Low bone mineral density in MSM irrespective of HIV status

Simon Collins, HIV i-Base

A recurring difficulty in interpreting the high rates of reduced bone mineral density (BMD) in HIV positive people is the lack of appropriate reference data.

Results from a Dutch study in gay men (MSM) published in the Journal of Infectious Diseases provided new data reporting that BMD may be reduced in gay men, irrespective of HIV status. This is important given the many factors that relate to bone health, including weight/BMI, diet, smoking, exercise, age, testosterone (TST) levels, in addition to HIV and ARV treatment.

Marlous Grijsen, from the Center for Infection and Immunity, Amsterdam, and colleagues compared the BMD in primary HIV infection (diagnosed within six months, of infection, n=41), with chronic HIV infection (n=106), and in HIV negative controls (gay men with comparable lifestyles, n=30). [1] The study wanted to explain a high prevalence of low BMD in MSM during primary HIV infection, and those patients contributed data to this study. [2]

This was a prospective study for all newly enrolled MSM from 2008-2011, who were treatment-naive and aged 20-55 years when presenting for care at a single centre in Amsterdam. All patients received a Dexa scan and low BMD was defined as a z-score (matched by age, sex and race) of >/=2.0 SDs below the mean at the lumbar spine or hip (using the US NHANES IV population dataset as reference). Medical conditions known to affect bone metabolism were an exclusion criteria, including IDU, renal disease and corticosteroid use. Smoking, alcohol, diet, and fracture history were also taken together with testing for a wide panel of bone-related biochemical markers including P1NP (bone formation) and CTX (bone resorption) for the participants in primary infection and the controls, but unfortunately not for those in chronic infection.

The three groups were matched for age and race: mean (+ SD) age 38 years (+ 8), and 80% were white. However, HIV negative men were heavier (82 vs 73 kgs, p=0.009) with higher BMI (24.4 vs 22.7, p=0.04).

Baseline characteristics relating to BMD-associated lifestyle factors were similar in the primary vs control patients, apart from a trend towards higher smoking rates (44% vs 23%, p=0.07). Biochemical markers were also similar, with the few statistically significant differences still broadly within reference ranges: lower phosphate 0.93 (+ 0.18) vs 1.32 (+ 0.23) mmol/L (reference range 0.7-1.45 mmol/L) and P1NP levels 42 (+ 16) vs 52 (+ 13) ug/L; reference range 22-87; p=0.009) and higher CTX 288 levels (+ 196) vs 154 (+ 93) ng/L; reference range <584; p=0.001).

Median CD4 and viral load in the primary vs chronic HIV positive groups were 543 (± 253) vs 438 (± 214) cells/mm3, and 5.3 (± 1.2) vs 4.5 (± 0.9) log copies/ mL, respectively.

DEXA results indicated significantly lower BMD in all three patient groups, compared to NHANES reference levels, with lower SD by both t-score and z-score for all sites (lumber spine, femoral neck and total hip; except hip in HIV negative MSM). Notably, lumbar spine z-scores were -1.0, -1.1 and -0.8 in the primary, chronic and control groups respectively, though at other sites z-scores were less marked (approximately -0.1 to -0.3). However, there were no significant differences between groups for either score at any site, either by HIV status or duration of infection.

Low BMD at one or more sites was reported in 20% (8/41), 22% (23/106) and 13% (4/30) of the primary, chronic and control groups (p=0.6, for between arm difference).

In multivariate analysis, BMI was associated with low BMD at all sites (p<0.001) but not HIV status.


These results are important in suggesting that individual health and lifestyle factors may impact on bone health prior to HIV infection and that this may have not been adequately accounted for in earlier studies. Nevertheless, this is still a small cross-sectional study.

The low BMD in HIV negative controls is similar to rates reported in PEP studies (10% of MSM in a San Francisco PEP study and 12% of men in the iPrEX study (lumbar spine) had low BMD (<2 SD), though background risk factors were also likely to be different in each study population. [3, 4]

The results highlight the importance of finding appropriate controls for HIV studies. The does not detract from the importance of optimal management of bone health and longitudinal data from at least two US cohort studies have reported that low BMD in HIV positive people is associated with increased fracture risk, perhaps at a younger age to HIV negative people. [5, 6]

Differences in biomarkers of bone metabolism warrant further research, especially to help understand the impact of HIV treatment, and this should be helped by the bone sub-study of the START trial.


  1. Grijsen M et al. Low bone mineral density in men who have sex with men regardless of HIV status. J Infect Dis. (2012) doi: 10.1093/infdis/jis687. First published online: November 12, 2012 (PDF full text)
  2. Grijsen ML et al. High prevalence of reduced bone mineral density in primary HIV-1-infected men. AIDS 2010; 24:(14):2233-8. (10 September 2010).
  3. Lui AY et al. Bone mineral density in HIV-negative men participating in a tenofovir pre-exposure prophylaxis randomized clinical trial in San Francisco. PLoS One. 2011; 6(8): e23688. Published online 2011 August 29. doi: 10.1371/journal.pone.0023688.
  4. Mulligan K, Glidden D, Gonzales P, et al. Effects of FTC/TDF on bone mineral density in seronegative men from 4 continents: DEXA results of the global iPrEx study. 18th Conference on Retroviruses and Opportunistic Infections, 27 February -2 March 2011, Boston. Late breaker abstract 94LB.
  5. Dao C et al. Higher and increasing rates of fracture among HIV-infected persons in the HIV Outpatient Study compared to the general US population,
1994 to 2008. 17th CROI 2010. Oral abstract 128.
  6. Womack J et al. HIV-infection and fragility fracture risk among male veterans. 17th CROI 2010. Oral abstract 129.

Links to other websites are current at date of posting but not maintained.