Osteoporosis among Patients with Chronic Obstructive Pulmonary Disease: Systematic Review and Meta-analysis of Prevalence, Severity, and Therapeutic Outcomes.

Chronic obstructive pulmonary disease (COPD) can be associated with systemic inflammatory trademarks and can coexist with other chronic debilitating diseases such as osteoporosis, which is considered among the most serious comorbidities of COPD. In this review, we aimed at finding answers for the following questions and tried to encapsulate the available literature: (1) how prevalent is osteoporosis among patients with COPD? (2) What are severity patterns of osteoporosis in case of COPD? (3) What are the therapeutic outcomes for patients with osteoporotic COPD? The total number of patients with COPD from all studies was 3815, majority of which were male (2658) representing 69.67% of patients. The mean ± standard deviation for percentage of forced expiratory volume in 1s (FEV1%) was 55.43 ± 14.62%, body mass index for almost 91.29% of patients was 24.4 ± 4.45 kg/m2, whereas fat-free mass index (FFMI) was 17 ± 0.93 kg/m2 for 17.66%. The percentage of patients with COPD having osteoporosis varied in the analyzed studies from 14% up to 66.6%. The mean prevalence of reported osteopenia from 14 studies (n = 2107) was 39.91%, whereas for osteoporosis, the mean prevalence was 37.62% for all included studies. Osteoporosis was highly prevalent among patients with COPD. It is reasonable to call for osteoporosis screening in patients with COPD who are above 65 years, in advanced stages, with BMI lower than 21 kg/m2 or with FFMI lower than 16 kg/m2 for males and 15 kg/m2 for females. There is a lack of research investigating severity and treatments of osteoporosis in patients with COPD. Copyright:

Introduction

Chronic obstructive pulmonary disease (COPD) is a well-known chronic lung disorder that can be identified by a progressive irreversible limitation of airflow, and characterized by repetitive episodic exacerbations. COPD is a life-threatening condition commonly associated with or caused by asthma, chronic bronchitis, and emphysema.[1] In patients with COPD, the limitation in airflow usually can be detected using spirometry test and then categorized according to the Global Initiative for Chronic Obstructive Pulmonary Disease (GOLD) standards.[23]

It has been appraised that this disease is affecting around three million patients in the UK alone,[5] and impacting the life of 329 million patients all around the world.[678] In fact, COPD can be associated with commodities that have systemic implications that can affect not only the pulmonary system but also other major vital systems. It can affect cardiovascular system by increasing the stiffness of arterial walls, affect muscular system by triggering atrophies, and can be associated with the increased severity of osteoporosis.[34]

Osteoporosis literally means permeable or pervious and it was described by the International Osteoporosis Foundation as a chronic systemic condition represented by a progressive bone density reduction, structural bone collapses, frailty, and a high risk of fractures.[910] Risk factors for osteoporosis include gender, lifestyle, advanced age, steroidal therapies, low body mass index (BMI), hormonal disorders, and some chronic conditions. Recently, it has been found that the presence of COPD significantly increases the risk for osteoporosis by two- to fivefold as compared to the normal population.[1144]

Osteoporosis and COPD are interconnected as there are many risk factors in common between both conditions. In fact, both conditions can be also associated with inflammatory manifestations and they can have an intertwined pathophysiology. Recently, it has been found that osteoporosis is more pervasive among patients with COPD than what it was thought.[1213]

Patients with COPD endure an extremely poor lung function with low vital capacity. Whenever such a chronic condition is associated with osteoporosis-caused fractures, the consequences will be unbearable. With reduced ability to move, the patient will be in a high risk of developing deep venous thrombosis and pulmonary embolism, which increases the risk of death and causes a drastic decline in patient’s health condition added to the already impaired quality of life.[141516]

The objective of this review was to conduct a proper search for the available literature to find suitable answers for the following questions and encapsulate the available literature about the presented topic: (1) How prevalent is osteoporosis among COPD patients? (2) What are severity patterns of osteoporosis in case of COPD? (3) What are the treatment outcomes for patients with osteoporotic COPD?

