Louis SG, Sato M, Geraci T, Anderson R, Cho SD, Van PY, Barton JS, Riha GM, Underwood S, Differding J, Watters JM, Schreiber MA. Correlation of Missed Doses of Enoxaparin With Increased Incidence of Deep Vein Thrombosis in Trauma and General Surgery Patients. JAMA Surg. 2014;149(4):365-370. doi:10.1001/jamasurg.2013.3963
Copyright 2014 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
Enoxaparin sodium is widely used for deep vein thrombosis (DVT) prophylaxis, yet DVT rates remain high in the trauma and general surgery populations. Missed doses during hospitalization are common.
To determine if missed doses of enoxaparin correlate with DVT formation.
Design, Setting, and Participants
Data were prospectively collected among 202 trauma and general surgery patients admitted to a level I trauma center.
Main Outcomes and Measures
Deep vein thrombosis screening was performed using a rigorous standardized protocol.
The overall incidence of DVT was 15.8%. In total, 58.9% of patients missed at least 1 dose of enoxaparin. The DVTs occurred in 23.5% of patients who missed at least 1 dose and in 4.8% of patients who did not (P < .01). On univariate analysis, the need for mechanical ventilation (71.8% vs 44.1%), the performance of more than 1 operation (59.3% vs 40.0%), and male sex (75% vs 56%) were associated with DVT formation (P < .05 for all). A bivariate logistic regression was then performed, which revealed age 50 years or older and interrupted enoxaparin therapy as the only independent risk factors for DVT formation. The DVT rate did not differ between trauma and general surgery populations or in patients receiving once-daily vs twice-daily dosing regimens.
Conclusions and Relevance
Interrupted enoxaparin therapy and age 50 years or older are associated with DVT formation among trauma and general surgery patients. Missed doses occur commonly and are the only identified risk factor for DVT that can be ameliorated by physicians. Efforts to minimize interrupted enoxaparin prophylaxis in patients at risk for DVT should be optimized.
Deep vein thrombosis (DVT) and venous thromboembolism (VTE) in critically ill or injured patients remain a major concern.1,2 Trauma and general surgery patients have a multitude of factors contributing to VTE, including venous stasis, endothelial injury, impaired fibrinolysis, and decreased levels of serum anticoagulants. Despite standard prophylaxis, the incidence of pulmonary embolus after trauma has been reported to be as high as 2% to 22%.3,4 It is the third most common cause of death among patients with major trauma surviving the first 24 hours and is the second most common medical complication among postoperative patients in the United States.5,6 The incidence of DVT is estimated to be as high as 60% among trauma patients not receiving prophylactic anticoagulation.6 In addition, VTE represents a significant cause of preventable health care expenditure, with estimated inpatient management costs of up to $17 000.7
Low-molecular-weight heparin (LMWH) has emerged as a safe and effective modality for the prevention of VTE. In patients undergoing abdominal and pelvic surgery, LMWH has been shown to reduce the risk of symptomatic VTE by 80%.8 Although little high-quality evidence exists in the trauma population, LMWH therapy has been estimated to prevent 4 to 10 times as many nonfatal VTE events as nonfatal bleeding complications in these patients.9 The LMWH enoxaparin sodium has been extensively studied and has been proven effective in the prevention of DVT.10
During hospitalization, intermittent interruption of enoxaparin therapy is common. Patients miss doses for various reasons, including preoperative interruption, patient refusal, and nursing errors. Interruption of therapy has recently been shown to correlate with increased incidence of DVT in patients with traumatic brain injury.11 To date, this has not been studied in the larger trauma and general surgery populations. We hypothesized that an increase in DVT incidence would be seen among trauma and general surgery patients with interrupted enoxaparin dosing during hospitalization.
This study was approved by the institutional review board at Oregon Health & Science University. The level I trauma center at the university abides by the current federal Health Insurance Portability and Accountability Act guidelines. Informed written consent to participate was obtained from each patient or from a legal representative.
Quiz Ref IDPatients admitted to the trauma center at Oregon Health & Science University who were ordered to receive prophylactic enoxaparin were eligible for enrollment. Patients received at least 1 dose of prophylactic enoxaparin to be considered for the study. All prophylactic dosing regimens were included, including once-daily, twice-daily, and dosing adjusted for renal impairment and obesity. Patients receiving therapeutic enoxaparin for nonthrombotic indications were eligible for enrollment. Patients were excluded for any known thromboembolic diagnoses or for the use of anticoagulants other than enoxaparin.
Patient characteristics were recorded, including age, sex, weight, body mass index (BMI), and ventilator support. Also collected were data on injuries, operations, significant comorbidities, trauma or surgical intensive care unit status, and the Acute Physiology and Chronic Health Evaluation II (APACHE II) score. Recorded were the time of blood draws, the time and dose of enoxaparin administration, and the type and location of intravenous access used to obtain blood.
Quiz Ref IDBilateral whole-leg duplex ultrasonography for DVT surveillance was performed per a rigorous institutional protocol, which included weekly ultrasonography for patients admitted to the intensive care unit and for patients admitted to the trauma ward. General surgery patients on the acute care ward were screened when DVT was suspected by the primary treatment team. Upper extremity and neck ultrasonography was performed on an as-needed basis when DVT was suspected. Computed tomographic angiography was performed when pulmonary embolus was suspected.
