Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Abnormal platelet counts are a common finding in critically ill patients. Whereas thrombocytopenia, defined as a platelet count less than 150*109/L, affects 13–60 % of Intensive Care Unit (ICU) patients [1] and has been extensively studied, the occurrence of thrombocytosis (platelet counts >400*109/L) is observed less frequently and has not been studied to the same extent.

In this chapter, the main causes of thrombocytopenia and thrombocytosis in critically ill patients will be illustrated, and their implications on morbidity and mortality will be discussed. Due to its importance in the ICU setting, a section in this chapter will be dedicated to heparin-induced thrombocytopenia (HIT).

6.1 Thrombocytopenia: A Classification

Before addressing the issues related to “true” thrombocytopenia, pseudo (or spurious) thrombocytopenia must be defined. In some conditions such as liver diseases, neoplasia, autoimmune disease, or in healthy subjects, antibodies mediated by anticoagulants such as EDTA are responsible for platelet clumping, which, not being detected by cell counters, will lead to falsely low platelet counts [2]. Pseudothrombocytopenia is not clinically significant and is diagnosed by microscopic examination of the blood smear (Fig. 6.1) and by repeating the whole blood count in tubes with a different anticoagulant (heparin- or citrate-based solutions).

Fig. 6.1
figure 1

Diagnostic algorithm based on blood smear (Adapted from Stasi [3])

Full size image

“True” thrombocytopenia, to a variable degree, affects all types of ICU patients in all parts of the world; adult medical ICU patients are mostly affected, but it is also observed in surgical and pediatric patients. These observations underlie the comment made by R.I. Parker in his recent review [1] that thrombocytopenia in ICU patients is “a truly universal occurrence.”

Although a threshold value of 150*109/L is generally accepted to indicate thrombocytopenia, stable platelet counts between 150 and 100*109/L are not necessarily considered pathological. Moreover, it is now recognized that the risk of clinically spontaneous bleeding is significantly high when platelet counts fall below 20–10*109/L [3].

The two main mechanisms responsible for thrombocytopenia are reduced production and increased destruction of platelets; less frequently, a reduced platelet count may also be due to sequestration and hemodilution [1, 2].

Table 6.1 summarizes the main classification criteria for thrombocytopenia, the most frequent pathological mechanisms and the associated clinical conditions. The table does not include causes of thrombocytopenia in pregnancy and postpartum, since these conditions go beyond the scope of this chapter.

Table 6.1 Causes of thrombocytopenia
Full size table

It should always be remembered that in a significant number of cases, thrombocytopenia is due to multiple factors, such as for example in sepsis.

The diagnostic workup for thrombocytopenia must include, in addition to laboratory tests discussed in this chapter, a family history for thrombocytopenia, the evaluation of its “dynamics,” meaning if it is a new finding, if it is chronic or whether it has a relapsing presentation. Information on bleeding episodes is also very important, as is the history of concomitant diseases such as infections, tumors, or autoimmune diseases. Finally, it is of paramount importance to collect the history related to recent medication (heparin, antibiotics) and blood transfusion since especially for hospitalized patients, drug-induced thrombocytopenia (DITP) is among the most common causes of low platelet counts. Since the aim of this chapter is to discuss thrombocytopenia in critically ill patients, it goes without saying that it is challenging to understand this condition in these patients also because a complete history may be difficult to obtain.

Whereas by definition, the Whole Blood Count is the basic laboratory test for diagnosing thrombocytopenia, the microscopic examination of the blood smear gives additional, important information on the pathogenetic mechanism involved [3]. Figure 6.1 illustrates an algorithm that guides the hematologist in the diagnosis of isolated thrombocytopenia. Other tests employed in the diagnosis of the causes of thrombocytopenia are liver and renal function tests, coagulation tests including d-dimers, lactate dehydrogenase, and bone marrow aspirate and biopsy.

Platelet antibody assays and other tests such as reticulated platelets have a limited specificity and therefore their use is debatable [16].

Before describing the clinical conditions associated with thrombocytopenia, the importance of the rate of decline in platelet counts must be pointed out. When a constant, slow reduction in platelet number is observed with minimum (nadir) counts falling below 20*109/L, a DITP due to marrow inhibition is the probable cause. On the other hand, when there is a fast rate of decline (24–48 h) in platelet numbers, an immune mechanism is suspected. A variable rate in platelet reduction is suggestive of consumptive coagulopathy [1].

