INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Survival in cystic fibrosis (CF) has improved during the last three decades (1). The reasons for this include care in specialized centers and greater attention to antibacterial therapy and nutrition (2, 3). Despite these improvements in management the major cause of death in adults with CF remains respiratory failure (1). The main indicators of a poor outcome are progressive loss of pulmonary function and failure to maintain body weight, though the basis of any interaction between these variables is unclear.

Lung structure is normal during fetal life, but early in life repeated infections lead to bacterial colonization and chronic inflammation with progressive lung destruction (4). An early clinical feature of lung disease is the development of hyperinflation, which increases with further lung injury (4). Progressive hyperinflation changes the shape of the thorax, putting the inspiratory muscles, particularly the diaphragm, at mechanical disadvantage. Additionally, decreased chest wall compliance increases the energy and oxygen costs of breathing (5). Together they may be a factor in the increase in resting energy expenditure (REE) associated with CF (6). Progression of this state secondary to chronic infection and continuing lung injury leads to the development of respiratory failure.

Short-term weight loss occurs with exacerbations of respiratory symptoms and may be reversed after antibiotic treatment. Longer term decline in nutritional status with a reduction in total body weight occurs in a high proportion of patients as the pulmonary disease progresses. A potential effect of weight loss in patients with an already reduced fat mass is that skeletal muscle, including inspiratory muscle, may be lost in the catabolic state associated with chronic pulmonary infection (7). This should affect the functional capacity of all skeletal muscle, including the inspiratory muscles, though comparison of leg muscle strength with inspiratory and expiratory muscle strength in patients with CF indicated preferential preservation of strength in the respiratory muscles (10). The loss of leg muscle strength was considered to be due to loss of muscle mass. Similar preservation of respiratory muscle strength in CF was reported to be related to body mass index (BMI) in a group of patients with a range of severity of pulmonary compromise (11). However, most of the patients had a near normal body mass or BMI, and the same measures made in a malnourished hyperinflated group of patients with CF demonstrated a loss of respiratory muscle strength (12).

Maximal inspiratory and expiratory pressures (MIP and MEP, respectively) give an indication of muscle strength, but do not indicate the endurance capabilities and work capacity of the respiratory muscles (10). In patients with non-steroid- dependent asthma, preservation of inspiratory muscle strength (IMS) and loss of endurance occurred and the dissociation was greater still in oral-steroid-dependent patients (13).

We hypothesized that the combination of chronic lung infection and poor nutritional status would alter body composition and have a negative influence on skeletal muscle function, including, in particular, the inspiratory muscles. Secondly, we hypothesized that the impact on the inspiratory muscles would be greater on the capacity to carry out work than on short duration tests of strength. To test this hypothesis in adult patients with CF we determined the relationship between body composition, IMS by MIP, and sustained maximal inspiratory pressure (SMIP) through the full inspiratory volume, a maneuver that reflects single-breath maximal work capacity.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

Survival study. To determine the effects of inspiratory muscle function, FEV₁, and nutritional status on survival we studied 49 adult patients (26 female) with a proven diagnosis of CF based on appropriate clinical features and sweat Na⁺ and Cl > 70 mml/L. At the time of study the mean age (SD) was 23.7 (4.4) yr. The results were analyzed 20 mo after the original collection of data.

Cross-sectional study. To determine the relationship of body composition with inspiratory muscle function we studied 25 patients (10 female) with proven CF; mean age (SD), 22.9 (3.8) yr. All were clinically stable and none was receiving oral corticosteroid treatment. The mean % predicted FEV₁ was 61.1%; range, 18 to 113%.

For both studies, patients were recruited from outpatients attending the Adult CF Center in Cardiff. Patients with diabetes mellitus or other metabolic diseases were excluded from the study. Twenty-five age- and sex-matched healthy subjects without CF were recruited for inspiratory muscle and lung function measurements. This study had Local Research Ethics Committee approval, and all subjects gave written informed consent.

