- Email: [email protected]
Multifetal pregnancy reduction in cases of threatened abortion of triplets
FERTILITY AND STERILITY威 VOL. 72, NO. 3, SEPTEMBER 1999 Copyright ©1999 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.
PREGNANCY AND BREAST-FEEDING
Multifetal pregnancy reduction in cases of threatened abortion of triplets Josef Shalev, M.D.,*†‡ Israel Meizner, M.D.,*†‡ Reuven Mashiach, M.D.,*†‡ Itai Bar-Chava, M.D.,*‡ and Zion Ben Rafael, M.D.*‡ Rabin Medical Center—Beilinson Campus, Petah-Tiqva, and Tel-Aviv University, Tel-Aviv, Israel
Objective: To investigate the course of pregnancy and fetal outcome after first-trimester multifetal pregnancy reduction (MFPR) in patients with triplet pregnancies and uterine bleeding. Design: Case series of patients with threatened triplet pregnancies considered for MFPR. Setting: Department of Obstetrics and Gynecology, Rabin Medical Center, Petah-Tiqva, Israel. Patient(s): Forty-two patients with triplet pregnancies and first-trimester uterine bleeding. Intervention(s): At 10 –15 weeks’ gestation, MFPR with intracardiac injection of potassium chloride was performed. The procedures were performed 7–10 days after cessation of bleeding (9 –13 weeks) or in the presence of minimal uterine bleeding (14 –15 weeks). In patients with heavy uterine bleeding, MFPR was postponed. Main Outcome Measure(s): Early- and late-pregnancy complications related to the procedure, pregnancy outcome, and fetal survival. Result(s): Performance of MFPR at 14 –15 weeks was associated with a higher abortion rate (38.5%), lower mean gestational age at delivery (30.6 weeks), and lower mean twin birth weight (1,376 ⫾ 218 g and 1,014 ⫾ 202 g) than was performance of MFPR at 10 –13 weeks (18.8%, 33.2 weeks, and 1,720 ⫾ 245 g and 1,596 ⫾ 170 g, respectively). Abortion occurred in four of the five patients with moderate to heavy uterine bleeding who did not undergo MFPR; the fifth patient gave birth prematurely at 28 weeks, and two of the newborns died. Conclusion(s): Pregnancy outcome and fetal mortality and morbidity in triplet pregnancy after MFPR are directly correlated with duration and amount of first-trimester bleeding. (Fertil Steril威 1999;72:423– 6. ©1999 by American Society for Reproductive Medicine.) Key Words: Multiple pregnancy, fetal reduction, abortion, ultrasound
Received January 19, 1999; revised and accepted April 20, 1999. Reprint requests: Josef Shalev, M.D., Ultrasound Unit, Department of Obstetrics and Gynecology, Rabin Medical Center—Beilinson Campus, Petah-Tiqva 49100, Israel (FAX: 972-3-9377597). * Department of Obstetrics and Gynecology, Rabin Medical Center—Beilinson Campus. † Ultrasound Unit, Rabin Medical Center. ‡ Sackler Faculty of Medicine, Tel-Aviv University. 0015-0282/99/$20.00 PII S0015-0282(99)00296-4
Selective multifetal pregnancy reduction (MFPR) is now being offered to many patients with triplet pregnancies. Although the procedure has proven safe and effective, there is controversy regarding whether MFPR improves fetal outcome in triplet pregnancies. Three reports have documented an increase in birth weight and gestational age at delivery and a decrease in perinatal mortality and shortterm morbidity in a total of 168 sets of triplets reduced to twins, compared with unreduced triplets (1–3). On the other hand, other studies involving a total of ⬎400 triplet pregnancies treated conservatively showed a good fetal outcome (mean birth weight, 1,796 g; mean gestational age at delivery, 33.5 weeks) (1, 4, 5). The management of triplet pregnancies with early bleeding is even more problematic. More
than 25% of multiple pregnancies are associated with some spotting or bleeding at the time MFPR is generally performed (10 –13 weeks). In these cases, clinicians face a dilemma: should MFPR be performed despite a threatened abortion, postponed to the second trimester, or canceled? In this report, we present our experience with selective fetal reduction in the first trimester in patients with triplet pregnancies at risk of spontaneous abortion. We reviewed the pregnancy course and fetal outcome in patients who underwent MFPR 7–10 days after vaginal bleeding stopped and in those managed conservatively because of intermittent, ongoing heavy uterine bleeding. On the basis of our pregnancy outcome findings, we recommend a tentative policy for MFPR. 423
MATERIALS AND METHODS Patients The study population consisted of 42 patients aged 27–38 years with triplet pregnancies following ovulation induction with hMG (Pergonal; Teva, Kefar Saba, Israel) or IVF. All had uterine bleeding during the first trimester.
Procedure Preliminary transvaginal sonographic scans (model 650.680; Aloka, Japan) were performed in all patients to determine the number and location of gestational sacs; fetal growth, pulse, and nuchal-fold dimensions; the amount of amniotic fluid; and whether blood had accumulated within the uterine cavity.
