Invited Commentary Indications for Splenectomy · splenectomy infection (OPSI). Mortality rates of OPSI may exceed 50 per cent in unvaccinated patients.' Therefore, removal of the

Invited Commentary

Indications for SplenectomySTEVEN C. KATZ, M.D., H. LEON PACHTER, M.D.

From the Department of Surgery, New York University Medical Center and Bellevue Hospital Center,Nezv York, NY

In the new millennium, indications for splenectomy have expanded. Proper patient selectionbased on an understanding of the biology of each individual's disease is essential for a favorableoutcome. We review the most common diseases for which surgeons may be called on to performsplenectomy and while highlighting potential pitfalls and caveats.

T HE SPLEEN MEDITATES important immunologic andhematologic functions as well as contributing to

numerous pathologic processes. When the spleen isinvolved in disease, splenectomy may be performedwith tbe intent of either altering tbe clinical course orproviding symptomatic relief. In the new millennium,indications for splenectomy have expanded, and sple-nectomy may be beneficial to patients with a broadspectium of benign and malignant diseases. Properpatient selection based on an understanding of the bi-ology of each individual's disease is essential for afavorable outcome.

Splenectomy is not without potential complications,the most feared of which being overwhelming post-splenectomy infection (OPSI). Mortality rates of OPSImay exceed 50 per cent in unvaccinated patients.'Therefore, removal of the spleen should be performedonly when the potential benefits to the patient clearlyexceed the risks. Recent advances in surgical tech-niques allow for a laparoscopic approach in many pa-tients who require splenectomy, thereby minimizingoperative morbidity and patient discomfort associatedwith open splenectomy. However, a minimally inva-sive approach lo splenectomy is not always feasible ordesirable and therefore familiarity with the traditionalapproach remains vital. We review the spectrum ofdisease processes for which splenectomy may be in-dicated. Prerequisites for successful therapeutic or pal-liative splenectomy are a thorough understanding ofsplenic anatomy, physiology, and pathophysiology.We discuss the most common diseases for which sur-geons may be called on to perform splenectomy andconsider tbe various surgical approaches while high-lighting potential pitfalls and caveats.

Address correspondence and reprint requests to H. LeonPachter. M.D.. HCC 6 6C, 550 First Avenue, New York. NY10016.

Development and Anatomy

The spleen forms as a mesenchymal condensationof the dorsal mesogastrium during the fifth week ofdevelopment. Initially, the spleen functions as a he-matopoietic organ and assumes its more mature lym-phoid characteristics at 15 to 18 weeks.^ The spleen isrelated to the posterior wall of the stomach and isconnected to the stomach and kidney by the gastro-splenic and splenorena! ligaments. With the exceptionof the hilum. the spleen is surrounded by peritoneum.On average, tbe spleen is 12 cm long and 7 cm wide.The splenic artery originates from the celiac axis anddivides into 5 or more terminal branches that enter thesplenic hilum. Several tributaries join to form thesplenic vein, which joins the superior mesenteric veinto form the portal vein posterior to the neck of thepancreas.

A dense fibrous capsule surrounds the spleen andtrabeculae form incomplete parenchymal compart-ments within the splenic pulp. Stretching of the cap-sule may cause pain in patients witb splenomegaly. Oncross-section, the spleen contains both red and whitepulp. Arteries within the white pulp, central arteries,are surrounded by a sheath of lymphocytes known asthe periarteriolar lymphatic sheatb (PALS).^ TbePALS is comprised primarily of T ceils, whereas ad-jacent lymphoid follicles are abundant in B cells. Themarginal zone lies between tbe white and red pulpcontaining dendritic cells, which capture and presentantigen to lymphocytes. Within the red pulp, the sinu-soids are lined by a fenestrated endothelium similar tothe lumen of the hepatic sinusoids."^

Normal Splenic Function

Although many functions of the spleen are redun-dant or can be assumed by other organs, splenectomycan lead to adverse consequences. An appreciation of

565

566 THE AMERICAN SURGEON July 2006 Vol. 72

normal spleen function is important for the physicianseeking to understand the potential consequences ofsplenectomy in a given disease process. Under normalconditions, the spleen contains less than 50 mL ofblood and does not serve as a depot for intravascularvolume, platelets, leukocytes, or erythrocytes. In thesetting of splenomegaly or portal hypertension, thestorage volume of the spleen expands and formed el-ements of Ihe blood are .sequestered."^ As much as onethird of the total platelet mass may be stored in thespleen and released during inHammatory states/'

Although active hematopoiesis occurs in the fetalspleen, it does not normally occur in postnatal life.However, in certain pathologic states such as myelo-fibrosis, extramedullary hematopoiesis may indeedtake place within the spleen. The splenic white pulp isthe largest accumulation of lymphoid tissue in thebody and is a site of lymphocyte production and acti-vation, from which cells migrate into the red pulp toreach the lumen of the splenic sinusoids. Dendriticcells and macrophages in the marginal zone are in-volved with antigen trapping, processing, and presen-tation. Splenic macrophages are particularly adept atrecognizing and clearing opsonized bacteria.^ Bothdendritic cells^ and T lymphocytes" within the spleenappear to have potent immunologic function.

In addition to its role in the immune system, thespleen is the site of senescent erythrocyte destruction.Macrophages in the splenic cords phagocytose eryth-rocytes and metabolize hemoglobin. Aged erythro-cytes are more sensitive to the relatively acidotic. hyp-oxic, and hypoglycemic milieu of the spleen. Theheightened sensitivity of senescent erythrocytes to thehostile splenic environment leads to alterations inmembrane carbohydrate moieties.'" thereby facilitat-ing recognition by macrophages and subsequent cull-ing or destruction of damaged cells. In addition toculling, pitting leads to removal of intracellular inclu-sions from erythrocytes. The loss of pitting after sple-nectomy accounts for the appearance of particulatematter such as Howell-Jolly bodies within erythro-cytes postoperatively. Furthermore, asplenia limitsbacterial clearance" and production of IgM and otheropsonin proteins.'^

Hematologic Conditions

Autoimmune Thrombocytopenia and Hemolytic Anemia

The pathogenesis of autoimmune thrombocytopeniaor idiopathic thrombocytopenic purpura (ITP) in-volves the formation of antibodies to several antigenicdeterminants, including glycoproteins llb/IIIa and hi/Ila.'"* The spleen is both a site of autoantibody pro-duction and platelet destruction. Although ITP tends tobe a self-limiting disorder in the pediatric population,

adults often require specific therapeutic intervention.In the event of persistent or recurrent disease aftertreatment with glucocorticoids. cytotoxic agents, orimmunoglobulin. patients with ITP may benefit fromsplenectomy. Demonstration ot megakaryocytes in thebone marrow and the absence of splenomegaly areessential for establishing the diagnosis of ITP beforesplenectomy is undertaken. Musser reported that aftersplenectomy for ITP, up to 77 per cent of patientsshow a complete response. In addition, 14 per cent ofpatients show a partial response, whereas only 9 percent failed to demonstrate a significant improvementin platelet count.'"^ Unfortunately, preoperative char-acteristics do not necessarily predict which patientswith ITP will respond to splenectomy.''^ Failure ofsplenectomy to improve the thrombocytopenia may bothe result of an unappreciated accessory spleen or intraabdominal implantation of splenic tissue shouldfragmentation occur during surgery or extraction altera iaparo.scopic approach. Particular attention should bepaid to the latter, especially during the "morseli/ation"process used to facilitate removal while limiting ihosize of the incision.

