Abstract
Objective
In frail elderly patients with Kümmell disease or osteoporotic burst fractures, cement augmentation relieves pain and partially restores height but achieves neither durable load-bearing support nor biological integration of the anterior column, whereas corpectomy carries substantial morbidity. We evaluated transpedicular placement of a three-dimensional (3D)-printed porous titanium cage for anterior column reconstruction.
Methods
We retrospectively reviewed six consecutive patients (mean age 69.3 years; 2 men, 4 women) who underwent transpedicular intrabody insertion of a 3D-printed porous titanium cage with posterior fixation for osteoporotic vertebral collapse. Three had Kümmell disease and three had osteoporotic burst fractures, including one proximal junctional failure. Back and leg pain (visual analog scale, VAS), disability (Oswestry Disability Index, ODI), and segmental angle were assessed.
Results
At a mean follow-up of 19.0 months (range, 12–36 months), mean back-pain VAS improved from 6.8 to 2.3 and ODI from 69.3% to 26.3%. Mean segmental kyphosis was corrected from 36.8° to 11.9° immediately after surgery and maintained at 15.8°. All neurological deficits improved, including complete recovery of sphincter function in one patient with conus medullaris syndrome. One severely osteoporotic patient showed loss of correction from cage subsidence; no symptomatic hardware failure occurred.
Conclusion
Transpedicular 3D-printed porous titanium cage augmentation appears to be a feasible, less-invasive option for anterior column reconstruction in osteoporotic vertebral collapse; however, these preliminary findings from a small, heterogeneous series require confirmation in larger, longer studies.
Keywords: Osteonecrosis; Osteoporotic fractures; Spinal fusion; Titanium; Printing, three-dimensional; Kyphosis
GRAPHICAL ABSTRACT

INTRODUCTION
Osteoporotic vertebral compression fractures are highly prevalent, and a critical subset progresses to severe collapse, anterior column insufficiency, and kyphotic deformity.7,11) Within this spectrum, Kümmell disease—the delayed collapse of a vertebral body that develops weeks to months after an apparently trivial injury and is identified radiographically by an intravertebral cleft, or vacuum phenomenon—and acute osteoporotic burst fractures share a common problem: a deficient, poorly load-bearing anterior column that, when severe, may cause canal compromise and neurological decline.19) Although Kümmell disease originates from ischemic necrosis, its advanced stages progress to vertebral comminution similar to burst fractures. Ultimately, both conditions leave the anterior column unable to resist ventral compressive loads in osteoporotic bone.17)
Because these elderly patients are frequently frail with significant comorbidities, conventional surgical options have distinct drawbacks. Cement augmentation relieves pain but fails to provide durable load-bearing reconstruction or biological integration, potentially risking adjacent-level fractures.4,5,6,24) Stand-alone posterior screw fixation, in turn, cannot counteract the ventral compressive loads across a deficient anterior column in osteoporotic bone, so correction is frequently lost and the instrumentation may loosen.2) At the other extreme, anterior corpectomy with reconstruction restores the load-bearing column reliably but at the cost of prolonged operative time, greater blood loss, and elevated perioperative risk in older adults.14,16,21) An intermediate strategy that reconstructs the anterior column with limited surgical insult is therefore attractive.
In this context, the intravertebral cleft or fractured cavity lies between relatively intact endplates that can anchor an interbody device without sacrificing adjacent discs. Consequently, several groups have repurposed interbody cages for off-label intravertebral placement via a posterior transpedicular corridor, achieving anterior support while avoiding an anterior approach.13,23) Seating a structural cage within the cleft restores the height of the collapsed vertebral body and reconstructs the anterior column, thereby correcting the local kyphotic deformity while preserving the adjacent discs. Building on this rationale and on the favorable osteoconductive behavior of additively manufactured porous titanium,20) we report the clinical and radiographic outcomes of transpedicular placement of a 3D-printed porous titanium cage for anterior column reconstruction in six patients with osteoporotic vertebral collapse.
