Vecuronium bromide

Study objective: This two-part study was designed to determine the effect of supplemental oxygen on the detection of hypoventilation, evidenced by a decline in oxygen saturation (Sp[O.sub.2]) with pulse oximetry.

Design: Phase 1 was a prospective, patient-controlled, clinical trial. Phase 2 was a prospective, randomized, clinical trial.

Setting: Phase 1 took place in the operating room. Phase 2 took place in the postanesthesia care unit (PACU).

Patients: In phase 1, 45 patients underwent abdominal, gynecologic, urologic, and lower-extremity vascular operations. In phase 2, 288 patients were recovering from anesthesia. Interventions: In phase 1, modeling of deliberate hypoventilation entailed decreasing by 50% the minute ventilation of patients receiving general anesthesia. Patients breathing a fraction of inspired oxygen (FI[O.sub.2]) of 0.21 (n = 25) underwent hypoventilation for up to 5 min. Patients with an FI[O.sub.2] of 0.25 (n = 10) or 0.30 (n = 10) underwent hypoventilation for 10 min. In phase 2, spontaneously breathing patients were randomized to breathe room air (n = 155) or to receive supplemental oxygen (n = 133) on arrival in the PACU.

Measurements and results: In phase 1, end-tidal carbon dioxide and Sp[O.sub.2] were measured during deliberate hypoventilation. A decrease in Sp[O.sub.2] occurred only in patients who breathed room air. No decline occurred in patients with FI[O.sub.2] levels of 0.25 and 0.30. In phase 2, Sp[O.sub.2] was recorded every min for up to 40 min in the PACU. Arterial desaturation (Sp[O.sub.2] < 90%) was fourfold higher in patients who breathed room air than in patients who breathed supplemental oxygen (9.0% vs 2.3%, p = 0.02).

Conclusion: Hypoventilation can be detected reliably by pulse oximetry only when patients breathe room air. In patients with spontaneous ventilation, supplemental oxygen often masked the ability to detect abnormalities in respiratory function in the PACU. Without the need for capnography and arterial blood gas analysis, pulse oximetry is a useful tool to assess ventilatory abnormalities, but only in the absence of supplemental inspired oxygen.

Key words: hypoventilation; pulse oximetry; supplemental oxygen

Abbreviations: FI[O.sub.2] = fraction of inspired oxygen; HR = heart rate; PAC[O.sub.2] = alveolar carbon dioxide tension; PACU = postanesthesia care unit; PA[O.sub.2] = alveolar oxygen tension; PETC[O.sub.2] = end-tidal carbon dioxide; RR = respiratory rate; Sp[O.sub.2] = oxygen saturation; VE = minute ventilation; V/Q = ventilation-perfusion; VT = tidal volume

**********

Although the physiologic consequences of moderate hypoventilation have not been clearly elucidated, profound hypoventilation with the development of carbon dioxide narcosis can cause coma, respiratory arrest, and circulatory failure. (1,2) Various studies (2-5) have reported the difficulty in detecting hypoventilation in patients undergoing sedation for GI, dental, and other endoscopic procedures. Moreover, several reports (6,7) have discussed the failure to diagnose severe hypoventilation in the perioperative period.

The early postoperative period may be associated with hypoventilation caused by respiratory depression and inability to maintain an adequate airway, (8,9) In addition, ventilation-perfusion (V/Q) mismatch also may occur secondary to atelectasis. Currently, accurate measurement of PaC[O.sub.2] or end-tidal carbon dioxide (PETC[O.sub.2]) to assess the adequacy of ventilation is not routine outside of the operating room environment, or in patients not intubated. (10) Although pulse oximetry is used widely to monitor arterial blood oxygenation, it is possible that pulse oximetry can be used to detect abnormalities in ventilation, by quantifying changes in oxygen saturation (Sp[O.sub.2]). (10,11) The objective of this study was to determine if pulse oximetry would indicate a decrease in ventilation, with and without administering supplemental oxygen.

