Objective:To compare the diagnostic value of three quantitative evaluation methods based on three-dimensional rapid fluid attenuation inversion recovery sequence (3D-FLAIR) vein-enhanced labyrinth images in endolymphatic hydrops.Methods:From October 2017 to April 2019, a retrospective study was conducted on 86 patients with unilateral otogenic vertigo who were admitted to Beijing Tongren Hospital, Capital Medical University. MRI was performed 8 h after the single-dose Gd-DTPA intravenously injection in all patients. Three evaluation methods were used to calculate the ratio of the endolymphatic area to the total lymphatic area, the ratio of the saccule to utricle area, and the ratio of the endolymphatic volume to the total lymphatic volume, respectively. The paired t test was used to compare the three ratios between the affected and healthy ears. With clinical diagnosis as the gold standard, the receiver operating characteristic (ROC) curve analysis was used to analyze the efficacy of three methods in diagnosing endolymphatic hydrops. Results:Totally 65 cases were finally diagnosed endolymphatic hydrops clinically. There were statistically significant differences of all the 3 ratios between the affected and healthy ears ( t=9.93, 7.22, 8.20, all P<0.001). The ROC curve showed that the area under the curve (AUC) of endolymph/total lymph area ratio, saccule/utricle area ratio, endolymph/total lymph volume ratio for diagnosis of endolymphatic hydrops were 0.882, 0.768, 0.884 (all P<0.001). And there were no significant differences between each paired AUCs (all P>0.05). Conclusions:All three methods of endolymph/total lymph area ratio, saccule/utricle area ratio, endolymph/total lymph volume ratio can quantitatively evaluate endolymphatic hydrops. The endolymphatic/total lymphatic area ratio method is still the most convenient method at present.

目的 探讨多参数MRI诊断突发性聋患者迷路出血的检出能力,为迷路出血患者提供优化的MRI扫描方案.方法 选取2016年1月至2020年8月首都医科大学附属北京同仁医院和首都医科大学大兴教学医院确诊的80例单侧突发性聋患者,收集临床和MRI资料,对符合迷路出血诊断标准患者的影像学资料进行分析,包括平扫T1WI、T2WI、T2-DRIVE-HR、增强T1WI及增强延迟3D-FLAIR 5个序列对迷路异常高信号的检出能力,比较分析平扫T1WI、增强T1WI及增强延迟3D-FLAIR序列的能力评分.结果 80例患者中有22例(27.5％)为迷路出血致突发性聋.22例(100.0％)增强T1WI和增强延迟3D-FLAIR均显示单侧迷路异常高信号,20例(90.9％)患侧迷路于平扫T1WI上显示异常高信号,5例(22.7％)T2WI显示更高信号.所有患者的T2-DRIVE-HR均未发现迷路异常信号.平扫T1WI、增强T1WI和增强延迟3D-FLAIR序列显示迷路异常高信号的能力评分分别为3(2,3)分、3(3,4)分和4(4,4)分,评分结果的Kappa系数分别为0.732、1.000和1.000,3个序列评分比较,差异有统计学意义(Z=37.520,P＜0.001),且增强延迟3D-FLAIR 序列显示迷路异常高信号的能力优于增强T1WI和平扫T1WI,差异有统计学意义(P＜0.05).结论 MRI能检出突发性聋患者的迷路出血,平扫T1WI可作为迷路出血致突发性聋患者迷路异常高信号的筛查序列,同时结合增强T1WI序列,尤其是增强延迟3D-FLAIR序列,能提高病变检出率.

