Osteoporosis in chronic obstructive pulmonary disease
- 1Section of Respiratory and Communicable Diseases, Dept of Medicine, University of Wales, College of Medicine, Llandough Hospital, Penarth, Wales, UK, 2Catharina Ziekenhuis, Afdeling Interne Geneeskunde, Eindhoven, The Netherlands
- A.A. Ionescu, Section of Respiratory and Communicable Diseases, Dept of Medicine, University of Wales, College of Medicine, Llandough Hospital, Penlan Road, Penarth, CF64 2XX, Wales, UK. Fax: 44 2920716416. E‐mail: ionescuaa@cardiff.ac.uk
Abstract
Osteoporosis is one of the systemic effects associated with chronic obstructive pulmonary disease (COPD). Potential risk factors of osteoporosis may be due to the lifestyle, genetics, treatment with corticosteroids, endocrine abnormalities or the impairment of the body composition and peripheral skeletal muscles. Evidence for the possible contribution of such factors is reviewed.
The occurrence of fractures, as a consequence of osteoporosis, can contribute to the disability and mortality of patients with COPD and add to the economic burden of the disease. The treatment with corticosteroids for the lung disease is associated with increased prevalence of fractures, but other factors may contribute.
There is a remarkable paucity of interventional studies targeting the osteoporosis in patients with COPD. The results of studies on the treatment of osteoporosis in chronic lung diseases, some including small numbers of patients with COPD, are reviewed in the paper.
Prospective longitudinal studies on the incidence of osteoporosis in chronic obstructive pulmonary disease need to assess patients with various degrees of disease severity and investigate the possible contribution of etiological factors. Randomised placebo-controlled trials are required to assess the effect of intervention, such as bisphosphonates, hormone replacement, calcium supplementation, on the prevention and treatment of osteoporosis and fractures in chronic obstructive pulmonary disease.
- chronic obstructive pulmonary disease
- fractures
- osteoporosis
- osteopenia
Chronic obstructive pulmonary disease (COPD) is a progressive disease of adulthood and older age. While the initial treatment is focused on relieving the symptoms due to the impairment of the lung function, a variety of systemic effects become obvious as the disease progresses 1–5. Osteoporosis has been recognised as one of the systemic effects of COPD and debate continues on the precise mechanisms involved and on the options for treatment 6–8. The etiology of osteoporosis in COPD is probably complex and various factors may contribute to its pathogenesis. Some of these are consequences of the chronic inflammatory lung disease and lung damage (reduced physical activity due to dyspnoea, reduced skeletal muscle mass and changes in body composition, systemic inflammation), of the therapy used during the disease (corticosteroid treatment), and of the natural changes due to ageing (hypogonadism, reduced muscle mass, inactivity). Environmental factors and habits from earlier in life also contribute to the pathogenesis of osteoporosis. When fractures occur as a complication of osteoporosis the quality of life of such patients, who are already restricted because of the lung disease, is further reduced. Therefore, awareness amongst healthcare providers and early diagnosis should trigger preventive and therapeutic measures that could avoid or reduce the consequences of osteoporosis.
Definitions
Osteoporosis is a systemic skeletal disease characterised by microarchitectural reduction of bone tissue leading to a low bone mass, increased bone fragility and thereby increased fracture risk 9. The preclinical state of osteoporosis is called osteopenia. Osteoporosis is commonly found in postmenopausal females and elderly subjects, or as a consequence of chronic disease or medical treatment. Different methods of bone mineral density (BMD) measurements can be used. Dual energy X‐ray absorptiometry (DXA) is currently the most frequently used and is accurate, reproducible and involves very low doses of radiation. BMD is expressed in standard deviation of means, the T and Z scores. The T score is a standard deviation compared to a young adult sex-matched control population. The Z score is a standard deviation compared to an age- and sex-matched control population. One standard deviation reduction in the BMD increases the fracture risk by 1.5–3 fold 10.
Epidemiology
Large epidemiological trials aimed to assess the incidence and prevalence of osteoporosis within populations of patients with COPD at various stages of disease severity are lacking. Information about the frequency of osteoporosis associated with COPD is available mainly from assessments before lung transplantation in patients with severe disease and from studies on the use of corticosteroid treatments in COPD (table 1⇓).
In a group of 15 patients with COPD referred for lung transplantation 7 (45%) had bone Z scores of >2 sd below the mean 11. Patients with cystic fibrosis (CF) referred for lung transplantation had greater frequency of osteoporosis (75%) than the COPD patients in the group studied by the same authors, while only 15% of those with other pulmonary disease had osteoporosis. Not surprisingly, of the 12 fractures that occurred in 45 patients after transplantation, six were found in patients with COPD and the rest in those with CF. A similar finding was reported in another group of 28 patients with COPD awaiting lung transplantation who had significantly lower BMD at the femoral neck site, which was only matched by patients with CF but not by those with other severe lung disease in the pre-transplantation group 12. The same authors reported a 29% prevalence of vertebral fractures in the pre-transplantation patients with COPD.
