- Email: [email protected]
Integration of the trans abdominal technique into an ongoing chorionic villus sampling program
Volume 161 I\umber 5
BG, Zaneveld LJD. Development of an assay to assess the functional integrity of the human sperm membrane and its relationship to other semen characteristics. J Reprod Fertil 1984;70:219-28. 8. Collins JA. Diagnostic assessment of the infertile male partner. Curr Probl Obstet Gynecol Infertil 1987; 10: 177224. 9. Van der Ven HH, Jeyendran RS, AI-Hasani S, PerezPelaez M, Diedrich K, Zaneveld LJD. Correlation between human sperm swelling in hypoosmotic medium (hypoos-
Semen predictors of in vitro fertilization
motic swelling test) and in vitro fertilization. J Androl 1986;7:190-6. 10. Yanagimachi R, Yanagimachi H, Rogers BJ. The use of zona-free animal ova as a test system for the assessment of the fertilizing capacity of human spermatozoa. Bioi Reprod 1976; 15:471-6. II. Chan SYW, Fox EJ, Chan MMC, et al. The relationship between the human sperm hypoosmotic swelling test, routine semen analysis, and the human sperm zona-free hamster ovum penetration assay. Fertil Steril 1985;44:668-72.
Integration of the transabdominal technique into an ongoing chorionic villus sampling program Karen L. Copeland, MS, Robert J. Carpenter, Jr., MD, Kurt R. Fenolio, MS, and David H. Ledbetter, PhD Houston, Texas Data are presented on 869 patients undergoing chorionic villus sampling procedures by one of two sampling techniques: 544 by a transcervical catheter aspiration method and 325 by a transabdominal two-needle aspiration method. The transcervical approach was the only procedure used in the first 330 cases, at which time the transabdominal technique was incorporated into our program. After an initial learning curve in the first 100 procedures the transcervical fetal loss rate stabilized at 2.7%, the number of patients requiring more than one catheter insertion decreased to 11 %, and tissue weights 2:10 mg were obtained in 88% of cases. The fetal loss rate for transabdominal chorionic villus sampling was 2.6%, indicating the addition of this new method did not significantly alter the fetal loss rate. Transabdominal chorionic villus sampling had an overall success rate of 99%, with only one insertion of the guide needle required for 98% of patients. Tissue weights of 2:10 mg were obtained in 99% of cases. These results demonstrate that the transabdominal procedure can be rapidly and effectively incorporated by an operator already experienced with transcervical chorionic villus sampling. Since several contraindications exist for either chorionic villus sampling method, the availability of both techniques at a single center greatly enhances the ability to offer first-trimester fetal diagnosis to a majority of patients. (AM J OBSTET GVNECOL 1989;161 :1289-94.)
Key words: Prenatal diagnosis, chorionic villus sampling, transcervical CVS, transabdominal CVS In the early 1980s reports first appeared on the ability to perform first-trimester prenatal diagnostic studies by chorionic villus sampling (CVS).1.2 Within a few years there were several large patient series reporting the successful application of CVS, indicating that chorionic villi can be used reliably for cytogenetic, biochemical, and deoxyribonucleic acid analyses. 3. 6 There are two From the InstItute for Molecular GenetIcs and the Department of Obstetrics and G.vnecology, Baylor College of MedIcine. This work was supported in part by grant No. HD19881 from the NatIOnal InstItute of Child Health and Human Development. Received for publicatIOn January 23,1989; revised May 2,1989; accepted May 25, 1989. Reprmt requests: Karen L. Copeland, MS, Institute for Molecular GenetICS, Baylor College of Medlcme. One Baylor Plaza, Houston,
TX 77030.
6/1 / 14236
major sampling techniques for obtaining chorionic villis currently in use: ultrasonographically guided transcervical catheter aspiration and ultrasonographically guided transabdominal needle aspiration (singleand double-needle systems).7 The transcervical technique has had the greatest utilization internationally and until recently was the only technique in use in the United States. Multicenter studies from the United States and Canada found a slightly increased rate of fetal loss (0.6% and 0.8%, respectively) for transcervical CVS over amniocentesis,"' 9 but these differences were not statistically significant. In addition, no significant maternal complications were noted after transcervical CVS. The transabdominal technique has been used by several groups in Europe and its use is rapidly growing in the U.S. IU 1289
1290 Copeland et al.
November 1989 Am J Obstct Gynecol
Table I. Transabdominal placental location Locatwn
Anterior Posterior Lateral Fundal Previa
n
164 117 16 16 12
Table II. Gestational age at CVS %
50 36 5
5 4
We report our experience with the transcervical and transabdominal techniques to illustrate that the transabdominal method can be readily incorporated into an ongoing transcervical program. Since contraindications exist for either technique, there are benefits for using both methods at a single center. Material and methods
The first 869 patients undergoing CVS procedures for prenatal diagnosis at our center are described, including 544 performed by the transcervical method and 326 by the transabdominal method . Risk of cytogenetic disorders was the indication in 91 % of cases, with maternal age the most frequent indication (80%). In 9% of cases, patients were at increased risk for a specific genetic disease in which either DNA analysis (7%) or biochemical analysis (2%) was performed. Cytogenetic analysis was attempted by both direct preparation 2 and long-term culture in all cases with ~ 10 mg. Direct and culture methods were successful in 85 % of cases, direct method only in 2%, and culture method only in 13%. Before the procedure, all women underwent an ultrasonographic evaluation to determine fetal viability, gestational age, and number of fetuses present; twin pregnancies were excluded from this study. Most patients in our referral area had ultrasonographic evaluation performed at the time of genetic counseling. According to a protocol approved by the Baylor College of Medicine Institutional Review Board, informed consent was obtained from all women before CVS. Our institution is one of eight centers participating in the Cooperative Clinical Comparison of Chorionic Villus Sampling and Amniocentesis sponsored by the National Institute of Child Health and Human Development to study the safety and accuracy of chorionic villus sampling. 8 Ultrasonographic evaluation at the time of the procedure established placental position and determined whether the transcervical or transabdominal approach could be used. Patients were excluded from either CVS procedure if there was active bleeding within I week of the scheduled procedure or preexisting rhesus alloimmunization. Transcervical CVS was not performed in women with active vaginal infections or cervical stenosis. Transabdominal CVS was not performed in
Weeks' gestatwn
9.0-9.9 10.0-10.9 11.0-11.9 12.0-12.9 13.0-13.9 Total no. of patients
Transcerv!cai n
135 277 116 15 1 544
I
Transabdominal
%
n
25 51 21 3
60 161 66 35 3 325
I
%
18 50 20 11 1
women with bowel visualized between the abdomen and uterine wall. Both CVS techniques were performed by the same operator and a team of ultrasonographers and nurses. The determination of whether to perform CVS and which method was to be used for each patient was made by the operator. Patients who could undergo CVS by either technique were randomized between the transcervical and transabdominal methods as a component of the collaborative study. The transcervical technique was performed with the patient in the lithotomy position. A short speculum was inserted into the vagina after the vagina and external genitalia were cleansed with Betadine solution. A single-toothed tenaculum was placed on the anterior cervix in one third of patients. With continuous real-time ultrasonographic guidance provided by an Acuson-128 scanner with a 3 .5 MHz linear transducer, a Portex catheter was inserted through the cervix into the chorion frondosum . A 20 ml Becton-Dickinson Luer lock syringe with 1 to 2 ml of aspiration media (minimal essential medium without fetal calf serum) was attached and villi were aspirated with hand suction (10 to 12 ml) as the catheter was withdrawn. The transabdominal technique was performed with the patient in the supine position. After selection of the needle insertion site, the area was cleansed with Betadine solution and draped with sterile towels. Xylocaine (3 ml of 1 % solution) was administered to decrease maternal discomfort. With continuous real-time ultrasonographic guidance provided by an Acuson-128 scanner with a 3.5 MHz sector transducer, we carried out a modificiation of the double-needle procedure described by Smidt-Jensen and HahnemannIl using Thin Wall Quinke spinal needles of 18 gauge (3.5 em) and 20 gauge (5.5 em). By freehand placement, the 18gauge spinal needle was directed through the maternal abdominal wall and myometrium with the tip inserted just beyond the decidual-chorion interface. The obturator was then removed and the 20-gauge needle was inserted through the outer (introducer) needle, into the trophoblastic tissue, and then moved back and forth. Villi were aspirated as described for the transcervical procedure. The transabdominal procedures early in
Transabdominal CVS
Volume 161 Number 5
1291
Table III. Insertions to obtain tissue* Transabdomnwl: Procedures 1-325
Transcervical Procedures 101-544
Procedures 1-100 (n)
No. of insertions
54 30
I
2
396 41 7
14 2
3 4
I
n
o
%
n
%
89 9 2
291
90 10
31
2 I
*Transcervical indicates catheter insertions. Transabdominal indicates inner-needle insertions.
