Letter to the Editor: “Why We Should Still Treat by Neurosurgery Patients With Cushing Disease and a Normal or Inconclusive Pituitary MRI”

We read with interest the study by Cristante et al. (1) on the outcome of transsphenoidal surgery in patients with Cushing disease (CD) with normal or inconclusive pituitary MRI. For years there has been a controversy on whether surgery should be performed in such cases (2). In this monocentric retrospective study based on 195 patients who underwent microscopic or endoscopic surgery in the last 25 years, no difference in remission rates was found between patients with positive (89 microadenomas, 18 macroadenomas) and negative pituitary MRI (44 inconclusive and 44 normal MRI). The authors thus suggest that patients with CD should undergo surgery regardless of MRI status. We would like to tone down these conclusions with the following points.

The lack of significant difference in remission between positive and negative MRI groups might be biased by the repartition into four MRI groups (microadenoma, macroadenoma, inconclusive, and normal). When only patients with obvious (107 patients) or doubtful (88 patients) MRI are considered, postsurgical remission rates seem rather different (89% vs 74%, respectively). Of note, 11 patients had to undergo repeat surgery during the first month because of a lack of corticotroph deficiency, including two who had a total hypophysectomy; 11 patients with a microadenoma on MRI finally had a partial hypophysectomy; and 50% of patients with negative MRI had partial or total hypophysectomy, compared with 14% with microadenoma. Precise results on endocrine side effects are not provided. Although no difference was found in terms of postsurgical endocrine side effects between positive and negative MRI groups, 30% of data on endocrine outcomes were actually missing. Surgery on negative MRI is likely to lead to an increased rate of hypopituitarism and diabetes insipidus (2), a point that must be taken into account in the management of such patients.

Indeed, a recent review suggested the possibility of treating with anticortisolic drugs patients with doubtful MRI for whom access to an expert center was difficult (3). We defended this short- to mid-term wait-and-see attitude when reporting the efficacy of ketoconazole in the management of CD (4). Although we consider that transsphenoidal surgery is the only first-line treatment aimed at curing the patient, we reported that a third of patients with initially negative or inconclusive MRI would actually present with a visible adenoma on MRI in the first 3 to 5 years of treatment. Similar results were suggested with the use of mitotane (5). This would undoubtedly make surgical management easier and decrease the rate of endocrine side effects.

Finally, the authors interestingly show the improvements in MRI over the last 25 years. However, 27% of patients still present a doubtful MRI during their diagnostic procedure. As recently shown, a way to improve these results might be to add to the diagnostic algorithm 11C-methionine positron emission tomography imaging, a promising procedure that could help locate the adenoma in such patients (6).

This should lead multidisciplinary expert teams to systematically discuss the best therapeutic option between surgery, additional imaging, or transient medical treatment in patients with doubtful MRI.

Additional Information

Disclosure Summary: F.C. and T.B. received research grants and speaker honoraria from Novartis and HRA Pharma. The remaining authors have nothing to disclose.

References and Notes