British Journal of Radiology (2004) 77, 308-311
© 2004 British Institute of Radiology
doi: 10.1259/bjr/21985061
Factors affecting clinical outcome of patients who undergo transcatheter arterial embolisation in splenic injury
Z Sekikawa, MD
1
S Takebayashi, MD
1
H Kurihara, MD
1
J Lee, MD
3
T Niwa, MD
1
M Kawamoto, MD
1
T Yamamoto, MD
2
J Suzuki, MD
2
M Sugiyama, MD
2 and
T Inoue, MD
3
Departments of 1 Radiology and 2 Critical and Emergency, Yokohama City University Medical Centre, 4–57 Urafunecho Minamiku, Yokohama-city and 3 Department of Radiology, Yokohama City University School of Medicine, 3–9 Fukuura Kanazawaku, Yokohama-city, Japan
 |
Abstract
|
|---|
Transcatheter arterial embolisation (TAE) offers a less invasive approach to traditional laparotomy for the management of bleeding in the context of blunt splenic injury. This is a retrospective review study to identify clinical factors associated with clinical outcome of the patients who underwent this procedure. Of 65 patients with splenic injuries at our institution, 26 patients underwent TAE for management of bleeding. The following factors were assessed to determine their relationship to procedure outcomes: American Association for the Surgery of Trauma (AAST) grade, complications, age, shock index, injury severity score (ISS), haemoglobin (Hb), haematocrit (Ht), prothrombin time (PT), activated partial thromboplastin time (APTT), systolic blood pressure (BP), BP changes during TAE, blood transfused before TAE and timing of TAE. The overall good clinical outcome rate was 73.1% (19/26). Of the factors we assessed, absence of concomitant pelvic injury, higher Hb, higher Ht, higher BP, greater increases in BP during TAE and a decreased requirement for blood transfusions before TAE were associated with good clinical outcome of the patients who underwent TAE in splenic injury.
|
Introduction
|
|---|
Several decades ago, splenectomy was the sole treatment for traumatic splenic injury [1], leaving the asplenic patient vulnerable to infection [2–4]. Recently, non-operative management of abdominal injury has been employed as an alternative in haemodynamically stable patients [1, 5–13]. Among non-operative approaches, transcatheter arterial embolisation (TAE) has been widely used to control bleeding in patients with abdominal injuries, as it can rapidly assure haemostasis [5–9]. Criteria for non-operative management include: (1) the restoration of haemodynamic stability with minimal fluid resuscitation; and (2) the absence of significant head injury, associated intra-abdominal injuries or concomitant injuries requiring surgical intervention. However, controversy remains regarding the indications for TAE [1, 10–13]. Several studies have suggested that various clinical factors should be used to guide the choice of treatment modality.
We retrospectively reviewed the medical records of patients with splenic injury and assessed the outcome of TAE to clarify clinical factors that are associated with clinical outcome of the patients who underwent this procedure.
|
Methods and materials
|
|---|
This project conformed to the guidelines of the Declaration of Helsinki.
Patients
We reviewed the records of 65 consecutive patients that were admitted to the Emergency Centre of Yokohama City University between January 1996 and December 2001 for splenic injuries with or without injuries of other organs. Patients that required emergency surgery for gastrointestinal tract injury or those with severe haemodynamic instability did not undergo angiography. Those patients with stable haemodynamics, implying a lack of significant bleeding, also did not undergo angiography. Thus, 32 of 65 patients underwent angiography. Of these 32 patients, 4 patients had injury grade that could not be classified due to incomplete documentation in the clinical chart, and those patients were excluded from study. An additional two patients were also excluded from study; in one case TAE was performed to stop pancreatic haemorrhage after splenectomy, and in the other case, the patient underwent TAE for delayed splenic rupture after a period of conservative management in another hospital. The remaining 26 patients (22 males and 4 females) were included in this study.