Similarly, autoimmune hemolytic anemia may re-sult in the need for splenectomy if the patient fails torespond to medical management, including oral corti-costeroids.""' Like with ITP, the pathogenesis of atito-immune hemolysis involves antibody-mediated cellu-lar destruction and complement activation within thesplenic substance. In patients with warm-reacting an-tibodies, favorable responses to spienectomy can beexpected in 50 per cent to 80 per cent.'^

Felty 's Syndrome

Felty's syndrome is the occurrence of neutropeniaand splenomegaly in patients with rheumatoid arthritis(RA). These manifestations are present in less than Iper cent of patients with RA. Patients with Felty'ssyndrome face an increased risk of infection as a resultof granulocytopenia. which results, in part, from in-trasplenic destruction of granulocytes.'^ Granulocyte-macrophage colony-stimulating factor may amelioratethe neutropenia in some cases.''^ Splenectomy is indi-cated only in patients with severe or recurrent neutro-penia or in tbose patients demonstrating recurrent andresistant infections.-" Splenectomy results in increasedgranulocyte levels in 80 per cent of patients, with 55per cent of patients experiencing no further infec-tions.-'

Thrombocytopenic Purpura

Thrombocytopenic purpura (TTP) presents with thepentad of fever, thrombocytopenia, hemolytic anemia,neurologic manifestations, and renal failure. TTP may

No. 7 INDICATIONS FOR SPLENECTOMY Katz and Pachter 567

result from an excess of subendothelial collagen, re-sulting in systemic platelet trapping. Splenectomy isindicated only for patients in whom plasmapheresisfails or who relapse when this therapy is discontinued.Excellent results may be achieved in refractorycases.̂ -̂ Over 50 per cent of patients with TTP whoundergo splenectomy may respond favorably.--^

Erythrocyte Membrane Disorders

Hereditary spherocytosis (HS) is the most commonerythrocyte membrane disorder for which splenectomyis indicated. An autosomal-dominant mutation resultsin a defective erythrocyte membrane resulting fromderangements in spectrin and ankyrin, rendering thecells less deformable and thus susceptible to intra-splenic destruction. As a result of erythrocyte destruc-tion within the spleen, patients with HS present withanemia. Jaundice, and splenomegaly. The anemia maybe relatively mild with jaundice being the only clinicalmanifestation.

The role of splenectomy in HS is to treat the under-lying anemia and prevent the development of biliarypathology, aplastic crises, and late hemachromato-sis."' Being that erythrocytes are destroyed within thespleen in patients with HS, splenectomy represents ahighly effective treatment modality. '̂* Splenectomyfor HS is generally deferred until the patient is 4 to 6years old to minimize the risks of OPSI. Splenectomyresults in correction of the anemia associated with HSand prevents subsequent hemolytic episodes.-^ Re-sponse rates as high as 90 per cent have been re-ported.''' Hereditary elliptocytosis is generally a non-pathologic trait, but when more than 90 per cent of theerythrocytes are affected, the disease manifestationsmay become significant. Indications for and results ofsplenectomy are similar to those of HS.̂ "̂

Erythrocyte Enzyme Deficiencies

Pyruvate kinase deficiency (PKD) is an autosomal-recessive disorder associated with hemolysis and re-duced red blood cell deformability. Consequently,erythrocytes are sequestered and destroyed in thespleen. The severity of disease ranges from fully com-pensated hemolysis to potentially fatal anemia andjaundice in the neonate. Although splenectomy doesnot eliminate hemolysis in patients with PKD. it doesresult in increased hemoglobin levels. As a result.transfusion requirements are greatly decreased or eveneliminated.-'' Patients witb PKD may also developcholelithiasis and associated biliary tract pathology asa result of chronic hemolysis. Therefore, cholecystec-tomy may be indicated in patients with PKD. In pa-tients with PKD undergoing cholecystectomy, con-comitant splenectomy merits serious consideration to

effectuate a decrease in the rate of bemolysis and sub-sequent transfusion requirements.-^

Glucose-6-phosphate dehydrogenase deficiency(G6PD) is an X-linked disorder that renders the patientsusceptible to hemolysis after exposure to certaindrugs or chemicals such as methylene blue and favabeans. Deficiency of glutathione leads to the destabi-lization of red blood cell membrane skeletons throughoxidative mechanisms.-*^ Splenectomy plays little, ifany, role in patients with G6PD deficiency and there-fore should not be undertaken. The cornerstone oftherapy in this unusual disease remains avoidance ofinciting drugs or physiological stressors, particularlyinfections.

Hemoglobin apathies

In sickle cell disease, a single mutation in the beta-hemoglobin chain of Hb A results in rigid, nondeform-able erythrocytes that are sequestered and destroyed inthe spleen. Although the spleen may be enlarged earlyin life, the organ tends to shrink and autoinfarct as aresult of recurring ischemic insults. Likewise, thalas-semia involves a defect in hemoglobin synthesis and ismost prevalent among Mediterranean populations,Thalassemia major, the homozygous form, presents inchildhood with failure to thrive, ulcers, and anL'mia.Fetal hemoglobin (Hb F) persists and a relative paucityof adult hemoglobin (Hb A) can be demonstrated.

The indications for splenectomy in sickle cell dis-ease and thalassemia include hypersplenism, seques-tration crises, splenic abscesses, and massive splenicinfarction.-'' Acute splenic sequestration crises arelife-thi-eatening events for which the transfusion re-quirement may be substantial and pain may he themost significant component of the overall presenta-tion."**' The features of acute splenic sequestration cri-ses include severe anemia, splenomegaly, and abdomi-nal pain. The patient's hemodynamic status maydeteriorate into frank circulatory collapse. These pa-tients require aggressive resuscitation with crystalloidand blood products before urgent splenectomy. In ad-dition to relieving the acute sequestration crises, sple-nectomy also alleviates mechanical symptoms result-ing from splenomegaly and may improve protein andenergy metabolism in children with sickle cell dis-ease.^' Postoperatively. although these patients benefitfrom decreased transfusion requirements and in-creased circulating erythrocyte mass,^^ they are par-ticularly at high risk for postsplenectomy infections.

Primary Hypersplenism

Hypersplenism refers to splenomegaly associatedwith a decrease in one or more of the formed bloodelements. Although hypersplenism is most often a sec-


ondary manifestation of an underlying disorder such a.sportal hypertension, primary or idiopathic hypersplen-istn may occur. Primary hypersplenism or hypersplen-ism in the absence of a specific disease process wasdiagnosed more frequently in the past as a result of ourlimited inability to detect the presence of lymphoma orleukemia. The sensitivity of modem diagnostic mo-dalities to detect lymphoma or leukemia has improvedto the extent that primary hypersplenism is now anuncommon diagnosis. In instances in which hyper-splenism is symptomatic and truly idiopathic. splenec-tomy is indicated to relieve the mechanical symptomsof splenomegaly, effect improvement of cytopenias,and eliminate the possibility of an underlying malig-nancy. Lymphoma has been pathologically detected in40 per cent to 70 per cent of patients undergoing sple-nectomy for presumed primary hypersplenism.-^^- "̂̂The frequency of this occurrence, as noted earlier, maybe significantly lower as a result of advances in diag-nostic imaging and the liberal use of fine-needle aspi-ration witb flow cytometry.

Neoplastic Conditions Requiring Splenectomy

Hodgkin 's Disease

The role of splenectomy in the management of pa-tients with Hodgkin's iymphoma has undergone sig-nificant modifications. In the past, staging laparotomy,which included splenectomy. was an important com-ponent in the evaluation of these patients. Detection ofdisease below the diaphragm has important prognosticand therapeutic implications. When splenectomy wasperformed as part of a staging laparotomy for Hodg-kin's disease, over one third of spleens removed werefound to be involved with Hodgkin's disease. As aresult, patients were upstaged by information obtainedduring staging laparotomy.'"^ However, improvementsin computed tomography, positron emission tomogra-phy, and magnetic resonance imaging techniques havelimited the need for staging laparotomy.^^ Moreover,given improved toxicity profiles of chemotherapeuticagents, patients with a high likelihood of stage 111 orIV disease may undergo systemic therapy without be-ing subjected to staging laparotomy. Yet. patients withearly disease and equivocal radiologie findings maystill be referred for operative staging. Surgical stagingshould be limited to patients with stage I or II diseasein whom a negative result may permit withholdingchemotherapy after radiation therapy.^^ Staging lapa-rotomy is now performed, at most, in 30 per cent ofpatient's with Hodgkin's disease. When surgery is re-quired, the laparoscopic approach is used with increas-ing frequency.^^

Non-Hodgkin 's Lymphoma

Splenectomy may be performed for non-Hodgkin"slymphoma (NHL) with either therapeutic or diagnosticintent. Indications for splenectomy include relief ofsymptoms resulting from splenic enlargement, cytope-nias. facilitation of chemotherapy administration, andto establish a tissue diagnosis. In a single study, sple-nectomy improved cytopenias in up to 72 per cent ofpatients and may contribute to durable remission.'"However, the surgical mortality and morbidity were2.9 per cent and 37 per cent, respectively. When lym-phoma primarily involves the spleen, cytopenias arefrequent and are reversed by early splenectomy in 82per cent of cases."*'* Berman and colleagues found thatin patients with hematologic malignancies, splenec-tomy resulted in sustained normalization of plateletlevels and thereby decreased the need for platelet in-fusions." '̂

Chronic Myelogenous Leukemia

Chronic myelogenous leukemia (CML) is a myelo-proliferative disease in which the normal bone marrowelements are progressively replaced by neoplastic my-eloid cells. CML is characterized by the Philadelphiachromosome, which involves fusion of fragmentsfrom chromosomes 9 and 22."̂ - CML is typically in-dolent in onset with a progressive phase heralded byfever and splenomegaly. The progressive phase mayculminate in a blast crisis with splenic sequestration ofblood elements contributing to the anemia, infection,and hemorrhagic complications.

