A 10-year-old boy (weight, 23 kg; height, 123 cm; and body mass index, 13.33 kg/m2) was referred to us in 2019 with a diagnosis of type-III GD. Thrombocytopenia was detected in the patient and a β-glucocerebrosidase (GCase) level of 1.54 nmol/h.mg was detected. He had undergone splenectomy in 2012 and had been administered 1200 units of imiglucerase once every 2 weeks since 2015. He had no history of bone crisis, pathologic fractures, or osteomyelitis. Deterioration of the thoracic kyphosis (TK) was noted in July 2017, but no rigid bracing was used. Thereafter, the deformity increased rapidly, leading to a kyphosis of 113° in 2019 (Fig. 1A). Pain developed over the apex of his deformity, and he started to have difficulty in walking with lower extremity weakness (3+/5 − 4/5). Moreover, the boy suffered from moderate horizontal gaze palsy and sternum deformity. Genetic testing indicated heterogeneity for gene mutations c.1448T > C and c.1205A > G. There were no signs of pulmonary dysfunction or myelopathy. The patient had mild anemia with normal platelet counts and blood clotting. Radiographs were obtained, which revealed a Cobb angle of TK as 113°(Fig. 1A). Computed tomography (CT) and magnetic resonance imaging (MRI) revealed inflammation of the mesenteric lymph nodes and anterior wedging of vertebrae from T7 to T10 (data not shown). The patient was diagnosed as : Spinal kyphosis deformity, type IIIb GD and sternal lordosis.
After anesthesia was induced, the child was placed in the prone position without compression of the sternum. The spinous processes, lamina, and facet joints of the instrumented segments were exposed using the posterior midline approach. After the pedicle screws had been implanted from T6 to L3, Smith-Peterson osteotomies were performed at five levels (T8-L1). We used gentle posterior compression along the bilateral rods to allow controlled anterior lengthening. The surgical time was 180 min, and blood loss was 300 ml. Intraoperative monitoring with neurogenic or transcranial motor-evoked potentials were used during every procedure. Correction of the segmental kyphotic angle was achieved (70°) after the first surgery, yielding a 61% correction rate (TK, 43°; lumbar lordosis [LL],38°; pelvic incidence [PI], 45°; PI-LL: −7°, sagittal vertical axis [SVA], 39 mm).
The patient went back to school without further brace protection. He administered 1200 units of imiglucerase once every 2 weeks as before. However, proximal junctional kyphosis (PJK) was observed at 4 months postoperatively (PJA,55°). He felt very hard to lift his head in his daily life. Revision surgery was performed to prevent the neurological impairment. During the revision surgery, we found that the pedicle screws were intact. We initially placed the screws in T2–T5 and then removed the screws of T6, after which laminectomy was performed, and a domino linkage system was used to link the internal fixation (Fig. 2D; TK:60°, LL:45°, PI: 45°, PI-LL:0°, SVA:67 mm). The surgical time was 195 min, and blood loss was 200 ml. Neuromonitoring signals during the operation were stable.We tried to mobilize the boy on the following day and discharged him on the fourth day after the revision surgery. He used brace protection for 5 months and was followed up at 3 months, 5 months, and 2 years. X-rays were repeated at the 2-year follow-up, which showed no obvious recurrence of PJK and solid fusion was achieved (PJA,19°, TK:60°, LL:50°, PI: 45°, PI-LL:-5°, SVA:-30 mm).
Hematoxylin-eosin staining and immunohistochemistry analysis were performed for CD68 (M1-type macrophages) and CD206 (M2 type macrophages) in resected lamina of the patient and compared with those of a 14 year-old healthy boy (control). Increased expression of CD68 but lower expression of CD206 was observed in the cancellous bone of the patient (Fig. 4).