Elsevier

The Lancet

Volume 348, Issue 9021, 20 July 1996, Pages 154-158
The Lancet

Articles
Double-blind placebo-controlled study of the hyperventilation provocation test and the validity of the hyperventilation syndrome

https://doi.org/10.1016/S0140-6736(96)02024-7Get rights and content

Summary

Background

Hyperventilation syndrome (HVS) describes a set of somatic and psychological symptoms thought to result from episodic or chronic hyperventilation. Recognition of symptoms during the hyperventilation provocation test (HVPT) is the most widely used criterion for diagnosis of HVS. We have investigated the validity of the HVPT and of the concept of HVS.

Methods

In a randomised, double-blind, crossover design, the ability of 115 patients with suspected HVS to recognise symptoms during the HVPT was compared with the ability to recognise symptoms during a placebo test (isocapnic overbreathing, with carbon dioxide levels maintained by manual titration). 30 patients who had positive results on the HVPT underwent ambulatory transcutaneous monitoring of pCO2 to ascertain whether they hyperventilated during spontaneous symptom attacks.

Findings

Of the 115 patients who underwent the HVPT and the placebo test, 85 (74%) reported symptom recognition during the HVPT (positive diagnosis HVS). Of that subset, 56 were also positive on the placebo test (false-positive), and 29 were negative on the placebo test (true-positive). False-positive and true-positive patients did not differ in symptom profile or in physiological variables. During ambulatory monitoring (15 true-positive, 15 false-positive) 22 attacks were registered. Transcutaneous end-tidal pCO2 decreased during only seven. The decreases were slight and apparently followed the onset of the attack, which suggests that hyperventilation is a consequence rather than a cause of the attack. There were no apparent differences between false-positive and true-positive patients.

Interpretation

The HVPT is invalid as a diagnostic test for HVS. Hyperventilation seems a negligible factor in the experience of spontaneous symptoms. The term HVS should be avoided.

Introduction

Hyperventilation is respiration that exceeds metabolic demands, resulting in a decrease in arterial partial pressure of carbon dioxide (pCO2) and an increase in pH of the body fluids (respiratory alkalosis). If hyperventilation is sustained for some time, secondary physiological changes take place and can lead to symptoms such as dizziness, paraesthesias, stiff muscles, cold hands or feet, and trembling.

In 1937, Kerr and colleagues1 introduced the term hyperventilation syndrome (HVS), because of the idea that hyperventilation may be the underlying mechanism in certain symptom patterns (in the absence of organic disease). Important evidence to suggest that HVS is a discrete clinical entity came from the demonstration that, in many patients, the usual symptoms could be reproduced by a brief period of voluntary hyperventilation.1, 2, 3 There has lately been an upsurge in the interest in HVS. Several studies have indicated high frequencies of HVS, 35–83% among patients with anxiety disorder4, 5, 6, 7 and 5–11% among various other medical populations.8, 9, 10

HVS is generally diagnosed on the basis of hyperventilation provocation test (HVPT), which requires voluntary hyperventilation for several minutes. The test is considered positive if the symptoms induced are recognised by the patient as similar to those occurring in daily life.3 Although this test is the most widely used criterion for diagnosis of HVS, the importance of additional psychological criteria has been emphasized, especially lowered arterial or end-tidal resting pCO2 values,11 and a delayed recovery of pCO2 after voluntary hyperventilation.12

Some studies have questioned the validity of symptom recognition during the HVPT. The main criticism is that the HVPT does not distinguish between symptoms produced by hypocapnia and those produced by other factors, such as the mechanical discomfort of rapid deep breathing, or the stressfulness of the procedure. Exposure of patients to other types of challenges can reproduce similar symptom patterns.13, 14, 15, 16 We and others15, 16 have shown that most patients positive on the HVPT also had symptoms during a stressful mental load task that did not involve or induce hypocapnia. This finding implies that the frequency of HVS may have been grossly overestimated. Despite these doubts, no placebo-controlled study has directly addressed the specificity and validity of the HVPT.