MATERIALS AND METHODS
Patients and study design
We retrospectively reviewed six consecutive patients who underwent transpedicular intrabody placement of a 3D-printed porous titanium cage for anterior column reconstruction of an osteoporotic thoracolumbar vertebral collapse by a single surgical team between 2022 and 2025. During this study period, a total of 9 patients underwent surgical treatment for osteoporotic vertebral collapse at our institution, of whom only these six patients met the strict anatomical and clinical criteria required for this specific technique. Candidates had progressive vertebral collapse with intravertebral instability, intractable mechanical pain, or canal compromise with neurological deficit, in whom stand-alone cement augmentation was judged insufficient and formal corpectomy was considered to carry excessive morbidity. A key prerequisite was preservation of the endplates adjacent to the collapsed cancellous bone, which provides a stable interface to anchor the cage and re-establish anterior load-sharing. Patients with active spinal infection, complete endplate destruction precluding stable cage anchorage, or pedicle anatomy unsuitable for transpedicular access were excluded. Selected revisions—e.g., junctional failure above a prior fusion needing anterior support—were also included, with the cage combined with extension of the existing construct. Bone mineral density (lowest T-score on dual-energy X-ray absorptiometry), the Thoracolumbar Injury Classification and Severity (TLICS) score, neurological status, operative details, and antiosteoporotic medication were recorded. This retrospective study was approved by the Institutional Review Board of the Korea University Anam Hospital (approval No. 2026AN0401), and the requirement for informed consent was waived owing to the retrospective design.
Surgical technique
All patients were treated through a posterior transpedicular approach. With the patient prone, the collapsed vertebra was reduced by positioning and manual pressure under fluoroscopy; a Jackson table was used preferentially, and when it was unavailable, two transverse bolsters were placed beneath the sternum and the anterior superior iliac spines to create a three-point bending effect. When canal compromise and neurological deficit were present, posterior decompression (pediculectomy or laminectomy) was performed to relieve neural pressure, and the same corridor was used for cage insertion. Necrotic or comminuted tissue within the cleft was curetted from the cranial and caudal aspects to expose bleeding endplate, analogous to endplate preparation in interbody fusion.13) To achieve a gradual and controlled vertebral height restoration, sequential trialing was performed using cage trials increasing in 1-mm increments, allowing for serial restoration of the collapsed vertebral body. Following this serial restoration process, the enlarged cleft was partially filled with allograft bone chips. The definitive cage size was meticulously selected based on the optimal trial size, pedicle diameter, and intravertebral cleft dimensions to minimize endplate violation while ensuring adequate anterior column support. Finally, a porous 3D-printed titanium cage (EIT Cellular Titanium; Emerging Implant Technologies, Wurmlingen, Germany), packed with demineralized bone matrix, was seated within the cleft to reconstruct the anterior column and correct local kyphosis; bone morphogenetic protein was not used. Posterior fixation was completed with pedicle screws without cement augmentation of the screws; a screw was placed at the fractured level through an intact pedicle when feasible to shorten the construct.15,18)
Outcome assessment
Back and leg pain were graded with a 0–10 visual analog scale (VAS) and function with the Oswestry Disability Index (ODI) preoperatively, at 1 month, and at final follow-up. On standing radiographs, the segmental (sagittal) angle of the index level (a positive value denotes kyphosis), lumbar lordosis (LL), sagittal vertical axis, and pelvic incidence-LL mismatch were measured. The surgical correction angle was defined as the difference between the preoperative and immediate postoperative segmental angles, and the loss of correction as the difference between the final and immediate postoperative angles. Osseous integration was assessed on computed tomography (CT) when available.
Statistical analysis
Given the small sample, data are summarized descriptively as mean ± standard deviation with ranges; no inferential statistical testing was performed.
RESULTS
Patient and operative characteristics
The cohort comprised two men and four women with a mean age of 69.3±11.4 years (range, 57–87). Three patients had Kümmell disease and three had osteoporotic burst fractures, one of which was a proximal junctional failure (PJF) above a previous lumbar fusion; the index level was T12 in two, L1 in two, and L2 in two patients. Three patients were osteoporotic (T-score ≤−2.5) and three osteopenic. A porous 3D-printed titanium cage was used in all patients (EIT Cellular Titanium), and posterior decompression by pediculectomy was performed in three. The mean follow-up was 19.0 months (range, 12–36 months), and all patients reached at least 12 months of follow-up. Demographic, operative, and clinical data are summarized in TABLE 1. One patient received an anabolic agent (romosozumab; case 4), three an antiresorptive alone (cases 1, 2, 5), and two none (cases 3, 6) (TABLE 1).
