Clinical Application Of Multi-parametric Magnetic Resonance Imaging In Diagnosis Of Posterior Scleritis

Background and Purpose:Posterior scleritis(PS)is a rare form of scleral inflammation that may involve adjacent structures,including the retina,choroid,and optic nerve,eventually resulting in visual impairment.The PS which has deep position,various clinical manifestations,and no specific ocular signs,is one of the most misdiagnosed and treatable ophthalmic diseases.PS is often misdiagnosed as fundus tumors such as choroidal melanoma,which may cause unnecessary enucleation.At present,the diagnosis of PS is mainly via the ultrasound findings and clinical history.However,the diagnostic stability resolution and soft tissue resolution of ultrasound is poor;and the ultrasound findings of some PS patients show no abnormalities.So PS is still lack of effective diagnostic methods.General or local application of long term,decreasing and enough doses of glucocorticoid can make most of PS patients that inflammation is quickly reduced and controlled.However,glucocorticoid efficacy is not good for a small number of patients with PS.The blind use of glucocorticoid will induce or aggravate the infection,cause metabolic disorder,osteoporosis and gastrointestinal bleeding,etc.Therefore,it is very important to accurately predict the glucocorticoid efficacy for patients with PS to guide clinical treatment.At present,the pathophysiological mechanism,and the clinical diagnosis and treatment of PS have been reported,but there is no effective means of prognostic evaluation.In recent years,magnetic resonance imaging(MRI)has been widely used in the study of PS.Readout segmentation of long variable echo-train diffusion-weighted imaging(RESOLVE-DWI)is an oninvasive method that can observe water molecular diffusion.Compared with conventional diffusion-weighted imaging(DWI),RESOLVE technology can improve the quality of DWI images,and reduce motion artifacts,distortion artifacts and blur effect.The technology has been widely used in various parts of human body and many diseases,including breast cancer,liver cancer and stroke.By analyzing the apparent diffusion coefficient(ADC)values of RESOLVE-DWI not only can reflect the internal organization characteristics of lesions,but also identify the different pathological types of lesions and predict the therapeutic effect.At the same time,conventional MRI and enhanced scan can intuitively and accurately display normal structure and pathological changes in anatomical structure.Multi-parametric MRI(Mp-MRI)as a multi-scan parameters,multiple imaging sequences,high soft tissue resolution and no radiation inspection technology,has been widely used in various types of disease pathology,diagnosis,treatment and prognosis research.There are few reports on MRI performance at home and abroad.Based on,the purpose of this study:(1)the diagnostic value of Mp-MRI combined with conventional MRI,enhanced MRI,RESOLVE-DWI scan and ultrasound was compared;(2)evaluate the value of Mp-MRI in predicting the efficacy of glucocorticoid in patients with PS.Materials and Methods:1.Case data:Part 1: Fifty-six patients(83 eyes)with ocular fundus occupying lesions were recruited from September 2014 to January 2017 in our hospital.There were 24 males and 32 females Cases,aged from 18 to 85 years,mean(52.8±19.3)years,duration of 3d ~ 2 years.56 patients were detected 83 single lesions.Confirmed by pathology or clinical follow-up,there are posterior scleritis in 30 cases(40 eyes),choroidal melanoma in 13 cases(22 eyes),in 5 cases(8 eyes)and orbital inflammatory pseudotumor in 8 cases(13 eyes)among them.Part 2: Twenty-eight patients(38 eyes)with PS who had not been treated with glucocorticoid and had primary treatment in ophthalmology outpatients in our hospital were recruited from the 30 PS patients in the first part of this study.There were 12 males and 16 females Cases,aged 18 to 83 years,mean(48.0±19.9)years,duration of 5d ~1 year.Equipment and methods:Part 1: Magnetic resonance examination used Siemens Tim Trio 3.0T magnetic resonance instrument and ultrasound examination used French IneScan S ophthalmic ultrasound system.All patients underwent routine MRI,enhanced MRI and RESOLVE-DWI and ultrasound.Three radiologists who have performed more than 5 years of diagnosis work were selected to observe and analyze the MRI results of the fundus lesions(including morphology,border,edge and internal enhanced performance)and ultrasound performance.With the surgical pathology or clinical follow-up results as the gold standard,the sensitivity,specificity,accuracy,positive likelihood ratio and negative likelihood ratio of PS were measured by multi-parameter MRI and ultrasound,respectively.We use MedCalc 16.0 software drawing receiver operating characteristic curve(ROC),calculate the area under the curve(AUC),and AUC was compared between the two groups by Z test,P <0.05 for the difference was statistically significant.Part 2: Magnetic resonance examination used Siemens Tim Trio 3.0T magnetic resonance instrument.One day before,2 weeks after treatment,4 weeks after treatment,12 weeks after treatment and 4 weeks after the end of treatment of glucocorticoid therapy(1 mg/kg of oral prednisone,10 mg every 2 weeks until 10 mg was maintained until the end of treatment),28 patients(38 eyes)of the 30 PS patients in the first part of this study underwent conventional MRI,enhanced MRI and RESOLVE-DWI examination.Three radiologists who have been working for more