Fever, shivering, pain, agitation, increased work of breathing, sepsis, and overfeeding are common causes of increased CO₂ production in the intensive care unit (ICU). In the weaning phase, cautious anxiolysis and pain relief can dramatically reduce the ventilatory requirement. Carbon dioxide production is also influenced by underlying nutritional status, as well as by the number and composition of the calories administered. The semistarvation that often precedes critical illness suppresses CO₂ production. Despite the importance
of adequate nutrition, patients should not be overfed. Excess calories may be converted to fat, generating CO₂ as a metabolic by-product unlinked to energy production. Carbohydrate evolves more CO₂ per calorie than fat or protein. However, even though large calorie loads can contribute to ventilatory failure, the importance of calorie composition to ventilator dependence remains to be shown convincingly. Overfeeding also may lead to abdominal distention and discomfort that may adversely impact a patient poised at the boundary of ventilatory failure.

Alveolar ventilation (V_A), the component of ventilation that is effective in eliminating carbon dioxide, is the total minute ventilation adjusted for the fraction of wasted ventilation. Breathing efficiency can be characterized by the following expression:

where V_D/V_T is the physiologic dead space fraction (see Chapter 5). Virtually, all the diverse processes that damage the lung or airways of the critically ill patient increase the wasted fraction of ventilation. Certain reversible factors unrelated to underlying lung pathology also can prove to be important. For example, thromboembolic or vasculitic pulmonary arterial occlusion, excessive PEEP, and hypovolemia may reduce perfusion to the ventilated lung, expanding the alveolar dead space. Small tidal volumes are characterized by a high anatomic dead space percentage. By adding “apparatus dead space,” disposable heat and moisture exchangers and other devices or tubing interposed between the ET tube and the “Y” of the ventilator circuit may contribute marginally to ventilatory inefficiency.

Although it is unusual for a patient to remain persistently ventilator dependent solely because of a lack of breathing effort, inappropriately depressed or enhanced drive to breathe may limit weaning progress. As a rule, old and debilitated patients are the most susceptible to drive suppression. Suppressed ventilatory drive may be explained by neurologic impairment, hypothyroidism, lingering sedation (very commonly the predominant cause), sleep deprivation, poor nutrition, and metabolic alkalosis (primary or compensatory). Starvation impairs hypoxic and, to a lesser extent, hypercapnic sensitivity. Interestingly, drive is restored rapidly within a few days of reinitiating adequate feeding, perhaps in advance of improved muscle function. The drive to breathe is generally higher, and the breathing pattern is different in the waking state. Minute ventilation occasionally falls markedly during sleep or sedation and accelerates with the return to alertness. (This pattern often applies to the recovering drug overdose victim.) When infused benzodiazepines are given continuously for days to weeks, sufficient drug and drug metabolites may store in brain and fat tissues to suppress consciousness and/or drive for long periods after they are discontinued (occasionally as long as 7 to 10 days). Such unintended effects argue forcefully for the practice of once daily interruption of sedation to assess the need for and rate of their ongoing use. The administration of pharmacologic stimulants such as doxapram or progesterone has been advocated when depressed drive impedes weaning. These drugs, however, usually are not helpful.

Enhanced central drive arising from neurogenic, psychogenic, reflex, or metabolic stimuli augments ventilatory demand and workload. In asthma and acute pulmonary edema, drive-stimulating reflexes arising from the lung or chest wall downregulate after correction of the underlying disorders. Hypoxemia, hypotension, developing sepsis, and acidosis also accentuate ventilatory demands. Correction of metabolic acidosis is one of the most important ways to reduce central drive. It is also important not to force PaCO₂ below the patient’s usual resting value, as the ensuing bicarbonate diuresis redefines the [V with dot above]_E needed to maintain (cerebral pH) homeostasis. In fact, PaCO₂ somewhat higher than normal for that patient may help minimize the [V with dot above]_E requirement. Anxiety and disorientation influence ventilatory demand and often can be addressed successfully by counseling, co-opting the patient into the weaning plan, and cautious use of anxiolytics or major tranquilizers. Haloperidol (Haldol) is a good first option but often proves inadequate. Quetiapine (Seroquel), risperidone, olanzepine (Zyprexa), and occasionally Depakote may work for delirious or combative patients when more usual approaches fail. The ECG must be monitored, however, to avert the possibility of dangerous QT prolongation and arrhythmia induction. Dexmedetomidine (Precedex) is an infused sedative that does not markedly suppress consciousness or drive to breathe. Given the difficulty of removing the dissociative sedatives (e.g., lorazepam, midazolam) without precipitating delirium, such properties often prove useful in weaning applications. Supplementing the inspired O₂ fraction is often an effective way to reduce drive and interrupt a panic reaction accompanied by worsening hypoxemia.

