A novel treatment option for Kummell’s disease using percutaneous kyphoplasty with a mesh-hold bone filling container Authors: Bingyan Mao, Wencong Wang, Qiang Wang, Lei Li Published: 2024 ABSTRACT AbstractPercutaneous kyphoplasty (PKP) is an effective, minimally invasive treatment for Kummell’s disease, but challenges like cement leakage, fracture, and displacement remain. This study explores the use of PKP with a mesh-hold bone filling container for this disease.A retrospective analysis was conducted on nine Kummell’s disease cases treated with this technique at a spinal surgery department from June 2016 to May 2022. Pre- and postoperative evaluations included X-rays, CT scans, and 3D reconstructions. Parameters like vertebral heights, kyphotic angle, operation time, blood loss, and bone cement issues were measured. Efficacy was assessed using ODI and VAS scores.Statistical analysis showed no significant difference in posterior vertebral height before and after treatment. All other indexes improved significantly postoperatively and at follow-up. No cement leakage or displacement occurred.PKP with a mesh-hold bone filling container is a promising treatment for Kummell’s disease. It enhances reduction quality, ensures adequate bone cement filling, and reduces issues related to cement leakage, fracture, and displacement.Keywords: Kummell’s disease, Percutaneous kyphoplasty (PKP), Polymethyl methacrylate, Cement leakage, Cement fracture, Cement displacementSubject terms: Diseases, Medical research Introduction Kummell’s disease, characterized by delayed collapse and avascular necrosis of a vertebral body resulting from compression fractures in osteoporotic thoracolumbar vertebrae, represents a rare manifestation of osteoporotic vertebral fractures1–4. First reported by German surgeon Hermann Kummel in 1895, this condition gained increasing attention over years5. Despite its historical recognition, the exact pathogenesis of Kummell’s disease remains unclear, with many attributing it to blood supply disorder and biomechanical instability6–8. Treatment options for Kummell’s disease predominantly involve conservative treatment and surgery. Conservative treatment, often necessitating prolonged bed rest, is usually associated with complications, some of which can be life threatening, particularly in elderly patients with osteoporosis. Granulation hyperplasia in necrotic bone tissues of the vertebral body, micromotion, blood supply disorders at fracture sites, and reduction in the activity and number of osteoblasts in osteoporosis patients contribute to the challenging nature of fracture healing in Kummell’s disease, leading to limited efficacy of conservative approaches9. Open surgery, while effective, poses risks and is less suitable for elderly patients with compromised bone quality and reduced surgical tolerance. Moreover, open surgery carries an increased risk of damaging adjacent vertebral body and complications associated with internal fixation10–14. Therefore, minimally invasive procedures have become a key research topic. In October 2015, we encountered a case of stage III Kummell’s disease in an elderly female patient. Her symptoms initiated by a minor sprain, included pain and restricted mobility in the thoracolumbar region, and alleviated when in a supine position. Postoperative CT reconstruction produced the model of necrotic vertebral body. Percutaneous kyphoplasty (PKP) was performed, involving vertebral body dilation and reduction using a balloon via a pedicle sleeve, followed by bone cement injection. Image reconstruction using reverse engineering software confirmed the absence of cortical bone obstruction in front of the vertebral body. Pedicle screw with tail wing were employed to anchor and prevent postoperative bone cement displacement. While the patient experienced significant relief from pain and neural symptoms postoperatively, a subsequent minor sprain at 6 months resulted in tolerable pain, revealing bone cement fracture and displacement on X-ray examination. The patient rejected further evaluation and surgical intervention, finding relief through the use of a fracture brace over 3 months (Fig. 1). This case prompted a critical reevaluation of PKP, specifically addressing challenges related to preventing bone cement leakage, ensuring sufficient filling of bone cement, and averting postoperative bone cement fracture and displacement. In response, this study introduced the use of a mesh-hold bone filling container in PKP and evaluated its efficacy in the context of Kummell’s disease. A 65-year-old female patient. (A1) Severe