Background
Complete coagulation of the vascular equator (as in the Solomon technique) has been suggested to reduce postoperative complications such as twin anemia polycythemia syndrome and the recurrence of twin-twin transfusion syndrome following fetoscopic laser coagulation of chorionic vessels for twin-twin transfusion syndrome.
Objective
We aimed to evaluate the benefit of this technique on perinatal outcomes compared with selective ablation of anastomoses.
Study Design
We conducted a monocentric retrospective study comparing selective laser coagulation of anastomoses to the Solomon technique from January 2006 to August 2020. To adjust for potential confounders, the cases operated by selective surgery were matched to the cases operated with the Solomon technique according to the gestational age at laser therapy, placental localization, and Quintero stage using propensity score matching.
Results
With a total of 994 cases, 399 matched pairs were included in the analysis. Compared with selective ablation, the Solomon technique was associated with significantly improved survival: the overall twin survival at delivery and discharge was 72% vs 79% ( P =.003) and 69% vs 75% ( P =.006), respectively; the double twin survival rate at discharge was 55% vs 65% ( P =.02), respectively, and the rate of intrauterine death dropped from 18% to 12% ( P =.003), respectively. The Solomon technique significantly reduced the rate of twin anemia polycythemia syndrome (10% vs 4%; P =.02), leading to fewer secondary rescue procedures (13% vs 7.3%; P =.01). However, the Solomon technique was associated with an increased risk of preterm rupture of membranes, especially at early gestational ages (3.8% vs 11%; P <.001 for preterm rupture of membranes <24 weeks). Among the survivors at delivery, both the groups had similar gestational ages at birth. Both neonatal mortality and severe neurologic morbidity were similar in both the groups. However, an increased risk of bronchopulmonary dysplasia was found in the Solomon group (4.5% vs 12%; P <.001).
Conclusion
Although the risk of preterm premature rupture of membranes has increased, the introduction of the Solomon technique has significantly improved perinatal outcomes in pregnancies affected with twin-twin transfusion syndrome.
Introduction
In monochorionic pregnancies, both the fetuses share a single placenta and are connected through intertwin vascular anastomoses. Both unidirectional (arteriovenous anastomoses) and bidirectional (venovenous and arterioarterial anastomoses) are usually present, even in uncomplicated pregnancies. , In some cases, an unequal distribution of anastomoses with uneven placenta sharing allows unidirectional blood shifts from one fetus (the donor twin) to the other (the recipient), leading to a hemodynamic imbalance. Although its physiopathology is still not fully understood, this unusual placental architecture appears to be one of the main mechanisms leading to the twin-to-twin transfusion syndrome (TTTS), which occurs in 10%–15% of monochorionic pregnancies. , The spontaneous outcome of TTTS is associated with extremely high rates of perinatal loss and subsequent neurologic lesions in survivors. Fetoscopic laser occlusion of placental anastomoses has proven to be effective and is now the undisputed first-line treatment for TTTS. The technique has evolved over the years. The initial nonselective coagulation of all vessels crossing through the membrane was first replaced by a selective identification and coagulation of intertwin anastomoses. , Despite an improvement in prognosis, perinatal morbidity remained significant.
Why was this study conducted?
This study aimed to compare the perinatal outcomes and complications following Solomon and selective fetoscopic laser photocoagulation for twin-twin transfusion syndrome.
Key findings
The Solomon technique was associated with a lower risk of postoperative twin anemia polycythemia syndrome and an improvement in perinatal survival compared with selective coagulation of anastomoses. However, the Solomon technique was also associated with a higher risk of early preterm rupture of membranes resulting in increased neonatal morbidity.
What does this add to what is known?
Compared with selective coagulation of placental anastomoses, the Solomon technique was associated with an improvement in perinatal survival without associated neurologic morbidity, but it was also associated with a higher risk of preterm rupture of membranes and neonatal extraneurologic morbidity.
More recently, the Solomon technique, involving the complete coagulation of the vascular equator along with the intertwin arteriovenous anastomoses, has been evaluated in an international randomized trial. It reduces the proportion of residual anastomoses and thus postoperative complications such as twin anemia polycythemia sequence (TAPS) or recurrent TTTS. Nevertheless, the impact of the Solomon technique on perinatal mortality and neonatal morbidity is unclear.
We aimed to compare the perinatal outcomes following selective coagulation of vascular anastomoses vs the Solomon technique.
Material and Methods
Study population
We reviewed all consecutive cases of monochorionic pregnancies affected by TTTS that were referred to our department for perinatal management between January 2006 and August 2020. An ultrasound diagnosis of TTTS was defined as follows: presence of oligohydramnios (maximal vertical pool [MVP] <2 cm) in the donor’s sac, polyhydramnios in the recipient’s sac (defined by an MVP >8 cm before 20 weeks of gestation and >10 cm after 20 weeks), and discordant bladder size between the 2 fetuses. Severity was assessed by the Quintero staging system. Triplet pregnancies, monoamniotic pregnancies, congenital malformations or chromosomal anomalies, and fetal death at referral (Quintero stage V) were excluded. We also excluded all the pregnancies affected by TTTS that did not undergo laser surgery.
