between menstrual and ultrasonographic dating was corrected for, no evidence of a multiple gestation, fetal anomaly, oligohydramnios, or fetal death was detected on diagnostic ultrasonographic evaluation. Patients with unexplained elevated levels were offered weekly nons tress testing, with biophysical profile follow-up if nonreactive, and monthly Doppler ultrasonography starting at 28 to 32 weeks' gestation. Adverse perinatal outcomes were defined as spontaneous loss at < 20 weeks, IUCR, preterm premature rupture of membranes, premature delivery at < 37 weeks, pregnancy-induced hypertension, abruptio placentae, congenital malformations other than neural tube and ventral wall defects, and stillbirth at ;;::: 20 weeks. Patients with second-trimester oligohydramnios were excluded from the unexplained elevated maternal serum AFP group, both because this finding is a known cause of elevated levels and because this was the protocol used in the Memphis study. II , 12 Patients used as controls for this study were obtained from perinatal outcome data for patients attending the same prenatal clinic during 1990 and 1991. All clinic patients with normal maternal serum AFP values during these 2 years were included. Statistical analysis was performed by X2 testing. Results
Among the 2764 patients who underwent maternal serum AFP screening between January 1988 and June 1992, 89 (3.2%) had elevated values ;;::: 2.0 multiples of the median; 57 (21.%) of these were unexplained. In nine cases a repeat test was normal, in five patients incorrect gestational dating was identified, in two pregnancies twins were found, and there was one case each of open spina bifida and early severe oligohydramnios.
1032 Brazerol, Grover, and Donnenfeld
October 1994 Am J Obstet Gynecol
Table I. Age, ethnic distribution, and obstetric risk factors of study and control patients Study (n = 57)
Age (mean) 2: 35 yr 18-34 yr < 18 yr Race Black White Hispanic Asian Other or unknown Bleeding, current pregnancy Mean prior pregnancy losses Previous IUGR infant History of prematurity Smoking Prior stillbirth
Control (n = 719)
No.
%
No.
%
23.8 2 51 4
3.5 89.5 7.0
24.4 35 632 52
4.9 87.9 7.2
80.7 lO.5 8.8 0 0 15.8
46 6 5 0 0 9 1.6 4 6 14 1
595 50 54 12 8 119 1.2 44 53 138 11
7.0 lO.5 24.6 1.8
Table II. Comparison of pregnancy outcome between study and control groups by Study (n = 57) Pregnancy outcome
No.
Preterm delivery PPROM Abruptio placentae IUGR SAB <20wk Stillbirth 2: 20 wk PIH Congenital malformations
4 4 4 6 3 4 7 1 33
TOTAL
J
x.
2
82.8 6.9 7.5 1.7 1.1 16.6 6.1 7.4 19.2 1.5
analysis
Control (n = 719)
%
No.
7.0 7.0 7.0 10.5 5.2 7.0 12.3 1.8 57.9
37 22 14 29 12 6 31 12 163
1
%
Significance
5.1 3.1 1.9 4.0 1.7 0.8 4.3 1.7 22.7
NS NS P < 0.025 P < 0.025 NS P < 0.001 P < 0.01 NS P < 0.001
NS, Not significant; PPROM, preterm premature rupture of membranes; SAB, spontaneous abortion; PIH, pregnancy-induced hypertension.
Fourteen patients were lost to follow-up (15.7%). The control group was composed of all 804 clinic patients who underwent screening between January 1990 and January 1992 and who were found to have normal values. Of this group obstetric outcome data were available on 719; 85 (10.6%) were lost to follow-up. The mean maternal age in the study and control groups was 23.8 and 24.4 years, respectively. Ethnic distribution and several possible confounding variables that might affect obstetric outcome were also similar between groups (Table I). . Pregnancy outcome data are summarized in Table II. Adverse perinatal outcome occurred in 57.9% (33/57) of patients with unexplained elevated maternal serum AFP levels compared with 22.7% (163/719) of controls. This difference was statisically significant (p < 0.0001). Furthemore, statistical differences were found between subjects and controls for abruptio placentae (7.0% [four of 57] vs 1.9% [141719], P < 0.025), IUGR (10.5% [six of 57] vs 4.0% [29/719], P < 0.025), stillbirth at
and pregnancy-induced hypertension (12.3% [seven of 57] vs 4.3% [31/719], P < 0.01). Differences that did not reach statistical significance were observed for preterm premature rupture of membranes, preterm delivery, spontaneous abortion at < 20 weeks, and congenital malformations. Although there was no statistically significant difference in deliveries at < 37 weeks between groups, there was a significant difference in deliveries at < 28 weeks (six of 57 ([11.1 %] in those with elevated maternal serum AFP levels vs nine of 719 [1.3%] among controls, p < 0.001). In those patients with an elevated level > 2.5 multiples of the median, 21 of 27 (72.9%) had adverse outcomes compared with 12 of 32 (37.5%) with values between 2.0 and 2.5 multiples of the median. Data on mode of delivery are summarized in Table III. There was a higher cesarean section rate in the study group compared with controls (33.3% vs 24.1%, respectively). However, this difference was not statistically significant. A comparison of the primary cesarean section rate was statistically significant at the p < 0.025
Brazerol, Grover, and Donnenfeld
Volume 171, Number 4 Am J Obstet Gynecol
1033
Table III. Comparison of mode of delivery by X2 analysis Study (n Mode of delivery
No.
Vaginal delivery Cesarean section Primary cesarean section Repeat cesarean section
38 19 16 3
=
Control (n
57)
I
%
No.
66.7 33.3 28.1 5.2
546 173 114 59
=
719)
I
% 75.9 24.1 15.8 8.2
Significance NS NS
P < 0.025 NS
NS, Not significant.
