Hide Sidebar

The interval of osteoporosis screening should be determined by baseline T scores.

The interval of osteoporosis screening should be determined by baseline T scores.

Critique date: Wed, 03/14/2012 - 9:04pm
Printer-friendly versionPrinter-friendly version
Institution: 
Mount Sinai School of Medicine
   
Article Citation
Citation: 
Gourlay ML, Fine JP, Preisser JS, May RC, Li C, Lui LY, Ransohoff DF, Cauley JA, Ensrud KE. Bone-Density Testing Interval and Transition to Osteoporosis in Older Women. NEJM 2012; 366:225-33.
   
Clinical Bottom Lines

Screening for osteoporosis is recommended for all patients 65 years and older. However, it is still not known how frequently bone density examinations should be repeated.

Nearly 5000 women were followed for up to 16 years for the development of osteoporosis with repeated bone density examinations performed during follow up. 10% of participants developed osteoporosis at 1 year, 4 years and 17 years according to baseline bone density scores showing severe osteopenia, moderate osteopenia and mild osteopenia or normal bone density, respectively. The authors concluded that for patients with normal bone density or mild osteopenia, repeat DXA should be performed at 10 years. For patients with moderate or severe baseline bone density, repeat DXA should be performed at 5 and 1-2 years respectively. However, the results also suggest that patients over 85 may need more frequent testing as they may progress to osteoporosis more quickly. 

Disclaimer: 
This is a review of the validity of a single study; the ‘bottom lines’ do not reflect comparison with the rest of the literature on this subject.
   
Methods
Type of Study: 
Prognosis
Study Design: 
Cohort study
Follow-up Period: 
16 years
Setting: 
Community-based in USA
Patient Population: 

9704 community dwelling women aged 65 and over were recruited. They were 99% Caucasian, 74% between the ages of 67-74, 17% Estrogen replacement users, 11% current or past glucocorticoid users.

Significant Exclusions: 

Participants were excluded from recruitment if they had bilateral total hip replacements or were unable to walk without assistance.

Of the original cohort, 4747 participants were not included in the final analysis for this study. These participants were excluded for the following reasons: missing baseline DXA (1190) or missing data on the baseline DXA (17), only had 1 DXA during follow up (1083), had osteoporosis at baseline (2197), or had a fracture or started osteoporosis treatment prior the second DXA (260).

Intervention/Exposure: 

Participants were originally recruited for a baseline medical exam and interview. At year 2 they were asked to complete a bone density examination. All participants who completed the baseline interview and DXA were then invited to participate in longitudinal follow up in which repeat DXAs were performed at years 6, 8, 10 and 16. 

Outcome Measures: 

The cumulative incidence of osteoporosis was measured over time and converted to a time to event statistical model using competing risk analysis. Subsequently the cohort was stratified by their baseline bone density into 4 groups: normal, mild osteopenia, moderate osteopenia and severe osteopenia. The duration from baseline to when 10% of each group of the cohort develops osteoporosis was calculated. In addition, the cumulative incidence of hip or vertebral fracture was also estimated and fitted in a time to event model. The time intervals for 2% of each group to develop hip or vertebral fracture were also calculated. The authors did not specify why the 10% and 2% thresholds were used. 

Participant Follow-up: 
51% of the original cohort.
   
Results
Results: 

The unadjusted estimated intervals for 10% of participants to transition to osteoporosis according to baseline bone density were as follows:

  • 17.4 years for normal bone density at baseline
  • 16.5 years for mild osteopenia at baseline
  • 4.6 years for moderate osteopenia at baseline
  • 1 year for severe osteopenia at baseline

Of note, when adjusted for osteoporosis risk factors (age, BMI, estrogen or glucocorticoid use, prior fracture, smoking status, rheumatoid arthritis) the results were similar (16.8, 17.3, 4.7 and 1.1 years according to the respective baseline DXA scores).

  • Women with older age progressed to osteoporosis more quickly. The transition from mild osteopenia to osteoporosis was 16.2 years for those aged 75 versus 11.8 years for those aged 85
  • Current estrogen use correlated with a slightly longer interval in both the moderate and severe osteopenia groups
  • BMI did not significantly affect outcome

The interval for 2% of participants to have a hip or vertebral fracture:

  • >15 years if had normal DXA or mild osteopenia at baseline
  • 5 years if had moderate to severe osteopenia at baseline 
   
Methodological Issues…
Methodological Issues and Applicability to Older Adults: 

Though this study certainly adds important knowledge to the natural history of osteopenia and the development of osteoporosis there are a number of limitations that need to be considered.

First, the poor overall follow up rate of this study is of particular concern. 2273 participants (23%) were not included in the final analysis due to incomplete DXA testing. Results from this large proportion of participants could significantly change the final results.

Second, the population used in this study was homogeneous with 99% white women. In clinical practice both men and women, and of all racial backgrounds are cared for with osteopenia and osteoporosis. This may affect the generalizability of the results. As well, women over 85 progressed to osteoporosis more quickly, thus rescreening intervals may need to be adjusted for the oldest old.

Third, though this study suggests that it is safe to wait 10 years prior to repeating DXA in women mild normal or mild osteopenia at baseline, the actual clinical effects of such a policy is unknown and needs to be investigated further. 

   
Topics
CAP Topics: 
   
Other Information
Funding Source and Role: 
None stated
Created By: 
Arielle Berger, Fellow, Brookdale Department of Geriatrics and Palliative Medicine, Mount Sinai School of Medicine

Disclaimer: This is a review of the validity of a single study; the ‘bottom lines’ do not reflect comparison with the rest of the literature on this subject.