Arthritis & Rheumatism, Volume 63,
November 2011 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Chicago, Illinois November 4-9, 2011.

The Association Between Erosive Hand Osteoarthritis and Systemic Bone Mineral Density.

Haugen1,  Ida K., Felson2,  David T., Englund3,  Martin, Wang2,  Ke, Aliabadi4,  Piran, Guermazi2,  Ali, Roemer2,  Frank

Diakonhjemmet Hospital, Oslo, Norway
Boston University, Boston, MA
Lund University, Lund, Sweden
Brigham and Women's Hospital, Boston


Erosive hand osteoarthritis (OA) is characterized by attrition of the joint plate, which could be due to inadequate bone quality. However, the knowledge about bone mineral density (BMD) in erosive hand OA is scarce. Thus, our aim was to examine the association between erosive hand OA and systemic BMD.


We included 2236 participants from the Framingham OA study with hand radiographs and femoral neck BMD (DEXA). The bilateral distal (DIP) and proximal interphalangeal (PIP), metacarpophalangeal (MCP), thumb base and wrist joints were scored for radiographic hand OA (Kellgren-Lawrence grade (KLG)>=2) and central erosions. Central erosions were present in DIP/PIP joints only. Thus, we defined erosive OA as >=1 DIP/PIP joint with KLG>=2 and central erosion in the same joint, and non-erosive OA as >=1 DIP/PIP joint with KLG>=2 and no central erosions. The unexposed group included those with no DIP/PIP OA (i.e., no OA or isolated MCP/thumb base/wrist OA). We used logistic regression to explore whether BMD (as continuous variable or age- and sex-specific BMD tertiles) was associated with presence of any DIP/PIP OA. We also performed linear regression with presence of erosive or non-erosive OA as independent variable and BMD as continuous outcome variable (one model). The analyses were repeated with the number of erosive and non-erosive joints as independent variables (one model). All analyses were adjusted for age, sex and BMI.


The 2236 participants (1281 women) had a mean (SD) age of 64.5 (9.0) years and a mean (SD) BMI of 28.6 (5.5) kg/m2. Erosive and non-erosive OA were present in 221 (9.9%) and 761 (34.0%) persons respectively. In those with erosive OA, the median (IQR) numbers of erosive and non-erosive joints were 3 (1–5) and 5 (3–8) respectively. In those with non-erosive OA, the median (IQR) number of OA joints was 2 (1–5).

In adjusted analyses, higher BMD was associated with higher odds of any OA (OR=1.90, 95%CI 0.91–3.98; p=0.09). The odds of any OA was significantly higher in those with the highest BMD tertile (OR=1.37, 95%CI 1.08–1.73; p=0.01), while not significant for those in the middle tertile (OR=1.21, 95%CI 0.96–1.53; p=0.11) compared to those in the lowest tertile.

There were small differences in BMD between the unexposed, erosive and non-erosive OA participants: The adjusted LS-mean (SD) BMD was 0.92 (0.01), 0.93 (0.02) and 0.93 (0.01), respectively. There was a trend towards higher BMD in participants with erosive and non-erosive DIP/PIP OA compared to the unexposed group (table).

Table. The association between erosive and non-erosive hand OA and BMD of the femoral neck (linear regression adj. for age, sex and BMI).

 Adjusted beta (95% CI); p-value
Presence of erosive and non-erosive DIP/PIP OA vs. no DIP/PIP OA as reference (one model) 
No DIP/PIP OA (reference)0.00
Erosive DIP/PIP OA0.01 (-0.01, 0.03); p=0.33
Non-erosive DIP/PIP OA0.01 (0.00, 0.02); p=0.08
Number of erosive and non-erosive DIP/PIP joints (one model) among all participants: 
Number of erosive joints-0.001 (-0.005, 0.003); p=0.65
Number of non-erosive joints0.002 (-0.001, 0.004); p=0.15

Analyses with exclusion of those with isolated MCP/thumb base/wrist OA from the unexposed group did not appreciably change the results.


There was a trend towards higher BMD in participants with DIP/PIP OA. Erosive hand OA was not associated with lower systemic BMD, suggesting that local structural impairment of periarticular bone quality may play a role in erosive hand OA not sufficiently covered by systemic BMD measurements.

To cite this abstract, please use the following information:
Haugen, Ida K., Felson, David T., Englund, Martin, Wang, Ke, Aliabadi, Piran, Guermazi, Ali, et al; The Association Between Erosive Hand Osteoarthritis and Systemic Bone Mineral Density. [abstract]. Arthritis Rheum 2011;63 Suppl 10 :1073

Abstract Supplement

Meeting Menu