All patients were observed for 2 hours after MFPR for signs of uterine contractions, increased uterine bleeding, or leakage of amniotic fluid. They were prescribed 1 week of home bed-rest and were instructed to watch for complications such as chorioamnionitis. Monitoring of cardiac activity in the remaining fetuses was performed 1 week after the procedure and, in patients in whom uterine bleeding ceased, monthly until delivery. The treating physician performed the pregnancy follow-up. We assessed pregnancy course and fetal outcome for all patients. The procedures were approved by the institutional review board in accordance with the Helsinki Declaration of 1975 on Human Experimentation.
Statistical Analysis
Patient selection for MFPR was performed according to the following schedule: All 42 patients were evaluated at 9 –10 weeks’ gestation. Sixteen were found eligible for selective reduction (group A). Reassessment of the remaining 26 patients at 11–12 weeks yielded 6 patients eligible for MFPR (group B); 19 were assigned for further follow-up and the remaining woman aborted at 11 weeks. A third assessment was performed at 13–14 weeks, the uppermost limit for first-trimester MFPR. Thirteen patients were found eligible (group C); all still had minimal, intermittent uterine bleeding (resembling the bleeding at the beginning or end of the menstrual period). One patient aborted at 14 weeks, before MFPR. Five patients with moderate to heavy uterine bleeding (group D) were advised to forego selective reduction and proceed with conservative management of pregnancy.
Variables were compared using the Wilcoxon test and the Kruskal-Wallis nonparametric test, as appropriate. All P values are exact (but not corrected for ties).
The patients were advised extensively about the potential risks of the reduction procedure, especially the risk of habitual abortion of pregnancies, and in each case, informed, written consent was obtained before MFPR.
Of the 16 procedures performed at 10 –11 weeks’ gestation (group A), 3 (18.75%) resulted in abortion before 24 weeks. One of these abortions was a septic abortion that occurred 48 hours after genetic amniocentesis. The mean gestational age at the time of delivery for the remaining 13 patients was 32.77 ⫾ 1.36 weeks, and the mean twin birth weights were 1,731.9 ⫾ 159.7 g and 1,558.5 ⫾ 115.1 g.
Technique Multifetal pregnancy reduction from triplets to twins was performed at 10 –15 weeks’ gestation after a bleeding-free period of 7–10 days (weeks 10 –13) or in the presence of minimal uterine bleeding (weeks 14 –15). The proximity of the fetus to the abdominal wall and its uppermost position in the uterine cavity were the criteria for selection for the procedure. The abdomen was cleaned with povidone-iodine and draped with a sterile sheet. No anesthetic medication was administered. Under guidance of direct, continuous ultrasound, a 20-gauge needle, 22 cm long with a stylet (Ismel Co., Ashkelon, Israel) was introduced transabdominally and through the uterine wall into the fetal thorax or, when feasible, the fetal heart. One to four millimoles of potassium chloride (15%, 20 mmol per ampule) were gradually injected; the needle was withdrawn after fetal cardiac asystole had been observed for 30 seconds. A second scan was then performed to determine heart activity in the surviving fetuses. 424
Shalev et al.
MFPR and threat of aborting triplets
RESULTS Of the 42 patients with triplet pregnancies and firsttrimester uterine bleeding, 35 underwent MFPR between 10 and 15 weeks’ gestation. Of the remaining patients, 5 had heavy to moderate uterine bleeding at the end of the first trimester and were advised to avoid selective reduction and 2 aborted in the first trimester while waiting for cessation of uterine bleeding. Pregnancy course and fetal outcome are shown in Table 1. The timing of uterine bleeding and of MFPR is illustrated in Figure 1.
In 6 of the remaining 26 patients, uterine bleeding stopped at 11 weeks and MFPR was performed at 12–13 weeks (group B). There was one abortion before 24 weeks, and five women delivered healthy twins (mean gestational age, 34.20 ⫾ 1.64 weeks; mean birth weights, 2,008.0 ⫾ 203.5 g and 1,816.0 ⫾ 226.8 g). One additional patient aborted at 11 weeks before MFPR was conducted. The other 19 patients had intermittent bleeding up to 13–14 weeks’ gestation, and the decision to perform MFPR was made based on the amount of bleeding. One patient aborted at 14 weeks and 13 underwent selective reduction at 14 –15 weeks (group C). Abortion occurred in 5 of the 13 patients before 24 weeks, one of the abortions occurring 2 weeks after amniocentesis. The remaining eight patients delivered at the beginning of the third trimester, at 30.0 ⫾ 1.31 weeks. Mean twin birth weights were 1,410.7 ⫾ 115.6 g and 1,071.9 ⫾ 106.2 g. Four Vol. 72, No. 3, September 1999
TABLE 1 Pregnancy and fetal outcome after multifetal pregnancy reduction or conservative treatment. Variable No. (%) of patients with pregnancy loss at ⬍24 wk No. of patients who underwent amniocentesis Gestational age at delivery (wk)*† Birth weight (g)‡ Fetus I Fetus II Fetus III
Group A (n ⫽ 16)
Group B (n ⫽ 6)
Group C (n ⫽ 13)
Group D (n ⫽ 5)
3 (18.75) 1 32.77 ⫾ 1.36
1 (16.7) 0 34.20 ⫾ 1.64
5 (38.5) 1 30.00 ⫾ 1.31
4 (80.0) 0 28
1,731.9 ⫾ 159.7 1,558.5 ⫾ 115.1 —
2,008.0 ⫾ 203.5 1,816.0 ⫾ 226.8 —
1,410.7 ⫾ 115.6 1,071.9 ⫾ 106.2 —
900 840 685
Note: Group A: bleeding at ⬍10 weeks, MFPR at 11–12 weeks; group B: bleeding at ⬍12 weeks, MFPR at 13–14 weeks; group C: mild bleeding at 12–14 weeks, MFPR at 15–16 weeks; group D: moderate to heavy bleeding at 12–14 weeks, no MFPR. MFPR ⫽ multifetal pregnancy reduction. * Values are means ⫾ SD or actual age. † The differences among groups A, B, and C were statistically significant (P⬍.0001) for the individual variables. ‡ Values are means ⫾ SD or actual weight. Shalev. Multifetal pregnancy reduction. Fertil Steril 1999.