TABLE 1. Demographic, operative, and clinical characteristics of the six patients.
| No. | Age/sex | Diagnosis (TLICS) | Level | BMD (T) | Anti-osteoporosis drugs | Procedure | Cage | Neurological status (pre → post) | F/U (month) |
|---|---|---|---|---|---|---|---|---|---|
| 1 | 87/M | Burst fx + Kümmell (4) | L2 | −2.9 | Denosumab | Transpedicular intrabody cage + PSF L1–3 | EIT | Rt HF/KE G4 → recovered | 12 |
| 2 | 62/F | Kümmell (3) | T12 | −2.2 | Risedronate | Pediculectomy + transpedicular cage + PSF T10–L2 | EIT | Intact | 36 |
| 3 | 79/M | Burst fx (5) | T12 | −1.6 | None | Bilateral pediculectomy + transpedicular cage + PSF T9–L2 | EIT | Conus syndrome → full recovery* | 12* |
| 4 | 68/F | Burst fx + PJF (4) | L2 | −1.4 | Romosozumab → denosumab → zoledronate | Instrumentation removal L2–5 + transpedicular cage + rod ext T10–L5 | EIT | HF/KE G3 → improved | 30 |
| 5 | 63/F | Kümmell (5) | L1 | −2.8 | Denosumab | Pediculectomy + transpedicular cage + PPSF T11–L3 | EIT | HF/KE G3, incontinence → improved | 12 |
| 6 | 57/F | Kümmell (4) | L1 | −3.2 | None | Transpedicular cage + PSF T11–L3 | EIT | Leg paresthesia → improved | 12 |
TLICS: Thoracolumbar Injury Classification and Severity, BMD: bone mineral density, F/U: follow-up, M: male, PSF: posterior screw fixation, EIT: EIT Cellular Titanium 3D-printed porous titanium cage, HF: hip flexor, KE: knee extensor, PJF: proximal junctional failure, PPSF: percutaneous pedicle screw fixation.
*Follow-up censored at death (unrelated comorbidity at 12 months; last radiographic follow-up at 6 months).
Clinical outcomes
All six patients tolerated surgery without intraoperative neurological injury. Mean back-pain VAS improved from 6.8±2.1 preoperatively to 2.3±2.4 at 1 month and 2.3±2.2 at final follow-up; leg-pain VAS from 4.5±3.7 to 2.2±2.9 and 2.2±2.4; and ODI from 69.3±19.0% to 32.7±22.4% and 26.3±11.6% (TABLE 2). Every patient with a preoperative neurological deficit improved, and the patient with conus medullaris syndrome regained independent ambulation with complete recovery of sphincter function.
TABLE 2. Sequential clinical outcomes (n=6).
| Outcome metric | Preoperative | Postoperative 1 month | Final follow-up |
|---|---|---|---|
| VAS, back | 6.8±2.1 | 2.3±2.4 | 2.3±2.2 |
| VAS, leg | 4.5±3.7 | 2.2±2.9 | 2.2±2.4 |
| ODI (%) | 69.3±19.0 | 32.7±22.4 | 26.3±11.6 |
Values are presented as mean ± standard deviation.
VAS: visual analog scale, ODI: Oswestry Disability Index.
Radiographic outcomes
Mean segmental kyphosis was 36.8° before surgery, 11.9° immediately afterward, and 15.8° at final follow-up, corresponding to a mean surgical correction of 24.9° and a mean loss of correction of 3.9° (net maintained correction, 20.9°). Per-patient values and sagittal parameters are presented in TABLE 3 and FIGURE 1. Where follow-up CT was available (four patients), osseous bridging was observed around the cage with preservation of the adjacent endplates and discs.
TABLE 3. Radiographic parameters (degrees; a positive value denotes kyphosis).
| No. | SA pre | SA postop | SA final | Correction (pre−postop) | Loss (final−postop) | LL (pre → final) | SVA final (mm) | PI−LL final |
|---|---|---|---|---|---|---|---|---|
| 1 | 26.9 | 8.1 | 25.3 | 18.8 | 17.2 | 21.7 → 2.1 | 98 | 36.0 |
| 2 | 38.4 | −2.2 | 7.7 | 40.5 | 9.9 | 79.0 → 58.1 | −19 | 7.5 |
| 3 | 37.6 | 21.2 | 16.6* | 16.4 | −4.6 | 42.7 → 31.7 | 197 | 19.4 |
| 4 | 27.3 | 6.2 | 7.3 | 21.1 | 1.0 | 17.2 → 37.7 | 21 | 19.8 |
| 5 | 41.6 | 19.5 | 16.2 | 22.1 | −3.3 | −14.2 → 11.7 | 129 | 27.8 |
| 6 | 49.0 | 18.6 | 22.0 | 30.4 | 3.4 | 25.9 → 26.1 | 21 | 33.2 |
| Mean | 36.8 | 11.9 | 15.8 | 24.9 | 3.9 | - | - | - |
The surgical correction angle is the preoperative minus the immediate postoperative SA; the loss of correction is the final minus the immediate postoperative SA (a positive value denotes recurrent kyphosis, a negative value further correction). A negative SA denotes segmental lordosis.
