Scoliosis is one of the most serious skeletal deformities in patients with cerebral palsy (CP). Surgery to correct scoliosis has been shown by several studies to provide improvements in quality of life for these patients. However, the complications of the surgery are more frequent, the techniques more varied, and the results of surgery are less uniform than in correcting idiopathic scoliosis curves in the same age group. Most prior reports in the CP population have studied corrective procedures performed at one center and with 1 or 2 techniques. However, there are several basic categories of implants used. One major distinction is between a unique one-piece, precontoured (“unit”) rod and those contoured by the surgeon (“custom-bent”) rods. The spinal and pelvic anchors and the process of curve correction differ fundamentally between the unit rod and custom-bent rods. This study used a multicenter design to examine the differences in intraoperative factors and postoperative results in spinal deformity correction for patients with CP using unit rods versus custom-bent rods.
Materials and Methods
One hundred fifty-seven children with CP who had a posterior spinal fusion at 1 of 9 centers participating in the Harms Study group were retrospectively analyzed, after approval at each of the Institutional Review Boards. All patients were under 18 years of age at surgery and all procedures were performed between 1995 and 2004. Minimum clinical follow-up of 2 years was required. Decisions regarding posterior versus combined approaches, and implant type were made by individual surgeons. Seventy-nine patients had unit rods (U) (Sofamor-Danek, Nashville, TN) (Figure 1) and 78 had custom-bent rods (C) of many different manufacturers. Most of the unit rods were inserted at 4 of the 9 centers. Four centers used primarily custom-bent rods; one center used approximately equal numbers of both implant types. Pelvic anchors in the custom-bent rods included iliac screws or rods, S-rods, or sacral screws. Clinical and radiographic data from the preoperative, intraoperative, and postoperative period were gathered by the individual sites, deidentified, and analyzed centrally. Pelvic obliquity was measured as described by Abel et al,1 using the angle between the T1–S1 line and a line between the tops of the iliac crests on upright radiographs. The absolute value of the difference between this angle and 90° was termed pelvic obliquity.2 Caregivers were asked to rate the clinical benefit of the surgery as excellent, good, fair, or poor.
For the continuous variables, analysis of variance was used to examine differences between the unit and custom-bent rod groups. If a dependent variable was not normally distributed or had unequal variances, a nonparametric Mann-Whitney U test was performed. χ2 test was used for the categorical variables. Significance was set at 0.05.
The mean age at surgery was 13.5 ± 2 and the mean follow-up was 3.5 ± 1.3 years (range, 2–12). Mean scoliosis curve size for the entire group was 77 ± 22° (range, 22–128). The mean preoperative curve size for patients instrumented with unit rods was 70 ± 19° and 82 ± 23° for custom-bent rods (P = 0.001). The pelvic obliquity was identical at 23° for each group. There were 109 posterior-only and 48 anterior and posterior surgeries. Of the anterior and posterior procedures, 39 were performed on the same day and 9 were staged. Unit rod surgeries had estimated blood loss of 2124 mL, which was not statistically different than the 1885 mL for custom rods (P = 0.3). Posterior-only surgical time averaged 311 minutes for the unit rods and 309 minutes for the custom rods (P = 0.5). Combined surgical time averaged 339 minutes for unit rods and 379 minutes for custom rods (P = 0.04). However, unit rod patients required more allogeneic blood replacement and spent more time in the intensive care unit (ICU) (4 vs. 3 days; P = 0.001) and in hospital (14 vs. 13 days; P = 0.006). Final major Cobb correction was similar for both groups (63% U, 66% C), and there was no difference in the percent correction between the implant types whether the curve had an apex above or below T10. Final pelvic obliquity correction was significantly greater with the unit rod: 74% U versus 22% C (P = 0.002) (Figure 2). The loss of correction of pelvic obliquity after surgery was also significantly different between the 2 implant types, with custom implants losing more correction over time (P = 0.004 for the first year and 0.03 for the second year). Numbers of patients did not permit comparison of the 4 different pelvic anchor types among the custom rods. Patients with unit rods had more instances of clinically apparent implant prominence at 2-year follow-up (12 vs. 2 instances, P = 0.03); most of these were proximal (Figure 3). Implant-related reoperations were 8 for each group. Neurologic complications were seen in 5 custom and 4 unit, which was not significantly different. Infections were seen in 15% of unit rods and 5% of custom rods (P = 0.03). The only factor which predicted overall complication rate for both groups was absolute curve magnitude (P = 0.04). The caregivers' assessment of the utility of surgery was higher for the patients who received custom-bent rods (P = 0.05). These findings are summarized in Table 1.