Table IV. Tissue weight obtained*
I
Transcervical Amount of tISSue
Failed <5 mg 5-9 mg 10-19 mg ~20 mg
Procedures 1-100 I
5
19 59 13
Procedures 101 -544
3 2 7 20 68
Table V. Fetal losses Transcervical
Transabdominal: Procedures
Procedures 1-100
1-325
93
*Values represent percent of patients. our experience were limited to anterior placentas and, as experience increased, more lateral and then posterior placentas were sampled. Placental localization is noted in Table 1. A transabdominal-trans vesical technique was performed in 21 patients. In these patients a transcervical approach was not possible because of a severely retroflexed uterus and anterior placenta. The maternal bladder was obstructing the only viable transabdominal approach, and emptying and filling of the maternal bladder did not alter the feasibility of either approach. The IS-gauge needle was inserted through the maternal abdominal wall, the maternal bladder, the myometrium, and into the chorion frondosum . Results
Procedures. The transcervical technique was performed in 544 patients and the transabdominal technique was performed in 325 patients between 9 and 13 weeks' gestation (Table II). While 70% of patients were between 10 and 11 weeks' gestation in both the transcervical and transabdominal groups, 12% of the transabdominal group were 2!:: 12 week's gestation, compared with 3% of the transcervical group. Early in the transcervical series (I to 100), multiple catheter insertions were necessary in 46% of cases to obtain adequate tissue samples (Table III). As operator experience increased (101 to 544), only II % of patients required multiple catheter insertions to obtain an adequate sample. There were 15 transcervical pro-
TAB-VTP Continuing <16 wk 16-20 wk 20·24 wk Total losses
9*
91 4 2 2 8 (8.8%)
I Procedures 101-544 31 413 5 4 2 II (2.7%)
Transabdominal: Procedures
1-325
17 308 4 I 3 8 (2.6%)
TAB-VTP, Therapeutic abortion-voluntary termination of pregnancy. *lncludes two miscarriages with genetic anomalies: trisomy 13 and male fetus affected with hemophilia.
cedures (2 .S%) in which no tissue was obtained. Eleven of the patients with transcervical failure underwent amniocentesis whereas four underwent transabdominal CVS. Tissue was obtained in 324 of 325 transabdominal cases. In 31S transabdominal cases (9S%) adequate tissue was obtained with only one insertion of the outer needle; 7 patients required a second insertion of the needle to obtain tissue. The inner needle was reinserted in 10% of cases to obtain additional tissue. Tissue obtained in two of the transabdominal procedures was of an unusual appearance and only 46,XX cells were seen after cytogenetic analysis; because of concerns of maternal cell contamination, repeat diagnostic studies were recommended. One patient underwent a repeat transabdominal procedure 2 weeks later and the other underwent amniocentesis; in both cases the 46,XX result was confirmed. One patient had a failed transabdominal CVS and underwent a successful transcervical procedure at a later date. All 21 transabdominaltransvesical procedures were successful. The transcervical method yielded small tissue weights in the first 100 procedures, with 24% of samples <10 mg and only 13% 2!::20 mg (Table tV). In the subsequent 444 procedures only 9% of samples were < 10 mg and 6S% were 2!::20 mg. The transabdominal
1292 Copeland el al.
November 1989 Am J Obslel Cynecol
Table VI. Pregnancy outcome > 24 weeks' gestation Transcervical
n
Continuing Live birth Term
<37 wk
<2500 gm Intrauterine growth retardation Stillbirth Lost to follow-up
485 479 443
I
23* 10
3
2 4
Transabdommal
%
n
I
Total
%
785
300
92
5 2
295
280 11 * 2 2
2 3
n
95
4
774 723
34
12 5 4 7
I
%
93 4
2
*Neonatal death in one transcervical procedure and one transabdominal procedure.
