logo_sociedad.jpg (4878 bytes)


Volumen 8, números 1-2, 1999

Página principal
Equipo directivo y comité científico
Información para los autores
Revistas Anteriores
Envio de artículos
Enlaces a revistas médicas
Congreso virtual de neurología


Respiratory Dysfunction Associated with Acute Cerebrovascular Events

Dr. Pablo R. Castillo, Dr. Mauricio A. Reinoso.

Pulmonary and Critical Care Medicine Section, Baylor College of Medicine and Houston Veterans Affairs Medical Center, Houston, TX, U.S.A.

Cerebrovascular events (CDVE) are a major cause of morbidity and mortality worldwide. Most patients with CVE do not develop significant respiratory problems but when present, they may be a marker of severe neurologic derangement. In one study (1), only 8% of patients presenting with acute carotid territory stroke were electively intubated and mechanically ventilated due to neurologic deterioration. "Good" outcome in terms of survival and neurological status of patients with hemispheric strokes who required mechanical ventilation have been reported in approximately 20% of cases (2). Respiratory disturbances associates with strokes can result from discrete or difuse lesions to key componets of the respiratory controller. The clinical spectrum of respiratory disordes in stroke include abnormal breathing patterns, hypoxemic and hypercapnic respiratory failure, aspiration pneumonia due to an inability to protect the airways and clear the airway by coughing, and acute pulmonary emblism due to prolonged inmobilization. There is a veriety of altered respiratory patterns associated with strokes. There is a variety of altered respiratory patterns associated with strokes (3-5). These changes are not only important in determining the location of the neuroanatomic lesion, but they havealso been regarded as outcome predictors in CVE. This paper reviews the evaluation, management, and effect of respiratory care interventions, management, and effect of respiratory care interventions on a variety of respiratory system problems in patients with CVE.



Hemis´jeroc oscje,oc strples jave beem associated with a modest reduction of the contralateral chest wall and diaphragm movements (6,7). On the other hand, bilateral hemispheric dysfunction has been typically associated with Cheyne-Stokes respiration (8); a periodic breathing pattern characterized by progresive hyperpnea followed by hypopnea and apnea. Cheyne-Stokes respiration is frequently described in patients with hemispheric CVE and significant cardiac and pulmonary diseases but it has also been noted in patients with brainstem strokes.

The exact mechanism of this form ob breathing pattern is not completely understood. In patients with bilateral hemispheric lessions, an increased response to carbon dioxide (CO2), presumably due to loss of the normal cortical inhibition, has been described (5). In patients with congestive heart failure and pulmonary disorders, circulatory slowing and hyperventilation with or without hypoxemia are instability. Cheyne- Stokes respiration is typically present in the tranition from wakefulness to sleep (drowsiness). Cheyne-Stokes respiration is typically present in stroke patients with reduced level of consciousness of in the transition from wakefulness to sleep (drowsiness) and the first two stages of non-rapid eye movement sleep (NREM). Cheyne-Stokes respiration with hypocapnia has been associated with high mortality (9,10). Conversely, intermitten Cheyene-Stokes respiration does not necesarily indicate a poor prognosis and it usually disappears with the improvement in neurological status.



Central neurogenic hyperventilation (CNH), is described in patients with midbrain and upper pontie tegmentum lesions (11,12). To be considered neurogenic in origin, such hyperventilation must be accompanied by a normal PaO2, as response to hypoxic drive can minic thies breathing pattern. True CNH seems to occur rarely and it has been associated with por clinical outcome. Most hyperventilation in stroke patients can be explain by reflex stimulation of the respiratory center by hypoxia, activation of lung and chest wall receptors, and metabolic factors 8i.e., metabolic acidosis with Kussmaulís breating).

Apneustic breathing (13)with its prolonged inspiratory pauses indicates a pontine stroke. Damage to the low pons or high medulla can produce an abnormal breathing pattern characterized by normal breaths separated by irregular pauses or cluster breathing. Grossly irregular breathing occurs in patient with medullary lesion. This ataxic breathing pattern is an ominous sign and usually precedes the patientis demise.

