6.4. PIIINP and PICP in serum, BALF and ELF in patients with sarcoidosis, fibrosing alveolitis and asbestos-related lung diseases (I-III)

The previous reports concerning the levels of serum and BALF-PIIINP in sarcoidosis, fibrosing alveolitis and asbestosis have been contradictory. Several studies have indi­cated that PIIINP is elevated in BALF of sarcoidosis patients (Low et al. 1983, Bjermer et al. 1986, Bjermer et al. 1987B, Bjermer et al. 1991, Schaberg et al. 1994, Straub et al. 1995, Tukiainen et al. 1994), although some studies have failed to reveal any difference between controls and sarcoidosis patients (Cantin et al. 1988, Milman et al. 1995). Serum levels of PIIINP have been found to be elevated compared with controls (Bacchella et al. 1996, Luisetti et al. 1990, Schoenfeld et al. 1996), but again, there is a study where no elevation was detected (Milman et al. 1995).

The elevated PIIINP in BALF in fibrosing alveolitis patients compared to controls is more like a rule than an exception (Harrison et al. 1993, Low et al. 1983, Low et al. 1992, Cantin et al. 1988, Bjermer et al. 1989, Kuroki et al. 1995, Fujimoto et al. 1995, Tukiainen et al. 1994). There is evidence of elevated levels of serum PIIINP (Low et al. 1983, Low et al. 1992, Harrison et al. 1993), but contradictory findings have also been reported (Watanabe et al. 1985). Information on BALF-PICP is limited to a study which showed elevated levels of BALF-PICP in sarcoidosis and fibrosing alveolitis compared to controls (Tukiainen et al. 1994).

Knowledge of the propeptide levels in human asbestos-related diseases is limited, most studies being experimental. According to the only study on BALF-PIIINP in asbestos-related diseases, the level of PIIINP in BALF was observed to be elevated in asbestosis compared to asbestos-exposed subjects without parenchymal disease (Begin et al. 1986). PICP but not PIIINP was elevated in BALF in asbestos-exposed subjects (Tukiainen et al. 1994). In addition, one study suggests that serum PIIINP may be a useful index of asbestosis (Cavalleri et al. 1988). There is also evidence of the elevation of PIIINP in BALF in progressive silicosis (Begin et al. 1987a).

In the present study, serum PIIINP was slightly elevated in 13% of the patients with sarcoidosis and 22% of the patients with fibrosing alveolitis, but not in asbestosis or asbestos-exposed subjects without parenchymal involvement. BALF and serum PIIINP had no reciprocal correlation. Based on these considerations, it is obvious that S-PIIINP cannot be used as a marker of local fibrogenesis in lungs. The levels of serum PICP were similar in sarcoidosis, fibrosing alveolitis, asbestos-exposed subjects, asbestosis and controls. This is in line with the fact that the majority of PICP is derived from bone (Melkko et al. 1990). It therefore seems evident that S-PICP is a poor indicator of any lung disease.

The level of BALF-PIIINP was highest in sarcoidosis and second highest in fibrosing alveolitis, but hardly detectable in the other groups. When evaluated in the epithelial lining fluid estimated by the urea method, PIIINP was elevated in sarcoidosis (121-fold) and fibrosing alveolitis (13-fold) compared to serum, suggesting active local synthesis or possibly degradation of type III collagen in the lower respiratory tract. Procollagen I carboxy­terminal propeptide in BALF was high in fibrosing alveolitis and asbestosis and was also elevated to a lesser extent in sarcoidosis, although the levels did not differ significantly. Detectable BALF-PICP was observed in some of the asbestos-exposured patients without parenchymal involvement but not in the controls. Again, the average concentrations of PICP were higher in ELF than in serum, being twofold in sarcoidosis and fourfold in fibrosing alveolitis and asbestosis, which can be considered a sign of active synthesis of type I collagen in lungs. BALF- and ELF-PICP were elevated in the patients who may have slowly progressive development of lung fibrosis, such as asbestosis or idiopathic fibrosis, and, to a lesser extent, in sarcoidosis. On the contrary, BALF-PIIINP was highest in parenchymal sarcoidosis with an obvious local inflammatory process, whereas the level of PIIINP in BALF was relatively low in fibrosing alveolitis and hardly detectable in asbestos-exposed patients and controls. This may suggest that BALF-PICP would be a better marker of local fibrogenesis than BALF-PIIINP.

One possible explanation for the contradictory results of the previous studies other than the methodological point of view discussed above is the heterogeneity in the status and course of the diseases investigated. At the time of the diagnosis, the patients can be in various stages of the disease. We would need large numbers of patients at each stage of the diseases, i.e. large national or international studies, to be able to make further conclusions.