目的 评价迷路炎所致单侧突发性聋患者MRI各序列迷路异常信号的显示能力,并探讨最佳阅片流程.方法 回顾性分析2016年1月至2021年8月就诊于首都医科大学附属北京同仁医院及首都医科大学大兴教学医院70例单侧突发性聋患者的MRI资料,其中符合本研究入组标准总计42例患者(男20例,女22例),平均年龄(47.3±13.9)岁;由两名医师分别评估患者5个MRI序列平扫T1WI、T2WI、水成像(T2-DRIVE-HR)、增强T1WI及增强延迟三维液体衰减反转恢复(3D-FLAIR)显示迷路异常信号的效能,并对平扫T1WI、增强T1WI、增强延迟3D-FLAIR三个序列进行评分,采用Kappa检验对结果行一致性分析;采用Friedman检验对平扫T1WI、增强T1WI、增强延迟3D-FLAIR三个序列的主观评分行统计学分析.结果 42例(100％,42/42)患侧迷路于增强T1WI及增强延迟3D-FLAIR上均显示异常高信号,28例(66.7％,28/42)患侧迷路于平扫T1WI上显示异常高信号,T2WI显示低信号3例(7.1％,3/42),T2-DRIVE-HR显示患侧低信号17例(40.5％,17/42),两名医师对平扫T,WI、T2WI、T2-DRIVE-HR、增强T1WI及增强延迟3D-FLAIR序列评估结果的Kappa系数分别为0.632、0.798、1.000、0.905、1.000;平扫T1WI、增强T1WI及增强延迟3D-FLAIR显示迷路异常高信号的能力评分中位数分别为2(2,3)、3(3,4)及4(4,4),两名医师对三个序列评分结果的Kappa系数分别为0.628、0.835、1.000,各序列总体评分有统计学差异(x2=72.641,P＜0.001),且两两间比较均有统计学意义(P均＜0.001).结论 对迷路炎所致突发性聋患者,平扫T1WI、增强T1WI能够发现迷路异常信号,但增强延迟3D-FLAIR相较于二者更敏感;T1WI和3D-FLAIR结合有助于该病的精准诊断.

Objective:To explore the clinical application value of each sequence by analyzing the characteristics of labyrinthine signal on MRI in patients with unilateral sudden deafness.Methods:Totally 52 patients of unilateral sudden deafness with inner ear MRI were analyzed retrospectively at Beijing Tongren Hospital, Capital Medical University from January 2016 to July 2019, all of which could find abnormalities in the labyrinth, including 17 cases of plain scan and 35 cases of enhanced scan, with sequences including plain T 1WI, enhanced T 1WI, plain and enhanced delayed 3D fluid attenuation inversion recovery (3D-FLAIR). The affected labyrinthine signal characteristics of each sequence were analyzed and the involvement sites were judged. The ability of each sequence to show labyrinthine abnormal signal was evaluated and scored. The Friedman test and Wilcoxon signed rank sum test were used to compare the subjective scores of the ability to show labyrinthine high signal in different sequences in plain and enhanced patients, respectively. Fisher′s exact probability method was used to analyze the relationship between the affected sites and the recovery of hearing, tinnitus and vertigo symptoms. Results:Fifty-two patients (100%, 52/52) showed labyrinthine high signal on T 1WI, 8 (15.4%, 8/52) showed higher signal and 3 (5.8%, 3/52) showed low signal on T 2WI. Thirty-five (100%, 35/35) showed high signal on enhanced T 1WI, among which 27 had enhancement (77.1%, 27/35). Fifty-two (100%, 52/52) showed significant high signal of the affected labyrinth on 3D-FLAIR (17 plain scan, 35 enhanced scan). The scores were 2 (2, 2), 3 (2, 3), 3 (3, 4) and 4 (4, 4) of T 1WI, enhanced T 1WI, plain and enhanced 3D-FLAIR respectively. The overall difference in subjective scores of plain T 1WI, enhanced T 1WI and enhanced 3D-FLAIR in enhanced patients was statistically significant (χ2=64.528, P<0.001), and the comparison between the two was statistically different (all corrected P<0.05). The plain 3D-FLAIR score was higher than the plain T 1WI in patients with a statistically significant difference ( Z=-3.729, P<0.001). Twenty-seven cases (51.9%, 27/52) exhibited high signal at the ampulla of semicircular canals, with a statistically significant difference in the distribution of hearing recovery or not ( P=0.001). Conclusions:Both T 1WI and 3D-FLAIR sequences can effectively identify the labyrinthine high signal, but the latter was better than the former of its ability to display, especially delayed enhanced 3D-FLAIR. The high signal at the ampulla of semicircular canals was a characteristic predictor of non-recovery of hearing.