The primary therapy for CML includes hydroxy-urea, interferon-alpha. and chemotherapy with bonemarrow transplantation. There is clearly a role for pal-liative spleneetomy to relieve compressive symptomsand improve cytopenias (Fig. l)."̂ -* However, splenec-tomy does not improve survival in patients with CML,and thromboembolic and vascular events are morecommon in patients with CML after splenectomywhen compared with those managed nonoperatively."*"^Therefore, a heightened suspicion for thrombotic com-plications is appropriate when splenectomy is per-formed in patients with CML and this concern is dis-cussed in greater detail subsequently.

Chronic Lymphocytic Leukemia

Splenectomy is performed for complications of chron-ic lymphocytic leukemia (CLL). including hypersplen-ism and the mechanical effects of splenomegaly.Anemia and thrombocytopenia may be significantlyimproved in patients with CLL after splenectomy.Moreover, in contrast to CML, in certain subgroups ofpatients such as those with hemoglobin levels


A.

B.

C.

FIG. 1. Spleen iiwoived with chronic myeiogenous leukemia (CML). This case il!u?,lrales [lie marked degree ot splenomegaly thatmay occur in patients with CML. Sptenecloniy was indicated in this case for relief of compressive sympionis. (A) Alter mobiii/ationof the spleen, it was delivered through a left upper quadrant incision and then the splenectomy was completed (B). A preoperativecomputed tomography scan demonstr"ates the degree of splenomegaly and how this may lead to compressive symploins (C).

less than 10 g/dL or platelet counts less than 50.000/L.splenectomy has been shown to improve long-termsurvival."*^

Hain' Cell Leukemia

Hairy cell leukemia (HCL), formerly known as"surgical leukemia," is a rare, chronic lymphoid leu-kemia that generally affects the elderly with spleno-megaly and pancytopenia. Neoplastic mononuclearcells with distinctive cytologic features are present inthe peripheral blood and bone marrow. Although aminority of patients present with asymptomaticsplenomegaly, most will require therapy to addressmanifestations of anemia, neutropenia, or thrombocy-topenia. Historically, splenectomy played a definitiverole in the treatment of patients with HCL."*̂ At pres-ent, splenectomy is infrequently performed for HCLbecause interferon-a, and more recently, purine ana-logs have proven to be highly effective in the man-agement of this type of leukemia.**"̂

Primary and Metastatic Tumors

The spleen is a relatively infrequent site of metas-tases of nonhematologic malignancies, and this maybe related to peculiarities of the splenic microcircula-

tion or immunologic function."*^ Splenic tumors mostcommonly present with splenomegaly and diffuse pa-renchymal replacement may result in hematologicconsequences. Moreover, a pathologic spleen is at riskfor either spontaneous rupture or rupture witb theslightest degree of external trauma, resulting in an in-traabdominal catastrophe.

The most common primary tumors of the spleen arevascular in nature.'*^ Benign primary vascular tumorsof the spleen include hemangiomas and lymphangio-mas. Both tumors are typically found incidtMitally.Hemangiomas are typically of the cavernous type andrarely rupture. On occasion, splenic hemangiomas areresponsible for a consumptive coagulopathy, whichmay necessitate extirpation of the spleen because Ibeprocess can be fatal.''" Lymphangiomas are endotbe-lial-lined lesions filled with proteinaceous material.Both hemangiomas and lymphangiomas may be singleor multiple.^' Littoral cell angiomas may present assingle or multiple lesions.-''̂ Splenic hamartomas arewell-circumscribed, nonencapsulated benign prolifera-tions of nonnal splenic tissue elements."**̂ Althoughmost commonly an incidental finding at autopsy or onabdominal imaging, diffuse splenic hamartomas mayresult in hypersplenism. In addition, splenic hamarto-mas may be the cause of intraabdominal hemorrhage


after rupture.̂ -^ Rarely, the spleen gives rise to lipo-mas'̂ '' and angiomyolipomas.'̂ -'̂

Hemangiosarcomas (Fig. 2) are the most commonprimary malignant tumor of the spleen and have re-ceived attention disproportionate to their incidencegiven the association with environmental factors suchas vinyl chloride and thorium dioxide.'^'' Splenic he-mangiosarcomas may present with splenomegaly, ane-mia, pleura! effusion, or spontaneous rupture. Theprognosis is generally poor irrespective of the type oftreatment rendered. Other sarcomas such as Kaposi'ssarcoma arise within the spleen infrequently. Whenthe spleen is the only site of disease or when splenicinvolvement by primary or metastatic sarcoma leads tosymptoms, splenectomy would appear to be a reason-able option.

Various lymphoid processes may involve the spleenprimarily. Although the spleen is more often a second-ary site of involvement in Hodgkin's and non-Hodgkin's lymphoma, both may arise from the splenicwhite pulp. Primary lymphoma of the spleen (Fig. 3)may replace the organ diffusely or present as a mui-tinodular process.^^ The spleen nnay also be affectedby histiocytic tumors and angiofollicular lymphoid hy-perplasia or Castleman's disease.*^"

The spleen may be involved by metastatic disease inup to 7 per cent of patients with malignant solid tu-mors.-̂ ^ Among the tumors thai metastasize to thespleen, melanoma, adenocarcinoma of the breast, andlung cancer are most common. When the spleen isinvolved with metastatic disease, it is rarely the solesite of disease and therefore the role of spienectomy isgenerally reserved for palliation of splenomegaly or

Fi(i. 2. Hemangiosarcoma of the spleen. Computed tomogra-phy scan of ihc abdomen with intravenous contrast demonstrates amass within the spleen with areas of hypoattenuation and enhance-ment. The areas of enhancement are typical for vascular tumorssuch as hemangiosarcoma. The specitnen measured approximately8 cm in diameter and was completely resected.

FKI. 3. Priniaiy splenic lymphoma. The patient presented willisymptoms related to splenomegaly and hypersplenism. This ab-dominal computed tomt>graphy scan with intravenous contrastdemonstrated a low-altenuating lesion in the spleen. After sple-nectomy and a search for synchronous sites ot" disease, the diag-nosis of primary splenic lymphoma was made.

hypersplenism. Meta.stases to the spleen are ustially anindicator of .systemic disease and aggressive lumorbiology, rendering splenectomy with curative intenigenerally inappropriate in these settings (Fig. 4).^"Like with any tumor involving the spleen, rupture orabscess formation may oceur and spienectomy may berequired on these bases.

FIG. 4. Melanoma meiastatic to the spleen. Several years afterthe excision of a cutaneous melanoma, this patient was found tohave numerous low-attenuating lesions in the spleen and liver(arrows). These lesions were consistent with metastatic melanoma.


