Intermediate-term clinical and radiographic outcomes with less invasive adult spinal deformity surgery: patients with a minimum follow-up of 4 years

Long-term Outcomes After Adult Spinal Deformity Surgery Using Lateral Interbody Fusion: Short Versus Long Fusion

STUDY DESIGN

Retrospective cohort study.

OBJECTIVE

To investigate long-term outcomes after short or long fusion for adult spinal deformity using lateral interbody fusion.

SUMMARY OF BACKGROUND DATA

Lateral interbody fusion is commonly used in adult spinal deformity surgery. Favorable short-term outcomes have been reported, but not long-term outcomes. Lateral interbody fusion with strong ability to correct deformity may allow the selection of short fusion techniques.

MATERIALS AND METHODS

We retrospectively reviewed adults who underwent this surgery with a minimum of 5 years of follow-up. Short fusion with the uppermost instrumented vertebra in the lumbar spine was performed in patients without degenerative changes at the thoracolumbar junction (S-group); others underwent long fusion with the uppermost instrumented vertebra in the thoracic spine (L-group). We assessed radiographic and clinical outcomes.

RESULTS

Short fusion was performed in 29 of 54 patients. One patient per group required revision surgery. Of the remainder, with similar preoperative characteristics and deformity correction between groups, correction loss (pelvic incidence-lumbar lordosis, P=0.003; pelvic tilt, P=0.005; sagittal vertical axis, P˂0.001) occurred within 2 years postoperatively in the S-group, and sagittal vertical axis continued to increase until the 5-year follow-up (P=0.021). Although there was a significant change in Oswestry disability index in the S-group (P=0.031) and self-image of Scoliosis Research Society 22r score in both groups (P=0.045 and 0.02) from 2- to 5-year follow-up, minimum clinically important differences were not reached. At 5-year follow-up, there was a significant difference in Oswestry Disability Index (P=0.013) and Scoliosis Research Society 22r scores (function: P=0.028; pain: P=0.003; subtotal: P=0.006) between the groups, but satisfaction scores were comparable and Oswestry Disability Index score (29.8%) in the S-group indicated moderate disability.

CONCLUSIONS

Health-related quality of life was maintained between 2- and 5-year follow-up in both groups. Short fusion may be an option for patients without degenerative changes at the thoracolumbar junction.

LEVEL OF EVIDENCE

III.

Defining the Post-Operative Progression of Degenerative Scoliosis: An Analysis of Cases without Instrument Failure

OBJECTIVE

The expected post-operative changes in radiographic alignment over time remain poorly defined in patients surgically treated for degenerative scoliosis without instrument failure. Here we aim to describe the optimal natural progression of radiographic degenerative scoliosis at multiple timepoints in patients treated with a transforaminal lumbar interbody fusion (TLIF).

METHODS

We identified an initial retrospective cohort of 114 patients treated with a TLIF for degenerative scoliosis between 2018 and 2022, with 39 patients ultimately meeting the imaging inclusion criteria. Patients who completed a primary or revision procedure with no evidence of instrument failure, proximal junctional kyphosis, or proximal junctional failure at last follow-up were included. Radiographic measurements of spinopelvic alignment were manually extracted from X-Ray scoliosis films.

RESULTS

Thirty-nine patients (mean age 62.6 ± 8.7, mean follow-up 2.9 years), of which 23 underwent a primary TLIF (Primary) and 16 a revision procedure (Revision), were analyzed. Patients in the Primary group experienced a durable improvement in Thoracolumbar Cobb angle (-25° ± 15°), Thoracic Kyphosis (10° ± 13°), and Pelvic Incidence/lumbar lordosis mismatch (PI/LL) (-19° ± 19°) through the first year of follow-up. In the Revision group, at one year follow-up, all measures of spinopelvic alignment except PI/LL mismatch had reverted to pre-operative levels. Thoracolumbar Cobb angle decreased to a significantly greater degree in the Primary group compared to the Revision group.

CONCLUSION

Primary TLIF operations without instrument failure consistently improve radiographic outcomes in three key measures through the first year. For revision procedures, there appears to be modest radiographic benefit at follow-up.

