Evaluation of Heparin Prophylaxis Protocol on Deep Venous Thrombosis and Pulmonary Embolism in Traumatic Brain Injury

There is currently no accepted standard for deep venous thrombosis (DVT) and pulmonary embolism (PE) prophylaxis in patients with traumatic brain injury (TBI). The objective of our study was to evaluate the effects of implementing a subcutaneous heparin prophylaxis protocol for patients with TBI that began in our hospital as of June 2009. In our retrospective cohort study, we examined 3812 TBI records between January 2007 and December 2011. A significant reduction in the risk of DVT/PE development was not demonstrated by comparing DVTand PE incidences before and after protocol implementation. A clear trend between heparin use and DVToccurrence could not be determined from a review of TBI records after June 2009. The use of heparin after initiation of our protocol among operative TBI cases without intracranial hemorrhage (ICH) based on admission head computed tomography was 58 per cent. ICH complication from heparin prophylaxis was 10.6 per cent for patients with TBI with ICH on admission (five of 47 cases) compared with 0.7 per cent for those without ICH on admission (four of 535 cases).

P ULMONARY EMBOLISM (PE) rarely occurs in the general population but may be fatal if not managed correctly.1 The risk of PE increases dramatically for hospitalized neurosurgical patients.2 Deep vein thrombosis (DVT) and PE are a major cause of morbidity and mortality among surgical patients, particularly those with traumatic brain injury (TBI).3 The American College of Chest Physicians' (AACP) most current guidelines on prevention of DVT/PE recommend heparin prophylaxis for major trauma patients at high risk for DVT/PE,4 although there are no existing standard recommendations for patients with TBI.

In this study, we examined the effects of implementation of a heparin prophylaxis protocol requiring 5000 units of subcutaneous heparin every 8 hours within 48 hours for patients with TBI without progression of clinical or radiological disease; patients with intracranial monitoring did not receive heparin within 48 hours. We compared the incidence of DVT and PE before and after implementation of the protocol and analyzed the relationship between DVT/PE occurrence and heparin use stratified by intracranial hemorrhage (ICH) status on admission and surgical status.

One of the main concerns for the use of heparin prophylaxis in TBI patients is the development of ICH complications.5 Although past studies have shown ICH complication rates up to 12 per cent from heparin prophylaxis,6 the reduction of DVT/PE incidence seems to outweigh the potential development of ICH.We also examined rate of ICH complications from heparin prophylaxis for patients with TBI after the implementation of the protocol.

Methods

We identified TBI patient records from our trauma registry between January 2007 and December 2011 for patients with a head Abbreviated Injury Severity scale (AIS) greater than 3. Brain AIS was chosen over Glasgow Coma Scale score because it was objectively scored by our Certified Trauma Registrars. Patients were pulled from the trauma registry and identified based on injury diagnostic codes (International Classification of Diseases, 9th Revision, Clinical Modification) and includes patients admitted to the hospital, died from traumatic injury, or transferred to another facility for higher level of care or per insurance request. DVT and PE cases were identified based on ultrasound and computed tomography (CT) diagnoses recorded in patient records. For the TBI cases after the implementation of heparin prophylaxis in June 2009, we further reviewed the individual records to identify ICH on admission based on head CT, ICH progression based on follow-up head CT, surgical status, and heparin use.

All statistical analyses were performed in STATA Version 12. The results were summarized in two-bytwo contingency tables separating patients into categories of before and after the implementation of the heparin prophylaxis and the incidence of DVT/PE. P values from two-tailed Pearson's x2 test were computed based on the entries in the contingency tables and were considered to be statistically significant if the values were < 0.05.

