Thrombocytopenia in the critically ill: considering pathophysiology rather than looking for a magic threshold

Excerpt

Thrombocytopenia is a sensitive marker of severity in acute inflammatory disorders and is therefore a common laboratory finding in critically ill patients. A number of prospective and retrospective studies have provided a crude epidemiological picture of thrombocytopenia in the intensive care unit (ICU) and have assessed its prognostic value [1]. Using common platelet count thresholds of 100 or 150 × 10⁹/L, the prevalence on ICU admission and the incidence of thrombocytopenia acquired during the ICU stay ranged from 8.3 to 67.1 % and from 13 to 41.2 %, respectively, depending on the type of ICU and the case mix of patients. Severe thrombocytopenia is defined as a platelet count <50 × 10⁹/L, with frequency ranging from 2.2 to 41.8 %. Thrombocytopenia in the ICU has been consistently associated with worse outcomes, and platelet count is part of some severity scores. Furthermore, when the platelet count declines over time, the so-called relative thrombocytopenia has also been suggested to be an accurate prognostic factor; For instance, Moreau et al. [2] already demonstrated that a 30 % decline in platelet count on day 4, but not the absolute platelet count at any time during ICU stay, was an independent predictor of mortality. Besides prognostic considerations, the significance of thrombocytopenia in critically ill patients in terms of specific diagnostic procedures, clinical consequences, and the resulting therapeutic implications remains largely unclear in the ICU. …