ARTERIOVENOUS MALFORMATION DRAINING VEIN PHYSIOLOGY AND DETERMINANTS OF TRANSNIDAL PRESSURE-GRADIENTS

ARTERIOVENOUS MALFORMATION (AVM) draining vein pressure (DVP) may have an influence on both the natural history of the disease and treatment outcome. The purposes of this study were to assess the relationship b etween DVP and other clinical and physiological variables and to chara cterize the transmission of arterial pressure across the AVM nidus, DV P measurements were carried out during elective AVM resection with iso flurane/nitrous oxide anesthesia with arterial carbon dioxide pressure of approximate to 30 mm tig. The gradient between the right atrium an d operative measurement site was noted. Pre-excision feeding mean arte rial pressure and DVP were measured with a 26-gauge needle simultaneou sly with systemic mean arterial pressure and central venous pressure ( CVP), DVP was tested with systemic mean arterial pressure increased to approximate to 20 mm Hg with phenylephrine or CVP increased with a Va lsalva maneuver. Finally, preresection and postresection DVP values we re compared. Relative to the site of measurement, DVP was 7 +/- 5 mm H g at a CVP of -4 +/- 5 mm Hg (n = 45), There was no influence of prese ntation, presence of deep venous drainage, size, location, or prior em bolization on DVP, In 19 patients, DVP decreased (8 +/- 4 to 5 +/- 3; P < 0.05) whereas CVP increased from pre- to postresection (-4 +/- 5 t o -2 +/- 5; P< 0.05). For the phenylephrine challenge (n = 11), there was no difference (P = 0.84) between the Delta DVP (2 +/- 1 mm Hg) and the Delta CVP (2 +/- 3 mm Hg). For the Valsalva maneuver challenge (n = 7), however, Delta CVP (8 +/- 4 mm Hg) was greater (P < 0.02) than Delta DVP (3 +/- 2 mm Hg). Feeding mean arterial pressure and DVP were positively correlated (y = 0.2x + 2.4; r = 0.59; n = 14; P < 0.05) wi thout any apparent influence of angiographic venous stenosis. Neverthe less, transnidal pressure drop or net cerebral perfusion pressure grad ient (i.e., the lowest possible perfusion pressure to which normal adj acent circulatory beds might be exposed) was inversely correlated with AVM size (y = -6.1x + 52.3; r = 0.68; n = 14; P < 0.01). Clinically r elevant changes in systemic mean arterial pressure and CVP affect DVP more as a venous than as an arterial structure. Most importantly, high er feeding mean arterial pressure is associated with higher DVP, but n otwithstanding, a lower transnidal pressure gradient is present in lar ger AVMs. A lower transnidal pressure gradient, which may be associate d with certain postoperative hemodynamic complications when transmitte d to adjacent capillary beds, also may protect against spontaneous int ranidal vessel rupture.