Results

Twin to Twin Transfusion Syndrome. Results

Between July 1997 and June 2002, 252 patients with the diagnosis of TTTS underwent surgical treatment at our center. UCO was performed on 36 (14.3%) of these patients. Ten cases were excluded from the analysis because they had undergone secondary UCO after an attempted SLPCV (technical success rate for TTTS 216/226, or 95.5%). 

One additional case was excluded because the ligated cord was that of a recipient twin that had died shortly before surgery. Thus, 25 cases of TTTS underwent primary UCO. Six (24%) had cord occlusion of the donor fetus (donor group) and 19 (76%) of the recipient fetus (recipient group). Six fetuses had the following discordant fetal malformations: 2 had neural tube defects (both donors), 1 had body-stalk anomaly (donor), 1 had pulmonary atresia (recipient), 1 had anencephaly (recipient), and 1 had intracranial hemorrhage (recipient). 

The two fetuses with stage I TTTS that underwent UCO had a lethal fetal anomaly (limb-body stalk anomaly, anencephaly). These data are summarized by stage in Table 10.1. There were no statistical differences between the donor vs the recipient groups in regards to stage of disease or fetal malformations. There were no significant preoperative differences between the donor and recipient groups that underwent UCO (Table 10.2). Table 10.3 shows operative characteristics in the donor vs recipient groups. All 19 recipient fetuses underwent UCL. Four of six (66%) in the donor group had UCL while two (33%) had UCP. This difference in surgical approach between the two groups was statistically significant (p = 0.05). 

There was a significantly higher rate of two trocars required to complete the procedure in the donor donor cord (83.4%) compared with the recipient cord (0.0%, p < 0.001). The median operating time in the donor group was almost twice as long as in the recipient group (75 vs 40 minutes, respectively). This difference was statistically significant (p = 0.04). Clinical outcome data are summarized in Table 10.4. 

There was no difference in gestational age at delivery between the donor group (median 34.8 weeks) and recipient group (median 33.8 weeks). Twenty-two of 25 (88%) patients with TTTS who had undergone a primary UCO had liveborn infants. There were no significant differences in total survival rates between the two groups. Four patients (16%) had preterm premature rupture of membrane (PPROM) within 21 days of the procedure (median 2.0 days; range 0–12). 

Two had improvement of group (50%) as compared to the recipient group (5.3%, p = 0.03). In all cases, the trocars entered the sac of the abnormal twin. There was a significantly higher rate of amnioinfusion for ligation of the PPROM spontaneously and one underwent amniopatch by the method previously reported.17 All three had liveborn infants. The fourth PPROM case miscarried at 20 weeks of gestation. Cord insertions were classified as central, marginal, or velamentous at surgical pathology. 

The incidence of the different cord insertion types were as follows: 

• donor – central 13 (52%), marginal 5 (20%), velamentous 7 (28%). 

• recipient – central 18 (72%), marginal 5 (20%), velamentous 2 (8%). 

There was no difference in the incidence of velamentous insertion between the donors or the recipients (7/25 donors, 2/25 recipients, p = 0.13 Fisher’s exact test). There was no difference in the incidence of velamentous cord insertion of the occluded cord: 3/6 donors, 2/19 recipients (p = 0.08, Fisher’s exact test). Perinatal survival relative to the placental insertion of the occluded cord was not statistically different: 15/17 central (88.2%), 2/3 marginal (66%), 5/5 (100%) velamentous, p = 0.37. There were no neurological or cardiac function abnormalities in any of the survivors. The total number and percentage of UCO performed per year compared with the number of laser surgeries for TTTS are shown in Table 10.5. A nearly fourfold decrease in the percentage of total UCOs was noted during the 5-year span of the study. 

DISCUSSION 

Selective feticide in TTTS involves occlusion of the umbilical cord of one of the fetuses with the hope that the co-twin will survive intact. By interrupting the vascular communications between the fetuses at the level of the umbilical cord, damage to the co-twin via feto–fetal hemorrhage may be avoided. The fundamental difference between SLPCV and UCO lies in that, although both methods interrupt the vascular communications between the two fetuses, SLPCV allows both twins a chance to survive. 

On the other hand, survival after SLPCV is unpredictable, sometimes resulting in the intrauterine demise of the ‘healthier’ twin, with survival of the seemingly most affected fetus. Patients are often distraught with the notion that the healthier fetus may die after SLPCV and, as a result, may request primary UCL. Parents may also request primary UCL in the case of twins discordant for a lethal anomaly. Secondary UCL, after a failed attempt at SLPCV, is rare in our experience, although always a counseling topic in any stage III or stage IV patient. After our original description of UCL for the treatment of TRAP sequence, several UCO methods have been developed, to include, umbilical cord ligation (UCL),18 laser photocoagulation of the umbilical vessels (UCP),12,19 electrocoagulation of the umbilical cord,13,14 and umbilical cord ultrasonic transection. 