In a study on 86 male patients with COPD receiving various regimes of corticosteroid treatments or none at all, osteoporosis was found in 21% at lumbar spine, 22% at the hip and 28% at femoral neck 13. In a study on 44 male patients with COPD, of which 19 received long-term oral corticosteroid treatment, a low BMD was found at the radial and vertebral bones in those treated with prednisolone, while a reduced trabecular BMD was also found in those who did not receive long-term corticosteroid treatment 8.
In a large multicentre randomised placebo-controlled trial (1,116 patients with COPD) on the effects of inhaled triamcinolone acetonide, in the group of 412 participants who had BMD measured at the beginning of treatment and 3 yrs later, those taking triamcinolone had significantly greater reduction in BMD at the lumbar spine (p<0.007) and femoral neck (p<0.001) than those receiving placebo 14. The 3 yrs assessment found a mean±sd 2.0±0.35% reduction in BMD at the femoral neck in those treated with triamcinolone compared to only 0.22±0.32% in those receiving placebo 14. In contrast to these findings, another multicentre double-blind randomised placebo-controlled study on 1,277 smokers with mild COPD found no significant change in BMD over 3 yrs in the 102 patients who had BMD measured and received inhaled budesonide compared to the 92 patients receiving placebo 15. A modest reduction in BMD at the trochanter site (0.38%) in the budesonide group compared to 0.04% in the placebo group (p=0.02) was reported 15.
The above research suggests that osteoporosis is found in a proportion of patients with COPD and confirms the view that long-term epidemiological studies which would take into account various stages of the disease are required in order to identify the patients who have a high risk to develop osteoporosis.
Potential risk factors to the development of osteoporosis in chronic obstructive pulmonary disease
It is likely that osteoporosis in patients with COPD is a consequence of various factors some having been present throughout the life of the patient, others due to the disease process itself and some specific to the treatment of the lung disease (table 2⇓).
Smoking and other lifestyle factors
Smoking has been recognised as a contributing factor to bone loss for more than two decades 17, 18. The risk of spinal osteoporosis was greater in males who smoked (relative risk 2.3, p=0.01) than in nonsmokers and the risk increased by 1.009 per pack-yr smoked 19. Moreover, the bone loss over a period of 16 yrs was greater for males who smoked any number of cigarettes (0.104 g·cm−1) than for nonsmokers (0.072 g·cm−1, p=0.03) and even greater for smokers of >2 packs per day (0.114 g·cm−1) 20. Notably, this study does not report the forced expiratory volume in one second (FEV1) for smokers, therefore it is not known if any of the subjects had COPD.
In 341 male subjects aged between 40–80 yrs smoking added to the bone loss due to ageing 21. In this wide range of ages, the current smokers had a lower per cent of bone cortical areas than those who never smoked. In a multiple regression analysis, after adjustment for the initial percentage of cortical area, both age (p=0.005) and smoking (p=0.03) were inversely related to the per cent cortical area 21.
In a prospective study over 3.5 yrs 84 peri- and postmenopausal females were followed up, of which 63 were nonsmokers, eight smoked <20 pack-yrs and 13 smoked >20 pack-yrs 22. Allowing for the small numbers of smokers included in the study, the authors found that after correcting for age and body mass index (BMI) heavy smokers had lower bone mineral content at the distal radius and lower BMD at the lumbar spine, while the same parameters were not different between light smokers and nonsmokers. The authors speculate that their findings together with previous reports of increased fractures in smokers would support the hypothesis that heavy smokers who start this habit early in their life have a deficit in the peak bone mass 22. This issue has been addressed in a prospective cohort study on the lifestyle factors which may influence the peak bone mass on 264 healthy subjects who were followed up from the ages of between 9 and 18 to the ages of 20 to 29 at completion of the study 23. The age of participants at the end of the study allowed assessment of the peak bone mass. Males who smoked had a 9.7% lower BMD at the femoral neck compared to nonsmokers. In addition to the effect of smoking on the deficit of peak bone mass, the level of regular exercise was the single most important determinant of peak bone mass (BMD at femoral neck was up to 10% greater in those who exercised regularly). In addition, increased calcium intake contributed by a 4.7% increase in BMD at the femoral neck only in females 23.
Increased alcohol intake is an independent risk factor for osteoporosis 20, 22. One hundred and eleven males had bone loss assessed over 16 yrs by single photon absorptiometry on the midshaft radius and a greater bone loss (0.105 g·cm−1) was found in those who had >1.5 drinks per day (mean 95.5 g alcohol per week) than in those who did not drink (0.062 g·cm−1) 20. The risk for vertebral fractures due to osteoporosis was greater for males who drank (relative risk 2.4, p=0.02) than for those who did not and it was 1.007 greater per ounce-year of drinking (p=0.02) 19.
The influence of lifestyle factors that determine the peak bone mass in early adulthood and affect the BMD during later life could explain why some patients with COPD have a greater risk to develop complications of severe osteoporosis when exposed to additional risk factors, such as corticosteroid treatments for the lung disease.
