Search

Additional file 1. About the Dutch Health and Youth care Inspectorate

Additional file 2. Interview format for the semi-structured interview

AimThe aim of this study was to determine whether an optimized 3.0-Tesla magnetic resonance imaging (MRI) protocol is sensitive and specific enough to detect patients with endometriosis. Material and MethodsThis was a prospective cohort study with consecutive patients. Forty consecutive patients with clinical suspicion of endometriosis underwent 3.0-Tesla MRI, including a T2-weighted high-resolution fast spin echo sequence (spatial resolution=0.75x1.2x1.5mm(3)) and a 3D T1-weighted high-resolution gradient echo sequence (spatial resolution=0.75x1.2x2.0mm(3)). Two radiologists reviewed the dataset with consensus reading. During laparoscopy, which was used as reference standard, all lesions were characterized according to the revised criteria of the American Fertility Society. Patient-level and region-level sensitivities and specificities and lesion-level sensitivities were calculated. ResultsPatient-level sensitivity was 42% for stage I (5/12) and 100% for stages II, III and IV (25/25). Patient-level specificity for all stages was 100% (3/3). The region-level sensitivity and specificity was 63% and 97%, respectively. The sensitivity per lesion was 61% (90% for deep lesions, 48% for superficial lesions and 100% for endometriomata). The detection rate of obliteration of the cul-the-sac was 100% (10/10) with no false positive findings. The interreader agreement was substantial to perfect (kappa=1 per patient, 0.65 per lesion and 0.71 for obliteration of the cul-the-sac). ConclusionsAn optimized 3.0-Tesla MRI protocol is accurate in detecting stage II to stage IV endometriosis.

The performance of a diagnostic test is often expressed in terms of sensitivity and specificity compared with the reference standard. Calculations of sensitivity and specificity commonly involve multiple observations per patient, which implies that the data are clustered. Whether analysis of sensitivity and specificity per patient or using multiple observations per patient is preferable depends on the clinical context and consequences. The purpose of this article was to discuss and illustrate the most common statistical methods that calculate sensitivity and specificity of clustered data, adjusting for the possible correlation between observations within each patient. This tutorial presents and illustrates the following methods: (a) analysis at different levels ignoring correlation, (b) variance adjustment, (c) logistic random-effects models, and (d) generalized estimating equations. The choice of method and the level of reporting should correspond with the clinical decision problem. If multiple observations per patient are relevant to the clinical decision problem, the potential correlation between observations should be explored and taken into account in the statistical analysis. (C) RSNA, 2012