Venous thromboembolism (VTE), including pulmonary embolism (PE) and deep venous thrombosis (DVT), is a common and severe complication of critical illness. Although well documented in the general population, the prevalence of PE is less known in the ICU, where it is more difficult to diagnose and to treat. Critically ill patients are at high risk of VTE because they combine both general risk factors together with specific ICU risk factors of VTE, like sedation, immobilization, vasopressors or central venous catheter. Compression ultrasonography and computed tomography (CT) scan are the primary tools to diagnose DVT and PE, respectively, in the ICU. CT scan, as well as transesophageal echography, are good for evaluating the severity of PE. Thromboprophylaxis is needed in all ICU patients, mainly with low molecular weight heparin, such as fragmine, which can be used even in cases of non-severe renal failure. Mechanical thromboprophylaxis has to be used if anticoagulation is not possible. Nevertheless, VTE can occur despite well-conducted thromboprophylaxis.

Background Severely ill patients might develop an alteration of their immune system called post-aggressive immunosuppression. We sought to assess the risk of ICU-acquired infection and of mortality according to the absolute lymphocyte count at ICU admission and its changes over 3 days.MethodsAdults in ICU for at least 3 days with a shock or persistent low blood pressure were extracted from a French ICU database and included. We evaluated the impact of the absolute lymphocyte count at baseline and its change at day 3 on the incidence of ICU-acquired infection and on the 28-day mortality rate. We categorized lymphocytes in 4 groups: above 1.5 × 103 cells/µL; between 1 and 1.5 × 103 cells/µL; between 0.5 and 1 × 103 cells/µL; and below 0.5 × 103 cells/µL.ResultsA total of 753 patients were included. The median lymphocyte count was 0.8 × 103 cells/µL [0.51–1.29]. A total of 174 (23%) patients developed infections; the 28-day mortality rate was 21% (161/753). Lymphopenia at admission was associated with ICU-acquired infection (p < 0.001) but not with 28-day mortality. Independently of baseline lymphocyte count, the absence of lymphocyte count increase at day 3 was associated with ICU-acquired infection (sub-distribution hazard ratio sHR: 1.37 [1.12–1.67], p = 0.002) and with 28-day mortality (sHR: 1.67 [1.37–2.03], p < 0.0001).ConclusionLymphopenia at ICU admission and its persistence at day 3 were associated with an increased risk of ICU-acquired infection, while only persisting lymphopenia predicted increased 28-day mortality. The lymphocyte count at ICU admission and at day 3 could be used as a simple and reproductive marker of post-aggressive immunosuppression.Electronic supplementary materialThe online version of this article (doi:10.1186/s13613-017-0242-0) contains supplementary material, which is available to authorized users.