Thrombocytopenia in septic shock patients--a prospective observational study of incidence, risk factors and correlation with clinical outcome.
The objectives of the study were to study the incidence of various degrees of severity of thrombocytopenia in septic shock, the risk factors for its development and the correlation with clinical outcome. Complete blood counts, chemistry panel, arterial lactate, serum cortisol, APACHE II score, logistic organ dysfunction score and SOFA score were determined in 69 septic shock patients within 24 hours of admission or onset of septic shock. We followed the patients until they died or for six months to determine the mortality rate. The incidence of thrombocytopenia in our study group was 55%. Patients with thrombocytopenia had significantly higher serum creatinine, SOFA score, vasopressor requirement, lower [P.sub.a][O.sub.2]/Fi[O.sub.2] ratio and higher mortality than those without thrombocytopenia (P <0.05). Higher SOFA score, low [P.sub.a][O.sub.2]/Fi[O.sub.2] ratio and high vasopressor dose were independent risk factors for development of thrombocytopenia. The presence of thrombocytopenia had significant correlation with SOFA score (P=0.008). On receiver-operator characteristic curve analysis, platelet count was found to be predictive of increased mortality (area under curve=0.56). Thrombocytopenic patients had 1.4 times the risk of mortality and lower survival probability at six months (log rank test P=0.03). In conclusion, thrombocytopenia is common in septic shock and is associated with worse clinical outcome. Higher SOFA score, low [P.sub.a][O.sub.2]/Fi[O.sub.2] ratio and high vasopressor dose are independent risk factors for development of thrombocytopenia in septic shock.
Key Words: thrombocytopenia, critically ill, sepsis, prognosis, septic shock, platelet count, mortality
Thrombocytopenia is a common finding in septic shock patients in the intensive care unit (ICU). Various mechanisms have been attributed to explain the occurrence of thrombocytopenia, including disseminated intravascular coagulation (DIC) (1-3), cytokine-driven haemophagocytosis of platelets (4,5), immune-mechanisms, such as elevated platelet-associated IgG (1,6), invasive catheters, especially pulmonary artery catheters (7,8), and medications, such as heparin and pencillin analogues. Thrombocytopenia in critically ill patients, including trauma patients and surgical patients, has been found to be an independent marker for poor prognosis, increased bleeding, longer ICU stay and increased mortality (7-11).
Most studies so far have explored the risk factors and clinical outcomes in critically ill medical, surgical, trauma or cardiac patients with thrombocytopenia. In this study, we looked at thrombocytopenic septic shock patients in a medical ICU. The objectives of the study were to study, 1) incidence of various degrees of severity of thrombocytopenia in septic shock, 2) risk factors for its development and 3) the correlation of thrombocytopenia with organ dysfunction, length of ICU stay and clinical outcome including mortality.
MATERIALS AND METHODS
This prospective, observational study was done from August 5, 2005 to March 10, 2006 in 69 consecutive septic shock patients at the 16-bed medical intensive care unit of Brookdale University Hospital and Medical Center, a tertiary academic medical centre affiliated with State University of New York, Brooklyn, U.S.A. The inclusion criteria included septic shock patients on vasopressors. The definition of septic shock was in accordance with current guidelines (12). The exclusion criteria included history of malignancy, past use of chemotherapeutic drugs, a history of platelet disorders, mechanical cardiac valves, splenectomy or hypersplenism, alcohol abuse, other causes of shock except sepsis and septic patients who were not in shock.
