[A Meta-Analysis of Efficacy of Carbon Dioxide Inhalation as a Challenge Test in Panic Disorder].
Level 3 - non-randomized controlled study
Systematic review and meta-analysis of non-randomized comparative challenge studies
What was done
A systematic review and meta-analysis of 33 studies (2,114 total participants) identified through PubMed, EMBASE, and PsycNET evaluating carbon dioxide (CO2) inhalation challenge tests. Studies were classified by CO2 concentration (5% and 35%) and sample characteristics (panic disorder patients, healthy controls, healthy first-degree relatives of panic disorder patients, and psychiatric controls). Heterogeneity was evaluated via Q and I2 statistics, publication bias via Begg-Mazumdar and Egger tests and funnel plots, and pooled odds ratios (ORs) were calculated using fixed- and random-effects models.
What was found
Participants with panic disorder (PD) had significantly higher rates of panic attacks than healthy controls (HC) following both 5% CO2 inhalation (OR = 14.713, 95% CI 7.532–28.739) and 35% CO2 inhalation (OR = 11.507, 95% CI 7.775–17.031). Healthy controls with a first-degree relative with PD were significantly more likely to experience panic attacks than HC without such relatives (OR = 2.658, 95% CI 1.678–4.212). Participants with PD also experienced significantly more panic attacks than patients with other psychiatric disorders (OR = 3.524, 95% CI 1.945–6.384).
Why it matters
This confirms that heightened CO2 sensitivity is a robust, relatively specific vulnerability marker for panic disorder that also manifests in unaffected first-degree relatives.
Limits
The abstract does not provide specific heterogeneity statistics (I2 or Q values), publication bias test scores, or quality assessments for the included studies. In addition, provocation tests in laboratory settings may not capture the full complexity of spontaneous panic attacks.
Cited by
- contradicts Inhaling carbon dioxide triggers a panic attack in approximately 50 percent of the general population.