Subarachnoid Hemorrhage is Followed by Pituitary Gland Volume Loss: A Volumetric MRI Observational Study

Subarachnoid hemorrhage (SAH) is a devastating disease associated with high mortality and morbidity. Besides neurological sequelae, neuropsychological deficits largely contribute to patients' long-term quality of life. Little is known about the pituitary gland volume (PGV) after SAH compared to healthy referents and the association of PGV with long-term outcome including cognitive function.

Sixty consecutive non-traumatic SAH patients admitted to the neurological intensive care unit between 2010 and 2014 were enrolled. 3-Tesla magnetic resonance imagining was performed at baseline (16 days) and 12 months after SAH to measure PGV semi-automatically using the software iPlan Net 3.5.0. PGV was compared to age and sex matched healthy referents. The difference between baseline and 1-year-PGV was classified as increase (> 20 mm³ PGV increase), stable (± 20 mm³), or decrease (> 20 mm³ PGV decrease). In addition, total intracerebral volume was calculated. Neuropsychological testing was applied in 43 SAH patients at 1-year follow up encompassing several domains (executive, attention, memory) and self-assessment (questionnaire for self-perceived deficits in attention [German: FEDA]) of distractibility in mental processes, fatigue and decrease in motivation. Multivariable regression with multivariable generalized linear models was used for comparison of PGVs and for subgroup analysis to evaluate a potential association between PGV and neuropsychological outcome.

Patients were 53 years old (IQR = 44-63) and presented with a median Hunt&Hess grade of 2 (IQR = 1-3). SAH patients had a significantly lower PGV both at baseline (360 ± 19 mm³, p < 0.001) and 1 year (367 ± 18 mm³p < 0.001) as compared to matched referents (mean 505 ± 18 mm³). PGV decreased by 75 ± 8 mm³ in 28 patients, increased by 120 ± 22 mm³ in 22 patients and remained stable in 10 patients at 1-year follow-up. PGV in patients with PGV increase at 12 months was not different to healthy referents (p = 0.062). Low baseline PGV was associated with impaired executive functions at 1 year (adjOR = 8.81, 95%-CI = 1.46-53.10, p = 0.018) and PGV decrease within 1 year was associated with self-perceived worse motivation (FEDA; Wald-statistic = 6.6, df = 1, p = 0.010).

Our data indicate significantly lower PGVs following SAH. The association of sustained PGV decrease with impaired neuropsychological long-term outcome warrants further investigations including neuroendocrine hormone measurements.