TY - JOUR T1 - Predictive significance of absolute lymphocyte count and morphology in adults with a new onset peripheral blood lymphocytosis JF - Journal of Clinical Pathology JO - J Clin Pathol SP - 1062 LP - 1066 DO - 10.1136/jclinpath-2014-202545 VL - 67 IS - 12 AU - Ping Sun AU - Emilia M Kowalski AU - Calvino K Cheng AU - Allam Shawwa AU - Robert S Liwski AU - Ridas Juskevicius Y1 - 2014/12/01 UR - http://jcp.bmj.com/content/67/12/1062.abstract N2 - Aims Lymphocytosis is commonly encountered in the haematology laboratory. Evaluation of blood films is an important screening tool for differentiating between reactive and malignant processes. The optimal lymphocyte number to trigger morphological evaluation of the smear has not been well defined in the literature. Likewise, the significance of lymphocyte morphology has not been well studied and there are no consensus guidelines or follow-up recommendations available. We attempt to evaluate the significance of lymphocyte morphology and to define the best possible cut-off value of absolute lymphocyte count for morphology review. Methods 71 adult patients with newly detected lymphocytosis of 5.0×109/L or more were categorised to either a reactive process or a lymphoproliferative disorder. We performed statistical analysis and morphology review to compare the difference in age, gender, lymphocyte count and morphological features between the two groups. Receiver operating characteristic analysis was performed to determine an optimal lymphocyte number to trigger morphology review. Results Lymphoproliferative disorders are associated with advanced age and higher lymphocyte count. Sensitivity, specificity, positive predictive value, negative predictive value and accuracy of lymphocyte morphology as a screening test were 0.9, 0.59, 0.60, 0.58 and 0.71, respectively. The optimal cut-off of lymphocyte number for morphology review was found to be close to 7×109/L. Conclusions We found a moderate interobserver agreement for the morphological assessment. ‘Reactive’ morphology was very predictive of a reactive process, but ‘malignant’ morphology was a poor predictor of a lymphoproliferative disorder. ER -