Detailed medical history, physical examination, complete blood counts, biochemistry panel, arterial lactate and serum cortisol were determined in all patients within 24 hours of admission or onset of septic shock as per the ICU protocol. We used three well-established scoring systems to estimate the clinical prognosis and organ dysfunction in our study: APACHE II score (13), LOD score (14) and SOFA score (15), which were determined in all patients at admission or onset of septic shock. Thrombocytopenia was defined as platelet count <150 x [10.sup.9]/l, the lower limit of platelet count in our clinical laboratory. We classified patients into various grades of severity according to the degree of thrombocytopenia: mild=101-149 x [10.sup.9]/l, moderate=51-100 x [10.sup.9]/l, severe=21-50 x [10.sup.9]/l and very severe=<20 x [10.sup.9]/l (15). Other data collected at admission included age, source of sepsis, use of medications including frusemide, heparin and beta-lactam antibiotics, presence of bacteraemic sepsis, [P.sub.a][O.sub.2]/Fi[O.sub.2] ratio and use of mechanical ventilation. A consultant haematologist determined the causes of thrombocytopenia in the study group after evaluation of potential causes like sepsis, medications, DIC (defined as the presence of elevated D-dimers and a [greater than or equal to] 25% reduction in two of the following markers: prothrombin time, antithrombin, fibrinogen level and platelet count within 24 hours) and antiplatelet factor 4-heparin antibodies in cases suspicious of heparin-induced thrombocytopenia. We followed the patients until they died or for six months to determine the mortality rate. Major bleeding episodes were recorded for each patient, including life-threatening episodes, and those compromising haemodynamic status or requiring transfusions. The study was approved by the Institutional Review Board of the hospital. We did not take informed consent as the study was merely observational and did not deviate from the usual standards of ICU care of septic shock patients.
Statistical analysis was performed using statistical software SPSS 14.0 (SPPS Inc., Chicago, IL, U.S.A.). Continuous variables were compared with Student's t test for normally distributed variables and Mann-Whitney test for non-parametrically distributed variables. The chi-square test and Fisher's exact test were used to compare categorical variables. Univariate regression analysis was done to identify candidate variables for multivariate regression analysis. Candidate variables were entered into a backward stepwise multivariate logistic regression model. Receiver-operator characteristic (ROC) curve analysis was done to see prediction of mortality by platelet count and to determine cut-off level for platelet count for determining increased mortality. Kaplan-Meier survival analysis was done to determine survival probability of patients with and without thrombocytopenia and log rank test was done to compare the two survival curves. Hazard ratio for predicting increased mortality by platelet count was calculated using Kaplan-Meier survival analysis. All P values were two-sided and were considered significant at [less than or equal to] 0.05.
We studied 69 septic shock patients with a mean age of 66.2 years (95% confidence interval=62.5-69.9, range=24 to 95 years). Major sources for sepsis included pneumonia (26 patients), urinary tract infection (16 patients), skin/decubitus (nine patients), intra-abdominal (five patients), central nervous system infection (three patients) and unknown source (10 patients). The characteristics of the study patients are demonstrated in Table 1.
Thrombocytopenia was present in 38 (55%) patients. The distribution of patients into various grades of severity of thrombocytopenia is shown in Table 1. Table 2 shows the causes of thrombocytopenia in the study group.
Table 3 shows the comparison of clinical data in patients with and without thrombocytopenia. Patients with thrombocytopenia had significantly higher serum creatinine, SOFA score and vasopressor requirement as well as lower [P.sub.a][O.sub.2]/Fi[O.sub.2] ratio (P <0.05). Patients with thrombocytopenia also had higher mortality rate compared with patients without thrombocytopenia (71% vs. 35.5%, P=0.001).
