Splenic Infarct

The arterial supply to the spleen consists of the splenic artery (a branch of the celiac axis) and the short gastric arteries (branches of the left gastroepiploic artery), which supply the upper pole of the spleen. Even with occlusion of the main splenic artery, collateral flow from the short gastric arteries often may preserve some or all of the splenic parenchyma.

Within the spleen, the arterial supply is segmental. Occlusion of these secondary branches results in the classic wedge-shaped infarct. Most commonly, these infarcts contract and fibrose over time, as demonstrated by the sickle hemoglobinopathies (in which repeated episodes of infarction ultimately result in autoinfarction of the spleen).[2]

Infiltrative hematologic diseases cause congestion of the splenic circulation by abnormal cells. For instance, the mechanism of splenic infarction in sickle cell disease is attributed to crystallization of the abnormal hemoglobin during periods of hypoxia or acidosis.[2, 3] The rigid erythrocyte leads to rouleaux formation and occlusion of the splenic circulation.

In homozygous sickle cell disease, multiple infarcts during childhood commonly result in a scarred, contracted, autoinfarcted spleen by adulthood. In individuals who are heterozygous for sickle trait, exposure to low-oxygen tension (eg, during unpressurized airplane travel) or vigorous activity (eg, skiing in high-altitude locations) can precipitate sickling and splenic infarction by the above-described mechanism.[4] In myelofibrosis, the splenic parenchyma is infiltrated by extramedullary hematopoiesis, causing congestion of the splenic circulation.

In malignant hematologic diseases (eg, chronic myeloid leukemia), increased splenic oxygen requirements secondary to an increased splenic mass, coupled with a decreased oxygen-carrying capacity secondary to the anemia of hypersplenism, lead to infarction.

Thromboembolism is another common cause of splenic infarcts.[5, 1] Splenic embolization may result from various cardiovascular conditions, including a left atrial or ventricular mural thrombus that formed as a result of acute myocardial infarction or atrial fibrillation or developed from complications of cardiac catheterization or bacterial endocarditis. A report on 108 patients with left-side endocarditis undergoing valvular surgery revealed a 19% incidence of splenic infarction; in almost half of the patients with infarction, the diagnosis was made incidentally on computed tomography (CT).[6]

Another mechanical cause of splenic infarct can be the injection of gastric varices in the setting of portal hypertension and gastric variceal bleeding.[7]

Hypercoagulable states can cause splenic infarction. For instance, hereditary protein C deficiency has been reported to cause splenic infarction.[8] Acquired hypercoagulable states include myeloproliferative disorders, lupus anticoagulant, and erythropoietin therapy.

Splenic infarct has also been reported in association with postpartum toxic shock syndrome.[9]

Splenic vein thrombosis, most commonly the result of pancreatitis or surgery, can result in venous infarction.

Unusual causes of splenic infarction include malaria,[10] pancreatitis, and cocaine use; it can also occur, uncommonly, as a late complication of liver transplantation.

The authors have treated one case of global splenic infarction in a person with multiple blunt injuries who required hepatic packing as part of a damage-control procedure. At initial exploration, the spleen was intact, perfused, and viable. The patient subsequently required reoperation for release of an intra-abdominal compartment syndrome. Follow-up CT after the second operation revealed global splenic infarction, attributed to an occlusion of splenic venous outflow due to severe intra-abdominal hypertension and the resultant impedance of venous return from the visceral circulation.

An anatomic variant that renders the spleen more susceptible to global infarction is that of the wandering spleen.[11] The spleen is attached by a long vascular pedicle, without the usual fixating ligaments to the diaphragm, colon, left kidney, and lateral abdominal wall. This allows torsion of the freely mobile spleen around its vascular pedicle, occluding the blood supply and leading to infarction.

