ORIGINAL ARTICLE
J Trauma Inj 2017;30(3):91-97
http://doi.org/10.20408/jti.2017.30.3.91
JOURNAL OF
TRAUMA AND INJURY
Treatment Option for High Grade
Spleen Injury and Predictive Factors
for Non-operative Management
Joung Won Na, M.D.1, Jung Nam Lee, M.D.2, Byung Chul Yu, M.D.2, Min A Lee, M.D.2,
Jae Jung Park, M.D.2, Gil Jae Lee, M.D.2
1
Department of Surgery, Gachon University Gil Medical Center, Incheon, Korea
2
Gachon University Gil Hospital Trauma Center, Incheon, Korea
Received: June 7, 2017
Revised: July 27, 2017 Purpose: Purpose: The prognostic factors of non-operative management (NOM) in
Accepted: July 28, 2017 high-grade spleen injuries have been extensively studied, but factors that would help
treatment decisions are lacking. We compared the characteristics of the patients to
identify the factors affecting treatment choices.
Methods: This is a review of 207 blunt spleen injury patients from January 2004 to De-
cember 2013. We compared clinical features and mortality between surgery and NOM,
and used multivariate regression analysis to find the factor most strongly associated
with prognosis.
Results: Of the 207 patients, 107 had high-grade spleen injury patents (grade III or
above). Of these, 42 patients underwent surgery and 65 patients underwent NOM. The
Correspondence to
mortality was 7% following surgery, 3% with NOM. The amount of packed red blood
Gil Jae Lee, M.D.
Gachon University Gil Trauma Center, cells transfused in the first 24 hours and spleen injury grade were associated with man-
Namdong-daero 774 beon-gil, Nam-
dong-gu, Incheon 21565, Korea
agement type, and mortality was highly associated with activated partial thromboplas-
Tel: +82-32-460-2299 tin time (aPTT) and spleen injury grade.
Fax: +82-32-460-3247
E-mail: nonajugi@gilhospital.com Conclusions: The grade of spleen injury was associated with management and mortali-
ty, so correctly assessing the spleen injury grade is important.
Keywords: Spleen injury; Blunt trauma; Non-operative management
INTRODUCTION
The spleen is the organ most often injured from blunt abdominal trauma [1]. Current
management trends are shifting from immediate splenectomy to non-operative man-
agement (NOM). NOM should be considered for patients who are hemodynamically
http://www.jtraumainj.org Copyright © 2017 The Korean Society of Trauma
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License
eISSN 2287-1683 (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted noncommercial use, distribution, and
pISSN 1738-8767 reproduction in any medium, provided the original work is properly cited.
Journal of Trauma and Injury Volume 30, Number 3, September 2017
stable, lack of peritoneal signs, and capable of treatment The same comparison was made in patients with spleen
by monitoring serial evaluations, with an operating room injury grade III or above, and variables that were signifi-
available for emergency laparotomy [2]. The benefits of cantly related to management and mortality were ana-
spleen conservation include preservation of spleen func- lyzed [9,10].
tion, avoiding complications after splenectomy, including Initial status, clinical features and mortality of the pa-
overwhelming post-splenectomy sepsis [3]. The selective tients and their values were compared by the management
application of embolization has resulted in increasing types (surgical patients [SP] and NOM). Among these,
success of NOM. However, in high grade spleen injuries, categorical variables were expressed using chi-square
prognostic factors to help with treatment decisions are analysis, and continuous items we are compared using
unclear. The variables considered important in many independent sample t-test. We used bivariate regression
studies are patient age, concurrent trauma, spleen injury analysis to assess factors associated with mortality and
grade, abbreviated injury score (AIS), injury severity score management type as the odds ratio [11]. We rejected null
(ISS), injury mechanism, initial vital signs, and initial hypotheses of no difference if p-values were less than 0.05.