compression fracture of the T12 vertebra by preoperative lateral X-ray with vacuum phenomenon; (A2-3). Necrotic bone in the T12 vertebra with pneumatosis by preoperative coronal and sagittal CT scan; (A4-6). T1-weighted hypointense signals, T2-weighted hyperintense signals and fat-suppressed T2-weighted hyperintense signals of hydrops by preoperative MRI; (B1-4). Status of bone cement filling and anchorage using screw with tail wing in the bone cement for the T1 vertebra after PKP upon 3D anteroposterior and lateral X-ray and coronal and sagittal CT scan; B5-6. Bone cement fracture and displacement in the T12 vertebra by anteroposterior and lateral X-ray due to minor trauma at 6 months after PKP; (C1-2). Main distribution of necrotic bone and pneumatosis for the thoracolumbar spine by preoperative CT-based reconstruction to plan for the puncture direction; (C3-4) .Distribution of bone cement and vertebral body reduction by postoperative CT reconstruction after PKP; (C5-6). Bone cement distribution of the T12 vertebra and anchorage of the screw with tail wing in the bone cement after removing the surrounding vertebral bodies; Red arrowhead (↑) indicates gas or liquid gaps generated by several separate injections of bone cement, which might influence overall stability of bone cement. Methods Patients From June 2016 to May 2022, 9 patients (2 males, 7 females; aged 62–75 years, average 67.2) diagnosed with stage III Kummell’s disease were treated with PKP using a mesh-hold bone filling container at our hospital. All surgeries were performed by the researcher. The primary inclusion criteria encompassed a diagnosis of primary osteoporosis (dual energy X-ray absorptiometry, T value < −2.5) and a history of minor trauma persisting for over 3 months without complete rest in bed for immobilization. Conservative treatment had proven ineffective, and symptoms either recurred or worsened. Radiographic evidence, including X-ray scans revealing cleft or vacuum phenomenon and CT scans displaying vertebral body collapse, pneumatosis, or effusion, confirmed the diagnosis. Reverse engineering software aided in modeling the distribution of necrotic bone for surgical planning. Patients with a history of spinal surgery, infections, or tumors were excluded. This work was approved by the Ethics Committee of Shimen County People’s Hospital (K2024002). Informed consent was obtained from all patients. Surgical technique Patients were informed of potential risks, including aggravated neurologic symptoms and the need for emergency surgery, and signed informed consent. Under local anesthesia supplemented with systemic analgesics, surgery was conducted with continuous monitoring of neurological status. With the affected vertebral body at the center, endplates were shown as linear shadows under the C-arm X-ray. Bilateral pedicles were symmetrical upon X-ray and had equal distance to the spinous process. Puncture needles were inserted into the pedicles under C-arm X-ray bilaterally, passing through the posterior vertebral body, and further delivered for 2–3 mm. Using a solid core drill, the vertebral body was dilated, reduction level and integrity of endplates were assessed by C-arm X-ray. An elastic strip-based dilation system was employed, achieving vertebral body dilation through rotation. Following successful reduction, the elastic metal strip was retracted, and the dilator was rotated an additional 30° to achieve further vertebral body dilation. A final dilation, forming a pattern of ✳, was achieved by a 180° rotation (Fig. 2). Using sagittal diameter measurements, an appropriate sized mesh-hold bone filling container (manufactured by Dragon Crown Medical Utensils Co., Ltd China) was selected for implantation. Bone cement was meticulously prepared and gradually injected, during the later stages of wire drawing, into the mesh through a bone cement injector. Continuous monitoring under C-arm X-ray ensured the mesh-hold bone filling container’s constant dilation and gradual filling with bone cement until optimal vertebral reduction was achieved (Fig. 3). In cases where pedicles remained intact, pedicle screws with tail wings were used to reinforce anchoring through the bone cement. Observations included bone cement distribution, the amount injected, and the duration of the procedure. In-situ observation continued for 7–10 min postoperatively. Continuous monitoring of vital signs and bilateral lower limbs mobility ensued. Patients were permitted to turn over in bed after 2 h and commenced exercises with a conventional external brace after 1 day of