|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Pulmonary Diseases (Dr. Dubois), Reinier de Graaf Groep Delft and Voorburg; the Department of Clinical Pharmacology (Dr. Röder), Academical Medical Center, University of Amsterdam; the Department of Pulmonary Diseases (Dr. Dekhuijzen), University Medical Center Nijmegen; Department of Clinical Epidemiology and Biostatistics (Dr. Zwinderman), Academic Medical Center, Amsterdam; and the Department of Internal Medicine and Endocrinology (Dr. Schweitzer), Reinier de Graaf Groep Delft and Voorburg, the Netherlands.
Correspondence to: Emile F. Dubois, MD, FCCP, Department of Pulmonary Diseases, Reinier de Graaf Groep Delft and Voorburg, Fonteynenburglaan 5, PBX 998, 2270 AZ Voorburg, the Netherlands; e-mail: dubois{at}rdgg.nl
| Abstract |
|---|
|
|
|---|
Design: Cross-sectional study.
Participants: Eighty-six white men with COPD selected from the outpatient clinic for pulmonary diseases.
Intervention: Data analysis from medical records, bone densitometry, and pulmonary function tests of consecutive selected patients. Inclusion period into the study was exactly 1 year.
Measurements and results: Ten patients received oral prednisolone daily (group 1). Eleven patients were treated for several exacerbations with multiple systemic prednisolone courses, up to a period of 2 weeks per course, with a cumulative dose of
1,000 mg (group 2). Likewise, 28 patients were treated with multiple systemic prednisolone courses, but with a cumulative dose < 1,000 mg (group 3). Thirty-seven patients were never treated with systemic prednisolone, and partly with inhaled corticosteroids (ICS) [group 4]. All groups were balanced for age and pack-years of smoking. In group 2, body mass index (BMI) and FEV1 were lowest and hyperinflation was highest. The cumulative systemic prednisolone dose was highest in group 1, irrespective of the additional ICS treatments. Dual energy x-ray absorptiometry scanning of the lumbar spine, total hip, and femoral neck regions revealed a T score
2.5 SD in 27 patients (31%), 31 patients (36%), and 34 patients (40%), respectively. BMD outcomes at any site were lower in patients receiving multiple systemic prednisolone courses > 1,000 mg, cumulatively (group 2), compared to the other groups, and these values were (mean ± 1 SD) 0.759 ± 0.238 g/cm2, 0.683 ± 0.115 g/cm2, and 0.686 ± 0.125 g/cm2, respectively (p < 0.0001). Multivariate regression analysis revealed a correlation between the cumulative dose of prednisolone in group 2 and BMD of the lumbar spine (adjusted r = 0.48; p < 0.01). At the total hip and femoral neck regions, only a correlation between BMI and BMD was observed (adjusted r = 0.65 and 0.58, respectively; p < 0.0001 for both sites).
Conclusions: Despite a far lower cumulative GC dose in comparison with patients treated with systemic corticosteroids continuously, after adjusting for BMI and lung function, osteoporosis of the lumbar spine was most frequent in patients receiving > 1,000 mg of prednisolone cumulatively, administered in multiple courses for the treatment of exacerbations of COPD.
Key Words: body mass index bone mineral density COPD glucocorticoid courses
| Introduction |
|---|
|
|
|---|
Fractures, as a result of osteoporosis, are one of the most serious complications of patients receiving GCs. The loss of bone mass and the associated risk of fracture appears to be strongly related to the magnitude of the GC dose received daily, and much less to the cumulative dose of GC that the patients received previously.2 In an older landmark study3 among asthma patients who were treated with at least 15 mg/d po of prednisolone for a minimum of 1 year, it was shown that the incidence of rib and vertebral fractures increased. Furthermore, the patients who received long-term GCs had significantly lower bone mineral densities (BMDs) at the distal and proximal radius.3 Similar outcomes were reported in older male patients with COPD receiving either inhaled or systemic GCs. The incidence of vertebral fractures rose compared to non-GC users (odds ratio, 1.38 and 2.15, respectively).4 Studies focusing on the safety of long-term inhaled GCs regarding the skeleton are difficult to interpret due to many confounding variables. In a previous study,5 however, evidence of a negative relationship between the total cumulative dose of inhaled GCs and BMD was demonstrated in patients with asthma; a doubling of the dose of inhaled GSs was associated with a decrease in spinal BMD at a rate of 0.16 SD per year.