Measurements

Anthropometric measurements. Weight and height, barefoot, were determined with a beam scale (kg) and stadiometer (cm), respectively. BMI was calculated from the ratio of weight:height² (kg/m²). Ideal body weight (IBW) was obtained from the Metropolitan Life Insurance Table (14). Skinfold thickness was measured (for patients in the cross-sectional study) with a Holtain caliper at four sites: biceps, triceps, iliac, and subscapular. The mean of three readings was recorded for each site (15). Fat mass was calculated according to the formulae (16).

For men:

For women:

Lean body mass (LBM) was determined as the difference between measured body weight and calculated fat mass (15). Mid-upper arm circumference was determined with a flexible tape at the midpoint of the left arm between the acromion process and the tip of the olecranon (15). The mid-upper arm muscle circumference (AMC) was calculated as follows: mid-upper arm circumference (cm)  ( × triceps skinfold) (16). The measured values for AMC were compared with standard values for age and sex (17). All anthropometric measurements were made by the same researcher (A.A.I.).

Nutritional depletion was defined as: a BMI < 20 kg/m² or a LBM:IBW < 69% for men and < 67% for women (18).

Inspiratory muscle function. The MIP was determined using an electronic manometer and computer software designed by K.C. The manometer had a fixed leak via a 2-mm-diameter aperture that prevented glottal closure during the inspiratory maneuver. The leak set a maximum flow during the inspiratory effort at 450 ml/s and allowed continuous measurement of pressure over a full range of lung volume changes until no further pressure could be generated. This measure of force was termed sustained maximum inspiratory pressure (SMIP).

The patient was asked to take a maximal and sustained inspiratory effort from residual volume (RV) to TLC, and pressure (cm H₂O) was recorded over time by a computer (Figure 1). MIP was the maximum pressure developed in the first second of the inspiration; SMIP was measured as the area under the pressure-time curve (Figure 2).

View larger version (125K): [in this window] [in a new window]   Figure 1.   Computer system used for the assessment of inspiratory muscle function.

View larger version (10K): [in this window] [in a new window]   Figure 2.   Pressure-time profiles generated on the computer screen by two female patients of similar age and lung function severity but different lean body mass (LBM). Although the maximum pressure generated in the first second (MIP) is similar, the patient with low LBM:IBW% has a lower capacity to maintain pressure over time, as suggested by the smaller area under the curve (SMIP).

MIP was expressed in absolute values (cm H₂O) and as % predicted (19). SMIP results were expressed in absolute values (pressure- time units) and as percent of normal, these values having been developed for age- and sex-matched healthy subjects without CF (20). MIP and SMIP were also expressed as a ratio to LBM.

Peripheral skeletal muscle function. For the cross-sectional study sustained handgrip (HG) force was measured using a hand dynamometer as an indicator of nonrespiratory skeletal muscle force. Results were expressed as kg force and related to LBM.

Lung function tests. The following were determined: FEV₁, FVC, and FEV₁/FVC ratio by dry wedge spirometry (Vitalograph Ltd., Buckingham, UK); TLC, RV, RV/TLC, and FRC by helium dilution technique (21); TL_CO by the single-breath method and the transfer coefficient (Kco) by the TL_CO corrected for the alveolar volume (21). All lung function variables were expressed as percent of the predicted value for age, height, and sex (22).

C-reactive protein. Serum C-reactive protein (CRP) was measured by double antibody sandwich enzyme-linked immunosorbent assay.

Statistical Analysis

Descriptive statistics were used for the mean and SD. Student's t test was used for comparison between groups. In non-normally distributed data, log₁₀ transformation was performed prior to applying the t test (23). Spearman's rank correlation was used for comparisons between lung function and body composition parameters. Chi-square test and distribution of percentiles were used for the analysis of difference in severity according to sex. ANOVA was used for the analysis of this relationship between the parameters of inspiratory muscle function and the nutritional indices. Data analysis was carried out using the SPSS software package. Differences were considered to be significant if p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Survival Study

Of the 49 patients studied, 29 were nutritionally depleted (BMI < 20 kg/m²) and the remaining 20 were nutritionally replete (BMI > 20 kg/m²). The patients with a low BMI had lower values for MIP and SMIP, although the MIP was better preserved than the SMIP (Table 1).