A rare breathing disorder in which the volitional breathing controller is preserverd but automatic breathing is impaired (Ondineis curse) (14m15) can be seen in ischemic lessions involving the lateral tegmentum points and medulla due to distal vertebral artey occlusion (lateral medullary stroke). Clinically, patients can maintain fair to adequate ventilation while awake but they develop severe hypoventilation and life-threatening central apneas during sleep (16). Invasive mechanical ventilation is often needed initially to maintain ventilation. Stable patients can be managed by noninvasive positive pressure ventilation during sleep. The opposite pattern, that is, preservation of atuomatic breathing with inability to voluntarily modify the breathing depth or frecuency has been described in patients with extensive midpontine infarcts ("locked-in syndrome"), lacunar infarcts in the upper pontine corticospinal tract, and presumably in bilateral hemispheric lesions (17).

Finally, increased intracranial pressure leading to transtentorial herniation produces a typical rostrocaudal deterioration in which normal breathing is replaced by Cheyne-Stokes respiration, followed by neurogenic hyperventilation and subsequetntly, atacic respiration that precedes apnea (18).



Patients with acute strokes may not be able to maintain adequate oxygenation and/or ventilation in the absence of an intracardiac shunt (acute respiratory failure) (19). In fact, respiratory arrest represent a primary cause of death in the first few days after a stroke (20). Depending on the pathophysiologic mechanism, patients may present with hypercapnic (pump failure) or hypoxemic respiratory failure. The hallmark of ventilatory pump failure is hypercapnia with acute respiratory acidosis. In acute stroke patients, this hypoventilation is most commonly due to a decreased total minute ventilation secondary to a reduced central respiratory drive. The alveolar-arterial oxygen gradient (P(A-a)O2) is useful in determining in the hypoxemia present in this form of respoiratory failure is only due to hypoventilation (normal gradient) or if there is additional parenchymal lung disease (elevated gradient). Hypercapnia may also result from concomitant abnormalities in one or several or the determinants of the carbon dioxide level in arterial blood [PaCO2]=k VCO2/VE(1-VD/VT)] (Table 1).

Tabla 1. Causes and pathogenesis of respiratory failure in acute stroke.
Increased CO2 production (VCO2) in patients with fever , sepsis, agitation or excessive carbohydrate load, associated with a limited ventilatory capacity (e.g., high VD/VT, low minute ventilation)

Increased dead space (VD/VT) in severe COPD, cystic fibrosis and severe asthma.

Decreased total minute ventilation (VE) due to other conditions associate with ventilatory pump dysfunction:

Decreased central respiratory drive: drugs (narcotics, sedatives, anesthetics), central hypoventilation, hypothyroidism.

Abdominal resoiratory efferents

Spinal Cord: traums, poliomyrliyis, amyotrophic lateral sclerosis, tetanus, rabies. Neuromuscualr: Myasthenia gravis, multiple sclerosis, botulism, Guillain-Barre Syndrome, hypophosphatemia, hypomagnesemia, drugs (streptomycin, amikacin, neuromuscular blockers), polyneuropathy of critical illness, bilateral phrenic nerve injury.

Abnormal chest walla and/or muscles: Severe kyphoscoliosis, ankylosing spondylitis, massive obesity, muscular dystrophy, polymyositis, respiratory muscles fatigue, acid maltase deficiency.

Airway obstruction: epiglotitis, fixed and variable upper airway obstrction due to tummors, post-extubation, tracheomalacia, bilateral vocal cord paralysis.


The hallmark or hypoxemic respoiratory failure is an inability to adequately oxygenate the blood. The main pathophysiologic mechanism involved are V/Q mismatch (respnd to 100% 02) and intrapulmonary shunting (no significant improvement with 100% O2). The patients typically exhibit a rapid shallow breathing pattern and a low or normalCO2. This form of respiratory failure is commonly the result of a diffuse acute lung injury with high permeability pulmonary edema (ARDS), severe pneumonic infiltrates, or cardiogenic pulmonary edema. Rarely, patients with acute brain injury, can present with a form of non-cardiogenic pulmonary edema termed neurogenic pulmonary edema (NPE) (21). The exact mechanism of NPE is unknown but massive addrenergic overacitivity with pulmonary vasoconstriction has been implicated (22,23).

