P&S Medical Review: Nov 1993, Vol.1, No.1
Acute Injury to the Adult Spleen: Evolution in Diagnosis and Management

MATTHEW J. KUEHNERT, M.D.
College of Physicians and Surgeons, Columbia University,
New York, NY.

Introduction

The spleen is the most commonly injured organ in blunt abdominal trauma, and unrecognized injury to intra-abdominal contents can lead to increased morbidity and mortality and longer hospital stays.[1],[2] The mean mortality for all patients presenting with splenic trauma ranges between 3-23%.[2] Over the last two decades, the diagnosis and management of splenic injury have changed dramatically.[3] In cases where laparotomy is clearly indicated due to severe blunt or penetrating trauma, direct visualization of the spleen makes diagnosis relatively straightforward. However, the advent of diagnostic peritoneal lavage (DPL), and more recently, CT and ultrasound, have made early diagnosis and conservative management for less severe injuries standard care. Imaging is also currently used to follow the evolution of a splenic injury. Although DPL still plays a crucial role in emergency room management, CT is currently the imaging study of choice for splenic injury. Ultrasound is gaining acceptance because it is less expensive, less invasive, and relatively portable. Management rests not only on the extent of splenic injury, but also on the existence of associated injuries and on the presence or absence of hemodynamic stability. New operative methods have been developed so that splenectomy is indicated for only the most severely injured spleens.

Mechanism Of Injury

Injury to the spleen often leads to massive hemorrhage into the abdominal cavity. Because of its uniquely open circulation, which enables it to play a vital role in the immunologic function of the body, it can bleed profusely with minimal injury.[4] The spleen is surrounded by a tough fibrous capsule, which, when ruptured, results in rapid hemorrhage and a 75% mortality if reparative surgery is not performed immediately.[5] Splenic injury occurs secondary to penetrating trauma and non-penetrating trauma. Rarely, the spleen can rupture spontaneously. Injury may involve disruption of the organžs parenchyma, capsule, or vasculature. While splenic injury occurs without any other intra-abdominal injury in as many as a third of patients, twenty-five percent have other associated injuries. The most commonly associated injuries are neurologic, thoracic, and orthopedic.[6] Associated injuries, patient age, and hemodynamic stability are important factors that will affect morbidity and mortality.[2]

At laparotomy, the spleen is found to be injured in 46% of patients who have sustained blunt trauma.[7] Blunt trauma accounts for 66-75% of all splenic trauma in adults and up to 97% in children.[2] In adults, 50-75% of these cases involve motor vehicle accidents (MVA), while in children and adolescents, injury incurred during contact sports is a more common mechanism.[1],[8]

Splenic injury may also occur during penetrating trauma. In this case, the most commonly associated organ injuries are to the stomach, left kidney, and pancreas. A trajectory may involve the flank or abdominal wall, and entry may occur transthoracically with potential lung, pleural, and diaphragmatic injury. Penetrating injury is associated with other serious abdominal injuries in over 90% of cases, and exploratory laparotomy is advocated for all gunshot injuries. In suspected stab wounds, however, local wound exploration, evaluation for intra-abdominal penetration, and serial abdominal exams are believed to be sufficient, provided hypotension and signs of peritoneal irritation are absent.[9]

Iatrogenic splenic injury can occur during surgical intervention and invasive diagnostic procedures and accounts for 20% of all splenectomies. Avulsion of the splenic capsule resulting from traction is the most common form of splenic injury that occurs at surgery. The spleen can also be directly injured as a result of manipulation, thoracentesis, or left renal biopsy. Since mortality after this so-called "incidental" splenectomy is above 10%, splenic salvage is strongly advocated in these cases.

Rarely, the spleen can rupture spontaneously without evidence of trauma. This is more likely to occur in patients with splenomegaly secondary to disease states which include infectious mononucleosis, malaria, leukemia, typhoid, sarcoidosis, hemolytic anemia, polycythemia vera, and splenic abscess[4]. However, unrecalled minor truama is still a more likely mechanism in patients with these diseases. Splenic rupture is also a rare complication of anticoagulant therapy.

While recent reports claim its incidence to be <1%, delayed rupture of the spleen after trauma can occur from either the release of tamponade by structures surrounding a minor laceration or a slowly enlarging subcapsular hematoma that ruptures.[4] The quiescent period is less than 2 weeks in the majority of these cases.