Patients with a BMI < 20 kg/m² had a lower mean FEV₁ than did replete subjects (Table 1). Female patients had greater impairment of FEV₁, than did male patients; mean (SD) % predicted FEV₁, 35.2 (4.2) and 47.4 (6.6), respectively, p < 0.01. Analysis of the distribution of % predicted FEV₁ by percentiles revealed lower values for female patients: 50th percentile, 32% compared with 38.5%, and at the 75th percentile, 51.5% compared with 75.5% for female and male patients, respectively.

Mortality was assessed 20 mo after the initial measurements. Ten patients died during this period (four female and six male) and two underwent heart-lung transplantation (one female and one male). In terms of severity of pulmonary disease, death or transplantation were considered as equivalent end points and the data from these two groups were pooled for analysis. The patients who died or received a transplant had lower BMI values at the time of assessment than did the survivors, mean (SD), 16.5 (3.1) kg/m² compared with 19.8 (3.4) kg/m², p < 0.01.

Both MIP and SMIP were lower in the patients who later died or received a transplant, but MIP was better preserved than SMIP: MIP mean % predicted (SD), 64.1 (23.9) and 85.5 (28.4); SMIP mean % predicted, 27.3 (19.9) and 51.5 (32.7), for non-survivors/transplanted and survivors, respectively. This corresponds with the greater impairment of lung function in those patients who subsequently died or received a transplant, compared with the survivors; FEV₁ mean % predicted (SD), 16.7 (6.9) and 54.4 (23.8), respectively.

Cross-sectional Study

Of the 25 patients, six had a body weight < 90% of IBW, 12 had a BMI < 20 kg/m², and eight had a low LBM:IBW%. The mean BMI (SD) of the group was 21.3 (3.8) kg/m². Patients with a low LBM:IBW ratio, mean (SD) 61.6 (5.2)%, were compared with those with a normal LBM:IBW ratio, 80.5 (4.9)%. The low LBM:IBW% group had a nonsignificantly lower FEV₁ (p = 0.07). When related to absolute LBM, no difference was found in FEV₁ between the low and normal LBM:IBW% groups; mean (SD) FEV₁% predicted/kg LBM was 1.29 (0.66) compared with 1.35 (0.46), respectively. There was no correlation between FEV₁ and LBM:IBW% (p = 0.1). Fat mass (FM:IBW%) was also reduced in the patient group compared with published values for young adults; mean (SD) female patients 21.8 (10)% compared with 25 (7)% and male patients 13.2 (8)% compared with 14 (6)% (24).

All parameters of inspiratory and arm muscle force were significantly less in patients with a low LBM:IBW% compared with those with a normal LBM:IBW% (Table 2). When related to absolute LBM, the MIP, SMIP, and HG force were not different between the low and normal LBM:IBW% groups; mean (SD) MIP/LBM, 0.45 (0.07) and 0.51 (0.13) cm H₂O/kg; SMIP/LBM, 7.78 (3.54) and 10.73 (3.13) pressure time units/ kg; and HG/LBM, 0.64 (0.06) and 0.73 (0.17) kg force/kg, respectively.

The mean MIP of the whole group was within normal limits, whereas the FEV₁ and SMIP were significantly less in the patients than in the subjects without CF (Table 3). Patients with a FEV₁ < 70% predicted had a lower SMIP (p = 0.03) than the patients with a normal FEV₁; differences in MIP were not related to FEV₁ (p = 0.1). SMIP, MIP, and HG correlated to LBM:IBW%, and SMIP was related to HG and to AMC (Figures 3-7). An AMC < 70% predicted was associated with a reduction in SMIP p < 0.01, MIP p < 0.05, and HG p < 0.05 when compared with an AMC > 70% predicted. LBM:IBW% and AMC were related r = 0.45, p = < 0.05. 

