Work of breathing and respiratory drive in obesity



    1. Division of Chest and Critical Care Medicine, Department of Internal Medicine, National Taiwan University Hospital, Hsin-Chu branch, Hsin-Chu
    Search for more papers by this author

    Corresponding author
    1. Chest Division, Department of Internal Medicine and Department of Medical Research, Mackay Memorial Hospital, Taipei, Taiwan
    Search for more papers by this author

  • The Authors: Dr Ching-Kai Lin is a Respiratory Physician at Hsin-Chu Branch of National Taiwan University Hospital, with research interests in sleep breathing disorder, asthma, bronchoscopy and critical care. Dr Ching-Chi Lin is a Respiratory Physician at Mackay Memorial Hospital, with research interests in sleep breathing disorder, obstructive airway disease and critical care.


Ching-Chi Lin, Chest Division, Department of Internal Medicine, Mackay Memorial Hospital, 92, Sec. 2, Chung Shan North Road, Taipei, Taiwan. Email:


Obesity, particularly severe central obesity, affects respiratory physiology both at rest and during exercise. Reductions in expiratory reserve volume, functional residual capacity, respiratory system compliance and impaired respiratory system mechanics produce a restrictive ventilatory defect. Low functional residual capacity and reductions in expiratory reserve volume increase the risk of expiratory flow limitation and airway closure during quiet breathing. Consequently, obesity may cause expiratory flow limitation and the development of intrinsic positive end expiratory pressure, especially in the supine position. This increases the work of breathing by imposing a threshold load on the respiratory muscles leading to dyspnoea. Marked reductions in expiratory reserve volume may lead to ventilation distribution abnormalities, with closure of airways in the dependent zones of the lungs, inducing ventilation perfusion mismatch and gas exchange abnormalities. Obesity may also impair upper airway mechanical function and neuromuscular strength, and increase oxygen consumption, which in turn, increase the work of breathing and impair ventilatory drive. The combination of ventilatory impairment, excess CO2 production and reduced ventilatory drive predisposes obese individuals to obesity hypoventilation syndrome.


Obesity has more than doubled worldwide since 1980. It is a preventable global problem that is associated with a number of comorbidities, including hypertension, type II diabetes mellitus, hyperlipidaemia and some cancers.1 It also affects many respiratory physiological parameters, including compliance, resistance, lung volumes, spirometric measures, bronchial hyperreactivity, upper airway mechanical function, neuromuscular strength, diffusing capacity and gas exchange. These, in turn, may affect the work of breathing (WOB), ventilatory drive and exercise capacity, and lead to sleep-breathing abnormalities.2

Obesity is associated with many respiratory diseases like chronic obstructive pulmonary disease, obstructive sleep apnoea syndrome (OSAS), asthma, pulmonary embolic disease and pneumonia.1,2 Various inflammatory cytokines have been linked to obesity and may contribute to synergistic systemic inflammatory effects in patients with obstructive airway disease and sleep apnoea syndrome.1,2 The purpose of this review is to elucidate the effects of obesity on the WOB and on ventilatory drive.


The obese patient has significantly increased oxygen consumption (VO2) and CO2 production (VCO2), and an impaired ventilatory system. Naimark and Cherniack reported WOB to be 540 kg.m/L in obese individuals and 227 kg.m/L in lean individuals,3 while Pelosi et al. reported values of 1.30 J/L and 0.52 J/L, respectively.4 Kress et al. found VO2 to be 355 mL/min in obese individuals and 221 mL/min in lean individuals who were sedated. After being anesthetized and paralyzed, VO2 declined by 16% in obese individuals compared with <1% in lean individuals. These findings indicate that the portion of VO2 required for spontaneous respiration is negligible in most lean individuals but is significant in obese patients, corroborating observations of increased WOB in obesity.5

Many factors may contribute to increased WOB in obese individuals, including impaired respiratory mechanics, upper airway mechanics, neuromuscular strength, gas exchange, neurohormonal influences and ventilatory drive.2

Work of breathing and respiratory function

The most common and consistent pulmonary function tests abnormality seen in obese individuals is a reduction in functional residual capacity (FRC) and expiratory reserve volume6 which is generally seen when the weight (kg)-to-height (cm) ratio is greater than 0.7.6 Because breathing occurs at low FRC and in the less compliant portion of the pressure–volume curve, increased effort is needed to overcome respiratory system elasticity. Thus, obese individuals need to do more respiratory work to maintain appropriate levels of ventilation.

