Aseptic lysis L2–L3 as complication of abdominal aortic aneurysm repair

An 80-year-old man, with multiple co-morbidities was admitted to the author’s hospital for severe low-back pain, lower limbs motor impairment and bilateral thigh pain. Nine months before admission to our department, he underwent endovascular aneurysm repair (EVAR) for an abdominal aortic aneurysm (Fig. 1). A computed tomography (CT) scan performed 1 month after the endovascular abdominal aortic aneurysm repair did not show any signs of lumbar vertebral or disc erosion (Fig. 2). Three months after the endovascular procedure, he developed a progressive lower back pain and bilateral thigh pain that did not respond to conservative treatments. At this time, in another hospital, MRI of the lumbar spine showed severe bone loss in the anterior half of the third lumbar vertebral body and L3 vertebroplasty was performed without any significant relief of the symptoms (Fig. 3). On admission in our hospital, 9 months after EVAR, clinical evaluation showed severe low-back pain, bilateral thigh pain, motor deficit of the lower limbs and the patient was not able to walk (Frankel C) [10]. Laboratory findings were within normal limits: haemoglobin, 12.5 g/dl (normal 12–16 g/dl), white blood cell count, 6.3 × 10⁹/l (normal 4.3–10.8 × 10⁹/l) with a normal differential, erythrocyte sedimentation rate 25 mm/h (normal 2–30 mm/h), C-reactive protein 2.4 mg/l (normal 0.00–3.00 mg/l). A new MRI of the lumbar spine, performed at admission, showed the false aneurysm and its relation to L2–L3 bodies causing vertebral and disc erosion (Fig. 4). Fluoroscopy guided biopsy was performed, but it was negative as regards microbiological and histopathological examination for tumours or infections. A more accurate evaluation of MRI of the lumbar spine uncovered an extensive abdominal aortic false aneurysm, corresponding with the prosthetic stents that had eroded the vertebral bodies of L2–L3. The vascular surgeon did not consider revision surgery necessary at that time. Laboratory findings, fluoroscopy guided biopsy and PET-CT (Fig. 5) had excluded infection, and for this reason we avoided draining the fluid at the L2–L3 disc space, but we opted for a posterior decompression of the central canal stenosis, between L2 and L3, to improve the neurologic condition of the patient. In addition a long and extensive instrumentation of T12–L5 was performed to avoid the risk of implant failure due to the presence of severe vertebral osteopenia (Fig. 6). Moreover, an autologous iliac bone graft was utilized to obtain better postero-lateral fusion. In the disc space L2–L3 we just found abundant serum-hematic fluid. Another biopsy of the L2–L3 disc performed during surgery confirmed the absence of tumours or infections. Postoperatively, laboratory findings were: haemoglobin 9.2 g/dl, white blood cell count 8.5 × 10⁹/l, erythrocyte sedimentation rate 35 mm/h, C-reactive protein 3.00 mg/l. After surgery the patient obtained excellent relief of low-back and thigh pain with a satisfactory regain of walking (Frankel D). Unfortunately, 1 month after surgery the patient died because of respiratory complications. We excluded sepsis as a possible cause of death because the patient in the postoperative period had no fever and laboratory findings were within normal limits.