Both traditional open and minimally invasive surgery (MIS) techniques may be implemented for the treatment of adult spinal deformity (ASD).1–4 Both techniques may satisfactorily achieve the goals of treating sagittal and coronal plane imbalance, instability, and neural compression.1,5,6 Because traditional open surgery for ASD has been associated with substantial morbidity,7–10 a number of surgeons have increasingly applied MIS techniques in ASD treatment,11–13 demonstrating both clinical and radiographic improvement with long-term follow-up.3,14,15
Less invasive techniques may employ either circumferential MIS (cMIS), which uses percutaneous posterior fixation, or hybrid techniques, which use MIS lateral or anterior interbody fusions with open posterior surgery. In studies of comparative effectiveness, both cMIS and hybrid techniques were associated with clinical improvement. However, in a comparison of the latest follow-up outcomes (mean 31.3–38.3 months), hybrid techniques were associated with a greater improvement in radiographic parameters at the cost of a higher complication rate.3 It remains unclear, however, if the clinical benefits achieved through MIS deformity correction are maintained over longer periods. The durability of this benefit is of interest when considering the possible diminishment of clinical improvement over time as observed in other studies evaluating surgical correction of ASD.15
Previous reports evaluating longitudinal outcomes following open ASD surgery have observed variability in the durability of clinical benefit over time.16,17 Furthermore, the reasons for reoperation vary based on time of follow-up.18,19 This variability underscores the importance of a longitudinal outcome evaluation following MIS ASD surgery as complications accumulate. Therefore, we utilized a multicenter database to compare the 2- and 3-year outcomes for those undergoing cMIS versus hybrid surgery for ASD.
Methods
Study Population
We retrospectively reviewed the multicenter prospectively maintained database of the Minimally Invasive Surgery International Spine Study Group (MIS-ISSG) that contains information on patients undergoing MIS for ASD. Data were collected on an annual basis from 2008 to 2017 and audited by individual sites. Enrollment was nonrandomized and nonconsecutive. All patients had pre- and postoperative anteroposterior (AP) and lateral 3-foot long cassette radiographs. Inclusion criteria included patients undergoing cMIS (Fig. 1) or hybrid surgery (Fig. 2), age ≥ 18 years, and at least one of the following: coronal Cobb angle (CC) ≥ 20°, sagittal vertical axis (SVA) > 5 cm, pelvic incidence–lumbar lordosis mismatch (PI-LL) ≥ 10°, or pelvic tilt (PT) > 20°. Surgeries were considered cMIS if they utilized MIS lateral or anterior interbody fusion and percutaneous posterior segmental pedicle screw instrumentation, and hybrid if they utilized MIS lateral or anterior interbody fusion and open posterior instrumentation, with or without posterior column osteotomies or releases. The operative technique used was at the discretion of the surgeon. We excluded patients undergoing surgery via open techniques alone. Participating centers (n = 9) obtained IRB approval.
Pre- (left) and postoperative (right) AP and lateral scoliosis radiographs in a patient undergoing ASD surgery via a cMIS approach.
Pre- (left) and postoperative (right) AP and lateral scoliosis radiographs in a patient undergoing ASD surgery via a hybrid approach.
Outcomes
We compared outcomes at 2 and 3 years using validated patient-reported outcome measures (PROMs): the Oswestry Disability Index (ODI), visual analog scale (VAS) for back pain, and VAS for leg pain. We assessed radiographic parameters at baseline and at the latest follow-up. Standing sagittal and coronal radiographic parameters were assessed and included CC, SVA, PI-LL, PT, and sacral slope (SS). Fusion was determined on plain radiographs as described previously.20 Fusion grades A or B, either anteriorly or posteriorly, were classified as fused. Other outcomes included complications, which were captured at 2- and 3-year time points. Complications have been defined previously.21 Infectious complications included pneumonia, urinary tract infection, sepsis, and superficial/deep surgical site infections. Radiographic complications included proximal/distal junctional kyphosis (DJK), pseudarthrosis, adjacent-segment disease, postoperative sagittal imbalance, postoperative coronal imbalance, and radiographic complications requiring reoperation.
Statistical Analysis
For univariate analyses, paired t-tests, Mann-Whitney U-tests, Fisher’s exact tests, and chi-square analyses were utilized. For adjusted analyses, ANOVAs and logistic regression models were used and results were adjusted for factors reaching p < 0.05 on univariate comparisons. All p values were two-tailed and an α < 0.05 was considered statistically significant.
Results
Overall, 197 patients (108 cMIS, 89 hybrid) were included (Table 1). Mean follow-up was similar between cohorts: cMIS 39.0 ± 13.3 months, range 22–74 months, versus hybrid 39.9 ± 16.8 months, range 22–94 months (p = 0.521). The percentage available for follow-up was similar at 3 years (cMIS 55.6% vs hybrid 57.3%, p = 0.805). At 2 years, both groups achieved greater than 90% follow-up, though the percentage was somewhat higher for the hybrid cohort (98.9% vs 91.7%, p = 0.049).
Baseline characteristics, perioperative data, and radiographic and clinical outcomes
| Variable | Hybrid (n = 89) | cMIS (n = 108) | p Value | |
|---|---|---|---|---|
| Unadjusted | Adjusted* | |||
| Baseline characteristics | ||||
| Mean age ± SD, yrs | 59.2 ± 12.8 | 63.3 ± 10.6 | 0.033 | — |
| Mean BMI ± SD, kg/m2 | 27.6 ± 5.3 | 27.5 ± 5.5 | 0.595 | — |
| Mean length of FU ± SD, mos | 39.9 ± 16.8 | 39.0 ± 13.3 | 0.521 | — |
| Median length of FU, mos | 36.0 | 36.0 | — | — |
| Mean ODI ± SD | 54.7 ± 16.6 | 49.7 ± 18.4 | 0.071 | — |
| Mean VAS back pain score ± SD | 6.9 ± 2.2 | 6.8 ± 2.1 | 0.544 | — |
| Mean VAS leg pain score ± SD | 5.5 ± 3.0 | 5.7 ± 2.8 | 0.813 | — |
| Mean CC ± SD, ° | 37.5 ± 18.6 | 31.5 ± 15.1 | 0.039 | — |
| Mean SVA ± SD, mm | 51.6 ± 62.3 | 39.9 ± 53.0 | 0.408 | — |
| Mean PI-LL ± SD, ° | 17.9 ± 17.6 | 14.6 ± 14.5 | 0.310 | — |
| Mean PT ± SD, ° | 23.1 ± 10.5 | 23.5 ± 10.5 | 0.944 | — |
| Mean SS ± SD, ° | 31.2 ± 13.0 | 29.7 ± 9.5 | 0.187 | — |
| Surgical characteristics | ||||
| Staged, n (%) | 46 (51.7) | 55 (50.9) | 0.515 | — |
| Mean levels ± SD | 8.1 ± 4.6 | 4.8 ± 2.7 | <0.001 | — |
| Mean levels w/ interbody fusion ± SD | 3.0 ± 1.4 | 3.4 ± 1.5 | 0.037 | — |
