INTRODUCTION — The respiratory system is dependent upon a complex system of ventilatory control to ensure appropriate and adequate ventilation in order to supply oxygen, remove carbon dioxide, and maintain acid-base homeostasis. Respiratory centers in the brain integrate input from neural and chemical receptors and provide neuronal drive to the respiratory muscles, which maintain upper airway patency and drive the thoracic bellows to determine the level of ventilation.
The abnormalities of ventilatory control that result from a variety of disorders, including chronic obstructive pulmonary disease (COPD), asthma, Ondine's curse, carotid body resection, Cheyne-Stokes respiration, myxedema, starvation, and neuromuscular disease, will be reviewed here. In addition, the effects of several pharmacologic agents on ventilation and ventilatory control will be reviewed. The physiologic aspects of ventilatory control and the evaluation of patients with disorders of ventilation are discussed separately. (See "Control of ventilation".)
CHRONIC OBSTRUCTIVE PULMONARY DISEASE — Development of hypercapnia in patients with chronic obstructive pulmonary disease (COPD) generally is associated with more severe disease, but is inconsistent among patients with similar degrees of airflow obstruction. Thus, patients with similar spirometric values can exhibit the "blue bloater" profile of hypercapnia (CO2 retention, with an elevated arterial carbon dioxide tension [PaCO2]) and hypoxemia or the "pink puffer" profile with eucapnia and relatively normal oxygen levels [1,2]. Those in the first group have reduced respiratory drive whereas the latter have increased drive.
The pattern of breathing in patients with chronic CO2 retention is characterized by a low tidal volume and high frequency, usually ≥22 breaths/minute. This respiratory pattern in combination with impaired matching of ventilation and perfusion leads to increased dead space ventilation and diminished alveolar ventilation, which contributes to CO2 retention [3].
Some hypercapnic, hypoxemic patients with COPD develop increased CO2 retention when O2 is administered. Such patients usually have both blunted hypercapnic and hypoxic drives [4]. Other factors contributing to CO2 retention during O2 breathing include worsening of ventilation-perfusion distribution secondary to relief of compensatory pulmonary vasoconstriction [5], and unloading of CO2 due to the Haldane effect [6]. (See "The evaluation, diagnosis, and treatment of the adult patient with acute hypercapnic respiratory failure".)
ASTHMA — In some patients with asthma, an increased risk of near fatal asthma exacerbations, characterized by severe hypoxemia and hypercapnia, is associated with a depressed ventilatory drive. Even during periods of disease quiescence when patients have no bronchospasm, patients with a history of near fatal asthma demonstrate depressed hypoxic and hypercapnic ventilatory responses as well as a reduced perception of dyspnea with added resistive loads (figure 1 and figure 2) [7]. (See "Identifying patients at risk for fatal asthma".)
Patients with asthma who have depressed chemosensitivity and inappropriately low sensations of dyspnea probably delay seeking medical treatment during an asthma attack, resulting in their increased risk of fatal asthma. Asthmatic subjects who develop CO2 retention during an asthma attack are likely to repeat the same in subsequent attacks [7-9].
ONDINE'S CURSE — The term Ondine's curse has been applied to patients with alveolar hypoventilation due to impaired autonomic control of ventilation, but whose voluntary control remains intact. Classically, they "forget to breathe" when they fall asleep, but maintain relatively normal blood gases while awake [10,11].
Ondine's curse is usually due to congenital central hypoventilation syndrome, but can also be caused by brainstem tumors and surgical incisions into the second cervical segment of the spinal cord to relieve intractable pain [12,13]. Paroxysmal central apnea has also been observed following medullary infarction [14,15]. (See "Complications of stroke: An overview", section on 'Pulmonary complications'.)
Congenital central hypoventilation syndrome — Congenital central hypoventilation syndrome (CCHS) is a rare genetic disorder caused by a defect in the PHOX2B (paired-like homeobox 2B) gene that typically presents with cyanosis occurring primarily during sleep [16]. CCHS is associated with a nearly absent respiratory response to hypoxia and hypercapnia, no respiratory discomfort during CO2 inhalation, mildly elevated arterial carbon dioxide tension (PaCO2) during wakefulness, and markedly elevated PaCO2 during non-REM sleep [17-20]. Patients with CCHS increase their ventilation and maintain relatively normal PaCO2 levels during exercise and lower their PaCO2 during passive leg cycling due to nonchemoreceptive inputs [21-24].
The evaluation and management of CCHS and its associations with Hirschsprung's disease and neural crest tumors are described separately. (See "Congenital central hypoventilation syndrome and other causes of sleep-related hypoventilation in children", section on 'Congenital central hypoventilation syndrome'.)
