Background Interstitial pneumonia (IP) is a serious complication associated with polymyositis (PM) and dermatomyositis (DM). Predictive factors for poor prognosis have not been fully identified. KL-6 and surfactant protein D (SP-D) are biomarkers for existence and severity of IP. Clinical usefulness of these markers in IP in PM/DM has not been fully evaluated.
Objectives To identify poor prognostic factors for IP in PM/PM. and to determine whether serum KL-6/SP-D levels predict patients with IP in PM/DM.
Methods Consecutive 50 PM/DM (P: 17, DM 33) patients with active IP, 6 of whom were died of respiratory failure, were enrolled in this study. Active IP was defined as IP that was judged as requiring intensive immunotherapy for IP by physicians managing patients, based on worsening dyspnea and, expanding ground glass opacities in HR-CT or decreasing PaO2. Serum KL-6 and SP-D levels were measured before treatment for IP and every 2-4 weeks after starting therapy. To find correlation between KL-6/SP-D levels and clinical feature including prognosis, medical records were reviewed retrospectively.
Results Serum KL-6 and SP-D levels were elevated in patients with active IP compared to those with inactive IP or without IP. Serum KL-6 and SP-D levels at the diagnosis of active IP were within normal range in 26% and 44% of patients with active IP, respectively. KL-6 and SP-D levels at the time of diagnosis failed to predict prognosis. Serum Kl-6 levels were increased up to 3 month after starting treatment and then decreased gradually to base lines, whereas SP-D levels were at peaks within 1st 4 weeks after the treatment and decreased rapidly to normal levels. In patients with poor prognosis showed in marked increased in KL-6/SP-D levels.
Comparison of clinical features between patients with poor and good prognosis showed significant difference in Kl-6 levels at 1st 4 weeks and increase of KL-6 and SP-D levels during during 1st 4 weeks, increased AaDO2 levels before treatment and existence of alveolar opacities. Multivariate logistic regression analysis showed only increase in KL-6/SP-D levels during during 1st 4 weeks as a poor prognostic factor. Cut off levels determined by ROC analysis were 1.70 and 1.75 of increase ratio of KL-6 and SP-D levels at 2-4 weeks after treatment to those at 0 week, respectively. Increase of KL-6 during 1st 4 weeks over 800U/ml was a poor prognostic factor.
Conclusions Increase in serum Kl-6/SP-D levels during 1st 4 weeks after starting therapy, but not their levels at any one time point, predicts poor prognosis. When marked increased of KL-6 and SP-D levels during the 1st 4 weeks, patients have risks of poor prognosis and additional therapy should be considered.
Disclosure of Interest None Declared