7% of adolescent idiopathic scoliosis (AIS) patients also present with a pars defect. To date, there are no available data on the results of fusion ending proximal to a spondylolysis in the setting of AIS. The aim of this study was to analyze the outcomes of posterior spinal fusion (PSF) in this patient cohort, to investigate if maintaining the lytic segment unfused represents a safe option.

Retrospective review of all patients who received PSF for AIS, presented with a spondylolysis or spondylolisthesis and had a min. 2-years follow-up. Demographic data, instrumented levels and preoperative radiographic data were collected. Mechanical complications, coronal or sagittal parameters, amount of slippage and pain levels were evaluated.

Data from 22 patients were available (age 14.4 ± 2.5 years), 18 Lenke 1–2 and four Lenke 3–6. Five patients (24%) had an isthmic spondylolisthesis, all Meyerding I. The mean preoperative Cobb angle of the instrumented curves was 58 ± 13°. For 18 patients the lowest instrumented vertebra (LIV) was the last touched vertebra (LTV); for two LIV was distal to the LTV; for two, LIV was one level proximal to the LTV. The number of segments between the LIV and the lytic vertebra ranged from 1 to 6. At the last follow-up, no complications were observed. The residual curve below the instrumentation measured 8.5 ± 6.4°, the lordosis below the instrumented levels was 51.4 ± 13°. The magnitude of the isthmic spondylolisthesis remained constant for all included patients. Three patients reported minimal occasional low back pain.

The LTV can be safely used as LIV when performing PSF for the management of AIS in patients with L5 spondylolysis

The purpose of this study is to evaluate risk factors for distal construct failure (DCF) in posterior spinal instrumented fusion (PSIF) in adolescent idiopathic scoliosis (AIS). We observed an increased rate of DCF when the pedicle screw in the lowest instrumented vertebra (LIV) was not parallel to the superior endplate of the LIV, however this has not been well studied in the literature. We hypothesise a more inferiorly angled LIV screw predisposes to failure and aim to find the critical angle that predisposes to failure.

A retrospective cohort study was performed on all patients who underwent PSIF for AIS at the Starship Hospital spine unit from 2010 to 2020. On a lateral radiograph, the angle between the superior endplate of the LIV was measured against its pedicle screw trajectory. Data on demographics, Cobb angle, Lenke classification, instrumentation density, rod protrusion from the most inferior screw, implants and reasons for revision were collected.

Of 256 patients, 10.9% (28) required at least one revision. The rate of DCF was 4.6% of all cases (12 of 260) and 25.7% of revisions were due to DCF. The mean trajectory angle of DCF patients compared to all others was 13.3° (95%CI 9.2° to 17.4°) vs 7.6° (7° to 8.2°), p=0.0002. The critical angle established is 11°, p=0.0076. Lenke 5 and C curves, lower preoperative Cobb angle, titanium only rod constructs and one surgeon had higher failure rates than their counterparts. 9.6% of rods protruding less than 3mm from its distal screw disengaged.

We conclude excessive inferior trajectory of the LIV screw increases the rate of DCF and a screw trajectory greater than 11° predisposes to failure. This is one factor that can be controlled by the surgeon intraoperatively and by avoiding malposition of the LIV screw, a quarter of revisions can potentially be eliminated.