- Email: [email protected]
Evaluation of pituitary microadenomas with dynamic MR imaging
European Journal of Radiology 41 (2002) 131– 135 www.elsevier.com/locate/ejrad
Evaluation of pituitary microadenomas with dynamic MR imaging Thomas Rand a,*, P. Lippitz b, E. Kink c, H. Huber d, B. Schneider e, H. Imhof a, S. Trattnig a a
Department of Radiology, Uni6ersity of Vienna, Wa¨hringer Gu¨rtel 18 -20, 1090 Vienna, Austria b Department of Radiology, E6ang KH Wien, Vienna, Austria c Department of Medicine, Uni6ersity of Vienna, Vienna, Austria d Department of Gynecology, Uni6ersity of Vienna, Vienna, Austria e Department of Medical Statistics, Uni6ersity of Vienna, Vienna, Austria Received 10 June 1999; received in revised form 30 August 2001; accepted 31 August 2001
Abstract Purpose: To evaluate the use of keyhole dynamic magnetic resonance (MR) imaging in the evaluation of women with borderline hyperprolactinemia. Patients and methods: We investigated 30 patients (mean age 31.6 9 4.3 years) with moderate hyperprolactinemia and clinically suspected microadenomas on T1-weighted spin echo (SE) sequences (repetition time (TR)/echo time (TE) 625/15 ms) with keyhole dynamic imaging and contrast enhancement of the pituitary gland. The probability of lesion presence according to MR criteria was ranked on a five-grade scale and compared with standard examinations of the hypophysis before and after intravenous administration of contrast media. Results: The probability for lesion presence showed a trend toward lower scores on standard SE sequences and higher scores with dynamic imaging (P =0.067). Conclusion: Dynamic keyhole imaging of the pituitary gland could help establish a diagnosis of microadenomas with a higher probability. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: MRI; Keyhole dynamic imaging; Pituitary gland; Microadenomas
1. Introduction MR imaging has provided excellent results in the evaluation of adenomas of the pituitary gland [1–3]. Dynamic magnetic resonance (MR) imaging of the normal pituitary gland and pituitary adenomas has been performed in several studies [4 – 13]. Although the results with regard to enhancement of pituitary adenomas in relation to the anterior lobe are somewhat disparate, a sequential MR enhancement pattern can be observed and microadenomas are best visualized at earlier phases of contrast enhanced dynamic images, with adenomas demonstrating lower signal intensity compared with normal pituitary parenchyma. An opposite enhancing pattern was rare. Yuh et al. [6] found pituitary adenomas to enhance earlier than the lobe and suggested that this might be * Corresponding author. Fax: + 43-1-40400-4898. E-mail address: [email protected] (T. Rand).
due to direct blood supply, similar to that of the posterior pituitary lobe. Technological advances have made available dynamic keyhole imaging with spin echo sequences (kSE) [8,14]. The aim of our study was to evaluate this new technique in the routine management of possible pituitary endocrinopathies, and particularly for the determination of which patients might benefit from pituitary imaging on the kSE basis for further diagnostic accuracy.
2. Patients and methods Our study group included 30 patients (mean age 31.69 4.3 years) with clinically suspected evidence of pituitary adenomas and borderline blood prolactin levels (HPRL) (64.19 57.0 ng/ml; mean9 S.D.; (min, 4 ng/ml; max, 224 ng/ml)). Blood sampling was per-
0720-048X/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 7 2 0 - 0 4 8 X ( 0 1 ) 0 0 4 1 2 - 0
132
T. Rand et al. / European Journal of Radiology 41 (2002) 131–135
formed because of symptoms such as sterility, galactorrhea, amenorrhea or oligomenorrhea. Hormone analysis was performed with enzyme linked immunosorbent assay (ELISA; Boehringer Mannheim). These patients had previously undergone evaluation of the pituitary gland in a routine examination using conventional SE images. Imaging was performed in the sagittal and coronal planes with 3-mm slice thickness, and no interslice gap. The field of view was 23 cm and the acquisition matrix was 256× 256. T1-weighted SE images were acquired with a repetition time (TR) of 500–625 ms and an echo time (TE) of 15 ms, before and after application of 0.2 ml/kg gadolinium DTPA (Magnevist, Schering). Pituitary adenomas were evaluated according to a five-grade scale under the following criteria (Fig. 1). 0, no evidence of microadenomas; I, slight probability: asymmetry of hypophysis morphology in the endosellar region; II, moderate probability: asymmetry and inhomogeneity in the endosellar region; III, high probability: demarcated focal hypointense area after contrast application; and IV, proof of lesion: focal hypointense area on preand postcontrast images associated with increase in volume of the hypophysis. Further evaluation of these patients was performed within 2– 4 weeks after the first investigation using
T1-weighted SE sequences (TR/TE, 100/20 ms) with keyhole dynamic imaging and contrast enhancement according to an SE protocol. Three slices in coronal orientation through the hypophysis in ten dynamic images per slice were acquired at 11-s intervals. All scans had 3-mm slice thickness and no interslice gap. Dynamic imaging was initiated simultaneously with the bolus injection of contrast material (gadopentetate dimeglumine, 0.2 ml/kg). FOV was 180× 180 cm. Probability or proof of the lesions was evaluated by two readers and ranked on a five-grade scale according to the following criteria. 0, No evidence of a lesion; I, poor evidence of lesion presence: inhomogeneity of enhancing hypophyseal tissue; II, moderate evidence of lesion presence: poor demarcation of a hypointense area on a single slice; III, high suspicion of lesion presence: well-demarcated lesions on a single slice; IV, proof of lesion: focal hypointense lesion well-demarcated on at least two slices. Results of conventional SE images and keyhole dynamic imaging were then compared using a test of symmetry (Bawker test) for statistical analysis.
