Thrombocytopenia in the ICU
August 1, 2014
Related Articles
SPECIAL FEATURE
Thrombocytopenia in the ICU
By Richard J. Wall, MD, MPH
Pulmonary Critical Care & Sleep Disorders Medicine, Southlake Clinic, Valley Medical Center, Renton, WA
Dr. Wall reports no financial relationships relevant to this field of study.
The term thrombocytopenia refers to platelet counts < 150,000/microL, with severe thrombocytopenia defined as a count < 50,000/microL. A normal platelet count is between 150,000 and 450,000/microL, but some healthy individuals have baseline counts outside that range. For this reason, low counts are not always a cause for alarm and repeat testing is warranted before acting on a value. Critically ill patients have thrombocytopenia for a variety of reasons and it is often a marker of physiological stress. Epidemiological studies have showed that ICU patients admitted with thrombocytopenia have higher severity of illness scores and more organ dysfunction.1 In this review, I will discuss the topic of thrombocytopenia in the critically ill patient. While a detailed review of workup and management of thrombocytopenia is beyond the scope of this article, I will touch on key concepts and common disease states that all ICU clinicians should know.
PLATELET PRODUCTION
A healthy individual produces platelets on the order of 35,000-50,000/microL each day, but megakaryocyte production can ramp up considerably during times of increased demand. Platelets survive in the circulation for 8-10 days, and then are removed by the reticuloendothelial system in the liver and spleen. Disease processes that upset this balance can lead to thrombocytopenia. In general terms, these processes include decreased bone marrow production, immune-mediated destruction, consumption in thrombi, and splenic sequestration (as with portal hypertension and splenomegaly). Typically, bone marrow disorders that reduce platelet production will also reduce other hematologic cell lines, causing pancytopenia. Normally, one-third of the body’s platelets reside in the spleen, in balance with the circulating pool. Conditions that increase splenic size or congestion will drop the platelet count without decreasing the total body platelet mass.
EPIDEMIOLOGY
Thrombocytopenia is a common finding in ICU patients. Hui et al conducted a systematic review of 24 studies (n = 6894 patients) looking at the epidemiology and consequences of thrombocytopenia in ICU patients.2 Patients were in medical, surgical, mixed, cardiac, and trauma ICUs. This insightful study showed several key themes. In general, the quality of evidence on this topic is fair. All 24 studies were observational, half were retrospective, most were single-center, most didn’t account for loss to follow-up, some didn’t specify ICU type, and the definition of "thrombocytopenia" differed between studies. In only two studies did researchers even have a protocol for re-verifying the platelet count.
Hui and colleagues found that the frequency of thrombocytopenia varied across studies. The prevalence of thrombocytopenia at ICU admission ranged from 8-68%. The incidence of developing new thrombocytopenia during the course of an ICU stay ranged from 13-44%. Nine studies reported risk factors for thrombocytopenia. The most common risk factors for the development of thrombocytopenia were sepsis, renal failure, shock, organ dysfunction, and high illness severity.
Hui et al found the rigor of studies looking at outcomes associated with thrombocytopenia in ICU patients to be mediocre. Of five studies reporting bleeding outcomes, only two used objective a priori definitions of "bleeding." Only one study looked at the risk of bleeding using multivariate methods to adjust for confounders and it found no association between thrombocytopenia and bleeding. Among the 16 studies examining mortality, only eight adjusted for confounders; six studies found thrombocytopenia was a risk factor for death, and two studies found no association. Two studies investigated the association between thrombocytopenia and antiplatelet medications: one found no association between aspirin and thrombocytopenia, and the other found that use of non-steroidal anti-inflammatory drugs was more common among thrombocytopenic patients (P = 0.012).1
RISKS OF THROMBOCYTOPENIA
Williamson et al examined risk factors and outcomes of ICU thrombocytopenia in a large and heterogeneous cohort of adult ICU patients.3 Subjects came from a 16-bed medical/coronary unit and a 14-bed surgical/trauma unit in a tertiary hospital in Quebec. A total of 20,696 patients were included. They defined thrombocytopenia as < 100,000 platelets/microL. They used multivariate regression to adjust for demographics, comorbidities, severity of illness, and numerous other confounders. Thrombocytopenia at admission was prevalent in 13% of patients. Another 8% of patients developed incident thrombocytopenia during their ICU stay. Overall, they found an increased ICU length of stay for both prevalent and incident thrombocytopenia (4.0 days and 6.1 days, respectively) compared to patients without thrombocytopenia (3.4 days; P < 0.001 for both). Major bleeding was more common for both prevalent and incident thrombocytopenia (20.3% and 19.3%, respectively) compared to patients without thrombocytopenia (12.5%; P < 0.001 for both). Prevalent and incident thrombocytopenia were associated with increased mortality (14.3% and 24.7%, respectively) compared to patients without thrombocytopenia (10.2%; P < 0.001 for both). In subgroup analyses, it appeared thrombocytopenia had the greatest impact on mortality among patients with cancer, genitourinary, digestive, respiratory, vascular, and infectious diagnoses. It had less of an impact on neurologic and musculoskeletal diagnoses.
