Conjoined twins--the Cape Town experience.

This paper records our experience in the management of 25 sets of conjoined twins seen over a 32 year period (1964 - 1996). The twins were classified into 14 complete and symmetrical sets and 11 incomplete or heteropagus. The 14 symmetrical sets included 9 thoracopagus, 2 ischiopagus, 1 craniopagus and 1 omphalopagus twins. In the incomplete heteropagus group there was 1 ischiopagus, one twin being anencepahlic, 2 dipygus, 5 parasitic, 2 fetus-in-fetu and I cranial and caudal. The management is detailed case by case. Overall 10 of 14 symmetrical sets underwent attempts at separation with 16 surviving the procedure, but there were 3 late deaths. In the incomplete group 10 of 11 were operated on with 9 survivors. The importance of a multi-disciplinary approach, the extensive investigations required pre-operatively to define areas of organ and bony conjunction, congenital anomalies of each twin and surgical teamwork is emphasized. Specific problems encountered were identified. In thoracopagus twins the hearts were of paramount importance as conjuction was usually fatal, being associated with major congenital defects. The greater the extent of thoracic cage fusion the greater the chance of associated severe anomaly. Skin expansion to assist coverage of the defects after separation was of great assistance, as was the use of collagen coated vicryl. Evaluation of the liver and pancreatico-biliary systems with isotope excretion scanning was crucial to pre-operative planning. Where there was fusion of the duodenum a single pancreatico-biliary system could be expected and prior strategies for separation and Roux-en-Y enteric drainage of both pancreatic and biliary secretion should be planned. Gastro-oesophageal reflux led to considerable morbidity in both twins of a thoraco-omphalopagus set. In ischiopagus and dipygus conjoined twins bilateral posterior iliac osteotomies were an essential component to anatomic reconstruction of the pelvic ring and wound closure. Also in this group, due to the frequency and extent of shared genital, urinary and ano-rectal structures, long-term morbidity was expected and a component of this might be due to spinal cord tethering, or as in one of our cases, a progressive hydrosyringomyelia. Timing of separation was ideally set at between 5 and 9 months with 6 to 8 weeks of prior tissue expansion but earlier operation was frequently required because of cardio-respiratory problems or organ failure in one twin. In most cases the goal of obtaining separate, independent and intact individuals was achieved.