technique yielded adequate sample sizes with only 3% of samples <10 mg and 93% ~20 mg. Twenty-three patients (2.6%) underwent a subsequent amniocentesis because of diagnostic failures or ambiguities. Nineteen procedures were performed for cytogenetic reasons: 14 cases of mosaicism of potential clinical signifciance (4 of which were confirmed), 1case of significant maternal cell contamination, 1 culture failure, and 3 cases with cytogenetic information limited to only one culture. One patient had repeatedly elevated serum et-fetoprotein levels. Three patients had additional studies for DNA or biochemical diagnoses. Immediate complications. For patients undergoing transcervical CVS, approximately 30% experienced spotting and bleeding after the procedure. In most patients this resolved within 24 hours and a brown discharge occurred for 1 to 2 days thereafter. Mild cramping occurred occasionally. For patients undergoing transabdominal CVS, cramping was common but resolved by the following morning. Bleeding or spotting was infrequent (5%) . Eleven patients (3%) had a vasovagal reaction immediately after the transabdominal procedure involving dizziness, bradycardia, flushing, and diaphoresis. This resolved in a few minutes if the patient was kept in a supine position. There were no recurrences of the vasovagal symptoms. Only one patient who underwent a transabclominal-transvesical procedure experienced a transient episode of minimal hematuria. Bleeding, cramping, and vasovagal episodes did not correlate with risk of fetal loss. Fetal loss. The transcervical procedure was performed in 544 patients, and 40 of these pregnancies were terminated because of a diagnosed genetic abnormality or for social reasons. In the continuing 504 pregnancies a total of 19 losses with normal chromosome analysis occurred from the time of the transcervical procedure to 24 weeks' gestation (Table V). Eight occurred in the first 100 procedures, for an uncorrected loss rate of 8.8% . In the subsequent 444 procedures there were 11 losses for an uncorrected fetal loss rate of 2.7%. In the 325 transabdominal procedures 17 pregnancies were terminated because of a
genetic abnormality or for social reasons. In the remaining 308 continuing pregnancies eight spontaneous losses occurred at <24 weeks, for an uncorrected fetal loss rate of2.6%. Four patients underwent unsuccessful transcervical CVS followed by a transabdominal procedure, and one had an unsuccessful transabdominal CVS followed by a transcervical procedure; none of these pregnancies miscarried. Laboratory findings. Twenty-one cases of cytogenetic abnormalities were detected : 14 with aneuploidy, 4 with unbalanced translocations, I with triploidy, and 4 with mosaicism confirmed in fetal tissue. Fifteen patients had abnormal results from DNA or biochemical studies. In all cases of abnormal diagnosis the pregnancies were terminated except for one case of sickle cell disease. Results were confirmed in all cases where tissue was obtained. There have been no false-positive or false-negative results in our 869 CVS cases. However, in the four cases of fetal mosaicism the mosaicism was not detected in the direct preparation and was seen only in the cultured cells. Outcome. All pregnancies have been delivered. Information is available for 99% of the 785 continuing pregnancies (Table VI) . There are no significant differences between the transcervical or transabdominal methods in the incidence of prematurity, low birth weight, intrauterine growth retardation, neonatal death, or stillbirth.
Comment Our experience with 325 transabdominal CVS procedures indicates it is a viable alternative for providing first-trimester prenatal diagnosis to operators currently using transcervical CVS. Our transabdominal fetal loss rate of 2.6% and success rate of 99% demonstrate that the addition of the transabdominal technique did not increase the overall risks of CVS at our center. Internationally, CVS has been performed primarily by the transcervical technique, which appears to be a reliable method of prenatal diagnosis. s." Transabdominal CVS was first described in 1984 by Smidt-Jensen and Hahnemann" and may have several theoretical
Volume 151 Number 5
and practical advantages. First, transabdominal CVS avoids the likelihood that organisms present in the vagina and cervix will be introduced into the uterus. Second, transcervical CVS is contraindicated in patients with active vaginal infections, in cases of cervical stenosis, or in cases in which the procedure would be very difficult because of placental position and uterine anatomy. These patients may be denied first-trimester diagnosis if the transcervical method is the only option available. 12 We initially attempted the transabdominal technique in two patients at high risk for genetic abnormality (cystic fibrosis and Becker muscular dystrophy) who could not undergo CVS transcervically because of a sharply retroverted uterus and an anterior placenta. After these two experiences the transabdominal procedure was performed in patients for whom the transcervical technique was not possible. However, the ability to provide diagnostic studies for a greater number of patients prompted us to incorporate the technique into our routine protocol. The transvesicaltransabdominal technique was incorporated for the same reasons; patients who could not undergo CVS by one of the standard techniques and who would be denied first-trimester testing were offered this alternative approach. The natural miscarriage rate for documented viable pregnancies between 9 and 24 weeks' gestation at different maternal ages is currently estimated at 2% to 4%.13-1; The actual risk of miscarriage for the two CVS techniques will be established from data collected from randomized trials that are underway in the United States and other countries. Our uncorrected loss rate in the first 100 transcervical procedures was 8.8%. After this learning period, the transcervical fetal loss rate decreased to 2.7%. For transabdominal CVS, the uncorrected fetal loss rate was 2.6%, indicating no significant increase with the addition of the procedure. Our findings are similar to published studies regarding the transabdominal technique. Smidt-Jensen and Hahnemann 16 reported on 100 patients undergoing transabdominal procedures with a 100% success rate, 98% of which were successful on the first attempt. They report an uncorrected fetal loss rate of 3.2%. Therkelsen et alY reported a 3.5% fetal loss rate in 350 patients undergoing transabdominal procedures that included the cases of Smidt-Jensen and Hahnemann. In a randomized study of 120 patients, Bovicelli et al. 18 reported a 1.7% failure rate for both transcervical and transabdominal techniques and identical fetal loss rates of 3.5%. In a later series of transabdominal procedures in 350 patients, Bovicelli et al. 19 reported only one failed procedure with 97% success on the first attempt. The fetal loss rate in this series was 2.7%. Lilford et al. tO reported a 98% success rate in 100 patients, 94% with
Transabdominal CVS
1293
one outer-needle insertion, and no fetal losses to date; 18 of the patients were past 14 weeks' gestation. Brambati et al. i l reported on 1159 patients undergoing transabdominal procedures with no failures and 96% success on the first attempt. The total fetal loss rate was 3.4%, but the authors judged only 2.4% of losses were procedure related. From the CVS registry maintained by Jackson,1O centers with more than 100 transabdominal cases reported success rates of 97.4% to 100% and uncorrected fetal loss rates of 0.7% to 4.3%. Small tissue weights have been a concern with transabdominal CVS. Both Bovicelli et al. 18 and Lilford et aUo commented on this, although no amounts were given. In their second patient series Bovicelli et al. 19 noted increased tissue weights with 64% of samples ~ 10 mg. However, 9.4% were <5 mg and only 5.6% were >20 mg. Brambati et aUI also reported difficulty early in their series with small tissue weights. Changing to a needle with a wider inner diameter and increasing the syringe suction produced tissue weights of ~ 10 mg in 97% of cases and ~20 mg in 75% of cases. The patient series of Smidt-Jensen and Hahnemann l6 and Therkelsen et al.17 showed weights of 10 to 50 mg, with a mean sample size of 27 to 28 mg. Our experience supports the latter findings with 89% of samples having tissue weights of ~20 mg. Our large tissue weights may be due to the thin-wall spinal needles, which have a wider inner diameter than standard spinal needles and allow easier aspiration. We experienced only one laboratory failure in the 869 CVS cases, even with samples as small as 5 mg. In general, however, amounts of 10 to 20 mg are optimal for direct analysis and to establish cultures!1 Tissue weights of at least 20 to 30 mg may be necessary for direct DNA or biochemical studies on chorionic villi and establishment of cytogenetic and backup cultures. If sufficient chorionic villus material is not available for direct DNA analysis, culturing of cells may delay results by 2 to 3 weeks. Use of the transabdominal technique did not reduce the tissue weights obtained or delay results in our patient series. In our experience the transabdominal approach was a quicker and less complicated procedure than the transcervical procedure. Sterile preparation of the abdominal insertion site is faster and more thorough than preparation of the vaginal vault and cervix. Patients are more comfortable lying supine than in the dorsal lithotomy position; therefore they are able to relax more and maintain a stationary position. Bovicelli et al. IB also reported that the transabdominal procedure was easier and quicker. Lilford et al!O stated there was a longer preparation for transabdominal CVS at their clinic and that 3% of patients required vaginal manipulation to reposition the uterus; this was not necessary in our transabdominal group. In a brief report by Monni et al."" of 72 women who had undergone transcervical