Late-onset central hypoventilation syndrome — Unlike congenital central hypoventilation syndrome (CCHS), which has a neonatal onset, late-onset central hypoventilation syndrome (LO-CHS) can develop anytime from infancy through adulthood. Like CCHS, these late-presenting cases are caused by PHOX2B mutations. The clinical features, diagnosis, and management of LO-CHS are described separately. (See "Congenital central hypoventilation syndrome and other causes of sleep-related hypoventilation in children", section on 'Late-onset central hypoventilation syndrome'.)
Rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation (ROHHAD) — ROHHAD is associated with rapid-onset obesity and central hypoventilation during sleep starting after 1.5 years of age, and also hypothalamic dysfunction and absence of a CCHS-related PHOX2B mutation (table 1) [25-27].
In addition to hypothalamic-pituitary abnormalities, patients develop symptoms of autonomic nervous system dysfunction (hyperthermia or hypothermia). About 40 percent develop neural crest tumors 7 to 16 years after the onset of obesity. At present there is no genetic testing available to diagnose ROHHAD, other than excluding the presence of a PHOX2B mutation. The evaluation and management of ROHHAD are discussed separately. (See "Congenital central hypoventilation syndrome and other causes of sleep-related hypoventilation in children", section on 'Rapid-onset obesity with hypothalamic dysfunction, hypoventilation, and autonomic dysregulation (ROHHAD)'.)
CAROTID BODY RESECTION — Carotid body resection, or glomectomy, was introduced as a treatment for asthma in Japan during the 1940s, and used in the United States until the 1960s, when controlled studies of the procedure did not support its efficacy [28]. Carotid body resection has been used for relief of dyspnea in patients with severe COPD, leading to improved dyspnea, but worse hypoxemia and hypercapnia [29].
Glomectomy depresses the hypoxic ventilatory response during exercise; minor abnormalities may also be seen in CO2 regulation during exercise, but eucapnia is generally present at rest [30].
Bilateral endarterectomy for carotid occlusive disease also may result in destruction of peripheral chemoreceptors, with diminution of hypoxic ventilatory response and slight elevation of resting arterial tension of carbon dioxide (PaCO2).
CHEYNE-STOKES RESPIRATION — Cheyne-Stokes respiration (CSR) describes cyclic breathing in which apnea is followed by gradually increasing respiratory frequency and tidal volume (ie, hyperpnea), then gradually decreasing respiratory frequency and tidal volume until the next apneic period [31]. It is considered a type of central sleep apnea syndrome. (See "Sleep-disordered breathing in heart failure".)
Mechanism — Delay between changes in ventilation and detection of the resulting arterial tension of carbon dioxide (ie, PaCO2) by the central chemoreceptors maintains a cyclic pattern of respiration. Factors believed to contribute to this delay include prolonged lung to brain circulation time, reduced tissue and lung CO2 and O2 stores, and increased ventilatory drive. The increased ventilatory drive is due, at least in part, to loss of effective damping factors.
In contrast to normal physiology, the lowest arterial oxygen saturation and highest PaCO2 occur near peak hyperpnea in CSR, suggesting that the level of PaCO2 at any given moment reflects the patient's stimulation to breathe and not the effectiveness of ventilation.
Comorbidities — Cheyne-Stokes respiration is commonly associated with cardiac disease; it can also accompany neurologic disease, sedation, normal sleep, acid-base disturbances, prematurity, and altitude acclimatization [32-34]. In one study of 42 patients with stable heart failure, 45 percent of patients had more than 20 episodes of apneas or hypopneas per hour of sleep [35]. (See "Sleep-disordered breathing in heart failure".)
Diagnosis — The diagnostic evaluation of suspected sleep-disordered breathing is the same for patients with or without heart failure. An in-laboratory overnight polysomnogram is the gold standard diagnostic test. (See "Central sleep apnea: Risk factors, clinical presentation, and diagnosis", section on 'Diagnostic evaluation' and "Sleep-disordered breathing in heart failure", section on 'Diagnosis'.)
Management — Management of CSR focuses on treatment of the underlying cause (eg, optimizing medical therapy of heart failure), but may also include continuous positive airway pressure (CPAP) during sleep, supplemental oxygen, or adaptive servoventilation (ASV) in selected patients. The management of CSR is discussed in more detail separately. (See "Sleep-disordered breathing in heart failure", section on 'Management'.)
MYXEDEMA — Hypoventilation can occur in patients with severe hypothyroidism [36]. It probably reflects both respiratory muscle weakness and depression of ventilatory drive [37,38]. Some patients will need noninvasive or invasive ventilatory support during the initial phases of thyroid replacement. The evaluation and management of myxedema are discussed separately. (See "Respiratory function in thyroid disease" and "Myxedema coma".)
