Khoi D. Than, Paul Park, Kai-Ming Fu, Stacie Nguyen, Michael Y. Wang, Dean Chou, Pierce D. Nunley, Neel Anand, Richard G. Fessler, Christopher I. Shaffrey, Shay Bess, Behrooz A. Akbarnia, Vedat Deviren, Juan S. Uribe, Frank La Marca, Adam S. Kanter, David O. Okonkwo, Gregory M. Mundis Jr., Praveen V. Mummaneni and the International Spine Study Group
Minimally invasive surgery (MIS) techniques are increasingly used to treat adult spinal deformity. However, standard minimally invasive spinal deformity techniques have a more limited ability to restore sagittal balance and match the pelvic incidence–lumbar lordosis (PI-LL) than traditional open surgery. This study sought to compare “best” versus “worst” outcomes of MIS to identify variables that may predispose patients to postoperative success.
A retrospective review of minimally invasive spinal deformity surgery cases was performed to identify parameters in the 20% of patients who had the greatest improvement in Oswestry Disability Index (ODI) scores versus those in the 20% of patients who had the least improvement in ODI scores at 2 years' follow-up.
One hundred four patients met the inclusion criteria, and the top 20% of patients in terms of ODI improvement at 2 years (best group, 22 patients) were compared with the bottom 20% (worst group, 21 patients). There were no statistically significant differences in age, body mass index, pre- and postoperative Cobb angles, pelvic tilt, pelvic incidence, levels fused, operating room time, and blood loss between the best and worst groups. However, the mean preoperative ODI score was significantly higher (worse disability) at baseline in the group that had the greatest improvement in ODI score (58.2 vs 39.7, p < 0.001). There was no difference in preoperative PI-LL mismatch (12.8° best vs 19.5° worst, p = 0.298). The best group had significantly less postoperative sagittal vertical axis (SVA; 3.4 vs 6.9 cm, p = 0.043) and postoperative PI-LL mismatch (10.4° vs 19.4°, p = 0.027) than the worst group. The best group also had better postoperative visual analog scale back and leg pain scores (p = 0.001 and p = 0.046, respectively).
The authors recommend that spinal deformity surgeons using MIS techniques focus on correcting a patient's PI-LL mismatch to within 10° and restoring SVA to < 5 cm. Restoration of these parameters seems to impact which patients will attain the greatest degree of improvement in ODI outcomes, while the spines of patients who do the worst are not appropriately corrected and may be fused into a fixed sagittal plane deformity.
Raqeeb M. Haque, Gregory M. Mundis Jr., Yousef Ahmed, Tarek Y. El Ahmadieh, Michael Y. Wang, Praveen V. Mummaneni, Juan S. Uribe, David O. Okonkwo, Robert K. Eastlack, Neel Anand, Adam S. Kanter, Frank La Marca, Behrooz A. Akbarnia, Paul Park, Virginie Lafage, Jamie S. Terran, Christopher I. Shaffrey, Eric Klineberg, Vedat Deviren and Richard G. Fessler
Various surgical approaches, including open, minimally invasive, and hybrid techniques, have gained momentum in the management of adult spinal deformity. However, few data exist on the radiographic outcomes of different surgical techniques. The objective of this study was to compare the radiographic and clinical outcomes of the surgical techniques used in the treatment of adult spinal deformity.
The authors conducted a retrospective review of two adult spinal deformity patient databases, a prospective open surgery database and a retrospective minimally invasive surgery (MIS) and hybrid surgery database. The time frame of enrollment in this study was from 2007 to 2012. Spinal deformity patients were stratified into 3 surgery groups: MIS, hybrid surgery, and open surgery. The following pre- and postoperative radiographic parameters were assessed: lumbar major Cobb angle, lumbar lordosis, pelvic incidence minus lumbar lordosis (PI−LL), sagittal vertical axis, and pelvic tilt. Scores on the Oswestry Disability Index (ODI) and a visual analog scale (VAS) for both back and leg pain were also obtained from each patient.
