M. J. Pommerening1, E. Rahbar1, K. M. Minei1, J. B. Holcomb1, M. A. Schreiber2, M. J. Cohen3, S. Underwood2, M. Nelson3, B. A. Cotton1 1University Of Texas Health Science Center At Houston,Houston, TX, USA 2Oregon Health And Science University,Portland, OR, USA 3University Of California – San Francisco,San Francisco, CA, USA
Introduction: Previous investigators have demonstrated that post-injury thrombocytosis is associated with an increase in thromboembolic (TE) risk. Increased rates of thrombocytosis have been found specifically in patients following splenectomy for trauma. We hypothesized that patients undergoing splenectomy (1) would demonstrate a more hypercoagulable profile during their hospital stay and (2) that this hypercoagulable state would be associated with increased TE events.
Methods: A 14-month, prospective, observational trial evaluating serial rapid thrombelastography (rTEG) was conducted at three ACS-verified, level-1 trauma centers. Inclusion criteria: highest-level trauma activation, arrival within 6 hours of injury, 18 years of age or older. Serial rTEG (ACT, k-time, α-angle, mA, LY-30) and traditional coagulation testing (PT, PTT, fibrinogen and platelet count) was obtained at 0, 3, 6, 12, 24, 48, 72, 96 and 120 hours. Thromboembolic complications were defined as in-hospital DVT, PE, acute MI, or ischemic stroke. Patients were stratified into splenectomy versus non-splenectomy cohorts. Univariate analysis was perfomed, followed by longitudinal analysis using an adjusted generalized estimating equations (GEE) to evaluate the effects of time, splenectomy, and group-time interactions on changes in rTEG and traditional coagulation testing. For TE risk, we employed a multiple logistic regression. Both models controlled for age, gender, injury severity, and admission blood pressure, base deficit, and hemoglobin.
Results: 1242 patients were enrolled. Of these, 605 patients had >24 hours of serial rTEG values post-admission and were analyzed (40 splenectomy, 565 non-splenectomy patients). Splenectomy patients were younger (median 30 vs. 38 years), more hypotensive (median systolic 100 vs. 130 mmHg) and more in shock on arrival (median base value -7 vs. -2); all p<0.001. While there was no difference in 24-hour (8 vs. 5%; p=0.348) or 30-day mortality (13 vs. 7%; p=0.129), splenectomy patients were more likely to develop TE events (17.5 vs. 7.5%; p=0.015). Logistic regression confirmed this risk, finding splenectomy was associated with an increased risk of TE events (OR 3.4, 95% C.I. 1.14-9.96, p=0.028). GEE modeling demonstrated that rTEG values of α-angle and mA are significantly higher (more hypercoagulable) in splenectomy patients at 48, 72, 96 and 120-hours; all p<0.05. The GEE model also demonstrated that platelet counts were significantly higher in splenectomy patients beginning at 72 hours and continuing through 120 hours; p<0.05.
Conclusion: This multicenter study demonstrates that patients undergoing splenectomy are more hypercoagulable than other trauma patients. This hypercoagulable state (identified through higher TEG α-angle and mA values) begins at approximately 48 hours post-injury and continues through at least day 5. Moreover, this hypercoagulable state is associated with 3-fold increased risk of TE complications.