Zu-Fang Wang1, Gui-Jun Li. 1. Clinical Laboratory, Jiaxing Maternal and Child Health Hospital, Jiaxing, Zhejiang 314051, China.
Abstract
OBJECTIVE: To study the value of follicle stimulating hormone (FSH), luteinizing hormone (LH) and LH/FSH ratio in the diagnosis of precocious puberty in girls by ROC curve analysis. METHODS: Gonadotropin-releasing hormone (GnRH) stimulation test was performed on 220 girls with pseudo-sexual precocity and 61 girls with true sexual precocity. Blood LH and FSH levels were measured before and after 30 and 60 minutes of taking the GnRH test. The ratio of LH to FSH was calculated. Sensitivity and best point for the diagnosis of precocity according to LH, FSH and LH/FSH ratio were analyzed by ROC curve analysis. RESULTS: The area under the ROC curve was 0.90 and 0.95 according to LH level and LH/FSH ratio respectively for the diagnosis of precocity. The best point for diagnosis by LH was 10.15 IU/L, with a sensitivity of 0.92 and specificity of 0.89. The best point for diagnosis by LH/FSH ratio was 0.60, with a missed diagnosis rate of 6.0% and specificity of 0.91. When true sexual precocity was diagnosed based on one index between LH>10.15 IU/L and LH/FSH ratio>0.60, sensitivity was 0.97 and specificity was 0.94. When the diagnosis of true sexual precocity was diagnosed based on both LH>10.15 IU/L and LH/FSH>0.60, sensitivity was 0.85 and specificity was 1.00. CONCLUSIONS: True sexual precocity can be diagnosed when both LH>10.15 IU/L and LH/FSH ratio>0.60. Only one of the two indexes for the diagnosis of true sexual precocity is presented, further observation is necessary to decrease missed diagnosis and misdiagnosis.
OBJECTIVE: To study the value of follicle stimulating hormone (FSH), luteinizing hormone (LH) and LH/FSH ratio in the diagnosis of precocious puberty in girls by ROC curve analysis. METHODS:Gonadotropin-releasing hormone (GnRH) stimulation test was performed on 220 girls with pseudo-sexual precocity and 61 girls with true sexual precocity. Blood LH and FSH levels were measured before and after 30 and 60 minutes of taking the GnRH test. The ratio of LH to FSH was calculated. Sensitivity and best point for the diagnosis of precocity according to LH, FSH and LH/FSH ratio were analyzed by ROC curve analysis. RESULTS: The area under the ROC curve was 0.90 and 0.95 according to LH level and LH/FSH ratio respectively for the diagnosis of precocity. The best point for diagnosis by LH was 10.15 IU/L, with a sensitivity of 0.92 and specificity of 0.89. The best point for diagnosis by LH/FSH ratio was 0.60, with a missed diagnosis rate of 6.0% and specificity of 0.91. When true sexual precocity was diagnosed based on one index between LH>10.15 IU/L and LH/FSH ratio>0.60, sensitivity was 0.97 and specificity was 0.94. When the diagnosis of true sexual precocity was diagnosed based on both LH>10.15 IU/L and LH/FSH>0.60, sensitivity was 0.85 and specificity was 1.00. CONCLUSIONS: True sexual precocity can be diagnosed when both LH>10.15 IU/L and LH/FSH ratio>0.60. Only one of the two indexes for the diagnosis of true sexual precocity is presented, further observation is necessary to decrease missed diagnosis and misdiagnosis.