Supplemental Digital Content is available in the text.
Summary:
Robust, reliable, and reproducible closure of lumbosacral myelomeningocele defects remains a challenge. In infants with lumbosacral myelomeningocele defects, multiple methods of soft tissue coverage have been described. These include various cutaneous, fascial, and muscle flaps and grafts. This is done with relative ease when ample soft tissue is present but becomes extremely difficult for large and distally located defects. We present here our closure technique of lumbosacral myelomeningocele defects in newborns, with associated short- and medium-term outcomes. We demonstrate the anatomy of this technique with fresh cadaver dissection and present a review of demographic and outcome data of 12 consecutive patients treated with this method from June 2014 to August 2019. No major intra- or postoperative complications have been encountered, with a mean follow-up of 22.2 months and median follow up of 18 months. After the neurosurgical repair of lumbosacral myelomeningocele, bilateral composite fascial flaps composed of thoracolumbar and gluteus maximus fascia are elevated in continuity. The paraspinous muscle flaps are then elevated, disinserted distally, and medialized to provide complete muscular coverage of the dural repair. The bilateral composite fascial flaps are medialized and closed over the deep paraspinous muscle flap repair. Two patients experienced areas of small, superficial skin necrosis, one of which healed by secondary intention and the other by debridement and full-thickness skin grafting. Use of bilateral paraspinous muscle flaps and bilateral composite fascial flaps composed of thoracolumbar and gluteus maximus fascia provides robust coverage of lumbosacral defects following myelomeningocele repair in infants.
INTRODUCTION
Myelomeningocele is a congenital malformation of the central nervous system, which is most frequently observed in the lumbosacral region.1 Robust, reliable, and reproducible closure of lumbosacral myelomeningocele defects remains a challenge. Surgical repair is done within the first 48 hours of life to decrease the risk of injury and central nervous system infection.2,3
Following neurosurgical myelomeningocele repair consisting of tubularization of the placode and dural closure, soft tissue coverage is performed. Multiple methods of soft tissue coverage have been described. These include various cutaneous, fascial, and muscle flaps and grafts.2,4,5 This is done with relative ease when ample soft tissue is present but becomes extremely difficult for large and distally located defects. We present here our multilayered closure technique of lumbosacral myelomeningocele defects in newborns, with associated short- and medium-term outcomes.
PATIENTS AND METHODS
Under institutional review board approval, this method of closure was performed in 12 consecutive patients at our institution from June 2014 to August 2019. Patient demographics, myelomeningocele defect characteristics, and complication profile were analyzed.
To better delineate the anatomy of our surgical repair, dissections were performed in 2 fresh cadavers. The anatomy and transition points of the composite thoracolumbar and gluteus maximus fascial flaps were demonstrated.
SURGICAL TECHNIQUE
After neurosurgical placode tubularization and dural repair, poor-quality skin surrounding the myelomeningocele defect is excised. Midline, costal margin, and iliac crests are marked preoperatively (see figure 1, Supplemental Digital Content 1, which shows preoperative myelomeningocele markings: solid black lines denote iliac crests, dashed lines indicate inferior costal margin, http://links.lww.com/PRSGO/B401). Bilateral composite flaps that include thoracolumbar fascia and gluteus maximus fascia are then elevated in continuity (Fig. 1). Fascia propria of the paraspinous muscle is included in the fascial dissection to increase the robustness of the medial portion of the flap. As the dissection is carried laterally, care is taken to avoid violation of the lateral abdominal wall muscles and retroperitoneum. The dissection is performed in a subperiosteal plane along the iliac crest to elevate the thoracolumbar and gluteal fascia flaps in continuity (see figure 2, Supplemental Digital Content 2, which shows composite flaps at sacral level: A, Cadaver dissection: gluteal fascial flaps are raised, leaving the underlying muscle fibers intact. Blue line indicates iliac crest. Dissection here is subperiosteal to maintain flap continuity. B, Cross-sectional diagram showing plane of dissection (black arrows), http://links.lww.com/PRSGO/B402).
Fig. 1.

Composite flap at lumbar level. A, Cadaver dissection: latissimus dorsi fasciocutaneous flaps (green asterisk) is raised in continuity with the thoracolumbar fascia (red asterisk) and the gluteus maximus fasciocutaneous flaps (white asterisk). Blue line indicates iliac crest. B, Cross-sectional diagram showing the plane of dissection (black arrows).
Bilateral paraspinous muscle flaps are then undermined and elevated based on the lateral row arterial perforators from the sacrum to the superior aspect of dural repair. Distally, the paraspinous muscles are completely disinserted from the sacrum and ilium. The paraspinous muscle flaps are advanced inferomedially, sutured to each other, and anchored to the bone for complete coverage of the dural repair (Fig. 2).