Patient age range was 13–64 years (mean±standard deviation (SD), 34.3±17.2). Causative events for splenic injury included traffic accidents (n=19), falls (n=5) and assaults (n=2). Radiologists examined all 26 patients using standard angiographic techniques. Associated hepatic (n=8, 4 of 8 were embolised), renal (n=9, 4 of 9 were embolised), and pancreatic (n=5, 3 of 5 underwent surgery after splenic TAE) injuries were also characterized. The decision to perform embolisation was ultimately decided by the radiologist. Indications included presence of extravasation (n=23) or pseudoaneurysm (n=1). The site of extravasation was beyond the splenic parenchyma (n=5), within the splenic parenchyma (n=16), and from hilar vessels (n=2). Even if there was no evidence of extravasation, patients proceeded to TAE if they had evidence of disruption of terminal arteries, and avascularity and irregularity in the accumulation of contrast medium (n=2).
Good clinical outcome of patients was defined as the ability to control bleeding successfully without use of ancillary methods. In three cases, patients underwent surgery after TAE secondary to continued haemodynamic instability. However, intraoperative observation documented haemostasis of the spleen, and haemodynamic instability was attributed to concomitant pancreatic injury or hepatic injury; thus, these patients were included in the final analysis. TAE that was performed after a haemostatic operation was also regarded as successful if no further bleeding occurred (n=1). Poor clinical outcome of patients was defined as inadequate haemostasis as documented by ultrasound (an expanding collection), intraoperative observations (visual bleeding), via clinical scenario (haemodynamic instability in spite of continuous blood transfusion) (n=4), and patients died within 6 h of TAE (n=3).
Transcatheter arterial embolisation
After initial stabilization in the emergency room, patients with suspected active intra-abdominal bleeding that did not require immediate surgery underwent angiographic investigations. Splenic artery angiography was obtained using intra-arterial administration of 61.24% iopamidol (Iopamiron 300; Nihon Schering, Osaka, Japan), 64.71% iohexol (Omnipaque 300; Daiichi, Tokyo, Japan), or 61.24% iomeprol (Iomeron 300; Eisai, Tokyo, Japan) at a rate of 3–4 ml s–1 (for a total of 10–15 ml) using a digital subtraction angiographic device (POLYSTAR T.O.P; Siemens, Munchen, Germany). We used 5 F catheters (MP-YT5.0F, MP-YT5.0F (1)-805-S, RM3 (Cathex, Tokyo, Japan)) and 3 F microcatheters, such as SP catheter (TERUMO, Tokyo, Japan) or FASTRACKER 325 (Boston Scientific, Cork, Ireland). Coil embolisation was performed with coils ranging from 3 mm to 10 mm in diameter. Gelatin sponges used in TAE were SPONGEL (Yamanouchi, Tokyo, Japan) or GELFORM (Pharmacia and Upjohn, Tokyo, Japan). TAE was performed by placing coils in the main trunk of the splenic artery between the origin of dorsal pancreatic branch and the next more distal pancreatic branch. Gelatin sponge particles were also injected into the branches of splenic artery in six cases with outstanding extravasation. The choice of stainless steel coils or microcoils depended on the size of the catheter used; for 5 F catheters, stainless steel coils were used, and for 3 F catheters, microcoils were used. Choice of coil size (3 mm, 5 mm, 6 mm, 8 mm or 10 mm diameter) was based on visual interpretation of the arterial diameter on angiogram. Stasis of contrast material at the proximal point of the splenic artery was demonstrated via fluoroscopy, and coil delivery was completed to ensure haemostasis. Perfusion to the spleen was usually maintained by the left gastric–short gastric route, or dorsal pancreatic artery. These routes were not embolised to maintain splenic perfusion. Finally, a coeliac arteriogram was obtained to confirm occlusion of the splenic artery.
