OBJECTIVE: To quantify the extent of hypothalamic damage after surgery for craniopharyngioma using magnetic resonance imaging (MRI) and to relate the findings to changes in body mass index (BMI). PATIENTS: Sixty-three survivors (36 males, 27 females) of childhood cramopharyngioma were treated surgically between 1973 and early 1994. METHODS: Cranial MRI was performed at a structured follow-up assessment 1.5-19.2 yr after the initial surgery. Hypothalamic damage was scored as 0 (no visible damage), 1 (intermediate), or 2 (severe). RESULTS: After surgery there was an increase in BMI standard deviation (SD) from diagnosis to study assessment in all but 7 patients. However, patients with MRI scores of 2 (n = 17) had a significantly greater increase in median BMI SD score at follow-up (+5.5 SD score), compared with +2.5 SD score and +1.1 SD score for patients with MRI scores of 1 or 0, respectively. Of the 17 cases with MRI scores of 2, 10 had a history of extreme weight loss or weight gain at presentation; preoperative neuroimaging demonstrated extensive hypothalamic infiltration by tumor in these cases. CONCLUSION: MRI gives sufficient anatomical definition to allow assessment of the extent of hypothalamic damage and, thereby, prediction of the patients most at risk for severe post-operative weight gain.
OBJECTIVE: To quantify the extent of hypothalamic damage after surgery for craniopharyngioma using magnetic resonance imaging (MRI) and to relate the findings to changes in body mass index (BMI). PATIENTS: Sixty-three survivors (36 males, 27 females) of childhood cramopharyngioma were treated surgically between 1973 and early 1994. METHODS: Cranial MRI was performed at a structured follow-up assessment 1.5-19.2 yr after the initial surgery. Hypothalamic damage was scored as 0 (no visible damage), 1 (intermediate), or 2 (severe). RESULTS: After surgery there was an increase in BMI standard deviation (SD) from diagnosis to study assessment in all but 7 patients. However, patients with MRI scores of 2 (n = 17) had a significantly greater increase in median BMI SD score at follow-up (+5.5 SD score), compared with +2.5 SD score and +1.1 SD score for patients with MRI scores of 1 or 0, respectively. Of the 17 cases with MRI scores of 2, 10 had a history of extreme weight loss or weight gain at presentation; preoperative neuroimaging demonstrated extensive hypothalamic infiltration by tumor in these cases. CONCLUSION: MRI gives sufficient anatomical definition to allow assessment of the extent of hypothalamic damage and, thereby, prediction of the patients most at risk for severe post-operative weight gain.