SA: segmental angle, LL: lumbar lordosis, SVA: sagittal vertical axis, PI−LL: pelvic incidence–lumbar lordosis mismatch.
*Final radiographic follow-up at 6 months.
FIGURE 1. Serial changes in the segmental (sagittal) angle for each patient across the preoperative, immediate postoperative, and final follow-up time points. A positive value denotes kyphosis and a negative value lordosis. All vertebrae were corrected immediately after surgery; the correction was well maintained in most cases, whereas case 1 showed near-complete recurrence of kyphosis (8.1° to 25.3°) consistent with cage subsidence.

Cases and complications
Case 1. An 87-year-old man (T-score −2.9) with an L2 burst fracture and Kümmell-type collapse (TLICS 4) and right hip-flexor/knee-extensor weakness underwent right transpedicular intrabody cage insertion with fixation L1–L3. Pain and weakness improved (back VAS 3 → 1; ODI 66 → 24%), but the segmental kyphosis corrected from 26.9° to 8.1° recurred to 25.3° by 12 months owing to cage subsidence (FIGURE 2).
FIGURE 2. The only case that lost the initially achieved correction (case 1). (A) Preoperative lateral radiograph showing 26.9° of segmental (L1–L3) kyphosis due to an L2 osteoporotic burst fracture with Kümmell-type collapse. (B) Immediate postoperative radiograph showing correction of the kyphosis to 8.1°. (C) At 1-year follow-up, the segmental kyphosis had recurred to 25.3°, consistent with cage subsidence.

Case 2. A 62-year-old woman (T-score −2.2) with T12 Kümmell disease (TLICS 3) underwent right pediculectomy, transpedicular intrabody cage insertion, and fixation T10–L2. Kyphosis improved from 38.4° to slight lordosis and was maintained at 7.7° at 36 months (back VAS 6 → 2; ODI 36 → 14%; FIGURE 3).
FIGURE 3. Durable correction in T12 Kümmell disease (Case 2). (A) Preoperative radiograph of T12 Kümmell disease showing compensatory lumbar hyperlordosis, with 38.4° of segmental kyphosis measured at T11–L1. (B) Radiograph at 1-year follow-up. (C) Radiograph at 3-year follow-up; the T11–L1 segmental kyphosis was corrected to approximately 7.7° and remained well maintained.

Case 3. A 79-year-old man (T-score −1.6) with a T12 burst fracture (TLICS 5) and conus medullaris syndrome underwent bilateral pediculectomy for decompression, transpedicular intrabody cage reconstruction, and fixation T9–L2. Sphincter function recovered completely and ambulation was restored by 6 months (back VAS 7 → 0; ODI 78 → 16%; kyphosis 37.6° → 16.6°). He died at 12 months from an unrelated medical comorbidity.
Case 4. A 68-year-old woman (T-score −1.4) with an L2 burst fracture and PJF at L1/2 above a prior oblique lumbar interbody fusion (TLICS 4) underwent removal of the instrumentation, L2 transpedicular intrabody cage reconstruction, and rod extension T10–L5. Kyphosis was corrected from 27.3° to 6.2° and maintained at 7.3° at 30 months (ODI 92 → 36%; FIGURE 4).
FIGURE 4. A case of PJF treated with intrabody cage augmentation to provide anterior support and correct the kyphosis. (A) Radiograph before the prior OLIF. (B) After the OLIF, a hyperacute PJF developed at the L2 body with severe, intractable back pain. (C) Radiograph at 1-year follow-up after L2 vertebral body reconstruction, showing the corrected kyphosis; the back pain resolved thereafter.

PJF: proximal junctional failure, OLIF: oblique lumbar interbody fusion.
Case 5. A 63-year-old woman (T-score −2.8) with L1 Kümmell disease (TLICS 5), gait disturbance, and urinary incontinence underwent left pediculectomy, L1 transpedicular intrabody cage insertion, and percutaneous fixation T11–L3. Gait and incontinence improved markedly (kyphosis 41.6° → 16.2°; ODI 78 → 44%). The early course was complicated by an epidural hematoma requiring evacuation, with uneventful recovery thereafter.
