© 2005 American Thoracic Society doi: 10.1164/rccm.2506007
Prescription of OxygenStill Problems after All These YearsUniversity of Edinburgh and NHS Lothian University Hospitals Division, Edinburgh, Scotland
It is almost 25 years since the two landmark trials of long-term oxygen therapy in chronic obstructive pulmonary disease (COPD) showed an improvement in survival in patients with COPD and chronic respiratory failure (1, 2). The levels of hypoxemia for inclusion in these trials were PaO2 less than or equal to 55 mg Hg in the Nocturnal Oxygen Therapy Trial (NOTT) and PaO2 40 to 60 mg Hg for the UK Medical Research Council (MRC) Trial. In the MRC trial the patients had hypercapnia, whereas this was not the case in the NOTT trial. In both the MRC and the NOTT studies, entry into the trials required the demonstration of hypoxemia in a period of clinical stability. These trials established the criteria for the prescription of domiciliary long-term oxygen therapy (LTOT) based on the presence of significant, chronic hypoxemia
There are variations between international approaches to the prescription of long-term oxygen therapy (5). In some countries, notably the United States and Canada, the demonstration of hypoxemia does not require measurement of arterial partial pressure of oxygen, because significant oxygen desaturation, measured by pulse oximetry Adherence to these guidelines remains problematic. Several studies have shown that a major problem is that most prescriptions for LTOT (60–82%) are made based on blood gases measured at the time of an exacerbation of COPD, and that relatively few patients (35–65%) have their blood gases reassessed when clinically stable, to reevaluate the need for LTOT (7–9). To address this problem, Guyatt and coworkers (pp. 573–580) undertook a randomized controlled trial of strategies for assessing eligibility for long-term domiciliary oxygen therapy, the results of which are published in this issue of the Journal (10). In this study all first-time applicants for LTOT in two Canadian centers (Toronto and Ottawa) were randomized to "conventional care," where the procedure for assessment continued in the usual way, with a decision to prescribe LTOT made by the home-oxygen program personnel on demonstration of significant hypoxemia, based on blood gases or oxygen saturation measurements. Patients could also be assigned to an "alternative assessment" group, in which those applicants who had experienced an exacerbation within the previous 2 months had oxygen therapy continued with a repeat assessment of the degree of hypoxemia performed 2 months after the period of instability. At 12 months the eligibility for LTOT was reevaluated based on blood gases or pulse oximetry. The study groups consisted of 276 applicants allocated to the conventional arm and 270 to the alternative assessment. The main results of the study were that the alternative assessment, which required reevaluation of oxygen therapy when patients were clinically stable, reduced the need for oxygen therapy, compared with those in the conventional assessment group. There were no differences in mortality, health-related quality of life, or resource use between the groups. Mortality was high, close to 20%, 1 year after randomization or approximately 500 days after beginning oxygen therapy. Indeed, of those randomized to the study, 45 in the conventional and 35 in the alternative groups died between the time the patient had started oxygen and the time of the proposed randomization. The study also showed that in the alternative assessment group, 148/270 patients who were eligible on the basis of information available at the time of randomization were shown to be clinically unstable when assessed. Of these, only 33% proved to be eligible on reassessment; 31% were found to be ineligible, 24% had oxygen discontinued, and 7% had died at reassessment at 2 months. These data support previous studies which suggested that LTOT could be discontinued in over a third of patients who were originally eligible for oxygen therapy when a reevaluation was performed (11). A fairly extensive health economic evaluation was performed between the two groups which showed that the alternative assessment group had higher costs, by Can $155 or U.S. $309 per applicant, but the authors suggested that these costs can be offset by the reduction in home oxygen costs. However, it would be difficult to extrapolate this cost analysis to other health care systems, and indeed there were differences in the economic evaluation comparing Canadian and American systems. In addition, even in the North American systems, the health care costs saved by central government may not be used to directly compensate the increased cost incurred by the alternative strategy. At 12 months after the initial assessment for LTOT, a large number of patients in both groups had their LTOT discontinued, with 15% more patients discontinuing LTOT in the conventional arm than in the alternative arm. The authors suggest the reason for this is that some patients may take more than 2 months to recover their "normal" arterial blood gas values after an exacerbation of COPD, although they base this suggestion on studies in which recovery was based on symptoms and health status rather than on blood gas values (12, 13, and 14). However, one previous study, based on arterial blood gases measured during recovery from an exacerbation of COPD, suggested that at least 3 months should elapse after an exacerbation before considering remeasuring blood gases at a time of clinical stability (15). The authors also suggest that patients should be reassessed 1 year after the prescription of LTOT, and that a proportion will have become ineligible due to improvements in blood gases and can have LTOT discontinued. There is controversy over whether oxygen therapy should be discontinued at this point if the original assessment was performed when the patient was in a stable condition. There are some indications that oxygen therapy may have reparative effects and that discontinuation of LTOT, consequent upon improving blood gases over time, can lead to subsequent deterioration in the patient's condition (16). Overall, the message from this study is that better adherence to recent guidelines (3) is needed, particularly ensuring that patients are assessed when in a stable state or, if this is not possible, then reassessed at least 2 months from the last exacerbation. This study shows that this approach will ensure that the correct patients continue to receive oxygen therapy. FOOTNOTES Am J Respir Crit Care Med Vol 172. pp 517–522, 2005 Internet address: www.atsjournals.org Conflict of Interest Statement: W.M. has been reimbursed for travel by GlaxoSmithKline (GSK), Zambon, AstraZeneca (AZ), Boehringer Ingelheim, Pfizer, and Micromet for attending conferences. He has received honoraria from GSK, AZ, Zambon, and Pfizer for participating as a speaker in scientific meetings. He serves on Advisory Boards for GSK, Pfizer, Almicall, Amgen, Bayer, and Micromet. He serves as a consultant for Pfizer and SMB Pharmaceuticals. Research grants to support work carried out in W.M.'s laboratory come from SMB, Pfizer, Ceremedix, GSK, Chugi, and Novartis. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
, with or without hypercapnia, in a patient with severe COPD (
, suffices for the prescription of LTOT, and the measurement does not need to be made in a period of clinical stability. In most European countries, on the other hand, measurement of arterial oxygen partial pressures are required during a period of clinical stability. Indeed, in the United Kingdom more than one measurement is necessary to confirm significant chronic hypoxemia (