(80%) of the five patients who underwent conservative treatment (group D) aborted before 24 weeks. The fifth patient delivered prematurely after active uterine contractions at 28 weeks. Birth weights were 900, 840, and 685 g. Only one of the three newborns (the one weighing 840 g) survived, but he remained severely handicapped. Overall, gestational age at delivery and twin birth weights significantly decreased (P⬍.001) as the gestational age at which MFPR was performed increased (Table 2, Fig. 1).
DISCUSSION In most published series of triplet pregnancies treated conservatively, the average gestational age at delivery is 33
FIGURE 1 Pregnancy course and dating of MFPR in 42 patients with triplet pregnancies at risk of abortion. Group A: bleeding at ⬍10 weeks, MFPR at 11–12 weeks; group B: bleeding at ⬍12 weeks, MFPR at 13–14 weeks; group C: mild bleeding at 12–14 weeks, MFPR at 15–16 weeks; group D: moderate to heavy bleeding at 12–14 weeks, no MFPR (N MFPR).
weeks (6); 5%–10% of triplets are delivered before 28 weeks. The average perinatal mortality rate is 77 per 1,000 live births. Little information is available on morbidity and long-term outcome of survivors. Evans et al. (7), who performed first-trimester MFPR in triplet pregnancies, achieved a decrease in the pregnancy loss rate at ⬍24 weeks and a decrease in the number of deliveries before 32 weeks. They also noted a reduction in perinatal mortality, neonatal respiratory complications, and neonatal intraventricular hemorrhage. In a related study, Berkowitz et al. (8) performed selective reduction of 102 sets of triplets to twins. Mean gestational age at birth was 35.9 weeks, with only 2% of patients undergoing delivery before 28 weeks. Loss rates before 24 weeks were reduced to 9%. All of these results were achieved in patients with normal, uncomplicated multiple pregnancies. In patients with uterine bleeding, however, the outcome of delaying MFPR until
TABLE 2 Statistical significance of differences in pregnancy outcome.
Variable Delivery (wk) Birth weight (g) Fetus I Fetus II
Group A versus group B
Group A versus group C
Group B versus group C
.120
.001
.002
.001
.026 .014
.001 .001
.002 .002
⬍.001 ⬍.001
Overall
Note: All P values are exact (but not corrected if identical). All P values were determined by the Wilcoxon nonparametric test. Group A: bleeding at ⬍10 weeks, MFPR at 11–12 weeks; group B: bleeding at ⬍12 weeks, MFPR at 13–14 weeks; group C: mild bleeding at 12–14 weeks, MFPR at 15–16 weeks. MFPR ⫽ multifetal pregnancy reduction. Shalev. Multifetal pregnancy reduction. Fertil Steril 1999.
FERTILITY & STERILITY威
Shalev. Multifetal pregnancy reduction. Fertil Steril 1999.
425
after the bleeding stops (i.e., into the second trimester) was unknown before the current study was performed.
and 88.2% of patients were discharged home with at least one infant weighing (mean ⫾ SD) 2,350 ⫾ 670 g (1).
In our series, 18 (42.7%) of the 42 patients had late uterine bleeding. We selected those patients with minimal bleeding for immediate MFPR and postponed the procedure in those with moderate to heavy bleeding. Our findings indicate that fetal outcome is significantly worse in cases of prolonged uterine bleeding in the first trimester of pregnancy—that is, the greater the duration and amount of uterine bleeding, the higher the fetal loss rate after MFPR. Indeed, in patients with mild bleeding at 13–14 weeks, the abortion rate after MFPR was nearly 40%.
In the current study, four of the five patients with moderate or heavy bleeding who did not undergo MFPR aborted, and the fifth patient lost two of the three newborns after preterm delivery. We assume that performance of MFPR despite continuing bleeding would not have worsened the poor pregnancy outcome of this group. Further studies involving a larger patient sample are needed to support or refute this assumption.