Benign Conditions Requiring Splenectomy

Gaucher's Disease

Gaucher's disease is an autosomal-recessive disor-der in which glucocerebrosidase deficiency results inthe systemic accumulation of glucocerebrosides,which may be accompanied by hepatosplenomegaly,anemia, thrombocytopenia, atid skeletal manifesta-tions. The spleen may be enlarged 60 to 70 times itsnormal size and 10 to 20 times its normal weight, andthe degree of splenomegaly may correlate with diseaseprogression.*'" Enzyme replacement therapy may re-sult in hematologic improvement and reduction in or-ganomegaly and bone pain for as long as 5 years.* '̂Before enzyme replacement therapy, most patientswith Gaucher's disease required splenectomy. Sple-nectotny is now reserved for persistetit pancytopeniasor pain, which may result from massive splenic ex-pansion with associated infarcts. However, total sple-nectomy may promote accumulation of glucocerebro-sides in other reticuloendothelial sites. Therefore,partial spienectomy is a reasonable option to limit theeffects of hypersplenism and splenomegaly whileminimizing the infectious and tnetabolic conse-quences 62

Wiskott-Aldrich Syndrome

Wiskott-Aldrich syndrome (WAS) is an X-linkedimmunodeficiency associated with autoimmune andinflammatory sequelae, including hemolytic anemia,thrombocytopenia. neutropenia, arthritis, va.sculitis.and eczema.''' Deranged natural killer T cell (NKT)function, which is known to possess immunoregulato-ry properties, may in part account for the autoimmunemanifestations.'^ Bone marrow transplantation is theonly definitive therapy for WAS but in some centers isreserved only for those with life-threatening compli-cations.*^ Although the hematologic manifestations ofWAS may respond to treatment with corticosteroids.immunosuppressive agents, or intravenous immuno-globulin. splenectomy is indicated for persistentthrombocytopenia.^"*

fChediak-Higashi Syndrome

Chediak-Higashi syndrome (CHS) is an autosomal-recessive immunodeficiency disorder involving defec-tive leukocyte chemotaxis and phagocytosis. CHSmay degenerate into an accelerated phase involvingpancytopenia and cellular infiltration of the spleen.'̂ '̂ ^Splenectomy may effectively address manifestationsof hypersptenism when they are unresponsive to othermodalities, including corticosteroids.

Splenic Cysts

Cystic lesions of the spleen may be divided intoprimary or true cysts and secondary or false cysts.True splenic cysts are parasitic, congenital, or neoplas-tic and typically have an epithelial lining. In contrast,false cysts lack a true epithelial lining and are mostcommonly a consequence of trauma. Splenic cystsmay cause abdominal pain, become secondarily in-fected, or rupture. The appropriate management ofsplenic cysts depends on an understanding of theirderivation and natural history.

Primary splenic cysts may originate as embryologicmesothelial inclusions*'-̂ or represent a neoplastic pro-cess in which the squamous epithelial lining may bekeratinized/'^ However, on a global basis, parasiticcysts are far more common than nonparasitic cysts ofthe spleen."̂ ^ Although splenectotny may be necessaryin the setting of excessively large cysts, splenic pres-ervation is often possible. Cyst enucleation, partialcystectomy, and partial splenectomy have proven to betechnically feasible. Splenic preservation may be pos-sible in over 40 per cent of patients with echinococcaldisease with recurrence tates comparable to those hav-ing undergone splenectomy.^^'

False cysts or pseudocysts of the spleen most com-monly arise after trauma and account for the majorityof nonparasitic splenic cysts.̂ "̂ Interestingly, only 30per cent of such patients can recall a traumatic event.Splenic pseudocysts likely result from ihc formation ofan inflammatory fibrous capsule surrounding the he-matoma, which ultimately resorbs. A calcified rim isoften present around the cyst (Fig. 5j and, in the ab-

FiG. 5. Splenic pseudocyst. Several months after blunt ab-dominal trauma, this patient presented with left upper quadrantpain. A cotiiputcd tomography scan of the abdomen demonstratinga large, well-defined cystic lesion of the spleen with a calcified rim(arrow). These imaging features are typical for a splenic pseudo-cyst.


sence of documented parasitic disease, is almost al-ways indicative of a trauma. Small pseudocysts (<4cm) tend to regress and should be observed in theabsence of specific symptoms. Symptomatic pseudo-cysts or those greater than 5 cm should be treatedsurgically. If possible, preservation of a portion of thespleen shouid be attempted. In this regard, surgicaloptions include cystectomy without splenectomy, par-tial splenectomy, or "decapsulization" of the cyst withor without omentopexy. Any of the aforementionedoptions may be approached traditionally or laparo-scopically. It has been our experience that splenicpseudocysts >6 cm, especially those involving the me-dial portion of the spleen, have a tendency to recurwhen unroofed laparoscopically despite excising alarge portion of the cyst wall. In our opinion, thesepseudocysts are best managed by either partial splenicresection (Fig. 6). Occasionally, total splenectomymay be required {Fig. 7). Although percutaneousdrainage of splenic pseudocysts has been described,the effectiveness of this approach is uncertain and the

risk of secondarily infecting the lesion is not insignifi-cant.

Splenic Abscesses

Perhaps owing to the potent immunologic capacityof resident .splenic leukocytes.''- ^^ abscesses are rarelyencountered in the spleen. The predominant manifes-tations are systemic such as ieukocytosis and fever.Few patients present with localized symptoms. Splenicabscesses are unilocular in 65 per cent, multilocular in8 per cent, and multiple in 27 per cent of cases/"** Theoverall mortality rate is 13 per cent."^" Like with othersplenic lesions, abscesses may rupture with the poten-tial for hemorrhage and disseminated peritonitis. Im-munodeficiency is a common underlying factor andpredisposes to fungal abscesses (Fig. 8). Other poten-tial causes include endocarditis, intravenous drugabu.se, pyogenic infections at distant sites, sickle cellanemia, and secondary infection of a traumatic pseu-docyst.''

B.

D,

Fic. 6. Excision of a splenic pseudocyst. (A) Computed tomography scan of the abdomen demonstrating ;t large, well-delined cysticlesion of the spleen, which was found several weeks after blunt trauma. (B) The border between the pseudocyst and normal splenicparenchyma was identified (arrow). A cystectomy was pertbrmed. and the (C) intact pseudocyst and (D) resid'ual spleen iu:c demon-strated. Note that the greaier omentum Is generally conveniently situated for use in eliminating the cavity in the spleen after cyst excision.


FiG. 7. Total splenectomy performed for pseudocyst. At times, splenic preservation is not possible given the si/e or location of apseudocyst. In this case, splenectomy was performed for the treatment of this symptomatic pseudocyst. Note that the normal splenicparenchyma was markedly compressed (arrows) as a consequence of the pseudocyst (A and B). _

multilocular abscess is encountered. Under these con-ditions, splenectomy would appear to be the most ef-fective definitive therapy.

Wandering Spleen and Splenic Torsion

Failure of the dorsal mesogastrium to fuse lo theposterior abdominal wall may result in the absence ofthe spleen's normal peritoneal attachments. The lackof normal attachments such as the splenorenal andgastrosplenic ligaments predisposes the spleen to tor-sion and ischemia. Acquired factors such as the hor-monal milieu of pregnancy may play a role as well.^-Although most patients are asymptomatic, splenic tor-sion may manifest as recurrent bouts of abdominalpain. Persistent pain with a palpable, mobile mass sug-gests ischemia and possible infarction. The diagnosismay be reached by noting absence of intravenous con-trast uptake within the spleen on computed tomogra-phy scanning. In the event of torsion with infarction,

. , . , , splenectomy isFIG. 8. Fungal abscesses ot the spleen. Multiple low-

attenuating lesions of the spleen representing abscesses resultingfrom Caiiduhi alhicans. In this case, drainage was not possible asa result of the number of abscesses. The patient was treated withsystemic antifungal agents and splenectomy was avoided.

Intervention, in some form, is mandatory for splenicabscesses because the mortality rate for untreatedcases approaches 100 per cent. Initial managementshould be nonoperative, and image-guided percutane-ous is particularly effective when the abscess is soli-tary and thick-walled.^' If appropriate antibiotics havebeen administered in conjunction with percutaneousdrainage in patients with unilocular abscesses and theclinical picture fails to improve within 48 hours, sple-nectomy should be undertaken without delay. Percu-taneous drainage is less likely to be effective when a

Trctuma

Although splenectotny is often necessary after trau-matic injury, nonoperative management of bluntsplenic injury in the stable patient is currently the stan-dard of care.'''* Spontaneous healing results in ana-totnic and functional restoration of the spleen.' Suc-cessful nonoperative management is also possible inpatients with intrinsic splenic pathology. Eleven pa-tients with HIV. acute leukemia, infectious mono-nucleosis, or sickle cell anemia were successfullymanaged without surgical intervention after splenicinjury, demonstrating their ability to heal parenchymaldisruption (Fig. 9).'^^ However, in the setting of ongo-ing hemorrhage, hemodynamic instability, or con-


t-Ki. y. SplcniL- rupitnc in (he setting of HIV infeciion. The spleen undergoes pathologic changes in the setting o\' HIV infection. Inthis HIV-positive paiient. a signihcant splenic injury was inctirred after a motor vehicle accident. The splenic injury (A) was managednonoperatively and a computed tomography scan I month after the injury reveals nearly complete healing.

comitant intraabdominal injury, splenectomy shouldbe undertaken without unnecessary delay.^^ Toachieve splenic preservation, a multidisciplinary ap-proach may be necessary. Adjunctive splenic emboli-zation may increase the chances for successful nonop-erative management in appropriate patients.^-" Whenoperative intervention is required, splenorrhaphyshould be considered depending in the stability of thepatient (Fig. 10).

latrogenic Injuries

Of note, up to 40 per cent of all splenectomies per-fomied in the United States are a result of iatrogenic

injuries caused during left hemicolectomies. antirefluxprocedures, and left nephrectomies.^^ With techniqtiesdescribed in this article, the overwhelming majority ofthese injuries can be managed without resorting sple-nectomy. Although preservation of splenic paren-chyma should be attempted, splenectomy is the pro-cedure of choice in the setting of excessive blood loss.