Vitamin D
Vitamin D regulates the absorption of calcium, parathyroid hormone secretion and bone resorption 74. A few studies reported low levels of vitamin D in some patients with COPD. In a group of 44 males with COPD 27% had serum concentrations of 25 hydroxyvitamin D (25 (OH) vitamin D) less than the lowest value for the healthy subjects, while the biochemical markers of bone metabolism did not differ in these patients compared to those with normal vitamin D levels 16. Another group reported reduced serum 25 (OH) vitamin D in 10 of their 28 patients with severe COPD awaiting lung transplantation but the serum level of the active metabolite 1,25 dihydroxyvitamin D was within the expected range in these patients and no difference in BMD was found between patients with a low or a normal 25 (OH) vitamin D level 12.
Genetic factors
There is evidence that monozygotic twins have greater concordance of the BMD than other siblings 24, 25. However, some environmental risk factors for osteoporosis such as the levels of physical activity, smoking and the amount of alcohol intake were similar in monozygotic twin pairs, while differences were found in dizygotic twin pairs 20. Therefore the assessment of the contribution of genetic factors and of environmental factors to the development of osteoporosis can be hazardous. Some researchers have investigated the polymorphism of the genes encoding receptors for vitamin D or oestrogens and the genes for collagen type I or interleukin (IL)‐6 (a cytokine involved in the differentiation of osteoclasts) 24, 26–28. While there is still considerable controversy on the genetic determinism of osteoporosis 24, 75, the most promising finding remains the identification of a polymorphism in the gene COLIA1 which encodes the type I collagen and mutations of this gene have been associated with a low BMD 29, 30.
Systemic and inhaled corticosteroid treatment
Treatment with corticosteroids is prescribed in patients with COPD in order to improve the obstructive syndrome, alleviate the shortness of breath and the associated discomfort and there is evidence that such treatment is not only beneficial but sometimes even lifesaving 76, 77. A meta-analysis on the use of oral corticosteroid treatment (prednisolone between 30 mg·day−1 for 7 days and 60 mg·day−1 tapered to 20 mg·day−1 over 8 weeks) in patients with stable COPD found an effect size from 0–38%, depending on the outcome measures (0% when the outcome measures were arterial blood gas, carbon monoxide diffusing capacity of the lung, or subjective changes, 38% when outcome was assessed by the 12‐min walking distance, peak expiratory flow rate and dyspnoea score) 76. When the authors of the meta-analysis averaged the effect size and used FEV1 as criterion, it became obvious that oral corticosteroid therapy had a 20% improvement effect (10% more than for the placebo group) 76.
A prospective, randomised controlled trial on the use of oral corticosteroids in patients admitted to hospital for exacerbations of COPD found that the per cent change of the post-bronchodilatation FEV1 was twice greater in the group treated with corticosteoids compared to placebo (p<0.001) and after 5 days of treatment the absolute FEV1 increased three times in the same group (p=0.039) 77. Moreover, the number of days in hospital was shorter for the corticosteroid-treated group (p=0.027). Despite these beneficial effects on the lung function and the well being of the patient, the use of corticosteroids is associated with various side effects, one of which is osteoporosis 1, 7, 31. Corticosteroids reduce the absorption of calcium in the gut, increase the renal excretion of calcium, and stimulate the bone resorption (probably through the effect of parathormone) 78–80. In addition, corticosteroids reduce the bone formation by directly inhibiting the osteoblastic line, as well as secondary to the hypogonadism associated with the excess of corticosteroids 78. Most of the corticosteroid-induced bone loss occurs in the trabecular bone (proximal femur, mainly Ward's triangle) and cortical rim of the vertebral bodies 80–82. Bone biopsies from patients receiving corticosteroid treatment showed reduced trabecular bone, increased resorption cavities and reduced numbers of osteoblasts 78, 83, 84.
Research into the COPD-related osteoporosis has assessed the use of oral corticosteroids, as well as the different types of inhaled corticosteroid treatments, the length, dosage and delivery systems, and have used as outcome measures BMD, the concentrations of bone formation and resorption markers, as well as the rate of fractures. In males with chronic bronchitis and a mean±sd FEV1 of 41±19% receiving oral corticosteoids for at least 1 yr the circulating levels of osteocalcin were reduced (p<0.01) compared to reference values, but the levels were also low (p<0.05) in those not taking long-term oral corticosteroids 8. The vertebral BMD was low in both COPD groups (patients who were not taking long-term oral corticosteroids had received intermittent courses), while those on oral corticosteroids had in addition a reduced BMD at the radius.
Patients who received repeated treatments with prednisolone for exacerbations of COPD, in a cumulative dose of >1 gram, had reduced BMD at the spine (p<0.001), after correction for FEV1 and BMI, compared to a control group 13. The patients who received continuous prednisolone or intermittent courses in a cumulative dose of <1 gram had a better preserved BMD at the spine and other sites. The authors did not measure bone resorption or formation markers, therefore their comment that in patients with repeated courses of systemic corticosteroids rapid bone loss occurs after each treatment while in the long-term treatment group a plateau might be achieved, needs to be assessed in further studies.
