SummaryWe evaluated the diagnostic usefulness of capnography curves during and following a hyperventilation provocation test (HVPT) in the hyperventilation syndrome (HVS). The diagnosis of HVS was based on the Nijmegen questionnaire and on the reproduction of symptoms during HVPT. Capnography curves of 40 HVS patients, 40 non-HVS patients with psycho-somatic complaints and 26 healthy controls were analyzed. There was no difference in baseline end-tidal CO2-level (FETCO2) between the 3 groups. The spontaneous fall of FETCO2during the adaption phase was clearly different in HVS patients versus non-HVS patients or controls: -0.12 mmmol/1 (95 % confidence limits-0.18 to-0.06) versus + 0.01 mmol/1 (95 % confidence limits-0.04 to + 0.16) (p = 0.002). The 3 minutes FETCO2recovery ratio and the 5 minutes ratio were not significantly different between the groups. In conclusion, in this study the spontaneous fall of FETCO2during the adaption phase of the HVPT was the only valuable part of the capnography test to discriminate between HVS and non-HVS patients.