The effects of different treatment regimens (continuous vs courses) on the amount of bone loss have not been studied extensively in humans. Biphasic loss of bone due to systemic GC treatment is comprised of a rapid initial bone loss of approximately 12% during the first few months, followed by a slower loss of approximately 2 to 5% annually.6 After initiation of high doses (three times 200 mg dexamethasone IV in 8 days for patients with rheumatoid arthritis and 1 g of methylprednisolone IV for 10 days for patients with multiple sclerosis), systemic GC-induced bone loss can be detected within a week. Most of the adverse effects of GCs occur in primarily trabecular bone, which is dominantly present in vertebrae.7 8
Treating acute airway inflammation in patients with COPD with short GC courses, usually lasting up to 2 weeks, is well accepted. Based on the pattern of GC-induced bone resorption, we hypothesize that treatment with repeated courses of systemic GCs leads to exceedingly high bone loss, probably as a result of repetitive, early-phase, osteoclast-mediated bone resorption. In the present study with patients with COPD, we compared BMD outcomes in different treatment groups and hypothesized that BMD values were lowest in those patients who received multiple systemic GC courses.
| Materials and Methods |
|---|
|
|
|---|
Subjects
Eighty-six men with documented COPD for at least 5 years were included in the study. The recruitment period lasted exactly 12 months. All patients were classified into one of four groups, according to their GC treatment history. Patient characteristics are listed in Table 1
.
|
Group 2: Group 2 consisted of 11 patients treated with multiple courses of oral or IV prednisolone, lasting up to 14 days, because of recurrent exacerbations. These patients had received at least 1,000 mg of prednisolone cumulatively, and were not receiving GC medication continuously.
Group 3: Group 3 consisted of 28 patients treated with multiple courses of oral or IV prednisolone. These patients had received < 1,000 mg prednisolone cumulatively, and were also not receiving continuous GCs.
Group 4: Group 4 consisted of 37 patients with mild-to-moderate COPD who were never treated with systemic GCs. Fifteen of these patients were daily ICS users when included in the study. These patients were considered the control group.
Smoking habits were expressed in pack-years. ICS prescribed in the study were beclomethasone and budesonide. These compounds were considered equivalent for the purpose of the study in terms of BMD impact.9 Estimates were drawn from the medical hospital records and expressed as "prescription years"; one prescription year with either ICS normalized to 800 µg/d during 1 year.
Pulmonary Function Tests
Pulmonary function test (PFT) data were obtained from the medical records and were all measured within 1 year prior to inclusion into the study. Function tests were established using Jaeger Masterlab (Erich Jaeger GmBH; Wuerzburg, Germany) and the (included) normal values were based on the European Respiratory Society standards.10
Total lung capacity (TLC), residual volume (RV), and FEV1 in absolute values and in percentages of predicted values were recorded.
Measurement of BMD
BMD was measured by dual energy x-ray absorptiometry using a densitometer (Expert-XL; Lunar Corporation; Madison, WI). The lumbar spine (L2L4), total hip, and femoral neck of each subject was evaluated. Calibration procedures were performed at least every day and again after each series of eight scans using the appropriate phantoms provided by the manufacturer. The coefficient of variation for BMD measurements was 1.3% at the lumbar spine and 1.6% at the total hip and femoral neck. Individual measurements for spinal, total hip, and femoral neck BMD were expressed as absolute values (grams per centimeter squared) as shown in Figure 1
.
|
- 2.5 SD) were studied with reference to a database of white male subjects provided by Lunar Corporation. The absolute BMD values corresponding with T score = - 1.0 SD and - 2.5 SD were, respectively, 1.121 g/cm2 and 0.941 g/cm2 at the lumbar spine, 0.975 g/cm2 and 0.780 g/cm2 at the total hip, and 0.968 g/cm2 and 0.773 g/cm2 at the femoral neck. Demographics, PFT, and BMD results were compared between groups using analysis of variance and the post hoc least-squares difference procedure. Correlations between BMI and PFT with BMD were analyzed and expressed using Pearson correlation coefficients. Relations between these variables were further analyzed using stepwise regression analysis. All p values < 0.05 were considered to be statistically significant.
| Results |
|---|
|
|
|---|
Smoking habits were 21 ± 19 pack-years, 30 ± 18 pack-years, 20 ± 15 pack-years, and 21 ± 15 pack-years in groups 1 through 4, respectively. These values did not differ significantly, but a trend for higher cigarette consumption was noted in group 2 (p = 0.09).
