Anda di halaman 1dari 8


PULSE OXIMETRY MONITORING AND LATE Eichhorn JH. Pulse oximetry monitoring and late postoperative
hypoxemia on the general care oor.
CARE FLOOR ABSTRACT. Hypoxemia has long been recognized as a risk to
John H. Eichhorn, MD patients in the operating room and postanesthesia care unit,
and hemoglobin oxygen saturation (HbO2 ) monitoring with
pulse oximetry has become a standard of care in these areas.
There is growing evidence, however, suggesting that later
postoperative hypoxemia also may play a role in organ dys-
function leading to morbidity and mortality. Economic pres-
sures to move patients earlier from expensive postanethesia
recovery and intensive care areas to the general care oor ^
where nurse-to-patient ratios are lower and lines of sight
and sound may be impaired by walls and curtains ^ may lead
to inadequate surveillance of at-risk patients. These patient-
management trends underscore the importance of improved
monitoring of respiratory status on the general care oor. In
this environment, telemetric pulse oximetry monitoring may
represent a cost-eective approach to maximizing quality of
care while enhancing risk management. This review discusses
late postoperative hypoxemia and identies areas for further
KEY WORDS. Hypoxemia, arterial oxygen saturation, pulse
oximetry, monitoring, respiratory monitoring, sleep apnea,
myocardial ischemia, cerebral ischemia.


Arterial hypoxemia, particularly that associated with

hypoventilation, is a major risk to patients during surgi-
cal anesthesia, especially for abdominal and thoracic
procedures, and it continues to be a risk through the
early postoperative period in the postanesthesia are unit
(PACU). While the risk of hypoxemia generally has
been thought to end by the time of discharge from the
PACU, a growing body of thought suggests that hypo-
xemia can persist in some patients on the general care
oor during the rst postoperative week. Although
monitoring of arterial oxygen saturation is mandated in
the operating room (OR) and PACU, such monitoring
generally is discontinued when the patient is discharged
to the general care oor, so that late postoperative
hypoxemia may go undetected and untreated. This
raises concerns, previously expressed in editorial com-
mentary [1, 2], that premature termination of respira-
tory monitoring is exposing already compromised pa-
tients to unnecessary risks. These concerns are timely
From the Department of Anesthesiology, University of Mississippi because increasing cost pressures are leading to earlier
Medical Center, Jackson, Mississippi, U.S.A. discharge from expensive postanesthesia and intensive
Received Oct 8, 1996. Accepted for publication Oct 11, 1997. care areas. This review considers evidence bearing on
Address correspondence to John H. Eichhorn, Department of Anes- these concerns.
thesiology, University of Mississippi Medical Center, 2500 N. State
Street, Jackson, MS 39216-4505, U.S.A.

Journal of Clinical Monitoring and Computing 14: 49^55, 1998.

1998 Kluwer Academic Publishers. Printed in the Netherlands.
50 Journal of Clinical Monitoring and Computing Vol 14 No 1 January 1998

