Improving the safety of emergency oxygen therapy

Oxygen (O2) is used in over two million (34%) ambulance journeys in the UK each year. About 17.5% of inpatients in UK hospitals receive emergency oxygen therapy at any one time, which equates to about 18,000 patients per day.(1) It is probably the most commonly used drug for the care of patients who present with medical emergencies. Despite this oxygen can often be the forgotten emergency medical therapy. Oxygen physiology is a multifactoral, complex process, and therefore any oxygen therapy must be flexible to adapt to the patients variable demands. The traditional approach of providing as much oxygen as possible for patients during the first stages of a medical illness is flawed. Too much, as well as too little, oxygen can be dangerous and even potentially fatal in certain patient groups.

In October 2008, the British Thoracic Society (BTS) published guidelines for emergency oxygen use in adult patient.(2) Shortly after this, the National Patient Safety Agency (NPSA) published a Rapid Response Report entitled Oxygen Safety in Hospitals.(3) The key message from both these reports is that O2 is a treatment for hypoxia. It is not used to treat breathlessness in non-hypoxic patients. O2 should be prescribed according to a target saturation range. Saturations must then be monitored and O2 adjusted to achieve targets. These targets were established to take account of the risk of carbon dioxide (CO2) levels in patients with COPD. This simple message is key to good oxygen treatment but can often get lost in talk about blood gases, O2 mask types and oxygen flow rates. Without a focus within organisations the prescribing and monitoring of O2 is often neglected. By championing appropriate oxygen use, the ongoing BTS audits have demonstrated that hospitals can improve the prescribing and administration of O2 to their patients.(4)

In the majority of patients, oxygen should be regarded as short-term treatment for hypoxia until the underlying cause of the acute illness can be corrected. On the whole, giving too much oxygen to patients in the short-term will have little effect. This can lead to complacency about using oxygen and does not take into account the fact that O2 has the potential to cause serious harm to a small group of patients.

CO2, although in low concentrations in exhaled breath, will cause sedation and even suppress respiration at elevated levels: so-called CO2 narcosis. Those most at risk to this problem are patients with COPD who normally run high CO2 levels (pCO2) as their respiratory centres have become insensitive to CO2, and therefore depend on a relative hypoxia to drive their breathing. COPD patients who receive titrated oxygen treatment in the pre-hospital setting have been shown to have a reduced mortality compared with those on high flow rate O2.(5) For this reason, the BTS guidelines advise that these patients’ oxygen saturations be maintained between 88% and 94%. This level of O2 is unlikely to increase the CO2 to dangerous levels in the short term, thereby giving emergency teams time to transport these patients to a medical centre where an arterial or capillary blood gas can be carried out, more to ensure the CO2 level is safe than to check if the O2 level is adequate.

Clinical trials have looked at using supplemental oxygen in strokes, cardiac arrests and intensive care. Although the results of these trials are less conclusive, there is still evidence that adjusting oxygen therapy to achieve target O2 saturations is beneficial. A retrospective study of over 36,000 intensive care admissions in The Netherlands demonstrated that both hypoxia and hyperoxia were independently associated with higher mortality within the first 24 hours of admission.(6) A study looking at patients post-resuscitation following cardiac arrest demonstrated that hyperoxia was associated with higher in-hospital mortality compared with either hypoxia or normoxia.(7) In a third study, patients with acute stroke were investigated, and high-flow oxygen was associated with increased mortality for mild-to-moderate stroke as classed by the Scandinavian Stroke Scale.(8) All these support the appropriate targeting and titrating of O2 therapy to achieve specific O2 saturations.

Much has been made about increasing the profile of appropriate antibiotic prescribing in the last decade. This focus has improved the use of antibiotics and reduced antibiotic associated complications. Key elements have been the involvement of all practitioners concerned and a clear leadership and championing of the cause.

At County Durham and Darlington Foundation Trust (CDDFT), concerns were raised in 2009 about how oxygen was being used in the acute setting. A two-week audit demonstrated that over 19% of adult acute medical patients were receiving emergency oxygen but only 73% of these had appropriate monitoring of oxygen saturations.(9) Less than 6% of patients had their oxygen saturations defined and no patients had a prescription for oxygen recorded. These poor results led a group of interested clinicians to resolve to promote good emergency oxygen practice. The group was multidisciplinary, involving pharmacists, medics, specialist nurses and ward-based nursing staff. This was a small group with no extra resources allocated to improve the situation. Using the example of antibiotic stewardship we designed our own programme to improve the safe use of emergency oxygen therapy.

A multi-pronged approach was developed looking at the reasons why oxygen was not prescribed and monitored appropriately. This was broadly split into two areas:

1. Improving policies and processes

2. Education and training.

Staff involved in the use and monitoring of oxygen were consulted. These included healthcare assistants, porters, nurses, medics and pharmacy. Particular attention was paid to the reasons why oxygen was not monitored or prescribed appropriately. What became very clear was that the paperwork and procedures either dis-incentivised or, in some cases, actually prevented proper prescribing and administration of oxygen.

Oxygen is classed as a medicinal product in the UK, with a general sales license. This means that oxygen can be sold or supplied without a prescription. The reason for this status is that oxygen can then be supplied direct to patients in the community by wholesalers without the involvement of community pharmacies. The legal status of oxygen is equally valid for patients in hospital. A prescription for oxygen is not legally required; however, clinical governance requires that the intentions of the clinician starting the oxygen are conveyed to the person administering the oxygen therapy. Similarly, the person administering the oxygen needs to keep clear records of what has been administered to the patient.