The flow volume loops from a healthy obese individual may show significant reductions in lung volumes while expiratory flows remain well preserved. Nonetheless, expiratory flow at 50% of vital capacity is low compared with the predicted value based on the predicted vital capacity.7 Rubinstein et al. reported that significant differences in expiratory flow at 25% of the reduced vital capacity persists after normalization of the predicted value, suggesting possible peripheral airway obstruction in obese men.7

Evidence also shows an increase in residual volume and residual volume-to-total lung capacity ratio and a higher total lung capacity by plethysmography than by helium dilution in obese individuals, indicating air trapping.7 Similarly, Douglas and Chong found that low expiratory reserve volume and FRC was associated with decreased frequency dependence of compliance and increased closing volume, suggesting small airway closure and air trapping.8 Thus, morbidly obese subjects may develop tidal expiratory flow limitation (EFL) that reaches the maximal expiratory flow rate during quiet breathing. Tidal EFL promotes dynamic pulmonary hyperinflation resulting in the development of intrinsic positive end expiratory pressure (PEEPi). PEEPi is the end expiratory elastic recoil pressure of the respiratory system due to incomplete expiration, and imposes a threshold load on inspiratory muscles before inspiratory flow is achieved during the next breathing cycle. PEEPi also imposes an additional mechanical load on the inspiratory muscles, thereby increasing WOB.8

Many aspects of respiratory function worsen in the supine position and during sleep in obese subjects, especially during the rapid eye movement stage of sleep.9 Some obese individuals develop PEEPi on the supine position, which may explain why residual volume is relatively well preserved in this population.8,10 Pankow et al. reported EFL in two of eight obese subjects in the upright position and in seven of eight obese subjects in the supine position. In obese subjects, PEEPi increased from the upright to the supine position and decreased in the right lateral position (compared with supine). In contrast, in the control subjects, PEEPi was smaller in each position. Compared with controls, the mean transdiaphragmatic pressure in each position was significantly larger in those with obesity. The mean transdiaphragmatic pressure increased from the upright to the supine position and decreased in the lateral position, while in control subjects, no significant changes occurred.11

Yap et al. examined total respiratory resistance and total respiratory reactance (Xrs) by forced oscillation. When seated, respiratory resistance at 6 Hz was higher in obese subjects than in control subjects, while Xrs at 6 Hz was lower. In obese subjects, supine respiratory resistance at 6 Hz markedly increased and Xrs fell when changing to the supine position, compared with control subjects.12 Steier et al. reported increases in neural respiratory drive and transdiaphragmatic pressure swings in obese individuals when changing from the sitting to the supine position, which were not seen in non-obese subjects. The application of continuous positive airway pressure (CPAP) reduced diaphragm electromyogram and inspiratory pressure swings by 40% and 25%, respectively, and markedly abolished PEEPi in the obese subjects when supine.13 Dyspnoea and EFL frequently occur in morbidly obese subjects in the supine position, and both supine EFL and low expiratory reserve volume values are related to orthopnoea, suggesting that dynamic pulmonary hyperinflation and PEEPi increase WOB, contributing to orthopnoea in morbidly obese subjects.14

In summary, for obese subjects, tidal breathing can be affected by EFL and PEEPi, while EFL and PEEPi are increased in the supine position. The increase in diaphragmatic load in the supine position is related to PEEPi. Obese patients have increased neural drive. Lastly, CPAP abolishes PEEPi and reduces neural respiratory drive in these patients.11–14

Respiratory muscle function

Effects of obesity on static inspiratory and expiratory muscle strength are variable and inconsistent.15 Respiratory muscle function may be more severely impaired with increasing obesity, and this impairment may be due to a myopathy or to the burden imposed on the diaphragm by obesity itself. Because FRC is low, breathing takes place at a less efficient portion of the pressure–volume curve, resulting in increased ventilation and higher flows needed to perform the maximum voluntary ventilation manoeuvre despite EFL.16

