| Home | Publications | Proceedings |

Healing Outcomes of Long Bone Fractures in Persons with Spinal Cord Injury

B. Jenny Kiratli, Gail D. O'Mara, Helen H. Dorra, Inder Perkash, and George E. Sims

Spinal Cord Injury Center, VA Palo Alto Health Care System, Palo Alto, CA 94304


Objectives: Long bone fractures in persons with spinal cord injury (SCI) are characterized as slow-healing, often resulting in malalignment and nonunion, with disorganized "exuberant" callus a common feature. Our clinical observations have been otherwise. This study was undertaken to examine a large series of long bone fractures occurring after SCI in order to determine timing and quality of fracture healing. Specifically, we evaluated the association between fracture location and morphology and healing outcome.

Clinical Relevance: With an improved understanding of the normal healing sequence, a more appropriate evaluation pathway can be implemented to effectively monitor SCI patients who sustain long bone fractures. If patients are scheduled for radiographic evaluation at prescribed intervals relevant to expected healing patterns, decisions can be made regarding continuation or cessation of treatment, and non-optimal healing may be identified and corrected. Further, treatment regimes can be implemented and monitored according to an outcome-based clinical evaluation system, rather than according to observation and clinical judgment, as these may vary with the experience of the personnel.

Methods: Systematic reviews were performed on available radiographic records of 170 long bone fractures in patients seen at the Spinal Cord Injury Center. Sequential x-ray films were evaluated for location and type of fracture, presence and appearance of callus, anatomic alignment, and healing outcome. Fracture morphology was determined according to an hierarchical classification system which codes bone, anatomic location (proximal, midshaft, distal), and fracture type and complexity. In addition, we developed a Healing Index in order to quantify the presence and extent of callus formation and healing endpoint (0=no callus, 1=callus present in one projection only, 2=callus present in more than one projection, 3=bony bridging, 4=remodeling). Bony union was defined by a healing score of 3 or 4.

Results: Overall, a high proportion of successful healing was observed in lower limb long bone fractures; 90% achieved bony union. The majority of non-unions occurred in the femur, but no bone segment was more likely to heal improperly. Overall, 57% of leg fractures healed in normal alignment with malalignment less frequent in the healed fractures of the tibial midshaft (33%) and distal tibia (27%). Approximately one half of the fractures healed with some degree of shortening with compaction more common in femoral than tibial fractures. Callus was apparent by 3 weeks for all bone segments, except the proximal tibia where it was observed as early as 10 days and the midshaft tibia where it was not observed before 5 weeks. Bridging bone was present by 4-6 weeks in all segments, except the proximal femur where it was observed at 10 weeks. Normal callus formation occurred in the majority of lower limb fractures appearing along stress lines. Exuberant callus was observed in less that 10% of all tibia fractures, in 30% of femoral midshaft and distal fractures, and in 60% of proximal femoral fractures although these were the least common fracture type.

Conclusions: In patients with SCI, fracture healing is rapid and bony union is usually achieved before 2 months. Contrary to common belief, exuberant callus formation is found infrequently and normal callus formation occurs along stress lines. This bone formation does not seem to require a program of mechanical loading (i.e., graded weight-bearing) considered necessary for optimal healing in able-bodied fracture patients.

Acknowledgments: VA Rehabilitation R & D Merit Review Grant #A2014-RA.