Table IV. Comparison of birth weight by X2 analysis Study (n = 57) Birth weight
No.
;::2500 gm 1500-2499 gm 750-1499 gm
36 9 5
j
Control (n = 719)
%
No.
63.1 15.8 8.8
63 88 4
I
% 83.9 12.2 0.6
Significance
p<
0.005
NS
P < 0.001
NS, Not significant.
level (28.1% in the study group vs 15.8% in controls). Birth weight data are tabulated in Table IV. In the study group 18 of 57 (31.6%) infants weighed <2500 gm versus 116 of 719 (16.1%) controls (JJ < 0.005). Also, five of 57 (8.8%) weighed <750 gm versus four of 719 (0.6%) controls (JJ < 0.001). Comment Our investigation, a historic cohort study using independent groups from an indigent urban clinic population, confirmed the previously reported association of unexplained elevated maternal serum AFP levels with an increased risk of adverse perinatal outcome. Interestingly, this is in contrast to a similar investigation performed in Memphis that found no apparent increased risk in indigent, urban clinic patients with unexplained elevated levels compared with matched controls. II A comparison of perinatal outcome between the Memphis investigation, which used a historic cohort study with matched pairs, and our investigation is found in Table V. Overall, there was a statisticaly significant difference for all adverse perinatal outcomes (57.9% [33/57] in Philadelphia vs 38.9% [28172] in Memphis, p < 0.05). When individual adverse outcome variables were compared, only stillbirth at > 20 weeks was found to be significantly different between the two populations (four of 57 in the Philadelphia group vs none in the Memphis group, p < 0.05). There are several notable differences beween these two investigations. The Memphis group used a small control group that was rigorously matched for 21 different variables defined as unequivocal risk factors for adverse perinatal outcome (Preexisting Risk Guide, Hollister, Inc., Libertyville, Ill.) and maternal age, race, and parity. A possible criticism
of our investigation is that all 21 of these confounding variables for poor obstetric outcome were not assessed because of unavailability of data, although several important factors such as bleeding during the current pregnancy, prior pregnancy losses, previous IVGR, history of premature delivery, smoking, and prior stillbirth were similar between our study and control patients. Surprisingly, the Memphis study did not control for bleeding during pregnancy, an independent risk factor both for adverse perinatal outcome and elevated maternal serum AFP level. 13 This may have introduced a significant bias in their study, especially because of the small number of control patients involved. In addition, it is noteworthy that the Memphis results are, in fact, consistent with a small increase in risk for patients with elevated maternal serum AFP levels when all adverse outcomes are considered (38.9% vs 31.9% in study and control patients, respectively), and it is possible that the lack of statistical significance was related to the small sample size. This is acknowledged in their study. II Alternatively, it is also possible that the results of our study may have been biased by risk factors that were uncontrolled for because of a lack of information (such as akohol and drug abuse) or that the high rate of adverse perinatal outcome in the uDexplained elevated maternal serum AFP level group was related to our relatively small sample size of 57 patients. The finding of a 72.9% adverse perintal outcome rate among pati~nts with maternal serum AFP values > 2.5 multiples of the median is in accordance with several previously published studies that document an increasing adverse perinatal outcome rate with higher values. 10, 14 The Memphis group concluded that unexplained
1034 Brazerol, Grover, and Donnenfeld
October 1994 Am J Obstet Gynecol
Table V. Comparison of adverse perinatal outcomes between Memphis and Philadelphia studies by X2 analysis Philadelphia
Memphis Study (n Adverse perinatal outcome
No.
Pre term delivery PPROM Abruptio placentae IUGR SAB at <20 wk Stillbirth at 2: 20 wk PIH Congenital malformations*
8 6 2 6 4 0 9 2 28
TOTAL
I
=
72)
Control (n
%
No.
ILl 8.3 2.8 8.3 5.6
12 4 2 1 0 0 10 1 23
12.5 2.8 38.9
I
=
72)
Study (n
%
No.
16.7 5.6 2.8 1.4
4 4 4 6 3 4 7 1 33
13.4 1.4 31.9
I
= 57)
Control (n
%
No.
7.0 7.0 7.0 10.5 5.2 7.0 12.3 1.8 57.9
37 22 14 29 12 6 31 12 163
1
=
719)
%
Significance
5.1 3.1 1.9 4.0 1.7 0.8 4.3 1.7 22.7
NS NS P < 0.025 P < 0.025 NS P < 0.001 P < 0.01 NS P < 0.001
NS, Not significant; PPROM, preterm premature rupture of membranes; SAB, spontaneous abortion; PIH, pregnancy-induced hypertension. *Excluding open neural tube and ventral wall defects.
