Fetoscopic laser coagulation for twin-to-twin transfusion syndrome: a comparison of flexible 1.0/1.2 mm fetoscopes with curved sheaths of 2.7/3.3 mm² vs. 2 mm fetoscopic lens technique with sheaths of 6.6/11.3 mm²

Introduction

Twin-to-twin transfusions syndrome (TTTS) is a potentially dangerous complication that affects approximately 10–15 % or monochorionic pregnancies [1], [2], [3], [4], [5], [6], [7]. The first description of the inter-twin hemodynamic imbalance and its pathogenesis has been attributed to a German gynecologist, Friedrich Schatz in 1882 [8].

Since the publication of the Eurofetus study in 2004, fetoscopic laser coagulation of the placental anastomoses has become the standard treatment for TTTS from the second trimester through to around 26 weeks’ gestation [6, 7].

The most common complication of fetoscopic laser surgery is preterm premature rupture of the membranes (PPROM). This results in increased risks for preterm delivery and exposes the fetuses to the risks of extreme prematurity. The rate of PPROM increases with an increasing size of the operative sheath [6, 9], [10], [11], [12], [13], [14]. Once damaged, the chorioamniotic membranes have a limited ability to repair themselves [6, 10], [11], [12], [13], [14], [15]. Gratecos et al. found no evidence of spontaneous membrane healing following fetoscopic surgery when the membranes underwent histologic evaluation following delivery [15].

Beck et al. found that the size of the fetoscopic instruments affects the incidence of iatrogenic PPROM. The larger the diameter of the instruments the higher the iatrogenic PPROM rate, the lower the gestational age at birth and the lower the neonatal survival [16]. Deliverer et al. concluded that the best way to prevent PPROM is to minimize the number and size of the membrane defects during invasive procedures like amniocentesis or fetoscopic surgery [17].

The obvious trade-off is that the smaller the scope, the more restricted the view and higher the likelihood of an incomplete ablation procedure resulting in recurrent TTTS and TAPS [13, 14].

The aim of this study was to compare neonatal outcomes after laser surgery using a flexible 1–1.2 mm scope within a sheath with an opening of 2.65–3.34 mm² compared to a standard 2 mm scope within a sheath with an opening of 6.63–11.27 mm²

Materials and methods

Ethical approvals for the study were received from both the Martin Luther University Halle-Wittenberg and the Johannes Guttenberg University. The involvement of the two centers reflects where the primary author (M.T.) was employed during the time intervals used for data collection. Prior data from the Center for Fetal Surgery, Johannes Gutenberg University have been previously published [6, 9]. The project was registered at the United States National Library of Medicine at the National Institutes of Health (Clinicaltrials.gov NCT03151915).

Staging for TTTS was carried out according to Quintero [18]. All patients diagnosed with TTTS between 16 and 26 weeks of gestation received fetoscopic laser therapy independent of the cervical length before surgery.

Data was collected after fetoscopic laser coagulation for TTTS. Where possible, missing data was collected from patient files in the obstetric clinic, from patients themselves or from referring physicians. Data for PPROM and umbilical artery pH was not available from fetoscopic surgery done using a 2 mm fetoscope.

Cases were divided based on the size of the fetoscope and sheath used for the fetoscopic surgery. All fetoscopes and sheaths were provided by Karl Storz GmbH & Co. Tuttlingen, Germany. There were four operating sets used during the study.

1. 2.0 mm diameter lens optic (26008AA 0°) used with the 2661U sheath (2.5 × 3.6 mm diameter, with a cross-sectional area of 6.63 mm²).

2. 2.0 mm diameter lens optic (26,008BU 30°) with the 22661UF sheath (3.1 × 4.3 mm diameter, with a cross-sectional are of 11.27 mm²). This set was used for cases with an anterior placenta.

3. 1.0 mm flexible fiber fetoscope (11510A 0°) used with a curved 11510KD sheath with two working channels (1.3 × 2.6 mm diameter with a cross-sectional area of 2.65 mm²) (Figure 1).