A database was maintained using a spreadsheet (Microsoft Excel; Microsoft Corporation). Statistical analyses were performed using a software program (SPSS 19; SPSS, Inc). Continuous data were analyzed using t test or Mann-Whitney test where appropriate. Categorical data were analyzed using χ2 test or Fisher exact test where appropriate. Logistic regression was performed in a standard bivariate fashion. Significance was defined as P < .05.
Data from 202 trauma and general surgery patients were collected during a 4-year period from February 2007 through September 2012. The mean (SD) values for the cohort were 56 (1) years for age, 14 (1) for the APACHE II score, and 33.7 (0.8) for BMI (calculated as weight in kilograms divided by height in meters squared). There were 73 trauma patients and 129 general surgery patients, and 58.9% were men. A review of the duplex ultrasonography examinations revealed a DVT rate of 15.8%. No significant differences were observed in age, BMI, or APACHE II score between the patients with vs without DVT. Men were more likely than women to develop DVT (Table 1). No pulmonary emboli were diagnosed among the cohort during the study period.
Patients who developed DVT and patients who did not were divided into 2 groups. Both groups had a median duration to initiation of enoxaparin therapy of 2 days. As summarized in Table 1, the group diagnosed as having DVT had a longer total hospital length of stay and a longer duration of enoxaparin therapy.
Quiz Ref IDAt least 1 enoxaparin dose was missed in 58.9% of patients. Most of those missed 4 or fewer doses; however, there was significant variability in the number of doses missed. Twenty-eight patients missed 1 dose, 17 missed 2 doses, 19 missed 3 doses, 10 missed 4 doses, 4 missed 5 doses, 5 missed 6 doses, 4 missed 7 doses, 2 missed 8 doses, 5 missed 9 doses, 3 missed 10 doses, and 22 missed more than 10 doses. The most common reason indicated for interruption of therapy was for a pending invasive procedure (Table 2). Trauma patients were more likely to have doses withheld because of concern for bleeding, but all other reasons for interruption of therapy were similar for trauma and general surgical patients. Patients who had uninterrupted enoxaparin therapy were compared with those who had therapy interrupted at least once. Quiz Ref IDA 4.8% DVT rate was observed in the uninterrupted therapy group, and a 23.5% DVT rate was observed in the interrupted therapy group (P < .01). Patients not diagnosed as having DVT were much less likely to miss doses than those diagnosed as having DVT (50.8% vs 88.2%) (P > .001). No differences were observed between patients with uninterrupted therapy and those with missed doses in measured coagulation variables, including the mean platelet count, fibrinogen level, partial thromboplastin time, and international normalized ratio (Table 3).
A univariate logistic regression was performed to assess the effect of increasing number of missed enoxaparin doses on the development of DVT. As shown in Table 4, missing 1 dose did not significantly alter the likelihood of DVT. However, a direct correlation was observed between the number of missed doses and the risk of DVT starting at 2 doses missed. After 18 doses missed, this relationship ceased to exist, likely because of the small sample size.
A total of 32 DVTs were diagnosed during the study. There were 27 DVTs in the lower extremity and 5 DVTs in the neck or upper extremity. In total, 92.8% of DVTs diagnosed in the interrupted therapy group were in the lower extremity, while 25.0% of DVTs diagnosed in the uninterrupted therapy group were in the lower extremity. Three of 4 DVTs diagnosed in patients with uninterrupted therapy were in the upper extremity or neck. All neck and upper extremity DVTs were associated with the use of indwelling central venous catheters or percutaneously inserted central catheters.
Additional univariate analyses of other common risk factors for the development of DVT were performed and are summarized in Table 5. Patients who required ventilator support outside of the operating room or who underwent more than 1 operation during their hospitalization had a higher incidence of DVT formation. Obesity (BMI >30), epidural catheter use, and renal failure (creatinine clearance <30 mL/min/1.73 m2) were not associated with increased DVT formation (to convert creatinine clearance to milliliters per second per meter squared, multiply by 0.0167).
Subgroup analysis was performed, separating trauma and general surgery patients; the results are summarized in Table 6. The groups had similar age distributions and obesity rates. Trauma patients were more likely to undergo more than 1 operation during their hospitalization and were more likely to be younger than 50 years than general surgery patients. Deep vein thrombosis occurred at a similar rate in the 2 groups, as did missed doses.
A bivariate logistic regression was then performed, including commonly accepted risk factors for DVT and those we found to be associated with DVT formation on univariate analysis (Table 7). Quiz Ref IDOnly interrupted enoxaparin therapy and age 50 years or older were independent risk factors for DVT. Trauma, male sex, obesity, multiple operations, and ventilator support were not found to be independent risk factors.
In this study, we hypothesized that interruption of pharmacologic prophylaxis after it was initiated would lead to an increase in the diagnosis of DVT. Interruption of prophylactic anticoagulation is common in patients admitted after major trauma and in patients undergoing major surgery. More than half of our patients had at least 1 dose missed during their hospitalization, with most of these patients missing more than 1 dose. When multivariate analysis was performed, interruption of therapy was a strong independent risk factor for the development of DVT.
