Minimal Clinically Important Differences in COPD: Body Mass Index and Muscle Strength

Abstract

Weight loss, muscle wasting, as well as muscle dysfunction are recognized as important problems in COPD, contributing to morbidity and mortality. This paper discusses body weight and muscle function as possible outcome parameters in the management of COPD. The relationship between these outcome measures and COPD-related management goals is discussed. Minimal clinically important differences (MCID) in the approach of patients suffering from COPD for these measures are discussed.

• Body weight
• Muscle mass
• Muscle function
• MCID
• COPD

Introduction

A minimal clinically important difference (MCID) is defined as being the smallest difference in score in the domain of interest that patients perceive as beneficial and that would mandate, in the absence of troublesome side effects and excessive cost, a change in the patient's management [[1]]. The MCID holds the promise of being able to define a threshold that distinguishes when a person or group has just begun to experience what is an important improvement [[2]].

Goals of effective COPD management are widely accepted and are indeed largely related to the experienced symptomatology of the patients' relief of symptoms, improvement in exercise capacity, improvement in health status, prevention of exacerbations and complications, prevention of disease progression, and reduction in mortality [[3]]. Despite these well-recognized outcome measurements from a patient's perspective, regulatory authorities largely rely on lung function parameters in order to determine positive outcome differences. Indeed, for many decades, COPD has been approached as a disease state characterized by airflow limitation. Expiratory airflow limitation is still now considered as the hallmark physiological change of COPD and the primary outcome in intervention strategies. However, in the literature, evidence is growing to consider COPD as a multi-component disease including systemic effects [[4]]. Weight loss and muscle wasting, as well as muscle dysfunction, are part of these systemic manifestations of COPD. Therefore, in a heterogeneous disease condition as COPD, selection of appropriate outcome measures depends on the disease component being addressed.

Body Mass Index in COPD

Weight loss is a phenomenon that has long been recognised in the clinical course of COPD patients. Attempts to describe different COPD classifications included body weight as an important discriminator [5&6]. In the 1960s several studies reported that low body weight and weight loss are negative predictive factors of survival in COPD [[7]]. At that time, weight loss was considered to be an integral part of the clinical picture of chronic bronchitis. Without adequate analysis of the underlying mechanisms or related functional consequences, nutritional depletion was considered as an inevitable and irreversible terminal event related to the severity of airflow obstruction.

In order to consider body weight as a valuable outcome parameter in COPD, measurement properties as well as the relationship between the outcome measure and COPD-related management goals have to be discussed.

Measurement Properties of Body Weight as an Outcome Measure

The most fundamental properties of an outcome measure, besides its validity, are the standardisation and the scalingproperties. If the human body is considered as a single compartment, the measurement of weight and height provides a simple assessment of nutritional status. Generally, the subject's weight is compared with a reference parameter. Actual body weight can be related to the “ideal body weight” (IBW) as derived from height, sex, and frame size, based on the Metropolitan Life Insurance tables [[8]]. Low body weight is generally and arbitrarily defined as body weight < 90% IBW. Another approach is the use of indices of relative weight as the Body Mass Index (BMI) or Quetelet Index, which is the ratio of body weight divided by height squared. In vivo measurements have demonstrated that the body mass index formula is a valid measure of changes in body mass [[9]].

The U.S. Department of Agriculture and the Department of Health and Human Services in their 2000 report on “Nutrition and Your Health” have proposed a BMI between 18.5 and 24.9 as a target for a healthy weight. A BMI lower than 18.5 kg/m² is considered underweight, a BMI greater than 25 kg/m² connotes overweight and a BMI greater than 30 kg/m² indicates obesity [[10]]. However, it was stated in the same report that weights above or below the healthy weight range may actually be healthy and that weights inside the healthy weight range may not be healthy stressing the assessment of body composition even in persons within the “normal' range. This is particularly important in clinical conditions where standardisation in the nomenclature of body composition in weight loss is important [[11]].

Especially in clinical conditions, differential changes in body composition have to be considered. Specific definitions for cachexia, wasting, and sarcopenia have been proposed. In cachexia, the loss of body cell mass, the total actively metabolising and contracting tissue, is greater than the loss of weight. In wasting, the decrease in body cell mass and weight are parallel. Cachexia is always accompanied by wasting, but wasting does not always lead to cachexia. Sarcopenia refers to an involuntary generalised loss of skeletal muscle mass and strength.