rest. Schematic diagram of vertebral dilation and reduction using elastic metal strip. Dilation was performed once after every rotation by 30°, A dilation in ✳ pattern was achieved after rotation by 180° (a→b→c→d). Schematic diagram of bone cement filling using mesh-hold bone filling container. Clinical evaluation Clinical evaluation was conducted at multiple time points, including pre-surgery, at 3 days and 3 months postoperative, and during the last follow-up (12–23 months). Visual Analogue Scale (VAS) scores (a 0–10 point scale where 0 point indicates no pain and 10 signifies extremely severe pain) were used to assess the control of lumbodorsal pain. The Oswestry Disability Index (ODI) was used to assess the impact of the procedure on patients’ daily lives. Radiographic examinations Upright anteroposterior X-ray scans were performed at key intervals: pre-surgery, at 3 days and 3 months postoperative, and during the last follow-up (12–23 months). Measurements included anterior, midline and posterior vertebral heights. The kyphotic angle was determined using the Cobb angle. To evaluate bone cement distribution and detect any leakage, CT reconstructions were generated using reverse engineering software. Results Nine patients received puncture of bilateral pedicles and PKP with a mesh-hold bone filling container. The operation time ranged from 52 to 68 min with an average of 58.89 min. The intraoperative blood loss was below 20 ml, and the exact volume was unknown. The injected amount of bone cement was 7.3–8.9 ml with an average of 8.27 ml. No instances of bone cement leakage were detected during intraoperative Pre- and postoperative X-ray, or CT scan with image reconstruction. All patients were followed up for over 12 months after surgery, averaging 17.9 months. The distribution of affected vertebral included 4 cases in T12 vertebra, 3 cases in L1 vertebra, and 2 cases in L2 vertebra. All cases had only one vertebra involved (Fig. 4). A 73-year-old female patient. (A1-2) Avascular necrosis of the T12 vertebra by preoperative coronal and sagittal cT scan; (A3-5) T1-weighted hypointense signals, T2-weighted hyperintense signals and fat-suppressed T2-weighted hyperintense signals of hydrops by preoperative MRI; (B1-4) Good status of bone cement filling in the T1 vertebra after PKP upon anteroposterior and lateral X-ray and coronal and sagittal CT scan; C1-2. Main distribution of avascular necrosis of T12 vertebra by preoperative CT reconstruction to plan for puncture direction; (C3-6) Status and distribution of bond cement filling in the T12 vertebra by postoperative CT reconstruction after PKP. VAS score: Preoperative VAS score averaged 8.89 ± 0.60 (range 8–10). At 3d postoperative, the average VAS score decreased to 2.78 ± 0.44 (2–3). At 3 months postoperative and at the end of the follow-up, the VAS scores were 2.22 ± 0.67 and 1.11 ± 0.60, respectively, indicating a continuous improvement. VAS scores were significantly lower post-treatment than pre-treatment (F = 413.043, P < 0.05). Pairwise comparisons (Bonferroni correction) indicated a significant reduction in VAS scores at all time points after surgery (P < 0.05), except for the observations at 3 days and 3 months postoperative (P = 0.080, > 0.05). ODI: Preoperative ODI was 70.62 ± 8.66, decreasing significantly to 28.64 ± 6.13 at 3d postoperative, 10.86 ± 3.59 at 3 months after surgery, and 8.89 ± 2.48 at the end of the follow-up (F = 183.917, P < 0.05). Pairwise comparison (Bonferroni correction) indicated no significant difference in ODI at 3 months after surgery and at the end of the follow-up (P = 0.724, > 0.05). ODI decreased significantly at any other time points after surgery (P < 0.05). Measurement of vertebral heights: Anterior and midline vertebral heights significantly increased at 3d postoperative, 3 months postoperative and at the end of the follow-up compared to pre-surgery (F = 73.143, 75.876, p < 0.05). No significant difference was observed in posterior vertebral height before and after surgery (F = 2.978, p > 0.05). Measurement of kyphotic angle: The kyphotic angle demonstrated a significant improvement at 3d and 3 months postoperative and at the end of the follow-up compared to pre-surgery (F = 382.160, p < 0.05) (Table 1). Table 1Comparison of indexes before and after operation ()Pre-PKP3 Days after PKP3 months after PKPAt Last Follow-upF valueP valuehp2VAS8.89 ± 0.602.78 ± 0.442.22 ± 0.671.11 ± 0.60413.0430.0000.981ODI(%)70.62 ± 8.6628.64 ± 6.1310.86 ± 3.598.89 ± 2.48183.9170.0000.958Ha(mm)12.07 ± 1.9119.14 ± 1.8018.68 ± 1.6418.35 ± 1.5973.1430.0000.901Hm(mm)12.74 ± 1.6318.26 ± 1.8418.10 ± 1.7517.93 ± 1.6675.8760.0000.905Hp(mm)21.22 ± 1.4721.95 ± 1.3521.79 ± 1.3721.55 ± 1.272.9780.1200.271Kyphotic angle(°)24.43 ± 1.6813.91 ± 1.0614.24 ± 1.0114.46 ± 0.91382.1600.0000.979VAS, indicates visual analogue scale; ODI, indicates Oswestry disability index; Ha, indicates anterior vertebral height; Hm, indicates midline vertebral height; Hp, indicates posterior vertebral height; PKP, indicates percutaneous kyphoplasty. Comparison of indexes before and after operation () VAS, indicates visual analogue scale; ODI, indicates Oswestry disability index; Ha, indicates anterior vertebral height; Hm, indicates midline vertebral height; Hp, indicates posterior vertebral height; PKP, indicates percutaneous kyphoplasty. In 1 case, postoperative CT scan and image reconstruction using reverse engineering software revealed less satisfactory bone cement filling, with a few air bubble shadows observed. This patient was advised to reduce physical activities and continue conservative treatment with bed rest. Discussion No unified diagnostic criteria have been formulated yet for Kummell’s disease. The diagnostic features include the following11,15,16: an elder age (generally in females), with or without history of minor trauma, non-standardized early treatment, thoracolumbar pain for over 3 months, aggravating pain under changing stress (which can be relieved by bed rest), and neurological injury of the corresponding vertebrae due to spinal nerve compression. Radiographically, distinct features aid in differentiation. Avascular necrosis of the vertebral body is a hallmark, accompanied by relative integrity of upper and lower endplates. Notably, avascular necrosis affecting the middle of the vertebral body leads to layering, observable through the opening and closing sign on X-rays during extension and flexion. CT scans reveal non-uniform or irregular distribution of gas or liquid, presenting as vacuum cleft, with heightened sensitivity to MRI. Specific MRI signals include hypointense signals during pneumatosis and effusion on T1W1 and T2W1 images, repsectively. In this study, all patients exhibited a T-score <-2.5 from bone density scans. The exclusion of infections and tumors further fortified the diagnostic validity. Osteoporosis emerges as a critical risk factor in Kummell’s disease17, associated with decreased bone mass and bone density. This study population, consisting of 2 males and 7 females aged 62–75 years with an average of 67.2, aligns with the higher incidence of osteoporosis in elderly females. The trabecular bone fracture resulting from minor trauma, coupled with occlusion or thrombosis of blood vessels nurturing the vertebral body, is considered a pivotal pathogenesis mechanism8,18. The anterior 1/3 of the vertebral body, supplied by a single terminal branch, is prone to happen avascular necrosis, contributing to non-healing of bone fractures. This agrees with the finding of the present study that the necrotic site was mainly located at anterior vertebral body. Another nonnegligible local factor is biochemical changes following vertebral fracture. The dynamic load concentration on the thoracolumbar segments plays a pivotal role, particularly when patients fail to adhere to complete bed rest. The consequence of this noncompliance manifests as a hinge-like opening and closing motion during repeated movement. These changes include local shear stress, presenting a significant impediment to the osteogenic process. This further aggravates bone necrosis, leading to the collapse or non-healing of the vertebral body, which finally leads to pseudarthrosis (Fig. 5)16,19,20. Schematic for the pathogenesis of Kummell’s disease. Simple vascular anatomy in front of the vertebral body can lead to post-fracture vascular damage and hence to avascular necrosis of the bone in front of the vertebral body. The stress is concentrated in the thoracolumbar spine, and repeated hinge-like opening and closing movement will aggravate vascular necrosis. Treatment goals in Kummell’s disease focus on stabilizing bone fracture, promoting fracture healing, restoring vertebral body height, and enhancing patients’ life quality. The choice of treatment depends on the influence from conventional culture, medical level and economic status. Conservative treatment which requiring prolonged bed rest, often yields poor results due to complications21–23. Open surgery, while effective, is traumatic, especially in elderly patients with reduced bone quality. PKP has demonstrated efficacy in fresh osteoporotic spinal compression fractures2,24–27. PKP