Kyriaki Savvidou1, Aglaia Dimitrakopoulou2, Nikolitsa Kafasi2, Konstantinos Konstantopoulos3, Theodore Vassilakopoulos3, Maria Angelopoulou3, Marina Siakantaris4, Penelope Korkolopoulou5, Panagiotis Kanavaros6, Panagiota Mikou1. 1. Department of Cytopathology, Laiko Hospital of Athens, Athens, Greece. 2. Flow Cytometry-Immunology Department, Laiko Hospital of Athens, Athens, Greece. 3. Hematology Clinic, Laiko Hospital, National and Kapodistrian University of Athens, Athens, Greece. 4. 1st Department of Internal Medicine, Laiko Hospital, National and Kapodistrian University of Athens, Athens, Greece. 5. Department of Pathology, Laiko Hospital, National and Kapodistrian University of Athens, Athens, Greece. 6. Department of Anatomy- Histology- Embryology, University of Ioannina, Athens, Greece.
Abstract
BACKGROUND: We investigated serous effusions occurring during the course of an already known hematologic neoplasia or as a first manifestation of it. We correlated cytology results with flow cytometry results, when available. In the absence of flow cytometry, our correlation was based on clinical follow up information obtained retrospectively. We evaluated our results in relation to the data of the literature and we considered some new suggestions for the improvement of cytology service. METHODS: Serous effusions in hematologic patients were retrieved from the files of the Department of Cytology, Laiko Hospital, for a period of 2 years. All patients had enrolled either a previous hematologic history, or a suspicious clinical and imaging status. Seventy-three serous effusions were included. Cytology reports consisting of morphology and immunocytochemistry assessment were correlated to flow cytometry results and, occasionally, to clinical follow-up. RESULTS: In the group of patients with previous history, sensitivity was 82.76%, positive predictive value was 100%, specificity 100%, and negative predictive value was 58.33%. In the group of patients without previous history, sensitivity and positive predictive value were both 91%, whereas specificity and negative predictive value could not be estimated. CONCLUSION: We provide evidence that the diagnostic accuracy of cytology with the adjunct of immunocytochemistry is high compared to flow cytometry for detecting hematologic malignancies. In order to improve clinical performance, it is suggested that a cytology triage of serous effusions in all patients with hematologic malignancy must be implemented.
BACKGROUND: We investigated serous effusions occurring during the course of an already known hematologic neoplasia or as a first manifestation of it. We correlated cytology results with flow cytometry results, when available. In the absence of flow cytometry, our correlation was based on clinical follow up information obtained retrospectively. We evaluated our results in relation to the data of the literature and we considered some new suggestions for the improvement of cytology service. METHODS: Serous effusions in hematologic patients were retrieved from the files of the Department of Cytology, Laiko Hospital, for a period of 2 years. All patients had enrolled either a previous hematologic history, or a suspicious clinical and imaging status. Seventy-three serous effusions were included. Cytology reports consisting of morphology and immunocytochemistry assessment were correlated to flow cytometry results and, occasionally, to clinical follow-up. RESULTS: In the group of patients with previous history, sensitivity was 82.76%, positive predictive value was 100%, specificity 100%, and negative predictive value was 58.33%. In the group of patients without previous history, sensitivity and positive predictive value were both 91%, whereas specificity and negative predictive value could not be estimated. CONCLUSION: We provide evidence that the diagnostic accuracy of cytology with the adjunct of immunocytochemistry is high compared to flow cytometry for detecting hematologic malignancies. In order to improve clinical performance, it is suggested that a cytology triage of serous effusions in all patients with hematologic malignancy must be implemented.