Neonatal morbidity was also higher in this group: both mean gestational age at delivery (30.00 ⫾ 1.31 weeks) and mean twin birth weights (1,410.7 ⫾ 115.6 g; 1,071.9 ⫾ 106.2 g) were significantly decreased. With regard to the effect of the increased dead fetal tissue following late MFPR, studies evaluating second- and third-trimester selective feticide in twins with one malformed fetus have found an uncomplicated pregnancy course (9, 10). Even in patients with uterine bleeding at ⬍12 weeks’ gestation, the abortion rate up to 24 weeks was increased to 16%–19%. This information may assist patients and their physicians when deciding whether to perform this procedure. A limitation of our study is that we did not divide our subjects into those with early-ceasing bleeding and late MFPR, those with early-ceasing bleeding and early MFPR, those with early MFPR despite bleeding, and those with late MFPR because of such bleeding. This limitation resulted from our being unable to recommend that our patients at risk of abortion, who conceived after long-standing infertility, undergo an invasive procedure at threatened abortion. Further data from other centers are requested so that recommendations can be made regarding the timing of MFPR in patients at risk of abortion. One of the main reasons for performing MFPR in triplet pregnancies is the high prematurity rate of 15%–20% in triplet pregnancies. There are high mortality and morbidity rates for infants delivered at 25–28 weeks. The risk-benefit ratios for performing MFPR despite heavy bleeding and for postponing MFPR until bleeding stops have yet to be determined. Data on the natural history of triplets during the first and early second trimesters are incomplete (11, 12). Furthermore, the number of patients with uterine bleeding who have undergone MFPR is small. On the basis of studies of similar procedures such as midtrimester selective feticide and amniocentesis that found pregnancy loss rates of 3.5%–9.3% to the 28th week of gestation (6 –9), we speculate that performing MFPR later than the second trimester may also increase the pregnancy loss rate by up to 10%. This is higher than the rate we found after MFPR in triplets during the first trimester (10 –13 weeks). In that study, the total pregnancy loss rate was ⬍9%, 426
Shalev et al.
MFPR and threat of aborting triplets
In conclusion, the present study shows that first-trimester bleeding in triplet pregnancies is associated with a significantly higher rate of abortion after MFPR. These findings, combined with study findings indicating a higher rate of uterine bleeding in the later stage of the first trimester and during the second trimester (11, 12), justify our management protocol of performing selective reduction as early as possible in patients with triplet pregnancies who are at risk of abortion. In patients with moderate to heavy bleeding, selective feticide remains questionable and should be performed with caution until further studies are completed.
Acknowledgments: The authors thank Mrs. Gloria Ginzach, Editorial Board, Rabin Medical Center, for her editorial assistance and Mrs. Iris Fahn for preparing the manuscript for publication.
References 1. Lipitz S, Reichman BN, Uval J, Shalev J, Achiron R, Barkai G, et al. A prospective comparison of the outcome of triplet pregnancies managed expectantly or by multifetal reduction to twins. Am J Obstet Gynecol 1994;170:874 –9. 2. Boulot P, Hedon B, Pelliccia G, Peray P, Laffargue F, Viala JL. Effects of selective reduction in triplet gestation: a comparative study of 80 cases managed with or without this procedure. Fertil Steril 1993;60: 497–503. 3. Depp R, Macones GA, Rosenn MF, Turzo E, Wapner RJ, Weinblatt VJ. Multifetal pregnancy reduction: evaluation of fetal growth in the remaining twins. Am J Obstet Gynecol 1996;174:1233– 8. 4. Jonas HA, Lumley L. Triplets and quadruplets born in Victoria between 1982 and 1990: the impact of IVF and GIFT on rising birthrates. Med J Aust 1993;158:659 – 63. 5. Newman RB, Hamer C, Clinton MM. Outpatient triplet management: a contemporary review. Am J Obstet Gynecol 1989;161:547–55. 6. Berkowitz RL, Lynch L, Stone J, Alvarez M. The current status of multifetal pregnancy reduction. Am J Obstet Gynecol 1996;174:1265–72. 7. Evans MI, Dommergues M, Timor Trish I, Zador IE, Wapner RJ, Lynch L. Transabdominal versus transcervical and transvaginal multifetal pregnancy reduction: international collaborative experience of more than one thousand cases. Am J Obstet Gynecol 1994;170:902–9. 8. Berkowitz RL, Lauren L, Lapinski R, Bergh P. First trimester transabdominal multifetal pregnancy reduction: a report of two hundred completed cases. Am J Obstet Gynecol 1993;169:17–21. 9. Lipitz S, Shalev E, Meizner I, Yagel S, Weinraub Z, Jaffa A, et al. Late selective termination of fetal abnormalities in twin pregnancies: a multicentre report. Br J Obstet Gynaecol 1996;103:1212– 6. 10. Shalev J, Meizner I, Rabinerson D, Mashiach R, Hod M, Bar-Hava I, et al. Improving pregnancy outcome in twin gestations with one malformed fetus by postponing selective feticide to the third trimester. Fertil Steril 1999; 72:257– 60. 11. Lipitz S, Reichman B, Paret G, Modan M, Shalev J, Serr DM, et al. The improving outcome of triplet pregnancies. Am J Obstet Gynecol 1989; 161:1279 – 84. 12. Danforth DN, Scott JR, ed. Obstetrics and gynecology. 5th ed. Philadelphia: JB Lippincott, 1986;474.
