Which measuring site in ankylosing spondylitis is best to detect bone loss and what predicts the decline: results from a 5-year prospective study

Table 4 Multiple linear regression analyses with ΔBMD (change in %) as a dependent variable

	Femoral neck		Total hip		Spine AP		Spine lateral		Spine vBMD		Total radius
Constant	20.4		1.5		–13.4		–47.3		–40.5		10.4
	β	p	β	p	β	p	β	p	β	p	β	p
Baseline
BMD at same site, g/cm²	–8.24	0.036	0.98	0.67	0.80	0.80	0.81	0.89	–16.3	0.42	2.89	0.60
Age, years	–0.12	0.006	–0.05	0.063	–0.04	0.31	–0.04	0.50	–0.07	0.22	–0.11	< 0.001
Sex	–1.61	0.11	–1.68	0.007	–2.06	0.051	–1.70	0.27	–1.47	0.25	–0.47	0.58
Pack years	0.13	0.004	0.08	0.003	0.09	0.054	0.12	0.086	0.09	0.12	–0.04	0.16
mSASSS, score	0.06	0.041	0.005	0.76	–0.03	0.27	–0.08	0.052	–0.06	0.10	0.02	0.26
During follow-up
Time between DXA, months	–0.19	0.41	–0.02	0.86	0.24	0.31	0.81	0.016	0.77	0.009	–0.14	0.36
Δweight, kg	0.03	0.72	0.24	< 0.001	0.37	< 0.001	0.49	< 0.001	0.51	< 0.001	0.08	0.18
NSAID-index, 0–100	0.02	0.064	0.01	0.097	0.01	0.42	0.01	0.56	0.00	0.89	0.01	0.26
Use of TNFi, 0–1	–1.53	0.11	1.23	0.086	3.15	0.012	2.17	0.22	1.06	0.49	–0.65	0.41
Use of bisphosphonate, 0–1	4.64	0.013	5.61	< 0.001	13.7	< 0.001	15.5	< 0.001	11.6	< 0.001	0.83	0.48
Mean CRP last 5 years, mg/L	–0.15	0.046	NA		NA		NA		NA		–0.09	0.072
ΔESR, mm/h	NA		–0.10	< 0.001	–0.12	0.004	–0.16	0.007	–0.13	0.012	NA
R ²	0.22		0.44		0.41		0.33		0.34		0.20

β are unstandardized coefficients
Coding for sex: man = 0, woman = 1
Significant β and p values are shown in bold typeface
AP anterior-posterior, BMD bone mineral density, CRP C-reactive protein, DXA dual-energy x-ray absorptiometry, ESR erythrocyte sedimentation rate, mSASSS modified Stoke Ankylosing Spondylitis Spine Score, NA not available, NSAID nonsteroidal anti-inflammatory drug, TNFi tumor necrosis factor inhibitor, vBMD volumetric BMD

ISSN: 1478-6362