HYPOXEMIA IN THE LATE POSTOPERATIVE PERIOD HbO2 sat as stable or unstable depending on the per-
centage of each postoperative hour spent at a given
Even after the direct respiratory depressant eects of oxygen desaturation. Unstable hypoxemia, with inter-
anesthetic medications and consequent ventilation-per- mittent episodes of less than 85% saturation, is most
fusion mismatch dissipate in the early postoperative frequent on the rst and second postoperative nights. In
period, persistent eects of surgical stress appear to all studies, episodic or unstable hypoxemia resolved by
compromise respiratory function for up to a week in the end of the rst postoperative week.
the late postoperative period [3^6]. Two patterns have Episodic hypoxemia is closely related to postsurgical
been identied [3^6] ^ constant, or persistent, hypoxe- problems of respiratory control and the impact of surgi-
mia is related to preoperative respiratory function while cal stress on sleep patterns [11] and sleep apnea [3, 5],
episodic noctural hypoxemia is closely related to episo- Reeder [8] and Rosenberg [9] each cite, and conrm,
dic sleep apnea that is aggravated by surgical stress and sleep studies showing that surgical stress suppresses both
postoperative physiological disturbances. Each pattern rapid eye movement and slow wave sleep, and that these
occurs with major (e.g., abdominal and thoracic), but both rebound coincident with the onset of episodic
not usually with minor surgical procedures [7], and hypoxemia. The role of opiate administration is unclear.
each is suspected of potentially contributing to the risk There is evidence that opiates do not prevent [8] and
of postsurgical morbidity and even mortality. may even contribute to [10, 12] episodic hypoxemia,
Constant, or persistent, hypoxemia is generally mild while other investigators nd little or no direct eect
and resolves in the course of the rst postoperative week. [9, 13]. Eects of patient obesity and advanced age in
The degree of hemoglobin desaturation is greatest on increasing the duration of hypoxemia have been re-
the second and third postoperative nights and is related ported [10, 14].
to baseline levels of hemoglobin oxygen saturation Administration of supplemental oxygen limits the
(HbO2 sat) during the immediate preoperative night extent of both constant and episodic hypoxemia [8, 13,
[6, 8] but is not predicted by preoperative spirometry 15] without, however, aecting the frequency of apneic
[8]. Constant relative hypoxemia is correlated with and hypoxemic episodes [7, 13, 16]. The degree and
decreased (by up to 20%) functional residual capacity frequency of hypoxemic events are not predicted by
(FRC) in the daytime and is greatest at night [3, 4]. The preoperative levels of HbO2 sat [5, 6].
decrease of FRC may be due to multiple anesthetic and
surgical factors [3, 4, 8] including atelectasis of depend-
ent lung regions, changes of chest wall muscle tone and CONSEQUENCES OF HYPOXEMIA
function, and, possibly, changes of bronchomotor and
pulmonary vascular tone. Following surgery, the de- Studies of late postoperative hypoxemia have been
crease of FRC and persistent hypoxemia are delayed motivated by the concern [17] that hypoxemia in this
in onset, are maximal after 16^24 hours, and require setting, especially the nocturnal episodic form, may
1 week to 10 days to resolve [4]. increase patients' risk of postoperative complications.
Episodic noctural hypoxemia can be superimposed Of particular concern is the possible contribution of
on a background of constant hypoxemia but primarily nocturnal episodic myocardial ischemia to diurnal (cir-
involves disruptions of normal sleep patterns induced cadian) variations of unexpected postoperative deaths
by the stress and physiologic disruption of anesthesia [18]. Some authors have found an association of episo-
and surgery. It is characterized by episodic, brief (usu- dic hypoxemia with cardiac ischemia in patients having
ally less than a minute) decreases in arterial hemoglobin risk factors for ischemic hart disease [19] and in patients
saturation during sleep. For instance, Rosenberg et al. during the rst week following an acute myocardial
[6, 9] found that following major abdominal surgery, infarction [20]. Hypoxemia and associated cardiac ische-
episodes of desaturation of 5% or more were most mia also have been linked with cardiac events during
frequent on the second postoperative night and often the perioperative period in surgical patients with [21,
included episodes of saturation falling to less than 80% 22] or without [23, 24] known cardiovascular disease.
(in the absence of oxygen therapy), representing a PaO2 Such a linkage to cardiovascular events, however, has
in the 40 s (mm Hg). In other studies, brief (30^60 s) not been found in all clinical studies [25, 26, 27].
episodes of desaturation were more frequent on the third The use of hemoglobin saturation monitoring to
postoperative night [5, 8] following major abdominal detect and treat hypoxemia reduced the incidence of
vascular surgery, although patients in these studies had myocardial ischema in the OR and PACU [28]. Such
received supplemental oxygen on the rst two nights. observations have led to suggestions [17, 19, 29] that
Other investigators [3, 10] have classied patterns of detection and treatment of episodic hypoxemia could
Eichhorn: Hypoxemia on the General Care Floor 51