Vol. 72, No. 3, September 1999
PREGNANCY AND BREAST-FEEDING
Multifetal pregnancy reduction in cases of threatened abortion of triplets Josef Shalev, M.D.,*†‡ Israel Meizner, M.D.,*†‡ Reuven Mashiach, M.D.,*†‡ Itai Bar-Chava, M.D.,*‡ and Zion Ben Rafael, M.D.*‡ Rabin Medical Center—Beilinson Campus, Petah-Tiqva, and Tel-Aviv University, Tel-Aviv, Israel
Objective: To investigate the course of pregnancy and fetal outcome after first-trimester multifetal pregnancy reduction (MFPR) in patients with triplet pregnancies and uterine bleeding. Design: Case series of patients with threatened triplet pregnancies considered for MFPR. Setting: Department of Obstetrics and Gynecology, Rabin Medical Center, Petah-Tiqva, Israel. Patient(s): Forty-two patients with triplet pregnancies and first-trimester uterine bleeding. Intervention(s): At 10 –15 weeks’ gestation, MFPR with intracardiac injection of potassium chloride was performed. The procedures were performed 7–10 days after cessation of bleeding (9 –13 weeks) or in the presence of minimal uterine bleeding (14 –15 weeks). In patients with heavy uterine bleeding, MFPR was postponed. Main Outcome Measure(s): Early- and late-pregnancy complications related to the procedure, pregnancy outcome, and fetal survival. Result(s): Performance of MFPR at 14 –15 weeks was associated with a higher abortion rate (38.5%), lower mean gestational age at delivery (30.6 weeks), and lower mean twin birth weight (1,376 ⫾ 218 g and 1,014 ⫾ 202 g) than was performance of MFPR at 10 –13 weeks (18.8%, 33.2 weeks, and 1,720 ⫾ 245 g and 1,596 ⫾ 170 g, respectively). Abortion occurred in four of the five patients with moderate to heavy uterine bleeding who did not undergo MFPR; the fifth patient gave birth prematurely at 28 weeks, and two of the newborns died. Conclusion(s): Pregnancy outcome and fetal mortality and morbidity in triplet pregnancy after MFPR are directly correlated with duration and amount of first-trimester bleeding. (Fertil Steril威 1999;72:423– 6. ©1999 by American Society for Reproductive Medicine.) Key Words: Multiple pregnancy, fetal reduction, abortion, ultrasound
Received January 19, 1999; revised and accepted April 20, 1999. Reprint requests: Josef Shalev, M.D., Ultrasound Unit, Department of Obstetrics and Gynecology, Rabin Medical Center—Beilinson Campus, Petah-Tiqva 49100, Israel (FAX: 972-3-9377597). * Department of Obstetrics and Gynecology, Rabin Medical Center—Beilinson Campus. † Ultrasound Unit, Rabin Medical Center. ‡ Sackler Faculty of Medicine, Tel-Aviv University. 0015-0282/99/$20.00 PII S0015-0282(99)00296-4
Selective multifetal pregnancy reduction (MFPR) is now being offered to many patients with triplet pregnancies. Although the procedure has proven safe and effective, there is controversy regarding whether MFPR improves fetal outcome in triplet pregnancies. Three reports have documented an increase in birth weight and gestational age at delivery and a decrease in perinatal mortality and shortterm morbidity in a total of 168 sets of triplets reduced to twins, compared with unreduced triplets (1–3). On the other hand, other studies involving a total of ⬎400 triplet pregnancies treated conservatively showed a good fetal outcome (mean birth weight, 1,796 g; mean gestational age at delivery, 33.5 weeks) (1, 4, 5). The management of triplet pregnancies with early bleeding is even more problematic. More
than 25% of multiple pregnancies are associated with some spotting or bleeding at the time MFPR is generally performed (10 –13 weeks). In these cases, clinicians face a dilemma: should MFPR be performed despite a threatened abortion, postponed to the second trimester, or canceled? In this report, we present our experience with selective fetal reduction in the first trimester in patients with triplet pregnancies at risk of spontaneous abortion. We reviewed the pregnancy course and fetal outcome in patients who underwent MFPR 7–10 days after vaginal bleeding stopped and in those managed conservatively because of intermittent, ongoing heavy uterine bleeding. On the basis of our pregnancy outcome findings, we recommend a tentative policy for MFPR. 423
MATERIALS AND METHODS Patients The study population consisted of 42 patients aged 27–38 years with triplet pregnancies following ovulation induction with hMG (Pergonal; Teva, Kefar Saba, Israel) or IVF. All had uterine bleeding during the first trimester.
Procedure Preliminary transvaginal sonographic scans (model 650.680; Aloka, Japan) were performed in all patients to determine the number and location of gestational sacs; fetal growth, pulse, and nuchal-fold dimensions; the amount of amniotic fluid; and whether blood had accumulated within the uterine cavity.