Risk Factors of Postoperative Coronal Balance Transition in Degenerative Lumbar Scoliosis

STUDY DESIGN

Retrospective radiographic study.

OBJECTIVE

To determine the potential risk factors influencing the transition of postoperative coronal balance in degenerative lumbar scoliosis (DLS) patients.

SUMMARY OF BACKGROUND DATA

As time passes after surgery, the spinal sequence of DLS patients may dynamically shift from coronal balance to imbalance, causing clinical symptoms. However, the transition of postoperative coronal balance and its risk factors have not been effectively investigated.

MATERIALS AND METHODS

We included 156 DLS patients. The cohort was divided into immediate postoperative coronal balance with follow-up balance (N=73) and follow-up imbalance (N=21), immediate postoperative coronal imbalance (CIB) with follow-up balance (N=23), and follow-up imbalance (N=39). Parameters included age, sex, classification of coronal balance, coronal balance distance, fusion of L5 or S1, location of apical vertebra, apical vertebral translation (AVT), Cobb angle of the main curve and lumbar-sacral curve, tilt and direction of L4/5, tilt and direction of upper instrumented vertebra (UIV), and Cobb angle of T1-UIV. Statistical testing was performed using chi-square/Fisher exact test, t tests or nonparametric tests, correlation testing, and stepwise logistic regression.

RESULTS

We identified a significant difference in preoperative AVT, preoperative Cobb angle, and immediate postoperative UIV tilt between patients with and without follow-up balance. Logistic regression analysis demonstrated factors associated with follow-up CIB included preoperative AVT ( P =0.015), preoperative Cobb angle ( P =0.002), and tilt of immediate postoperative UIV ( P =0.018). Factors associated with immediate postoperative CIB in patients with follow-up coronal balance were sex, correction ratio of the main curve, and direction of L4. Logistic regression analysis further identified a correction ratio of main curve ≤0.7 ( P =0.009) as an important predictive factor.

CONCLUSION

Patients with immediate postoperative coronal balance and higher preoperative AVT, preoperative Cobb angle, and tilt of immediate postoperative UIV were more likely to experience follow-up CIB. A correction ratio of the main curve ≤0.7 was an independent predictor of follow-up CIB.

LEVEL OF EVIDENCE

3.

Patient outcomes after circumferential minimally invasive surgery compared with those of open correction for adult spinal deformity: initial analysis of prospectively collected data

OBJECTIVE

Circumferential minimally invasive spine surgery (cMIS) for adult scoliosis has become more advanced and powerful, but direct comparison with traditional open correction using prospectively collected data is limited. The authors performed a retrospective review of prospectively collected, multicenter adult spinal deformity data. The authors directly compared cMIS for adult scoliosis with open correction in propensity-matched cohorts using health-related quality-of-life (HRQOL) measures and surgical parameters.

METHODS

Data from a prospective, multicenter adult spinal deformity database were retrospectively reviewed. Inclusion criteria were age > 18 years, minimum 1-year follow-up, and one of the following characteristics: pelvic tilt (PT) > 25°, pelvic incidence minus lumbar lordosis (PI-LL) > 10°, Cobb angle > 20°, or sagittal vertical axis (SVA) > 5 cm. Patients were categorized as undergoing cMIS (percutaneous screws with minimally invasive anterior interbody fusion) or open correction (traditional open deformity correction). Propensity matching was used to create two equal groups and to control for age, BMI, preoperative PI-LL, pelvic incidence (PI), T1 pelvic angle (T1PA), SVA, PT, and number of posterior levels fused.