Fig. 2.

Paraspinous muscle dissection. A, Release of the distal paraspinous muscles (pickups), asterisk denotes dural repair. B, Approximation of paraspinous muscles in the midline, and flap inset completely covering dural repair.
The composite thoracolumbar and gluteus muscle fascial flaps are then medialized and sutured to each other over the paraspinous muscle flap repair (see figure 3, Supplemental Digital Content 3, which shows approximation of composite fascial flap in the midline over the paraspinous muscle flaps, http://links.lww.com/PRSGO/B403). The overlying skin is then closed in layers (see figure 3, Supplemental Digital Content 3, which shows skin approximation in the midline, http://links.lww.com/PRSGO/B403). No additional maneuvers such as back cutting of the muscle or lateral relaxing incisions are required to achieve complete muscular coverage with paraspinous flaps and composite thoracolumbar and gluteus muscle fascial flaps. Postoperatively, the infants are kept in the prone position for 6 weeks to protect the repair except for during feeding, when they are held in an upright position with no pressure on the repair site.
RESULTS
Baseline characteristics of the 12 patients are featured in Table 1. All of the patients had defects in the lumbosacral region. No major intra or postoperative complications have been encountered, with a mean follow-up of 22.2 months and median follow-up of 18 months.
Table 1.
Outcome Data
| Patient | Birth Weight, g | Size of Defect, cm2 | Length of Follow-up, mo | Complications | Received VP Shunt Postoperatively |
|---|---|---|---|---|---|
| 1 | 3270 | 25 | 3 | 15 days postoperative: 2 cm superficial wound dehiscence over central third of wound, healed by secondary intention | Yes |
| 2 | 3260 | 25 | 3 | None | No |
| 3 | 3730 | 25 | 6 | 14 days postoperative: 2 cm superficial dehiscence, managed with skin graft | Yes |
| 4 | 3380 | 25 | 9 | None | Yes |
| 5 | 2580 | 25 | 17 | None | Yes |
| 6 | 4095 | 12 | 17 | None | No |
| 7 | 3232 | 9 | 18 | None | No |
| 8 | 3200 | 12 | 20 | None | Yes |
| 9 | 3615 | 25 | 28 | None | No |
| 10 | 2550 | 36 | 36 | None | No |
| 11 | 3380 | 16 | 48 | None | No |
| 12 | 3870 | 42 | 61 | None | Yes |
| Mean: 3346 g | Mean: 23 cm2 | Mean: 22.2 mo | 2/12 patients | 6/12 patients |
VP, Ventriculoperitoneal shunt.
All of the patients had robust soft tissue coverage at the level of the previous defect with minimal scarring (see figure 4, Supplemental Digital Content 4, which shows typical 1-year postoperative follow-up appearance of the myelomeningocele closure, http://links.lww.com/PRSGO/B404). Two of the patients had partial dehiscence of the skin closure. Importantly, muscular coverage was adequately deep to the skin dehiscence, negating the need for any neurosurgical intervention. In the first case, a 2 cm2 area of skin breakdown was allowed to heal by secondary intention. In the second case, skin necrosis measuring 2 cm2 was debrided in the operating room on postoperative day14 with the placement of Integra followed by full-thickness skin grating 4 weeks later. Both patients in subsequent follow-up had good skin integrity and have not required further intervention.
DISCUSSION
In our prior experience, we used a 2-layered closure comprising paraspinous muscle fascia and skin over the dural repair. In the first 4 cases of the senior author’s (B.G.) experience, 2 of the patients developed wound dehiscence and cerebrospinal fluid (CSF) leak, which necessitated additional interventions, protracted hospital courses, and delayed definitive closure. This was observed mostly at the caudal end of lumbosacral defects where fascial coverage was the least robust. Given the high complication rate (50% CSF leak) with the initial method, we sought a technique that would provide a reliable and more robust closure.
Compared with skin closure alone, it is generally agreed upon that the addition of fascial or muscle flaps over dura provides a more robust repair, less risk of CSF leak, and pseudomeningocele formation.6–8 Prior studies have reported a CSF leak after myelomeningocele closure to be approximately 1%–3%.6,9 In a medium-term follow-up with our novel technique, no patients have experienced a CSF leak.