Clinical analysis
Follow up CT scans were obtained in 15 patients out of 19 good clinical outcome group, all of them revealed good perfusion in the spleen. The following factors were assessed to determine their relationship to procedure outcomes: injury grade, complications, age, shock index, injury severity score (ISS), haemoglobin (Hb), haematocrit (Ht), prothrombin time (PT), activated partial thromboplastin time (APTT), systolic blood pressure (BP), BP changes during TAE, blood transfused before TAE and the intervals between the injury and arrival at the hospital and performance of TAE. Other considerations included pelvic injury (pelvic fractures or extravasation from the iliac artery), head injury (brain contusions, subarachnoid haemorrhage and skull fractures) and lung injury (lung contusions, pneumothorax and haemothorax). Patients were classified using the American Association for the Surgery of Trauma (AAST) Organ Injury Scale-Spleen, based on CT findings or intraoperative observation. Statistical analysis was performed using Fisher's 2 x 2 exact tests or the Mann-Whitney U-test. Statistical significance was set at p<0.05.
|
Results
|
|---|
Of the 28 patients with blunt splenic injury that underwent TAE, 26 qualified for this study. Of these 26 patients, we confirmed good clinical outcome in 19 patients (73.1%). Grade of injury was not a significant predictor of clinical outcome (Table 1
). While clinical outcome of the patients was not affected by the presence of lung or head injury, the absence of pelvic injury was significantly associated with good clinical outcome (Table 2). Types of pelvic injury included vertical shear (n=2), lateral compression (n=2) and anteroposterior compression (n=1). There was no significant association between patient age and clinical outcome or between shock index (0.44 to 2.50) or ISS (4 to 50) and clinical outcome (Table 3).
Higher values of Hb and Ht were significantly associated with good clinical outcome (Table 3
), but differences in PT and APTT were not (Table 3). BP range was 50–150 mmHg just before TAE and 40–170 mmHg after TAE. BP elevation during TAE ranged from –50 mmHg to 70 mmHg. Higher values of each of these parameters were significantly associated with good clinical outcome (Table 3). Blood transfusion requirements before TAE (range 0–40 units) were significantly associated with clinical outcome of the patients (Table 3), with lower blood transfusion requirements favouring good clinical outcome.
The interval from injury to arrival at hospital ranged from 20 min to 6 h, and the interval from injury to the beginning of TAE ranged from 80 min to 12.5 h. The period from arrival at hospital to TAE ranged from 35 min to 7 h. No statistical association was observed with respect to timing of the procedure and clinical outcome of the patients (Table 3).
|
Discussion
|
|---|
Several studies have previously assessed success rate of TAE for management of bleeding in splenic injury. In this series, we confirmed good clinical outcome in 73.1% of patients, which is lower than previously reported success rates of over 80% [6–8]. However, the wide indications for and strict definition of good clinical outcome in the present study might contribute to this lower rate. Furthermore, mean ISSs in our study (26.8±10.7) were much higher in the present study than in previous reports [6–8]. For example, Hagiwara et al [7] described a success rate of 86.7% in a series of patients with mean ISS of 22.2±12.8, but excluded patients with severe abdominal organ injury (such as hepatic or renal trauma) of those that underwent surgical treatments. Sclafani et al [6] reported a success rate of 95%, but patients in their study had a mean ISS of 18, and the rate of associated complications such as pelvic injury was lower than that in the present study (4/60 vs 5/26). Finally, Davis et al [8] reported a success rate of 80%, but the ISS of patients in their study was only 21±10. In the present study, splenic injury was distributed between all AAST grades (I–V), and injury grade did not emerge as a significant factor in influencing clinical outcome of the patients who underwent TAE. In contrast, Brasel et al [12] reported that injury grade was the only factor related to the success rate of non-operative management. However, as Moore et al [15] point out, the AAST splenic injury scale was not developed to assign prognostic value.
In grading splenic injury, we used the established practice of CT. However, CT findings often show no correlation to the severity of the splenic injury [16, 17] and are notoriously poor in identifying vascular injuries [5–7]. Indeed, Sutyak et al [16] reported that CT findings were a poor predictor of operative findings of the degree of adult splenic injury. In the present study, CT correctly identified bleeding in only 2 of 6 cases (33.3%).
