Alterations of coagulation and fibrinolysis in patients with blunt splenic injury after splenic artery embolization.

BACKGROUND: Thrombotic complications following splenectomy have been documented. However, there has been sparse literature regarding thrombotic complications following splenic artery embolization (SAE).The objective of this study was to determine changes in coagulation and fibrinolysis and assess the thrombotic risk after SAE in patients with blunt splenic injury (BSI).
METHODS: This study included 38 BSI patients who were hemodynamically stable on admission. SAE was performed if the splenic injury was classed as grade III or greater and had no requirement of immediate surgery. Platelet (PLT), fibrinogen (FIB), D-dimers (D-D), fibrinogen/fibrin degradation products (FDP), antithrombin III (AT III), prothrombin time (PT), activated partial thromboplastin time (APTT), thrombin time (TT), hemoglobin (Hb), and hematocrit (Hct) were measured before SAE procedures and then 1d, 3d, and 7d after SAE.
RESULTS: The technical success rate of SAE and the splenic salvage rate were 100%. There was no mortality. Compared with pre-SAE values, the levels of PLT, FIB, D-D, and FDP increased significantly at 3 days and 7 days after SAE (p < 0.05). However, AT III, PT, APTT, TT, Hb, and Hct showed no statistically significant difference at 1d, 3d, and 7d after SAE (p > 0.05).
CONCLUSION: Alterations in PLT and hemostatic parameters might contribute to the increased risk of thrombotic complications in BSI patients undergoing SAE. Thromboembolism following SAE should be considered and thrombotic prophylaxis should be recommended.

INTRODUCTION

The most frequently injured solid organ following blunt abdominal trauma is spleen. Splenectomy was a standard surgical procedure for splenic injury until the mid‐20th century. However, thrombotic and infectious complications following splenectomy were demonstrated by many studies. , Postsplenectomy thrombosis is life‐threatening complications, including pulmonary embolism(PE), portal‐splenic mesenteric venous thrombosis (PSMVT) and deep venous thrombosis (DVT). Several factors may contribute to the occurrence of thromboembolic events following splenectomy, involving thrombocytosis, hypercoagulability due to imbalance between coagulation and fibrinolysis, endothelial damage and hemodynamic changes in portal flow.

During the past three decades, management of blunt splenic injury (BSI) has evolved from surgical procedure to nonoperative management (NOM). NOM has become the standard of care in patients with stable hemodynamics. Splenic artery embolization (SAE), as an valuable adjunct, has increased the success rate of NOM, approaching 98%. However, since most of published data have focused on the efficacy and benefits of SAE, there have been limited studies considering the complications of the procedure. Reported complications arising from SAE include rebleeding, contrast‐induced renal insufficiency, splenic abscesses, splenic infarction, pleural effusions, and fever. Nevertheless, there has been sparse literature on thrombotic complications following SAE.

Thrombocytosis, alterations of coagulation and fibrinolysis contribute to hypercoagulability and thromboembolism. Platelet (PLT) count, fibrinogen (FIB), D‐dimers (D‐D), fibrinogen/fibrin degradation products (FDP), antithrombin III (AT III) are key markers of coagulation and fibrinolytic activity. Prothrombin time (PT), activated partial thromboplastin time (APTT), and thrombin time (TT) are used for routine coagulation testing. We conducted this study to evaluate the changes in the hemostatic parameters and the risk of thromboembolism by measuring the levels of PLT, FIB, D‐D, FDP, AT III, PT, APTT, and TT before and after SAE in BSI patients.

MATERIALS AND METHODS

Study population

Splenic injury patients treated with SAE in the First Affiliated Hospital of USTC between January 2017 and July 2019 were retrospectively studied. Laboratory tests (i.e., complete blood count, hemostatic parameters) and contrast‐enhanced computed tomography (CT) of the patients were performed promptly on admission. Laboratory tests were repeated on schedule (i.e., 1d, 3d, 7d, after SAE). Inclusion criteria were as follows: splenic injury classed as grade III or greater according to American Association of Surgery for Trauma (AAST) classification, no sign of immediate surgery, contrast extravasation or pseudoaneurysms or arteriovenous fistulas on CT scans (Figure 1 A,B), patients with stable hemodynamics. Exclusion criteria included recent thrombotic events (PE, PSMVT, or DVT), antiplatelet or anticoagulation treatment, coagulopathy, cancer, infection, or hepatic dysfunction.