Each technique is associated with varying degrees of invasiveness and operative skill. No surgical approach has been shown to have superior outcomes with regard to co-twin survival thus far. At our center, UCO was offered to TTTS patients carrying a fetus with a lethal discordant fetal anomaly, or if stage III or IV disease was diagnosed and the patient desired UCO of the critically ill fetus, after extensive counseling regarding other possible therapeutic modalities. An overall 88% survival rate of the co-twin was noted in this study. 

UCL was the method of choice. This method allows for complete and instantaneous occlusion of all umbilical vessels, which may optimally prevent feto–fetal hemorrhage. UCL requires adequate accessibility to the target umbilical cord. If access to the target cord is limited, our secondary approach is UCP. The advantage of laser occlusion of the umbilical cord is that this procedure may be performed even if access to the cord is limited. However, cord occlusion is not as rapid and simultaneous as UCL. 

Selective feticide in pregnancies complicated by TTTS poses the additional operative challenge of varying approach based on if the target fetus is the donor or the recipient twin. In our study, no clinical or preoperative differences were noted between the two groups (see Tables 10.1 and 10.2). However, there were some operative differences, such as requirement for two trocars and amnioinfusion and operating time, which suggest that cord occlusion of the donor twin may be technically more demanding than that of the recipient twin (see Table 10.3). 

This is not unexpected in that additional operative maneuvers are required to allow for optimal visualization and access to the umbilical cord of the oligohydramniotic donor fetus. Also, because a major requirement of UCO is to enter the gestational sac of the target cord, direct access to the cord of the recipient within polyhydramnios is technically easier. Although in this study all target cord were handled within their own respective gestational sac, we now have experience with transmembranous laser photocoagulation (selective thermophotolysis). 

This latter technique may overcome many of the identified difficulties associated with UCL of the donor twin within its own sac. Despite the technical differences described above, UCO via UCL/UCP resulted in similar outcome data in regards to survival of the co-twin, gestational age at delivery of the co-twin, birthweight, and PPROM rate of the donor vs the recipient fetus (see Table 10.4). The single survivor statistics for TTTS treated by cord occlusion via UCL/UCP appear similar to the results after SLPCV therapy,20 and may be higher than those treated by serial amniodrainage if stage of disease is taken into account. For this reason, primary UCO is not recommended any longer for otherwise uncomplicated stage III or IV TTTS. 

Outcome results of UCL/UCP appear favorable compared with other methods of selective feticide techniques performed in monochorionic multiples. Embolization using varying thrombogenic substances had relatively high rates of demise of the co-twin, presumably due to passage of the substance to the co-twin via incompletely obliterated vascular communications.6–11 A multicenter study of 50 consecutive umbilical cord occlusions via UCP and/or bipolar electrocoagulation for varying indications has recently been reported.21 UCP done primarily and bipolar electrocoagulation done secondarily if necessary was performed in 37 cases, and primary bipolar electrocoagulation was performed in 13 cases. 

The overall survival rate was 75%, and overall PPROM rate was 52%. The relatively high PPROM rate in that study, particularly the 25% persistent PPROM that occurred before 30 weeks’ gestation, is of concern. Newer techniques such as umbilical cord ultrasonic transection are currently under investigation, but are unlikely to provide any additional advantages.14 Direct comparison of each technique in regards to co-twin survival must be performed to determine optimal surgical method. Table 10.5 describes a decreasing percentage of primary and secondary UCOs compared with laser surgery over time for the treatment of TTTS at our center. 

This may be a reflection of surgeon experience as well as patient reassurance in regards to the comparable outcome results of SLPCV vs UCO. Indeed, the likelihood of at least one survivor is approximately 85% in both techniques, with the obvious difference that with SLPCV both fetuses may have a chance to survive. 

This is particularly important in view of the recent availability of bipolar coagulation, which is perceived as a simpler method of achieving separation of the circulations of both twins. Unfortunately, such technological advance may result in the unnecessary demise of many stage III or IV fetuses that could have otherwise had an opportunity to survive with SLPCV. Nonetheless, patients may sometimes request UCO as a primary therapeutic technique in TTTS stages III or IV, as laser occlusion of the vascular anastomoses cannot predict which of the two fetuses will survive. 

Our experience shows that of those stage III–IV patients with a single intrauterine fetal death (IUFD) after SLPCV, the demised fetus may be the one with normal preoperative Dopplers approximately 20% of the time. Considering the pros and cons of each approach, we currently do not advocate primary UCO for TTTS stages III or IV, unless there are additional complications, such as discordant anomalies or a terminally ill fetus, that may compromise the health of the co-twin in utero. This chapter addresses the issues of umbilical cord occlusion in TTTS via UCL/UCP. 

Because the approach varies based on whether the target fetus is the donor or the recipient, different techniques must be utilized to successfully perform this procedure. This may include using more than one trocar and/or amnioinfusion to obtain access to the donor fetus. Operator experience with more than one cord occlusive method is required. In this study of TTTS patients that underwent primary UCO, we showed that donor cord occlusion may be technically more difficult, yet overall outcome results are similar between the donor and recipient twins.