3. Results Investigations with conventional SE images revealed no evidence of microadenomas in nine patients, poor probability in 12 patients, moderate probability in two patients, and high probability of lesion(s) in one patient. In six patients, lesions were regarded as positive by the readers. Investigations with keyhole dynamic imaging according to an SE protocol revealed no evidence of microadenomas in four patients, poor probability in three patients, moderate probability in four patients, high probability in six patients and proof of lesion presence in 13 patients (Fig. 2). Statistical analysis revealed a trend for higher grades of lesion presence with dynamic keyhole images (P= 0.067).
4. Discussion
Fig. 1. Slight asymmetry of the pituitary gland with minimal volume increase and inhomogeneity of the right endosellar portion.
Our findings reflect the potential of dynamic kSE imaging for the evaluation of microadenomas. Although the results were not statistically significant, we found higher grades of evidence for the presence of lesions and more positive results with kSE imaging. For dynamic imaging of the pituitary gland MR techniques has been found superior to computed tomography (CT) scanning and optimization of the contrast application has been defined by several authors [8,9,11– 13].
T. Rand et al. / European Journal of Radiology 41 (2002) 131–135
133
Fig. 2. (a – d) Dynamic keyhole imaging demonstrates enhancement patterns of the pituitary region with demarcation of a microadenoma on Fig. 2b and c.
Kucharczyk et al. suggested the use of dynamic kSE imaging for the evaluation of pituitary microadenomas, and compared conventional unenhanced, and contrastenhanced dynamic keyhole fast spin-echo (kSE) images. The authors suggest this technique to be a useful supplemental sequence in MR examinations with suspected pituitary microadenomas [14]. Other authors questioned critically the use of increasingly sophisticated methods in pituitary MR imaging [16], particularly for patients where the combination of clinical history and measurements of hormone serum levels allows a reliable diagnosis, but stressed the importance of advanced imaging techniques in patients where the endocrinologic diagnosis is less certain, such as in nonfunctional weakly-secreting tumors or for accurate preoperative localization. Therefore, in our study we used a different approach and investigated a group of patients with equivocal radiological and endocrinological findings in order to analyze the efficacy of dynamic kSE imaging in this particular patient group. The technique of kSE imaging is based on the con-
cept that repetitive samples are obtained from the central 25% (keyhole) of the k-space to acquire a series of rapid, dynamic low spatial-frequency data, with an approximately constant signal throughout the echo train. The high frequency profiles determine the spatial resolution, whereas the contrast is determined by the low frequency profiles [15]. First, a reference sample is evaluated using all profiles. Second, after contrast application only the central profiles are analyzed, whereas the outer regions of the k-space, representing the missing high spatial-frequency data, are additionally interpolated. The advantage of this technique compared with fast GE sequences is a reduction of susceptibility artifacts. The sellar region in particular is affected by susceptibility artifacts since interfaces of tissues with highly different density such as air, bone, and soft tissue create local field inhomogenities [17]. Moreover, rapid SE imaging could be performed without reducing TR [14,18]. Elster et al. estimated that dynamic keyhole MR imaging for the diagnosis of pituitary adenomas provided an additional yield of 9% concerning the detection of lesions
134
T. Rand et al. / European Journal of Radiology 41 (2002) 131–135
Fig. 2. (Continued)
[16]. However, the increased sensitivity must also be weighed against a loss of specificity, as many normal pituitary glands harbor incidental adenomas, which might be unrelated to the clinical findings, but are revealed with advanced MR imaging techniques [19]. Since patients in our study were preselected under the premise of clinically and endocrinologically suspected microadenomas, we minimized the possibility of detecting incidental adenomas by application of kSE. Reading in our study was not completely blinded, as kSE images are obviously different from conventional SE images; however, we tried to minimize the bias by not providing the readers with serum prolactin levels or other clinical information. The relationship between serum prolactin blood levels (HPRL) and pituitary adenomas has been discussed in the literature [20,21]. Serum hormone assay tests are a basic tool in the evaluation of pituitary adenomas and generally allow a confident clinical diagnosis of functioning pituitary adenomas. Tomcak et al. suggested in their study the use of dynamic MR imaging for patients where laboratory results suggest the presence of a microadenoma and conventional MRI is unable to localize a lesion [10].
