Introduction

Patients who sustain a solid organ injury after blunt trauma present many clinical management challenges. Of these, the timing of venous thromboembolism (VTE) prophylaxis initiation remains a controversial issue in contemporary clinical practice. Blunt intra-abdominal solid organ injuries (liver, spleen, and kidney) are frequently managed nonoperatively and therefore carry a risk of bleeding. This may be exacerbated by VTE prophylaxis and could contribute to failure of nonoperative management. Conflictingly, blunt trauma patients are also at high risk for VTE and may benefit from prompt initiation of chemoprophylaxis [1]. The optimal timing for VTE prophylaxis initiation among patients with blunt solid organ injury is not well defined by the current literature, although a small number of retrospective studies suggest that initiation within 48 h may be safe [2,3,4,5]. Prospective validation of these retrospective findings is required.

The primary objective of this study was to prospectively determine the optimal timing of VTE prophylaxis initiation among patients with blunt solid organ injury managed nonoperatively. Our hypothesis was that initiation within 48 h would result in a lower rate of VTE without an increased risk of bleeding or failure of nonoperative management.

Materials and methods

In this prospective observational study, all adult (>15 years) blunt trauma patients who presented to LAC + USC Medical Center between December 1, 2016, and November 30, 2017, and sustained a solid organ injury (liver, spleen, and/or kidney) managed nonoperatively were screened for inclusion. Patients were excluded if they were transferred from an outside hospital, died in the emergency department (ED), had a pre-existing bleeding disorder, were on home antiplatelet or anticoagulation medication, or if they received no VTE prophylaxis during their hospital admission. Eligible patients were identified and enrolled on admission. Nonoperative management was defined by a documented plan in the ED by the trauma team for nonoperative management and the lack of laparotomy within 4 h of admission. Failure of nonoperative management was defined by exploratory laparotomy ≥4 h after admission. Angioembolization was not an exclusion criterion. Institutional Review Board approval was obtained from the University of Southern California. Due to the observational nature of the study, consent was waived.

Patient demographics (age, gender), injury data [type and American Association for the Surgery of Trauma (AAST) grade of solid organ injury, abbreviated injury scale (AIS) score by body region, injury severity score (ISS)], and clinical data [vital signs and Glasgow coma scale (GCS) score on presentation] were recorded. The type of VTE prophylaxis, dosing schedule, and timing of the first dose, measured to the nearest minute, were recorded. The primary outcome was VTE event rate [deep vein thrombosis (DVT) and pulmonary embolism (PE)]. Secondary outcomes included hospital length of stay (LOS), intensive care unit (ICU) LOS, need for and volume of post-prophylaxis blood transfusion, need for delayed (post-prophylaxis) interventional radiology (IR) or operative intervention, failure of nonoperative management, and mortality. All patients without contraindication (e.g., lower extremity fracture) received sequential compression devices to bilateral lower extremities until ambulation. We do not routinely screen for VTE at LAC + USC. Symptomatic patients with DVT were diagnosed with duplex ultrasonography, and those with PE were diagnosed with computed tomographic pulmonary angiography (CTPA).

The decision to initiate VTE prophylaxis was at the discretion of the attending trauma surgeon. Patients were dichotomized into study groups based on the timing of the first dose of VTE prophylaxis from the time of admission to hospital: early prophylaxis (≤48 h) and late prophylaxis (>48 h). Univariate analysis was used to compare baseline characteristics and outcomes between groups. Continuous variables were compared using the Mann–Whitney U test and are presented as median [interquartile range (IQR)]. Categorical variables were compared using Fisher’s exact test and are presented as number (%). Multivariate analysis with logistic regression was performed to determine independent risk factors for late prophylaxis initiation. The covariates for the logistic regression were selected a priori and included gender, traumatic brain injury (TBI), lower extremity fractures, pelvic fractures, ICU LOS, and ISS. Although obesity is a known risk factor for VTE, the number of obese patients in this study (n = 2) was too low to include obesity as a covariate. A receiver operating characteristic (ROC) curve was performed to determine the time-to-prophylaxis cutoff point. Statistical significance was defined as p < 0.05.

Results

Patient demographics and clinical data

Over the study period, 2191 patients presented to LAC + USC after blunt trauma (Fig. 1). Of these, 238 (11%) sustained a solid organ injury. After excluding those who were managed operatively (n = 49), died in the ED (n = 8), transferred from an outside hospital (n = 15), had a pre-existing bleeding disorder (n = 5), those on home antiplatelet or anticoagulation medications (n = 3), and those who received no VTE prophylaxis during hospital admission (n = 40), 118 patients remained for analysis.