elevated maternal serum AFP levels and adverse perinatal outcome may not apply to all obstetric populations but suggested further studies from institutions similar to theirs. Our study attempted to respond to their request and documented a different result; for indigent urban clinical patients an unexplained elevated maternal serum AFP value confers an increased risk of adverse perinatal outcome over and above the already increased risk related to socioeconomic status. The explanation for the association between elevated maternal serum AFP levels and adverse pregnancy outcome remains elusive, although placental dysfunction, including partial abruptio placentae, pathologic findings, fetomaternal bleeding, and abnormal implantation are often implicated. Additionally, the optimal perinatal management strategy for patients with unexplained elevated levels is not known. IS·" Our population of pregnant women with unexplained elevated maternal serum AFP values were advised to have weekly nonstress tests and monthly Doppler velocimetry studies beginning between 28 and 32 weeks. In spite of this recommend~tion for increased surveillance, an extremely high adverse perinatal outcome rate was identified. In this population, however, there was a high frequency of unkept appointments for perinatal testing that may have contributed to the elevated adverse outcome rate. In addition, early delivery because of nonreassuring fetal testing was performed on several occasions during this investigation. These pregnancies may have thus benefited from the increased surveillance, inasmuch as fetal hypoxia or intrauterine death may have been prevented. However, both prematurity and fetal death were categorized as adverse perinatal outcomes in this study; thus substituting prematurity for stillbirth would not have altered the adverse perinatal outcome rate. It is unfortunate that our study design did not permit analysis of whether increased perinatal surveillance prevents
fetal morbidity and death. It is clear that only a large, prospective, randomized study using a well-defined protocol will answer this question. It is also of note that 11 % of our study patients were delivered before 28 weeks, the earliest that testing was initiated. The finding of adverse perinatal outcome before viability in pregnancies with unexplained elevated maternal serum AFP levels has been previously observed. In fact, one group suggested that prenatal surveillance is not useful in patients with unexplained elevated levels both because the adverse outcomes occur before viability and because the relative risk of preventable fetal death is low. IS Our data do not support this conclusion, because 27 of 33 of the patients with adverse outcomes in the group with unexplained elevated maternal serum AFP levels were delivered after 28 weeks. In spite of the many studies examining perinatal outcome with unexplained elevated maternal serum AFP levels, further work is required to determine the prognostic value of this screening test and the effect of variables such as socioeconomic status on pregnancy outcome. This is necessary to gain insight into the underlying pathophysiologic mechanisms responsible for adverse perinatal outcomes and to assist in identifYing patients at increased risk for IUGR, abruptio placentae, prematurity, preterm premature rupture of membranes, and fetal death. We thank Greg Maislin for his expert statistical assistance and critical review of the study design. REFERENCES
1. Bergstrand CG, Czar B. Demonstration of a new protein fraction in serum from the human fetus. J Clin Lab Invest 1956; 174:8. 2. Seller MJ, Coltart TM, Campbell S, Singer JD. Early termination of anencephalic pregnancy after detection by raised a-fetoprotein levels. Lancet 1973;1:73. 3. Merkatz IR, Nitowsky HM, Macri IN, Johnson WE. An
Cameron et al.
Volume 171, Number 4 Am J Obstet Gynecol
4.
5. 6. 7. 8. 9.
10. 11.
association between low maternal serum a-tetoprotein and fetal chromosome abnormalities. AM ] OBSTET GYNECOL 1984; 148:886-90. American College of Obstetricians and Gynecologists technical bulletin no 154. a-Fetoprotein. Washington, DC: American College of Obstetricians and Gynecologists, 1991. Wald N, Cuckle H, Stirrat GM, et al. Maternal serum a-fetoprotein and low birth weight. Lancet 1977;2:268-70. Brock D], Barron L, Duncan P. Significance of elevated mid-trimester maternal plasmaa-fetoprotein values. Lancet 1979;2:1281-2. Smith ML. Raised maternal serum a-fetoprotein levels and low birth weight babies. Br] Obstet Gynaecol 1980;87: 1099-102. Burton BK. Outcome of pregnancy in patients with unexplained elevated or low levels of maternal serum a-fetoprotein. Obstet Gynecol 1988;72:709-13. Davis RO, Goldenberg RL,· Boots L, et al. Elevated levels of midtrimester maternal serum a-fetoprotein are associated with preterm delivery but not with fetal growth retardation. AM] OBSTET GYNECOL 1992;167:596-601. Robinson L, Grau P, Crandall BF. Pregnancy outcomes after increasing maternal serum a-fetoprotein levels. Obstet Gynecol 1989;74:17-9. Phillips OP, Simpson ]L, Morgan CD, et al. Unexplained elevated maternal serum a-fetoprotein is not predictive of
adverse perinatal outcome in an indigent urban population. AM] OBSTET GYNECOL 1992;166:978-82. 12. Richards DS, Seeds JW, Katz VL, Lingley LH, Albright SG, Cefalo RC. Elevated maternal serum a-fetoprotein with oligohydramnios: ·ultrasound evaluation and outcome. Obstet Gynecol 1988;72:337-41. 13. Williams MA, Hickok DE, Zingheim RW, et al. Low birth weight and preterm delivery in relation to early-gestation vaginal bleeding and elevated maternal serum a-fetoprotein. Obstet Gynecol 1992;80:745-9. 14. Nelson LH, Bensen], Burton BK. Outcomes in patients with unusually high maternal serum a-fetoportein levels. AM] OBSTET GYNECOL 1987;157:572-6. 15. Burton BK. Outcome of pregnancy in patients with unexplained elevated or low levels of maternal serum a-fetoprotein. Obstet Gynecol 1988;72:709-13. 16. Milunsky A, lick SS, Bruell CL, et al. Predictive values, relative risks, and overall benefits of high and low maternal serum a-fetoprotein screeing in singleton pregnancies: new epidemiologic data. AM] OBSTET GYNECOL 1989;161: 291-7. 17. Garver KL. Update 011 MSAFP policy statement from the American Society of Human Genetics. Am] Hum Genet 1989;45:332-4. 18. Nagey DA, Klebanoff M, Blitzer MG, et al. a-Fetoprotein as a predictor of preventable pregnancy loss [Abstract]. AM ] OBSTET GYNECOL 1993;168:315.