Objective: To explore the value of speculating etiology of the magnetic resonance imaging (MRI) T1 weighted imaging (T1WI) labyrinthine high signal ratio in patients with unilateral sudden deafness accompanied by vertigo and tinnitus and its relationship with hearing prognosis. Methods: Fifty-two patients with unilateral sudden deafness accompanied by vertigo and tinnitus who were admitted to Beijing Tongren Hospital Affiliated to Capital Medical University from January 2016 to July 2019 were collected, including 27 males and 25 females, aged (47.7±15.1) years. The inner ear MRI data of 52 patients (17 plain scan, 35 enhanced scan) with unilateral sudden deafness were retrospectively analyzed. Two radiologists independently measured the labyrinthine high signal intensity of the affected side and the contralateral normal side on T1WI and enhanced T1WI and calculated the signal ratio (the normal labyrinth signal was subtracted from the affected signal and then divided by the normal signal). The etiology of the enhanced group was judged based on two methods, including whether the abnormal high signal was enhanced or not (unenhancement indicated hemorrhage and enhancement indicated inflammation), and the locations of labyrinthine involvement on enhanced three-dimensional fluid attenuated inversion recovery (3D-FLAIR) (inflammation usually involved the perilymph spaces, while hemorrhage involved the perilymph and endolymph spaces). In the plain group, the locations of labyrinthine involvement on 3D-FLAIR was applied to infer the potential etiology. Results: The two methods presumed that 8 cases might be hemorrhage (22.9%, 8/35) and 27 be inflammation (77.1%, 27/35) in the enhanced group, which had a high consistency, while it was speculated that 7 patients might be hemorrhage (7/17) and 10 patients be inflammation (10/17) in the plain group. The measurement results of the two radiologists were highly consistent within and between the groups [the intraclass correlation coefficient (ICC) values were greater than 0.800]. The area under the receiver operating characteristic (ROC) curve (AUC) of the T1WI high signal ratio in the enhanced group for speculating etiology was 0.949 (P<0.01), when the predictive threshold value was 0.467, with a sensitivity of 96.3% and a specificity of 87.5%. It might be hemorrhage when the ratio was higher than the threshold value, otherwise it was inflammation. The T1WI labyrinthine high signal ratio was higher in the hemorrhage group than that of the inflammation group, and the hearing prognosis was worse (all P<0.05). The T1WI labyrinthine high signal ratio of the unrecovered group was higher than that of the recovered group (P=0.034). Conclusions: The etiology of labyrinthine high signal formation can be inferred by quantitative values combined with the involved sites. The high signal in the labyrinth indicates poor hearing prognosis, the higher the signal intensity, the greater the possibility of hemorrhage and the worse the hearing prognosis.

Objective:To investigate the imaging findings of incomplete partition type Ⅲ (IP-Ⅲ) cochlea malformation using high-resolution CT (HRCT) and MRI, and to measure the key anatomical structures, providing the accurate qualitative and quantitative data for cochlear implantation (CI).Methods:Totally 14 patients (28 ears) with IP-Ⅲ cochlea malformation who underwent cochlear implantation at Beijing Tongren Hospital, Capital Medical University from February 2012 to March 2019 were retrospectively collected. All the patients were male, aged 7 months to 27 years old, with the median age as 3 years old. All the 14 patients underwent preoperative HRCT and 9 of them underwent preoperative MRI. The development of inner ear structure, including cochlea, vestibule, semicircular canals, vestibular aqueduct and internal auditory canal (IAC) was reviewed and analyzed. The travel route and position of labyrinthine, tympanic and mastoid segment of facial nerve canal were evaluated; the width of labyrinthine segment of facial nerve canal and superior vestibular nerve canal, as well as the angle between first and second parts of the facial nerve canal were measured on HRCT. The shape of stapes and the development of cochlear nerve were analyzed on MRI.Results:All the 14 cases (28 ears) showed nearly normal shape of the cochlea, with the bony interscalar septa presenting while the modiolus completely absent. The lateral portion of the IAC was dilated, and the septum was absent between the base of the cochlea and the IAC, appearing as a"gourd-like"shape. A small saclike protrusion was formed in the vestibule and protruded into the upper semicircular canal in 10 cases (20 ears) (71.4%, 20/28). The beginning of the vestibular aqueduct enlarged in 12 cases (24 ears) (85.7%, 24/28). All the 14 cases (28 ears) showed that labyrinthine segment of facial nerve canal was located almost above the cochlea and showing stiffly. The labyrinthine segment of facial nerve canal widened in 7 cases (14 ears) (50.0%, 14/28) and the superior vestibular canal widened in 13 cases (26 ears) (92.9%, 26/28). The width of labyrinthine segment of facial nerve canal and the superior vestibular canal were (1.14±0.37) mm and (1.66±0.35) mm, respectively. The angle between the first and second parts of the facial nerve canal was 96.83°±15.63°. Eleven cases (22 ears) (78.6%, 22/28) showed thickened footplate of stapes and poor oval window, but the round window was clear. Nine cases (18 ears) showed normal development of the cochlear nerve on MRI.Conclusion:IP-Ⅲ cochlea malformation has the characteristic imaging features. Preoperation accurate assessment of the shape and location of important anatomical structures such as cochlea, internal auditory canal and facial nerve can provide valuable information for CI.