MATERIALS AND METHODS

The Institutional Review Board of the University of South Florida College of Medicine, Tampa, approved the study protocol, and consent was obtained in patients scheduled to undergo surgical procedures.

Phase 1

Forty-five patients gave informed consent to undergo a trial of controlled deliberate hypoventilation receiving general anesthesia and mechanical ventilation. Patients underwent abdominal, gynecologic, urologic, and lower-extremity vascular operations. We used a pulse oximeter, capnograph, arterial catheter, and ECG to measure Sp[O.sub.2], PETC[O.sub.2], BP, heart rate (HR), and arterial blood gas data. General anesthesia was induced with thiopental, 3 to 5 mg/kg IV. IV administration of succinylcholine

chloride, 1 mg/kg, or vecuronium bromide, 0.1 mg/kg, facilitated tracheal intubation. Anesthesia was maintained with isoflurane using a semiclosed-circle absorber system. IV vecuronium bromide was administered to maintain muscle relaxation.

Patients received mechanical ventilation with a tidal volume (VT) of 8 mL/kg and a respiratory rate (RR) sufficient to produce an PETC[O.sub.2] of 30 to 40 mm Hg before initial data collection. Data collection started 10 min after administering the desired fraction of inspired oxygen (Floe). FI[O.sub.2] was determined by measurement of the inspired oxygen concentration on the oxygram. Three patient groups with FI[O.sub.2] levels of 0.21, 0.25, or 0.30, respectively, were studied. We recorded HR, BP, Sp[O.sub.2], PETC[O.sub.2], RR, and VT. Minute ventilation (VE) was determined as the product of RR and VT. Hypoventilation was instituted by reducing RR to decrease the VE by 50%. Patients breathing room air (FI[O.sub.2] of 0.21) underwent induced hypoventilation for up to 5 min, or until Sp[O.sub.2] was < 90%. Patients with FI[O.sub.2] levels of 0.25 or 0.30 underwent hypoventilation for up to 10 min.

Intraoperative data were collected each minute until the end of the hypoventilation trial, or until Sp[O.sub.2] fell below 90%, whichever occurred first. Arterial blood was sampled to measure pH, Pa[O.sub.2], and PaC[O.sub.2] before hypoventilation and during final data collection. Intraoperative data were summarized as mean [+ or -] SD and evaluated with Student t test for paired observations, or with analysis of variance for repeated measurements and Scheffe test, when appropriate. A Pearson [chi square] test with Yates correction for continuity was used to determine probability under the null hypothesis of increase or decrease from initial to final value of Sp[O.sub.2], arterial pH, PaC[O.sub.2], and PETC[O.sub.2].

Phase 2

Patients gave informed consent to be randomized to breathe room air, or to receive supplemental oxygen on arrival to the postanesthesia care unit (PACU). All surgical patients except those undergoing thoracotomy were eligible for participation. Baseline Sp[O.sub.2] measurements were recorded preoperatively in all patients breathing room air, prior to entering the operating room. Patients who received general anesthesia, regional anesthesia, or monitored anesthesia care were included in the study. Patients who received general anesthesia and who were not extubated in the operating room were excluded from the study. On leaving the operating room, patients with Sp[O.sub.2] [greater than or equal to] 90% breathing room air were transported to the recovery room without supplemental oxygen. On arrival in the PACU, while breathing room air, patients with Sp[O.sub.2] [greater than or equal to] 90% were randomized to continue to breathe room air, or to receive 30% oxygen with a facemask. Patients who experienced Sp[O.sub.2] < 90% before or on arrival to the PACU were administered supplemental oxygen and discontinued from the study.