1294 Copeland et al.
CVS in one pregnancy and transabdominal CVS in a subsequent pregnancy, 71 of the women preferred the transabdominal method. We believe there are two primary reasons for the high success rate qnd low miscarriage rate in our transabdominal group. First, before starting the transabdominal CVS method the operator had extensive experience with ultrasonographically directed needle procedures, including amniocentesis, percutaneous umbilical cord sampling, and fetal blood transfusion. Second, the transabdominal method was incorporated after the operator had already performed more than 300 transcervical CVS procedures in continuing pregnancies. Some of the difficulties operators experience when starting CVS may not be a result of the sampling method used (transcervical versus transabdominal) but may reflect limited experience with early pregnancies. Ultrasonographic visualization, uterine position, and placental localization are important varjab~es in determining how the sampling instrument (catheter or needle) is placed in the chorion. Correct placement is established only when the material is aspirated and villi are seen. The operators' experience with transcervical CVS clearly facilitated the use of the transabdominal method. In summary, the transabdominal technique was readily incorporated into our CVS program without an incr~ase in fetal loss or diagnostic failures. Operators with experience in transcervical CVS and ultrasonographically directed needle procedures (includ5ng amniocentesis, fetal· blood sampling, and intrauterine transfusions) should be able to incorporate tQe transabdominal method into their CVS progr,!m rapidly and effectively. We do not believe transabdominal CVS will completely replace transcervical CV~. Operators should consider using both techniques to increase the availability of CVS to the maximum number of patients. We thank Kenneth Moise, MD, Fred Elder, PhD, Mildred Crosby, RN, Cathy Rayburn, RN, and Carolyn Karp for their assistance in th~ Baylor CVS project.
November 1989 Am J Obstet Gynecol
4. 5.
6. 7.
8.
9. 10. 11. 12. 13.
14. 15.
16. 17.
18. 19.
REFERENCES 1. GosdenJR, Mitchell AR, Gosden CM, Rodeck CH, Morsman JM. Direct vision chorion biop~y and chromosome specific DNA probes for determination of fetal sex in first trimester prenatal diagnosis. Lancet 1982;2:1416-9. 2. Simoni G, Brambati B, Danesino C, et al. Efficient direct chromosQme analy~is and enzyme determin'!tion from chorionic villi samples in the first trimester of pregnancy. Hum Genet 1983;63:349-57. 3. Elias S, Simpson JL, Martin AO, et al. Chorionic villus
20. 21. 22.
sampling in continuing pregnancies. I. Low fetal loss rates in initial 109 cases. AMJ OBSTET GYNECOL 1986; 154: 124952. Hogge WA, Schonberg SA, Golbus MS. Chorionic villus sampling: experience of the first 1000 cases. AM J OBSTET GYNECOL 1986; 154: 1249-52. Czpepulkowski Bli, Heaton DE, Kearney LU, Rodeck CH, Coleman DV. Chorionic villus culture for first trimester diagnosis of chromosome defects: evaluation by two London centers. Prenat Diagn 1986;6:271-82. Brambati B, Oldrini A, Ferrazzi E, Lanzani A. Chorionic villus sampling: an analysis of the obstetric experience of 1000 cases. Prenat Diagn 1987;7:157-69. Brambati B, Oldrini A, Ferazzi E, Lanzani A. Chorionic villus sampling: general methodological and clinical approach. In: Fraccaro M, Simoni G, Brambati B, eds. First trimester fetal diagnosis. Berlin: Springer-Verlag, 1985:718. Rhoads GG, Jackson LG, Schlesselman SE, et al. The safety and efficacy of chorionic villus sampling for early prenatal diagnosiS of cytogenetic abnormalities. N Engl J Med 1989;320:609-17. Canadian collaborative CVS-amniocentesis clinical trial group. Multicentre randomised clinical trial of chorion villus sampling and amniocentesis. Lancet 1989; 1: 1-6. Jackson L. Chorionic Villi Sampling Newsletter. Philadelphia: Jefferson Medical College of Thomas Jefferson University, 1988:26. Smidt-Jensen S, Hahnemann N. Transabdominal fine needle biopsy frol11 chorionic villi in the first trimester. Prenat Diagn 1984;4: 163-9. Brambati B, Simoni G, Danesino C, et al. First trimester fetal diagnosis of genetic disorders: clinical evaluation of 250 cases. J Med Genet 1985;22:92-9. Liu DTY,Jeavons B, Preston C, Pearson D. A prospective study of spontaneous miscarriage in ultrasonographically normal pregnancies and relevance to chorionic villus sampling. Premit Diagn 1987;7:223-7. Gilmore DH, McNay MB. Spontaneous fetal loss rate in early pregnancy. Lancet 1985; 1: 107. Wilson RD, Kendrick V, Wittman NBK, McGillivray BC. Spontaneous abortion and pregnancy outcpmc; after normal first trimester ultrasound examination. Obstet Gynecol 1986;67:352-5. . Smidt-Jensen S, Hahnemann N. Transabdominal chorionic villus sampling for fetal genetic diagnosis. Prenat Diagn 1988;8:7-17. Therkelsen AJ,Jensen PKA, HertzJM, Smidt-Jensen S, Hahnemann N. Prenatal cytogenetic diagnosis after transabdominal chorionic villus sampling in the first trimester. Prenat Diagn 1988;8:19-31. Bovicelli L, Rizzo N, Montacuti V, Morandi R. Transabdominal versus transcervical routes for chorionic villus sampling. Lancet 1986;2:290. Bovicelli L, Rizzo N, Montacuti V, et al. Transabdominal chorionic villus sampling: analysis of 350 consecutive cases. Pr!,!nat Diagn 1988;8:495-500. Lilford RJ, Irving HC, Linton G, Mason MK. Transabdominal chorion villus bippsy: 100 consecutive cases. Lancet 1987; 1: 1415~17. Brambati B, Lanzani A, Oldrini A. Transabdominal chorionic villus sampling: clinical experience of 1159 cases Pre nat Diagn 1988;8:609-17. Monni G, Olla G, Cao A. Patient's choice between transcervical and transabdominal chorionic villus sampling. Lancet 1988;1:1057.