reduce the risk of postoperative myocardial events. This thesiologists (ASA) [2]. These standards were instituted
possibility demands further clinical investigation, par- to help maximize successful tolerance of and recovery
ticularly in the elderly and in others who may already from ventilatory and respiratory perturbations associ-
have compromised coronary function and vulnerability ated with anesthesia and surgery, and they have received
to even moderate myocardial ischemia. widespread support and compliance within the anes-
Another area of concern [17] is the potential contri- thesiology community. Since its introduction into clin-
bution of hypoxemia to postsurgical cerebral dysfunc- ical practice in the mid-1980s, pulse oximetry moni-
tion, including confusion and delirium [30] especially toring has enabled continuous measurement of blood
in susceptible elderly patients. As with episodic hypo- hemoglobin saturation (HbO2 sat) in a variety of clinical
xemia, confusion is more likely to be seen in surgical settings and in virtually all of the studies cited above. Its
patients after major surgery than minor surgery with clinical and scientic utility stems from the fact that it
comparable general anesthetics [31]. Confusion is max- provides a noninvasive and continuous estimation of an
imal on the third postoperative day, is signicantly otherwise dicult-to-measure clinical parameter. The
associated with arterial oxygen desaturation on the instrumentation is relatively inexpensive, the technology
second postoperative night, and is diminished by sup- is stable, and it is amenable to computer networking for
plemental oxygen treatment [32]. Contributory roles of central monitoring, as described in recent comprehen-
surgical trauma and age are suggested by the observa- sive reviews [39^41].
tion that only minor changes of cognitive function are As with all clinical instrumentation, there are limita-
found in normal volunteers made hypoxemic to an tions that must be recognized and mitigated for eective
hemoglobin saturation level of 78% [33]. use of pulse oximetry in clinical settings [39, 42, 43].
It has also been suggested that, in patients with already Pulse oximetry monitoring (POM) provides a spectro-
compromised cerebral circulation, hypoxemia may play scopic estimate of hemoglobin oxygen saturation rather
a role in the development of ischemic stroke [34]. This than a direct measure of the total oxygen content of the
connection has been previously been demonstrated in blood (oxemia), yet hemoglobin saturation fraction is
non-surgical patients [35] and warrants further inves- generally taken to reect changes of ventilatory/respi-
tigation. ratory status rather than changes of hemoglobin con-
Impairment of wound healing during periods of centration that would aect total oxygen content. It is
hypoxemia has also been raised as a concern [17], given important to remember, also, that registered oximetry
the dependence on tissue oxygenation of wound heal- values can be confounded by congenital or pharmaco-
ing rates and clinical outcomes [36]. Although to date logical methemaglobinemias, certain vital dyes, exces-
there are no studies directly relating episodic hypoxe- sive ambient light, and by circumstances where tissue
mia to wound healing or infection, subcutaneous tissue perfusion is limited. In addition, false alarms may occur
oximetry revealed that postoperative oxygen therapy due to artifacts of patient motion, transducer displace-
increased would oxygenation [37]. ment, or when there is venous congestion or pulsation
While the clinical consequences of constant and epi- [44, 45]. Recently developed artifact rejection algo-
sodic hypoxemia in the postoperative period are yet to rithms can better distinguish arterial pulse waveforms
be fully elaborated, the history of widespread adoption (i.e., rapid onset, slow decay) from movement-induced
of pulse oximetry in other clinical settings suggests that pulse (generally slow onset and decay) and can substan-
it may oer a resource-eective approach to detecting tially reduce the incidence of false alarms and move-
hypoxemia on the general care oor, whether used ment artifacts [46, 47]. Although these technical limita-
around-the-clock, or intermittently during the day (as tions can be controlled [38, 39], the clinical eectiveness
a ``5th vital sign,'' [38] for example) and continuously of pulse oximetry depends on establishing clear guide-
only during the night. lines for using POM and proper training for medical
and nursing sta in its theory and practice [48^51].



The risk of hypoxemia and the clear need for pulse The use of pulse oximetry monitoring has been con-
oximetry monitoring of hemoglobin saturation in both sidered a minimum standard of care in ORs and PACUs
the operating room and the postanesthesia care unit are in the United States since its mandate by ASA standards
well-recognized and have been formalized in treatment adopted in 1989. These standards evolved largely to
standards adopted by the American Society of Anes- codify and validate existing anesthesiology practices as
52 Journal of Clinical Monitoring and Computing Vol 14 No 1 January 1998

