A probability prediction rule for malignant cervical lymphadenopathy usingsonography

Background. Our purpose was to weigh various sonographic parameters as pred icting malignant cervical lymphadenopathy and build a reliable prediction r ule. Methods. One hundred and eighty-nine cervical lymph node lesions from 125 c onsecutive patients were used for building the prediction model. Sonographi c variables, including 15 morphologic features of B-mode, 5 vascular parame ters of color Doppler mode, along with age and sex, were analyzed with mult ivariate logistic regression to evaluate the joint effect of a set of indep endent variables. A prediction rule for malignant lymphadenopathy was estab lished, and prospective validation was assessed on a new group consisting o f 100 lymph nodes from another 60 consecutive patients. Results. The association of heterogeneous content, long transverse diameter , pathologic vascular pattern, high Vascular density, and older age provide d the most robust prediction value. Scoring scale was designed as Ix (age) + 2x (vascularity index) + 3x (short axis) + 4x (vascular pattern) + 4x (in ternat echo) according to the parameter estimates of multivariate logistic regression analysis. Gut-off Value of score greater than or equal to 10 as malignancy resulted in 89.2% sensitivity and 85.2% specificity. Prospective validation also showed satisfactory results (sensitivity, 82.9%, specifici ty, 86.2%). Conclusions. By measuring only 4 sonographic parameters and age, this predi ction rule could provide the physician a nonconfusing and reliable probabil ity reference far managing cervical lymphadenopathy. (C) 2000 John Wiley & Sons, Inc.