Diagnosis and Initial Management

In the past, diagnosis of splenic injury was based solely on the patient's history and physical signs. However, the signs and symptoms of intra-abdominal injury are often unreliable and are often masked by head injury, major fractures, and the effects of substance abuse. As the role of the spleen in cell-mediated and humoral immunity has become clearer, so has the importance of ascertaining whether the extent of injury allows nonoperative management.[5] Toward this end, the advent of a number of diagnostic methods has provided the clinician with an increasing number of alternative methods to assess patients. Laparotomy is indicated for all gunshot wounds, hypotension on hospital arrival, and the presence of major chest trauma, as these parameters are strongly correlated with major abdominal visceral injury.[9] An arterial base deficit > 3 meq/L is also correlated with major trauma but by itself is not universally considered an absolute indication for operative intervention.

It is important to construct as detailed a history as possible. If the patient is not able to respond, specific details concerning the trauma can be ascertained from bystanders, police, and emergency personnel who were at the scene of injury. Crushing injuries associated with lap seat belts alone, for instance, are distinct from those associated with a lap belt and shoulder harness.[1] Lateral impact collisions may increase suspicion of splenic trauma. A sudden deceleration injury from a fall or MVA may cause tearing of the splenic artery and its branches.[8] Witnesses can also often provide information regarding the patient's level of consciousness immediately after the accident as well as likelihood of substance abuse.[1] If family or friends are present, a quick survey of current medications, past medical history, and allergies (especially to antibiotics and contrast media) is important in deciding subsequent management.

Patients may experience left upper quadrant (LUQ) tenderness, referred left shoulder pain (Kehržs sign, commonly enhanced by Trendelenburg position), or rarely, a fixed area of percussible dullness in the LUQ (Balancežs sign). Inspection for ecchymoses and abrasions can provide clues to internal hemorrhage. Point tenderness is an important sign becuase there is a 20% chance of splenic injury with fractures of any of the left six ribs.[1] Abdominal tenderness and guarding is indicative of peritonitis due to intraperitoneal blood or visceral perforation. Patients with systolic blood pressures under 100 mm Hg should be classified as hemodynamically unstable and their blood should be typed in preparation for possible transfusion. If hypotension continues after infusion of 2 liters of crystalloid, packed red blood cells should be infused until hemodynamic stability is restored.[9] Baseline hematocrit, electrolytes, BUN/creatinine and a coagulation screen should be obtained. Elevated white count and amylase are often nonspecific in the setting of intra-abdominal trauma[1]. Initial management in the emergency room should include frequent monitoring of vital signs and hemocrit. A urinalysis should be performed to detect blood and toxic metabolites in the urine. Placement of a nasogastric tube as well as insertion of a Foley catheter is recommended to decompress the stomach and bladder, and is a requirement if DPL is to be performed. Intravenous antibiotics can be given prophylactically in anticipation of both aerobic and anaerobic infections that can accompany splenic hematoma[10].

Findings on plain radiographs, such as left lower rib fractures, pleural effusion, displacement of the gastric air bubble, or elevation of the left hemidiaphragm, can help in the diagnosis of splenic trauma. A minimum of 800 ml of intraperitoneal blood is required to be evident on an abdominal film. Therefore clinical signs of abdominal hemorrhage are usually more sensitive.[1] However, evidence of previous splenic trauma may be detected in the form of calcification from old hematomas.[5]

Diagnostic peritoneal lavage (DPL) was developed in the 1960's as an adjunct to history and physical exam in the detection of hemoperitoneum at a time when exploratory laparotomy was the only other alternative. Physical exam has an accuracy of 65% in the diagnosis of hemoperitoneum overall, and of 15% in patients who have no symptoms suggestive of injury. [1],[9] DPL is inexpensive, rapid and has a high specificity and very high sensitivity. Analysis of fluid recovered should include red and white blood cell counts, amylase, alkaline phosphatase, and examination for bile, bacteria, or food fibers. Controversy exists as to the thresholds for a positive result, but an intraperitoneal bleed is present in 95% of patients whose RBC count is greater than 100,000/mm[1] and is absent in 95% of patients whose RBC count is less than 20,000 /mm[3]. [1] Guidelines for the interpretation of DPL results are presented in table 1. Relative contraindications include previous abdominal surgery (as adhesions can create loculations), pregnancy, and massive obesity.