Midtrimester chorionic villus sampling: An alternative approach? Alan D. Cameron, MD: Karl W. Murphy, MD: Margaret B. McNay, MPhil: Alan M. Mathers, MH, ChB,b John Kingdom., MD,b David A. Aitken, PhD: Jennifer Crossley, PhD: Stuart Imrie, BSc: and Gordon Lowther, PhDc
Glasgow, United Kingdom OBJECTIVE: Our purpose was to audit midtrimester chorionic villus sampling after a positive maternal serum screening test for autosomal trisomy. . STUDY DESIGN: From January 1990 until July 1993 chorionic villus sampling was offered to all screened positive women. RESULTS: Five hundred fifty-one mothers had chorionic villus sampling. The mean age was 31.7 years. The mean gestational age was 18.2 weeks. The mean time for direct karyotyping was 4.4 days and for culture results 20.2 days. Results were obtained in 99.6% of samples: direct plus culture results in 94%, direct results alone in 2.3%, and culture results alone in 3.3%. Fourteen pregnancies had abnormal karyotypes. There were five cases of placental mosaicism and one false-positive result. The loss rate was 0.4%. CONCLUSION: Midtrimester chorionic villus sampling, which is easier to perform than cordocentesis, provides a rapid and reliable karyotype. The complication rate is comparable to that of other invasive procedures. (AM J OBSTET GVNECOL 1994;171 :1035-7.)
Key words: Serum screening, chorionic villus sampling, karyotype, abortion
From the Department of Obstetrics and Gynaecology, the QJ.teen Mother's Hospital,· Glasgow Royal Maternity Hospital,h and the Duncan Guthrie Institute of Medical Genetics.' Presented at the Fourteenth Annual Meeting of the Society of Perinatal Obstetricians, Las Vegas, Nevada, January 24-29, 1994.
Reprint requests: Alan D. Cameron, MD, Queen Mother's Hospital, Yorkhill, Glasgow, United Kingdom. Copyright © 1994 by Mosby-Year Book, Inc.
0002-9378/94 $3.00 + 0 6/6/57825
1035
Brazerol, Grover, and Donnenfeld
9. 10.
11. 12.
aspiration in labor using amnioinfusion. Obstet Gynecol 1989;73:647-51. Sadovsky Y, Amon E, Bade ME, Petrie RH. Prophylactic amnioinfusion during labor complicated by meconium: a preliminary report. AM J OBSTET GYNECOL 1989; 161 :613-7. Macri CJ, Schrimmer DB, Leung A, Greenspoon JS, Paul RH. Prophylactic amnioinfusion improves outcome of pregnancy complicated by thick meconium and oligohydramnios. AM J OBSTET GYNECOL 1992;167:117-21. Uhing MR, Bhat R, Philobos M, Raju TN. Value of amnioinfusion in reducing meconium aspiration syndrome. Am J Perinatol 1993;10:43-5. Wen TS, Eriksen NL, Blanco JD, Graham JM, Oshiro BT,
Am J Ohstet Gynecol
13. 14. 15. 16.
Prieto JA. Association of clinical intra-amniotic infection and meconium. Am J Perinatol 1993; 10:438-40. Miyazaki FS, Nevarez F. Saline amnioinfusion for relief of repetitive variable decelerations: a prospective randomized study. AM J OBSTET GYNECOL 1985;153:301-6. Meis PJ, Hall M, Marshall JR, Hobel CJ. Meconium passage: a new classification for risk assessment during labor. AM J OBSTET GYNECOL 1978;131:509-13. Katz VL, Bowes WA. Meconium aspiration syndrome: reflections on a murky subject. AM J OBSTET GVNECOL 1992;166:171-83. Jovanovic R, Nguyen HT. Experimental meconium aspiration in guinea pigs. Obstet Gynecol 1989;73:652-6.
Unexplained elevated maternal serum a-fetoprotein levels and perinatal outcome in an urban clinic population William F. Brazerol, MD, Steven Grover, MD, and Alan E. Donnenfeld, MD Philadelphia, Pennsylvania OBJECTIVE: Our purpose was to determine whether obstetric patients with unexplained elevated maternal serum a-fetoprotein levels from an indigent clinic population are at increased risk for adverse perinatal outcome compared with similar patients with normal values. STUDY DESIGN: Perinatal outcomes from inner-city obstetric patients with unexplained elevated maternal serum a-fetoprotein levels (> 2.0 multiples of the median) were compared with patients from the same clinic with normal values. The frequency of adverse outcomes in the two groups was subjected to x2 analysis. RESULTS: Adverse perinatal outcomes occurred in 33 of 57 (58%) of the subjects with unexplained elevated maternal serum a-fetoprotein levels compared with 163 of 719 (23%) patients with normal values (p < 0.001). Statistically significant differences were observed for abruptio placentae (p < 0.025), intrauterine growth retardation (p < 0.025), stillbirth at >20 weeks (p < 0.001), and pregnancy-induced hypertension (p < 0.01). Differences in the frequencies of preterm premature rupture of membranes, preterm delivery, pregnancy loss < 20 weeks, and congenital malformations were not statistically significant. CONCLUSION: In contrast to a previous report, we found that unexplained elevated maternal serum a-fetoprotein levels confer an increased risk of adverse perinatal outcome in an urban clinic population over and above the already increased risk related to socioeconomic status. (AM J OaSlET GYNECOL 1994;171 :1030-5.)