Contrary to our expectations, most concomitant injuries did not affect clinical outcome of the patients. Cathy et al [11] reported that while presence of lung injury did not affect outcomes of operative vs non-operative management in patients with blunt splenic injury, the presence of injury to the cervical spine, head, or pelvis was a significant factor. The present study showed a significant influence of pelvic injury on clinical outcome, but no influence from the presence of head injury. Concomitant head injury has been reported to be associated with failure of non-operative management in patients with blunt splenic injury [10]. Although experimentally unproven, Gando et al [18] suggested that the release of tissue thromboplastin from the injured brain may impair systemic coagulation. However, other authors dispute the concept that failure of non-operative management is associated with head injury. For example, Keller et al [19] concluded that associated head injury in children is not a contraindication for non-operative management. Sartorelli et al [14] reported that the presence of head injury or solid abdominal organ injury should not preclude non-operative management in haemodynamically stable patients. Data from the present study supports the use of non-operative management in patients with head injury. Good clinical outcome of the patients who underwent TAE with blunt splenic injury with associated pelvic injury was significantly lower than that in patients without pelvic injury (Table 2), regardless of the type of injury. As patients with pelvic injury had significantly lower Hb and Ht, this may or may not be an independent phenomenon. Good clinical outcome was independent of patient age. Smith et al [1] suggested that patients over 55 years of age require surgical management, while Karen et al [12] concluded that there was no correlation between age and the success rate of non-surgical management. In general, the concept that patient age does not affect the success rate of non-operative management seems to have prevailed recently [12, 13].
Previous studies have demonstrated operative management of patients with blunt splenic injury was employed for patients with significantly higher ISS than for those managed non-operatively [8, 16, 19, 20]. However, few authors have examined the relationship between ISS and clinical outcome of TAE. Velmahos et al [9] compared the ISS of patients with positive angiograms to those with negative angiograms and found no significant difference. Furthermore, Davis et al [8] reported that a higher ISS does not automatically predict failure of non-operative management. Mean shock index and mean ISS had no significant influence on clinical outcome of the patients in the present study. Although it is reasonable to expect that low Hb, low Ht, impairment of coagulation system and low BP would contribute to poor clinical outcome, there is no published data to confirm this supposition. In the present study, we demonstrated that low Hb, low Ht, and low BP were associated with poor clinical outcome of the patients who underwent TAE. However, impairments in coagulation (PT and APTT) showed no association with clinical outcome.
Our institution dictates the availability of radiology services and consultation within 30 min of an emergency call. The interval from injury to arrival at hospital and to commencement of TAE showed no significant effect on clinical outcome. Similarly, interval from arrival at hospital to the commencement of TAE had no significant effect on clinical outcome of the patients. This is consistent with previous reports by Velmahos et al [9].
|
Conclusion
|
|---|
We retrospectively reviewed the medical records of 65 patients with splenic injury and assessed 26 patients with regard to TAE clinical outcome. Concomitant pelvic injury was associated with poor clinical outcome; however this relationship was not seen with concomitant lung or head injury. Low Hb, low Ht, low BP before and after TAE, decreases in BP during the procedure, and increased transfusion requirements before TAE were associated with poor clinical outcome. Injury grade, patient age, shock index, ISS, PT, APTT and timing of the procedure from injury or from arrival at the hospital did not significantly affect clinical outcome of the patients who underwent TAE in splenic injury.
Received for publication June 23, 2003.
Revision received November 17, 2003.
