Br. J. Surg. 1992, Vol. 79, December, 1330-1 333

W. Schweizer, L. Bohlen, A. Dennison and L. H. Blumgart Department of Visceral and Transplantation Surgery, University of Berne, lnselspital, CH-3010 Berne. Switzerland Correspondence t o : Dr W. Schweizer

Prospective study in adults of splenic preservation after traumatic rupture Seventy-Jive adults with splenic injury were evaluated prospectively over 45 months to examine the possibility of splenic preservation. Haemodynamically unstable patients underwent surgery with the intent of splenic preservation. Stable patients received non-operative treatment regardless of the grade of splenic injury determined by ultrasonography and computed tomography. Thirty-seven patients required splenectomy and in 38 the organ was preserved (20 operative preservation, 18 non-operative treatment). Of 22 patients initially receiving non-operative treatment, there were four secondary haemorrhages after 7, 7, 10 and 13 days making surgery necessary. Three of these patients underwent splenectomy and in one the spleen was preserved by partial resection. After splenectomy four patients required reoperation because of rebleeding or f o r evacuation of a haematoma. Patients who had undergone splenectomy had a significantly increased infection rate (P < 0.005) compared with those in whom the spleen was preserved, even when patients were matched with respect to multiple trauma using the Injury Severity Score ( P < 0.01).

The overwhelming postsplenectomy sepsis syndrome' was first described in 1952. Higher infection rates in children and adultszL4and haematological disturbance^^.^ after splenectomy have since been recognized. This has led to attempts at splenic autotransplantation5-' and splenic preservation in situ*L22. Autotransplantation has been shown experimentally'8~20*21 and ~ l i n i c a l l y ~ to ' ~ be ~ ~of- ~little ~ value; organ preservation in situ is, in contrast, increasingly advocated and appears to be safe8-' '. There have recently been reports of partial splenic resection after trauma8-' ' . Improved diagnostic facilities, better knowledge of the methods of non-operative treatment in adults12,22,and close follow-up have led to an increasing number of splenic preservation procedures. In this study, different diagnostic and therapeutic regimens after splenic injury were evaluated prospectively and early complication rates compared. '3'

Patients and methods Between October 1986 and June 1990,75 patients (18 women, 57 men; mean age 35 (range 16-78) years) with splenic injuries from blunt abdominal trauma were studied prospectively. Haemodynamic status on admission was used to determine the choice of treatment. Stable patients received non-operative treatment; unstable patients underwent splenorrhaphy, partial splenic resection or splenectomy. In stable patients with a shock index (pulse rate/systolic blood pressure) i 1 on admission and an initial stable phase (12000 ml electrolyte solution and/or < 2 units packed red cells administered) of at least 2 h a conservative approach was adopted. Ultrasonography and/or computed tomography (CT) were used for accurate assessment of the severity of splenic and concomitant injuries. Peritoneal lavage was used in patients who required operative procedures for extra-abdominal injuries because it could be repeated in theatre if the patient became clinically unstable. It was employed in a few patients when ultrasonography was not immediately available. On the basis of the ultrasonography, CT and/or intraoperative findings, the splenic injury was graded I-V (Table I ). The severity of multiple trauma was assessed using the Injury Severity Score (ISS). Patients unsuitable for non-operative treatment underwent laparotomy with the intention of splenic preservation by splenorrhaphy or partial resection. The decision was left to the operating surgeon. The surgical technique of splenic preservation has been described in a technical protocolz6 and recorded in a videotape". The hilum was clamped with a soft bowel clamp and methods for repair depended

on the grade of injury. Resorbable collagen platelets, resorbable gauze, polytetrafluoroethylene strips and a variety of methods to achieve haemostasis (infra-red photocoagulation, haemostatic material and supportive mesh ) were used. In patients undergoing partial resection with lower pole injuries extending into the hilum, the spleen was preserved on the short gastric vessels with or without preservation of an upper pole artery. The variable anatomy of the arteries at the hilum determined the type of resection16 (Figure I ). Single-dose antibiotic prophylaxis with 2 g cephazolin was given immediately before surgery. In patients undergoing splenectomy, cephazolin was continued for 3-5 days (1 g four times daily), when one dose of benzathine benzylpenicillin (2.4 x lo6 units) was given intramuscularly. Polyvalent Pneumococcus vaccine was given on the tenth day after operation. Antibiotics were otherwise given only for proven infections. All patients were assessed for early infection and only those with pathognomonic clinical and radiological findings supported by positive bacteriological culture were included. Identical bacteriological surveillance was undertaken in all groups. Statistical analysis was performed using the x2 test.