A reduction in the blood concentration of osteocalcin occurred in healthy subjects treated with high or low doses of prednisolone or with inhaled budesonide after the first week of treatment, and no further reduction after a second week of the same treatment was found 32. This suggests a rapid effect of both oral and inhaled corticosteroids on the blood concentration of osteocalcin. A rapid reduction of bone formation assessed by the blood concentration of osteocalcin was confirmed in healthy subjects taking 15 mg prednisolone or 1 mg inhaled beclomethasone for 1 week 33. After treatment a mean reduction of 42% of the circulating osteocalcin occurred in the prednisolone group, the concentration being close to that measured in asthma or COPD patients on long-term prednisolone (>7.5 mg daily), while in the beclomethasone group a 17% fall in the osteocalcin concentration was reported 33.
Despite some concern on the lack of international standards for the assays used to measure osteocalcin concentrations 85, the above studies support the view that for doses of corticosteroids which have an effect on bone resorption markers, this occurs very soon after the exposure to the drug has started. Controversy still exists on the relationship between these rapid changes in bone markers and the long-term bone loss. Osteocalcin and procollagen type I carboxyl terminal propeptide (PICP) (bone formation markers) and type I collagen carboxy terminal telopeptide (ICTP) (marker of bone resorption) were measured after 4 weeks of treatment with at least 800 µg inhaled budesonide or beclomethasone in a group of patients with mild asthma or COPD 34. The authors report a decline in osteocalcin, a rise of the concentration of PICP and of the ratio PICP to ICTP 34. However, the same researchers reported no difference from baseline in PICP after 2.5 yrs of treatment with beclomethasone 800 µg·day−1 for airways obstruction 34, 35.
Another group studied patients with moderate-to-severe asthma treated with 2 different regimes of inhaled corticosteroids or inhaled and oral corticosteroids and measured markers of bone formation and resorption every 3 months for 24 months 36. The reported 3 monthly measurements of osteocalcin (bone formation) and deoxypyridinoline (bone resorption) show considerable variability between some assessment times. The study concluded that neither the bone formation nor the bone resorption markers changed significantly after 2 yrs compared to baseline 36. This suggests that short-term changes in such bone markers need to be interpreted cautiously.
While controversy persists on the relationships between short- and long-term changes in the bone turnover markers several studies have reported a loss of BMD after treatment with corticosteroids. When the effect of inhaled corticosteroids on the bone density and bone markers were assessed, the type, dose and possibly form of delivery of the treatment had an effect on the BMD 14, 15, 31, 34–42. However, results from various studies are difficult to compare with each other because of the study population (asthma and COPD pooled together in some), different outcome measurements and length of time between initial and final assessments. In patients with asthma quantitative computed tomography analysis of vertebral trabecular BMD found a reduced BMD after 12 (p=0.006) and 24 months (p=0.004) treatment with beclomethasone 2 mg·day−1 compared with fluticasone 1 mg·day−1, while BMD at the spine and femoral neck by DXA were unchanged in both treatment groups 36. The BMD measured on patients with asthma treated with high dose of inhaled beclomethasone (1–2 mg·day−1 for at least 1 yr) and repeated short courses of oral corticosteroids for exacerbations was similar to the BMD of patients who were treated with inhaled corticosteroids and regular oral prednisolone (dose 5–30 mg·day−1) for between 1 and 25 yrs 37.
However, another group reported no decline in the BMD in patients with moderate or severe asthma treated with daily inhaled 1 mg fluticasone or 1.6 mg budesonide delivered through volume spacer, for 12 months 38. Patients with mild-to-moderate COPD treated with inhaled triamcinolone acetonide 1.2 mg·day−1 were followed up for a mean of 40 months 14. A significant reduction in BMD at the lumbar spine (p=0.007) and femoral neck (p<0.001) compared to placebo was reported 14. Patients with mild COPD treated with 800 µg budesonide for 3 yrs had no reduction in BMD compared to placebo 15.
A review of the long-term studies on the association of bone loss in patients with asthma treated with inhaled corticosteroids concluded that while the cumulative dose for adverse effects on bone may vary according to the type of drug used, daily doses <800 µg of beclomethasone or budesonide and <750 µg of fluticasone have limited effects on bone metabolism 43. In a study on 71 male patients with moderate-to-severe COPD who did not receive any formulation of corticosteroids, hypercapnic patients had lower total and trabecular BMD (p<0.001) and greater serum concentration of carboxy-terminal cross-linked telopeptide (bone resorption marker, p<0.05) than those with normocapnia and comparable FEV1/vital capacity and BMI 44.
Treatment with corticosteroids is a contributory factor to the pathogenesis of osteoporosis in COPD, but other aspects of the disease are probably linked to the loss of BMD.
