The treatment options for early onset scoliosis (EOS) for the growing child are vast and include nonoperative therapies such as casting and bracing as well as a variety of surgical techniques. Skaggs et al1 described a classification of growth-friendly surgical treatments of EOS that included distraction based such as growing rods2 or vertebral expandable prosthetic titanium ribs,3 growth guided such as Shilla4 or Luque trolley constructs,5 and tension-based techniques such as vertebral stapling or tethers.6 Another growth-friendly surgical treatment was introduced in an attempt to reduce the number of surgical procedures in distraction-based techniques, namely a magnetically controlled growing rods (MCGR).7 Cheung et al8 were the first authors to describe early clinical results of the MCGR technique that did not require open lengthening like traditional growing rods (TGR). Since then several published studies have reported similar results demonstrating promising clinical effectiveness of MCGR and its ability to reduce the number of open procedures.8 However, postoperative complication rates in MCGR surgery remain largely unknown. Previous studies have shown wound and implant complication rates for TGR procedures to be as high as 58%.9,10 Deep surgical site infection (SSI) in pediatric deformity is known to be a risk for curve progression if implants need to be removed.11 It is unclear whether the complication rate in MCGR patients will differ from the traditional TGR technique. This study was performed to characterize postoperative complications in EOS patients who underwent the MCGR procedure.
Five centers involved in an international EOS study obtained ethics approval from their respective institutional review boards or ethic committees to participate in this retrospective review. Inclusion criteria were: (1) diagnosis of EOS of any etiology; (2) 10 years and younger at the time of index surgery; (3) preoperative major curve size of >30 degrees; (4) preoperative thoracic spine height (T1-T12) of <22 cm. Patients who had MCGR surgery from November 2009 to November 2012 and were followed for a minimum of 1 year after the initial implantation procedure were included in the study. Patients with conversion from TGR systems to MCGR instrumentation were also included. The surgical technique for implantation of the MCGR was as described by Cheung et al.8 Typical construct included an upper thoracic proximal foundation including hooks and/or pedicle screws and mid-lumbar or low-lumbar spine pedicle screws.
Deep SSI was defined according to the criteria of the Center for Disease Control and Prevention as modified by Horan et al.10 For the purpose of this study, we also defined deep SSI as any infection that required additional surgical intervention. Complications were categorized as wound related and implant related. Implant-related complications were classified as rod breakages, rod malfunction, and anchor pullouts. Complications were also classified as early (<6 mo from index surgery) and late (>6 mo from index surgery). The technique for magnetically controlled distraction and the interval of distractions were based on surgeon preference without standardization across sites. Since there is no consensus on complications classification and the only available classification system reported by Smith et al12 does not cover MCGR, the treating surgeons and their respective research teams reported the complications as defined by the study protocol. Before analyzing any data the principal investigator and the study’s monitoring board performed an audit of all patient data including complications.
Univariate analysis was used to assess the difference between patients with complications and those without complications.
A total of 115 patients from 15 sites were screened. Fifty-four patients (32 females and 22 males) who had MCGR surgery (35 first-generation MCGR and 19 second-generation MCGR) between 2009 and 2012 met the inclusion criteria. Twenty-one of 35 first-generation MCGR were primary and 14 were conversion cases. Nine of the 19 second-generation cases were primary and the remaining 10 were conversion cases.
At the time of data collection 34 patients were due for 2-year follow-up of which 24 (70%) patients actually completed 2-year follow-up (16 primary and 8 conversion cases). Mean age at initial surgery was 7.0 years (SD=2.3; range, 2.4 to 10.7 y) in primary cases and 7.8 years (SD=2.4; range, 3.6 to 11 y) in conversion cases. Mean duration of follow-up was 21.2 months (SD=8.9) in primary group and 17.0 months (SD=5.8) in conversion cases.
Twenty-three (42%) of 54 patients had at least 1 complication (14 primary and 9 conversion). Fifteen patients (28%) had at least 1 revision surgery. Seven rod breakage occurred in 6 patients (11%) including 2 primary and 4 conversion cases; 2 patients with 4.5 mm diameter rods (both conversion cases) and 4 patients with 5.5 mm rods (2 primary and 2 conversion). Of the 6 broken rods, 2 (33%) failed early (4 mo; both 5.5 mm) and 4 (66%) failed late (mean=14.5 mo; two 4.5 mm and two 5.5 mm rods).
