Falls in patients with chronic obstructive pulmonary disease: a call for further research

Summary

Chronic obstructive pulmonary disease (COPD) is a respiratory disease that results in airflow limitation and respiratory distress. The effects of COPD, however, are not exclusively limited to respiratory function and people with COPD face many non-respiratory manifestations that affect both function and mobility. Deficits in function and mobility have been associated with an increased risk for falling in older adults. The purpose of this study was to provide a theoretical framework to identify risks factors for falls in people with COPD. We have analyzed the literature to identify possible relationships between pathophysiological changes observed in COPD and common risk factors for falls. Well-established fall risk factors in people with COPD include lower limb muscle weakness and impaired activities of daily living. Other intrinsic risk factors such as gait and balance deficits, nutritional depletion, malnutrition, depression, cognitive impairments and medications are possible risk factors that need to be confirmed with more studies. There is no evidence that visual deficits are common in COPD. The role that precipitating factors such as syncope and postural hypotension may have on fall risk is unclear. Exacerbations and dyspnea do not have a precipitating effect on fall risk but they contribute to the progressive physical deterioration that may theoretically increase the risk for falls. While these results suggest that people with COPD might have an increased susceptibility to fall compared to their healthy peers, further research is needed to determine the prevalence of falls and specific risk factors for falls in people living with COPD.

Chronic obstructive pulmonary disease (COPD) is a respiratory disease characterized by progressive, partially reversible airflow limitation, which results from an emphysematous destruction of the lung parenchyma and increased airway resistance due to inflammation, bronchospasm and increased mucous production. Despite being formerly considered a disease affecting the lungs, it is now well recognized that people living with COPD also suffer from many non-respiratory manifestations including skeletal muscle dysfunction, systemic inflammation, nutritional depletion and malnutrition.^1–4 Cardiovascular problems (e.g., coronary heart disease),⁵ neurological impairments (e.g., polyneuropathy),^6–8 and metabolic disorders (e.g., osteoporosis)⁹ are also commonly associated with COPD as are psychological alterations such as depression and anxiety.^10–13 While the presentation, development and manifestations of COPD are highly variable from patient to patient, the severity of the disease typically progresses as the patient ages.¹⁴ As such, age-related physiological changes and co-morbidities can compound the pathophysiological challenges generally observed in people with COPD.

Falls, which are common among older people, can have devastating effects on overall function, health related quality of life (HRQoL) and life expectancy. Although chronic conditions have been commonly associated with a higher risk for falling,¹⁵ the potential implications of both respiratory and non-respiratory manifestations of COPD on the fall risk of these individuals have not been systematically examined. Furthermore, studies specifically analyzing the incidence of falls among people with COPD are lacking. To date, only one fall-risk study has included people diagnosed with COPD.¹⁶ In this large cross-sectional retrospective study (4050 older women), a linear trend of increasing fall risk with increasing number of chronic diseases was found (OR=1.37; 95% CI: 1.25 to 1.49). More importantly, COPD was found to be one of the leading conditions associated with number of falls, second only to osteoarthritis. Because this study did not specifically include neurological conditions commonly associated with high prevalence of falls such as stroke¹⁷ or Parkinson’s disease,¹⁸ whether people with COPD have an increased risk for falling compared to these patients is currently unknown.

Recent research has highlighted the importance of identifying risk factors for falls in order to enhance prevention.¹⁹ Given the scarcity of studies specifically assessing contributing factors for falls in people with COPD, this review aims to provide a theoretical framework to identify potential risk factors that could increase fall incidence among these individuals. We report the pathophysiological manifestations (i.e., respiratory and non-respiratory) commonly observed in COPD and well-established risk factors for falls to study any potential association. In an attempt to evaluate the relevance of independent risk factors for falls in COPD, the evidence behind each risk factor will be categorized into three different levels: established, possible and theoretical. The number of studies and their methodological quality will determine the level of evidence. The clinical relevance of this review lies in identifying specific risk factors for falls in people with COPD, which may facilitate the design of appropriate rehabilitation strategies to prevent falls.

Risk Factors for Falls

To analyze the different aspects that can increase fall occurrence in people with COPD, we propose the use of a framework (Figure 1) that includes intrinsic and precipitating factors for falls.²⁰ Intrinsic factors include those related to the physical and psychological status of the individual: muscle weakness, gait, balance, visual deficits, use of an assistive device, impaired activities of daily living (ADLs), depression and cognitive impairments.²¹ Medications are also considered an intrinsic factor due to their inextricable impact on the physical and psychological well-being of the individual.²² In addition, nutritional depletion and malnutrition, which are common in people with COPD, are also included as potential intrinsic factors. Precipitating factors are acute episodes (e.g., syncope and postural hypotension)^23,24 that might increase the risk for falls. Exacerbations and dyspnea, which are common in COPD, are also included as potential precipitating factors. Although falls can also be affected by extrinsic factors such as environmental hazards (e.g., stairs and poor lighting), this framework only analyzes intrinsic and precipitating factors that may increase fall risk in people with COPD. Due to the ensuing multitude of possible risk factors for falls in COPD, the proposed framework outlines potential interactions among these factors (Figure 1). This framework may prove useful to identify potential risk factors for falls in both clinical practice and research.