The management of acute respiratory failure is initally supportive aimed at the correction of hypoxemia or hypercapnia until specific actions are implemented to correct, if possible, the factors that lead to the respiratory failure. The main goal in managing hypercapnic respiratory failure. The main goal in managing hypercapnic respiratory failure is to improve alveolar ventilation through the use of mechanical ventilation. This is most comonly done through an endotracheal tube using a volume-controlled ventilator (usual initial settings are: tidal volume (VT) 10 ml/kg, assist-control mode (A/C), rate 8-10 breath per minute, and fractional concentration of inspired oxygen (FiO2) 1.0). Arterial blood gases (ABGs) must be checked 10-20 min later to detect inadverted and potentially life-threatening acute alkalosis secondary to overcorrection of the hypercapnia and to adjust the FiO2. Non-invasive mechanical ventilation could only be tried for cooperative, hemodynamically stable stroke patients with intact upper airway function without excessive secretions, regurgitation, or vomiting.

Parients with V/Q mismatch abnormalities without significant intrapulmonary shunt will usually respond to non-invasive O2 suplementation (i.e., nasal cannula, Venturi mask). In patients with cardiogenic pulmonary edema, the use of CPAP (5-10 cmH20) via a face mask in addition tooxygen supplementation can be beneficial by reducing the transmural pressure of the left ventricle, and therefore afterload but also by decreasing the preload. The ventilatory management of patients with diffuse acute lung injury (i.e., ARDS) requires machanical vantilation and should be viewed as a balance between adequate oxygenation on one hand, and the risk for barotrauma and cardiovascular compromise on the other. And detailed discussion of the ventilatory management of acute lung injury/ARDS is beyond the scope of this review, but a special effort should be made to deliver enough oxygenation and maintain adequate mean arterial pressure in stroke patients to avoid further increased intracraneal pressure and ischemia.

Adequate arterial oxygen saturation is usually archieved by raising the FiO2 and increasing the end-expiratory lung volume to recruit collapsed or flooded alveolI. This can be achieved by adding extrinsic end –expiratory positive pressure (PEEP) and/or setting the ventilator to create dynamic hyperinflation (auto-PEEP). It is unclear if one strategy is more effective that the other in ARDS but in stroke patients with increased intrathoracic pressure with its impedance of cerebral and systemic venous return (25,26). Since the effects of PEEP are not easily estimated, intracranial pressure monitoring is recommended if levels > 10 cm H20 are used. Concomitant elevation of the upper part of the body 30 (with the head in neutral position also help to optimize cerebral venous return. To avoid potiential oxygen toxicity, it is recommended not to use 100% O2 for more than a few hours and to maintain an FiO2 for more than a few hours and to maintain an FiO2 SYMBOL 163\f "Symbol" 0.6. To avoid barotrauma, it is recommended to maintain plateau pressure (end-inflation hold pressure) less than 25 cm H20. The VT chosen shoud be one tha preents lung overinflation (i.e., plateau pressure <35 cm H20) and, alveolar derecruitment at the end of expiration (inadequate oxygenation). This usually means an initial VT of 7-8 ml/Kg that may need to be reduced to 5 ml/kg.

Allowing CO2 retention and respiratory acidemia in an effort to limit barotrauma (through reductions in rate and VT) is referred as "Permisseve Hypercapnia". Although, widely used for the majority or ARDS patients, this strategy should be avoided in CVA patients due to the deleterous effect of CO2 and accidemia on cerebral blood flow (27).



Strokes resulting in depressed level of consciousness or blbar dysfunction may cause abnormal swallowing mechanism and altered cough and gag reflexes that compromise the airway patency. Early identification of such high-risk patients may avoid aspiration of gastric or oropharyngeal contents. Simple routine measures to reduce the risk of aspiration include the use of frequent suctioning of secretions, elevation of the head of the bed, and avoidance of oral intake, at least in the acute period. Nearly 30% of patients experience neurogenic dysphagia following stroke (28,29). After the acute period resolves, swallowing study with modified barium protocol should be obtained if abnormal voluntary cough and gag reflexes or dysphonia are present (30,31), or if supervised feeding is associated with coughing, regurgitation or respiratory distress. According to the degree of dysphagia, oral feeding can still be possible if eating is done only in the upright position, optimal dietary texture is selected, and small size boluses (i.e., <= 1 teaspoon) are given under direct supervision. With severe swallowing problems or episodes of recurrent aspiration, placement of a gastric (i.e., percutaneous endoscopic gastrostomy PEG) or small bowel tubes is indicated.

