Relationships between exercise-induced pulmonary hypertension and nocturnal desaturation

To the Editors:

We read with interest the excellent article by Christensen et al. 1, which failed to show a consistent relationship between hypoxaemia and pulmonary hypertension (PH) during exercise equivalent to activities of daily living (ADL) in chronic obstructive pulmonary disease (COPD) patients.

This important observation prompted us to report here the results of a prospective study that we conducted during the 1990s, which has only been published in abstract form 2. Nine male COPD patients with mild hypoxaemia at rest (six current and three ex-smokers) were included in the study (age: 66.9±5.8 yrs, forced expiratory volume in one second/forced vital capacity: 43.2±15.5% predicted, arterial oxygen tension (P_a,O2): 8.66±0.47 kPa (65.1±3.5 mmHg), carbon dioxide arterial tension (P_a,CO2): 5.93±0.77 kPa (44.6±5.8 mmHg)). Haemodynamic measurements were performed with a Swan-Ganz thermodilution catheter at rest and at the end of a bicycle exercise test (constant workload of 60 W for 10 min). This load is equivalent to a 10-min walk on flat ground and can be assimilated to ADL for these COPD patients. Two out of nine patients developed exercise-induced pulmonary hypertension (EIPH) with pulmonary artery pressure (P_pa) >30 mmHg. For the group as a whole, the results were similar to those reported by Christensen et al. 1, with a significant increase in P_pa during exercise: 28.4±7.9 mmHg versus 16.6±4.0 mmHg at rest (p<0.05) without any correlations between P_pa during exercise and P_a,O2 (at rest and during exercise). P_pa during exercise was not correlated with airflow obstruction and P_a,CO2 levels. No left ventricular dysfunction was noted in any patients (either on echocardiography or Swan-Ganz measurements).

All COPD patients underwent polysomnography. No associated sleep apnoea syndrome was diagnosed. No correlations were found between haemodynamic (P_pa at rest and exercise) and polysomnographic data (mean and minimal arterial oxygen saturation). Based on analysis of saturation when asleep, only the two patients with EIPH could be considered to be “desaturators” according to the definition proposed by Fletcher et al. 3, and EIPH was significantly more frequent in desaturator than in non-desaturator patients (p<0.05).

Survival is known to be related to arterial oxygen desaturation on exercise both in patients with primary pulmonary hypertension 4 and in patients with COPD and mild hypoxaemia 5. In the latter group, the reasons for reduced survival have not been elucidated 6. The role of hypoxaemia-induced PH has been suggested, but Chaouat et al. 7 found no differences in P_pa at rest between desaturator and non-desaturator COPD patients. To the best of our knowledge, no data are available concerning the relationship between EIPH and desaturation during sleep in mild COPD patients.

Although exercise-induced pulmonary hypertension may occur in the absence of significant activities of daily living-induced oxygen desaturation, exercise-induced pulmonary hypertension was found in all “desaturator” and never in “non-desaturator” patients. Moreover, we found a significant positive correlation between blood carbon monoxide levels and nocturnal desaturation (p<0.05), supporting the hypothesis that exercise-induced pulmonary hypertension is due to remodelling of pulmonary arteries caused by low-grade inflammation related to cigarette smoking. These findings support the hypothesis that pulmonary hypertension may be the consequence of pulmonary vascular remodelling and is not due to exercise- or sleep-induced hypoxaemia. As exercise-induced pulmonary hypertension has been found to be predictive of the development of persistent pulmonary hypertension 8, it is important to: 1) diagnose remodelling of pulmonary arteries as early as possible; and 2) propose specific treatment. As shown in our study 2, remodelling could be assessed either by pulmonary haemodynamic measurements during exercise, but also by polysomnographic recordings. Treatment of remodelling may not be based on correction of exercise and/or nocturnal desaturation 9, but must target the primum movens that is cigarette smoking.