Figure 1. Theoretical framework of the risk factors for falls in people with COPD and potential interactions.

Intrinsic and precipitating factors are depicted by white or shaded rectangles respectively. Since cardiovascular co-morbidities are common in people with COPD,¹⁵¹ the use of cardiac and hypertensive medications is conceivably high among these individuals. The use of these medications appears to increase the frequency of episodes of postural orthostatic hypotension and syncope,¹⁵⁰ which are precipitating risk factors for falls. Other medications commonly used by people with COPD include corticosteroids and possibly antidepressants.^122,123 The use of corticosteroids is associated with an increased risk of developing visual problems (e.g., cataracts).⁹⁰ In addition, the use of antidepressants is thought to increase reaction time and attention deficits.²⁰ The number of medications¹³⁴ as well as visual deficits^81,85 and cognitive impairments¹²⁴ increase balance deficits and risk for falling. Nutritional depletion, characterized by a reduction of muscle mass (i.e., atrophy),⁹⁷ contributes to muscle weakness^99,100 and, by extension, to balance and mobility deficits (e.g., lower gait speed).¹⁰¹ Malnutrition, common in people with COPD,^103,104 may increase balance deficits and impair activities of daily living (ADLs), which lead to progressive deconditioning and a reduction of functional capacity.¹¹⁴ Deconditioning is further aggravated by dyspnea (i.e., breathlessness), the most common and activity-limiting symptom in people with COPD.¹⁵⁸ Exacerbations of COPD usually require periods of bed rest (e.g., hospitalization) and aggressive pharmaceutical interventions (e.g., systemic corticosteroids)¹⁵³ that increase muscle weakness^156,157 and contribute to a progressive reduction of ADLs.

Intrinsic Factors

Muscle Weakness

Clinical trials^25,26 and systematic reviews²⁷ have shown strong correlations between fall incidence and lower limb muscle weakness. Muscle strength is thought to be an essential factor in maintaining postural control and minimizing postural sway,^28,29 both of which are important in preventing older people from falling.²³ Muscle power, which appears to be more closely related to postural control (e.g., single-leg stand and postural sway)³⁰ and functional performance (e.g., stair climbing and comfortable gait speed)^31–34 than muscle strength, might be even more important to prevent falls.^35,36

Muscle strength and endurance are reduced in people with COPD compared to healthy controls. Lower limb muscles, which are actively involved in fall avoidance strategies, are preferentially impaired in these individuals.³⁷ For example, Mathur et al., found that a group of people with moderate to severe COPD (FEV₁=51 ± 17% predicted) had reduced volume (~25%), strength (~25%), and increased intramuscular fat (~35%) of the thigh muscles, when compared to a healthy control group matched for age, gender and body mass.^38,39 All these factors (i.e., reduced muscle volume, strength and increased intramuscular fat) have been associated with an increased loss of function and mobility in older adults.⁴⁰

Muscle endurance is reduced in people with COPD compared to healthy controls. For example, one study assessed muscle endurance through a protocol in which participants had to perform 100 sub-maximal knee extensions.⁴¹ While the control group was able to complete the protocol (i.e., 100 sub-maximal knee extensions), a large percentage (~43%) of people with COPD were not able to do so. Reduced muscle endurance in COPD is important because it reflects increased muscle fatigability,⁴² which has been associated with impaired postural control assessed by single-leg stand in young adults.^43,44 At the functional level, muscle activation measured with magnetic stimulation has been found to be significantly lower in people with severe COPD (FEV₁=29 ± 9% predicted) compared to controls (89 ± 20% versus 109 ± 6%; p<0.01).⁴⁵ Other common structural muscle abnormalities in COPD also include muscle fibre atrophy (most notably in fast twitch type II fibres), reduced capillary density, mitochondrial dysfunction and a lower proportion of oxidative enzymes.⁴⁶ Taken together these abnormalities indicate an overall decline in both the contractile and oxidative capacity of lower limb muscles in people with COPD.⁴⁷

Muscle power, defined as the product of force and velocity of muscle contraction, has not been investigated in people with COPD. However, given that strength and power are so highly correlated,⁴⁸ it is unlikely that people with COPD, with compromised muscle strength, could have normal values of muscle power. The preferential atrophy of fast twitch type II fibres in the lower limb muscles of these individuals⁴⁹ is indicative of decreased ability to perform activities involving fast and powerful muscle contractions.⁵⁰ Given the importance of muscle function in the prevention of falls, it is possible that the reductions of muscle strength, and presumably power, in people with COPD could predispose this population to fall (Table 1).

