In singleton pregnancies with an absence of end-diastolic flow (AEDF) in the umbilical artery (UA) Doppler, betamethasone administration had been reported as associated with a transient return of positive UA end-diastolic flow in 63% (Robertson et al., Reference Robertson, Murila, Tong, Baker, Yu and Wallace2009) and 81.8% of cases (Thuring et al., Reference Thuring, Malcus and Marsal2011). But the etiology of betamethasone-induced return of positive end-diastolic UA flow is unclear (Robertson et al., Reference Robertson, Murila, Tong, Baker, Yu and Wallace2009). Selective intrauterine growth restriction (sIUGR) occurs in about 12% of monochorionic twin (MC) pregnancies (Gaziano et al., Reference Gaziano, Lia and Kuhlmann2000; Gratacos et al., Reference Gratacos, Carreras, Becker, Lewi, Enriquez, Perapoch, Higueras, Cabero and Deprest2004; Sebire et al., Reference Sebire, Snijders, Hughes, Sepulveda and Nicolaides1997); MC twins have a higher incidence of respiratory and neurological complications than their dichorionic counterparts (Acosta-Rojas et al., Reference Acosta-Rojas, Becker, Munoz-Abellana, Ruiz, Carreras and Gratacos2007; Fick et al., Reference Fick, Feldstein, Norton, Fyr, Caughey and Machin2006; Gonzalez-Quintero et al., Reference Gonzalez-Quintero, Luke, O'Sullivan, Misiunas, Anderson, Nugent, Witter, Mauldin, Newman, D'alton, Grainger, Saade, Hankins and Macones2003). Antenatal corticosteroids are often administered to enhance fetal lung maturity due to the frequent need for delivery earlier than 34 weeks of gestation in order to avoid the intrauterine fetal demise of the sIUGR twin, which could subsequently lead to brain injury of the co-twin arising from agonal transfusion (Russell et al., Reference Russell, Quintero and Kontopoulos2007).
As revealed in our previous studies, MC pregnancies with sIUGR and UA-AEDF saw more severe placenta share discordance than those where the sIUGR twin showed normal UA Doppler (Chang et al., Reference Chang, Chang, Chao, Hsieh, Wang and Wang2009). So, the etiology of UA-AEDF in sIUGR twin could be partly attributed to severe placenta share discordance (Chang et al., Reference Chang, Chang, Chao, Hsieh, Wang and Wang2009). However, would a similar relationship also exist between the betamethasone-induced return of positive UA end-diastolic flow and the placental share discordance in MC pregnancy with sIUGR and UA-AEDF? The purpose of this study was aimed at addressing this issue.
Materials and Method
Data were collected prospectively between March 2006 and July 2011 from women who gave birth at Chang Gung Memorial Hospital to MC twins with sIUGR and UA-AEDF. MC was diagnosed by ultrasound as having a single placenta, a thin dividing membrane and lacking a twin peak sign, and was confirmed by postpartum placenta examination.
An MC pregnancy with sIUGR was defined as a combination of the co-twin with an estimated fetal weight below the 10th percentile and fetal weight discordance between the twin pair >25% (Acosta-Rojas et al., Reference Acosta-Rojas, Becker, Munoz-Abellana, Ruiz, Carreras and Gratacos2007; Ananth et al., Reference Ananth, Vintzileos, Shen-Schwarz, Smulian and Lai1998). Intermittent UA-AEDF is not recognized as abnormal Doppler in this study. Pregnancies with signs of twin–twin transfusion syndrome (TTTS), presenting at the time of betamethasone administration, were not included in this study. The study was approved by the local institutional ethics committee.
UA Doppler examinations directly prior to (D0) and after 24 hours (D1) and 48 hours (D2) after administration of the first dose of the treatment cycle (including two doses of 12 mg betamethasone, injected 24 hours apart) were recorded. Pulsed and color Doppler was detected by a multifrequency sector array transabdominal transducer (Voluson 730 Pro, GE Medical Systems, Milwaukee, WI, USA). The system operated at output intensities of <100 mW/cm spatial peak temporal average in both imaging and Doppler modes, and the high-pass filter was set at 100 Hz. The sample volume was positioned about 3 cm from the placental insertion of the umbilical cord. All recordings were taken during episodes of fetal inactivity. If positive end-diastolic flow was detected after betamethasone administration, then the Doppler examination was performed on a daily basis in order to record the duration of betamethasone-induced positive end-diastolic UA flow.