In a former study we tried to analyze the relationship between HPR levels and the estimated probability of microadenomas. We suggested a serum prolactin level of 100 ng/ml as a useful level for the investigation of the pituitary gland, with an estimated probability of 55% for the diagnosis of a microadenoma [22]. In our current study group, HPR levels were in the borderline range (64.139 57.03; mean9 S.D.). According to our results, dynamic kSE appears to increase the estimated probability for the detection of microadenomas. Consequently, pituitary imaging in this patient collective might achieve greater accuracy with a decrease in the number of equivocal MR findings. However, because of the conservative treatment no surgical proof could be achieved, which was another limitation of our study. Nevertheless, the clinical consequences from an improved detection of prolactin-secreting adenomas are questionable, as therapeutic management of these patients would be the same. Thus, the routine use of kSE imagimg might appear unwarranted; however, no additional contrast application is necessary and only a short
T. Rand et al. / European Journal of Radiology 41 (2002) 131–135
additional investigation time is required. Moreover, although our patient collectively consisted only of patients with prolactin-secreting adenomas, keyhole imaging might be important in patients with Cushing’s disease or acromegaly, where improved lesion detection of microadenomas is mandatory for surgery. However, the proof or exclusion of a tumor as a secondary cause of amenorrhea is necessary for classification according to the WHO criteria [23]. According to the WHO classification of endocrinologically caused amenorrhea, WHO V is defined as hyperprolactinemic amenorrhea with evidence of a tumor, whereas WHO VI is defined as hyperprolactinemic amenorrhea without tumor evidence. However, in normoprolactinemic patients, WHO I and WHO VII are differentiated by the evidence of a tumor. Thus, dynamic kSE imaging with an improved tumor evaluation might also improve consequently the classification of these patients according to the WHO criteria. In conclusion, we suggest that dynamic kSE imaging might be a useful tool even for the routine evaluation of pituitary adenomas. Although dynamic kSE imaging does not add relevant information in patients with clinically and/or radiologically evident prolactin-secreting adenomas, this technique can reveal otherwise invisible microademas, in particular in patients with equivocal clinical and laboratory findings. Because of its potentially short acquisition time and no need for additional contrast application, kSE might prove to be a valuable tool in the routine examination of microadenomas, rather than incurring much higher expense as a second investigation in only preselected cases.
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
Acknowledgements This work was supported by Ludwig Boltzmann Institute for Radiologic Tumor Diagnosis.
