| Part ⅠAnalysis of the diagnosis and etiological diagnosis of Cushing’s SyndromeObjective:To explore the diagnosis value and optimal cut-off points in a variety of qualitative and etiological diagnosis process of Cushing’s syndrome (CS), develop and validate diagnostic criteria applied to our hospital, and promote the establishment of suitable diagnostic criteria for Chinese CS patients.Methods:Patients who visited the Department of Endocrinology of Huashan Hospital from January 2011 to November 2013 with suspected Cushing’s syndrome were analyzed. All patients underwent revaluation procedure according to updated guidelines and obtained confirmed diagnosis eventually. The diagnosis of Cushing’s disease was confirmed by histopathology examination or on the basis of long-term remission after surgical or radiation therapy despite the absence of histopathology evidence. Ectopic ACTH Syndrome and adrenal Cushing’s syndrome were all diagnosed by histopathology results. Patients excluded from CS after a series of screening tests and follow-up were analyzed as non-Cushing’s syndrome patients. Data were collected retrospectively and used for statistical analysis.Results:A total of 90 CS patients and 42 non-CS patients were included, of 71 CD patients,16 adrenal Cushing’s syndrome patients and 3 EAS patients. ROC analysis showed that the optimal threshold for 24h UFC was 116.27μg/24h(sensitivity 95.6%,specificity 90.9%,AUC 0.986) in the diagnosis of CS, whereas 129.13μg/24h,the upper limit of 24hUFC Normal range in our hospital, yielded sensitivity 92.1% and specificity 90.9%. the optimal threshold of the ratio of Midnight serum cortisol(F0am) to Morning serum cortisol(F8am) was 55.6%(sensitivity 97.8%,specificity 84.2%,AUC 0.943).The optimal threshold of F0am was 10.11μg/dl(sensitivity 93%,specificity 100%,AUC 0.994),while the cutoff value of 5μg/dl showed sensitivity of 100%,specificity of 79.5% and 7.5μg/dl showed sensitivity of 97.8%,specificity of 87.2%. The optimal threshold of serum cortisol after lmg Dexamethasone suppression test (1mgDST) in the diagnosis of CS was 4.185μg/dl (sensitivity 98.9%, specificity 100%, AUC 1.000), however, other cutoff values like 1.8μg/dl (sensitivity 100%, specificity 90.3%) and 5μg/dl (sensitivity 95.6%, specificity 100%) yielded. The optimal threshold of serum cortisol after lmg DST in the differential diagnosis of CD and EAS was 39.86μg/dl(sensitivity 100.0%,specificity 94.4%,AUC 0.981).Meanwhile, combination of any 2 or 3 of these screening tests achieved an excellent sensitivity of 100.0% in diagnosis of CS.The optimal cut-off point of 8am morning plasm ACTH(ACTHs) was 17.8 pg/ml(sensitivity 100.0%,specificity 100.0%,AUC 1.000)in diagnosing ACTH dependent CS and 179pg/ml (sensitivity 100.0%,specificity92.3%,AUC 0.923) in differential diagnosis of CD and EAS. The optimal threshold of 24hUFC Suppression ratio in standard HDDST was 48.46%(sensitivity 92.6%,specificity100%,AUC 0.956),in accordance with diagnostic performance of cut-off point 50%.The best threshold for ACTH ratio(CEN:PER) in BIPSS for differential diagnosis of CD and EAS at baseline and after DDAVP infusion were 1.26(sensitivity 96%,specificity100%,AUC 0.980) and 1.42(sensitivity 100%,specificity 100%,AUC 1.000),respectively. Using the cut-off value of 2, sensitivity was increased from 78.6% at baseline to 85.7% 5min after infusion. DDAVP also improved the diagnostic accuracy of CSS (from 78.9% at baseline to 89.5% after stimulation). In 19 CD patients who underwent CSS, accuracy of CSS through micro catheter was superior to IPSS at baseline (78.9% and 73.7%, respectively).However, of 28 CD and 2 EAS patients who underwent BIPSS, accuracy at baseline was better than that of 19 CD patients who underwent CSS (80.0% and 78.9%, respectively), while inferior to CSS after DDAVP stimulation (86.7% and 89.5%, respectively).Conclusion:24UFC, F0am, 1mgDST are all ideal approaches to the diagnosis of CS, of which 1mgDST achieves highest sensitivity and specificity, F0am shows high sensitivity and specificity, while 24hUFC yields an ordinary specificity. The ratio of F0am to F8am can also be an indicator in diagnosis of CS. Combination of multiple screening tests can improve sensitivity significantly. Results of CS screening tests can also be predictive of etiological diagnosis,especially lmg DST showed high diagnostic value in differentiate CD from EAS.ACTH,HDDST,BIPSS or CSS with DDAVP all appear to be ideal methods of etiological diagnosis.ACTHg achieves fair diagnostic value in primary identification of ACTH dependent and independent CS and in differential diagnosis of CD and EAS. Given its poor sensitivity yet not so satisfactory specificity, HDDST is more suitable as a confirmatory test for the diagnosis of CD.BIPSS combined with DDAVP achieves the most extraordinary diagnostic performance, DDAVP further enhances the diagnostic accuracy of BIPSS and CSS, yet high false negative rate occurs when baseline or peak ACTH ratio(CEN:PER)>2 are used as diagnostic criteria of CD. There’s no evidence suggesting significant superiority of CSS to BIPSS. Each test has its limitation and overclaim for diagnostic sensitivity will in turn reduce specificity inevitably. Decision of the most appropriate