All patients were observed for 2 hours after MFPR for signs of uterine contractions, increased uterine bleeding, or leakage of amniotic fluid. They were prescribed 1 week of home bed-rest and were instructed to watch for complications such as chorioamnionitis. Monitoring of cardiac activity in the remaining fetuses was performed 1 week after the procedure and, in patients in whom uterine bleeding ceased, monthly until delivery. The treating physician performed the pregnancy follow-up. We assessed pregnancy course and fetal outcome for all patients. The procedures were approved by the institutional review board in accordance with the Helsinki Declaration of 1975 on Human Experimentation.
Statistical Analysis
Patient selection for MFPR was performed according to the following schedule: All 42 patients were evaluated at 9 –10 weeks’ gestation. Sixteen were found eligible for selective reduction (group A). Reassessment of the remaining 26 patients at 11–12 weeks yielded 6 patients eligible for MFPR (group B); 19 were assigned for further follow-up and the remaining woman aborted at 11 weeks. A third assessment was performed at 13–14 weeks, the uppermost limit for first-trimester MFPR. Thirteen patients were found eligible (group C); all still had minimal, intermittent uterine bleeding (resembling the bleeding at the beginning or end of the menstrual period). One patient aborted at 14 weeks, before MFPR. Five patients with moderate to heavy uterine bleeding (group D) were advised to forego selective reduction and proceed with conservative management of pregnancy.
Variables were compared using the Wilcoxon test and the Kruskal-Wallis nonparametric test, as appropriate. All P values are exact (but not corrected for ties).
The patients were advised extensively about the potential risks of the reduction procedure, especially the risk of habitual abortion of pregnancies, and in each case, informed, written consent was obtained before MFPR.
Of the 16 procedures performed at 10 –11 weeks’ gestation (group A), 3 (18.75%) resulted in abortion before 24 weeks. One of these abortions was a septic abortion that occurred 48 hours after genetic amniocentesis. The mean gestational age at the time of delivery for the remaining 13 patients was 32.77 ⫾ 1.36 weeks, and the mean twin birth weights were 1,731.9 ⫾ 159.7 g and 1,558.5 ⫾ 115.1 g.
Technique Multifetal pregnancy reduction from triplets to twins was performed at 10 –15 weeks’ gestation after a bleeding-free period of 7–10 days (weeks 10 –13) or in the presence of minimal uterine bleeding (weeks 14 –15). The proximity of the fetus to the abdominal wall and its uppermost position in the uterine cavity were the criteria for selection for the procedure. The abdomen was cleaned with povidone-iodine and draped with a sterile sheet. No anesthetic medication was administered. Under guidance of direct, continuous ultrasound, a 20-gauge needle, 22 cm long with a stylet (Ismel Co., Ashkelon, Israel) was introduced transabdominally and through the uterine wall into the fetal thorax or, when feasible, the fetal heart. One to four millimoles of potassium chloride (15%, 20 mmol per ampule) were gradually injected; the needle was withdrawn after fetal cardiac asystole had been observed for 30 seconds. A second scan was then performed to determine heart activity in the surviving fetuses. 424
Shalev et al.
MFPR and threat of aborting triplets
RESULTS Of the 42 patients with triplet pregnancies and firsttrimester uterine bleeding, 35 underwent MFPR between 10 and 15 weeks’ gestation. Of the remaining patients, 5 had heavy to moderate uterine bleeding at the end of the first trimester and were advised to avoid selective reduction and 2 aborted in the first trimester while waiting for cessation of uterine bleeding. Pregnancy course and fetal outcome are shown in Table 1. The timing of uterine bleeding and of MFPR is illustrated in Figure 1.
In 6 of the remaining 26 patients, uterine bleeding stopped at 11 weeks and MFPR was performed at 12–13 weeks (group B). There was one abortion before 24 weeks, and five women delivered healthy twins (mean gestational age, 34.20 ⫾ 1.64 weeks; mean birth weights, 2,008.0 ⫾ 203.5 g and 1,816.0 ⫾ 226.8 g). One additional patient aborted at 11 weeks before MFPR was conducted. The other 19 patients had intermittent bleeding up to 13–14 weeks’ gestation, and the decision to perform MFPR was made based on the amount of bleeding. One patient aborted at 14 weeks and 13 underwent selective reduction at 14 –15 weeks (group C). Abortion occurred in 5 of the 13 patients before 24 weeks, one of the abortions occurring 2 weeks after amniocentesis. The remaining eight patients delivered at the beginning of the third trimester, at 30.0 ⫾ 1.31 weeks. Mean twin birth weights were 1,410.7 ⫾ 115.6 g and 1,071.9 ⫾ 106.2 g. Four Vol. 72, No. 3, September 1999
TABLE 1 Pregnancy and fetal outcome after multifetal pregnancy reduction or conservative treatment. Variable No. (%) of patients with pregnancy loss at ⬍24 wk No. of patients who underwent amniocentesis Gestational age at delivery (wk)*† Birth weight (g)‡ Fetus I Fetus II Fetus III
Group A (n ⫽ 16)
Group B (n ⫽ 6)
Group C (n ⫽ 13)
Group D (n ⫽ 5)
3 (18.75) 1 32.77 ⫾ 1.36
1 (16.7) 0 34.20 ⫾ 1.64
5 (38.5) 1 30.00 ⫾ 1.31
4 (80.0) 0 28
1,731.9 ⫾ 159.7 1,558.5 ⫾ 115.1 —
2,008.0 ⫾ 203.5 1,816.0 ⫾ 226.8 —
1,410.7 ⫾ 115.6 1,071.9 ⫾ 106.2 —
900 840 685
Note: Group A: bleeding at ⬍10 weeks, MFPR at 11–12 weeks; group B: bleeding at ⬍12 weeks, MFPR at 13–14 weeks; group C: mild bleeding at 12–14 weeks, MFPR at 15–16 weeks; group D: moderate to heavy bleeding at 12–14 weeks, no MFPR. MFPR ⫽ multifetal pregnancy reduction. * Values are means ⫾ SD or actual age. † The differences among groups A, B, and C were statistically significant (P⬍.0001) for the individual variables. ‡ Values are means ⫾ SD or actual weight. Shalev. Multifetal pregnancy reduction. Fertil Steril 1999.