blood test results [4,5]. If clearly applicable factors for
management choices, whether NOM or surgery, can be
described, that will be very useful. RESULTS
The object of this study is to review 10 years of expe-
rience of treatment for blunt spleen injury and identify During the study period, 247 patients were hospitalized
predictive factors. and among them, patients under 15 years of age were
excluded. Children tend to have a slight degree of injury
compared to adults at the same trauma. For example,
METHODS in the case of in car traffic accident (TA), an adult often
seats in the front seat as a driver, while children often
Our study was a retrospective study using medical records wear seat belts in the back seat, resulting in a low degree
of 247 patients admitted to a university hospital between of injury to children. Capsules of the spleen are thicker
January 2004 and December 2013 for blunt spleen injury than adults, and the elastin and smooth muscle contents
who had contrast enhanced computed tomography scans of the blood vessels and capsules of the spleen are large, so
and confirmed spleen injuries. We excluded patients they are more resistant to injury and more resilient than
under 15 years old. The exposure variable of interest was adults. Therefore, the treatment direction is different for
surgery versus NOM. We compared patient character- adults and children, and basic treatment methods are also
istics (age, sex, injury mechanism), AIS, ISS, and spleen different. For this reason, we excluded children younger
injury grade. Spleen injury grading follows the American than 15 years (n=40) [12]. Of the 207 patents, there were
Association for the Surgery of Trauma (AAST) Splenic 160 (77.3%) NOM patients, mean age of 41.6±15.8 years.
Injury Scale [6-8] (Table 1). There were 40 (22.7%) surgical patients, with mean age
Table 1. Grading of splenic injury (American Association for the Surgery of Trauma Organ Injury Scale)
Gradea Description
I Sub capsular hematoma <10% of surface area; laceration <1cm parenchymal depth
II Sub capsular hematoma,10-50% of surface area, <5 cm in diameter; laceration, 1-3 cm in depth
Sub capsular hematoma >50% of surface area or expanding; laceration >5 cm depth or expanding; intraparenchymal hematoma, >3 cm or
III
expanding
IV Laceration involving segmental or hilar vessels producing major revascularization
V Completely shattered spleen; hilar injury that devascularizes the entire spleen
a
Advance 1 grade for multiple injuries up to grade III.
92 http://doi.org/10.20408/jti.2017.30.3.91
Joung Won Na, et al. Treatment Option for High Grade Spleen Injury
of 41.1±16.1. Table 2 compares clinical characteristics by grade I NOM patients. Among grade II patients, four
treatment type. The most common injury mechanism patients (8.5%) underwent surgery and 49 (30.6%) were
was an in-car crash (72 patients) and 17 patients (37.2%) NOM. Among grade III, there were 20 (42.6%) SP and
underwent surgery. The second most common cause was 58 (36.3%) NOM, while grade IV included 15 (31.9%)
out-of-car crash, 55 patients (34.4%), of whom 13 (27.7%) SP patients and seven (4.4%) NOM. All seven grade V
were surgical and 26.9% were NOM. Other mechanisms patients had surgery. This means the higher spleen injury
were slip, sports injury, and violence, and the SP or NOM grades are more likely to be treated surgically [13].
ratios of these patients were 12 (25.5%) and 32 (20.0%), The AIS did not differ significantly in the head, neck,
respectively. One patient (2.1%) underwent surgery for chest, and extremity, but the abdomen AIS grade dif-
grade I, and splenectomy was performed due to mes- fered significantly between surgical and NOM patients.
entery bleeding, with spleen and vascular injury detected The overall ISS in surgical patients was 22.5±25.7, and in
during emergency laparotomy. There were 46 (28.8%) NOM cases 16.4±9.4 (p=0.001).
