The Experimental Study And Clinical Application Of MR Imaging On Articular Cartilage Disease In The Knee

Part 1:Comparative study of MR sequences on articular cartilage in the kneeObjectiveA group of young healthy volunteers were performed with MRI to explore the best magnetic resonance(MR) sequence showing articular cartilage in the knee.Materials and Methods1.Subjects20 healthy young volunteers were selected for MRI study of articular cartilage in their knees,including 10 cases male and 10 cases female.All the volunteers had no symptoms of knee pain,neither any other medical records of knee joint disease.2.MR scanningAll images were acquired with a 3.0-T MR imaging unit(GE Signa EXCITE HD; GE healthcare,USA).After routine MR sequence scanning,sagittal imaging were performed with 3D-FS-SPGR sequence,FS-GRE sequence and FS-FSE sequence.3.MR images analysisAll images were transported into the AW4.3 workstation and PACS system to be analyzed.All the images were co-checked by two radiologists.In the AW4.3 workstation, the signal intensity of subcartilage bone tissue,muscle,fluid surrounding the articular cartilage and background were detected in 3D-FS-SPGR sequence,FS-GRE sequence and FSE sequence,respectively.The area of region of interest(ROI) was about 4mm~2. and Every ROI was detected repeatedly for three times and the average signal intensity was calculated.The SNR,SNR efficiency(SNR divided by the square root of the imaging time) for cartilage,subcartilage bone tissue,muscle and fluid surrounding the articular cartilage were calculated,respectively.Then contrast-to-noise ratio(CNR) of cartilage to the surrounding tissues was also acquired.4.Statistical analysisAll the data were expressed as Mean±SD and statistical software SPSS 13.0 was used.One-way ANOVA was used to evaluate the differences of SNR,SNR efficiency and CNR between the different MR sequences in cartilage,subcartilage bone tissue, muscle and fluid surrounding the articular cartilage.If equal variances assumed, Least-significant Difference(LSD) test was applied,otherwise Kruskal-Wallis test or/and Dunnett's T3 test was used.P≤0.05 was considered to indicate significant difference.Results1.Sensitive MR sequence for articular cartilage in normal knee1.1 Analysis of SNR and SNR efficiency for the normal articular cartilage in the knee and its surrounding tissuesSNR values of cartilage,subcartilage bone and muscle were significant different among three MR sequences,including 3D-FS-SPGR sequence,FS-GRE sequence and FSE sequence(P＜0.05).There was significant difference in SNR of fluid between FS-FSE sequence and FS-GRE sequence,or FS-FSE sequence and 3D-FS-SPGR sequence(P＜0.05).However,there were not significant difference between 3D-FS-SPGR sequence and FS-GRE sequence(P=0.776).Thus as to showing fluid,FS-FSE sequence was more sensitive than the other two,while 3D-FS-SPGR sequence was best to show cartilage.SNR efficiency of cartilage,subcartilage bone and fluid were significantly different among three MR sequences(P＜0.05).But as to the SNR efficiency of muscle,except that no difference was found between FS-GRE sequence and 3D-FS-SPGR sequence(P＞0.05),others were significantly different(P＜0.01).Since 3D-FS-SPGR sequence having a long scanning time,the SNR efficiency in various tissues would be decreased,but the SNR efficiency of cartilage was still higher than those in other two sequences.SNR efficiency of fluid in 3D-FS-SPGR sequence was lowest among three sequences.In muscle,it was close to that in FS-GRE sequence. But as to the subcartilage bone,it was in the medium among three sequences.1.2 Analysis of CNR for the normal articular cartilage in the knee and its surrounding tissuesSignificant differences of the CNR as for cartilage to subcartilage bone,and cartilage to fluid,could be found among the three sequences(P＜0.05).Except for those between 3D-FS-SPGR sequence and FS-FSE sequence,others were obviously different(P＜0.05) in CNR of cartilage to muscle.CNR of cartilage to bone in 3D-FS-SPGR sequence was the highest,while as to FS-FSE sequence,it was the lowest.In addition,CNR of cartilage to fluid in FS-FSE sequence was the highest, while as to FS-GRE sequence,the lowest.ConclusionArticular cartilage was shown as high signal intensity among the three sequences, including 3D-FS-SPGR sequence,FS-GRE sequence and FSE sequence.And 3D-FS-SPGR sequence was the best to show the articular cartilage,due to the highest SNR values and CNR of cartilage to surrounding tissues.Part 2:The comparison between MR anatomy and the body specimen of the articular cartilage in the kneeObjective1.To compare and evaluate the differences of the articular cartilage thickness, detected by two different methods:direct detection from corpse sample