There are numerous etiologies of splenic infarct. The vast majority (88%), however, are either infiltrative hematologic diseases that cause congestion of the splenic circulation by abnormal cells, or thromboembolic conditions that produce obstruction of larger vessels.[12] The etiologies of splenic infarct may be categorized as follows:

• Malignant hematologic disorders - Leukemia, lymphoma (ie, Hodgkin, non-Hodgkin), myelofibrosis
• Benign hematologic disorders -  Hypercoagulable states (protein C or protein S deficiency), oral contraceptives, lupus anticoagulant; erythropoietin therapy; idiopathic venous thrombosis; polycythemia vera; sickle hemoglobinopathies ^[2]
• Embolic disorders -  Endocarditis, atrial fibrillation, prosthetic mitral valve, paradoxical emboli from right heart, left ventricular mural thrombus following myocardial infarct, infected thoracic aortic graft, HIV-associated mycobacterial infections
• Vascular disorders
• Autoimmune/collagen vascular disease
• Trauma - Blunt trauma, torsion of the wandering spleen, left-heart catheterization via femoral artery approach, sclerotherapy of bleeding gastric varices, vasopressin infusion, embolization for splenic bleeding
• Operative etiologies ^{[13, 14]} - Pancreatectomy, liver transplant
• Miscellaneous - Splenic vein thrombosis, pancreatitis, amyloidosis, sarcoidosis, pancreatic cancer, acute respiratory distress syndrome (ARDS), postpartum toxic shock syndrome

A few cases of splenic infarct assocated with COVID-19 have been documented, suggesting that splenic infarct should be considered as one of the potential causes of abdominal pain occurring in COVID-19 patients.

Often a clinically silent condition, splenic infarct is associated most commonly with hematologic disorders. Although splenic infarct rates of 50% and 72% have been reported in chronic myelogenous leukemia and myelofibrosis, respectively, few large series describing this entity exist.

In 1998, Nores et al[12] reported 59 cases treated over a 30-year period at the University of California, Los Angeles (UCLA) and at the Cedars-Sinai Medical Center. In 1986, Jaroch et al[15] identified 75 patients through clinical or autopsy reports at the Cleveland Clinic and found only an additional 77 cases in the literature. Most of the current literature consists of case reports only.

The frequency of visualized splenic infarcts may be rising because of the following factors:

• Increased radiologic imaging of patients (with subsequent increased incidental detection of splenic infarcts) ^[16]
• Current standard of nonoperative management of blunt splenic injuries
• Increased use of angiographic embolization for vascular splenic injuries ^[17]

The prognosis for splenic infarction varies according to the underlying disease process responsible for the infarct. Splenectomies for infarction of massively enlarged spleens accompanying hematologic malignancies reportedly are associated with mortalities as high as 35%. At the other end of the spectrum, many infarcts are clinically occult, with no significant long-term sequelae.

Asplenic individuals have an increased lifetime risk for developing overwhelming postsplenectomy sepsis (OPSS), with the highest rate in the pediatric age group. Patients should be counseled to seek medical attention even for seemingly minor infections, because these can progress to fatal bloodstream infection within hours.

These considerations have proved to be the impetus for splenic preservation.

The clinical spectrum ranges from asymptomatic infarction (discovered incidentally on radiologic or postmortem studies or at laparoscopy or laparotomy for another indication) to hemorrhagic shock (secondary to massive subcapsular hemorrhage with free rupture into the peritoneal cavity). Approximately one third of splenic infarcts are clinically occult.

The most common presenting symptom is left-upper-quadrant abdominal pain (up to 70%). Additional symptoms include fever and chills, nausea and vomiting, pleuritic chest pain, and left shoulder pain (Kehr sign).

Septic thromboemboli may result in splenic abscesses, which present with sepsis and left upper abdominal pain.

In the series of Nores et al, most of the patients with thromboembolic infarction were symptomatic; 70% of patients with emboli were febrile, and 86% of individuals with thrombosis had abdominal pain.[12]

In a 10-year retrospective study, Antopolsky et al examined clinical presentations in 49 episodes of acute splenic infarction. The most common symptom was either abdominal or left flank pain (80% of episodes), while the most common sign was left-upper-quadrant tenderness (35% of episodes). In 16.6% of patients, splenic infarction was the presenting symptom of an underlying disease. No in-hospital mortality or serious complications occurred in the cohort.[18]

No specific diagnostic laboratory studies for splenic infarction exist, though an elevated white blood cell count is not infrequent.