In addition to adverse clinical outcomes, thrombocytopenia has other consequences.4 First, fear of bleeding may dissuade clinicians from performing necessary interventions or procedures. Second, thrombocytopenia may prompt platelet transfusion, which has been associated with infectious and noninfectious complications and an increased risk of death in certain populations, such as liver transplant patients.5 Furthermore, thrombocytopenia often leads to additional investigations and tests (e.g., heparin-induced thrombocytopenia (HIT) panel).
When to worry about bleeding? The absolute platelet count is not a good predictor of bleeding risk. Rather, one should consider all factors that increase bleeding risk. Accordingly, the literature does not advocate a specific threshold for transfusing patients. Prior bleeding at a certain platelet count is a better predictor in a given patient, and clinical judgment is useful in this regard. A minimum count of 50,000/microL for surgical procedures seems wise. Severe spontaneous bleeding is likely with counts < 10,000/microL. One exception is patients with idiopathic thrombocytopenic purpura (ITP), who rarely have severe spontaneous bleeding even with very low counts.
APPROACH TO ICU THROMBOCYTOPENIA
When a critically ill patient has thrombocytopenia, the differential is broad (see Table). Always repeat the test to confirm it’s true, review the peripheral smear (to exclude clumping or lab artifact), and assess other hematologic parameters. If the drop occurs after admission to the hospital, it’s likely from infection or drugs. A new reduction is more concerning than a chronic low count, as it suggests an active process. Other abnormal hematologic labs should prompt a diligent workup for a serious diagnosis.
The differential for a thrombocytopenic patient who is actively bleeding or has purpura depends on whether he has other symptoms. If the patient lacks evidence of systemic illness and other hematologic labs are normal, the likely diagnosis is either immune thrombocytopenia (ITP) or drug-induced thrombocytopenia. If the patient has other symptoms, however, the differential is broad (see Table). Depending on clinical judgment, consider coagulation studies, liver function tests, cultures, and bone marrow aspiration. Getting a detailed history is key, including past platelet counts, new medications, bleeding history, and family history. Examination should focus on lymphadenopathy and hepatosplenomegaly.
