Pituitary microadenomas

Neuroradiology (1989) 31:293 298
Neuro — radiology © Spfinger-Verlag 1989
Originals

Does Gadolinium enhance their demonstration?
P. Macpherson 1, D. M. Hadley 1, E. Teasdale 1, and G. Teasdale 2
Departments of 1 Neuroradiology and 2 Neurosurgery, Institute of Neurological Sciences, Southern General Hospital, Glasgow, Scotland
Summary. Ten patients with biochemical evidence of a hormonally active pituitary adenoma were exam- ined by dynamic contrast enhanced computed tomo- graphy (CT) and then by pre and post Gadolinium- DTPA (Gd-DTPA) enhanced magnetic resonance imaging (MRI). Excluding one false positive case, CT and unenhanced MRI were comparable in the detection of microadenoma. Post Gd-DTPA exam- ination gave more clear evidence of the actual adenoma in two patients and aided in the demon- stration of a third. However, in two others all im- aging techniques failed to demonstrate the microadenoma subsequently found at surgery. On the post enhancement MRI it was easier to assess the relationship of a tumour to the cavernous sinus and to visualise the relationships of the parasellar carotid arteries.
Key words: Pituitary microadenoma – Magnetic res- onance imaging – Gadolinium DTPA – Computed tomography
Controversy exists as to whether computed tomo- graphy (CT) or non-enhanced magnetic resonance imaging (MRI) is more accurate in demonstrating pi- tuitary microadenoma [1-4]. Gadolinium-DTPA Dimeglumine (Gd-DTPA) enhances the normal pituitary gland by causing up to 60% increase in signal intensity on Tl-weighted MR images. Most adenomas show as a region of hypoin- tensity more apparent after enhancement of the nor- mal part of the gland [5]. With approval from the Hospital Ethics Committee and the informed con- sent of the patients we have conducted a trial of pre and post Gd-DTPA enhanced MRI versus dynamic high resolution CT to ascertain which of the imaging modalities corresponded more accurately with the subsequent surgical findings.
Patients and methods
Ten patients diagnosed by conventional biochemical criteria as having a hormonally actiw: pituitary tu- mour were studied. Five had evidence of prolactino- ma and five of acromegaly. There were six females and four males with age range 23-54 with a mean of 40 years. CT had confirmed the absence of mac- roadenoma, but had demonstrated a focal abnor- mality in only one, indirect evidence of a lesion in seven and had been normal in two. In all cases in this rather inconclusive group, trans-sphenoidal adeno- mectomy was planned so that they were deemed suitable for inclusion in the trial. Patients younger than 18, pregnant or with hepatic or renal insuffi- ciency, would have been excluded, but none of these factors were present in the patients presenting. Iodi- nated contrast medium was not administered within 24 h before or after injection of Gd-DTPA. Dynamic CT scanning was performed with a Philips Tomoscan 350 in the coronal plane, 3 mm overlapping slices being taken after the injection of 50 ml of iopamido1370 and during the injection of a further 50 ml. The voxel size was 0.43 x 0.43 x 3 mm. MR examination was performed on a 0.15 T Picker Vista 1100 resistive imager. From a sagittal pilot, TI- weighted (SE 500/32) 5 mm thick contiguous coro- nal sections with a 27 cm FOV and 256 x 256 matrix (voxel size 1 x 1 x 5 mm), were centred through the pituitary gland. After the preliminary investigation and without changing their position five patients were re-examined after receiving Gd-DTPA in a dose of 0.1 mmol/kg body weight and five after 0.2 mmol/kg. The Tl-weighted sequences were car- tied out immediately after the injection and repeated sequentially up to six times in the first hour and once again at 5 h. The CT and MR images were reported inde- pendently and prospectively. For both imaging mo-
294
Table 1. Biochemical, imaging and surgical findings: Patients 1-5 received 0.1 mmol/kg and 6-10 0.2 mmol/kg Gd-DTPA
Patient Sex Age Biochemistry CT MRI Surgical findings of adenoma Pre Gd Post Gd
1 M 30 Prolactin 4-5 700-1000 Normal Normal Normal left 2 F 39 Prolactin Expanded Expanded Expanded 5-6 2000 left left left left 3 F 25 Prolactin Expanded Expanded Focal hyperintensity 3-4 2500 right fight initially: later right isointense fight 4 F 24 Prolactin Expanded Focal hyper- Focal region became 500-700 centre intensity centre isointense: normal? 5 M 42 HGH Expanded Expanded Focal hypointensity 100 right right fight 6 F 35 Prolactin Expanded Expanded Expanded 2500 left left left 7 M 46 HGH 7.1-10 Normal Normal 8 F 54 HGH Expanded Expanded 11-20 centre and right centre and right 9 F 28 HGH Focal low Expanded 38-59 attenuation left left 10 M 54 HGH Expanded Focal hypo- 23 46 left intensity left Normal Expanded centre and right Expanded left Focal hypointensity and ring enhancement left
