Quantitative Assessment Of Pituitary Lesions With Dynamic Contrast-enhanced MRI

Purpose: To identify nonfunctioning adenomas,microprolactinoma,Rathke cleft cyst(RCC),hypopituitarism and normal pituitary with semi-quantitative analysisof the time intensity curve(TIC)by dynamic contrast enhanced MRI(DCE-MRI).To explore the application value of TIC parameters to differential diagnosis the pituitary lesions,including maximum intensity(SImax),time to peak(Tmax),the enhancement rate(E1-7)and peak enhancement rate(PER).Also with the fast volume acquisition sequence(liver acquisition with volume acceleration,LAVA)scan for large pituitary adenoma,we try to use the quantitative analysis of A and B softwares to explore the difference between two postprocessing softwares.Methods: The first part is an retrospective analysis of 88 cases with pituitary lesions and 23 controls,whose MRI of pituitary and hormone function tests were normal.Twenty-seven patients with nonfunctioning adenomas(14 men and 13 women;mean age=55±25 years),twenty-six patients of microprolactinoma(6 men and 20 women;mean age=37±6 years),sixteen patients with RCC(6 men and 10 women;mean age=52±24 years),nineteen patients with hypopituitarism(13 men and 6 women;mean age=30±37 years),twenty-three controls with normal pituitary form and function called normal group 1(10 men and 13 women;mean age=52±29 years).Due to the age of hypopituitarism and microprolactinoma,we generated normal group 2,including 19 cases(8 men and 11 women;mean age=52±29 years).The second part collected 15 cases of pituitary disease prospectively(8 men and 7 women,mean age=49±12 years).Imaging was performed on a 3.0 Tesla GE scanner using an 8-channel head coil.The first part acquired 7 phase imaging.We used Functool to postprocess the data of nonfunctioning adenomas,microprolactinoma,RCC,hypopituitarism and normal pituitary.The TIC was extracted from DCE-MRI.Ten parameters,including gmaximum intensity(SImax),time to peak(Tmax),the enhancement rate(E1-7)and peak enhancement rate(PER).The second part is to postprocesse the data from LAVA scaning(30 phase images)with A and B softwares,extracting Ktrans,Ve and kep.The delayed phase rate was extracted from early and delay phase of coronal T1 imaging.The collected data were compared through the SPSS 19.0.The analysis of count data used chi-square test,two independent samples t-test and ANOVA,also K-S and K-W tests were used to compare the mean values of ten parameters in nonfunctioning adenomas,microprolactinoma,RCC,hypopituitarism and normal pituitary groups.Receiver operating characteristic(ROC)test was used to assess the ability of parameters between nonfunctioning adenomas and RCC,hypopituitarism and normal pituitary groups.The correlation between PRL values and ten parameters,Ktrans,Ve,kep and the delayed phase rate were tested using Spearman’s correlation.All statistical results were P<0.05 as statistically significant.Results: 1.TIC type: Normal pituitary gland showed a fast upward plat curve(Ⅰtype);hypopituitarism,microprolactinoma and nonfunctioning adenomas showed a slowupward curve(Ⅱtype)and RCC showed a aflat curve(Ⅲtype).The signal values were decreased by normal pituitary,hypopituitarism,microprolactinoma,nonfunctioning adenomas and RCC.2.The TIC parameters among nonfunctioning adenomas,RCC and normal pituitary showed: SImax,Tmax,E2-7and PER had statistical significance(P<0.05).Compared with normal pituitary,there were significant differences in SImax,E3-7 and PER in nonfunctioning adenomas(P<0.05).Compared with RCC,normal pituitary and nonfunctioning adenomas,there were significant difference(P<0.05)between PER,Tmax,E2-7 and SImax.3.Compared with normal group 2,SImax,E1,E3-7 and PER in hypopituitarism had statistical difference(p<0.05);the SImax of microprolactinoma had statistical difference(P =0.001).4.Compared with 10 parameters,the AUC(0.965)of joint variable was the best one to identify the nonfunctioning adenomas and RCC,the best diagnostic value was 0.838,the sensitivity value was 0.875,the specificity value was 0.963.Also the AUC(0.872)of joint variable to identify hypopituitarism and normal pituitary was the best one,the best diagnostic value was 0.650,the sensitivity was 0.696,and the specificity was 0.955.5.Spearman analysis of serum PRL and values of the 10 parameter in patients with microprolactinoma were not had correlated(all P>0.05).6.A and B softwares showed Ktrans、Ve had statistical difference(p=0.000,0.009),while the difference between,Kep had no statistically difference(p=0.566).7.The values of Ktrans,Ve,Kep were not correlated with the delayed phase rate(all P>0.05).Conclusion: 1.Pituitary diseases,including nonfunctioning adenomas,microprolactinoma,RCC,hypopituitarism and normal pituitary can be identified by TIC and semi-quantitative parameters based on 7 phase imaging.2.The joint variable is the best parameter to identify the nonfunctioning adenomas and RCC,hypopituitarism and normal pituitary.3.The existing quantitative analysis of DCE-MRI softwares contain difference,so it’s essential to standardize the influencing factors.