Tomographic Findings are not Always Predictive of Failed Nonoperative Management in Blunt Hepatic Injury.

Tomographic Findings are not Always Predictive of Failed Nonoperative

Management in Blunt Hepatic Injury

1

Yi-Chieh Huang, MD,

Shih-Chi Wu, MD,

Chih-Yuan Fu

, MD,

Yung-Fang Chen,

MD,

Chi-Hsun Hsieh, MD,

Yu-Chun Wang, MD,

Hung-Chang Huang, MD,

2

Jui-Chien Huang, MD,

Chih-Wei Lu, MD

China Medical University, Taichung, Taiwan

Trauma and Emergency Center, China Medical University Hospital, Taichung, Taiwan

Wanfang Medical Center, Trauma and Emergency Surgery Department

Department of Radiology, China Medical University Hospital, Taichung, Taiwan

*Corresponding author.

Chih-Yuan Fu, MD.

Wanfang Medical Center, Trauma and Emergency Surgery Department

Xinglong Rd, Sec 3, No.111, Wenshan District, Taipei, Taiwan

E-mail: [email protected]

Tel: 886-2-29307930

Running title: Nonoperative management in blunt hepatic injury

Key word: blunt hepatic injury, angioembolization, nonoperative management,

contrast extravasation

Yi-Chieh Huang, MD

E-mail: [email protected]

Shih-Chi Wu, MD.

E-mail: [email protected]

Chih-Yuan Fu, MD.

E-mail:[email protected]

Yung-Fang Chen, MD.

E-mail: [email protected]

Chi-Hsun Hsieh, MD

E-mail: [email protected]

Yu-Chun Wang, MD.

E-mail: [email protected]

Hung-Chang Huang, MD.

E-mail: [email protected]

Jui-Chien Huang, MD.

E-mail: [email protected]

Chih-Wei Lu, MD.

E-mail: [email protected]

Background: The nonoperaitve management(NOM) has become the standard

treatment of blunt hepatic injury(BHI) for stable patients. Contrast extravasation(CE)

on computed tomographic(CT) scan had been reported as a sign that was associated

with NOM failure. The goal of this study was to further investigate the risk factors of

NOM failure in patients with CE on CT scan.

Method: From January 2005 to September 2009, patients with CE noted on CT scan

due to blunt hepatic injury were retrospectively studied. Physiological parameters,

severity of injury, amount of transfusion as well as treatment outcome were compared

between patients with NOM failure and NOM success.

Results: A total of 130 patients were enrolled. ISS score, amount of blood transfusion

before hemostatic procedure, and grade of liver injury were significantly higher in

NOM failure than NOM success patients. There was no statistical difference in the

NOM successful rate between patients with contrast leakage into peritoneum and

those with contrast confined in hepatic parenchyma.

Conclusions: Higher ISS, more blood transfusion, and higher grade of liver injury are

factors that correlate with NOM failure in patients with BHI. Contrast leakage into

peritoneum is not always a definite sign of NOM failure in BHI. Early and aggressive

AE is an effective adjunct of NOM in BHI patients even with contrast leakage into

peritoneum.

Introduction

Liver is one of the most commonly injured solid organs in blunt abdominal trauma

that approximately 5% of all trauma admissions are associated with liver injuries.

Traditionally, operative intervention was thought to be the mainstay of controling

hemorrhage in severe hepatic injuries but treatment outcomes remained poor with a

mortality rate of 50-80% .

,

The principles to manage blunt hepatic injury (BHI) has

changed dramatically during the past three decades following the advancement of

treatment concepts and improvements in diagnostic modalities that nonoperative

management (NOM) has become the standard treatment of choice for stable

patients.

Currently, more than 80% of patients with BHI are treated nonoperatively with

reported success rates of over 90%.

Studies have also shown that with the

application of angioembolization (AE) the overall mortality rates in patients with

severe hepatic trauma were as low as 8-22%.

Nowadays AE has been

widely-accepted as a safe and effective therapeutic modality in selected patients with

liver injury.

Although NOM has been widely applied and was highly successful, failure were

not uncommon on some high-risk patients and these patients usually required surgical

interventions. Risk factors of NOM failure included hemodynamic instability, high

grade of liver injury and periportal tracking.

Besides, contrast extravasation (CE)

on computed tomographic (CT) scans, a sign of active bleeding, was also been

reported as a risk factor for NOM failure. It has been proved that application of AE

may significantly reduce NOM failure rate in those patients.