Soft
Soft
Soft
Fibrous centre adenoma not found 6-7 Fibrous wall right soft centre 3-4 left Soft 3-4 left Soft 8-9 centre Soft 5-7 left Soft 6-7 left Soft/solid
Normal levels: prolactin ~< 360 mU/1; HGH < 5 mU/1; Adenoma size in mm
dalities the findings were first divided into normal and abnormal. The latter group were then sub- divided into those in whom there was direct evidence of adenoma; focal low attenuation on CT; focal low or high signal on MRI and in those in whom there was only indirect evidence: localised increased bulk and/or displacement of stalk. The images were sub- sequently reviewed in the light of the surgical find- ings. Post Gd-DTPA enhancement was assessed visually and also calculated by normalising the tissue signal intensities to an area of normal white matter at the level of the pituitary gland. White matter has been shown to be unaffected by Gd-DTPA, thus eliminating small differen- ces caused by tuning and attenuation changes be- tween sequences. The difference between pre and post Gd-DTPA images were expressed as a percentage of the pre-Gd-DTPA signal intensity as follows:
Post-contrast adenoma Pre-contrast adenoma Post-contrast WM reference Pre-contrast WM reference 100 N– Pre-contrast adenoma 1 Pre-contrast WM reference
(WM = white matter)
Serum iron and renal and liver function tests were performed before Gd-DTPA injection and after- wards at 4 and 24 h in all and at 48 h in eight of the ten patients.
Results
The overall results are given in Table 1,
Pre-Gd-D TPA comparisons
In two patients both the CT and MRI appearances were normal. In the remaining eight patients each examination showed deviations from normal, but these were not always identical. MRI was considered to be equivalent to CT in respect of lesion detection in seven of the ten patients, better in two and poorer in one. In Patient 10, CT showed indirect evidence of a tumour while MR demon- strated a mass with low intensity centre (Fig. 1 a, b). In Patient 4, CT also demonstrated indirect evidence of an adenoma while MRI showed focal hyperintensity (not confirmed surgically) (Fig. 2). In Patient9, CT showed a focal low attenuation region while MR demonstrated indirect evidence only.
Post Gd-D TPA comparisons
a) With pre Gd-DTPA MRI. In two patients the post enhanced images remained normal. Of the eight whose initial examination had been abnormal, im-
295
proved information was obtained in three. In Pa- tientl0, whose pre Gd-DTPA examination was more informative than CT, the post enhancement image was even more helpful in that it showed a ring of enhanced signal round the low signal component,
Fig.la-e. Patient 10. a CT showing in- crease of pituitary gland bulk on left, but no focal region of low attenuation, b MRI shows hypointense region on the left. c Postenhancement MRI demonstrates a ring of enhancing tissue round the low sig- nal component
Fig.2a, b. Patient 4. a CT showing in- creased bulk centrally and to the left caus- ing erosion of the floor of the pituitary fossa, b MRI demonstrating focal hyperin- tensity on the left of the gland (No tumour found at surgery)
making it stand out more clearly (Fig. 1 c). Patient 5’s plain image had shown indirect evidence of a lesion but after Gd-DTPA the gland enhanced giving direct evidence of a low intensity lesion (Fig. 3). In the third (Patient3), enhancement of the actual tumour
296
Fig.3a, b. Patient 5. a MRI demonstrating isointense increased bulk of pituitary gland, b Post Gd-DTPA image showing a relati- vely hypointense region centrally and to the fight
Fig.4a, b. Patient3. a MRI showing increased bulk centrally. b Post Gd-DTPA examination demonstrating relative enhance- ment on an early image of the adenoma found at surgery
relative to the normal pituitary tissue was seen on the early images but it merged with the rest of the gland by 20 min (Fig.4). These last two patients had re- ceived the lower dose of Gd-DTPA. By visual assessment, the infundibulum, nasal mucosa and intracavernous segments of the cranial nerves III-VI enhanced in all patients, confirming that effective blood levels of the agent had been achieved. A typical graph of percentage enhance- ment in the gland and lesion is shown in Fig.5. Figure 6 depicts the finding in Patient3 (Fig.4) whose tumour showed visual enhancement on the first post Gd-DTPA image but became isointense at 20 min. Enhancement of the cavernous sinus allowed easier assessment of the relationship of a tumour to this structure, and facilitated visualisation of the flow void of the carotid arteries.
