特发性肺纤维化与非特异性间质性肺炎的非创伤性鉴别诊断方程

A non-invasive equation in diagnosis of idiopathic pulmonary fibrosis and nonspecific interstitial pneumonia

目的 为提高临床非创伤性区分特发性肺纤维化（IPF）与非特异性间质性肺炎（NSIP）的水平,建立IPF与NSIP的非创伤性鉴别诊断方程.方法 选病理确诊的14例IPF和18例NSIP患者为研究对象,记录患者临床资料、胸部高分辨率CT评分、支气管肺泡灌洗等结果.将各个变量分别赋值,筛选,建立方程,确定临界值.结果 （1）IPF组患者比NSIP组年龄偏大,男性较多,吸烟者较多.（2）与NSIP组比,IPF组患者胸部高分辨率CT评分网格影、蜂窝影评分高,磨玻璃影评分低.（3）IPF组淋巴细胞占支气管肺泡灌洗液中细胞总数大于20%低于NSIP组.（4）建立方程：Y=0.9＋0.123x1-0.045x2＋0.009x3＋0.033x4（x1、2、3、4分别为影像学显示蜂窝影、影像学显示磨玻璃影、年龄、影像学显示网格影）,方程临界值为1.5.结论 回归方程y=0.9＋0.123x1-0.045x2＋0.009x3＋0.033x4可协助临床医生区分IPF与NSIP.

Objective Idiopathic pulmonary fibrosis （IPF） and nonspecific interstitial pneumonia （NSIP） are the two largest subsets of idiopathic interstitial pneumonia （IIP）. They have a sharply differences in therapy and survival, however, the identification of them is difficult. In general, NSIP has a much better responsiveness to therapy and good survival. On the other hand, in IPF, was demonstrated to be associated with poorer survival and responsiveness to therapy. To diagnosis of IPF and NSIP need histopathologic classification which however, requires surgical lung biopsy. Therefore, we try to set up an equation using those data from non-invasive methods to identify IPF and NSIP. Methods A retrospective review of 32 patients with IPF （ n = 14 ） and NSIP （ n = 18 ） was carried out. General clinical data, pulmonary function, chest radiographic, and bronchoalveolar lavage fluid were recorded. The statistical methods of logistic regression and multiple linear regression are used to establish the equation. The method of curve fitting is used to find the critical value of the equation to the differential diagnosis between the IPF and NSIP. Results （ 1 ） Compare to NSIP, patients of IPF are older, more males than females, having a bigger proportion of smokers. （2） Compared with NSIP patients, IPF patients have higher CT score of ＂grid shadow＂, ＂cellular shadow＂ and lower CT score of ＂ground-glass shadow＂. （ 3 ） Bronchoalveolar lavage （BAL） lymphocytosis was more frequently observed in NSIP than IPF. （4） We get an equation： y = 0. 9 ＋ 0. 123x1 -0. 045x2 ＋0. 009x3 ＋0. 033x4 and it can improve the differential diagnosis between the IPF and NSIP in this group of patients. Conclusion Clinical,high resolution computerized tomography and BAL are useful non-invasive tools in diagnosis IPF and NSIP. The equation： y = 0. 9 ＋ 0. 123x1 - 0. 045x2 ＋ 0. 009x3 ＋ 0. 033x4 can help us to distinguish the IPF and NSIP.