Laser coagulation of anastomoses is the first-line treatment of pregnancies affected by TTTS in all cases presenting with Quintero stages II–IV. In Quintero stage 1, surgery is performed in the case of symptomatic polyhydramnios (maternal respiratory discomfort, cervical shortening, or contractions). Between April 2011 and March 2018, some of the cases presenting with Quintero stage 1 were included in a randomized controlled trial (RCT) comparing immediate surgery and expectant management. Cases that were allocated to laser surgery and those requiring rescue therapy were also included in this retrospective analysis.
Procedures
The procedures for percutaneous surgeries were performed under local anesthesia, tocolysis (indomethacin or atosiban), antibiotic prophylaxis (cephazolin), and maternal sedation (remifentanyl). An 8–10 Fr trocar (Cook Medical) was inserted in the polyhydramniotic cavity under ultrasound guidance using the Seldinger method. Placental vessel coagulations were performed using a 1.3- or 2-mm semirigid fetoscope or a 3.3-mm rigid 3-channel fetoscope (Karl Storz SE & Co KG, Tuttlingen, Germany) and a diode laser fiber (Dornier Med-Tech GmbH, Wessling, Germany).
In the case of selective coagulations, the crossing vessels were visualized to identify and coagulate the intertwin vascular anastomoses only. Starting 2013, procedures were performed according to the Solomon technique. In addition to coagulation of all the visible anastomoses, a thin line of placental surface was coagulated from one edge to the other, connecting all the anastomoses and thus creating a virtually dichorionic placenta, which is supposed to reduce any potentially remaining anastomoses. This interanastomosis line is called the vascular equator. An amnioreduction is performed at the end of the procedure.
Perinatal management
Following laser surgery, patients were hospitalized for a short period (generally 2 days). Evaluation of twin viability, bladder filling, amniotic fluid deepest pools, integrity of the intertwin membranes, and Doppler measurements were performed twice a day. The patients were discharged in the absence of immediate complications, and follow-up consisted of weekly ultrasound investigations for at least a month. Thereafter, ultrasounds were performed every 2 weeks by the referring perinatologist in the case of complete recovery. Delivery was either spontaneous or elective (cesarean or vaginal) according to the local protocols. We recommended delivery at approximately 34 weeks’ gestation.
Outcomes
Prenatal outcomes
We analyzed survival at delivery, intrauterine demise (IUD), spontaneous miscarriage (defined as birth before 24 weeks with at least 1 fetus alive before delivery), cord occlusion, and termination of pregnancy. Postoperative complications included recurrence of TTTS, TAPS, rescue therapy rates, and early preterm rupture of membranes (PROM) (occurring before 24 weeks and 28 weeks). Recurrence of TTTS was defined as persisting or recurring discordance in amniotic fluid levels and bladder volumes after the procedure. Postoperative TAPS was defined as discordance in the middle cerebral artery pic systolic velocity (MCA-PSV) measurements, with one twin presenting with an MCA-PSV of at least >1.5 MoM and the other with that <1 MoM (stage I TAPS). Rescue therapy was defined by the need for another invasive procedure at a minimum following an initial laser surgery, including repeat laser photocoagulation, in-utero transfusion, amnioreduction, or cord occlusion.
Postnatal outcomes
Among twins alive at delivery, neonatal death was defined as death within 28 days of birth or discharge. The gestational age at birth was divided into 3 groups (24–28 weeks, 28–32 weeks, and ≥32 weeks). Birthweight difference was reported as a percentage of difference between the largest and smallest twin. Neonatal morbidity included neurologic morbidity (defined by either intraventricular hemorrhage grade 3 or 4 on postnatal ultrasound and magnetic resonance imaging (MRI) or periventricular leukomalacia on postnatal ultrasound and MRI), bronchopulmonary dysplasia (BPD), necrotizing colitis grade ≥ 2, and severe neonatal sepsis.
Statistical analysis
We performed a matching of the Solomon cases (laser coagulation of the vascular equator) and selective cases (selective coagulation of intertwin anastomoses) using a propensity score to account for both the potential confounding factors influencing the choice of the technique and the factors associated with the outcome. The propensity score was defined as the conditional probability of being treated with either one technique or the other, given the preoperative perinatal characteristics. Before 2012, the selective technique was the only technique performed. The Solomon technique was progressively introduced in our unit starting in 2012 and was definitively adopted in 2013 for all cases of TTTS. To calculate the propensity score, we performed a logistic regression analysis adjusted on prenatal characteristics that are associated with the prognosis and perinatal outcomes. Our model included the gestational age at laser therapy, the Quintero stage, and the placental localization. Matching was then performed using the nearest-neighbor method.