The risk factors associated with thrombocytopenia in septic shock according to multivariate logistic regression analysis are demonstrated in Table 4. Higher SOFA score, low [P.sub.a][O.sub.2]/Fi[O.sub.2] ratio and high vasopressor dose were independent risk factors for development of thrombocytopenia after adjusting for age and gender. Presence of renal failure (as measured by serum creatinine), serum albumin, LOD score, APACHE II score and use of heparin, frusemide and beta-lactam antibiotics were not independent risk factors for development of thrombocytopenia after adjusting for age and gender.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
The presence of thrombocytopenia had significant correlation with SOFA score (r=-0.32, P=0.008) as shown in Figure 1, but had no significant correlation with APACHE II score (r=-0.09, P=0.47), LOD score (r=-0.004, P=0.97) and number of ICU days (r=-0.07, P=0.54). On ROC curve analysis, platelet count was found to be predictive of increased mortality (area under curve=0.56, 95% confidence interval=0.43-0.68) as shown in Figure 2. A platelet count of <95 x [10.sup.9]/l had a sensitivity of 39.5% and specificity of 77.4% in predicting increased mortality as demonstrated in Figure 3 on ROC curve analysis. Kaplan-Meier survival analysis (Figure 4) showed that thrombocytopenic patients had 1.4 times increased risk of mortality (hazard ratio=1.4, 95% confidence interval=0.72-2.95) and lower survival probability at six months (log rank test P=0.03). Five patients (7.2%) had bleeding at some time during their ICU stay, requiring packed red cell transfusion.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Our study results demonstrate that thrombocytopenia is common in septic shock and sepsis is the major causative factor. Higher SOFA score, low [P.sub.a][O.sub.2]/Fi[O.sub.2] ratio and high vasopressor dose are independent risk factors for development of thrombocytopenia. Patients with thrombocytopenia had worse clinical outcome determined by higher serum creatinine, SOFA score, vasopressor requirement, lower [P.sub.a][O.sub.2]/Fi[O.sub.2] ratio and increased mortality.
The incidence of thrombocytopenia was 55% in our study patients. This incidence is higher than that reported in previous studies which have reported an incidence of 23 to 27% in ICU patients (7,16), up to 41% in trauma patients (17) and 35 to 58% in patients with sepsis or septicaemia (18,19). This could be explained by the reason that our study population was more critically ill as we included only patients with septic shock. This variation could also be explained by different definitions of thrombocytopenia (platelet counts <100 x [10.sup.9]/l, <150 x [10.sup.9]/l or <200 x [10.sup.9]/l) to that used in some previous studies.
Thrombocytopenia in septic shock has various contributing causes. Various mechanisms have been given to explain the effect of sepsis on platelet count as explained earlier (1-8). In our study we found that sepsis was the most contributing factor for thrombocytopenia. Sepsis was also the major independent risk factor for thrombocytopenia in another study (1). Medications contributed to only 13% cases of thrombocytopenia while no case of heparin-induced thrombocytopenia was found in our study.
We found that higher SOFA score, low [P.sub.a][O.sub.2]/[F.sub.i][O.sub.2] ratio and high vasopressor dose were independent risk factors for development of thrombocytopenia after adjusting for other factors. Presence of renal failure (as measured by serum creatinine), serum albumin, LOD score, APACHE II score and use of heparin, frusemide and beta-lactam antibiotics were not independent risk factors for development of thrombocytopenia after adjusting for age and gender. A previous study also found that sepsis and elevated creatinine level were independent risk factors for severe thrombocytopenia (7). Heparin use was not a significant risk factor for thrombocytopenia in other studies as well (7,20). Beta-lactam antibiotics were not an independent risk factor for development of thrombocytopenia in our study. This finding is also supported by other studies (1,16,17,20). Since drug-induced thrombocytopenia represents only a minority of such cases in the ICU, it has been suggested that clinicians should regard evolving thrombocytopenia in critically ill patients as an early warning sign of sepsis, rather than attributing it to medications (21).
