Delayed resolution of high-pressure pulmonary edema or capillary leak

Both clinical and experimental evidence suggest that the time course of edema formation is different from that of edema resolution. To better describe and quantify this difference, we followed the accumulation of high-pressure pulmonary edema in live dogs with the thermal-green dye (TGD) double-indicator technique at steady-state levels of lung liquid. We raised left atrial pressure (PLa) in steps of 5 to 10 mm Hg as high as 25 mm Hg and followed edema to steady-state levels. Lung water was then measured as PLa was lowered to initial values. By plotting steady-state edema against PLa, pressure-volume relationships were constructed. There was little change in edema until PLa reached approximately 15 mm Hg, at which point further changes in PLa were associated with large increases in lung liquid. At PLa = 25 mm/kg, the average lung water had increased by 10 ml/kg. In each animal there was slow resolution of edema with decreases in PLa from its peak back to its initial value, but in no animal was edema fully reabsorbed even though PLa was maintained at about 5 mm Hg for as long as 10 hours. Several possible explanations account for these observations. Water could be trapped in alveolar and central interstitial spaces. In addition, vessel closure in edematous lung units could further influence water reabsorption. These observations raise the possibility that pulmonary edema in the presence of normal filling pressures may represent resolution of a transient high-pressure edema as opposed to a capillary leak syndrome.