Results Diagnosis of intra-abdominal trauma with suspected splenic injury was made by ultrasonography in 44 patients and by C T in 20. Peritoneal lavage was carried out in 22 patients and in four laparotomy was performed on clinical grounds. The diagnosis and grading of splenic injury in patients receiving non-operative treatment was made by C T in 16 and by ultrasonography in six, the latter mainly with minor injuries of grades I and 11, where ultrasonography seemed reliable and

Table 1 The Splenic Injury Scale Grade

Injury

I I1 IIIa IIIb

Subcapsular haematoma Capsular tear Superficial parenchymal tear ( i1 cm) Deep parenchymal tear ( 1 cm) without involvement of hilum Parenchymal tear with involvement of hilum Fragmentation

IV V

L

1330

~

_

_

~ ~ ~ _ _ _ _ _ _ _ _ _ _ _ _

0007-1123,92 12133CL04

(

1992 Butterworth-Heinemann Ltd

__

Splenic preservation after traumatic rupture: W . Schweizer et al.

sufficient. Follow-up of all patients treated conservatively was by repeated C T examination. The management oCthe 75 patients is shown in Figure 2. A total of 37 patients required spleaectomy, 34 performed primarily and three secondarily. In 22 patients, primary treatment was non-operative. Secondary bleeding occurred after 7, 7, 10 and 13 days in four such patients, three of whom underwent splenectomy. The fourth patient underwent partial splenic preservation. A total of 38 spleens were preserved, 18 without operation and 20 following surgery. In 13 patients the

Br. J. Surg., Vol. 79, No. 12, December 1992

spleen was preserved as a whole and in seven partial resection was necessary. In three patients, only the upper pole of the spleen was preserved, in two supplied only by the short gastric vessels and in the third by the upper polc artery (Figure I ) . Partial splenic resection was performed in one grade 11, two grade 111, three grade 1V and one grade V injuries. The mean increase in the operative time for preservation measured from clamping of the vessels at the hilum until repositioning of the spleen in the subdiaphragmatic cavity was 18 (range 9-26) min. There were nine complications from bleeding. Four were

1331

-

Splenic preservation after traumatic rupture: W . Schweizer et at.

Splenic injuries (75)

I Non-operative treatment ( 2 2 )

I

I

Operative preservation (21)

S p l e n e c t o m y (32)

+ l

18

Secondary bleeding (4)

I

19 ( s e c o n d a r y b l e e d i n g , same a n a e s t h e t i c , in 1)

lntraoperative complications (2)

1,1\ N o n - o p e r a t i v e t r e a t m e n t (18)

Operative preservation (20)

of 75 patients

Splenectomy (37) Splenectomy (37)

Splenic preservation (38)

Figure 2 Treatment

I

3

with splenic trauma

Table 2 Infections und their causative organisms in patients with splenic trauma Treatment

Infection

Organism

Splenectomy ( 15 )

Respiratory tract (10)

Haemophilus influenme ( 5 ) Staphylococcus aureus ( 2 ) Enrerobncter ( 2 ) Klebsiella ( 2 ) Pro teus m irabilis Pseudomonas aeruginosa Citrobacter Serratia marcescens Pneumococcus

Sepsis ( 4 ) (positive blood culture)

Pseudomonas aeruginosa ( 2 ) Enterococcus (Streptococcus faecalis Klebsiella

Infected abdominal haematoma ( 2 )

Serratiu marcescens Pseudomonas ueruginosu Actinobacter Enterococcus (Streptococcus jaeca1i.r Proteus mirabilis

Urinary tract ( 1 )

Escherichia coli

Respiratory tract ( 1 )

Haemophilus injuenzae

Urinary tract ( 1 )

Staphylococcus epidermidis

Respiratory tract ( 2 )

Haemophilus injuenzae Pseudomonas cepacia Staphylococcus aureus

Operative preservation ( 2 )

Non-operative ( 2 )

20

2

._ 5 15

5m

VI t-

c

m

a 10 L

0

.

