Impact of prenatal testing on maternal-fetal bonding: Chorionic villus sampling versus amniocentesis

Duerbeck et al.

October 1991

Am

in larger series of fetuses with congenital anomalies the incidence of abnormal systolic-to-diastolic ratios is 50%.11 Three of the thirteen fetuses studied had abnormal systolic-to-diastolic ratios for a rate of 23%; this incidence is nearly 10 times the rate cited in control populations. 1 It is interesting to note that in a situation in which the number of umbilical arteries was reduced from two to one, Doppler flow velocity ratios through these single umbilical arteries were normal in 77% of cases. REFERENCES 1. Schulman H, Winter D, Farmakides G. Pregnancy surveillance with uterine and umbilical arteries. AM J OBSTET GYNECOL 1989;160:192-6. 2. Erskine RLA, Ritchie JWK. Umbilical artery blood flow characteristics in growth retarded fetuses. Br J Obstet GynaecoI1985;92:605-IO. 3. Trudinger BJ , Giles WB, Cook CM. Umbilical and uterine artery flow wave forms in pregnancy associated with major fetal abnormality. Br J Obstet Cynaecol 1985;92:666-70.

J Obstet Gynecol

4. Trudinger BJ, Giles WB, Cook CM . Flow velocity waveform in the maternal uteroplacental and fetal umbilical placental circulation. AM J OBSTET GYNECOL 1985; 152: 155-63. 5. Fleischer A, Schulman H, Farmakides G, Bracerol L. Umbilical velocity waveforms in intrauterine growth retardation. AM J OBSTET GYNECOL 1985; 15 1:502-5. 6. Rochelson B, Schulman H, Farmakides G. The significance of absent end-diastolic velocity in umbilical artery velocity wave forms. AM J OBSTET GYNECOL 1987; 156: 1213-8. 7. Peckham CH, Yerushalmy J. Aplasia of one umbilical artery: incidence by race and certain obstetric factors. Obstet Gynecol 1965;26:359-63. 8. Bernirschke K, Dodds JP. Angiomyxoma of the umbilical cord with atrophy of an umbilical artery. Obstet Gynecol 1967;30:99-103. 9. Froehlich LA, Fujikura T. Follow-up of infants with single umbilical artery. Pediatrics 1973;52:22-5 . 10. Bryan EM, Kohler HG. The missing umbilical artery: paediatric follow-up. Arch Dis Child 1985;50:7 14-8. II. Meizner 1, Katz M, Lunenfeld E, Insler V. Umbilical and uterine flow velocity waveforms in pregnancy complicated by major fetal anomalies. Prenatal Diag 1987;7 :49 1-6.

Impact of prenatal testing on maternal-fetal bonding: Chorionic villus sampling versus amniocentesis N. Caccia, BSc, J.M. Johnson, MD, G.E. Robinson, MD, and T. Barna, BSc Toronto, Ontario, Canada The process of maternal-fetal attachment, considered vital for normal infant development, begins during pregnancy and can be affected by a number of external factors. In this study the impact of prenatal testing on maternal-fetal bonding was evaluated in 253 women undergoing either first-trimester chorionic villus sampling (n = 101) or second-trimester genetic amniocentesis (n = 152). The women were evaluated by means of a modification of the Cranley Maternal-Fetal Attachment Scale, administered before and after the results of the prenatal diagnostic testing were made known to them (mean gestational ages of 10.6 and 15.7 weeks for the chorionic villus sampling group and 16.5 and 21.1 weeks for the amniocentesis group). The results showed: (1) that maternal-fetal attachment begins as early as 10 weeks' gestation and increases significantly as the pregnancy progresses, (2) that maternal-fetal attachment increases significantly, once the results are known to be normal, for both groups (p < 0.001). (3) that this increase occurs about 5 weeks earlier for patients with chorionic villus sampling in comparison to those undergoing amniocentesis (p < 0.001). Thus, with regard to the process of maternal-fetal attachment. first-trimester chorionic villus sampling appears to be preferable to second-trimester amniocentesis. (AM J OBSTET GYNECOL 1991;165:1122-5.)

Key words: Maternal-fetal bonding. prenatal diagnosis. chorionic villus sampling, amniocentesis

From the Department of Obstetrics and Gynecology and the Department of Psychiatry, University of Toronto. Presented at the Eleventh Annual Meeting of the Society of Perinatal Obstetricians. San Francisco, Califomia. January 28-February 2.