| L5–S1 ALIF, n (%) | 18 (20.2) | 10 (9.3) | 0.023 | — |
| UIV | <0.001 | — | ||
| T1–6 | 19 (21.6) | 4 (3.7) | ||
| T7–12 | 45 (51.1) | 32 (29.9) | ||
| Below T12 | 24 (27.3) | 71 (66.4) | ||
| LIV | <0.001 | — | ||
| Above S1 | 25 (28.4) | 55 (51.4) | ||
| S1 | 21 (23.9) | 39 (36.4) | ||
| Ilium | 42 (47.7) | 13 (12.1) | ||
| Use of bmp, n (%) | 64 (71.9) | 71 (68.9) | 0.248 | — |
| Operative and fusion data | ||||
| Mean EBL ± SD, ml | 1512.7 ± 1467.4 | 464.3 ± 508.5 | <0.001 | — |
| Mean operative time ± SD, mins | 621.8 ± 260.8 | 426.3 ± 181.3 | <0.001 | — |
| Mean radiographic outcomes ± SD | ||||
| CC, ° | ||||
| Baseline | 37.5 ± 18.6 | 31.5 ± 15.1 | 0.039 | — |
| Postop | 16.5 ± 11.0 | 17.2 ± 11.1 | 0.703 | 0.017 |
| Change | −20.9 ± 13.6 | −14.5 ± 12.6 | 0.003 | 0.022 |
| SVA, mm | ||||
| Baseline | 51.6 ± 62.3 | 39.9 ± 53.0 | 0.408 | — |
| Postop | 48.9 ± 56.6 | 37.7 ± 56.6 | 0.150 | 0.152 |
| Change | −5.9 ± 60.0 | 2.8 ± 48.7 | 0.530 | 0.884 |
| PI-LL, ° | ||||
| Baseline | 17.9 ± 17.6 | 14.6 ± 14.5 | 0.310 | — |
| Postop | 11.2 ± 16.5 | 10.4 ± 13.5 | 0.691 | 0.472 |
| Change | −6.8 ± 16.8 | −3.8 ± 14.7 | 0.354 | 0.467 |
| PT, ° | ||||
| Baseline | 23.1 ± 10.5 | 23.5 ± 10.5 | 0.944 | — |
| Postop | 23.4 ± 10.4 | 23.9 ± 10.2 | 0.749 | 0.812 |
| Change | −0.1 ± 7.9 | 0.6 ± 7.1 | 0.926 | 0.381 |
| SS, ° | ||||
| Baseline | 31.2 ± 13.0 | 29.7 ± 9.5 | 0.187 | — |
| Postop | 31.7 ± 11.5 | 28.5 ± 9.8 | 0.043 | 0.107 |
| Change | 1.1 ± 8.6 | −0.5 ± 6.9 | 0.214 | 0.163 |
| Clinical outcomes | ||||
| Mean ODI ± SD | ||||
| 24 mos | 35.1 ± 17.3 | 28.2 ± 21.1 | 0.011 | <0.001 |
| 36 mos | 33.8 ± 17.0 | 33.2 ± 21.3 | 0.931 | 0.501 |
| 24-mo change | −19.3 ± 18.0 | −21.1 ± 21.1 | 0.497 | 0.146 |
| 24-mo percentage change | −33% ± 33% | −42% ± 43% | 0.051 | 0.006 |
| 36-mo change | −18.5 ± 17.4 | −19.7 ± 23.3 | 0.656 | 0.805 |
| 36-mo percentage change | −34% ± 30% | −33% ± 48% | 0.669 | 0.838 |
| Mean VAS back pain score ± SD | ||||
| 24 mos | 4.0 ± 2.7 | 3.2 ± 2.7 | 0.022 | 0.006 |
| 36 mos | 4.2 ± 2.4 | 3.5 ± 2.7 | 0.115 | 0.106 |
| 24-mo change | −2.8 ± 2.9 | −3.6 ± 2.8 | 0.091 | 0.073 |
| 24-mo percentage change | −37% ± 45% | −52% ± 44% | 0.019 | 0.001 |
| 36-mo change | −3.1 ± 2.5 | −3.4 ± 2.6 | 0.326 | 0.335 |
| 36-mo percentage change | −41% ± 34% | −50% ± 42% | 0.121 | 0.110 |
| Mean VAS leg pain ± SD | ||||
| 24 mos | 2.9 ± 2.7 | 2.9 ± 2.8 | 0.828 | 0.398 |
| 36 mos | 2.9 ± 3.0 | 1.9 ± 2.5 | 0.114 | 0.032 |
| 24-mo change | −2.7 ± 3.6 | −2.7 ± 3.4 | 0.938 | 0.891 |
| 24-mo percentage change | −19% ± 131% | −31% ± 109% | 0.749 | 0.254 |
| 36-mo change | −3.0 ± 3.4 | −4.6 ± 2.5 | 0.007 | 0.074 |
| 36-mo percentage change | −42% ± 60% | −71% ± 55% | 0.003 | 0.097 |
bmp = bone morphogenetic protein; EBL = estimated blood loss. Percentages reported are considered from available information. Frequencies may not add up to the total cohort size where there are missing data. Boldface type indicates statistical significance.
Adjusted for age, preoperative CC, levels of surgery, levels of interbody fusion, presence of L5–S1 ALIF, UIV, and LIV.
Radiographic Outcomes
Table 1 reports the radiographic outcomes for the cohorts. For the hybrid cohort, significant improvements were observed for maximum CC (p < 0.001) and PI-LL (p = 0.001) at the latest follow-up compared to baseline. There were no significant changes in SVA, PT, and SS when comparing follow-up to baseline for the hybrid cohort (p > 0.05). For the cMIS cohort, significant improvements were observed for maximum CC (p < 0.001) and PI-LL (p = 0.018) at the latest follow-up compared to baseline. There were also no significant changes in SVA, PT, and SS when comparing follow-up to baseline for the cMIS cohort (p > 0.05).
On univariate comparisons, the hybrid cohort was associated with a greater change in CC (−20.9° vs −14.5°, p = 0.003) and higher postoperative SS (31.7° vs 28.5°, p = 0.043) compared to the cMIS cohort. After adjustment for age, preoperative CC, number of levels operated, number of levels with an interbody fusion, presence of L5–S1 anterior lumbar interbody fusion (ALIF), level of upper instrumented vertebrae (UIV), and level of lower instrumented vertebrae (LIV), the hybrid cohort had a lower postoperative CC (16.5° vs 17.2°, adjusted p = 0.017) and greater change in CC when compared to baseline (−20.9° vs −14.5°, adjusted p = 0.022).
Clinical Outcomes
Figure 3 depicts the mean ODI, VAS back pain, and VAS leg pain scores at baseline and at 2- and 3-year time points. Both procedures were associated with significant improvements at 2- and 3-year time points for the ODI and for VAS back and leg pain compared to baseline (p < 0.001 for all comparisons). For the cMIS cohort, there was further improvement in leg pain at 3 years compared to year 2 (mean VAS leg pain 2.9 to 1.9, p = 0.030). In contrast, there was no significant change in mean ODI between the 2- and 3-year time points (mean ODI 28.2 to 33.2, p = 0.059) for the cMIS cohort. Other 3-year results were not significantly different from 2-year results for the cMIS group with regard to VAS back pain, and the hybrid group with regard to the ODI and VAS back and leg pain.
Comparisons of mean ODI (A), VAS back pain (B), and VAS leg pain (C) at baseline (BL) and 2 (Y2) and 3 (Y3) years following surgery. *p < 0.05, **p < 0.001.
Table 1 compares the 2- and 3-year outcomes between the cohorts. At 2 years, cMIS was associated with superior mean ODI (adjusted p < 0.001) and ODI change as a percentage of baseline (adjusted p = 0.006). Furthermore, cMIS was associated with superior mean back pain scores (adjusted p = 0.006) and back pain change as a percentage of baseline (adjusted p = 0.001). At 3 years, these differences were no longer apparent (adjusted p > 0.05, all comparisons). At 3 years, but not 2 years, cMIS was associated with superior mean leg pain scores (adjusted p = 0.032). For ODI, the null 3-year finding was driven by a mean increase in 3-year ODI in cMIS compared to the hybrid cohort: 1) mean 2- to 3-year change, hybrid −2.1 ± 11.6 versus cMIS +5.7 ± 17.8, adjusted p = 0.095; and 2) mean 2- to 3-year percentage change, hybrid −2.0% ± 47% versus cMIS +50.0% ± 134%, adjusted p = 0.177. For VAS back and leg pain, there were no significant differences for 2- to 3-year change (adjusted p = 0.850 and 0.213, respectively) and percentage change (adjusted p = 0.652 and 0.163, respectively).
Complications
Overall, there was a significantly higher rate of complications in the hybrid cohort (57.3% vs 40.7%, adjusted p = 0.006; Table 2). However, there was a higher rate of pseudarthrosis in the cMIS cohort (9.3% vs 1.1%, adjusted p = 0.039).