Surgical Therapy

Medical therapy may be inadequate or, at times,inapptopriate in some patients with diseases involvingthe spleen. Patients may experience persistent abdomi-nal pain, left shoulder pain, or early satiety. Massively

_ FIG. 10. Spicnorrhaphy splenic preservation alter iniiimatic injury lA) to the s|>kvn is o!k-n possible and is desirable to ;noiil ilicrisk of overwhelming postspleneciomy infection with its associated high monality rate. In this patient, suture splenorrhaphy wasperformed with the edges of the injured splenic parenchyma reapproximated, whereas compression of the intervening tissue achievedhemostasis (B).


enlarged spleens may also cause an alteration of bowelhabits or urinary frequency resulting from bladdercompression. Nonoperative interventions may also failto control anemia, thrombocytopenia. or neutropenia.leaving the patient susceptible to a variety of second-ary complications. In some instances, splenectomytnay also be necessary to reach a diagnosis such as incases of idiopathic hypersplenisrn. As noted earlier,rnany of these patients harbor occult malignancy andtherefore pathologic analysis of the splenic paren-chytna is important when a cause for splenomegaly isnot apparent. The importance of the aforementionedstatement cannot be overemphasized, especially if atiiinimally invasive approach is used. Under these cir-cumstances, the extraction site should be large enoughso that adequate splenic parenchymal tissue is pre-served for pathologic analysis.

Open Splenectomy

After the decision to perform splenectomy on a pa-tient, the surgeon must ensure that the patient has re-ceived the appropriate immunizations and that hema-tologic deficiencies have been sufficiently corrected.Patients should be immunized against pneumococcus.Haemoplnliis infhienzae type B. and group C menin-gococci.''̂ ^ Influenza vaccination may be considered aswell. In addition, all patients should be prepared inanticipation of open splenectomy given the chance ofconversion even if a laparoscopic approach is initiallyundertaken.

When performing open splenectomy (OS), adher-ence to sound principles of splenic surgery will maxi-mize the chance of a successful outcome. Moreover,most of the concepts used in performing open sple-nectomy are applicable to the laparoscopic approach.The major pitfalls of splenectomy include hemor-rhage, injury to the greater curvature of the stomach,pancreatic injury, failure to detect accessory spleens,and iatrogenic rupture of the spleen with subsequentimplantation of splenic tissue within the peritonealcavity. Ensuring complete hemostasis in the bed of thespleen at the conclusion of the operation is crucial.Areas deserving careful attention include the tail of thepancreas, the left adrenal gland, the posterior abdomi-nal wall, and the left diaphragmatic surface. Diffuseoozing may require the administration of blood prod-ucts, depending on the patient's underlying pathologiccondition. To avoid pancreatic injury, the surgeonshould clearly identify the pancreatic tail when man-aging the vasculature at the splenic hilum. In addition,the short gastric vessels tnust be defined and ligatedwith precision to avoid injury to the stomach and apotential gastric fistula.^"'

Early control and ligation of the arterial inflow is

itiiportant not only to reduce the size of the spleenwhen it is enlarged, but it also allows for the safetransfusion of platelets without the fear that they willbe rapidly consumed by the spleen. As the spleen de-compresses and shrinks in size, further manipulation isfacilitated and the risk of iatrogenic is significantlyreduced. Moreover, maintaining the integrity of thesplenic capsule is advisable for both oncologic andhemostatic reasons to minimize the chance of tumordissemination or bleeding from the parenchyma.

Minimally Invasive Splenectomy

Laparoscopic splenectomy (LS) was initially de-scribed in 1991.''' Because of a constellation of fac-tors, including the advent of ultrasonic dissectors, im-proved optical technology, and use of endoscopicstapling devices, laparoscopic splenectomy is cur-rently the preferred approach for benign splenic dis-eases.^' The conversion rate for laparoscopic splenec-tomy performed for hematologic disease has beenreported to be 5 per cent.^' Although LS is associatedwith increased operative titnes. the duration of hospitalstays are shorter and splenectomy-related morbidity isdecreased.'̂ ^ Specifically, when splenectomy was per-formed for hematologic disease, the laparoscopic ap-proach resulted in earlier initiation of oral feeding,earlier discharge, fewer blood transfusions, and a de-creased analgesia requirement.^^

Patients with ttialignant disease requiring splenec-totny may have longer operative times and more bloodloss during LS.**' This is most likely related to thehigher splenic volume typically presetit in patientswith malignant disease in contrast to the benign pro-cesses such as ITP. Despite the increased operativetime and blood loss, when LS is performed in thesetting of splenomegaly, there were no statistically dif-ferences in conversion rates, titne to discharge, or tiior-bidity.**"* Splenomegaly cannot be considered an abso-lute contraindication to a minimally invasive approachbut warrants caretul patient selection and preoperativeplanning.'̂ ^ Hand-assisted laparoscopy may be of ad-ditional benefit when performing splenectomy in pa-tients with splenomegaly and allow for removal of theintact organ when a histologic diagnosis is required.*^^

One criticism of the laparoscopic approach to sple-nectomy has been the lack of tactile feedback andperhaps an impaired ability to identify accessoryspleens. The incidence of accessory spleens is ap-proximately 15 per cent.^^ Failure to detect and ablateaccessory splenic tissue tnay lead to treatment failurein cases in which the spleen is responsible for thedestruction of platelets, erythtocytes. or leukocytes.These concerns, however, have not proven to be asignificant itnpediment. Like in all of laparoscopic


surgery, lack of tactile sense is compensated for byexperience. Furthermore, laparoscopy may instead fa-cilitate detection of accessory splenic tissue becausevisualization is usually better than with an open ap-proach. In a large meta-analysis. the rate of accessoryspleen identification was similar among patients un-dergoing LS and OS.**̂

Splenic-Preserving Techniques

Partial splenectomy may be appropriate for treat-ment for certain benign conditions in which residualsplenic tissue will not have undue hematologic or on-cologic consequences. In most situations, preservationof sufficient normal splenic tissue lowers the risks ofinfectious complications. Between 30 per cent and 50per cent of splenic tissue should be preserved alongwith an identifiable blood supply to preserve mean-ingful splenic function.' Subtotal splenectomy per-formed in children for hereditary spherocytosis canresult in a prolonged decrease in hemolysis while of-fering the benefit of retained reticuloendothelial func-tion in the splenic remnant.̂ ** As discussed earlier,splenic cysts have been treated by partial excision ofthe cyst wall or with "unroofing" procedures, and pa-tients with Gaucher's disease may benefit from partialsplenectomy. Partial cystectomy with omentopexy hasbeen used in the management of hydatid disease**'̂ andtraumatic pseudocysts have been effectively treatedwith subtotal splenectomy.''^ Approaches to splenicpreservation in the setting of trauma include the use ofsuture splenorrhaphy. mesh splenorrhaphy. fibrin glue,and argon beam laser coagulation.'

Splenic Bed Drainage

Drainage of the splenic bed remains a controversialissue. The theoretic rationale for drainage of thesplenic bed after splenectomy includes the removal ofaccumulated fluid, which could become a nidus forinfection. Yet, it is uncertain if drainage of the splenicbed is of any significant benefit. Associated gastroin-testinal injury and duration of drainage seem to be ofgreater significance in terms of postoperative morbid-ity than the presence or absence of drains."" If drainsare to be used, then the closed suction variety such asa Jackson-Pratt should be used. To minimize infec-tious complication, drains should be removed asquickly as possible.

