An audit of level two and level three checks of anaesthesia delivery systems performed at three hospitals in South Australia.

Anaesthetists may subject patients to unnecessary risk by not checking anaesthetic equipment thoroughly before use. Numerous adverse events have been associated with failure to check equipment. The Australian and New Zealand College of Anaesthetists and anaesthetic delivery system manufactures have made recommendations on how anaesthetic equipment should be maintained and checked before use and for the training required for staff who use such equipment. These recommendations are made to minimise the risk to patients undergoing anaesthesia. This prospective audit investigated the adherence of anaesthetic practitioners to a selection of those recommendations. Covert observations of anaesthetic practitioners were made while they were checking their designated anaesthetic machine, either at the beginning of a day's list or between cases. Structured interviews with staff who check the anaesthetic machine were carried out to determine the training they had received. The results indicated poor compliance with recommendations: significantly, the backup oxygen cylinders' pressure/contents were not checked in 45% of observations; the emergency ventilation device was not checked in 67% of observations; the breathing circuit was not tested between patients in 79% of observations; no documentation of the checks performed was done in any cases; and no assessment or accreditation of the staff who performed these checks was performed. It was concluded that the poor compliance was a system failing and that patient safety might be increased with training and accrediting staff responsible for checking equipment, documenting the checks performed, and the formulation and use of a checklist.

Identification of movement artefact by the Nellcor N-200 and N-3000 pulse oximeters.

OBJECTIVES: The Nellcor N-3000 pulse oximeter is designed to be able to identify signal artefact related to movement of the body part to which the probe is attached. It may therefore provide a reliable means of monitoring arterial oxyhemoglobin saturation (SpO2) in awake, moving patients. This study compared the Nellcor N-3000 and N-200 pulse oximeters in terms of their ability to identify readings associated with movement, in a group of volunteers making standardized movements. METHODS: Thirty-six volunteers were studied. Volunteers breathed room air throughout the study. SpO2 of each volunteer was monitored by both a Nellcor N-200 and a Nellcor N-3000 simultaneously on both hands. Volunteers made a series of five standardized movements, each lasting one minute, with each hand during the monitoring session, while SpO2 and oximeter status were recorded from all four oximeters. The mean SpO2 reading was calculated during each movement. SpO2 readings which the oximeter identified as being associated with movement, pulse search not locked, sensor not attached, or break in communications were excluded from analysis. RESULTS: The N-3000 rejected from 17 to 78% of readings taken during movement, compared to 0 to 2% with the N-200. Although the remaining readings of both types of oximeters were subject to some movement artefact, which led to spuriously low SpO2, this was significantly less with the N-3000. CONCLUSIONS: The Nellcor N-3000 pulse oximeter is able, to some extent, to identify movement artefact. It should offer an advantage over the N-200 when monitoring moving patients.

Evaluation of the influence of movement on saturation readings from pulse oximeters.

This study aimed to develop a protocol for assessing the influence of movement on oxyhaemoglobin saturation readings from pulse oximeters. Thirty-six volunteers took part in the study. In each volunteer, each hand was monitored by both a Nellcor N200 oximeter using a disposable probe and by a Datex Satlite DS103 oximeter using a clip-on finger probe. Volunteers made five standardised movements during which output was recorded from all four oximeters. All movements were associated with apparent decreases in oxyhaemoglobin saturation which were statistically significant for two movements with the Nellcor equipment and for four movements with the Datex equipment. Movement was associated with increases in the magnitude of pulse amplitude, but this was not quantitatively associated with magnitude of artefactual changes in saturation. Use of this standardised movement protocol allows quantification of movement artefact from pulse oximeters and should facilitate the development of equipment less affected by movement.

The effect of fentanyl administered epidurally by patient-controlled analgesia, continuous infusion, or a combined technique of oxyhaemoglobin saturation after abdominal surgery.

The aims of this study were to determine the effect of three different modes of epidural administration of fentanyl on oxyhaemoglobin saturation and pain control. Forty-three patients undergoing elective abdominal surgery were randomly allocated to the following groups: (1) continuous infusion of fentanyl at a rate of 50 micrograms.h-1 with additional epidural boluses (25 micrograms) as required; (2) patient-controlled analgesia using a 25 microgram epidural bolus of fentanyl with a 15 min lock-out period; (3) a combination of patient-controlled analgesia and continuous infusion. Oxyhaemoglobin saturation was measured by continuous computerised pulse oximetry for 48 h after operation together with pain and sedation scores. In the first 24 h after surgery patients in the continuous infusion group spent a significantly greater proportion of time below oxygen saturations of 94% and 85% than those in the other two groups. On day 2 all oxygen saturation measurements were worse than during day 1, but differences between groups were not significant. Those patients receiving patient-controlled analgesia required significantly less fentanyl than patients in either of the other groups (p < 0.05). However, the mean pain and sedation scores did not differ significantly between the three treatment groups. There was no association between total fentanyl dose and oxygen saturation values. Overall, self-administered fentanyl appeared to cause less oxyhaemoglobin desaturation than nurse-administered analgesia without any loss of analgesic effect.

Oxyhaemoglobin saturation following elective abdominal surgery in patients receiving continuous intravenous infusion or intramuscular morphine analgesia.

Oxygen saturation was continuously measured using computerised pulse oximetry for 8 h overnight pre-operatively and for the first 24 h postoperatively in 40 patients receiving intermittent intramuscular morphine or continuous infusion of morphine following elective upper abdominal surgery. The proportion of time with an oxygen saturation less than 94% was used as an index of desaturation. Patients receiving continuous infusion analgesia received a larger morphine dose and achieved better analgesia than the intramuscular group. Postoperatively, the duration of desaturation increased 10-fold over pre-operative values, 'intramuscular' patients spending 39.0% (SD, 37.0%) and 'continuous infusion' patients 40.0% (SD, 37.5%) of the time below 94% saturation. Although newer therapies (e.g. epidural analgesia and patient-controlled analgesia) are currently receiving greater attention, the sequelae of these more traditional analgesic techniques warrant further study.

An evaluation of intravenous infusion pumps and controllers.

The accuracy, safety, reliability and cost of use of 35 intravenous infusion pumps and 3 flow controllers were assessed. When infusing saline 11 out of 17 syringe pumps, 3 out of 5 peristaltic pumps, 1 out of 2 roller pumps and all 14 cassette pumps tested were accurate to within 5% over their full ranges of operation. There was no significant change in the performance of any of the pumps tested when saline was infused through a standard resistance, except in the cases of the 3 flow controllers which were unable to infuse at all against the resistance. When 50% dextrose was infused, delivery by two peristaltic pumps was reduced by 23 and 38%. No pump cut out or alarmed at pressures of up to 200 mmHg and 21 pumps continued to infuse against pressures of 750 mmHg or greater. Surges of up to 0.5 ml occurred after release of an outlet obstruction. One device was fitted with a variable high pressure alarm. This device could also measure pressure in the infused vessel and was found to be accurate for measurements of central venous pressure. The cost of consumables for a single use for syringe pumps ranges from A$2 to $5, for peristaltic and roller pumps from A$1 to $10, and for cassette pumps from A$7 to $12, with an additional A$2 for a burette. Accurate delivery of intravenous fluids and drugs is available but is expensive and requires the operator to be specially trained. No simple, cheap, accurate device is yet available.