Outcomes of Primary and Conversion Magnetically Controlled Growth Rods Are Different at Two-Year Follow-up: Results of North American Release

Actuator Size of Magnetic Controlled Growth Rod (7 cm vs. 9 cm) is Not Predictive of Unplanned Return to the Operating Room: A Retrospective Multicenter Comparative Cohort Study

BACKGROUND

Magnetic controlled growth rods (MCGR) are the most common type of implant used for operative treatment of patients with early-onset scoliosis (EOS). Rods can have either a 7-cm actuator, allowing 2.8 cm of potential expansion, or a 9-cm actuator which allows 4.8 cm potential expansion. We hypothesized that the rate of unplanned return to the operating room (UPROR) will be increased when the 9-cm actuator is implanted in smaller patients. In addition, we aimed to identify a cutoff for spine length between planned upper and lower instrumented MCGR levels that best differentiated between patients having a high versus low risk of UPROR.

METHODS

We identified 167 patients from a prospectively collected registry of EOS patients who began MCGR treatment at 9 years of age or younger, with greater than 1 year of follow-up, and had adequate radiographs. Demographic, clinical, and surgical characteristics were analyzed for 7-cm and 9-cm actuator patients. Chi-square tests and Student t tests were used to test for differences between the 2 actuator rod groups. A predictive model for UPROR within 2 years was developed based on variables significantly predictive of UPROR.

RESULTS

The average follow-up was 2.6 years (range, 1 to 5 y) in both the 7 cm (n=74) and 9 cm (n=93) groups. Twenty-five complications in 14 patients led to UPROR within 2 years of MCGR insertion, 8% incidence (95% CI, 4%-13%). Device-related complications (n=15) were the most common reason for UPROR, followed by wound complications (n=4), pain-related complications (n=3), junctional kyphosis (n=2), and incarcerated umbilical hernia (n=1). After adjusting for age, spine height, number of spine anchors, sex, and diagnosis, there was no significant difference in UPROR rates between groups. Fewer proximal anchors, smaller T1-S1 height, and more caudal mid-point of primary coronal curvature were significantly associated with UPROR in the predictive model.

CONCLUSION

MCGR actuator size is not a significant factor in predicted UPROR. Smaller height, fewer anchors, and caudal apex increased UPROR risk.

LEVEL OF EVIDENCE

This is a retrospective, multicenter comparative cohort study (Level III therapeutic).

Lengthening Behavior of Magnetically Controlled Growing Rods in Early-Onset Scoliosis: A Multicenter Study

BACKGROUND

The "law of diminishing returns" is described for traditional growing rods. Magnetically controlled growing rods (MCGRs) have become a preferred implant for the surgical treatment of early-onset scoliosis (EOS). We examined a large cohort of patients with EOS to determine whether the law of diminishing returns applies to MCGRs.

METHODS

A prospectively collected, multicenter registry was queried for patients with EOS treated with MCGRs. Patients with only spine-based implants and a minimum of 2 years of follow-up were included; patients with congenital scoliosis, single rods, <3 lengthenings, or >25% missing data were excluded. Patients were analyzed in 3 cohorts: primary MCGR (pMCGR) had first-time MCGR implants, secondary MCGR (sMCGR) were converted from an MCGR to a new MCGR, and conversion MCGR (cMCGR) were converted from a non-MCGR implant to MCGR.

RESULTS

A total of 189 patients in the pMCGR group, 44 in the cMCGR group, and 41 in the sMCGR group were analyzed. From post-MCGR placement to the most recent follow-up or pre-definitive procedure, there were no differences in the changes in major Cobb angle, T1-S1 height, or T1-T12 height over time between the pMCGR and cMCGR groups. There was a decrease in length achieved at subsequent lengthenings in all cohorts (p < 0.01), and the sMCGR group had a significantly poorer ability to lengthen at each subsequent lengthening versus the pMCGR and cMCGR groups (p < 0.02). The 1-year survival rate was 90.5% for pMCGR, 84.1% for sMCGR, and 76.4% for cMCGR; 2-year survival was 61.5%, 54.4%, and 41.4%, respectively; and 3-year survival was 37.6%, 36.7%, and 26.9%, respectively. Excluding MCGRs still expanding, 27.6% of pMCGRs, 8.8% of sMCGRs, and 17.1% of cMCGRs reached the maximum excursion. Overall, 21.7% reached the maximum excursion. Within the pMCGR cohort, idiopathic and neuromuscular etiologies had a decline in lengthening achieved over time (p < 0.001), while syndromic EOS demonstrated a preserved ability to lengthen over time (p = 0.51). When the etiological groups were compared with each other, the neuromuscular group had the least ability to lengthen over time (p = 0.001 versus syndromic, p = 0.02 versus idiopathic).

