Conjoined twins are a rare congenital anomaly with an incidence of 1 in 10,000 to 50,000 live births.1 Conjoined twins are classified based on the most prominent site of union; the prefix “pygo” is used for twins joined at the sacrum, “omphalo” for those connected at the umbilicus, and “ischio” for those joined at the hip.2 An interprofessional approach is necessary for separation surgery because of the various and complicated anatomies of conjoined twins. If separation of the shared anatomy is feasible and safe, the reconstruction of the conjoined area is always a challenge for plastic surgeons, especially when the defect is large.3 Many plastic surgeons have reported a variety of methods to manage wound closure for conjoined twins, including tissue expanders,4,5 skin grafts,6 mesh,7 and local skin flaps.8 Here, the authors report the use of opposing triangle flaps for separating three different kinds of conjoined twins and propose a mathematical theory for the feasibility of flap design.
Informed consent for the separation surgeries was obtained, and the Ethics Committee of the Children’s Hospital of Fudan University approved these surgeries. Permission was obtained from the twins’ parents to publish the case details and related images. Further details on patient imaging and the general and urologic separation surgeries have been previously published.9
Patients and Preoperative Examination
Female pygopagus twins were delivered at 38 weeks’ gestation. The physical examination revealed the twins were joined at the sacrum, buttocks, and perineum. Each twin had a separate urethra and vagina. Barium contrast imaging of the gastrointestinal tract showed that the twins shared a 0.5-cm common anus. MRI revealed that the dural sacs were continuous. A three-dimensional printed model of the conjoined area was made to assist plastic surgeons in calculating and designing the flap.
Male omphalopagus twins were delivered at 34 + 5 weeks’ gestation. The twins were joined from the xiphoid to the umbilicus. A Doppler cardio-ultrasound showed that the hearts were separate without life-threatening malformations, but their livers were conjoined. A contrast-enhanced abdominal computed tomography (CT) scan was obtained separately for each patient, and the images were transferred to the Computer-Assisted Surgery Planning System (Hisense, Qingdao, China) to generate a three-dimensional model of the liver. The model revealed that the livers were fused, and certain vessels passed through the joined livers that would need to be ligated during the separation surgery. A plastic three-dimensional printed model of the joined abdomens was created to help design a flap.
Male ischiopagus twins weighing 5,140 g were delivered at 36 + 6 weeks’ gestation. The initial physical examination revealed that the twins were joined from the umbilicus to the perineum, and the lateral boundary of the joined area was the bilateral inguinal region. The perineal structures were ambiguous, and there were four orifices and exstrophic mucosa. After considering the results of an iohexol (Omnipaque) contrast study and a three-dimensional model based on enhanced CT images, the orifices were identified. Both patients had duplicated bladders, and each baby’s duplicated bladders were separated and united with one of the bladders of the other. The twins shared a colon, which passed through the middle of the four bladders. A plastic three-dimensional printed model of the twins was made for flap design.
After the physical examinations and simulating surgery on the three-dimensional printed models, the plastic surgery team decided to use opposing triangle flaps to close the skin defects from separating the three pairs of conjoined twins. Plastic surgeons calculated the approximate coverage of the triangle flap for each of the three conjoined twins before separation based on the mathematic theorem that the sum of any two sides of a triangle is greater than the third side. The authors designated the length of the conjoined area a, the width b, and the length of the skin bridge e. The extendable length of skin was Δd, which was determined by the amount of skin that could be gently pinched by the surgeon minus the double thickness of the skin. The length of each triangle flap was equal to the width of the conjoined area b, the pedicle width of the flap was equal to the length of the conjoined area a, and the length of one side of the triangle was c. If the twins were separated from the middle of the skin bridge, then each of the twins received two halves of the width of the skin bridge. As a result, the area of the skin bridge for one twin could cover S = 2 × a × (Δd + 0.5e) (Figure 1A). However, for the opposing triangle flap, each twin not only had an extension on one side of the skin bridge, but also had an extended area from the two sides of the triangle. Therefore, the area of the flap covered St = 2 × c × Δd + a × (e + Δd) (Figure 1B). The difference of the two areas was ΔS = St – S = Δd (2 × c − a). Again, given the fact that the sum of any two sides of a triangle is greater than the third side, 2 × c was always greater than a, so the opposing triangle flap could cover a greater area.
For the conjoined area of the pygopagus twins, a = 6 cm, b = 5 cm, and e = 2 cm. The length of the side of the flap was c = 5.8 cm, and Δd = 1 cm. Therefore, the conjoined area that needed to be covered was 30 cm2, and the area of the triangle flap was 29.6 cm2, which was nearly equal to the conjoined area (Figure 2A).