4. 1.2 mm lens (11,530AA 0°) with a curved 11,530 KB sheath with two working channels (1.6 × 2.9 mm diameter with cross-sectional diameter of 3.34 mm²). This scope set was used for cases with severe polyhydramnios or maternal BMI>35.

On the evening before the surgery and directly before the operation, all women received 100 mg indomethacin suppositories. The vascular architecture on the surface of the monochorionic placenta was carefully examined and all anastomoses between the two fetal circulations were identified. All visible anatomoses were selectively coagulated using the Medilas Fibertom 8100 Nd:YAG laser (Dornier MedTech, Wessling, Germany) with a 0.6 mm laser fiber and with power set at 50–60 W. In cases with a placental net of anastomoses, the laser coagulation was performed through the vascular net according to the Solomon method [19]. In cases with a cloudy and/or slightly opaque amniotic fluid an amnioexchange was performed using a Amniotic Flush Solution, Serumwerk Bernburgh AG, CE 0483, Germany. Amniotic fluid was drained following completion of the laser coagulation to a deepest vertical pocket of 4–5 cm using a suction pump in the 1 mm scope group and passive drainage through the sheath in the 2 mm scope group.

Survival was defined as survival to discharge from hospital. Statistical analysis was done using SPSS^® for Windows^® (German version 17.0). For the analysis data from scope sets one and two were combined as they both used 2.0 mm diameter scopes and scope sets three and four were combined as they both used 1.0/1.2 mm diameter scopes. To ensure data protection, the data was anonymized. A two-tailed t-test was used for continuous variables that were normally distributed and a chi-square test was used for categorical variables. A p-value of 0.05 was used to determine statistical significance. A Kaplan-Meyer survival analysis was used to compare latency between surgery and delivery. A literature search was carried out on PubMed^® and Embase^® using keywords of twin twin transfusion syndrome, TTTS, outcome, fetoscopic laser therapy, fetoscopic laser surgery, fetoscopic laser photocoagulation, treatment and management. These were combined using Boolean operators “and” and “or”. From the studies found, those that described more than 100 laser surgeries from a single center were selected for review. Those that contained information on gestational age at the time of surgery, gestational age at delivery, survival of at least one fetus, survival of two fetuses, PPROM<32 weeks, recurrent TTTS and diameter of the fetoscope were included. No effort was made to differentiate between different methods of laser therapy. The papers included and their relevant data are included in Table 3.

Results

A total of 247 patients were treated with laser therapy between 2006 and 2019 with a total of 251 procedures performed. A second procedure was required in 3 cases due to development of recurrent TTTS (Re-TTTS) [11]. In 1 case due to poor visibility of an anterior placenta, the initial procedure was discontinued, and a second attempt made the next day. An overview of the study groups is shown in Figure 2.

Figure 1:

Flexible fiber 11510A 1 mm fetoscope and curved 11,510 KD sheath with two working channels (1.3 × 2.6 mm, sectional area of 2.65 mm²).

Figure 2:

Overview of the study groups.

There was no difference in maternal age or gestational age at the time of surgery between the groups. There was a difference in the TTTS staging between the two groups. More cases of Stage II TTTS were performed in the 2 mm scope group, while more cases of Stage III and Stage IV were done in the 1/1.2 mm scop group. More anastomoses were coagulated, and less amniotic fluid removed in the 1/1.2 mm scope group compared to the 2 mm scope group (Table 1).

Re-TTTS occurred in 3 cases (1.9 %) in the 1.0/1.2 mm scope group. In these cases, a second procedure was performed within 21 days. In 2 cases, two anastomoses were re-coagulated and in 1 case, one anastomosis was re-coagulated. The Solomon technique was performed at the second procedure. All 3 cases resulted in live births of both fetuses. No Re-TTTS was reported in the 2.0 mm scope group. This rate is well below the average of 4.7±3.9 % determined from the literature review.