(80%) of the five patients who underwent conservative treatment (group D) aborted before 24 weeks. The fifth patient delivered prematurely after active uterine contractions at 28 weeks. Birth weights were 900, 840, and 685 g. Only one of the three newborns (the one weighing 840 g) survived, but he remained severely handicapped. Overall, gestational age at delivery and twin birth weights significantly decreased (P⬍.001) as the gestational age at which MFPR was performed increased (Table 2, Fig. 1).
DISCUSSION In most published series of triplet pregnancies treated conservatively, the average gestational age at delivery is 33
FIGURE 1 Pregnancy course and dating of MFPR in 42 patients with triplet pregnancies at risk of abortion. Group A: bleeding at ⬍10 weeks, MFPR at 11–12 weeks; group B: bleeding at ⬍12 weeks, MFPR at 13–14 weeks; group C: mild bleeding at 12–14 weeks, MFPR at 15–16 weeks; group D: moderate to heavy bleeding at 12–14 weeks, no MFPR (N MFPR).
weeks (6); 5%–10% of triplets are delivered before 28 weeks. The average perinatal mortality rate is 77 per 1,000 live births. Little information is available on morbidity and long-term outcome of survivors. Evans et al. (7), who performed first-trimester MFPR in triplet pregnancies, achieved a decrease in the pregnancy loss rate at ⬍24 weeks and a decrease in the number of deliveries before 32 weeks. They also noted a reduction in perinatal mortality, neonatal respiratory complications, and neonatal intraventricular hemorrhage. In a related study, Berkowitz et al. (8) performed selective reduction of 102 sets of triplets to twins. Mean gestational age at birth was 35.9 weeks, with only 2% of patients undergoing delivery before 28 weeks. Loss rates before 24 weeks were reduced to 9%. All of these results were achieved in patients with normal, uncomplicated multiple pregnancies. In patients with uterine bleeding, however, the outcome of delaying MFPR until
TABLE 2 Statistical significance of differences in pregnancy outcome.
Variable Delivery (wk) Birth weight (g) Fetus I Fetus II
Group A versus group B
Group A versus group C
Group B versus group C
.120
.001
.002
.001
.026 .014
.001 .001
.002 .002
⬍.001 ⬍.001
Overall
Note: All P values are exact (but not corrected if identical). All P values were determined by the Wilcoxon nonparametric test. Group A: bleeding at ⬍10 weeks, MFPR at 11–12 weeks; group B: bleeding at ⬍12 weeks, MFPR at 13–14 weeks; group C: mild bleeding at 12–14 weeks, MFPR at 15–16 weeks. MFPR ⫽ multifetal pregnancy reduction. Shalev. Multifetal pregnancy reduction. Fertil Steril 1999.
FERTILITY & STERILITY威
Shalev. Multifetal pregnancy reduction. Fertil Steril 1999.
425
after the bleeding stops (i.e., into the second trimester) was unknown before the current study was performed.
and 88.2% of patients were discharged home with at least one infant weighing (mean ⫾ SD) 2,350 ⫾ 670 g (1).
In our series, 18 (42.7%) of the 42 patients had late uterine bleeding. We selected those patients with minimal bleeding for immediate MFPR and postponed the procedure in those with moderate to heavy bleeding. Our findings indicate that fetal outcome is significantly worse in cases of prolonged uterine bleeding in the first trimester of pregnancy—that is, the greater the duration and amount of uterine bleeding, the higher the fetal loss rate after MFPR. Indeed, in patients with mild bleeding at 13–14 weeks, the abortion rate after MFPR was nearly 40%.
In the current study, four of the five patients with moderate or heavy bleeding who did not undergo MFPR aborted, and the fifth patient lost two of the three newborns after preterm delivery. We assume that performance of MFPR despite continuing bleeding would not have worsened the poor pregnancy outcome of this group. Further studies involving a larger patient sample are needed to support or refute this assumption.