5

0

=

n

12

10

8

Q

6

c O

4

z " 2

n

Figure 3 a Injury Severity Score ( I S S ) and b splenic injury grade for patients undergoing non-operatitie treatment ( ), operative preservation (B ) or splenectomy (0) for splenic trauma

the secondary splenic ruptures in patients receiving nonoperative treatment. In one patient undergoing partial splenic resection, rebleeding occurred immediately after closure of the abdomen; this was treated by immediate reoperation. Four rebleeds after splenectomy were from the splenic bed; no patient requiring reoperation suffered a complication as a consequence.

1332

Two patients developed a pancreatic fistula after splenectomy but both settled with conservative treatment within a few days. Septic and infective problems are shown in Table2. The infection rate was significantly lower in patients undergoing splenic preservation (four of 38 uersus 15 of 37 patients having splenectomy, P < 0.005). After matching patients for ISS, 12 proven infections occurred in 33 submitted to splenectomy compared with four of 38 having splenic preservation ( P < 0.01). Five patients in the series died: three after splenectomy and two after operative preservation. Two patients died after splenectomy from acute respiratory distress syndrome (ARDS ) and one from a cerebral injury. Of the patients with splenic preservation, one died from ARDS and the other from a pulmonary embolus. Details for all 75 patients divided into the different treatment groups according to ISS and grade of splenic injury are shown in Figure 3.

Discussion There is increasing evidence of a raised late mortality rate after splenectomyz-4, but early complications after splenic injury have rarely been studied2,28.

Br. J. Surg., Vol. 79, No. 12, December 1992

Splenic preservation after traumatic rupture: W. Schweizer e t al.

The present study demonstrates the ability to preserve an increasing number of traumatized spleens in adults by surgical preservation or non-operative therapy. This strategy has previously been used mainly in children' ' * 1 6 . This change has occurred through increasing experience with operative preservation and an ability and confidence to pursue a nonoperative approach based on accurate diagnostic methodsz2. In haemodynamically unstable patients, operative preservation is the treatment of choice. Partial resection is employed in cases where the lower pole of the spleen is more traumatized because of its relatively exposed position. An upper pole artery is present in about 60 per cent of patients29330,making upper pole splenic preservation technically feasible (Figure I ). Splenic preservation is a safe technique' 3,'8,20-z2,29 , as this study confirms; there are no more bleeding complications than after splenectomy. A small number of patients treated without operation develop secondary bleeding. In the present series this occurred in four of 22 patients receiving primary non-operative treatment. Patients in the group so treated who developed secondary rupture were all older than 40 years and the non-operative approach may not be so safe in those over this age. The secondary rupture is often dramatic, requiring immediate surgery; most occur within the first 2 weeks following injury".' '. All patients treated without operation should be kept in hospital for at least 10 days and routine ultrasonography or CT carried out before discharge. If there is persisting intra-abdominal fluid collection, additional C T with intravenous contrast should be performed. The patient should not play strenuous sports for at least 3 months and must report immediately if there is sudden left upper abdominal pain. Relatives and general practitioners should be similarly informed. Some 3-6 months after injury, further ultrasonography and/or CT should be carried out. In this series even when patients were matched for TSS there was a significant difference in the rate of early postoperative infections, which were caused by capsulated bacteria (Tahlr 2). In view of the early complications after splenectomy demonstrated by the present study and and the late complications' 4 . 1 4 ~ ' 7 ~ ' 8 . 2 0 ~ 2preservation 1, is the treatment of choice for splenic trauma. The increase in operating time is minimal. Non-operative treatment may be valuable in younger patients with limited injuries. Experienced teams should be able to preserve the spleen in 60-80 per cent of all patients.

'

5.

6. 7. 8 9. 10.

11.

12. 13. 14.

15.

16. 17. 18. 19. 20. 21.

22. 23. 24.

Acknowledgements

25.

The authors thank Mr H. Holzherr for drawing Figure I and Mrs J. Andres and Mrs U. Whincop for secretarial help.

26. 27.

References I.

2. 3. 4

King H, Schumacker HP. Spleen studies. I : Susceptibility to infection after splenectomy performed in infancy. A n n Surg 1952; 136: 239-49. 0'Ncal BJ, McDonald JC. The risk or sepsis in asplenic adults. A m Surq 1981; 194: 775-8. Pimp1 E, Dapunt 0, Kaindl H. Thalhamcr J . Incidence of septic and thromboembolic-related deaths after splenectomy in adults. Br J Suriq 1989; 76: 517-21. Robinette C D , Fraumeni JF. Splenectomy and subsequent mortality in veterans of the 1939 1945 war. Lirnwl 1977; ii: I27--9.