FIGURE 1

Images in a 56‐year‐old patient with grade IV splenic injury. (A) and (B) Contrast‐enhanced CT shows splenic laceration and extensive subcapsular hematoma (black arrows) with contrast material extravasation (arrowheads). Free intraperitoneal fluid is seen around the liver (white arrows).(C) Splenic artery angiography shows intraparenchymal bleeding (arrowheads) and splenic contusion (arrows). (D) Splenic artery embolization arrested the bleeding

The study was approved by the ethics committee of the First Affiliated Hospital of USTC (No.2021‐RE‐007). The requirement for written informed consent was waived as this was a retrospective study.

Treatment protocol

Under local anesthesia, coeliac axis, and splenic artery angiography was performed with a 5‐French catheter via right femoral artery puncture, a 3‐French microcatheter was used if necessary. Indications for transcatheter embolization included intraparenchymal bleeding, contrast extravasation, pseudoaneurysms, or arteriovenous fistulas on splenic arteriogram (Figure 1C,D). Proximal splenic artery embolizations with gelatin sponge particles was done if diffuse splenic parenchyma injury or excessive vessel abnormalities were seen. Distal embolizations were performed when there were limited bleeding sites in spleen. In addition, other managements including bed rest, vital sign monitoring, hemoglobin (Hb) and hematocrit (Hct) determination, abdominal paracentesis drainage were used in these patients. Color Doppler Ultrasound examination and computed tomographic pulmonary angiography were performed for the patients clinically suspected of having DVT, PSMVT or PE.

Laboratory tests

Blood samples were collected in tubes containing EDTA for PLT count and Hb, Hct determination, which were determined using Sysmex XE‐5000 automated hematology analyzer (Sysmex, Kobe, Japan). Samples were collected in tubes containing 3.2% sodium citrate and then centrifuged at 2000 rpm for 20 min to obtain plasma, which was used for determining coagulation and fibrinolysis parameters. FIB, D‐D, FDP, AT III, PT, APTT, and TT were measured on the STA‐R Evolution automated coagulation analyzer with identical commercial kit (Diagnostica Stago, Asnieres, France). All tests were in accordance with the manufacturer's specifications.

Statistical analysis

Continuous variables were described as mean ±standard deviation. Categoric variables were described as number and percentage. The hemostatic parameters pre‐ and post‐embolization were compared using Student t‐test. Statistical calculations were performed using SPSS software. p < 0.05 was regarded statistically significant.

RESULTS

Clinical characteristics

The demographic and clinical characteristics of the patients are presented in Table 1. A total of 38 BSI patients treated with SAE were recruited in this study. There were 23 males and 15 females with a mean age of 47.13 ± 13.44 years. The AAST splenic injury grades were classified as follows: III, 22(58%); IV, 15(39%) and V, 1(3%). There were 27 proximal (71%) and 11 distal (29%) embolizations. The technical success rate of SAE and the splenic salvage rate were 100%. The mean packed red blood cell and fresh frozen plasma transfusion were 1.15 ± 0.91 and 1.6 ± 1.02 units. There was no thrombotic complication in any of the SAE‐treated patients during hospitalization. Other complications included pleural effusions occurring in 29 patients (76%), fevers occurring in 20 patients (52%), splenic infarctions in 11 patients (28%). Thoracocentesis was used for the patients with pleural effusions. The treatment of fever included physical cooling and nonsteroidal anti‐inflammatory drugs (NSAIDs). Antibiotic therapy was administered in case of suspected or proven infection. Splenic infarctions were asymptomatic and did not require further intervention. Splenic abscess or re‐bleeding after SAE were not observed. Moreover, additional procedures such as splenectomy or repeated angioembolization, were not performed. There was no mortality.