===========================================================
TABLE 1.  INTERPRETATION OF DPL DATA
===========================================================
PARAMETER                     POSITIVE            EQUIVOCAL

Red Blood Cell  count         >100,000            >50,000  
  (mm-[3])
White Blood Cell count        >500                >200
  (mm-[3])
Amylase                       >=20             
  (IU/L)
Alk Phos                      >=3                           
  (IU/L)
Bile                          +         
  (confirmed biochemically)
===========================================================

CT is currently the imaging study of choice for the assessment of splenic injury in patients who are hemodynamically stable. CT scanning can identify injuries to all abdominal organ systems, the retroperitoneum, and bones in a patient in whom intra-abdominal injury is likely. Thus, CT has a clear advantage over DPL as it can identify specific injury and is also noninvasive. However, as with DPL, criteria for a positive study must be agreed upon before the study is useful. In the setting of blunt abdominal trauma, the CT is positive if there is hemoperitoneum, retroperitoneal hematoma, or splenic, hepatic, pancreatic, renal, or hollow viscus injury.[8] Hemoperitoneum can be detected on CT with blood volume > 100 ml[3].[8] Blood reaches maximum attenuation at approximately 20 hours after injury and then stabilizes at a lower attenuation over the next 48 hours. However, repeat scans with thinner slices may have to be obtained to give accurate staging of the injury. Hemodynamic changes caused by trauma and resuscitation can delay the appearance of IV contrast in the spleen and create a mottled enhancement unless there is a one minute delay between infusion and scanning.[6] CT is accurate in determining injury but cannot reliably predict the best course of management without clinical correlation.[2]

Studies comparing CT to DPL have been primarily retrospective and without a "gold standard" to compare the two if the patients are managed nonoperatively. An exception is a study by Mayer et al in which 301 hemodynamically stable blunt trauma patients with equivocal abdominal exams underwent CT followed by DPL and laparotomy.[11] Fifty-one patients had both a positive CT and DPL as defined above, and of these, 7 had additional injuries not detected on initial CT scan, including splenic injuries. Furthermore, 19 patients with positive DPL and negative CT had significant abdominal injuries found on laparotomy. There were three false negative lavages and two complications attributed to DPL. DPL was significantly more sensitive than CT (95.9% vs. 74.3%) while specificity was similar (99% for DPL vs. 99.5% for CT) for significant hemoperitoneum and/or parenchymal injury. In light of the controversy surrounding the sensitivity of DPL vs. CT, some have suggested that the two be used in a complementary fashion in the stable patient with CT indicated if DPL is equivocal, especially if it is likely that there are multiple injuries present.[8],[2]

Ultrasound is gaining acceptance in detecting splenic injury, especially in situations in which DPL is relatively contraindicated. It is noninvasive, fully portable, and is particularly suited to detect abnormal fluid collections. Criteria for splenic injury include splenic enlargement, change in the contour of the splenic border, and the appearance of perisplenic fluid. Abnormal fluid collections, including hemoperitoneum, can be readily detected with few false positives and rare false negatives. One European study found that ultrasound was able to detect intraperitoneal blood in 25 of 25 patients and isolated parenchymal injuries in 22 of 25 patients.[11] Ultrasound can also be used to follow patients managed nonoperatively. It is ideal for patients who have had previous abdominal surgery, where DPL may be nondiagnostic. It is also ideal for the pregnant patient as it involves neither radiation nor an invasive procedure.[11] Ultrasound, however, is a poor study for hollow viscus injuries because of the presence of air.