We classified the patients by NOM and SP, and com-
Table 2. Clinical features of 207 patients with blunt spleen
injury
Table 3. Initial clinical features and management, outcomes
OP NOM of 207 patients
p-value
(n=47) (n=160)
OP NOM
Age 41.1±16.1 41.6±15.8 0.839 p-value
(n=47) (n=160)
Sex 0.295
Hb 11.6±2.7 12.7±2.3 0.006
Male (n=165 [79.7]) 40 (85.1) 125 (78.1)
SBP 107.8±27.8 117.5±29.2 0.051
Female (n=42 [20.3]) 7 (14.9) 35 (21.8)
Platelet (×103) 187.7±82.1 218.1±80.4 0.024
Injury mechanism 0.577
aPPT (sec) 35.2±15.8 29.5±6.3 0.020
In car TA 17 (36.2) 55 (34.4) INR 1.49±1.8 1.2±1.3 0.285
Out car TA 13 (27.7) 43 (26.9) Lactate 3.4±2.7 2.7±2.3 0.113
Fall down 5 (10.6) 30 (18.8) BE -5.1±5.4 - 4.5±4.7 0.489
Othersa 12 (25.5) 32 (20.0) Management <0.001
AIS Emergency OP 43 (91.5) 0 (0.0)
Head 2.7±0.5 2.7±0.8 0.869 Delayed OP 4 (8.5) 0 (0.0)
Neck 2.0±0.8 1.8±0.8 0.707 Embolization 0 (0.0) 9 (5.6)
Chest 2.9±0.7 2.7±0.7 0.070 Conservative mx. 0 (0.0) 151 (94.4)
Abdomen 3.4±0.8 2.4±0.7 0.000 Transfusion
Extremity 3±0.5 2.9±1.6 0.771 pRBC 19.5±16.1 7.2±8.8 0.001
ISS 22.49 ± 25.82 16.38±17.86 0.001 PC 10.7±14.8 6.8±9.6 0.192
Spleen injury grade <0.001 FFP 9.4±9.1 4.5±6.2 0.020
I 1 (2.1) 46 (28.7) ICU stay (days) 9.8±13.5 9.4±10.3 0.872
II 4 (8.5) 49 (30.6) Hospital stay (days) 25.9±24.9 27.7±26.2 0.853
III 20 (42.6) 58 (36.2) In hospital mortality (%) 7 (14.9) 5 (3.1) 0.036
IV 15 (31.9) 7 (4.4) Values are presented as mean±deviation or number (%).
V 7 (14.9) 0 (0.0) OP: operation, NOM: non-operative management, Hb: hemoglobin, SBP:
Values are presented as mean±deviation or number (%). systolic blood pressure, aPPT: activated partial thromboplastin time, INR:
OP: operation, NOM: non-operative management, TA: traffic accident, international normalized ratio, BE: base excess, mx.: management, pRBC:
AIS: abbreviated injury scale, ISS: injury severity score. packed red blood cell, PC: platelet, FFP: fresh frozen plasma, ICU: intensive
a
Others: slip, sports injury, violence. care unit.
http://www.jtraumainj.org 93
Journal of Trauma and Injury Volume 30, Number 3, September 2017
pared their initial hemoglobin level, systolic blood patients with spleen grade III or higher high spleen inju-
pressure, international normalized ratio (INR), platelet, ry. Their mean age was SP 40.3±16.6, NOM 39.7±15.5,
activated partial thromboplastin time (aPTT), lactate, and in-car TA was the most frequent injury mechanism,
Base-excess (Table 3). The results showed that the NOM similar to overall patients, but there are no significant
patients’ values were closer to normal than the SP pa- differences between the two management types. In high
tients. Of the surgical patients, 91.5% (43 patients) re- grade spleen injury, abdomen AIS was significantly higher
ceived emergency surgery (immediate splenectomy) and in patients managed by surgery (Table 4).
four patients (8.5%) were switched to surgery (delayed Most test results of high grade patients in SP case were
splenectomy) during non-operative management. The more abnormal than NOM cases, but the differences
activated PTT of NOM patients 29.5±6.3 sec was signifi- were not significant. With activated PTT, the surgical
cantly lower than in the SP patients 35.2±15.8 seconds were 36.2±16.4 seconds and non-surgical patients were
(p=0.020). Significantly more blood in first 24 hours was 29.6±5.6 seconds (p<0.017). The surgical patients con-
needed in patients managed surgically (SP 19.5±16.2 sumed an average of 20.1±17.1 packs of packed red blood
packs, NOM 7.2±8.8 packs, p<0.001). This was similar cells (RBCs) in the first 24 hours, more than the NOM pa-
to the amount of fresh frozen plasma (FFP) and platelet tients: a significant difference compared with other blood
transfusions. Intensive care unit stay was 9.8±13.5 days products (Table 5).