Reduced peripheral skeletal muscle mass and strength
Skeletal muscle dysfunction in COPD is probably multifactorial. The reduced mobility due to shortness of breath, the myopathy due to corticosteroid treatments and metabolic factors generate a vicious circle more obvious in patients with severe disease 45–47. The force generated by the skeletal muscles during voluntary contractions has an important role in the postnatal development of the bone. The muscle strength increases with growth in childhood and declines in adulthood. It is usually followed by age-related changes in the bone mass 48. The physiological process of ageing is accompanied by a decline in the skeletal muscle mass, referred to as sarcopenia, where the progressive reduction of the diameter of the muscles and of the number of muscle fibres is associated with a decline in the muscle strength 49, 50. Both the muscle mass and strength are related to BMD 50, 51. In the elderly sarcopenia and the muscle weakness are associated with increased rates of falls and hip fractures 52, 53. The close relationship between the skeletal muscle strength and the bone mass is supported by the beneficial effects of resistance training on BMD 86. The greater the stress on a bone area, the greater the bone mass, therefore the mechanical stress applied to the bone during training increases the BMD 87. Resistance training for 1 yr increased BMD by 1% at the lumbar spine and femoral neck in postmenopausal females, while in the control group a decrease of 1.8% (lumbar spine) and 2.5% (femoral neck) occurred 88.
Patients with COPD are at risk to develop osteoporosis due to a reduced skeletal muscle mass and strength, both secondary to the disease and due to the natural process of ageing. It remains for future research to assess if training of various skeletal muscle groups improve BMD or prevent the progressive loss of bone mass.
Body mass index and changes in body composition
Weight loss and a low BMI are predictors of mortality in patients with COPD 2, 17. A preferential depletion of fat-free mass (FFM) has been reported in COPD and associated with increased exacerbations and hospital admission rates 3, 89, 90. Moreover, a regional loss of muscle mass is predictive of mortality in patients with COPD and a FEV1 of <50% 91. Bone mass is related to FFM in males 92, while in females some investigators reported a relationship between fat mass (FM) and bone density 54, 55 and others between both FM and FFM and bone density 56, 57.
The relationship between FM and/or FFM and bone mass suggests that the load of soft tissues is important in the preservation of bone mass. In underweight elderly the bone mineral content was reduced compared to age-matched subjects with a normal BMI 58, 59. Both the BMI and the mid arm muscle circumference (an index of FFM) were associated with a reduced BMD in a cross sectional study on 100 patients with COPD 60. Other groups reported positive relationships between the BMI and the bone mass in patients with COPD, with correlation coefficients between 0.34 (p<0.05) and 0.57 (p<0.01) 61, 62. Patients with severe COPD admitted for a rehabilitation programme had similar weight and height to a healthy age-matched control group. However, the patients had a low FFM by DXA (p<0.05) and this was accompanied by a reduced whole-body BMD (p<0.001) and BMD of the arms (p<0.05), legs (p<0.01) and trunk (p<0.01) 63.
Such studies support the view that weight loss and mainly the depletion of FFM are factors contributing to the loss of BMD in some patients with COPD.
Hypogonadism and other endocrine abnormalities
Hypogonadism and the reduced availability of sex hormones, either due to ageing or to the effect of corticosteroid treatment, contribute to the development of osteoporosis. Oestrogen deficiency in females increases the bone loss after menopause, and the decline in circulating free oestrogen in elderly males has also been related to a reduction in bone mass 64–67, 75. Oestrogen regulates both bone resorption and formation, while testosterone regulates bone formation 65, 68, 75. In addition to these age-related changes, corticosteroids alter the secretion of the gonadal hormones by reducing the secretion of the pituitary luteinizing hormone and of the circulating oestrogen and testosterone levels 80. Males with severe COPD receiving oral corticosteroids (mean dose of prednisolone 9.4 mg·day−1) had serum levels of free testosterone below normal and these were related to the dose of corticosteroid treatment (r=−0.54, p=0.007) and to BMI (r=0.24, p=0.037) 68. Treatment with testosterone for 12 months in asthmatic males receiving long-term oral corticosteroids improved the BMD at the lumbar spine by 5% (p=0.005 compared to the placebo group) and the FFM by 0.9 kg (p=0.02 compared to placebo) 69.
Another group of hormones with a potential role in osteoporosis are the insulin-like growth factors (IGF‐1, its tissue ligands, the binding proteins IGFBP 1–6 and receptors) 70. IGF‐1 and IGFBP 5 stimulate the differentiation and proliferation of osteoblasts 93. The bone concentrations of IGF‐1 and IGFBP‐5 (which enhances the effect of IGF‐1) decline with age and there is evidence of a parallel decline of their serum concentrations, despite the continuing controversy on the relationship between serum and bone concentration of these hormones 94, 95. In addition, IGFs stimulate the synthesis of proteins and there is a reduced expression of IGF‐1 gene during fasting and catabolism 2, 96. It has been hypothesised that the down-regulation of IGF‐1 messenger ribonucleic acid (mRNA) expression found in the skeletal muscle is related to the muscle atrophy in some patients with COPD 71. Low serum levels of IGF‐1 have been reported in elderly with hip fractures and a low BMD 72. These studies suggest that IGF‐1 may influence bone mass directly or through its role in the preservation of the skeletal muscle mass. Both mechanisms may be of relevance in COPD and further research is needed to find out if in addition to the age-related decline in IGF‐1 there is any link between the IGF‐1 activity on the bone and skeletal muscles and the pathogenesis of osteoporosis.