and from MR imaging with the sensitive sequences;2.To analyze and evaluate the location of different elements of cartilage in articular genus by means of special staining for the articular cartilage in the knee;3.To examine the cartilage thickness and T2~*relaxation time of the normal articular genus,by using the new 3T MR technology:3D-FS-SPGR sequence and T2~*GRE multiple echo sequence.1 Materials1.1 Sample group:The fresh articular genu samples from two normal adult young male corpses,medium stature,from the local country,were obtained(keeping fresh in a cold preservation).Under the situation that defrosted well,MRI scanning was carried on these 2 cases according to the routine MR sequences.After defrosting,the knees were dissected longitudinally at a thickness of 2cm each.1.2 Control group:20 cases,same as that in Part 1.2 Experimental instruments2.1 The MR instrument and scan parametersAll images were acquired with a 3.0-T MR imaging unit(GE signa VH/I;GE healthcare,Milwaukee,WI,USA).Sagittal MR scan was administrated in all the specimens following various MR sequences,including FS-FSE sequence,FS-GRE sequence,FS-SPGR sequence,having the same scanning parameters as those in part 1, and also following T2~*GRE multiple echo sequence at the same time.3 Corpse patellae dissection and histological processing3.1 Corpse patellae dissectionThe corpse patellae were preserved below -30℃,keeping in a stretching position. After being taken out and defrosted,they were dissected longitudinally using electronic saw by a professional technician in the department of anatomy.Each layer has a thickness of 2cm.Clean water was used to keep their surfaces clear and an absorbing cloth to keep them dry.Then screen following the anatomical position. Finally,the articular cartilage was acquired by a professor for the preparation of histological examination.3.2 histological processingThe articular cartilage specimens were sent to the department of pathology in Nanfang University.Routine procedures of neutral formalin fixation,alcohol dehydration,paraffin embedding and slicing were then completed.Subsequently,the slices were stained with a compound dyes,Victorial blue and Ponceau red.After xylene clearing,they were mounted with neutral gum for visualizing the collagen fibres in the cartilage matrix.4.MR images analysisIn the AW4.3 workstation,the images of 3D-FS-SPGR sequence were remodified, and the thickness of articular cartilage was examined.At the same time,T2~*map of articular cartilage was also remodified according to T2~*GRE multiple echo sequence, and T2~*relaxation time of articular cartilage was detected in T2~*map.5.Comparative analysis of articular cartilage between corpse sample and MR imaging Post-dissected specimens of articular genus were kept dry and photoed.The thickness of articular cartilage in some local region was examined with sliding caliper. And that in the same region of MR images was also examined.Then the differences between anatomical specimens and MRI were evaluated.6.Statistical analysisAll the data were expressed as Mean±SD and statistical soft SPSS 13.0 was used. One-way ANOVA was used to evaluate the differences of articular cartilage thickness and T2~*relaxation time in the normal volunteers.If equal variances assumed, Least-significant Difference(LSD) test was applied,otherwise Kruskal-Wallis test or/and Dunnett's T3 test was used.P≤0.05 was considered to indicate significant difference.Results1.Cartilage histologyWith the microscope,it was found that collage fibre was rich at the periphery of the articular cartilage,while the cartilage matrix was less and the chondrocytes were immature at this site.However,in the center of the articular cartilage,collage fibre was less,while the cartilage matrix increased and chondrocytes became mature.The running diversity of collagenous fibres resulted in the differences of MR images.2.Comparison between cartilage anatomy and MR imagingThe differences of cartilage thickness detected with these two methods were not obvious,which demonstrated the examination in MRI according to 3D-FS-SPGR sequence could represent the actual cartilage thickness of articular genus.3.A range of detecting data for cartilage thickness and T2~* relaxation time for the normal group could be obtained.Conclusion1.By comparing cartilage anatomy and MR imaging,3D-FS-SPGR sequence could represent the actual morphology and cartilage thickness of articular genus.2.By means of special staining for the articular cartilage in the knee,collage fibre was rich at the periphery of the articular cartilage,while the cartilage matrix was less. and the chondrocytes were immature at this site.However,in the center of the articular cartilage,collage fibre was less,while the cartilage