References
[17] [18]
[19] [20]
[1] Elster AD. Modern imaging of the pituitary. Radiology 1993;187:1 – 14. [2] Kucharczyk W, Davis DO, Kelly WM, Sze G, Norman D, Newton TH. Pituitary adenomas: high resolution MR imaging at 1,5 T. Radiology 1986;161(1):761 – 5. [3] Kulkarni MV, Lee KF, McArdle CB, Yeakley JW, Haar FL. 1,5-T MR imaging of pituitary microadenomas: technical considerations and CT correlation. Am J Neuroradiol 1987;9:5 – 11. [4] Sakamoto Y, Takahashi M, Korogi Y, Bussaka H, Ushio Y. Normal and abnormal pituitary glands: gadopentetate dimeglumine-enhanced MR imaging. Radiology 1991;178:441 – 5. [5] Miki Y, Matsuo M, Nishizawa S, Kuroda Y, Keyaki A, Makita Y, Kawamura J. Pituitary adenomas and normal pituitary tissue:
[21]
[22]
[23]
135
enhanced patterns on gadopentetate-enhanced MR imaging. Radiology 1990;177:35 – 8. Yuh WT, Fisher DJ, Nguyen HD, Tali ET, Gao F, Simonson TM, Schlechte JA. Sequential MR enhancement pattern in normal piuitary gland and in pituitary adenoma. Am J Neuroradiol 1984;15:101 – 8. Stadnik T, Stevenaert A, Beckers A, Luypaert R, Buisseret T, Osteaux M. Pituitary microadenomas: diagnosis with two and three-dimensional MR imaging at 1,5 T before and after injection of gadolinium. Radiology 1990;176:419 – 28. Stadnik T, Spruyt D, van Binst A, Luypaert R, d’Haens J, Osteaux M. Pituitary microadenomas: diagnosis with dynamic serial CT, conventional CT and T1-weighted MR imaging before and after injection of gadolinium. Eur J Radiol 1994;18:191 –8. Bartynski WS, Lin L. Daynamic and conventional spin-echo MR of pituitary microlesions. Am J Neuroradiol 1997;18:965 – 72. Tomczak R, Merkle E, Fiala S, Rilinger N, Brambs HJ, Rieber A. Ranking of dynamic magnetic resonance imaging in the diagnosis of pituitary microadenomas. Fortschr Ro¨ ntgenstr 1998;168(5):488 – 92. Tabarin T, Laurent F, Catargi B, Olivier-Puel F, Lescene R, Berge J, Galli FS, Drouillard J, Roger P, Guerin J. Comparative evaluation of conventional and dynamic magnetic resonance imaging of the pituitary gland for the diagnosis of Cushing’s disease. Clin Endocrinol 1998;49:293 – 300. Hayashi S, Ito K, Shimada M, Tsubuku M, Kaneko I, Kogure T, Kusama K, Nanba T, Senoh A, Tsauchida S. Dynamic MRI with slow injection of contrast material for the diagnosis of pituitary adenoma. Radiat Med 1995;13(4):167 – 70. Bishop J, Henkelman RM, Plewes DB. Dynamic spin-echo imaging: theoretical assessment and implementation. J Magn Reson Imaging 1994;4(6):843 – 52. Jones RA, Haraldseth O, Mu¨ ller TB, Rinck PA, Oksendal AN. K-space substitution: a novel dynamic imaging technique. Magn Reson Med 1993;29:830 – 4. Kucharczyk W, Bishop JE, Plewes DB, Keller MA, George S. Detection of pituitary microadenomas: comparison of dynamic keyhole fast spin-echo, unenhanced, and conventional contrastenhanced MR imaging. Am J Radiol 1994;163:671 – 9. Elster AD. High-resolution, dynamic pituitary MR imaging: standard of care or academic pastime? Am J Radiol 1994;163:680 – 2. Elster AD. Sellar susceptibility artifacts: theory and implications. Am J Neuroradiol 1992;14:129 – 36. van Vaals J, Brummer M, Dixon WT, Tuithof HH, Engels H, Nelson RC, Gerety BM, Chezmar JI, Den Boer JA. Keyhole method for accelerating imaging of contrast agent uptake. J Magn Reson Imaging 1993;3:671 – 5. Mollit ME, Russel EJ. The pituitary incidentaloma. Ann Intern Med 1990;112:925 – 31. Lundin P, Nyman R, Burman P, Lundberg PO, Muhr C. MRI of pituitary macroadenomas with reference to hormonal activity. Neuroradiology 1992;34:43 – 51. Pojunas KW, Daniels DL, Williams AL, Haughton VM. MR imaging of prolactin-secreting microadenomas. Am J Neuroradiol 1985;7:209 – 13. Rand T, Trattnig S, Kink E, Sator M, Schneider B, Huber J, Imhof H. MRI of microadenomas in patients with hyperprolactinemia. Neuroradiology 1996;38:744 – 6. World Health Organization Scientific Group Report 1976. Agent stimulating gonadal function in the human. The WHO consultation on the diagnosis and treatment of endocrine forms of female infertility, chaired by B. Lunenfeld. WHO Tech Rep Ser, pp. 514.