well as to accelerate a trend toward increasing reliance Given the inferred increased risk to vulnerable pa-
on oximetry in the PACU. The consensus among anes- tients in these situations, current methods of respiratory
thesiologists is that pulse oximetry is critical for mini- monitoring appear to be inadequate for maintaining
mizing the risk of catastrophic events through its early traditionally high standards of patient care. Physical
warning function. In support of this, pulse oximetry is evaluation using respiratory rate, labor of breathing,
credited with contributing to a drop in injury claims and possible cyanosis is too insensitive and too casually
against anesthesiologists and in reduced malpractice costs applied to be useful in identifying all but the most
since hemoglobin saturation monitoring came into severely hypoxemic patients. Other routine clinical pa-
common use in the 1980s [52]. Pulse oximetry also is rameters and radiographic ndings also are of little use
credited, in part, with reducing unanticipated intensive in predicting or detecting hypoxemia [54]. The count-
care unit admissions associated with anesthesia in a large ing of respirations is, at best, a dicult-to-use and very
teaching hospital [53]. Despite this consensus support, indirect measure of hypoxemia. The ``gold standard'' ^
the actual impact of pulse oximetry use on limiting arterial blood gas analysis ^ while accurate, involves
cardiac and respiratory events continues to be debated several familiar drawbacks. It is, rst of all, invasive and
largely due to an understandable lack of controlled trials often suciently painful so that hyperventilation may
demonstrating an oximetry monitoring-related reduc- artifactually raise arterial oxygen tension during the
tion of morbidity and mortality. procedure itself. Furthermore, the time involved, ex-
Moller et al. [25], in the only large, multi-institution pense, demand for skilled sta, and reporting delays
prospective controlled clinical trial of pulse oximetry, make it poorly suited for routine use, particularly if
found that the use of pulse oximetry reduced the inci- episodic hypoxemia is to be detected. Pulse oximetry,
dence of recognized hypoxemic and other respiratory on the other hand, provides noninvasive, potentially
events. As noted above, pulse oximetry monitoring also continuous, and readily determined levels of arterial
reduced the incidence of recognized myocardial ische- hemoglobin saturation in most clinical situations. Com-
mia. Although the number of cardiac arrests was de- bined with modern telemetry networking (telemetric
creased from 11 to 4, there were far too few patients pulse oximetry ^ TPON) and enhanced artifact rejec-
even in this large study to demonstrate a statistically tion algorithms, oximetry can provide continuous res-
signicant preventive benet of pulse oximetry mon- piratory monitoring [14] displayed at nursing stations
itoring with regard to extremely rare catastrophic as well as at the bedside.
events [25, 28]. It is very likely that this fact will persist Principles and guidelines for use of pulse oximetry
permanently due to the obvious logistic and ethical on the general care oor have been recently published
problems of conducting such a clinical investigation. [51] as have detailed nursing guidelines for respiratory
What remains, therefore, is the strongly supported, monitoring of postsurgical patients [50]. In each, it is
inherently logical, inferential conclusion that the ear- emphasized that POM is likely to be of use in a variety
liest possible warning of developing hypoxemia aords of high-risk patients including those with pre-existing
the maximum possible time for diagnostic and thera- pulmonary disease, obese patients, and patients receiving
peutic intervention, thus reducing the likelihood of an ventilation-depressing analgesics. Once hemoglobin
untoward patient outcome. desaturation is detected and measurement artifacts have
been ruled out, physicians may order conrmatory
arterial blood gas measurements, administer supple-
MONITORING RESPIRATORY STATUS ON THE GENERAL CARE mental oxygen, and review treatment protocols to
FLOOR identify iatrogenic causes.

Several patient management treads call for improved

surveillance of respiratory status on the general care RESOURCE-EFFECTIVENESS ISSUES
oor. As noted above, pressures for early discharge
from PACU, ICU, and coronary care units may result Although additional and more denitive data would be
in the premature (compared with prior practices) release desirable, pulse oximetry monitoring on the general
to the general care oor of at-risk patients. Pressures to care oor has been advocated for at-risk patients as a
increase work loads and decrease stang on nursing cost-eective approach to maximizing quality of care
teams likely result in less intensive scrutiny of patients. while enhancing risk management. The relatively low
As postsurgical patients are moved from the PACU to expense of pulse oximetry systems may even be oset
private and semiprivate rooms, surveillance becomes by operational cost savings if adequate guidelines for
much more dicult due to long hallways, closed-door use and suitable sta training are in place [42]. Because
rooms, and blocked lines of sight and sound.
Eichhorn: Hypoxemia on the General Care Floor 53