BG, Zaneveld LJD. Development of an assay to assess the functional integrity of the human sperm membrane and its relationship to other semen characteristics. J Reprod Fertil 1984;70:219-28. 8. Collins JA. Diagnostic assessment of the infertile male partner. Curr Probl Obstet Gynecol Infertil 1987; 10: 177224. 9. Van der Ven HH, Jeyendran RS, AI-Hasani S, PerezPelaez M, Diedrich K, Zaneveld LJD. Correlation between human sperm swelling in hypoosmotic medium (hypoos-
Semen predictors of in vitro fertilization
motic swelling test) and in vitro fertilization. J Androl 1986;7:190-6. 10. Yanagimachi R, Yanagimachi H, Rogers BJ. The use of zona-free animal ova as a test system for the assessment of the fertilizing capacity of human spermatozoa. Bioi Reprod 1976; 15:471-6. II. Chan SYW, Fox EJ, Chan MMC, et al. The relationship between the human sperm hypoosmotic swelling test, routine semen analysis, and the human sperm zona-free hamster ovum penetration assay. Fertil Steril 1985;44:668-72.
Integration of the transabdominal technique into an ongoing chorionic villus sampling program Karen L. Copeland, MS, Robert J. Carpenter, Jr., MD, Kurt R. Fenolio, MS, and David H. Ledbetter, PhD Houston, Texas Data are presented on 869 patients undergoing chorionic villus sampling procedures by one of two sampling techniques: 544 by a transcervical catheter aspiration method and 325 by a transabdominal two-needle aspiration method. The transcervical approach was the only procedure used in the first 330 cases, at which time the transabdominal technique was incorporated into our program. After an initial learning curve in the first 100 procedures the transcervical fetal loss rate stabilized at 2.7%, the number of patients requiring more than one catheter insertion decreased to 11 %, and tissue weights 2:10 mg were obtained in 88% of cases. The fetal loss rate for transabdominal chorionic villus sampling was 2.6%, indicating the addition of this new method did not significantly alter the fetal loss rate. Transabdominal chorionic villus sampling had an overall success rate of 99%, with only one insertion of the guide needle required for 98% of patients. Tissue weights of 2:10 mg were obtained in 99% of cases. These results demonstrate that the transabdominal procedure can be rapidly and effectively incorporated by an operator already experienced with transcervical chorionic villus sampling. Since several contraindications exist for either chorionic villus sampling method, the availability of both techniques at a single center greatly enhances the ability to offer first-trimester fetal diagnosis to a majority of patients. (AM J OBSTET GVNECOL 1989;161 :1289-94.)
Key words: Prenatal diagnosis, chorionic villus sampling, transcervical CVS, transabdominal CVS In the early 1980s reports first appeared on the ability to perform first-trimester prenatal diagnostic studies by chorionic villus sampling (CVS).1.2 Within a few years there were several large patient series reporting the successful application of CVS, indicating that chorionic villi can be used reliably for cytogenetic, biochemical, and deoxyribonucleic acid analyses. 3. 6 There are two From the InstItute for Molecular GenetIcs and the Department of Obstetrics and G.vnecology, Baylor College of MedIcine. This work was supported in part by grant No. HD19881 from the NatIOnal InstItute of Child Health and Human Development. Received for publicatIOn January 23,1989; revised May 2,1989; accepted May 25, 1989. Reprmt requests: Karen L. Copeland, MS, Institute for Molecular GenetICS, Baylor College of Medlcme. One Baylor Plaza, Houston,
TX 77030.
6/1 / 14236
major sampling techniques for obtaining chorionic villis currently in use: ultrasonographically guided transcervical catheter aspiration and ultrasonographically guided transabdominal needle aspiration (singleand double-needle systems).7 The transcervical technique has had the greatest utilization internationally and until recently was the only technique in use in the United States. Multicenter studies from the United States and Canada found a slightly increased rate of fetal loss (0.6% and 0.8%, respectively) for transcervical CVS over amniocentesis,"' 9 but these differences were not statistically significant. In addition, no significant maternal complications were noted after transcervical CVS. The transabdominal technique has been used by several groups in Europe and its use is rapidly growing in the U.S. IU 1289
1290 Copeland et al.
November 1989 Am J Obstct Gynecol
Table I. Transabdominal placental location Locatwn
Anterior Posterior Lateral Fundal Previa
n
164 117 16 16 12
Table II. Gestational age at CVS %
50 36 5
5 4
We report our experience with the transcervical and transabdominal techniques to illustrate that the transabdominal method can be readily incorporated into an ongoing transcervical program. Since contraindications exist for either technique, there are benefits for using both methods at a single center. Material and methods
The first 869 patients undergoing CVS procedures for prenatal diagnosis at our center are described, including 544 performed by the transcervical method and 326 by the transabdominal method . Risk of cytogenetic disorders was the indication in 91 % of cases, with maternal age the most frequent indication (80%). In 9% of cases, patients were at increased risk for a specific genetic disease in which either DNA analysis (7%) or biochemical analysis (2%) was performed. Cytogenetic analysis was attempted by both direct preparation 2 and long-term culture in all cases with ~ 10 mg. Direct and culture methods were successful in 85 % of cases, direct method only in 2%, and culture method only in 13%. Before the procedure, all women underwent an ultrasonographic evaluation to determine fetal viability, gestational age, and number of fetuses present; twin pregnancies were excluded from this study. Most patients in our referral area had ultrasonographic evaluation performed at the time of genetic counseling. According to a protocol approved by the Baylor College of Medicine Institutional Review Board, informed consent was obtained from all women before CVS. Our institution is one of eight centers participating in the Cooperative Clinical Comparison of Chorionic Villus Sampling and Amniocentesis sponsored by the National Institute of Child Health and Human Development to study the safety and accuracy of chorionic villus sampling. 8 Ultrasonographic evaluation at the time of the procedure established placental position and determined whether the transcervical or transabdominal approach could be used. Patients were excluded from either CVS procedure if there was active bleeding within I week of the scheduled procedure or preexisting rhesus alloimmunization. Transcervical CVS was not performed in women with active vaginal infections or cervical stenosis. Transabdominal CVS was not performed in
Weeks' gestatwn
9.0-9.9 10.0-10.9 11.0-11.9 12.0-12.9 13.0-13.9 Total no. of patients
Transcerv!cai n
135 277 116 15 1 544
I
Transabdominal
%
n
25 51 21 3
60 161 66 35 3 325
I
%
18 50 20 11 1
women with bowel visualized between the abdomen and uterine wall. Both CVS techniques were performed by the same operator and a team of ultrasonographers and nurses. The determination of whether to perform CVS and which method was to be used for each patient was made by the operator. Patients who could undergo CVS by either technique were randomized between the transcervical and transabdominal methods as a component of the collaborative study. The transcervical technique was performed with the patient in the lithotomy position. A short speculum was inserted into the vagina after the vagina and external genitalia were cleansed with Betadine solution. A single-toothed tenaculum was placed on the anterior cervix in one third of patients. With continuous real-time ultrasonographic guidance provided by an Acuson-128 scanner with a 3 .5 MHz linear transducer, a Portex catheter was inserted through the cervix into the chorion frondosum . A 20 ml Becton-Dickinson Luer lock syringe with 1 to 2 ml of aspiration media (minimal essential medium without fetal calf serum) was attached and villi were aspirated with hand suction (10 to 12 ml) as the catheter was withdrawn. The transabdominal technique was performed with the patient in the supine position. After selection of the needle insertion site, the area was cleansed with Betadine solution and draped with sterile towels. Xylocaine (3 ml of 1 % solution) was administered to decrease maternal discomfort. With continuous real-time ultrasonographic guidance provided by an Acuson-128 scanner with a 3.5 MHz sector transducer, we carried out a modificiation of the double-needle procedure described by Smidt-Jensen and HahnemannIl using Thin Wall Quinke spinal needles of 18 gauge (3.5 em) and 20 gauge (5.5 em). By freehand placement, the 18gauge spinal needle was directed through the maternal abdominal wall and myometrium with the tip inserted just beyond the decidual-chorion interface. The obturator was then removed and the 20-gauge needle was inserted through the outer (introducer) needle, into the trophoblastic tissue, and then moved back and forth. Villi were aspirated as described for the transcervical procedure. The transabdominal procedures early in
Transabdominal CVS
Volume 161 Number 5
1291
Table III. Insertions to obtain tissue* Transabdomnwl: Procedures 1-325
Transcervical Procedures 101-544
Procedures 1-100 (n)
No. of insertions
54 30
I
2
396 41 7
14 2
3 4
I
n
o
%
n
%
89 9 2
291
90 10
31
2 I
*Transcervical indicates catheter insertions. Transabdominal indicates inner-needle insertions.
