Impact of Selective Fetal Growth Restriction on Laser Therapy Outcomes in Twin-Twin Transfusion Syndrome

What does this study add to current knowledge?

• Contrary to current literature, our study demonstrates that twin-twin transfusion syndrome (TTTS) cases with concomitant selective fetal growth restriction (sFGR) present with a higher TTTS stage, but they do not represent a more severe TTTS at initial presentation; rather, they are a more severe type of sFGR.

What are the main clinical implications?

• Our study shows that sFGR, as defined by ISUOG, coexists in approximately 30% of TTTS cases. Furthermore, sFGR is independently associated with decreased double survivorship at the expense of donor survival. Those with type 2 or 3 sFGR and undergoing laser therapy at early gestational age are especially at risk.

Monochorionic diamniotic (MCDA) twins represent 15–30% of all twin gestations [1]. As MCDA twins share a single placenta, vascular anastomoses and at times unequal placental territories are responsible for their high complication rates. Complications specific to MCDA twins include twin-twin transfusion syndrome (TTTS) and selective fetal growth restriction (sFGR). TTTS affects approximately 15% of MCDA pregnancies and is responsible for most of the mortality and morbidity in this group [2]. The management of TTTS is clearly described in the literature: fetoscopic laser photocoagulation has been demonstrated to be superior to amnioreduction and is now considered the gold standard for severe TTTS. It is usually offered for Quintero stage II–IV between 16+0 and 26+0 weeks [3]. sFGR affects approximately 10–15% of MCDA pregnancies and is classified per the Gratacós et al. [4] criteria, where type I has continuous anterograde flow in the umbilical arteries (UAs) of the sFGR fetus, type II has persistently absent or reversed UA flow and type III has intermittently absent or reversed UA flow. For type I sFGR, expectant management is the favored option because of its excellent prognosis. However, for type II and type III, the optimal management is less clear. Fetoscopic laser photocoagulation has been proposed, but with suboptimal results and not without significant risks, and radiofrequency ablation or cord occlusion is often offered [5]. Complicating the overall picture is the fact that sFGR occurs in approximately 50% of TTTS cases [6, 7]. The management of these pregnancies is controversial and is not frequently discussed in the literature. With advancing technologies, outcomes of cases treated with fetoscopic laser photocoagulation are improving and dual survival is now the goal of treatment.

The aim of our study was to compare survival outcomes of fetoscopic laser photocoagulation in MCDA pregnancies complicated by TTTS with and without concomitant sFGR. We hypothesized that fetoscopy leads to better outcomes for TTTS alone. In our opinion, obliterating the vascular anastomoses in cases with concomitant sFGR would leave the smaller fetus with a very small residual placental territory, therefore, increasing the risk of intra-uterine demise.

This was a retrospective cohort study of all MCDA pregnancies treated with fetoscopic laser photocoagulation at the Sainte-Justine Mother and Child University Hospital Center between February 2006 and January 2020. This study was approved by the Institutional Ethics Committee. Women were considered eligible if they were at least 18 years of age at the time of the pregnancy. Patients were excluded if they had higher order multiple pregnancies, preoperative twin anemia polycythemia sequence (TAPS) or if they were planned for fetoscopic laser photocoagulation but underwent bipolar coagulation or radiofrequency ablation instead. Demographic data collected included age, body mass index (BMI), ethnicity, and parity. Gestational age at TTTS diagnosis and treatment, preterm premature rupture of membranes, placental abruption, chorioamnionitis, and cervical insufficiency following fetal therapy as well as timing, mode, and indication for delivery were reviewed. Maternal outcome data analyzed included pulmonary edema, mirror syndrome, and obstetrical intermediate-level care unit admission. Fetal outcome data analyzed included intra-uterine demise and post-laser TAPS.

TTTS was diagnosed according to the standard criteria, meaning a deepest vertical pocket of amniotic fluid of less than 2 cm in the donor gestational sac and of 8 or more cm in the recipient sac (10 cm if 20 weeks’ gestation or more) [8]. The severity of TTTS was classified per the Quintero staging system with one caveat [9]: if there were signs of cardiac failure on fetal echocardiography, the severity of TTTS assigned was automatically stage 3 according to the CHOP scoring system [10]. Diagnosis of sFGR was by an estimated fetal weight (EFW) less than the 10th percentile in one fetus and an EFW discordance of 25% or more between fetuses [11]. Severity of sFGR was classified according to UA Doppler velocimetry in the smaller twin.