Table. Common Causes of Thrombocytopenia in ICU Patients |
||
Cause |
Other Possible Findings |
Comment |
Drugs |
Fever, thrombosis (HIT) |
See text |
Infections |
Fever, lymphadenopathy |
#1 cause in ICU patients; viral, bacterial, parasitic |
Liver disease |
Hepatosplenomegaly |
May be initial presentation of liver disease |
Rheumatologic |
Thrombosis (APS) |
See "APS" below |
Post-trauma/surgery |
Typical nadir at post-op day 4 |
Consumption is proportional to intraoperative tissue and blood loss |
Pregnancy |
5% of pregnant woman develop mild gestational thrombocytopenia which resolves after delivery; if count is < 70,000/microL, consider other causes such as preeclampsia, TTP, or HELLP |
|
Nutrient deficiency (e.g., B12, folate, copper) |
Neurologic symptoms |
Consider if history of bariatric surgery or restricted diet |
Myelodysplasia |
Other cytopenias |
|
Aplastic anemia |
Other cytopenias |
|
DIC |
Fever, thrombosis |
Clots usually venous; most common with infection and malignancy |
Bone marrow infiltration (e.g., leukemia, lymphoma) |
Lymphadenopathy, other |
|
Paroxysmal nocturnal |
Thrombosis, hemolytic anemia, cytopenias |
Clots of intrabdominal and cerebral veins |
TTP-HUS |
Neurologic symptoms, |
Clots of small vessels; plasma exchange may be life-saving; many patients present only with anemia and thrombocytopenia |
Antiphospholipid syndrome |
Thrombosis, history of recurrent pregnancy loss, lupus |
Clots either venous and arterial; can occur in isolation or with other processes (SLE, infection, cancer, medications) |
Congenital thrombocytopenia |
May see large platelets on smear; usually diagnosed in childhood |
|
APS = antiphospholipid syndrome; HIT = heparin-induced thrombocytopenia; TTP-HUS = thrombotic thrombocytopenic purpura-hemolytic uremic syndrome; SLE = systemic lupus erythematosis; DIC = disseminated intravascular coagulation; HELLP = hemolysis, elevated liver enzymes, low platelets |
||
A non-bleeding ICU patient admitted with chronic isolated asymptomatic thrombocytopenia likely has ITP, liver disease, HIV infection, or a myelodysplastic syndrome. The most common cause of new onset thrombocytopenia in ICU patients is sepsis, accounting for approximately half of cases.6 After sepsis, the most common causes are liver disease (hypersplenism), disseminated intravascular coagulation (DIC), primary hematologic disorders, medications, massive transfusion, and alcoholism.
When to worry about clotting? A few rare thrombocytopenia conditions are associated with increased risk of thrombosis. These include HIT, antiphospholipid antibody syndrome, disseminated intravascular coagulation (DIC), thrombotic thrombocytopenic purpura-hemolytic uremic syndrome (TTP-HUS), and paroxysmal nocturnal hemoglobinuria. Treatment of each generally focuses on the underlying problem (e.g., stopping heparin, anticoagulation, treating infection, plasma exchange, etc.). In general, appropriate use of anticoagulants or thromboprophylaxis should not be withheld if the platelet count if > 50,000/microL, especially if the patient is at high risk (e.g., postoperative).
PLATELET COUNT TRENDING
Platelet count courses vary among different ICU patient populations.7 For surgical patients, platelet counts frequently dip on days 1-4 as a result of perioperative consumption. The magnitude of the drop reflects the extent of tissue trauma and blood loss. Thereafter, platelet counts rise on days 5-7, and peak at day 14. The reactive count seen at 2 weeks is a downstream response to the acute consumption and subsequent thrombopoietin surge during surgery. If the recovery is blunted after 4 days in a postoperative patient, consider the presence of continuing critical illness. A new drop after day 4 suggests an acute pathology such as infection or drug-induced bone marrow suppression.
For medical ICU patients, the platelet course depends on the underlying disease state and is useful for prognostic purposes. Akca et al performed a prospective observational study of 1449 patients from 40 ICUs in 16 countries.8 Thrombocytopenia was defined as < 150,000/microL. Platelet counts and other measures of organ dysfunction were measured daily. In general, platelet counts dropped significantly in the first days of acute illness, reaching a nadir on day 4. Among survivors, the platelet count returned to its admission value by the end of week 1 and subsequently rose higher than its admission value by day 9. In non-survivors, the platelet count also returned to its admission value at week 1, but then plateaued and there was no subsequent increase. In other words, thrombocytopenia at any time is associated with increased ICU mortality, but prolonged thrombocytopenia with a blunted platelet recovery confers additional risk of death.
Moreau et al looked at hospital mortality among 1077 ICU patients in nine ICUs who subsequently spent > 5 days in the unit.9 All had normal admission platelet counts. At ICU admission, there was no difference in platelet counts between survivors and non-survivors. Like Akca, they found that platelet counts reached a nadir on day 4. After adjusting for severity of illness and other factors, they determined that a 30% decline in platelet count during the first 4 days strongly and independently predicted hospital mortality. Taken together, these two studies suggest that a severe platelet drop in the first 4 days and/or a blunted recovery response are both poor prognostic indicators.