pulse oximetry can easily indicate whether a patient REFERENCES

needs oxygen therapy or not, prophylactic and thera-
peutic oxygen use can be rationalized and minimized. 1. Hanning CD. Prolonged postoperative oxygen therapy.
In addition, use of oximetry can reduce the number of Br J Anaesth 1992; 69: 115^116
diagnostic blood gas analyses performed [55]. Further, 2. Eichhorn JH. Pulse oximetry as a standard of practice in
telemetric pulse oximetry systems can aid in the ecient anesthesia. Anesthesiology 1993; 78: 423^426
3. Jones JG, Sapsford DJ, Wheatley RG. Postoperative hy-
utilization of nursing resources. By identifying the con- poxaemia: Mechanisms and time course. Anaesthesia
tribution of hypoxemia to the disruptive and distressing 1990; 45: 566^573
eects of postoperative delirium, oximetry monitoring 4. Clinton CW, Van Heerden PV, Roux A. Postoperative ^
can also improve patient care and even patients' quality its causes and prevention. South African Journal of Sur-
of life. gery 1992; 30: 95^99
Perhaps more signicantly, oximetry use in late post- 5. Reeder MK, Goldman MD, Loh L et al. Late post-
operative patients could potentially decrease the risk of operative nocturnal dips in oxygen saturation in patients
undergoing major abdominal vascular surgery. Predic-
catastrophic late sequellae of anesthesia and surgery as tive value of pre-operative overnight pulse oximetry.
well as the specic dangers of ventilatory depresssion Anaesthesia 1992; 47: 110^115
from pain therapy opiates, whether administered intra- 6. Rosenberg J, Ullstad T, Rasmussen J et al. Time courses
venously, epidurally, or intrathecally, with virtually no of postoperative hypoxaemia. European Journal of Sur-
added risk of its own. Furthermore, this potential bene- gery 1994; 160: 137^143
t appears to be available at a favorable cost. One early 7. Rosenberg J, Oturai P, Erichsen CJ, Pedersen MH, Kehlet
cost-benet analysis estimated that one fatal hypoxemic H. Eect of general anesthesia and major versus minor
surgery on late postoperative episodic and constant hypo-
episode in 40,000 moderately hypoxemic patients xemia. Journal of Clinical Anesthesia 1994; 212^216
would justify the cost of the oximetry [56]. 8. Reeder MK, Goldman MD, Loh L et al. Postoperative
hypoxaemia after major abdominal vascular surgery. Br
J Anaesth 1992; 68: 23^26
CONCLUSION 9. Rosenberg J, Wildschiodtz G, Pedersen MH, von Jessen
F, Kehlet H. Late postoperative nocturnal episodic hypo-
The combination of constant and episodic hypoxemia xaemia and associated sleep pattern. Br J Anaesth 1994;
72: 145^150
in the late postoperative period may represent a poten- 10. Entwistle MD, Roe PG, Sapsford DJ, Berrisford RG,
tially serious risk for postsurgical patients. Given increas- Jones JG. Patterns of oxygenation after thoractomy. Br
ing cost pressures to accelerate transfer of patients from J Anaesth 1991; 67: 704^711
intensive care areas of the hospital to the general care 11. Knill RL, Moote CA, Skinner MI, Rose EA. Anesthesia
oor, it is valid to maximize methods of patient mon- with abdominal surgery leads to intense REM sleep
itoring to guard against the potential complications of during the rst postoperative week. Anesthesiology
late postoperative hypoxemia. 1990; 73: 52^61
12. Nozaki-Taguchi N, Oka T, Kochi T, Taguchi N, Mizu-
Pulse oximetry became a standard of care in the guchi T. Apnoea and oximetric desaturation in patients
operating room and postanesthesia care unit because it receiving epidural morphine after gastrectomy: a compar-
warns very early in the development of desaturation ison of intermittent bolus and patient controlled admin-
and also detects clinically inapparent episodes of arterial istration. Anaesthesia and Intensive Care 1993; 21: 292^
hypoxemia. Its nearly unanimous acceptance and sup- 297
port by the anesthesiology community its testimony to 13. Rosenberg J, Pedersen MH, Gebuhr P, Kehlet H. Eect
its clinical utility. Current evidence suggests that hypo- of oxygen therapy on late postoperative episodic and
constant hypoxaemia. Br J Anaesth 1992; 68: 18^22
xemia may play a role in postoperative morbidity and 14. Eichhorn JH, Cohen MS, Darvish AH. Hypoxemia
possibly even mortality in the late postoperative period during the rst postanesthetic 24 hours demonstrated
as well. Given recent advances in the technological reli- by continuous telemetric artifact-free pulse oximetry.
ability of pulse oximetry, as well as its incorporation of Anaesth Analg 1992; 74: S45
telemetry features, further clinical and scal study of 15. Jones JG, Jordan C, Scudder C, Rocke DA, Barrowclie
the problem of hypoxemia on the general care oor is M. Episodic postoperative oxygen desaturation: The
warranted to identify patients at risk, assess its eect on value of added oxygen. Journal of Royal Society of
Medicine 1985; 78: 1019^1922
patient outcomes, develop methods of problem preven- 16. Raemer DB, Warren DL, Morris R, Philip BK, Philip
tion and treatment, and determine more fully the e- JH. Hypoxemia during ambulatory gynecologic surgery
cacy and resource-eectiveness of pulse oximetry mon- as evaluated by the pulse oximeter. J Clin Monit 1987; 3:
itoring in this setting. 244^248
54 Journal of Clinical Monitoring and Computing Vol 14 No 1 January 1998

