Short-term outcomes of stage II and III Lenke-Silva degenerative scoliosis in osteoporotic patients treated with augmented pedicle screws: A prospective case series study

INTRODUCTION

Adult degenerative scoliosis (ADS) involves degenerative changes in the intervertebral discs and facet joints, resulting in a progressive spinal deformity that affects the sagittal, coronal, and axial planes of the spine.^[1] Patients with ADS commonly present with symptoms indicative of spinal stenosis, including exacerbated back pain, postural imbalance, radiculopathy, or a combination of these clinical manifestations.^[2]

Formulating an appropriate surgical plan requires consideration not only of the patient’s symptoms and clinical findings but also of multiple additional factors. Surgical decision-making is influenced by age, medical comorbidities, prior surgical interventions, as well as social, environmental, psychological, and overall life expectancy considerations.^[3]

Patients with ADS constitute a distinct subset of elderly individuals often burdened with multiple medical comorbidities. In such cases, extensive fusion procedures may unnecessarily elevate the risk of perioperative complications, including significant blood loss and fixation-related issues. Management guided by the Lenke and Silva classification emphasizes the importance of correlating pain distribution with the anatomical location of the scoliotic curve. In many instances, symptoms can be effectively addressed through short-segment decompression and fusion confined to the superior and inferior vertebral margins of the affected curvature.^[4]

In Level II Lenke-Silva classification, management involves limited instrumentation restricted to the decompression zone in patients exhibiting the aforementioned symptoms – typically those requiring extensive decompression – with scoliotic curves <30°, subluxation exceeding 2 mm, and absence of anterior osteophytes at the decompression site. Importantly, candidates for this approach should not exhibit significant back pain, deformity-related symptoms, or thoracic hyperkyphosis, and must be relatively well-balanced overall.^[1]

In the Level III Lenke-Silva classification, instrumented fusion encompasses the entire lumbar curve along with the required decompression levels when primary back pain is attributed to the underlying spinal deformity. These cases typically involve curves exceeding 45°, subluxation >2 mm, and the absence of anterior osteophytes within the operative region, despite acceptable coronal and sagittal alignment.^[1]

This study aimed to assess the short-term clinical and radiological outcomes of stage II and III Lenke-Silva classified degenerative scoliosis in osteoporotic patients managed with augmented pedicle screw fixation.

The authors hypothesize that short-segment fixation with the augmented screws technique in the management of stage II and III Lenke-Silva degenerative scoliosis has shorter operative time, perioperative complications, especially implant failure, blood loss, wound infection, less surgical trauma, and a shorter hospital stay duration.

MATERIALS AND METHODS

This prospective case series study was carried out at our department between January 2021 and January 2024. Participants were eligible for inclusion in this study if they met the following criteria: Adult patients aged 55 years or older diagnosed with ADS, classified as stage II or III according to the Lenke-Silva system. Eligible participants demonstrated a coronal Cobb angle >10°, significant lumbar spine degeneration, lumbar spinal stenotic claudication, radiating lower limb pain, and chronic back pain. All included patients had confirmed osteoporosis with a T-score ≤−2.5 standard deviation and presented with medical comorbidities such as diabetes mellitus and ischemic heart disease. In addition, all cases exhibited a clear clinical correlation and documented failure of adequate conservative management. Exclusion criteria encompassed revision spinal surgeries, spinal tumors, and current or prior spinal infections.

Forty-six patients were included in the study. Each patient underwent a thorough clinical evaluation, including a comprehensive medical history and a detailed physical examination. Diagnostic imaging consisted of plain radiographs of the lumbosacral spine in both static and dynamic views, dual-energy X-ray absorptiometry (DEXA) scans, and magnetic resonance imaging (MRI) of the lumbosacral region, all of which were assessed in correlation with the clinical findings to confirm the diagnosis. Pre-operative assessment included the visual analog scale (VAS)^[5] for both back and leg pain, as well as the Arabic version of the Oswestry disability index (ODI)^[6] questionnaire for low back pain, administered to patients who fulfilled the study’s inclusion criteria.

