INTRODUCTION
It is well-known that osteoporosis is associated with complications and poor outcome after spine surgery. The gold standard of bone mineral density measurement is dual energy X-ray absorptiometry (DXA), but the results of DXA scan are not always available for spine surgery patients. In such situations, surgeons may try to make a guess about bone strength using available imaging modalities; however, there has been no scientific validation of the accuracy of this intuitive diagnosis. Therefore, the purpose of this study is to examine the accuracy of the orthopedic surgeon's intuitive diagnosis of low bone mineral density using lumbar spine radiographs and CT sagittal images.
METHODS
Institutional review board approval was obtained for this study. We recruited volunteer board-certified orthopedic surgeons, and non-orthopedic specialists, including physicians specialized other than orthopedic or radiology, medical students, non-specialized interns, and orthopedic residents in our university and affiliated hospitals. The participants were asked to respond to a binomial questionnaire on whether each image corresponded to osteoporosis patient or not, using 45 lateral lumbar spine radiographs and 45 lumbar spine CT sagittal reconstruction images. The years of clinical experience, frequency of encountering osteoporosis patients, and frequency of lumbar spine image evaluation were collected. For the diagnostic criteria of osteoporosis, the T-score cutoff value of -2.5 was used in accordance with the guidelines, and the diagnostic performance of osteopenia, the T-score cutoff of -1.0 was also investigated.
RESULTS
Sixty-two evaluators (25 orthopedic surgeons and 37 non-orthopedic surgeons) were included in the final analysis. The overall percentage of correct answers at T-score cutoff -2.5 was 54.0±8.5% for radiographs and 51.8±10.2% for CT images. Comparing orthopedic specialists and non-specialists, the correct response rates for osteoporosis (T score cutoff -2.5) were slight but statistically significantly better for specialists: radiographs (56.8±6.8% vs 52.1±1.4%; p=0.03) and CT images (58.2±8.6% vs 47.4±9.1%; p<0.001). There was no correction between years of clinical experience and percentage of correct answers among board-certified orthopedic specialists.
DISCUSSION
Although orthopedic surgeons demonstrated better scores in intuitive correct diagnosis of osteoporosis than non-orthopedic specialists, the percentage of correct diagnosis was only slightly better than completely random choice. Surgeons should be aware of their own limitation for intuitive diagnosis, and we recommended that all high risk patients should undergo formal bone assessments, such as DXA, quantitative CT scan, before spine surgery.