Table IV. Tissue weight obtained*
I
Transcervical Amount of tISSue
Failed <5 mg 5-9 mg 10-19 mg ~20 mg
Procedures 1-100 I
5
19 59 13
Procedures 101 -544
3 2 7 20 68
Table V. Fetal losses Transcervical
Transabdominal: Procedures
Procedures 1-100
1-325
93
*Values represent percent of patients. our experience were limited to anterior placentas and, as experience increased, more lateral and then posterior placentas were sampled. Placental localization is noted in Table 1. A transabdominal-trans vesical technique was performed in 21 patients. In these patients a transcervical approach was not possible because of a severely retroflexed uterus and anterior placenta. The maternal bladder was obstructing the only viable transabdominal approach, and emptying and filling of the maternal bladder did not alter the feasibility of either approach. The IS-gauge needle was inserted through the maternal abdominal wall, the maternal bladder, the myometrium, and into the chorion frondosum . Results
Procedures. The transcervical technique was performed in 544 patients and the transabdominal technique was performed in 325 patients between 9 and 13 weeks' gestation (Table II). While 70% of patients were between 10 and 11 weeks' gestation in both the transcervical and transabdominal groups, 12% of the transabdominal group were 2!:: 12 week's gestation, compared with 3% of the transcervical group. Early in the transcervical series (I to 100), multiple catheter insertions were necessary in 46% of cases to obtain adequate tissue samples (Table III). As operator experience increased (101 to 544), only II % of patients required multiple catheter insertions to obtain an adequate sample. There were 15 transcervical pro-
TAB-VTP Continuing <16 wk 16-20 wk 20·24 wk Total losses
9*
91 4 2 2 8 (8.8%)
I Procedures 101-544 31 413 5 4 2 II (2.7%)
Transabdominal: Procedures
1-325
17 308 4 I 3 8 (2.6%)
TAB-VTP, Therapeutic abortion-voluntary termination of pregnancy. *lncludes two miscarriages with genetic anomalies: trisomy 13 and male fetus affected with hemophilia.
cedures (2 .S%) in which no tissue was obtained. Eleven of the patients with transcervical failure underwent amniocentesis whereas four underwent transabdominal CVS. Tissue was obtained in 324 of 325 transabdominal cases. In 31S transabdominal cases (9S%) adequate tissue was obtained with only one insertion of the outer needle; 7 patients required a second insertion of the needle to obtain tissue. The inner needle was reinserted in 10% of cases to obtain additional tissue. Tissue obtained in two of the transabdominal procedures was of an unusual appearance and only 46,XX cells were seen after cytogenetic analysis; because of concerns of maternal cell contamination, repeat diagnostic studies were recommended. One patient underwent a repeat transabdominal procedure 2 weeks later and the other underwent amniocentesis; in both cases the 46,XX result was confirmed. One patient had a failed transabdominal CVS and underwent a successful transcervical procedure at a later date. All 21 transabdominaltransvesical procedures were successful. The transcervical method yielded small tissue weights in the first 100 procedures, with 24% of samples <10 mg and only 13% 2!::20 mg (Table tV). In the subsequent 444 procedures only 9% of samples were < 10 mg and 6S% were 2!::20 mg. The transabdominal
1292 Copeland el al.
November 1989 Am J Obslel Cynecol
Table VI. Pregnancy outcome > 24 weeks' gestation Transcervical
n
Continuing Live birth Term
<37 wk
<2500 gm Intrauterine growth retardation Stillbirth Lost to follow-up
485 479 443
I
23* 10
3
2 4
Transabdommal
%
n
I
Total
%
785
300
92
5 2
295
280 11 * 2 2
2 3
n
95
4
774 723
34
12 5 4 7
I
%
93 4
2
*Neonatal death in one transcervical procedure and one transabdominal procedure.
