The risk of thrombosis in children with acute lymphoblastic leukemia (ALL) reportedly ranges between 1% and 37%. Epidemiologic studies have usually been hampered by small numbers, making accurate estimates of thrombosis risk in ALL patients very difficult. The aim of this study was to better estimate the frequency of this complication and to define how the disease, its treatment, and the host contribute to its occurrence. We made an attempt to combine and analyze all published data on the association between pediatric ALL and thrombosis, by using a meta-analytic method. The rate of thrombosis in 1752 children from 17 prospective studies was 5.2% (95% CI: 4.2-6.4). The risk varies depending on several factors. Most of the events occurred during the induction phase of therapy. Lower doses of asparaginase (ASP) for long periods were associated with the highest incidence of thrombosis, as were anthracyclines and prednisone (instead of dexamethasone). The presence of central lines and of thrombophilic genetic abnormalities also appeared to be frequently associated with thrombosis. In conclusion, the overall thrombotic risk in ALL children was significant, and the subgroup analysis was able to identify high-risk individuals, a finding that will hopefully guide future prospective studies aimed at decreasing this risk.

Recent studies suggest that thrombophilic abnormalities and the use of oral contraceptives (OCs) are the leading causes of cerebral vein thrombosis (CVT). The purpose of this study was to assess the association between CVT and thrombophilic states, OCs, and their interaction. For data sources, we used the MEDLINE, EMBASE, and Cochrane Library databases (January 1994 to March 2005), reference lists of retrieved articles, and contact with content experts. We selected studies comparing the prevalence of OC use and the prevalence of prothrombitic abnormalities in patients with CVT compared with healthy controls. Two reviewers independently selected studies and extracted study characteristics, quality, and outcomes. Odds ratios (ORs) were calculated for each trial and pooled using the Mantel-Haenszel method. Seventeen studies were included. There was an increased risk of CVT in patients using OCs (OR 5.59; 95% confidence interval [CI] 3.95 to 7.91; P < .001), and in patients with factor V Leiden (OR 3.38; 95% CI 2.27 to 5.05; P < .001), with mutation G20 210A of prothrombin (OR 9.27; 95% CI 5.85 to 14.67; P < .001) and with hyperhomocysteinemia (OR 4.07; 95% CI 2.54 to 6.52; P < .001). We concluded that OC users, and patients with factor V Leiden, the prothrombin G20 120A mutation, and hyperhomocysteinemia are at a significantly increased risk of CVT.

Heparin-induced thrombocytopenia (HIT) is an uncommon but potentially devastating complication of anticoagulation with unfractionated heparin (UFH) or low-molecular-weight heparin (LMWH). Our objective was to determine and compare the incidences of HIT in surgical and medical patients receiving thromboprophylaxis with either UFH or LMWH. All relevant studies identified in the MEDLINE database (1984-2004), not limited by language, and from reference lists of key articles were evaluated. Randomized and nonrandomized controlled trials comparing prophylaxis with UFH and LMWH and measuring HIT or thrombocytopenia as outcomes were included. Two reviewers independently extracted data on thromboprophylaxis (type, dose, frequency, and duration), definition of thrombocytopenia, HIT assay, and rates of the following outcomes: HIT, thrombocytopenia, and thromboembolic events. HIT was defined as a decrease in platelets to less than 50% or to less than 100 x 10(9)/L and positive laboratory HIT assay. Fifteen studies (7287 patients) were eligible: 2 randomized controlled trials (RCTs) measuring HIT (1014 patients), 3 prospective studies (1464 patients) with nonrandomized comparison groups in which HIT was appropriately measured in both groups, and 10 RCTs (4809 patients) measuring thrombocytopenia but not HIT. Three analyses were performed using a random effects model and favored the use of LMWH: (1) RCTs measuring HIT showed an odds ratio (OR) of 0.10 (95% confidence interval [CI], 0.01-0.2; P = .03); (2) prospective studies measuring HIT showed an OR of 0.10 (95% CI, 0.03-0.33; P < .001); (3) all 15 studies measured thrombocytopenia. The OR was 0.47 (95% CI, 0.22-1.02; P = .06). The inverse variance-weighted average that determined the absolute risk for HIT with LMWH was 0.2%, and with UFH the risk was 2.6%. Most studies were of patients after orthopedic surgery.