The continuous variables are presented as median and interquartile range (IQR) and were compared using the Wilcoxon–Mann–Whitney test. Frequencies and percentages were used for categorical variables and compared using the chi-squared test or Fisher’s exact test. The odds ratios are presented for both the pregnancy and fetal and neonatal outcomes. To account for the correlation between pairs of twins, the binary outcomes defined at the infant level were analyzed using generalized estimating equations.
All the analyses were carried out using R ( http://www.r-project.org ; R Foundation, Vienna, Austria).
Ethical statement
The study protocol was approved by our institutional review board (CERAPHP.centre #00011928, ref 2021-07-14)
Results
Study population
Between January 2006 and August 2020, 994 monochorionic diamniotic pregnancies presenting with TTTS were treated by laser ablation of anastomoses in our department ( Figure ). 15 cases were excluded because of incomplete data. Thus, 979 pregnancies were evaluated in the analysis: 399 were treated with the Solomon technique and 580 with selective coagulation of anastomoses. Each case treated by the Solomon technique was matched with 1 control treated by selective coagulation of anastomoses on the basis of their propensity score. The remaining 181 cases were unmatched and thus excluded from the analysis. The outcome at delivery or discharge was lost to follow-up in 16 of 399 (4%) pregnancies in the selective group and in 9 of 399 pregnancies (2.25%) in the Solomon group. A total of 578 neonates were delivered alive in the selective group, and 550 remained alive at 28 days or at discharge (ie, 28 neonatal deaths took place). In the Solomon group, 630 neonates were delivered alive, and 600 remained alive at 28 days or at discharge (ie, 30 neonatal deaths took place within 28 days of birth or discharge).
The characteristics of the population are displayed in Table 1 . The matching procedure based on the Quintero stages, placental location, and gestational age at surgery resulted in similar prenatal characteristics in both the groups, except from maternal age, which was not included in the matching procedure. A description of the unmatched cases is given in the appendix ( Supplemental Table ).
| Population characteristics | Selective, N=399 | Solomon, N=399 | P value a |
|---|---|---|---|
| Maternal age | 29.0 (26.0–33.0) | 31.0 (28.0–34.0) | <.001 |
| Nulliparous | 171 (43) | 192 (48) | .2 |
| Cervical length <15 mm | 12 (3.0) | 12 (3.0) | >.9 |
| GA at laser | 20.10 (18.30–22.60) | 20.10 (17.80–22.60) | .4 |
| MVP (cm) | 10.00 (9.00–12.50) | 10.00 (8.00–11.35) | .057 |
| Quintero stage | >.9 | ||
| 1 | 98 (24) | 97 (24) | |
| 2 | 130 (33) | 130 (33) | |
| 3 | 162 (41) | 162 (41) | |
| 4 | 9 (2) | 10 (2) | |
| Anterior placenta | 177 (44) | 171 (43) | .7 |
a Statistical tests performed: Wilcoxon rank-sum test; chi-square test of independence.
Perinatal outcomes according to the surgical technique
The perinatal outcomes according to the surgical procedure are presented in Table 2 . Survival was significantly different between the groups: the Solomon technique significantly increased the overall fetal survival compared with the selective technique, both at birth (578/798 [72%] vs 630/798 [79%]; P =.003), and at discharge (550/798 [69%] vs 600/798 [75%]; P =.006). Selective surgery was associated with a higher rate of intrauterine fetal death (140/798 [18%] vs 97/798 [12%]; P =.003; odds ratio [OR], 1.56 [1.18–2.07]). A significant improvement in dual survival was also found in the Solomon group, both at birth (276/399 [71%] vs 231/399 [60%]; P =.007) and at discharge (253/399 [65%] vs 211 [55%]; P =.02). Among the pregnancies delivering after 24 weeks with at least 1 twin alive, the median gestational age at birth was 33.00 weeks (30.15–34.30) in the selective group and 32.40 weeks (29.90–34.10) in the Solomon group, P =.043. The preterm birth rates were similar: delivery between 24 and 28 weeks occurred in 29 of 399 (8.6%) pregnancies in the selective group and in 41 of 399 (12%) in the Solomon group, whereas delivery after 32 weeks occurred in 205 of 399 (61%) and 197 of 399 (55%) pregnancies, respectively. The difference was not significant. Lastly, among pregnancies with both fetuses alive at delivery, we found a higher birthweight difference between the largest and smallest twin in the selective group (median, 13%; interquartile range [IQR], [5–23]) than in the Solomon group (median, 10%; IQR, [5–18]), P =.013.