In 6 patients (11%, 5 primary and 1 conversion) rods did not lengthen at least in 1 episode. Four of the 6 lengthened at subsequent office visits with repeated magnetically controlled distraction, and 2 of the 6 had their rod exchanged. Seven (13.0%) patients (3/7, 43% de novo vs. 4/7, 57% revision) had either proximal or distal implant-related complication either from pedicle screw pullout, loosening or breakage, dacron mesh support failure, or hook migration at an average of 8.4 months. Two patients (3.7%, 1 primary and 1 conversion case) had infections requiring irrigation and debridement. One patient presented with early postoperative wound drainage (2 wk) and was treated with oral antibiotics (amoxicillin and clavulanic acid). The second patient (conversion case) presented with skin penetration of the rod and was diagnosed as having a late methicillin-sensitive Staphylococcus aureus infection (8 mo). The latter case required removal of one of the dual rods and intravenous antibiotic treatment (cefazolin and teicoplanin). Table 1 summarizes all complications by type and frequency.
Several studies have shown that TGR surgery is effective in the treatment of EOS.2,13–15 In the growing child, this may require several open posterior distraction procedures until skeletal maturity is met and as frequent as every 6 months between procedures. The recently introduced MCGR device allows the implantable rods to be distracted in the outpatient setting with a hand-held external magnetic remote controller.
A younger age and the likelihood of having more open posterior distractions have shown to be risk factors for higher complications in traditional rod treatment of EOS.9 One method to decrease the number of complications is to delay implantation of TGR in the younger patient to reduce the number of open surgeries in the treatment period. In theory, MCGR may be suitable for implantation in younger patients because the number of open distraction surgeries is reduced in favor of outpatient closed distraction procedures.
Our experience at early to intermediate follow-up showed that compared with TGR, MCGR results in a lower infection rate (4% in MCGR vs. 11% in TGR).10 In contrast, MCGR does not appear to prevent common implant-related complications such as rod or foundation failure. As powerful as pedicle screw instrumentation is in establishing stronger foundation of rods since Moe concept of instrumentation without fusion 32 years ago, we observed foundation-related problems, specifically pedicle screw fixation problems in surgical management of EOS. Rod breakage also occurred in both the 4.5 and 5.5 mm groups. Rod breakage and implant pullout are already well known to occur with the use of TGR.9 No rod types or foundation anchors appear to be immune to this problem. This would suggest that this is a problem with instrumentation spanning unfused spinal segments that likely will not be overcome with any type of implant. There is some belief that more frequent lengthenings that are possible with MCGR may allow the rods to more closely match spinal growth and possibly decrease stress across implants.16 Currently this study is too early in its experience to support or refute this concept.
Skaggs and colleagues did report on the concept of decreasing gains in thoracic height associated with progressive lengthenings. He defined this concept as the “law of diminishing returns.”17 Likely the etiology to this observation is the progressive stiffness that the spinal tissues undergo with both repeated violation of tissue from surgical procedures and the prolonged effects of a stiff implant. Some authors have reported on their findings during final fusion of ankylosis or autofusion of the spine between the anchor sites where growth and mobility were thought to be preserved.18 The lack of lengthening in the MCGR patients do not appear to follow the same principle as the “law of diminishing returns.” In 4 of the 6 patients, lengthening achieved in subsequent visits. This lack of lengthening may be related to residual tissue stiffness that may be diminished with subsequent spinal growth. It may also be technical and related to the improper placement of the controller above the MCGR actuator. However, a lack of lengthening can be concerning because until future lengthening is attained it is not known whether there is a problem with the MCGR actuator. It is our recommendation that a delayed lengthening attempt be made before any surgical exploration for MCGR failure is considered. As for the “law of diminishing returns” there is some thought that the increased frequency of lengthening that can be done with the MCGR may decrease the development of autofusion by more closely matching the spinal growth. Unfortunately, long-term studies with greater numbers will be needed to assess this idea.
There are limitations to this study. As this is an early experience with the MCGR, a longer follow-up is needed for a more complete report on a complication rate to be subsequently compared with those registered with TGR and reported in the literature. We believe the uniqueness of the device and its application in an otherwise challenging population require the early reporting of potential problems. A second limitation is the inability to standardize the application of the device between centers and patients. Some cases were primary, patients that never had any previous growing rods, whereas others were conversions that had the device implanted after an exchange from a TGR. In addition, some cases benefitted from first-generation MCGR implant, whereas others were implanted with second-generation devices. This may affect the complication profile of the device. As more patients are being treated with MCGR and the current patients have longer follow-up with first-generation device, future studies can highlight the effects of different techniques and patient types.