The potential role of the chronic systemic inflammation
Increased concentrations of the circulating inflammatory mediators (tumour necrosis factor (TNF)‐α, IL‐6) have been reported in COPD, mainly in patients who lose weight and in those with a low FFM and skeletal muscle mass 3, 5, 97, 98. Moreover, peripheral monocytes from patients with COPD who lose weight had an increased ability to produce TNF‐α 99. Leukocyte-derived IL‐1α and TNF‐α stimulate bone resorption 64, 100, 101 and IL‐6 stimulates the formation of osteoclasts 102. Peripheral macrophages from patients with idiopathic osteoporosis and rapid bone turnover produce increased amounts of IL‐1 73. There is some evidence for a possible role of the inflammatory mediators on bone metabolism in COPD, which has been described in other diseases associated with weight loss, depletion of FFM and systemic inflammation such as chronic heart failure, cystic fibrosis or cancer 103–105.
Further research is required to evaluate if there is a relationship between the increased blood concentrations of some inflammatory mediators and the bone loss in COPD, and if such circulating inflammatory mediators are active at a tissue level and involved in other effects such as the muscle wasting and cachexia.
Consequences of osteoporosis in chronic obstructive pulmonary disease: the risk of fractures
Debate continues on the precise mechanisms and pathophysiology of osteoporosis in COPD, but most physicians are concerned about the risk of fractures, which would add to the disability of such patients (table 1⇑). In a general population, a British epidemiological study reported that the two main independent factors for increased risk of hip fractures in the elderly were inactivity and muscle weakness 106. The American National Health and Nutrition Examination Survey (NHANES) III found that weight loss of at least 10%, a low phalangeal bone density and the presence of any chronic condition were main factors, with smoking and low physical activity levels additional risk factors for hip fractures in White males 107. Such risk factors reported in a general population are likely be found in patients with COPD, who smoked or continue to smoke, are inactive and have weak skeletal muscles.
Fifty per cent of the patients with various disease treated with corticosteroids suffer fractures and the proportion is close to 100% in those treated for rheumatoid arthritis 78, 108–110. A report on the use of oral corticosteroids and the risk of fractures, where 40% of patients were treated for respiratory disease found an increased rate of nonvertebral and hip fractures compared to the control group (relative rates (95% confidence interval (CI)) were 1.33 (1.29–1.38) and 1.61 (1.47–1.76), respectively) 108. The relative rate of nonvertebral fractures increased with the dose of corticosteroids from 1.17 in the low dose (<2.5 mg·day−1) to 1.64 in the high dose group (at least 7.5 mg·day−1) and most of the excess risk of fractures disappeared within 1 yr when the treatment was stopped 108.
A retrospective study on the risk of fractures in patients receiving inhaled corticosteroids, which included a 6% sample of the UK population and excluded patients with any confounding variables for fractures, reported a similar risk in patients who used inhaled corticosteroids and in those who received inhaled bronchodilators only, which suggests a role played by the chronic respiratory disease itself as a risk factor for osteoporosis 111. In both groups the risk of fractures was higher than in the control group, and the relative risk increased with the dose of inhaled corticosteroid treatment (beclomethasone in 86.4% of cases) from 0.95 (95% CI 0.67–1.34) if <300 µg·day−1 to 1.77 (1.31–2.40) if >700 µg·day−1 111. In asthma the use of continuous long-term oral corticosteroids was associated with 11% fractures of ribs and vertebrae, compared with no fractures in a group treated with intermittent oral corticosteroids 112. Compared to the risk of fractures in post-menopausal females where for one sd reduction in BMD the risk doubles 113, in patients with asthma treated with long-term corticosteroids the risk of fractures was associated with greater BMD (osteopenia range) 114.
A cross-sectional study on 312 males with COPD assessed the prevalence of fractures in three groups: 1) patients who never used corticosteroids, 2) patients who received inhaled corticosteroids and 3) those receiving systemic steroids (range of the duration of corticosteroid treatments in group 3 was 2.5 to 1,300 weeks) 7. The prevalence of at least one vertebral fracture was 48.7% in group 1, 57.1% in group 2 and 63.3% in group 3, with ≥6 vertebral fractures found only in group 3. The high risk of fractures in the group who never used corticosteroids suggests that factors others than the treatment may be involved in the pathogenesis of osteoporosis. However, a study on a smaller number of patients with COPD and severe lung disease found no evidence of an increased risk of fractures in patients who were not receiving long-term corticosteroid treatment 16, which suggests the need for further large prospective studies in order to investigate which patients with COPD are more at risk of fractures due to osteoporosis.