In the absence of a two-compartment analysis of body composition, dividing body weight into a fat and a fat-free compartment, depletion in fat-free mass can be masked by an increase in fat mass resulting in a normal body weight, or, a low body weight can be present with preservation of fat-free mass [12&13]. Therefore, especially in patients with low body weight, further assessment of body compositional changes has to be advocated.

Prevalence of Low Body Weight in COPD

Presence of low body weight is a frequent finding in COPD patients. Among 779 men included in the National Institutes of Health clinical trial of intermittent positive-pressure breathing, 25% weighed < 90% IBW: 51% of the more obstructed group (FEV₁< 35%) and 20% in the less obstructive patients (FEV₁ > 47% predicted) [[14]].

Schols et al. [[12]] reported data of body weight and body composition in a group of 255 patients with moderate to severe COPD, consecutively admitted to a pulmonary rehabilitation programme. An IBW < 90% predicted was found in 27% of patients with a 35 < FEV₁< 50% pred, 41% of patients with a FEV₁< 35% predicted and 46% of patients with hypoxemia (PaO₂ < 55 Torr). This study indicates that low body weight commonly occurs in COPD patients eligible for pulmonary rehabilitation and that low body weight significantly contributes to the experienced functional impairment in these patients [[12]].

Engelen et al., [[13]], studying body composition in relation to respiratory and peripheral skeletal muscle function in a group of COPD out-patients, reported a low body weight in 17% of the study group.

Prevalence of low body weight was studied in a cohort of more than 4000 patients with COPD treated with long-term oxygen therapy. The prevalence of low body weight as defined by a BMI < 20, was 23% in men and 30% in women [[15]]. The prevalence of low body weight was also reported to be very high in patients with acute respiratory failure and in patients accepted for lung transplantation, with prevalence rates up to 60% and 72% [16&17].

Besides measurement of body weight or BMI, nutritional status can be defined by unintentional weight loss. Unintentional weight loss > 10% during the past 6 months is categorised as severe malnutrition [[18]]. A recent prevalence study conducted in the Netherlands reported in 501 patients admitted to the pulmonary ward for non-oncological illnesses > 10% weight loss in 13% of the patients and 5–10% weight loss in 14% of the patients. A weight loss between 5–10% over a 6-month period is considered as a risk for malnutrition. Remarkably, a significant number of patients had manifested unintentional weight loss with BMI values within the normal range [[19]].

Significance of BMI as Outcome Measure

BMI and Disease Characterisation

Filley et al. [[6]] in the 1960s already described two contrasting types of patients with chronic airway obstruction based on clinical criteria and body weight: the pink puffer and the blue bloater. The pink puffer, the emphysematous patient, was more breathless with marked hyperinflation, thin in appearance with major weight loss. The blue bloater had more severe central cyanosis and was frequently obese; this blue bloater type has often been considered as the chronic bronchitis subtype of COPD [[6]].

Different authors have attempted to correlate the degree of weight loss to the presence or severity of emphysema [20&21]. The relationship between nutritional status and COPD subtypes was further evidenced in the paper of Openbrier et al. [[22]] demonstrating that patients with emphysema were somatically depleted in comparison with patients with chronic bronchitis. They found that in patients with emphysema, a good correlation was present with the degree of airflow limitation as well as with the single-breath diffusing capacity.Others reported that transfer coefficient of diffusing capacity (Kco) was significantly lower in those COPD patients with fat-free mass (FFM) depletion versus patients without FFM depletion [[13]]. These data indicate that weight loss and nutritional depletion is a particular problem in those patients with impaired diffusing capacity. Engelen et al. further elaborated these data by analysing body weight and body composition in COPD patients subdivided into an emphysematous group and a bronchitis group based on high resolution computed tomography; this imaging procedure allows direct assessment of the presence, extent and severity of emphysema. Body weight and BMI as well as FFM and fat mass were significantly lower in emphysematous patients compared to the bronchitis group [23&24].

These data support evidence for a relationship between COPD subtypes and particular patterns of body composition.

Functional Performance and Health Status

Dyspnoea and exercise intolerance are prominent symptoms in COPD patients. In addition to airflow limitation and impaired diffusing capacity, it has become evident during the last decade that respiratory and skeletal muscle weaknesses are important determinants of these symptoms [25&26]. Although muscle dysfunction is highly related to muscle wasting [27&28], many papers have demonstrated that BMI can be considered as an indicator of functional disability. In underweight COPD patients, impaired respiratory muscle strength has been demonstrated [[12]],[29-31]. This impaired respiratory muscle strength contributes to a higher level of dyspnoea in these underweight patients [[32]]. These data are supported by autopsy data, which demonstrated that body weight and muscularity are strongly related to diaphragm muscle mass [[33]]. In addition, changes in BMI associated with muscle tissue depletion significantly impair skeletal muscle strength in COPD patients [[28]].

