Cardiac autonomic control in obstructive sleep apnea: effects of long-term CPAP therapy.

To determine how long-term treatment with continuous positive airway pressure (CPAP) affects cardiac autonomic function, we measured R-R interval (RRI), respiration, and blood pressure in 13 awake patients with moderate-to-severe obstructive sleep apnea (OSA) in both supine and standing postures, before and after 3 to 9 mo of home therapy. Using visual feedback, the subjects controlled their respiration to track a randomized breathing pattern. From the RRI spectrum, we computed high-frequency power and the ratio of low-frequency to high-frequency power (LHR). To correct for differences in breathing, the average transfer gain relating respiration to RRI changes (G(RSA)) and the modified low-frequency to high-frequency ratio (MLHR) were also derived. CPAP therapy did not change the conventional spectral indices of heart rate variability (HRV). However, G(RSA) increased with average nightly CPAP use in supine (p < 0.01) and standing (p < 0.03) postures, whereas MLHR decreased with CPAP compliance during standing (p < 0.03). Supine mean heart rate decreased with compliance (p < 0.03). None of the estimated parameters was correlated with duration of therapy when actual CPAP use was not taken into account. These results suggest that CPAP treatment improves vagal heart rate control in patients with OSA and that the degree of improvement varies directly with compliance level.

Patient-ventilator asynchrony.

The basic mechanism of patient-ventilator asynchrony is the mismatching between neural inspiratory and mechanical inspiratory time. Alterations in respiratory drive, timing, respiratory muscle pressure, and respiratory system mechanics influence the interaction between the patient and the ventilator. None of the currently available partial ventilatory support modes are exempt from problems with patient-ventilator asynchrony. Ventilator triggering design in the trigger phase and the set variables in the post-trigger phase contribute to patient-ventilator interaction. The set inspiratory flow rate in the post-trigger phase for assist-control volume cycled ventilation affects patient-ventilator asynchrony. Likewise, the initial pressure rise time, the pressure support level, and the flow-threshold for cycling off inspiration for pressure support ventilation are important factors affecting patient-ventilator asynchrony. Current investigations have advanced our understanding in this area; however, its prevalence and the extent to which patient-ventilator asynchrony affect the duration of mechanical ventilation remain unclear.

Respiratory alkalosis.

Respiratory alkalosis is an extremely common and complicated problem affecting virtually every organ system in the body. This article reviews the various facets of this interesting problem. Respiratory alkalosis produces multiple metabolic abnormalities, from changes in potassium, phosphate, and calcium, to the development of a mild lactic acidosis. Renal handling of the above ions is also affected. The etiologies may be related to pulmonary or extrapulmonary disorders. Hyperventilation syndrome is a common etiology of respiratory alkalosis in the emergency department setting and is a diagnosis by exclusion. There are many cardiac effects of respiratory alkalosis, such as tachycardia, ventricular and atrial arrhythmias, and ischemic and nonischemic chest pain. In the lungs, vasodilation occurs, and in the gastrointestinal system there are changes in perfusion, motility, and electrolyte handling. Therapeutically, respiratory alkalosis is used for treatment of elevated intracranial pressure. Correction of a respiratory alkalosis is best performed by correcting the underlying etiology.

Effects of sodium bicarbonate administration on the exercise tolerance of normal subjects breathing through dead space.

STUDY OBJECTIVE: The purpose of this study was to determine whether the administration of sodium bicarbonate to normal individuals would increase their PaCO2 and thereby decrease the ventilatory requirements at a given workload. DESIGN: In this double-blind crossover study, six normal men ingested either 3 mEq/kg NaHCO3 or 1 mEq/kg NaCl once a day for 5 days, in addition to 40 mg of furosemide and 40 mEq KCl. After each 5-day treatment, the subjects underwent a symptom-limited maximal bicycle ergometer exercise test while breathing through external dead space (with a volume of approximately 50% of their FEV1), a second exercise test without any external dead space, and an assessment of their respiratory response to hypercapnia. RESULTS: The administration of the NaHCO3 resulted in a significant increase in the arterial HCO3- from 20.8 to 24.0 mEq/L and a significant increase in the PaCO2 from 31.7 to 36.9 mm Hg at rest that persisted during exercise. During exercise periods with the added dead space, the Borg scores were significantly lower at each workload after the subjects received bicarbonate, but the maximal exercise level did not increase. The mean (+/-SD) slope of the mouth occlusion pressure response to hypercapnia was significantly lower after the administration of NaHCO3 than after NaCl, respectively: 0.73+/-0.41 vs 1.27+/-0.97 cm H2O/mm Hg. CONCLUSION: From this study we conclude that the administration of NaHCO3 results in a significant increase in the PaCO2, decreases the ventilation and the Borg score at equivalent workloads, and decreases the hypercapnic response in normal individuals.

