Increased microvascular water permeability in patients with septic shock, assessed with venous congestion plethysmography (VCP).

Abstract

To investigate microvascular water permeability (filtration capacity, Kf) in patients with septic and non-septic shock using a new non-invasive method for studying microvascular parameters in man. Intensive Care Unit of a university hospital. We investigated 28 patients, presenting with cardio-vascular instability due to either septic shock, or non-septic shock (haemorrhage, multiple trauma, respiratory and/or cardiac failure). We used standard invasive methods of monitoring (in-dwelling arterial lines and pulmonary artery flotation catheters) in combination with computer assisted venous congestion plethysmography (VCP) measurements, for a parallel assessment of peripheral microcirculatory parameters. On admission to the ICU, patients with septic shock revealed a significantly higher mean value of filtration capacity Kf = 6.1 +/- 0.4 x 10(-3) (mean value +/- standard error of the mean, ml.min-1.100 ml tissue-1.mmHg-1 = KfU) than non-septic patients Kf = 3.5 +/- 0.3 KfU (p < 0.02). The Kf values of the septic patients were significantly higher than those from age-matched patients with peripheral vascular disease (4.1 +/- 0.2 KfU, p < 0.001) and those of healthy controls (4.3 +/- 0.2 KfU, p < 0.001); the Kf values of the non-septic patients, however, were not significantly different. The highest mean Kf value observed during the stay on ICU was Kfmax 11.6 +/- 0.2 KfU in the septic group and 5.7 +/- 0.1 KfU in the non-septic group (p < 0.001). Pvi, a value reflecting the balance of hydrostatic and oncotic forces in the microcirculation, was elevated in both patient groups. On admission, in septic patients Pvi was 39.2 +/- 3.3 mmHg and in non-septic patients 35.1 +/- 2.7 mmHg, these values were not significantly different, but significantly higher than the Pvi value of healthy controls (Pvi 21.5 +/- 0.8) (p < 0.001). A weak, however significant, positive correlation was found between Kf and Pvi in both patient groups. No correlations were found between Kf, as well as Pvi, and cardiac index (CI), oxygen delivery index (DO2I), oxygen consumption index (VO2I) and systemic vascular resistance index (SVRI). An increase in permeability of microvessels will cause a loss of intravascular fluid and may therefore partially explain the large fluid requirements of patients in shock. It will also favour the development of oedema, which is often found in septic patients. We propose that changes in Kf are useful indices of microvascular malfunction and that VCP allows the non-invasive assessment of these parameters.