Prophylactic endotracheal intubation should be considered if patients are not able to protect their airwais. This decision should take into consideration the severity and prognosis of the neurological damage, co-morbid conditions, and prior patientsi and proxyis wishes regarding invasive support (1,2,32,33). Other common indications for endotracheal intubation in stroke patients are tonic-clonic seizures or status epilepticus, respiratory failure due to pneumonia and pulmonary edema and therapeutic hyperventilation (1,2).

Physicians should be aware of the potential risks of perfoming endotracheal intubation in neurological patients who are prone to develop or already have increased intracranial pressure (34-36). In this patients, use of thiopental, nebulized or intravenous lidocaine, and an alpha/beta blocker such as labetalol would prevent further raising in intracranial pressure, coughing and sympathetic response to intubation (37).



Mechanical hyperventilation will acutely and transiently loweir intracranial pressure (24). Cerebral blood flow is linearly related to PaCOO2 between 25 and 80 mm. Hg in normal brain. Lowering the PaCO2 level to 25-30 mm Hg remains a useful emergency therapy for acutely elevated intracranial pressure although, worsening of ischemia due to vasoconstriction in infarct penumbral areas has been postulated as a potential risk (1). The effects of mechanical hyperventilation are short-lived, probably up to 24 hours. Return to normocarbia should be done gradually over 24-48 hours. In patients with otherwise normal lungs, noninvasive continuous monitoring with an infrared capnograph in line with the ventilator can avoid multiple ABGs determinations.



The output of the respiratory center, clinically expressed as respiratory muscles contraction, can be altered by an external stimuous such as mechanical ventilation. Most patients intubated and mechanically ventilated continue to make spontaneous respiratory efforts. If the magnitude of the inspiratory efforts is sufficent, each effort triggers a machine breath, i.e., the respiratory pump entrains the ventilator. On the other hand, when teh amplitude os spontaneous inspiratory efforts is small relative to the sensitivity of the machine triggering mechanism, increases in spontaneous respiratory frequency are not expressed as changes in actual machine rate and minute ventilation (Figure 1). This is often the case in stroke patients requiring mechanical ventilation who have depresed sensorium with reduced respiratory drive. These "wasted respiratory efforts" repesent a form of patient-ventilator asynchrony that is not only associated with increased in oxygen consumption and patient discomfort but they can also produce an irregular breathing pattern which could mislead the evaluthing pattern wich could mislead the evaluating physician.

Figura 1. Airway pressure (PAW) and flow (V) tracings in an intubated mechanically ventilated patient. Nota asynchronous spontaneous respiratory efforts (arrows) during machine deflation. (Dots: mechanical inflation)

Interactions betwen the ventilator and the patient’s respiratory system can mimic breathing patterns observed in non-intubated patients with different levels of cenral newrological dysfunction. Figure 2 shows airway pressure and flow tracing in one of our patients with brainstem ischemic stroke. At first glance, it may seem that he exhibits a form of irregular periodic breathing. Careful evaluation of respiratory muscles contraction by physical exam and analysis of airway and flow tracings during the presumed "apneic" intervals demonstrate a regular tachypneic intrinsic pattern (Figure 2-arrows). Here, some inspiratory efforts are able to trigger machine breaths (dots). With each subsequent mechanical inflation, the operating lung volume increases (manifested by a progressive increase in peak pressures) to a point that the patientis effort cannot overcome the lung elastic recoil pressure plus the machine triggering sensitivity and there is a failure to trigger a mechanical breath. When this patient was allowed to breath spontaneously with the assitence of continuous positive airway pressure (CPAP), the underlying tachypneic pattern was unvovered (Figure 3). This tachpnea was persistent despite normal oxygenation and lung mechanics, leading to a diagnosis of central neurogenick hyperventilation.