Some reports

suggested that the type of CE could predict clinical outcomes following hepatic

injuries.

However, whether the types of CE played a role in NOM of BHI has not yet

been determined. Therefore, we tried to determine the risk factors associated with

NOM failure in patients with CE on CT scans. The characteristics of patients who had

a failed NOM and their outcomes were also discussed.

Materials and Methods

We reviewed the trauma registry and medical records of blunt hepatic injury

patients at the China Medical University Hospital (CMUH) from January 2005 to

September 2009. CMUH serves as a major trauma referral center in the central part of

Taiwan and serves a population of six million people. During the 57-month

investigation period, patients with BHI were treated according to our established

algorithm (Fig. 1). Patients with unstable hemodynamics [systolic blood pressure

(SBP) <90 mmHg] and concomitant hemoperitoneum received immediate celiotomies

without further study. Routine CT scans were performed in patients with stable

hemodynamics (SBP ≧ 90 mmHg) and suspected blunt abdominal injury.

Angioembolization was performed on all patients with CE on CT scans. In our

institution, AE was 24 hours available and was performed by a board-certified

radiologist. However, those patients who deteriorated during the AE preparation

period was converted to celiotomy immediately without delay. All patients were sent

to an intensive care unit for close observation after AE or operaions. A secondary

operation was indicated for those patients who could not achieve hemodynamic

stability or required further blood transfusions following the primary intervention.

The exclusion criteria of this study were as follows: (1) patients who received

immediate celiotomies due to unstable hemodynamics on arrival; (2) patients were

operated due to indications other than liver injury (3) patients with other major causes

of hemorrhage other than liver injury.

Patients with CE on CT scans were the focus of the current study. These patients

were expected to be managed by angioembolization (AE) for hemostasis. However,

some patients deviated from the algorithm and required surgical interventions. In the

current study, NOM failure was defined as receiving a celiotomy in the initial 24

hours of admission either before or after AE. Patients who did not require surgical

intervention after AE were defined as NOM successes. The demographic

characteristics, initial SBP values, abbreviated injury scales (AIS), injury severity

scores (ISS), amounts of blood transfused before hemostatic procedures

(angioembolization or celiotomy), grades of liver injury, and prothrombin time

(PT)/international normalized ratio (INR) values were compared between patients

with NOM failure and success (Table 1). The differences between patients with

contrast leakage into peritoneum (Fig 3-1) or with contrast confined to the hepatic

parenchyma (Fig. 3-2) were also compared (Table 2). Table 3 lists the characteristics

of the patients who experienced NOM failure.

Results

From January 2005 to September 2009, 148 patients with BHI were admitted to the

CMUH trauma center. The patient distribution is listed in Fig. 2. Eight patients who

received immediate celiotomies due to unstable hemodynamics and ten patients who

received celiotomies due to associated intraabdominal injury other than liver were

excluded from the study. Among the 130 enrolled patients with BHI, 119 (91.5%)

were treated conservatively without surgical interventions. Thirty-two of these 130

patients were noted to have CE on their CT scans. In these 32 patients, 11 received

celiotomies due to hemodynamic deterioration before (N = 8) or after AE (N = 3). The

remaining 21 patients received AE and NOM successfully without further

complications.

The mean age of the 148 BHI patients was 32.1 ±15.6 years. Of these patients, 96

were male (64.9%) and 52 were female (35.1%). In patients with CE on their CT

scans, there were no significant differences with regard to age, gender, SBP, PT/INR

or AIS between NOM success and NOM failure patients (Table 1). ISS values (29.5

±12.5 vs. 19.7 ±8.5, p < 0.05), amounts of transfused blood prior to hemostatic

procedures (4.1 ±2.8 vs. 1.8 ±2.6 U, p < 0.05), and grades of liver injury (p < 0.05)

were significantly higher in NOM failure than in NOM success patients (Table 1).

Table 2 compared the differences between different types of CE. There were no

statistical differences in age, gender, AIS values, and ISS values between patients

with contrast leakage into the peritoneum and with contrast confined to the hepatic

parenchyma (Table 2). For those patients who had contrast leaked into the peritoneum,

they were transfused with more before hemostatic procedures were started (3.6 ±3.4

vs. 1.3 ±1.5 U, p < 0.05) than those patients with contrast confined to the hepatic

parenchyma. In patients with contrast leakage into the peritoneum, the NOM success

rate was 52.9% compared to 80% of the corresponding patients who had contrast

confined to the hepatic parenchyma. There was no statistical difference in NOM

success rates between these two groups of patients.