Results

As summarized in Table 1, a total of 3812 TBI cases were identified from the trauma registry between January 2007 and December 2011. There were 1970 TBI cases before June 2009 of which 19 cases of DVT and one case of PE were identified (Tables 1 and 2). After the implementation of the heparin prophylaxis protocol in June 2009, there were 22 cases of DVT and one case of PE from the 1842 TBI cases (Tables 1 and 2). This translated to a DVT risk of 0.97 per cent before the protocol and a DVT risk of 1.21 per cent after the protocol. The relative risk ratio for DVT (1.241, P 4 0.492) and PE (1.070, P 4 0.962) after the implementation of the heparin prophylaxis protocol was not statistically significant (Tables 1 and 2).

The 1842 TBI cases after June 2009 were further reviewed and it was determined that 1146 cases did not have ICH based on admission head CT. Of these 1146 cases, 330 patients underwent surgery, of which 190 received heparin prophylaxis within 48 hours. A stratified analysis (Fig. 1) showed 58 per cent of these patients received heparin after implementation of the heparin protocol. A time series plot with DVT incidence and heparin use in these 330 cases (Fig. 2) found no clear association between heparin use and DVT occurrence.

To determine the incidences of ICH complications, the 1842 TBI cases underwent a different stratified analysis, shown in Figure 3, beginning with separation of 1146 cases without ICH and 565 cases with ICH as determined by the admission head CT. One hundred thirty-seven TBI cases were excluded because no record of admission CT was found in the patient records. Of the 1146 TBI cases without ICH at admission, 535 cases had received heparin of which four cases (0.7%) developed ICH as determined by subsequent head CT. For the 565 TBI cases with ICH at admission, 47 cases had received heparin within 48 hours of admission of which five cases (10.6%) had increased bleeding as determined by subsequent head CT. Of note, most TBI cases that developed DVT (n 4 17 [77%]) received heparin within 48 hours versus those (n 4 5 [23%]) that did not.

Discussion

The heparin prophylaxis protocol was not demonstrated to have a significant effect on reducing risk of DVT and PE among patients with TBI. Multiple factors may contribute to this finding: 1) low incidence of DVT/PE before and after protocol implementation; 2) poor protocol compliance; 3) differences in dose, type, and timing of DVT prophylaxis before and after protocol implementation; 4) patients who died, were discharged, or transferred before 48 hours; and 5) differences in severity of coexisting injuries before and after protocol initiation. Our hospital follows ACCP guidelines4 and does not use surveillance ultrasound routinely; diagnosis is confirmed with ultrasound and/ or spiral chest CT initiated by clinical suspicion.

An implicit assumption in our retrospective cohort study design is compliance to the heparin prophylaxis protocol after its implementation. The determination of actual compliance rate proved to be difficult as a result of insufficient data in the electronic medical records. Based on a proxy of surgical TBI cases without ICH on admission, a 58 per cent heparin use was shown, which appears to reflect a fair compliance rate given that the protocol excludes patients with ICH progression, intracranial monitoring, or other bleeding risk. Regardless, plotting this information did not show a clear trend between noncompliance and DVT occurrences (Fig. 2).

The dose, timing, and type of DVT prophylaxis used before 2009 could also explain the lack of significance in our results. Heparin was the only drug accepted by all members of our trauma committee when this protocol was developed; before this time, different drugs were used with different standard doses at the discretion of the clinician, which may or may not be more effective than the 5000-units three-times-a-day dosing of heparin. The effect of prophylaxis on the results outside the use of heparin should be small, because our neurosurgeons were adamant that low-molecularweight heparin or factor X inhibitors were to be avoided. It has been suggested that shortening thewindow of heparin prophylaxis initiation from 48 to 24 hours may increase the effect size.7 A future prospective comparison is needed to determine drug efficacy in patients with TBI.

Loss to follow-up could be another cause for the lack of significance in our results, because our data included patients who died, were transferred, or were discharged. However, there is no known dramatic change of patient demographic that would cause the rate of loss to follow-up to differ before and after June 2009. Similarly, differences in the severity of coexisting injuries can potentially influence the results. We defined our populations based on brain AIS and did not collect the overall Injury Severity Score (ISS). In reviewing our trauma database, there was no significant difference in the ISS score of our patients sustaining blunt trauma between the two periods. Therefore, the effect from loss to follow-up and severity of coexisting injuries should affect both populations before and after implementation of the heparin prophylaxis protocol equally.