b) With CT. MRI was equivalent to CT in six, better in three and poorer in one. Surgical correlations. In Patient 4, CT showed in- direct evidence of a central adenoma, routine MRI a central focal hyperintensity and on post enhance- ment an isointense gland. At surgery the centre of the gland was noted to be fibrous but no adenoma was found (Fig. 2). An adenoma was found in the other nine patients. In two of these the imaging had been normal, but in the other seven the lesion was at, or within, the site considered by CT and MR to be ab- normal. Side effects. None of the patients expe- rienced any untoward effect from Gd-DTPA, either at the time of the injection or in the ensu- ing 48h. A few minor biochemical changes occurred but they were not considered to be clinically relevant. In one patient, the pre Gd- DTPA serum iron was 19 mol/1, 18 at 24 h reducing to 7 at 48h (normal range in our laboratory 10-30 tool/l). Measurement of the hepatocellular enzymes re- vealed a baseline AST of 24 U/1 in one patient, with the level falling to 12 at 24 h and rising to 45 at 48 h (normal range 10/40 U/l). No contrast reactions oc- curred in relation to dynamic CT.
100 – Gadolinium – DTPA 0.1 m mol/Kg IV 100 Gadolinium – DTPA 0.1 m mol/Kg IV
297
75-
E ~g 5o- .1=
25-
0 ~x i 10 15 20 25 30 35 40 45 50 55 5
5
Pituitary Gland 8 Adenoma
#
0 5
Time / min h
75
=
~ 5o
25
0
6
rp I I I I I I I I 1 I ~ I 0 5 10 15 20 25 30 35 40 45 50 55 5
T[me / rain h
Fig.5. Typical graph of percentage enhancement in the pituitary gland and adenoma after the administration of Gd-DTPA
Fig. 6. The percentage enhancements in Patient 3 whose tumour showed relative visual enhancement on the early post Gd-DTPA images only
Discussion
Davis et al. [1] stated that no series had convincing- ly shown unenhanced MR to be superior or even equivalent to contrast enhanced CT for the detec- tion of microadenomas. Pojunas et al. [2] reported that in llpatients with CT evidence of microadenoma (surgically proven in eight), MR (].3-1.5 T; 3 mm coronal slices) demonstrated a le- sion in only six. Davis et al. [3] investigated 25 pa- tients with suspected pituitary lesions (confirmed histologically in nine) and concluded that for mac- roadenoma MR is the investigation of choice but that CT is better for the demonstration of microadenoma. Conversely, Kulkarni [4] reported that in all patients considered clinically and endo- crinologically to have a microadenoma, a focal sig- nal abnormality was demonstrated by MR (1.5 T) in 83% and by CT in 42%. Davis et al. [1] evaluated three patients with microadenoma (mean diameter 7.9 mm). Each had a focal lesion of reduced attenuation on enhanced CT, corresponding with the surgical site of the adenoma. MR imaging in one patient demonstrated a hypointense focal lesion which was unchanged after Gd-DTPA, while the surrounding normal gland enhanced. The other two patients had homo- geneous sellar contents initially but a hypointense lesion was detected after Gd-DTPA in one. Dwyer et al. [6] examined by CT and MR a series of ten patients with Cushing’s Disease in whom a microadenoma was subsequently confirmed surgi- cally. CT had demonstrated the adenoma in three and pre Gd-DTPA examination (0.5 T: 3 mm sec- tions) in six (all five of the cystic lesions; one of the five solid lesions). Serial TI imaging after the ad- ministration of Gd-DTPA was performed. They
found that there was early enhancement of the gland and delayed enhancement of an adenoma when cystic (1 h). Two of the five solid lesions be- came apparent after the injection, in one because of enhancement of the normal pituitary gland and in the other because of relative enhancement of the adenoma. In the main, a solid adenoma and the pi- tuitary gland had similar enhancement characteris- tics. Doppman et al. [7] reported a series of eight patients with Cushing’s Disease in all of whom CT had failed to demonstrate pituitary adenoma sub- sequently confirmed surgically and/or histologi- cally in seven. In three, microadenomas were unde- tected both with and without Gd-DTPA, two were visible as hypointense foci following the agent but not on the unenhanced images, while in three microadenomas were seen both before and after Gd-DTPA on Tl-weighted images (1.5 T). We found CT and pre Gd-DTPA MRI to be comparable in the demonstration of pituitary microadenoma. After the administration of Gd- DTPA additional information was obtained in three of the 10 patients. In all of the patients, the gland showed early enhancement which gradually tailed off over five hours, while in all but two, tumours