for SP and NOM 9.4±10.3 days, not significantly differ- Mortality of surgically managed high grades patients
ent. The mortality rate of NOM 3.1% was significantly was 16.7% and NOM was 4.6%. This was similar to the
lower than SP patients 14.9% (p=0.036). overall mortality (SP 14.9%, NOM 3.1%), meaning high-
The same comparative analysis was performed in 107 er mortality rates at higher ratings, but the NOM never-
Table 4. Clinical features of grade ≥III patients Table 5. Initial clinical features and management , outcomes
OP NOM of grade ≥III patients
p-value
(n=42) (n=65) OP
NOM (n=65) p-value
Age 40.3±16.6 39.7±15.5 0.845 (n= 42)
Sex 0.197 Hb 11.6±2.7 12.1±2.0 0.267
Male (n=85) 36 (85.7) 49 (75.4) SBP 107.9±28.3 112.4±27.4 0.445
Female (n=22) 6 (14.3) 16 (24.6) Platelet (×103) 181.8±78.8 211.3±84.9 0.074
Injury mechanism 0.053 PTT (sec) 36.2±16.4 29.6±5.6 0.017
In car TA 15 (35.7) 18 (27.7) INR 1.2±0.3 1.3±1.7 0.806
Out car TA 13 (31.0) 19 (29.2) Lactate 3.4±2.8 2.6±1.6 0.093
Fall down 2 (4.8) 16 (24.6) BE -5.1±5.5 -4.4±4.8 0.528
a
Others 12 (28.6) 12 (18.5) Transfusion (packs)
AIS pRBC 20.1±17.1 2.6±1.6 0.003
Head 2.7±0.5 2.7±0.7 0.878 PC 11.1±15.2 4.6±8.6 0.109
Neck 2.0±0.8 2.1±0.9 0.800 FFP 9.7±9.6 5.1±8.2 0.097
Chest 2.9±0.7 2.9±0.6 0.963 ICU stay (days) 9.8±14.3 11.7±13.2 0.631
Abdomen 3.6±0.7 2.9±0.4 <0.001 Hospital stays (days) 27.4±25.7 30.0±26.5 0.625
Extremity 3.0±0.5 2.9±0.6 0.506 In hostpital mortality (%) 7 (16.7) 3 (4.6) 0.036
ISS 23.7±10.9 19.7±10.4 0.063 Values are presented as mean±deviation or number (%).
Values are presented as mean±deviation or number (%). OP: operation, NOM: non-operative management, Hb: hemoglobin, SBP:
OP: operation, NOM: non-operative management, TA: traffic accident, systolic blood pressure, PTT: partial thromboplastin time, INR: interna-
AIS: abbreviated injury score, ISS: injury severity score. tional normalized ratio, BE: base excess, pRBC: packed red blood cell, PC:
a
Others: Slip, sports injury, violence. platelet, FFP: fresh frozen plasma, ICU: intensive care unit.
94 http://doi.org/10.20408/jti.2017.30.3.91
Joung Won Na, et al. Treatment Option for High Grade Spleen Injury
Table 6. Factors associated with operative management and adults, and found that failure of NOM AIS ≥4 was up to
mortality for blunt spleen injury 54.6%. Other studies on the relationship between failure
Exp (B) 95% CI p-value of the NOM and the ISS note higher failure rates of NOM
Factors associated if the ISS is greater than 25 [19].
with management Rossaint et al. reported that it is important to monitor
RBC transfusion 0.918 0.858-0.982 0.012 initial hemoglobin and coagulation factors in major trau-
aPTT 1.101 1.052-1.153 <0.001 ma where massive bleeding is expected, such as spleen in-
Spleen injury grade ≥III 0.207 0.061-0.700 0.011 jury. Lactate and base deficit are sensitive tests to estimate
Factors associated with mortality and monitor the extent of bleeding and shock [20,21].
In our study, the factors affecting treatment and prog-
aPTT 1.226 1.063-1.413 0.005
nosis of blunt spleen injury were age, injury mechanism,
Spleen injury grade ≥III 9.253 1.779-48.123 0.008
spleen injury grade, initial vital status, CBC, coagulation
CI: confidence interval, RBC: red blood cell, aPPT: activated partial throm- battery, and the amount of transfusion in first 24 hours.
boplastin time.
These variables were compared by treatment type (NOM
or SP), followed by bivariate regression analysis of the
theless had a lower mortality rate. This means that even factors that were significantly different. The spleen injury
higher-grade patients can be considered for conservative grade, activated PTT, and the amount of transfused red
management, and this is our main opinion. blood cells were associated with surgical treatment of high
We conducted bivariate regression analysis to identify grade spleen injury. We concluded that it could be more
factors associated with management types and mortal- beneficial to consider NOM because the mortality of
ity of blunt spleen injury patients. Variables associated NOM was lower than surgery in even high spleen injury
with management type were the amount of packed RBC grade patients.
transfused within 24 hours, activated PTT and high grade Our study had considerable limitations. The patient
spleen injury. The mortality-related variables were acti- records of this study included 10 years of data before es-
vated PTT and spleen in patients with high grade spleen tablishment of the trauma center in this hospital, so there
injury (Table 6). could be many omissions. The study sample size was
small. Most patients were of similar age, making compar-
isons of age differences difficult. Thus the importance of
DISCUSSION age, as mentioned in the literature, could be undervalued.