A close relationship between poor muscle condition, reflected by reduction in percent ideal body weight or BMI and maximal aerobic capacity has been observed in several studies [34-37]. Furthermore, reduced body mass has an independent negative effect on muscle aerobic capacity in COPD patients, as manifested by a decrease in maximal oxygen consumption, reduction in the lactate threshold and slowing of the oxygen consumption kinetics [[37]].

The functional consequences of being underweight, and particularly of FFM depletion, have also been reflected as a decreased health status as measured by the St. George Respiratory Questionnaire [[38]]. In another study, depletion of FFM had a greater impairment in the activity and impact scores of the SGRQ irrespective of body weight [[39]].

Health Care Utilisation

There is growing evidence in the literature that low body weight in COPD is related to a higher utilisation, especially of in-patient services. Kessler et al. [[40]] reported that, besides gas exchange impairment and pulmonary haemodynamic worsening, the risk of being hospitalised with COPD was significantly increased in patients with a low BMI and in patients with a limited 6-minute walking distance. Low body weight also increased the risk of early non-elective readmissions in patients previously admitted for an exacerbation [[41]]. A prospective cohort study of 1016 patients admitted to hospital for an acute exacerbation of COPD described that survival time after the exacerbation was independently related to BMI as well as to severity of illness, age, prior functional status, PaO₂.FiO₂, congestive heart failure, serum albumin, and the presence of cor pulmonale [[42]].

A low BMI was found in approximately 50% of patients with emphysema who were undergoing lung volume reduction surgery (LVRS). This low BMI was associated with increasing morbidity following LVRS, manifested by prolonged ventilator support and increased hospital length of stay [[43]]. Similar data were reported in lung transplant candidates: duration of mechanical ventilation and time spent in the ICU was significantly related to initial body composition [[17]]. In severe COPD patients treated with home long-term oxygen therapy (LTOT), Chailleux recently reported that low BMI was the most powerful predictor of duration and rate of hospitalisation, independently of blood gas levels and respiratory function. The lowest hospitalisation rates were observed in the obese patients in this study [[15]].

BMI and COPD Related Mortality

The relationship between weight loss and being underweight with mortality has been the subject of investigation since the 1960s. Several retrospective studies using different COPD populations provided evidence for a relationship between low BMI and mortality, independent of FEV₁[44&45]. In their Danish population study, Landbo et al. [[46]] reported that the relative risk of death adjusted for smoking, chronic mucus hypersecretion, FEV₁ and gender was significantly increased in underweight subjects (BMI < 20 kg/m²) but decreased in overweight and even obese subjects with mild and moderate disease. In COPD patients randomly allocated to LTOT or medical treatment, Gorecka and colleagues [[47]] found that BMI was a significant predictor of survival, independent of FEV₁. In a cohort of more than 4000 patients treated with LTOT, Chailleux et al. [[15]] reported that low body weight was an independent risk factor for mortality: the 5-year survival rates were 24%, 34%, 44%, and 59%, respectively, for patients with BMIs < 20, 20–24, 25–29, and ≥ 30. The best prognosis was again observed in overweight and obese COPD patients on LTOT.

Towards Definition of MCID in COPD

Despite the growing evidence that body weight can be considered as a global or summative outcome to assess systemic effects of the COPD disease condition, cohort studies with longitudinal follow-up of different possible outcome parameters in an anchor-based approach are lacking. Theproportion of patients who will benefit from treatment can be directly estimated from the effect size (the average change divided by the baseline standard deviation). An effect size of 0.2 is considered small; a medium effect size is 0.5 and a large effect size is 0.8 [[48]].

Different intervention studies reported an increase in body weight after interventional studies [49-55]. These effects in body weight were related to changes in physiological variables: in most studies effects on skeletal as well on respiratory muscle function were evaluated. Based on the reported data, effect sizes for the different intervention studies are calculated (Table 1).

Table 1.  Outcome of Nutritional Intervention Strategies in COPD.