Discussion

As the liver is a commonly injured organ in blunt abdominal trauma, mortality rates

of blunt hepatic injury can be as high as 50-80%.

With improvements in diagnostic

technologies and revoluting treatment concepts, NOM has become the standard

treatment in BHI patients with stable hemodynamics.

However, there are still

some patients who had a failed NOM. Several risk factors that predict NOM failure in

blunt hepatic injuries include hemodynamic instability, high AAST grade of liver

injury, periportal tracking, and contrast extravasation on CT scan.

Improved outcomes have been achieved with the application of AE, which has been

reported as a safe and effective adjunct in NOM.

At our institution, AE can be performed within a short period of time and is used in

our established algorithm for patients with BHI. In our series, the overall success rate

of NOM is 91.5% when AE is used as an adjunct. On the other hand, the NOM

success rate is 65.5% for those patients with CE on CT scans. Abdominal AIS values

were not significantly different between NOM failure and NOM success patients.

However, ISS values were significantly higher (29.5 ±12.5 vs. 19.7 ±8.5, p < 0.05) in

NOM failure patients compared to NOM success patients (Table 1), suggesting that

NOM failures might be related to associated injuries of other body regions. As a result,

associated injuries play a significant role on NOM success. The current literature also

indicates that concomitant injuries are major determinants in BHI patient outcomes.

Moreover, a significantly higher grade of liver injury was observed in NOM failure

patients that 64% of NOM failure patients had high-grade liver injuries (grades IV &

V) (Table 1). This difference was statistically significant and indicates that the grade

of liver injury is an important predictive factor for NOM failure in BHI patients in the

current study. As described in other studies, most patients required surgical

interventions had grade IV or V injuries and underwent urgent celiotomies due to

hemodynamic instability.

It has been reported that the amount of blood

transfusions were thought to be an independent predictor of increased mortality in

blunt hepatic and splenic injuries. Patients who required either urgent or delayed

operations received more blood transfusions than those who were managed

nonoperatively and it has been suggested that nonoperative management should be

deemed a failure if a transfusion of more than two units were required.

In our

study, more blood was transfused prior to hemostatic procedures in NOM failure

patients compared to NOM success patients (4.1 ±2.8 vs. 1.8 ±2.6 U, respectively, p <

0.05) (Table 1). Furthermore, 81.8% of NOM failure patients had a blood transfusion

of greater than three units. This cutoff value was statistically significant, suggesting

that blood transfusions of greater than three units seemed to be a risk factor for NOM

failure.

CT scan is an extremely valuable tool for evaluation of abdominal injury. Many

studies have reported CE on CT scans as a specific sign of vascular injury and active

bleeding

and the scale of CE is related to the possibility of NOM failure.

Fang et al. described different types of CE on CT scans and made a prediction of

patient outcomes after hepatic injury using his classification. He found that contrast

leakage into the peritoneum (Fig. 3-1) represented massive or direct active bleeding

into a relatively low-pressure space and was associated with 100% of NOM failures.

In contrast, contrast confined to the hepatic parenchyma (Fig. 3-2) had a high

probability of spontaneous hemostasis.

However, in the current study, we found that

there was no difference on NOM failure rate between different types of CE. Our study

showed that 9/17 (52.9%) of the patients with contrast leakage into the peritoneum

were managed successfully without operative interventions. This is a significant

improvement compared to that reported by Fang et al. which can be explained by:

First, improvements in CT resolution have made CE more easily visible. A

state-of-the-art multidetector CT scan can precisely locate active bleeders accurately

via optimized contrast injection protocol and provide valuable information for the

selection of appropriate treatment.

Second, a strategy to apply AE early and

aggressively has contributed significantly to the improved NOM success rate not only

in liver injury but also in other solid organ injuries such as splenic, renal injury as

well as unstable pelvic fracfture.

Third, the constantly improving interventional

radiology instruments and techniques in the past 20 years may be helpful in

conjunction with other treatment modalities to improve NOM success rate.

Finally,

the integrity of the trauma care system, including pre-hospital phase and in-hospital

critical care, also helps to increase NOM success rate. Taken together, these factors

has made successful NOM in patients with bleeding high grade liver injuries possible.

As a result, different types of CE did not seem to significantly affect the success rate

of NOM in current study.

The characteristics of patients who received celiotomies (N = 11) in the current

study are described in Table 3. Eight patients received celiotomies due to

hemodynamic deterioration during the AE preparation period. The other three

received celiotomies after AE due to ongoing bleeding or hemodynamic deterioration.