The correction of spinal deformity in CP has undergone numerous reassessments and refinements. Early reports from Minneapolis showed improved efficacy after anterior and posterior fusion for neuromuscular scoliosis using implants of the time. This was followed by widespread adoption of circumferential fusion for scoliosis in CP. The merits of single-session versus staged approaches are still debated.3–5 More recent reports6,7 have shown that posterior fusion alone can often be effective in controlling curves and even in preventing crankshaft phenomenon if a rigid posterior implant such as the unit rod is used. Exactly where this boundary should be drawn in terms of age, curve size, and rigidity remains to be defined.8 Nevertheless, the concept of rigid spinal instrumentation first exemplified by the unit rod and followed by other forms of segmental fixation, has ushered in a new era of spine correction for the neuromuscular population. Long iliac anchors provide superior control of pelvic obliquity, while necessitating more dissection to insert. The standardized shape of the unit rod produces a relatively uniform coronal and sagittal profile. It does not lend itself to use of hooks or screws as originally designed. Estimation of correct implant length is difficult in patients with large curves. Dissection at the cranial end of the spine to pass wires may produce a tendency to junctional kyphosis. We also had a hypothesis that unit rods would not produce adequate control of curves with an apex in the higher thoracic spine, although this was not borne out in this study.
In parallel, the potential morbidity of the surgery and the difficulty in assessing its clinical benefit have been highlighted. In 2003, Tsirikos et al studied outcomes including life expectancy in a sizeable group (n = 288) of patients who underwent spinal fusion for neuromuscular scoliosis with spasticity. A host of variables were reviewed, including ambulatory status, magnitude of spinal deformity, intraoperative blood loss, surgical time, and length of hospital stay. Survival rates were analyzed and revealed that the only contributing factors were (1) length of time spent in the intensive care unit, and (2) the degree of preoperative hyperkyphosis.9 Through studies such as this, the concept has been established that this is a vulnerable population for whom the choice of the safest and most effective surgical techniques can be life-saving. In the current study, the main variable which predicted an increased risk of complications was curve magnitude. A retrospective study by Sponseller et al10 and prospective study by Jones et al,11 the latter using a validated outcomes database, showed that parents and caregivers felt that spine surgery was beneficial in most but not all patients with CP. The current study showed that the caregivers rated clinical benefit to be higher with custom than unit rods. Further study will be required to validate and explain this finding.
Several studies have discussed the often extensive intraoperative blood loss in neuromuscular patients, but did not compare the difference between implant types.12–18 A broader look at this and other common intraoperative and perioperative complications with different implant types was undertaken in this study. This report looked at the 2 major instrumentation strategies for scoliosis in CP. The study showed that the unit rod provided superior correction of pelvic obliquity. This may be explained by the fixed, 90° angle between the longitudinal rods versus the pelvic limbs. The unit rod generates very powerful cantilever forces by virtue of its hold on the pelvis. This virtually guarantees that the pelvis will be perpendicular to the spine if the fixation holds. Furthermore, it assures that the primary curve and the compensatory or fractional curves will be in balance with the head over the pelvis in the frontal plane. In other words, the correction of the primary curve and the compensatory or fractional curves is balanced so that one segment does not achieve greater or less correction and the trunk is perpendicular to the pelvis. This balance is believed to be the most important functional outcome of neuromuscular spine surgery. In contrast, custom rods may allow the surgeon to correct the major curve to a greater degree than the secondary curves, potentially leading to imbalance.
The negative findings in patients treated with the unit rod in this study were the increased blood loss, ICU, and hospital stay. There was an increased rate of infection and of rod prominence proximally. The increased blood loss may be a result of the multiple laminotomies, wire passage, and wire manipulation, although this is speculative. We hypothesize that the increased incidence of infection after unit rod surgery may be due to the dissection of the subcutaneous tissue and muscle needed to make the transition from the iliac wings to the spine. For insertion of a unit rod, subcutaneous flaps must be elevated bilaterally to the iliac crests, and the paraspinous muscle over the midline is incised or stretched to where the rod joins the iliac posts to the midline. These steps are not needed for insertion of an S-rod or sacral screws. For iliac screws, although the lateral dissection is the same, a connector may be used to join the screw to the rod with less tissue trauma. The increased incidence of clinically significant proximal implant prominence may be a result of the shape of the rod, which has a dorsally-directed bend at the top, as well as the supralaminar dissection needed to pass the wires proximally. Further development of implants may be able to address these issues.
