Comparative Study On The Value Of MRI And Needle Biopsy Of Sacroiliac Joints In Diagnosis Of Sacroiliitis

OBJECTIVETo study the measure agreement of MRI and needle biopsy of sacroiliac joints in diagnosis of sacroiliitis, and explore how to diagnose sacroiliitis with the two techniques.METHODSFrom 2000 to 2006 there were 94 patients with inflammatory low back pain were examined by MRI, CT and needle biopsy on sacroiliac joints (SIJs), 85 cases of whom were scanned by both plain MRI and dynamic MRI, 9 cases of whom were scanned by plain MRI only. Retrospectively analyzed the data of MRI, CT, biopsy pathology of the SIJs and the clinical features of all the patients.RESULTS188 SIJs of all the 94 patients were classed by CT into four grades basing on the 1884 modified New York criteria, including grade 0(n=80), gradeⅠ(n=29), gradeⅡ(n=32), gradeⅢ(n=44) and gradeⅣ(n=3). There were 109 joints being＜gradeⅡversus 79 joints being≥gradeⅡ.1. group 1 (joints being＜gradeⅡin CT): 59% (64/109) of the joints had abnormal changes (one abnormal change or more) of the SIJs at MRI in group one. 107 specimens were collected successfully of all the 109 joints, 63% (67/107) of the specimens had abnormal changes (one abnormal change or more) of the SIJs detected under microscopy. The results of MRI and pathology were compared with McNemar test and measure of agreement was calculated. The value of the kappa coefficient was 0.626, indicating medium association.54% (59/109) of the cartilages had abnormal changes of the SIJs at MRI in group one. 104 cartilage specimens were collected successfully of all the joints, 64% (67/104) of the cartilages had abnormal changes detected under microscopy. The cartilages data of MRI and pathology were compared with McNemar test and measure of agreement was calculated. The value of the kappa coefficient was 0.33, indicating low association.19% (21/109) of the sub-chondral bone had abnormal changes of SIJs at MRI in group one. 88 sub-chondral bone specimens were collected successfully of all the joints, 64% (56/88) of the sub-chondral bone had abnormal changes detected under microscopy. The sub-chondral bone data of MRI and pathology were compared with McNemar test. The positive percentage of biopsy was significantly higher than that of MRI (P=0.00).Edema signal was found in 14% (15/109) of the joints at MRI in group one, 4 of those were around the cartilages, 8 of those were in the peri-articular marrow, 2 of those were in the marrow of spina iliaca posterior superior, and 1 of those suffused most of the articular area. The marrow tissue was found in only 11 of the joints biopsy specimens, 82% (9/11) of the marrow were abnormal detected under microscopy.2. group two (joints being≥gradeⅡin CT):The positive percentage of MRI was 100% (79/79) in group two which were≥gradeⅡin CT, versus 90 % (71/79) that of biopsy specimens detected under microscopy. 7 of the 79 joint biopsy specimens were normal detected under microscopy, 4 of those are from joints being gradeⅡin CT, 3 of those are from the joints being gradeⅢ. 38 patients were definitely diagnosed as AS by clinical features and CT founding. 4 of them were failure of diagnosis by the SIJs biopsy, the diagnosis missing percentage was 11%.96% (76/79) of the cartilages had abnormal changes of the SIJs at MRI in group two. 69 cartilage specimens were collected successfully of all the joints, 93% (64/69) of the cartilages had abnormal changes detected under microscopy. The cartilages data of MRI and pathology were compared with McNemar test and measure of agreement was calculated. The value of the kappa coefficient was 0.26, indicating low association.97% (77/79) of the sub-chondral bone had abnormal changes of SIJs at MRI in group two. 70 sub-chondral bone specimens were collected successfully of all the joints, 90% (63/70) of the sub-chondral bone had abnormal changes detected under microscopy. The sub-chondral bone data of MRI and pathology were compared with McNemar test. There was no significant difference between the positive percentages of the two methods. The value of the kappa coefficient was 0.19, the statistics test of the kappa coefficient showed there was no significant association between the two methods (P=0.06).Edema signal was found in 84% (66/79) of the joints at MRI in group two, as 15% (12/79) of the joints show fatty accumulation abroad the joint area. 6 joints show fatty accumulation and marrow edema simultaneously. The marrow tissue was found in only 7of the joints biopsy specimens, 100% (7/7) of the marrow were abnormal detected under microscopy.3. There were 85 patients (170 joints) scanned by dynamic MRI, only 164 joints had nice MRI images and biopsy specimens as well. We divided the 164 joints into two groups basing on the pathology appearances, group A was with inflammatory cells infiltration and pannus formation detected under microscopy, and group B was without. There were 5, 82, 18 joints being no enhancement, low enhancement and high enhancement respectively in group one, versus 21, 36, 2 joints in group two. The enhancement of group A was significantly higher than that of group B (P=0.00).CONCLUSION1. The abnormal changes of cartilage, sub-chondral bone and bone marrow of SIJs can be detected by MRI and biopsy as well. MRI and biopsy have a certain extent association on detection the abnormal changes in cartilages. And the positive percentage of the biopsy of sub-chondral bone in Joints being＜gradeⅡin CT is significantly higher than that of MRI. Combining the MRI and biopsy techniques can provide more useful messages about SIJs. 2. Joints being＜gradeⅡin CT has high positive percentage in MRI and biopsy. The results of MRI and biopsy have medium association, they are both helpful to the diagnosis of early AS.3. Some patients with unilateral sacroiliitis of gradeⅡin CT maybe have bilateral sacroiliitis detected by MRI or biopsy. As for these patients, the MRI and biopsy of SIJs are helpful to the diagnosis of AS.4. Biopsy has a certain false negative ratio and we can not negate the diagnosis of sacroiliitis only by the results of biopsy.5. The enhancement of SIJs maybe associate with the inflammatory cells infiltration and pannus formation in the joints. The dynamic MRI can detect the early sacroiliitis and measure the activity degree of the sacroiliitis.