matrix increased and chondrocytes became mature.This indicated the distribution of tissue elements in the articular cartilage was related with its function.3.A range detecting data of cartilage thickness and T2~*relaxation time for the normal group could be obtained.Part 3 The clinical application of MR imaging on the disease of articular cartilage in the kneeObjective1.The cartilage thickness at various phases of OA in the articular genus is to be detected by using MR 3D-FS-SPGR sequence.And its changes during the pathological progress of OA are also evaluated after being compared with those in the normal group;2.By comparing with the showing from anthroscope,the specificity and sensitivity of imaging cartilage are to be evaluated with some sensitive MR sequences for cartilage, and the preponderance of MRI in diagnosing the disease of articular cartilage is also to be estimated;3.By the detection of T2~* relaxation time for cartilage in knee OA of various phases and in normal group through T2~*GRE multiple echo sequence,the correlation of T2~* relaxation time changes with the development of pathological changes of OA patients is to be estimated.And through the analysis of T2~* relaxation time for that in acute OA,the changes of T2~* relaxation time are examined at the early time of cartilage injury.4.By the detection ofT2~* relaxation time for the pathological articular cartilage in the knee through T2~* GRE multiple echo sequence with multi-echoes,the progress of cartilage pathology was analyzed,in order to find the concealed pathological changes in cartilage injury as early as possible.Material and MethodSubjects1.Selected criterion:The subjects were selected from the outpatients of Nanfang Hospital ranged in the period between the June in 2006 and the January in 2008, having the symptoms of knee pain or functional impairment in their knees.All the cases had no history of arthroscopy and surgery in either knee.If the patients had a history of trauma,it was necessary that no signs of bone fracture in the knee would be shown from X-ray.According to the emergent time of trauma,these cases were classified into two stages.If the trauma happened within two months,it was in the acute period.However,the contrary was in the chronic period.Since having the similar pathogenesis,the chronic trauma-induced osteoarthritis would be incorporated into the degenerative osteoarthritis as one group.62 cases with the above criterion were included and their articular cartilage thickness was detected in this research.Among of them,27 cases were male and 35 cases female,which ranged in age between 19 and 65 years,with an average age of 37 years.36 cases of them performed with arthroscopy and surgery in knee after MRI examination,including 19 males and 17 females,which ranged in age between 21 and 61 years,with an average age of 44 years.56 of 62 cases carried on the measure of T2~*relaxation time,including 25 males and 31 females,with an average age of 36 years old. 2.Classification of casesAccording to the OA index made by WOMAC and K-L radiological diagnosis standard,the cases with OA in the knee could be further classified into gentle, medium and serious injury groups.3.Grading standard according to the cartilage injuryGradeⅠ:The cartilage is intact;GradeⅡ:The cartilage is lack in local area but the bone at this site is not exposed;GradeⅢ:The cartilage is broken with exposed subcartilage bone tissue,but it is integrate with no abnormal signals;GradeⅣ:The whole layer of cartilage is loss,and the bone underlying the articular surface has the abnormal signals.Unlike the classification based on the clinical symptoms,this grading standard is accord with the degree of cartilage injury,which helps to compare the MR images with those from anthroscope.InstrumentsIt is same as part 2.Image processingThe image processing is same as part 1.Estimation of the cartilage injury:Before the surgery,the images based on the sensitive MR sequences for cartilage were checked by three professors of radiology. Then the injury of articular cartilage and its grading are estimated from MRI.On the other hand,its grading from the anthroscope was judged by the clinical specialists of anthroscope.Both MRI and anthroseope grading were based on the double-blinded principle.Statistical analysisAll the data were expressed as Mean±SD and statistical soft SPSS 13.0 was used. One-way ANOVA was used to evaluate the differences of articular cartilage thickness and T2~* relaxation time according to 3D-FS-SPGR and T2~*GRE sequences between the normal volunteers and OA patients.If equal variances assumed,Least-significant Difference(LSD) test was applied,otherwise Kruskal-Wallis test or/and Dunnett's T3 test was used.P≤0.05 was considered