Table 1. Qualitative analysis of different evidence levels of intrinsic risk factors for falls in people with COPD.

Evidence is categorized into three different levels as established (E), possible (P) and theoretical (T). A rationale to explain the different levels of evidence is provided.

Risk Factor	Evidence as risk factor for falls in older people	Rationale	Evidence in people with COPD	Rationale
Muscle weakness
Strength	E	Deficits in muscle strength are associated with increased fall risk.	E	Deficits in muscle strength and endurance are common in people with COPD.
Endurance	P		E
Power	P	There are no studies reporting on muscle endurance and power and their effect on fall risk. However, strength, endurance and power are highly correlated	P	Muscle power has not been investigated in people with COPD but muscle power is highly correlated with strength
Gait deficits
Speed	E	Lower gait speed has been associated with increased fall risk. There are no studies reporting on walking distance and fall risk. However, reduced walking distance and gait speed are correlated	P	Only one study has are reported reduced gait speed in people with COPD. Lower walking distance is common in people with COPD
Distance	P		E
Balance deficits	E	Deficits in postural control are associated with increased fall risk	P	Only three studies have reported balance deficits in people with COPD
Visual deficits	E	Visuals deficits are associated with increased fall risk	T	There are no studies reporting on visual deficits in people with COPD but the use of corticosteroids in people with COPD is common and their visual side-effects are well established
Use of assistive device	E	The use of an assistive device is associated with increased fall risk. However, the underlying mechanisms behind this association are unknown	-	The prevalence of assistive device use among people with COPD compared to healthy peers is unknown
Impaired ADLs	E	Impaired ADLs are associated with increased fall risk.	E	Impaired ADLs are common in people with COPD and contribute to physical deterioration and reduced mobility
Depression	E	Depression is associated with increased fall risk. However, the underlying mechanisms behind this association are unknown	P	Depression is common in people with COPD. However, the exact prevalence is equivocal
Cognitive impairments	E	Cognitive impairments are associated with increased fall risk.	P	Cognitive impairments are common in people with COPD. However, the estimate of prevalence vary
Medications (type)
Psychotropic	E	Psychotropic and cardiac drugs are associated with increased fall risk. The side-effects of corticosteroid use on muscle weakness and vision are well established	T	The use of psychotropic and cardiac medications might be high in people with COPD. However, the exact prevalence is unknown. The use of corticosteroids is common in people with COPD.
Cardiac	E		T
Corticosteroids	P		P
Bronchodilators	-		-
Theophylline	-		-
Medications (number)	E	The number of medications is associated with increased fall risk	E	The use of multiple medications is common in people with COPD

Gait Deficits

Since falls are reported to occur mainly during walking activities,^51,52 it is not surprising that gait deficits had been associated with disability and falls in older adults.⁵³ Gait deficits have been commonly assessed both quantitatively (e.g., speed and distance) and qualitatively (e.g., gait variability and instability). Quantitatively, gait speed has usually been assessed by measuring the time required to walk a number of meters (5–8 m) either at comfortable (self-selected) or fast gait speeds.⁵⁴ Lowered comfortable gait speed (<1m·s⁻¹) measured over an 8 m walkway, has been found to be strongly associated with an increased risk for falls (RR=5.4; 95% CI: 2.0 to 14.3) in seniors aged 75 years old and older.⁵⁵ Consistent with reduced gait speed, the time required to walk 400m has also shown to be significantly larger in a group of fallers compared to non-fallers (338 ± 65.8s vs. 330.5 ± 60s; p<0.05).⁵⁶

Only a few studies have reported gait speed impairments in people with COPD. For example, Butcher et al.,⁵⁷ reported that fast gait speeds measured over a 6 m walkway were significantly lower (~28%) in oxygen-dependent people with COPD (FEV₁=29.9 ± 3.7% predicted) compared to a healthy control group. Although the values obtained by this group of people with COPD (1.53 ± 0.11m·s¹) are well above the threshold for fall risk established in older adults (<1m·s⁻¹),⁵⁵ it should be taken into account that in this study,⁵⁷ gait speed tests were performed at fast, not comfortable gait speeds. Average values of fast gait speed for healthy elderly (<60 years) range from 0.84 m·s⁻¹ to 2.1 m·s¹.⁵⁴ Since normative data on comfortable gait speed in people with COPD are not yet available, more studies are required to establish average measures of gait speed and specific values prognostic of an increased fall risk in COPD.

Another quantitative approach commonly used to evaluate gait deficits consists of measuring walking distance. Although walking distance is intrinsically related to gait speed and stride length,⁵⁸ the association between reduced walking distance and increased fall risk in older adults has not been directly investigated. Depending on the level of impairment, walking distance tests such as the six-minute walk test (6MWT) or the shuttle walk test (SWT) provide good estimates of endurance/aerobic capacity⁵⁹ and overall functional performance.⁶⁰ For example, walking distance in the 6MWT has been correlated to peak oxygen consumption (r=0.63 to 0.79) in people with COPD⁶¹ and measures of muscle function commonly associated with increased fall risk such as strength (R²=0.38) and peak power (R²=0.48) of lower limbs in older adults.⁶² Given these associations, it is likely that shorter walking distance might be correlated to increased fall risk.

