An increased microvascular water permeability has been reported after ischemia/reperfusion both in animal models and in human studies. We studied the changes in fluid filtration capacity (FFK) after ischemia/reperfusion due to tourniquet in patients undergoing arthroscopy of the knee. METHOD: Healthy male volunteers (n = 24, mean age 46.9 +/- 3.5) were studied prior to, 1 and 6 hours after arthroscopy of the knee, during which a tourniquet was applied to the thigh. FFK, isovolumetric venous pressure (Pvi) and arterial blood flow in the limb was measured in both legs (tourniquet leg and control leg) using computer assisted venous congestion plethysmography. Venous blood samples were obtained from a cubital vein prior to and from the femoral vein 2 mins after deflation of the tourniquet cuff. 12 patients received preoperatively an infusion of 6% Dextran (D) and 12 patients 500 ml of electrolyte solution (VE) an. RESULTS: The mean duration of the tourniquet was (D) 56.0 +/- 6.9 min and (VE) 53.9 +/- 4.2 min which resulted in a significant increase in venous lactate concentration from (D) 1.4 +/- 0.1 mmol.l-1 to 2.7 +/- 0.3 mmol.l-1 and (VE) 1.3 +/- 0.1 mmol.l-1 to 2.7 +/- 0.3 mmol.l-1 (p < 0.001). A significant decrease in pH from (D) 7.39 +/- 0.01 to 7.32 +/- 0.01 (p < 0.001) and from (VE) 7.39 +/- 0.01 to 7.32 +/- 0.01 (p < 0.001) was also seen. Preoperatively no significant differences in the FFK values of the tourniquet leg (D = 5.3 (4.8-10.7) ml. x 10(-3) min-1.100 ml tissue-1 mmHg-1 = FFKU) and the control leg (5.2 (4.7-8.6 FFKU)) were observed. The maximum FFK value in D was seen 1 hour after ischemia/reperfusion in both, the tourniquet leg (7.5 (4.6-14.2 FFKU) and the control leg (7.8 (5.5-9.4 FFKU). In VE however the maximal FFK value were measured 6 hours after ischemia/reperfusion with an increase in the tourniquet leg from 5.2 (4.2-6.2 FFKU) to 8.1 (3.7-10.4 FFKU) and the control leg from 6.1 (3.6-7.0 FFKU) to 7.3 (6.1-8.3 FFKU) (Median (Range). One-way ANOVA). There were no significant differences in the FFK values between D and VE except for a lower Pvi in VE at the third measurement. No significant changes in the arterial blood flow were seen perioperatively as well as between the operated and non-operated leg. CONCLUSION: A tourniquet of < 1 hour does impair tissue oxygenation as indicated by the increase in lactate and the decrease in pH. The duration of the tourniquet was however to short to have caused sufficient microvascular damage result in a more pronounced increase in fluid filtration capacity.