6.1.1 Thrombocytopenia Due to Reduced Production

Thrombocytopenia caused by bone marrow suppression may be due to acquired or congenital conditions. In the latter category are comprised Fanconi’s anemia, congenital amegakaryocytic thrombocytopenia, thrombocytopenia, and absent radii syndrome; a comprehensive review of these clinical conditions has been recently published by Parikh and Bessler [4]. The inherited bone marrow failure syndromes are genetic disorders affecting blood cell lineages. They are characterized by a wide spectrum of symptoms ranging from aplastic anemia to symptoms related to the suppression of one or two cell lines. Congenital amegakaryocytic thrombocytopenia is an inherited bone marrow failure syndrome usually diagnosed at birth, and characterized by insufficient production of megakaryocytes due to a defect in the thrombopoietin receptor [5].

Acquired bone marrow failure is often due to myelodysplastic syndromes, a heterogeneous group of clonal bone marrow disorders characterized by ineffective hematopoiesis, morphological and functional abnormalities of hematopoietic cells, and increased risk of malignant transformation. The prevalence of thrombocytopenia in these diseases varies from 40 to 65 % [6], and together with platelet dysfunction, is responsible for the increased hemorrhagic risk in these patients.

Sepsis is a condition affecting a significant number of patients admitted to hospitals; a recent review reports that in the USA, 2 % of patients corresponding to 750,000 per year are septic, half of which are admitted to ICUs [8]. Clinical signs of sepsis are diverse and depend on the microorganism, site of original infection, and health condition of the patient. Thrombocytopenia in sepsis is a common finding and severe forms of sepsis are associated with coagulation disorders that can lead to disseminated intravascular coagulation (DIC).

Thrombocytopenia can also be caused by drugs that suppress the bone marrow, and in particular megakaryocyte proliferation and maturation. Whereas antimetabolytes, cytotoxic drugs, and alkylating agents exert a toxic effect on all bone marrow cell lines, some antibiotics such as linezolid, may cause a selective suppression of platelet cell lines [11].

Other causes of thrombocytopenia due to decreased production (Table 6.1) are storage disorders [10], infiltration of bone marrow due to neoplastic diseases [9] and radiation therapy [12].

Thrombocytopenia due to reduced production is not a frequent cause of admission to the ICUs, since it is more often preexistent.

6.1.2 Thrombocytopenia Due to Enhanced Destruction or Consumption

6.1.2.1 Thrombocytopenia Due to Enhanced Destruction: Nonimmune Mechanisms

Medical devices such as mechanical heart valves, left-ventricular assistance devices, and aortic balloon pumps may be responsible for the destruction of platelets. In a study on 1,302 patients who underwent percutaneous coronary intervention (PCI) with baseline normal platelet counts (≥150*109/L), 3.1 % developed post-PCI thrombocytopenia. Multivariate analysis showed that the use of intra-aortic balloon pump was an independent predictor of thrombocytopenia, with an odds ratio of 2.8, confidence intervals 1.1–6.8, p = 0.024. Post-PCI thrombocytopenia was significantly associated with major adverse cardiovascular events at 6 months (hazard ratio 2.7, CI 1.3–5.5, p = 0.0069) [13].

Microangiopathic processes such as thrombotic thrombocytopenic purpura (TTP), hemolytic uremic syndrome (HUS), and disseminated intravascular coagulation (DIC) may be responsible for thrombocytopenia due to enhanced platelet destruction.

TTP is characterized by microvascular platelet clumping, which leads to thrombocytopenia and microangiopathic hemolytic anemia. Common findings are “broken” erythrocytes or schistocytes (see algorithm reported in Fig. 6.1), neurological disorders, renal failure, and fever [14]. The disease is due to a congenital or acquired deficiency in ADAMTS13, a metalloprotease which cleaves von Willebrand factor. ADAMTS13 deficiency is responsible for microvascular thrombosis and thrombocytopenia. Plasma exchange is the optimal therapy, and its effectiveness is probably due to the removal of anti-ADAMTS13 autoantibodies and large von Willebrand factor multimers.

HUS is similar to TTP in that microvascular thrombosis, thrombocytopenia, microangiopathic hemolytic anemia, renal insufficiency, and altered mental status are common features. However, ADAMTS13 is normal and the disease is generally due to endothelial cell damage caused by a toxin produced by pathogenic strains of Escherichia or Shigella. In HUS, thrombocytopenia is usually not severe but dialysis may be required to treat renal insufficiency [23].

DIC does not occur as an isolated event but is practically always associated with an underlying condition such as tissue damage (trauma, burns, hemolytic transfusion reaction, acute transplant rejection), neoplasia, systemic infection, obstetric conditions (abruption placentae, placenta previa, amniotic fluid embolism), and other clinical conditions such as shock, cardiac arrest, and aortic aneurysm. DIC is the result of an overstimulation of the coagulation system and its clinical presentation varies from severe hemorrhage to thrombosis (or both simultaneously). Thrombocytopenia, abnormal prothrombin time and activated partial thromboplastin time (PT and aPTT), decreased fibrinogen and elevated fibrinogen degradation products are common laboratory features of DIC. DIC-associated mortality is mostly due to the original disease, which is complicated by hemorrhage or thrombosis. Multiorgan dysfunction syndrome is a frequent consequence of DIC and is due to hemorrhagic or thrombotic events in liver, heart, kidneys, central nervous system, and lungs [15].