DRUG-INDUCED THROMBOCYTOPENIA
Any drug can cause thrombocytopenia via platelet-reactive antibodies. However, certain culprits are commonly implicated. These include heparin, sulfonamides, beta-lactams, piperacillin, vancomycin, rifampin, carbamazepine, phenytoin, and quinine. Thrombocytopenia will develop within hours of exposure if the patient has been previously exposed, or within 2 weeks if it is a new drug. Upon drug removal, the thrombocytopenia usually resolves within 1 week.
HIT is a potentially lethal condition in which anti-platelet antibodies activate the platelets, thereby both depleting the platelet supply but also increasing the risk of thrombosis. Clots can be either venous or arterial, with venous being more common. Thrombosis occurs in up to half of individuals. Onset is usually 5-10 days after heparin initiation. Rapid early onset of HIT (within 24 hours) can occur if the patient was exposed to heparin in the past 3 months. Upon drug removal, the thrombocytopenia resolves within 1 week. However, the antibodies can persist for 2-3 months. Rarely, "delayed-onset HIT" can occur (with or without thrombosis) several days after heparin is withdrawn because the patients have very high HIT antibody titers. In one series, it occurred a median of 9 days after the drug was stopped.10 The probability of HIT as a cause for a low platelet count can be estimated using the validated "4Ts Score" (Thrombocytopenia, Timing, Thrombosis, and oThers).11 A score < 4 can essentially exclude HIT with a negative predictive value of 99.8%. Patients with a presumptive HIT diagnosis should have immediate heparin discontinuation and administration of a non-heparin anticoagulant (unless there is a bleeding risk). A HIT antibody test should be sent. Readers interested in learning more about this condition are referred to an excellent review by Lee et al.12
SUMMARY
Thrombocytopenia in ICU patients is a common finding with numerous potential causes. Common risk factors include sepsis, renal failure, shock, organ dysfunction, and high illness severity. Surgical ICU patients frequently exhibit thrombocytopenia postoperatively, but a delayed recovery or acute platelet drop merits close scrutiny. Studies looking at outcomes suggest that mortality among thrombocytopenic patients is not due primarily to bleeding.
REFERENCES
- Strauss R, et al. Thrombocytopenia in patients in the medical intensive care unit: Bleeding prevalence, transfusion requirements, and outcome. Crit Care Med 2002;30:1765-1771.
- Hui P, et al. The frequency and clinical significance of thrombocytopenia complicating critical illness: A systematic review. Chest 2011;139:271-278.
- Williamson DR, et al. Thrombocytopenia in the critically ill: Prevalence, incidence, risk factors, and clinical outcomes. Can J Anaesth 2013;60:641-651.
- Crowther MA, et al. Thrombocytopenia in medical-surgical critically ill patients: Prevalence, incidence, and risk factors. J Crit Care 2005;20:348-353.
- Pereboom IT, et al. Platelet transfusion during liver transplantation is associated with increased postoperative mortality due to acute lung injury. Anesth Analg 2009;108:1083-1091.
- Vanderschueren S, et al. Thrombocytopenia and prognosis in intensive care. Crit Care Med 2000;28:1871-1876.
- Thiele T, et al. Thrombocytopenia in the intensive care unit-diagnostic approach and management. Semin Hematol 2013; 50:239-250.
- Akca S, et al. Time course of platelet counts in critically ill patients. Crit Care Med 2002;30:753-756.
- Moreau D, et al. Platelet count decline: An early prognostic marker in critically ill patients with prolonged ICU stays. Chest 2007;131:1735-1741.
- Warkentin TE, Kelton JG. Delayed-onset heparin-induced thrombocytopenia and thrombosis. Ann Intern Med 2001;135: 502-506.
- Lo GK, et al. Evaluation of pretest clinical score (4 T’s) for the diagnosis of heparin-induced thrombocytopenia in two clinical settings. J Thromb Haemost 2006;4:759-765.
- Lee GM, Arepally GM. Diagnosis and management of heparin-induced thrombocytopenia. Hematol Oncol Clin North Am 2013;27:541-563.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.