Download Yearly Zoledronic Acid in Postmenopausal Osteoporosis

The
n e w e ng l a n d j o u r na l
of
m e dic i n e
c or r e sp ondence
Yearly Zoledronic Acid in Postmenopausal Osteoporosis
To the Editor: In their report on the Health Outcomes and Reduced Incidence with Zoledronic
Acid Once Yearly (HORIZON) Pivotal Fracture Trial,
Black et al. (May 3 issue)1 conclude that a onceyearly infusion of zoledronic acid reduces the risk
of vertebral, hip, and other fractures. Although
this is a well-designed and well-done study, we
are concerned about the increase in atrial fibrillation in patients treated with zoledronic acid.
Hypocalcemia as a cause was ruled out. There is
no report of hypokalemia during the study. Another electrolyte imbalance, hypomagnesemia, is
reported as a side effect of zoledronic acid2 and
can cause cardiac arrhythmias.3,4 It would be interesting to know the incidence of hypomagnesemia during the HORIZON study, mainly among
the patients with atrial fibrillation.
Forty-nine patients in the zoledronic-acid group
had transient hypocalcemia at days 9 to 11 after
the infusion. There are several case reports describing hypocalcemia during treatment with zoledronic acid, inducing a compensatory increase in
parathyroid hormone.5 In the study by Black et al.,
however, we are not informed about the mean
change in calcium and parathyroid hormone concentrations during treatment.
Ron N.J. de Nijs, M.D., Ph.D.
Antonius A.A. Westgeest, M.D., Ph.D.
Máxima Medical Center
5631 BM Eindhoven, the Netherlands
[email protected]
To the Editor: Black et al. do not have a satisfactory explanation for their report of an increased
incidence of serious atrial fibrillation in patients
receiving zoledronic acid. Rapid atrial fibrillation,
severe hypocalcemia, and a marked increase in
the serum parathyroid hormone level were presenting features in a woman with hypovitaminosis D who had been treated with 60 mg of pamidronate 3 weeks before the episode, as reported
in the Journal in 2003.1 Clinically, hypocalcemia
may lower the ejection fraction, leading to myocardial dysfunction. The report by Black et al.
contains little evidence that serum calcium was
maintained at pretreatment levels after the administration of zoledronic acid, beyond the statement that the drug had “little or no effect” on
levels 9 to 11 days after infusion.
We previously reported that zoledronic acid (at
a dose of 4 mg) lowered serum calcium levels,
usually into the lowest quartile of the reference
range.2 This finding was evident from the second
day (Fig. 1). Calcium levels remained suppressed
this week’s letters
711
Yearly Zoledronic Acid in Postmenopausal
­Osteoporosis
715
Antiretroviral Drugs and Myocardial Infarction
717
A Medical Mystery: High Blood Pressure —
The Answer
718
Management of Pulmonary-Artery Dissection
719
Wine and Sensitization to Hymenoptera Venom
720
Gadolinium and Nephrogenic Systemic Fibrosis
723
Thunderstorms and iPods — Two Reports
of the Same Case
1. Black DM, Delmas PD, Eastell R, et al. Once-yearly zole-
dronic acid for treatment of postmenopausal osteoporosis.
N Engl J Med 2007;356:1809-22.
2. Sorscher SM. Electrolyte abnormalities with zoledronic acid
therapy. Cancer J 2002;8:348-9.
3. Abbott LG, Rude RK. Clinical manifestations of magnesium
deficiency. Miner Electrolyte Metab 1993;19:314-22.
4. Whang R, Hampton EM, Whang DD. Magnesium homeostasis and clinical disorders of magnesium deficiency. Ann Pharmacother 1994;28:220-6.
5. Peter R, Mishra V, Fraser WD. Severe hypocalcaemia after
being given intravenous bisphosphonate. BMJ 2004;328:335-6.
n engl j med 357;7 www.nejm.org august 16, 2007
Downloaded from www.nejm.org at MAXIMA MEDISCH CENTRUM on August 1, 2008 .
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
711
Mean Serum Calcium Level (mg/dl)
The
n e w e ng l a n d j o u r na l
9.2
8.8
8.4
Placebo (16 patients)
8.0
Zoledronic acid (15 patients)
0
2
4
6
8
10
Day
Figure 1. Serial Changes in Mean Serum Calcium Levels in Patients with
Stroke Treated with Zoledronic Acid or Placebo.