than 5 years of diagnostic work,at 1 day before treatment and One day before,2 weeks after treatment,4 weeks after treatment,12 weeks after treatment and 4 weeks after the end of treatment,avoiding the cystic and necrotic areas in the ADC maps without knowing the patient grouping,sketched out the lesion region of interest(ROI)of PS lesions,measured 3 times ADC value and took the average.PS lesions border referred to routine MRI and enhanced MRI images,manually outline the border of the lesions and measured the largest diameter of the lesions before and after treatment of glucocorticoid,maximum diameter retraction rate = [(maximum diameter before of treatment-maximum diameter after treatment)/ maximum diameter before treatment)× 100%.According to the evaluation criteria for the efficacy of glucocorticoid,all patients should be divided into effective group and ineffective group.Calculating the rate of change of ADC value(?ADC),?ADC= [(average ADC value after of treatment/ the average ADC value before treatment)/ the average ADC value before treatment] ×100%.Statistical analysis was performed using IBM SPSS Statistics Version 19.0 software.The measured data were expressed in the form of ?x±s,P <0.05 was statistically significant.Before and after glucocorticoid treatment,the maximum diameter and ADC value of the PS lesions of effective group and ineffective group were analyzed using two independent samples t test;before and after glucocorticoid treatment,the maximum diameter and ADC value of the PS lesions of effective group and ineffective group were compared by paired samples t test;The ADC value of the PS lesion before treatment with glucocorticoid and the maximal diameter before treatment;the ADC value and the maximal diameter retraction rate of the PS lesion before treatment;the ADC value of the PS lesion after treatment and the maximal diameter after treatment and the ?ADC and maximal diameter retraction rate after treatment were analyzed by Spearman correlation analysis;using MedCalc 16.0 statistical software to draw ROC curve analysis of glucocorticoid treatment before ADC value prediction of glucocorticoid efficacy to determine the threshold,P <0.05 for the difference was statistically significant.Results:Part 1: MP-MRI results showed that PS lesions were nodular(13 cases of 18 eyes)and diffuse type(10 cases of 15 eyes).Thirty-one lesions showed isointensity on T1 WI,hyporintensity on T2 WI,obviously homogeneous enhancement on contrast-enhanced MRI,slightly limited on RESOLVE-DWI and isointensity on ADC maps;two lesions showed isointensity on T1 WI,hyperintensity on T2 WI,obviously homogeneous enhancement on contrast-enhanced MRI,slightly limited on RESOLVE-DWI and hypointensity on ADC maps.The typical "T" signs could be seen in 15 cases(24 eyes),accounting for 72.7%.Ultrasound showed that PS lesions were nodular(6 cases of 7 eyes)and diffuse type(10 cases of 12 eyes).All lesions showed moderate or strong echo,a few blood flow signal and non-obvious sound attenuation.The typical "T" signs could be seen in 10 cases(12 eyes),accounting for 70.6%.The sensitivity,specificity,accuracy,positive likelihood ratio and negative likelihood ratio of Mp-MRI and ultrasound were 82.5%,90.7%,83.1%,8.87,0.19 and 47.5%,88.4%,35.9%,4.08,0.59,respectively.The area under ROC curve(AUC)of MP-MRI and were 0.883 and 0.726,respectively.The difference was statistically significant(Z = 2.542,P = 0.011).Part 2: 28 patients(38 eyes)with PS were divided into effective group group(17 cases of 24 eyes)and ineffective group(11 cases of 14 eyes)according to the evaluation criteria of glucocorticoid efficacy.There was no significant difference in gender,age and course of PS between the effective group and the ineffective group(P> 0.05).The ADC value(1.36 ± 0.11)× 10-3 mm2/s was significantly higher than that of the ineffective group(1.12 ± 0.41)×10-3mm2/s,the difference was statistically significant(P =0.036).The ADC value of the effective group was(1.45±0.14)×10-3mm2/s at 2 weeks after treatment,which was significantly higher than that before treatment(P<0.01).After 4 weeks,12 weeks and 4 weeks after the end of treatment,the ADC value of the invalid group was(1.14 ± 0.37)× 10-3 mm2 / s,which was slightly higher than that before treatment(P = 0.285).PS values were(1.53±0.13)×10-3 mm2/s,(1.60±0.13)×10-3 mm2/s and(2.01±0.14)×10-3 mm2 / s,respectively,compared with those before treatment(P <0.01),but there was no significant change in the ineffective group.The AUC of the ROC curve of the ADC value before treatment of glucocorticoid was 0.747 and the ADC value was 1.33×10-3 mm2 / s as the effective threshold for the treatment of glucocorticoid,and the sensitivity and specificity of the predicted treatment were 87.5% and 75.0 %.Conclusion:Part 1: By comparing the diagnostic value of Mp-MRI and ultrasound on PS,Mp-MRI is superior to ultrasound in diagnosis of PS,which can not only diagnose early,but also can accurately determine the extent of inflammatory exudation.At the first suspicions of this disease in clinical work,Mp-MRI can be used as an important examination method.Part 2: By evaluating the value of Mp-MRI in predicting the efficacy of glucocorticoids in PS patients,the ADC value of the PS lesion in the treatment of glucocorticoid was earlier than that of the maximal diameter.MP-MRI can predict the efficacy of glucocorticoid in the early stage by monitoring the ADC value of the PS lesion and the ADC value of 2 weeks after the treatment of glucocorticoid.