Fig. 1
figure 1

Flow of patients through study. Adult, defined as age > 15 years. ED emergency department, OSH outside hospital. Early prophylaxis ≤48 h after hospital admission. Late prophylaxis >48 h after hospital admission

Full size image

The median age was 36 years [IQR 27–55], and 66% of patients (n = 78) were male (Table 1). Few patients were hypotensive (SBP < 90 mmHg) (n = 13, 11%) or tachycardic (HR > 120 bpm) (n = 17, 14%) on admission. Median GCS was 15 [IQR 14–15]. Median ISS was 22 [IQR 14–26]. In terms of solid organ injuries, liver injuries occurred in 57 patients (48%), splenic injuries in 43 (36%), and kidney injuries in 34 (29%). Overall, 16% of patients (n = 19) had more than one solid organ injury. Associated injuries were common (Table 1).

Table 1 Patient demographics, clinical data, and injury data
Full size table

Early versus late VTE prophylaxis initiation

Of the 118 study patients, 61 (52%) received early prophylaxis and 57 (48%) received late prophylaxis. Patient demographics, admission vital signs, GCS, AIS in each body region, solid organ injury type and AAST grade, pelvic fractures, and lower extremity fractures did not vary between groups (p > 0.05) (Tables 1, 2). There were significantly more TBIs in the late prophylaxis group (n = 18, 32% vs n = 5, 8%, p = 0.002). Need for angioembolization did not vary between groups (n = 10, 16% vs n = 14, 14%, p = 0.821).

Table 2 AAST grade of solid organ injuries
Full size table

Patients in the early prophylaxis group were started on prophylaxis at a median of 2105 min [IQR 1830–2512], or approximately 35 h after admission. Patients in the late prophylaxis group were initiated on prophylaxis at a median of 4320 min [IQR 3420–5760], or approximately 72 h, after admission. Prophylaxis was delivered as enoxaparin or heparin. The majority of patients received enoxaparin, with similar rates of heparin use in both groups (n = 5, 8% vs n = 6, 11%, p = 0.667).

The solid organ injuries sustained by both groups are delineated in Table 2. The median AAST grade of solid organ injury did not vary between groups (p > 0.05). The type of solid organ injured (liver, spleen, or kidney) did not vary between groups (p > 0.05). There was no difference between groups in the number of patients with >1 solid organ injured (p = 0.323) (Table 1).

Outcomes

Overall, 10 VTEs occurred in 8 patients. There were 3 patients (3%) with isolated DVTs, 3 (3%) with isolated PEs, and 2 (2%) with concomitant DVTs and PEs, both of which occurred in the late prophylaxis group. Univariate analysis of outcomes revealed a lower rate of DVT among early prophylaxis patients (n = 0, 0% vs n = 5, 9%, p = 0.024). There was no difference between early and late prophylaxis groups in VTE rate (n = 2, 3% vs n = 6, 11%, p = 0.153) or PE rate (n = 2, 3% vs n = 3, 5%, p = 0.672) (Table 3). No patient in either study group failed nonoperative management. No patient in either group required angioembolization or operative intervention after initiation of VTE prophylaxis. There was no difference in the volume of post-prophylaxis blood transfusion between groups (median 0 [IQR 0–0] vs 0 [IQR 0–0], p = 0.180). There was no difference in mortality (n = 2, 3% vs n = 1, 2%, p = 1.000). Hospital and ICU LOS were significantly longer in the late prophylaxis group (14 days [7–35] vs 6 days [4–11], p < 0.001; 7 days [4–12] vs 3 days [2–6], p < 0.001) (Table 3). There was no difference between groups in need for ICU admission (n = 52, 85% vs n = 52, 91%, p = 0.398).

Table 3 Univariate analysis of outcomes
Full size table

Multivariate analysis with logistic regression was performed to examine independent predictors of late VTE prophylaxis initiation (Table 4). TBI was significantly associated with late prophylaxis initiation (adjusted OR 0.22, 95% CI 0.07–0.74, p = 0.015).

Table 4 Multivariate analysis of risk factors for late (>48 h) initiation of VTE prophylaxis
Full size table

ROC curve analysis demonstrated that a time-to-prophylaxis cutoff of 56 h had a sensitivity of 1.00 and specificity of 0.65 for the development of VTE (AUROC 0.825) (Fig. 2).

Fig. 2
figure 2

Receiver operating characteristic (ROC) curve for time-to-prophylaxis cutoff point. Threshold, 56 h. Sensitivity 1.00, specificity 0.65. AUROC 0.825

Full size image

Discussion

Care of the injured patient involves careful consideration of the risks and benefits of intervention. One major consideration surrounds the optimal timing of VTE prophylaxis initiation. Trauma patients are known to be at high risk for VTEs, with rates that approach 15% [6]. This is significant since VTEs are an important cause of morbidity and mortality among patients who survive beyond the first 72 h [7, 8]. Pharmacologic VTE prophylaxis exists and is effective at decreasing the rates of VTE. One study demonstrated that a delay in VTE prophylaxis initiation beyond 4 days resulted in a threefold increase in the risk of VTE [1]. However, patients with nonoperatively managed blunt solid organ injury present a competing concern for the possible precipitation of bleeding after VTE prophylaxis administration. With the rise in rates of nonoperative management over the past decade, determination of the appropriate time to initiate VTE prophylaxis in these high risk patients is of paramount importance. Major trauma society guidelines indicate that pharmacologic prophylaxis can be initiated among patients with blunt solid organ injury without increasing the rate of failure of nonoperative management [9, 10]. However, because only a small number of retrospective studies have examined this issue thus far [2,3,4,5], these guidelines are unable to make an evidence-based recommendation about the optimal timing for initiation of therapy. One excellent thromboelastography (TEG)-based study demonstrated that patients with blunt solid organ injury transition into a hypercoagulable state at approximately 48 h post injury, suggesting that VTE prophylaxis initiation before that time may be wise [11].

