One Year Results of the Randomized BiPOWR Trial Comparing the Spring Distraction System (SDS) and the One Way Self-Expanding Rod (OWSER) for the Correction of Neuromuscular and Syndromic Early Onset Scoliosis

Abstract

Background: Current "growth-friendly" implants for treatment of Early Onset Scoliosis (EOS) have limitations that reduce their efficacy and cost-effectiveness. Recently, two systems have been developed that mitigate many of these limitations, the Spring Distraction System (SDS) and the One Way Self-Expanding Rod (OWSER). The purpose of the multicenter BiPOWR trial was to compare 1-year efficacy and -safety of both strategies in the treatment of neuromuscular or syndromic EOS.

Methods: Non-ambulant, neuromuscular/syndromic EOS patients were included in three academic hospitals. They were randomized to treatment with SDS or OWSER and were blinded until after surgery. Outcomes were coronal curve, spinal growth and the occurrence of (serious) adverse events ((S)AEs). In addition, spinal growth and implant lengthening were calculated. Data were collected pre-operatively, immediately post-operatively, and at 1-, 3-, 6-, and 12-month follow-up.

Results: Thirty patients were included. Two patients passed away during follow-up, and these patients were replaced. All collected data were used for analysis. Mean age at surgery was 9.0 years, and 20/30 patients were male. Mean coronal curve decreased from 74.9° pre-operatively, to 37.6° post-operatively, remaining stable at 37.7° at the 1-year follow-up, with no group differences. T1-T12 length increased by 18 mm/year for SDS and 9 mm/year for OWSER. For T1-S1 length, this was 26 mm/year (SDS) and 18 mm/year (OWSER). Five (S)AEs occurred in the SDS group and 11 (S)AEs in the OWSER group. Two SDS patients passed away, unrelated to the surgery or implant. One (S)AE in the SDS group and 6 (S)AEs in the OWSER group were implant-related.

Conclusions: The SDS and the OWSER achieved coronal curve correction of 50%, which was maintained at 1-year follow-up. Spinal length increase was excellent for both systems. The (S)AE rate was 30%/patient/year for SDS and 78%/patient/year for OWSER.

Key concepts: (1) The current study is the first RCT that compares two "growth-friendly" implants in a neuromuscular early onset scoliosis (EOS) population.(2) The Spring Distraction System (SDS) and the One Way Self-Expanding Rod (OWSER) both achieve around 50% curve correction which is maintained at 1 year follow-up.(3) Both systems achieve excellent T1-T12- and T1-S1 height increase, without the need for repetitive lengthenings.(4) The (S)AE rate of SDS was 30%/patient/year. For OWSER, the (S)AE rate was 78%/patient/year.

Study design: RCT.

Level of evidence: Level 1.

Keywords: BiPOWR; Early onset scoliosis; Neuromuscular; One way self-expanding rod; Spring distraction system.

Figure 1

Spring Distraction System. The SDS consists of three components that are added to standard growing rods. It provides a continuous distraction force during follow-up, without the need for repeated lengthenings. In the BiPOWR trial, the distal anchors are iliosacral screws. Green. A side-to-side connector with one oversized hole through which a CoCr rod can slide freely. Gold. Ti6Al4V springs which can be compressed over the rod. Blue. The buttress compresses the spring against the side-to-side connector.

Figure 2

One Way Self-expanding Rod. The OWSER is a growing rod that passively lengthens one way as the spine grows. In the BiPOWR trial, the distal anchors are iliosacral screws instead of pedicle screws. Blue. the proximal fixation consists of hooks positioned in a claw configuration. Two crosslinks are added for torsional stability (green). Red. The growing domino, combined with a rod that is serrated across its distal length, allows for lengthening. The reserve length can be 50 mm or 80 mm long. Movement in the other direction is prevented by a split retaining ring system inside the domino.

Figure 3

CONSORT patient flow diagram.

Figure 4

Coronal- and sagittal curve changes over time. Mean or median and 95% confidence interval of each timepoint are plotted for each group.

Figure 5

Spinal height/length changes over time. Mean and 95% confidence interval of each timepoint are plotted for each group.

Figure 6

SDS complication. Example of an SDS patient that showed almost complete implant expansion within 1 year. Note the initial lengthening between insertion and the first radiograph of 20 mm. As there was still at least 3 years of expected spinal growth, the choice was made to perform a re-operation in which the springs were re-tensioned and the distal anchor rod construct was lengthened. The colored segment of the sliding rod denotes how much implant growth is left.

Figure 7

OWSER complication. In this patient, the set screws that connect the OWSER domino to the anchor rod failed, causing the domino to erroneously move up on the serrated rod, without achieving any spinal lengthening or additional correction. In this patient, the OWSER rod was replaced during a re-operation.

Figure 8

Kaplan-Meier curves for (S)AE-free survival. Survival curves for both groups with 95% confidence intervals. Ticks denote censored patients.

figs1