Open Collections

A prospective observational study of risk indicators associated with the development of thrombocytopenia in a community based ICU/CCU

University of British Columbia

INTRODUCTION: Thrombocytopenia is a common complication in .critically ill patients and can present a challenging clinical problem. Its incidence has been reported to range between 13% to 41% and is associated with an increased length of hospital stay and mortality. Information is needed to clarify risk indicators associated with the development of thrombocytopenia in critically ill patients and to improve clinical decision-making when addressing this common problem. OBJECTIVES: 1. To estimate the incidence of thrombocytopenia in a community based intensive and coronary care unit (ICU/CCU). 2. To compare the incidence of thrombocytopenia in ICU and CCU patients 3. To identify risk indicators associated with the development of thrombocytopenia in ICU/CCU patients using logistic regression modelling. 4. To compare clinical outcomes among patients who did and did not develop thrombocytopenia during their ICU/CCU stay. DESIGN: A prospective, observational, study. SETTING: The Intensive/Coronary Care Unit (ICU/CCU) at Lions Gate Hospital (LGH), which is a 350 bed community-based hospital in North Vancouver, British Columbia, Canada. PATIENTS: The target population for this study included all patients over the age of 18 years who had 2 or more platelet counts recorded, at least 12 hours apart, during an ICU/CCU admission. All patients were included unless they met any of the exclusion criteria, which included an admission platelet count < 150 x 109/L, repeat admission to the unit, concurrent involvement in another study, indication of hypersplenism, and particular disease states associated with the development of thrombocytopenia. METHODS: Data were obtained prospectively during each patient's ICU/CCU stay through daily review of the medical record, patient interviews, and discussion with the medical team. Most responsible diagnoses, clinical outcomes, and any missing data were obtained retrospectively from the patients' medical charts approximately six weeks after discharge from hospital. Thrombocytopenia was defined as two consecutive platelet counts < 150 x 10⁹/L at least 12 hours apart. Descriptive analysis was used to summarize baseline demographic characteristics of the study sample, as well as to select potential variables for logistic regression analysis. Univariate analyses identified variables (p < 0.25) potentially associated with thrombocytopenia, which were then subjected to multivariate backward stepwise logistic regression using (p[sub out] > 0.10 and p[sub in] < 0.05) to generate two different models. The first model was an admission or baseline model that identified risk indicators independently associated with thrombocytopenia upon admission to the ICU/CCU. The second was a model that included indicators present on admission and those that patients were exposed to in the ICU/CCU. RESULTS: Of the 362 patients who met the inclusion criteria, 68 (18.8%; 95% CI: 14.8% - 22.8%) developed thrombocytopenia during their ICU/CCU stay. Thrombocytopenia developed more often in patients with an ICU (29.7%; 95% CI: 22.9% - 36.5%) than CCU (8.9%; 95% CI: 4.9% - 12.9%) most responsible diagnosis. Baseline multivariate logistic regression analysis identified eight risk indicators independently associated with the development of thrombocytopenia: sepsis, gastrointestinal diagnosis, GI bleed diagnosis, respiratory non-surgery diagnosis, musculoskeletal/connective tissue diagnosis, age1, APACHE II Score, and admission platelet count1. The ICU/CCU model identified nine risk indicators independently associated with thrombocytopenia: sepsis, gastrointestinal diagnosis, respiratory non-surgery diagnosis, musculoskeletal/connective tissue diagnosis, packed red blood cell (PRBC) transfusion, fresh frozen plasma (FFP) transfusion, Swan-Ganz catheter insertion, acetylsalicylic acid (ASA)¹, and admission platelet count1. Exploratory analysis identified bleeding episodes as a possible risk indicator for thrombocytopenia. No medications, including heparin, were found to be associated with increased risk of developing thrombocytopenia following multivariate logistic regression analysis. Clinicians discontinued heparin in 18% of the patients who developed thrombocytopenia, apparently due to concern regarding HIT. Mean length of ICU/CCU and hospital stays, and mortality were greater among patients who developed thrombocytopenia. CONCLUSIONS: Thrombocytopenia developed in approximately 19% of patients admitted to a community based ICU/CCU. Indicators associated with an increased risk for thrombocytopenia included markers for severity of illness (e.g. sepsis, APACHE II score, or respiratory non-surgery diagnosis), foreign surfaces (e.g. Swan-Ganz catheters), and, based on an exploratory finding, episodes of bleeding. The identified risk indicators should be considered when treatment decisions are made in critically ill thrombocytopenic patients.

Thesis/Dissertation

application/pdf

2009-07-20

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/11003

Pharmaceutical Sciences

University of British Columbia

UBCV

DSpace