Key words: Unexplained elevated maternal serum a-fetoprotein levels, adverse perinatal outcome a-Fetoprotein (AFP) was first identified in 1956,1 and in 1973 elevated amniotic fluid levels were initially used clinically to diagnose anencephaly.2 Subsequently, elFrom the Section of Genetics, Department of Obstetrics and Gynecology, Pennsylvania Hospital. Presented at the Fourteenth Annual Meeting of the Society of Perinatal Obstetricians, Las Vegas, Nevada, January 24-29, 1994. Reprint requests: Alan E. DonnenfeId, MD, Section of Genetics, Department of Obstetrics and" Gynecology, Pennsylvania Hospital, 800 Spruce St., Philadelphia, PA 19107. Copyright © 1994 by Mosby-Year Book, Inc. 0002-9378/94 $3.00 + 0 6/6/57824
1030
evated matenal serum values became the standard of care as a screening test for neural tube defects. In 1984 low maternal serum AFP values were found to be associated with fetal aneuploidy." As a result of two decades of clinical investigation, maternal serum AFP has become an indispensible screening test that aids in the identification of pregnant women from the general obstetric population who are at risk for a number of disorders, including open neural tube defects, abdominal wall defects, and chromosomal abnormalities. For these reasons, offering maternal serum AFP screening
Volume 171, Number 4 Am ] Obstet Gynecol
to all pregnant women has been recommended! Abnormally elevated values are routinely followed by ultrasonographic examination to confirm gestational age, establish viability, rule out multiple gestation, or identity another cause for the elevated level. In approximately 5% of patients ultrasonographic evaluation will identity an open neural tube or ventral wall defect. In those patients whose ultrasonography confirms dates and no explanation for the elevated maternal serum AFP level is found, many patients will opt to undergo amniocentesis, Most will have a normal amniotic fluid AFP in spite of the elevated maternal serum AFP. This group comprises those with unexplained elevated maternal serum AFP levels, representing approximately 2% to 3% of the population undergoing maternal serum AFP screening. Initially this group of patients had been reassured that there was no evidence of an open fetal defect, indicating a favorable pregnancy outlook. In the past several years a number of investigators have demonstrated an increased risk of adverse perinatal outcome associated with unexplained maternal serum AFP levels, including spontaneous abortion, stillbirth, prematurity, intrauterine growth restriction (IUCR), abruptio placentae, premature rupture of membranes, pregnancy-induced hypertension, preeclampsia, and preterm labor. 5-9 The rate of adverse perinatal outcome has been reported to range from 16% to 47% and has been found to be directly correlated with the degree of elevation of maternal serum AFP. 10 Most previous studies have examined the perinatal outcome of patients with unexplained elevated maternal serum AFP levels in private obstetric populations at low risk. A single study from Memphis has evaluated the implications of an unexplained elevated test result in an indigent urban population and compared the perinatal outcome with matched-paired controls from the same clinic with similar obstetric risk factors.ll The authors found that unexplained elevated levels offered little, if any, predictive value for adverse perinatal outcome in populations already at high risk on the basis of obstetric or socioeconomic criteria. The purpose of this study was to perform a similar investigation of indigent urban patients from Philadelphia in a historic cohort analysis with independent groups to determine whether unexplained elevated maternal serum AFP levels identity a subset of clinical patients at further increased risk for adverse perinatal outcome above that related to obstetric or socioeconomic status. Methods
Patient data from women attending the Pennsylvania Hospital prenatal clinic in Philadelphia from January 1988 to June 1992 were analyzed. All patients were offered maternal serum AFP screening between 15 and 20 weeks' gestation, usually in association with simulta-
Brazerol, Grover, and Donnenfeld
1031
neous gestational dating ultrasonography to avoid false-negative and false-positive results because of inaccurate gestational dating. A total of 2764 women had maternal serum AFP screening during this period, with a 3.2% screen-positive rate. An elevated maternal serum AFP level was defined as ;;::: 2.0 multiples of the median adjusted for race, weight, and diabetes mellitus. An abnormal test was repeated in 1 week (unless the patient was past 18 weeks or the multiples of the median value was;;::: 3.0) and a gestational dating ultrasonography was obtained, if not previously performed. If two values were;;::: 2.0 multiples of the median, patients were offered genetic counseling, diagnostic ultrasonography and amniocentesis for amniotic fluid AFP level, chromosome analysis, and, when applicable, acetylcholinesterase level. If amniocentesis was declined, high-resolution ultrasonography was performed. An elevated maternal serum AFP level was considered unexplained if, after erroneous gestational dating (i.e., a ;;::: 10-day discrepancy between menstrual and ultrasonographic dating was corrected for, no evidence of a multiple gestation, fetal anomaly, oligohydramnios, or fetal death was detected on diagnostic ultrasonographic evaluation. Patients with unexplained elevated levels were offered weekly nons tress testing, with biophysical profile follow-up if nonreactive, and monthly Doppler ultrasonography starting at 28 to 32 weeks' gestation. Adverse perinatal outcomes were defined as spontaneous loss at < 20 weeks, IUCR, preterm premature rupture of membranes, premature delivery at < 37 weeks, pregnancy-induced hypertension, abruptio placentae, congenital malformations other than neural tube and ventral wall defects, and stillbirth at ;;::: 20 weeks. Patients with second-trimester oligohydramnios were excluded from the unexplained elevated maternal serum AFP group, both because this finding is a known cause of elevated levels and because this was the protocol used in the Memphis study. II , 12 Patients used as controls for this study were obtained from perinatal outcome data for patients attending the same prenatal clinic during 1990 and 1991. All clinic patients with normal maternal serum AFP values during these 2 years were included. Statistical analysis was performed by X2 testing. Results
Among the 2764 patients who underwent maternal serum AFP screening between January 1988 and June 1992, 89 (3.2%) had elevated values ;;::: 2.0 multiples of the median; 57 (21.%) of these were unexplained. In nine cases a repeat test was normal, in five patients incorrect gestational dating was identified, in two pregnancies twins were found, and there was one case each of open spina bifida and early severe oligohydramnios.