Postsplenectomy Complications

Early Morbidity

After splenectomy for hematologic conditions, thepostoperative complication rate may reach 24 per cent.Patients having received corticosteroids are at highest

risk for subsequent infectious complications.'-* In ad-dition to commonly recognized postoperative compli-cations such as wound infection, pneumonia, and sep-sis, patients may develop a pancreatic or gastricfistula. When splenectomy is performed for hemato-logic disease, particular attention must be paid to theblood and platelet coutits in the perioperative period.

Splenectomy tnay also be associated with an el-evated risk for adverse cardiovascular events. Patientswho undergo splenectomy for hereditary spherocytosismay experience a higher incidence of tnyocardial in-farction, stroke.* '̂ and pultnonary hypertension.'*- Inaddition, splenectomy places patients at risk for otherthromboembolic cotnplications^'' such as splenic orportal vein thrombosis. Thrombotic prophylaxis withheparin products"^ and screening Doppler sonogra-phy"*"*̂ should be considered in high-risk patients.When screened with ultrasonography. poiial or splenicvein throtnbosis occurred in 55 per cent of patientsafter LS and 19 per cent after OS.'^'' Furthermore, pa-tietits with fever or abdotninal pain after splenectomyshould be suspected of having portal and splenic veinthrombosis should be treated with anticoagulationtherapy if this condition is documented.****

An additional clinical manifestation of splenectomymay be thrombocytosis. Theoretically, thrombocytosisafter splenectomy may increase blood viscosity andcreate a more thrombogenic milieu. However, the re-lationship between postsplenectomy thrombocytosisand venous thrombosis is unclear.'**' The use of aspirinto treat postsplenectotny thrombocytosis must be con-sidered on an individual basis taking into account theoverall thrombotic diathesis, particularly the presenceof an underlying myeloproliferative process. Most sur-geons initiate antiplatelet or antithrombotic therapy ata platelet count greater than 750.00/tnL.'" Despite thopotential for these complications, splenectomy can beperformed safely in the vast majority of patients withbenign and tnalignant disease who may benefit fromremoval of the spleen.

LiUe Morbidity

The immunologic effects of spienectomy includeimpaired responsiveness to new antigens, itnpairedphagocytic clearance of opsonized or unopsonizedbacteria, and impaired production of the opsonin pro-teins tuftsin and properdin.'**^ Therefore, patietits hav-ing undergone splenectomy are at lifelong risk frombacterial sepsis.

Encapsulated bacteria such as Streptococcus pneu-moniae are the most common pathogens involved inpostsplenectomy sepsis.'*'̂ The precise incidence ofOPSI is not known because published reports varyconsiderably, which is in part the result of variations in


the study populations and definitions of OPSL Theannual incidence of OPSI has been reported to be ashigh as 0.23 per cent to 0.42 per cent'"" per year witha lifetime risk being 5 per cent.̂ ** In contrast. CuUing-ford repotted the incidence of OPSI to be 0.04 per 100person-years.'"' Ahhough it is difficult to determinethe precise incidence for OPSI, it is clear that its trueoccurrence has been often overstated.

The clinician must assess the individual patient'stisk based on factors including the time interval fromsplenectomy. age, and underlyitig medical conditions.The risk of infection is highest in the first 2 years aftersplenectomy.'"- although it is unclear whether the riskof frank OPSI declines over time. '"^ The risk of OPSIis elevated in children, in those with underlying he-matologic or malignant processes, and those otherwiseimmunosuppressed.'"-^ Moreover, in patients withthaiassemia major and sickle cell anemia, the inci-dence of infection and associated mortality is notablyhigher.'"^ Less controversial is the lethal nature orOPSI. After the occurrence of OPSI, the mortalityrates reported to range from 38 per cent to 69 percent.''" Therefore, the primary fear with OPSI may liein its lethality rather than the frequency with whichthis complication occurs.

Several strategies to reduce the risk of OPSI havebeen used. Preservation of splenic tissue after traumaor utilization of splenic-sparing surgical techniqueswhen possible should prevent OPSI when an adequatevolume of parenchyma remains. Ideally, to maintainsplenic immunoiogic function, at least 30 per cent ofsplenic parenchyma attached to an identifiable bloodsource should be preserved. Although autotrans-planted splenic tissue has been shown to be both viableand functional, the true clinical impact of this proce-dure is uncertain^^ because the response to a specificbacterial antigenic challenge in humans is presentlyunknown. However, given its conceptual appeal andlow risk of complications, autotransplantation ofsplenic tissue, if feasible, should be attempted.

If the entire spleen must be removed, the patientshould be immunized with the Pneumovax vaccineagainst S. pneumoniae. Ideally, vaccinations should begiven before splenectomy because the humoral re-sponse may be better with the spleen intact. Boosterimmunization should be given at 3- to 5-year inter-vals.'̂ Vaccination against Haemophilus influenzaetype b. Neisseha meningitidis type C. and influenza isgenerally recointnended as well. Ahhough prophylac-tic antibiotics should be given in the event of anydental or invasive medical procedure, routine antibi-otic administration should be limited to 5 years inchildren and 2 years in adults after splenectomy.'"^

Any sign of infection in an asplenic individual mustbe considered a true medical emergency. The patient

should receive immediate broad-spectrum intravenousantibiotic coverage subsequent to the withdrawal ofblood samples for culture. Bacteremia can be demon-strated in nearly all cases of OPSI and bacteria areoften present on a peripheral smear of the patient'sblood. '̂̂ Although intravenous immunoglobulin istheoretically appealing, evidence for its routine use islacking.

I I

, Summary

The spleen is an important component of the im-mune and hematologic systems. When the spleen isresponsible for or involved with pathologic processes,or when splenomegaly produces significant symp-toms, operative therapy may be indicated. Given thepotenfial complications of splenectomy, includingOPSI, preservation of a critical mass of splenic paren-chyma should be attempted when possible. In in-stances in which splenectomy is indicated, the laparo-scopic approach may offer significant benefits. Asound understanding of the underlying pathophysiol-ogy of the process necessitating splenectomy and an-ticipation of the potential postsplenectomy complica-tions will optimize the likelihood of successful clinicaloutcomes.

REFERENCES

1. Pachter HL, Grau J. The current status of splenic preserva-tion. Adv Surg 2000;34:L^7-74.

2. Larsen W. Htiman Embryology. New York: Churchill Liv-ingstone; 1993.

3. Groom AC. The Microcirculatory Society Eugene M. Landisaward lecture. Microcirculation of the spleen: new concepts, newchallenges. Microvasc Res 1987:34:269-89.

4. Katz SC. Piilarisetty VG. Bleier Jl. et al. Liver sinusoidalendothelial cells are insufficient to activate T cells. J Immunol2OO4;173:23O-5.

5. Jandl JH. Aster RH. Increased splenic pooling and the patho-genesis of hypersplenism. Am J Med Sci l967;253:383-98.

6. Aster RH. Pooling of platelets in the spleen: rule in thepathogenesis of 'hypersplenic' thrombocytopenia. J Clin Invest1966:45:645-57.

7. Jandl JH. Greenberg MS. Yonemoto RH. Castle WB. Clini-cal determination of the sites of red cell sequestration in hemolyticanemias. J Clin Invest 1956;35:842-67.

8. Piilarisetty VG. Shah AB. Miller G, et al. Liver dendriticcells are less immunogenic than spleen dendritic ceils because ofdifferences in subtype composition. J Immunol 2004;I72:1009-17.

9. Katz SC, Piilarisetty VG, Bleier JL et al. Conventional liverCD4 T cells are functionally distinct and suppressed by environ-mental factors. Hepatology 2005:42:293-300.

10. Lux SE, John KM. Isolation and partial characterization ofa high molecular weight red cell membrane protein complex nor-mally removed by the spleen. Blood 1977;50:625-4L

11. Brown El. Hosea SW, Frank MM. The role of the spleen in


experimental pneumococcal bacteremia. J Clin Invest 1981:67:975-82.

12. Hammarstrom L. Smith Cl. Development of anti-poly saccharide antibodies in asplenic children. Clin Exp Immunol1986:66:457-62.

13. McMillan R. Autoantibodies and autoantigens in chronicimmune thrombocytopenic purpura. Semin Hematol 2()()():37:239^8.

14. Mu.sser G, Lazar G, Hocking W. Busuttil RW. Splenectomyfor hematologic disease. The UCLA experience with 306 patients.Ann Surg 1984:200:40-5.