1991.

Reprint requests: J.M. Johnson, MD, Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, 200 Elizabeth St., 6EN-216, Toronto, Ontario, Canada M5G 2C4. 6/6/31198

1122

The process of maternal-infant bonding, considered vital for normal infant development. has been shown to begin during pregnancy and can be affected by a number of external factors .'·3 While the bonding process in the later stages in pregnancy (>20 weeks' gestation) has been well investigated ,3·4 the extent to which bonding is manifest in early pregnancy «20 weeks'

Volume 165 Number 4, Part I

gestation) and the factors affecting this bond remain unclear. Prenatal diagnosis has been shown to have a significant effect on the emotional and psychologic components of pregnancy,S-' but its impact on the bonding process in early pregnancy has not been previously investigated. Since the period of undergoing prenatal diagnosis and awaiting the results is associated with significant parental anxiety,s-Io it is possible that prenatal diagnosis may interfere with or delay this bonding process. The purpose of this study was to evaluate maternalfetal bonding in early pregnancy to determine: (1) whether maternal-fetal bonding can be demonstrated before 20 weeks' gestation and (2) the effect of prenatal diagnosis on maternal-fetal bonding, specifically, whether bonding is delayed in patients who undergo amniocentesis as opposed to chorionic villus sampling (CVS). Material and methods

The study group consisted of women with singleton pregnancies, confirmed by ultrasonography, with appropriate indications for prenatal testing according to the Canadian Guidelines for Prenatal Genetic Diagnosis." Eligible patients underwent nondirective genetic counseling regarding the risks, benefits, and limitations of chorionic villus sampling and/or amniocentesis and provided informed consent for their procedure of choice and for participating in the study. The women were evaluated by means of a modification of the Cranley Maternal-Fetal Attachment Scale,12 administered before (preresult questionnaire) and after (postresult questionnaire) their results were made known to them. The Maternal-Fetal Attachment Scale allows patients to rank statements of their feelings about the pregnancy and the fetus in five areas of attachment: interaction with the fetus, giving of self, roletaking, differentiation from self, and attributing characteristics to the fetus. The statements are ranked on a scale from 1 to 5, with 5 being the most attached or bonded. In this study the scale was modified by omitting the category of "attributing characteristics to the fetus" because the questions were not appropriate for a gestational age <20 weeks. The effect of ultrasonography on maternal attitudes toward the pregnancy and fetus were examined by asking the patient to rank a series of statements on a scale of 1 to 5 with 5 indicating the greatest effect. The data were analyzed and statistical analysis, where appropriate, was by paired two-tailed Student t test with p< 0.05 being considered significant. Results

Over the 4-month period beginning May 1989, 253 consecutive women undergoing chorionic villus sam-

Prenatal diagnosis and maternal-Ietal bonding

1123

Table I. Indications for prenatal testing

Indication

Late maternal age Previous abnormality Family history of genetic disease Abnormal maternal serum a-fetoprotein level

Amniocentesis (n = 151) (%)

(n

CVS = 102) (%)

77 4 4

71 19 10

15

N/ A

NA, Not available.

pIing (n = 101) or second-trimester genetic amniocentesis (n = 152) met the entry criteria. The groups did not differ significantly with respect to age, parity, socioeconomic status, or level of postsecondary education. Their indications for prenatal testing are described in Table l. A 100% follow-up rate was achieved in both groups. The preresult and postresult questionnaires were administered to the patients at an average gestational age of 10.6 and 15.7 weeks for the CVS patients and at 16.5 and 2l.1 weeks for the amniocentesis patients. This resulted in three comparison groups being generated: the preresult and postresult CVS group (group 1), the postresult CVS and preresult amniocentesis group (group 2), and the preresult and postresult amniocentesis group (group 3). In all three groups the level of bonding was significantly higher in the postresuit group than in the preresult group (p < 0.001) (Table II). This increase in bonding occurred approximately 5 weeks earlier in patients undergoing CVS compared with those undergoing amniocentesis (Fig. 1). There was an appreciable level of bonding (3.8 on a scale of 5) present in all preresult patients with no statistical difference between amniocentesis and CVS patients, in spite of a difference in gestational age of almost 6 weeks (Fig. 1). Similarly, the postresult levels of bonding did not differ significantly, with the postresult CVS patients achieving the same level of bonding by 15 weeks as the postresult amniocentesis patients achieved at 21 weeks (p < 0.001) (Table II). All of the patients underwent one or more ultrasonographic examination(s) in association with their procedures. In 126 of 253 patients (50%), seeing the fetus on ultrasonography made them feel less anxious about the pregnancy and closer to the baby (72%), made the baby seem more real (87%) and in some cases made them feel "more like parents" (27%). As may be expected, positive feelings also made a significant percentage of the parents (55%) feel more anxious about the possibility of an abnormal result. These reactions were stronger at earlier gestational ages (~12 weeks'