Unlike gait speed, shorter walking distance has been extensively reported in people with COPD.⁶³ For example, people with end-stage emphysema have shown a lower 6MWT walking distance (235.1 ± 92 m)⁶⁴ compared to reference data (~535 m) from healthy older adults (~61 years).⁶⁵ While these studies indicate that gait impairments are not uncommon in people with COPD, and therefore, that increased fall risk associated with gait deficits might be possible (Table 1), qualitative studies⁵⁸ assessing gait kinematics (e.g., variability and instability) associated with fall risk in these individuals are lacking.

Balance Deficits

Postural control is mainly determined by the interaction of three major sensory systems: vestibular, somatosensory and visual.^66,67 Postural control is critical in preventing older people from falling.²⁰ A recent meta-analysis of 16 controlled studies ranked balance disorders as the second most important individual risk factor for falling following lower limb muscle weakness.⁶⁸ Not surprisingly, balance training has shown to be effective in lowering fall risk in this population.⁶⁹

Unlike walking distance and muscle strength, postural control has not been extensively investigated in COPD. A prospective cohort study investigating postural control in people with COPD (FEV₁=62 ± 23% predicted) found a significantly reduced performance (9%; p<0.0017) in the Functional Reach Test when the COPD group (n=1202) was compared with the control group (n=302).⁶³ The Functional Reach Test is a clinical measure of balance that measures how far a subject can reach forward beyond arm’s length using a fixed base of support (i.e., feet on the floor) without losing balance.⁷⁰ Another study showed significantly lower scores in the Community Balance and Mobility Scale in two groups of people with moderate (FEV₁=45.7 ± 3.7% predicted) and severe (FEV₁=29.9 ± 3.7% predicted) COPD compared to a healthy control group.⁵⁷ However, differences among groups in postural sway measured through dynamic posturography were not significant. Chang and colleagues investigated standing postural sway before and after a sub-maximal exercise task (6MWT) in a group of 19 people (69 ± 9 years) with COPD (FEV₁=46% predicted).⁷¹ Postural sway was significantly increased after the 6MWT, which suggests that, as shown in other studies with healthy individuals,^43,44 fatigue may worsen postural control in people with COPD. More importantly, when compared to a fall-risk study that used similar balance tests in 156 older adults (76.5 ± 5.1 years),⁷² Chang et al., found that the values of lateral postural sway measured at baseline in the COPD group were similar to those with a history of falls. In spite of the small number of studies assessing postural control in COPD, these preliminary results suggest that increased fall risk secondary to balance deficits in these individuals may be present (Table 1).

The underlying mechanisms for the increased postural sway observed in people with COPD are currently unknown. Although vestibular deficits are unlikely,^73,74 somatosensory deficits could contribute to impaired balance in this group. For example, studies evaluating polyneuropathy in people with COPD have shown that the presence of nerve conduction abnormalities is not uncommon, especially in the most disabled individuals. Appenzeller et al., reported reduced peroneal nerve conduction velocity in people with severe COPD.⁷⁵ Another study revealed electrophysiological disturbances in the motor and sensory nerves of 23 people with severe COPD (FEV₁ < 50% predicted).⁶ In a more recent study, 5 out of 30 people with moderate COPD (FEV₁ < 80% predicted) showed signs of axonal neuropathy (smaller amplitude potentials, increased latency, decreased conduction velocity) compared to an age-matched group⁸ but these anomalies were only significant in the afferent nerves of the most impaired patients. Interestingly, the majority of studies assessing nerve conduction in COPD show that polyneuropathy affects preferentially the sensory nerves of the lower limbs.

Polyneuropathy has a profound impact on the somatosensory system and people with polyneuropathy show deficits in postural control as assessed by single-leg stance,⁷⁶ gait (i.e., speed, kinematics and efficiency),⁷⁷ lower limb muscle function (i.e., peak velocity and strength),⁷⁸ in addition to a high prevalence of falls.^79,80 Unfortunately, to date no studies have specifically examined the prevalence of polyneuropathy in a large cohort of people with COPD. Moreover, the impact that polyneuropathy may have on proprioception, reaction time, muscle activation and postural control in people with COPD is unknown.