We found that septic shock patients with thrombocytopenia at admission or onset of septic shock had significantly higher serum creatinine, SOFA score, vasopressor requirement as well as lower [P.sub.a][O.sub.2]/[F.sub.i][O.sub.2] ratio. Patients with thrombocytopenia also had higher mortality rate compared to patients without thrombocytopenia. On ROC curve analysis, platelet count was found to be predictive of increased mortality. Platelet count of <95 x [10.sup.9]/l had a sensitivity of 39% and specificity of 77% in predicting increased mortality. Thrombocytopenic patients also had 1.4 times increased risk of mortality and lower predicted survival at six months on Kaplan-Meier survival analysis. This is the first study exploring the clinical outcome of thrombocytopenic septic shock patients to the best of our knowledge. Thrombocytopenia has been found to be an independent marker for mortality in previous studies in critically ill medical or surgical patients (7-11) and correlates with a prolonged ICU stay, more bleeding and worse survival (1,7). However, thrombocytopenia appears to be an early marker and not a causative factor for increased mortality as apparent by the low bleeding incidence in our study patients. It appears that thrombocytopenia is a manifestation of the severity of underlying septic process, which contributes to increased mortality. Correction of the underlying septic process is therefore paramount in correcting thrombo-cytopenia.
In conclusion, this is the first study of incidence, risk factors for development of thrombocytopenia and clinical outcome in thrombocytopenic patients in the subgroup of septic shock patients in the ICU. Thrombocytopenia is common in this subgroup and is associated with worse clinical outcome. Thus, thrombocytopenia at admission or at onset of septic shock in septic shock patients can be used as an early marker for risk stratification to identify patients at risk of complicated clinical course and increased mortality. Clinicians should be aggressive in diagnosing and treating the cause of thrombocytopenia in septic shock patients in view of their poor clinical outcome.
Accepted for publication on June 8, 2007.
(1.) Stephan F, Hollande J, Richard O, Heffi A, Maier-Redelsperger M, Flahault A. Thrombocytopenia in a surgical ICU. Chest 1999; 115:1363-1370.
(2.) Corrigan JJ Jr, Ray WL, May N. Changes in the blood coagulation system associated with septicemia. N Engl J Med 1968; 279:851-856.
(3.) Neame PB, Kelton JG, Walker IR, Stewart IO, Nossel HL, Hirsh J. Thrombocytopenia in septicemia: the role of disseminated intravascular coagulation. Blood 1980; 56:88-92.
(4.) Stephan F, Thioliere B, Verdy E, Tulliez M. Role of hemophagocytic histiocytosis in the etiology of thrombocytopenia in patients with sepsis syndrome or septic shock. Clin Infect Dis 1997; 25:1159-1164.
(5.) Francois B, Trimoreau F, Vignon P, Fixe P, Praloran V, Gastinne H. Thrombocytopenia in the sepsis syndrome: role of hemophagocytosis and macrophage colony-stimulating factor. Am J Med 1997; 103:114-120.
(6.) Poskitt TR, Poskitt PK. Thrombocytopenia of sepsis. The role of circulating IgG-containing immune complexes. Arch Intern Med 1985; 145:891-894.
(7.) Baughman RP, Lower EE, Flessa HC, Tollerud DJ. Thrombocytopenia in the intensive care unit. Chest 1993; 104:1243-1247.
(8.) Kim YL, Richman KA, Marshall BE. Thrombocytopenia associated with Swan-Ganz catheterization in patients. Anesthesiology 1980; 53:261-262.
(9.) Strauss R, Wehler M, Mehler K, Kreutzer D, Koebnick C, Hahn EG. Thrombocytopenia in patients in the medical intensive care unit: bleeding prevalence, transfusion requirements, and outcome. Crit Care Med 2002; 30:1765-1771.
(10.) Vanderschueren S, De Weerdt A, Malbrain M, Vankersschaever D, Frans E, Wilmer A et al. Thrombocytopenia and prognosis in intensive care. Crit Care Med 2000; 28:1871-1876.
(11.) Sprung CL, Peduzzi PN, Shatney CH, Schein RM, Wilson MF, Sheagren JN et al. Impact of encephalopathy on mortality in the sepsis syndrome. The Veterans Administration Systemic Sepsis Cooperative Study Group. Crit Care Med 1990; 18:801-806.
(12.) American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992; 20:864-874.
(13.) Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med 1985; 13:818-829.