21-2.jpg (7294 bytes)

Figura 2. Patient-ventilator asynchrony in a patient with brainstem stroke mimicking irregular periodic breathing. (Dots: mechanical inflation; arrows: spontaneous respiratory efforts. See text for explanation)


21-3.jpg (9314 bytes)

Figura 3. Tachypneic breathing pattern in an intubated patient with brainstem stroke while breathing spontaneously with CPAP assistance. See text for explanation.

An intrinsic periodic breathing pattern can be precipitated by the way physicians set the mechanical ventilator. This is particulary seen in patients with central neurologic damage and some degree of altered level of consciousness or during sleep. Most mechanically ventilated patients are maintained at settings resulting in nearly "normal" PaCO2 which in turn maintain recruitment of the respiratory muscles. Following the transition to NREM sleep stage 1, the normal response s an incrase in the recruitment threshold of respiratory muscles for CO2. If the ventilator settings are unchanged during sleep time, it may be possible that PaCO2 now falls below the operating recruitment threshold for CO2 and apnea develop. If the patientis backup ventilator rate is low enough as it usual in stable ventilator dependent patients, anough as it is usual in stable ventilator dependt patients (e.g., SIMV 4 breaths per minute or using pressure support ventilation (PSV mode)) this prolonged central apnea can cause cortical arousal and therefore, sleep disruption. The easiest way to overcome this ventilator-triggered periodic breathing is to increase the ventilatoris frequency at night when using SIMV or A/C modes of ventilation and avoiding using PSV at night as the sole mode of ventilation.



The prevalence of deep venous thrombosis in patients with ischemic strokes is aroudn 42%. This number has been reduced to 26% with the use of low-dose unfractionated heparin and further down to 17% with prophylactic use of low-molecular-weight heparin (38). The efficacy of pneumatic leg compression and elastic stockings in ischemic stroke lis not well establish. But in neurosurgical patients pneumatic compression has shown to be effective. In high risk is contrindicated, an inferior vena cava filter (e.g., Greenfield filter) should be considered.

Data on the prevalence of pulmonary embolism (PE) after strokes is incomplete but it is estimated that 13% of all stroke patients or 22% of strokes patients with DVT will develp PE. In one series (39), half of the stroke patients who developed PE died from this complication. Moreover, when DVT was diagnosed, this was invariably present in the paralyzed leg.