not only alleviates lower back pain but also broadens the range of motion, thereby improving the overall quality of life for patients. This procedure also establishes relative stability in the injured vertebral body, mitigating instability at the fractured site and partially restoring vertebral body height. Crucially, PKP impedes further necrosis and collapse of the vertebral body, averting the progressive aggravation of kyphosis deformity that could potentially lead to neurological injury. The intricate nature of non-healing of intravertebral fractures associated with Kummell’s disease precipitates osteosclerosis in the lower and upper endplates. Long-term hinge-like micromotion accentuates the risk of severe vertebral collapse and deformation, necessitating precision in puncture skills and balloon dilation degrees. In our study, the reconstruction of the vertebral body model using reverse engineering software facilitated a meticulous approach. This allowed for the observation of the distribution of avascular necrosis in bone tissues, aiding in the strategic planning of the puncture procedure. Model reconstruction not only augmented the success rate of initial puncture attempts but also served as a preventive measure against pedicle and nerve root injuries resulting from repeated punctures. Galibert et al. pioneered the use of polymethylmethacrylate (PMMA) bone cement in vertebral hemangioma. Subsequently, percutaneous vertebroplasty (PVP) and PKP have become widely used in osteoporotic vertebral compression fractures, distinguished by their minimal invasiveness, short operation durations, swift postoperative recovery, and immediate pain relief after surgery. PKP stands out for its ability to enhance the mechanical stability of fractured vertebral bodies, offering partial restoration of vertebral body height and alleviation of spinal deformities28. The use of PKP with balloon dilation has been reported as a promising approach for treating Kummell’s disease. Balloon dilation, when coupled with PKP, proves more effective in restoring vertebral body height and correcting kyphosis. Despite the global adoption of PVP and PKP since the 1980s, the persistent challenge of bone cement leakage remains a critical concern29–35. While rare, instances of bone cement leakage can have catastrophic consequences. Intraoperative bone cement leakage directly impacts the adequacy of cement filling, influencing vertebral stability and the restoration of vertebral body height. PMMA, when intracystically injected post balloon dilation, shows limited binding with surrounding bone tissues compared to trabecular bone before balloon dilation. PMMA, primarily a space-occupying material, lacks mechanical interlock or biocompatibility. Consequently, support and fixation through bone cement filling may be compromised, leading to displacement or fracture of the cement36. To mitigate the risk of bone cement leakage, multiple injections of bone cement have proven effective27,37. However, challenges arise with potential effusion (blood) or gases affecting the overall stability of the bone cement during this process. The presented case of bone cement fracture and failure underscores the delicate balance required in addressing concerns about leakage. Excessive apprehension may lead to insufficient injection, while obstacles posed by effusion and gases during the injection process can culminate in the failure of bone cement fixation. In our procedure of PKP, we introduce a novel technique using a mesh-hold bone filling container. Characterized by intact lower and upper endplates often presenting sclerosis in Kummell’s disease, the mesh-hold bone filling container incorporates an elastic strip-based dilation system, proving more effective in achieving dilation between the lower and upper endplates compared to traditional balloons. The mesh-hold bone filling container has a reticular, braided structure designed to restrict bone cement movement. This structure allows for the accumulation of bone cement within its confines. As the reticular structure undergoes dilation, the bone cement undergoes a transformation from random diffusion to initial concentration and subsequent radial diffusion. This strategic evolution in the diffusion process serves to enhance the overall precise and effective bone cement distribution. To further fortify the anchoring of the bone cement within the vertebral body, several techniques are employed: (1) Mesh size selection. Tailoring the mesh size of the bone filling container based on the extend of the bone defect at the anterior and posterior