Complications
| Variable | Hybrid (n = 89) | cMIS (n = 108) | Unadjusted p Value | Adjusted OR (95% CI)* | Adjusted p Value* |
|---|---|---|---|---|---|
| Any complication, n (%) | 51 (57.3) | 44 (40.7) | 0.015 | 3.2 (1.4–7.5) | 0.006 |
| Reoperation | 26 (29.2) | 24 (22.2) | 0.169 | 1.8 (0.7–4.4) | 0.190 |
| Major complication | 31 (34.8) | 20 (18.5) | 0.007 | 2.4 (1.0–5.9) | 0.050 |
| Minor complication | 42 (47.2) | 37 (34.3) | 0.045 | 3.7 (1.5–9.1) | 0.004 |
| Infectious | 11 (12.4) | 3 (2.8) | 0.009 | 17.8 (3.1–104.0) | <0.001 |
| Implant-related | 13 (14.6) | 14 (13.0) | 0.448 | 0.8 (0.3–2.6) | 0.712 |
| Radiographic | 21 (23.6) | 24 (22.2) | 0.476 | 1.5 (0.6–3.8) | 0.360 |
| PJK | 16 (18.0) | 7 (6.5) | 0.011 | 4.7 (1.4–16.6) | 0.015 |
| DJK | 2 (2.2) | 4 (3.7) | 0.437 | 1.8 (0.2–15.7) | 0.580 |
| Pseudarthrosis | 1 (1.1) | 10 (9.3) | 0.012 | 0.1 (0.003–0.9) | 0.039 |
| Adjacent-segment disease | 3 (3.4) | 7 (6.5) | 0.256 | 1.3 (0.2–7.0) | 0.790 |
| Sagittal imbalance | 2 (2.2) | 2 (1.9) | 0.613 | 6.9 (0.5–106.2) | 0.166 |
| Coronal imbalance | 1 (1.1) | 1 (0.9) | 0.701 | 3.6 (0.1–91.2) | 0.444 |
| Radiographic complication requiring reoperation | 16 (18.0) | 17 (15.7) | 0.409 | 1.4 (0.5–3.6) | 0.521 |
| Surgical site infection | 4 (4.5) | 3 (2.8) | 0.394 | 1.8 (0.2–15.1) | 0.605 |
| Neurological | 20 (22.5) | 12 (11.1) | 0.025 | 4.1 (1.4–11.8) | 0.009 |
| Cardiopulmonary | 13 (14.6) | 1 (0.9) | <0.001 | 154.6 (0.9–27940) | 0.057 |
| Gastrointestinal | 4 (4.5) | 1 (0.9) | 0.130 | 3.0 (0.2–44.3) | 0.422 |
Boldface type indicates statistical significance.
Adjusted for age, preoperative CC angle, levels of surgery, levels of interbody fusion, presence of L5–S1 ALIF, UIV, and LIV.
Considering the waning durability of outcomes from 2 to 3 years following cMIS corrections when compared to hybrid techniques, we compared the prevalence of specific complications between the cMIS cohorts with and without ODI and back pain worsening between 2 and 3 years (Table 3). The loss of benefit in ODI at 3 years for cMIS was driven by reoperation occurrence (p = 0.018) and radiographic complications requiring reoperation (p = 0.017). Similarly, the loss of benefit in back pain at 3 years for cMIS was driven by pseudarthrosis (p = 0.034) and radiographic complications requiring reoperation (p = 0.047). For the hybrid cohort, there were no differences in complication rates between the cohorts with and without clinical worsening in ODI and back pain between 2 and 3 years (p > 0.05).
cMIS complications
| Variable | ODI* | p Value | VAS Back Pain Score* | p Value | ||
|---|---|---|---|---|---|---|
| Same or Better, n = 20 | Worse, n = 29 | Same or Better, n = 20 | Worse, n = 29 | |||
| Any complication, n (%) | 11 (55.0) | 11 (37.9) | 0.187 | 10 (50.0) | 12 (41.4) | 0.380 |
| Reoperation | 4 (20.0) | 9 (31.0) | 0.018 | 4 (20.0) | 9 (31.0) | 0.301 |
| Major complication | 4 (20.0) | 7 (24.1) | 0.082 | 3 (15.0) | 8 (27.6) | 0.248 |
| Minor complication | 9 (45.0) | 8 (27.6) | 0.364 | 8 (40.0) | 9 (31.0) | 0.364 |
| Infectious | 1 (5.0) | 2 (6.9) | 0.361 | 2 (10.0) | 1 (3.4) | 0.361 |
| Implant-related | 4 (20.0) | 3 (10.3) | 0.609 | 2 (10.0) | 5 (17.2) | 0.391 |
| Radiographic | 4 (20.0) | 7 (24.1) | 0.082 | 2 (10.0) | 9 (31.0) | 0.080 |
| PJK | 2 (10.0) | 2 (6.9) | 0.543 | 1 (5.0) | 3 (10.3) | 0.457 |
| DJK | 0 (0) | 2 (6.9) | 0.162 | 1 (5.0) | 1 (3.4) | 0.655 |
| Pseudarthrosis | 2 (10.0) | 4 (13.8) | 0.176 | 0 (0) | 6 (20.7) | 0.034 |
| Adjacent-segment disease | 0 (0) | 2 (6.9) | 0.162 | 1 (5.0) | 1 (3.4) | 0.655 |
| Sagittal Imbalance | 0 (0) | 0 (0) | — | 0 (0) | 0 (0) | — |
| Coronal Imbalance | 0 (0) | 0 (0) | — | 0 (0) | 0 (0) | — |
| Radiographic complication requiring reoperation | 2 (10.0) | 7 (24.1) | 0.017 | 1 (5.0) | 8 (27.6) | 0.047 |
| Surgical site infection | 1 (5.0) | 2 (6.9) | 0.361 | 0 (0) | 3 (10.3) | 0.198 |
| Neurological | 1 (5.0) | 2 (6.9) | 0.361 | 1 (5.0) | 2 (6.9) | 0.639 |
| Cardiopulmonary | 1 (5.0) | 0 (0) | 0.592 | 1 (5.0) | 0 (0) | 0.408 |
| Gastrointestinal | 1 (5.0) | 0 (0) | 0.592 | 0 (0) | 1 (3.4) | 0.592 |
Boldface type indicates statistical significance.
Between 2 and 3 years.
Despite the diminishment in outcomes between 2 and 3 years postoperatively after cMIS correction, it is important to note that the cMIS group still achieved similar, if not greater, mean improvements than the hybrid cohort from baseline to the 3-year time point (i.e., 36-month change) for disability, back pain, and leg pain (Table 1).
Subgroup Analysis
Five or Fewer Levels
We repeated the radiographic, clinical, and complication analyses, comparing cMIS (n = 65) and hybrid (n = 23) surgeries with 5 or fewer levels only (Table 4). Regardless of approach, both procedures were associated with significant improvements at the latest follow-up for CC (but not SVA, PI-LL, PT, and SS) and for 2- and 3-year ODI, VAS back pain, and VAS leg pain compared to baseline (p < 0.05 for all comparisons).
