Geniculate Artery Embolization in Patients With Recurrent Hemarthrosis After Knee Arthroplasty: A Retrospective Study.

BACKGROUND: Recurrent hemarthrosis after knee arthroplasty is an uncommon and disabling complication of this frequently performed procedure. Selective endovascular embolization of the geniculate arteries is one of the therapeutic options to manage this complication. The purpose of this study is to analyze the effectiveness of this treatment in patients suffering from recurrent hemarthrosis after knee arthroplasty. METHODS: We performed a retrospective study of 31 patients (39 embolization procedures) with recurrent hemarthrosis after knee arthroplasty. There were 17 men and 14 women with a median age of 67 years (range 48-90). All patients were referred for geniculate artery embolization between January 2007 and November 2016. RESULTS: Twenty-seven procedures were executed on the right side and 12 on the left side. Total knee arthroplasty was performed on 29 patients, only 2 patients underwent unicompartmental knee arthroplasty. Embolization of the superior geniculate arteries was achieved in all patients. In 12 of 39 procedures (31%), at least 1 of the inferior geniculate arteries could not be catheterized, therefore embolization was achieved through collaterals. Symptomatic improvement was observed in 26 of 31 patients (84%). Discomfort or mild postprocedural pain was observed in most patients, needing only minor pain medication, mostly resolving within 24 hours. Two patients presented with a severe complication: a 48-year-old male patient developed septic arthritis and an 85-year-old hypertensive female patient treated with anticoagulants showed aseptic necrosis of the femoral condyles. CONCLUSION: Embolization of geniculate arteries is a safe and effective treatment in recurrent hemarthrosis post knee arthroplasty. Clinical improvement was seen in most patients.

A spinal arteriovenous fistula in a 3-year old boy.

We present a case of a 3-year-old boy with neurodegeneration. Family history reveals Rendu-Osler-Weber disease. Magnetic resonance imaging (MRI) of the spinal cord and spinal angiography showed a spinal arteriovenous fistula with venous aneurysm, causing compression of the lumbar spinal cord. Embolisation of the fistula was executed, resulting in clinical improvement. A week after discharge he was readmitted with neurologic regression. A second MRI scan revealed an intraspinal epidural haematoma and increase in size of the aneurysm with several new arterial feeders leading to it. Coiling of the aneurysm and fistulas was performed. Postoperative, the spinal oedema increased despite corticoids, causing more extensive paraplegia of the lower limbs and a deterioration of his mental state. A laminectomy was performed and the aneurysm was surgically removed. Subsequently, the boy recovered gradually. A new MRI scan after two months showed less oedema and a split, partly affected spinal chord. This case shows the importance of excluding possible arteriovenous malformations in a child presenting with progressive neurodegeneration. In particular when there is a family history for Rendu-Osler-Weber disease, scans should be performed instantly to rule out this possibility. The case also highlights the possibility of good recovery of paraplegia in paediatric Rendu-Osler-Weber patients.

Navigation with three-dimensional rotational radiographic data for transpedicular percutaneous needle introduction: feasibility and comparison with fluoroscopic guidance.

PURPOSE: To investigate the feasibility of navigation with three-dimensional (3D) rotational radiographic data for transpedicular percutaneous needle introduction and to compare navigation with 3D rotational radiographic data with conventional fluoroscopic guidance. MATERIALS AND METHODS: A navigation system was coupled to a 3D rotational radiographic imaging system. In a cadaver study, 60 biopsy needles were introduced into vertebral bodies with fluoroscopic guidance or navigation with 3D rotational radiographic data by two interventionalists with different levels of experience in percutaneous procedures. Radiation exposure, fluoroscopy and introduction times, and needle position were evaluated and compared. RESULTS: For 3D rotational radiographic navigation, the needle position was equivalent for both interventionalists (7 mm). For fluoroscopic guidance, the interventionalist with less experience in percutaneous procedures had significantly more pedicle cortex violations than the other interventionalist (eight vs one). Radiation exposure measured at the operators' hand was lower for 3D rotational radiography-guided needle introductions. Radiation exposure measured at the body wall of the cadaver was equivalent between modalities (9 mGy). CONCLUSIONS: Navigation with 3D rotational radiographic image data is feasible for transpedicular percutaneous needle introduction and has two advantages compared with fluoroscopic guidance: it poses less strict requirements on the expertise of the interventionalist, and it reduces radiation exposure to the interventionalist.

Prospective clinical follow-up after percutaneous vertebroplasty in patients with painful osteoporotic vertebral compression fractures.