During our patient`s treatment periods from 2004 to
Non-operative management of hemodynamically stable 2014, the intervention team for emergency embolization
blunt spleen injury is currently accepted as a standard had not full-time activated. All grade V patients under-
option. NOM means surgical observation with serial went surgery; therefore we could not obtain information
physical examination, serial computed tomography, or about high grade spleen injury angiographic embolization
angiographic embolization. Research reports the success results. Gaarder et al. [22] analyzed the effect of angio-
rate of NOM is 78% to 98% [14]. With the development graphic embolization in severe spleen injuries on out-
of angiographic intervention, the success rate increased comes measured by laparotomy and splenectomy rates for
up to 98% [15]. In high grade spleen injury of grade III mandatory embolization in grade 3 to 5 whenever posi-
and above, the prognostic consideration for treatment tive angiographic findings. They found that angiographic
decision is unclear [16]. In a study by Olthof et al. [17], embolization use resulted in an increase of patients select-
in hemodynamically stable patients, the factors affecting ed for NOM from 57% to 73% and failure rate decrease
the failure of NOM were age 40 and older, spleen injury from 21% to 4% [22]. Another shortfall of our study was
grade ≥III, ISS ≥25, abdominal AIS ≥3, trauma and in- having no data about in high grade spleen injury patients
jury score (TRISS) <0.8. Watson et al. [18] reviewed 3,085 converted from NOM to delayed operation. The infor-
http://www.jtraumainj.org 95
Journal of Trauma and Injury Volume 30, Number 3, September 2017
mation was limited to surgery management or NOM, MA, Champion HR. Organ injury scaling: spleen and liver (1994
making it impossible to assess prognostic factors.Despite revision). J Trauma 1995;38:323-4.
these limitations, our study’s significance is confirmation 8. Tinkoff G, Esposito TJ, Reed J, Kilgo P, Fildes J, Pasquale M,
of the importance of spleen injury grade in blunt trau- et al. American Association for the Surgery of Trauma Organ
ma of spleen and analysis of definitive factors that affect Injury Scale I: spleen, liver, and kidney, validation based on the
treatment options. Further study analyzed by the cost National Trauma Data Bank. J Am Coll Surg 2008;207:646-55.
effectiveness of treatment options, patient heterogeneity, 9. Tohira H, Jacobs I, Mountain D, Gibson N, Yeo A. Comparisons
and post NOM complications like pseudo-aneurysm or of the outcome prediction performance of injury severity scor-
delayed hemorrhage is needed [23-25]. ing tools using the abbreviated injury scale 90 Update 98 (AIS
98) and 2005 Update 2008 (AIS 2008). Ann Adv Automot Med
2011;55:255-65.
CONCLUSION 10. Lesko MM, Woodford M, White L, O’Brien SJ, Childs C, Lecky
FE. Using Abbreviated Injury Scale (AIS) codes to classify com-
In conclusion, mortality of high grade spleen injury pa- puted tomography (CT) features in the Marshall System. BMC
tients was higher than low grade injuries, and mortality Med Res Methodol 2010;10:72.
of NOM patients was lower than in surgically managed 11. Rosati C, Ata A, Siskin GP, Megna D, Bonville DJ, Stain SC.
patients. The factors that affect the treatment options in Management of splenic trauma: a single institution’s 8-year ex-
high grade splenic injury were aPTT and spleen injury perience. Am J Surg 2015;209:308-14.
grade. 12. Powell M, Courcoulas A, Gardner M, Lynch J, Harbrecht BG,
Udekwu AO, et al. Management of blunt splenic trauma:
significant differences between adults and children. Surgery
REFERENCES 1997;122:654-60.
13. Aseervatham R, Muller M. Blunt trauma to the spleen. ANZ
1. Davis JJ, Cohn I Jr, Nance FC. Diagnosis and management of Journal of Surgery. 2000;70:333-7.
blunt abdominal trauma. Ann Surg 1976;183:672-8. 14. Watson GA, Rosengart MR, Zenati MS, Tsung A, Forsythe
2. Stassen NA, Bhullar I, Cheng JD, Crandall ML, Friese RS, Guil- RM, Peitzman AB, et al. Nonoperative management of severe
lamondegui OD, et al. Selective nonoperative management of blunt splenic injury: are we getting better? Aust N Z J Surg
blunt splenic injury: an Eastern Association for the Surgery of 2006;61:1113-9.