The maximum voluntary ventilation reflects ventilatory mechanics as well as respiratory muscle function. It declines with increasing body mass index (BMI) at a rate greater than the decline in forced expiratory volume in 1 s and forced vital capacity (FVC), suggesting impaired respiratory muscle function. Following bariatric surgery and a reduction in BMI from 41.5 kg/m2 to 31.7 kg/m2, one study found improvements in inspiratory (PImax) and expiratory (PEmax) respiratory muscle strength by 21% and 22%, respectively, while respiratory muscle endurance increased by 13%. These improvements were greater than those seen in FVC (9%), forced expiratory volume in 1 s (3%) or lung volumes (7–10%), implying a direct effect of weight reduction on respiratory muscle function.17

As stated previously, morbidly obese subjects may develop EFL such that PEEPi and dynamic pulmonary hyperinflation occur especially in the supine position. Dynamic pulmonary hyperinflation will also overstretch the diaphragm (secondary to the cephalad displacement by weight of the abdomen), causing it to operate on the descending limb of its length–tension curve, further impairing diaphragmatic function.18

Chlif et al. showed that, at rest, PImax was significantly lower in obese subjects than in controls.18 Effective inspiratory impedance (P0.1/tidal volume/inspiratory time) is a combination of P0.1 (the pressure generated in the first 100 ms of an occluded breath) and the mean inspiratory flow of non-occluded breaths. The increase in effective inspiratory impedance in obese suggests that the mechanical load on the respiratory muscles is increased. The P0.1/PImax ratio is significantly higher in obese patients than in lean controls, indicating increased neural drive to inspiratory muscles in obese patients. The energy cost of breathing, the ratio of the inspiratory power of breathing at rest to the critical inspiratory power (Wrest/Wcrit), which reflects the oxygen cost of the respiratory muscles, is also greater in obese subjects than in controls.18

The higher breathing frequency to tidal volume ratio observed in obese subjects may be a manifestation of incipient respiratory muscle weakness. An increase in frequency to tidal volume ratio is associated with an increase in the ratio of mean inspiratory pressure to maximal inspiratory pressure and the ratio of inspiratory time to the total time of the respiratory cycle. These changes reflect the impairment in respiratory muscle function, respiratory muscle endurance, respiratory mechanics and increased WOB, and produce a defective breathing pattern in obese individuals, leading to dyspnoea and impending ventilatory failure especially in the supine position.18

Obesity and upper airway mechanical factors

Obesity may elevate the passive critical pressure by fatty deposits and through increased soft tissue load on the pharynx leading to increases in pharyngeal collapsibility. These fatty deposits are particularly pronounced in men with central obesity compared with women with peripheral obesity.19 Obesity, particularly central obesity, imposes mechanical loads on the upper airway, and is a major risk factor for increased upper airway resistance and collapsibility.

Obesity may also affect upper airway neural control. Schwartz et al. showed that weight loss leads to a decrease in both passive and active passive critical pressure in apnoeic subjects, suggesting a concomitant recovery in active neuromuscular control with weight loss.20

Obesity, particularly central obesity, in addition to snoring and sleep apnoea, has been associated with local inflammation of the upper airway structures. In vivo studies have shown that repeated upper airway collapse may increase the expression of several inflammatory genes, including macrophage inflammatory protein-2, TNF-α, interleukin-1β and P-selectin, which have somnogenic central nervous system activity.21,22 In contrast, levels of anti-inflammatory adipokines decrease. These cytokines may trigger an inflammatory reaction in the upper airway, leading to inflammatory cell infiltration and remodelling.23 Ultra-structural changes in the pharyngeal mucosa have been reported with neurosensory deficits to pinprick two-point discrimination.24,25 Sensory deficits in protective reflex responses to negative epiglottic intraluminal pressures may impair neuromuscular responses to upper airway collapse during sleep.26

As obesity progresses and sleep apnoea develops, nocturnal disturbances in sleep and gas exchange can trigger further elevations in oxidative stress, inflammatory cytokines and humoral factors, further aggravating pharyngeal neuromuscular dysfunction.21,22,26 Recent studies have shown that obesity and leptin deficiency are associated with marked elevations in passive passive critical pressure, potentially implicating these factors in the pathogenesis of upper airway obstruction during sleep.27,28 Thus, local as well as systemic inflammatory responses and humoral effects may contribute to disturbances in upper airway neuromuscular control, thereby increasing upper airway resistance, WOB and sleep apnoea susceptibility in obese patients.

