Objective
The selectivity index (SI) has been proposed as a measure of technical success in laser surgery for twin–twin transfusion syndrome. Surgeries with an index >–0.25 have been considered highly selective. The purpose of this study was to evaluate the applicability of this index in our patient population.
Study Design
The SI was assessed in 314 consecutive laser surgeries and correlated with perinatal survival.
Results
A total of 310 patients (98.7%) underwent a completely selective procedure. The SI was 0.8 in the selective laser photocoagulation of communicating vessels group vs 0.3 in the nonselective laser photocoagulation of communicating vessels group ( P = .001). In the selective group perinatal survival of at least 1 twin (92.6% vs 50%) and survival of the donor (75.4% vs 0%) was significantly better ( P = .05).
Conclusion
The SI as originally proposed is misleading and of limited use as it does not differentiate selective from nonselective procedures. We propose instead using a ratio of selective/nonselective procedures, and selectively coagulated/total number of coagulated vessels to appraise center-specific and patient-specific surgical performance of laser surgery for twin–twin transfusion syndrome.
Twin–twin transfusion syndrome (TTTS) is thought to occur from a net imbalance of blood flow from the donor twin to the recipient twin via placental vascular anastomoses. The intraoperative identification of the anastomotic vessels had been an issue before 1998. Selective laser photocoagulation of communicating vessels (SLPCV) was described by Quintero et al in 1998 as a superior method of therapy, targeting only the placental vascular anastomoses and sparing any normally perfused areas in the placenta. SLPCV proved to be superior to a nonselective technique. Some authors have reported that a selective technique may not be feasible in every case due to technical factors such as placental insertion, accessibility, and fetal position, with most of the procedures being a mix of selectively coagulated (SC) and nonselectively coagulated (NSC) placental vessels. A selectivity index (SI), defined as log [(SC + 1)/(NSC + 1)], (where NSC could also be presumed anastomoses not followed to their terminal end) was proposed by Stirnemann et al as a measure of performance in laser surgery for TTTS. A high SI defined as >–0.25 was reported by the authors to correlate with improved postnatal survival at 28 days of at least 1 twin and both twins.
The purpose of our study was to evaluate the applicability of SI in a patient population undergoing laser therapy for TTTS by our group.
Materials and Methods
Our database was queried for all patients with monochorionic diamniotic twin pregnancies diagnosed with TTTS undergoing laser therapy in a 6-year period. Triplet pregnancies and monoamniotic pregnancies were excluded from this analysis. Surgeries had been approved by the respective institutional review boards and all patients gave written consent.
Inclusion criteria in the study were: (1) monochorionic diamniotic twins diagnosed with TTTS, defined as polyhydramnios in the sac of the recipient twin with a maximum vertical pocket (MVP) of ≥8 cm and oligohydramnios in the donor twin with an MVP of ≤2 cm; and (2) gestational age at the time of the procedure between 16-26 weeks.
Preoperative ultrasound
All patients underwent a comprehensive ultrasound examination that included fetal anatomy, biometry, measurements of the MVP in each sac, and cervical length. Doppler waveform analysis of the umbilical artery, umbilical vein, ductus venosus, middle cerebral artery, and tricuspid valve were obtained whenever possible. Disease severity was classified according to Quintero staging system (stage I: poly/oligohydramnios with bladder of the donor still visible and normal Doppler; stage II: bladder of the donor not visible, normal Doppler; stage III: critically abnormal arterial or venous Doppler in either twin, defined as absent or reverse end-diastolic velocity in the umbilical artery, absent or reverse flow in the ductus venosus and pulsatile umbilical venous flow; stage IV: hydrops fetalis in either twin; stage V: demise of 1 or both twins).
Operative procedure
SLPCV was the intended surgery in all cases and was performed as previously described. Namely, all placental vascular anastomoses were identified and photocoagulated, sparing any normally perfused areas of the placenta. Procedures were considered nonselective (NS-LPCV) if the terminal end of at least 1 lasered vessel could not be identified, even if it was suspected as being a bona fide vascular anastomosis. The number of lasered vessels (NS and NSC) was prospectively recorded in each case. The SI was calculated as per Stirnemann et al.
Follow-up ultrasounds
All patients underwent a follow-up ultrasound the next day after the surgery (between 16-24 hours postoperatively). Postoperative ultrasound assessment included documentation of the fetal heart rate; assessment of the MVPs; repeated Doppler studies of the middle cerebral artery, umbilical vessels, and ductus venosus; and repeated cervical length and assessment of possible membrane detachment. Patients were typically discharged on postoperative day 1. Weekly follow-up ultrasounds were arranged through the referring physician for 1 month with monthly ultrasounds thereafter.
Outcome data
Perinatal outcomes were obtained using preprinted sheets completed by the referring physicians. Patients diagnosed with in utero fetal demise (IUFD) who subsequently miscarried were categorized as IUFD. Patients who miscarried without prior diagnosis of IUFD were not included in the IUFD numbers, however their outcomes were reflected in the perinatal survival. After delivery, a total of 214 (68%) available placentas were assessed in surgical pathology of St Joseph’s Women’s Hospital (Tampa, FL) and Tampa General Hospital (Tampa, FL) and patent vascular anastomoses were identified through air injection as previously described. The main outcomes were perinatal survival at 30 days of at least 1 twin and of both twins and their correlation with the SI. Another outcome assessed was surgical failure, defined as the incidence of persistence or reversal of TTTS after the surgery. Additional outcomes analyzed included gestational age at delivery, rate of delivery before the threshold for viability (defined as gestational age at delivery of <24 weeks), incidence of preterm premature rupture of membranes within 21 days after the procedure, and intrauterine fetal demise of the recipient and donor. The SC and NSC vessels had been prospectively documented and were used to calculate the SI in each case.
Statistical analysis
Analyses were conducted using software (SPSS, Version 17.0; SPSS Inc, Chicago, IL). Categorical variables were analyzed using χ 2 and Fisher exact test. Continuous variables were assessed for normality of distribution using The Kolmogorov-Smirnov test and analyzed with Student t test, analysis of variance, and Mann-Whitney U test as indicated. A P value of ≤ .05 was considered statistically significant.
Results
Between April 2000 and September 2005, 324 patients with TTTS who had undergone laser photocoagulation were identified. Ten patients were excluded from the analysis. Of these, 3 patients were excluded due to elective termination of pregnancy. Six patients were excluded due to previous amniodrainage with unintentional septostomy and a resulting MVP of the donor >2 cm. In 1 patient, endoscopic laser photocoagulation was aborted due to intraoperative bleeding and diminished visualization of the placental vessels. A total of 314 patients were therefore included in the final analysis; 98.7% underwent SLPCV. Only 4 (1.3%) patients underwent NS-LPCV.
Preoperative characteristics
Table 1 illustrates the clinical characteristics of the patients at presentation in the 2 surgical groups (SLPCV vs NS-LPCV). There were no significant differences in gestational age at the procedure, MVP and estimated fetal weight of donors and recipients, Quintero stage, incidence of anterior placenta, cervical length, or presence of cerclage between the 2 groups.
| NS-LPCV | SLPCV | ||
|---|---|---|---|
| Characteristic | n = 4 | n = 310 | P |
| GA at surgery, wk median (range) | 20.3 (17.9–22.3) | 20.1 (16.3–25.9) | .9 |
| MVP R, cm median (range) | 10.4 (9–12.3) | 10.7 (8–18.7) | .6 |
| MVP D, cm median (range) | 0.35 (0–1.9) | 0 (0–2) | .6 |
| EFW D, g median (range) | 237 (81–343) | 260 (91–712) | .6 |
| EFW R, g median (range) | 388 (202–557) | 356 (153–1170) | .8 |
| Quintero stage 1, n (%) | 0 (0) | 39 (12.6) | > .05 |
| Quintero stage 2, n (%) | 0 (0) | 91 (29.4) | |
| Quintero stage 3, n (%) | 4 (100) | 152 (49) | |
| Quintero stage 4, n (%) | 0 (0) | 28 (9) | |
| Anterior placenta, n (%) | 3 (75) | 139 (44.8) | .3 |
| Cervical length, cm median (range) | 3.8 (3.1–4.2) | 4 (0.8–6.2) | .5 |
| Cerclage, n (%) | 1 (25) | 29 (9.4) | .3 |
Intraoperative characteristics
Table 2 demonstrates that there was a statistically significant difference in the operative times between the 2 groups, with a median value of 40 minutes in SLPCV procedures compared with 63 minutes in NS-LPCV procedures ( P = .013). There was no difference between the numbers of SC vessels in the 2 groups. The SI was significantly higher in the SLPCV group (0.8) compared with the NS-LPCV group (0.03) ( P = .001). Although all the patients in the NSC group had SI >–0.25 (0.22, –0.24, 0.47, –0.15, and 0.39) and could be therefore categorized as high selectivity procedures, there was a statistically significant increase in the donor IUFD in this group. The median values of the SI were >–.25 in both groups.