Evaluation of pituitary microadenomas with dynamic MR imaging Thomas Rand a,*, P. Lippitz b, E. Kink c, H. Huber d, B. Schneider e, H. Imhof a, S. Trattnig a a
Department of Radiology, Uni6ersity of Vienna, Wa¨hringer Gu¨rtel 18 -20, 1090 Vienna, Austria b Department of Radiology, E6ang KH Wien, Vienna, Austria c Department of Medicine, Uni6ersity of Vienna, Vienna, Austria d Department of Gynecology, Uni6ersity of Vienna, Vienna, Austria e Department of Medical Statistics, Uni6ersity of Vienna, Vienna, Austria Received 10 June 1999; received in revised form 30 August 2001; accepted 31 August 2001
Abstract Purpose: To evaluate the use of keyhole dynamic magnetic resonance (MR) imaging in the evaluation of women with borderline hyperprolactinemia. Patients and methods: We investigated 30 patients (mean age 31.6 9 4.3 years) with moderate hyperprolactinemia and clinically suspected microadenomas on T1-weighted spin echo (SE) sequences (repetition time (TR)/echo time (TE) 625/15 ms) with keyhole dynamic imaging and contrast enhancement of the pituitary gland. The probability of lesion presence according to MR criteria was ranked on a five-grade scale and compared with standard examinations of the hypophysis before and after intravenous administration of contrast media. Results: The probability for lesion presence showed a trend toward lower scores on standard SE sequences and higher scores with dynamic imaging (P =0.067). Conclusion: Dynamic keyhole imaging of the pituitary gland could help establish a diagnosis of microadenomas with a higher probability. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: MRI; Keyhole dynamic imaging; Pituitary gland; Microadenomas
1. Introduction MR imaging has provided excellent results in the evaluation of adenomas of the pituitary gland [1–3]. Dynamic magnetic resonance (MR) imaging of the normal pituitary gland and pituitary adenomas has been performed in several studies [4 – 13]. Although the results with regard to enhancement of pituitary adenomas in relation to the anterior lobe are somewhat disparate, a sequential MR enhancement pattern can be observed and microadenomas are best visualized at earlier phases of contrast enhanced dynamic images, with adenomas demonstrating lower signal intensity compared with normal pituitary parenchyma. An opposite enhancing pattern was rare. Yuh et al. [6] found pituitary adenomas to enhance earlier than the lobe and suggested that this might be * Corresponding author. Fax: + 43-1-40400-4898. E-mail address: [email protected] (T. Rand).
due to direct blood supply, similar to that of the posterior pituitary lobe. Technological advances have made available dynamic keyhole imaging with spin echo sequences (kSE) [8,14]. The aim of our study was to evaluate this new technique in the routine management of possible pituitary endocrinopathies, and particularly for the determination of which patients might benefit from pituitary imaging on the kSE basis for further diagnostic accuracy.
2. Patients and methods Our study group included 30 patients (mean age 31.69 4.3 years) with clinically suspected evidence of pituitary adenomas and borderline blood prolactin levels (HPRL) (64.19 57.0 ng/ml; mean9 S.D.; (min, 4 ng/ml; max, 224 ng/ml)). Blood sampling was per-
0720-048X/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 7 2 0 - 0 4 8 X ( 0 1 ) 0 0 4 1 2 - 0
132
T. Rand et al. / European Journal of Radiology 41 (2002) 131–135
formed because of symptoms such as sterility, galactorrhea, amenorrhea or oligomenorrhea. Hormone analysis was performed with enzyme linked immunosorbent assay (ELISA; Boehringer Mannheim). These patients had previously undergone evaluation of the pituitary gland in a routine examination using conventional SE images. Imaging was performed in the sagittal and coronal planes with 3-mm slice thickness, and no interslice gap. The field of view was 23 cm and the acquisition matrix was 256× 256. T1-weighted SE images were acquired with a repetition time (TR) of 500–625 ms and an echo time (TE) of 15 ms, before and after application of 0.2 ml/kg gadolinium DTPA (Magnevist, Schering). Pituitary adenomas were evaluated according to a five-grade scale under the following criteria (Fig. 1). 0, no evidence of microadenomas; I, slight probability: asymmetry of hypophysis morphology in the endosellar region; II, moderate probability: asymmetry and inhomogeneity in the endosellar region; III, high probability: demarcated focal hypointense area after contrast application; and IV, proof of lesion: focal hypointense area on preand postcontrast images associated with increase in volume of the hypophysis. Further evaluation of these patients was performed within 2– 4 weeks after the first investigation using
T1-weighted SE sequences (TR/TE, 100/20 ms) with keyhole dynamic imaging and contrast enhancement according to an SE protocol. Three slices in coronal orientation through the hypophysis in ten dynamic images per slice were acquired at 11-s intervals. All scans had 3-mm slice thickness and no interslice gap. Dynamic imaging was initiated simultaneously with the bolus injection of contrast material (gadopentetate dimeglumine, 0.2 ml/kg). FOV was 180× 180 cm. Probability or proof of the lesions was evaluated by two readers and ranked on a five-grade scale according to the following criteria. 0, No evidence of a lesion; I, poor evidence of lesion presence: inhomogeneity of enhancing hypophyseal tissue; II, moderate evidence of lesion presence: poor demarcation of a hypointense area on a single slice; III, high suspicion of lesion presence: well-demarcated lesions on a single slice; IV, proof of lesion: focal hypointense lesion well-demarcated on at least two slices. Results of conventional SE images and keyhole dynamic imaging were then compared using a test of symmetry (Bawker test) for statistical analysis.