Case 6. A 57-year-old woman (T-score −3.2) with L1 Kümmell disease (TLICS 4) and bilateral leg paresthesia but intact strength underwent transpedicular intrabody cage insertion with fixation T11–L3. Back pain and disability improved (back VAS 8 → 1; ODI 66 → 24%), with kyphosis corrected from 49.0° to 18.6° and mild settling to 22.0°.
No hardware failure or cage migration occurred; the main adverse radiographic event was subsidence in case 1. Mean operative time was 217.5 minutes and mean blood loss 683 mL; two patients were transfused. Two had complications—pneumonia with urinary tract infection (case 3, resolved) and an epidural hematoma needing evacuation (case 5). No anterior-approach or hardware complication occurred (TABLE 4).
TABLE 4. Operative data and perioperative morbidity of the six patients.
| No. | OP time (minutes) | EBL (mL) | Transfusion | Complication |
|---|---|---|---|---|
| 1 | 185 | 1,500 | RBC 2 U, FFP 2 U | None |
| 2 | 185 | 300 | None | None |
| 3 | 290 | 500 | None | Pneumonia, UTI |
| 4 | 240 | 800 | RBC 1 U | None |
| 5 | 160 | 300 | None | Epidural hematoma → evacuation |
| 6 | 245 | 700 | None | None |
| Mean | 217.5 | 683 | 2/6 | - |
OP: operation, EBL: estimated blood loss, RBC: red blood cell, U: unit, FFP: fresh frozen plasma, UTI: urinary tract infection.
DISCUSSION
In elderly patients with osteoporotic vertebral collapse, treatment must reconcile three competing aims—correcting deformity, protecting neural elements, and limiting surgical insult to a fragile host. Transpedicular intrabody cage augmentation occupies a middle ground that captures much of the mechanical benefit of anterior reconstruction while retaining the lower morbidity of a posterior-only operation.21) By using the intravertebral cleft as a ready-made cavity, the surgeon can restore vertebral height, correct local kyphosis, and re-establish anterior load-sharing without performing a corpectomy or sacrificing the adjacent discs.12,13) Our six-patient experience is consistent with this rationale: pain and disability improved substantially, every neurological deficit recovered, and most of the correction obtained at surgery was preserved at a mean of 19.0 months. These observations, however, remain preliminary and should be interpreted within the constraints discussed below. The posterior-only exposure avoids the thoracotomy or retroperitoneal approach of anterior reconstruction, which carries greater blood loss and up to one-third respiratory complications.22,27) Still, our series had real morbidity (one 1,500-mL loss, two transfusions, two complications); the advantage lies in avoiding the anterior approach, not in eliminating perioperative risk.
A biologically active cage versus cement
Bone cement-augmented short-segment fixation can relieve pain and partially restore height, but polymethylmethacrylate does not integrate biologically and is prone to stress concentration and osteolysis at the cement–bone interface, which may lead to delayed collapse or screw loosening in severely porotic bone and sometimes to revision surgery.4,14,24) A cage filled with osteoconductive graft instead provides immediate mechanical support while serving as a scaffold for bone ingrowth.9) The interconnected porosity of additively manufactured titanium approximates native trabecular architecture and, unlike the hydrophobic surface of polyetheretherketone, promotes cellular attachment and vascular ingrowth—an advantage in the poorly vascularized milieu of a necrotic Kümmell body.1,20,26) In our series, follow-up CT showed osseous bridging around the cage with preserved adjacent discs, supporting maturation into a biologically active fusion mass rather than an inert spacer (FIGURE 5). Anabolic agents may aid ingrowth into the porous cage. The only patient given one (case 4) maintained correction, whereas the subsided patient (case 1) received only an antiresorptive; still, two maintained correction without any drug, so this is only hypothesis-generating.
FIGURE 5. Osseointegration of the porous titanium cage on CT. (A, B) Preoperative CT of Kümmell disease showing a collapsed vertebral body with an intravertebral vacuum cleft and accompanying segmental kyphosis: (A) sagittal and (B) coronal images. (C, D) One-year postoperative CT showing osteoconduction within and around the porous three-dimensional-printed titanium cage, with physiological osseous integration and reconstruction of the vertebral body: (C) sagittal and (D) coronal images.

CT: computed tomography.