Visual Deficits

Visual deficits have been associated with an increased risk for falling in older adults.^81,82 For example, in a large prospective cohort study with 2002 older women, Coleman et al., reported that the odds of experiencing a fall were approximately two times greater for those with loss of vision.⁸³ In other studies, the relative fall risk in people with visual deficits compared to people without visual problems has been estimated to be 1.45; 95% CI: 1.08 to 1.94.⁸⁴ Associations between impaired vision and falls might be mediated by the role that visual inputs, particularly stereopsis and contrast sensitivity, play in the stabilization of posture.^81,85 This is especially relevant in older people who are more dependent on vision to maintain a standing posture.⁸⁶

There are no studies reporting on the prevalence of visual impairments in people with COPD. Several studies, however, have found strong dose-dependent associations between inhaled corticosteroid use and risk of cataracts^87–89 as well as associations between oral corticosteroids and risk of glaucoma.⁹⁰ Cataracts and glaucoma reduce visual acuity, impair postural control, and have been associated with an increased rate of falls.^91,92 Since inhaled corticosteroids are frequently prescribed in people with COPD,⁹³ the deleterious effects these medications can have on visual function and falls should not be underestimated. Smoking has also been correlated with an increased risk of cataracts, particularly nuclear cataracts.⁹⁴ Although the associations of corticosteroids and smoking with cataracts and glaucoma are well established, there are no studies analyzing the prevalence of either corticosteroid or smoking induced visual deficits in a large cohort of patients with COPD. Similarly, there are no studies exploring associations between visual deficits and risk for falls in people with COPD. Likewise, the potential role that visual deficits have as a risk factor for falls in people with COPD is only theoretical (Table 1).

Nutritional Depletion and Malnutrition

Nutritional depletion, which is common among people with COPD,^95,96 is characterized by body weight loss, in particular, muscle mass.⁹⁷ The loss of muscle mass in people with COPD has been associated with increased mortality,⁹⁸ impaired muscle function,^99,100 reduced exercise capacity¹⁰¹ and poorer HRQoL.¹⁰² Even though there are no studies reporting on the association between nutritional depletion and risk for falls, leg muscle mass assessed by dual-energy X-ray absorptiometry (DEXA) has been found to be significantly lower (p<0.01) in a group of older adults fallers compared to non-fallers.⁵⁶ The deleterious consequences of muscle depletion on function and mobility are well established.⁴⁰

Malnutrition is not uncommon in people with COPD. For example, recent studies have reported that people with COPD show low levels of circulating vitamin D.^103,104 Vitamin D deficiency is important because it has been associated with an increased risk for falling.¹⁰⁵ The physiological mechanisms underlying this correlation are possibly related to the role that vitamin D plays in muscle function.¹⁰⁶ In support of this, vitamin D supplementation has shown to be effective in improving neuromuscular function and postural control in older adults.^107,108 and higher vitamin D serum levels (25-hydroxyvitamine D) have been correlated with increased lung function in a large (n=14091) cohort study.¹⁰⁹ Whether vitamin D supplementation improves respiratory function or reduces the incidence of falls in people with COPD is yet to be determined, however, the low levels of circulating vitamin D and the high prevalence of osteoporosis among people with COPD¹¹⁰ provide some indication that vitamin D supplementation might be beneficial. Although the effects that nutritional depletion and malnutrition have on the risk for falls in people with COPD are uncertain, the evidence of decreased muscle mass and nutritional deficits in COPD is strong. Due to the well established association between muscle mass and strength,¹¹¹ as well as the impact of muscle weakness on the risk for falls,²⁷ it is possible that nutritional depletion may contribute as a risk factor for falls in COPD (Table 1).

Use of an Assistive Device

The use of an assistive device has been associated with an increased risk for falling (OR=2.6; 1.2 to 4.6).²⁰ The reasons why people that use an assistive device fall more frequently are unclear and it is possible that the use of these devices could simply reflect impaired mobility, gait and balance, all of which have been associated with increased fall risk.^20,68 The use an of assistive device may also reflect fear of falling, which is thought to impact on the physical performance and mobility of older adults.¹¹² Some people with COPD that are oxygen dependent use portable oxygen tanks to reduce breathlessness. However, these devices have now become so portable and small that they do not seem to interfere with the normal mobility of these individuals. In addition, some people with severe COPD use wheel-walkers and scooters to facilitate mobility, however, it is unclear whether the use of these devices is more frequent in this group than in older adults. There are no studies indicating that the use of any assistive device in people with COPD could increase the risk for falling (Table 1).

Impaired Activities of Daily Living (ADLs)

Impaired ADLs have been associated with an increased risk for falling in older adults (OR=2.3; 95% CI: 1.5 to 3.1).^20,23 However, the order of causation of impaired ADLs and falls is uncertain. Impaired ADLs are normally accompanied by reduced mobility and poor physical performance, both of which have been associated with increased risk for falling in older adults.¹¹³ On the other hand, injurious falls are sometimes followed by a period of immobilization, which has detrimental effects on mobility, physical conditioning and ADLs. Numerous studies employing different quantification methods have confirmed that people with COPD have impaired ADLs compared to healthy individuals. For example, Pitta et al., used triaxial accelerometers to evaluate physical activity levels in 50 people with moderate to severe COPD (FEV₁=43 ± 18% predicted) compared to 25 healthy controls.¹¹⁴ Compared to the control group, people with COPD showed significantly lower levels of physical activity (i.e., lower walking/standing time and higher sitting/lying time) and movement intensity (during walking). Similar findings have been obtained in several other studies using either motion sensors or physical activity questionnaires.^115,116 Therefore, impaired performance of ADLs in people with COPD is well-established (Table 1).