  1. Grotta JC, Pasteur N, Khwaja G, Hamel T, Fisher M, Ramirez A. Elective intubation for neurologic deterioration after stroke. Neurology 1995; 45:640-4
  2. Wijdicks EF, Scott JP. Causes and outcome of mechanical ventilation in patients with hemispheric ischemic stroke. Mayo Clin Proc 1997; 72:210-213
  3. LeeNC, Klassen AC, Resch JA: Respiratory Pattern disturbances in ischemic cerebral vascular disease. Stroke 1974; 5:612-6
  4. Vingerhoets F, Bogousslavsky J. Respiratory dysfunction in stroke. In: Clinics in chest medicine 1994; 15:4
  5. Heyman A, Birchfield RI, Sieker HO. Effects of bilateral cerebral infarction on respiratory center sensitivity. Neurology 1958; 8:694
  6. Fluck DC: Chest movements in hemiplegia. Clin Sci 1966;31:383
  7. Keltz H, Kaplan S, Stone DJ: Effects of quadriplegia and hemidiaphragmatic paralysis on thoracoabdominal pressure during respiration. Am J Phys Med. 1969; 48:109
  8. Brown HW, Plum F. The neurologic basis of Cheyne –Stokes. Respiration. Amer J Med 1961;30:849-851
  9. Rout MW, Lane DJ, Wollner. Prognosis in acute CVA in realtion to respiratory pattern and blood gas tensions. Brit Med J 1971;3:7
  10. Raudem E, Zupping, Kaasik AE. Disturbances of gas metabolism in patients with occlusive cerebrovaslar disease. Prog Brain Res 1968; 30:3113-2
  11. Plum F. Mechanism of central hyperventilation (letter). Ann Neurol 1982; 39:636
  12. Plum F. Hiperpnea, hyperventilation (letter). An Int Med 1972; 76:328-331
  13. Plum F. Apneustic breathing in main. Arch Neurol 1964; 10:115-6
  14. Krieger AJ, Rosomoff HL. Sleep induced apnea: A respiratory and autonomic dysfunction syndrome followin bilateral cervical cordotomy. J Neurosurg 1974; 40:168-171
  15. Askenasy JJM, Goldhammer I. Sleep apnea as a feature or bulbar stroke. Stroke 1988; 19:637-640
  16. Caplan LR, Pessin BM, Yarnell P. Poor outcome after lateral medullary infarcts. Neurology 1986; 36:1510-14
  17. Feldman MH. Physiological observations in a chronic case of "Locked-in" syndrome. Neurology 1971; 21:459-462
  18. Plum F, Posner JB. The diagnosis of stupor and coma. Philadelphia: FA Davis, 1982; 87-90
  19. Reinoso MA. Pulmonary Disorders. In Practical guide to the Care of the Critically ill Patient. St. Louis: Mosby-YearBook 1994 (269-291)
  20. Oppenheimer S, Hachinski V: Complications of acute stroke. Lancet 1992; 339:721
  21. Demling R, Riessen R: Pulmonary dysfunction after cerebral injury. Crit Care Med. 1990; 18:768
  22. Maron M, Dawson C: Pulmonary venoconstriction caused by elevated cerebrospinal fluid pressure in the dog. J Appl Physiol 1980; 49:73
  23. Dempsey R, Kumada M, Reis D: Central neural mechanism of the cerebral ischemic response. Cir Res 1979; 44:48
  24. American College of Chest Physicians consensus conference onmechanical ventilation. Chest 1993; 104:1833-1859
  25. Apuzzo NLJ, Weiss MH, Petersons V, et al. Effect of PEEP on intracranial pressure inman. J Neurosurg 1977; 46:227-233
  26. Aidinis SJ, Lafferty J, Shapiro HM. Intracranial responses to PEEP. Anesthesia 1976; 45:275-286
  27. Tuxen DV Permissive Hypercapnic Ventilation. Am J Respir Crit Care Med. 1994; 150:870-874
  28. Veis SL, Logemann JA. Swalloving disorders in persons in persons with cerebravascular
  29. Horner J, Massey EW, Brazer S.Aspiration in bilateral stroke patients. Neurology 1990; 4G: 1686-89
  30. Horner J, Massey EW. Silent aspiration following stroke. Neurology 1998; 38:317-320
  31. Meadows JC. Dysphagia in unilateral cerebral lesions. J Neurol Neurosurg psychiatry 1973; 26:8 53-5
  32. EI-M. B, Bornstein NM, Futs P, Korczyn AD. Mechanical ventilation in stroke patients-is it worthwhile?. Neurology 1996; 47:657-9
  33. Wijdicks EF, Scott JP. Outcome in patients with acute basilar artery occlusion requiring mechanical ventilation. Stroke 1996; 27:1301-1303
  34. Hacke W. Neurocritical care. Berlin: Springer-Verlag 1994:139-156
  35. Borel CO, Guy J. Ventilatory management in critical neurologic illiness. Neurologic Clinics 1995; 13:3, 627-631
  36. Hund E, Grau A, Hacke W. Neurocritical care for acute ischemic stroke. Neurologic Clinics 1995; 13:3,5I1-9
  37. Kaplan JD, Schuster DP. Physiologic consequences of tracheal intubation. Clinics Chest Med. 1991; 12(3):425-32
  38. American College of Chest Physicians Consensus conference on Antihrombotic Therapy. Chest 1995; 108(Suppl): 312S-334S
  39. Wijdicks EFM, Scott JP. Pulmonary embolism Associated with Acute Stroke. Mayo Clin Proc 1997; 72:301-308



   Haga click Aquí o en el logo para volver a medicosecuador.com


   HAGA CLICK AQUI PARA VOLVER A medicosecuador.com