Measurement of cardiac output with non-invasive Aesculon impedance versus thermodilution

Background: This study compared the non‐invasive thoracic electrical bioimpedance Aesculon^® technique (TEB_Aesculon) with thermodilution (TD) to evaluate whether TEB_Aesculon may offer a reliable means for estimating cardiac output (CO) in humans. Material and method: Cardiac output was measured with TD and TEB_Aesculon in 33 patients, with a mean age ± SEM of 59 ± 2·7 years, that underwent right heart catheterization for clinical investigation of pulmonary hypertension or severe heart failure. Four to five CO measurements were performed with each technique simultaneously in 33 patients at rest, 11 during exercise and seven during NO inhalation. Result: Cardiac output correlated poorly between TEB_Aesculon and TD at rest (r = 0·46, P<0·001), during exercise (r = 0·35, P<0·013) and NO inhalation (r = 0·41, P<0·017). CO was higher for TEB_Aesculon than TD with 0·86 ± 0·14 l min^?1 at rest (P<0·001) and 2·95 ± 0·69 l min^?1 during exercise (P<0·003), but similar during NO inhalation, with a tendency (P<0·079) to be 0·44 ± 0·19 l min^?1 higher for TEB_Aesculon than TD. CO increased from rest to exercise for TEB_Aesculon and TD with 6·11 ± 0·6 l min^?1 (P<0·001) and 3·91 ± 0·36 l min^?1 (P<0·001), respectively; an increase that was higher (P<0·002) for TEB_Aesculon than TD. During NO inhalation, compared to rest, CO decreased for TEB_Aesculon with 0·62 ± 0·11 l min^?1 (P<0·002), but not significantly for TD with 0·21 ± 0·12 l min^?1 (P<0·11). Bland–Altman analysis showed a poor agreement between TEB_Aesculon and TD. Conclusion: TEB_Aesculon overestimated CO compared to TD with ～17% at rest and ～34% during exercise, but the techniques showed similar results during NO inhalation. CO, furthermore, correlated poorly between TEB_Aesculon and TD. TEB_Aesculon may at present not replace TD for reliable CO measurements in humans.