Objective: To investigate the visualization of endolymph in patients with otogenic vertigo by intravenous administration of single dose of gadolinium contrast agents and magnetic resonance three-dimensional fluid-attenuated inversion recovery sequence (3D-FLAIR MRI), and further assess the extent of endolymphatic hydrops. Methods: From Beijing Tongren Hospital of Capital Medical University between October 2017 and June 2018, 30 patients (16 males, 14 females) with unilateral otogenic vertigo were involved in this study, with the age of 30 to 68 years, mean age of (53±10) years. Eight hours after intravenous administration of single dose (0.1 mmol/kg, body weight) of gadopentetate (Gd-DTPA), 3D-FLAIR sequence was performed in 30 patients. The location of endolymphatic hydrops was observed and then the degree of hydrops was quantitatively elevated by two radiologists. The consistency test was used to analyze the location and degree of endolymphatic hydrops in the two radiologists and the paired t-test was used to compare the difference between the affected and healthy side of endolymphatic spaces of the patients with otogenic vertigo. Results: In 30 patients, the gadolinium distributed in all parts of the perilymph inside the inner ear, and can accurately outline the boundaries of the peri-and endolymph. Twenty-six patients (26/30, 86.7%) were found to have unilateral endolymphatic hydrops, including 18 mild hydrops, 8 significant hydrops. The two radiologists had a very good agreement on the assessment of endolymphatic hydrops(kappa=0.864, ICC=0.959). In the 3D-FLAIR MR images of 26 patients with endolymphatic hydrops, the saccule (26/26, 100%) had a higher rate of hydrops than the cochlea and utricle(16/26, 61.5%; 14/26, 53.8%), and two radiologists had a very good agreement on the location of endolymphatic hydrops(kappa=0.820). Moreover, there was a significantly statistical difference between the affected and healthy area of the endolymphatic space in this study (P<0.01). Conclusion: The technique of 3D-FLAIR MR imaging through single dose intravenous gadolinium injection is feasible, which can estimate endolymphatic hydrops in patients with otogenic vertigo, and accurately classify the degree of hydrops.

目的 探讨扩散加权成像(DWI)在鉴别泪腺IgG4相关性病变(IgG4-RLD)与易混淆泪腺淋巴瘤、炎性假瘤的价值.方法 搜集2010年9月至2017年12月经证实10例IgG4-RLD、10例淋巴瘤及9例炎性假瘤患者的MRI资料;以IgG4-RLD为研究对象,淋巴瘤和炎性假瘤为对照组.采用方差分析、LSD检验、秩和检验、x2检验、Fisher精确检验比较三组结果[DWI信号、表观扩散系数(ADC)信号、ADC值、病灶与ADC脑桥值的比值(ADC-R)、SI病灶值与背景噪声标准差的比值(SNR)]差异以及采用受试者工作特征(ROC)曲线分析上述结果预测IgG4-RLD效能,P＜0.05认为差异有统计学意义.结果 IgG4-RLD和淋巴瘤、炎性假瘤在ADC-R、SNR均有统计学差异(P＜0.05);淋巴瘤和炎性假瘤在DWI和ADC信号上有统计学意义(P＜0.0125);淋巴瘤和炎性假瘤的ADC值有统计学差异(P＜0.05).预测IgG4-RLD效能:DWI信号:曲线下面积(AUC)=0.563,P=0.5303;ADC信号:AUC=0.532,P=0.7459;信号强度(SI)值:AUC=0.532,P=0.8106;SI病灶与SI脑桥比值(SIR):AUC=0.589,P=0.4605;SNR:AUC=0.532,P=0.7909;ADC值:AUC=0.468,P=0.7799;ADC-R:AUC =0.5,P=I.0;上述四个定量指标(SIR、SI、ADC值、ADC-R)联合:AUC =0.658,P=0.1492.结论 SNR、ADC-R有助于IgG4-RLD和淋巴瘤、炎性假瘤的鉴别;DWI信号、ADC信号、ADC值有助于淋巴瘤和炎性假瘤的鉴别.单一指标预测IgG4-RLD效能较低,而联合指标预测IgG4-RLD效能有所提高.