Although popular in Europe, ultrasonography has been slow in gaining acceptance in the United States, in part because of the amount of training required of its users. In one study comparing ultrasound to DPL, ultrasound and DPL both had a false positive rate of 1.9% while ultrasound had a false negative rate of 4.7% compared to 2.8% for DPL.[12] In a followup two years later, prospective analysis showed DPL to have a sensitivity of 93% while that of ultrasonography was found to be 88%. However the sensitivity of ultrasound rose from 84 to 96% in the three years during the study. [7] Jarowenko et al found that out of 1,038 blunt trauma patients evaluated by ultrasound, there were 6 false negative studies and no false positives.[5] Oller et al recommend that a negative ultrasound study be followed by other diagnostic methods if splenic injury is suspected by clinical signs and symptoms.[10]

Nuclear scintigraphy with technetium sulphur colloid is also used to locate injury. Although not the diagnostic tool of choice, it may still have a role in the stable but uncooperative patient with local trauma to the LUQ. It may be especially useful in a patient with coexisting contrast allergy and skin injury, which would preclude both CT and ultrasound. Nuclear scintigraphy can also be used to follow return of splenic function.[2] The resolution of this method, however, is significantly less than that of CT.[9] Angiography is also used to diagnose injury, but less invasive studies are more efficient and accurate. MRI provides the best anatomical resolution but is contraindicated in an emergency setting.

In summary, in the acute setting for diagnosis of adult splenic trauma, DPL is recommended for the stable patient with equivocal physical exam for hemoperitoneum and for patients needing immediate treatment for associated injuries. CT is indicated for the stable patient in whom DPL and physical exam are equivocal. Ultrasound is controversial as an initial screening tool for splenic injury.[8] Ultrasound is recommended presently as an adjunct to CT where DPL is relatively contraindicated, but is being accepted increasingly as a cost effective alternative to CT in the patient with isolated injury.[10]

Classification of Injury

Because classification of splenic injury determines subsequent management, many distinctions based on both CT imaging and operative information have been artificially erected to attempt to predict which injuries should be best managed surgically or conservatively. After reviewing the numerous and often conflicting grading systems used to stage these injuries, the Organ Injury Scaling Committee of the American Association for the Surgery of Trauma proposed a standardized classification system (Table 2).[9]

Management

In the past, splenic injury had been an absolute indication for splenectomy. However, the newly recognized role of the spleen in intravascular antigen clearance, especially for encapsulated organisms, has resulted in a chnage in the recommendations for management.[9] The high incidence of both overwhelming post-splenectomy infection (OPSI) and late septic morbidity in both children and adults after splenectomy has led experts to develop new techniques for splenic salvage and guidelines for nonoperative management.

OPSI can occur at any time after surgery. The patient typically presents with sepsis, with the prominent signs being sudden onset of nausea, vomiting, and mental confusion. Disseminated intravascular coagulaton, coma, and death can occur within 24 hours. OPSI has a mortality of 50-75%[2]. The risk of OPSI is lower in adults than in children. Incidence in adults ranges from 0-2.2% post-splenectomy.[2] It is controversial whether the incidence of OPSI after splenectomy due to trauma significantly differs from this.[9] When splenectomy is performed due to splenic injury, morbidity and mortality from infection is believed to be lower because of intraperitoneal implantation of splenic material seeded during the traumatic event. This process, called splenosis, occurs in 26-66% of cases,[2] and the incidence of OPSI ranges from .28 to 1% in these patients.[9] Pneumococcus is the organism responsible for OPSI in the majority of cases but meningiococcus and H. influenza are also often involved.[4] Administration of polyvalent pneumococcal vaccine is recommended, although its benefits are controversial, especially where traumatic injury is involved.[2] Studies involving patients post-splenectomy for all indications, have shown prophylactic penicillin to be of benefit in reducing the incidence of pneumococcal sepsis only when taken for at least 3 years after splenectomy, which creates a potential problem for patient compliance.[4],[2]

=================================================================
TABLE 2.  CLASSIFICATION OF SPLENIC INJURY,
          ORGAN INJURY SCALING COMMITTEE - 1987
=================================================================
GRADE          INJURY DESCRIPTION

I
  Hematoma     Subcapsular, nonexpanding, <10% surface area
  Laceration   Capsular tear, nonbleeding, <1 cm parenchymal depth 
II
  Hematoma     Subcapsular, nonexpanding, 10-50% surface area or
               intraparenchymal, nonexpanding <2 cm in diameter
  Laceration   Capsular tear, active bleeding or 1-3 cm
               parenchymal depth that does not involve a
               trabecular vessel
III
  Hematoma     Subcapsular, either nonexpanding >50% surface area
               or expanding
  Laceration   >3 cm parenchymal depth or involving trabecular
               vessels
IV
  Hematoma     Ruptured intraparenchymal hematoma with active
               bleeding
  Laceration   Laceration involving segmental or hilar vessels
               producing major devascularization (>25% of spleen)
V
  Laceration   Completely shattered spleen
  Vascular     Hilar vascular injury that devascularizes spleen
=================================================================