technique yielded adequate sample sizes with only 3% of samples <10 mg and 93% ~20 mg. Twenty-three patients (2.6%) underwent a subsequent amniocentesis because of diagnostic failures or ambiguities. Nineteen procedures were performed for cytogenetic reasons: 14 cases of mosaicism of potential clinical signifciance (4 of which were confirmed), 1case of significant maternal cell contamination, 1 culture failure, and 3 cases with cytogenetic information limited to only one culture. One patient had repeatedly elevated serum et-fetoprotein levels. Three patients had additional studies for DNA or biochemical diagnoses. Immediate complications. For patients undergoing transcervical CVS, approximately 30% experienced spotting and bleeding after the procedure. In most patients this resolved within 24 hours and a brown discharge occurred for 1 to 2 days thereafter. Mild cramping occurred occasionally. For patients undergoing transabdominal CVS, cramping was common but resolved by the following morning. Bleeding or spotting was infrequent (5%) . Eleven patients (3%) had a vasovagal reaction immediately after the transabdominal procedure involving dizziness, bradycardia, flushing, and diaphoresis. This resolved in a few minutes if the patient was kept in a supine position. There were no recurrences of the vasovagal symptoms. Only one patient who underwent a transabclominal-transvesical procedure experienced a transient episode of minimal hematuria. Bleeding, cramping, and vasovagal episodes did not correlate with risk of fetal loss. Fetal loss. The transcervical procedure was performed in 544 patients, and 40 of these pregnancies were terminated because of a diagnosed genetic abnormality or for social reasons. In the continuing 504 pregnancies a total of 19 losses with normal chromosome analysis occurred from the time of the transcervical procedure to 24 weeks' gestation (Table V). Eight occurred in the first 100 procedures, for an uncorrected loss rate of 8.8% . In the subsequent 444 procedures there were 11 losses for an uncorrected fetal loss rate of 2.7%. In the 325 transabdominal procedures 17 pregnancies were terminated because of a
genetic abnormality or for social reasons. In the remaining 308 continuing pregnancies eight spontaneous losses occurred at <24 weeks, for an uncorrected fetal loss rate of2.6%. Four patients underwent unsuccessful transcervical CVS followed by a transabdominal procedure, and one had an unsuccessful transabdominal CVS followed by a transcervical procedure; none of these pregnancies miscarried. Laboratory findings. Twenty-one cases of cytogenetic abnormalities were detected : 14 with aneuploidy, 4 with unbalanced translocations, I with triploidy, and 4 with mosaicism confirmed in fetal tissue. Fifteen patients had abnormal results from DNA or biochemical studies. In all cases of abnormal diagnosis the pregnancies were terminated except for one case of sickle cell disease. Results were confirmed in all cases where tissue was obtained. There have been no false-positive or false-negative results in our 869 CVS cases. However, in the four cases of fetal mosaicism the mosaicism was not detected in the direct preparation and was seen only in the cultured cells. Outcome. All pregnancies have been delivered. Information is available for 99% of the 785 continuing pregnancies (Table VI) . There are no significant differences between the transcervical or transabdominal methods in the incidence of prematurity, low birth weight, intrauterine growth retardation, neonatal death, or stillbirth.
Comment Our experience with 325 transabdominal CVS procedures indicates it is a viable alternative for providing first-trimester prenatal diagnosis to operators currently using transcervical CVS. Our transabdominal fetal loss rate of 2.6% and success rate of 99% demonstrate that the addition of the transabdominal technique did not increase the overall risks of CVS at our center. Internationally, CVS has been performed primarily by the transcervical technique, which appears to be a reliable method of prenatal diagnosis. s." Transabdominal CVS was first described in 1984 by Smidt-Jensen and Hahnemann" and may have several theoretical
Volume 151 Number 5
and practical advantages. First, transabdominal CVS avoids the likelihood that organisms present in the vagina and cervix will be introduced into the uterus. Second, transcervical CVS is contraindicated in patients with active vaginal infections, in cases of cervical stenosis, or in cases in which the procedure would be very difficult because of placental position and uterine anatomy. These patients may be denied first-trimester diagnosis if the transcervical method is the only option available. 12 We initially attempted the transabdominal technique in two patients at high risk for genetic abnormality (cystic fibrosis and Becker muscular dystrophy) who could not undergo CVS transcervically because of a sharply retroverted uterus and an anterior placenta. After these two experiences the transabdominal procedure was performed in patients for whom the transcervical technique was not possible. However, the ability to provide diagnostic studies for a greater number of patients prompted us to incorporate the technique into our routine protocol. The transvesicaltransabdominal technique was incorporated for the same reasons; patients who could not undergo CVS by one of the standard techniques and who would be denied first-trimester testing were offered this alternative approach. The natural miscarriage rate for documented viable pregnancies between 9 and 24 weeks' gestation at different maternal ages is currently estimated at 2% to 4%.13-1; The actual risk of miscarriage for the two CVS techniques will be established from data collected from randomized trials that are underway in the United States and other countries. Our uncorrected loss rate in the first 100 transcervical procedures was 8.8%. After this learning period, the transcervical fetal loss rate decreased to 2.7%. For transabdominal CVS, the uncorrected fetal loss rate was 2.6%, indicating no significant increase with the addition of the procedure. Our findings are similar to published studies regarding the transabdominal technique. Smidt-Jensen and Hahnemann 16 reported on 100 patients undergoing transabdominal procedures with a 100% success rate, 98% of which were successful on the first attempt. They report an uncorrected fetal loss rate of 3.2%. Therkelsen et alY reported a 3.5% fetal loss rate in 350 patients undergoing transabdominal procedures that included the cases of Smidt-Jensen and Hahnemann. In a randomized study of 120 patients, Bovicelli et al. 18 reported a 1.7% failure rate for both transcervical and transabdominal techniques and identical fetal loss rates of 3.5%. In a later series of transabdominal procedures in 350 patients, Bovicelli et al. 19 reported only one failed procedure with 97% success on the first attempt. The fetal loss rate in this series was 2.7%. Lilford et al. tO reported a 98% success rate in 100 patients, 94% with
Transabdominal CVS
1293
one outer-needle insertion, and no fetal losses to date; 18 of the patients were past 14 weeks' gestation. Brambati et al. i l reported on 1159 patients undergoing transabdominal procedures with no failures and 96% success on the first attempt. The total fetal loss rate was 3.4%, but the authors judged only 2.4% of losses were procedure related. From the CVS registry maintained by Jackson,1O centers with more than 100 transabdominal cases reported success rates of 97.4% to 100% and uncorrected fetal loss rates of 0.7% to 4.3%. Small tissue weights have been a concern with transabdominal CVS. Both Bovicelli et al. 18 and Lilford et aUo commented on this, although no amounts were given. In their second patient series Bovicelli et al. 19 noted increased tissue weights with 64% of samples ~ 10 mg. However, 9.4% were <5 mg and only 5.6% were >20 mg. Brambati et aUI also reported difficulty early in their series with small tissue weights. Changing to a needle with a wider inner diameter and increasing the syringe suction produced tissue weights of ~ 10 mg in 97% of cases and ~20 mg in 75% of cases. The patient series of Smidt-Jensen and Hahnemann l6 and Therkelsen et al.17 showed weights of 10 to 50 mg, with a mean sample size of 27 to 28 mg. Our experience supports the latter findings with 89% of samples having tissue weights of ~20 mg. Our large tissue weights may be due to the thin-wall spinal needles, which have a wider inner diameter than standard spinal needles and allow easier aspiration. We experienced only one laboratory failure in the 869 CVS cases, even with samples as small as 5 mg. In general, however, amounts of 10 to 20 mg are optimal for direct analysis and to establish cultures!1 Tissue weights of at least 20 to 30 mg may be necessary for direct DNA or biochemical studies on chorionic villi and establishment of cytogenetic and backup cultures. If sufficient chorionic villus material is not available for direct DNA analysis, culturing of cells may delay results by 2 to 3 weeks. Use of the transabdominal technique did not reduce the tissue weights obtained or delay results in our patient series. In our experience the transabdominal approach was a quicker and less complicated procedure than the transcervical procedure. Sterile preparation of the abdominal insertion site is faster and more thorough than preparation of the vaginal vault and cervix. Patients are more comfortable lying supine than in the dorsal lithotomy position; therefore they are able to relax more and maintain a stationary position. Bovicelli et al. IB also reported that the transabdominal procedure was easier and quicker. Lilford et al!O stated there was a longer preparation for transabdominal CVS at their clinic and that 3% of patients required vaginal manipulation to reposition the uterus; this was not necessary in our transabdominal group. In a brief report by Monni et al."" of 72 women who had undergone transcervical
