TY - JOUR
T1 - Thrombotic complications in adult patients with lymphoma
T2 - A meta-analysis of 29 independent cohorts including 18 018 patients and 1149 events
AU - Caruso, Vanesa
AU - Di Castelnuovo, Augusto
AU - Meschengieser, Susana
AU - Lazzari, Maria A.
AU - De Gaetano, Giovanni
AU - Storti, Sergio
AU - Iacoviello, Licia
AU - Donati, Maria Benedetta
PY - 2010/7/1
Y1 - 2010/7/1
N2 - Thrombotic complications in hematologic malignancies have important clinical implications. In this meta-analysis we sought to obtain accurate estimates of the thrombotic risk in lymphoma patients. Articles were searched in electronic databases and references. Eighteen articles were identified (29 cohorts, 18 018 patients and 1149 events). Pooled incidence rates (IRs) were calculated by the use of a method based on the exact maximum likelihood binomial distribution. The global IR of thrombosis was 6.4% (95% confidence interval [CI] 6.0%-6.8%). The global IRs of venous or arterial events were 5.3% (95% CI, 5.0%-5.7%) and 1.1% (95% CI, 0.9%-1.2%), respectively. The IR of thrombosis observed in subjects with non-Hodgkin lymphoma (NHL) was 6.5% (95% CI, 6.1%-6.9%), significantly greater than that observed for patients with Hodgkin lymphoma (4.7%; 95% CI, 3.9%-5.6%). Within NHL, patients with high-grade disease had a greater risk of events (IR 8.3%; 95% CI, 7.0%-9.9%) than low-grade disease (IR 6.3%; 95% CI, 4.5%-8.9%). This meta-analysis shows that the IR of thrombosis in lymphoma patients is quite high, especially in those with NHL at an advanced stage of the disease. These results may help better defining lymphoma populations at high thrombotic risk, to whom prophylactic approaches could be preferentially applied.
AB - Thrombotic complications in hematologic malignancies have important clinical implications. In this meta-analysis we sought to obtain accurate estimates of the thrombotic risk in lymphoma patients. Articles were searched in electronic databases and references. Eighteen articles were identified (29 cohorts, 18 018 patients and 1149 events). Pooled incidence rates (IRs) were calculated by the use of a method based on the exact maximum likelihood binomial distribution. The global IR of thrombosis was 6.4% (95% confidence interval [CI] 6.0%-6.8%). The global IRs of venous or arterial events were 5.3% (95% CI, 5.0%-5.7%) and 1.1% (95% CI, 0.9%-1.2%), respectively. The IR of thrombosis observed in subjects with non-Hodgkin lymphoma (NHL) was 6.5% (95% CI, 6.1%-6.9%), significantly greater than that observed for patients with Hodgkin lymphoma (4.7%; 95% CI, 3.9%-5.6%). Within NHL, patients with high-grade disease had a greater risk of events (IR 8.3%; 95% CI, 7.0%-9.9%) than low-grade disease (IR 6.3%; 95% CI, 4.5%-8.9%). This meta-analysis shows that the IR of thrombosis in lymphoma patients is quite high, especially in those with NHL at an advanced stage of the disease. These results may help better defining lymphoma populations at high thrombotic risk, to whom prophylactic approaches could be preferentially applied.
UR - http://www.scopus.com/inward/record.url?scp=77955869841&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77955869841&partnerID=8YFLogxK
U2 - 10.1182/blood-2010-01-258624
DO - 10.1182/blood-2010-01-258624
M3 - Article
C2 - 20378755
AN - SCOPUS:77955869841
VL - 115
SP - 5322
EP - 5328
JO - Blood
JF - Blood
SN - 0006-4971
IS - 26
ER -