Prediction of portal vein thrombosis after hepatectomy for hepatocellular carcinoma

Diagnostic accuracy of D-dimer in patients at high-risk for splanchnic vein thrombosis: A systematic review and meta-analysis

“…1, while the citations of the excluded articles are available upon request). Finally, 12 studies were included in this systematic review [21][22][23][24][25][26][27][28][29][30][31][32], since 2 patients cohorts were reported in more than one publication [33,34]. No additional studies were identified from the reference lists of the 14 retrieved reports, from 2 systematic review published on this topic [10,35] and from searching into the grey literature.…”

“…In fact, 7 studies enrolled patients undergoing surgical procedures (e.g. splenectomy, hepatectomy, or gastroesophageal devascularisation) [22][23][24]27,[30][31][32], often in combination with other risk factors (such as liver cirrhosis [23,24,27,30,31] or HCC [32]); 4 studies enrolled patients with liver cirrhosis only [21,25,26,28] and one study enrolled patients with HCC only [29]. In addition, 9 studies evaluated patients with different degree of liver dysfunction, as expressed by Child-Pugh classes A, B or C [21,[23][24][25][26]28,[30][31][32]; however, separate sensitivity and specificity data were available only in 7 of them [21,23,25,26,28,31,32].…”

“…The reference standard imaging was Doppler US in 8 studies [21][22][23][24]27,[29][30][31] and CT or MR scan in 4 studies [25,26,28,32]. Imaging tests and D-dimer were performed on the same day in 2 studies [21,26].…”

“…Imaging tests and D-dimer were performed on the same day in 2 studies [21,26]. With the exception of one study that did not report details of the kits/reagents [22] and one study that used a semi-quantitative D-dimer assay [23], the others used quantitative D-dimer assays (5 latex enhanced immunoassays or immunoturbidimetric assays [21,27,[30][31][32], 1 immunofiltration assay [26] and 4 enzyme-linked immunosorbent assays [24,25,28,29]). Eight studies used the manufacturers' D-dimer cut-offs [22][23][24]27,[29][30][31][32], 3 studies used post-hoc optimal cut-offs [21,25,28], and 1 study presented accuracy results with both the manufacturer's and an optimal cut-off [26].…”

“…The risk of bias for patient selection was high in 9 studies which did not enrol consecutive patients or used inappropriate exclusions [23][24][25][27][28][29][30][31][32]. The risk of bias for the index test was high in 3 studies which assessed only post-hoc D-dimer thresholds [21,25,28].…”