STARVATION — The hypoxic ventilatory response falls approximately 40 percent among normal volunteers following 10 days of 500 kcal/day diet restriction [39]. There is little change in the hypercapnic ventilatory response. Prolonged caloric restriction can have other effects on pulmonary function, including the development of hyperinflation combined with emphysematous radiographic changes [40]. (See "Anorexia nervosa in adults and adolescents: Medical complications and their management", section on 'Pulmonary'.)
NEUROMUSCULAR DISEASE — Impairment of the ventilatory response to CO2 occurs in patients with a wide variety of neuromuscular diseases [41]. The reduced ventilatory response is due to neuromuscular weakness, and ventilatory drive, as measured by mouth occlusion pressure, is usually well preserved [42-44]. (See "Control of ventilation", section on 'Mouth occlusion pressure'.)
Patients with neuromuscular disease show different first-breath ventilatory responses to graded elastic and resistive loads [45]. These abnormalities in respiratory timing suggest that neuromuscular diseases impair respiratory perception as well as respiratory muscle strength. (See "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation".)
DRUGS AFFECTING VENTILATORY DRIVE — A variety of drugs can affect central ventilatory drive. These medications may either depress or stimulate ventilation, and may be useful therapeutically in some clinical situations.
Central nervous system depressants — Several classes of drugs suppress central respiratory drive, including opiates, barbiturates, and benzodiazepines. These medications should be used judiciously in patients with preexisting hypoventilation, but may be useful in patients with hyperventilation secondary to pain, anxiety, or malignant pleural effusion [46]. (See "Pain control in the critically ill adult patient" and "Sedative-analgesic medications in critically ill adults: Selection, initiation, maintenance, and withdrawal".)
Medroxyprogesterone — Medroxyprogesterone increases the ventilatory drive in normal males, leading to about a 5 mmHg (0.67 kPa) fall in arterial tension of carbon dioxide (PaCO2) [47]. The drug has been used to treat patients with excessive polycythemia of high altitude and central hypoventilation syndromes, such as the obesity-hypoventilation syndrome [48-50]. (See "Treatment and prognosis of the obesity hypoventilation syndrome".)
Its effectiveness in treating hypercapnia associated with COPD is unclear; reductions in PaCO2 are not generally accompanied by improvements in other objective or subjective outcome measures. As an example, one trial of medroxyprogesterone in seven patients with hypercapnic chronic bronchitis found that treatment lowered the mean PaCO2 from 51 to 44 mmHg (6.78 to 5.85 kPa), but had no effect upon sensations of dyspnea or 12-minute walking distance [51].
Theophylline — Theophylline increases the hypoxic ventilatory response and prevents the fall in hypoxic ventilatory response that normally occurs after 15 minutes of hypoxemia [52]. Like medroxyprogesterone, theophylline may have some utility in the treatment of central hypoventilatory states, but its utility in hypercapnia secondary to obstructive sleep apnea is limited.
In a case report, theophylline was used successfully to treat a woman with diabetes mellitus, neuropathy, and end stage renal disease who had multiple cardiorespiratory arrests secondary to severe Cheyne-Stokes respiration [53].
Acetazolamide — Acetazolamide, an inhibitor of carbonic anhydrase, has several effects relevant to ventilation.
●In the brain, by blocking CO2 conversion to bicarbonate in tissue capillaries, acetazolamide can acutely raise local tissue partial pressure of carbon dioxide (PCO2). Acetazolamide also lowers cerebrospinal fluid bicarbonate in normal individuals at both 3000 and 14,000 feet [54]. Locally elevated PCO2 and lower pH in the brain would increase central ventilatory drive and lower PaCO2.
●In the kidney, acetazolamide increases retention of hydrogen ion and increases renal excretion of bicarbonate, causing a metabolic acidosis over several hours; the metabolic acidosis further increases respiratory drive.
●Opposing the effects that promote increased ventilation, acetazolamide blocks the reciprocal conversion of bicarbonate to CO2 in pulmonary capillaries, thereby impairing the lung's ability to excrete CO2. This could lead to an increase in PaCO2 if ventilation remained constant. However, in normal individuals acetazolamide leads to increased minute ventilation resulting in lower end-tidal PCO2, lower PaCO2, and higher arterial oxygen tension (PaO2) [54,55].
Patients receiving acetazolamide who are able to increase their ventilation will decrease their PaCO2 because their increased ventilatory drive will surpass their impaired lung excretion of CO2. In contrast, patients who are unable to augment their ventilation (eg, those with severe COPD) may develop a severe respiratory acidosis on acetazolamide [56]. A review of the use of acetazolamide in patients with COPD raises concerns of its use in those with severe COPD [57].