Decompression through a posterior corridor
Four patients had motor or sphincteric deficits, including one with conus medullaris syndrome and sphincter dysfunction, and all recovered. Restoring anterior height tensions the posterior wall and posterior longitudinal ligament, producing indirect reduction of retropulsed fragments (ligamentotaxis); combined with direct posterior decompression where indicated, this allowed neurological recovery without an anterior approach.8,10,13,25) The complete return of sphincter function in our patient with conus syndrome illustrates that meaningful recovery is achievable through a single posterior corridor even in advanced collapse.
Construct length and the role of fusion
Whether posterolateral fusion is required in addition to screw fixation remains debated. In acute trauma, percutaneous fixation alone may provide short-term stability,3) but in Kümmell disease and osteoporotic collapse with pseudarthrosis, screws acting without anterior support are prone to recurrent kyphosis and loosening.2) When a cage reconstitutes the anterior column, posterior fixation mainly provides temporary stability until the index body consolidates, so percutaneous fixation without formal fusion may suffice in selected patients. Placing a screw at the fractured level through an intact pedicle is a practical way to shorten the construct while preserving stability,15,18) although our heterogeneous fixation lengths reflect the absence of consensus. Fusion remains prudent when endplates are fragile, osteoporosis is severe, or adjacent segments are degenerate.
Loss of correction and subsidence
Our results were not uniform. In case 1, the earliest case, in a patient with a T-score of −2.9, nearly all of the initial correction was lost by 12 months through subsidence. This mirrors the principal limitations of the technique: pedicle-diameter constraints on cage size and the risk of endplate violation and subsidence in porotic bone. Meticulous endplate preparation, careful cage sizing, and expandable cages that can be delivered through a narrow pedicle and expanded in situ are the main strategies to mitigate this risk.26) The contrast between case 1 and the well-maintained corrections in cases 2 and 4 suggests that patient selection and technical refinement, rather than the concept itself, govern durability. Bone quality alone does not explain this: case 6 (T-score −3.2) fared better. Case 1 was the only short-segment construct (one level above and below); such constructs concentrate load in porotic bone and invite subsidence. We therefore favor long-segment fixation at the thoracolumbar junction and in severe osteoporosis.
Extension beyond Kümmell disease and future directions
The same principle—anterior column support when corpectomy is too morbid—applies to osteoporotic burst fractures and, potentially, to osteolytic or pathological collapse.3) Half of our patients had burst fractures rather than classic Kümmell disease, including a PJF salvaged by intrabody cage reconstruction and rod extension, suggesting that transpedicular cage augmentation is versatile across the spectrum of anterior column insufficiency in fragile bone. Further refinement with percutaneous or endoscope-assisted cage delivery may reduce soft-tissue disruption and broaden applicability.23) Therefore, this technique may be particularly useful in patients who are not ideal candidates for corpectomy due to frailty but in whom cement augmentation alone is insufficient to provide durable anterior column support. Case 4, however, was a revision for PJF—distinct from the primary collapses—and its revision context should be kept in mind when interpreting its outcome.
Limitations
This is a small, retrospective, single-arm series without a control group, and the cohort is heterogeneous in the underlying pathology, level, fixation length, and antiosteoporotic regimen. Although the cohort is small, it comprised all consecutive eligible patients treated with this technique during the study period—including the cases complicated by cage subsidence and by unrelated death—which limits selection bias; the modest number reflects the narrow, anatomically defined indication for the procedure and is consistent with comparable published series.13,26) All patients reached at least 12 months of follow-up (range, 12–36 months), which meets or exceeds the minimum follow-up of comparable published case series;26) nevertheless, 12 months remains relatively short for confirming the long-term durability of a load-bearing reconstruction. One patient died of an unrelated cause at 12 months, with radiographic follow-up available to 6 months. These factors preclude firm conclusions about comparative efficacy, and the findings should be regarded as preliminary and hypothesis-generating; larger prospective comparative studies with standardized constructs and longer follow-up are warranted.
CONCLUSION
Transpedicular placement of a 3D-printed porous titanium cage provided pain relief, neurological recovery, and maintained kyphosis correction in most of these elderly patients with osteoporotic vertebral collapse, with the lower access-related morbidity of a posterior-only approach. These findings are preliminary, drawn from a small and heterogeneous series with relatively short follow-up, and the loss of correction from cage subsidence in one severely osteoporotic patient underscores the importance of meticulous technique and patient selection. This procedure appears to be a promising intermediate option between the two conventional extremes, but prospective comparative studies with longer follow-up are required to define its definitive role.