(14.) Le Gall JR, Klar J, Lemeshow S, Saulnier F, Alberti C, Artigas A et al. Logistic Organ Dysfunction system. A new way to assess organ dysfunction in the intensive care unit. ICU Scoring Group. JAMA 1996; 276:802-810.
(15.) Vincent JL, de Mendonca A, Cantraine F, Moreno R, Takala J, Suter PM et al. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Working group on "sepsis-related problems" of the European Society of Intensive Care Medicine. Crit Care Med 1998; 26:1793-1800.
(16.) Bonfiglio MF, Traeger SM, Kier KL, Martin BR, Hulisz DT, Verbeck SR. Thrombocytopenia in intensive care patients: a comprehensive analysis of risk factors in 314 patients. Ann Pharmacother 1995; 29:835-842.
(17.) Hanes SD, Quarles DA, Boucher BA. Incidence and risk factors of thrombocytopenia in critically ill trauma patients. Ann Pharmacother 1997; 31:285-289.
(18.) Riedler GF, Straub PW, Frick PG. Thrombocytopenia in septicemia. A clinical study for the evaluation of its incidence and diagnostic value. Helv Med Acta 1971; 36:23-28.
(19.) Brun-Buisson C, Doyon F, Carlet J, Dellamonica P, Gouin F, Lepoutre A et al. Incidence, risk factors, and outcome of severe sepsis and septic shock in adults. A multicenter prospective study in intensive care units. French ICU Group for Severe Sepsis. JAMA 1995; 274:968-974.
(20.) Shalansky SJ, Verma AK, Levine M, Spinelli JJ, Dodek PM. Risk markers for thrombocytopenia in critically ill patients: a prospective analysis. Pharmacotherapy 2002; 22:803-813.
(21.) Drews RE. Critical issues in hematology: anemia, thrombocytopenia, coagulopathy, and blood product transfusions in critically ill patients. Clin Chest Med 2003; 24:607-622.
(22.) Kitchens CS. Disseminated intravascular coagulation. Curr Opin Hematol 1995; 2:402-406.
B. SHARMA *, M. SHARMA [[dagger]], M. MAJUMDER [[double dagger]], W. STEIER [[section]], A. SANGAL *, M. KALAWAR **
Departments of Pulmonary and Critical Care Medicine and Hematology-Oncology, Brookdale University Hospital and Medical Center, State University of New York, Brooklyn, New York, United States of America
* M.D., Senior Resident in Internal Medicine.
[[dagger]] M.D., Research Associate, Internal Medicine.
[[double dagger]] M.D., FCCP, Assistant Professor in Pulmonary and Critical Care Medicine.
[[section]] M.D., Associate Professor in Hematology and Oncology and Program Director, Hematology and Oncology Fellowship Program.
** M.D., Assistant Professor of Hematology-Oncology
Address for reprints: Dr B. Sharma, Department of Internal Medicine, Brookdale University Hospital and Medical Center, One Brookdale Plaza, Brooklyn, New York 11212, U.S.A.