1032 Brazerol, Grover, and Donnenfeld
October 1994 Am J Obstet Gynecol
Table I. Age, ethnic distribution, and obstetric risk factors of study and control patients Study (n = 57)
Age (mean) 2: 35 yr 18-34 yr < 18 yr Race Black White Hispanic Asian Other or unknown Bleeding, current pregnancy Mean prior pregnancy losses Previous IUGR infant History of prematurity Smoking Prior stillbirth
Control (n = 719)
No.
%
No.
%
23.8 2 51 4
3.5 89.5 7.0
24.4 35 632 52
4.9 87.9 7.2
80.7 lO.5 8.8 0 0 15.8
46 6 5 0 0 9 1.6 4 6 14 1
595 50 54 12 8 119 1.2 44 53 138 11
7.0 lO.5 24.6 1.8
Table II. Comparison of pregnancy outcome between study and control groups by Study (n = 57) Pregnancy outcome
No.
Preterm delivery PPROM Abruptio placentae IUGR SAB <20wk Stillbirth 2: 20 wk PIH Congenital malformations
4 4 4 6 3 4 7 1 33
TOTAL
J
x.
2
82.8 6.9 7.5 1.7 1.1 16.6 6.1 7.4 19.2 1.5
analysis
Control (n = 719)
%
No.
7.0 7.0 7.0 10.5 5.2 7.0 12.3 1.8 57.9
37 22 14 29 12 6 31 12 163
1
%
Significance
5.1 3.1 1.9 4.0 1.7 0.8 4.3 1.7 22.7
NS NS P < 0.025 P < 0.025 NS P < 0.001 P < 0.01 NS P < 0.001
NS, Not significant; PPROM, preterm premature rupture of membranes; SAB, spontaneous abortion; PIH, pregnancy-induced hypertension.
Fourteen patients were lost to follow-up (15.7%). The control group was composed of all 804 clinic patients who underwent screening between January 1990 and January 1992 and who were found to have normal values. Of this group obstetric outcome data were available on 719; 85 (10.6%) were lost to follow-up. The mean maternal age in the study and control groups was 23.8 and 24.4 years, respectively. Ethnic distribution and several possible confounding variables that might affect obstetric outcome were also similar between groups (Table I). . Pregnancy outcome data are summarized in Table II. Adverse perinatal outcome occurred in 57.9% (33/57) of patients with unexplained elevated maternal serum AFP levels compared with 22.7% (163/719) of controls. This difference was statisically significant (p < 0.0001). Furthemore, statistical differences were found between subjects and controls for abruptio placentae (7.0% [four of 57] vs 1.9% [141719], P < 0.025), IUGR (10.5% [six of 57] vs 4.0% [29/719], P < 0.025), stillbirth at
and pregnancy-induced hypertension (12.3% [seven of 57] vs 4.3% [31/719], P < 0.01). Differences that did not reach statistical significance were observed for preterm premature rupture of membranes, preterm delivery, spontaneous abortion at < 20 weeks, and congenital malformations. Although there was no statistically significant difference in deliveries at < 37 weeks between groups, there was a significant difference in deliveries at < 28 weeks (six of 57 ([11.1 %] in those with elevated maternal serum AFP levels vs nine of 719 [1.3%] among controls, p < 0.001). In those patients with an elevated level > 2.5 multiples of the median, 21 of 27 (72.9%) had adverse outcomes compared with 12 of 32 (37.5%) with values between 2.0 and 2.5 multiples of the median. Data on mode of delivery are summarized in Table III. There was a higher cesarean section rate in the study group compared with controls (33.3% vs 24.1%, respectively). However, this difference was not statistically significant. A comparison of the primary cesarean section rate was statistically significant at the p < 0.025
Brazerol, Grover, and Donnenfeld
Volume 171, Number 4 Am J Obstet Gynecol
1033
Table III. Comparison of mode of delivery by X2 analysis Study (n Mode of delivery
No.
Vaginal delivery Cesarean section Primary cesarean section Repeat cesarean section
38 19 16 3
=
Control (n
57)
I
%
No.
66.7 33.3 28.1 5.2
546 173 114 59
=
719)
I
% 75.9 24.1 15.8 8.2
Significance NS NS
P < 0.025 NS
NS, Not significant.
Table IV. Comparison of birth weight by X2 analysis Study (n = 57) Birth weight
No.
;::2500 gm 1500-2499 gm 750-1499 gm
36 9 5
j
Control (n = 719)
%
No.
63.1 15.8 8.8
63 88 4
I
% 83.9 12.2 0.6
Significance
p<
0.005
NS
P < 0.001
NS, Not significant.