Computed tomography (CT), performed with intravenous (IV) nonionic contrast, is the current diagnostic modality of choice (see the image below). Before the CT era, the diagnosis of splenic infarction was most commonly made at laparotomy for an intra-abdominal catastrophe or on postmortem examination.

Magnetic resonance imaging (MRI), preferably performed with IV gadolinium contrast, is another useful modality that clearly identifies infarcted splenic parenchyma. Contrast-enhanced MRI and multidetector CT allow three-dimensional reconstructions in any plane (coronal, sagittal, or axial) to better visualize the classic appearance of wedge-shaped infarctions within the spleen.

Given the dual blood supply of the spleen, contrast-enhanced studies should be performed during a delayed phase so that the normal early archiform pattern of arterial splenic enhancement does not mask lesions or create pseudolesions.

Contrast studies should be performed during an appropriate delay (for patients with good cardiac reserve, 50 seconds is an acceptable scan delay) when most spleens will be in the uniform phase of enhancement.[19] These postcontrast scans clearly depict the classic segmental, wedge-shaped, low-attenuation defect.[20] Less frequently, the entire spleen may be infarcted, leaving only a rim of contrast-enhancing capsule.[21]

Other modes of diagnosis include radioisotope scans and ultrasonographic[22] (US) evaluation of the spleen. Angiography is indicated when a vascular lesion is suspected as the etiologic cause, as in cases of arterial embolization, or when it is necessary to manage segmental bleeding by embolization (see the images below).

US is useful in cases where the splenic parenchyma can be visualized.[23] Significant amounts of luminal bowel gas, as well as morbid obesity, render this modality less useful. In a retrospective study of 49 episodes of acute splenic infarction, Antopolsky et al found ultrasonographic scanning to be diagnostically useful in only 18% of patients.[18]

Pathologic examination of the resected spleen may provide information regarding the pathogenesis of the infarct (eg, evidence of septic or atheromatous emboli or the presence of an infectious etiology).

The indications that warrant surgical intervention include hematologic complications for the underlying disease process and persistent symptoms, such as sepsis, abscess, hemorrhage, or persistent pseudocyst formation.

Analysis of operative indications in the series of Nores et al revealed that patients with hematologic conditions were typically explored for complications of those conditions (69%), especially patients with nonmalignant conditions (90%), whereas patients with hematologic malignancies often (47%) underwent surgery for acute symptoms.[12] Complications of splenic infarction were a frequent indication (60%) for operation in patients with septic emboli.[12]

Pseudocysts usually are the result of prior trauma, with organization and liquefaction of the resultant hematoma, surrounded by a fibrous pseudocapsule. Small (< 4 cm) asymptomatic pseudocysts usually regress with time, whereas larger (>5 cm) splenic cysts require treatment because the risk of rupture can be as high as 25%.[24]

Splenic abscess can result from septic emboli or superinfection of a prior infarct. Traditionally, such abscesses have been managed by means of splenectomy, which can be technically challenging because of the intense surrounding inflammation and because of the scar tissue that surrounds the abscess. Indwelling percutaneous catheter drainage and appropriate antibiotic therapy can be used to treat selected cases of well-formed, unilocular abscesses.

An asymptomatic infarct without complication does not require surgical intervention. In fact, overall, most splenic infarcts do not require surgical intervention.

Surgery is indicated only in the presence of complications. Otherwise, the infarcted spleen can be left in situ, and the patient can be observed. Because of the rarity of this disorder and the largely anecdotal character of many reports, the roles of antibiotics and antiplatelet agents (for the treatment of thrombocytosis) have not been formally addressed. Similarly, no scientifically supported information exists regarding the possible increase in susceptibility to overwhelming postsplenectomy sepsis (OPSS) in these patients.

The principal mainstay of nonoperative therapy is analgesia with either narcotics or nonsteroidal anti-inflammatory drugs (NSAIDs) and close follow-up. Many of the altitude-related episodes in patients with sickle cell disease can be safely treated with supportive care rather than with splenectomy.[4] In certain circumstances, isolated splenic abscess can be treated with percutaneous drainage alone, and splenectomy can be avoided. However, a patient with multiple splenic abscesses usually will require splenectomy.