Table 2 shows the PPROM data for the 1.0/1.2 mm group with an overall rate of PPROM<32 weeks of 16.3 %. Data was not collected for PPROM in the cases where a 2.0 mm scope was used. The rate of at least one survivor was significantly higher in the 1.0 mm scope group (97.2 vs. 88.3 %). There was also a higher gestational age at delivery in the 1.0/1.2 mm scope group (33.1±4.0 vs. 31.6±4.7 week) and a longer latency between surgery and delivery (12.7±5.0 vs. 10.2±5.1 weeks).

The latency period between intervention and delivery was also evaluated using a survival analysis that is shown in Figure 3. This shows a statistically significant difference in the time between surgery and delivery between the two groups. The curves for both groups are initially similar but start to diverge approximately 9 weeks after surgery with the curve for the 1.0 mm group showing a longer latency compared to the 2.0 mm group from that point onwards.

Figure 3:

Latency between laser procedure and delivery. Red curve: n=93, classic 2.0 mm diameter lens optic with the 2661U sheath for posterior placenta (2.5 × 3.6 mm; sectional area, 6.63 mm², Karl Storz GmbH & Co. KG^®), 2.0 mm diameter 26008BUA lens optic (30°) with the 2661UF sheath (3.1 × 4.3 mm; sectional area, 11.27 mm²) was used in cases with anterior placenta localization. Blue curve: n=147, flexible fiber 11510 A fetoscope of 1 mm (curved sheath 2.65 mm²) and a 1.2 mm 11530AA optic with a curved sheath 11,530 KB with two working channels (1.6 × 2.9 mm; cross section 3.34 mm²).

Discussion

This study shows significant improvements in the latency period between surgery and delivery in the group where a smaller fetoscope and operative sheath were used to perform fetoscopic laser surgery. This resulted in a mean increase in gestational age at delivery of 2 weeks in that group. While there was no difference in that rate of both fetuses surviving there was an increase in at least one fetus surviving until discharge from hospital.

The use of a smaller diameter fetoscope and a curved sheath that has a smaller opening area result in a smaller hole being created in the chorioamniotic membranes with the potential of reduced iatrogenic injury to the membranes. The size of the membrane defect created by the operative sheath is reduced by as much as four-fold when a 1 mm diameter fetoscope is used compared to a 2 mm fetoscope, depending on the sheath used.

Gratacos et al. examined the membrane defect in 19 cases after fetoscopy was done through a 10 French cannula with a 3.3 mm diameter [15]. In all cases, the defect in the membranes was identified on evaluation following delivery. There was no evidence of membrane healing. The defect in the amnion was larger than in the chorion with the mean area of the amniotic defect measuring 14.4 mm² compared to 11.1 mm² in the chorion [15]. Papanna et al. similarly described a defect of 19.6 mm² from using a 12 French cannula with a 4 mm diameter opening [14]. Tchirikov et al. measured a defect in the amnion after fetoscopic laser coagulation with a 1 mm scope and a sheath with an opening of 2.65 mm² and found a defect of 3 mm² [11].

In 2009, Ruano et al. reported a series of 19 cases of laser photocoagulation for Stage 3 and 4 TTTS using a 1.0 mm scope with a straight sheath with 2.2 mm2 area opening. PPROM occurred in 2 cases (10.5 %) [20]. The overall mean gestational age at delivery was 32.1 (range: 26.0–38.0) weeks. The overall survival rate was 52.6 %, with survival of at least one twin of 78.9 % and both twins in 26.3 % [20]. While this study spans a longer time and has larger numbers of cases, the overall survival rates are much higher and the gestational age at delivery was also higher. In the group using a 1.0 mm scope the overall survival was 97.2 % with 81 % dual survivors and the average gestational age at delivery was 33.1±4 weeks [20].