In the PACU, Sp[O.sub.2] measurements were recorded every minute. Patients who experienced Sp[O.sub.2] < 90% for 2 consecutive min in the PACU received a "stir-up" regimen consisting of verbal and tactile stimulation. Patients in the supplemental oxygen group received a stir-up regimen, in addition to receiving oxygen by facemask. Patients in the room-air group initially received a stir-up regimen, without oxygen administration. If the Sp[O.sub.2] did not increase to [greater than or equal to] 90% within 2 min, patients then were administered supplemental oxygen with a facemask. Patients who initially breathed room air and subsequently received oxygen were classified as room-air dropouts. However, data collected until the time of dropout were recorded. Logistic regression analysis was used to study the effects of age, gender, and weight. Intergroup comparisons of the incidence of desaturation, use of intraoperative narcotics and muscle relaxants, and use of narcotics in the PACU were made with a Pearson [chi square] test.

RESULTS

Phase 1

There were no intergroup differences in age or weight (Table 1). There were no differences between the VT measured at the start of the hypoventilation trial (initial) and the VT measured at the end of the trial (final). The final RR and VE were approximately 50% of initial values. Initial and final arterial blood analysis data collected during induced hypoventilation are shown in Tables 2-4. Every patient had an increase in PaC[O.sub.2] and PETC[O.sub.2] and a decrease in arterial pH and Pa[O.sub.2] (p < 0.001). Patients with an FI[O.sub.2] of 0.21 had a significant decrease in Sp[O.sub.2] when initial and final Sp[O.sub.2] values were compared (97 [+ or -] 2% vs 91 [+ or -] 3%, p < 0.001). All patients with an FI[O.sub.2] of 0.21 had an immediate decrease in mean Sp[O.sub.2] and increase in mean PETC[O.sub.2] during hypoventilation (p < 0.001) [Fig 1]. Over half of these patients had Sp[O.sub.2] < 90% within 5 rain of hypoventilation, which accounts for the variable number of patients after 3 min of hypoventilation. Nine of 10 patients with an FI[O.sub.2] of 0.25 maintained Sp[O.sub.2] > 90% throughout the 10-min study period. Every patient with an FI[O.sub.2] of 0.30 maintained Sp[O.sub.2] > 90% throughout the study. Changes in Sp[O.sub.2] during hypoventilation in patients with FI[O.sub.2] levels of 0.25 or 0.30 were insignificant (Fig 1).

[FIGURE 1 OMITTED]

Phase 2

Three hundred eleven surgical patients consented to participate in this phase of the study. Six patients had cancellation of the operation, or were transported to the ICU postoperatively and were not included. Another 16 patients were not able to participate in the study because they remained intubated (n = 8) or received supplemental oxygen (n = 8) on arrival to the PACU. The thoracic surgeon would not permit inclusion of his patients in the study. Of the 289 eligible patients, 155 patients were randomized to breathe room air and 134 patients were randomized to receive supplemental oxygen. One patient who was randomized to receive supplemental oxygen was dropped from the study due to noncompliance and refusal to wear the oxygen mask.

There were no intergroup differences in age, gender, use of intraoperative narcotics or muscle relaxants, and the use of narcotics in the PACU (Tables 5, 6). There was an intergroup difference in weight. There were no intergroup differences in preoperative Sp[O.sub.2] (98 [+ or -] 2%) or in the Sp[O.sub.2] measurement on arrival to the PACU (97 [+ or -] 3%). Seventeen of 289 patients experienced episodes of Sp[O.sub.2] < 90% (5.9%). Fourteen of 155 patients who breathed room air had episodes of Sp[O.sub.2] < 90% compared to 3 of 133 patients who breathed 30% oxygen (p = 0.02). The three patients who experienced Sp[O.sub.2] < 90% while receiving supplemental oxygen had immediate restoration of Sp[O.sub.2] [greater than or equal to] 90% with the stir-up regimen. Of the 14 patients who inhaled room air and experienced desaturation, 9 patients had an immediate increase in Sp[O.sub.2] with the stir-up regimen. The other five patients received supplemental oxygen to restore Sp[O.sub.2] [greater than or equal to] 90%. All patients ultimately had Sp[O.sub.2] [greater than or equal to] 90% (Table 7). anesthesia and mechanical ventilation. Increases in PETC[O.sub.2] and PaC[O.sub.2] occurred in every patient who underwent deliberate hypoventilation. These changes were accompanied by an immediate decrease in Sp[O.sub.2] in patients with FI[O.sub.2] of 0.21, but not in patients with FI[O.sub.2] levels of 0.25 or 0.30. Although a significant decrease in mean Pa[O.sub.2] occurred during 10 min of hypoventilation in patients with FI[O.sub.2] levels of 0.25 or 0.30, mean Pa[O.sub.2] remained sufficiently elevated to prevent a detectable decrease in Sp[O.sub.2]. When such a decline in Sp[O.sub.2] occurred, it was not consistent or sufficient to detect a significant change. Thus, administration of even small amounts of supplemental oxygen masked our ability to detect hypoventilation with pulse oximetry in anesthetized patients receiving mechanical ventilation.