RESULTS

A total of 154 patients (77 underwent open procedures and 77 underwent cMIS) were included after matching for age, BMI, PI-LL (mean 15° vs 17°, respectively), PI (54° vs 54°), T1PA (21° vs 22°), and mean number of levels fused (6.3 vs 6). Patients who underwent three-column osteotomy were excluded. Follow-up was 1 year for all patients. Postoperative Oswestry Disability Index (ODI) (p = 0.50), Scoliosis Research Society-total (p = 0.45), and EQ-5D (p = 0.33) scores were not different between cMIS and open patients. Maximum Cobb angles were similar for open and cMIS patients at baseline (25.9° vs 26.3°, p = 0.85) and at 1 year postoperation (15.0° vs 17.5°, p = 0.17). In total, 58.3% of open patients and 64.4% of cMIS patients (p = 0.31) reached the minimal clinically important difference (MCID) in ODI at 1 year. At 1 year, no differences were observed in terms of PI-LL (p = 0.71), SVA (p = 0.46), PT (p = 0.9), or Cobb angle (p = 0.20). Open patients had greater estimated blood loss compared with cMIS patients (1.36 L vs 0.524 L, p < 0.05) and fewer levels of interbody fusion (1.87 vs 3.46, p < 0.05), but shorter operative times (356 minutes vs 452 minutes, p = 0.003). Revision surgery rates between the two cohorts were similar (p = 0.97).

CONCLUSIONS

When cMIS was compared with open adult scoliosis correction with propensity matching, HRQOL improvement, spinopelvic parameters, revision surgery rates, and proportions of patients who reached MCID were similar between cohorts. However, well-selected cMIS patients had less blood loss, comparable results, and longer operative times in comparison with open patients.

Two- and three-year outcomes of minimally invasive and hybrid correction of adult spinal deformity

OBJECTIVE

Previous studies have demonstrated the short-term radiographic and clinical benefits of circumferential minimally invasive surgery (cMIS) and hybrid (i.e., minimally invasive anterior or lateral interbody fusion with an open posterior approach) techniques to correct adult spinal deformity (ASD). However, it is not known if these benefits are maintained over longer periods of time. This study evaluated the 2- and 3-year outcomes of cMIS and hybrid correction of ASD.

METHODS

A multicenter database was retrospectively reviewed for patients undergoing cMIS or hybrid surgery for ASD. Patients were ≥ 18 years of age and had one of the following: maximum coronal Cobb angle (CC) ≥ 20°, sagittal vertical axis (SVA) > 5 cm, pelvic incidence-lumbar lordosis mismatch (PI-LL) ≥ 10°, or pelvic tilt (PT) > 20°. Radiographic parameters were evaluated at the latest follow-up. Clinical outcomes were compared at 2- and 3-year time points and adjusted for age, preoperative CC, levels operated, levels with interbody fusion, presence of L5-S1 anterior lumbar interbody fusion, and upper and lower instrumented vertebral level.

RESULTS

Overall, 197 (108 cMIS, 89 hybrid) patients were included with 187 (99 cMIS, 88 hybrid) and 111 (60 cMIS, 51 hybrid) patients evaluated at 2 and 3 years, respectively. The mean (± SD) follow-up duration for cMIS (39.0 ± 13.3 months, range 22-74 months) and hybrid correction (39.9 ± 16.8 months, range 22-94 months) were similar for both cohorts. Hybrid procedures corrected the CC greater than the cMIS technique (adjusted p = 0.022). There were no significant differences in postoperative SVA, PI-LL, PT, and sacral slope (SS). At 2 years, cMIS had lower Oswestry Disability Index (ODI) scores (adjusted p < 0.001), greater ODI change as a percentage of baseline (adjusted p = 0.006), less visual analog scale (VAS) back pain (adjusted p = 0.006), and greater VAS back pain change as a percentage of baseline (adjusted p = 0.001) compared to hybrid techniques. These differences were no longer significant at 3 years. At 3 years, but not 2 years, VAS leg pain was lower for cMIS compared to hybrid techniques (adjusted p = 0.032). Those undergoing cMIS had fewer overall complications compared to hybrid techniques (adjusted p = 0.006), but a higher odds of pseudarthrosis (adjusted p = 0.039).

CONCLUSIONS

In this review of a multicenter database for patients undergoing cMIS and hybrid surgery for ASD, hybrid procedures were associated with a greater CC improvement compared to cMIS techniques. cMIS was associated with superior ODI and back pain at 2 years, but this difference was no longer evident at 3 years. However, cMIS was associated with superior leg pain at 3 years. There were fewer complications following cMIS, with the exception of pseudarthrosis.