Because of the risk of fetal demise in the sIUGR twin with abnormal UA Doppler, all MC pregnancies with sIUGR and UA-AEDF would be suggested to deliver no later than 34 gestational weeks in our institution. Therefore, the indications of betamethasone administration for studying cases included the following situations: (1) deterioration of fetal condition such as reverse of end-diastolic flow of UA Doppler, (2) threatening preterm delivery before gestational age of 34 weeks, (3) two days before planned delivery. For clinical purposes, final decisions regarding delivery were made depending on clinical fetal condition or obstetric situation at that time.
The placenta was examined carefully after delivery. Following draining of the blood from the umbilical cord vessels, the placenta was washed to remove all the clots to ensure the completeness of cotyledons and membranes. Then, after removal of the umbilical cord and membranes, the placenta was freshly cut along the vascular equator into two territories between both cord insertions. The vascular equator was defined as a border drawn in the middle of the avascular zone on the chorionic fetal surface where there were no intertwining vascular anastomoses, or on the anastomosis points where twin–twin communicating vessels met. Each placental portion was weighed separately, which yielded an estimated individual placental mass (eIPM). The placenta share discordance was calculated as the difference between the eIPM of the appropriated for gestational age (AGA) and the sIUGR twin divided by the eIPM of the larger twin, the equation expressed as [((eIPM of AGA – eIPM of sIUGR)/eIPM of AGA) × 100]. Birth weight discordance was calculated in a similar fashion.
Statistical analysis was conducted with SPSS software (version 11.0 for Windows; SPSS Inc, Chicago, IL, USA). Two-sample Student t test or Mann–Whitney U test was used to compare between groups for the continuous variables. Qualitative data were compared by means of χ 2 test. A probability value of less than 0.05 was considered statistically significant.
Results
There were 26 MC pregnancies with sIUGR and UA-AEDF who received antenatal betamethasone for accelerating fetal lung maturation and the D0, D1, and D2 UA Doppler examination data were obtained during the study period. Part of the data on fetal UA Doppler have been published before (Chang et al., Reference Chang, Chang, Chao, Hsieh, Wang and Wang2011). The characteristics of those MC pregnancies are listed in Table 1. Six cases (23.1%) saw the sIUGR twin UA Doppler returning to positive end-diastolic flow after maternal betamethasone administration, all happening at D1 examination. Three of them had their positive end-diastolic UA flows last for 7 days, 5 days, and 1 day, respectively, and then became AEDF again; the other three cases whose flow was still detectable after 48 hours needed delivery due to either preterm labor or scheduled cesarean section, so the exact duration of return of UA positive end-diastolic flow could not be determined. In the remaining 20 patients with persistent UA-AEDF after betamethasone administration, there were seven neonatal deaths: two twin pairs losing both fetuses due to extreme prematurity and the other three experiencing a single neonatal death. One case that turned into TTTS one week after betamethasone administration (the reason of betamethasone administration is preterm labor) received amnioreduction and was delivered at gestational age of 33 weeks. So with respect to the neonatal death rate, seven (17.5%) of the 40 fetuses without betamethasone-induced return of positive end-diastolic flow died, whereas no death was observed in the 12 babies with return of UA positive end-diastolic velocity; the difference, however, was not statistically significant (p = .18).
Characteristics of Monochorionic Twins With Selective Intrauterine Growth Restriction With Abnormal Umbilical Artery Doppler That Received Antenatal Betamethasone for Accelerating Fetal Lung Maturation
sIUGR = selective intrauterine growth restriction; UA = umbilical artery; AGA = appropriate for gestational age.
The characteristics of MC pregnancies with sIUGR, based on with or without betamethasone-induced return of UA positive end-diastolic flow were listed in Table 2. Placental share discordance and birth weight discordance were not statistically different between return of positive UA end-diastolic flow group (n = 6) and persistent UA-AEDF group (n = 20): 63.3% versus 62.5% and 39.0% versus 38.3%, respectively. Their gestational age of betamethasone administration and gestational age at delivery, were also comparable in comparison between these two groups of MC pregnancies.