Materials and Methods

In this review, we have tried as much as possible to adhere to Cochrane Effective Practice and Organisation of Care guidelines,[42] the US National Institutes of Health recommendations for reviews, and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement guidelines on systematic reviews protocols.[17] To collect the required articles for this review, a systematic electronic search strategy was conducted using PubMed/MEDLINE, USM e-resources, and EBSCOhost.

The inclusion criteria included the following: (1) Relatively stable patients at any COPD stage, (2) screened patients for comorbidities (osteoporosis should be present among comorbidities), (3) tested for bone mineral density (BMD) and screened for osteoporosis with T-score, (4) analyzed osteoporosis treatment among patients with COPD and/or its outcomes, (5) published in English language, and (6) published in the last 10 years. The exclusion criteria included the following: (1) Review articles, editorial articles, abstracts, congress abstracts, case reports, case series, and current clinical trials; (2) gray literature materials and republished manuscripts, (3) incomplete or current research, and (4) end-of-life or palliative care studies.

Selection strategy

The main search terms that were used included osteoporosis among patients with COPD, prevalence of osteoporosis among them, severity of osteoporosis in COPD, and treatment outcomes of osteoporosis in COPD, whereas MeSH (Medical Subjects Headings) terms included COPD, comorbidities, osteoporosis, osteopenia, pulmonary diseases, obstructive diseases, chronic bronchitis, emphysema, chronic obstructive disease, asthma, chronic diseases, therapeutics, therapy, management, treatment effect, and bone density. From each of the conducted searches on the aforementioned databases, keywords were selected and combined using “AND,” and within each individual group, they were joined and searched using “OR.” In addition, reference list of some of the recruited papers were screened and searched to handpick any relevant articles that were missed in the preliminary searches, and then articles were put for full-text examination and assessment according to the designed strategy and criteria.

Osteoporosis prevalence among patients with chronic obstructive pulmonary disease

The recruited studies were only those, which had included prevalence of osteoporosis as one of its results based on BMD test (dual-energy X-ray absorptiometry [DXA]) and the patients having osteoporosis can be separated independently from study population.

Osteoporosis severity in chronic obstructive pulmonary disease

Only studies, which were exploring the correlation of these two conditions, assessing the severity of osteoporosis in COPD and where patients with osteoporosis can be identified from other comorbidities and diseases were accepted.

Therapeutic outcomes

Only studies that included clear prescribed bone medication, which can be independently identified in case of patients with osteoporotic COPD were accepted. Randomized controlled trials, cohort studies, or clinical studies that are about other means of treatment (such as lifestyle adjustments and physical exercise) for osteoporosis, which investigate the outcomes of treatments among patients with COPD were also accepted.

Results

The initial search results for general terms such as osteoporosis among patients with COPD showed 703 results, of which, only 72 met the selection criteria for general terms; then for more specific terms such as prevalence, severity, and treatment outcomes, 295 results were found and only 65 were primarily suitable so the total number of selected articles for primarily assessment was 137 articles. Later, 118 articles were excluded after a strict screening for inclusion criteria, removal of duplicates, congress abstracts, unclear articles, and review articles. A total of 19 articles were included of 998, later, one more article was excluded due to lack of definition of patients with osteoporosis and lack of useful information for this review, so the final number of articles included in this review was 18 articles [Figure 1].

Figure 1

Flowchart of search strategy and articles collection process

The total number of patients with COPD from the recruited studies was 3815.

Majority of which were male (n = 2658) representing 69.67% of all patients. Moreover, the mean ± standard deviation (SD) for patient’s age if reported was 65.84 ± 5.76 years, the percentage of forced expiratory volume in 1s (FEV1%) predicted for 3290 (86.23%) patients from 12 articles was 55.43% ± 14.62%, which was clearly low if compared to normal population. For 3483 (91.29%) patients from 13 studies, the mean ± SD for BMI was 24.4 ± 4.45 kg/m2, whereas fat-free mass index (FFMI) was reported in three studies only for 674 (17.66%) patients and it was 17 ± 0.93 kg/m2.

The percentage of patients with COPD having osteopenia or osteoporosis varied in the analyzed studies (18%–65% and 14%–66%, respectively). Despite the fact that multiple methods were used, DXA was the method of choice to test patients’ BMD, and T-score was the standard for osteoporosis screening in all studies [Table 1]. The mean prevalence of reported osteopenia for 14 (n = 2107) of the recruited studies was 39.91%, whereas the mean prevalence of osteoporosis for all included studies was 37.62%.

Table 1

Analysis of the recruited studies in the systematic review

First author (Ref)	Subjects (n)	Setting	Sex M/F (n)	FEV1% pred†	Other comorbidity(ies)	BMI kg/m2†	FFMI kg/m2†	Measurement	Osteopenia	Osteoporosis
Sakurai-Iesato et al.[18]	50	Chiba University Hospital, Japan	50 M*	50.40 ± 17.0	CVD, HTN, DM, dyslipidemia, hyperuricemia, GERD	23.2 ± 3.4	N/A	BMD (DXA), T-score, PFT, FRAX, CT scan, CAT	46%	14%
Duckers et al.[19]	30	4 Cardiff and Vale GP clinics, UK	30 M*	63.70 ± 17.9	N/A	26.4 ± 5.1	18.3 ± 2.4	BMD (DXA), T-score, PFT, bone formation/sorption biomarkers	57%	17%
Gupta et al.[20]	49	King Fahd Hospital, KSA	44/5	66.58 ± 15.0	N/A	N/A	N/A	Spirometry, BMD, T-score	65.30%	28.57%
Silva et al.[21]	95	Hospital De Clínicas de Porto Alegre, Brazil	65/30	41 ± 16	N/A	25.8 ± 4.7	N/A	BODE index, T-score, TLC, DLCO	42.10%	42.10%
Hattiholi and Gaude[22]	70	Prabhakar Kore Hospital and Medical Research Centre, India	45/25	42.70 ± 16.9	N/A	21.4 ± 4.7	N/A	BMD (DXA), Z-score, T-score	18.6%	65.70%
Camiciottoli et al.[23]	412	University Hospital of Florence, Italy	299/113	69 ± 25	ADS, DM, HF, IAH, PVD, IHD	26 ± 5	N/A	mMRC dyspnea scale, PFT, BMD (DXA), T-score	N/A	17%
Abbasi et al.[24]	90	Lung clinic of Bouali Hospital in Qazvin, Iran	90 M*	N/A	N/A	22.2 ± 4.17	N/A	BMD (DXA), T-score, PFT	31.5%	52.80%
Lee et al.[25]	1081	National Health and Nutrition Examination Survey, South Korea	810/271	77.21 ± 1.49a	HTN, DM hyperlipidemia, depression, malignancies	22.79 ± 0.15	N/A	BMD (DXA), T-score, PFT	N/A	17.70%
Ramachandran et al.[26]	60	Respiratory Medicine Hospital, India	51/9	0.49 L	N/A	N/A	N/A	Spirometry, BMD (DXA), T-score	33%	35%
Nayyaret al.[27]	84	Gandhi Medical College, India	42/42	N/A	N/A	N/A	N/A	BMD (DXA), T-score	41.6%	45.2%
Bhattacharyya et al.[28]	37	Institute of Pulmocare and Research, Kolkata, India	35/2	0.81 ± 0.37 L	N/A	N/A	N/A	BMD (DXA), T-score	51.35%	21.62%
Lin et al.[29]	125	Chia-Yi Christian Hospital, Taiwan	123/2	48.05 ± 19	DM, HTN, hyperlipidemia, CAD, CHF, CKD, CLD	21.8 ± 3.4	N/A	mMRC dyspnea scale, St. George’s Respiratory Questionnaire, depression scale, BMD (DXA), CAT, T-score	N/A	40%
Graat-Verboom et al.[30]	90	Catharina Hospital in Eindhoven, Netherland.	54/36	63.40± 1.9	N/A	26.8±0.50	16.80±0.3	BMD (DXA), X-spine, T-score, mMRC dyspnoe scale, Charlson score.	N/A	61%
Gaude and Hattiholi[31]	102	Prabhakar Kore Hospital, Belgaum, India	64/38	N/A	Pneumothorax, CHF, pulmonary thromboembolism	N/A	N/A	BMD (DXA), T-score PFT, GOLD criteria	19.6%	66.6%
Graat-Verboom et al.[32]	554	The Centre for Integrated Rehabilitation of Organ Failure (CIRO), Horn, The Netherlands	344/210	42.10 ± 0.7	N/A	24.7 ± 0.2	15.90 ± 0.1	BMD (DXA), T-score, 6MWD, body composition	41%	21%
Ferguson et al.[37]	658	88 Medical Center in the USA	382/276	44 ± 13	N/A	26.7 ± 6.2	N/A	BMD (DXA), T-score, SGRQ activity score	42%	23%
Watanabe et al.[38]	49‡	Teikyo University Chiba Medical Center and Yarita Hospital, Japan	49 M*	66 ± 12..4	Vertebral fracture	21.5 ± 3.2	N/A	CAT score, PFT, BMD (DXA), T-score, Z-score	38.8%b	42.9%b
Li et al.[39]	179	Temple Lung Center, USA	81/98	0.71 ± 0.30 L	Bone fracture	25.3 ± 5.3	N/A	Modified BODE score, 6MWD, PFT, BMD (DXA), T-score	31%	66%

M = male, F = female, N/A = not assessed, FEV1 = forced expiratory volume in 1s, FEV1% pred = FEV1% predicted, which is a derived value of FEV1%, BMI = body mass index, FFMI = fat-free mass index, BMD = bone mineral density, TLC = total lung capacity, DLCO = diffusing capacity of the lung for carbon monoxide, DXA/DEXA = dual-energy absorptiometry, CAT = COPD Assessment Test, PFT = pulmonary function test, mMRC = Modified Medical Research Council, FRAX = WHO Fracture Risk Assessment Tool, 6MWD = 6-min walk distance test, ADS = anxious depressive syndrome, IAH = idiopathic arterial hypertension, IHD = ischemic heart disease, PVD = peripheral vascular disease, CAD = coronary artery disease, HTN = hypertension, CHF = congestive heart failure, CKD = chronic kidney diseases, CLD = chronic liver disease, DM = diabetes mellitus, GERD = gastroesophageal reflex disease, CVD = cardiovascular disease

aPatients with osteoporosis only, bpercentage of the included screened subjects for osteoporosis only

†Values are mean ± SD, ‡BMD scan was available only in 49 patients of 136 patients, *male patients only

In this table, results are presented as mean and standard error of mean unless otherwise indicated

Five studies were conducted in India, investigating the prevalence of osteoporosis among patients with COPD. They included 385 patients, of which majority were males representing 61.55% (n = 237). Majority of the patients were in III or IV stage of COPD, representing 77.40% (n = 298), the available data for BMI were from Hattiholi and Gaude[22] and the mean ± SD was 21.4 ± 4.7 kg/m2 for 70 patients only. Most of patients were in advanced stages of COPD (III and IV) and with more than two exacerbations per year. Almost 19% of patients were osteopenic and around 66% were osteoporotic.

In a study by Ramachandran et al.,[26] they found that 67% of patients had abnormal bone density, which was correlated to the severity of airways obstruction with P < 0.001. Nayyar et al.[27] found that of 116 patients, 87.06% of them were either osteopenic or osteoporotic and they had significantly lower bone density as compared to normal group. Bhattacharyya et al.[28] found that of 37 patients, 51.35% were osteopenic and 21.62% were osteoporotic. According to Gaude and Hattiholi,[31] after using multivariate logistic regression analysis (odds ratio [OR] = 34.48, 95% confidence interval [CI]: 1.59–1,000, P < 0.02), it was found that patients with stage IV COPD and patients with more than three exacerbations per year were at higher risk of osteoporosis.

From Japan and Southeast Asia, four studies were detected. The most important of them was Lee and Kwon,[25] as it had the largest sample size with 1081 patients representing alone 28.33% of all patients included in this review. In this, they found that 17.7% of patients with COPD were diagnosed with osteoporosis. Patients with osteoporosis had lower euro quality of life five-dimensional system scores as compared to the non-osteoporotic group. After multivariate analyses, they found that older patients were at a higher risk of osteoporosis (OR = 1.10, P < 0.001), male patients with higher BMI were at lower risk as compared to female ones (OR = 0.74, P < 0.001), and low health-related quality of life was associated with osteoporosis (β = −0.21, P = 0.023).

Lin et al.[29] found that 40% of patients included in the study were osteoporotic. After multivariate logistic regression, they concluded that BMI and FEV1 were linked to increased risks of osteoporosis among patients with COPD with (OR = 0.824, 95% CI: 0.73–0.93; P = 0.002) and (OR = 0.360, 95% CI: 0.13–0.98; P = 0.046), respectively, and patients with both conditions had a significantly higher Modified Medical Research Council (mMRC) dyspnea scores, higher COPD Assessment Test (CAT) scores, and lower oxygen-cost diagram scores as compared to normal COPD population with (1.7 ± 0.8 vs. 1.4 ± 0.8; P = 0.046), (14.7 ± 8 vs. 11.5 ± 7; P = 0.019), and (4.8 ± 1.8 vs. 5.4 ± 1.6; P = 0.045), respectively.

Sakurai-Iesato et al.[18] concluded that 46% of participants were osteopenic and 14% were osteoporotic. They found that CAT and FEV1 scores were correlated with BMD of patients with COPD. T-scores for GOLD category D were particularly lower in patients of COPD with osteoporosis (A [−0.98], B [−1.06], C [−1.05], and D [−2.19]; P < 0.05). Watanabe et al.[38] included 136 subjects in the study. The study suggested that osteoporosis was extremely underdiagnosed among patients with COPD as only four medical diagnoses for osteoporosis were encountered, despite the fact almost 50% of patients were osteoporotic.

From Europe and North America, six studies met the criteria for this review. Combined they represented 50.4% (n = 1923). According to Duckers et al.,[19] 57% of patients in the study had osteopenia and 17% had osteoporosis, in patients with COPD, bone biomarkers were inversely proportional to hip BMD as compared to control group (r = −0.51, r = −0.67, r = −0.57; P < 0.05). However, they did not indicate a significant relation to lumbar and spine BMD. In a study by Camiciottoli et al.,[23] they compared comorbidities of COPD and showed that osteoporosis had a significant impact on patients with COPD and accounted for almost 30% of all comorbidities. Graat-Verboom et al.[30] concluded that osteoporosis is more prevalent among patients with COPD than what was anticipated and they detected a significant increase of osteoporosis among them with 14% in 3 years and lower 25-OH Vitamin D was associated with 7.5 times increased risk of osteoporosis.

Graat-Verboom et al.[32] conducted a study in the Netherlands. Of all participated patients, 41% had osteopenia and 21% had osteoporosis. Even though 21% of patients were on oral corticosteroids and 75% used inhalation form, only 18% of all patients were using bisphosphonates, calcium supplementation, vitamin D, or a combination of them. In fact, osteoporosis was extremely undertreated among COPD, that is, approximately 80% of patients with osteoporosis did not receive or take any bone medication. The BMI and FFMI were irregular in 85% of patients with BMI mean ± SD (osteopenic = 24.3 ± 0.3 kg/m2 and osteoporotic = 21.2 ± 0.3 kg/m2) and FFMI mean ± SD (osteopenic = 15.75 ± 0.2 kg/m2 and osteoporotic = 14.5 ± 0.2 kg/m2).

The presented meta-analysis in Figure 2 was conducted using inverse variance random effect model. The meta-analysis favored osteoporotic-positive group and the diamond of the point estimate was fairly distant from the line of no effect of the pooled results, which indicates that the data were statistically significant. It was also evident from the P-value test of the overall effect (Z = 2.37 [P = 0.02]) that the data were statistically significant. However, the meta-analysis also showed a considerable heterogeneity among the included studies. Confidence interval seemed to be quite scattered around and did not overlap sufficiently, and the studies with a considerable heterogeneity seemed to differ by more than what is usually anticipated to be caused by a chance.

Figure 2

Meta-analysis for osteoporosis among patients with COPD

Subgroup analysis was conducted using RevMan-5 to investigate the presented heterogeneity [Figure 3]. It was clear that there are major differences among the studies, for example, the size extremely varied from only 30 participants up to more than a thousand of them; some studies recruited a few patients with severe COPD, some studies included patients who were on corticosteroids treatments and/or on osteoporosis medications, whereas others excluded them; some studies assessed patients with multiple chronic complications, whereas some did not; and some studies were more precise and took daily follow-ups, whereas others did not.

Figure 3

Subgroup analysis for osteoporosis among patients with COPD

In fact, the detected statistical heterogeneity can be caused by too many factors. Factors such as clinical heterogeneity, which is always present in clinical studies, variation in weight, differences in sample size, study design differences, patients’ status and condition, research setting, research quality, study outcomes, type of intervention, and methodological/structural differences.

Population differences, ethnic diversity, and sample size played a major role in the presented heterogeneity. Among the interventional studies, the study by Ferguson et al.[37] was significantly larger in size, which made it stand out as an outlier, which negatively impacted the homogeneity even though it had a good quality. The removal of this study from analysis markedly abated the heterogeneity from (τ² = 0.29, χ² = 31.49, df = 2 [P < 0.00001], I² = 94%) to (τ² = 0.01, χ² = 1.23, df = 1 [P = 0.27], I² = 18%).

It is also well known that age can influence the progression of both conditions. In fact, the studies extremely varied in terms of age of participant. In a study by Sakurai-Iesato et al.,[18] patients on average were older than 70 years, and in a study by Lin et al.,[29] most of the patient were around 73, whereas in a study by Li et al.,[39] the majority of patients were younger than 60 years and in a study by Nayyar et al.,[27] all of them were younger than 60 years. Generally, younger patients tend to have a better overall health status and a higher BMI as compared to the older ones.

Osteoporosis positive studies were homogeneous (heterogeneity test: τ² = 0.00, χ² = 5.66, df = 5 [P = 0.34], I² = 12%) and they overlapped sufficiently, which represented a consistent, harmonized, and statistically significant results (test for overall effect: Z = 27.64; P < 0.00001) and which can be attributed to the structural similarities among the studies as all of them were noninterventional. Except for the study by Duckers et al.,[19] all of them were conducted at tertiary care settings, and most of them were stratified. Also patients’ age categories seemed to be not so scattered and remained around 63 and 73 years. Even though osteoporosis negative studies were also homogeneous, in combination, they represented much smaller sample size (n = 490, 12.8%) as compared to interventional and osteoporosis positive studies (n = 802, 21% and n = 2164, 56.7%, respectively).