The dashed line indicates the lower limit of the laboratory reference range
for calcium at 8.4 mg per deciliter (P<0.001 for the comparison between
zoledronic acid and placebo for the interaction between treatment and
RETAKE
1st
AUTHOR: Black
ICM
time when modeled
with
random effects by repeated-measures analysis of
2nd
1 of 1
REGdenote
F FIGURE:
variance). I bars
95% confidence
intervals.
3rd
CASE
EMail
m e dic i n e
dronic acid in postmenopausal women with low bone mineral
density. N Engl J Med 2002;346:653-61.
4. Kim HW, Park CW, Shin YS, et al. Calcitriol regresses cardiac hypertrophy and QT dispersion in secondary hyperparathyroidism on hemodialysis. Nephron Clin Pract 2006;102:c21-c29.
9.6
0
of
Revised
ARTIST: ts
Line
4-C
H/T
H/T
Combo
partly
recovered
SIZE
22p3
To the Editor: Black et al. report a significant
increase in the risk of atrial fibrillation, classified
as a serious adverse event, among patients treated with intravenous zoledronic acid. In a letter in
the same issue of the Journal, Cummings et al.
report a trend toward an increased risk of atrial
fibrillation among patients treated with oral alendronate.1
To determine whether there was a similar effect with oral risedronate, we evaluated the incidence of nonadjudicated adverse events of atrial
fibrillation and cerebrovascular accident and of
death from these events in placebo-controlled,
phase 3 clinical trials of risedronate for the treatment of osteoporosis. These trials followed
approximately 15,000 patients for up to 3 years
(Table 1).
In the risedronate group, as compared with
the placebo group, there was no significant difference in the incidence of atrial fibrillation (classified as an adverse event or a serious adverse
event), cerebrovascular accident, or death associated with cardiovascular adverse events. The difference in the rate of death from cerebrovascular
accident (P = 0.003) was consistent with findings
reported previously.2 These data do not support
a causal association between atrial fibrillation and
the use of risedronate.
Roger Karam, M.D.
and had only
10 days later. Two
AUTHOR,
PLEASE patients
NOTE:
thirds of
our treated
also had phosphate
Figure has been redrawn and type has been reset.
levels below
lower
limit of the normal range
Pleasethe
check
carefully.
within 10 days.
JOB: 356xx Reid et al. reported decreased
ISSUE: 08-26-07
calcium and
phosphate levels in women with osteoporosis
1 month after infusion of zoledronic acid.3
Women who were treated with quarterly infusions
(at a dose of 1 mg) had mean values of parathyroid hormone that were 30% higher than baseline
values 3 months after their last infusion. Prolonged
relative hypocalcemia and associated secondary
hyperparathyroidism might underlie cardiac ar- Procter & Gamble Pharmaceuticals
rythmogenesis, as seen in the secondary hyper- Mason, OH 45040
parathyroidism of patients undergoing hemodi- [email protected]
alysis.4 Further studies are needed to document John Camm, M.D.
the duration of relative hypocalcemia and its likely St. George’s University of London
preventable determinants — for instance, vita- London SW17 0RE, United Kingdom
Michael McClung, M.D.
min D insufficiency — before treatment.
Oregon Osteoporosis Center
Kenneth E.S. Poole, B.M., Ph.D.
Portland, OR 97213
Stephen Kaptoge, Ph.D.
Dr. Camm reports receiving consulting fees from Procter
Jonathan Reeve, D.M., D.Sc.
& Gamble, Merck, and Novartis; and Dr. McClung, research
Enon
University of Cambridge
Cambridge CB2 2QQ, United Kingdom
[email protected]
1. Rosen CJ, Brown S. Severe hypocalcemia after intravenous
bisphosphonate therapy in occult vitamin D deficiency. N Engl J
Med 2003;348:1503-4.
2. Poole KE, Loveridge N, Rose CM, Warburton EA, Reeve J.
A single infusion of zoledronate prevents bone loss after stroke.
Stroke 2007;38:1519-25.
3. Reid IR, Brown JP, Burckhardt P, et al. Intravenous zole-
712
grants and consulting fees from Amgen, Lilly, Merck, Novartis,
Procter & Gamble, Roche, and Sanofi-Aventis and lecture fees
from Lilly, Merck, and Sanofi-Aventis. No other potential conflict of interest relevant to this letter was reported.
1. Cummings SR, Schwartz AV, Black DM. Alendronate and
atrial fibrillation. N Engl J Med 2007;356:1895-6.