Hemothorax and chylothorax.

Hemothorax and chylothorax remain perplexing medical problems. The primary cause of hemothorax is trauma, whereas the primary cause of chylothorax is cancer. Most patients with hemothorax can be treated with chest tube drainage only. Early thoracotomy with thoracic duct ligation is recommended for patients with chylothorax when conservative treatment with chest tube drainage and hyperalimentation fails. Radiation therapy is the mainstay of treatment for chylothorax related to cancer. Video-assisted thoracoscopy may play an increasing role in the surgical treatment of both hemothorax and chylothorax.

O2-induced change in ventilation and ventilatory drive in COPD.

We examined the role of respiratory control during O2-induced hypercarbia in patients with chronic obstructive pulmonary disease (COPD), by comparing the observed change in ventilation (delta VEobs) with the delta VE predicted (delta VEpred) from the patients' ventilatory drive and the O2-induced delta PaCO2 and delta SaO2. Eleven stable hypoxemic COPD patients (mean +/- SD: FEV1 = 1.00 +/- 0.25 L, FVC = 2.33 +/- 0.38 L; room air PaCO2 = 52.7 +/- 7.9 mm Hg, SaO2 87.7 +/- 5.1%) were studied. Using standard rebreathing methods, we measured the ventilatory responses to hypercapnia (delta VE/PCO2 = 0.76 +/- 0.55 L/min/mm Hg) and to hypoxia (delta VE/delta SaO2 = -0.74 +/- 0.31 L/min/%). After breathing 100% O2 for 15 min, the mean delta VEobs was -0.08 +/- 0.62 (SEM) L/min (p = NS), the delta SaO2 was 7.6 +/- 3.6% (p < 0.001), and the delta PaCO2 was 6.6 +/- 3.3 mm Hg (p < 0.001). The delta VEpred was expressed as the sum of a decrease in ventilation due to suppression of hypoxic drive [calculated as the product (delta VE/SaO2) x delta SaO2] and an increase in ventilation due to the O2-induced hypercarbia [calculated as the production (delta VE/delta PCO2) x delta PaCO2]. The mean delta VEpred [-0.96 +/- 0.68 (SEM)] did not differ significantly from mean delta VEobs. We conclude that the O2-induced delta VEobs is equal to that expected from the ventilatory drives and the changes in PaCO2 and SaO2; and that O2-induced hypercarbia does not indicate a failure of respiratory control mechanisms in the maintenance of PaCO2 homeostasis.

Breathing pattern during acute respiratory failure and recovery.

The objective of this study was to compare the breathing pattern of patients who failed to wean from mechanical ventilation to the pattern during acute respiratory failure. We hypothesized that a similar breathing pattern occurs under both conditions. Breathing pattern, mouth occlusion pressure (P[0.1]) and maximum inspiratory pressure (P[I,max]) were measured in 15 patients during acute respiratory failure, within 24 h of the institution of mechanical ventilation, and in 49 patients during recovery, when they were ready for discontinuation from mechanical ventilation. The following indices were calculated: rapid shallow breathing index (respiratory frequency/tidal volume (fR/VT)); rapid shallow breathing-occlusion pressure index (ROP = P[0.1 x fR/VT]); P(0.1)/P(I,max); and effective inspiratory impedance (P[0.1]/VT/(inspiratory time (tI)). Patients who failed to wean (n=11) had a similar ROP,fR/VT and P(0.1)/P(I,max) to those with acute respiratory failure despite a significantly reduced P(0.1)/VT/tI, the value of which was comparable to that of patients who weaned successfully (n=38). The P(I,max) of patients who failed to wean was similar to that of patients who weaned successfully. We conclude that patients who failed to wean had a breathing pattern similar to that during acute respiratory failure, despite a reduced mechanical load on the respiratory muscles and a relatively adequate inspiratory muscle strength. This suggests that strategies that enhance respiratory muscle endurance may facilitate weaning.