The conjoined area of the omphalopagus twins was large (a = 15 cm, top width = 3 cm, bottom width = 6 cm). The conjoined area that needed to be covered was 67.5 cm2. Two triangle flaps of different sizes were designed for this case to avoid the skin on the ribs. For the upper small triangle, the length of the bottom was 5 cm, the height of the triangle was 4 cm, and the length of side c = 4.7 cm. For the lower, larger rectangle flap, the height was 7 cm, and the lengths were 3 and 5 cm; the Δd was 1 cm. The skin bridge was 1 cm on the top, 2 cm on the middle, and 3 cm on the bottom. The total area of the two flaps and skin bridge was 68.4 cm2, which was slightly larger than the conjoined area (Figure 2B).
For these twins, the length of the conjoined area was 9 cm, which included the abdomen and scrotum. Because the skin of the scrotum was very loose, the conjoined area could be covered even if the scrotums were cut in the middle. As a result, the length of the conjoined area of the abdomen (a) was 6 cm, and the width (b) was 6 cm. The width of the skin bridge (e) was 2 cm. The length of side c of the triangle flap was 6.7 cm; Δd = 1 cm. Accordingly, the conjoined area that needed to be covered was 36 cm2, and the area of the triangle flap was 31.4 cm2. To overcome the 4.6 cm2 deficit, plastic surgeons planned to bring together the edges of the distal aspect of the triangle flap because the skin on the waist, at the distal aspect of the exposed area made by the flap, was loose (Figure 2C).
The separation took place when the twins were 3 months old. The operation was begun by plastic surgeons raising the pedicle opposing triangle flaps on both twins as previously designed. The neurosurgeons separated the conjoined coccyges and the dural sacs, which were closed with artificial dura. The general surgeons incised the blind-end orifice and the common anal canal longitudinally and reconstructed each anal sphincter by suturing the muscles of the common anus and around the blind-end orifice together under guidance by electrical stimulation, followed by a sigmoid colostomy. The wounds of the twins were completely closed without difficulty with the triangle flaps (Figure 3A, B).
The omphalopagus twins presented for separation surgery at 3 months of age. After anesthesia, the plastic surgeons incised the skin to raise the opposing triangle/rectangle flaps to the previously designed size. Exploration confirmed that the livers were fused, with an obvious fine white line on the surface. The fused livers were therefore divided along the white line, and the veins and arteries separated. The joined sternums and xiphoids were also separated. After evaluation, the myofascial structure tension was not high, so surgeons closed the muscular layers directly, without the use of mesh. Complete closure of the wound with the flaps was achieved (Figure 4A, B).
Surgery took place at 3 months of age. The plastic surgeons raised the double-opposing triangle flap as designed. Abdominal exploration revealed that the intestines were joined at Meckel diverticulum, and the colon was shared. The general surgeons divided the single rectum into halves and reconstructed two rectums. The shared colon was allocated for the twins as previously reported.9 Urologic exploration confirmed the result of the three-dimensional model regarding duplicated bladders; twin B had an exstrophic bladder. Urologists reconstructed the bladder and performed a cystostomy. With the aid of electrical stimulation, an anoplasty was performed at the anal dimples. A two-cavity ileostomy was performed for twin B, and a terminal ileostomy and proximal colostomy was performed for twin A. Plastic surgeons used a triangle flap to close the wounds without tension. As suspected, the area of that flap was approximately 10 cm2 smaller than the exposed area, so surgeons sutured the distal skin edges together horizontally, achieving primary wound closure (Figure 3C, D).
The twins recovered uneventfully without any complications and were discharged 18 days after surgery. The colostomies were closed 4 months later. Two years after separation, the twins were healthy and had normal bowel movements, and the parents were satisfied with the appearance of the wounds.
Four days after separation, the twins’ providers noted wound infections at the distal end of the triangle flap (15% of the lower flap) and in the rectangle flaps, with purulent discharge culture positive for Staphylococcus aureus. Blood cultures were negative. Because the infection manifested as skin necrosis without affecting fat or fascia, the wound closed after daily irrigation with sterile saline and silver-containing dressing changes for 1 week. During the 1-year follow-up period, the twins exhibited normal liver function and development, and the wounds healed well (Figure 4C, D).
These twins had a quick recovery, and the wounds healed well after separation. Because of their complicated presentation, the twins had a series of surgeries after separation, including an Achilles tendon lengthening procedure for clubfoot and a urethroplasty, straightening of the penis, and second phalloplasty 1 year later. After closure of the enterostomy, the defect and wounds healed well without complications.
Separation of conjoined twins usually requires extensive interprofessional cooperation. The goal of the plastic surgeons on the team is to reconstruct the skin and tissue defects after separation; numerous strategies have been suggested for this purpose. Tissue expanders have been reported on most often for covering the conjoined area; however, complications of tissue expanders are common, such as skin ulcers, expander rupture, and infections.7 Further, expanders require patients to undergo anesthesia twice, and pediatric patients with implanted expanders require extra caution to avoid complications. Skin grafts are a possibility when the deficit is large and primary closure is not feasible.10–12
In contrast, opposing flaps have wide-ranging applications and are usually used to avoid or relieve linear scar contraction.13–15 Opposing flaps have been reported previously in the separation of pygopagus twins. Although primary wound closure was achieved, the rationale for the flap design was not clarified.8,16 This work justified opposing triangle flaps based on basic mathematic truths. Given the extensibility of skin, opposing triangle flaps could cover more area than an incision through the middle of the skin bridge. Particularly important, plastic surgeons measured the conjoined areas during physical examination and calculated the exact area that the design flap could cover before surgery to affirm that patients could achieve primary closure without tension or a skin graft.
