Influence of pulse oximeter settings on the frequency of alarms and detection of hypoxemia: Theoretical effects of artifact rejection, alarm delay, averaging, median filtering or a lower setting of the alarm limit

OBJECTIVE

The potential benefit of a reduced frequency of false pulse oximeter low oxyhemoglobin saturation (SpO2) alarms is that the attention of personnel is only directed to patients who experience hypoxemia. The present study was undertaken to better understand the effects of different settings of the pulse oximeter on false (artifact) and true (hypoxemia) alarms.

METHODS

Using the original SpO2 data of 200 postoperative patients, we calculated off-line the effects of five methods (artifact rejection, alarm delay (2-44 s, 2 s increments), averaging (10-90 s), median filtering (10-90 s) and decreasing the alarm limit from 90% to 85%) on the number of (true- and false) alarms.

RESULTS

830 episodes of hypoxemia (SpO2 < or = 90%) and 73 episodes of severe hypoxemia (SpO2 < or = 85%) occurred. With a SpO2 alarm limit of 90%, the alarm was triggered 1535 times (830 true, 705 false). Artifact rejection reduced alarms by almost 50%. An alarm delay of 6 s or an averaging or median filtering epoch of 10 s resulted in an alarm reduction of almost 50%. No differences were found in the reduction of alarms between averaging and median filtering. Changing the alarm limit to 85% reduced the number of alarms by 82%. A similar reduction of alarms was obtained with either an alarm delay of 18 s or an averaging or median filtering epoch of 42 s. However, an alarm limit of 85% reduced the number of false alarms less than the other three algorithms (p < 0.01).

CONCLUSIONS

The data from the present study suggest that in order to effectively suppress false alarms caused by pulse oximeter artifacts, it may be preferable to use a longer filtering epoch of approximately 40 s, rather than to decrease the lower alarm limit.

Influence of pulse oximeter lower alarm limit on the incidence of hypoxaemia in the recovery room

In a prospective, randomized study, we have investigated the effects of two arbitrary pulse oximeter lower alarm limit (LAL) settings (90% = group 90, n = 320 and 85% = group 85, n = 327) on the incidence of hypoxaemia in the recovery room. In group 90, we calculated the theoretical effect of elimination of transient episodes of low pulse oximeter oxyhaemoglobin saturation (SpO2) by introducing a time delay between the onset of the alarm condition and triggering of the alarm. When only hypoxaemic episodes lasting more than 1 min were included, SpO2 < or = 90% occurred in 11% of patients in group 90 and in 20% in group 85 (relative risk (RR) 1.84, confidence interval (CI) 1.26-2.69; P < 0.01). Hypoxaemia < or = 85% occurred in 2% of patients in group 90 and in 6% in group 85 (RR 3.10, CI 1.32-7.28; P < 0.01). In group 90, 1007 alarms (33% false) occurred, whereas in group 85, 395 alarms (28% false) occurred. Introducing a theoretical delay of 15 s in group 90 between crossing the alarm threshold and triggering the alarm would have reduced the number of alarms by 60%. The results of the study suggest that decreasing the alarm limit in an attempt to reduce frequent false alarms may lead to an increase in more relevant episodes of hypoxaemia and setting the LAL at 85% cannot be recommended routinely. Introducing a 15 s delay in group 90 would reduce the number of alarms by the same amount as changing the LAL from 90% to 85%.

Influence of motivation of care providers on the incidence of postoperative hypoxaemia in the recovery room

We have studied the influence of motivation of care providers on the incidence and duration of postoperative hypoxaemia in the recovery room. In a prospective, switch-back designed cohort study, we have compared the incidence of low pulse oximeter saturation values (SpO2) during pre-intervention, intervention and post-intervention phases. Low SpO2 values were classified as either hypoxaemia (SpO2 < or = 90%, minimum duration 1 min) or artefact. Pulse oximetry trend data from 1350 patients, 450 in each group, were analysed. During the intervention phase, motivation was increased by adding an explicit instruction to prevent and treat low SpO2 values and making personnel aware that they were being studied (Hawthorne effect). The incidence of hypoxaemia decreased significantly from 17.8% during the pre-intervention phase to 11.6% during the intervention phase (relative risk (RR) 0.65, 95% confidence interval (CI) 0.47-0.90; P < 0.01). The incidence of severe hypoxaemia (SpO2 < or = 85%, 1 min) decreased from 7.8% to 3.3% (RR 0.43, CI 0.24-0.76; P < 0.01). The number of patients who had severe hypoxaemia for more than 5 min decreased from 13 to 1 (RR 0.08, CI 0.02-0.36; P < 0.01). In the post-intervention period, the incidence of hypoxaemia returned to pre-intervention values. The results of this study suggest that motivation of care providers to prevent and treat low SpO2 is an important determinant of postoperative hypoxaemia in the recovery room.