All participants were hospitalized 1 day before the scheduled surgical procedure. Following our hospital’s infection control protocol, a single dose of intravenous Cefazolin (1 g) was administered 30–60 minutes before the surgical procedure.

All patients underwent general anesthesia and were positioned in the prone position. A standard posterior midline approach was utilized in all cases. The fenestrated screws were inserted into the pedicles of the upper and lower vertebrae for the fixation in some cases. In the concave side, in the other cases, and other solid screws were inserted in all other pedicles. To minimize the risk of cement leakage into the spinal canal, careful verification of the screw length and the orientation of the fenestrated screw holes – ensuring they are positioned as far anteriorly as possible from the posterior vertebral wall – was essential. Anteroposterior (AP) and lateral intraoperative imaging were employed to confirm accurate screw placement [Figure 1].

Close

Figure 1:

Anteroposterior image to check the proper screws position.

Export to PPT

Decompression was performed according to the stenotic levels. Complete laminectomy and wide neural decompression were performed in cases presenting with radicular symptoms and neurogenic claudication, accompanied by radiological evidence of neural element compression on MRI. Because osteophytes develop more in the concavity of any curvature, this was taken into consideration if curve correction was tried, as distraction of the concavity side in the presence of those osteophytes might tether nerve roots in the foramina.

Fenestrated adapters were connected to the fenestrated screws. Common vertebroplasty cement is used. After the cement is mixed, the bone filler device is loaded immediately after checking the viscosity of the cement. The bone filler device and the plunger portion are inserted into the outer sheath, and cement is injected by depressing the plunger until the desired volume of cement is delivered. Injection for each fenestrated screw is repeated. The adapter is left in place until the cement is fully cured to avoid pulling cement into the saddle of the screw. The volume of polymethylmethacrylate (PMMA) injected per screw ranged from 1.5 to 3 cc. Cement injection was performed under continuous lateral fluoroscopic guidance using an image intensifier, with subsequent verification of cement distribution through AP and lateral fluoroscopic projections [Figure 2].

Close

Figure 2:

Cement injection under image intensifier visualization.

Export to PPT

Correction of the scoliosis is achieved by applying compression to the convex side and distraction to the concave side. Osteophytes are more at the concavity; be sure to decompress roots, especially at the concavity, before any correction. Rigid curves require mobilization, especially through multiple facet osteotomies; additional osteotomies may be necessary if there are connecting anterior osteophytes.

Cages were inserted at unstable levels following decompression, particularly in segments demonstrating significant lateral translation or in L5/S1 levels requiring sacral fixation. The cage cavities were packed with autogenous bone fragments to promote fusion. Posterolateral fusion was concurrently performed by placing bone grafts posterior to the transverse processes and adjacent to the fixation rods. A Redivac drain was routinely applied in all cases and maintained for 48 h postoperatively.

Intraoperative parameters, including operative duration, radiation exposure time, estimated blood loss, and any complications, such as instrument malfunction, dural breaches, or injury to neural elements, were meticulously recorded.

Post-operative evaluation of the patient’s neurological status. Immediate post-operative lumbosacral spine (AP and lateral) standing radiographs, including hip joint, were performed for all patients. Lumbopelvic parameters are calculated postoperatively. Post-operative treatment for all patients included oral antibiotics for 5 days, non-steroidal anti-inflammatory drugs, and muscle relaxants for 2 weeks. Treatment of osteoporosis started early after surgery using denosumab 60 mg once every 6 months, calcium supplements, and Vitamin D. Back muscle strengthening exercises were initiated gradually, starting on the 2^nd postoperative day. Patients were mobilized out of bed on the same day, aided by a lumbar support brace. Post-operative rehabilitation included unrestricted ambulation with brace protection, while activities involving sitting for more than 30 minutes and forward flexion were restricted for a duration of 6 weeks.

VAS scores for back and leg pain were recorded at the initial follow-up, 2 weeks postoperatively. Subsequent evaluations were conducted at 3, 6, 12, 18, and 24 months post-surgery, using both the VAS for back and leg pain and the ODI questionnaire to assess clinical outcomes.