Patient-ventilator interactions.

Patient-ventilator synchrony is important in the management of the ventilator-dependent patient. Factors inherent to the patient and the ventilator influence patient-ventilator synchrony. Detection of patient-ventilator synchrony may require monitoring of airway pressure and flow waveforms.

Temporal relationships of ventilatory failure, pump failure, and diaphragm fatigue.

The time course of ventilatory failure, pump failure, and diaphragm peripheral fatigue was determined during the application of external inspiratory resistive loads (IRL) in anesthetized rabbits. Pump failure is defined as the inability of the diaphragm to sustain the expected force under IRL. To assess contractile fatigue, transdiaphragmatic pressures (Pdi) generated by bilateral phrenic nerve stimulation at 75 Hz (Pdi-75) and 20 Hz (Pdi-20) were measured. The amplitude of evoked diaphragm electromyographic (EMG) signals was measured to assess neurotransmission failure. The rate of rise of spontaneous diaphragm EMG was used as an index of respiratory drive. Ventilation was evaluated together with arterial blood gases. During IRL the rate of rise of spontaneous diaphragm EMG increased, and there was a progressive hypercapnic acidosis and hypoxemia, indicating ventilatory failure. In contrast, Pdi-75 and Pdi-20 were stable until the time of respiratory arrest (apnea), when they decreased by 34 and 45%, respectively. The amplitude of evoked diaphragm EMG signals remained unchanged throughout the IRL and decreased only slightly at the time of apnea. We conclude that IRL induces progressive ventilatory failure long before any contractile fatigue of the diaphragm or pump failure occurs. This suggests that ventilatory failure is due to central fatigue, whereas pump failure (apnea) is attributable to multiple factors.

Measurement of lung mechanics at different lung volumes and esophageal levels in normal subjects: effect of posture change.

Lung elastance and resistance increase in the supine posture. To evaluate the effects of change in posture on regional lung mechanics at different lung volumes, lung elastance and resistance were measured at graded volume subdivisions and three esophageal levels at seated and supine body positions, using the esophageal balloon technique. Volumes were adjusted to be the same in both postures. In general, lung elastance (both static and dynamic) tended to be higher in supine posture and uniform at all lung volumes, except at 80% vital capacity, where it increased sharply. The ratio of dynamic to static lung elastance was slightly higher at the cephalad esophageal level, where regional flow rates and relative volume expansion are lower. Lung resistance varied inversely with lung volume but was higher at corresponding volume subdivisions in the supine posture. It decreased at more cephalad esophageal levels, where volume expansion and flow are less. Thus, the increase in regional flow at low volume subdivisions (most marked in the supine position) also contributed to higher lung resistance at these volumes. These findings are explained on the basis of a combination of Newtonian physics as well as nonlinear viscoelastic properties of the lung as applied to regional flow and volume expansion.

Comparison of mitoxantrone and tetracycline as pleural sclerosing agents in rabbits.

Bleomycin is the antineoplastic agent used most commonly for the treatment of malignant pleural effusion. It is absorbed rapidly from the pleural space and does not elicit pleurodesis in the normal rabbit pleura. Mitoxantrone is a new antineoplastic that differs from bleomycin in that it binds to membranes. Accordingly it might remain in the pleural space for a longer period and produce a pleurodesis. The objective of this project was to determine whether mitoxantrone is an effective sclerosant in an experimental model in rabbits. The following medications were instilled intrapleurally in anesthetized male rabbits: 35 mg/kg tetracycline or 0.5, 1.0, or 2.0 mg/kg mitoxantrone. The animals were killed at 28 days and the pleural spaces assessed grossly for pleurodesis and microscopically for fibrosis and inflammation. The mean degree of gross pleurodesis did not differ significantly in the rabbits that received tetracycline (3.8 +/- 0.4) and in the rabbits that received 2 mg/kg mitoxantrone (3.2 +/- 1.3). The degree of pleural and lung inflammation was significantly greater after mitoxantrone than after tetracycline, both ipsilaterally and contralaterally. The mortality after the highest dose of mitoxantrone was 50%. From this study we conclude that the intrapleural administration of mitoxantrone in rabbits can produce a pleurodesis. The histologic picture after mitoxantrone administration differs markedly from that after tetracycline injection. After mitoxantrone injection there are many more inflammatory cells present on the side that received the injection, and there is much more fibrosis and inflammation in the contralateral pleura and lung. The model of pleural fibrosis following intrapleural mitoxantrone may be useful for the study of pleural fibrosis.