Ventilatory power is the product of [V with dot above]_E and mechanical work of breathing per liter of ventilation. The quotient of this mechanical workload and neuromuscular efficiency defines how much energy must be expended in breathing. Once tidal volume and inspiratory time are set, the frictional and elastic properties of the respiratory system determine the pressure generated per breath, as well as the external work output per liter of ventilation. The average inspiratory pressure developed by the respiratory system per breath can be approximated by a simple formula:

P = R (V_T /t_i) + V_T/2C_RS) + auto-PEEP

where R and C_RS are the inspiratory resistance and compliance of the respiratory system, V_T is tidal volume, and t_i is the time required for inspiration (see Chapter 5 and Fig. 10-2). Therefore, for the same level of [V with dot above]_E, more external work must be done if C_RS falls or if V_T, auto-PEEP, mean inspiratory flow (V_T/t_i), or R increase. Bronchospasm, retained secretions, and mucosal edema are the primary reversible factors that increase R. Retained secretions greatly amplify the inspiratory workload in patients with already narrowed airways. Lung edema and infiltration, high lung volumes, pleural effusions, abdominal distention, and supine posture reduce C_RS. Air trapping and auto-PEEP are highly dynamic phenomena influenced powerfully by minute ventilation and the ease with which gas empties from the lungs (see Chapter 25). Auto-PEEP varies with changes in the breathing impedance (e.g., bronchospasm, secretion retention, minute ventilation, body position, alteration of inspiratory time fraction, and expiratory muscle activity).

These are not the only factors that determine the energy needed by the respiratory muscles, however. For the same tidal volume, the respiratory muscles consume more oxygen (become less effective) when they begin contraction from a mechanically disadvantageous high lung volume or when the pattern of muscle contraction is poorly coordinated. For example, patients with diaphragmatic weakness may allow the abdominal contents to be drawn upward with each inspiratory effort, so that much of the tension developed by the inspiratory muscles of the chest cage fails to translate into the negative pleural pressure that draws air into the lungs.

The external properties of the ventilator circuit also can play an important role in determining the ventilatory workload. ET tube resistance exceeds the normal resistance of the upper airway. Frictional pressure losses increase rapidly when high flows are driven through small-caliber tubes. Kinks and encrusted secretions may encroach on an otherwise adequate lumen. In some instances, it may be wise to exchange an orotracheal tube for a larger one (e.g., via a tube changer; see Chapter 6) or replace a nasotracheal tube with an orotracheal one. Tracheostomy is worth considering to reduce resistance and to facilitate extraction of airway secretions in particularly difficult patients. In theory, all pressure-support ventilation (PSV) and CPAP should be withdrawn before extubation to test the marginal patient’s ventilatory reserve (the “T-piece trial,” see following). Even in marginal candidates however, extubation from modest pressure support and CPAP often succeeds, especially when the patient bites the tube or experiences extreme discomfort or a narrow bore tube hinders airflow.

Tracheostomy offers larger tube diameter, shorter axial length, lower resistance and improved secretion clearance. The resistance of other circuit components varies widely and can add significantly to the ventilatory burden. Although the valves of older machines frequently presented problems, they are rarely limiting when using modern equipment.

Although it is unusual for a patient to remain persistently ventilator dependent solely because of a lack of breathing effort, multiple interacting factors can suppress the output of the ventilatory drive center. As a rule, old and debilitated patients are the most susceptible to drive suppression. Sedatives and neurologic impairment generally receive adequate clinical attention. However, other important causes of drive suppression may be potentially reversible. For example, chronic loading of ventilation (as during a severe bout of asthma) can condition the ventilatory center to tolerate hypercapnea. Metabolic alkalosis, hypothyroidism, and sleep deprivation are commonly overlooked causes of impaired ventilatory drive. Because the output of the ventilatory center tends to parallel metabolic rate and caloric intake, nutritional status is important. Starvation impairs hypoxic and, to a lesser extent, hypercapnic sensitivity. Interestingly, drive is restored rapidly within a few days of reinitiating adequate feeding. The drive to breathe is generally higher, and the breathing pattern is different in the waking state. Minute ventilation occasionally falls markedly during sleep or sedation and accelerates impressively with the return to alertness. (A florid example of this pattern is often provided by the recovering drug overdose victim.) The administration of pharmacologic drive stimulants such as doxapram or progesterone has been advocated when depressed drive impedes weaning. These drugs, however, are seldom helpful or used.