View larger version (9K): [in this window] [in a new window]   Figure 3.   Correlation between SMIP and LBM (r = 0.7; p < 0.001).

View larger version (9K): [in this window] [in a new window]   Figure 4.   Correlation between MIP and LBM (r = 0.5; p < 0.05).

View larger version (8K): [in this window] [in a new window]   Figure 5.   Correlation between handgrip force and LBM (r = 0.5; p < 0.05).

View larger version (9K): [in this window] [in a new window]   Figure 6.   Correlation between SMIP and handgrip force (r = 0.8; p < 0.001).

View larger version (10K): [in this window] [in a new window]   Figure 7.   Correlation between SMIP and AMC (r = 0.6; p < 0.05).

No difference in the RV was found between the patients with low LBM:IBW% and those with normal LBM:IBW%; mean (SD) RV% predicted, 128.4 (41.3) versus 123 (50.8), respectively (p = 0.72).

Analysis of variance showed a significant effect of LBM: IBW% on SMIP (p < 0.001) even after allowance for spirometry. Taking SMIP as the dependent variable, the effect of LBM:IBW% (p < 0.001) on SMIP was greater than either BMI (p = 0.52) or FEV₁ (p = 0.44). Using multivariate analysis to compare the source of variation with sex allowed for as a covariable, LBM:IBW% contributed 39.4% to the variation of SMIP compared with 2.7% by the FEV₁. For MIP the contribution of LBM:IBW% was 26.6% and of the FEV₁ 0.1%, whereas for HG the contribution of LBM:IBW% was 36% compared with 2.6% for the FEV₁.

The patients with a low LBM:IBW% had a greater circulating concentration of CRP than did those with a normal LBM:IBW% (p < 0.01), although all patients were clinically stable when studied (Table 2). Circulating CRP concentrations were negatively correlated with LBM:IBW% (r = 0.71, p < 0.01).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The survival and cross-sectional studies showed a clear relationship between body composition and impairment of inspiratory muscle function in adults with CF. The significant association between low LBM, impaired inspiratory muscle function, and reduced HG force suggests a general reduction in skeletal muscle function, rather than a problem limited purely to the inspiratory muscles. Reduction in the BMI and the LBM:IBW% separated those patients with a lower MIP and low SMIP from those with normal inspiratory muscle function. However, in both studies MIP was better preserved than SMIP, which suggests inspiratory muscle strength and capacity for work are differentially affected by the loss of skeletal muscle mass. Supporting the interpretation of the loss of LBM causing a general deficit in skeletal muscle mass is the finding that MIP, SMIP, and HG related to absolute LBM showed no separation between the deplete and replete subjects. This view is further emphasized by the high level of association between SMIP and LBM, SMIP and HG, and SMIP with AMC, the latter a further index of skeletal muscle mass, and the lesser relationships between MIP and these indices. These findings suggest that the loss of inspiratory muscle function is part of a general reduction in skeletal muscle mass rather than a change in the quality of muscle function. This interpretation is supported by the reported correlation between muscle volume and peak work rate (25).