Patient outcomes at the final follow-up were assessed according to the modified MacNab classification.^[7] Global outcomes were categorized into four distinct groups based on the modified MacNab criteria: “Excellent” denoted complete resolution of pain, unrestricted mobility, and a full return to pre-morbid occupational activity. “Good” described patients who experienced intermittent, non-radicular pain with substantial symptom relief and the capacity to resume modified work duties. “Fair” outcomes referred to individuals who achieved partial functional improvement yet remained limited in daily activities and/or were unemployed. “Poor” outcomes were assigned to cases showing no clinical improvement, persistent objective symptoms, or new root involvement, necessitating further surgical intervention.

Post-operative complications were recorded. Follow-up radiograph of the lumbosacral spine performed at 3, 6, 12, 18, and 24 months. A computed tomography scan of the lumbosacral spine was performed after 12 months.

The collected data were reviewed, coded, and entered into a computer for statistical analysis using the Statistical Package for the Social Sciences, version 25. Data presentation and appropriate statistical analyses were performed based on the nature of each variable.

The McNemar test was employed to evaluate the statistical significance of changes in categorical variables assessed at 2 time points within the same cohort. For continuous variables measured on more than two occasions in the same group, repeated measures analysis of variance was applied to determine statistically significant differences in mean values.

Case presentation

A 66-year-old female patient presented with back pain and bilateral lower limb pain, and claudication pain for a 6-year duration. The patient failed conservative treatment.

The patient underwent a standing lumbosacral spine radiograph, which was demonstrated with ADS, with a Cobb angle of 19.2° [Figure 3]. MRI of the lumbosacral spine showed spinal canal stenosis and degenerative lumbar spine from L2 to L5 [Figure 4]. DEXA scan with T score −2.6 (osteoporosis).

Close

Figure 3:

Pre-operative radiographs of the lumbosacral spine showing adult degenerative scoliosis with a Cobb angle of 19.5°.

Export to PPT

Close

Figure 4:

Pre-operative T2 magnetic resonance imaging showing spinal canal stenosis from L2 to L5.

Export to PPT

Posterior spinal fusion (PSF) from L2 to L5 using fenestrated screws augmented on the most cranial (L2) and caudal (L5) vertebrae with wide decompression and posterolateral and interbody fusion using interbody cages [Figures 5 and 6]. Immediate post-operative radiograph [Figure 7]. Patient has improved after surgery with great improvement of the bilateral lower limb pain; she was allowed to freely walk with a lumbosacral support orthosis on the 2^nd day of surgery and discharged from the hospital after 2 days.

Close

Figure 5:

Intraoperative lateral images showing L2 to L5 posterior spinal fusion and cement injection of upper and lower vertebrae.

Export to PPT

Close

Figure 6:

(a and b) Intraoperative anteroposterior and lateral images showing L2 to L5 posterior spinal fusion with selective upper and lower screws augmentation.

Export to PPT

Close

Figure 7:

(a and b) Immediate post-operative anteroposterior and lateral radiographs showing correction of the pre-operative Cobb angle and good cementation of the vertebrae.

Export to PPT

Follow-up at 2 weeks, 3, 6, 12, 18, and 24 months [Table 1].

Table 1: Results of pre-operative and follow-up VAS for back and leg pain and ODI scale.

	Pre-operative	2 weeks follow-up	3 months follow-up	6 months follow-up	12 months follow-up	18 months follow-up	24 months follow-up
VAS of back pain	8.5	6.2	4.8	3.6	2.1	2	1.8
VAS of leg pain	7.6	4.8	3.2	2.9	1.7	1 0.5	1.4
ODI scale	74	-	30	22	16	16	14

VAS: Visual analog scale, ODI: Oswestry disability index

The patient’s Cobb angle was corrected from 19.2° to 0.8° after surgery and remained stable for 24 months. The patient had full fusion at 6 months. The patient returned to her pre-operative level of activities after 3 months. Modified MacNab criteria were excellent. Follow-up radiographs were taken at 3, 6, 12, 18, and 24 months [Figure 8], which showed good construct position without failure of fixation or progression of the post-operative Cobb angle.