The class I and a proportion of the less severe class II injuries can often be managed conservatively unless there are associated injuries that mask examination for acute abdomen or which create hemodynamic instability. Non-operative management is now preferred for these injuries as current imaging studies have become better able to detect and evaluate injury in the stable patient. In adults, only 15-20% of splenic injuries can be managed non-operatively. Vital signs and hematocrit of these patients need to be followed closely in the ICU, with a physical exam repeated every 4-6 hours. The requirement of more than two units of blood to keep the hemoglobin above 8 g/dl in the initial 48 hours is an indication for operative management.[4] A CT scan should be repeated at 12 hours, on day 3, and day 7 after admission. CT may show an abdominal visceral injury not appreciated on the initial scan, and may be used to estimate whether there has been continued intraperitoneal bleeding. Strict bed rest for one week should be followed by avoidance of both strenuous activity for 6-8 weeks and contact recreation for at least 6 months or until CT shows resolution of the lesion.[4],[9] Ultrasound is gaining acceptance as an alternative to CT in following resolution, especially in Europe.[10]

Success rates of nonoperative management have been reported to be as high as 83% using these recommendations with no related morbidity or mortality on followup.[2] However, the overall success rate varies from 20-100% depending on the particular protocol used. Some point to the lack of a seven-day strict bed rest order to explain the high failure rate of conservative management in some reports.[10] Conservative management is successful in over 90% of pediatric cases. This high success rate is thought to be partially due to the hemostatic properties of a proportionally thicker splenic capsule as compared to adults.[9]

The benefits of conservative management are negated if the potential risks of hepatitis C, HIV, and transfusion reaction outweigh the potential risk of the post-splenectomy state. Luna and Dellinger have calculated a risk benefit analysis and have suggested that the risk of death from transfusion related complications exceeds that of post splenectomy infection when more than 2 units of blood are given.[13] Although children have traditionally been allowed multiple transfusions, some experts disagree and favor operative management over repeated transfusion. Non-operative management is also contraindicated in patients with previous splenic disease, mental status change, or associated intra-abdominal injuries. In addition, the introduction of autotransfusers or "cell savers" for use during laparotomy should make operative management more attractive, as these devices allow the recovery of intraperitoneal blood.[3] Controversy exists whether advanced age is a relative contraindication to conservative mangagement.[10]

The severe Class II and the Class III injuries have better outcomes with splenorrhaphy, and class IV injuries often require segmental splenectomy. The splenic artery divides into between 6 and 36 branches before entering the splenic parenchyma, providing segmental perfusion and allowing for partial resection in focal injury. The splenic cords mark the termination of the central arteries and lack a vascular endothelium. This creates an optimal biological filter, but it also can make bleeding profuse and nonfocal.[9] Salvage can be accomplished in a variety of ways including suture repair, electrocautery, hemostatic agents, Dexon mesh, and "fibrin glue."[4] The omentum can also be sutured to the spleen as a tamponade. The mean success rate for repair is 45%, with a wide range reported.[2]

In a study of operative splenic salvage in 107 adults over ten years at the Denver General Hospital,[7] all Grade I injuries and 57% of Grade II injuries were repaired using hemostatic agents only. The rest of the Grade II injuries and all Grade III injuries required suture or mesh bag. Of the grade IV injuries, 80% were treated with segmental splenectomy, 8% with total splenectomy, and 12% with suture repair. All Stage V patients underwent total splenectomy. There was a 3% rate of splenic rebleeding, and otherwise there were no complications related to splenic salvage. Four patients undergoing splenorrhaphy (4%) died, two from massive head injury, one from cardiac arrhythmia, and one from multiple organ failure.