Among these 11 patients, 7 (63.6%) had high-grade (grades IV & V) injuries and the

other 4 patients (36.4%) had low-grade injuries. Eight patients (72.7%) presented with

contrast leakage into the peritoneum. There were two mortalities in this study. One

patient was involved in a traffic accident and had a hemorrhagic shock due to a

juxtahepatic vein tear, liver laceration, and renal laceration (ISS = 50). He was dead

during emergency celiotomy. Another patient was also injured in a traffic accident

and had an ISS of 50. He received two celiotomies for hemostasis in the initial three

days and was complicated with increased intracranial pressure after surgery. The

patient died on the fourth day due to brainstem dysfunction caused by diffuse brain

swelling with uncal herniation. One limitation of the current study is its

retrospective nature and small number of examined cases. A possible selection bias

may limit our conclusions. Nonetheless, we demonstrated improved results of NOM

by timely and effective AE for those patients with severe BHI. Further studies with

larger sample size and prospective designs are needed to update the current algorithms

for prompt diagnoses and precise treatment of patients with blunt abdominal trauma.

Conclusion

In the current study, a higher ISS score, more blood transfusions, and a higher

grade of liver injury are factors that correlate with NOM failure in patients with BHI.

Although contrast leakage into the peritoneum indicates severe injury and active

bleeding, it is not longer a sign that predicts NOM failure in BHI. Early and

aggressive AE is an effective adjunct of NOM in BHI patients even for those patients

with contrast leakage into the peritoneum.

Acknowledgements

The authors thank Ying Chi Lin and the Department of Health Service

Administration, China Medical University for their assistance in the statistical

analysis.

References

1. Croce MA, Fabian TC, Menke PG. Nonoperative management of blunt hepatic

trauma is the treatment of choice for hemodynamically stable patients. Ann Surg

1995; 221:744-753.

2. Fabian TC, Bee TK. Trauma, 6

edition. McGraw-Hill 2008; p. 638.

3. Moynihan BG. Abdominal Operations. Philidelphia: W.B. Saunders 1905; p. 473.

4. Pringle JH. Notes on the arrest of hepatic hemorrhage due to trauma. Ann Surg

1908; 48:541-548.

5. Cogbill TH, Moore EE, Jurkovich GJ, et al. Severe hepatic trauma: a

multi-center experience with 1335 liver injuries. J. Trauma 1988; 28:1433-1438.

6. Meredith JW, Young JS, Bowling J, et al. Nonoperaitve management of blunt

hepatic trauma: the exception or the rule? J. Trauma 1994; 36:529-535.

7. Letoublon C, Chen Y, Arvieux C. et al. Delayed celiotomy or laparoscopy as part

of the nonoperaitve management of blunt hepatic trauma. World J. Surg. 2008;

32:1189-1193.

8. Pachter HL, Kundson MM, Esrig B, et al. Status of non-operative management

of blunt hepatic injuries in 1995: a multicenter experience with 404 patients. J.

Trauma 1996; 40:31-38.

9. Asensio JA, Demetriades D, Chahwan S, et al. Approach to the management of

complex hepatic injuries. J. Trauma 2000; 48:66-69.

10. Richardson JD. Change in the management of injuries to the liver and spleen. J

Am Coll Surg 2005; 200:648-669.

11. Johnson JW, Graciac VH, Gupta R, et al. Hepatic angiography in patients

undergoing damage control laparotomy. J. Trauma 2002; 52:1102-1106.

12. Carrillo EH, Spain DA, Wohltmann CD, et al. Interventional techniques are

useful adjuncts in nonoperative management of hepatic injuries. J. Trauma 1999;

46:619-624.

13. Ciraulo DL, Luk S, Palter M, et al. Selective hepatic arterial embolization of

grade IV and V blunt hepatic injuries: an extension of resuscitation in the

nonoperative management of traumatic hepatic injuries. J. Trauma 1998;

45:353-359.

14. Hagiwara A, Yukioka T, Ohta S, et al. Nonsurgical management of patients with

blunt hepatic injury. AJR Am J Roentgenol 1997; 169:1151-1156.

15. Asensio JA, Roldán G, Petrone P, et al. Operative management and outcomes in

103 AAST-OIS grades IV and V complex hepatic injuries: trauma surgeons still

need to operate, but angioembolization helps. J. Trauma 2003; 54:647-654.