To evaluate whether there was an abnormal increase in upper airway resistance in the sitting and supine positions in obesity hypoventilation syndrome (OHS) patients compared with subjects with eucapnic obesity and OSAS or normal controls, Lin et al. used impulse oscillometry during wakefulness. In morbidly obese eucapnic OSAS subjects and controls, upper airway resistance was normal in sitting but increased in the supine position in the obese eucapnic OSAS group. In contrast, patients with OHS had increased upper airway resistance in both sitting and supine positions.29 Lee et al. evaluated WOB in the sitting and supine positions, both awake and during stage 2 sleep, in morbidly obese patients with eucapnic OSAS or OHS. While WOB was normal in the sitting position in the eucapnic group, it increased significantly during wakefulness and sleep when supine. In contrast, OHS patients had an abnormally high WOB in both sitting and lying positions regardless of being awake or asleep.9 The results support the hypothesis that changes in body position and sleep may contribute to an increase in airway resistance in eucapnic obese patients when lying supine, and that an increase in upper airway resistance can be a factor contributing to increased WOB and predisposition to daytime hypercapnea in OHS.

Sutherland et al. examined middle-aged obese men with moderate-to-severe OSAS who underwent weight loss, and found that weight loss increases velopharyngeal airway volume and that changes in upper airway length appear to have a significant influence on the reduction in apnoea frequency.30

Physiological changes and the WOB during exercise in obesity

As highlighted previously, FRC and expiratory reserve volume are reduced in the resting state in obese subjects, forcing them to breathe close to residual volume. Breathing at low lung volumes places the tidal flow volume loop in a region where it may encroach on the maximal flow volume envelope, increasing the propensity to limit expiratory flow, resulting in an inability to decrease FRC with exercise.31–33

On the other hand, by increasing FRC during peak exercise, obese patients may actually pseudonormalize their FRC and allow tidal volume to become positioned on a more compliant portion of the pressure–volume curve of the respiratory system. Because of dynamic hyperinflation and increased respiratory system elasticity, obese subjects use a rapid, shallow breathing pattern to minimize elastic work and oxygen cost of breathing, and to optimize breathing comfort.32,33

During exercise, minute ventilation is increased such that the ventilatory reserve may be reduced during peak exercise. The VO2 and VCO2 at rest and during exercise are increased in obesity.31–33 The increased metabolic requirements in obesity are due to the additional energy needed to move heavier body parts during exercise, reduce the mechanical efficiency of peripheral muscles, and increase WOB and VO2.32–34 The VO2 peak is decreased when viewed in terms of actual body weight but is normal or increased if corrected for ideal weight. Thus, a healthy obese subject may have good cardiovascular fitness despite higher metabolic cost and reduced work capacity to perform even modest activity.35,36

Increased VO2 for a cycle ergometer is predictable. In severely obese subjects, the VO2–work rate relationship is displaced upward by approximately 6 mL/min/kg of extra body weight without a change in the slope.35,36 Even during maximal exercise on a cycle ergometer, both peak work rate and VO2 are normal in healthy obese subjects.33,35 Obese subjects can increase their ventilation sufficiently to avoid hypercapnea, while the ventilatory response to inhaled CO2 is not different from that seen in a normal-weight control group.33,34

Dyspnoea at rest may be reported by obese individuals even without demonstrable lung disease.16,32 Dyspnoea is common during exercise, and breathlessness may be related to an increase in the oxygen cost of breathing, which increases parabolically with breathing frequency.37 Increased respiratory muscle force in obesity due to either abnormal respiratory mechanical factors or increased metabolic demand imposes an increased WOB and/or increased respiratory drive, and are important factors contributing to the sensation of breathlessness in obese individuals.16,32,38 Sahebjami found that healthy obese men who reported breathing difficulties at rest had higher BMI than subjects without dyspnoea. Forced expiratory flow at 75% vital capacity, maximum voluntary ventilation, Plmax and PEmax were lower in subjects with dyspnoea, many of whom were also smokers.16 Large airway function (including FVC, forced expiratory volume in 1 s and forced expiratory volume in 1 s/FVC ratio), lung volumes and gas exchange parameters were similar between the two groups. Obese but otherwise healthy subjects may experience dyspnoea at rest. Risk factors for dyspnoea in these individuals include reduced Plmax, PEmax and maximum voluntary ventilation, increased BMI and small airway dysfunction.16