Depression and Cognitive Impairments

A recent study has reported that depression is associated with the number of falls in older adults.²³ Although depression and falls share common risk factors (e.g., impaired ADLs and impaired cognitive function),¹¹⁷ the underlying mechanisms governing this association and the order of causation are unclear. A potential explanation might be that the use of antidepressant medications can have deleterious effects on the physical status of people with depression.¹¹⁸ Others have suggested that depression may result in inattention to potential environmental hazards.²⁰

Chronic obstructive pulmonary disease has recently been associated with a number of psychiatric disorders,¹¹⁹ including depression and anxiety.^120,121 Two reviews have reported on the prevalence of depression and anxiety in COPD, which could affect between 2%–57% and 2%–51% of these people respectively.^122,123 Compared to control subjects,¹² this high prevalence of depression may suggest that people with COPD have an increased risk for falling (Table 1). However, there are no studies assessing associations between depression and fall occurrence in people with COPD. Furthermore, although medications commonly prescribed for anxiety disorders (e.g., benzodiazepines and neuroleptics) have been associated with an increased risk for falls,¹¹⁸ the potential association between anxiety and falls has not yet been determined.

Cognitive impairments have been identified as an important factor of increased fall risk in older adults. For example, a 2 year prospective study showed that people with dementia fell four times per year,¹²⁴ which is a much higher rate than commonly reported rates from older community-dwelling adults.¹²⁵ Another prospective study with 336 old adults (>75 years) showed that people with cognitive impairments fell 5 times more often (OR=5; 95% CI: 1.8 to 13.7) than people without cognitive impairments.¹²⁵ The major involvement of associative cortical areas in the integration of sensory and executive tasks¹²⁶ may explain the link between reduced cognitive capacity and increased risk for falls. For example, a recent cross-sectional study found significantly increased postural sway and lower performance of executive functions (i.e., set shifting, updating and response inhibition) in people with mild cognitive impairment compared with people without cognitive deficits.¹²⁷

Several studies have reported that cognitive impairments are not uncommon among people with COPD, especially in those with hypoxemia and hypercapnia.¹²⁸ For example a clinical study showed that 42% of hypoxemic people with COPD (PaO₂ 51mmHg) had severe cognitive impairments, assessed through neuropsychological tests, compared with only 14% of controls.¹²⁹ These seminal findings have been recently confirmed with more advanced brain imaging techniques. For example, compared to healthy and non-hypoxemic people with COPD, hypoxemic people with COPD have shown reduced brain perfusion in both anterior and sub-cortical areas of the dominant hemisphere evaluated through single photon emission computerized tomography (SPECT).¹³⁰ More importantly, such changes in brain perfusion have been correlated to poorer performance in neuropsychological tests in people with COPD.¹³¹ Even though these studies suggest that cognitive impairment is not rare in people with COPD, the exact prevalence of cognitive impairments and their effects on fall risk in COPD are currently unknown (Table 1).

Medications

Numerous studies have shown that certain types of medications are associated with increased frequency of falls in older people. In a systematic review, Leipzig and colleagues found that cardiac and psychotropic drugs increased the risk for falls with OR ranging from 0.9 to 1.6 and 1.5 to 1.7 respectively.^132,133 Similar findings have been replicated in other studies investigating the action of central nervous system medications (e.g., narcotics and antidepressants) on the risk for falls and fractures in older women.¹¹⁸ For example, Tinetti and colleagues found that sedative medications (i.e., benzodiazepines, phenothiazines and antidepressants) significantly increased the risk for falls in older people (OR=28.3; 95% CI: 3.4 to 39.4).¹²⁵ The number of medications has also been independently correlated to impaired postural control assessed by a composite of balance measures (i.e., side by side balance, sternal nudge, tandem stand, one leg standing balance).¹³⁴

Medications commonly used in the management of COPD include: bronchodilators (anticholinergics, β₂-agonists, methylxanthines), oral and inhaled corticosteroids, and theophylline. Other less frequent pharmacologic treatments include: vaccines, mucolytics, antitussives, α-1 antitrypsin augmentation therapy, antioxidants, immunoregulators, vasodilators, narcotics, and antibiotics in cases of acute infection.¹³⁵ While some of these medications may have adverse side effects including dizziness, postural hypotension, altered vision and long-term cognitive changes that might theoretically increase the risk for falling²⁴ the frequency and severity of these side effects and any impact they may have on fall risk in people with COPD is still to be elucidated.