In this study, we prospectively examined patients who presented to our Level I trauma center over a 1-year period with a blunt solid organ injury managed nonoperatively. We found that initiation of VTE prophylaxis within 48 h of admission resulted in a significantly lower rate of DVT among blunt trauma patients with solid organ injury. Based on ROC curve analysis, the risk of VTE appeared to increase most notably when prophylaxis was not initiated in the first 56 h. Importantly, no patient required operative intervention or angioembolization for bleeding after initiation of prophylaxis. There was no difference in the volume of post-prophylaxis transfusion between patients in the early and late prophylaxis groups. Our findings prospectively validate previous retrospective studies on VTE prophylaxis among patients with solid organ injury which suggested that prophylaxis can safely be initiated within 48 h without increasing the need for post-prophylaxis transfusion or failure rates of nonoperative management [2,3,4,5].

Therefore, in the absence of contraindications related to associated injuries (for example, expanding intracranial hemorrhage), the best available evidence supports that VTE prophylaxis can safely be initiated within 48 h of admission for blunt solid organ injury, regardless of injury grade, without concern for exacerbation of bleeding or failure of nonoperative management.

Approximately half of our patients were initiated on VTE prophylaxis >48 h after hospital admission. The higher incidence of TBI among the late prophylaxis group provides a plausible explanation for the delay in VTE prophylaxis initiation in these patients. In fact, after controlling for potential confounders, the presence of TBI was independently associated with late initiation of VTE prophylaxis. Although there was no institutional policy at the time of this study for VTE prophylaxis among TBI patients, many clinicians withhold prophylaxis for a period of time after an interval CT scan of the head demonstrates stability of the intracranial bleeding. The greater incidence of concomitant TBI in the late prophylaxis group is a potential confounding factor which must be considered in the interpretation of these results.

The limitations to this study must be acknowledged. First, it is a single-center study. This limits study size and event detection rates. The possibility that the study may have been underpowered, and thus that a type II error may exist, must therefore be considered. Second, we do not routinely screen for DVTs at our institution. Previous studies have shown that DVT surveillance increases the number of diagnosed DVTs [12]. It is possible that early prophylaxis has a different effect on symptomatic and asymptomatic DVTs, and this has not been addressed by the present study. Third, this study captured few patients with grade IV–V injuries, likely reflecting an increased need for operative management among these patients. The paucity of high-grade injuries in this cohort likely contributed to the finding that no patient failed nonoperative management of a blunt solid organ injury. The small number of high-grade injuries limits the generalizability of our findings, and caution must therefore be used when applying the results of this study to grade IV and V injuries. Fourth, although we defined early prophylaxis as ≤48 h on the basis of existing retrospective [2,3,4,5] and TEG-based [11] literature, the optimal time to initiate prophylaxis may be early within this window. We also did not capture missed doses of VTE prophylaxis, which may confound the data. Further study will be required to determine the ideal initiation time more precisely and to define the impact of missed doses of VTE prophylaxis. Fifth, this study captured only VTEs that were diagnosed in hospital. Because the risk of VTE after trauma is at its highest for 3 months post-injury [13], it is possible that the incidence of VTE is underestimated by this study. Next, the generalizability of our data to other centers with different patient demographics is unclear. Our extremely low obesity rate, for example, is unlikely to be consistent across centers. Finally, because the time of VTE prophylaxis initiation was at the discretion of the attending trauma surgeon, there may be selection bias between the early and late prophylaxis groups wherein patients with injuries at lower risk for bleeding were initiated on prophylaxis earlier. Although an attempt was made to control for potential confounders, it is impossible to account for all such confounders. We must acknowledge that there may be differences in the study groups that we are unaware of. These limitations can be addressed in the future with a multicenter study including institutions with a variety of practice patterns.

In conclusion, this study provides a prospective analysis of the timing of VTE prophylaxis initiation among blunt trauma patients with solid organ injury managed nonoperatively. After adjusting for potential confounders, we found that commencement of VTE prophylaxis within 48 h of admission was associated with significantly fewer VTEs. Importantly, early prophylaxis appeared to be safe across AAST injury grade and type of solid organ injury (liver, spleen, and/or kidney), without an increased risk of bleeding that necessitated intervention or blood transfusion. In clinical practice, this has encouraged earlier initiation of VTE prophylaxis among these patients at our institution. We recommend initiation of VTE prophylaxis within 48 h of hospital admission among patients with blunt solid organ injury without contraindication to prophylaxis.