In healthy volunteers, PaC[O.sub.2] has been shown to rise at a logarithmic rate of 8 to 25 mm Hg/min after the onset of apnea. (12) A decrease in alveolar ventilation of at least 50%, such as we produced, caused a small, but significant increase in PaC[O.sub.2] during the brief period of data collection. Given sufficient time for equilibrium, PaC[O.sub.2] would have doubled, at least. In contrast, Pa[O.sub.2] fell relatively precipitously. Acute hypoventilation is known to decrease the volume of oxygen delivered to gas-exchanging lung units. However, the rate at which oxygen is removed from the lung by pulmonary capillary blood proceeds at a normal rate. During hypoventilation with room air, the disequilibrium between oxygen extracted from the alveoli and oxygen delivered to the alveoli causes a concentrating effect of nitrogen and carbon dioxide, which further exaggerates the decrease in alveolar oxygen tension (PA[O.sub.2]). This sequence explains why, during hypoventilation with room air, mean Pa[O.sub.2] fell rapidly and markedly by 30 mm Hg while mean PaC[O.sub.2] increased only 5 mm Hg. The differences in the rate of change of PA[O.sub.2], Pa[O.sub.2], and PaC[O.sub.2] during hypoventilation have clinical importance. The rate of decline in Pa[O.sub.2] and Sp[O.sub.2] is greater than the increase in PaC[O.sub.2]. Therefore, changes in oxygenation, as measured by pulse oximetry, will provide an earlier indication of hypoventilation than will capnography, but only when breathing room air. (11)

Changes in oxygenation during hypoventilation may be further clarified by examining the equation used to estimate alveolar gas content:

PA[O.sub.2] = PI[O.sub.2] - PaC[O.sub.2] [(FI[O.sub.2] + (1 - FI[O.sub.2])/R)]

where PI[O.sub.2] is the product of FI[O.sub.2] and barometric pressure minus water vapor pressure, and R is the respiratory gas exchange ratio. This ratio results when the amount of carbon dioxide eliminated by alveolar ventilation is divided by the amount of oxygen taken up by the pulmonary capillary blood, which, under normal metabolic and ventilatory conditions, is typically 0.8. As indicated by the alveolar gas equation, PA[O.sub.2] will decrease in response to increasing PaC[O.sub.2]. During hypoventilation, elimination of carbon dioxide decreases transiently, causing a rapid and significant increase in the volume of oxygen removed relative to the volume of carbon dioxide eliminated. Pa[O.sub.2] varies directly with PA[O.sub.2]. Under unsteady-state conditions, there is a temporary decrease in the respiratory gas exchange ratio to produce a marked fall in Pa[O.sub.2] accompanied by a modest rise in PaC[O.sub.2]. (11)

