Intended for healthcare professionals


Dysfunctional breathing and asthma

BMJ 2001; 322 doi: (Published 05 May 2001) Cite this as: BMJ 2001;322:1075
  1. Duncan Keeley, general practitioner,
  2. Liesl Osman, senior research fellow
  1. Health Centre Thame, Oxfordshire OX9 3JF
  2. Chest Clinic, Aberdeen Royal Infirmary, Aberdeen AB25 2ZN

    It is important to tell the difference

    General practice p 1098

    General practitioners and emergency departments from time to time see patients with asthma who appear very breathless, with fast deep breathing and wheeziness, who complain of tingling lips and hands and who recover quite rapidly after breathing in and out of a paper bag and then using a few puffs of salbutamol. Asthma and anxiety with dysfunctional breathing are both common conditions and they often coexist. Indeed, a paper in this week's issue suggests a very high prevalence of dysfunctional breathing among patients with asthma.1 There are reasons to doubt the prevalence suggested by this paper, but the overlap between anxiety and asthma nevertheless creates a problem for patients and their doctors since we seem not to be very good at telling the difference.

    Several studies have shown that patients with asthma have significantly higher anxiety scores than normal and are more likely to have clinically diagnosed panic disorder. 2 3 Conversely, patients with panic disorders, hyperventilation, or “overbreathing” may have unidentified airways reversibility.4 Demeter investigated 47 patients referred for hyperventilation syndrome using methacholine challenge and reversibility testing and judged 38 of them to have asthma.5 The hyperventilation symptoms were eliminated in 29 with a combination of explanation and bronchodilators.

    Thomas et al sent the Nijmegen questionnaire, an instrument designed to identify the so called hyperventilation syndrome, to all adults in one general practice with a diagnosis of asthma and at least one prescription for an asthma drug in the previous year (p 1098).1 Of those who responded 29% had scores indicative of dysfunctional breathing. The authors suggest that “a large minority of patients may be experiencing avoidable morbidity because of inappropriate diagnoses and ineffective treatment.” Their study reminds us of a very real and important problem but overestimates its size: we do not believe that nearly a third of patients in general practice with a diagnosis of asthma have been wrongly diagnosed.

    Firstly, the 16 item Nijmegen questionnaire was not validated in an asthmatic population. Its 91% sensitivity and 95% specificity for physician diagnosed hyperventilation syndrome were shown in a study comparing a group of physician diagnosed non-asthmatic hyperventilators with a group of non-asthmatic normal controls.6 Several of the questions relate to symptoms such as shortness of breath, pain and constriction in the chest, and feeling tense—symptoms common to asthma and dysfunctional breathing. This necessarily impairs the ability of the questionnaire to make the latter distinction.

    Secondly, the very existence of a discrete “hyperventilation syndrome” has been questioned by research looking for, and failing to find, reliable correlations between panic, overbreathing, and hypocapnia. One reviewer of this work suggests that hyperventilation syndrome is a chimera.7 Clinicians may respond that, though unable to define a chimera, they know one when they see one. Nevertheless, our understanding of the interaction between physical symptoms, physiological disturbances, and cognitive perceptions in anxiety—and in asthma—remains limited.

    What should we do about the overlap between the symptoms of asthma and of anxiety? Firstly, because straightforward misdiagnosis is possible we must perform careful and repeated history taking, examination, and physiological measurements—particularly peak flow diaries. We should not assume that an earlier diagnosis was correct, especially when computerised records carry terse definitive-looking diagnoses from earlier years without providing the information on which the diagnosis was based.

    Secondly, we must routinely assess the extent and effect of the anxieties of our asthmatic patients. We can then seek to allay them and avoid stepping up asthma treatments inappropriately when anxiety, hyperventilation, or laryngeal dysfunction are the problem, not worsening asthma: Hyland showed that a higher rate of corticosteroid prescribing was significantly associated with higher levels of panic or fear independent of lung function.8 Fears about treatments are especially important since they may impair self management of both continuing symptoms (“inhaled steroids are bad for you”) and acute attacks (“it is dangerous to exceed the stated dose of a bronchodilator”). Confidence in self management is vital if the inevitable anxiety associated with having asthma is to be minimised. Screening questions for excessive anxiety should be non-respiratory (Do you ever feel that something awful is about to happen? How often do worrying thoughts go through your mind?), and we should ask about patients' family and social backgrounds to learn of the predicaments that may cause their anxiety.

    Thirdly, for some patients, hyperventilation with symptoms of hypocapnia is part of their experience of asthma. We must explain the similarities and differences between the symptoms of worsening asthma and hypocapnia and the side effects of increased doses of bronchodilators—all of which may be experienced during an attack. The guidance should be: if in doubt, treat for asthma but try also to slow down the breathing rate to avoid hypocapnia. Some patients may find peak flow measurement helpful in distinguishing between bronchospasm and hyperventilation. All need to know that the symptoms of hypocapnia and of higher dose bronchodilators, though unpleasant, are not dangerous.

    Thomas et al suggest that breathing therapy is appropriate for some patients. But there is no good evidence that breathing therapy benefits patients with asthma. The studies they describe were carried out among patients without asthma. In 1990 Howell rejected management of behavioural breathlessness by breathing training and recommended sympathetic explanation aimed at giving patients reassurance and insight and at “removing the frightening element of the experience.”9 This approach may still offer the most practical way of helping patients with asthma cope with anxiety. Specialist referral will be appropriate where there are continuing uncertainties over diagnosis or management—ideally to a unit with psychological as well as medical expertise.


    • Competing interests DK has received occasional consultancy fees and help with organisation of or travel to meetings from companies including Allen and Hanburys, Astra-Zeneca, MSD, 3M, and Boots.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    View Abstract