Close

Figure 8:

(a and b) 24 months follow-up radiographs of the lumbosacral spine showing good construct position and good interbody fusion.

Export to PPT

RESULTS

Of the 60 patients with ADS initially evaluated for eligibility, 46 fulfilled the inclusion criteria and were enrolled in the study [Table 2].

Patients’ medical comorbidities, pre-operative foot drop, the mean duration of the complaint, and DEXA scan T-score were recorded [Table 3].

Four cases received L1 to L5 fusion, two cases received L2 to L4 fusion, 13 cases received L2 to L5 fusion, 12 cases received L3 to L5 fusion, seven cases received L3 to S1 fusion, two cases received L4 to L5 fusion, four cases received L4 to S1 fusion, and two cases received L2 to S1 fusion [Table 4].

Pre-operative VAS scores for leg and back pain were documented and subsequently reassessed at post-operative follow-up intervals of 2 weeks, 3 months, 6 months, 12 months, 18 months, and 24 months. ODI score showed significant improvement at 3 months; follow-up values were also calculated at 6, 12, 18, and 24 months of follow-up. The mean Cobb’s angle was measured preoperatively, immediately postoperatively, and at follow-up intervals of 6, 12, 18, and 24 months [Table 5].

All patients achieved successful fusion at 6 months postoperatively, except for 3 patients who experienced delayed fusion [Table 6].

One patient developed proximal-level adjacent segment disease (ASD) that needed revision surgery. Another patient developed proximal-level adjacent segment degeneration with very mild back pain, and close follow-up was recommended. After this complication occurred, the spinopelvic parameters were revised for the postoperative radiographs of these patients, and hypolordosis and pelvic incidence (PI)-lumbar lordosis (LL) mismatch were documented, which predispose to adjacent segment degeneration. Three patients presented with proximal junctional kyphosis (PJK), one patient had mild back pain with documented very good posterolateral fusion on radiographs, and we recommended a follow-up after 6 months. Two patients needed revision surgery, but they refused the surgery. Ten screws out of 184 augmented screws have anterior cement leakage without any symptoms or complications [Table 7].

According to the modified MacNab criteria at final follow-up, a total of 89.1% of patients achieved either excellent or good clinical outcomes [Table 8].

DISCUSSION

In this study, short-segment fixation using augmented screws demonstrated favorable clinical outcomes in the management of ADS. This approach proved highly effective, as evidenced by significant improvements in VAS scores for leg and back pain, as well as in the ODI, by the 24-month follow-up. According to the modified MacNab criteria, excellent or good outcomes were achieved in 89.1% of the patients. Furthermore, in this study, the short-segment fixation had good results in ADS correction, and the postoperative Cobb’s angle was found to be markedly improved in most cases without noticeable progression for 24 months of follow-up, so it succeeds in degenerative scoliosis correction without the need for long-segment fixation, which carries a higher complication rate. In addition, in this study, very good results of fusion were obtained in all cases after 6 months, except for three cases with delayed fusion, which were fused after 12 months. The possible cause of delayed fusion in the three cases could be attributed to noncompliance with osteoporosis treatment that was prescribed postoperatively for the patients. Hence, it is very important in osteoporotic patients to emphasize the treatment of osteoporosis, even with augmented screws fixation.

A systematic review and meta-analysis by Phan et al. (2017) about outcomes of short fusion versus long fusion for ADS.^[8] The study found no statistically significant differences in terms of coronal Cobb’s angle and LL correction associated with a short or long fusion, with reduced perioperative time and costs, but also some complications.^[8]

Alvarez-Galovich et al. (2020) studied 89 patients with degenerative lumbar disease with poor bone quality.^[9] All patients underwent posterior spinal fixation with augmented screws. The authors determined that augmentation of pedicle screws with PMMA is both safe and effective, serving as a credible alternative to conventional fixation methods for enhancing spinal stability.^[9]

Guo et al. (2020) investigated the effects of selective cement augmentation of cranial and caudal pedicle screws.^[10] They concluded that this targeted approach offers stability comparable to non-selective augmentation in lumbosacral degenerative diseases, while significantly reducing the risk of cement leakage.^[10]

In comparison with similar published articles, in this study, intraoperative blood loss, operation time, and hospital stay were average, mostly due to short-segment fixation with subsequent lower morbidity. The low incidence of screws cut out could be attributed to screw augmentation.