Baseline and surgical characteristics
| Characteristic | ≤5 Levels Operated (n = 88) | ≥6 Levels Operated (n = 109) | ||||
|---|---|---|---|---|---|---|
| Hybrid (n = 23) | cMIS (n = 65) | p Value | Hybrid (n = 66) | cMIS (n = 43) | p Value | |
| Baseline, mean ± SD | ||||||
| Age (yrs) | 60.7 ± 8.2 | 64.1 ± 8.1 | 0.080 | 58.7 ± 14.0 | 61.9 ± 13.6 | 0.234 |
| BMI (kg/m2) | 29.0 ± 3.0 | 27.7 ± 5.2 | 0.214 | 27.1 ± 5.8 | 27.4 ± 6.0 | 0.796 |
| Length of FU (mos) | 39.2 ± 17.8 | 36.6 ± 13.2 | 0.464 | 40.2 ± 16.6 | 42.6 ± 12.8 | 0.428 |
| ODI | 55.1 ± 17.2 | 48.3 ± 19.4 | 0.140 | 54.1 ± 16.5 | 51.8 ± 16.9 | 0.411 |
| VAS back pain | 7.2 ± 2.0 | 6.9 ± 2.1 | 0.591 | 6.8 ± 2.3 | 6.5 ± 2.2 | 0.533 |
| VAS leg pain | 6.5 ± 3.1 | 5.8 ± 3.0 | 0.392 | 5.2 ± 3.0 | 5.6 ± 2.5 | 0.438 |
| CC (°) | 19.1 ± 9.6 | 25.9 ± 12.4 | 0.053 | 41.9 ± 17.5 | 38.9 ± 15.4 | 0.379 |
| SVA (mm) | 55.3 ± 59.9 | 42.3 ± 50.2 | 0.364 | 50.4 ± 63.5 | 37.0 ± 56.7 | 0.283 |
| PI-LL (°) | 16.7 ± 14.0 | 13.4 ± 13.6 | 0.349 | 18.3 ± 18.8 | 16.1 ± 17.5 | 0.545 |
| PT (°) | 21.1 ± 10.1 | 22.7 ± 9.7 | 0.535 | 23.9 ± 10.6 | 24.6 ± 11.6 | 0.749 |
| SS (°) | 37.1 ± 10.8 | 32.0 ± 9.3 | 0.049 | 29.3 ± 13.1 | 26.8 ± 9.0 | 0.292 |
| Surgical | ||||||
| Staged, n (%) | 6 (26.1) | 26 (40.0) | 0.238 | 40 (60.6) | 29 (67.4) | 0.544 |
| Mean levels ± SD | 2.9 ± 1.3 | 3.2 ± 1.3 | 0.367 | 9.9 ± 3.8 | 7.1 ± 1.7 | <0.001 |
| Mean levels w/ interbody fusion ± SD | 1.9 ± 1.2 | 2.9 ± 1.1 | 0.001 | 3.4 ± 1.3 | 4.3 ± 1.7 | 0.002 |
| L5–S1 ALIF, n (%) | 8 (34.8) | 8 (12.3) | 0.005 | 10 (15.2) | 4 (9.3) | 0.583 |
| UIV, n (%) | 0.256 | 0.012 | ||||
| T1–6 | 0 (0) | 0 (0) | 19 (28.8) | 4 (9.3) | ||
| T7–12 | 1 (4.5) | 0 (0) | 44 (66.7) | 32 (74.4) | ||
| Below T12 | 21 (95.5) | 64 (100) | 3 (4.5) | 7 (16.3) | ||
| LIV, n (%) | 0.004 | <0.001 | ||||
| Above S1 | 5 (22.7) | 41 (64.1) | 20 (30.3) | 14 (32.6) | ||
| S1 | 14 (63.6) | 19 (29.7) | 7 (10.6) | 20 (46.5) | ||
| Ilium | 3 (13.6) | 4 (6.3) | 39 (59.1) | 9 (20.9) | ||
| Use of bmp, n (%) | 11 (47.8) | 34 (53.1) | 0.808 | 53 (80.3) | 37 (86.0) | 0.307 |
FU = follow-up.
Percentages reported are considered from available information. Frequencies may not add up to the total cohort size where there are missing data. Boldface type indicates statistical significance.
Adjusting for baseline differences, at 2 years the cMIS cohort was associated with lower 2-year ODI (24.1 ± 18.7 vs 35.6 ± 19.6, adjusted p = 0.048). Additionally, the cMIS cohort was associated with a greater 2-year back pain improvement (−4.0 ± 2.8 vs −2.7 ± 3.8, adjusted p = 0.039) and greater 2-year back pain change as a percentage of baseline (−56.1% ± 36% vs −29.5% ± 49%, adjusted p = 0.019). At 3 years, the cMIS cohort remained associated with a greater 3-year back pain improvement (−4.1 ± 1.9 vs −1.7 ± 2.8, adjusted p = 0.007) and greater 3-year back pain change as a percentage of baseline (−58.5% ± 27% vs −18.5% ± 42%, adjusted p = 0.004). The cMIS cohort was associated with less 3-year leg pain (1.9 ± 2.7 vs 3.0 ± 3.2, adjusted p = 0.029), greater 3-year leg pain improvement (−5.2 ± 1.9 vs −3.9 ± 4.0, adjusted p = 0.005), and greater 3-year leg pain change as a percentage of baseline (−80.2% ± 25% vs −31.8% ± 90%, adjusted p = 0.009). Comparing 2- versus 3-year change, cMIS was associated with continued improvement in leg pain as a percentage of 2-year leg pain (−34.1% ± 75%) compared to an observed worsening in the hybrid cohort (84.5% ± 277%, adjusted p = 0.019). Otherwise, there were no significant differences for the remaining 2-year, 2-year change, 3-year, 3-year change, and 2- to 3-year change scores for ODI, back pain, and leg pain (adjusted p > 0.05 for all comparisons).
Adjusting for baseline differences, the cMIS cohort was associated with lower postoperative PI-LL (10.5° ± 13.2° vs 16.9° ± 14.3°, adjusted p = 0.044) compared to the hybrid cohort. Similarly, the cMIS cohort was associated with lower postoperative SVA (39.5 ± 52.2 mm vs 67.9 ± 58.6 mm, adjusted p = 0.037) compared to the hybrid cohort. The two cohorts otherwise did not differ significantly for the remaining radiographic parameters and postoperative complication rates (adjusted p > 0.05 for all comparisons).
Six or More Levels
We repeated the radiographic, clinical, and complication analyses, comparing cMIS (n = 43) and hybrid (n = 66) surgeries with 6 or more levels only (Table 4). Regardless of cMIS or hybrid approach, both procedures were associated with a significant improvement at the latest follow-up for CC and PI-LL (but not SVA, PT, and SS) and for 2- and 3-year ODI, VAS back pain, and VAS leg pain compared to baseline (p < 0.05 for all comparisons).
Adjusting for these baseline differences, the cMIS cohort was associated with less 2-year back pain (2.6 ± 2.3 vs 4.0 ± 2.7, adjusted p = 0.002), greater 2-year back pain improvement (−4.0 ± 2.6 vs −2.8 ± 2.6 adjusted p = 0.038), and greater 2-year back pain reduction as a percentage of baseline (−61.6% ± 36% vs −38.0% ± 45%, adjusted p = 0.004) compared to the hybrid cohort. Additionally, the cMIS cohort was associated with lower 2-year ODI (27.5 ± 21.3 vs 33.6 ± 16.2, adjusted p = 0.006) and greater 2-year ODI reduction as a percentage of baseline (−47.0% ± 40% vs −35.6% ± 30% adjusted p = 0.033). Comparing 2- versus 3-year change, cMIS was associated with worsening in ODI from 2 to 3 years as a percentage of 2-year ODI (72.2% ± 169%) compared to an observed continued improvement in the hybrid cohort (−5.7% ± 51%, adjusted p = 0.042). Otherwise, there were no significant differences for the remaining 2-year, 2-year change, 3-year, 3-year change, and 2- to 3-year change scores for ODI, back pain, and leg pain (adjusted p > 0.05 for all comparisons).
Adjusting for baseline differences, the cohorts did not differ significantly for postoperative radiographic parameters (adjusted p > 0.05 for all comparisons). The hybrid cohort was associated with an increased odds of overall complication (odds ratio [OR] 3.1, 95% confidence interval [CI] 1.2–7.8, adjusted p = 0.020), minor complication (OR 3.1, 95% CI 1.1–8.4, adjusted p = 0.026), infectious complication (OR 13.9, 95% CI 1.4–144.3, adjusted p = 0.027), and neurological complication (OR 4.5, 95% CI 1.2–17.8, adjusted p = 0.031). Otherwise, there were no significant differences for reoperation, major complications, implant-related complications, radiographic complications, surgical site infections, cardiopulmonary complications, and gastrointestinal complications (adjusted p > 0.05 for all comparisons).
Given the waning benefit durability of the cMIS technique compared to the hybrid technique between 2 to 3 years, we compared the occurrence of complications between the cMIS cohorts with and without ODI and back pain worsening between 2 and 3 years. The loss of benefit in ODI was not associated with different rates of complication occurrence. However, the loss of benefit in back pain at 3 years for cMIS was driven by pseudarthrosis (p = 0.024) and radiographic complications requiring reoperation (p = 0.046).
Loss to Follow-Up
To assess the impact of loss to follow-up on outcomes following cMIS or hybrid techniques for ASD, we compared complications and clinical and radiographic outcomes for those who reached 3-year follow-up (n = 111) versus those who did not (n = 86). At baseline (Table 5), those lost to follow-up had a lower proportion of staged procedures (41.9% vs 58.6%, p = 0.020) and a mean lower number of levels with interbody fusion (2.9 ± 1.4 vs 3.5 ± 1.5, p = 0.002). Otherwise, the groups were similar for the remaining baseline characteristics. Table 5 compares the clinical and radiographic outcomes of the two cohorts. Adjusting for baseline differences, those lost to follow-up had lower VAS leg pain change (−2.7 ± 3.7 vs −3.8 ± 3.1, adjusted p = 0.039) and VAS leg pain percentage change (−17.4% ± 149% vs −55.3% ± 59%, adjusted p = 0.044). Otherwise, loss to follow-up was not associated with different clinical outcomes for ODI and VAS back pain, radiographic outcomes, or complication occurrence.