Collet et al. found that breathlessness was higher in an obese group with a BMI >49 kg/m2 compared with those with a BMI ≤49 kg/m2. Dyspnoea in these subjects was related to lung function and inspiratory muscle function. Subjects with BMI >49 kg/m2 had a significantly higher partial arterial carbon dioxide concentration (PaCO2) and significantly lower vital capacity, inspiratory capacity and PImax values compared with those with a BMI ≤49 kg/m2. Inspiratory muscle performance correlated more significantly with dyspnoea in those in whom BMI ≤49 kg/m2.39

Babb et al. also found that dyspnoea on exertion is common in otherwise healthy obese women. There was no difference in peak exercise capacity between obese women with and without exertional dyspnoea. However, dyspnoea was associated with a greater increase in the oxygen cost of breathing during exercise.32 El-Gamal et al. evaluated ventilatory drive by the carbon dioxide re-breathing technique and assessed dyspnoea using the dyspnoea score of the Chronic Respiratory Disease Questionnaire. Severely dyspnoeic obese subjects had higher respiratory drive parameters and lower lung volumes compared with the mild-to-moderate dyspnoea group. After body weight reduction surgery (gastroplasty), there were significant reductions in BMI, dyspnoea score and respiratory drive parameters.40 The results suggest that dyspnoea commonly observed in obese patients may be related to increased ventilatory drive and reduced static lung volumes. Scano et al. suggested that during exercise, decreased respiratory muscle efficiency or deconditioning may contribute to dyspnoea in hyperinflated obese subjects.41


The VO2, VCO2 and WOB are increased in obese subjects.5,42 Central respiratory drive and minute ventilation are increased to maintain eucapnia despite abnormal respiratory system mechanics and high WOB.5,13,42 In contrast, some morbidly obese individuals who lack this compensatory increase in respiratory drive are at high risk of developing respiratory failure when load increases.43 The development of awake hypercapnea in the morbidly obese is defined as OHS.

Sampson and Grassino showed that subjects with simple obesity have increased hypercapnic ventilatory responses compared with normal subjects, suggesting that subjects with eucapnic obesity attempt to maintain ventilatory homeostasis in the presence of excessive ventilatory load. In contrast, patients with OHS have decreased neuromuscular responses to hypercapnea when compared with either simple obesity or normal subjects. This is the result of a blunted central ventilatory drive and not due to worse mechanical limitations.44 Lopata and Onal demonstrated that in eucapnic obese subjects without sleep apnoea, diaphragmatic electromyogram responses to hypercapnea were greater than in control subjects, eucapnic obese subjects with sleep apnoea and OHS patients. Non-obese healthy subjects had increased diaphragmatic electromyogram and mouth occlusion pressure responses to hypercapnea when an external weight load was added to their upper abdomen. In contrast, mouth occlusion pressure responses to hypercapnea were decreased in eucapnic obese patients with sleep apnoea and in OHS patients.45

In the study by Steier et al., ventilatory load and neural drive were higher in simple obese subjects than in non-obese healthy subjects.13 Burki and Baker found that eucapnic obese subjects had increased resting inspiratory neuromuscular drive, ventilatory responsiveness to hypoxia and increased minute ventilation primarily from an increase in respiratory rate. A decreased expiratory time per breath produced an increase in the inspiratory-to-expiratory time ratio at rest, suggesting that morbidly obese subjects maintain eucapnia through alterations in central breath timing.42 Work by Chlif et al. led them to conclude that in obese individuals, impaired respiratory mechanics, decreased maximal inspiratory pressure, increased ventilatory drive and increased WOB alter the breathing pattern, resulting in a rapid shallow breathing, predisposing them to ventilatory failure.18

Ventilatory responses to hypercapnea and hypoxaemia are reduced in OHS subjects compared with simple obesity and obese eucapnic sleep apnoea patients.43,46 Zwillich et al. found that hypoxic and hypercapnic ventilatory drives were markedly reduced in OHS subjects compared with normal controls, and may contribute to the alveolar hypoventilation characteristic of these individuals.46 After weight reduction, VO2 and respiratory drive are decreased in obese eucapnic sleep apnoea patients but are increased in OHS.40,43 Similarly, ventilatory responsiveness to CO2 is reduced in obese eucapnic sleep apnoea patients after weight loss but is increased in OHS patients.40,43 These findings suggest that in obese individuals, ventilatory demand can be significantly reduced and ventilatory capacity increased by weight loss.

