Article Text


Organisation of respiratory care
P223 One blood gas is not enough to assess a patient for LTOT—how to kiss goodbye to circa £10 million in England
  1. C Loutsios1,
  2. G Logan2,
  3. C Deeming2,
  4. D Powrie1,
  5. A Davison1
  1. 1Southend University Hospital, Southend on Sea, UK
  2. 2South East Essex PCT, Southend on Sea, UK


It is recommended in the recent NICE Clinical Guidelines on COPD (June 2010) that stable patients should be offered LTOT if the PaO2 <7.3 kPa or >7.3 and <8 kPa with associated features; assessment should be made by measuring arterial blood gases on two occasions at least 3 weeks apart in confirmed stable COPD with optimum management. This was the criteria for entry in the MRC and NOTT LTOT trials. This is currently our practice. It has also been recently suggested as part of the COPD National Strategy that only one measurement of blood gases may be necessary. The annual spend on tariffs including LTOT in England in 2006 was £35 500 000. 38 patients were started on LTOT by the South East Essex Oxygen Service in 1 year from March 2009. In addition 11 stable COPD patients had a blood gas measurement in respiratory outpatients and had a PaO2 <7.3 kPa (Mean 6.79 SD 0.4) and when repeated by the oxygen team was above 7.3 kPa. These patients therefore did not meet the criteria for LTOT. One patient subsequently did meet the criteria within the year. This suggests that at least an extra 28% of patients would have been prescribed LTOT if only one initial blood gas below 7.3 kPa was used to assess for eligibility for LTOT. This could produce an extra spend of approximately £10 million on LTOT if extrapolated across the whole of England. These results suggest that there is considerable variability in PaO2 in hypoxic patients over time. These results also support the current NICE Clinical Guideline which recommends two measurements of arterial blood gases at least 3 weeks apart should be made before prescribing oxygen. This was the evidence base for starting LTOT in randomised controlled trials.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.