CONCLUSIONS

The MCGR experiences the law of diminishing returns in patients with EOS. We found that only 21.7% of rods expanded to within 80% of the maximum excursion.

LEVEL OF EVIDENCE

Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Magnetically Controlled Growing Rods Graduation: Deformity Control with High Complication Rate

STUDY DESIGN

A multicenter retrospective review of consecutive series of patients.

OBJECTIVE

Long-term experience with using the magnetically controlled growing rods (MCGR) to treat patients with deformity in the growing spine to the conclusion of treatment with posterior spine fusion.

SUMMARY OF BACKGROUND DATA

MCGR treatment for growing spine gained popularity with paucity of long-term follow up data. We hypothesized that final fusion might be more effective in bringing additional correction of the spine deformity after treatment with MCGR than that reported after traditional growing rods (TGR) due to less scarring and auto-fusion.

METHODS

Retrospective review of 47 patients with varied etiology, treated between 2011 and 2017 which graduated treatment were followed in five academic medical centers for average of 50 months (range, 10-88).

RESULTS

The initial mean coronal deformity of 69.6° (95% CI 65-74) was corrected to 40° (95% CI 36-40) immediately after the MCGR implantation but progressed to 52.8° (95% CI 46-59) prior to the final surgery (P < 0.01). Nevertheless, thoracic spine growth (T1-T12 height) improved from 187.3 mm (95% CI 179-195) following index surgery to 208.9 mm (95% CI 199-218) prior to final fusion (P < 0.01). Significant correction and spinal length were obtained at final fusion, but metallosis was a frequent observation (47%, 22/47). The average growth rate was 0.5 mm/month (95% CI 0.3-0.6). The overall complication rate within our cohort was 66% (31/47) with 45% (21/47) of unplanned returns to the operating theater. 32% (15/47) of the patients had an implant related complication. Unplanned surgery was highly correlated with thoracic kyphosis greater than 40° (OR 5.42 95% CI 1.3-23).

CONCLUSION

Treatment of growing spine deformities with MCGR provides adequate control of spine deformity it is comparable to previously published data about TGR. The overall high complications rate over time and specifically implant related complications.Level of Evidence: 4.

Spinal Deformity in Sotos Syndrome: First Results of Growth-friendly Spine Surgery

BACKGROUND

Sotos syndrome (SS), or cerebral gigantism, describes children with macrocephaly, craniofacial abnormalities, general overgrowth, ligamentous laxity, developmental delay, and neurological disabilities. Fewer than 500 cases have been reported since Sotos and colleagues described the condition in 1964 and no literature exists on the management of spinal deformity in children under 10 years old.The aims of this study were: (1) to characterize the presentation of spinal deformities in patients with SS; and (2) to provide preliminary results of growth-friendly instrumentation (GFI) in these children.

METHODS

Thirteen children (9 boys) with SS and minimum of 2-year follow-up were identified from 2 multicenter early-onset scoliosis (EOS) databases (1997-2017). Mean age at index surgery and follow-up duration were 5.0 years (range, 1.8 to 10 y) and 7.2 years (range, 2.1 to 14.9 y), respectively. Patients underwent GFI for a mean of 5.7 years (range, 2 to 10.2 y), with an average of 9 lengthenings (range, 2 to 18). Definitive spinal fusion was performed in 4 patients (31%). Major curve magnitude, T1-T12 and T1-S1 lengths, thoracic kyphosis, and lumbar lordosis were evaluated preindex, postindex, latest GFI, and postfusion, when possible.

RESULTS

Five thoracolumbar (38%), 4 double major (31%), 2 main thoracic (15%), and 2 double thoracic curves (15%) were seen that spanned a mean of 6.8 levels (5 to 9). Major curves improved 36% (range, 5% to 71%), from a mean of 71 degrees (range, 48 to 90 degrees) to 46 degrees (range, 20 to 73 degrees) postindex surgery (P<0.001). Major curves remained stable at a mean of 52 degrees (range, 20 to 87 degrees) at latest GFI (P=0.36). True T1-T12 and T1-S1 growth velocities during GFI were 0.5 mm/mo (range, 0.4 to 0.8 mm/mo) and 0.8 mm/mo (range, 0.1 to 2.1 mm/mo), respectively. Twenty-six complications occurred in 9 patients (69%) averaging 2 complications per patient (range, 0 to 7).

CONCLUSIONS

This is the first study to evaluate the outcomes of GFI in children with SS and EOS. Compared with published data for outcomes of GFI in EOS, children with SS may have less major curve correction. Growth-friendly surgery remains an effective treatment method for EOS in patients with SS.

LEVELS OF EVIDENCE

Level IV-retrospective case-series.