ABSTRACT
Objective: To determine the positive predictive value of magnetic resonance spectroscopy in diagnosing malignant thyroid nodules by taking histopathology as a gold standard
Methodology: This descriptive cross-sectional study was undertaken at the department of Radiology, CMH Multan from October 2014 to March 2015. 77 patients with malignant thyroid nodules on ultrasonography between ages 30-70 years, of either gender were included. Patients with previous thyroid surgery, already biopsy proven malignant thyroid nodules and, those having contraindication to magnetic resonance spectroscopy [MRS] were excluded. All the patients then underwent MRS for choline peak and choline /creatine [Cho/Cr] ratio. Findings were correlated with histopathology
Results: Mean age of the patients was 46.53 +/- 9.15 years. Out of these 77 patients, 62 [80.52%] were female and 15 [19.48%] were males with female to male ratio of 4:1. MRS supported the diagnosis of malignant thyroid nodules in 60 patients. Histopathology confirmed malignant thyroid nodules in 49 [true positive] cases where as 11 [False Positive] had no malignant lesion on histopathology. Positive predictive value of magnetic resonance spectroscopy [MRS] in diagnosing malignant thyroid nodules was 81.67%
Conclusion: Magnetic resonance spectroscopy [MRS] is a non-invasive modality of choice with high positive predictive value in diagnosing malignant thyroid nodules. It has not only dramatically improved our ability of diagnosing thyroid lesions pre- operatively but also helps the surgeons for proper decision making
ABSTRACT
Objective: To determine the diagnostic accuracy of ratio of platelet count to spleen size for prediction of esophageal varices in patients of liver cirrhosis, keeping upper GI endoscopy as gold standard
Materials and Methods: This cross-sectional validation study was carried out in Radiology Department, Combined Military Hospital, Peshawar from February, 2015 to August, 2015.One hundred and fifty patients of either sex, having liver cirrhosis with no episode of gastrointestinal bleeding, scheduled to undergo upper GI endoscopy were selected.Ultrasound abdomen of these patients was carried out and spleen size was determined in millimeters. Platelet count if already not performed was also carried out.Platelet count was divided by the spleen size to obtain the platelet count to spleen size ratio. Close follow up of the patient was done until he/she underwent upper GI endoscopy for diagnosis of esophageal varices. A correlation was performed between the platelet count to spleen size ratio and findings of upper GI endoscopy
Results: Collected data was analyzed through computer software SPSS11.0.The ratio of platelet count to spleen size as a predictor of esophageal varices in patients of liver cirrhosis demonstrated sensitivity of 92.5%, specificity of 87.5%, positive predictive value of 93.3%, negative predictive value of 86.1% and diagnostic accuracy of 90.8%
Conclusion: The ratio of platelet count to spleen size as a predictor of esophageal varices in patients of liver cirrhosis is found to be high