Previous paraspinous muscle closure has been described but only as bipedicled flaps for defects located in the high lumbar region.6,7 However, no reports comment upon distal disinsertion of the paraspinous muscles for muscular coverage of caudal spinal defects. In our experience, the paraspinous muscles (once disinserted distally) allow robust and complete coverage of lumbosacral defects. We contend that our layered closure with well-vascularized muscle in addition to robust composite fascial flaps mitigates breakdown complications. If skin breakdown occurs over the midline as we saw in 2 of our patients, there is still well-vascularized muscle buttressing the dural repair from the external environment. The resulting defect can be easily managed via healing by secondary intention, skin graft, or local re-elevation of cutaneous flaps.
An additional potential benefit of this technique is its midline closure. Midline scars are advantageous in the long term because they facilitate exposure for potential secondary spinal surgeries. One-third of patients with myelomeningocele undergo detethering of the spinal cord and 21% will need spinal fusion for scoliosis.8 As such, we avoid placing suture lines away from the midline neural repair (bipedicled, Limberg, rotation, V-Y advancement, and perforator flaps) to limit scarring and skin necrosis of the posterior trunk.
One potential disadvantage of disinserting the paraspinous muscles distally is partial defunctionalization. Fifty-nine percent of patients born with myelomeningocele will be community ambulators, 16% household ambulators and 25% will be non-ambulatory.10 To our knowledge, no long-term studies specifically examine the functional status changes of myelomeningocele patients after paraspinous muscles are disinserted for flap coverage. We have yet to see any functional differences in our patients after using this technique but acknowledge that our outcomes are limited by short- and medium-term outcomes.
CONCLUSIONS
The use of bilateral paraspinous muscle flaps, composite thoracolumbar, and gluteus maximus fasciocutaneous flaps provides robust, reproducible coverage of lumbosacral myelomeningocele defects in infants. Complication rates for this method are minimal, and this technique should have a place in the surgeon’s armamentarium for the treatment of low lumbosacral myelomeningocele defects.
PATIENT CONSENT STATEMENT
Parents or guardians provided written consent for the use of the patients’ image.
Supplementary Material
Footnotes
Published online 17 June 2020.
Presented at the Mountain West Society of Plastic Surgeons Meeting, March 8, 2018, Jackson Hole, Wyo., and at Plastic Surgery The Meeting, September 28, 2018, Chicago, Ill.
Disclosure: The authors have no financial interest to declare in relation to the content of this article.
Related Digital Media are available in the full-text version of the article on www.PRSGlobalOpen.com.
Drs. Holoyda and Kim equally contributed to this work.
REFERENCES
- 1.Lanigan MW. Surgical repair of myelomeningocele. Ann Plast Surg. 1993;31:514–521. [DOI] [PubMed] [Google Scholar]
- 2.Park HS, Morrison E, Lo C, et al. An application of keystone perforator island flap for closure of lumbosacral myelomeningocele defects. Ann Plast Surg. 2016;77:332–336. [DOI] [PubMed] [Google Scholar]
- 3.Shinde S, Singhal S. Meningomyelocele: the tip of the iceberg. BMJ Case Rep. 2009;2009:bcr04.2009. 1811. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Iacobucci JJ, Marks MW, Argenta LC. Anatomic studies and clinical experience with fasciocutaneous flap closure of large myelomeningoceles. Plast Reconstr Surg. 1996;97:1400–1408; discussion 1409. [DOI] [PubMed] [Google Scholar]
- 5.Moore TS, Dreyer TM, Bevin AG. Closure of large spina bifida cystica defects with bilateral bipedicled musculocutaneous flaps. Plast Reconstr Surg. 1984;73:288–292. [DOI] [PubMed] [Google Scholar]
- 6.Lien S, Maher C, Garton H, et al. Local and regional flap closure in myelomeningocele repair: a 15 year review. Child’s Nervous Syst. 2010;26:1091–1095. [DOI] [PubMed] [Google Scholar]
- 7.Patel KB, Taghinia AH, Proctor MR, et al. Extradural myelomeningocele reconstruction using local turnover fascial flaps and midline linear skin closure. J Plast Reconstr Aesthet Surg. 2012;65:1569–1572. [DOI] [PubMed] [Google Scholar]
- 8.Zide BM, Epstein FJ, Wisoff J. Optimal wound closure after tethered cord correction. Technical note. J Neurosurg. 1991;74:673–676. [DOI] [PubMed] [Google Scholar]
- 9.Talamonti G, D’Aliberti G, Collice M. Myelomeningocele: long-term neurosurgical treatment and follow-up in 202 patients. J Neurosurg. 2007;1075 Suppl368–386. [DOI] [PubMed] [Google Scholar]
- 10.Swank M, Dias L. Myelomeningocele: a review of the orthopaedic aspects of 206 patients treated from birth with no selection criteria. Dev Med Child Neurol. 1992;34:1047–1052. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