15. Kojouri K. Vesely SK. Terrell DR. George JN. Splenec-tomy for adult patients with idiopathic thromhocytopenic purpura:a systematic review to assess long-term platelet count responses,prediction of response, and surgical complications. Blood 2004:104:2623-34.

16. Wilheim MC, Jones RE. McGehee R, et al. Splenectomy inhematologic disorders. The ever-changing indications. Ann Surg1988:207:581-9.

17. Bowdler AJ. The role of the spleen and splenectomy inautoimmune hemolytic disease. SemIn Hemato! 1976:13:335-48.

18. Balint GP, Balint PV. Felty's syndrome. Best Pract ResClin Rheumatol 2004:18:631^5.

19. Pereira J. Velloso ED. Loterio HA. et al. Long-term remis-sion of neutropenia in Eelty's syndrome after a shon GM-CSFtreatment. Acta Haematol 1994:92:154-6.

20. Moore RA. Brunner CM. Sandusky WR. Leavell BS. Fel-ty's syndrome: long-term follow-up after splenectomy. Ann InternMed 1971:75:381-5.

21. Rashba EJ, Rowe JM, Packman CH. Treatment of the neu>tropenia of Felty syndrome. Blood Rev l996;]0:177-84.

22. Bukowski RM. Thrombotic thrombocytopenic purpura: areview. Prog Hemost Thromb 1982:6:287-337.

23. Bukowski RM. Hewlett JS. Reimer RR. et al. Therapy ofthrombotic thrombocytopenic purpura: an overview. SeminThromb Hemost 1981:7:1-8.

24. Schwartz SL Role of splenectomy in hematologic disorders.World J Surg 1996:20:1156-9.

25. Schwartz SI. Splenectomy for hematologic disease. SurgClin North Am l981;61:l!7-25.

26. Zanella A, Fermo E. Bianchi P, Valentini G. Red cell py-ruvate kinase deficiency: molecular and clinical aspects. Br J Hae-matol 2005:130:11-25.

27. Watanabe Y, MiyauchI K. Horiuchi A. et al. Concomitantlaparoscopic splenectomy and cholecystectomy as an effective andminimally invasive treatment of pyruvate kinase deficiency withgallstones. Surg Bndosc 2002:16:1495.

28. Johnson RM, Ravindranath Y, el-Alfy M. Goyette G Jr.Oxidant damage to erythrocyte membrane in glucose-6-phosphatedehydrogenase deficiency: correlation with in vivo reduced gluta-[hione concentration and membrane protein oxidation. Blood1994;83:l 117-23.

29. Al-Sa!em AH, NaseruUah Z, Qaisaruddin S. et al. Spleniccomplications of the sickling syndromes and the role of splenec-tomy. J Pediatr Hematol Oncol 1999:21:401-6.

30. al-Salem AH, Qaisaruddin S, Nasserallah Z. al Dabbous 1.al Jam'a A. Splenectomy in patients with sickle-cell disease. AmJ Surg 1996:172:254-8.

31. Badaloo AV. Singhal A. Forrester TE, et al. The effect ofsplenectomy for hypersplenism on whole body protein turnover.

resting metabolic rate and growth in sickle cell disease. Eur J ClinNtitr l9%:50:672-5.

32. Pinna AD, Argiolu F. Marongiu L, Pinna DC. Indicationsand results for splenectomy for beta thaiassemia in two hundredand twenty-one pediatrlc patients. Surg Gynecol Obstet 1988:167:109-13.

33. Carr JA. Shurafa M. Velanovich V. Surgical indications inidinpatliic splenomegaly. Arch Surg 2(H)2:l37:64-8.

34. Cronin CC, Brady MP, Murphy C, et al. Splenectttmy inpatients with undiagnosed splenomegaly. Postgrad Med J 1994:70:288-91.

35. Taylor MA. Kaplan HS. Nelsen TS. Staging laparotomywith splenectomy for Hodgkin's disease: the Stanford experience.World J Surg 1985:9:449-60.

36. Bangerter M. Griesshammer M. Bergmann L. Progress inmedical imaging of lymphoma and Hodgkin"s disease. Curr OpinOncol 1999:11:339-42.

37. Sombeck MD. Mendenhall NP. Kaude J V, et al. Correlationof lymphangiography, computed tomography, and laparotomy inthe staging of Hodgkin's disease. Int J Radiat Oncol Bitil Phys1993:25:425-9.

38. Martinet O, Bettschart V. Scholl B. Suter M. Value of lap-aroscopic staging for Hodgkin disease. Surg Laparosc Endosc Per-cutan Tech 2000:10:335-7.

39. Lehne G. Hannisdal E, Langholm R, Nome O. A 10-ycarexperience with splenectomy in patients with malignant noii-Hodgkin's lymphoma at the Norwegian Radium Hospital. Cancer1994:74:933-9.

40. Morel P. Duprie? B. Gosselin B. et al. Role of early sple-nectomy in malignant lymphomas with prominent splenic involve-ment (primary lymphomas of the spleen). A study of 59 cases.Cancer 1993:71:207-15.

41. Berman RS. Feig BW. Hunt KK, et al. Platelet kinetics anddecreased transfusion requirements after splenectomy for hemato-logic malignancy. Ann Surg 2004:240:852-7.

42. Khouri 1, Sanchez F, Deisseroth A. Leukemias. In: DeViuiV. Heliman S. Rosenberg S. eds. Cancer: Principles and Practiceof Oncology. Philadelphia: Lippincott-Raven; 1997:2285-321.

43. Bouvet M, Babiera GV. Termtihlen PM. et al. Splenectomyin the accelerated or blastic phase of chronic myelogenous leuke-mia: a single-institution. 25-year experience. Surgery 1997:122:20-5.

44. Results of a prospective randomized trial of early spienec-tomy in chronic myeloid leukemia. The Italian Cooperative StudyGmup on Chronic Myeloid Leukemia. Cancer 1984:54:333-8.

45. Cusack JC Jr. Seymour JF. Lemer S. et al. Role of sple-nectomy in chronic lymphocytic leukemia. J Am Coll Surg 1997:185:237^3.

46. Au WY, Klasa RJ, Gallagher R. et al. Second malignanciesin patients with hairy cell leukemia in British Columbia: a 20-yearexperience. Blood 1998:92:1160-1.

47. Kraut EH. Grever MR. Bouroncle BA. Long-term follow-up of patients with hairy cell leukemia after treatment with 2'-deoxycoformycin. Blood 1994:84:4061-3.

48. Morgensterii L. Rosenberg J. Geller SA. Tumors of thespleen. World J Surg 1985:9:468-76.

49. Burke JS. Surgical pathology of the spleen: an approach tnthe differential diagnosis of splenic lymphomas and leukemia.-..Part I. Diseases of the white pulp. Am J Surg Pathol551-63.


50. Goyal A. Babu SN. Kim V. et ai. Hemangiocndothelioma ofliver and spleen: trauma-induced consumptive coagulopathy. J Pe-diatr Surg 2OO2;37:E29.

51. Chan KW. Saw D. Distinctive, multiple lymphangiomas ofspleen. J Pathol 1980:131:75-81.

52. Falk S. Stutte HJ. Frizzera G. Littoral cell angioma. A novelsplenic vascular lesion demonstrating histiocytic differentiation.Am J Surg Pathol 1991:15:1023-33.

53. Morgenstem L, McCafferty L. Rosenberg J. Michel SL.Hamartomas of the spleen. Arch Surg 1984; 119:1291-3.

54. Easier RH. Dowlin WM. Primary lipoma of the spleen.Report of a case. Arch Pathol 1969:88:557-9.

55. Huibert JC, Graf R. Involvement of the spleen by renalangiomyolipoma: metastasis or multicentricity? J Urol 1983:130:328-9.

56. Popper H. Thomas LB. Alterations of liver and spleenamong workers exposed to vinyl chloride. Ann N Y Acad Sci1975:246:172-94.

57. Rapparot H. Tumors of the hematopoieiic system. In: Pa-thology AFIo. ed. Atlas of Tumor Pathology. Washington, DC;1966.

58. Bowne WB. Lewis JJ. Filippa DA. et al. The managementof uniccntric and multicentric Castleman's disease: a report of 16cases and a review of the literature. Cancer I999;85:7()6-17.

59. Macheers SK. Mansour KA. Management of isolatedsplenic metastases from carcinoma of the lung: a case report andreview of the literature. Am Surg 1992:58:683-5.