PFTs
PFT data are summarized in Table 2
. There were significant differences between patients in groups 1 and 4 compared to patients in groups 2 and 3. Patients in group 2 had the lowest FEV1 values. Compared to patients in group 3, their FEV1 outcomes were also lower (p < 0.01). Regarding TLC and RV, patients in groups 2 and 3 had a marked hyperinflation, which was most severe in group 2.
|
|
|
|
Multivariate regression analysis was also used to study correlations between BMI, PFT, group (GC dose per treatment group), and BMD per skeletal site. Spinal BMD was associated with "group" (adjusted r = 0.48; p < 0.001), whereas BMD at total hip and femoral neck were associated with BMI (adjusted r = 0.65 and 0.58, respectively; p < 0.0001).
| Discussion |
|---|
|
|
|---|
In the current study, it was observed that the majority of patients in group 2 had osteoporosis (82%). We wondered whether this could be attributed primarily to the regimen of GC administration. It is of note that the BMI of the patients in group 2 was lowest and PFTs revealed a marked hyperinflation and a significantly lower FEV1. After including these factors in a multivariate regression model, we found no independent association between these parameters and spinal BMD. Low spinal BMD appeared to be confined to multiple GC courses > 1,000 mg, cumulatively. The finding of GC use and associated spinal bone loss is in agreement with other earlier reports.12 13 In other studies,14 15 16 17 however, GC associated bone loss at the hip was shown to be similar or even worse, compared to that of the spine.
The mechanism of GC-induced bone loss is thought to be mediated through the induction of apoptosis of osteoblasts and osteocytes, leading to a reduction of bone formation and associated bone loss.6 The initial rapid onset of bone loss is caused by osteoclast-mediated bone resorption, probably as the result of extended osteoclast life span.6 Looking at effects due to GCs for at least 1 year reveals an initial substantial bone loss, which later on attenuates and plateaus. In a longitudinal study, using bone biopsy, treatment with intermediate doses of oral prednisolone resulted in 27% decrease in trabecular bone in 6 months, without any evidence of further bone loss later on.18 Other bone density studies19 20 revealed a similar pattern of bone loss within the first 12 months of treatment, and then also reaching a plateau following a longer duration of treatment. On the basis of these observations, we speculate that multiple courses of systemic prednisolone are correspondingly associated with multiple episodes of rapid bone loss. This hypothesis requires further study with prospective analyses of biochemical markers of bone turnover.
In the current study, groups were similar with respect to age, gender, and smoking habits. There are limitations, however, some of which are related to the cross-sectional study design. Firstly, while there were no differences in PFT results in groups 1 and 4, these results were lower in groups 2 and 3. Secondly, all PFTs were measured in patients with stable disease and not during exacerbations. Thirdly, measurements of the production of proinflammatory cytokines were not available from the medical records. Fourthly, patient numbers, particularly those of groups 1 and 2, were small. Alternatively, it should be emphasized that, since only one consulting lung physician saw all of the patients (thereby practicing uniform care), we believed that maximal homogeneity in the treatment of patients could be achieved.
It was decided to include all male COPD patients consecutively seen by one pulmonary physician during exactly 12 months into the study. Patients in group 4, the control subjects, were not subclassified despite daily ICS use in 15 of 37 subjects. Therefore, group 4 remained balanced for age, BMI, and smoking habits. Subanalysis in this group was performed, although the design of the current study was inappropriate for detecting any small differences in BMD between ICS users and nonusers. The ICS users in group 4 appeared to be older than the nonusers. In this respect, the similar spinal BMD outcomes can be explained by age-related degenerative changes.21 The BMD outcomes of total hip and femoral neck in ICS users were lower, but similar after adjusting for age. There are no data in the literature indicating that ICS can cause extra bone loss in elderly men. Several reports, particularly in women, have shown contradictory results with respect to ICS use and BMD.22 Effects of inhaled budesonide and beclomethasone were studied in a large French multicenter study, revealing no differences in BMD outcomes.9 In contrast with this study, two recent longitudinal studies23 24 in premenopausal and early postmenopausal women with asthma have shown that ICS use lead to a dose-related bone loss.
