Four studies (448 twins) were included: one RCT (142 twins) and three observational controlled studies (306 twins).
Two of the three observational studies adjusted for confounding. In one of the observational studies, the data were collected retrospectively. Two studies did not report consecutive enrolment of the patients. The interventions and the duration of follow-up were generally poorly described.
Laser coagulation was associated with the highest overall survival (rates ranged from 56.3 to 62.4%), survival of at least one twin (range: 76.4 to 83.2%), and differential survival of donor and recipient (respective ranges: 52 to 63.2% and 58.3 to 70%). Lower rates of survival were associated with serial amnioreduction: overall survival rates were 38.6 to 64.3%, survival of at least one twin was 51.4 to 85.7%, and differential survival of donor and recipient ranged from 40 to 57.7% and from 37.1 to 57.7%, respectively.
Laser photocoagulation versus serial amnioreduction (three studies): all three studies found improved overall survival with laser photocoagulation, with the difference statistically significant in two studies. The studies all found laser coagulation to be associated with significantly improved survival of at least one twin. All three studies found increased survival of the donor twin with laser photocoagulation, but in none of the studies was the increase statistically significant. They also found increased survival of the recipient twin with laser photocoagulation, but the increase was only statistically significant in one study. The three studies all found significantly reduced overall neurological morbidity with laser photocoagulation; neurological morbidity was reduced for donors in two studies reporting differential outcomes (significantly reduced in one of these) and was significantly reduced for recipients in both studies.
Serial amnioreduction versus septostomy (one observational study, n=7): there was no significant difference between treatments for overall survival or the survival of at least one twin.
Failures occurred in 1.67% (4 out of 240) of laser photocoagulation procedures, 3.54% (7 out of 198) of serial amnioreduction procedures and 42.9% (3 out of 7) of septostomy procedures. Laser coagulation was associated with fewer terminations of pregnancy (1.4% versus 10.6%) and neonatal deaths (7.5% versus 25.4%) than amnioreduction (based on three studies), but was associated with more spontaneous miscarriages (13.4% versus 11.0%) and stillbirths (23.1% versus 13.0%).