PURPOSE: To prospectively assess short-term, midterm, and long-term pain relief in patients with painful osteoporotic vertebral compression fractures (VCFs) treated with percutaneous vertebroplasty (PV). MATERIALS AND METHODS: Visual analog scale (VAS) scores for pain at the treated vertebral level, analgesic use, and satisfaction with outcome were assessed in 112 patients after PV of 168 VCFs. Serial follow-up was performed at 24 hours and 3, 6, and 12 months and in a small number of patients at 1-3 years. Procedure-related complications were evaluated by physical examination and computed tomography of treated levels. RESULTS: After PV, VAS scores for pain at the individual vertebral levels treated and use of analgesic agents were significantly reduced compared with before treatment at every follow-up period. Within 24 hours after the procedure, the decreases in all scores were less compared with scores at later follow-up periods, but this was not significant. The preprocedural mean VAS score was 8.8 (range, 5-10). At follow-up, mean VAS scores ranged from 2.5 to 3.3 (range, 0-10). In the short term after PV, patients used significantly less analgesic drugs and 86% of patients were satisfied with the outcome. At midterm and long-term follow-up, patients used even less analgesic drugs and 95%-100% of patients were satisfied with the outcome of PV. Procedure-related complications with clinical consequences occurred in three patients (2.7%): one patient experienced a cardiovascular reaction, one patient had a pedicle chip fracture, and one had a rib fracture. CONCLUSION: PV of painful osteoporotic VCFs provides significant pain reduction in nearly all treated patients.

Three-dimensional rotational X-ray navigation for needle guidance in percutaneous vertebroplasty: an accuracy study.

STUDY DESIGN: The position of a needle tip displayed on a navigation system after transpedicular introduction into a vertebral body is compared with the real position of the needle tip when using a direct navigation coupling between a three-dimensional rotational X-ray (3DRX) system and a navigation system. OBJECTIVES: To assess whether the needle tip position displayed by the navigation system corresponds to the real needle position and to quantitatively determine needle navigation accuracy in a clinically relevant setting. SUMMARY OF BACKGROUND DATA: Image-guided navigation has reportedly increased the accuracy and safety of pedicle screw insertion and decreased complication rates. In former studies, the result of image-guided navigation was mainly compared qualitatively with the result of conventional fluoroscopy-guided procedures. Previously, a direct navigation coupling between a 3DRX system and a standard navigation system was introduced that bypasses the need for explicit patient-to-image registration necessary for image-guided orthopedic surgery. In a phantom experiment, the reported accuracy of navigation with the coupling to a 3DRX system was approximately 1 mm. However, in a clinical setting, additional errors can be introduced. METHODS: Twenty-three needles were placed transpedicularly into vertebral bodies of embalmed human trunks using 3DRX-guided navigation. The navigated needle tip positions were compared with the real needle tip positions manually extracted from 3DRX volumes acquired after completion of the introduction. RESULTS: The average distance between the navigated needle tip and the real position of the needle tip extracted from a postprocedure 3DRX volume was 2.5 +/- 1.5 mm. CONCLUSIONS: Accuracy of 3DRX-guided navigation is 2.5 +/- 1.5 mm in a clinically relevant setting, which is less than the accuracy determined in phantom experiments.

The risk of new osteoporotic vertebral compression fractures in the year after percutaneous vertebroplasty.

PURPOSE: To prospectively assess the incidence, location, and possible causative mechanisms of new vertebral compression fractures (VCFs) in 66 symptomatic patients with osteoporotic VCFs treated with percutaneous vertebroplasty (PV) and to study the relation between new VCFs and back pain symptoms. MATERIALS AND METHODS: Sixty-six patients with 102 painful symptomatic VCFs were treated with PV. All patients had baseline total spinal magnetic resonance (MR) imaging. Follow-up MR imaging was performed at 3, 6, and 12 months to locate new VCFs. Visual analog scales for pain and pain medication consumption were used to assess clinical outcomes. The following characteristics were compared in patients with new VCFs after PV versus patients without new VCFs: patient age, sex, presence of secondary osteoporosis, bone mineral density, number of preexisting VCFs, shape and grade of VCFs, type of bone cement used for PV, volume of injected cement, and cement leakage in intervertebral disc spaces. RESULTS: Sixteen of 66 patients had 26 new VCFs during 1 year of follow-up after PV. Most new VCFs occurred within 3 months of PV, half of new VCFs appeared in levels adjacent to treated levels, and half of the new VCFs were symptomatic. The presence of more than two preexisting VCFs was the only independent risk factor for the development of a new VCF. CONCLUSIONS: New VCFs occurred after PV in 24% of patients. Half of new VCFs occurred in levels adjacent to treated levels and half were symptomatic. The presence of more than two preexisting VCFs was the only independent risk factor for the development of a new VCF.