The main therapeutic goal in DIC is that of treating the underlying condition. As far as transfusion of blood products is concerned, there has been a lot of debate on its benefit and potential harm; generally, platelet counts should be kept more than 20*109/L in presence of mild bleeding and more than 50*109/L when there is active bleeding. Plasma or cryoprecipitate should be considered when bleeding is associated with low fibrinogen levels. The aim of fibrinogen replacement is to maintain levels more than 100 mg/dl to prevent or treat bleeding [24].

6.1.2.2 Thrombocytopenia Due to Enhanced Destruction: Immune Mechanisms (Except HIT)

In addition to Heparin-Induced Thrombocytopenia (HIT) which will be discussed in the following section, primary Immune Thrombocytopenia (ITP), post-transfusion purpura (PTP), and drugs may lead to immune platelet destruction.

ITP is an acquired disorder mediated by immunological mechanism, characterized by low platelet counts in the absence of any possible known cause of thrombocytopenia. It affects children and adults (with a slight prevalence in women) and symptoms range from massive bleeding (gastrointestinal, skin–mucosal, and intracranial) to minimal bruising or only alterations in whole blood count. Evaluation of the blood smear is important in the diagnosis of ITP (Fig. 6.1) and antiplatelet antibody assays are not routinely performed due to the low specificity of this test. Adult ITP is treated with corticosteroids or IVIg and platelet transfusions are recommended only for emergency cases in presence of active bleeding [16].

PTP is a rare complication of transfusion occurring 7–10 days after a red blood cell or platelet transfusion and is characterized by a dramatic fall in platelet count reaching a nadir less than 10*109/L. Thrombocytopenia is caused by platelet alloantibodies in the recipient which at first destroy the transfused platelets, but successively also react with self-platelets. PTP is managed by administering IVIg or if available, compatible platelets (usually HPA-1a negative) [19].

Drug-induced thrombocytopenia (DITP) may either be caused by drugs suppressing bone marrow (see previous section) or by drugs eliciting diverse types of antibodies. Table 6.2 summarizes the main types of antibodies implicated in DITP [11]. DITP may be hard to diagnose in critically ill patients, since thrombocytopenia may become evident several days after the beginning of therapy, and has to be distinguished from other causes of thrombocytopenia.

Table 6.2 Antibodies and mechanisms implicated immune drug-induced thrombocytopenia
Full size table

Other causes of thrombocytopenia include platelet sequestration and hemodilution. Thrombocytopenia is a common feature of liver cirrhosis and is attributable to portal hypertension with sequestration of platelets in the enlarged spleen [21].

In massive transfusion, defined as the transfusion of one blood volume in 24 h, coagulation abnormalities are almost always present and are in part due to hemodilutional thrombocytopenia. However, coagulopathy associated with massive transfusion has many additional components, among which are coagulation factor dilution, hypothermia, type of solutions used for volume replacement, and DIC [22].

6.1.2.3 Clinical Significance of Thrombocytopenia in Critically Ill Patients

Hui and coworkers published a review in 2011 [25] aimed at better understanding the clinical role of thrombocytopenia in critically ill patients. It analyzed 24 studies for a total of 6,894 patients; whereas 8.3–67.4 % of patients had low platelet counts at admission, the proportion of patients which developed thrombocytopenia during their stay in the ICU ranged from 13 to 44 %. Major risk factors for the development of thrombocytopenia were high illness severity, organ dysfunction, sepsis, and renal failure. The review was unable to show convincing evidence for an association between thrombocytopenia and bleeding, but multivariate analysis conducted by six studies indicated that thrombocytopenia was an independent predictor of mortality. This finding is confirmed by Stansbury and coworkers [26] in their study on the prognostic significance of platelet counts in the first 24 h after severe injury.

6.2 Heparin-Induced Thrombocytopenia (HIT)

Critically ill patients are often suspected of having HIT, because both thrombocytopenia and heparin treatment are common in the ICU setting. Nevertheless, a recent study demonstrated that the diagnosis of HIT was confirmed in only 0.5 % of these patients [27].

Heparin-Induced Thrombocytopenia (HIT) is a particular type of drug-induced thrombocytopenia that is associated with a pro-thrombotic condition, despite a low circulating platelet count. Although this disorder may occur with any molecular-weight heparin, the incidence of HIT is higher with unfractionated heparin compared to low-molecular-weight heparin [28, 29]. Other risk factors are host-related, with the female sex more affected than the male [30] and the surgical population more affected than the medical [31].