28. 29. 30.

Cooney DR, Swanson SE, Dearth JC, Dewanjee M K , Telander RL. Splenic auto-transplantation in prevention of overwhelming postsplenectomy infection. J Pediarr Sury 1979; 14: 336-42. Vega A, Howell C , Krasna I et a/. Splenic autotransplantation: optimal functional factors. J Pediutr Surg 1981; 16: 898-904. Hohenberger W, Haupt W, Kalden JR, Simon M , Mahlstedt J. Die Autologe Replantation von Milzpartikeln - ein etabliertes Verfahren? Chirurg 1985; 56:659-62. Feliciano DV, Bitondo CG, Mattox KL, Ramisek JD, Burch J M , Jordan G L . A four-year experience with splenectomy LIersus splenorrhaphy. Ann Surg 1985; 201: 568-75. Giulkano AE, Lim RC. Is splenic salvage safe in the traumatized patient? Arch Sury 1981; 116: 651-6. Gosh S, Symes J M , Walsh T H . Splenic repair for trauma. Br J Surg 1988; 75: 1139-40. Lambrecht W, Heller M . Organerhaltende Therapie der Kindlichen Milzruptur. UnfUNchirurgie 1984; 10: 66-72. The Splenic Injury Study Group. Splenic injury: a prospective multicentre study on non-operative and operative treatment. Br J Surg 1987; 74: 310-13. Biiyiikiinal C, Danismend N, Yeker D. Spleen-saving procedures in paediatric splenic trauma. Br J Surg 1987; 74: 350-2. Di Cataldo A. Puleo S, Li Destri G a/. Splenic trauma and overwhelming postsplenectomy infection. Br J Surg 1987; 74: 343-5. Bisteff EL, Benton Adkins R. Splenic traum management. South Med J 1984; 77: 1284-90. Roth H, Daum R, Benz G . Stadieneinteilung der Milzruptur chirurgische Konsequenzen im Kindesalter. Chirurg 1986; 57: 194-7. Diirig M, Harder F. Auswirkungen der Splenektomie. Cliirurg 1986; 57: 189-93. Diirig M, Harder F. Die Splenektomie und ihre Alternativen. In: Burri C, Harder F, Bauer R, eds. Aktue4e Prohkenie in C/iiruriqif,undOr/liopuc/ir.Stuttgart: Hans Huber, 1985: 8-1 16. Cooper MJ, Williamson RCN. Splenectomy : indications, hazards and alternatives. Br J Surcj 1984; 71: 173-80. Oaks D D . Splenic trauma. Curr Prohl SurU 1981; 18: 342-401. Perry JF. Injuries of the spleen. Curr f r o b / Suuq 1988; 25: 757-859. Williams MD. Young DH, Schiller WR. Trend toward nonoperative management of splenic injuries. A m J Sung 1990: 160: 588-93. Moore GE, Stevens RE, Moore EE, Aragon G E . Failure of splenic implants to protect against fatal post-splenectomy infection. An' J Surg 1983; 146: 413-14. Traub A, Giebink GS, Smith C c / d.Splenic reticuloendothelial function after splenectomy, spleen repair and splenic autotransplantation. N Enill J M i d 1987; 317: 1559-64. Seufert RM. Die Milztransplantation Standortbestimmung. Cliirury 1986; 57: 182-8. Schweizer W, Bohlen L,Gilg M, Blumgart LH. Technical aspects of splenic preservation in splenic injury. Hi+ Chir A ~ ~ r t1991: r 58: 137-41. Schweizer W, Bohlen L, Gilg M. Kipfer B. Blumgart L H . Splcwic Prcscwuriori iri Truirnitr. Zurich: Videothek Swiss Surgical Society, 1990. Ziemski JM, Rudowski WJ. Jaskowiak W, Rusiniak L, S c h d R . Evaluation of early post-splenectomy complications. S r q GI~ricwdOhstct 1987; 165: 507-14. Strcicher H-J. Anatomiegcrechte Chirurgie der Milz. C/iirur