Of the 234 patients with adult spinal deformity, 184 patients had pre- and postoperative radiographs and were thus included in the study (MIS, n = 42; hybrid, n = 33; open, n = 109). Patients were a mean of 61.7 years old and had a mean body mass index of 26.9 kg/m2. Regarding radiographic outcomes, the MIS group maintained a significantly smaller mean lumbar Cobb angle (13.1°) after surgery compared with the open group (20.4°, p = 0.002), while the hybrid group had a significantly larger lumbar curve correction (26.6°) compared with the MIS group (18.8°, p = 0.045). The mean change in the PI−LL was larger for the hybrid group (20.6°) compared with the open (10.2°, p = 0.023) and MIS groups (5.5°, p = 0.003). The mean sagittal vertical axis correction was greater for the open group (25 mm) compared with the MIS group (≤ 1 mm, p = 0.008). Patients in the open group had a significantly larger postoperative thoracic kyphosis (41.45°) compared with the MIS patients (33.5°, p = 0.005). There were no significant differences between groups in terms of pre- and postoperative mean ODI and VAS scores at the 1-year follow-up. However, patients in the MIS group had much lower estimated blood loss and transfusion rates compared with patients in the hybrid or open groups (p < 0.001). Operating room time was significantly longer with the hybrid group compared with the MIS and open groups (p < 0.001). Major complications occurred in 14% of patients in the MIS group, 14% in the hybrid group, and 45% in the open group (p = 0.032).
This study provides valuable baseline characteristics of radiographic parameters among 3 different surgical techniques used in the treatment of adult spinal deformity. Each technique has advantages, but much like any surgical technique, the positive and negative elements must be considered when tailoring a treatment to a patient. Minimally invasive surgical techniques can result in clinical outcomes at 1 year comparable to those obtained from hybrid and open surgical techniques.
Juan S. Uribe, Donald A. Smith, Elias Dakwar, Ali A. Baaj, Gregory M. Mundis, Alexander W. L. Turner, G. Bryan Cornwall and Behrooz A. Akbarnia
In the surgical treatment of spinal deformities, the importance of restoring lumbar lordosis is well recognized. Smith-Petersen osteotomies (SPOs) yield approximately 10° of lordosis per level, whereas pedicle subtraction osteotomies result in as much as 30° increased lumbar lordosis. Recently, selective release of the anterior longitudinal ligament (ALL) and placement of lordotic interbody grafts using the minimally invasive lateral retroperitoneal transpsoas approach (XLIF) has been performed as an attempt to increase lumbar lordosis while avoiding the morbidity of osteotomy. The objective of the present study was to measure the effect of the selective release of the ALL and varying degrees of lordotic implants placed using the XLIF approach on segmental lumbar lordosis in cadaveric specimens between L-1 and L-5.
Nine adult fresh-frozen cadaveric specimens were placed in the lateral decubitus position. Lateral radiographs were obtained at baseline and after 4 interventions at each level as follows: 1) placement of a standard 10° lordotic cage, 2) ALL release and placement of a 10° lordotic cage, 3) ALL release and placement of a 20° lordotic cage, and 4) ALL release and placement of a 30° lordotic cage. All four cages were implanted sequentially at each interbody level between L-1 and L-5. Before and after each intervention, segmental lumbar lordosis was measured in all specimens at each interbody level between L-1 and L-5 using the Cobb method on lateral radiography.
The mean baseline segmental lordotic angles at L1–2, L2–3, L3–4, and L4–5 were –3.8°, 3.8°, 7.8°, and 22.6°, respectively. The mean lumbar lordosis was 29.4°. Compared with baseline, the mean postimplantation increase in segmental lordosis in all levels combined was 0.9° in Intervention 1 (10° cage without ALL release); 4.1° in Intervention 2 (ALL release with 10° cage); 9.5° in Intervention 3 (ALL release with 20° cage); and 11.6° in Intervention 4 (ALL release with 30° cage). Foraminal height in the same sequence of conditions increased by 6.3%, 4.6%, 8.8% and 10.4%, respectively, while central disc height increased by 16.1%, 22.3%, 52.0% and 66.7%, respectively. Following ALL release and placement of lordotic cages at all 4 lumbar levels, the average global lumbar lordosis increase from preoperative lordosis was 3.2° using 10° cages, 12.0° using 20° cages, and 20.3° using 30° cages. Global lumbar lordosis with the cages at 4 levels exhibited a negative correlation with preoperative global lordosis (10°, R = −0.756; 20°, −0.730; and 30°, R = −0.437).
Combined ALL release and placement of increasingly lordotic lateral interbody cages leads to progressive gains in segmental lordosis in the lumbar spine. Mean global lumbar lordosis similarly increased with increasingly lordotic cages, although the effect with a single cage could not be evaluated. Greater global lordosis was achieved with smaller preoperative lordosis. The mean maximum increase in segmental lordosis of 11.6° followed ALL release and placement of the 30° cage.