Two types of HIT are described with different clinical features. Type 1 HIT is likely induced by a nonimmune mechanism, with circulating platelet clumping in the presence of heparin and their sequestration in the spleen. The consequent thrombocytopenia develops usually in 2–3 days after starting heparin, is mild and resolves spontaneously with no thrombotic or hemorrhagic complications. Unlike the former, Type 2 HIT is an immunomediated disorder, in which the anticoagulant binds to Platelet Factor 4 (PF4), a protein released from activated platelets, and triggers the development of specific antibodies [32]. The macromolecular complex constituted by the antibody and heparin-PF4 binds a specific receptor on the platelet surface leading to further platelet activation [33] and to thrombin generation [34]. Activated platelets are cleared from circulation with consequent thrombocytopenia and a paradoxical enhanced risk for arterial and especially venous thrombosis.

Different laboratory methods are available to identify the presence of HIT antibodies:

  • Functional assays with the HIT patient serum activating normal platelets in the presence of heparin

  • Antigen assays to detect the binding of HIT antibodies to their target heparin/PF4

Functional assays are more specific for clinically relevant antibodies, but require specialized personnel, so antigen assays are the most widely used [35].

A typical feature of Type 2 HIT is the reduction of more than 50 % in the platelet count, leading to a moderate thrombocytopenia with a median platelet nadir of 50–60*109/L; unlike other drug-mediated thrombocytopenias, a platelet number less than 20*109/L is very uncommon. In naïve patients, the typical onset of thrombocytopenia is 5–14 days after the beginning of heparin exposure; in patients treated in the past 3 months, it may occur early within 24 h (early onset). Seldom platelet counts begin to fall after more than 15 days from the beginning of heparin treatment, sometimes after heparin discontinuation with a delay onset [36].

When HIT is strongly suspected, any heparin treatment (even exposure to heparin flushes or lines washing procedure) must be discontinued and replaced with another anticoagulant, for example, direct thrombin or activated FX inhibitors [37]. In a few days, platelet count returns to normal values or to pretreatment values.

Actual guidelines suggest a clinical evaluation with a scoring system to test the likelihood of the disorder [38, 39]. The most widely used is the 4 T’s score, based on clinical traits of HIT such as the degree of thrombocytopenia, the timing of the onset, the presence of a new or enlarged thrombosis, and an eventual different cause of platelet count decrease, as is shown in Table 6.3 [38].

Table 6.3 The 4 T’s score (adapted from [38])
Full size table

With a low score (≤3), HIT can be excluded without any laboratory assay and the heparin treatment may be continued; if the score is moderate or high (4–6), all heparin exposure should be discontinued to avoid HIT complications and an alternative anticoagulant should be chosen [40]. Recently, two new methods have been proposed for assessing the clinical probability of HIT in the early management of patients suspected of having HIT [41, 42] but they need further validation. Whichever method is used, a careful evaluation is necessary in HIT exclusion or confirmation in order to prevent bleeding risks in thrombocytopenic patients.

6.3 Thrombocytosis in Critically Ill Patients

Elevated platelet counts (>400*109/L) are not a common finding among critically ill patients and contrary to thrombocytopenia, thrombocytosis in hospitalized patients has not been investigated at great length. From the etiological point of view, thrombocytosis may be classified as primary or secondary. Whereas the former group includes myeloproliferative or myelodysplastic syndromes, the latter may be either secondary or paraneoplastic. In the ICU patient, the main underlying clinical conditions responsible for thrombocytosis are infection, trauma, splenectomy, hemolysis, bleeding, and drugs such as antifungals, amoxicillin/clavunate, enoxaparin [43]. Two other main conditions leading to thrombocytosis are familial (hereditary), due to a mutation responsible for an increase in the production of thrombopoietin [44], and essential thrombocythemia, a condition which may eventually lead to myelofibrosis or leukemia.

Differential diagnosis between primary and secondary thrombocytosis is not always straightforward in the ICU setting. Generally speaking, if thrombocytosis occurs during the stay in an ICU, it is most probably of secondary nature. However, if the patient is admitted urgently and no previous whole blood count is available, hematological consultation and further testing (e.g., JAK2) might be useful for the characterization of thrombocytosis [43]. For patients affected by secondary thrombocytosis, risk of thrombotic or hemorrhagic complications is <2 %, irrespective of platelet count.

As far as therapy is concerned, there is no threshold above which platelet removal by apheresis or antiaggregation therapy should be initiated. Risk of thrombosis in these patients must consider associated clinical conditions such as sepsis, trauma, and rheumatic disease, which themselves predispose to venous clot formation. Platelet apheresis is able to reduce platelet counts significantly and is used primarily in patients with myeloproliferative diseases in which either a thrombotic or bleeding event has occurred. Aspirin is administered in thrombocytosis (both primary and secondary) patients who have had a thrombotic event.