When supplemental oxygen is administered, hypoventilation will have a similar, but far less significant effect on PA[O.sub.2] and Pa[O.sub.2]. As we observed, only small increases in inspired oxygen are needed to alleviate any desaturation that might occur secondary to hypoventilation. With supplemental oxygen, relatively less nitrogen in the alveolar gas mixture partially negates the concentrating effect of nitrogen and carbon dioxide in the alveoli, as oxygen is consumed. The effect of supplemental oxygen on masking hypoventilation can be demonstrated further with mathematical modeling of PA[O.sub.2] and PC[O.sub.2] as a function of alveolar ventilation and varying FI[O.sub.2] (Fig 2). With an FI[O.sub.2] of 0.30, PA[O.sub.2] is approximately 100 mm Hg (point a) when alveolar carbon dioxide tension (PAC[O.sub.2]) approaches 90 mm Hg (point b), thereby making detection of profound hypoventilation impossible with pulse oximetry. But with an FI[O.sub.2] of 0.21, as PAC[O.sub.2] rises > 65 mm Hg (point c), PA[O.sub.2] will decrease to < 60 mm Hg (point d), resulting in a Sp[O.sub.2] < 90%. While breathing room air, a patient cannot hypoventilate sufficiently to elevate PaC[O.sub.2] > 70 mm Hg without a pulse oximeter reading < 90%, thus precluding the possibility of carbon dioxide narcosis and undetected apnea. (1) With supplemental inspired oxygen as low as 0.25, PAC[O.sub.2] could be nearly 100 mm Hg (point e) when PA[O.sub.2] approaches 60 mm Hg (point f). Supplementation of inspired oxygen with an FI[O.sub.2] > 0.25 could put a patient at risk for carbon dioxide narcosis, before Sp[O.sub.2] would fall below 90%.

[FIGURE 2 OMITTED]

In phase 1, all patients with an FI[O.sub.2] of 0.30 maintained Sp[O.sub.2] > 90% throughout 10 min of hypoventilation during general anesthesia. In addition, 9 of 10 patients with an FI[O.sub.2] of 0.25 maintained Sp[O.sub.2] > 90%. Based on our modeling of induced hypoventilation in anesthetized patients receiving mechanical ventilation, we hypothesized that in spontaneously breathing patients, the detection of hypoventilation with Sp[O.sub.2] monitoring may be ineffective when patients are administered supplemental oxygen. Prior publications support this concept. (2,5) The clinical applicability of using pulse oximetry to detect hypoventilation has been limited by the use of supplemental oxygen. (10,13,14) Therefore, in the second phase of the study, we sought to evaluate the clinical utility of pulse oximetry to detect hypoventilation in spontaneously breathing patients in the PACU, with and without administering supplemental oxygen.

Patients who were able to maintain Sp[O.sub.2] [greater than or equal to] 90% during room-air breathing on arrival to the PACU were randomized for further data collection. There was an intergroup difference in mean body weight that was small but statistically significant. It is unlikely that this difference confers any clinical significance. The vast majority of the patients who breathed room air did not experience Sp[O.sub.2] < 90%, yet the administration of supplemental oxygen did not maintain Sp[O.sub.2] [greater than or equal to] 90% in all patients. The incidence of Sp[O.sub.2] < 90% was four times higher in patients who breathed room air (9.0% vs 2.3%). All but a few patients who breathed room air had immediate restoration of Sp[O.sub.2] following the stir-up regimen, suggesting that hypoventilation may be an etiologic factor in the decline in Sp[O.sub.2]. Measurements to confirm hypoventilation with the presence of hypercarbia during episodes of desaturation in the PACU were not undertaken. Drawing arterial blood to measure PaC[O.sub.2] would have been impractical from a clinical standpoint, since episodes of desaturation were transient. Other methods that quantify carbon dioxide, such as capnography or transcutaneous carbon dioxide, also are fraught with difficulties in the PACU setting. (3,4) Since all other variables and anesthetic management were equivalent, we surmise that respiratory depression and hypoventilation occurred with equivalent frequency in both groups of patients. However, the lower incidence of desaturation in the group of patients who received supplemental oxygen likely was due to a masking effect by the increased FI[O.sub.2]. In all patients, the transient decrease in Sp[O.sub.2] easily was managed.