The etiology and treatment of spontaneous pneumothorax.

In the past year, studies on spontaneous pneumothorax have focused on etiology and treatment. Chronic obstructive pulmonary disease remains the most common cause of secondary spontaneous pneumothorax. However, Pneumocystis carinii infections in patients who have AIDS have become the leading cause of spontaneous pneumothorax in a population where its prevalence is high. One of the treatment modalities of spontaneous pneumothorax is tube thoracostomy with the instillation of tetracycline as the sclerosing agent. Tetracycline is no longer available. Fortunately, its derivatives doxycycline and minocycline are equally effective. Talc in slurry or insufflated appears to be more effective than tetracycline derivatives. Experience with talc in slurry for the treatment of spontaneous pneumothorax is still limited. Another treatment modality for spontaneous pneumothorax is thoracoscopy, more recently termed video-assisted thoracic surgery, and it has warranted renewed interest due to the advent of improved endoscopic technology. In the treatment of spontaneous pneumothorax, video-assisted thoracic surgery is nearly as effective as thoracotomy.

Talc slurry is an effective pleural sclerosant in rabbits.

Insufflated talc is probably the most effective agent for creating a pleurodesis both in the clinical situation and in animals. However, the insufflation of talc requires an invasive procedure such as thoracoscopy or thoracotomy. Recently, there have been reports that talc in a slurry was effective in the clinical situation. The objective of this project was to determine whether talc in a slurry at varying doses is an effective sclerosant in an experimental model in rabbits. Talc, 50, 100, 200, and 400 mg/kg, in a 2-mL slurry was injected intrapleurally through a small catheter in male rabbits. Eleven rabbits received each dose. Twenty-eight days after the instillation, the animals were killed. The pleural spaces were assessed grossly for evidence of pleurodesis and microscopically for evidence of fibrosis and inflammation. The degree of pleurodesis (on a scale of 0 to 4) after the injection of 50, 100, 200, and 400 mg/kg of talc was 1.1 +/- 0.9, 1.5 +/- 1.1, 2.7 +/- 0.6, and 3.4 +/- 0.5, respectively. The degree of microscopic fibrosis similarly increased with increasing doses of talc. These scores were similar to those we have reported with the tetracycline derivatives. In contrast to the results with tetracycline derivatives, none of the rabbits developed fibrothorax or hemothorax. From this study, we conclude that talc in a slurry is a very effective pleural sclerosant in rabbits and does not produce hemothoraces as do the tetracycline derivatives.

Temporal evolution of pleural fibrosis induced by intrapleural minocycline injection.

Minocycline is as effective as tetracycline in inducing pleural fibrosis, but the long-term pleural changes induced by minocycline are unknown. The objective of this study was to evaluate in rabbits the evolution of the pleural changes induced by the intrapleural instillation of minocycline. Under light anesthesia, minocycline at 10 mg/kg in a total volume of 2 ml of bacteriostatic saline solution was injected into the right pleural space of 25 male rabbits. The animals were sacrificed in groups of five at 15 d and 1, 2, 4, and 6 mo. Macroscopic and microscopic examinations of the pleura were performed for evidence of pleural fibrosis and inflammation. Similarly, the underlying lung was also examined for microscopic alveolar fibrosis and inflammation. During the 6-mo observation period there was no significant change in the degree of pleural fibrosis. In contrast, microscopic pleural inflammation, alveolar fibrosis, and alveolar inflammation all decreased significantly over the observation period. In conclusion, intrapleural minocycline injection results in persistent pleural fibrosis at 6 mo. It remains to be determined whether the pleural fibrosis will persist beyond this period.

Characteristics of the ventilator pressure- and flow-trigger variables.