The strength of our study lies in the combination of a double-blind placebo-controlled study on the HVPT and a follow-up study with ambulatory monitoring of pCO2. The latter technique has the potential to be a gold standard, since it allows measurement of pCO2 during spontaneous symptom episodes.17, 18 Our hypothesis was that if hyperventilation is an important symptom-producing factor in some patients, these patients must experience symptom recognition during the HVPT but not during the placebo test. We predicted that these true-positive patients would have more typical hyperventilation symptoms than the other groups, that a larger proportion would meet physiological HVS criteria, and most importantly, that they would show decreases in pCO2 during spontaneously developing symptom attacks.

Section snippets

Patients and methods

The study was approved by the medical ethics committee of Slotervaart Hospital. The sample consisted of 115 consecutive outpatients (62 male, 53 female; mean age 38·8 years [SD 15.0], range 16–72), referred by medical specialists and general practitioners because of medically unexplained somatic symptoms that suggested HVS. Inclusion criteria were adequate command of the Dutch language, no anxiolytic medication above the regular dose (ie, no special medication taken out of fear for the test),

Results

End-tidal pCO2 and respiratory rate were averaged over the last 30 s of each period (baseline, overbreathing, and recovery). During the HVPT, end-tidal pCO2 fell by more than 50% (table 1). During the placebo test, brief changes in end-tidal pCO2 could not be prevented, but overall the baseline value was successfully maintained. There were no differences between the HVPT and the placebo test in respiration rate or depth of overbreathing (data not shown; available from authors).

The HVPT induced

Discussion

Our results clearly demonstrate the low validity of the HVPT when used for diagnosis of HVS. Two-thirds of the patients who were positive on the HVPT were also positive on the placebo test. The hypothesis that the placebo test would be useful in distinguishing true-positive from false-positive patients was not confirmed: the groups did not differ on any of the symptom variables, physiological variables, or pCO2 during spontaneous symptom episodes. A high percentage of false-positive results

References (31)

  • M Soley et al.

    The aetiology of effort syndrome

    Am J Med Sci

    (1938)
  • RA Lewis et al.

    Definition of hyperventilation syndrome

    Bull Eur Physiopathol Respir

    (1986)
  • R Rapee

    Differential response to hyperventilation in panic disorder and generalized anxiety disorder

    J Abnorm Psychol

    (1986)
  • RuiterC et al.

    The hyperventilation syndrome in panic disorder, agoraphobia, and generalized anxiety disorder

    Behav Res Ther

    (1989)
  • RE Brashear

    Hyperventilation syndrome

    Lung

    (1983)
  • Cited by (129)

    • The psychophysiology of the sigh: II: The sigh from the psychological perspective

      2022, Biological Psychology
      Citation Excerpt :

      Breathing patterns resulting in deviations from normoxia and normocapnia, may cause respiratory symptoms. However, respiratory symptoms caused by dysfunctional breathing patterns can also occur in the presence of efficient gas exchange (Courtney & Cohen, 2006; Hornsveld & Garssen, 1997; Hornsveld, Garssen, Dop, Spiegel, & Haes, 1996; Vlemincx, Van Diest, & Van Den Bergh, 2012). While the term ‘dysfunctional breathing’ has been long used for a variety of maladaptive breathing patterns, recently, a classification system has been proposed to better understand and define dysfunctional breathing patterns (Boulding, Stacey, Niven, & Fowler, 2016).

    • Comparative analysis between available challenge tests in the hyperventilation syndrome

      2021, Respiratory Medicine
      Citation Excerpt :

      The most interesting finding of our study was that, for similar minute-ventilation, HVPT was a better method in term of symptoms reproducibility in patients with HVS (≥3 NQ items) than CPET. This significant difference could be the consequence of different PETCO2 levels during HVPT and CPET, several symptoms from NQ such as tingling, numbness, dizziness, and neuromuscular symptoms being clearly related to hypocapnia [2,40,47]. There was no significant difference between asthmatics and non-asthmatics with HVS during HVPT and CPET.

    View all citing articles on Scopus
    View full text