Even though a minority of COPD patients will have improvements in pulmonary function,25 systemic GCs are preferentially used for exacerbations of COPD. The Veterans Affairs Cooperative Study Group26 showed fewer treatment failures for patients who received different regimens of high-dose GCs compared to control subjects. A prospective study of severe COPD patients showed a benefit of systemic GC use at the onset of exacerbations.13 Based on these findings, the current practice of administering a course of GCs for exacerbations in COPD is widely believed to be justified.
Information about GC prescription patterns was obtained from a large database in the United Kingdom, revealing that respiratory disease was the most frequent indication (ie, 40%). Compared to patients with chronic pain syndromes, long-term daily oral GC treatment of COPD patients was less likely to occur. Furthermore, it was shown that cessation of long-term GCs exposure was advantageous to fracture risk.2 Results of several other studies27 28 29 30 31 32 33 34 focusing on GC-induced bone loss indicate that bone recovery occurs after cessation of GCs treatments. In contrast with these results, current data provide evidence against reversible bone loss. This discrepant finding can be explained, firstly, by the repetitive administration of prednisolone courses, and secondly by the interaction of nonstable disease activity on bone loss. Prospective studies involving this category of patients may help to resolve these controversies.
In COPD patients, several variables are thought to affect bone, but due to patient heterogeneity, their relevance remains unclear. Most data on cigarette smoking and other risk factors, like body weight and vitamin D status, come from healthy women. For example, after adjusting for body weight, a negative association between tobacco smoking on total body and femoral neck BMD was demonstrated in a large controlled cohort study35 of healthy premenopausal Danish women. In this particular study,35 there was only a small, but negative, effect of current smoking or vitamin D status in a subpopulation. This was also shown in another study among elderly men.36 These studies indicate that these risk factors contribute to bone loss, particularly when abundantly present like in patients with COPD.
Frequent usage of GCs is inevitable in the course of COPD, particularly during severe exacerbations, which are occasionally life-threatening conditions throughout the course of the disease. The present, cross-sectional study supports the hypothesis that administration of multiple GC courses would bring forth more harmful effects against bone, compared to continuous GC treatment regimens. Whether the loss of bone is mainly caused by "multiple GC interventions" and/or by "illness with all its negative biological effects and risk factors" should be confirmed with longitudinal studies in healthy subjects or laboratory animals. Longitudinal studies comparing bone loss either caused by continuous or multiple GC courses may confirm our findings and may lead to new recommendations for future clinical management of COPD patients treated with different GC regimes.
Since it remains difficult to predict which COPD patient will lose critical amounts of bone when exposed to multiple courses of GCs, care and vigilance by the physician remains a prerequisite for treating this category of patients. An average GC course consists of approximately 250 mg of prednisolone equivalent. The data reported in our study indicate a cumulative dose threshold of 1,000 mg of prednisolone, above which alarming bone loss can be expected. Based on this, we would stress that those patients receiving multiple GC courses should have their BMD measured in an early phase of the management of this disease.
| Footnotes |
|---|
Received for publication March 20, 2001. Accepted for publication November 27, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. F. Vondracek and B. A. Hemstreet Retrospective evaluation of systemic corticosteroids for the management of acute exacerbations of chronic obstructive pulmonary disease Am. J. Health Syst. Pharm., April 1, 2006; 63(7): 645 - 652. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. C. Mineo, V. Ambrogi, D. Mineo, A. Fabbri, E. Fabbrini, and R. Massoud Bone Mineral Density Improvement After Lung Volume Reduction Surgery for Severe Emphysema Chest, June 1, 2005; 127(6): 1960 - 1966. [Abstract] [Full Text] [PDF] |
||||
![]() |
A A Ionescu When should we suspect osteoporosis in patients with chronic airways disease? Chronic Respiratory Disease, January 1, 2005; 2(1): 1 - 2. [PDF] |
||||
![]() |
P. D. Scanlon, J. E. Connett, R. A. Wise, D. P. Tashkin, T. Madhok, M. Skeans, P. C. Carpenter, W. C. Bailey, A. S. Buist, M. Eichenhorn, et al. Loss of Bone Density with Inhaled Triamcinolone in Lung Health Study II Am. J. Respir. Crit. Care Med., December 15, 2004; 170(12): 1302 - 1309. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |