Patients

The study population was derived from a cohort of 151 adults with cystic fibrosis who were participating in a prospective study documenting the natural history of cystic fibrosis related bone disease.^2,12 Of the 123 patients who attended for measurements of total body BMC on recruitment to the study, 100 patients (54 male) re-attended for further BMC measurements a median (25–75% interquartile range (IQR)) of 12.0 (12.0–12.8) months later. The mean (SD) baseline total body BMC in patients with follow up BMC measurements was 2133 (390) g compared with 2086 (424) g in patients without follow up BMC measurements (p = 0.61). Of the 23 patients who were not rescanned, eight had died, four had received a double lung transplant, and 11 (9%) had chosen to withdraw from the study. The study was approved by the South Manchester research ethics committee and each patient provided informed written consent.

Bone mineral content measurements

Total body BMC was measured on recruitment to the study and at follow up by dual energy x ray absorptiometry (QDR 4500 Acclaim densitometer, Hologic, Bedford, Mass, USA). The short term precision for the total body BMC measurement, expressed as the coefficient of variation, was 0.92%. The BMC results were expressed as absolute values (in grams) from which the annual change and annual percentage change in BMC were calculated.

Biochemical measurements

Morning blood and urine samples were taken from patients on recruitment to the study and at follow up, at a time when their lung disease was stable—that is, not prescribed oral or intravenous antibiotics for the treatment of a pulmonary exacerbation. Urine samples were not necessarily first void as participants attended as outpatients from all over the North West region. Plasma IL-6 levels were measured using Quantikine immunoassays (R&D Systems Europe Ltd, Abingdon, UK) and the intra- and inter-assay coefficients of variation were 2.6% and 4.5%, respectively. The sensitivity of the assay was 0.7 pg/ml (range 3.1–300). Serum concentrations of C reactive protein (CRP), osteocalcin, bone specific alkaline phosphatase (BSAP), and urinary deoxypyridinoline crosslinks were measured as described previously.²

Clinical assessment

The best recorded FEV₁, weight, and height in the 4 months before the recruitment and follow up visits were used to calculate the percentage predicted FEV₁¹³ and body mass index (BMI). Sputum microbiology and prescriptions of inhaled corticosteroids, oral corticosteroids, nebulised antibiotics and intravenous antibiotics were recorded throughout the study as these factors could modify the inflammatory response to infection. Levels of physical activity were assessed using the Baecke questionnaire which scores levels of physical activity from a minimum of 3 to a maximum of 15, as described previously.²

Statistical analysis

Data were analysed using SPSS version 9.0 (SPSS Inc, Chicago, IL, USA) and variables were summarised using the mean (SD) or median (25–75% IQR). The annual change in BMC was calculated for each patient to correct for the varying time between recruitment and follow up. For the biochemical indices of bone turnover together with CRP and IL-6, the mean value of the two readings obtained from each individual was used for statistical analysis. A one sample t test was used to determine whether the annual change in BMC was significantly different from zero. The independent samples t test or the Mann Whitney U test was used to determine differences between patient groups. Spearman’s rank correlations were performed to explore the relationship between markers of systemic inflammation, markers of bone turnover, and the change in BMC over 1 year. Linear regressions (single and multiple variables) were used to characterise the relationship between the annual change in BMC (dependent variable) and the following predictors: age at baseline, sex, B cepacia colonisation, daily prednisolone use, annual antibiotic use, annual change in BMI, annual change in FEV₁, and mean concentrations of IL-6, CRP, deoxypyridinoline crosslinks, osteocalcin, and BSAP. Forward and backward selection methods were used to identify independent predictors in the multiple variable model. Residuals were examined to check for normality.

Patient characteristics

None of the patients included in this analysis were taking any specific bone sparing agents and none had received a transplant. Four patients were pancreatic sufficient and all pancreatic insufficient patients were prescribed 22.5 μg (900 IU) of oral vitamin D (calciferol) daily, but seven patients admitted to not taking their vitamin supplements on a regular basis. Sixty one patients were homozygous for the delta F508 genotype, 31 patients were heterozygous for the delta F508 genotype, and eight patients had another genotype.

On recruitment to the study the mean (SD) age, percentage predicted FEV₁, BMI, and physical activity score of the patients were 25.6 (7.1) years, 61.8 (24.1)%, 21.3 (2.3) kg/m² and 7.8 (1.3), respectively. Over the follow up period the mean (SD) annual change in FEV₁ and BMI were –2.1 (8.2)% predicted (p = 0.01) and 0.2 (1.1) kg/m² (p = 0.05), respectively. There was no significant change in the physical activity score over the follow up period.

Seventy three patients were colonised with Pseudomonas aeruginosa, 22 patients were colonised with B cepacia and five patients were colonised with Staphylococcus aureus, Haemophilus influenzae or no organism. Of the 73 patients colonised with P aeruginosa, 38 were prescribed daily nebulised antipseudomonal antibiotics (colomycin or gentamicin). In the whole group over the follow up period the median (25–75% IQR) intravenous antibiotic usage was 23 (0–58) days per year. Twenty six patients received no intravenous antibiotic therapy and, in the remaining 74 patients, the median (25–75% IQR) intravenous antibiotic usage was 38 (19–71) days per year.

Nine patients were prescribed long term continuous oral corticosteroids (prednisolone 5–15 mg daily) during the follow up period, three for allergic bronchopulmonary aspergillosis and six for unstable lung disease. In the whole group the median (25–75% IQR) oral corticosteroid usage during the follow up period was 0.6 (0.0–2.3) mg prednisolone per day. Thirty six patients received no oral corticosteroid therapy and in the remaining 64 patients the median (25–75% IQR) oral corticosteroid usage was 1.6 (0.7–4.3) mg/day. Sixty eight patients had been prescribed inhaled glucocorticoid therapy (budesonide 200–400 μg twice daily or fluticasone propionate 250–500 μg twice daily).

Biochemical results

Blood and urine samples were taken at the recruitment and follow up visits from 75 of the 100 patients who re-attended for dual energy x ray absorptiometry. There was no significant difference in baseline BMC between patients with (2136 (399) g) and without biochemical data (2121 (371) g, p = 0.87). Due to the inherent variability of the inflammatory and bone biochemical markers, averaged values from the recruitment and follow up visits were calculated. The median (25–75% IQR) concentrations of these variables were: IL-6 5.5 (2.6–11.5) pg/ml (normal range 0.2–2.9 pg/ml, fig 1⇓), CRP 10.3 (5.0–26.8) mg/l (normal range <15 mg/l), BSAP 18.3 (15.3–24.4) U/l (normal range 10.0–23.0 U/l), osteocalcin 9.5 (7.1–11.6) ng/ml (normal range 3.4–10.0 ng/ml), and deoxypyridinoline crosslinks 6.1 (4.7–8.2) nM/mM (normal range 2.3–7.4 nM/mM).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

 Mean interleukin-6 concentrations in 75 adults with cystic fibrosis (mean value for each patient calculated from the baseline and follow up measurements; dashed lines denote the upper and lower limits of the normal range).

The urinary excretion of crosslinks was positively related to IL-6 (r_s  =  0.27, p = 0.02) but was not related to any other clinical or biochemical indices of systemic inflammation. Bone formation, as measured by bone specific alkaline phosphatase and osteocalcin, was not related to any of the indicators of inflammation.

Median (25–75% IQR) levels of IL-6 were significantly higher in patients colonised with B cepacia (9.5 (5.2–19.5) pg/ml) than in those colonised with P aeruginosa (4.7 (1.9–9.0) pg/ml), p = 0.01. However, there was no significant difference in the CRP, annual intravenous antibiotic use, oral corticosteroid use, levels of physical activity, change in FEV₁, or change in BMI between these microbiological groups (table 1⇓).

Men had significantly higher median (25–75% IQR) levels of IL-6 (7.2 (4.1–13.6) pg/ml) than women (3.9 (1.8–6.4) pg/ml, p = 0.007). Men also had significantly higher levels of CRP (20.0 (7.5–28.0) mg/l) than women (5.5 (5.0–23.5) mg/l, p = 0.01). However, there was no significant difference in annual intravenous antibiotic use, oral corticosteroid use, levels of physical activity, change in FEV₁, or change in BMI between men and women (table 2⇓).

Bone mineral content results

The mean (SD) total body BMC on recruitment to the study was 2133 (390)g in the whole group, 2311 (367)g in male patients and 1919 (303)g in female patients. The mean (SD) annual change in total body BMC was –16.1 (62.1) g in the whole group (p = 0.01), −22.5 (75.1) g in men (p = 0.03) and –8.4 (41.2) g in women (p = 0.18). The mean (SD) annual percentage change in total body BMC was –0.6 (2.8)% in the whole group (p = 0.03), −0.8 (3.3)% in men (p = 0.07) and −0.4 (2.2)% in women (p = 0.21).

The mean (SD) annual reduction in BMC was significantly greater in patients colonised with B cepacia (−64.7 (67.6) g) than in those colonised with P aeruginosa (−3.0 (54.0) g), p<0.001. In the patients colonised with P aeruginosa there was no difference in the annual change in BMC between those prescribed and those not prescribed nebulised antibiotics. In the whole group there was no significant difference in the annual change in total body BMC between patients prescribed oral corticosteroids or inhaled corticosteroids and those not.

The percentage predicted FEV₁ (r_s = 0.21, p = 0.04) and BMI (r_s = 0.32, p = 0.001) at the time of recruitment to the study were significantly related to the baseline total body BMC, which confirms the cross sectional relationship between clinical markers of cystic fibrosis disease severity and BMC. The annual change in BMI was weakly related to the annual change in BMC (r_s = 0.20, p = 0.05) but there was no relationship between the annual change in FEV₁ and the annual change in BMC (r_s = −0.05, p = 0.65).

There were significant relationships between the change in total body BMC and the markers of systemic inflammation (fig 2⇓). In the univariate analysis there was a significant negative relationship between change in total body BMC and mean levels of IL-6 (r_s =  −0.39, p<0.001) and CRP (r_s = −0.34, p = 0.002). There was also a significant negative relationship between change in total body BMC and the clinical indicators of systemic inflammation—namely, days of intravenous antibiotic use (r_s = −0.27, p = 0.006) and oral corticosteroid use (r_s = −0.20, p = 0.045). Mean levels of physical activity were not related to change in total body BMC. When multiple linear regression analysis was used to allow for the interrelationships between variables, IL-6 (coefficient −2.2 (95% CI –3.4 to –1.0) per pg/ml, p = 0.001), colonisation with B cepacia (coefficient −46.8 (95% CI −75.5 to −18.1), p = 0.002), and the annual change in BMI (coefficient 15.4 (95% CI 3.6 to 27.2), p = 0.011) were independently significant predictors of annual change in BMC. Sex was not a significant predictor of annual change in BMC.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

 Univariate relationship between change in total body bone mineral content (BMC) and the mean concentration of interleukin (IL)-6, mean concentration of C reactive protein (CRP), annual intravenous antibiotic use, and oral prednisolone use.