Neonatal morbidity was also higher in this group: both mean gestational age at delivery (30.00 ⫾ 1.31 weeks) and mean twin birth weights (1,410.7 ⫾ 115.6 g; 1,071.9 ⫾ 106.2 g) were significantly decreased. With regard to the effect of the increased dead fetal tissue following late MFPR, studies evaluating second- and third-trimester selective feticide in twins with one malformed fetus have found an uncomplicated pregnancy course (9, 10). Even in patients with uterine bleeding at ⬍12 weeks’ gestation, the abortion rate up to 24 weeks was increased to 16%–19%. This information may assist patients and their physicians when deciding whether to perform this procedure. A limitation of our study is that we did not divide our subjects into those with early-ceasing bleeding and late MFPR, those with early-ceasing bleeding and early MFPR, those with early MFPR despite bleeding, and those with late MFPR because of such bleeding. This limitation resulted from our being unable to recommend that our patients at risk of abortion, who conceived after long-standing infertility, undergo an invasive procedure at threatened abortion. Further data from other centers are requested so that recommendations can be made regarding the timing of MFPR in patients at risk of abortion. One of the main reasons for performing MFPR in triplet pregnancies is the high prematurity rate of 15%–20% in triplet pregnancies. There are high mortality and morbidity rates for infants delivered at 25–28 weeks. The risk-benefit ratios for performing MFPR despite heavy bleeding and for postponing MFPR until bleeding stops have yet to be determined. Data on the natural history of triplets during the first and early second trimesters are incomplete (11, 12). Furthermore, the number of patients with uterine bleeding who have undergone MFPR is small. On the basis of studies of similar procedures such as midtrimester selective feticide and amniocentesis that found pregnancy loss rates of 3.5%–9.3% to the 28th week of gestation (6 –9), we speculate that performing MFPR later than the second trimester may also increase the pregnancy loss rate by up to 10%. This is higher than the rate we found after MFPR in triplets during the first trimester (10 –13 weeks). In that study, the total pregnancy loss rate was ⬍9%, 426
Shalev et al.
MFPR and threat of aborting triplets
In conclusion, the present study shows that first-trimester bleeding in triplet pregnancies is associated with a significantly higher rate of abortion after MFPR. These findings, combined with study findings indicating a higher rate of uterine bleeding in the later stage of the first trimester and during the second trimester (11, 12), justify our management protocol of performing selective reduction as early as possible in patients with triplet pregnancies who are at risk of abortion. In patients with moderate to heavy bleeding, selective feticide remains questionable and should be performed with caution until further studies are completed.
Acknowledgments: The authors thank Mrs. Gloria Ginzach, Editorial Board, Rabin Medical Center, for her editorial assistance and Mrs. Iris Fahn for preparing the manuscript for publication.
References 1. Lipitz S, Reichman BN, Uval J, Shalev J, Achiron R, Barkai G, et al. A prospective comparison of the outcome of triplet pregnancies managed expectantly or by multifetal reduction to twins. Am J Obstet Gynecol 1994;170:874 –9. 2. Boulot P, Hedon B, Pelliccia G, Peray P, Laffargue F, Viala JL. Effects of selective reduction in triplet gestation: a comparative study of 80 cases managed with or without this procedure. Fertil Steril 1993;60: 497–503. 3. Depp R, Macones GA, Rosenn MF, Turzo E, Wapner RJ, Weinblatt VJ. Multifetal pregnancy reduction: evaluation of fetal growth in the remaining twins. Am J Obstet Gynecol 1996;174:1233– 8. 4. Jonas HA, Lumley L. Triplets and quadruplets born in Victoria between 1982 and 1990: the impact of IVF and GIFT on rising birthrates. Med J Aust 1993;158:659 – 63. 5. Newman RB, Hamer C, Clinton MM. Outpatient triplet management: a contemporary review. Am J Obstet Gynecol 1989;161:547–55. 6. Berkowitz RL, Lynch L, Stone J, Alvarez M. The current status of multifetal pregnancy reduction. Am J Obstet Gynecol 1996;174:1265–72. 7. Evans MI, Dommergues M, Timor Trish I, Zador IE, Wapner RJ, Lynch L. Transabdominal versus transcervical and transvaginal multifetal pregnancy reduction: international collaborative experience of more than one thousand cases. Am J Obstet Gynecol 1994;170:902–9. 8. Berkowitz RL, Lauren L, Lapinski R, Bergh P. First trimester transabdominal multifetal pregnancy reduction: a report of two hundred completed cases. Am J Obstet Gynecol 1993;169:17–21. 9. Lipitz S, Shalev E, Meizner I, Yagel S, Weinraub Z, Jaffa A, et al. Late selective termination of fetal abnormalities in twin pregnancies: a multicentre report. Br J Obstet Gynaecol 1996;103:1212– 6. 10. Shalev J, Meizner I, Rabinerson D, Mashiach R, Hod M, Bar-Hava I, et al. Improving pregnancy outcome in twin gestations with one malformed fetus by postponing selective feticide to the third trimester. Fertil Steril 1999; 72:257– 60. 11. Lipitz S, Reichman B, Paret G, Modan M, Shalev J, Serr DM, et al. The improving outcome of triplet pregnancies. Am J Obstet Gynecol 1989; 161:1279 – 84. 12. Danforth DN, Scott JR, ed. Obstetrics and gynecology. 5th ed. Philadelphia: JB Lippincott, 1986;474.
Vol. 72, No. 3, September 1999