Per hospital protocol, a level II ultrasound was performed in the 24 h prior to fetal surgery for all patients using a GE Voluson E8. During this ultrasound, biometry, detailed anatomy, placental and cord insertion, amniotic fluid volume, vitality, and bladder visualization were documented for both fetuses, UA, middle cerebral artery, and ductus venosus (DV) Doppler velocimetry. A fetal echocardiography was also performed whenever possible.

All patients provided written informed consent before undergoing fetoscopic laser photocoagulation. The procedure was performed in an operating room under local anesthesia and sedation. The Solomon technique was used as the laser ablation technique, as previously described [12, 13].

Immediately after the intervention, patients received a 100 mg suppository of indomethacin and 30 mg of extended release oral nifedipine unless contraindicated, to reduce the risk of uterine contractions. Patients were hospitalized for at least 24 h following surgery. Postoperatively, an obstetrical ultrasound was done within 24–72 h.

Data were stored in a database and analyzed with the STATA software (version 13, College Station, TX, USA). Data are presented as median with interquartile range or n (%). Qualitative data were compared by the χ² test or Fisher’s exact test as appropriate. Continuous variables were compared by one-way ANOVA or Kruskal-Wallis test, and Student’s t test or Mann-Whitney test for pairwise comparisons as appropriate. Differences were considered significant when p < 0.05. A univariate linear regression model was performed to identify potential prenatal predictors of double survivorship. All significant associations found in the univariate analysis (p < 0.10) were included in the multivariate linear regression model.

In total, we reviewed the obstetrical and ultrasound charts of 240 consecutive women with pregnancies complicated by TTTS (with or without sFGR). Ninety-one patients were excluded. A total of 149 patients were included in our study. Forty-seven patients (31.5%) had a pregnancy complicated by TTTS and sFGR (Fig. 1). Twenty-eight of the concomitant sFGR were type 1 sFGR and 19 were type 2 or 3.

Maternal and fetal baseline characteristics are summarized in Table 1. Donor UA Doppler velocimetry anomalies tended to be more frequent in the TTTS + sFGR group (34% vs. 13.7% p = 0.007), as were Donor DV Doppler velocimetry anomalies (13.6% vs. 4.5% p = 0.083). Recipient Doppler velocimetry anomalies, either UA or DV, did not differ between the two groups. At treatment, TTTS distribution was significantly different between the two groups. While TTTS + sFGR presented predominately with TTTS stage 3 (42.6%), most TTTS alone had a stage 2 at treatment (50.9%) (p = 0.026). Mean gestational age at diagnosis did not differ between the two groups. However, mean gestational age at treatment was higher in the concomitant group (21+2 weeks GA vs. 20+6 weeks GA p = 0.03). Pregnancy outcomes did not differ between the two groups (Table 2).

Survival outcomes differed between the two groups (Table 3). Double survival (DS) was lower in the concomitant sFGR group (48.9% [23/47] vs. 68.6% [70/102] in the TTTS-only group [p = 0.021]). Fetal donor survival was 59.6% (28/47) in the TTTS + sFGR group and 84.3% (86/102) in the TTTS-only group (p = 0.001), whereas fetal recipient survival did not differ between groups (83% [39/47] in the TTTS + sFGR group vs. 74.5% [76/102] in the TTTS-only group [p = 0.252]). Survivorship of at least one twin did not differ between the two groups: 93.6% (44/47) in the TTTS + sFGR versus 92.2% (94/102) in TTTS-only group (p = 0.751). In univariate analysis, 2 factors were associated with lower probability of dual survival: the presence of a type 2 or 3 sFGR (OR = 0.27; 95% CI 0.09–0.77 p = 0.009) and an EFW discordance of more than 30% (OR = 0.42; 95% CI 0.20–0.88, p = 0.021). However, after multivariate analysis, only sFGR type 2–3 (OR = 0.56; 95% CI 0.32–0.96, p = 0.033) was independently associated with lower probability of DS. Interestingly, higher gestational age at laser therapy (OR = 1.17; 95% CI 1.01–1.36, p = 0.036) was independently associated with higher probability of dual survival (Table 4).