The measurement of SMIP appears to be a better indicator of survival and more related to LBM changes than MIP. SMIP is a measure of pressure generation over the full range of volumes from RV to TLC and is an indicator of single-breath work capacity. The pressure profile is inversely related to lung volume. Hence, the peak negative pressure generated by the inspiratory muscles occurs as the subject inspires maximally from RV, pressure will gradually fall as inspiration continues through a fixed leak until TLC is reached giving the characteristic pressure-time curve (Figure 2). The SMIP curves for CF probably reflect a combination of loss of inspiratory muscle mass and altered pulmonary mechanics. With progression of the disease the FEV₁ decreases further and is probably associated with an increased RV and further compromise of IMF in addition to nutritional factors. Diaphragmatic muscle mass is related to total body weight, and loss of as much as 30% of predicted body weight was associated with a reduction in diaphragmatic mass, thinning of the muscle, and a reduction in its area, which was related to sarcomere loss in patients with COPD (26). In animal models, diaphragmatic endurance was reduced in nutritionally depleted animals (27). In humans, the transdiaphragmatic area is reduced in malnutrition and may be a factor in the loss of muscle contractile force per unit of area in this state (26). Similar studies in CF are not available for comparison, but if the same happened in malnourished patients with CF then the capacity to maintain a relatively normal initial tension in the muscle, but not a sustained force, could be expected. This would explain why SMIP is differentially affected in patients with a low LBM:IBW% compared with a preservation of MIP. Such findings also support our view that muscle mass loss is the likely cause of reduced muscle function. A similar dissociation between MIP and sustained inspiratory effort occurred in patients with asthma, but not in patients with COPD who showed a reduction in both inspiratory muscle strength and endurance (13). The loss of endurance was even greater in patients with steroid-dependent asthma, suggesting that the metabolic-myopathic effects of corticosteroids had their major effect on endurance.

Studies of respiratory muscle function in CF are difficult to compare because of differences in the age of subjects studied, the methodologies used, and the indices of muscle function used. Inspiratory muscle work has been infrequently studied, making our findings with SMIP difficult to compare with other data. MIP or MEP have been the main measures used (28). The tendency has been for MIP, in particular, to appear to be well preserved in patients with cystic fibrosis (10, 27), which is in keeping with our findings, though the patients have tended to be younger than ours and the degree of malnutrition has been modest, making direct comparisons difficult. In hyperinflated malnourished young adults with CF, MIP and MEP were reported to be reduced and were related to arm muscle circumference, an indicator of skeletal muscle mass, which supports our findings in patients with low LBM:IBW% and AMC (12). The effect of sex in our study was probably related to severity of pulmonary disease and had a greater impact on SMIP than on MIP. The reduction in pressure generation at higher lung volumes was also associated with greater mortality in the survival study.

Support for our results comes from studies in COPD where hyperinflation and muscle wasting may be cofactors in the loss of inspiratory muscle function (32). Hyperinflation, causing a training effect, was postulated to be the mechanism of preservation of the MIP and MEP in CF (10). In our study there was no separation on the basis of RV in patients with normal or low LBM:IBW%. Hence, in our malnourished adult patients with CF, the contribution of pulmonary mechanics to loss of inspiratory muscle function was effectively controlled for and appears to be of less importance than loss of skeletal muscle mass, though the method of assessing RV and TLC in our study was not robust. However, SMIP measures pressure generation at a full range of absolute lung volumes and therefore included in the measurement is the impact of both the degree of hyperinflation and its mechanical consequences, which strengthens our view of the importance of loss of skeletal muscle mass.

This raises the question as to how muscle mass is lost. Weight loss in CF is multifactorial and includes, possibly, the energy cost of the gene defect itself, poor energy intake, increased REE, less efficient pulmonary mechanics, and catabolic intermediary metabolism secondary to pulmonary infection and inflammation (5, 7). These mechanisms are likely to lead to a negative energy balance with consumption of host tissues as substrates. This is most likely to be noted during periods of worsening respiratory symptoms and may be responsible for the associated weight loss. The main impact of weight loss in adults with CF is likely to fall on the skeletal muscle mass as there is usually a reduced fat mass. The inverse relationship between the circulating CRP and maintenance of LBM supports a link between changes in body composition and pulmonary infection. CRP is a downstream product of the activity of cytokines such as IL-6 and possibly IL-1 and TNF-. These pro-inflammatory cytokines are also potentially metabolically active and tend to cause catabolic changes and have been implicated in wasting states.

Accepting the limitations of this study, it still provides evidence to support the hypothesis of a link between inspiratory muscle function and body composition in CF. The measurement of pressure generation throughout the range of inspiratory volumes gives new insights into inspiratory muscle function and allows conclusions to be drawn regarding inspiratory muscle bulk and the severity of pulmonary impairment, which cannot be made from measurement of MIP at RV or FRC.