level (28.1% in the study group vs 15.8% in controls). Birth weight data are tabulated in Table IV. In the study group 18 of 57 (31.6%) infants weighed <2500 gm versus 116 of 719 (16.1%) controls (JJ < 0.005). Also, five of 57 (8.8%) weighed <750 gm versus four of 719 (0.6%) controls (JJ < 0.001). Comment Our investigation, a historic cohort study using independent groups from an indigent urban clinic population, confirmed the previously reported association of unexplained elevated maternal serum AFP levels with an increased risk of adverse perinatal outcome. Interestingly, this is in contrast to a similar investigation performed in Memphis that found no apparent increased risk in indigent, urban clinic patients with unexplained elevated levels compared with matched controls. II A comparison of perinatal outcome between the Memphis investigation, which used a historic cohort study with matched pairs, and our investigation is found in Table V. Overall, there was a statisticaly significant difference for all adverse perinatal outcomes (57.9% [33/57] in Philadelphia vs 38.9% [28172] in Memphis, p < 0.05). When individual adverse outcome variables were compared, only stillbirth at > 20 weeks was found to be significantly different between the two populations (four of 57 in the Philadelphia group vs none in the Memphis group, p < 0.05). There are several notable differences beween these two investigations. The Memphis group used a small control group that was rigorously matched for 21 different variables defined as unequivocal risk factors for adverse perinatal outcome (Preexisting Risk Guide, Hollister, Inc., Libertyville, Ill.) and maternal age, race, and parity. A possible criticism
of our investigation is that all 21 of these confounding variables for poor obstetric outcome were not assessed because of unavailability of data, although several important factors such as bleeding during the current pregnancy, prior pregnancy losses, previous IVGR, history of premature delivery, smoking, and prior stillbirth were similar between our study and control patients. Surprisingly, the Memphis study did not control for bleeding during pregnancy, an independent risk factor both for adverse perinatal outcome and elevated maternal serum AFP level. 13 This may have introduced a significant bias in their study, especially because of the small number of control patients involved. In addition, it is noteworthy that the Memphis results are, in fact, consistent with a small increase in risk for patients with elevated maternal serum AFP levels when all adverse outcomes are considered (38.9% vs 31.9% in study and control patients, respectively), and it is possible that the lack of statistical significance was related to the small sample size. This is acknowledged in their study. II Alternatively, it is also possible that the results of our study may have been biased by risk factors that were uncontrolled for because of a lack of information (such as akohol and drug abuse) or that the high rate of adverse perinatal outcome in the uDexplained elevated maternal serum AFP level group was related to our relatively small sample size of 57 patients. The finding of a 72.9% adverse perintal outcome rate among pati~nts with maternal serum AFP values > 2.5 multiples of the median is in accordance with several previously published studies that document an increasing adverse perinatal outcome rate with higher values. 10, 14 The Memphis group concluded that unexplained
1034 Brazerol, Grover, and Donnenfeld
October 1994 Am J Obstet Gynecol
Table V. Comparison of adverse perinatal outcomes between Memphis and Philadelphia studies by X2 analysis Philadelphia
Memphis Study (n Adverse perinatal outcome
No.
Pre term delivery PPROM Abruptio placentae IUGR SAB at <20 wk Stillbirth at 2: 20 wk PIH Congenital malformations*
8 6 2 6 4 0 9 2 28
TOTAL
I
=
72)
Control (n
%
No.
ILl 8.3 2.8 8.3 5.6
12 4 2 1 0 0 10 1 23
12.5 2.8 38.9
I
=
72)
Study (n
%
No.
16.7 5.6 2.8 1.4
4 4 4 6 3 4 7 1 33
13.4 1.4 31.9
I
= 57)
Control (n
%
No.
7.0 7.0 7.0 10.5 5.2 7.0 12.3 1.8 57.9
37 22 14 29 12 6 31 12 163
1
=
719)
%
Significance
5.1 3.1 1.9 4.0 1.7 0.8 4.3 1.7 22.7
NS NS P < 0.025 P < 0.025 NS P < 0.001 P < 0.01 NS P < 0.001
NS, Not significant; PPROM, preterm premature rupture of membranes; SAB, spontaneous abortion; PIH, pregnancy-induced hypertension. *Excluding open neural tube and ventral wall defects.
elevated maternal serum AFP levels and adverse perinatal outcome may not apply to all obstetric populations but suggested further studies from institutions similar to theirs. Our study attempted to respond to their request and documented a different result; for indigent urban clinical patients an unexplained elevated maternal serum AFP value confers an increased risk of adverse perinatal outcome over and above the already increased risk related to socioeconomic status. The explanation for the association between elevated maternal serum AFP levels and adverse pregnancy outcome remains elusive, although placental dysfunction, including partial abruptio placentae, pathologic findings, fetomaternal bleeding, and abnormal implantation are often implicated. Additionally, the optimal perinatal management strategy for patients with unexplained elevated levels is not known. IS·" Our population of pregnant women with unexplained elevated maternal serum AFP values were advised to have weekly nonstress tests and monthly Doppler velocimetry studies beginning between 28 and 32 weeks. In spite of this recommend~tion for increased surveillance, an extremely high adverse perinatal outcome rate was identified. In this population, however, there was a high frequency of unkept appointments for perinatal testing that may have contributed to the elevated adverse outcome rate. In addition, early delivery because of nonreassuring fetal testing was performed on several occasions during this investigation. These pregnancies may have thus benefited from the increased surveillance, inasmuch as fetal hypoxia or intrauterine death may have been prevented. However, both prematurity and fetal death were categorized as adverse perinatal outcomes in this study; thus substituting prematurity for stillbirth would not have altered the adverse perinatal outcome rate. It is unfortunate that our study design did not permit analysis of whether increased perinatal surveillance prevents
fetal morbidity and death. It is clear that only a large, prospective, randomized study using a well-defined protocol will answer this question. It is also of note that 11 % of our study patients were delivered before 28 weeks, the earliest that testing was initiated. The finding of adverse perinatal outcome before viability in pregnancies with unexplained elevated maternal serum AFP levels has been previously observed. In fact, one group suggested that prenatal surveillance is not useful in patients with unexplained elevated levels both because the adverse outcomes occur before viability and because the relative risk of preventable fetal death is low. IS Our data do not support this conclusion, because 27 of 33 of the patients with adverse outcomes in the group with unexplained elevated maternal serum AFP levels were delivered after 28 weeks. In spite of the many studies examining perinatal outcome with unexplained elevated maternal serum AFP levels, further work is required to determine the prognostic value of this screening test and the effect of variables such as socioeconomic status on pregnancy outcome. This is necessary to gain insight into the underlying pathophysiologic mechanisms responsible for adverse perinatal outcomes and to assist in identifYing patients at increased risk for IUGR, abruptio placentae, prematurity, preterm premature rupture of membranes, and fetal death. We thank Greg Maislin for his expert statistical assistance and critical review of the study design. REFERENCES
1. Bergstrand CG, Czar B. Demonstration of a new protein fraction in serum from the human fetus. J Clin Lab Invest 1956; 174:8. 2. Seller MJ, Coltart TM, Campbell S, Singer JD. Early termination of anencephalic pregnancy after detection by raised a-fetoprotein levels. Lancet 1973;1:73. 3. Merkatz IR, Nitowsky HM, Macri IN, Johnson WE. An
Cameron et al.