Trauma practice management guideline. J Trauma Acute Care 15. Peitzman AB, Heil B, Rivera L, Federle MB, Harbrecht BG, Clan-
Surg 2012;73(5 Suppl 4):S294-300. cy KD, et al. Blunt splenic injury in adults: multi-institutional
3. Wani MA, Shah M, Malik AA. Non-operative treatment of study of the Eastern Association for the Surgery of Trauma. J
splenic injury in patients with blunt abdominal trauma. Int Trauma 2000;49:177-87; discussion 187-9.
Surg J 2017;4:278-81. 16. Scarborough JE, Ingraham AM, Liepert AE, Jung HS, O’Rourke
4. Hildebrand DR, Ben-Sassi A, Ross NP, Macvicar R, Frizelle AP, Agarwal SK. Nonoperative management is as effective as
FA, Watson AJ. Modern management of splenic trauma. BMJ immediate splenectomy for adult patients with high-grade
2014;348:g1864. blunt splenic injury. J Am Coll Surg 2016;223:249-58.
5. Olthof DC, Joosse P, van der Vlies CH, de Haan RJ, Goslings JC. 17. Olthof DC, Joosse P, Bossuyt PM, de Rooij PP, Leenen LP,
Prognostic factors for failure of nonoperative management in Wendt KW, et al. Observation versus embolization in patients
adults with blunt splenic injury: a systematic review. J Trauma with blunt splenic injury after trauma: a propensity score analy-
Acute Care Surg 2013;74:546-57. sis. World J Surg 2016;40:1264-71.
6. Moore EE, Shackford SR, Pachter HL, McAninch JW, Browner 18. Watson G, Hoffman M, Peitzman A. Nonoperative manage-
BD, Champion HR, et al. Organ injury scaling: spleen, liver, and ment of blunt splenic injury: what is new? Eur J Trauma Emerg
kidney. J Trauma 989;29:1664-6. Surg 2015;41:219-28.
7. Moore EE, Cogbill TH, Jurkovich GJ, Shackford SR, Malangoni 19. Sondeen JL, Prince MD, Kheirabadi BS, Wade CE, Polykratis IA,
96 http://doi.org/10.20408/jti.2017.30.3.91
Joung Won Na, et al. Treatment Option for High Grade Spleen Injury
de Guzman R, et al. Initial resuscitation with plasma and other 2006;61:192-8.
blood components reduced bleeding compared to hetastarch 23. Leeper WR, Leeper TJ, Ouellette D, Moffat B, Sivakumaran T,
in anesthetized swine with uncontrolled splenic hemorrhage. Charyk-Stewart T, et al. Delayed hemorrhagic complications in
Transfusion 2011;51:779-92. the nonoperative management of blunt splenic trauma: early
20. Rossaint R, Bouillon B, Cerny V, Coats TJ, Duranteau J, Fernán- screening leads to a decrease in failure rate. J Trauma Acute Care
dez-Mondéjar E, et al. Management of bleeding following major Surg 2014;76:1349-53.
trauma: an updated European guideline. Crit Care 2010;14:R52. 24. Muroya T, Ogura H, Shimizu K, Tasaki O, Kuwagata Y, Fuse T,
21. Rossaint R, Bouillon B, Cerny V, Coats TJ, Duranteau J, Fernán- et al. Delayed formation of splenic pseudoaneurysm following
dez-Mondéjar E, et al. The European guideline on management nonoperative management in blunt splenic injury: multi-in-
of major bleeding and coagulopathy following trauma: fourth stitutional study in Osaka, Japan. J Trauma Acute Care Surg
edition. Crit Care 2016;20:100. 2013;75:417-20.
22. Gaarder C, Dormagen JB, Eken T, Skaga NO, Klow NE, Pill- 25. Bruce PJ, Helmer SD, Harrison PB, Sirico T, Haan JM. Nonsur-
gram-Larsen J, et al. Nonoperative management of splenic gical management of blunt splenic injury: is it cost effective?
injuries: improved results with angioembolization. J Trauma Am J Surg 2011;202:810-6; discussion 815-6.
http://www.jtraumainj.org 97