The occurrence of fractures has implications for the morbidity and mortality. Severe osteoporosis with thoracic vertebral fractures and wedging with hyperkyphosis was associated with about 10% reduction in forced vital capacity (FVC) in females, with a cumulative effect of the number of fractures on the decline in FVC 115. Such a reduction in FVC would add to the lung function impairment and disability in patients with chronic lung disease. Mortality after hip fractures in the elderly is about 20% in the first year (highest mortality rate for any fractures) and 19% of these patients require residential care when discharged from hospital, which adds to the economic burden of the disease 116, 117.
Therapeutic interventions for osteoporosis in patients with chronic lung disease
Despite the remarkable lack of interventional studies targeting osteoporosis in patients with COPD, some conclusions can be drawn from the treatment of corticosteroid-induced osteoporosis, postmenopausal osteoporosis and the few studies on patients with asthma and other chronic lung diseases.
Calcium and vitamin D supplements
In postmenopausal females calcium and vitamin D supplements may be beneficial if the dietary intake of calcium is low, which has been reported in some elderly females 118–120. Some studies on corticosteroid-induced osteoporosis have reported that calcium and vitamin D supplements may reduce the rate of bone loss in the short term, but no increase of bone mass in the long term was found. Current recommendations suggest a minimum daily intake of 1,200 mg of calcium in adults older than 50 yrs and 1,500 mg per day with vitamin D3 400 IU·day−1 if glucocorticoid-induced osteoporosis is diagnosed 43, 80, 121.
Hormone replacement
In postmenopausal females oestrogen-replacement therapy reduced the rate of fractures and this effect was more beneficial if the treatment was initiated early when menopause had started 118, 120, 122, 123. In glucocorticoid-induced osteoporosis in females oestrogen and progesteron treatment reduced the speed of bone loss 124. In males with asthma receiving long-term corticosteroid treatment, depot injections with testosterone improved the BMD at the lumbar spine and increased the FFM 69. Such studies suggest that when hypogonadism is diagnosed replacement therapy is beneficial for the preservation of bone mass. The same recommendations should be applied to patients with COPD and postmenopausal or corticosteroid-induced hypogonadism, only after considering potential risks of such treatments.
Calcitonin
Calcitonin is a peptide hormone secreted by the C cells of the thyroid, which has an inhibitory effect on the osteoclasts 118. Studies on the bone-protective effect of calcitonin have not been conclusive 43. In a study on 44 asthmatics with severe symptoms, intra-nasal calcitonin improved BMD at the lumbar spine compared to the calcium-treated group where BMD decreased 125. However, about one third of the patients stopped the treatment with calcitonin because of side effects and because of exacerbations of asthma 125.
Bisphosphonates
Bisphosphonates are metabolically stable analogues of pyrophosphate with affinity for the hydroxyapatite in the bone, where they inhibit the action of osteoclasts, and therefore inhibit bone resorption 126. A few studies have assessed the effect of different bisphosphonates on the preservation of BMD in patients with chronic lung disease, most of which received corticosteroid treatments (table 3⇓).
In 28 asthmatics receiving inhaled beclomethasone 2–2.2 mg·day−1, treatment with calcium supplements (1000 mg·day−1) or cyclical etidronate 400 mg·day−1 in addition to calcium 1000 mg·day−1 for 18 months, equally improved BMD compared to a group where no supplements were given 127. In another study the effect on BMD of cyclical etidronate for 12 months was compared with calcium supplements (500 mg·day−1) in patients who continued oral corticosteroid treatment (at least 10 mg·day−1) 128. Etidronate improved BMD at the lumbar spine by 5.7% and total hip by 6.8%, while the calcium supplemented group lost 3.4% BMD at the lumbar spine and 4.1% at the total hip 128. In 15 patients with asthma and two with sarcoidosis treated with prednisolone for a range of 3–30 yrs, with a range of doses between 7.5 and 40 mg·day−1, treatment with 30 mg of pamidronate every 3 months for 1 yr produced a mean gain in BMD at the lumbar spine of 3.4% without any change at the femoral neck 129.
Two studies on greater number of patients, one on 60 with chronic asthmatic bronchitis treated with inhaled beclomethasons or fluticasone 130 and another on 74 patients with asthma treated with oral and inhaled corticosteroids 131, reported improvement in BMD after 12 months treatment with clodronate. Another study on females with asthma (n=70) or COPD (n=8) receiving between 800 µg and >1600 µg inhaled corticosteroid treatment (and 50% intermittent additional oral corticosteroid treatment) assessed the effect of oral alendronate 10 mg·day−1 and calcium supplements (500 mg·day−1) compared to placebo and calcium supplements (500 mg·day−1) 132. While the placebo and calcium-treated groups lost a mean 0.8% BMD at the spine and 0.37 at the hip, the alendronate group gained 3% and 1.6% BMD, respectively 132.
