Osmolality and respiratory regulation in humans: respiratory compensation for hyperchloremic metabolic acidosis is absent after infusion of hypertonic saline in healthy volunteers.

Abstract

Several animal studies show that changes in plasma osmolality may influence ventilation. Respiratory depression caused by increased plasma osmolality is interpreted as inhibition of water-dependent thermoregulation because conservation of body fluid predominates at the cost of increased core temperature. Respiratory alkalosis, on the other hand, is associated with a decrease in plasma osmolality and strong ion difference (SID) during human pregnancy. We investigated the hypothesis that osmolality would influence ventilation, so that increased osmolality will decrease ventilation and decreased osmolality will stimulate ventilation in both men and women. Our study participants were healthy volunteers of both sexes (ASA physical status I). Ten men (mean 28 years; range 20-40) and 9 women (mean 33 years; range 22-43) were included. All women participated in both the follicular and luteal phases of the menstrual cycle. Hyperosmolality was induced by IV infusion of hypertonic saline 3%, and hypoosmolality by drinking tap water. Arterial blood samples were collected for analysis of electrolytes, osmolality, and blood gases. Sensitivity to CO2 was determined by rebreathing tests performed before and after the fluid-loading procedures. Infusion of hypertonic saline caused hyperchloremic metabolic acidosis with decreased SID in all subjects. Analysis of pooled data showed absence of respiratory compensation. Baseline arterial PCO2 (PaCO2) mean (SD) 37.8 (2.9) mm Hg remained unaltered, with lowest PaCO2 37.8 (2.9) mm Hg after 100 minutes, P = 0.70, causing a decrease in pH from mean (SD) 7.42 (0.02) to 7.38 (0.02), P < 0.001. Metabolic acidosis was also observed during water loading. Pooled results show that PaCO2 decreased from 38.2 (3.3) mm Hg at baseline to 35.7 (2.8) mm Hg after 80 minutes of drinking water, P = 0.002, and pH remained unaltered: pH 7.43 (0.02) at baseline to pH 7.42 (0.02), P = 0.14, mean difference (confidence interval) = pH -0.007 (-0.017 to 0.003). Our results indicate that osmolality has an influence on ventilation. Respiratory compensation for hyperchloremic metabolic acidosis was suppressed during hyperosmolality. Water loading caused a decrease in plasma osmolality and metabolic acidosis, and although the decrease in SID was smaller compared with salt loading, the expected respiratory compensation was observed. Ventilation was also stimulated in men, therefore independently of progesterone levels. We propose that the influence of osmolality on ventilation consists mainly as depression in conditions of hyperosmolality and that this depression is absent during hypoosmolality.