There are six physiologic/pathologic conditions that may lead to arterial oxyhemoglobin desaturation: (1) FI[O.sub.2] < 0.21; (2) diffusion defect; (3) barometric pressure < 760 mm Hg; (4) right-to-left intrapulmonary shunting of blood; (5) low, but finite, V/Q ratio; and (6) hypoventilation. In our investigation, only the last three are potential etiologies of Sp[O.sub.2] < 90%. Since arterial desaturation was transient and reversed by either a stir-up regimen, or an increase in FI[O.sub.2], it is unlikely that right-to-left intrapulmonary shunting was a significant cause. Only by analysis of an arterial blood sample could we verify hypoventilation, rather than decreased V/Q ratio, as the source of arterial desaturation. Nevertheless, increase in FI[O.sub.2] to 0.25 or 0.3 will increase Pa[O.sub.2] and Sp[O.sub.2], thereby masking the detection of respiratory abnormalities, either hypoventilation or low V/Q. Thus, whether the decrease in Sp[O.sub.2] is secondary to low V/Q ratio, a form of "regional hypoventilation," or global hypoventilation resulting in hypercarbia, pulse oximetry monitoring during room air breathing will permit earlier detection of gas exchange abnormalities.

Our findings concur with others (6,13) who have reported the limitation of pulse oximetry in monitoring ventilatory status when supplemental oxygen is administered. Hypercarbia secondary to respiratory depression and not reliably detected by pulse oximetry has been reported during GI endoscopy and bronchoscopy. (2,5) One case report (7) describes a patient receiving morphine patient-controlled anesthesia with high-flow oxygen administration by facemask in whom carbon dioxide narcosis and apnea developed. Pulse oximetry readings between 92% and 95% were recorded, despite a PaC[O.sub.2] of 102 mm Hg and an arterial pH of 7.08. The authors (7) concluded that pulse oximetry failed to permit detection of opioid-induced respiratory depression, in the presence of supplemental oxygen.

Based on our findings, we advocate the application of supplemental oxygen only in patients who are unable to maintain an acceptable Sp[O.sub.2] while breathing room air. In patients able to maintain Sp[O.sub.2] > 90% on an FI[O.sub.2] of 0.21, pulse oximetry monitoring during room air breathing is a useful tool to assess ventilation, without the need for capnography or arterial blood gas analysis. While our data were obtained in the operating room and PACU setting, our results suggest that this type of monitoring also could be utilized in any environment where monitoring of ventilation is needed, such as procedural suites for bronchoscopy and GI endoscopy, where sedation is utilized. Pulse oximetry during room-air breathing also will be useful in guiding and/or limiting the administration of opioids and other respiratory-depressant drugs. Assessment of ventilatory abnormalities in patients receiving epidural, intrathecal, and IV patient controlled anesthesia narcotics could be achieved with pulse oximetry, but only during room-air breathing.

Historically, an Sp[O.sub.2] < 90% has been used to define "arterial hypoxemia." (15,16) Accordingly, clinicians often will administer supplemental oxygen out of habit to ensure "adequate" oxygenation and to avoid reaching the 90% threshold. (15) But is this clinical practice warranted? Currently, there is no consensus in the literature regarding recommendations on the prophylactic administration of supplemental oxygen to all postoperative patients, and some communications have stressed the dangers of masking severe hypoventilation with supplemental oxygen. (6,7,13) We suggest that the decision to administer supplemental oxygen not be based on routine, but should entail consideration of the risk of masking undetected hypoventilation, or mismatching of ventilation and perfusion, in accordance with the patient's need for increased Sp[O.sub.2]. If persistent, decreased Sp[O.sub.2] may indicate the need for arterial blood analysis to determine if the arterial hypoxemia is due to hypoventilation, or mismatching of ventilation and pulmonary perfusion. Then, appropriate treatment may be administered.