TABLE 1

Characteristics of patients who underwent SAE

Characteristics	N (or mean ±SD)	%
Age (years)	47.13 ± 13.44	‐
Sex
Male	23/38	61
Female	15/38	39
BMI(kg/m 2)	21.75 ± 2.52	‐
Mobidity after SAE
Thromboembolism	0	0
Pleural effusion	29/38	76
Fever	20/38	52
Splenic infarction	11/38	28
Splenic abscess	0	0
Re‐bleeding	0	0
Mortality	0	0
Splenic injury grade
III	22/38	58
IV	15/38	39
V	1/38	3
Packed red blood cell transfusion(units)	1.15 ± 0.91	‐
Fresh frozen plasma transfusion(units)	1.6 ± 1.02	‐
Proximal splenic artery embolizaion	27/38	71
Distal embolizaion	11/38	29

Abbreviations: BMI, body mass index; SAE, splenic artery embolization.

Comparison of laboratory tests pre‐ and post‐embolization

As shown in Table 2, D‐D, FDP and PLT levels presented statistically significant increase at 3d and 7d after SAE when compared to the pre‐embolization(baseline) values (p < 0.05). The FIB levels increased significantly at 1d, 3d, and 7d after SAE compared to baseline values (p < 0.05). Compared to the baseline level, the values of ATIII, PT, APTT, TT, Hb, and Hct showed no statistically significant difference 1d, 3d, and 7d after SAE (p > 0.05).

TABLE 2

Comparison of hemostatic parameters pre‐ and post‐embolization

	N	Pre SAE	1d Post SAE	3d Post SAE	7d Post SAE	P1	P2	P3
D‐D (µg/ml)	38	3.99 ± 2.27	3.74 ± 2.41	5.65 ± 2.58	6.32 ± 2.54	0.517	<0.001	<0.001
FDP (µg/ml)	38	10.74 ± 7.85	9.36 ± 6.59	16.95 ± 10.21	18.95 ± 12.99	0.06	<0.001	<0.001
FIB (g/l)	38	2.42 ± 0.79	3.5 ± 0.89	5.44 ± 1.4	5.7 ± 1.19	<0.001	<0.001	<0.001
PLT (109/l)	38	195.73 ± 63.86	196.36 ± 49.38	381.42 ± 136.59	672.73 ± 220.01	0.949	<0.001	<0.001
PT (s)	38	12.3 ± 0.96	12.16 ± 1.21	12.13 ± 0.91	12.14 ± 0.92	0.403	0.323	0.417
APPT (s)	38	37.54 ± 5.05	36.76 ± 5.89	38.16 ± 7.02	37.77 ± 5.61	0.052	0.304	0.666
TT (s)	38	15.69 ± 1.12	15.51 ± 1.01	15.57 ± 1.08	15.88 ± 1.51	0.149	0.528	0.384
AT Ⅲ (%)	38	91.06 ± 11.74	87.11 ± 11.85	89.53 ± 9.14	94.69 ± 13.14	0.067	0.409	0.106
Hb (g/l)	38	105.97 ± 14.47	104.34 ± 14.89	103.68 ± 15.98	106.1 ± 17.35	0.058	0.067	0.921
Hct (%)	38	30.97 ± 4.2	30.34 ± 4.18	30.35 ± 4.72	31.21 ± 4.71	0.081	0.108	0.601

P1, 1d Post SAE vs Pre SAE; P2, 3d Post SAE vs Pre SAE; P3, 7d Post SAE vs Pre SAE.

Abbreviations: APPT, activated partial thromboplastin time, TT, thrombin time; AT III, Antithrombin III; D‐D, D‐dimers; FDP, fibrinogen/fibrin degradation products; FIB, fibrinogen; Hb, hemoglobin; Hct, hematocrit; PLT, Platelet; PT, prothrombin time; SAE, splenic artery embolization.

DISCUSSION

Postsplenectomy thrombosis is a severe complication and has been documented, with an incidence of 7.7% or even higher. However, the thromboembolic complication in patients with BSI treated by SAE are still unclear. Our results showed significant increase in PLT and statistical difference in FIB, D‐D, FDP pre‐ and post‐embolization in BSI patients. This theoretically increases the risk of thromboembolism after SAE because thrombocytosis and alterations in hemostasis contribute to thromboembolic events. To the best of our knowledge, this is the first study assessing PLT and hemostatic parameters pre‐ and post‐embolization in BSI patients.