For an infarcted spleen with any of the above-mentioned complications, splenectomy is required. Because of the (admittedly small) risk of fatal OPSS, splenic preservation is preferable whenever possible.

In cases of torsion of a wandering spleen, splenopexy with splenic salvage is the procedure of choice in the well-perfused, noninfarcted spleen. Techniques include suturing the spleen to the surrounding structures, wrapping the organ in omentum or mesh prior to suture fixation, or placing it in a surgically created retroperitoneal pouch. This has been reported laparoscopically.[25]

Complications, such as bleeding or pseudocyst formation, also may be amenable to splenic salvage using techniques of partial splenectomy.

Whereas a unilocular abscess can be managed successfully in select cases with percutaneous catheter drainage, some authors advocate splenectomy in all cases of splenic infarct and abscess, questioning the utility of preserving the residual, partially functioning spleen. This may be accomplished by using traditional open techniques or with laparoscopic techniques.

Perisplenic inflammation and dense adhesions can make splenectomy difficult. Another choice is to perform preoperative splenic artery embolization; this purposely infarcts the remaining spleen and minimizes blood loss, which otherwise can be quite profuse in these difficult dissections. Intraoperative ligation of the splenic artery at the superior margin of the pancreas in the lesser sac is another alternative to minimize blood loss if the spleen is enlarged.

Preparation for surgery

The choice of preoperative antibiotics should be guided by the patient's clinical status and associated comorbidities. A first-generation cephalosporin usually is adequate.

Appropriate preoperative hydration is essential in anticipation of potential blood loss.

If splenectomy (rather than splenic preservation) is planned, the patient should receive the pneumococcal and Haemophilusinfluenzae vaccine at least 2 weeks prior to the operation, if feasible.

Neoadjuvant chemotherapy in patients with hematologic malignancies (eg, chronic myeloid leukemia) can reduce spleen size and potentially increase the safety of the procedure.

Preoperative sequential compression devices and subcutaneous heparin can be used for deep vein thrombosis (DVT) prophylaxis.

Operative details

The approach (eg, laparoscopy, midline laparotomy, or left subcostal incision) is determined by the underlying pathology/clinical scenario, as well as by the surgeon's preference. The presence of splenic infarct (or abscess) is not per se a contraindication for laparoscopy. Generally, the laparoscopic approach is contraindicated in emergency situations, such as hemorrhagic shock and spontaneous rupture. The surgeon's experience determines the point at which the enlarged spleen is too large to remove laparoscopically.

Because of the risk of lifelong susceptibility to infectious complications after splenectomy (particularly the rare but highly fatal OPSS), laparoscopic or open partial splenectomy should be attempted when feasible. Laparoscopic spleen-preserving distal pancreatectomy, with or without splenic-vessel preservation, has yielded promising results.[26]

Laparoscopic approach

For the laparoscopic approach, patient position is key to the success of this operation. Choices include the "hanging spleen" technique, in which the patient is in the lateral decubitus position (as described for laparoscopic adrenalectomy). Another option is the "leaning spleen" technique, in which the patient is at a 45º tilt. Both methods require a beanbag or jelly roll positioning pad. Once pneumoperitoneum is established, the splenocolic ligament is usually divided with ultrasonic shears. Next, the lesser sac is entered, and the short gastric vessels are divided.

If the surgeon prefers to "preligate," the splenic artery can be dissected along the superior border of the pancreas and divided with the vascular stapler. Otherwise, if feasible, the splenic vein and artery are exposed first and are dissected 1 cm proximal to the hilum, and they are usually divided with the vascular stapler. Then, the remainder of the spleen is released by dividing the peritoneal attachments with the ultrasonic scalpel.

In other scenarios, the attachments are divided first and the spleen is rolled anteriorly to expose the tail of the pancreas and the hilar vessels posteriorly. These vessels are then transected with the laparoscopic vascular stapler. The spleen is placed into an EndoCatch bag and is morcellated and extracted through the largest trocar site.