Acetazolamide is effective in the prophylaxis and treatment of acute and chronic mountain sickness [58-60]. This effect is likely due to increased ventilation and decreased oxygen desaturation with acetazolamide. (See "High-altitude illness: Physiology, risk factors, and general prevention".)
In addition, acetazolamide reduces central sleep apneas in patients with heart failure and Cheyne-Stokes respiration, although not without adverse effects. A beneficial effect was shown in a small cross-over study of 12 patients with heart failure who had a reduction in episodes of central sleep apnea when acetazolamide was given an hour before bedtime [61]. Another study found that four days of acetazolamide led to reduced Cheyne-Stokes respiration and a blunted hypoxic ventilatory response, while it increased hypercapnic ventilatory response and lowered the workload achievable during cardiopulmonary exercise testing [62]. The efficacy and side effects of long term treatment are unknown.
Antioxidants — A combination of antioxidants – vitamins E, A, and C for two months, allopurinol for 15 days, and N-acetylcysteine for three days – increases the sensitivity of the ventilatory response to CO2 in normal subjects during hyperoxic unloaded breathing and after resistive breathing [63]. The effect of antioxidants on ventilatory response to CO2 in non-hyperoxic conditions or in patients with lung disease is unknown.
SUMMARY AND RECOMMENDATIONS
●A variety of disorders, including chronic obstructive pulmonary disease, asthma, Ondine's curse, carotid body resection, Cheyne-Stokes respiration, myxedema, starvation, and neuromuscular disease, are associated with abnormal ventilatory control. (See 'Introduction' above.)
●Some patients with chronic obstructive pulmonary disease (COPD) develop hypercapnia (ie, elevated arterial carbon dioxide tension [PaCO2] or CO2 retention) and hypoxemia. The development of hypercapnia in COPD usually occurs in patients with more severe COPD who also have blunted hypoxic and hypercapnic respiratory drives. Some hypercapnic, hypoxemic patients with COPD develop increased CO2 retention when O2 is administered. (See 'Chronic obstructive pulmonary disease' above.)
●Some patients with asthma demonstrate depressed hypoxic and hypercapnic ventilatory responses as well as a reduced perception of dyspnea. These patients appear to be at an increased risk of near fatal asthma exacerbations, characterized by severe hypoxemia and hypercapnia. It is thought that the reduced perception of dyspnea contributes to delays in seeking medical attention during asthma exacerbations. (See 'Asthma' above and "Identifying patients at risk for fatal asthma", section on 'Poor perception of dyspnea'.)
●The term Ondine's curse has been applied to patients with alveolar hypoventilation due to impaired autonomic control of ventilation, but normal voluntary control. The most common cause is congenital central hypoventilation; other etiologies include brainstem injury or tumor and late-onset central hypoventilation syndrome. (See 'Ondine's curse' above.)
●Congenital central hypoventilation syndrome (CCHS) is associated with a nearly absent respiratory response to hypoxia and hypercapnia, no respiratory discomfort during CO2 inhalation, mildly elevated PaCO2 during wakefulness, and markedly elevated PaCO2 during non-REM sleep. CCHS can occur in association with Hirschsprung's disease, a condition characterized by abnormalities of the cholinergic innervation of the gastrointestinal tract. (See 'Congenital central hypoventilation syndrome' above and "Congenital central hypoventilation syndrome and other causes of sleep-related hypoventilation in children", section on 'Congenital central hypoventilation syndrome'.)
●Cheyne-Stokes respiration (CSR) describes cyclic breathing in which apnea is followed by gradually increasing respiratory frequency and tidal volume (ie, hyperpnea), then gradually decreasing respiratory frequency and tidal volume until the next apneic period. The diagnosis and management of CSR are discussed separately. (See 'Cheyne-stokes respiration' above and "Sleep-disordered breathing in heart failure".)
●A wide variety of neuromuscular diseases are associated with impairment of the ventilatory response to CO2. The reduced ventilatory response is usually due to neuromuscular weakness; ventilatory drive is usually well-preserved. (See 'Neuromuscular disease' above.)
●Rare causes of hypoventilation include carotid body resection or injury, myxedema, and starvation. (See 'Carotid body resection' above and 'Myxedema' above and 'Starvation' above.)
●A variety of drugs can affect central ventilatory drive; central nervous system depressants can decrease ventilatory drive, while medroxyprogesterone, theophylline, acetazolamide, and antioxidants can modestly increase ventilatory drive. However, the clinical use of these drugs to adjust ventilatory drive is limited. (See 'Drugs affecting ventilatory drive' above.)