Accepted for publication December 16, 2003.
|
References
|
|---|
- Smith JS Jr, Wengrovitz MA, Delong BS. Prospective validation of criteria, including age, for safe, nonsurgical management of the ruptured spleen. J Trauma 1992;33:363–9.[Medline]
- Posey DC, Marks C. Overwhelming postsplenectomy sepsis in childhood. Am J Surg 1983;145:318–21.[CrossRef][Medline]
- Malangoni MA, Dillon LD, Klamer TW, Condon RE. Factors influencing the risk of early and late serious infection in adults after splenectomy for trauma. Surgery 1984;96:775–83.[Medline]
- Green JB, Shackford SR, Sise MJ, Fridlund P. Late septic complications in adults following splenectomy for trauma: a prospective analysis in 144 patients. J Trauma 1986;26:999–1004.[Medline]
- Sclafani SJA, Weisberg A, Scalea TM, Phillips TF, Duncan AO. Blunt splenic injuries: nonsurgical treatment with CT, arteriography, and transcatheter arterial embolization of the splenic artery. Radiology 1991;181:189–96.[Abstract/Free Full Text]
- Sclafani SJA, Shaftan GW, Scalea TM, Patterson LA, Kohl L, Kantor A, et al. Nonoperative salvage of computed tomography-diagnosed splenic injuries: utilization of angiography for triage and embolization for hemostasis. J Trauma 1995;39:818–27.[Medline]
- Hagiwara A, Yukioka T, Ohta S, Nitatori T, Matsuda H, Shimazaki S. Nonsurgical management of patients with blunt splenic injury: efficacy of transcathter arterial embolization. AJR Am J Roentgenol 1996;167:159–66.[Abstract/Free Full Text]
- Davis KA, Fabian TC, Croce MA, Gavant ML, Flick PA, Minard G, et al. Improved success in nonoperative management of blunt splenic injuries: embolization of splenic artery pseudoaneurysms. J Trauma 1998;44:1008–15.[Medline]
- Velmahos GC, Chahwan S, Falabella A, Hanks SE, Demetriades D. Angiographic embolization for intraperitoneal and retroperitoneal injuries. World J Surg 2000;24:539–45.[CrossRef][Medline]
- Smith JS Jr, Cooney RN, Mucha P Jr. Nonoperative management of the ruptured spleen: a revalidation of criteria. Surgery 1996;120:745–51.[CrossRef][Medline]
- Cathy KL, Brady WJ Jr, Butler K, Blow O, Cephas GA, Young ES. Blunt splenic trauma: characteristics of patients requiring urgent laparotomy. Am Surg 1998;64:450–5.[Medline]
- Brasel KJ, Delisle CM, Olson CJ, Borgstrom DC. Splenic injury: trends in evaluation and management. J Trauma 1998;44:283–6.[Medline]
- Gaunt WT, McCarthy MC, Lambert CS, Anderson GL, Barney LM, Dunn MM, et al. Traditional criteria for observation of splenic trauma should be challenged. Am Surg 1999;65:689–91.[Medline]
- Sartorelli KH, Frumiento C, Rogers FB, Osler TM. Nonoperative management of hepatic, splenic, and renal injuries in adults with multiple injuries. J Trauma 2000;49:56–62.[Medline]
- Moore EE, Cogbill TH, Jurkovich GJ, Shackford SR, Marangoni MA, Champion HR. Organ injury scaling: spleen and liver (1994 revision). J Trauma 1995;38:323–4.[Medline]
- Sutyak JP, Chiu WC, D'Amelio LF, Amorosa JK, Hammond JS. Computed tomography is inaccurate in estimating the severity of adult splenic injury. J Trauma 1995;39:514–8.[Medline]
- Kohn JS, Clark DE, Isler RJ, Poge CF. Is computed tomographic grading of splenic injury useful in the nonsurgical management of blunt trauma? J Trauma 1994;36:385–9.[Medline]
- Gando S, Tedo I, Kubota M. Posttrauma coagulation and fibrinolysis. Crit Care Med 1992;20:594–600.[Medline]
- Keller MS, Sartorelli KH, Vane DW. Associated head injury should not prevent nonoperative management of spleen or liver injury in children. J Trauma 1996;41:471–5.[Medline]
- Cocanour CS, Moore FA, Ware DN, Marvin RG, Clark JM, Duke JH. Delayed complications of nonoperative management of blunt adult splenic trauma. Arch Surg 1998;133:619–25.[Abstract/Free Full Text]
Top
Abstract
Introduction