diagnostic test and optimal cutoff value must be made based on integrated consideration of clinical situation.Part IISerum cortisol curve of healthy Chinese adultsObjective:To investigate circadian rhythm of cortisol in health Chinese adults, and try to offer reference value and guidance to the diagnosis and treatment of primary and secondary adrenal disease.Methods:Healthy Chinese subjects who participated in the phase I clinical drug research "pharmacokinetics research of sitafloxacin in healthy Chinese subjects" at antibiotic research institute of Huashan Hospital in July 2013 and meet the inclusion and exclusion criteria were analyzed. Blood were drawn off for serum cortisol measurement at 7:30.8:00.8:30.9:00、9:30、10:30、11:30、13:30、15:30、19:30 on the first day and 7:30,19:30 of the second day and 7:30 of the third day. Data was collected and used for statistical analysis.Results:A total of 36 subjects were included, of 18 males and 18 females. The mean age for all patients was 22.47±13.83 years. The mean BMI for all patients was 21.87±1.36 Kg/m2. Our result shows that serum cortisol at 7:30 was significantly higher than that at other time points. Cortisol levels decreased slowly, yet during the 4 phases of 8:00-9:00,8:30-9:30,9:30-13:30 and 15:30-19:30 respectively, cortisol levels hardly changed. No meal-induced cortisol stimulation was observed. The maximum serum cortisol at 7:30am of all subjects was more than 13.0μg/dl, while serum cortisol at 8:00am in 20 (55.6%) subjects was less than 13.0μg/dl. The ratio (FRatio) of serum cortisol at 15:30 to morning serum cortisol at 7:30 of the subjects ranged from 0.27 to 0.83,26 of which (72.2%) were less than 0.50. The ratio (FRatio) of serum cortisol at 17:30 to morning serum cortisol at 7:30 of the subjects ranged from 0.17 to 0.85,29 of which (80.6%) were less than 0.50.Conclusion:Of the healthy adults in our research, serum cortisol peaked at 7:30am and decreased by phases. Serum cortisol at 8:00am is not as ideal as that at 7:30am in predicting adrenal function. It seems that serum cortisol at 15:30-17:30 less than 50% of peak cortisol of the day is not an accurate criterion to identify the normal circadian rhythm of cortisol. These results provide reference value and guidance for the diagnosis and treatment of adrenal insufficiency.Part ⅢAssessment of Hypothalmus-Pituitary-Adrenal Axis InsufficiencyObjective:To investigate a variety of tests in diagnosing HPA axis insufficiency in terms of diagnostic value, cut-off point selection, conditions of use. And try to contribute to its early and accurate diagnosis as well as timely treatment.Methods:Patients who visited the Department of Endocrinology of Huashan Hospital from April 2011 to July 2013 with suspected adrenal insufficiency were analyzed. All patients underwent dynamic tests like Insulin Tolerance Test or Low Dose Corticotrophin Test and obtained confirmed diagnosis eventually. Data of baseline examination and dynamic tests was collected retrospectively and used for statistical analysis.Results:A total of 62 patients who underwent ITT and 22 patients who underwent Low Dose Corticotrophin Test were included. There were 17 patients who didn’t achieve adequate hypoglycemia eventually, yet 10 of them showed normal serum cortisol response(more than 18μg/dl),meanwhile, among those 45 patients with sufficient hypoglycemia,35 patients had normal response.13 of 22 patients who underwent low dose corticotrophin test had peak serum cortisol more than 18μg/dl. The optimal cutoff value for basal serum cortisol in predicting adrenal function is 12.15μg/dl(sensitivity 52.1%, specificity95.2%, AUC 0.767) when ITT results were taken as gold standard. Other useful cutoff values were 12.81μg/dl (sensitivity 46.5%, specificity 100%) and 0.99μg/dl (sensitivity 100%, specificity 4.8%). There was a weak positive correlation between nadir blood glucose and age, basal ACTH in ITT(R=0.248 and 0.320, respectively).Most (95.2%) nadir blood glucose occurred at 30min, while most (97.1%) serum cortisol peaked at 90min. Bland-Altaian analysis showed that 94.9% of venous blood glucose and capillary blood glucose measured in our department simultaneously were in the 95% limits of agreement, however, mean difference(d)and standard deviation(Sd)of capillary blood glucose minus venous blood glucose is 0.289 and 0.586mmol/L, respectively. Serum cortisol usually (90.9%) peaked at 30min in Low Dose Corticotrophin Test.Conclusion:Basal serum cortisol can be used as first-line test in the assessment of HPA axis. While performing ITT, relevant factors such as age and basal ACTH should be taken into account to apply the appropriate insulin dose for successful induced hypoglycemia. If blood glucose fails to drop to 2.2mmol/L at 30min-45min yet, additional insulin dose can be applied. Shortened duration of ITT will increase the risk of lower specificity and need to be carefully considered. Capillary blood glucose can be used to predict venous blood glucose; however, bias exited that capillary blood glucose tended to be higher than venous blood glucose. Low Dose Corticotrophin Test is a promising alternative to ITT.Further research is needed on its diagnostic consistency with ITT. |
PDF Full Text Request