Two types of medications commonly used in COPD warrant special attention for their potential adverse effects on fall risk: corticosteroids and psychotropics. Corticosteroid use in people with COPD has been estimated to be ~61.5% and ~8.3% for inhaled and oral corticosteroids respectively.⁹³ In spite of their known benefits,¹³⁶ these medications have several detrimental side effects that can potentially increase the risk for falling. Side effects of corticosteroid use include stereoid induced myopathy, hyperglycemia, hypertension, loss of bone mineral density, glaucoma and cataracts.¹³⁷ Some of these side effects have been specifically observed in people with lung disease. For example, Nava et al. reported muscle weakness after 5 days of treatment with high dosage (12mg·kg⁻¹) of methylprednisolone in patients with acute lung rejection.¹³⁸ Although not all studies have observed such decline in muscle function,¹³⁹ the detrimental effects of corticosteroids, in particular oral corticosteroids, on both respiratory and peripheral muscle function in people with COPD are well established.¹⁴⁰

Long term use of corticosteroids in people with COPD has also been associated with metabolic disorders such as osteoporosis and osteopenia¹¹⁰ and a subsequent increased risk of bone fracture in some,¹⁴¹ but not all,^142,143 studies. The severity of side effects such as muscle weakness, risk of fracture and visual deficits appears to be related to the dose, duration, and type of administration (i.e., oral corticosteroids produce more side effects).^{89,135,144,145} Differences in these parameters might well explain conflicting results among studies. Given the deleterious effects that visual impairments,^81,82 and muscle weakness,^25,26 have in the risk for falling, the use of corticosteroids in patients with COPD needs to be weighed against the potential risks.¹⁴⁶

Because people with COPD experience many psychological effects such as depression, anxiety and sleep disorders,¹¹⁹ the use of psychotropic medications including antidepressants and benzodiazepines is conceivably high among these patients. Anxyolitics, narcotics and antidepressants have been associated with an increased risk for falling in several studies.^16,118 The underlying mechanisms explaining the association between the use of these medications and increased fall risk might be related to commonly reported side effects such as dizziness, fatigue, daytime sleepiness, slowed postural reflexes, postural hypotension and impaired postural control.¹⁴⁷ Since no specific studies have quantified the exact prevalence of use of these substances in people with COPD, the importance that these medications may exert on their risk for falling in these subjects remains unknown (Table 1).

Precipitating Factors

Syncope and Postural Orthostatic Hypotension (POH)

Syncope has been defined as a transient, self-limited loss of consciousness, usually leading to a fall.¹⁴⁸ The multitude of possible aetiologies and potential recall bias (e.g., retrograde amnesia is common after syncope) make the prevalence of syncope difficult to determine. However, a one year prospective study including 711 older people with a mean age of 87 years, reported an incidence of 6% per year, with a two year prevalence and recurrence rates of 10% and 30% respectively.¹⁴⁹ Recently, syncope has been identified as a potential risk factor for falls.²³ The percentage of falls related to syncope has been estimated to range between 0.5%–3%, however, the existing overlap between falls and syncope¹⁵⁰ and the fact that syncope is excluded in most studies of falls, suggest that this percentage might be higher.³⁰ There are no studies investigating syncope in COPD, however, both conditions share some common aspects that suggest that the risk of syncope in people with COPD might be significant. The likelihood of syncope is increased with cardiovascular diseases (i.e., coronary heart disease and aortic stenosis) and the use of hypertensive medication.¹⁵¹ Due to the multiple effects of smoking and other shared predisposing factors (e.g., systemic inflammation), the risk of cardiovascular disease (OR=2.4; 95% CI: 1.9 to 3) and hypertension (OR=1.6; 95% CI: 1.3 to 1.9) in people with COPD is high.¹⁵² Nevertheless, given that no studies have examined at the association between syncope and COPD, whether there is an increased risk for falls due to a transient loss of consciousness in people with COPD needs to be investigated further (Table 2).

Table 2. Qualitative analysis of different evidence levels of precipitating risk factors for falls in people with COPD.

Evidence is categorized into three different levels as established (E), possible (P) and theoretical (T). A rationale to explain the different levels of evidence is provided.

Risk Factor	Evidence as risk factor for falls in older people	Rationale	Evidence in people with COPD	Rationale
Syncope	P	Syncope is associated with increased fall risk. However, the prevalence of syncope and its effect on fall risk is unknown	-	There are no studies reporting on the prevalence of syncope in people with COPD
Postural orthostatic hypotension	T	Postural orthostatic hypotension may produce transient dizziness. The effects of dizziness on fall risk are unknown	T	There are no studies reporting on the prevalence of postural orthostatic hypotension in people with COPD. However, cardiovascular co-morbidities and use of hypertensive medications, both of which are associated with increased fall risk, might be high in people with COPD
Exacerbations	-	There are no studies reporting on the role of exacerbations on fall risk	E	Exacerbations are common in people with COPD. Exacerbations reduce muscle strength and physical activity levels
Dyspnea	-	There are no studies reporting on the role of dyspnea on fall risk	E	Dyspnea is the most common symptom in people with COPD. Dyspnea reduces physical activity levels and contributes to physical deconditioning