Splenectomy has been a standard treatment for adult patients with idiopathic thrombocytopenic purpura (ITP) for more than 50 years. However, the durability of responses, the ability to predict who will respond, and the frequency of surgical complications with splenectomy all remain uncertain. To better interpret current knowledge we systematically identified and reviewed all 135 case series, 1966 to 2004, that described 15 or more consecutive patients who had splenectomy for ITP and that had data for 1 of these 3 outcomes. Complete response was defined as a normal platelet count following splenectomy and for the duration of follow-up with no additional treatment. Forty-seven case series reported complete response in 1731 (66%) of 2623 adult patients with follow-up for 1 to 153 months; complete response rates did not correlate with duration of follow-up (r = -0.103, P = .49). None of 12 preoperative characteristics that have been reported consistently predicted response to splenectomy. Mortality was 1.0% (48 of 4955 patients) with laparotomy and 0.2% (3 of 1301 patients) with laparoscopy. Complication rates were 12.9% (318 of 2465) with laparotomy and 9.6% (88 of 921 patients) with laparoscopic splenectomy. Although the risk of surgery is an important consideration, splenectomy provides a high frequency of durable responses for adult patients with ITP.

Factor V Leiden (FVL) is associated with venous thrombosis; however, an association between FVL and arterial thrombosis remains controversial. We investigated FVL as a risk factor for myocardial infarction (MI), ischemic stroke (IS), or non-MI ischemic heart disease (non-MI-IHD). The design was 3 case-control studies and 3 prospective studies with 21 years' follow-up. The setting was the general population in Copenhagen, Denmark. The participants for The Copenhagen City Heart Study were 20- to 95-year-old participants without cardiovascular disease (control population, n = 7907) or participants diagnosed with MI (n = 469), IS (n = 231), or non-MI-IHD (n = 365). In addition, 3 independent patient populations from Copenhagen University Hospital with MI (n = 493), IS (n = 231), or non-MI-IHD (n = 448) were included. We measured FVL genotype; major cardiovascular risk factors; and MI, IS, and non-MI-IHD incidence and prevalence. Prevalences of FVL heterozygotes and homozygotes in control subjects from the general population were 7.7% and 0.2%. Odds ratios and relative risks of MI in FVL carriers (heterozygotes + homozygotes) versus noncarriers were 1.24 (95% confidence interval [CI], 0.91-1.69) and 0.83 (0.58-1.20) in case-control and prospective studies, respectively. Corresponding risks for IS were 0.92 (95% CI, 0.56-1.53) and 0.68 (0.45-1.04), and for non-MI-IHD 1.01 (95% CI, 0.71-1.44) and 0.97 (0.66-1.42). Findings from The Copenhagen City Heart Study suggest that FVL is not associated with MI, IS, or non-MI-IHD.

This meta-analysis focuses on 2 prospective studies in patients with heparin-induced thrombocytopenia (HIT) and thromboembolic complication (TEC) who were treated with lepirudin (n = 113). Data were compared with those of a historical control group (n = 91). The primary endpoint (combined incidence of death, new TEC, and limb amputation) occurred in 25 lepirudin-treated patients (22.1%; 95% CI, 14.5%-29.8%): 11 died (9.7%; 95% CI, 4.9%-16.8%), 7 underwent limb amputation (6.2%; 95% CI, 2.5%-12.3%), and 12 experienced new TEC (10.6%; 95% CI, 5.8%-18.3%). The risk was highest in the period between diagnosis of HIT and the start of lepirudin therapy (combined event rate per patient day 6.1%). It markedly decreased to 1.3% during lepirudin treatment and to 0.7% in the posttreatment period. From the start of lepirudin therapy to the end of follow-up, lepirudin-treated patients had consistently lower incidences of the combined endpoint than the historical control group (P =.004, log-rank test), primarily because of a reduced risk for new TEC (P =. 005). Thrombin-antithrombin levels in the pretreatment period (median, 43.9 microg/L) decreased after the initiation of lepirudin (at 24 hours +/- 6 hours; median, 9.18 microg/L.) During treatment with lepirudin, aPTT ratios of 1.5 to 2.5 produced optimal clinical efficacy with a moderate risk for bleeding, aPTT ratios lower than 1. 5 were subtherapeutic, and aPTT levels greater than 2.5 were associated with high bleeding risk. Bleeding events requiring transfusion were significantly more frequent in patients taking lepirudin than in historical control patients (P =.02). In conclusion, this meta-analysis provides further evidence that lepirudin is an effective and acceptably safe treatment for patients with HIT.