2. Steinbuch M, D’Agostino RB, Mandel JS, et al. Assessment of
mortality in patients enrolled in a risedronate clinical trial program: a retrospective cohort study. Regul Toxicol Pharmacol
2002;35:320-6.
n engl j med 357;7 www.nejm.org august 16, 2007
Downloaded from www.nejm.org at MAXIMA MEDISCH CENTRUM on August 1, 2008 .
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
correspondence
Table 1. Demographic Characteristics and Incidence of Atrial Fibrillation, Cerebrovascular Accident, and Death
among Patients with Osteoporosis Receiving Risedronate.*
Placebo
(N = 5048)
Variable
Risedronate
P Value†
2.5 mg
(N = 4998)
5 mg
(N = 5020)
Age — yr
73.4±9.2
73.6±9.1
73.5±9.3
0.44
Duration of drug exposure — mo
23.8±13.2
21.5±12.6
24.0±13.2
0.50
Adverse event
70 (1.4)
66 (1.3)
70 (1.4)
1.0
Serious adverse event
24 (0.5)
24 (0.5)
29 (0.6)
0.49
Adverse event
77 (1.5)
71 (1.4)
70 (1.4)
0.62
Associated death
24 (0.5)
15 (0.3)
7 (0.1)
0.003
96 (1.9)
83 (1.7)
80 (1.6)
0.25
Atrial fibrillation — no. (%)
Cerebrovascular accident — no. (%)
Death associated with cardiovascular adverse
event — no. (%)
*Plus–minus values are means ±SD. Data are from the Vertebral Efficacy with Risedronate Therapy Trial, the Bone
Mineral Density Trial, the Risedronate 5 mg Daily Prevention Trial, the Glucocorticosteroid-Induced Osteoporosis
Prevention and Treatment Trials, and the Hip Intervention Program Trial.
†P values are reported only for the comparison between the group receiving 5 mg of risedronate and the placebo
group, since 5 mg is the approved dose and the 2.5-mg dose was discontinued early in some trials.
To the Editor: The report by Black et al. adds an
important new dimension to the management of
postmenopausal osteoporosis. However, a number of trials have already shown the efficacy of
bisphosphonates in preventing osteoporotic fractures.1-3 Since Black et al. have been in the forefront of research in this area,1,2 we are concerned
about their choice of a placebo-controlled trial to
show the usefulness of zoledronic acid in osteoporosis. In our view, only a lack of evidence regarding the effectiveness of a trial drug should
justify conducting a placebo-controlled trial. We
believe that the data available before the start of
the study by Black et al., in 2002,4 should have led
to a trial comparing oral (daily, weekly, or monthly) bisphosphonates with yearly parenteral zoledronic acid. Such a trial would have been more
pertinent, both clinically and ethically.
Marwan A. Najib, M.R.C.P.
Imran Aziz, F.R.C.P.
Royal Albert Edward Infirmary
Wigan WN1 2NN, United Kingdom
[email protected]
1. Black DM, Cummings SR, Karpf DB, et al. Randomised trial
of effect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet 1996;348:1535-41.
2. Cummings SR, Black DM, Thompson DE, et al. Effect of
alendronate on risk of fracture in women with low bone density
but without vertebral fractures: results from the Fracture Intervention Trial. JAMA 1998;280:2077-82.
3. Harris ST, Watts NB, Genant HK, et al. Effects of risedronate treatment on vertebral and nonvertebral fractures in women
with postmenopausal osteoporosis: a randomized controlled
trial. JAMA 1999;282:1344-52.
4. Harris ST. Bisphosphonates for the treatment of postmenopausal osteoporosis: clinical studies of etidronate and alendronate. Osteoporos Int 2001;12:Suppl 3:S11-S16.
To the Editor: In the editorial by Compston1
accompanying the article by Black et al., there is a
sentence that is misleading. With respect to zoledronic-acid treatment in postmenopausal women,
Compston writes that “the absence of long-term
adverse effects on renal function is reassuring,”
citing a letter I wrote to the Journal in 2003.2 In
that letter, I wrote that some patients taking zoledronic acid had long-term renal effects — namely
renal deterioration progressing to renal failure,
the need for dialysis, and even death. The renal
failure was caused by acute tubular necrosis, which
is not always reversible.3
A re-review of the cases I discussed in my
letter showed that 22 patients continued to have
elevated serum creatinine levels at the time of
recovery, as compared with the levels at baseline.
Five of these patients were treated with dialysis.
Thirteen patients had received only one dose of
n engl j med 357;7 www.nejm.org august 16, 2007
Downloaded from www.nejm.org at MAXIMA MEDISCH CENTRUM on August 1, 2008 .
