OBJECTIVE Premature ovarian failure is associated with a reduction in
bone mineral density. As survival rates following treatment for haemat
ological malignancies improve, chemotherapy-induced ovarian failure is
becoming more common. However, there are few data concerning the impa
ct of this on bone mineral density (BMD), We have therefore measured t
he BMD in 33 women with ovarian failure following treatment with cytot
oxic chemotherapy. PATIENTS AND DESIGN We studied 33 women who receive
d combination chemotherapy for Hodgkin's disease (n=27), non-Hodgkin's
lymphoma (n=4), sarcoma (n=1) and acute myeloid leukaemia (n=1), The
mean (range) age of the subjects at the time of BMD measurement was 37
.5 (24-50) years and the mean (median: range) duration of amenorrhoea
was 49 (24: 5-277) months. Eleven women had received hormone replaceme
nt therapy (HRT) for a mean (range) duration of 25 (1-62) months. BMD
was measured by single photon absorptiometry or single X-ray absorptio
metry, and dual energy X-ray absorptiometry at the distal and proximal
radius, the femoral neck and the lumbar spine, respectively. BMD was
expressed as Z-scores and statistical analysis was performed using the
Wilcoxon matched-pairs signed-rank test. RESULTS There was no signifi
cant reduction in BMD at the hip, spine or forearm in the cohort as a
whole, although there was a trend to reduce bone density at all sites.
When patients who had received HRT were excluded from the analysis th
ere were small reductions in mean BMD at all sites, but this was only
statistically significant at the proximal forearm (Z-score=-0.65; P=0.
03). Mean BMD of the HRT-treated patients was normal at all sites. Onl
y seven patients (21%) had a BMD Z-score< -2 at any site. CONCLUSION I
t is inappropriate to assume that ovarian failure from different aetio
logies has a similar deleterious impact on the skeleton. Untreated pre
mature ovarian failure following cytotoxic chemotherapy results in som
e reduction in bone mineral density, but this is of a minor degree and
is less than that observed in other hypo-oestrogenic states. The reas
on for this is unclear but studies of residual hormone production in t
he cytotoxic-damaged ovary may provide an answer.