1124 Caccia et al.

October 1991 Am J Obstet Gynecol

5r-----------------------------------------------,

B 0

n

d I

4.5

n g

S

c a

4

I

e

3.5L-----~------~------~----~------~----~

10

16 14 18 Gestational Age (wks)

12

20

22

Fig. 1. Maternal-fetal bonding: CVS versus amniocentesis. Maternal-fetal bonding increased significantly in both CVS patients (thin line) and amniocentesis patients (thick line) once results of testing were reported to be normal (15.7 weeks' gestation for CVS patients, 21.1 weeks for amniocentesis patients). This increase in bonding occurred approximately 5 weeks earlier in CVS patients. Bonding scale is from 1 to 5 with 5 representing greatest bonding.

Table II. Maternal-fetal bonding: CVS versus amniocentesis Group 1 Attachment task* I. Interaction with fetus

Pre-CVSt (n = 102)

2. Giving of self 3. Role-taking 4. Differentiation from self

2.7 4.2 3.8 4.6

Mean attachment

3.8

I

Group 2

Post-Cvst (n = 102)

Post-CVS (n = 102)

3.5 4.8 4.5 4.8 4.4§

3.5 4.8 4.5 4.8 3.8

1

Group 3

Pre-At (n = 151) 2.8 4.3 4.0 4.4 4.4§

(n

Pre-A = 151) 2.8 4.3 4.0 4.4 3.8

J

Post-At (n = 151) 3.6 4.8 4.7 4.9 4.5§

*Modified Cranley Maternal-Fetal Attachment Scale (scale of 1 to 5, with 5 being most bonded). tPreresult and postresult CVS patients. :f:Preresult and postresult amniocentesis patients. §p < 0.001, when compared with preresult bonding (paired two-tailed Student t test).

gestation), although they also were present at later stages (13 to 20 weeks' gestation). We found that the first sonogram tended to have the greatest effect on their feelings toward the pregnancy, although subsequent sonograms also had an effect. Comment

Our results indicate that in patients undergoing prenatal diagnosis, maternal-fetal bonding can be demonstrated as early as 10 weeks' gestation and increased significantly once the test results are known to be normal. This is consistent with previous studies that have shown that women undergoing prenatal diagnosis are reluctant to become "involved" in their pregnancies until their fetuses are known to be normal. 5 This reluctance also may explain why, in these patients, bonding

appears more a function of the prenatal diagnosis experience than of gestational age. In this study the postresult CVS patients were shown to be significantly more bonded than were the pre result amniocentesis patients, in spite of being at a similar gestational age. Once the results were known to be normal, the level of bonding in CVS patients and amniocentesis patients was similar, suggesting that this factor plays a major role in the development of bonding in these patients. Previous investigators hve shown that CVS and amniocentesis patients have similar levels of attachment by the late second trimester,g suggesting that CVS offers no long-term benefit in maternal-fetal bonding compared with amniocentesis. However, amniocentesis patients have been shown to be more bonded than gestational agematched controls once they know their fetuses are