17. Kehlet H, Rosenberg J. Late post-operative hypoxaemia 35. Palomaki H, Partinen M, Erkinjuntti T, Kaste M. Snor-
and organ dysfunction. European Journal of Anaes- ing, sleep apnea syndrome, and stroke. Neurology 1992;
thesiology 1995; 10: 31^34 42: 75^82
18. Rosenberg J, Pedersen MH, Ramsing T, Kehlet H. Cir- 36. Van Esbroeck G, Gys T, Hubens A. Evaluation of tissue
cadian variation in unexpected postoperative death. Br J oximetry in perioperative monitoring of colorectal sur-
Surg 1992; 79: 1300^1302 gery. Br J Surg 1992; 79: 584^587
19. Goldman MD, Reeder MK, Muir AD et al. Repetitive 37. Rosenberg J, Ullstad T, Larsen PN, Moesgaard F, Kehlet
nocturnal arterial oxygen desaturation and silent myocar- H. Continuous assessment of oxygen saturation and sub-
dial ischemia in patients presenting for vascular surgery. cutaneous oxygen tension after abdominal operations.
Journal of American Geriatrics Society 1993; 41: 703^709 Acta Chirurgica Scandinavica 1990; 156: 585^590
20. Galatius-Jensen S, Hansen J, Rasmussen V et al. Noctur- 38. Porter R, Merlin M, Heller M. The fth vital sign.
nal hypoxaemia after myocardial infarction: association Emergency 1990; 22: 37^41
with nocturnal myocardial ischaemia and arrhythmias. 39. Hanning CD, Alexander-Williams JM. Pulse oximetry:
British Hart Journal 1994; 72: 23^30 A practical review. British Medical Journal 1995; 311:
21. Gill NP, Wright B, Reilly CS. Relationship between 367^370
hypoxaemic and cardiac ischaemic events in the perioper- 40. Lindberg LG, Lennmarken C,Vegfors M. Pulse oximetry
ative period. Br J Anaesth 1992; 68: 471^473 ^ clinical implications and recent technical developments.
22. Reeder MK, Muir AD, Foex P et al. Postoperative Acta Anaesthesiologica Scandinavica 1995; 39: 279^287
myocardial ischaemia: Temporal association with noc- 41. Severinghaus JW. History and recent developments in
turnal hypoxaemia. Br J Anaesth 1991; 67: 626^631 pulse oximetry. Scandinavian Journal of Clinical and
23. Pateman JA, Hanning CD. Postoperative myocardial in- Laboratory Investigation 1993; 214 (Suppl): 105^111
farction and episodic hypoxaemia. Br J Anaesth 1989; 63: 42. Inman KJ, Sibbald WJ, Rutledge FS et al. Does imple-
648^650 menting pulse oximetry in a critical care unit result in
24. Rosenberg J, Rasmussen V, von Jessen F, Ullstad T, substantial arterial blood gas savings? Chest 1993; 104:
Kehlet H. Late postoperative episodic and constant hypo- 542^546
xaemia and associated ECG abnormalities. Br J Anaesth 43. Mardirossian G, Schneider RE. Limitations of pulse
1990; 65: 684^691 oximetry. Anesth Prog 1993; 39: 194^196
25. Moller JT, Pedersen T, Rasmussen LS et al. Randomized 44. Pan PH, Gravenstein N. Intraoperative pulse oximetry:
evaluation of pulse oximetry in 20,802 patients: I. Design, Frequency and distribution of discrepant data. J Clin
demography, pulse oximetry failure rate, and overall Anesth 1994; 6: 491^495
complication rate. Anesthesiology 1993; 78: 436^444 45. Wiklund L, Hok B, Stahl K, Jordeby-Jonsson A. Post-
26. Entwistle MD, Sommerville D, Tandon AP, Jones JG. anesthesia monitoring revisited: Frequency of true and
Eect of hypoxaemia on the resting electrocardiogram false alarms from dierent monitoring devices. J Clin
