A Retrospective Study Of Factors Related To Surgical Outcome Of Growth Hormone-secreting Pituitary Adenomas

Objective:This study systematically reviewed the relationship between clinical features and surgical outcome,aiming to provide experience for surgical treatment and postoperative monitoring of growth hormone-secreting pituitary adenoma.Methods:A total of 54 patients with growth hormone-secreting pituitary adenoma were systematically reviewed who underwent surgical resection in the Department of Neurosurgery,Qilu Hospital,Shandong University from June 2016 to June 2021.The samples were divided into "total resection" and "partial resection" groups according to surgical outcome.In addition to the analysis of general information(age,gender,BMI),univariate and multivariate analysis of the relationship between the four key factors and surgical outcome was focused.These four key factors include hormone levels,image features based on magnetic resonance imaging,postoperative pituitary function,and pathological features.To find statistically significant and stable indicators for evaluating the outcome of surgery,we constructed the receiver operating characteristic curve to find the best cut-off value.Finally,the evaluation effect of the evaluation model of surgical outcome constructed by different indicators was compared.Results:Univariate analysis showed that baseline growth hormone ratio was a risk factor for total resection of growth hormone secreting pituitary adenoma,and it was easier to achieve total resection of growth hormone adenomas when baseline growth hormone ratio was less than 15.27(P<0.05).No significant association was found between baseline insulin-like growth factor and postoperative hormone remission rate(P>0.05).Among image features based on magnetic resonance imaging,univariate analysis found that total resection was more likely to be achieved when the maximum tumor diameter was less than 2.15cm or the tumor volume was less than 3.72cm3(P<0.05).The total resection rate of adenomas with lower Knosp grade(0-2)was significantly higher than that in patients with higher Knosp grade(3-4),with total resection rate of 100%and 66.67%,respectively,and the difference was statistically significant(P<0.001).There was no significant correlation between normalized T2-weighted signal ratio,postoperative thyroid function,postoperative urine volume and surgical outcome(P>0.05).However,intraoperative partial excision was more likely to result in early postoperative cortisol reduction(below the lower limit of the reference range)(P<0.05).Univariate Logistic regression analysis and continuously corrected Chi-square test both supported that Ki-67 proliferation index was an risk factor for total resection of GH-secreting adenoma.The percentage of total resection in patients with Ki-67 proliferation index greater than 3%was 69.23%,which was lower than 97.56%in patients with Ki-67 proliferative index less than 3%.(P<0.05).However,the expression of P53 could not well evaluate the surgical outcome(P>0.05).The results of multivariate analysis showed that only baseline growth hormone and Ki-67 proliferation index were stable and reliable indexes for evaluating surgical outcome.After comparing the models that using baseline growth hormone ratio or Ki-67 proliferation index alone and the combination of these two indicators to evaluate surgical outcome,we found no statistical difference in the area under the receiver operating characteristic curves corresponding to each model(P>0.05),but all the three models have higher evaluation efficiency.Conclusion:(1)Baseline growth hormone ratio and Ki-67 proliferation index are the most stable and reliable indexes to evaluate the surgical outcome of growth hormone-secreting adenoma.(2)when the baseline growth hormone ratio was less than 15.27,adenoma was more likely to achieve total resection.(3)Ki-67 proliferation index≥3%indicated that partial resection was more likely.(4)Both the baseline growth hormone ratio,Ki-67 proliferation index and the combined application of these two indexes showed high efficacy in evaluating the outcome of surgery,and there was no statistical difference between the evaluation effect of each model.