to indicate significant difference.Given multiple factors affecting articular cartilage thickness and T2~* relaxation time,such as weight,height,body mass index(BMI),age and sex,non-parametric correlation analysis(Spearman's test) was used to evaluate the effects of these factors on the articular cartilage.P≤0.05 was considered to indicate significant difference.As to the degree of cartilage injury,the kappa coefficient analysis was used to evaluate the sensitivity and specificity of the three MR sequences,taking the results from anthroscope as a golden standard.Result1.Comparison of articular cartilage thickness between in the OA cases of various degrees and in the healthy volunteersIn the gentle OA group,there was no statistical difference of articular cartilage thickness with that in the normal volunteers,while that in a bulk of articular cartilage in serious OA group was significant different from that in the normal volunteers.In addition,multiple factors,such as sex,age,weight and BMI,might affect the thickness of articular cartilage in the knee.2.Comparison of T2~* relaxation time in cartilage between various OA groups and healthy volunteersIn the gentle OA group,the T2~* relaxation time for a bulk of articular cartilage region was obviously different from that in the healthy volunteers.However,there was no statistical difference of T2~* relaxation time in serious OA group,when compared with the healthy volunteers.The change of T2~* relaxation time for cartilage was also associated with the age,weight and height.In addition,other multiple factors might affect it.3.The images of cartilage pathological changes in three MR sequencesIn the FS-FSE sequence,the accumulated fluid in the cartilage defection appeared as an obvious high signal,and the cartilage was so as well.In the 3D-FS-SPGR sequence,both the accumulated fluid in the cartilage defection and cartilage defection or injury show as low signal,while the cartilage was in a high signal.In the FS-GRE sequence,the accumulated fluid in the cartilage defection appeared as low signal,and the cartilage was high signal.However,in the site of cartilage injury,slightly low signal or similar signal as the normal cartilage was shown.4.Evaluation of the FS-GRE,FS-FSE and 3D-FS-SPGR sequences on the pathological changes of articular cartilageIn the images of various MR sequences,there were 288 cartilage surfaces in 36 patients.The results of anthroscope indicated that there were 32 cartilage pathological changes in 21 cases.The 3D-FS-SPGR sequence was best to show the pathological changes of cartilage.Its sensitivity arrived at 99.2%and its specificity was 94.4%. Both were higher than other two MR sequences,including FS-GRE and FS-FSE sequences.The kappa value of the 3D-FS-SPGR sequence was 0.937＞0.75,which demonstrated the good consistency as the anthroscope when showing the pathological changes of cartilage.5.Evaluation of T2~* relaxation time on the pathological changes of articular cartilage According to the different incidence time and degree of OA in the knee,the value of T2~* relaxation time was also different.In the early stage of OA,the cartilage signal displayed a limit increase;while in the mid or late stage,it was similar as or slightly lower than the signal of surrounding normal cartilage.In the acute injury,the cartilage signal in the injury showed a high T2~* relaxation time.However,in the chronic injury period,T2~* relaxation time of cartilage decreased to or below the level of surrounding normal cartilage.Conclusion3.1 By comparing of articular cartilage thickness between in the OA cases and in the healthy volunteers,the articular cartilage thickness in a bulk of cartilage weigh-loading surface attenuated gradually,following with the aggregated OA disease.Except for the degree of disease,it might be affected by multiple factors, such as weight,age,sex,and so on.3.2 The sensitivity and specificity of 3D-FS-SPGR sequence for showing cartilage pathological changes were higher than other two sequences.In addition,it presented a good consistency with the anthroscope in showing the pathological changes of cartilage.3.3 The difference of T2~* relaxation time for cartilage was mainly between the gentle OA cases and normal group,which indicated the shape of articular cartilage varied not very much in the early time of OA,while the change of tissue structure and elements was true.3.4 In the acute phase of articular cartilage injury,the pathological changes mainly took place in the cartilage weight-loading surface of tibial condyle of femur or in the patella.At the same time,the detection of T2~* relaxation time could detect some earlier pathological change or some native change in the cartilage,which couldn't be found by anthroscope.