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
713
The
n e w e ng l a n d j o u r na l
zoledronic acid before having renal failure. In this
group, six patients underwent dialysis. Two of
these patients died, in addition to a third patient.
Furthermore, five patients continued to have elevated serum creatinine levels after discontinuation of zoledronic acid.
The Food and Drug Administration continues
to have concerns about the safety of zoledronic
acid. In January 2005, the product labeling was
updated to include medication dosing based on
the baseline creatinine clearance.4
Jennie T. Chang, Pharm.D.
Food and Drug Administration
Silver Spring, MD 20993
[email protected]
1. Compston J. Treatments for osteoporosis — looking beyond
the HORIZON. N Engl J Med 2007;356:1878-80.
2. Chang JT, Green L, Beitz J. Renal failure with the use of
zoledronic acid. N Engl J Med 2003;349:1676-8.
3. Markowitz GS, Fine PL, Stack JI, et al. Toxic acute tubular
necrosis following treatment with zoledronate (Zometa). Kidney
Int 2003;64:281-9.
4. Zometa product label. East Hanover, NJ: Novartis Pharmaceuticals, 2002. (Accessed July 26, 2007, at http://www.us.
zometa.com/info/about/index.jsp.)
of
m e dic i n e
phosphorous: 9 to 11 days after the initial infusion, there was a slightly larger decrease in the
zoledronic-acid group than in the placebo group,
but in the zoledronic-acid group, phosphorus levels
did not differ between women with atrial fibrillation and those without atrial fibrillation. Magnesium and potassium levels did not differ between the two study groups and were similar in
women with and those without atrial fibrillation.
As discussed in our article, episodes of atrial
fibrillation did not cluster in time immediately after any infusion, when serum electrolytes are most
affected. Taken together, these findings support
our speculation that if the observed increase in
atrial fibrillation was in fact related to the administration of zoledronic acid, it was probably not
due to changes in serum electrolytes.
We welcome the data from Karam et al. on
risedronate and look forward to the publication
of data on other bisphosphonates to further elucidate this finding.
The investigators and industry sponsor of our
trial incorporated ethical considerations regarding placebo controls, such as those raised by
Najib and Aziz, into the study design. The protocol specified that patients had to be unable or
unwilling to take oral bisphosphonates. All women
were counseled regarding the risk of fracture and
the availability of approved osteoporosis medications. One innovation of the trial was that women
who were receiving nonbisphosphonate treatments
for osteoporosis (including hormone therapy,
raloxifene, and calcitonin) were included (“stratum 2”), and all patients were free to begin any
of these treatments during the study while continuing to receive the study treatment. Furthermore, we monitored patients for excessive bone
loss or multiple fractures, and women who met
either criterion were again counseled about alternative treatment options. We agree that a trial
with an active comparator would be of great clinical interest, but the numbers required for such a
trial to demonstrate equivalence or superiority
with respect to the risk of fracture are prohibitive (20,000 to 30,000 patients).
The authors reply: Poole et al. and de Nijs and
Westgeest note the potential relationship between
clinically important changes in electrolytes and
the occurrence of cardiac arrhythmias. We also
evaluated electrolyte changes in the HORIZON
trial. There were significant differences between
study groups in the change in serum calcium from
the level before the first infusion of zoledronic
acid to the level 9 to 11 days after the infusion,
but the magnitude of the difference was relatively small (a reduction of 0.2±0.5 mg per deciliter
in the zoledronic-acid group vs. an increase of
0.03±0.4 in the placebo group). No difference
was evident at 12 months, and no significant
changes occurred with subsequent doses. In the
zoledronic-acid group, the change in calcium levels after the first infusion did not differ between
women with atrial fibrillation and those without
atrial fibrillation. Over the 3 years of the study,
mean serum calcium levels increased in both
study groups, although the mean increase was
slightly larger in the placebo group than in the Dennis Black, Ph.D.
zoledronic-acid group. The difference in the mean University of California, San Francisco
serum calcium level between study groups 9 to San Francisco, CA 94107
11 days after the initial infusion (0.2 mg per deci- [email protected]
liter) was less than that observed by Poole et al. at Erik Eriksen, M.D.
10 days (0.6 mg per deciliter).1
Novartis Pharma
In our study, a similar pattern was seen for CH-4002 Basel, Switzerland
714
n engl j med 357;7 www.nejm.org august 16, 2007
Downloaded from www.nejm.org at MAXIMA MEDISCH CENTRUM on August 1, 2008 .