showed no intensity change relative to the gland at any time. In the one patient whose tumour enhanced early relative to the remainder of the gland the tu- mour was found to be soft, while in the patient with the ring enhancement, this did occur in a part found at surgery to be the solid rim of a mixed solid and soft lesion. The results of the present study and previous re- ports indicate that Gd-DTPA enhancement can sometimes increase the information obtained on MRI of patients suspected of having pituitary microadenoma. If one could rely on an imaging technique to demonstrate conclusively whether or not an adenoma was present, this would obviate the rare instances where clinical and biochemical assess- ment indicate the probable presence of an adenoma, but in whom surgical exploration proves negative. Unfortunately, this state has not yet been reached. We have previously shown that with CT, a suspected
298
lesion of < 5 mm in diameter is just as likely to be part of the normal pituitary as a microadenoma [8]. Further, in the present study, a gland found at oper- ation not to contain an adenoma (although fibrous) had shown false positive appearances on both CT and MRI while two homogeneously enhancing glands contained a tumour. In addition to the detection of an adenoma, in the imaging of patients with pituitary disorders evidence is sought of the vascular relationships of the gland. Scotti et al. [9] stated that no feature permitted cer- tain distinction between invasion and nonqnvasion of the cavernous sinus by microadenoma. Davis et al. [1] considered that the normal cavernous sinus en- hancement by Gd-DTPA made identification of the cavernous sinus extension by an adenoma difficult. On the contrary, because the cavernous sinus enhan- ces relative to any adjacent tumour or infiltration we found it easier to assess the relationship of an adeno- ma to the cavernous sinus. Furthermore, the en- hancement made it easier to appreciate the flow void in the carotid arteries and therefore facilitated the demonstration of the relationship between the gland and/or tumour and these vessels, a pre-requisite for trans-sphenoidal surgery.
Conclusion
Contrast enhanced CT and plain MRI are com- parable and either may be used routinely for the im- aging of suspected pituitary microadenoma. If the result is equivocal or normal, then Gd-DTPA (0.1 mmol/kg) enhanced MRI may give further lo- calising information pre-operatively. However, even with enhancement, MRI is not completely reliable in demonstrating whether or not a microadenoma is present and biochemistry remains the more defini- tive diagnostic investigation.
Acknowledgement. We are grateful to Schering Health Care Limited for the supply of Gadolinium-DTPA.
References
1. Davis PC, Hoffman JC, Malko JA, Tindall GT, Taker T, Av- ruch L, Braun IF (1987) Gadolinium-DTPA and MR imaging of pituitary adenoma: a preliminary report. AJNR 8:817-823 2. Pojunas KW, Daniels DL, Williams AL, Haughton VM (1986) MR imaging of prolactin-secreting nficroadenomas. AJNR 7: 209-213 3. Davis PC, Hoffman JC, Spencer T, Tindall GT, Braun IF (1987) MR imaging of pituitary adenoma: CT, clinical and surgical correlation. AJNR 8:107-112 4. Kulkarni MV, Lee KF, McArdle CB, Yeakley JW, Haar FL (1988) 1.5-T MR imaging of pituitary microadenomas: techni- cal considerations and CT correlation. AJNR 9:5-11 5. Kilgore DP, Breger RK, Daniels DL, Pojunas KW, Williams AL, Haughton VM (1986) Cranial tissues: normal MR appear- ance after intravenous injection of Gd-DTPA. Radiology 160: 757-761 6. Dwyer AJ, Frank JA, Doppman JL, Oldfield EH, Hickey AM, Cutler GB, Loriaux DL, Schiable TF (1987) Pituitary adenomas in patients with Cushing’s Disease: initial experience with Gd- DTPA-enhanced MR imaging. Radiology 163:421-426 7. Doppman JL, Frank JA, Dwyer AJ , Oldfield EH, Miller DL, Nieman LK, Chrousos GP, Cutler GB Jr, Loriaux DL (1988) Gadolinium DTPA enhanced MR imaging of ACTH-secreting microadenomas of the pituitary gland. JCAT 12:728-735 8. Teasdale E, Teasdale G, Mobsen F, Macpherson P (1986) High- resolution computed tomography in pituitary microadenoma: is seeing believing? Clin Radiol 37:227-232 9. Scotti G, Yu CY, Dillon WP, Norman D, Colombo N, Newton TH, de Groot J, Wilson CB (1988) MR imaging of cavernous sinus involvement by pituitary adenomas. AJNR 9:657-664
Received: 1 February 1989
Dr. P. Macpherson Department of Neuroradiology Institute of Neurological Sciences Southern General Hospital 1345 Govan Road Glasgow G51 4TF Scotland

PLACE THIS ORDER OR A SIMILAR ORDER WITH US TODAY AND GET AN AMAZING DISCOUNT