We also observed that severe TBI may be protective for DVT, because the incidence of DVT is lower among those patients who did not receive prophylaxis. This is especially true in our protocol, in which prophylaxis is held if there was no evidence of clinical or radiological brain injury progression, selecting out the most severe patients. One explanation may be the result of early coagulopathy seen in patients with major TBI secondary to release of tissue thromboplastin.8 Stratification of head injury with initial coagulopathy should be included in future studies to clarify this.

Based on our chart review, we were able to quantify the risk of ICH complications. Both the incidence of ICH development and increased ICH were consistent with prior findings5 and match the natural progression of ICH in the published literature. There may be significant underreporting of the incidence with patients who have intraventricular monitoring, because it precluded some patients from receiving a routine followup head CT, although this represents a small subset. The lack of more detailed information from cases before June 2009 makes the determination of relative risks for ICH complications difficult.

Conclusions

A significant reduction in DVT/PE incidences after implementation of heparin prophylaxis protocol cannot be demonstrated by the current study at our hospital. Severe brain injury is a heterogeneous group. Severe brain injury defined by AIS greater than 3 without ICH has a very low complication rate with the use of heparin prophylaxis as opposed to the subset with ICH on admission. There has not been a significantly powered study in TBI to define the risk/benefit ratio of prophylaxis in the ICH group. In addition, this study suggests that the subset of patients with brain injury who did not receive heparin prophylaxis, likely as a result of existing coagulopathy and progression of ICH, may in fact have a DVT protective effect.

REFERENCES

1. Dalen JE, Paraskos JA, Ockene IS, et al. Venous thromboembolism. Scope of the problem. Chest 1986;89(suppl 5): 370S-3S.

2. Danish SF, Burnett MG, Stein SC. Prophylaxis for deep venous thrombosis in patients with craniotomies: a review. Neurosurg Focus 2004;17:E2.

3. Hamilton MG, Hull RD, Pineo GF. Venous thromboembolism in neurosurgery and neurology patients: a review. Neurosurgery 1994;34:280-96.

4. Gould MK, Garcia DA, Wren SM, et al. Prevention of VTE in nonorthopedic surgical patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(suppl):e227S-77S.

5. Browd SR, Ragel BT, Davis GE, et al. Prophylaxis for deep venous thrombosis in neurosurgery: a review of the literature. Neurosurg Focus 2004;17:E1.

6. Patel NY, Hoyt DB, Nakaji P, et al. Traumatic brain injury: patterns of failure of nonoperative management. J Trauma 2000; 48:367-75.

7. Norwood SH, McAuley CE, Berne JD, et al. Prospective evaluation of the safety of enoxaparin prophylaxis for venous thromboembolism in patients with intracranial hemorrhagic injuries. Arch Surg 2002;137:696-702.

8. Pathak A, Dutta S, Marwaha N, et al. Change in tissue thromboplastin content of brain following trauma. Neurol India 2005;53:178-82.

MAYIN LIN, D.O., M.P.H.,*[dagger] JOSEPH VIVIAN DAVIS, D.O.,[dagger] DAVID T. WONG, M.D.[dagger]

From *Western University of Health Sciences and the [dagger]Department of General Surgery, Arrowhead Regional Medical Center, Colton, California

Presented at the 24th Annual Scientific Meeting of the Southern California Chapter of the American College of Surgeons, January 18-20, 2013, in Santa Barbara, California.

Address correspondence and reprint requests to David T. Wong, M.D., Department of General Surgery, Arrowhead Regional Medical Center, 2285 Cartier Circle, Colton, CA 92324. E-mail: [email protected].