Coagulopathy in patients following trauma and major surgery remains incompletely characterized. These patients have significant disruptions in the normal homeostatic balance of procoagulants and anticoagulants.12- 14 The introduction of pharmacologic anticoagulation further contributes to an already deranged system in ways that are only partially understood. It has been suggested that initiation and subsequent interruption of pharmacologic anticoagulation may exacerbate a hypercoagulable state following trauma.11,15 A recent study11 performed in patients with traumatic brain injury demonstrated higher VTE rates in patients who had interrupted therapy compared with patients who had prophylaxis withheld until all expected surgical procedures were completed.
Our data regarding catheter-associated DVT are also compelling. Of 4 DVTs diagnosed in the patients who underwent uninterrupted therapy, 3 were provoked by the use of an indwelling catheter (central venous catheter or percutaneously inserted central catheter). This leaves a single DVT that was not caused by a catheter when enoxaparin was not interrupted. The patient had undergone total hip arthroplasty and developed ipsilateral DVT.
The formation of DVT does not occur in a vacuum, and a significant body of literature is dedicated to attempts at isolating risk factors for DVT. Variable characteristics between patients with vs without DVT herein were similar to risk factors acknowledged by several large studies.16- 18 These included multiple operations, age 50 years or older, and ventilator support required outside of the operating room. In addition, the patients with DVT had a longer overall hospital length of stay and a longer duration of enoxaparin therapy. We also found that men had a higher likelihood of DVT diagnosis, which is likely due to a type I statistical error owing to our small sample size because this has not been found in previous larger studies.
Multivariate analysis was performed to attempt to isolate the effect of interruption of prophylaxis on DVT formation. We found age 50 years or older and interrupted enoxaparin therapy to be the only independent risk factors for DVT formation. This study argues against the hypothesis that it is not the missed doses themselves but rather the reasons behind the missed doses (eg, for the operating room) that lead to increased incidence of DVT.
Pending invasive procedures were the most common documented reason for withholding doses. However, we found that surprising numbers of doses were withheld without a specified cause. Other reasons for missed doses included patient refusal, epidural catheter removal, nursing and pharmacy errors, and the patient being absent from the room. Doses withheld for ambiguous reasons decreased during the study because implementation of the electronic medical record at our institution required reasons to be specified. Because of the significant findings of this study, missed doses of enoxaparin now represent a quality improvement measure at our hospital, and we have educated surgeons about the adverse consequences of missing doses. These interventions have already resulted in a reduced percentage of doses withheld.
Our study has several acknowledged limitations. First, while the small sample size suggests the possibility of a type I error, the study was adequately powered. Second, many patients received short courses of anticoagulation therapy. Although we attempted to enroll severely ill patients (as evidenced by the mean APACHE II score of 14), a significant number of patients underwent 5 days of therapy or less. Third, despite a rigorous screening protocol, there was variance in screening between the trauma patients and the general surgery patients on the ward. Trauma patients were screened on a weekly basis, while general surgery patients were screened based on symptoms. This could have introduced surveillance bias, as has been noted in other studies.11,19
Although it is generally agreed by clinicians that pharmacologic prophylaxis should begin as soon as possible based on the physician’s assessment of a patient’s bleeding risk, the decision to withhold a dose of enoxaparin is often undertaken without comprehensively assessing the consequences. While we do not recommend continuing anticoagulation during high-risk operations such as intracranial or spinal procedures, the decision to withhold prophylaxis before multiple low-risk operations should not be taken lightly. In addition, optimization of systems to eliminate unnecessary missed doses is imperative. Through nursing and patient education, doses missed because of nursing errors or patient refusal should be eradicated. Missed doses are the only identified risk factor for DVT that can be ameliorated by physicians, and efforts to minimize interrupted enoxaparin prophylaxis in patients at risk for DVT should be optimized.
Accepted for Publication: June 28, 2013.
Corresponding Author: Scott G. Louis, MD, Trauma Research Institute of Oregon, Oregon Health & Science University, 3181 SW Sam Jackson Park Rd, Mail Code L223A, Portland, OR 97239 (email@example.com).
Published Online: February 26, 2014. doi:10.1001/jamasurg.2013.3963.
Author Contributions: Dr Louis had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Cho, Underwood, Differding, Watters, Schreiber.
Acquisition of data: Louis, Sato, Geraci, Anderson, Cho, Van, Barton, Riha, Underwood, Differding, Watters.
Drafting of the manuscript: Louis.
Critical revision of the manuscript for important intellectual content: Schreiber.
Statistical analysis: Louis.
Administrative, technical, or material support: Louis, Sato, Geraci, Anderson, Cho, Van, Barton, Riha, Underwood, Differding, Watters.
Study supervision: Underwood, Differding, Schreiber.
Conflict of Interest Disclosures: None reported.
Previous Presentation: This study was presented at the Pacific Coast Surgical Association 84th Annual Meeting; February 17, 2013; Kauai, Hawaii.