Validation of a clinical prediction rule to reduce preoperative type and screen procedures

BACKGROUND

We have developed a prediction rule for the occurrence of perioperative red blood cell transfusion to help to reduce the number of unnecessary preoperative type and screen procedures. We evaluated the robustness of this prediction rule in patients from another hospital.

METHODS

The rule was retrospectively applied to 1282 consecutive patients ('validation set') who underwent similar surgical procedures to the patients in the derivation study. The outcome was similarly defined as any allogeneic transfusion on the day of surgery or during the first postoperative day. The predictive value of the rule was assessed using a Receiver Operating Characteristic curve (ROC) and compared with the results of the derivation study. Subsequently, the number of correctly predicted transfusions was compared.

RESULTS

The patient characteristics did not differ between the two sets, except for the incidence of transfusion (derivation study: 18%; present study: 8%). In the validation set, the ROC area of the prediction rule was 0.78 (95% confidence intervals [CI]: 0.73-0.82), which was within the CI of the ROC area found in the derivation study (0.75; 95% CI: 0.72-0.79). In total, 35% of the type and screen procedures could be omitted (derivation study: 50%), with 13% missed transfused patients (derivation study: 20%).

CONCLUSIONS

After comparing the results of this validation study with that of the derivation study, the prediction rule was robust and may work in other clinics as well.

Advanced pulse oximeter signal processing technology compared to simple averaging. II. Effect on frequency of alarms in the postanesthesia care unit

STUDY OBJECTIVE

To determine the effect of a new pulse oximeter (Nellcor Symphony N-3000, Pleasanton, CA) with signal processing technique (Oxismart) on the incidence of false alarms in the postanesthesia care unit (PACU).

DESIGN

Prospective study.

SETTING

Nonuniversity hospital.

PATIENTS

603 consecutive ASA physical status I, II, and III patients recovering from general or regional anesthesia in the PACU.

INTERVENTIONS

We compared the number of alarms produced by a recently developed "third"-generation pulse oximeter (Nellcor Symphony N-3000) with Oxismart signal processing technique and a conventional pulse oximeter (Criticare 504, Waukesha, WI). Patients were randomly assigned to either a Nellcor pulse oximeter or a Criticare with the signal averaging time set at either 12 or 21 seconds. For each patient the number of false (artifact) alarms was counted.

MEASUREMENTS AND MAIN RESULTS

The Nellcor generated one false alarm in 199 patients and 36 (in 31 patients) "loss of pulse" alarms. The conventional pulse oximeter with the averaging time set at 12 seconds generated a total of 32 false alarms in 17 of 197 patients [compared with the Nellcor, relative risk (RR) 0.06, confidence interval (CI) 0.01 to 0.25] and a total of 172 "loss of pulse" alarms in 79 patients (RR 0.39, CI 0.28 to 0.55). The conventional pulse oximeter with the averaging time set at 21 seconds generated 12 false alarms in 11 of 207 patients (compared with the Nellcor, RR 0.09, CI 0.02 to 0.48) and a total of 204 "loss of pulse" alarms in 81 patients (RR 0.40, CI 0.28 to 0.56). The lower incidence of false alarms of the conventional pulse oximeter with the longest averaging time compared with the shorter averaging time did not reach statistical significance (false alarms RR 0.62, CI 0.3 to 1.27; "loss of pulse" alarms RR 0.98, CI 0.77 to 1.3).

CONCLUSIONS

To date, this is the first report of a pulse oximeter that produced almost no false alarms in the PACU.

Advanced pulse oximeter signal processing technology compared to simple averaging. I. Effect on frequency of alarms in the operating room

STUDY OBJECTIVE

To determine the effect of a new signal processing technique (Oxismart, Nellcor, Inc., Pleasanton, CA) on the incidence of false pulse oximeter alarms in the operating room (OR).

DESIGN

Prospective observational study.

SETTING

Nonuniversity hospital.

PATIENTS

53 ASA physical status I, II, and III consecutive patients undergoing general anesthesia with tracheal intubation.

MEASUREMENTS AND MAIN RESULTS

In the OR we compared the number of alarms produced by a recently developed third generation pulse oximeter (Nellcor Symphony N-3000) with Oxismart signal processing technique and a conventional pulse oximeter (Criticare 504). Three pulse oximeters were used simultaneously in each patient: a Nellcor pulse oximeter, a Criticare with the signal averaging time set at 3 seconds (Criticareaverage3s) and a similar unit with the signal averaging time set at 21 seconds (Criticareaverage21s). For each pulse oximeter, the number of false (artifact) alarms was counted. One false alarm was produced by the Nellcor (duration 55 sec) and one false alarm by the Criticareaverage21s monitor (5 sec). The incidence of false alarms was higher in Criticareaverage3s. In eight patients, Criticareaverage3s produced 20 false alarms (p < 0.01).

CONCLUSIONS

Our study did not show a beneficial effect in the OR on the incidence of false alarms of the Nellcor monitor with Oxismart signal processing compared with the Criticare monitor with the longer averaging time of 21 seconds.