oxygen therapy, then and now

Can Respir J Vol 12 No 2 March 2005 67

Dr Warren (pages 81-85) has written a delightful historyabout the pioneers in oxygen therapy and the twoCanadian heroes of medicine who had different views

about the topic. Osler and Meakins sound like the voices oftwo schools of medical thinking that go back as far as we cansee in history. The story of oxygen therapy is part of the historyof evidence-based medicine.

Ancient Greek physicians were divided into rival schools.Most, like the pneumatists and dogmatists, had elaborate theo-ries about health and disease from which they confidentlyderived recommendations for treatment. Opposing them werethe empiricists, who were unconvinced by explanationsinvolving elements, humours, pneuma, seasons, climate, com-plexions and diet. They rejected the theories of pneumatistsand dogmatists and insisted that treatments be based purely onexperience. Dogmatists and pneumatists in turn scorned thevalidity of conclusions drawn from experience. Modern physi-cians would agree with both critiques. We now see the ancienttheories as worthless, and the empiricists’ evidence cannothave been better than level 4.

Both our theory and our ability to draw correct conclusionsfrom experience have advanced somewhat since the time ofGalen, but it is still true that there are few clinical situationswhere we understand the biology so well that we can give anovel treatment with the confident expectation that it willprove safe or effective, and few situations where we have out-come data that are so good that we can be sure a given patientwill benefit from the treatment. As a result, the old discussionbetween schools of thought is still going on. Nowadays, theempiricists have turned into evidence-based medicine doctorswhile the pneumatists and dogmatists have turned into scien-tists (physiologists, biochemists, cellular and molecular biolo-gists and pharmacologists). Most of us are now trained in bothevaluation of clinical outcome data and biological science, butwe are all either empiricists or scientists at heart. When neithertheory nor outcome give clear guidance, the empiricists amongus are tempted to try treatments for which outcomes data areonly ‘suggestive’, whereas scientists are tempted to try treat-ments that current biology theory predicts will work. (Thereare also the skeptics, who fall back on the primum non nocereprinciple.)

In the oxygen therapy story, Meakins sounds like a scientist.He was impressed by data showing that oxygen saturation wentup with oxygen therapy. Osler, on the other hand, sounds likean empiricist. He had heard about oxygen toxicity and worried

that oxygen might do more harm than good. He was waiting tosee evidence of benefit – not proof of a change in oxygen satu-ration, but proof that death rates go down. If they had livedanother 80 years, Osler would be a hard-headed supporter ofclinical trials, rather like the current editor of this Journal, whileMeakins would continue to produce physiological data thatgive very good reasons to perform new clinical trials.

The early studies of oxygen therapy make a catalogue of pit-falls in clinical investigation. Joseph Priestly, who isolated oxygenfor the first time and knew some of its biological effects,inhaled some himself and noted an agreeable glow and light-ness of the chest. Beddoes, a chemist, took oxygen and notedthat his fingertips turned carmine in color. It is possible that hehad severe, chronic respiratory disease with cyanosis and hisobservation was valid, but it is more likely that, given what weknow now about ‘central and peripheral’ cyanosis and the dif-ficulty of perceiving cyanosis, he either imagined the colourchange or his excitement caused blood flow in his fingers toincrease. Both scientists illustrate the fallacy of trusting physi-ological observations reported by subjects who are themselvestheorists. JS Haldane and colleagues, having discovered theprinciple of negative feedback homeostatic control of carbondioxide in the blood, performed on themselves the experimentof hyperventilating and then watching their own relaxedbreathing to see if they went apneic. They did, as expected.This is a reproducible experiment. I performed it on myself as afirst year medical student during the respiratory physiologycourse. Subsequent research showed that posthyperventilationapnea occurs only in stroke patients and physiologists whobelieve the simple theory of feedback control of ventilation bychemoreceptors. Similarly, Davy found that oxygen eased alltypes of breathlessness, whereas most properly controlled modernexperiments have shown little effect of oxygen on dyspnea inlung disease. Dr Kellog’s claim about the “wonderful vitalizingand invigorating influence of oxygen” can be trusted about asmuch as his conviction that oxygen enemas were good for liverdisease. He may have been a quack, but his evidence was asvalid as that of Davy and Beddoes. He might have been quotingthem.