The authors acknowledge the limitations of this small case series. First, the omphalopagus twins experienced wound infections and necrosis. This might have occurred because the entire flap was so large that the distal part had poor blood supply. Although the ratio of the width and the length was nearly 1:1 for the lower part of the flap, which conformed to the flap design principle, there may have been variation in abdominal vessels in the twins. In future, preoperative vascular Doppler examination and CT angiography could provide detailed and accurate vascular distribution information for enhanced flap design. Further, flap delay could help establish better blood supply of the large flap to diminish the possibility of flap failure.17 In addition, a smaller flap could have been designed, and the skin edges could have been approximated over the exposed area, as with the ischiopagus case.
Second, although surgeons printed three-dimensional models, they calculated the area of soft tissue needed and the flap using a two-dimensional method that treated the conjoined area or flap as a triangle or rectangle. For more precise flap design, it may have been better to use software combined with imaging to calculate the area in three dimensions.18
Finally, because of the small sample size and the variations in conjoined twin type, this work could not establish a precise formula for calculating the exact required flap area to achieve primary closure.
With careful measurement and calculation, opposing triangle flaps may be a good choice for plastic surgeons looking to achieve primary wound closure in conjoined twin separation surgeries. Future research involving larger samples, different types of conjoined twins, and a precise calculation model would be beneficial to determine the feasibility and safety of opposing triangle flaps for this purpose.
1. Tannuri AC, Batatinha JA, Velhote MC, et al. Conjoined twins—twenty years' experience at a reference center in Brazil. Clinics 2013;68:371–7.
2. Spitz L. Conjoined twins. Prenat Diagn 2005;25(9):814–9.
3. Jackson OA, Low DW, LaRossa D. Conjoined twin separation. Plast Reconstr Surg 2012;129(4):956–63.
4. Sun J, Zhang Y, Ruan Q, et al. Gains from the separation of conjoined twins. Plast Reconstr Surg 2008;122(2):87e–8e.
5. McDowell BC, Morton BE, Janik JS, et al. Separation of conjoined pygopagus twins. Plast Reconstr Surg 2003;111(6):1998–2002.
6. Kesan K, Gupta A, Gupta RK, et al. Gluteal flap for omphalocele repair in a case of epigastric heteropagus: a novel approach for surgical management. Indian J Plast Surg 2013;46(1):127–9.
7. Dasgupta R, Wales PW, Zuker RM, et al. The use of Surgisis for abdominal wall reconstruction in the separation of omphalopagus conjoined twins. Pediatr Surg Int 2007;23(9):923–6.
8. Matta H, Jacobsz A, Auchincloss J, et al. Successful separation of pygopagus conjoined twins. J Pediatr Surg 2006;41(3):586–8.
9. Liu X, Dong K, Zheng S, et al. Separation of pygopagus, omphalopagus, and ischiopagus with the aid of three-dimensional models. J Pediatr Surg 2018;53(4):682–7.
10. Wong C, Lin C, Fu B, et al. Reconstruction of complex abdominal wall defects with free flaps: indications and clinical outcome. Plast Reconstr Surg 2009;124(2):500–9.
11. Park JS, Park JJ, Kim DY, et al. Emergency transplantation of free flap between separated thoraco-omphalopagus conjoined twins. Arch Plast Surg 2016;43(1):97–9.
12. Canty TS, Mainwaring R, Vecchione T, et al. Separation of omphalopagus twins: unique reconstruction using syngeneic cryopreserved tissue. J Pediatr Surg 1998;33(5):750–3.
13. Ertaş NM, Borman H, Deniz M, et al. Double opposing rectangular advancement elongates tension line as much as Z-plasty: an experimental study in the rat inguinal skin. Burns 2008;34(1):114–8.
14. Ertaş N, Borman H. Double opposing rectangular advancement flap is an alternative technique in the treatment of wide linear postburn scar contractures. Burns 2011;37(8):1449–57.
15. Moulton-Barrett R, Vanderschelden B. Double-opposing unilobar rotation flaps in the reconstruction of moderate-to-large defects of the scalp. J Craniofac Surg 2015;26(6):e523–5.
16. Zhi-Cheng Z, Tian-Sheng S, Fang LI, et al. The application of computer aided design and computer aided engineering technique in separation of pygopagus conjoined twins. Natl Med J China 2009;89(19):1322–5.
17. Hudson DA. Missed closed degloving injuries: late presentation as a contour deformity. Ann Plast Surg 1996;36(3):238–42.
18. Trost JG, Lin LO, Clark SJ, et al. Separation of thoraco-omphalo-ischiopagus conjoined twins. Plast Reconstr Surg 2016;138(5):1064–72.