Volume 171, Number 4 Am J Obstet Gynecol
4.
5. 6. 7. 8. 9.
10. 11.
association between low maternal serum a-tetoprotein and fetal chromosome abnormalities. AM ] OBSTET GYNECOL 1984; 148:886-90. American College of Obstetricians and Gynecologists technical bulletin no 154. a-Fetoprotein. Washington, DC: American College of Obstetricians and Gynecologists, 1991. Wald N, Cuckle H, Stirrat GM, et al. Maternal serum a-fetoprotein and low birth weight. Lancet 1977;2:268-70. Brock D], Barron L, Duncan P. Significance of elevated mid-trimester maternal plasmaa-fetoprotein values. Lancet 1979;2:1281-2. Smith ML. Raised maternal serum a-fetoprotein levels and low birth weight babies. Br] Obstet Gynaecol 1980;87: 1099-102. Burton BK. Outcome of pregnancy in patients with unexplained elevated or low levels of maternal serum a-fetoprotein. Obstet Gynecol 1988;72:709-13. Davis RO, Goldenberg RL,· Boots L, et al. Elevated levels of midtrimester maternal serum a-fetoprotein are associated with preterm delivery but not with fetal growth retardation. AM] OBSTET GYNECOL 1992;167:596-601. Robinson L, Grau P, Crandall BF. Pregnancy outcomes after increasing maternal serum a-fetoprotein levels. Obstet Gynecol 1989;74:17-9. Phillips OP, Simpson ]L, Morgan CD, et al. Unexplained elevated maternal serum a-fetoprotein is not predictive of
adverse perinatal outcome in an indigent urban population. AM] OBSTET GYNECOL 1992;166:978-82. 12. Richards DS, Seeds JW, Katz VL, Lingley LH, Albright SG, Cefalo RC. Elevated maternal serum a-fetoprotein with oligohydramnios: ·ultrasound evaluation and outcome. Obstet Gynecol 1988;72:337-41. 13. Williams MA, Hickok DE, Zingheim RW, et al. Low birth weight and preterm delivery in relation to early-gestation vaginal bleeding and elevated maternal serum a-fetoprotein. Obstet Gynecol 1992;80:745-9. 14. Nelson LH, Bensen], Burton BK. Outcomes in patients with unusually high maternal serum a-fetoportein levels. AM] OBSTET GYNECOL 1987;157:572-6. 15. Burton BK. Outcome of pregnancy in patients with unexplained elevated or low levels of maternal serum a-fetoprotein. Obstet Gynecol 1988;72:709-13. 16. Milunsky A, lick SS, Bruell CL, et al. Predictive values, relative risks, and overall benefits of high and low maternal serum a-fetoprotein screeing in singleton pregnancies: new epidemiologic data. AM] OBSTET GYNECOL 1989;161: 291-7. 17. Garver KL. Update 011 MSAFP policy statement from the American Society of Human Genetics. Am] Hum Genet 1989;45:332-4. 18. Nagey DA, Klebanoff M, Blitzer MG, et al. a-Fetoprotein as a predictor of preventable pregnancy loss [Abstract]. AM ] OBSTET GYNECOL 1993;168:315.
Midtrimester chorionic villus sampling: An alternative approach? Alan D. Cameron, MD: Karl W. Murphy, MD: Margaret B. McNay, MPhil: Alan M. Mathers, MH, ChB,b John Kingdom., MD,b David A. Aitken, PhD: Jennifer Crossley, PhD: Stuart Imrie, BSc: and Gordon Lowther, PhDc
Glasgow, United Kingdom OBJECTIVE: Our purpose was to audit midtrimester chorionic villus sampling after a positive maternal serum screening test for autosomal trisomy. . STUDY DESIGN: From January 1990 until July 1993 chorionic villus sampling was offered to all screened positive women. RESULTS: Five hundred fifty-one mothers had chorionic villus sampling. The mean age was 31.7 years. The mean gestational age was 18.2 weeks. The mean time for direct karyotyping was 4.4 days and for culture results 20.2 days. Results were obtained in 99.6% of samples: direct plus culture results in 94%, direct results alone in 2.3%, and culture results alone in 3.3%. Fourteen pregnancies had abnormal karyotypes. There were five cases of placental mosaicism and one false-positive result. The loss rate was 0.4%. CONCLUSION: Midtrimester chorionic villus sampling, which is easier to perform than cordocentesis, provides a rapid and reliable karyotype. The complication rate is comparable to that of other invasive procedures. (AM J OBSTET GVNECOL 1994;171 :1035-7.)
Key words: Serum screening, chorionic villus sampling, karyotype, abortion
From the Department of Obstetrics and Gynaecology, the QJ.teen Mother's Hospital,· Glasgow Royal Maternity Hospital,h and the Duncan Guthrie Institute of Medical Genetics.' Presented at the Fourteenth Annual Meeting of the Society of Perinatal Obstetricians, Las Vegas, Nevada, January 24-29, 1994.
Reprint requests: Alan D. Cameron, MD, Queen Mother's Hospital, Yorkhill, Glasgow, United Kingdom. Copyright © 1994 by Mosby-Year Book, Inc.
0002-9378/94 $3.00 + 0 6/6/57825
1035