Pressure- and flow-triggering are available in the Puritan Bennett 7200ae and Siemens SV 300. Using a mechanical lung model, we described the characteristics of the pressure- and flow-triggered continuous positive airway pressure (CPAP) of both ventilators. In the Puritan Bennett 7200ae, the pressure-triggered CPAP is characterized by the relatively insufficient flow delivery after the triggering, resulting in a greater lung pressure-time product (total PTP) than the flow-triggered CPAP. Pressure support of 5 cmH2O results in total PTP less than that with flow-triggered CPAP. In the Siemens SV 300, total PTP with pressure- or flow-triggered CPAP is comparable. Total PTP is less with pressure- or flow-triggered CPAP of the Siemens SV 300 than that of the Puritan Bennett 7200ae, respectively. The application of small pressure- or flow-triggered pressure support in the Puritan Bennett 7200ae eliminates the difference. The impact of these differences on patient inspiratory muscle work remains to be determined.

Influence of pressure- and flow-triggered synchronous intermittent mandatory ventilation on inspiratory muscle work.

OBJECTIVE: To determine the effect of pressure- and flow-triggered synchronous intermittent mandatory ventilation on inspiratory muscle work. DESIGN: Consecutive clinical, prospective, randomized trial. SETTING: Medical intensive care unit (ICU) of a U.S. Veterans Affairs Medical Center. PATIENTS: Eight patients recovering from acute respiratory failure of various etiologies. INTERVENTIONS: Assist control, followed by randomized application of pressure- and flow-triggered synchronous intermittent mandatory ventilation at 60%, 40%, 20% of the assist-control rate, and flow-triggered continuous positive airway pressure. A total of eight settings were maintained for 10 mins each. MEASUREMENTS AND MAIN RESULTS: Total work rate (joules/min), inspiratory muscle work (joules/L), and pressure time-product per breath (cm H2O-sec) were measured. During pressure- or flow-triggered synchronous intermittent mandatory ventilation, total work rate increased as the mandatory rate was decreased. The method of ventilator triggering had a significant effect on the total work rate. With pressure-triggered synchronous intermittent mandatory ventilation, the total work rate at 60% of the assist-control rate was similar to that with assist-control; whereas with flow-triggered synchronous intermittent mandatory ventilation, this result was achieved at 40% of the assist-control rate. At a machine support level of 20%, total work rate with pressure-triggered synchronous intermittent mandatory ventilation was significantly greater than with flow-triggered synchronous intermittent mandatory ventilation. The method of ventilator triggering had no significant effect on the inspiratory muscle work of the mandatory breaths. This finding was in contrast to the effect on inspiratory muscle work of spontaneous breaths. With pressure-triggered synchronous intermittent mandatory ventilation, inspiratory muscle work of the spontaneous breaths was greater than with the flow-triggered at machine support of 40% and 20%. With either pressure- or flow-triggered synchronous intermittent mandatory ventilation, inspiratory muscle work of the mandatory breaths was not significantly different from that of the corresponding spontaneous breaths, except at the lower machine support levels with the pressure-triggered synchronous intermittent mandatory ventilation. Pressure-time product followed a trend similar to that of inspiratory muscle work. CONCLUSIONS: During synchronous intermittent mandatory ventilation, the method of ventilator triggering has a significant effect on the total work rate and inspiratory muscle work of the spontaneous breaths, particularly at lower machine support levels. Conversely, the method of ventilator triggering has no significant effect on inspiratory muscle work of the mandatory breaths.

Comparison of the effectiveness of tetracycline and minocycline as pleural sclerosing agents in rabbits.

Parenteral tetracycline, one of the most commonly used agents for producing pleurodesis, is no longer available because of stricter regulations governing the manufacturing process. The objective of this project was to determine whether minocycline, a tetracycline derivative, is an effective sclerosant in an experimental model in rabbits. We also studied the relationship of the dose and the volume injected to the degree of pleurodesis. The following medications were instilled intrapleurally in anesthetized male rabbits: tetracycline, 35 mg/kg; or minocycline, 4, 7, 10, or 20 mg/kg, diluted to a total volume of 1 or 2 ml of bacteriostatic saline solution; or minocycline, 40 mg/kg, diluted to a total volume of 2 ml of the solution. Twenty-eight days after the instillation, the animals were killed. The pleural spaces were assessed grossly for evidence of pleurodesis and microscopically for evidence of fibrosis and inflammation. The degree of pleurodesis grossly and microscopically after the injection of 7, 10, 20, or 40 mg/kg of minocycline was comparable to that after the injection of 35 mg/kg of tetracycline, while the dose of 4 mg/kg was less effective. In the animals who received the higher doses of minocycline intrapleurally (ie > or = 20 mg/kg), there was an excess mortality both early (chi 2 = 3.61, 0.05 < p < 0.10) and late (chi 2 = 11.0, p < 0.005) which appeared to be related to the development of hemothorax. The intrapleural injection of the tetracycline derivatives was significantly (p < 0.05) more effective when the total volume of the solution was 2 ml rather than 1 ml. The present study demonstrates that minocycline is an effective agent for producing pleurodesis in the rabbit. Minocycline given intrapleurally at doses of 7 mg/kg or above is comparable to tetracycline, 35 mg/kg. Higher doses of minocycline (> or = 20 mg/kg) produce a high mortality that seems to be related to hemothorax. Since, in humans, a large experience confirms only 20 mg/kg of tetracycline is needed to produce adequate pleurodesis safely, we recommend a dose of 4 mg/kg of minocycline for the production of pleurodesis.