Thrombocytosis, defined by a PLT count of more than 500 × 109/L, includes primary thrombocytosis caused by myeloproliferative disease, and reactive (secondary) thrombocytosis which is related to infection, hemorrhage, inflammation, malignancy, major surgery, and asplenia. The positive correlation between primary thrombocytosis and thrombosis is well established. Reactive thrombocytosis has also been shown to be associated with high risk of thrombosis, although some studies do not support this direct association. A prospective study, including 147 patients who underwent splenectomy, confirmed that higher PLT count was directly associated with PSMVT, with approximately 5% incidence. Two clinical studies from Ho et al. demonstrated that reactive thrombocytosis increased the risk of venous thromboembolism or acute PE in a linear “dose‐related” fashion. , Reactive thrombocytosis was also associated with an increased in vitro thrombotic tendency in critically ill patients. Not only venous thrombosis, but also carotid artery thrombosis was proven to be related to reactive thrombocytosis recently. PLT count was found to be a predictor of thrombotic tendency. Our study indicates that PLT significantly increased at 3d and 7d after SAE compared with baseline. This might have contributed, at least in part, to the occurrence of thromboembolism in BSI patients treated by SAE.

Markers of hemostasis and fibrinolysis would increase in patients at risk of thromboembolism. Also, concentrations of these markers are a measure of thrombotic tendency. A prospective randomized study, assessing PT, APTT, PLT, FIB, FDP, and D‐D, suggested greater thrombotic risk in patients undergoing open surgery than laparoscopic procedures. Another study, assessing markers of coagulation, fibrinolysis and PLT activation evaluated the association between hypobaric hypoxia and thromboembolism risk.

Fibrinogen is considered as the direct coagulation marker. The FIB levels affect fibrin network and elevated FIB results in tight fibrin cross‐linking, which results in clot formation. Plasma concentration of FIB reflects thrombin generation and coagulation activation, which results in hypercoagulability. Watters et al studied coagulation parameters and showed that FIB levels were higher in the splenectomy group than control group. They concluded that the splenectomy led to persistent hypercoagulability, developing elevated incidence of DVT (7%) in splenectomy patients compared with that (0%) in nonsplenectomy patients. In our study, the FIB levels are found to be increased significantly after SAE compared with pre‐embolization values. This indicates that coagulation activation takes place post‐embolization.

D‐D and FDP are used as biomarkers of activated coagulation and secondary fibrinolysis. D‐D is generated through the sequential reaction of thrombin, factor XIII, plasmin and finally to the cross‐linked fibrin degradation product. This process is triggered by coagulation activation. Therefore, D‐D is an indirect indicator of thromboembolism. A few studies have established the close relationship between elevated D‐D and thromboembolic disease too. FDP are degradation products of fibrin and fibrinogen activated by plasmin, and are related to coagulopathy and fibrinolysis activation. Canonico et al found that plasma levels of fibrinogen, D‐D, fibrinopeptides 1 and 2, and PLT activating inhibitor were higher in splenectomy patients compared to control group, indicating that splenectomy contributes to hypercoagulable state. The results of our study show that D‐D and FDP levels are greater at 3d and 7d after SAE compared to pre‐embolization. These findings indicate that coagulation and secondary fibrinolysis system are activated in patients following SAE.

Antithrombin III blocks the action of thrombin on fibrinogen through 1:1 binding with thrombin. Therefore, AT III is an important anticoagulant marker. The PT and APTT are conventional, broader screening tests of extrinsic and intrinsic coagulation disorders. Short PT and APTT predict hypercoagulable state. The TT reflects the coagulation and anticoagulation cascade and fibrinolysis. However, our study failed to find a significant difference in AT III, PT, APTT, and TT between pre‐ and post‐embolization in BSI patients.

The main limitations of our study are its retrospective nature and absence of control population. Moreover, the relatively small sample size may not represent all variations of hemostatic changes in patients undergoing SAE. Finally, the follow‐up periods may not have been long enough to establish the incidence of thromboembolic events in post‐embolization patients.

In conclusion, results of the current study demonstrated that PLT significantly increased and activation of coagulation and fibrinolysis took place in BSI patients undergoing SAE. These alterations might have contributed to the increased risk of thrombotic complications, especially 3d and 7d after SAE. Despite SAE as a critical treatment is used extensively in BSI patients and has increased the success rate of NOM, thromboembolism following SAE should be considered and thrombotic prophylaxis should be recommended. However, a further study with larger sample size and longer follow‐up is necessary to confirm our findings.

CONFLICT OF INTEREST

No potential conflict of interest was reported by the authors.