Robert D Rifkin1, David B Mernoff. 1. Division of Cardiology, Department of Medicine, Barnes-Jewish Hospital, Washington University School of Medicine, St Louis, MO 63110, USA.
Abstract
BACKGROUND: Measurement of the hematocrit (HCT) and total protein (TP) of pericardial fluid can be of critical importance in the management of pericardial effusion (PE) but requires pericardiocentesis. There has been no systematic study of computed tomography (ct) hounsfield density (hd) measurement for the noninvasive evaluation of pe composition. METHODS: We performed CT-guided pericardiocentesis in 53 patients to obtain simultaneous measurement of the PE HD, HCT, and TP. We determined the relationship of PE HCT to the following variables: (1) the average of the mean HD in 5 regions of interest (ROIs) within the PE; (2) the maximum PE HD; (3) the difference between the average effusion region of interest HD and the ventricular blood pool HD. We also correlated PE HD with PE HCT and TP to assess their contribution to the PE CT HD. RESULTS: The average of the HD measured in up to 5 regions within the PE was the best predictor of PE HCT ( r = 0.84). Addition of ventricular blood pool HD and venous HCT to the regression did not improve prediction of PE HCT over average PE HD alone. Pericardial effusion HD was unrelated to PE TP. There was a statistically significant difference in the PE HD among 8 categories of underlying disease etiology in the population. Excluding postcardiotomy cases (n = 22) in which PE etiology is known before pericardiocentesis, visceral or vascular rupture was associated with the highest HD values. Hounsfield density >30 HU had a sensitivity of 100%, specificity 70%, and predictive value 33% for this condition in the remaining cases (n = 31). CONCLUSIONS: Computed tomography is a rapid and accurate method for noninvasive estimation of pericardial fluid HCT and may be helpful in guiding both the acute management and differential diagnosis of PE.
BACKGROUND: Measurement of the hematocrit (HCT) and total protein (TP) of pericardial fluid can be of critical importance in the management of pericardial effusion (PE) but requires pericardiocentesis. There has been no systematic study of computed tomography (ct) hounsfield density (hd) measurement for the noninvasive evaluation of pe composition. METHODS: We performed CT-guided pericardiocentesis in 53 patients to obtain simultaneous measurement of the PE HD, HCT, and TP. We determined the relationship of PE HCT to the following variables: (1) the average of the mean HD in 5 regions of interest (ROIs) within the PE; (2) the maximum PE HD; (3) the difference between the average effusion region of interest HD and the ventricular blood pool HD. We also correlated PE HD with PE HCT and TP to assess their contribution to the PE CT HD. RESULTS: The average of the HD measured in up to 5 regions within the PE was the best predictor of PE HCT ( r = 0.84). Addition of ventricular blood pool HD and venous HCT to the regression did not improve prediction of PE HCT over average PE HD alone. Pericardial effusion HD was unrelated to PE TP. There was a statistically significant difference in the PE HD among 8 categories of underlying disease etiology in the population. Excluding postcardiotomy cases (n = 22) in which PE etiology is known before pericardiocentesis, visceral or vascular rupture was associated with the highest HD values. Hounsfield density >30 HU had a sensitivity of 100%, specificity 70%, and predictive value 33% for this condition in the remaining cases (n = 31). CONCLUSIONS: Computed tomography is a rapid and accurate method for noninvasive estimation of pericardial fluid HCT and may be helpful in guiding both the acute management and differential diagnosis of PE.
Authors: Eduardo M Vilela; Catarina Ruivo; Claudio E Guerreiro; Marisa P Silva; Ricardo Ladeiras-Lopes; Daniel Caeiro; Gustavo P Morais; João Primo; Pedro Braga; Nuno Ferreira; José Pedro L Nunes; Vasco Gama Ribeiro Journal: Ther Adv Cardiovasc Dis Date: 2018-08-15
Authors: Farbod Sedaghat-Hamedani; Edgar Zitron; Elham Kayvanpour; Hanns-Martin Lorenz; Hugo A Katus; Benjamin Meder Journal: Clin Res Cardiol Date: 2014-04-29 Impact factor: 5.460