Using the templates provided by the BTS, both inpatient prescriptions and patient monitoring charts were redesigned. This was to allow the intentions of the healthcare professional to be clear when the oxygen therapy was initiated. Nurses could then effectively monitor and record the patients’ response to oxygen therapy. Nurses were empowered to adjust the patients’ oxygen to achieve target saturations using flow charts developed and included in hospital guidelines. The flow charts provided direction on how to titrate up or wean patients off oxygen depending upon how they responded. Also included in this guide was advice about the most appropriate choice of oxygen delivery device.

Clear roles and responsibilities were established for all staff involved in the use of oxygen. A standard operating procedure was developed to ensure that nursing staff and healthcare assistants were appropriately supported. An oxygen patient group direction (PGD) was also developed to encourage nursing staff to take ownership of the adjustment of the oxygen to achieve target saturations. Although technically not required, a PGD was used because nursing staff were familiar with the format and they felt most comfortable with its use to support their practice.

With the introduction of a raft of new paperwork and policies, there was clearly a need for an extensive training programme. Barriers to effective implementation of good oxygen stewardship were those common to the roll out of best antibiotic management in the last decade. Lessons learnt from antimicrobial stewardship included local cooperation, sustained education, good care bundles and executive level planning.(10) Local cooperation was developing well with the enthusiastic work of the oxygen group. The challenges were how to ensure effective and sustained education with no extra resources. The decision was to focus on one distinct area and then to expand from there.

The acute medical admissions unit at Darlington Memorial Hospital was chosen because it was the highest user of O2 and had a motivated team to implement the required changes. Using the BTS emergency O2 teaching aid for nurses, a short e-learning style package was developed; this was supported by the pharmacist and respiratory nurses carrying out one-on-one teaching on an ad hoc basis when the clinical situation allowed. Healthcare assistants were also included because they are often the people who perform the majority of patient monitoring roles. Emphasis within the teaching was on appropriate monitoring of patients and the selection of the most suitable oxygen delivery method.  Also basic emergency O2 use and practice was built into the nurse annual essential training program.

The more challenging area for education was medical staff. The junior staff are often working on short rotations or on call infrequently, dependent upon their other training commitments. The transient nature of the medical workforce means that training needs to sustained, as highlighted by lessons from antibiotic stewardship. Instead of focussing on just those working on medical admissions, oxygen training was included in the induction of all junior medical staff. The learning was reinforced on ward rounds, and during clinical conversations on the ward by respiratory specialists, medics, nurses and pharmacists. Ward nursing staff also questioned and encouraged good practice as they had been advised during their training.

The medical admissions unit was re-audited following the introduction of the oxygen stewardship programme. The percentage of patients on emergency oxygen had fallen to 7.3% with all 100% of these patients having regular four hourly oxygen saturation monitoring. Over 90% of patients had an oxygen target saturation defined and 79% had oxygen prescribed on their drug chart. The stewardship principles have been in place for over a year now and there have been no untoward incidents where hypoxia has contributed to the problem. It is important that the emphasis on good oxygen management does not lead to patients who are hypoxic not receiving the oxygen treatment they need.

Within the current climate of healthcare, cost implications need to be considered, and some cost benefits have been realised by optimising the treatment of hypoxic patients. Not only has there been a reduction in the use of oxygen but there has also been one associated saving with oxygen apparatus. The appropriate use of oxygen has reduced the amount of tubing and number of masks required. Anecdotally there has also been a reduction in the number of patient transfers requiring bottled oxygen, although this has not been recorded.

The next stage of the project is to ensure that standards are kept high and that 100% of patients have oxygen prescribed appropriate. The message is now being rolled out to the wider Trust. By implementing a comprehensive oxygen stewardship programme, including guidelines, education and audit, the appropriate use of oxygen is improving at the study hospital. Benefits include improved patient safety, reduced cost and engender a multidisciplinary approach to the management of the hypoxic patient.

• Oxygen is a treatment for hypoxia not breathlessness.
• Oxygen should be prescribed to achieve a target oxygen saturation.
• Patients on should be monitored and have their oxygen saturation kept within target range by adjusting therapy.
• Care must be exercised for all patients at risk of hypercapnic respiratory failure
• An oxygen stewardship programme has the potential to improve the management of acutely ill hypoxic patients.

1. NHS Improvement. Emergency Oxygen. 2013 www.improvement.nhs.uk/lung/NationalImprovementProjects/Emergencyoxygen.aspx (accessed 20 May 2013).
2. O’Driscoll BR, Howard LS, Davison AG. Guideline for emergency oxygen use in adult patients. Thorax 2008 63:Suppl VI.
3. National Patient Safety Agency. NPSA Rapid response report. Oxygen safety in hospitals. NPSA/2009/RRR006.  September 2009. www.nrls.npsa.nhs.uk/resources/?EntryId45=62811 (accessed 20 May 2013).
4. British Thoracic Society. BTS baseline audit on oxygen use. September 2008. www.brit-thoracic.org.uk (accessed 20 May 2013).
5. Austin MA et al. Effect of high flow oxygen on mortality in chronic obstructive pulmonary disease patients in prehospital setting: randomised controlled trial. BMJ 2010;341:c 5462.
6. de Jonge E et al. Association between administered oxygen, arterial partial oxygen pressure and mortality in mechanically ventilate intensive care unit patients. Crit Care 2008;12:R156.
7. Kilgannon JH et al. Association between arterial hyperoxia following resuscitation from cardiac arrest and in-hospital mortality. JAMA 303:2165-71.
8. Ronning OM, Guldvog B. Should stroke victims routinely receive supplemental oxygen? : A quasi-randomised controlled trial. Stroke 1999;30: 2033-7.
9. Gibson D, Tedd H, Foden AP.  In house audit on appropriate emergency O2 administration to acutely ill medical patients.  CDDFT 2009.
10. Bal AM, Gould IM. Antibiotic stewardship: overcoming implementation barriers. Curr Opin Infect Dis 2011;24:357-62.