This article is more than 5 years old. Information may no longer be current.
Height loss in postmenopausal women may indicate spinal fracture
Briot K. CMAJ. 2010 Apr 6;182(6):558-62.
Click Here to Manage Email Alerts
Loss of height in postmenopausal women may indicate vertebral fracture, according to the results of a recently published study in the Canadian Medical Association Journal.
Between December 2007 and May 2008, French researchers asked 1,779 general practitioners to each recruit their first five female patients who were older than 60 years, regardless of the reason for consultation. A questionnaire was used to collect data on demographic and clinical variables, history of osteoporosis and current treatments used to prevent osteoporosis.
The researchers used three assessments of height — tallest height in early adulthood recalled by the patient, the estimated current height reported by the patient at the time of their visit and the patient’s current measured height. Loss of height was defined as the difference between the patient’s tallest recalled height and current measured height.
The mean loss of height among the 8,610 patients included in the analysis was 4.5 cm, with a mean current reported height measuring 2.1 cm lower than the tallest recalled height and 2.4 cm lower than the measured current height.
“We observed a mean loss of height of 4.5 cm since early adulthood in a large population of postmenopausal women in primary care practices,” Karine Briot, MD and colleagues wrote in their study.
“We found that the risk of an existing vertebral fracture was significantly higher among patients with height loss of at least 4 cm,” they wrote.
According to the findings, the best predictors of a loss of height measuring 3 cm or more were age, previous vertebral or nonvertebral fracture, thoracic kyphosis, scoliosis, back pain and osteoporosis. The authors concluded that the measurement of height loss could be an accurate method for detecting vertebral fractures.
The analysis of radiographic reports of fractures demonstrated that a height loss of >4 cm had a 74% specificity and a 49% sensitivity for predicting an existing vertebral fracture. The flatness of the receiver operator curve demonstrates the poor performance characteristics of height loss as a predictor of fracture. The authors did not report a positive predictive value of >4 cm of height loss.
In the interpretation, the authors suggested that height loss may be an accurate method to predict the presence of vertebral fractures. This conclusion is not well supported by the data. Less than 50% of patients with an existing fracture will be identified with a 4 cm threshold.
The value of screening for asymptomatic vertebral fractures is also unclear. Screening studies are most useful when identification of a disease or disorder may affect natural history or may direct treatment that will be more effective in earlier stages of disease.
Information on age, back pain, osteoporosis, and previous fracture are already known in this patient population, and these factors together are more sensitive and specific in identifying vertebral fractures. The authors do demonstrate effectively that measuring height and weight is more accurate than patient self-reporting.
– Sigurd H. Berven, MD
Orthopedics Today Editorial Board Member
Follow OrthoSuperSite.com
on Twitter