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
correspondence
Deborah Sellmeyer, M.D.
effects associated with zoledronic acid but, rather,
to provide a context in which the long-term renal
safety in the HORIZON study could be regarded
1. Poole KE, Loveridge N, Rose CM, Warburton EA, Reeve J. as reassuring. As Dr. Chang points out, long-term
A single infusion of zoledronate prevents bone loss after stroke. adverse effects of zoledronic acid on renal funcStroke 2007;38:1519-25.
tion have been documented in a small number of
patients and remain a potential concern.
The editorialist replies: In my editorial, the Juliet Compston, M.D., F.R.C.P.
reference to Dr. Chang’s letter was not intended University of Cambridge School of Clinical Medicine
to support the absence of long-term adverse renal Cambridge CB2 2QQ, United Kingdom
University of California, San Francisco
San Francisco, CA 94107
Antiretroviral Drugs and the Risk of Myocardial Infarction
To the Editor: We agree with the Perspective
article by Hughes and Williams (April 26 issue)1
that a more detailed presentation of data from the
well-conducted Data Collection on Adverse Effects
of Anti-HIV Drugs (DAD) Study Group collaboration2 may help to disentangle several variables that
potentially affect the risk of myocardial infarction among persons infected with human immunodeficiency virus type 1 (HIV-1). First, data on the
nadir CD4+ lymphocyte count, but not the recent
CD4+ count, were presented in relation to the risk
of myocardial infarction. Inflammation may trigger an acute myocardial infarction by activating
prothrombotic mechanisms, and the patient’s
immune status at the time of the myocardial infarction is most relevant in this regard. Moreover, inflammation increases markedly with the
onset of clinical AIDS.3 Therefore, threshold models relating a low CD4+ count (i.e., <200 cells per
cubic millimeter) with myocardial infarction may
be more biologically plausible than linear models.
Second, what was the hazard ratio associated
with the duration of nucleoside reverse-transcriptase inhibitor use? As with protease inhibitors,
previous data have implicated constituents of the
nucleoside reverse-transcriptase inhibitor class in
metabolic abnormalities.4 Controlling for the use
of nucleoside reverse-transcriptase inhibitors attenuated the hazard ratio relating the use of
protease inhibitors with the risk of myocardial
infarction; this suggests the presence of an association between the use of nucleoside reversetrancriptase inhibitors and myocardial infarction.
Robert C. Kaplan, Ph.D.
Albert Einstein College of Medicine
Bronx, NY 10461
[email protected]
Phyllis C. Tien, M.D.
University of California, San Francisco
San Francisco, CA 94143
Jason Lazar, M.D.
State University of New York Downstate Medical Center
Brooklyn, NY 11203
1. Hughes MD, Williams PL. Challenges in using observational
studies to evaluate adverse effects of treatment. N Engl J Med
2007;356:1705-7.
2. The DAD Study Group. Class of antiretroviral drugs and the
risk of myocardial infarction. N Engl J Med 2007;356:1723-35.
3. Lau B, Sharrett AR, Kingsley LA, et al. C-reactive protein is a
marker for human immunodeficiency virus disease progression.
Arch Intern Med 2006;166:64-70.
4. Brown TT, Li X, Cole SR, et al. Cumulative exposure to nucleoside analogue reverse transcriptase inhibitors is associated
with insulin resistance markers in the Multicenter AIDS Cohort
Study. AIDS 2005;19:1375-83.
To the Editor: The DAD Study Group observed
a relationship between exposure to protease inhibitors and myocardial infarction. No association was found between HIV-related factors, such
as the nadir CD4+ lymphocyte count or the peak
HIV-1 RNA level, and the risk of myocardial infarction. However, HIV-related factors might have
been confounded with treatment effects. Patients
with myocardial infarction had a lower rate of
viral suppression, although more of them had received antiretroviral treatment. Therefore, the
analysis should have been adjusted for a treatment effect on viral suppression. Can the lower
rate of viral suppression among patients with myocardial infarction at least partly be explained by a
higher rate of interruption of treatment with protease inhibitors?1
In addition, we think it is not appropriate to
incorporate high-density lipoprotein (HDL) cholesterol levels into the multivariate model only in
a time-updated manner. HDL cholesterol levels
n engl j med 357;7 www.nejm.org august 16, 2007
Downloaded from www.nejm.org at MAXIMA MEDISCH CENTRUM on August 1, 2008 .
Copyright © 2007 Massachusetts Medical Society. All rights reserved.
715