F-18-fluoride PET for early diagnosis and evaluation of therapeutic outcome in patients with heterotopic ossification (HO) after recent paraplegia due to spinal cord injury

Aim: Heterotopic ossification (HO) is the presence of bone in soft tissue. The acquired form of HO most frequently is seen with either musculoskeletal trauma, spinal cord injury or central nervous system injury. Fever, swelling, erythema, and occasional joint tenderness seen in early HO can be difficult to distinguish from cellulitis, osteomyelitis or thrombophlebitis. As compared to paraplegia alone, combination of HO and paraplegia, especially excessive delay of diagnosis, is associated with a significantly higher incidence for thrombosis, immobilization, decubitus leading to a reduced expectation and quality of life. To evaluate the role of F-18-fluoride PET for the early diagnosis and the evaluation of the therapeutic outcome, 38 patients (56 PET examinations) were analyzed prospectively. Material and Methods: Within 8 weeks after acquired paraplegia, each patient was studied by F-18-fluoride PET in addition to clinical, serologic and conventional radiographic examinations. Whole-body PET studies (ECAT Exact 47, attenuation corrected, iterative reconstruction) were obtained 150 min. after injection of 12 MBq F-18-fluoride/kg body weight. For semi-quantitative analysis, standardized-uptake values (SUV) and the metabolic transverse diameters (MTD) of the lesions were assessed. To prevent post-traumatic neurogenic HO, patients received physiotherapy and NSA (indomethacin 3 x 50 mg per day for 4 months). In addition, external beam radiation therapy (EBRT, 1x7 Gy according to ICRU) was administered, if PET demonstrated HO. Therapeutic outcome was compared with a historical group of patients receiving only physiotherapy. Kaplan-Meier-Method, log-rank-, chi-square- and Wilcoxon-test were used for statistical analyses. Results: In the course of HO, 4/38 patients received EBRT with 1x7 Gy for a second time due to an increasing SUV and/or MTD and rising levels of alkaline phosphatase. Within a follow-up period of at least 30 months, none of the patients showed clinical signs of HO as compared to a historical group which had clinical symptoms in 79% (p<0.05). Median SUV of HO was 9.8 (range 7.2-33). After radiation therapy, SUV decreased by up to 23% as compared to pre-therapeutic values (p<0.05). There was no correlation between SUV and clinical symptoms. Conclusion: Our preliminary results show that intra-individual changes of SUV permit early detection of HO and/or HO-recurrence and may predict therapeutic outcome after EBRT. The metabolic diameter of HO in projection to the osseous structures may influence the size of the treatment portals for radiation therapy