In conclusion, according to our experience, early to intermediate follow-up demonstrates a low infection rate (4%) with MCGR. However, MCGR does not appear to prevent common implant-related complications, such as rod or foundation failure. The lack of lengthening seen at some visits is unique to MCGR but does not always indicate a problem with the device. Ultimately, future studies with longer follow-up will be needed for a more complete report on the complication profile of this promising device (Figs. 1, 2).
1. Skaggs DL, Akbarnia BA, Flynn JM, et al. A classification of growth friendly spine implants. J Pediatr Orthop. 2014;34:260–274.
2. Akbarnia BA, Breakwell LM, Marks DS, et al. Dual growing rod technique followed for three to eleven years until final fusion: the effect of frequency of lengthening. Spine. 2008;33:984–990.
3. Campbell RM Jr, Adcox BM, Smith MD, et al. The effect of mid-thoracic VEPTR opening wedge thoracostomy on cervical tilt associated with congenital thoracic scoliosis in patients with thoracic insufficiency syndrome. Spine. 2007;32:2171–2177.
4. Horan TC, Gaynes RP, Martone WJ, et al. CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections. J Infect Control. 1992;20:271–274.
5. Ouellet J. Surgical technique: modern Luque trolley, a self-growing rod technique. Clin Orthop Relat Res. 2011;469:1356–1367.
6. Betz RR, D’Andrea LP, Mulcahey MJ, et al. Vertebral body stapling procedure for the treatment of scoliosis in the growing child. Clin Orthop Relat Res. 2005;434:55–60.
7. Akbarnia BA, Mundis GM Jr, Salari P, et al. Innovation in growing rod technique: a study of safety and efficacy of a magnetically controlled growing rod
in a porcine model. Spine. 2012;37:1109–1114.
8. Cheung KM, Cheung JP, Samartzis D, et al. Magnetically controlled growing rods for severe spinal curvature in young children: a prospective case series. Lancet. 2012;379:1967–1974.
9. Bess S, Akbarnia BA, Thompson GH, et al. Complications of growing-rod treatment for early-onset scoliosis: analysis of one hundred and forty patients. J Bone Joint Surg Am. 2010;92:2533–2543.
10. Kabirian N, Akbarnia BA, Pawelek JB, et al. Deep surgical site infection following 2344 growing-rod procedures for early-onset scoliosis: risk factors and clinical consequences. J Bone Joint Surg Am. 2014;96:e128.
11. Cahill PJ, Warnick DE, Lee MJ, et al. Infection after spinal fusion for pediatric spinal deformity: thirty years of experience at a single institution. Spine. 2010;35:1211–1217.
12. Smith JT, Johnston C, Skaggs D, et al. A new classification system to report complications in growing spine surgery: a multicenter consensus study. J Pediatr Orthop. 2015;35:798–803.
13. Akbarnia BA, Marks DS, Boachie-Adjei O, et al. Dual growing rod technique for the treatment of progressive early-onset scoliosis: a multicenter study. Spine. 2005;30(suppl):S46–S57.
14. Elsebai HB, Yazici M, Thompson GH, et al. Safety and efficacy of growing rod technique for pediatric congenital spinal deformities. J Pediatr Orthop. 2011;31:1–5.
15. Thompson GH, Akbarnia BA, Campbell RM Jr. Growing rod techniques in early-onset scoliosis. J Pediatr Orthop. 2007;27:354–361.
16. Agarwal AAA, Jayaswal A, Goel V. Smaller interval distractions may reduce chances of growth rod breakage without impending desired spinal growth: a finite element study. Spine Deform. 2014;2:430–436.
17. Sankar WN, Skaggs DL, Yazici M, et al. Lengthening of dual growing rods and the law of diminishing returns. Spine. 2011;36:806–809.
18. Flynn JM, Tomlinson LA, Pawelek J, et al. Growing-rod graduates: lessons learned from ninety-nine patients who completed lengthening. J Bone Joint Surg Am. 2013;95:1745–1750.