The limitations of this study lie in its retrospective nature and variable indications for combined versus posterior procedures. There was no significant difference in the experience level of surgeons between the centers. All surgeons had been practicing at least 5 years from fellowship and had significant experience with the surgical treatment of scoliosis in CP. The unit rod was used more commonly by surgeons having pediatric orthopedic fellowship training, whereas custom-bent rods were used more often by surgeons with spinal fellowships. In addition, surgical techniques have evolved over time, which was the main reason for limiting it to a 10-year interval.
Unit rod provides superior correction of pelvic obliquity but more proximal implant prominence. It may be associated with an increased transfusion requirement, infection rate, ICU, and hospital stay.
- Unit Rod provided better correction of pelvic obliquity than custom rods.
- Patients treated with unit rods experienced more proximal implant prominence.
- There was no difference in correction of the major curve between patients treated with unit or custom-bent rods.
1. Abel MF, Blanco JS, Pavlovich L, et al. Asymmetric hip deformity and subluxation in cerebral palsy
: an analysis of surgical treatment. J Pediatr Orthop
2. Gupta MC, Wijesekera S, Sossan A, et al. Reliability of radiographic parameters in neuromuscular scoliosis
3. Yazici M, Asher MA, Hardacker JW. The safety and efficacy of Isola-Galveston instrumentation and arthrodesis in the treatment of neuromuscular spinal deformities. J Bone Joint Surg Am
4. Lipton GE, Letonoff EJ, Dabney KW, et al. Correction of sagittal plane spinal deformities with unit rod instrumentation in children with cerebral palsy
. J Bone Joint Surg Am
5. Tsirikos AI, Chang WN, Dabney KW, et al. Comparison of one-stage versus
two-stage anteroposterior spinal fusion
in pediatric patients with cerebral palsy
and neuromuscular scoliosis
6. Smucker JD, Miller F. Crankshaft effect after posterior spinal fusion
and unit rod instrumentation in children with cerebral palsy
. J Pediatr Orthop
7. Westerlund LE, Gill SS, Jarosz TS, et al. Posterior-only unit rod instrumentation and fusion
for neuromuscular scoliosis
8. Thacker M, Hui JHP, Wong HK, et al. Spinal fusion
and instrumentation for paediatric neuromuscular scoliosis
: retrospective review. J Orthop Surg (Hong Kong)
9. Tsirikos AI, Chang WN, Dabney KW, et al. Life expectancy in pediatric patients with cerebral palsy
and neuromuscular scoliosis
who underwent spinal fusion
. Dev Med Child Neurol
10. Sponseller PD, Whiffen JR, Drummond DS. Interspinous process segmental spinal instrumentation for scoliosis
in cerebral palsy
. J Pediatr Orthop
11. Jones KB, Sponseller PD, Shindle MK, et al. Longitudinal parental perceptions of spinal fusion
for neuromuscular spine deformity in patients with totally involved cerebral palsy
. J Pediatr Orthop
12. Brenn BR, Theroux MC, Dabney KW, et al. Clotting parameters and thromboelastography in children with neuromuscular and idiopathic scoliosis
undergoing posterior spinal fusion
13. Edler A, Murray DJ, Forbes RB. Blood loss during posterior spinal fusion
surgery in patients with neuromuscular disease: is there an increased risk? Paediatr Anaesth
14. Shapiro F, Sethna N. Blood loss in pediatric spine surgery. Eur Spine J
15. Cassidy C, Craig CL, Perry A, et al. A reassessment of spinal stabilization in severe cerebral palsy
. J Pediatr Orthop
16. Sponseller PD, LaPorte DM, Hungerford MW, et al. Deep wound infections after neuromuscular scoliosis
surgery. A multicenter study of risk factors and treatment outcomes. Spine
17. Thometz JG, Simon SR. Progression of scoliosis
after skeletal maturity in institutionalized adults who have cerebral palsy
. J Bone Joint Surg Am
18. Edwards BT, Zura R, Bertrand S, et al. Treatment of neuromuscular scoliosis
with posterior spinal fusion
using the Galveston technique: a retrospective review and results of 62 patients. J Long Term Eff Med Implants