Significance of iatrogenic pneumothoraces.

The purpose of this study was to review the cases of iatrogenic pneumothorax that occurred between October 1983 and December 1988 at the Veterans Administration Medical Center, Long Beach, Calif, to determine the treatment and complications. During this time period, 106 patients were identified with iatrogenic pneumothorax, and the charts of 98 were available for review. There were 90 cases of spontaneous pneumothorax at this institution during the same time period. The most common cause of iatrogenic pneumothorax was transthoracic needle aspiration (35), followed by thoracentesis (30), subclavian venipuncture (23), and positive pressure ventilation (7). In 11 cases, the cause was due to miscellaneous triggers. The majority of the patients (65 of 98) were treated with chest tubes. The chest tubes were in place 4.7 +/- 3.9 days. Nine of the patients required a second chest tube. Aspiration of the pneumothorax only was attempted in five patients, and all patients subsequently received a chest tube. Two patients died from iatrogenic pneumothorax. One patient receiving positive pressure ventilation developed an unrecognized tension pneumothorax. The other patient developed a pneumothorax after thoracentesis and was treated with a chest tube, which led to a staphylococcal empyema and death. From this study, we conclude that the incidence of iatrogenic pneumothorax exceeds that of spontaneous pneumothorax and that there is substantial morbidity and some mortality from iatrogenic pneumothorax.

Variability of cardiac output over time in medical intensive care unit patients.

OBJECTIVES: To determine the amount of spontaneous variability of cardiac output over time in critically ill patients, and to determine the effect of mechanical ventilation on cardiac output variability over time. DESIGN: Case series. SETTING: Medical intensive care unit in a Veterans Affairs Medical Center. PATIENTS: Twenty-two patients with indwelling pulmonary artery flotation catheters were studied. Two patients were studied twice. INTERVENTIONS: During a 1-hr time period in which no interventions were required or made, thermodilution cardiac output was determined at baseline and then every 15 mins for 1 hr. At each time point, five individual cardiac output measurements were made and a mean was computed. The covariables of heart rate, respiration rate, mean arterial pressure, mean pulmonary arterial pressure, pulmonary artery occlusion pressure, and temperature were also recorded at each time point. MEASUREMENTS AND MAIN RESULTS: The variability of the five cardiac output measurements made at each time point was expressed by calculating for each patient a coefficient of variation of the measurements. The overall mean coefficient of variation of the measurements was 5.8%. The variability of the cardiac output measurements over time was expressed by calculating for each patient a coefficient of variation over time. The overall mean coefficient of variation over time was 7.7%. A subgroup of 15 "covariable stable" patients (defined as those patients with covariables within +/- 5% of the mean covariable values during the hour) had a mean coefficient of variation over time of 6.4%, whereas "covariable unstable" patients (with > +/- 5% changes in any covariable) had a mean coefficient of variation over time of 9.9% (p < .05). Patients breathing spontaneously had a mean coefficient of variation over time of 10.1%, whereas mechanically ventilated patients had a mean coefficient of variation over time of 6.3% (p < .05). CONCLUSIONS: The spontaneous variability of cardiac output should be considered when interpreting two cardiac output determinations made at separate times. Due to spontaneous variability alone, a patient with a baseline cardiac output of 10.0 L/min would be expected (95% confidence interval) to have a cardiac output range of 9.2 to 10.8 L/min if covariables were stable, and a range of at least 8.8 to 11.2 L/min if covariables were unstable. Patients who were mechanically ventilated displayed less variability than patients who were breathing spontaneously.