Symptoms and chest radiograph findings are suitable clinical indicators of IP activity, though their precise quantification is difficult. Serum lactate dehydrogenase may also be a rough indicator of disease activity in conditions unassociated with other organ dysfunctions, including liver disease. Therefore, in the present study, changes in arterial O2 tension or saturation were chosen as parameters of disease activity since BAL and open-lung biopsy were not performed in all patients, and since local and small transbronchial lung biopsy specimens may not reliably reflect disease activity. In the active group, LI was significantly higher than in the stable group (Fig 6), suggesting that noninvasive measurement of LI may reliably reflect disease activity in IP. Cardinal pathophysiologic characteristics of active IP consist of alveolar epithelial cell injury due to inflammation (alveolitis) and progressive fibrotic changes. The alveolar septum consists of epithelium, interstitium, and endothelium, in close contact with each other. Therefore, it is likely that alveolar inflammation spreads easily to the capillary endothelium and, consequently, that pulmonary vascular leakiness assessed by LI did indeed reflect disease activity in this study. Serial LI measurements made in the same patients will be important to confirm these results. In addition, studies of larger populations with other types of gamma-emitting radiolabeled proteins, such as 99mTc-human serum albumin, should also be performed to validate these preliminary observations.
Methodologic Considerations and Possible Study Limitations
In the present study, the exponential equilibration coefficient of Ga-circulating transferrin between intravascular and pulmonary interstitial compartments (LI) in the control group (11.37 ± 0.30 X 10_ 3/min) was nearly identical to that measured with In-labeled transferrin by Gorin et al in normal subjects (13.36 X 10_ 3/min). Furthermore, Raijmakers et al applied an in vivo Ga-circulating transferrin labeling technique and found the pulmonary LI useful and sensitive to distinguish cardiogenic from noncardiac pulmonary edema. The main difference between our methodology and theirs is the absence, in our studies, of a blood volume marker, such as 99mTc-RBCs, in addition to the protein marker. The incorporation of a blood marker seems particularly desirable to assess vascular leakiness in acute lung injury due to systemic inflammatory events, since pronounced hemodynamic changes and significant leakage of protein marker into other organs may take place during the measurements. However, such important systemic events were absent in our patient population, and the mean LI value measured in our control group was close to that measured by Raijmakers et al (10.2 X 10_ 3/min).
In conclusion, an increase in pulmonary vascular leakiness in patients with IP was detectable by calculating the exponential equilibration coefficient of Ga-circulating transferrin. Further studies are warranted to examine the contribution of this noninvasive assessment of vascular leakiness as a method to monitor disease activity and the effects of treatment in IP.
Figure 6. Differences in LI and gallium index between stable and active phases of IP. The average of six ROIs for each patient is shown.