1294 Copeland et al.
CVS in one pregnancy and transabdominal CVS in a subsequent pregnancy, 71 of the women preferred the transabdominal method. We believe there are two primary reasons for the high success rate qnd low miscarriage rate in our transabdominal group. First, before starting the transabdominal CVS method the operator had extensive experience with ultrasonographically directed needle procedures, including amniocentesis, percutaneous umbilical cord sampling, and fetal blood transfusion. Second, the transabdominal method was incorporated after the operator had already performed more than 300 transcervical CVS procedures in continuing pregnancies. Some of the difficulties operators experience when starting CVS may not be a result of the sampling method used (transcervical versus transabdominal) but may reflect limited experience with early pregnancies. Ultrasonographic visualization, uterine position, and placental localization are important varjab~es in determining how the sampling instrument (catheter or needle) is placed in the chorion. Correct placement is established only when the material is aspirated and villi are seen. The operators' experience with transcervical CVS clearly facilitated the use of the transabdominal method. In summary, the transabdominal technique was readily incorporated into our CVS program without an incr~ase in fetal loss or diagnostic failures. Operators with experience in transcervical CVS and ultrasonographically directed needle procedures (includ5ng amniocentesis, fetal· blood sampling, and intrauterine transfusions) should be able to incorporate tQe transabdominal method into their CVS progr,!m rapidly and effectively. We do not believe transabdominal CVS will completely replace transcervical CV~. Operators should consider using both techniques to increase the availability of CVS to the maximum number of patients. We thank Kenneth Moise, MD, Fred Elder, PhD, Mildred Crosby, RN, Cathy Rayburn, RN, and Carolyn Karp for their assistance in th~ Baylor CVS project.
November 1989 Am J Obstet Gynecol
4. 5.
6. 7.
8.
9. 10. 11. 12. 13.
14. 15.
16. 17.
18. 19.
REFERENCES 1. GosdenJR, Mitchell AR, Gosden CM, Rodeck CH, Morsman JM. Direct vision chorion biop~y and chromosome specific DNA probes for determination of fetal sex in first trimester prenatal diagnosis. Lancet 1982;2:1416-9. 2. Simoni G, Brambati B, Danesino C, et al. Efficient direct chromosQme analy~is and enzyme determin'!tion from chorionic villi samples in the first trimester of pregnancy. Hum Genet 1983;63:349-57. 3. Elias S, Simpson JL, Martin AO, et al. Chorionic villus
20. 21. 22.
sampling in continuing pregnancies. I. Low fetal loss rates in initial 109 cases. AMJ OBSTET GYNECOL 1986; 154: 124952. Hogge WA, Schonberg SA, Golbus MS. Chorionic villus sampling: experience of the first 1000 cases. AM J OBSTET GYNECOL 1986; 154: 1249-52. Czpepulkowski Bli, Heaton DE, Kearney LU, Rodeck CH, Coleman DV. Chorionic villus culture for first trimester diagnosis of chromosome defects: evaluation by two London centers. Prenat Diagn 1986;6:271-82. Brambati B, Oldrini A, Ferrazzi E, Lanzani A. Chorionic villus sampling: an analysis of the obstetric experience of 1000 cases. Prenat Diagn 1987;7:157-69. Brambati B, Oldrini A, Ferazzi E, Lanzani A. Chorionic villus sampling: general methodological and clinical approach. In: Fraccaro M, Simoni G, Brambati B, eds. First trimester fetal diagnosis. Berlin: Springer-Verlag, 1985:718. Rhoads GG, Jackson LG, Schlesselman SE, et al. The safety and efficacy of chorionic villus sampling for early prenatal diagnosiS of cytogenetic abnormalities. N Engl J Med 1989;320:609-17. Canadian collaborative CVS-amniocentesis clinical trial group. Multicentre randomised clinical trial of chorion villus sampling and amniocentesis. Lancet 1989; 1: 1-6. Jackson L. Chorionic Villi Sampling Newsletter. Philadelphia: Jefferson Medical College of Thomas Jefferson University, 1988:26. Smidt-Jensen S, Hahnemann N. Transabdominal fine needle biopsy frol11 chorionic villi in the first trimester. Prenat Diagn 1984;4: 163-9. Brambati B, Simoni G, Danesino C, et al. First trimester fetal diagnosis of genetic disorders: clinical evaluation of 250 cases. J Med Genet 1985;22:92-9. Liu DTY,Jeavons B, Preston C, Pearson D. A prospective study of spontaneous miscarriage in ultrasonographically normal pregnancies and relevance to chorionic villus sampling. Premit Diagn 1987;7:223-7. Gilmore DH, McNay MB. Spontaneous fetal loss rate in early pregnancy. Lancet 1985; 1: 107. Wilson RD, Kendrick V, Wittman NBK, McGillivray BC. Spontaneous abortion and pregnancy outcpmc; after normal first trimester ultrasound examination. Obstet Gynecol 1986;67:352-5. . Smidt-Jensen S, Hahnemann N. Transabdominal chorionic villus sampling for fetal genetic diagnosis. Prenat Diagn 1988;8:7-17. Therkelsen AJ,Jensen PKA, HertzJM, Smidt-Jensen S, Hahnemann N. Prenatal cytogenetic diagnosis after transabdominal chorionic villus sampling in the first trimester. Prenat Diagn 1988;8:19-31. Bovicelli L, Rizzo N, Montacuti V, Morandi R. Transabdominal versus transcervical routes for chorionic villus sampling. Lancet 1986;2:290. Bovicelli L, Rizzo N, Montacuti V, et al. Transabdominal chorionic villus sampling: analysis of 350 consecutive cases. Pr!,!nat Diagn 1988;8:495-500. Lilford RJ, Irving HC, Linton G, Mason MK. Transabdominal chorion villus bippsy: 100 consecutive cases. Lancet 1987; 1: 1415~17. Brambati B, Lanzani A, Oldrini A. Transabdominal chorionic villus sampling: clinical experience of 1159 cases Pre nat Diagn 1988;8:609-17. Monni G, Olla G, Cao A. Patient's choice between transcervical and transabdominal chorionic villus sampling. Lancet 1988;1:1057.