Depressed chemosensitivity in morbidly obese subjects may arise from genetic factors, obesity, sleep deprivation, sleep disordered breathing, nocturnal hypoxaemia and sleep fragmentation. All of these may result in impaired compensation for mass loading, which prevents them from initiating the predicted rise in respiratory muscle drive and predispose some to the development of hypercapnea. However, the blunted respiratory response is not likely to be familial because the ventilatory response to hypercapnea is similar between first degree relatives of patients with hypercapnic OSAS and control subjects.47 Similarly, Javaheri et al. found that familial respiratory chemosensitivity does not predict hypercapnea in OHS patients.48

Support for the notion that weight load is an important mechanism underlying blunted respiratory drive in the morbidly obese arises from the observation that weight reduction improves PaCO2 in those with hypercapnea. Chapman et al. demonstrated that in eucapnic morbid obesity without OSAS, the hypercapnic and isocapnic hypoxic ventilatory response fell after weight reduction surgery (gastroplasty) such that for a given oxygen saturation, the mean ventilation was significantly lower in the less obese state.49 Gold et al. reported that the awake hypercapnic ventilatory response was lower among non-hypercapnic sleep apnoea patients than in obesity-matched non-apnoeic controls. Sleep apnoea patients had lower awake partial arterial oxygen concentration and higher awake PaCO2. The lower hypercapnic ventilatory response, higher waking PaCO2 and lower total lung capacity in the sleep apnoea patients are similar to that seen in OHS patients. The results suggest that disturbances in lung function and ventilatory control in obese eucapnic sleep apnoea patients are intermediate along a continuum from simple obesity to OHS.50

Verbraecken et al. evaluated patients with various types of sleep apnoea and found that a depressed hypercapnic ventilatory response could only be observed in chronic hypercapnic obese OSAS patients but not in obese eucapnic sleep apnoea patients or the overlap group. One month of nocturnal CPAP therapy in obese eucapnic sleep apnoea patients did not change CO2 drive.51 Lin et al. showed that eucapnic obese OSAS patients had normal hypercapnic and hypoxic ventilatory drives while awake before nasal CPAP treatment, and there was no change in ventilatory drive following nasal CPAP treatment. In contrast, both hypercapnic and hypoxic ventilatory drives during wakefulness were significantly impaired in OHS subjects before nasal CPAP treatment, and improved after 2 weeks of nasal CPAP therapy.52

Lastly, Moura et al. found that the respiratory drive of obese normocapnic OSAS patients was within the range seen in controls. There was a significant increase in awake ventilatory drive after three consecutive nights of nasal CPAP therapy.53 As individuals with eucapnic obesity often have similar BMI as those with hypercapnea,40,43 it suggests that other mechanisms aside from weight are involved in the development of hypercapnea.

Respiratory drive and sleep-breathing abnormalities

Sleep deprivation can lead to reduced ventilatory response to hypercapnea.54 Obese individuals with OSAS frequently have poor sleep efficiency, which can impact on ventilatory responsiveness and the development of hypercapnea. In a large study of patients with OSAS, Laaban and Chailleux found 11% to be hypercapnic. Risk factors for hypercapnea in this study included higher BMI and reduced FVC.55 However, only a small portion of obese patients, even those with severe OSAS, develop daytime hypercapnea. Kaw et al. reported the severity of obesity and OSAS, and the degree of restrictive pulmonary function as being significant predictors of chronic hypercapnea.56 Akashiba et al. found that low mean nocturnal arterial oxygen saturation was the factor most strongly associated with hypoventilation, suggesting that recurrent nocturnal hypoxaemia may blunt central respiratory drive.57 Chouri-Pontarollo et al. identified two groups of OHS individuals: those with CO2 responses within the normal range and those who were low responders. Those with blunted responses were objectively sleepier and had more severe rapid eye movement hypoventilation than those with more normal responses.58

Banerjee et al. showed that extremely obese subjects (BMI ≥50 kg/m2) without OHS are still able to maintain eucapnia even in the presence of OSAS. Despite similar BMI, severities of apnoea/hypopnoea index, arousal indices and sleep architecture, patients with OHS exhibited more severe oxygen desaturation compared with the eucapnic obese group. One-night CPAP treatment significantly improved the apnoea/hypopnoea index and sleep architecture abnormalities. In the OSAS without OHS group, only 9% continued to spend >20% of total sleep time with SpO2 <90%, compared with 43% of OHS subjects.59