Around the line of no effect, inconsistent results were detected representing only 9.4% of patients. There are some variations but the effect in most cases seemed to fall in a reasonably narrow range except for the study by Abbasi et al.;[24] furthermore, the studies presented a moderate heterogeneity (τ² = 0.04, χ² = 7.47, df = 3 [P = 0.06], I² = 60%) and when we pulled the study by Abbasi et al.[24] out of the analysis, the studies were completely homogeneous and favored osteoporosis-exposed group (τ² = 0.00, χ² = 1.10, df = 2 [P = 0.58], I² = 0%) with statistically significant test results for overall effect (Z = 3.88; P = 0.0001).

To recapitulate, overlapping was evident among some studies on both sides but studies with higher weight and better quality were in favor of osteoporotic-positive group. The point estimate and confidence interval of the pooled results were statistically significant; however, the true effect size varied from one study to another. In fact, we might be introducing a little bet of error by pooling all the studies, yet it will be less error than if we work with groups of three, four, or five studies because that will be a much potent source of errors as the smaller the number of the included participants the more likely to get extreme results.

Discussion

Osteoporosis prevalence among COPD was higher than that among healthy subjects and varied widely (osteoporosis, 14%–66% and osteopenia, 18%–65%), and those rates were close to previous work more than 10 years ago.[33] Osteoporosis in patients with COPD was correlated to the disease severity and low BMD, and the treatments for osteoporosis among patients with COPD were extremely underinvestigated.

Sakurai-Iesato et al.[18] detected a low prevalence rate for osteoporosis, yet in case of osteopenia, it showed a high percentage of patients with osteopenia with 14% and 46%, respectively [Table 1]. This is due to the low number of patients from category D 18%, which had the lowest T-scores (−2.19), P < 0.05) as compared to other categories such as B with 46% and C with 36%. In a study by Watanabe et al.,[38] they confirmed the link between COPD and reduced BMD and they attributed high prevalence rates to the recruited sample as 79.41% of patients had vertebral fractures.

However, even though Rubinsztajn et al.[45] detected an abnormal BMD in almost 75% of patients with COPD who were screened for osteoporosis, they noticed no difference in the frequency of bone metabolism among the included groups.[45] Although Gazzotti et al.[46] noticed that the frequency of osteoporosis in the COPD group was 29.7% and 18.3% in control group (P = 0.007), the frequency of vertebral fractures was 18.6% in the COPD group and 9.0% in control group (P = 0.06).

The highest prevalence for osteopenia was found in a study by Gupta et al.[20] In this study, 93.87% of patients were identified as either osteopenic or osteoporotic, which was surprisingly high and can be explained by small sample size and exclusion of patients receiving any type of bone loss medication (DXA for osteoporosis, −2.1 ± 0.9 and for osteopenia, −1.50 ± 0.35, P < 0.001). Studies represented the older population[222431] were also relatively high in the prevalence of osteoporosis among patients with COPD (65.7%, 61%, and 66.6%, respectively).

Despite the importance of FFMI, it was only assessed in three studies of the 18 studies recruited in this review. All of the studies that assessed FFMI were conducted in Europe,[193032] with 17 ± 0.93 for FFMI representing 17.66% of patients. Duckers et al.[19] after investigating patients’ BMD at hip area found it lower than control group (P < 0.05), whereas at lumbar area, it showed no difference. FFMI (r = 0.51, P = 0.004) in those patients was related to lower hip BMD, also both FEV1 and FFMI (P < 0.05) were correlated to low hip BMD with R2 = 0.35.

In a three-year long prospective clinical study, the progression of osteoporosis among patients with COPD was analyzed by Graat-Verboom et al.[30] They found 44% of participants reported having more than one comorbidity with COPD and the prevalence of osteoporosis increased from 47% at baseline to 61% at the end of the trial. The mean FEV1% predicted for patients declined in follow-ups from 63.4 ± 1.9 to 59.5 ± 2.1 (P = 0.001) and the percentage of patients with COPD in III and IV stage combined increased from 28.88% to 44.44%. The increased severity of COPD was associated with increased percentage of patients diagnosed with osteoporosis.