Baseline characteristics and radiographic and clinical outcomes at last follow-up
| Characteristic | 3-Yr FU Achieved (n = 111) | Lost to FU (n = 86) | p Value | |
|---|---|---|---|---|
| Unadjusted | Adjusted* | |||
| Baseline, mean ± SD | ||||
| Age (yrs) | 62.0 ± 10.2 | 60.7 ± 13.6 | 0.588 | — |
| BMI (kg/m2) | 27.7 ± 5.2 | 27.3 ± 5.7 | 0.628 | — |
| Length of FU (mos) | 47.6 ± 13.9 | 28.8 ± 8.1 | <0.001 | — |
| ODI | 52.2 ± 17.3 | 51.5 ± 18.5 | 0.306 | — |
| VAS back pain | 7.0 ± 2.0 | 6.6 ± 2.3 | 0.178 | — |
| VAS leg pain | 5.8 ± 2.7 | 5.3 ± 3.1 | 0.106 | — |
| CC (°) | 34.7 ± 16.5 | 33.8 ± 17.9 | 0.364 | — |
| SVA (mm) | 40.0 ± 54.4 | 53.7 ± 62.0 | 0.547 | — |
| PI-LL (°) | 15.8 ± 17.4 | 16.7 ± 15.2 | 0.540 | — |
| PT (°) | 23.3 ± 10.2 | 23.4 ± 11.0 | 0.655 | — |
| SS (°) | 30.2 ± 10.9 | 30.7 ± 11.9 | 0.471 | — |
| Surgical | ||||
| Staged, n (%) | 65 (58.6) | 36 (41.9) | 0.020 | — |
| Mean levels ± SD | 6.4 ± 3.8 | 6.2 ± 4.2 | 0.511 | — |
| Mean levels w/ interbody fusion ± SD | 3.5 ± 1.5 | 2.9 ± 1.4 | 0.002 | — |
| L5–S1 ALIF, n (%) | 14 (12.6) | 14 (16.3) | 0.298 | — |
| UIV, n (%) | 0.237 | — | ||
| T1–6 | 11 (10.0) | 12 (14.1) | ||
| T7–12 | 49 (44.5) | 28 (32.9) | ||
| Below T12 | 50 (45.5) | 45 (52.9) | ||
| LIV, n (%) | 0.258 | — | ||
| Above S1 | 43 (39.1) | 37 (43.5) | ||
| S1 | 39 (35.5) | 21 (24.7) | ||
| Ilium | 28 (25.5) | 27 (31.8) | ||
| Use of bmp, n (%) | 77 (70.0) | 58 (67.4) | 0.409 | — |
| Radiographic, mean ± SD | ||||
| CC (°) | ||||
| Postop | 16.6 ± 11.5 | 17.2 ± 10.2 | 0.994 | 0.844 |
| Change | −18.3 ± 14.1 | −16.4 ± 12.4 | 0.351 | 0.916 |
| SVA (mm) | ||||
| Postop | 41.2 ± 53.8 | 46.4 ± 61.9 | 0.259 | 0.414 |
| Change | 1.7 ± 54.4 | −6.9 ± 54.7 | 0.794 | 0.195 |
| PI-LL (°) | ||||
| Postop | 10.6 ± 15.0 | 11.0 ± 15.0 | 0.958 | 0.969 |
| Change | −5.1 ± 17.2 | −5.4 ± 13.2 | 0.141 | 0.641 |
| PT (°) | ||||
| Postop | 24.0 ± 9.9 | 23.1 ± 10.9 | 0.927 | 0.550 |
| Change | 0.3 ± 8.1 | 0.1 ± 6.3 | 0.081 | 0.714 |
| SS (°) | ||||
| Postop | 29.8 ± 10.4 | 30.4 ± 11.5 | 0.273 | 0.779 |
| Change | 0.02 ± 8.4 | 0.6 ± 6.7 | 0.103 | 0.584 |
| Clinical, mean ± SD | ||||
| ODI | ||||
| Latest | 34.2 ± 19.0 | 31.4 ± 20.1 | 0.343 | 0.587 |
| Latest change | −18.6 ± 19.7 | −19.6 ± 18.7 | 0.711 | 0.939 |
| Latest percentage change | −32.3% ± 40% | −38.7% ± 39% | 0.287 | 0.496 |
| VAS back pain | ||||
| Latest | 3.9 ± 2.5 | 3.6 ± 2.9 | 0.470 | 0.467 |
| Latest change | −3.2 ± 2.5 | −3.0 ± 3.1 | 0.662 | 0.915 |
| Latest percentage change | −44.2% ± 38% | −39.9% ± 53% | 0.525 | 0.701 |
| VAS leg pain | ||||
| Latest | 2.4 ± 2.8 | 2.7 ± 2.9 | 0.536 | 0.202 |
| Latest change | −3.8 ± 3.1 | −2.7 ± 3.7 | 0.044 | 0.039 |
| Latest percentage change | −55.3% ± 59% | −17.4% ± 149% | 0.033 | 0.044 |
Percentages reported are considered from available information. Frequencies may not add up to the total cohort size where there are missing data. Boldface type indicates statistical significance.
Adjusted for surgical characteristics reaching p < 0.05 on univariate comparisons (surgical staging, number of levels with an interbody fusion).
Discussion
In a study of 197 patients undergoing hybrid or cMIS for ASD, both techniques were associated with significant clinical improvements from baseline at 2- and 3-year time points. cMIS was associated with superior 2-year clinical outcomes for ODI and back pain, in terms of both absolute mean ODI and VAS back pain scores and ODI and VAS back pain change as a percentage of baseline. These differences were no longer apparent at 3 years. cMIS was associated with superior 3-year leg pain scores compared to hybrid procedures, but no difference was observed at 2 years. At the latest follow-up, both procedures were associated with significant improvements in CC and PI-LL, but no changes in SVA and PT. Hybrid procedures were associated with a greater improvement in CC, but were associated with higher odds of complications including minor complications, proximal junctional kyphosis (PJK), and infectious and neurological complications. However, there were higher odds of pseudarthrosis following cMIS compared to hybrid procedures.
A prior study by Park et al.3 including 43 cMIS and 62 hybrid surgeries for ASD investigated outcomes with a minimum of 1-year follow-up (mean follow-up 31.3–38.3 months). The study found no between-group differences at the latest follow-up with regard to clinical outcomes (ODI or pain scores) or degree of radiographic correction for CC, LL, SVA, or PI-LL mismatch. What the study did not address is if there are between-group differences at intermediate time points. This is especially concerning given the described deterioration of clinical benefit following both traditional open16 and less invasive surgical techniques for ASD15 after 2-year postoperative follow-up, indicating the dynamic nature of clinical outcomes over time.
Our present study provides needed granularity in addressing this knowledge gap. We found that at the interim 2-year clinical time point, cMIS procedures were associated with superior benefit for disability and back pain. This benefit appears to depreciate over time, into equivalence at 3 years, consistent with the findings of Park et al.3 The equivalency in 3-year ODI outcomes appears to be driven by a worsening in ODI between 2 and 3 years for the cMIS cohort rather than a continued improvement in the hybrid cohort. Similarly, although both cohorts demonstrated a mean worsening in back pain between 2- and 3-year time points, the equivalence in 3-year back pain outcomes appears to be due to a larger mean worsening in back pain in the cMIS cohort. This waning benefit durability of cMIS may be a result of: 1) the higher rate of pseudarthrosis following cMIS; 2) the higher rate of reoperation and radiographic complications requiring reoperation in the cMIS subgroup experiencing ODI worsening (compared to those not experiencing ODI worsening) between 2 and 3 years; and/or 3) the higher rate of pseudarthrosis and radiographic complications requiring reoperation in the cMIS subgroup experiencing back pain worsening (compared to those not experiencing back pain worsening) between 2 and 3 years. The percutaneous posterior approaches in cMIS are less amenable to meticulous decortication during posterolateral and laminar arthrodesis, which may account for the higher rates of pseudarthrosis.22
Investigations of durability are particularly relevant for ASD, given the common reporting of 2-year outcomes or heterogeneous evaluations of outcomes at the latest follow-up. We demonstrate that clinical outcomes continue to evolve between 2 and 3 years. For example, although there was no difference in leg pain outcomes at 2 years, cMIS was associated with superior leg pain outcomes at 3 years compared to hybrid approaches. This was driven by a continued improvement in leg pain between 2 and 3 years in the cMIS cohort as opposed to a plateauing of leg pain after 2 years for the hybrid cohort. The factors driving this discrepancy are likely multifactorial, although the higher rate of complications following hybrid procedures, including PJK and neurological deficits, may be contributory.
