Cerebrospinal fluid sodium concentration and osmosensitive sites related to arterial pressure in anaesthetized rats

Intracerebroventricular injection of hypertonic saline induces experimental hypertension. To measure [Na] in the vicinity of osmosensitive sites, we continuously measured [Na] in cerebrospinal fluid ([Na]csf) in the lateral ventricle (LV, n = 6), in the third ventricle (V3, n = 6) and in the medial preoptic nucleus (MPO, n = 6) ([Na]MPO) with a Na-sensitive electrode together with mean arterial pressure (MAP) during infusion of hypertonic artificial cerebrospinal fluid (ACSF, [Na] = 1,000 meq/kg H2O) at 5 "mu"l/min for 3 min into the LV in urethane- anaesthetized rats. [Na]csf in the LV began to increase at the beginning of infusion, reaching a peak of 48 +/- 9 meq/kg H2O (mean +/- SE) around the end of infusion, then recovering to the pre-infusion level by 17 min. [Na]csf in V3 changed similarly to that in the LV without any delay, although the peak value was reduced (61% , P < 0.05). In the MPO, in contrast the increase in [Na]MPO was delayed (3 min, P < 0.002) and the peak reduced even further (to 37%, P < 0.01) compared with that in V3. Thereafter, it remained higher than the pre-infusion level until the end of recovery (P < 0.05). MAP began to increase at the onset of infusion (P < 0.05); the maximum increase of 16 +/- 2 mm Hg (n = 18) was reached at the end of infusion, whereafter this level was almost sustained until the end of the 22-min recovery period. To analyse quantitatively the relationship between MAP and [Na]csf, hypertonic ACSF was infused at 2.5 "mu"l/min into the LV. [Na]csf in the LV and MAP increased at half the rates seen with 5 "mu"l/min. These results suggest that the first increase in MAP after hypertonic infusion into the LV is due to the increase in [Na] in the LV and V3, and that the subsequent sustained increase in MAP is related to the delayed increase in [Na] in the periventricular tissues of the V3.