60. Niederau C. Haussinger D. Gaucher's disease: a review forthe internist atid hepatologist. Hepatogastroenterology 2000;47:984-97.

61. Weinreb NJ. Charrow J. Andersson HC, et al. Effectivenessof enzyme replacement therapy in 1028 patients with type I Gau-cher disease after 2 to 5 years of treatment: a report from theGaucher Registry. Am J Med 2002;l 13:112-9,

62. Rubin M. Yampolski I. Lambro/o R. et al. Partial splenec-tomy in Gaucher's disease. J Pediatr Surg 1986:21:125-8.

63. Dupuis-Girod S. Medioni J. Haddad E. et al. Autoimmunityin Wiskott-Aldrich syndrome: risk factors, clinical features, andoutcome in a single-center cohort of 55 patients. Pediatrics 2003;lll:e622-7.

64. Schurman SH. Candotti F. Autoimmunity in Wiskott-Aldrich syndrome. Curr Opin Rheumatol 2003:15:446-53.

65. Asian Y. Erduran E. Gedik Y. et al. The role of high dosemethylprednisolone and splenectomy in the accelerated phase ofChediak-Higashi syndrome. Acta Haematol 1996:96:105-7.

66. Atmatzidis K. Papaziogas B, Mirelis C, et al. Splenectomyversus spleen-preserving surgery for splenic echinococcosJs. DigSurg 2(H)3;2O:527-3I.

67. Pachter HL, Hofstetter SR, Elkowitz A. et al. Traumatic-cysts of the spleen—the role of cystectomy and splenic preserva-tion: experience with seven consecutive patients. J Trauma 1993:35:430-6.

68. Piilarisetty VG. Katz SC. Bleier Jl, et al. Natural killerdendritic cells have both antigen presenting and lytic function andin response to CpG produce IFN-gamma via autocrine IL-12. JImmunol 2005:174:2612-8.

69. Ooi LL, Leong SS. Splenic abscesses from 1987 to 1995.Am J Surg 1997:174:87-93.

70. Nelken N. Ignatius J. Skinner M. Christensen N. Changing

clinical spectrum of splenic abscess. A multicenter study and re-view of the literature. Am J Surg [987:154:27-34.

71. Ulhaci N. Meteoglu I. Kacar F, Ozbas S. Abscess of thespleen. Pathol Oncol Res 2004:10:234-6.

72. Buehner M. Baker MS. The wandering spleen. Surg Gyne-col Obstet 1992; 175:373-87.

73. Sayeed S, Koniaris LG, Kovach SJ. Hirokawa T. Torsion ofa wandering spleen. Surgery 2002:132:535-6.

74. Sharma OP. Oswanski MF. Singer D, et al. Assessment ofnonoperative management of blunt spleen and liver trauma. AmSurg 2005:71:379-86.

75. Guth AA. Pachter HL, Jacobowitz GR. Rupture of thepathologic spleen: is there a role for nonoperative therapy? JTrauma 1996:41:214-8.

76. Cogbili TH. Moore EE. Jurkovich GJ. et al. Nonoperativemanagement of blunt splenic trauma: a multicenter experience. JTrauma 1989:29:1312-7.

77. Haan JM, Bochicchio GV. Kramer N, Scalea TM. Nonop-erative management of blunt splenic injury: a 5-year experience. JTrauma 2005:58:492-8.

78. Cassar K, Munro A. latrogenic splenic injury. J R Coll SurgEdinb2OO2;47:731^L

79. Davies JM. Bames R, Milligan D. Update of guidelines forthe prevention and treatment of infection in patients with an absentor dysfunctional spleen. Clin Med 2(K)2:2:440-3.

80. Scott-Conner C. Chassin's Operative Strategy in GeneralSurgery. 3rd ed. Iowa City: Spring Science & Business Media:2002.

81. Rosen M, Brody F, Walsh RM. et al. Outcome of laparo-scopic splenectomy based on hematologic indication. Surg F.ndosc2002:16:272-9.

82. Winslow ER. Brunt LM. Perioperative outcomes of laparo-scopic versus open splenectomy: a meta-analysis with an emphasison complications. Surgery 2003:134:647-53.

83. Donini A, Baccarani U, Terrosu G. et al. Laparoscopic vsopen splenectomy in the management of hematologic diseases.Surg Endosc 1999;l3:1220-5.

84. Heniford BT. Park A. Walsh RM. et ai. Laparoscopic sple-nectomy in patients with normal-sized spleens versus splenomeg-aly: does size matter? Am Surg 2001:67:854-7.

85. Napoli A. Catalano C, SHecchia G. et ai. Laparoscopic sple-nectomy: multi-detector row CT for preoperative evaluation. Ra-diology 2004:232:361-7.

86. Smith L. Luna G, Merg AR, et al. Laparoscopic spienec-tomy for treatment of splenomegaly. Am J Surg 2004:187:618-20.

87. Rudowski WJ. Accessory spleens: clinical significance withparticular reference to the recurrence of idiopathic thrombocyto-penic purpura. World J Surg 1985:9:422-30-

88. Bader-Meunier B, Gauthier F. Archambaud F. et al. Long-term evaluation of the beneficial effect of subtotal splenectomy formanagement of hereditary spherocytosis. Blood 2001:97:399-403.

89. Ozdogan M. Baykal A. Keskek M. et al. Hydatid cyst of thespleen: treatment options. Int Surg 2001:86:122-6.

90. Pachter HL, Hofstetter SR. Spencer FC. Evolving conceptsIn splenic surgery: splenorrhaphy versus splenectomy and post-splenectomy drainage: experience in 105 patients. Ann Surg 1981;194:262-9.

91. Schilling RF. Spherocytosis, splenectomy, strokes, and heatattacks. Lancet 1997:350:1677-8.


92. Hoeper MM. Niedermeyer J, Hoffmeyer F. et al. Pulmonaryhypertension after splenectomy? Ann Intern Med 1999:130:506-9.

93. Winslow ER, Brunt LM, Drebin JA, et al. Portal veinthrombosis after splenectomy. Am J Surg 2002:184:631-5.

94. van't Riet M, Burger JW, van Muiswinkel JM. et al. Diag-nosis and treatment of portal vein thrombosis following splenec-tomy. Br J Surg 2000:87:1229-33.

95. Ikeda M. Sekimoto M. Takiguchi S, et al. High incidence ofthrombosis of the portal venous system after laparoscopic splenec-tomy: a prospective study with contrast-enhanced CT scan. AnnSurg 2005:241:208-16.

96. Randi ML, Fabris F, Dona S, Glrolami A. Evaluation ofplatelet function in postsplenectomy thrombocytosis. Folia Hae-matol Int Mag Klin Morphol Blutforsch 1987:114:252-6.

97. Pimpl W, Dapunt O, Kaindl H. Thalhamer J. Incidence ofseptic and throniboembolie-related deaths after splenectomy inadults. Br J Surg 1989,76:517-21.

98. Lynch AM, Kapila R. Overwhelming postsplenectomy in-fection. Infect Dis Clin North Am 1996;10:693-707.

99. Waghom DJ. Overwhelming infection in asplenic patients:current best practice preventive measures are not being followed.J Clin Patho! 2001:54:214-8.

100. Ejstrud P. Kristensen B, Hansen JB. et al. Risk and pat-terns of bacteraemia after splenectomy: a population-based study.Scand J Infect Dis 2000:32:521-5.

101. Cullingford GL, Watkins DN, Watts AD, Mallon DF. Se-vere late postsplenectomy infection. Br J Surg 1991:78:716-21.

102. Holdswonh RJ. Irving AD, Cuschieri A. Postsplenectomysepsis and its mortality rate: actual versus perceived risks. Br .1Surg 199l;78:lO3]-8.

103. Davidson RN, Wall RA, Prevention and management ofinfections in patients without a spleen. Clin Microbiol Itifect 2001:7:657-60.

104. Bisharat N. Omari H, Lavi I, Raz R. Risk of infection anddeath among post-splenectomy patients. J Infect 2001:43:182-6.

105. de Montalembert M, Lenoir G. Antibiotic prevention ofpneumococcal infections in asplenic hosts: admission of insuffi-ciency. Ann Hematol 2004:83:18-21.

Documents

Invited Commentary Indications for Splenectomy · splenectomy infection (OPSI). Mortality rates of OPSI may exceed 50 per cent in unvaccinated patients.' Therefore, removal of the