Yasuhiko Hayashi1, Daisuke Kita2, Takuya Watanabe2, Issei Fukui2, Yasuo Sasagawa2, Masahiro Oishi2, Osamu Tachibana3, Fumiaki Ueda4, Mitsutoshi Nakada2. 1. Department of Neurosurgery, Graduate School of Medical Science, Kanazawa University, Kanazawa, 13-1 Takara-machi, Kanazawa, Ishikawa, 920-8641, Japan. yahayashi@med.kanazawa-u.ac.jp. 2. Department of Neurosurgery, Graduate School of Medical Science, Kanazawa University, Kanazawa, 13-1 Takara-machi, Kanazawa, Ishikawa, 920-8641, Japan. 3. Department of Neurosurgery, Kanazawa Medical University, Kanazawa, Japan. 4. Department of Radiology, Graduate School of Medical Science, Kanazawa University, Kanazawa, Japan.
Abstract
PURPOSE: Diabetes insipidus (DI) remains a complication of transsphenoidal surgery (TSS) for sellar and parasellar tumors. Antidiuretic hormone (ADH) appears as hyper intensity (HI) in the pituitary stalk and the posterior lobe of the pituitary gland on T1-weighted magnetic resonance (MR) imaging. Its disappearance from the posterior lobe occurs with DI, indicating a lack of ADH. The appearance of HI in the pituitary stalk indicates disturbances in ADH transport. METHODS: This retrospective study included 172 patients undergoing TSS for sellar tumors at our institute from 2006 to 2014. Sequential T1-weighted MR images without enhancement were evaluated for HI in the pituitary stalk and the posterior lobe to assess the localization of ADH before and at intervals after TSS. DI was assessed pre- and postoperatively. HI in the pituitary stalk showed the following morphology: (1) ovoid in the distal end of the pituitary stalk (group A), (2) linear in the distal part of the pituitary stalk (group B), (3) linear in the whole pituitary stalk (group C). RESULTS: Preoperative DI occurred in 6 patients (3.5 %) with no HI observed in the posterior lobe. Postoperative DI was transient in 82 patients (47.7 %), and permanent in 11 (6.4 %). One week after surgery, HI was absent in the posterior lobe in 74 patients (43.0 %), and present in the pituitary stalk in 99 patients (57.6 %); both were significantly correlated with postoperative DI (p < 0.001). The absence of HI in the posterior lobe (A, 48.9 %; B, 68.3 %; C, 92.3 %), persistence of DI (A, 3.7 days; B, 45.9 days; C, 20.5 months), and duration until HI recovery in the posterior lobe (A, 3.6 months; B, 6.8 months; C, 22.9 months) were greatest in group C, followed by group B, and then group A. Fourteen group A patients did not have postoperative DI despite having HI in the pituitary stalk and the posterior lobe. Four group C patients developed permanent DI with persistence HI in the pituitary stalk. CONCLUSION: HI in the pituitary stalk and its absence in the posterior lobe indicated postoperative DI, which was transient if HI was detected in the pituitary stalk. DI duration could be predicted by the length of HI in the pituitary stalk, which corresponded to the degree of ADH transport obstruction.
PURPOSE:Diabetes insipidus (DI) remains a complication of transsphenoidal surgery (TSS) for sellar and parasellar tumors. Antidiuretic hormone (ADH) appears as hyper intensity (HI) in the pituitary stalk and the posterior lobe of the pituitary gland on T1-weighted magnetic resonance (MR) imaging. Its disappearance from the posterior lobe occurs with DI, indicating a lack of ADH. The appearance of HI in the pituitary stalk indicates disturbances in ADH transport. METHODS: This retrospective study included 172 patients undergoing TSS for sellar tumors at our institute from 2006 to 2014. Sequential T1-weighted MR images without enhancement were evaluated for HI in the pituitary stalk and the posterior lobe to assess the localization of ADH before and at intervals after TSS. DI was assessed pre- and postoperatively. HI in the pituitary stalk showed the following morphology: (1) ovoid in the distal end of the pituitary stalk (group A), (2) linear in the distal part of the pituitary stalk (group B), (3) linear in the whole pituitary stalk (group C). RESULTS: Preoperative DI occurred in 6 patients (3.5 %) with no HI observed in the posterior lobe. Postoperative DI was transient in 82 patients (47.7 %), and permanent in 11 (6.4 %). One week after surgery, HI was absent in the posterior lobe in 74 patients (43.0 %), and present in the pituitary stalk in 99 patients (57.6 %); both were significantly correlated with postoperative DI (p < 0.001). The absence of HI in the posterior lobe (A, 48.9 %; B, 68.3 %; C, 92.3 %), persistence of DI (A, 3.7 days; B, 45.9 days; C, 20.5 months), and duration until HI recovery in the posterior lobe (A, 3.6 months; B, 6.8 months; C, 22.9 months) were greatest in group C, followed by group B, and then group A. Fourteen group A patients did not have postoperative DI despite having HI in the pituitary stalk and the posterior lobe. Four group C patients developed permanent DI with persistence HI in the pituitary stalk. CONCLUSION:HI in the pituitary stalk and its absence in the posterior lobe indicated postoperative DI, which was transient if HI was detected in the pituitary stalk. DI duration could be predicted by the length of HI in the pituitary stalk, which corresponded to the degree of ADH transport obstruction.
Entities:
Keywords:
Diabetes insipidus; Magnetic resonance imaging; Pituitary stalk; Posterior lobe; Transsphenoidal surgery
Authors: I Fujisawa; R Asato; M Kawata; Y Sano; K Nakao; T Yamada; H Imura; Y Naito; K Hoshino; S Noma Journal: J Comput Assist Tomogr Date: 1989 May-Jun Impact factor: 1.826
Authors: I Fujisawa; K Kikuchi; K Nishimura; K Togashi; K Itoh; S Noma; S Minami; T Sagoh; T Hiraoka; T Momoi Journal: Radiology Date: 1987-11 Impact factor: 11.105