It is generally believed that arterial hypoxaemia from apnoea results from simple alveolar hypoventilation similar to that of voluntary breath holding. The ultimate degree of desaturation is determined by the duration of breath holding, lung volume at onset and blood oxygen at apnoea onset because of the shape of the oxyhaemoglobin dissociation curve. If the patient is a CO2 retainer because of decreased central drive or increased dead space, alveolar PO2 will be lower, leading to lower alveolar oxygen stores and more rapid desaturation.43,56,60 It is possible that persistent hypoxaemia is accompanied by defective ventilatory load compensation or that the ventilatory response to hypoxia and/or hypercapnea is so depressed in OHS subjects that inadequate alveolar ventilation is accompanied by blood gas derangements.59,60

In obese individuals with OSAS, there may be a difference in ventilatory pattern following apnoeic events between individuals who are able to maintain eucapnia and those who are hypercapnic, suggesting that under certain conditions, repetitive abnormal breathing during sleep may induce a depression of ventilatory drive.43,61–63 When apnoeas or hypopnoeas occur during sleep, ventilation is intermittently reduced, permitting acute episodes of hypercapnea to arise. However, most obese individuals can sufficiently hyperventilate following apnoea to eliminate the accumulated CO2, and therefore maintain overall eucapnia.63 In contrast, when there is inadequate compensatory hyperventilation (failure to sufficiently increase inter-event ventilation or reduce inter-event duration) immediately after each episode or if the response to the accumulated CO2 is blunted, increased PaCO2 during sleep can occur.43 Thus, in order for hypercapnic OSAS to develop, other factors (e.g. chronic obstructive pulmonary disease, hypoxia, heart failure, kyphoscoliosis, infection, sedatives or diuretic therapy) need to be present to impair the acute ventilatory compensation for transient sleep hypercapnea.

In order to explain the development of awake hypercapnea in some obese individuals, Norman et al. proposed a model outlining how progression from acute intermittent nocturnal hypercapnea during apnoeic events to persistent daytime hypoventilation is possible. This involves impairment of the renal bicarbonate excretion aside from sleep disordered breathing.64

The transient accumulation of CO2 following apnoeic/hypopnoeic events initiate the kidneys to retain HCO3-. As the time constant for HCO3- excretion is longer than that for PCO2, the small rise in bicarbonate during a single night may not be completely excreted prior to the next sleep period, and elevated HCO3- concentration will blunt the change in [H+] for a given change in PCO2, thereby blunting respiratory drive.64,65 Persistently high HCO3- not only enables the development and maintenance of this state of hypoventilation via blunting of the respiratory drive, but also leads to chronic daytime hypercapnea.65 In contrast, when the ventilatory response to CO2 and renal excretion of HCO3- is normal, there are no increases in PCO2 and HCO3-. This is because bicarbonate excretion during the day will compensate for that retained during the night. Therefore, most eucapnic obese subjects with or without sleep breathing disorders do not produce sustained hypercapnea despite increased mechanical loading and VCO2 secondary to mass loading, and intermittent nocturnal upper airway obstruction.63

Neurohormonal influences on respiratory drive

Wakefulness provides a tonic input into ventilation via the reticular activating system. Stimulation of the reticular activating system facilitates breathing. Horner et al. found that a transiently aroused state immediately after spontaneous sleep is neurophysiologically distinct from sustained wakefulness. The abrupt electrographical changes associated with substantial brief cardiorespiratory activation at awakening suggest that the transiently aroused awake state may contribute to the altered magnitude of the post-event ventilatory response in non-obese healthy individuals, and obese individuals with or without sleep disordered breathing.66 Sustained nocturnal hypoxaemia increases the arousal threshold, and therefore may impair the normal defence mechanisms that operate to minimize the insult of abnormal breathing and gas exchange during sleep.67 In animal models, hypoxia has been shown to impair the synthesis and turnover of many neurotransmitters, including γ-aminobutyric acid, dopamine and adenosine.68–70 Sustained hypoxia has also been shown to impair the arousal response to external resistive loading and airway occlusion from sleep even in healthy, non-obese men.67 OHS subjects exhibit more severe prolonged oxygen desaturation during sleep than eucapnic morbidly obesity subjects with OSAS. Chronic hypoxaemia, poor sleep quality and possibly the influence of leptin that alters hypoxic and hypercapnic ventilatory responses may contribute in OHS development.56,59 Residual desaturation in many OHS subjects during initial CPAP treatment when compared with extremely severely obese subjects without OHS probably reflects a continuously impaired ventilatory drive and hypoventilation despite the reversal of apnoeic events.59

