Reoperation for growth hormone-secreting pituitary adenomas: report on an endonasal endoscopic series with a systematic review and meta-analysis of the literature.

OBJECTIVE Surgery is generally the first-line therapy for acromegaly. For patients with residual or recurrent tumors, several treatment options exist, including repeat surgery, medical therapy, and radiation. Reoperation for recurrent acromegaly has been associated with poor results, with hormonal control usually achieved in fewer than 50% of cases. Extended endonasal endoscopic approaches (EEAs) may potentially improve the results of reoperation for acromegaly by providing increased visibility and maneuverability in parasellar areas. METHODS A database of all patients treated in the authors' center between July 2004 and February 2016 was reviewed. Cases involving patients with acromegaly secondary to growth hormone (GH)-secreting adenomas who underwent EEA were selected for chart review and divided into 2 groups: first-time surgery and reoperation. Disease control was defined by 2010 guidelines. Clinical and radiological characteristics and outcome data were extracted. A systematic review was done through a MEDLINE database search (2000-2016) to identify studies on the surgical treatment of acromegaly. Using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, the included studies were reviewed for surgical approach, tumor size, cavernous sinus invasion, disease control, and complications. Cases were divided into reoperation or first-time surgery for comparative analysis. RESULTS A total of 44 patients from the authors' institution were included in this study. Of these patients, 2 underwent both first-time surgery and reoperation during the study period and were therefore included in both groups. Thus data from 46 surgical cases were analyzed (35 first-time operations and 11 reoperations). The mean length of follow-up was 70 months (range 6-150 months). The mean size of the reoperated tumors was 14.8 ± 10.0 mm (5 micro- and 6 macroadenomas). The patients' mean age at the time of surgery was younger in the reoperation group than in the first-time surgery group (34.3 ± 12.8 years vs 49.1 ± 15.7 years, p = 0.007) and the mean preoperative GH level was also lower (7.7 ± 13.1 μg/L vs 25.6 ± 36.8 μg/L, p = 0.04). There was no statistically significant difference in disease control rates between the reoperation (7 [63.6%] of 11) and first-time surgery (25 [71.4%] of 33) groups (p = 0.71). Univariate analysis showed that older age, smaller tumor size, lower preoperative GH level, lower preoperative IGF-I level, and absence of cavernous sinus invasion were associated with higher chances of disease control in the first-time surgery group, whereas only absence of cavernous sinus invasion was associated with disease control in the reoperation group (p = 0.01). There was 1 case (9%) of transient diabetes insipidus and hypogonadism and 1 (9%) postoperative nasal infection after reoperation. The systematic review retrieved 29 papers with 161 reoperation and 2189 first-time surgery cases. Overall disease control for reoperation was 46.8% (95% CI 20%-74%) versus 56.4% (95% CI 49%-63%) for first-time operation. Reoperation and first-time surgery had similar control rates for microadenomas (73.6% [95% CI 32%-98%] vs 77.6% [95% CI 68%-85%]); however, reoperation was associated with substantially lower control rates for macroadenomas (27.5% [95% CI 5%-57%] vs 54.3% [95% CI 45%-62%]) and tumors invading the cavernous sinus (14.7% [95% CI 4%-29%] vs 38.5% [95% CI 27%-50%]). CONCLUSIONS Reoperative EEA for acromegaly had results similar to those for first-time surgery and rates of control for macroadenomas that were better than historical rates. Cavernous sinus invasion continues to be a negative prognostic indicator for disease control; however, results with EEA show improvement compared with results reported in the prior literature.