In adults with cystic fibrosis and BMD <2.0 at inclusion in the study (some on long-term treatment with oral corticosteroids), treatment with 30 mg pamidronate every 3 months and calcium supplements (1000 mg·day−1; n=13) was compared with calcium supplements only (n=15) 133. After 6 months the mean (95% CI) gain of BMD in the pamidronate group compared with the calcium-supplemented group was 5.8 (2.7–8.9)% at the lumbar spine and 3.0 (0.3–5.6)% at the hip, while there was a reduction of BMD at the distal forearm of 1.7 (0.3–3.7)%. Patients who were not treated with coticosteroids while they received pamidronate, frequently experienced bone pain. Another randomised, controlled, non-blinded trial compared the effect of 30 mg i.v. pamidronate every 3 months for 2 yrs with vitamin D (800 IU·day−1) and calcium supplements (1000 mg·day−1) with the same doses of vitamin D and calcium in 34 adults with cystic fibrosis following lung transplantation 134. Post-transplantation patients received immunosuppressants which included cyclosporin A, prednisolone and azathioprine. BMD increased by a mean±sd 8.8±2.5% at the spine and 8.2±3.8% at the femur in patients treated with pamidronate, compared to 2.6 ±3.2% and 0.3±2.2% respectively, for the vitamin D and calcium group (p<0.015).
The British Thoracic Society has recently completed a prospective multicentre placebo-controlled trial in over 700 patients treated for asthma with oral and/or inhaled corticosteroids comparing etidronate, calcium supplementation and the combination of etidronate with calcium supplementation over 5 yrs. The results of this trial are expected in the near future.
Prevention and therapy: suggestions for the future
In order to assess the efficacy of treatments, more information on the risk factors and the pathogenesis of COPD-induced osteoporosis is needed. Such information could be gathered through prospective studies designed to assess the rate of decline of the BMD and the contributing factors such as the type of corticotherapy, the presence of hypogonadism, ongoing smoking, reduced physical activity and the weakness of the skeletal muscles.
In view of the relationship between the skeletal muscle mass and the BMD 50, 51, 87, 88 it is likely that training programmes and conditioning will have beneficial effects on the maintenance of BMD in patients with COPD. However, the types of training and the specific programmes of rehabilitation need to be designed.
Hormone replacement therapy is likely to be beneficial in patients with COPD and hypogonadism. In view of the age group and exposure to corticosteroid treatment, which are associated with hypogonadism, the assessment of the hormonal status should be part of the general investigation of osteoporosis in patients with COPD.
The intake of calcium and vitamin D should be assessed, in view of some reports that supplementation is beneficial for the preservation of the bone mass, mainly in subjects with a reduced intake 118–120. According to the current nutritional recommendations the daily intake of calcium (1,200–1,500 mg·day−1) and vitamin D (at least 400 IU·day−1) should be ensured 43, 80, 121.
More promising in view of the available research in patients with chronic lung disease is the therapy with bisphosphonates. The patients with COPD and osteoporosis treated with long-term systemic corticosteroids should be considered for such treatments. For patients with osteopenia, those on long-term inhaled corticosteroids and/or intermittent courses of oral corticosteroids without florid osteoporosis, regular monitoring of BMD by DXA scanning should be undertaken.
Suggested future research
Prospective studies on the incidence of osteoporosis and osteopenia in COPD will need to assess patients with all degrees of disease severity, including those with mild impairment of the lung function, which would allow assessment of patients who have not been exposed to corticosteroid treatments. Such longitudinal studies should investigate the contribution of potential factors leading to osteoporosis such as the nutrition, smoking, levels of physical activity, body habitus and composition, peripheral skeletal muscle mass and function.
Future research will explore the incidence of hypogonadism in patients with COPD and if hormone manipulation has long-term effects on the preservation of bone mass. Other possible hormonal changes will need to be investigated, such as the potential imbalance between anabolic (IGF-1) and catabolic (corticosteroids) hormones that may contribute to the loss of skeletal muscle and bone mass in COPD.
The relationship between the persistent systemic inflammation and the depletion of skeletal muscles and bone in some patients with COPD will need to be further explored, in order to clarify if circulating inflammatory mediators have an effect at the tissue level or if other factors stimulate increased inflammation in the peripheral tissues, including the bone tissue.
Randomised placebo-controlled trials are required to assess the effects of bisphosphonates on the prevention and treatment of osteoporosis and fractures in various groups of patients with chronic obstructive pulmonary disease (exposed to oral or inhaled corticosteroids, affected by hypogonadism, as well as those who might have low bone mineral density due to other causes). Such trials should have clear outcome measures, such as the change in bone mineral density or the assessment of the fracture rates and should investigate potential short-term and long-term side effects of the treatment.
- Received June 30, 2003.
- Accepted June 30, 2003.
- chronic obstructive pulmonary disease
- fractures
- osteoporosis
- osteopenia
- © ERS Journals Ltd