Deficits in hypoxic and hypercapnic drives are corrected in many OHS patients after positive airway pressure treatment, but are not changed by positive airway pressure in those with eucapnic OSAS.52,71 Correcting nocturnal oxygenation and reversing daytime hypercapnea by effective positive airway pressure treatment supports the hypothesis that blunted central respiratory drive in hypercapnic obesity is related to hypoxaemia and hypercapnea during sleep. Furthermore, the reversibility of the blunted central drive suggests that it is a secondary effect of the syndrome, not its origin.

Leptin, a 16 kD satiety hormone produced by adipocytes, plays a key role in regulating energy intake and energy expenditure.72 Obesity leads to increased VCO2 and load. Leptin reduces appetite and weight via receptors in the hypothalamus, and increases ventilation by stimulating central respiratory centres.27,72 In non-obese individuals, serum leptin levels are reduced by sleep deprivation and increased by perceived emotional stress.73,74 Leptin levels rise in proportion to body fat, and obese individuals generally exhibit unusually high circulating leptin concentration.27,75 Hence, the presence of high levels of circulating leptin in obesity are thought to increase ventilation to compensate for increased work and VCO2, explaining why most severely obese individuals do not develop hypercapnea. On the other hand, hyperleptinaemia is associated with lower respiratory drive76 and the presence of hypercapnic respiratory failure in obesity.77,78 These data suggest that in some obese individuals, resistance to the ventilatory stimulatory effects of leptin may develop.

Observations also suggest that OHS patients may have central leptin resistance or reduced cerebrospinal fluid penetration of leptin. For leptin to affect the respiratory centre and increase minute ventilation, it has to penetrate the blood–brain barrier. A leptin receptor has been cloned from the choroid plexus.79 Caro et al. found that mean serum leptin is 318% higher in obese than in lean individuals, while the leptin cerebrospinal fluid-to-serum ratio is fourfold higher in lean individuals compared with obese subjects. Thus, the capacity of leptin transport across the blood–brain barrier may be lower in obese individuals.79 This difference may explain why some obese individuals with severe OSAS develop OHS while others do not. However, the absolute cerebrospinal fluid leptin level is still 30% higher in the obese group, which contradicts the hypothesis that low leptin levels in cerebrospinal fluid leads to hypoventilation. Central leptin resistance may, therefore, play a role in the development of hypercapnea.

Insulin-like growth factor 1 has a pleiotropic role in metabolism, ventilatory control, muscle function and cardiovascular protection. In a case control study of morbidly obese subjects with OSAS, Monneret et al. reported that serum insulin-like growth factor 1 levels were inversely and closely associated with PaCO2. Reduced inspiratory capacity and FVC, reflecting decreased respiratory muscle strength, were seen in subjects with low insulin-like growth factor 1. Hence, deficient insulin-like growth factor 1 secretion might play a role in the development of hypercapnea in obese individuals through altered ventilatory drive and respiratory muscle weakness.80 Summarized possible mechanisms by which obesity can lead to blunted ventilatory drive and chronic daytime hypercapnea is shown in Figure 1.

Figure 1.

Possible mechanisms by which obesity can lead to blunted ventilatory drive and chronic daytime hypercapnea. ERV, expiratory reserve volume; FRC, functional residual capacity; VCO2, CO2 production; VDVT, dead space-tidal volume ratio; VO2, oxygen consumption.

In conclusion, obesity, particularly severe central obesity, affects respiratory system mechanics, upper airway mechanical function and VCO2. Despite the increased WOB imposed by the excessive weight of most individuals, even the morbidly obese can maintain eucapnia through increased respiratory drive. However, if the compensatory mechanisms that normally operate to achieve this are impaired or overwhelmed, awake hypercapnea emerges.


The authors thank Ms Shwu-Fang Liaw, Ms Mei-Wei Lin and Ms Feng-Ting Chang for their help in preparing this review of literature.