Beck et al. reported a systematic review on PPROM and fetal survival after minimally invasive fetal surgery found that the larger the diameter of the instruments used, the higher the rate of PPROM, the lower the gestational age at delivery and the lower the fetal survival rates [16]. In this study, the PPROM rate before 32 weeks was 16.3 % in the group where a 1.0 mm scope was used. Unfortunately, PPROM rates were not recorded in the 2.0 mm group and comparisons between the groups were not possible. From the studies included from the literature review, the average rate of PPROM was 19.6±15 %. Thus, the smaller scope and sheath sizes may be associated with a decreased rate of PPROM but this requires further investigation.

There were differences between the groups in the frequency of Stages on TTTS treated, with the 1.0/1.2 mm group having significantly more advanced stages compared to the 2.0 mm group. This may be a function of an increase in the number of patients referred from abroad over time. Travelling from abroad has become more difficult due to financial or bureaucratic reasons resulting in patients presenting having already developed to a more advanced stage of TTTS.

Some patients were referred after being considered inoperable in their home countries because of being in an advanced stage of TTTS. It is important in laser surgery to coagulate all vascular anastomoses to prevent the risk for recurrent TTTS or TAPS. Incomplete laser surgery and residual anastomoses may be related to various factors, including different laser surgery techniques, experience of the operator (learning curve effect) and location of the placenta (anterior or posterior). Several authors have reported a rate of residual anastomoses up to 32 % when placentas were examined using conventional fetoscopic instruments for treatment of TTTS [21], [22], [23]. Knijnenburg et al. reported on a series of 371 placentas evaluated over a period of 2002–2016 and found an overall incidence of 21 %. The rate decreased from 38 % in the initial time period to 11.7 % in the final time period [22]. This demonstrates the impact of experience and the development of the Solomon technique [19, 20]. It might be expected that using a smaller scope might increase the frequency of residual anastomoses due to challenges with visualization associated with a smaller field of view.

While the placentas were not directly evaluated for residual anastomoses, the gestational age at delivery and survival were higher than in the 2.0 mm group. The incidence of repeat TTTS was low at 1.9 % with no TAPS seen in the 1.0/1.2 mm scope group compared to no cases of either in the 2.0 mm group. There would appear to be a trade-off of using a smaller diameter fetoscope. Visualization may be more difficult, and procedures may take more time to perform a thorough evaluation of the placental surface with increased risk for residual anastomoses [10].

However, a larger scope diameter may result in fewer residual anastomoses but is associated with higher rates of PPROM, lower gestational age at delivery and increased neonatal morbidity and mortality.

The increased number of anastomoses coagulated in the 1.0/1.2 mm group may be a function of there being more advanced stages of TTTS, but there is currently no evidence that there are more vascular anastomoses in higher stages of TTTS. The increased volume of amniotic fluid removed at the end of the procedure in the 2.0 mm group is not well explained. It may in part be attributed to patients coming from abroad who may have had an amnioreduction prior to their referral to attempt to temporize their symptoms during travel. The use of passive drainage in the 2.0 mm group was also less accurate than using a suction system. A significant amount of amniotic fluid escaped accurate measurement and had to be estimated.

From the Kaplan-Meyer survival analysis there is an overall difference in latency between the operative procedure and delivery. It is interesting to note that the divergence in the curves doesn’t start until some 9 weeks after the surgery. While the size of the scope and sheath may be the important factor in the overall difference, it is interesting that the effect doesn’t become evident until a time distant from the actual surgery. This raises the possibility that there may be other co-factors responsible for some of the differences observed. Since most of the cases using the 2.0 mm scopes occurred earlier in the study, it is important to consider what other changes may have been made that may have affected the timing of delivery. Not all patients were delivered at the study center and delivery decisions were, in those cases, made by the local attending physicians. Thus, different patterns of practice may have influenced delivery timing over time. One example might be the management of twin gestations affected by sFGR particularly as delivery timing can be influenced by the results of Doppler findings.

The strength of this study is a relatively large sample size with care provided by a single operator. There are limitations related to the retrospective nature of the study and the fact that some information was not available to allow for some important comparisons.