Multiple investigations also support the notion of evolving clinical outcomes over time following surgery for ASD. In a study of 53 patients undergoing cMIS or hybrid surgery for ASD who were followed yearly for 4 years postoperatively, Wang et al. found that clinical improvement peaked at 2-year follow-up, and that by 4-year follow-up patients had lost an average of 24.4% of ODI improvement.15 In a study of 118 patients undergoing open thoracolumbar deformity correction who were followed for 5 or more years, Adogwa et al. found that outcomes peaked at 2-year follow-up and were reduced by 5-year follow-up.16 Combined with our findings of decreased ODI from 2 to 3 years following cMIS for ASD, these studies yield important insights on durability. First, this may help place postoperative follow-up at 2- and 3-year time points in an appropriate context for surgeons and patients. Second, this may also encourage closer follow-up of patients past the 2-year time point, especially in cohorts receiving cMIS procedures. Complications, including pseudarthrosis and adjacent-segment disease, will accumulate with time and have an escalating role in outcome deterioration as cohorts are followed longitudinally. Other problems unrelated to the initial surgery, including accumulation of age-related changes (e.g., vestibular dysfunction,23 sarcopenia24), medical conditions (e.g., diabetes,25 osteoarthritis,26 polypharmacy27), and decrements in physical function (e.g., decreased walking speed and mobility disability28) over time, are also important to consider. These other processes challenge the health and physical function of ASD patients longitudinally, further impacting disability and pain postoperatively. These other issues may adversely confound results from some PROMs.
Correcting coronal and sagittal radiographic parameters is a goal of ASD surgery. In our study, patients undergoing cMIS had fewer severe deformities, characterized by lower mean baseline CC, SVA, and PI-LL mismatch. Both techniques were associated with radiographic improvements in CC and PI-LL, and both had similar postoperative SVA and PT. This latter finding may be a function of the baseline values of SVA and PT for both cohorts not indicating severe derangement preoperatively.
The radiographic change in CC and PI-LL was of greater magnitude for the hybrid cohort, resulting in ultimately similar mean postoperative CC and PI-LL. Upon adjustment for baseline differences between the cohorts, hybrid procedures were associated with greater CC improvement at the latest follow-up, consistent with the investigation by Park et al.3 These results reinforce the notion that cMIS approaches may be less suitable for larger magnitudes of coronal29 and sagittal correction, as compared to traditional open30 or (as observed presently) hybrid deformity surgery. The selection of patients requiring less major deformity correction in cMIS, as opposed to a hybrid technique, is consistent with the MIS approach selection algorithms (MISDEF31 and MISDEF24) published by Mummaneni et al. MIS techniques should be used judiciously and may not be appropriate for patients with more severe deformity requiring long-segment fusions or with preexisting multilevel instrumentation.4
Consistent with prior investigations on complications following ASD surgery in general,32–37 both cMIS and hybrid procedures were associated with a modest complication rate (40.7%–57.3%). As reported previously,3 cMIS procedures were associated with fewer complications (40.7%) compared to hybrid procedures (57.3%). In patients less tolerant of complications (such as the elderly or frail), cMIS techniques may be preferentially used when feasible. However, this must be balanced with the higher rate of pseudarthrosis observed in the cMIS cohort (9.3% vs 1.1%).
Although we accounted for the number of levels in our adjusted analyses, we conducted further subgroup analyses grouping patients into those with 5 or fewer levels operated and those with 6 or more levels operated. The findings in the subgroup undergoing 6 or more levels of reconstruction were consistent with the overall analyses; namely, cMIS procedures were associated with superior back pain and disability profiles at 2 years, and equivalence in the same parameters at 3 years when compared to hybrid procedures. The 3-year equivalence in PROMs between hybrid and cMIS subgroups suggests a waning durability of these early, more optimal outcome measures within the cMIS subgroup. As in the overall analysis, this waning durability in the subgroup with 6 or more levels was driven by higher rates of pseudarthrosis and radiographic complications requiring reoperation following cMIS compared to hybrid approaches. For the cohort with 5 or fewer levels operated, cMIS procedures were associated with superior 2-year back pain improvement and disability measures. However, at 3 years for the cohort with 5 or fewer levels operated, the cMIS cohort demonstrated durable superiority of back pain improvement and less leg pain when compared to the hybrid cohort. This result was driven by a divergence in leg pain durability between 2- and 3-year time points, wherein leg pain continued to improve for cMIS but worsened for the hybrid cohort. The back pain finding stands in contrast to the overall analysis, in which we found that cMIS exhibited waning back pain durability beyond 2 years. It appears that cMIS retains durability in comparison to hybrid constructs with shorter-segment fusions alone. This may be due to the similar rates of pseudarthrosis and radiographic complications following cMIS and hybrid procedures of 5 or fewer levels. These findings suggest that the clinical durability of cMIS and hybrid procedures may be considered distinctly for short- and long-segment operations. The reasons for these observed differences are multifactorial and may also include factors leading to the decision to pursue short- versus long-segment surgery,38 such as the patient’s presenting symptoms, patient radiographic characteristics, patient frailty, and surgeon philosophy.
Limitations
This is a retrospective study of a multicenter database and therefore has some associated limitations. Other limitations warrant discussion. First, alternative causes of lost durability should be considered. Specifically, compared to 2-year follow-up, there was lower follow-up at 3 years. This may result in selection bias stemming from patients with worse symptomatology disproportionately continuing to follow-up. Of note, however, is that the proportion of follow-up at 3 years was similar for both cohorts in our study. Additionally, our analysis comparing those reaching 3-year follow-up and those lost to follow-up revealed similar radiographic outcomes, complication profiles, and clinical outcomes—except for leg pain change and leg pain percentage change—at the latest follow-up. Furthermore, an asymmetrical distribution of patients with concomitant osteoarthritic conditions or who developed new pathologies (e.g., hip/knee osteoarthritis, sacroiliac joint degeneration/pain, trauma, tumor, etc.)39 should be considered. Second, our cohorts differed in several key baseline variables and surgical parameters. However, we attempted to account for these factors in our adjusted analyses. Third, compared to traditional open surgery, hybrid and cMIS techniques are relatively new. As surgeons’ experiences with each technique mature13 and technology improves, the findings reported here may evolve. Lastly, although we attempted to stratify less invasive techniques to treat ASD into hybrid and cMIS cohorts, these remain somewhat heterogeneous entities that may be employed distinctly by surgeons. To this point, our group has published an algorithm to aid in surgical selection for minimally invasive deformity correction that involves progressively more invasive surgical recommendations for progressively more severe and rigid deformities, among other characteristics.4 This provides an additional framework for selection of the cMIS versus hybrid approach for minimally invasive ASD surgery. Hybrid and cMIS procedures are not similarly appropriate for all cases of ASD. However, the comparison of cMIS and hybrid techniques holds precedent1–3,15,40 and remains clinically relevant. For the latter, there remains an overlap of patients who may receive either cMIS or hybrid surgery, such as those with moderate multiplanar deformity with severe foraminal stenosis along the concavity of a scoliotic curve. For these patients, neural decompression and deformity correction can both be achieved indirectly via disc/foraminal height restoration with a cMIS procedure or directly with osteotomies and facetectomies via a hybrid procedure.