The Characteristics of Monochorionic Twins With sIUGR With and Without Betamethasone Induced Return of the UA Doppler End-Diastolic Velocity
N = case number; sIUGR = selective intrauterine growth restriction; UA = umbilical artery; TTTS = twin–twin transfusion syndrome; GA = gestational age; AGA = appropriate for gestational age.
Discussion
Our previous study found that MC pregnancies with a sIUGR twin exhibiting UA-AEDF would have more severe unequal placenta share than those with a sIUGR twin demonstrating positive end-diastolic UA Doppler (Chang et al., Reference Chang, Chang, Chao, Hsieh, Wang and Wang2009). Owing to the fact that these MC pregnancies with one sIUGR often require maternal corticosteroid injection because of the pressing need for early delivery, we are interested to know if the severity of placenta share discordance will have an effect on UA Doppler change after antenatal betamethasone administration. From our study results shown in Table 2, we discovered that both the discordances of placenta share and birth weight did not correlate with the betamethasone-induced return of sIUGR twin UA positive end-diastolic flow.
The incidence of betamethasone-induced return of positive UA end-diastolic velocity in MC with sIUGR and UA-AEDF found by our present study was 23.1% (6/26). It has elsewhere been reported as 63% (58/92) (Robertson et al., Reference Robertson, Murila, Tong, Baker, Yu and Wallace2009) and 81.8% (9/11) (Thuring et al., Reference Thuring, Malcus and Marsal2011) in singleton pregnancies with UA-AEDF. So the incidence of betamethasone-induced return of positive end-diastolic flow in singleton pregnancies is higher than those reported for multiple pregnancies. There are two theories explaining the etiology of the betamethasone-induced return of UA positive end-diastolic flow in the IUGR fetuses: one is that exogenous steroid increases fetal blood pressure (Derks et al., Reference Derks, Giussani, Jenkins, Wentworth, Visser, Padbury and Nathanielsz1997); the other is corticosteroid reducing placental resistance by altering the placental tone (Clifton et al., Reference Clifton, Wallace and Smith2002; Korebrits et al., Reference Korebrits, Yu, Ramirez, Marinoni, Bocking and Challis1998). Due to the fact that the current dose of betamethasone administration for twin pregnancy is suggested as the same for the singleton pregnancy and studies have shown that maternal and umbilical cord blood serum betamethasone concentrations after antenatal betamethasone administration are not different between twin and singleton pregnancy (Ballabh et al., Reference Ballabh, Lo, Kumari, Cooper, Zervoudakis, Auld and Krauss2002; Gyamfi et al., Reference Gyamfi, Mele, Wapner, Spong, Peaceman, Sorokin, Dudley, Johnson, Leveno, Caritis, Mercer, Thorp, O'Sullivan, Ramin, Carpenter, Rouse, Miodovnik and Sibai2010). So the impact of betamethasone-induced increasing fetal blood pressure is not sufficient enough to explain the different incidence of betamethasone-induced return of UA positive end-diastolic flow between singleton and multiple pregnancies. Thus, we favor the second theory that different degree of betamethasone-induced placenta vascular dilatation may be responsible for the different incidence of betamethasone-induced return of UA positive end-diastolic flow between these two groups.
Recent studies have revealed that the current standard dose or interval for antenatal corticosteroid administration in singleton pregnancy, whether it be single or multiple courses, has not helped much in reducing respiratory distress in twins (Choi et al., Reference Choi, Song, Seo, Oh, Kim and Roh2009). And also the effect of antenatal corticosteroids decreases with increasing plurality (Blickstein et al., Reference Blickstein, Shinwell, Lusky and Reichman2005). So, combining our finding of a lower rate of betamethasone-induced return of UA positive end-diastolic flow in twin pregnancies than in singleton pregnancy, the optimal dose and interval of antenatal corticosteroid to enhance the fetal lung maturation in multiple pregnancies may need further studies to evaluate.
In conclusion, although more severe unequal placenta share correlates with the sIUGR twin abnormal UA Doppler in MC twins, the severity of placenta share discordance was not different between MC twins with and without return of positive sIUGR twin UA end-diastolic flow after betamethasone administration.