In the current study, one patient developed a post-operative infection that was managed through debridement and retention of implants. In cases of acute early infections with intact soft-tissue coverage, a stable implant, and appropriate antibiotic therapy, implant retention may be feasible following thorough surgical debridement.^[11]

Three out of 46 patients (6.52%) developed PJK after 48 months of follow-up, which is less than the percentage in Wang et al.’s study, which reported that PJK developed in 17 of 98 patients (17.3%) with a minimum 2-year follow-up following long instrumented PSF.^[12]

By reviewing the patients who developed PJK, the three patients were severely osteoporotic according to DEXA scan, with non-compliance with osteoporosis treatment after surgery. This predisposed them to an increased risk of PJK development, and all of them had delayed fusion for 6 months, with fusion achieved at 12 months of follow-up.

Ye et al. (2023) investigated prospectively the predictive value of global sagittal alignment and upper instrumented vertebrae level in the development of PJK among patients with adult spinal deformity.^[13] Their findings indicated that individuals who developed PJK exhibited a smaller PI-LL mismatch compared to those without PJK.^[13]

In the present study, after 12 months, 1 patient (case 3) developed recurrence of back pain and bilateral lower limbs radicular pain. This patient developed ASD that needs revision and cranial extension to D12, and another patient (case 7) developed adjacent segment degeneration with mild back pain and no radicular symptoms. When the spinopelvic parameters were reviewed in these two patients, there was a post-operative hypolordosis in both cases and a spinopelvic mismatch, which predisposed them to ASD.

Mohamed et al. (2010) conducted a study on 53 patients with degenerative spine diseases and followed them up for a mean of 35 months.^[14] The authors observed that restoration of lordotic alignment during lumbar fusion for degenerative conditions reduced the incidence of ASD. Conversely, disruption of normal lordosis was associated with increased rates of ASD.^[14]

A notable disadvantage of augmented screw fixation is the increased radiation exposure to the patient, surgeon, and operating room staff, as cement injection requires continuous fluoroscopic imaging during the procedure. This risk can be mitigated by implementing appropriate protective barriers and radiation safety measures.

In the present study, four cases had anterior cement leakage. In the first case, cement leakage was noted anteriorly in the sacral screw. Hence, it is better to avoid cementation of sacral screws, because, usually in sacral screws, we need to have anterior anchorage of the screw, or we can use a too-short fenestrated screw to avoid anterior cement leakage. Use other techniques of increasing screw strength, like bicortical screw, use a large diameter screw, or use another technique of cementation, like retrograde cementation with intact anterior wall. In the other three cases, cement leakage was noted anteriorly in a vein, and it was non-symptomatic.

To minimize the risk of cement leakage, an optimized protocol may include reducing the volume of cement injected per pedicle tract and limiting the number of augmented screws. In addition, using low- to medium-viscosity bone cement with a high barium sulfate content (30%), applying the cement at low pressure, and positioning the fenestrated screw tip near the lateral aspect of the vertebral body can further reduce the leakage risk.^[15]

The present study is subject to several limitations, including a limited sample size, a relatively short follow-up duration, a lack of spinopelvic parameters, and the absence of a control group. In addition, the use of DEXA scanning at the lumbar spine represents a methodological constraint, as bone mineral density measurements are influenced by factors associated with deformity, such as degenerative and sclerotic changes, vertebral rotation, and vertebral compression.

CONCLUSION

Short-segment fusion combined with decompression, utilizing PMMA-augmented pedicle screws, has demonstrated highly favorable clinical outcomes in the treatment of stage II and III ADS as classified by Lenke and Silva in osteoporotic patients. This approach resulted in significant improvements in VAS scores for both back and leg pain, as well as in ODI scores. Radiographic outcomes revealed satisfactory correction of the Cobb angle and high fusion rates. They maintained screw and construct stability at 24-month follow-up, with a notably lower incidence of pedicle screw loosening and revision surgeries.