3. Results Investigations with conventional SE images revealed no evidence of microadenomas in nine patients, poor probability in 12 patients, moderate probability in two patients, and high probability of lesion(s) in one patient. In six patients, lesions were regarded as positive by the readers. Investigations with keyhole dynamic imaging according to an SE protocol revealed no evidence of microadenomas in four patients, poor probability in three patients, moderate probability in four patients, high probability in six patients and proof of lesion presence in 13 patients (Fig. 2). Statistical analysis revealed a trend for higher grades of lesion presence with dynamic keyhole images (P= 0.067).
4. Discussion
Fig. 1. Slight asymmetry of the pituitary gland with minimal volume increase and inhomogeneity of the right endosellar portion.
Our findings reflect the potential of dynamic kSE imaging for the evaluation of microadenomas. Although the results were not statistically significant, we found higher grades of evidence for the presence of lesions and more positive results with kSE imaging. For dynamic imaging of the pituitary gland MR techniques has been found superior to computed tomography (CT) scanning and optimization of the contrast application has been defined by several authors [8,9,11– 13].
T. Rand et al. / European Journal of Radiology 41 (2002) 131–135
133
Fig. 2. (a – d) Dynamic keyhole imaging demonstrates enhancement patterns of the pituitary region with demarcation of a microadenoma on Fig. 2b and c.
Kucharczyk et al. suggested the use of dynamic kSE imaging for the evaluation of pituitary microadenomas, and compared conventional unenhanced, and contrastenhanced dynamic keyhole fast spin-echo (kSE) images. The authors suggest this technique to be a useful supplemental sequence in MR examinations with suspected pituitary microadenomas [14]. Other authors questioned critically the use of increasingly sophisticated methods in pituitary MR imaging [16], particularly for patients where the combination of clinical history and measurements of hormone serum levels allows a reliable diagnosis, but stressed the importance of advanced imaging techniques in patients where the endocrinologic diagnosis is less certain, such as in nonfunctional weakly-secreting tumors or for accurate preoperative localization. Therefore, in our study we used a different approach and investigated a group of patients with equivocal radiological and endocrinological findings in order to analyze the efficacy of dynamic kSE imaging in this particular patient group. The technique of kSE imaging is based on the con-
cept that repetitive samples are obtained from the central 25% (keyhole) of the k-space to acquire a series of rapid, dynamic low spatial-frequency data, with an approximately constant signal throughout the echo train. The high frequency profiles determine the spatial resolution, whereas the contrast is determined by the low frequency profiles [15]. First, a reference sample is evaluated using all profiles. Second, after contrast application only the central profiles are analyzed, whereas the outer regions of the k-space, representing the missing high spatial-frequency data, are additionally interpolated. The advantage of this technique compared with fast GE sequences is a reduction of susceptibility artifacts. The sellar region in particular is affected by susceptibility artifacts since interfaces of tissues with highly different density such as air, bone, and soft tissue create local field inhomogenities [17]. Moreover, rapid SE imaging could be performed without reducing TR [14,18]. Elster et al. estimated that dynamic keyhole MR imaging for the diagnosis of pituitary adenomas provided an additional yield of 9% concerning the detection of lesions
134
T. Rand et al. / European Journal of Radiology 41 (2002) 131–135
Fig. 2. (Continued)
[16]. However, the increased sensitivity must also be weighed against a loss of specificity, as many normal pituitary glands harbor incidental adenomas, which might be unrelated to the clinical findings, but are revealed with advanced MR imaging techniques [19]. Since patients in our study were preselected under the premise of clinically and endocrinologically suspected microadenomas, we minimized the possibility of detecting incidental adenomas by application of kSE. Reading in our study was not completely blinded, as kSE images are obviously different from conventional SE images; however, we tried to minimize the bias by not providing the readers with serum prolactin levels or other clinical information. The relationship between serum prolactin blood levels (HPRL) and pituitary adenomas has been discussed in the literature [20,21]. Serum hormone assay tests are a basic tool in the evaluation of pituitary adenomas and generally allow a confident clinical diagnosis of functioning pituitary adenomas. Tomcak et al. suggested in their study the use of dynamic MR imaging for patients where laboratory results suggest the presence of a microadenoma and conventional MRI is unable to localize a lesion [10].