Postural orthostatic hypotension (POH) is clinically defined as a fall in blood pressure of at least 20 mmHg systolic or 10mmHg diastolic when standing or during head-up tilt testing.¹⁵³ The prevalence of POH in older adults is estimated to lie between the range of 5%–30%.¹⁵³ Age, the use of some medications (e.g., antihypertensives, diuretics, antidepressants) and some clinical conditions (e.g., diabetes, Parkinson’s disease, autonomic disorders) are contributing factors of POH.¹⁵³ Transient dizziness after changing posture, the most commonly reported symptom of POH, is thought to increase the risk for falling, however the association between POH and fall occurrence remains speculative.³⁰ From our clinical experience, the prevalence of POH in people with COPD is possibly high. However, there are no formal studies investigating POH in COPD, and as such, its potential implication in fall occurrence among these patients cannot be determined (Table 2).

Exacerbations

An exacerbation is defined as “a sustained worsening of the patient’s condition, from the stable state and beyond normal day-to-day variations, necessitating a change in regular medication in a patient with underlying COPD”.¹⁵⁴ The exact prevalence of exacerbations in people with COPD is difficult to calculate because it varies substantially depending on the severity of the disease as well as the prior number of exacerbations.¹⁵⁵ However, recent studies have shown that people with severe and moderate COPD (according to the classification of the Global Initiative for Chronic Obstructive Lung Disease) might experience an average of 3.43 and 2.63 exacerbations per year respectively.¹⁵⁶

Since exacerbations usually require hospitalization, prolonged bed rest and aggressive pharmacological interventions (e.g., high doses of oral corticosteroids), muscle atrophy and weakness associated with exacerbations are likely. For example, Spruit et al.,¹⁵⁷ observed that quadriceps peak torque was reduced in people with COPD (FEV₁=66 ± 22 % predicted) hospitalized due to an exacerbation, compared to a group of clinically stable COPD patients (FEV₁=86% ± 16 predicted). More importantly, this accentuated weakness was still apparent 3 months after discharge from the hospital. Peripheral muscle weakness could contribute to the reduced levels of physical activity observed in people with COPD even one month after exacerbations.¹⁵⁸ Due to the negative effects of exacerbations on muscle function and physical activity levels, gait and balance deficits after exacerbations cannot be missed. Thus, an increased risk for falls secondary to exacerbations is possible (Table 2).

Dyspnea

Dyspnea is the most common activity-limiting symptom of COPD. Because the level of dyspnea is strongly correlated to the intensity of the activity performed, people with COPD tend to progressively reduce physical activity and ADLs. This further contributes to overall physical deconditioning, which increases the level of inactivity and, in the long term, social isolation.¹⁵⁹ There are no studies analyzing the potential impact of dyspnea on falls occurrence but it seems unlikely that dyspnea could contribute as a precipitating (acute) factor for falls in people with COPD. Paradoxically, dyspnea leads people with COPD to adopt a much more sedentary lifestyle that reduces mobility and that could eventually “protect” these patients from falling. However, the long-term consequences of dyspnea and its detrimental impact on physical activity and further deconditioning in COPD are well established.¹¹⁷ Although dyspnea as a precipitating factor does not appear to increase the risk for falling, the long-term effects of dyspnea have important repercussions on the overall physical status of people with COPD that can increase fall risk (Table 2).

Conclusions

The results of this review suggest that people with COPD may have an increased risk for falling compared to their healthy peers. Most risk factors for falls identified in older people are common in people with COPD. There is evidence that lower limb muscle weakness and impaired activities of daily living are well-established risk factors for falls in people with COPD. Other intrinsic factors such as gait and balance deficits, nutritional depletion, malnutrition, depression, cognitive impairments and medications are possible risk factors that need to be confirmed with further research. Due to the lack of studies, the potential implication of visual deficits on fall occurrence in people with COPD is only theoretical. There is no evidence to suggest that people with COPD may have an increased risk for falls associated with the use of an assistive device. While the role that precipitating factors such as syncope and postural hypotension may have on fall occurrence in COPD is unclear, the long-term effects of dyspnea and exacerbations on limited mobility and progressive physical deterioration might theoretically increase the risk for falls.

In summary, we have used a physiological approach to present a theoretical framework of potential risk factors for falls in COPD (Figure 1). Even though our results suggest that fall risk in people with COPD may be high, the lack of epidemiological studies assessing the prevalence of falls and the scarcity of clinical trials analyzing specific risk factors for falls in people with COPD indicates that any conclusions from this review should be cautious. Further research is needed to determine the prevalence of falls and specific risk factors for falls in people living with COPD. In the meantime, we expect that the proposed framework will provide a basis to generate relevant research questions regarding the risk for falls among people with COPD.