ACKNOWLEDGMENTS
No benefits in any form have been or will be received from a commercial party directly or indirectly related to the subject of this manuscript. The manuscript submitted contains information about a medical device, but the authors have no financial or proprietary interests in the product or the manufacturer.
Footnotes
Funding: No funding was obtained for this study.
Conflict of Interest: The authors have no financial conflicts of interest.
Informed Consent: The requirement for informed consent was waived owing to the retrospective design.
Ethics Approval: This retrospective study was approved by the Institutional Review Board of the Korea University Anam Hospital (approval No. 2026AN0401).
References
- 1.Bae J, Ang CY, Syed I, Jeong SK, Shin SH, Lee SH. Minimally invasive surgery transpedicular intrabody cage technique for the management of Kummell disease. Int J Spine Surg. 2024;18:73–80. doi: 10.14444/8570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Chen C, Gao X, Li H, Pan X, Wang S. Intravertebral insertion of interbody fusion cage via transpedicular approach for the treatment of stage III Kümmell disease: a technical note and case presentation. Br J Neurosurg. 2023;37:1909–1914. doi: 10.1080/02688697.2021.1892590. [DOI] [PubMed] [Google Scholar]
- 3.Chen L, Liu H, Hong Y, Yang Y, Hu L. Minimally invasive decompression and intracorporeal bone grafting combined with temporary percutaneous short-segment pedicle screw fixation for treatment of thoracolumbar burst fracture with neurological deficits. World Neurosurg. 2020;135:e209–e220. doi: 10.1016/j.wneu.2019.11.123. [DOI] [PubMed] [Google Scholar]
- 4.Ha KY, Kim KW, Kim YH, Oh IS, Park SW. Revision surgery after vertebroplasty or kyphoplasty. Clin Orthop Surg. 2010;2:203–208. doi: 10.4055/cios.2010.2.4.203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kim CH, Kim P, Ju CI, Kim SW. Acute paraparesis caused by spinal epidural fluid after balloon kyphoplasty for traumatic avascular necrosis: a case report. Korean J Neurotrauma. 2023;19:398–402. doi: 10.13004/kjnt.2023.19.e34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kim CH, Kim P, Ju CI, Kim SW. Open surgery for osteoporotic compression fracture within one month of single level balloon kyphoplasty. Korean J Neurotrauma. 2023;19:348–355. doi: 10.13004/kjnt.2023.19.e33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ko MJ, Lee BJ. Kummell’s disease is becoming increasingly important in an aging society: a review. Korean J Neurotrauma. 2023;19:32–41. doi: 10.13004/kjnt.2023.19.e13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kweon S, Kwon S, Kim C, Lee C, Kim I, San Ko Y. Indirect decompression of osteoporotic vertebral compression fracture using intraoperative motor evoked potential monitoring-guided ligamentotaxis. Korean J Neurotrauma. 2023;19:258–265. doi: 10.13004/kjnt.2023.19.e21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lee BJ, Seok MC, Koo HW, Jeong JH, Ko MJ. Bone substitute options for spine fusion in patients with spine trauma-part I: fusion biology, autografts, allografts, demineralized bone matrix, and ceramics. Korean J Neurotrauma. 2023;19:446–453. doi: 10.13004/kjnt.2023.19.e62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lee J, Song KS. Transpedicular intravertebral cage augmentation in a patient with neurologic deficits after severely collapsed Kummell disease: minimum 2-year follow-up. World Neurosurg. 2020;135:146–155. doi: 10.1016/j.wneu.2019.11.131. [DOI] [PubMed] [Google Scholar]
- 11.Lee JT, Ko MJ, Lee BJ, Lee YS, Lee SH. Pain intervention for osteoporotic compression fracture, from physical therapy to surgery: a literature review. Korean J Neurotrauma. 2024;20:159–167. doi: 10.13004/kjnt.2024.20.e32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lee S. Letter to the editor: commentary on Kummell's disease is becoming increasingly important in an aging society: a review (Korean J Neurotrauma 2023;19:32-41) Korean J Neurotrauma. 2024;20:77–78. doi: 10.13004/kjnt.2024.20.e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lee SH, Jain V, Lee SH, Baek OK, Moon KH. Transpedicular intrabody cage insertion with posterior spine stabilization in Kummell disease: report of 2 cases. World Neurosurg. 2018;116:236–240. doi: 10.1016/j.wneu.2018.05.133. [DOI] [PubMed] [Google Scholar]