In a former study we tried to analyze the relationship between HPR levels and the estimated probability of microadenomas. We suggested a serum prolactin level of 100 ng/ml as a useful level for the investigation of the pituitary gland, with an estimated probability of 55% for the diagnosis of a microadenoma [22]. In our current study group, HPR levels were in the borderline range (64.139 57.03; mean9 S.D.). According to our results, dynamic kSE appears to increase the estimated probability for the detection of microadenomas. Consequently, pituitary imaging in this patient collective might achieve greater accuracy with a decrease in the number of equivocal MR findings. However, because of the conservative treatment no surgical proof could be achieved, which was another limitation of our study. Nevertheless, the clinical consequences from an improved detection of prolactin-secreting adenomas are questionable, as therapeutic management of these patients would be the same. Thus, the routine use of kSE imagimg might appear unwarranted; however, no additional contrast application is necessary and only a short
T. Rand et al. / European Journal of Radiology 41 (2002) 131–135
additional investigation time is required. Moreover, although our patient collectively consisted only of patients with prolactin-secreting adenomas, keyhole imaging might be important in patients with Cushing’s disease or acromegaly, where improved lesion detection of microadenomas is mandatory for surgery. However, the proof or exclusion of a tumor as a secondary cause of amenorrhea is necessary for classification according to the WHO criteria [23]. According to the WHO classification of endocrinologically caused amenorrhea, WHO V is defined as hyperprolactinemic amenorrhea with evidence of a tumor, whereas WHO VI is defined as hyperprolactinemic amenorrhea without tumor evidence. However, in normoprolactinemic patients, WHO I and WHO VII are differentiated by the evidence of a tumor. Thus, dynamic kSE imaging with an improved tumor evaluation might also improve consequently the classification of these patients according to the WHO criteria. In conclusion, we suggest that dynamic kSE imaging might be a useful tool even for the routine evaluation of pituitary adenomas. Although dynamic kSE imaging does not add relevant information in patients with clinically and/or radiologically evident prolactin-secreting adenomas, this technique can reveal otherwise invisible microademas, in particular in patients with equivocal clinical and laboratory findings. Because of its potentially short acquisition time and no need for additional contrast application, kSE might prove to be a valuable tool in the routine examination of microadenomas, rather than incurring much higher expense as a second investigation in only preselected cases.
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
Acknowledgements This work was supported by Ludwig Boltzmann Institute for Radiologic Tumor Diagnosis.