Our study demonstrates that sFGR coexists in approximately 31.5% of pregnancies complicated by TTTS. The prevalence of coexisting sFGR and TTTS was lower than described in the literature. This could be explained by the different criteria used to define sFGR. Van Winden et al. [14] used only an EFW below the 10th percentile to define sFGR, whereas we used the diagnostic criteria of an EFW below the 10th percentile and a 25% inter-twin growth discordance as recommended by the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) and Delphi Consensus published by Khalil et al. [11, 15]. It is also worthwhile to note that we only included TTTS cases that qualified for surgery, which could have altered the proportion of cases complicated with sFGR.

Mean gestational age at delivery in both groups was approximately 30+4 weeks, which compares to what has been published in the literature and is largely due to complications related to laser (see Table 2) as well as maternal and fetal indications for delivery [16‒19]. We found that DS was significantly diminished by the presence of concomitant sFGR, at the expense of the donor twin.

In accordance with previous studies, our results show that sFGR seems to be an independent predictor of twin survival after laser photocoagulation, in particular donor twin survivorship [14, 20, 21]. Research by Van Winden et al. [22] has shown that there is not only unequal placental sharing when sFGR coexists with TTTS but also a proportion of placentas with a smaller total mass, suggesting placental insufficiency. Therefore, as hypothesized, obliterating the vascular anastomoses leaves the donor twin with a very small residual placental territory leading to a higher risk of intra-uterine demise. Moreover, as in previously published studies, the distribution of TTTS stages between the two groups was significantly different. In our study, the proportion of stage 3 TTTS was significantly higher in the concomitant TTTS + sFGR group. This is explained by the fact that there is an overlap in definitions between stage 3 TTTS and type 2/3 sFGR. Indeed, most concomitant TTTS + sFGR are classified as stage 3 based on abnormal UA Doppler velocimetry, which is inherent to the sFGR classification. Contrary to other studies, TTTS stage did not impact fetal survival in our cohort. It is important to stress that we did not expect to find a difference in donor survivorship based on TTTS stage. Per the hypothesized pathophysiology in concomitant sFGR, diminished donor survival is based on a lesser residual placental mass; therefore, TTTS stage would not impact donor survivorship. However, we would have expected to see a difference in double survivorship based on UA Doppler velocimetry anomaly [20]. It is well known that type 2 sFGR carries a poor prognosis, whereas type 3 sFGR has better outcomes [23, 24]. Therefore, type 2 sFGR should have a worse donor survivorship than type 3. We could not distinguish between type 2 and 3 due to the small number of patients in our cohort, however, we did demonstrate that having type 2 or 3 sFGR negatively impacted double survivorship. Doppler velocimetry anomalies alone did not affect dual survival.

Finally, although a small EFW discordance between fetuses is associated with better outcomes in MCDA pregnancies, our results did not show this to be an independent predictor of survival [23, 25]. This all goes to show that contrary to what previous studies have stated, the most important prognostic factors in twin DS following laser therapy are the presence and type of sFGR.

Several limitations should be considered while interpreting our results. The retrospective nature of this study led to some incomplete data, possibly introducing bias in our results. Nonetheless, this study was conducted in a single tertiary center where there is standardized management of pregnancies complicated by TTTS and/or sFGR with a low loss to follow-up rate.

In summary, our study shows that sFGR, as defined by ISUOG, coexists in approximately 30% of TTTS cases and that it is independently associated with decreased double survivorship at the expense of the donor twin. Those with type 2 or 3 sFGR and those undergoing laser therapy at early gestational age are especially at risk. Importantly, although cases with concomitant sFGR present for treatment with a higher TTTS stage, it may represent a more severe type of sFGR rather than severe TTTS. Additional studies are needed to validate our results and to further evaluate the specific impacts of type 2 and type 3 sFGR in donor survival following laser therapy for TTTS.

This was conducted ethically in accordance with the World Medical Association Declaration of Helsinki. Patients have given their written informed consent. This study protocol was reviewed and approved by the CHU Sainte-Justine Research Ethics Board, number 2016-1075.

The authors have no conflicts of interest to declare.

The authors did not receive any funding for this research.

All authors contributed to all aspect of the study design. Laurence Sophie Carmant: collected the data, did the data analysis and wrote the first draft. François Audibert: performed the statistical analysis. Elisabeth Codsi: contributed to the first draft. Laurence Sophie Carmant, François Audibert, Katherine Thériault, Sandrine Wavrant, and Elisabeth Codsi: helped revised the article up to the final draft.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.