Long-term outcomes of transsphenoidal surgery for management of growth hormone-secreting adenomas: single-center results

OBJECTIVE Transsphenoidal surgery is advocated as the first-line management of growth hormone (GH)-secreting adenomas. Although disease control is defined by strict criteria for biochemical remission, the length of follow-up needed is not well defined in literature. In this report, the authors present their long-term remission rate and identify various predictive factors that might influence the clinical outcome. METHODS The authors conducted a single-institute retrospective analysis of all transsphenoidal procedures for GH-secreting adenomas performed from January 2000 to June 2016. The primary outcome was defined as biochemical remission according to the 2010 consensus criteria and measured at the 1-year postoperative mark as well as on the last recorded follow-up appointment. Secondary variables included recurrence rate, patterns of clinical presentation, and outcome of adjuvant therapy (including repeat surgery). Subgroup analysis was performed for patients who had biochemical or radiological "discordance"-patients who achieved biochemical remission but with incongruent insulin-like growth factor 1 (IGF-1)/GH or residual tumor on MRI. Recurrence-free survival analysis was conducted for patients who achieved remission at 1 year after surgery. RESULTS Eighty-one patients (45 female and 36 male) with confirmed acromegaly treated with transsphenoidal surgery were included. In 62 cases the patients were treated with a pure endoscopic approach and in 19 cases an endoscopically assisted microscopic approach was used. Primary biochemical remission after surgery was achieved in 59 cases (73%) at 1 year after surgery. However, only 41 patients (51%) remained in primary surgical remission (without any adjuvant treatment) at their last follow-up appointment, indicating a recurrence rate of 31% (18 of 59 patients) over the duration of follow-up (mean 100 +/- 61 months). Long-term remission rates for pure endoscopic and endoscopically assisted cases were not significantly different (48% vs 52%, p = 0.6). Similarly, no significant difference in long-term remission was detected between primary surgery and repeat surgery (54% vs 33%, p = 0.22). Long-term remission was significantly influenced by extent of resection, cavernous sinus invasion (radiologically as well as surgically reported), and preoperative and early postoperative GH and IGF-1 levels (within 24-48 hours after surgery) as well as by clinical grade, with lower remission rates in patients with dysmorphic features and/or medical comorbidities (grade 2-3) compared to minimally symptomatic or silent cases (grade 1). CONCLUSIONS The long-term surgical remission rate appears to be significantly less than "early" remission rates and is highly dependent on the extent of tumor resection. The authors advocate a long-term follow-up regimen and propose a clinical grading system that may aid in predicting long-term outcome in addition to the previously reported anatomical factors. The role of repeat surgery is highlighted.