Volume 165 Number 4, Part I

normal'; this seems to indicate that prenatal diagnosis may increase maternal-fetal bonding earlier in gestation, Maternal-fetal attachment has been shown to alter maternal life-styles to be more beneficial to the fetus 3 ; therefore it is possible that an earlier increase in bonding may be beneficial by positively influencing maternal behavior at an earlier stage in pregnancy, when it may have a more significant effect on fetal outcome. One of the factors that likely affects maternal-fetal bonding is maternal anxiety. Anxiety has been shown to correlate with the somatic symptoms of pregnancy and gestational age, as well as inversely with maternal education. 13 Studies evaluating the effect of prenatal diagnosis on maternal anxiety have shown that CVS patients demonstrate an earlier reduction in maternal anxiety than amniocentesis patients. g · 10 The reduction parallels the increase in attachment we found and may have been a factor facilitating the ease and degree of bonding observed in our patients. The effect of prenatal diagnosis on third-trimester anxiety is unclear, although Marteau et aU reported that in patients who have undergone prenatal diagnosis, anxiety levels in the third trimester fall or remain the same as in the second trimester, whereas in the general population anxiety levels increase in the third trimester. Preresult anxiety levels have been shown to be lower in patients whose primary indication for prenatal diagnosis is late maternal age compared with other indications. l4 In this study later maternal age was the primary indication for prenatal diagnosis in 117 of 151 amniocentesis patients (77%) and 72 of 102 CVS patients (71 %). Preresult binding levels in this subpopulation were evaluated and compared with levels in the remaining patients. No significant differences were found, suggesting that maternal-fetal bonding in prenatal diagnosis patients is not influenced by the indication for the procedure in spite of possible differences in preresult anxiety levels. The influence of ultrasonography on maternal attitudes toward the fetus were significant, with patients reporting that seeing the fetus on a sonogram made them feel less anxious about the fetus but more anxious about the possibility of an abnormality; overall the ultrasonography engendered positive feelings toward the pregnancy. It is likely that the ultrasonographic examination performed immediately before the preresult bonding evaluation may have contributed to the significant bonding levels found at the time of the procedure.

Prenatal diagnosis and maternal-fetal bonding

1125

We conclude that in patients undergoing prenatal genetic testing, maternal-fetal bonding can be demonstrated as early as 10 weeks' gestation and increases significantly once test results are known to be normal. Seeing the fetus on ultrasonography and being informed of normal results were important factors influencing the development of this bonding. The increase in bonding occurred approximately 5 weeks earlier in patients undergoing CVS compared with those undergoing amniocentesis, suggesting CVS may be preferable to amniocentesis with respect to the process of maternal-fetal bonding. We acknowledge the Angus Reid Group, 140 4th Ave. S.W., Calgary, Alberta, Canada, for their assistance in the data analysis, and Ms. Karen Tustin for her technical assistance.

REFERENCES 1. Grace JT. Development of maternal fetal attachment during pregnancy. Nurs Res 1989;34:228-32. 2. Lerum CW, LoBiondo-Wood G. The relationship of maternal age, quickening, and physical symptoms of pregnancy to the development of maternal fetal attachment. Birth 1989;16:13-7. 3. Davis MS, Akridge KM. The effect of promoting intrauterine attachment in primiparas on postdelivery attachment. J Obstet Gynecol Neonatal Nurs 1987; 16:430-7. 4. Kemp VH, Page CK. Maternal prenatal attachment in normal and high-risk pregnancies. J Obstet Gynecol Neonatal Nurs 1987; 16: 179-84. 5. Beeson D, Golbus MS. Anxiety engendered by amniocentesis. Birth Defects 1979;15:191-7. 6. Marteau TM, Johnston M, Shaw RW, et al. The impact of prenatal screening and diagnostic testing upon the cognitions, emotions and behaviour of pregnant women. J Psychiatr Res 1989;33:7-16. 7. Phipps S, Zinn AB. Psychological response to amniocentesis: mood state and adaptation to pregnancy. Am J Med Genet 1986;25:131-42. 8. Fava GA, Trombini G, Michelacci L, et al. Hostility in women before and after amniocentesis. ] Reprod Med 1983;28:29-34. 9. Robinson GE, Garner DM, Olmsted MP, et al. Anxiety reduction after chorionic villus sampling and genetic amniocentesis. AM] OBSTET GYNECOL 1988;158:953-6. 10. Spencer ]W, Cox DN. Emotional responses of pregnant women to chorionic villi sampling or amniocentesis. AM ] OBSTET GYNECOL 1987;157:1155-9. 11. Canadian recommendations for prenatal diagnosis of genetic disorders. Bull Soc Obstet Gynaecol 1983;5:5. 12. Cranley MS. Development of a tool for the measurement of maternal attachment during pregnancy. Nurs Res 1981;30:281-4. 13. Lubin B, Gardener SH, Roth A. Mood and somatic symptoms during pregnancy. Psychosom Med 1975;37: 136-46. 14. Evers-Kiebooms G, Swerts A, et al. Psychological aspects of amniocentesis: anxiety feelings in three different risk groups. Clin Genet 1988;33:196-206.