Conclusions
Both cMIS and hybrid procedures were associated with clinical improvement in disability and back pain at 2 and 3 years. Although cMIS was associated with superior outcomes for ODI and back pain at 2 years, this diminished over time with no difference between groups at 3 years. This waning durability was partly driven by a higher rate of pseudarthrosis following cMIS compared to hybrid approaches. Conversely, cMIS was associated with superior outcomes for leg pain at 3 years, but not at 2 years, compared to hybrid procedures. Both procedures were associated with significant improvements in maximum CC and PI-LL mismatch. Hybrid procedures were associated with superior CC improvement compared to cMIS, but the two approaches shared similar sagittal radiographic outcomes. Compared to cMIS, hybrid procedures were associated with a higher rate of complications with the exception of pseudarthrosis.
Disclosures
Dr. Chan reports support of non–study-related research from Orthofix Medical Inc. Dr. Eastlack reports direct stock ownership in SI Bone, SeaSpine, Alphatec, NuVasive, and Spine Innovation; being a consultant to Spinal Elements, Aesculap, NuVasive, SeaSpine, SI Bone, Baxter, Stryker, Carevature, and Medtronic; being a patent holder for NuTech, Globus, Seaspine, SI Bone, Spine Innovation, and Stryker; being on the speakers bureau for Radius; receiving royalties from Globus, SI Bone, SeaSpine, Aesculap, and NuVasive; and receiving support of non–study-related clinical or research effort from SeaSpine, NuVasive, Medtronic, and SI Bone. Dr. Fessler reports royalties from DePuy-Synthes, outside the submitted work. Dr. Than reports being a consultant to Bioventus, Integrity Implants, and DePuy-Synthes, and receiving honoraria from LifeNet Health and DJO. Dr. Chou reports being a consultant to Globus and Medtronic, and receiving royalties from Globus. Dr. Fu reports being a consultant to DePuy-Synthes, Globus, Johnson & Johnson, SI Bone, and Atlas Spine. Dr. Park reports being a consultant to Globus and NuVasive, receiving royalties from Globus, receiving honorarium from Depuy, and receiving support of non–study-related clinical or research effort from DePuy, ISSG, SI Bone, and Cerapedics. Dr. Wang reports being a consultant for DePuy-Synthes, Spineology, Medtronic, Globus, and Stryker; being a patent holder for DePuy-Synthes; having direct stock ownership in ISD, Kinesiometrics, and Medical Device Partners; receiving royalties from DePuy-Synthes Spine, Children’s Hospital of Los Angeles, Springer Publishing, and Quality Medical Publishing; receiving grants from the Department of Defense; receiving personal fees from DePuy-Synthes Spine, Stryker Spine, K2M, and Spineology; being an advisory board member for Vallum; and owning stock in Spinicity and Innovative Surgical Devices, outside the submitted work. Dr. Kanter reports receiving royalties from NuVasive and Zimmer Biomet; being a consultant for NuVasive; and being a patent holder for Zimmer Biomet. Dr. Okonkwo reports receiving royalties from and being a consultant to NuVasive and Zimmer Biomet, and being a patent holder for Zimmer Biomet. Dr. Nunley reports royalties from Spineology Inc., Stryker, Zimmer, Camber Spine/IMSE, and Accelus; being on the speakers bureau for Spineology Inc., Zimmer, Camber Spine/IMSE, Instrinisic Therapeutics, Providence Medical, and Neo Spine; being a consultant for Spineology Inc., Stryker, Zimmer, Camber Spine/IMSE, Accelus, Centinel Spine, Instrinisic Therapeutics, Providence Medical, Neo Spine, NG Medical, and Regeltec; receving research support from Spineology Inc., Centinel Spine, Providence Medical, NuVasive, and Surgalign; and receiving stock/stock options in Spineology Inc., Camber Spine/IMSE, Regeltec, and Surgalign. Dr. Anand reports being a consultant to Medtronic, DePuy-Synthes, Spinal Balance, Spinal Simplicity, and Viseon; receiving royalties from Medtronic, Globus Medical, and Elsevier; being a patent holder for Medtronic; and having direct stock ownership in Globus Medical, Atlas Spine, Paradigm Spine, Spinal Balance, Spinal Simplicity, Viseon, and Theracell. Dr. Uribe reports being a consultant to NuVasive, SI Bone, Misonix, and Mainstay. Dr. Mundis reports being a consultant to NuVasive, Viseon, SeaSpine, and Carlsmed; having direct stock ownership in NuVasive, Viseon, SeaSpine, Alphatec, and Suralign; being a patent holder for K2M; and receiving royalties from NuVasive, Stryker, and K2M. Dr. Bess reports grants to the ISSG from Stryker, NuVasive, ISSGF, Globus, Medtronic, SI Bone, SEA Spine, and DePuy-Synthes; and receiving royalties from Stryker and NuVasive. Dr. Shaffrey reports direct stock ownership in NuVasive; being a consultant to NuVasive, Medtronic, and SI Bone; receiving royalties from NuVasive, Medtronic, and Zimmer Biomet; and being a patent holder for NuVasive, Medtronic, and Zimmer Biomet. Dr. Mummaneni reports being a consultant to Stryker Spine, DePuy Synthes, and Globus; having direct stock ownership in Spinicity/ISD; receiving statistical analysis for study/writing or editorial assistance on the manuscript from ISSG; receiving royalties from DePuy-Synthes, Thieme Publishers, and Springer Publishers; and receiving support of non–study-related clinical or research effort from AO Spine and NREF.
Author Contributions
Conception and design: Eastlack, Chan, Fessler, Park, Wang, Mummaneni. Acquisition of data: Eastlack, Fessler, Than, Chou, Fu, Park, Wang, Kanter, Okonkwo, Nunley, Anand, Uribe, Mundis, Bess, Shaffrey, Le, Mummaneni. Analysis and interpretation of data: Eastlack, Chan, Fessler, Chou, Mummaneni. Drafting the article: Eastlack, Chan, Fessler. Critically revising the article: Eastlack, Chan, Fessler, Than, Chou, Fu, Park, Wang, Kanter, Okonkwo, Nunley, Anand, Uribe, Mundis, Le, Mummaneni. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Eastlack. Statistical analysis: Le. Study supervision: Eastlack, Mummaneni.
References
- 1↑
Haque RM, Mundis GM Jr, Ahmed Y, El Ahmadieh TY, Wang MY, Mummaneni PV, et al. Comparison of radiographic results after minimally invasive, hybrid, and open surgery for adult spinal deformity: a multicenter study of 184 patients. Neurosurg Focus. 2014;36(5):E13.
- 2
Uribe JS, Deukmedjian AR, Mummaneni PV, Fu KM, Mundis GM Jr, Okonkwo DO, et al. Complications in adult spinal deformity surgery: an analysis of minimally invasive, hybrid, and open surgical techniques. Neurosurg Focus. 2014;36(5):E15.
- 3↑
Park P, Wang MY, Lafage V, Nguyen S, Ziewacz J, Okonkwo DO, et al. Comparison of two minimally invasive surgery strategies to treat adult spinal deformity. J Neurosurg Spine. 2015;22(4):374–380.
- 4↑
Mummaneni PV, Park P, Shaffrey CI, Wang MY, Uribe JS, Fessler RG, et al. The MISDEF2 algorithm: an updated algorithm for patient selection in minimally invasive deformity surgery. J Neurosurg Spine. 2019;32(2):221–228.
- 5↑
Choi SH, Son SM, Goh TS, Park W, Lee JS . Outcomes of operative and nonoperative treatment in patients with adult spinal deformity with a minimum 2-year follow-up: a meta-analysis. World Neurosurg. 2018;120:e870–e876.
- 6↑
Lenke LG, Fehlings MG, Shaffrey CI, Cheung KM, Carreon L, Dekutoski MB, et al. Neurologic outcomes of complex adult spinal deformity surgery: results of the prospective, multicenter Scoli-RISK-1 study. Spine (Phila Pa 1976).2016;41(3):204–212.
- 7
Booth KC, Bridwell KH, Lenke LG, Baldus CR, Blanke KM . Complications and predictive factors for the successful treatment of flatback deformity (fixed sagittal imbalance). Spine (Phila Pa 1976).1999;24(16):1712–1720.
- 8
Glassman SD, Hamill CL, Bridwell KH, Schwab FJ, Dimar JR, Lowe TG . The impact of perioperative complications on clinical outcome in adult deformity surgery. Spine (Phila Pa 1976).2007;32(24):2764–2770.