These early modern scientists, and others like them, hadsuch confidence in their theories about oxygen that they feltno need to ask questions about clinical outcomes. Theirauthoritative pronouncements were probably very importantin establishing the myth of oxygen (ie, a strong and widespreadbelief that oxygen can both treat disease and improve health in

Oxygen therapy, then and now

William Whitelaw MDCN PhD FRCPC

GUEST EDITORIAL

Department of Medicine, University of Calgary, Calgary, AlbertaCorrespondence: Dr Whitelaw, Heritage Medical Research Building, 3330 Hospital Drive Northwest, Calgary, Alberta T2N 1N4.

Telephone 403-220-6868, fax 403-270-8928, e-mail [email protected]

©2005 Pulsus Group Inc. All rights reserved

whitelaw_8617.qxd 3/4/2005 2:54 PM Page 67

ways that go far beyond any evidence base) that prevails today.The oxygen myth is the foundation of industries that sell oxygen-enriched air in ‘oxygen bars’, oxygen from tanks to revive tiredfootball players on the bench, hyperbaric chambers for allkinds of illnesses, and oxygenated water as an invigoratingdrink. On my desk is a recently manufactured bottle ofantiphlogistic tablets, proof of the living influence of JosephPriestly, who called his oxygen dephlogisticated air.Fortunately for the vulnerable public, antioxidant tablets arealso sold in abundance.

The first reports on the use of oxygen in patients from the1895 British Medical Association meeting described in Dr Warren’s article are more in keeping with expectationsbased on modern experience. Oxygen was given to patientswith cardiac or cardiorespiratory disease, but only in desperatecases with cyanosis. It had only a palliative effect, reducingsymptoms, but only “maintaining life a little longer”. Patientswith pneumonia or heart failure severe enough to causecyanosis are expected to show improvement in cerebral func-tion, change in colour and reduction of pulse rate with oxygen.The next step might have been to study the biology of oxygentreatment or to assess its clinical effectiveness more thoroughly.

Meakins concentrated his attention on measurements ofoxygen, first showing that cyanosis in severe lung disease wasnot due to methemoglobin but, in fact, to low oxygen.Studying the treatment, he showed that oxygen administrationcould correct saturation in patients with low oxygen. Hisadmirable preoccupation was with measurable physiologicalvariables. Having seen oxygen levels rise with oxygen treat-ment he was certain that oxygen treatment was important. Helived in the era before modern empiricists gave their school thenew evidence-based medicine name and made it commonplaceto ask the ‘so what’ question that Osler was interested in. Inthe paper reporting 10 cases of pneumonia with moderate

desaturation, of whom only three had improvements in symp-toms with oxygen, Meakins concluded that oxygen “isabsolutely indicated to relieve or remove the ill effects of avery dangerous condition”. Although one of the first large-scale comparative outcome studies in medicine was performedon pneumonia patients in Paris by PCA Louis in the 1820s toassess the value of bloodletting, such studies were never donefor oxygen in acute disease. Barach’s paper promoted oxygentherapy for pneumonia because it made the patients feel betterbut five of his 16 patients died and he thought oxygen had noeffect on prognosis, because that was about the death rate hewas used to seeing without oxygen. At the end of the story, asDr Warren points out, oxygen became accepted treatment forpneumonia and various other acute respiratory diseases with-out benefit of any convincing systematic evidence of effective-ness. It also became accepted treatment for acute coronarysyndromes with normal oxygen and cancer palliation, wherethe myth of oxygen makes it a powerful placebo.

Where are we now? For chronic respiratory disease we havegood evidence from long-term oxygen trials for the use of oxy-gen treatment in stable, hypoxemic, chronic obstructive pul-monary disease patients. But an unplanned spin-off of thosetrials has had a huge influence on oxygen treatment in allkinds of patients. The 85% oxygen saturation value, the cutoffvalue chosen for entry into those trials, has somehow becomea magic number, the saturation criterion for prescribing oxygenin any acute or chronic case. Patients, families, and health careprofessionals of all kinds have concerns about the hour-to-hoursafety of any patient with 84% or less oxygen saturation. Thenumber drives discharge dates, short-term oxygen needs andpatient anxiety. We thus prescribe oxygen for our chronicobstructive pulmonary disease patients according to the princi-ples of empiricists, but rely on arguments by analogy and theprinciples of scientists for the rest of our patients.

Can Respir J Vol 12 No 2 March 200568

Editorial

whitelaw_8617.qxd 3/4/2005 2:54 PM Page 68

Submit your manuscripts athttp://www.hindawi.com

Stem CellsInternational

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Immunology ResearchHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinson’s Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttp://www.hindawi.com

oxygen therapy, then and now

Documents