Lower T-scores at trochanter regions were associated with vitamin D deficiency, and patients who developed osteoporosis had declined FEV1 after conducting pulmonary function test (PFT) (8%; P = 0041). In fact, the increased systemic inflammatory mediators among patients with COPD might act as a stimulator or a trigger for bone resorption process,[3435] which can contribute to the development of osteoporosis in COPD and can increase the risk for bone fractures.[36]

In contrast, Ferguson et al.[37] concluded after adjustment for age and gender that no association was observed between patient’s spirometry results and impaired BMD. They attributed the dissimilarity to the differences in the applied methodologies, as epidemiologic studies usually obtain data from clinical surveys based on population and use clinical information for the determination of severity. However, the majority of patients were in severe stages of COPD and 57% of patients did not complete the trial, which had extremely limited the power to detect differences in BMD in this study.

Only two article very briefly addressed the treatment of osteoporosis among patients with COPD. Graat-Verboom et al.[32] found that only 18% of patients with COPD in the study were treated with bone medication, 23% of patients with osteopenia were prescribed bone medication, and only 8.1% of patients with COPD received bone pharmacological therapy. After conducting univariate analysis, it was noticed that patients with COPD without bone medication have had three times higher risk for osteoporosis between the age of 55–65 years, and 4.5 times higher risk to become osteoporotic when the patient is older than 65 years of age. The risk for osteopenia was three times higher for patients older than 65 years of age. The link between these two conditions does not seem to be strange, especially if the systemic inflammatory issues were taken into consideration.[32] Also both conditions have interconnected overlapping characteristics and might share some clinical features.[37]

The study by Ferguson et al.[37] was part of TORCH International project and was conducted in USA. The overall prevalence of osteoporosis at baseline was high (65%), yet over three years, the changes in patients’ BMD were not so significant. However, before participating in the study, 30% of patients were on medication for BMD loss (7% bisphosphonate and 23% other drugs) and while conducting the study, 20% more developed BMD loss and started bone therapy (11% bisphosphonate and 9% other drugs), so the number of patients with BMD loss was significantly higher than the global safety population (10% before and 7% more during the trial).[43]

In contrast to other studies,[4041] Silva et al.[21] noticed that white patients were less prone to osteoporosis than other races, whereas Li et al.[39] concluded that in patients with severe COPD, osteoporosis showed no difference in terms of prevalence between Caucasian and African American, and they recommended osteoporosis screening to all patients with severe COPD regardless of ethnicity. However, Lee et al.[47] concluded that COPD comorbidities like osteoporosis are affected by race and ethnic group and suggested that different strategies to be adapted for the optimal management of COPD considering the ethnic group of patients. In fact, the racial factor in osteoporosis is a very complex and controversial topic and applying “one-size-fits-all” approach will not be suitable.

Conclusion

Osteopenia and osteoporosis were highly prevalent among the patients with COPD. Even though the specific pathophysiological interaction between both conditions is still unclear, it is very reasonable to call for osteoporosis screening in patients with COPD, especially those who are above 65 years of age, in advanced COPD stages, with BMI lower than 21 kg/m2, or with FFMI lower than 16 kg/m2 for males and 15 kg/m2 for females.

There is a lack of information as well as a lack of studies investigating severity and treatment of osteoporosis in patients with COPD. Underassessment of osteoporosis among patients with COPD seemed to be a health-care weakness rather than a COPD-related problem.

Financial support and sponsorship

The authors acknowledge that this project was sponsored by Bridging/ Bridging-Incentive Grants by Universiti Sians Malaysia (Grant Number: 304.PFARMASI.6316508).

Conflicts of interest

There are no conflicts of interest.