Author	Design1	No of patients	Duration	Weight	Effect size	Other outcome measures
Lewis 1987	R–C	21	8 weeks	↑1.1 kg	0.12	-Pi-max/Pe-max
						-Respiratory muscle endurance
Knowles 1988	R–C	25	8 weeks	↑1.1 kg	0.1	-Pi-max/Pe-max
						-Respiratory muscle endurance
Efthimiou 1988	R–C	14	12 weeks	↑4.2 kg	1.17	↑Pi-max/Pe-max Sternomastoid muscle function
Goldstein 1989	NR–NC	10	2 weeks	↑2.0 kg	0.60	↑Respiratory muscle strength hamstring/endurance strength
Otte 1989	R–C–P	28	13 weeks	↑1.5 kg	0.28	-MVV;-immune response
						-12 minute walking distance
Whittaker 1990	R–C	10	16 days	↑2.4 kg	0.26	-Respiratory muscle strength/endurance
						-Adductor pollicis function
Schols 1995	R–C		8 weeks	↑2.6 kg	0.39	↑Respiratory muscle strenght

¹NR = nonramdomized, NC = noncontrolled, R = randomized, C = controlled, P = placebo.

Substratifying these reports depending on the outcome effects on skeletal or muscle function, it can be hypothezised that an effect size of at least 0.3 is associated with improvement in functional status.

Creutzberg et al. [[56]] reported more recently the outcome of nutritional supplementation therapy implemented in a pulmonary rehabilitation program in depleted patients with COPD. A rise in body weight of 2.1 kg after 8 weeks of intervention was reported (effect size: 0.30): this rise was lower than the predicted value of 2.6 kg. A computer simulation model taking into account the patient's sex, age, height, body composition, and dietary intake was used for the estimation of the weight and FFM response after nutritional therapy: the changes in body weight and FFM were predicted on the base of the actual net rise in dietary intake after nutritional therapy (expected effect size: 0.37) [56&57]. However, the actual gain in fat free mass after eight weeks was similar to the predicted gain (1.1 kg vs 0.9 kg). Expressed in BMI, nutritional intervention in this depleted COPD group resulted in an increase in BMI from 20.24 ± 1.69 to 21.02 ± 1.88 or a net increase of 0.78 ± 0.79 (p < 0.001) (effect size BMI: 0.46).

Compared to a historical placebo group of depleted COPD patients, the increases in body weight and FFM were significantly greater after nutritional supplementation therapy compared to placebo. These effects on body composition were associated with significant effects on skeletal muscle function, exercise performance, and health status.

The item “sense of well-being” on the Medical Psychological Questionnaire for Chronic Lung patients improved significantly after 8 weeks of treatment. Otherwise, on baseline, no significant correlations could be demonstrated between BMI and health status or sense of well-being, neither between changes in BMI and change of these parameters after intervention.

Based on the concept that preferential increase in actively metabolizing tissue and particularly muscle mass importantly contributes to an improved functional performance in COPD patients, specific interventions aim to preferentially increase muscle mass or fat-free mass. Creutzberg et al. [[58]] recently reported the outcome of a double blind, placebo controlled trial using anabolic steroids as additional intervention in patients undergoing a pulmonary rehabilitation program including nutritional intervention for depleted COPD patients. Treatment with nandrolon decanoate relative to placebo resulted in higher increases in fat-free mass owing to a rise in intracellular mass. However, despite the difference of 1.4 kg FFM (effect size for fat free mass changes: 0.78) between both groups, muscle function, exercise capacity and health status were not different between both treatment groups. It has to be realized that interventions with nandrolone also resulted in an net decrease in fat mass.

Another approach to assess MCID for BMI is to evaluate the prognostic importance of weight changes in unselected subjects with COPD. Prescott et al. [[59]] reported these data in COPD patients participating in the Copenhagen City Heart Study. Participants attended two examinations 5 years apart and were followed-up for 15 years for COPD related and all cause mortality. The proportion of subjects who lost > 1 unit BMI between the 2 examinations was significantly associated with severity of COPD by spirometric criteria, reaching approximately 30% in subjects with severe COPD. After adjusting for age, smoking habits, baseline BMI and lung function, weight loss was associated with higher mortality inCOPD patients for weight loss > 3 BMI units. Risk of COPD-related mortality increased with weight loss but not with weight gain. In subjects with severe COPD, there was a significant risk ratio modification between effect of baseline BMI and weight change in the normal-to-underweight (BMI < 25). Best survival was seen in those who gained weight, whereas for the overweight and obese (BMI ≥ 25), best survival was seen in stable weight.

The prognostic significance of body weight changes was further unraveled in the study of Schols et al. [[45]]. She demonstrated that after stratification of the COPD patients into BMI guidelines, a threshold value of 25 kg/m² was identified below which the mortality risk was clearly increased. A weight gain > 2 kg over the course of an 8-week rehabilitation program in depleted and non-depleted patients with COPD as well as an increase in maximal inspiratory mouth pressure were identified as significant predictors of survival. Summarizing these data, definition of a MCID of fat-free mass or muscle mass instead of a MCID of BMI must be advocated in COPD patients. Exact definition of the desirable values of this MCID requires further evaluation.