- 9
Street JT, Lenehan BJ, DiPaola CP, Boyd MD, Kwon BK, Paquette SJ, et al. Morbidity and mortality of major adult spinal surgery. A prospective cohort analysis of 942 consecutive patients. Spine J. 2012;12(1):22–34.
- 10
Yadla S, Maltenfort MG, Ratliff JK, Harrop JS . Adult scoliosis surgery outcomes: a systematic review. Neurosurg Focus. 2010;28(3):E3.
- 11
Anand N, Baron EM, Thaiyananthan G, Khalsa K, Goldstein TB . Minimally invasive multilevel percutaneous correction and fusion for adult lumbar degenerative scoliosis: a technique and feasibility study. J Spinal Disord Tech. 2008;21(7):459–467.
- 12
Wang MY, Mummaneni PV . Minimally invasive surgery for thoracolumbar spinal deformity: initial clinical experience with clinical and radiographic outcomes. Neurosurg Focus. 2010;28(3):E9.
- 13↑
Wang MY, Tran S, Brusko GD, Eastlack R, Park P, Nunley PD, et al. Less invasive spinal deformity surgery: the impact of the learning curve at tertiary spine care centers. J Neurosurg Spine. 2019;31(6):865–872.
- 14↑
Anand N, Baron EM, Khandehroo B, Kahwaty S . Long-term 2- to 5-year clinical and functional outcomes of minimally invasive surgery for adult scoliosis. Spine (Phila Pa 1976).2013;38(18):1566–1575.
- 15↑
Wang MY, Park P, Tran S, Anand N, Nunley P, Kanter A, et al. Intermediate-term clinical and radiographic outcomes with less invasive adult spinal deformity surgery: patients with a minimum follow-up of 4 years. Acta Neurochir (Wien). 2020;162(6):1393–1400.
- 16↑
Adogwa O, Karikari IO, Elsamadicy AA, Sergesketter AR, Galan D, Bridwell KH . Correlation of 2-year SRS-22r and ODI patient-reported outcomes with 5-year patient-reported outcomes after complex spinal fusion: a 5-year single-institution study of 118 patients. J Neurosurg Spine. 2018;29(4):422–428.
- 17↑
Passias PG, Bortz CA, Lafage V, Lafage R, Smith JS, Line B, et al. Durability of satisfactory functional outcomes following surgical adult spinal deformity correction: a 3-year survivorship analysis. Oper Neurosurg (Hagerstown). 2020;18(2):118–125.
- 18↑
Daniels AH, Bess S, Line B, Eltorai AEM, Reid DBC, Lafage V, et al. Peak timing for complications after adult spinal deformity surgery. World Neurosurg. 2018;115:e509–e515.
- 19↑
Mok JM, Cloyd JM, Bradford DS, Hu SS, Deviren V, Smith JA, et al. Reoperation after primary fusion for adult spinal deformity: rate, reason, and timing. Spine (Phila Pa 1976).2009;34(8):832–839.
- 20↑
Tsuchiya K, Bridwell KH, Kuklo TR, Lenke LG, Baldus C . Minimum 5-year analysis of L5-S1 fusion using sacropelvic fixation (bilateral S1 and iliac screws) for spinal deformity. Spine (Phila Pa 1976).2006;31(3):303–308.
- 21↑
Buell TJ, Smith JS, Shaffrey CI, Kim HJ, Klineberg EO, Lafage V, et al. Multicenter assessment of surgical outcomes in adult spinal deformity patients with severe global coronal malalignment: determination of target coronal realignment threshold. J Neurosurg Spine. 2021;34(3):399–412.
- 22↑
Kim YJ, Bridwell KH, Lenke LG, Rhim S, Cheh G . Pseudarthrosis in long adult spinal deformity instrumentation and fusion to the sacrum: prevalence and risk factor analysis of 144 cases. Spine (Phila Pa 1976).2006;31(20):2329–2336.
- 24↑
Dodds RM, Granic A, Davies K, Kirkwood TB, Jagger C, Sayer AA . Prevalence and incidence of sarcopenia in the very old: findings from the Newcastle 85+ Study. J Cachexia Sarcopenia Muscle. 2017;8(2):229–237.
- 25↑
Kirkman MS, Briscoe VJ, Clark N, Florez H, Haas LB, Halter JB, et al. Diabetes in older adults. Diabetes Care. 2012;35(12):2650–2664.
- 26↑
Collerton J, Davies K, Jagger C, Kingston A, Bond J, Eccles MP, et al. Health and disease in 85 year olds: baseline findings from the Newcastle 85+ cohort study. BMJ. 2009;339:b4904.
- 27↑
Wang R, Chen L, Fan L, Gao D, Liang Z, He J, et al. Incidence and effects of polypharmacy on clinical outcome among patients aged 80+: a five-year follow-up study. PLoS One. 2015;10(11):e0142123.
- 28↑
Jaul E, Barron J . Age-related diseases and clinical and public health implications for the 85 years old and over population. Front Public Health. 2017;5:335.
- 29↑
Eastlack RK, Mundis GM Jr, Wang M, Mummaneni PV, Uribe J, Okonkwo D, et al. Is there a patient profile that characterizes a patient with adult spinal deformity as a candidate for minimally invasive surgery? Global Spine J. 2017;7(7):703–708.
- 30↑
Uddin OM, Haque R, Sugrue PA, Ahmed YM, El Ahmadieh TY, Press JM, et al. Cost minimization in treatment of adult degenerative scoliosis. J Neurosurg Spine. 2015;23(6):798–806.
- 31↑
Mummaneni PV, Shaffrey CI, Lenke LG, Park P, Wang MY, La Marca F, et al. The minimally invasive spinal deformity surgery algorithm: a reproducible rational framework for decision making in minimally invasive spinal deformity surgery. Neurosurg Focus. 2014;36(5):E6.
- 32
DeWald CJ, Stanley T . Instrumentation-related complications of multilevel fusions for adult spinal deformity patients over age 65: surgical considerations and treatment options in patients with poor bone quality. Spine (Phila Pa 1976).2006;31(19 suppl):S144–S151.
- 33
Daubs MD, Lenke LG, Cheh G, Stobbs G, Bridwell KH . Adult spinal deformity surgery: complications and outcomes in patients over age 60. Spine (Phila Pa 1976).2007;32(20):2238–2244.
- 34
Cho SK, Bridwell KH, Lenke LG, Yi JS, Pahys JM, Zebala LP, et al. Major complications in revision adult deformity surgery: risk factors and clinical outcomes with 2- to 7-year follow-up. Spine (Phila Pa 1976).2012;37(6):489–500.
- 35
Smith JS, Klineberg E, Lafage V, Shaffrey CI, Schwab F, Lafage R, et al. Prospective multicenter assessment of perioperative and minimum 2-year postoperative complication rates associated with adult spinal deformity surgery. J Neurosurg Spine. 2016;25(1):1–14.
- 36
Soroceanu A, Burton DC, Oren JH, Smith JS, Hostin R, Shaffrey CI, et al. Medical complications after adult spinal deformity surgery: incidence, risk factors, and clinical impact. Spine (Phila Pa 1976).2016;41(22):1718–1723.
- 37
Kelly MP, Lenke LG, Godzik J, Pellise F, Shaffrey CI, Smith JS, et al. Retrospective analysis underestimates neurological deficits in complex spinal deformity surgery: a Scoli-RISK-1 Study. J Neurosurg Spine. 2017;27(1):68–73.
- 38↑
Amara D, Mummaneni PV, Ames CP, Tay B, Deviren V, Burch S, et al. Treatment of only the fractional curve for radiculopathy in adult scoliosis: comparison to lower thoracic and upper thoracic fusions. J Neurosurg Spine. 2019;30(4):506–514.
- 39↑
Wang MY, Uribe J, Mummaneni PV, Tran S, Brusko GD, Park P, et al. Minimally invasive spinal deformity surgery: analysis of patients who fail to reach minimal clinically important difference. World Neurosurg. 2020;137:e499–e505.
- 40↑
Hamilton DK, Kanter AS, Bolinger BD, Mundis GM Jr, Nguyen S, Mummaneni PV, et al. Reoperation rates in minimally invasive, hybrid and open surgical treatment for adult spinal deformity with minimum 2-year follow-up. Eur Spine J. 2016;25(8):2605–2611.