Chronic myeloid leukemia (CML) is characterized by the chromosomal translocation t(9;22) resulting in the chimeric bcr-abl oncogene that encodes the P210 fusion protein, which contains a unique amino acid sequence. If peptides derived from the leukemia-specific part of P210 are expressed in HLA molecules on the cell membrane of leukemic cells, an immunological response may occur. Recent studies using synthetic peptides identical to the bcr-abl fusion region showed that some peptides are capable of binding to HLA-A3, -A11, and -B8 molecules. Cytotoxic T-cell responses have been induced against bcr-abl-derived synthetic peptides bound to HLA-A3 and -B8. We hypothesized that if antigen processing of the P210 fusion protein leads to presentation of peptides from the fusion region by major histocompatibility complex (MHC) molecules in vivo, this may be reflected in a diminished incidence of CML in individuals expressing HLA-A3, -A11, or -B8. Consequently, lower frequencies of these antigens would be expected in patients with CML compared with unaffected individuals. A case-control study and a meta-analysis were performed to test this hypothesis. The multicenter case-control study compared patients with CML from the data base of the European Group for Blood and Marrow Transplantation (EBMT) with unaffected individuals from the registry of Bone Marrow Donors Worldwide. Patients and controls were matched per country. The meta-analysis consisted of five studies reported in the literature. The multicenter case-control study consisting of 1,899 patients and 512, 363 bone marrow donors as controls yielded odds ratios (ORs) of 0.90 (95% confidence interval [CI], 0.80 to 1.00) for HLA-A3, 1.16 (95% CI, 1.02 to 1.33) for HLA-A11, and an OR of 0.73 (95% CI, 0.65 to 0. 82) for HLA-B8. Coexpression of HLA-A3 and HLA-B8 gave an OR of 0.51 (95% CI, 0.40 to 0.67). This can be translated in a protective effect of 27% for HLA-B8, 10% for HLA-A3, and 49% protection for the combination of HLA-A3 and HLA-B8. The meta-analysis comprising 463 CML patients and 4,912 controls showed a 29% risk reduction for individuals expressing HLA-B8 (OR of 0.71; 95% CI, 0.52 to 0.97), but an OR of 1.19 (95% CI, 0.90 to 1.56) for HLA-A3 and an OR of 1. 09 (95% CI, 0.80 to 1.50) for HLA-A11. In conclusion, these results indicate that HLA-B8 expression, in particular when HLA-A3 is coexpressed, is associated with a diminished incidence of CML. A biological mechanism may be that presentation of bcr-abl breakpoint peptides in these HLA molecules can induce a protective immune response.

Among 4,760 acute lymphoblastic leukemia (ALL) patients enrolled from 1986 to 1995 in two subsequent trials of the BFM and AIEOP study group, 61 patients were found to have Philadelphia chromosome-positive (Ph+) ALL. These patients were analyzed for presenting features and treatment outcome to identify specific prognostic factors. Treatment stratification was based on initial cell mass and early response as determined by blast count in peripheral blood after a 7-day induction prephase with prednisone and one dose of intrathecal methotrexate on day 1. All patients were treated by similar intensive Berlin-Frankfurt-Münster (BFM) protocols. The median age of Ph+ patients was 7.5 years, the median white blood cell count (WBC) was 75 x 10(9)/L, 77% of patients had common ALL, and 29% coexpressed myeloid markers. After a median observation time of 4.2 years, 29 of 61 patients are alive (survival probability [pSUR] at 4 years, 0.49; standard error [SE], 0.06), and 24 of 61 are in first complete remission (CR1; probability of event-free survival [pEFS] at 4 years, 0.38; SE, 0.06). Twenty (35%) of 57 evaluable patients had >/=1,000 leukemic blasts per microliter of blood on day 8 of induction (defined as prednisone-poor-response [PPR]). These patients were older (10.0 v 6.88 years; P = .02) and had a higher WBC (144 v 29 x 10(9)/L; P = .0016) as compared with patients with prednisone good response (PGR; <1,000 blasts/microL at day 8). Only 2 of 20 patients (10%) with PPR remained in CR1 and alive: 6 patients with PPR did not survive after allogeneic bone marrow transplantation (BMT) due to recurring disease (n = 3) and toxicity (n = 3), and 12 nontransplanted patients died due to progression (n = 5) or relapse (n = 7). In contrast, 26 (70%) of the 37 patients with PGR are alive. Of 18 patients transplanted by allo-BMT, 1 relapsed (now in CR2) and 4 died after BMT. Among the 19 patients with PGR treated by chemotherapy alone, 8 remained in CR1 and 11 relapsed, of which 4 are in CR2 or CR3. The prednisone response emerged as the only independent prognostic factor for survival in Cox regression analysis. Thus, two thirds of Ph+ childhood ALL cases can be identified early by PGR, which, when treated with intensive BFM chemotherapy, with or without BMT, have a significantly lower risk of treatment failure. With a median continuous complete remission (CCR) time of 4.1 years, pEFS for PGR is 0.55 (SE, 0.08) compared with 0.10 (SE, 0.07) in patients with PPR (P = .0001). PGR is also an indicator for treatment responsiveness and durable second remission after relapse, which in turn may provide a second chance for BMT.