Cement Capping of Prominent Spinal Implants to Prevent Skin Breakdown

Early-onset scoliosis refers to a scoliotic deformity that presents in children under the age of 10 and thus represents a wide range of etiologies. The MURCS association (Müllerian, Renal, Cervicothoracic, and Somite abnormalities) is a rare development disorder—a likely variant of the more common Mayer-Rokitansky-Küster-Hauser syndrome—often presents with early-onset, progressive cervicothoracic scoliosis and/or kyphosis. When external interventions fail to halt the progression of the spinal deformity, growth-friendly internal implants, such as growing rods, are often utilized to stabilize the spine and allow for continued thoracic growth. Traditional growing rods (TGR) require multiple lengthening procedures, often through the same incision site, which may lead to thinning of the soft tissue envelope and predispose the skin to breakdown. This may be exacerbated by implant prominence in relatively thin patients and those with impaired nutritional status, which can accompany non-idiopathic scoliosis. Additional alteration of the stress forces at the proximal aspect of the construct may result in adjacent segment degeneration and the development of proximal junctional kyphosis (PJK). Previous studies report rates of PJK up to 26% in patients with early-onset scoliosis treated with dual rod constructs and increasing rates of PJK with a pre-existing hyperkyphotic deformity.1,2 Patients with MURCS association may be uniquely predisposed to the development of PJK as the deformity is centered around the cervicothoracic junction, and posterior instrumentation often does not extend into the cervical spine to preserve cervical motion. We present the case of a patient with MURCS association and severe, progressive cervicothoracic kyphoscoliosis undergoing treatment with spinal growing rods. The treatment course was complicated with the development of PJK and prominent apical hardware threatening the skin integrity. This was managed with cement capping of the prominent pedicle screw to prevent skin breakdown and implant exteriorization. TECHNIQUE The patient presented at age 5 months with torticollis, 20-degree right thoracic scoliosis, and MURCS association: right renal agenesis, total anomalous pulmonary venous return, congenital duplication of the uterus, accessory spleen, and conductive hearing loss. Physical therapy was initiated. Scoliosis progressed to 40 degrees at age 8 months, bracing with a custom spine orthosis was initiated before cardiac surgery. At age 12 months, serial manipulative spine casting was started. At age 23 months, bipolar right sternocleidomastoid lengthening was performed in conjunction with a casting procedure because of resistant torticollis. Casting was performed over a 2-year period, but the deformity gradually progressed to a 72-degree right thoracic scoliosis with an associated 75-degree kyphosis. At age 3 years, halo gravity traction was performed for 6 weeks, followed by implantation of traditional growing rods from T2-L3. The rods were lengthened every 6 months until the longitudinal connector needed to be replaced because of growth. She was then converted to magnetically controlled growing rods at age 5 years. In-office rod lengthening then proceeded every 3 months. Over the next 2 years, progressive torticollis combined with PJK at the cervicothoracic junction led to increasing neck pain and unacceptable cosmesis. The rods were then removed, repeat sternocleidomastoid lengthening was performed, and halo gravity traction was reinitiated. At age 7 years, she then underwent growing rod revision, with extension of the proximal fixation to C5 to control the cervicothoracic deformity, and a TGR construct was reimplanted caudally to allow for continued growth between T5 and L1. Magnetically controlled growing rods were not used during the revision because of her underlying anxiety and difficulty tolerating the magnetic lengthening procedures in the clinic. The plan was to continue with operative lengthening of the TGR until sufficient growth had occurred and then convert to a definitive posterior spinal fusion. At 2 months postoperative, it was noted that there had been significant thinning of the skin and soft tissues over a prominent upper thoracic pedicle screw at the apex of the residual kyphotic deformity (Fig. 1). At 6 months postoperative, she underwent routine lengthening of the growing rods. At that time, full thickness flaps were elevated, exposing the prominent apical pedicle screw to facilitate soft tissue approximation and closure. Antibiotic-impregnated bone cement was mixed with vancomycin powder and molded around the screw and rod to create a smooth contour (Fig. 2). The paraspinous muscles were mobilized by the plastic surgery team, and the incision was closed in layers. The incision was managed with routine wound care.FIGURE 1: Severe thinning of the soft tissues overlying a prominent screw at the apex of the cervicothoracic kyphosis (A, B) and preoperative lateral x-ray (C) with an arrow pointing at the prominent screw.FIGURE 2: Intraoperative images before (A) and after cement capping (B) of the prominent pedicle screw.OUTCOMES The 10-year-old patient is now 2-years postoperative following cement capping of the prominent apical pedicle screw. She had been undergoing routine operative lengthening of the TGR without pain or skin breakdown at the level of the prominent apical pedicle screw. The skin is intact over the cement cap without a sharp edge from the implant (Fig. 3). Definitive posterior spinal fusion with upper thoracic vertebral column resection and cephalad extension of the instrumentation is now being considered to improve the residual deformity and optimize head position.FIGURE 3: Clinical appearance at 2-years postoperative (A, B) and postoperative lateral spine radiograph (C) demonstrating the intact cement cap (arrow).COMPLICATIONS Complications related to the application of the cement cap over the prominent implant were not observed. Before hardening, the cement was carefully molded around the implants in an effort to prevent migration of the cement. At the time of subsequent revision instrumentation, we anticipate that the cement will not significantly impede access to the implants and will be easily removed with standard instruments such as curettes and osteotomes. There is a potential concern that the implant may remain prominent even after the application of the cement bolster. Care was taken to shape the cement cap to ensure a smooth contour was achieved, and soft tissue mobilization was used to facilitate adequate closure over the construct. There is also a theoretical systemic toxicity posed from the use of antibiotic-impregnated cement with tobramycin and vancomycin. Previous studies have shown that topical vancomycin applied after instrumented posterior spinal fusion does not produce serum levels near the toxic threshold, and serum tobramycin levels remain low following implantation with antibiotic-impregnated cement.3,4 The risk of elevated serum levels may increase in smaller children, and appropriate drug dosing should be followed. Alternatives to manage the prominent apical pedicle screw were considered. One option would be to remove the prominent screw. This would have required a larger intervention with proximal extension of the exposure to temporarily remove the rod from the upper implants. If the fixation point is no longer needed and no further surgery is planned, a prominent pedicle screw cradle can be removed with a metal cutting burr without disrupting the associated rod. This leaves the shaft of the screw in the pedicle but removes the prominent metal edges. Another option would be the advancement of paraspinous muscles alone with a multilayered plastic surgery wound closure. In our experience, isolated muscle flap advancement over a prominent pedicle screw at the apex of a kyphotic deformity does not provide sustained soft tissue coverage. The tension at the apex of the deformity allows the relatively sharp edges of the prominent screw to gradually wear through a thick muscular wound closure. In summary, cement capping of prominent spinal implants may be selectively considered in the armamentarium of a spinal deformity surgeon to reduce the potential for skin breakdown. This technique may be particularly useful in patients without robust soft tissue coverage over spinal implants.