Muscle Strength in COPD

Muscle performance is largely characterized by strength and endurance. Strength is defined as the capacity of the muscle to develop maximal force and endurance is defined as the capacity of the muscle to maintain a certain force over time, thus to resist fatigue. Loss of either one of these aspects results in muscle weakness and, hence, in impaired muscle performance. Numerous studies have convincingly demonstrated that COPD is commonly associated with muscle weakness [[25]][60-62]. Probably the most extensive study on the influence of muscle weakness on symptomatology and exercise performance in cardiorespiratory disorders was done by Hamilton and co-workers [[26]]. They found significantly reduced strengths of both peripheral and respiratory muscles in patients suffering from chronic respiratory disorders, including COPD.

The potential role of muscle weakness in the outcome of COPD is reported by several authors. In COPD patients, quadriceps strength is an important determinant of exercise performance, quality of life, and direct medical costs [[26]],[[35]][[60]]. In one study, loss of muscle mass had a strong impact on survival, emphasizing the multi-systemic pathogenesis of COPD [[63]]. Future prospective evaluation will have to be undertaken to confirm these retrospective data.

Absolute muscle strength in normal persons largely depends on muscle mass. Therefore, strength is generally corrected for muscle cross-sectional area. It can be questioned whether the loss in strength in COPD can be attributed to muscle atrophy or whether a pathological process impairing the contractile function or the excitation contraction coupling is involved in these patients. If muscle atrophy is the sole factor involved, loss of strength would be proportional to the decrease in muscle cross-sectional area. Furthermore, it can be hypothesized that differences in whole-body fat-free mass or in sub-regional FFM can contribute to the observed muscle weakness. Engelen et al. indeed reported that skeletal muscle weakness was associated with wasting of extremity FFM in COPD patients, independent of airflow obstruction or the presence of CT confirmed emphysema. Interestingly, muscle function per kilogram of FFM was not significantly different between COPD patients and age-matched controls [[62]].

Others reported that in patients with COPD the ratio of quadriceps strength to mid-thigh cross-sectional area measured by computed tomography is similar to that of normal subjects [[63]]. Furthermore, preservation of the contractile properties of the skeletal muscles was demonstrated by in vitro assessment of the contractility of muscle strips from the m. vastus lateralis in COPD. All contractile properties measured, corrected for muscle bundle cross-sectional area, were similar in COPD and in control subjects [[62]]. This fiber cross-sectional area is also strongly related to fat-free mass [[64]]. Maximal isometric tetanic tension corrected for muscle bundle cross-sectional area was also independent of the difference in fiber-type distribution present between COPD patients and healthy control subjects [[65]]. Therefore, it seem that the reduction in muscle strength can be explained by the muscle atrophy rather than by alterations in the contractile apparatus in COPD patients not using systemic steroids. The impact of muscle mass on muscle endurance is yet a matter of research. Franssen et al. reported impaired endurance capacity of the lower extremities but a preservation of endurance of the upper extremities. No relationship could be demonstrated between skeletal muscle endurance and fat-free mass in that study, indicating the need to approach strength and endurance differently in COPD [[66]]. Indeed, future studies are needed to unravel the possible role of morphological and metabolic changes in skeletal muscles in relation to endurance capacity as part of the impaired muscle performance in COPD.

Methodological issues also limit the use of muscle function as an outcome parameter in COPD. Muscular strength, or more precisely, the maximum force or tension generated by a muscle, can be measured using different methods: tensiometry, dynamometry, one-repetition maximum or computer-assisted force and work output determinations. Besides selection of methodology, standardization of the procedure as well as reliability of the measurements have to be evaluated before muscle strength can be considered as a possible outcome parameter in COPD.

Based on the observed relationship between muscle atrophy and muscle strength, the absence of myopathic changes at least in systemic steroid non-users and the observation that contractile properties are independent of the fiber type distribution in COPD, it seem that assessment of muscle mass can be considered as a substitute of muscle function, considering the present methodological issues with respect to muscle function as an outcome measure. Further studies are needed to evaluate the differential effects of strength and endurance on experienced symptomatology by COPD patients. Specific targeting of muscle mass can be considered as afuture treatment goal as a considerable amount of progress has been made in our understanding of the signaling pathways that mediate skeletal muscle hypertrophy and atrophy. These findings provide hope that novel drug targets might be found in the future to block muscle atrophy and to improve the well-being and health status of COPD patients.