(ECG) in patients with cardiac ischaemia. Annals of Acad- Anesth 1994; 6: 182^188
emy of Medicine, Singapore 1994; 23: 460^464 46. Lie C, Kehlet H, Rosenberg J. Comparison of the Nellcor
27. Keyl C, Lemberger P, Rodig G, Damgbacher M, Frey A. N-200 and N-3000 pulse oximeters during simulated
Hypoxaemia and myocardial ischaemia on the night post-operative activities. Anaesthesia 1997; 52: (in press)
before coronary bypass surgery. Br J Anaesth 1994; 73: 47. Plummer JL, Ilsley AH, Fronsko RRL, Owen H. Identi-
157^161 cation of movement artefact by the Nellcor N-200 and
28. Moller JT, Johannessen NW, Espersen K et al. Random- N-3000 pulse oximeters. J Clin Monit 1997 (in press)
ized evaluation of pulse oximetry in 20,802 patients: II. 48. Stoneham MD, Saville GM, Wilson IH. Knowledge
Perioperative events and postoperative complications. about pulse oximetry among medical and nursing sta.
Anesthesiology 1993; 78: 445^453 Lancet 1994; 344: 1339^1342
29. Cunningham AJ. Myocardial ischemia ^ association with 49. Bowton DL, Scuderi PE, Harris L, Haponik EF. Pulse
perioperative cardiac morbidity. Yale Journal of Biology oximetry monitoring outside the intensive care unit:
and Medicine 1993; 66: 339^348. progress or problem? Annals of Internal Medicine 1991;
30. Parikh SS, Chung F. Postoperative delirium in the elderly. 115: 450^454
Anesth Analg 1995; 80: 1223^1232 50. McGagan PA. Hazards of hypoxemia: How to protect
31. Rosenberg J, Kehlet H. Postoperative mental confusion ^ your patient from low oxygen levels. Nursing 1996; 26:
association with postoperative hypoxemia. Surgery 1993; 41^46
114: 76^81 51. Consortium on Respiratory Monitoring on the General
32. Aakerlund LP, Rosenberg J. Postoperative delirium: Care Floor. Principles and guidelines for respiratory
Treatment with supplementary oxygen. Br J Anaesth monitoring on the general care oor. J Clin Monit 1996;
1994; 72: 286^290 12: 411^416
33. Noble J, Jones JG, Davis EJ. Cognitive function during 52. Tinker JH, Dull DL, Caplan RA, Ward RJ, Cheney FW.
moderate hypoxemia. Anaesthesia and Intensive Care Role of monitoring devices in prevention of anesthetic
1993; 21: 180^184 mishaps: A closed claims analysis. Anesthesiology 1989;
34. Rosenberg J. Late postoperative hypoxemia: Mechanisms 71: 541^546
and clinical implications. Danish Medical Bulletin 1995; 53. Cullen DJ, Nemeskal AR, Cooper JB, Zaslavsky A,
42: 40^46 Dwyer MJ. Eect of pulse oximetry, age, and ASA
Eichhorn: Hypoxemia on the General Care Floor 55

physical status on the frequency of patients admitted

unexpectedly to a postoperative intensive care unit and
the severity of their anesthesia-related complications.
Anesth Analg 1992; 74: 181^188
54. Bowton DL, Scuderi PE, Haponik EF. The incidence and
eect on outcome of hypoxemia in hospitalized medical
patients. American Journal of Medicine 1994; 97: 38^46
55. Bierman MI, Stein KL, Snyder JV. Pulse oximetry in the
postoperative care of cardiac surgical patients: A random-
ized controlled trail. Chest 1992; 102: 1367^1370
56. Stausholm K, Kehlet H, Rosenberg J. Oxygen therapy
reduces postoperative tachycardia. Anaesthesia 1995; 50:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.