vertebral body proves instrumental. When the anterior vertebral body exhibits a larger defect than the posterior counterpart, a countainer with a smaller mesh size at the anterior vertebral body is utilized. This intential choice raises the difficulty of bone cement flow at the anterior vertebral body, thereby minimizing the risk of anterior leakage. Moreover, this approach allows for an increased injection volume of bone cement, ensuring full contact between bone cement and bone tissues. (2) Pedicle screw reinforcement. In cases where the posterior vertebral body and pedicles remain intact, the insertion of a pedicle screw with a tail wing is implemented. This is accompanied by a small amount of bone cement injection into the pedicle, serving to prevent forward sliding of the bone cement. Notably, this technique proved effective in a case where the bone cement on the right side fractured in the middle, leading to forward sliding at the right anterior vertebral body. In contrast, no displacement of bone cement occurred at the right posterior vertebral body due to the use of a pedicle screw with a tail wing. This result demonstrated the efficacy of using a pedicle screw to augment the anchoring of bone cement in the vertebral body. (3) Mitigation of gaps through controlled injections. Continuous injections of bone cement using a bone cement injector may induce gaps in bone cement, attributed to the presence of air bubbles or effusion (blood). Postoperative CT scans revealed air bubble shadows in the vertebral body in a singular case. In response, this patient was advised to reduce physical activity and adhere to prolonged bed rest. At the end of the last follow-up, none of the cases exhibited fracture of the bone cement. The injured vertebral body and its adjacent normal vertebral bodies were reconstructed both pre- and post-surgery using reverse engineering software. From the reconstructed model, the region with avascular necrosis of bone tissues was located and the direction of intraoperative puncture was determined. The decision to employ a pedicle screw with a tail wing for anchoring the bone cement hinged on the integrity of the cortical bone at the posterior vertebral body and within the pedicles. Employing 3D measurement, the extent of the necrotic region within the vertebral body was gauged, facilitating an approximate assessment of the necessary bone cement volume. The administration of bone cement was meticulously observed through the reconstructed model of the vertebral body. This data served as the foundation for determining the optional duration of postoperative bed rest, strategically aimed at minimizing the risk of bone cement fixation failure. Conclusions In conclusion, the utilization of a mesh-hold bone filling container in PKP demonstrates notable enhancements in the quality of reduction achieved. This improvement is attributed to the augmented volume of injected bone cement as well as the strengthened anchoring capacity within the vertebral body. More importantly, this technique can mitigate the potential risk of bone cement leakage, fractures, and displacement. Thus, it stands as a new and minimally invasive treatment option for Kummell’s disease. However, as a novel medical device, the mesh-hold bone filling container indwells within the human body without removal, thus raising the need to perform large-scale clinical trials and long-term follow-up studies to confirm its safety and efficacy profiles. Acknowledgements None. Abbreviations Percutaneous kyphoplasty Oswestry disability index Visual analogue scale Polymethylmethacrylate Percutaneous vertebroplasty Author contributions Bingyan Mao and Wencong Wang carried out the studies, participated in collecting data, and drafted the manuscript. Qiang Wang and Lei Li performed the statistical analysis and participated in its design. Bingyan Mao and Lei Li participated in acquisition, analysis, or interpretation of data and draft the manuscript. All authors read and approved the final manuscript. Funding None. Data availability All data generated or analysed during this study are included in this published article. Declarations Ethics approval and consent to participate This work has been carried out in accordance with the Declaration of Helsinki (2000) of the World Medical Association. This work was approved by the Ethics Committee of Shimen County People’s Hospital (K2024002). Informed consent was obtained from all patients. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement All data generated or analysed during this study are included in this published article.