Other laparoscopic approaches are the supine approach (with set-up similar to laparoscopic Nissen fundoplication) and the hand-assisted approach. Laparoscopic partial splenectomy also has been described, including for splenic abscess, using the ultrasonic scalpel or radiofrequency ablation for parenchymal division.[27, 28]

Open approach

For the open approach, a standard midline incision or left subcostal incision (with possible Kehr extension up to the xiphocostal junction) may be used. Placing a large roll under the left flank often aids in exposure.

Usually, the gastrocolic ligament is opened outside the gastroepiploic arcade and the splenic artery is identified along the superior border of the pancreas and then ligated. Next, the spleen is completely mobilized by dividing the avascular splenophrenic and splenorenal ligaments. The splenocolic ligament is divided, and the spleen is mobilized into the wound. The tail of the pancreas is separated from the splenic artery and vein, the splenic vessels are ligated, and the spleen is extracted.

There are several techniques described for splenic repair.[24] The critical intraoperative maneuver regarding intraoperative splenic salvage entails fully mobilizing the spleen into the wound; this requires the division of one or two of the short gastric vessels combined with gentle dissection posteriorly, so that the capsule is not torn in the mobilization process.[17]

The use of postoperative drainage depends on the operative findings, including the presence of a frank abscess, and the proximity of the dissection to the pancreatic tail. The authors' preference is for closed suction drainage to be placed in the splenic bed, with early removal (ie, within 24 hours). There is some evidence that the routine use of drainage of the splenic bed in the absence of extrasplenic abscess or pancreatic injury is associated with an increased risk of local infection.

If concern exists regarding possible impingement upon the gastric wall during short gastric vessel division, a nasogastric tube may be left in place.

Following splenectomy or global infarct, administer pneumococcal vaccine and vaccine against Haemophilus (if not given preoperatively).

Postoperative thrombocytosis occurs frequently. Antiplatelet therapy is initiated for platelet counts higher than 1,000,000/μL. However, antiplatelet therapy is controversial in this circumstance, and its benefit has not been demonstrated with a randomized, prospective, controlled trial.

The authors' current preferences for DVT prophylaxis are the use of sequential compression devices prior to incision and the postoperative employment of early ambulation and subcutaneous heparin.

Postoperative pain management may include IV patient-controlled analgesia and possible epidural analgesia for cases requiring laparotomy.

Pulmonary complications (eg, pneumonia, atelectasis) were the most frequent complication (20%) in the series reported by Nores et al.[12] With the current trend toward a laparoscopic approach to splenectomy and partial splenectomy, this complication rate should be significantly reduced.

Postoperative abscess, fever, and wound infection constitute the second most common category of complications (~10%). The first line of treatment is radiologically guided percutaneous drainage and antibiotics. A missed injury to a viscus (eg, stomach or colon) should be ruled out.

Hemorrhage, though relatively infrequent, can follow splenectomy as a consequence of the intense perisplenic inflammation.

Because of the intimate association of the pancreatic tail and the splenic hilum, pancreatic injury can occur, especially in the setting of intense inflammation and/or abscess. The majority of these injuries resolve with nonoperative management, which includes wide drainage, the use of a somatostatin analogue to decrease exocrine pancreatic function, and either total parenteral nutrition (TPN) or enteral alimentation distal to the ligament of Treitz.

As a consequence of the intense inflammatory reaction that can accompany splenic abscess, dissection of the spleen from the greater curvature of the stomach can be difficult, and inadvertent, unrecognized injuries to the greater curve of the stomach do occur. With adequate external drainage and with no obstruction to normal gastric emptying, these can be treated expectantly with TPN or distal luminal alimentation and nasogastric tube decompression.

OPSS is the most serious postsplenectomy complication; it occurs rarely (0.5%) in adult patients but carries a prohibitive mortality in unvaccinated patients. For this reason, a trend away from splenectomy and toward splenic conservation has emerged.

Follow-up is directed by the underlying cause of the infarction or abscess rather than by the absence of the spleen. Patients undergoing nonoperative management of splenic infarct, particularly a large infarct (when there is a concern that it will rupture), should be followed closely, with a low threshold for serial imaging.