TABLE 1 Characteristic Value Mean (95% confidence interval) Age, years 66.2 (62.5-69.9) APACHE II score 27.6 (26.2-29.1) LOD score 8.2 (7.7-8.7) SOFA score 11.1 (10.3-11.9) Platelet count, x[10.sup.9]/l 198.7 (162.2-235.3) Creatinine, mg/dl 2.7 (2.2-3.2) Lactate, mmol/l 4.1 (3.2-4.9) Albumin, g/dl 2.4 (2.3-2.6) Number of ICU days 15.1 (11.3-19) Number (%) Gender Male 34 (49.3) Female 35 (50.7) Source of sepsis Pneumonia 26 (37.7) Urinary tract infection 16 (23.2) Skin/decubiti 9 (13) Intra-abdominal 5 (7.3) Central nervous system 3 (4.4) Unknown 10 (14.5) Thrombocytopenia Total 38 (55) Mild 8 (11.6) Moderate 11 (15.9) Severe 6 (8.7) Very severe 13 (18.8) Mortality rate 38 (55) Characteristics of study patients (n=69) TABLE 2 Cause No. (%) Sepsis (all) 21 (55.3) Bacteraemic sepsis 12 (31.6) DIC 9 (23.7) Other medications 5 (13.2) Unknown cause 4 (10.5) Heparin-induced thrombocytopenia 0 (0) Multifactorial 26 (68.4) Causes of thrombocytopenia (number of thrombocytopenic patients=38). DIC=disseminated intravascular coagulation. TABLE 3 Variable Thrombocytopenia No thrombocytopenia (n=38) (n=31) Age, years 65.3 (60.4-70.2) 67.3 (61.5-73.1) APACHE II 28.7 (26.8-30.6) 26.3 (24.2-28.5) SOFA score 12.7 (11.6-13.8) 10.3 (9.2-11.5) LOD score 8.3 (7.7-9) 8.03 (7.3-8.8) Albumin, g/dl 2.4 (2.2-2.6) 2.4 (2.2-2.7) Bicarbonate, mEq/l 20.8 (18.6-23) 22.4 (20.1-24.6) Hydrocortisone, [micro]g/dl 18.6 (14.5-22.8) 17.2 (13.9-20.5) Creatinine, mg/dl 3.1 (2.3-3.9) 2.2 (1.5-2.8) Number of ICU days 14.2 (8.1-20.3) 16.3 (11.5-21.1) Lactate, mmol/l 4.3 (3.2-5.4) 3.8 (2.4-5.2) Noradrenaline dose 9.3 (8.1-10.3) 7.4 (6.2-8.7) ([micro]g/min) [P.sub.a][O.sub.2]/ 210.5 (169.3-253.4) 290.8 (251.3-334.6) Fi[O.sub.2] ratio Number of deaths 27 11 Variable P value Age, years 0.59 APACHE II 0.95 SOFA score 0.03 * LOD score 0.53 Albumin, g/dl 0.96 Bicarbonate, mEq/l 0.72 Hydrocortisone, [micro]g/dl 0.61 Creatinine, mg/dl 0.04 * Number of ICU days 0.59 Lactate, mmol/l 0.62 Noradrenaline dose 0.01 * ([micro]g/min) [P.sub.a][O.sub.2]/ 0.02 * Fi[O.sub.2] ratio Number of deaths 0.001 * Comparison of clinical data in patients with and without thrombocytopenia. All values are expressed as mean (95% confidence intervals). * significant P values. TABLE 4 Variable Odds ratio 95 % CI P value APACHE II score 0.95 0.83-1.09 0.45 LOD score 0.78 0.54-1.17 0.24 SOFA score 1.46 1.09-1.95 0.01 * Serum creatinine 1.2 0.91-1.59 0.19 Serum albumin 0.88 0.36-2.18 0.79 Vasopressor dose 1.21 0.96-1.56 0.02 * [P.sub.a][O.sub.2]/ 1.27 0.97-1.59 0.02 * Fi[O.sub.2] ratio Beta-lactam antibiotics 0.89 0.58-1.18 0.31 Heparin use 0.91 0.62-1.21 0.53 Frusemide 1.12 0.93-1.52 0.23 Multivariate logistic regression analysis model analysing various risk factors for development of thrombocytopenia in septic shock. * significant P values.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Original Papers|
|Author:||Sharma, B.; Sharma, M.; Majumder, M.; Steier, W.; Sangal, A.; Kalawar, M.|
|Publication:||Anaesthesia and Intensive Care|
|Article Type:||Clinical report|
|Date:||Dec 1, 2007|
|Previous Article:||Methods of endotracheal tube placement in patients undergoing pelviscopic surgery.|
|Next Article:||A comparison of 4% succinylated gelatin solution versus normal saline in stable normovolaemic sheep: global haemodynamic, regional blood flow and...|
|Pre-existing cardiac disease, troponin I elevation and mortality in patients with severe sepsis and septic shock.|