Sedation may cause profound respiratory depression and hypoventilation. Thus, accurate monitoring of ventilatory status of sedated patients is desirable. Methods to detect hypoventilation in the spontaneously breathing patients receiving respiratory-depressant drugs are limited. Pulse oximetry primarily has been used to assess oxygenation, but not ventilation. A decline in Sp[O.sub.2] during room-air breathing appears to be a reliable indicator of ventilatory abnormalities, whether occurring at a global or regional level; the presence of such abnormalities will go undetected in the presence of supplemental oxygen. Without the need for capnography and arterial blood gas analysis, pulse oximetry is a useful tool to assess ventilation in the absence of supplemental inspired oxygen.

REFERENCES

(1) Sieker HO, Hickam JB. Carbon dioxide intoxication: the clinical syndrome, its etiology and management with particular reference to the use of mechanical respirators. Medicine 1956; 35:389-423

(2) Nelson DB, Freeman ML, Silvis SE, et al. A randomized controlled trial of transcutaneous carbon dioxide monitoring during ERCP. Gastrointest Enclose 2000; 51:288-295

(3) Bennett J, Petersen T, Burleson JA. Capnography and ventilatory assessment during ambulatory dentoalveolar surgery. J Oral Maxillofac Surg 1997; 55:921-925

(4) Anderson JA, Clark PJ, Kafer ER. Use of capnography and transcutaneous oxygen monitoring during outpatient general anesthesia for oral surgery. J Oral Maxillofac Surg 1987; 45:3-10

(5) Evans EN, Ganeshalingam K, Ebden P. Changes in oxygen saturation and transcutaneous carbon dioxide and oxygen levels in patients undergoing fiberoptic bronchoscopy. Respir Med 1998; 92:739-742

(6) Davidson JA, Hosie HE. Limitations of pulse oximetry: respiratory insufficiency--a failure of detection. BMJ 1993; 307:372-373

(7) Smyth E, Egan TD. Apneic oxygenation associated with patient-controlled analgesia. J Clin Anesth 1998; 10:499-501

(8) Marshall BE, Wyche MQ Jr. Hypoxemia during and after anesthesia. Anesthesiology 1972; 37:178-209

(9) Hines R, Barash PG, Watrous G, et al. Complications occurring in the postanesthesia care unit: a survey. Anesth Analg 1992; 74:503-509

(10) Downs JB. Prevention of hypoxemia: the simple, logical, but incorrect solution. J Clin Anesth 1994; 6:180-181

(11) Lumb AB. Nunn's applied respiratory physiology, 5th ed. Oxford, UK: Butterworth Heinemann, 2000; 289-290

(12) Stock MC, Downs JB, McDonald JS, et al. The carbon dioxide rate of rise in awake apneic humans. J Clin Anesth 1988; 1:96-103

(13) Hutton P, Clutton-Brock T. The benefits and pitfalls of pulse oximetry. BMJ 1993; 307:457-458

(14) Wiklund L, Hok B, Stahl K, et al. Postanesthesia monitoring revisited: frequency of true and false alarms from different monitoring devices. J Clin Anesth 1994; 6:182-188

(15) DiBenedetto RJ, Graves SA, Graventstien N, et al. Pulse oximetry monitoring can change routine oxygen supplementation practices in the postanesthesia care unit. Anesth Analg 1994; 78:365-368

(16) Scuderi PE, Mims GR, Weeks DB, et al. Oxygen administration during transport and recovery after outpatient surgery does not prevent episodic arterial desaturation. J Clin Anesth 1996; 8:294-300

* From the H. Lee Moffitt Cancer Center and the Department of Anesthesiology, University of South Florida College of Medicine, Tampa, FL.

This work was done at the H. Lee Moffitt Cancer Center and the University of South Florida College of Medicine.

Support was provided solely by departmental sources. Manuscript received July 23, 2003; revision accepted May 10, 2004.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: permissions@chestnet.org).

Correspondence to: John B. Downs, MD, FCCP, H. Lee Moffitt Cancer Center, 12902 Magnolia Dr, Suite 2194 Anesthesia, Tampa, FL 33612; e-mail: jdowns@hsc.usf.edu

COPYRIGHT 2004 American College of Chest Physicians
COPYRIGHT 2004 Gale Group