Class V splenic injuries require splenectomy. Splenectomy is also favored over splenorrhaphy for patients with lesser injuries but with medical contraindications to prolonged surgery. The most common complications are pneumonia, pneumothorax, pulmonary effusion, subphrenic abscess, and thromboembolism. Of the above complications, only the subphrenic abscess has a greater incidence with splenorrhaphy than with splenectomy.[2] Initial management includes packing and pressure with rapid mobilization of the spleen to the incision site. Drainage is controversial and is recommended only after massive contamination or questionable hemostasis as there is a high rate of subphrenic abscess from skin flora.[9] Some splenic function can be preserved by autotransplantation. One of the more innovative methods involves cutting the spleen into small fragments and placing them into an omental pocket secured by through and through sutures, creating a "splenic omelet."[4] If we exclude OPSI, the rate of late sepsis in all patients following splenectomy is around 2.5%, and the risk of late major septic morbidity has been shown to be increased in patients having had splenectomy for trauma. After following 144 trauma patients for an average of five years, it was estimated that the risk of septicemia following splenectomy in trauma patients is 140 times that of normals.[14] The potential late complications of splenectomy are also increased in trauma patients.[9] These complications include increased morbidity and mortality from pneumonia, purulent pyelonephritis, and a fourfold increase in the incidence of pulmonary embolism.[2],[9] Differences in perioperative mortality between splenectomy and salvage may reflect associated injuries in the Class IV injury rather than success or failure of the procedure itself.[2]

Comparison between operative and nonoperative management is difficult due to the multitude of grading schemes for both CT-guided and surgical management. Most grading schemes guiding nonoperative management have been based on CT. However, since Mirvis et al found that CT is accurate at identifying injury but not at reliably predicting outcome, clinical evaluation rather than staging has been used in management.[15] In fact, some authors have urged that the hemodynamically stable victim of blunt trauma with a documented isolated splenic injury be treated nonoperatively regardless of stage of injury.[13]

In summary, the physician must base diagnosis and management of a patient with splenic injury primarily on clinical grounds. DPL, CT, and ultrasound are important diagnostic tools which can be critical in determining the presence of isolated and associated injuries, and hence, mode of management. The stable patient, although a candidate for nonoperative management, must be monitored carefully for indications for operative intervention. There are risks and benefits of both nonoperative and operative management which must be carefully considered before this difficult decision is made.

REFERENCES

1. McAnena OJ, Moore EE, Marx JA. Initial evaluation of the patient with Blunt abdominal trauma. Surg Clin of North Am 1990;70:495.

2. Wilson RH, Moorehead RJ. Management of splenic trauma. Injury 1992;23:5.

3. Witte CL, Esser MJ, Rappaport WD. Updating the management of salvageable splenic injury. Ann Surg 1992;215:261.

4. Lucas CE. Splenic trauma: choice of management. Ann Surg 1991;213:98.

5. Rolfes RJ, Ros PR. The Spleen: An integrated imaging approach. Crit Rev Diagn Imaging 1990;30:41.

6. Gay SB, Sistrom CL. Computed tomographic evaluation of blunt abdominal trauma. Radiol Clin North Am 1992;30:367.

7. Pickhardt B, Moore EE, Moore FA, McCroskey BL, Moore GE. Operative splenic salvage in adults: a decade perspective. J Trauma 1989;29:1386.

8. Thal ER, Meyer DM. The evaluation of blunt abdominal trauma: CT, lavage, or sonography? Adv Surg 1991;24:201.

9. Shackford SR, Molin M. Management of splenic injuries. Surg Clin North Am 1990;70:595.

10. Oller B, Armengol M, Camps I, et al. Nonoperative management of splenic injuries. Am Surg 1991;57:409.

11. Halbfass HJ, Wimmer B, Hauenstein K, et al. Ultrasound diagnosis of blunt abdominal injuries. Fortschr Med 1981;99:1681.

12. Peiper HJ. Significance of traumatology in abdominal and vascular surgery. Jpn J Surg 1985;15:93.

13. Villalba MR, Howells GA, Lucas RJ, et al. Nonoperative management of the adult ruptured spleen. Arch Surg1990;125:836.

14. Green JB, Shackford SR, Sise MJ, et al. Late septic complications in adults following splenectomy for trauma: a prospective analysis of 144 patients. J Trauma 1986;26:999.

15. Mirvis SE, Whitley NO, Gens DR. Blunt splenic tauma in adults: CT-based classification and correlation with prognosis and treatment. Radiology 1989;171:33.