16. Mohr AM, Lavery RF, Barone A, et al. Angiographic embolization for liver

injuries: low mortality, high morbidity. J. Trauma 2003; 55:1077-1082.

17. Ochsner MG. Factors of failure for nonoperative management of blunt liver and

splenic injuries. World J. Surg. 2001; 25:1393-1396.

18. Moore EE, Cogbill TH, Jurkovich GJ, et al. Organ injury scaling: spleen and

liver (1994 revision). J. Trauma 1995; 38:323-324.

19. Fang JF, Chen RJ, Wong YC, et al. Classification and treatment of pooling of

contrast material on computed tomographic scan of blunt hepatic trauma. J.

Trauma 2000; 49:1083-1088.

20. DiGiacomo JC, McGonigal MD, Haskal ZJ, et al. Arterial bleeding diagnosed by

CT in hemodynamically stable victims of blunt trauma. J. Trauma 1996;

40:249-252.

21. Renz BM, Bott J, Filiciano DV. Failure of nonoperative treatment of a gunshot

wound to the liver predicted by computed tomography. J. Trauma 1996;

40:191-193.

22. Knudson

Lim

Jr,

D, et al. Non-operative

of blunt

liver injurie in adults: the need for continued surveillance.

1990; 30:

1494-1500.

23. Davis KA, Brody JM, Cioffi WG. Computed tomography in blunt hepatic trauma.

Arch. Surg. 1996; 131:255-260.

24. Ciraulo D, Luk S, Palter M, et al. Selective hepatic arterial embolization of grade

IV and V blunt hepatic injuries: an extension of resuscitation in the nonoperative

management of traumatic hepatic injuries. J. Trauma 1998; 45:353– 359.

25. Schnüriger B, Inderbitzin D, Schafer M, et al. Concomitant injuries are an

important determent of outcome of high-grade hepatic trauma. Br. J. Surg. 2009;

96:104-110.

26. Chen RJ, Fang JF, Lin BC, et al. Factors determining operative mortality of

grade V blunt hepatic trauma. J. Trauma 2000; 49:886-891.

27. Robinson III WP, Ahn J, Stiffler A. et al. Blood transfusion is an independent

predictor of increased mortality in nonoperatively managed blunt hepatic and

splenic injuries. J. Trauma 2005; 58:437-445.

28. Malhotra AK, Fabian TC, Corce MA, et al. Blunt hepatic injury: a paradigm shift

from operative to nonoperative management in the 1990s. Ann Surg. 2000;

231:804-813.

29. Malhortra AK, Latifi R, Fabian TC, et al. Multiplicity of solid organ injury:

influence on management and outcomes after blunt abdominal trauma. J. Trauma

2003; 54:925-929.

30. Fang JF, Wong YC, Lin BC, et al. The CT risk factors for the need of operative

treatment in initially hemodynamically stable patients after blunt hepatic trauma.

J. Trauma 2006; 61:547-554.

31. Wu SC, Chow KC, Lee KH, et al. Early selective angioambolization improves

success of nonoperative management of blunt splenic injury. Am Surg 2007;

73:897-902.

32. Wei B, Hemmila MR, Arbabi S, et al. Angioembolization reduces operative

intervention for blunt splenic injury. J. Truama 2008; 64 1472-1477.

33. Fu CY, Wu SC, Chen RJ, et al. Evaluation of pelvic fracture stability and the

need for angioembolization: pelvic instabilities on plain film have an increased

probability of requiring angioembolization. Am J Emerg Med 2009; 27:792-796.

34. Hagiwara A, Sakasi S, Goto H, et al. The role of interventional radiology in the

management of blunt renal injury: a practical protocol. J. Trauma 2001;

51:523-531.

Legends:

Figure 1: The algorithm for the management of patients with blunt hepatic injuries.

Figure 2: The patient distribution in the current study.

Figure 3-1: Contrast leakage into the peritoneum. The large arrow indicates contrast

extravasation into the hepatic parenchyma, and the small arrow indicates contrast

pooling into the hemoperitoneum.

Figure 3-2: Contrast confined to the hepatic parenchyma. The large arrow indicates

contrast extravasation into the hepatic parenchyma.

Table 1. The NOM success patients had higher ISS, AIS and larger amount of blood

transfusion before hemostatic procedure.

Table 2. The NOM successful rate was not significantly different between contrast

leakage into peritoneum and contrast confined in hepatic parenchyma

Table 3. There were two mortalities in NOM failure patients due to multiple organ

failure. There were three patients received celiotomy after angioembolization