- 14.Lee SM, Oh HS, Lee SH, Lee HC, Hwang BW. Cement augmented anterior reconstruction and decompression without posterior instrumentation: a less invasive surgical option for osteoporotic thoracolumbar fracture with cord compression. Korean J Neurotrauma. 2020;16:190–199. doi: 10.13004/kjnt.2020.16.e37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Li C, Zhou Y, Wang H, Liu J, Xiang L. Treatment of unstable thoracolumbar fractures through short segment pedicle screw fixation techniques using pedicle fixation at the level of the fracture: a finite element analysis. PLoS One. 2014;9:e99156. doi: 10.1371/journal.pone.0099156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Li KC, Li AF, Hsieh CH, Liao TH, Chen CH. Another option to treat Kümmell’s disease with cord compression. Eur Spine J. 2007;16:1479–1487. doi: 10.1007/s00586-006-0094-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Li KC, Wong TU, Kung FC, Li A, Hsieh CH. Staging of Kummell’s disease. J Musculoskelet Res. 2004;8:43–55. [Google Scholar]
- 18.Liao JC, Chen WJ. Short-segment instrumentation with fractured vertebrae augmentation by screws and bone substitute for thoracolumbar unstable burst fractures. Biomed Res Int. 2019;2019:4780426. doi: 10.1155/2019/4780426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lim J, Choi SW, Youm JY, Kwon HJ, Kim SH, Koh HS. Posttraumatic delayed vertebral collapse: Kummell’s disease. J Korean Neurosurg Soc. 2018;61:1–9. doi: 10.3340/jkns.2017.0505.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.McGilvray KC, Easley J, Seim HB, Regan D, Berven SH, Hsu WK, et al. Bony ingrowth potential of 3D-printed porous titanium alloy: a direct comparison of interbody cage materials in an in vivo ovine lumbar fusion model. Spine J. 2018;18:1250–1260. doi: 10.1016/j.spinee.2018.02.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Oh Y, Hwang S, Kwon SW, Shin HJ, Kim S. Clinical and radiological efficacy of short-segment in situ fixation vs. long-segment fixation and corpectomy in young patients with traumatic thoracolumbar burst fractures. Korean J Neurotrauma. 2025;21:303–316. doi: 10.13004/kjnt.2025.21.e37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ohba T, Oba H, Oda K, Tanaka N, Haro H. Surgical outcomes after minimally invasive direct lateral corpectomy with percutaneous pedicle screws for osteoporotic thoracolumbar vertebral collapse with neurologic deficits in the thoracolumbar spine compared with those after posterior spinal fusion with vertebroplasty. Spine (Phila Pa 1976) 2021;46:1271–1278. doi: 10.1097/BRS.0000000000003988. [DOI] [PubMed] [Google Scholar]
- 23.Park HJ, Kim HB, You KH, Kang MS. Percutaneous transpedicular intracorporeal cage grafting for Kümmell disease. Acta Neurochir (Wien) 2022;164:1891–1894. doi: 10.1007/s00701-022-05211-z. [DOI] [PubMed] [Google Scholar]
- 24.Park SJ, Kim HS, Lee SK, Kim SW. Bone cement-augmented percutaneous short segment fixation: an effective treatment for Kummell’s disease? J Korean Neurosurg Soc. 2015;58:54–59. doi: 10.3340/jkns.2015.58.1.54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rezvani M, Asadi J, Sourani A, Foroughi M, Tehrani DS. In-fracture pedicular screw placement during ligamentotaxis following traumatic spine injuries, a randomized clinical trial on outcomes. Korean J Neurotrauma. 2023;19:90–102. doi: 10.13004/kjnt.2023.19.e9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Song KS, Yoon BI, Ham DW. Transpedicular intravertebral cage augmentation using expandable cage in Kummell disease: technical note and case series. World Neurosurg. 2024;184:119–124. doi: 10.1016/j.wneu.2024.01.079. [DOI] [PubMed] [Google Scholar]
- 27.Sudo H, Ito M, Kaneda K, Abumi K, Kotani Y, Nagahama K, et al. Anterior decompression and strut graft versus posterior decompression and pedicle screw fixation with vertebroplasty for osteoporotic thoracolumbar vertebral collapse with neurologic deficits. Spine J. 2013;13:1726–1732. doi: 10.1016/j.spinee.2013.05.041. [DOI] [PubMed] [Google Scholar]