References
[17] [18]
[19] [20]
[1] Elster AD. Modern imaging of the pituitary. Radiology 1993;187:1 – 14. [2] Kucharczyk W, Davis DO, Kelly WM, Sze G, Norman D, Newton TH. Pituitary adenomas: high resolution MR imaging at 1,5 T. Radiology 1986;161(1):761 – 5. [3] Kulkarni MV, Lee KF, McArdle CB, Yeakley JW, Haar FL. 1,5-T MR imaging of pituitary microadenomas: technical considerations and CT correlation. Am J Neuroradiol 1987;9:5 – 11. [4] Sakamoto Y, Takahashi M, Korogi Y, Bussaka H, Ushio Y. Normal and abnormal pituitary glands: gadopentetate dimeglumine-enhanced MR imaging. Radiology 1991;178:441 – 5. [5] Miki Y, Matsuo M, Nishizawa S, Kuroda Y, Keyaki A, Makita Y, Kawamura J. Pituitary adenomas and normal pituitary tissue:
[21]
[22]
[23]
135
enhanced patterns on gadopentetate-enhanced MR imaging. Radiology 1990;177:35 – 8. Yuh WT, Fisher DJ, Nguyen HD, Tali ET, Gao F, Simonson TM, Schlechte JA. Sequential MR enhancement pattern in normal piuitary gland and in pituitary adenoma. Am J Neuroradiol 1984;15:101 – 8. Stadnik T, Stevenaert A, Beckers A, Luypaert R, Buisseret T, Osteaux M. Pituitary microadenomas: diagnosis with two and three-dimensional MR imaging at 1,5 T before and after injection of gadolinium. Radiology 1990;176:419 – 28. Stadnik T, Spruyt D, van Binst A, Luypaert R, d’Haens J, Osteaux M. Pituitary microadenomas: diagnosis with dynamic serial CT, conventional CT and T1-weighted MR imaging before and after injection of gadolinium. Eur J Radiol 1994;18:191 –8. Bartynski WS, Lin L. Daynamic and conventional spin-echo MR of pituitary microlesions. Am J Neuroradiol 1997;18:965 – 72. Tomczak R, Merkle E, Fiala S, Rilinger N, Brambs HJ, Rieber A. Ranking of dynamic magnetic resonance imaging in the diagnosis of pituitary microadenomas. Fortschr Ro¨ ntgenstr 1998;168(5):488 – 92. Tabarin T, Laurent F, Catargi B, Olivier-Puel F, Lescene R, Berge J, Galli FS, Drouillard J, Roger P, Guerin J. Comparative evaluation of conventional and dynamic magnetic resonance imaging of the pituitary gland for the diagnosis of Cushing’s disease. Clin Endocrinol 1998;49:293 – 300. Hayashi S, Ito K, Shimada M, Tsubuku M, Kaneko I, Kogure T, Kusama K, Nanba T, Senoh A, Tsauchida S. Dynamic MRI with slow injection of contrast material for the diagnosis of pituitary adenoma. Radiat Med 1995;13(4):167 – 70. Bishop J, Henkelman RM, Plewes DB. Dynamic spin-echo imaging: theoretical assessment and implementation. J Magn Reson Imaging 1994;4(6):843 – 52. Jones RA, Haraldseth O, Mu¨ ller TB, Rinck PA, Oksendal AN. K-space substitution: a novel dynamic imaging technique. Magn Reson Med 1993;29:830 – 4. Kucharczyk W, Bishop JE, Plewes DB, Keller MA, George S. Detection of pituitary microadenomas: comparison of dynamic keyhole fast spin-echo, unenhanced, and conventional contrastenhanced MR imaging. Am J Radiol 1994;163:671 – 9. Elster AD. High-resolution, dynamic pituitary MR imaging: standard of care or academic pastime? Am J Radiol 1994;163:680 – 2. Elster AD. Sellar susceptibility artifacts: theory and implications. Am J Neuroradiol 1992;14:129 – 36. van Vaals J, Brummer M, Dixon WT, Tuithof HH, Engels H, Nelson RC, Gerety BM, Chezmar JI, Den Boer JA. Keyhole method for accelerating imaging of contrast agent uptake. J Magn Reson Imaging 1993;3:671 – 5. Mollit ME, Russel EJ. The pituitary incidentaloma. Ann Intern Med 1990;112:925 – 31. Lundin P, Nyman R, Burman P, Lundberg PO, Muhr C. MRI of pituitary macroadenomas with reference to hormonal activity. Neuroradiology 1992;34:43 – 51. Pojunas KW, Daniels DL, Williams AL, Haughton VM. MR imaging of prolactin-secreting microadenomas. Am J Neuroradiol 1985;7:209 – 13. Rand T, Trattnig S, Kink E, Sator M, Schneider B, Huber J, Imhof H. MRI of microadenomas in patients with hyperprolactinemia. Neuroradiology 1996;38:744 – 6. World Health Organization Scientific Group Report 1976. Agent stimulating gonadal function in the human. The WHO consultation on the diagnosis and treatment of endocrine forms of female infertility, chaired by B. Lunenfeld. WHO Tech Rep Ser, pp. 514.