sign in sign up
<<
Home , Splenectomy

World J. Surg. 9, 437--443, 1985

1985 by the Soci~t~ lnternationale de Chirurgie

Splenectomy for and Secondary Hypersplenism

William W. Coon, M.D.

Department of , The University of Michigan Hospitals, Ann Arbor, Michigan, U.S.A.

Splenomegaly and secondary hypersplenism may be associ- The separation of hypersplenism into primary ated with acute and chronic infections, autoimmune states, and secondary categories is also imprecise. The portal hypertension or splenic vein thrombosis, and a diseases usually included under "primary" hyper- number of infiltrative and neoplastic conditions involving splenism are those in which the fundamental defect the . Our experience and that of others with these is thought to be related to a congenital or acquired various conditions demonstrates that the decision to per- alteration in cell membrane or structure of form splenectomy should be based on well-defined and hematopoietic cells (idiopathic thrombocytopenic often strictly limited indications. Except for idiopathic purpura, acquired , some of the splenomegaly, the presence and severity of secondary congenital hemolytic anemias, etc.). hypersplenism or severely symptomatic splenomegaly This discussion will be confined to selected enti- should be well documented. In each case, the potential for ties usually considered to be associated with "sec- palliation and known mean duration of expected response ondary" hypersplenism in which splenomegaly and must be weighed against the increased morbidity and altered splenic microcirculation may bring about mortality of splenectomy (as compared to operation for splenic sequestration of erythrocytes, neutrophils, "primary" hypersplenism). or . Here, again, the specificity of this terminology is debatable since, in several second- ary hypersplenic conditions (Felty's syndrome, Hypersplenism is a nonspecific term indicating that hairy cell ), altered cell structure may also the spleen may be functioning in a manner having a play a role in the pathophysiologic mechanism of deleterious effect on the blood cells of the host. The cytopenia. usual criteria utilized to indicate the presence of this Secondary hypersplenism has been associated condition are: (a) splenomegaly; (b) a clinically with certain acute and chronic infections, autoim- significant decrease in one or more of the cellular mune states (Felty's syndrome, systemic lupus elements of the blood; (c) the presence of a normal erythematosus), infiltrative and neoplastic condi- or hypercellular ; and (d) correction or tions involving the spleen (sarcoidosis, Gaucher's improvement of the cytopenia by splenectomy. disease, several myeloproliferative disorders, and However, in clinical application, all of these ele- ) and occasionally with "congestive ments may not be satisfied. Splenomegaly is not and splenomegaly" brought about by portal hyperten- sion or splenic vein thrombosis. should not be present in idiopathic thrombocy- topenic purpura. A cytopenia present in several of the myeloproliferative disorders may respond to Leukemia splenectomy in spite of the presence of decreased cellular precursors in the bone marrow. Splenectomy has been performed for several types of leukemia for more than 100 years, but its value and limitations have only been defined in the past Reprint requests: William W. Coon, M.D., Section of few years. General Surgery, D2214, Medical Professional Building, Only in the relatively rare is University of Michigan Hospitals, Ann Arbor, Michigan splenectomy the initial treatment of choice. Al- 48109, U.S.A. though this condition was recognized as a distinct 438 World J. Surg. Vol. 9, No. 3, June 1985 entity more than 25 years ago, in the past 10 years operation has been performed primarily for acute other related conditions have been appropriately symptoms and fear of splenic rupture, severe excluded as a result of a better definition of marrow , or highly symptomatic spleno- and splenic pathological findings and utilization of megaly [1]. Of 11 patients, 8 were operated on in the the tartrate-resistant acid phosphatase stain which "accelerated phase" of their disease [8] in which is positive in 85-90% of cases. This usually slowly median survival is only 3-6 months. Three died progressive disease is probably a variant form of postoperatively. In 2 patients splenectomy was per- chronic lymphocytic leukemia since the cell of formed as a preliminary to bone marrow transplan- origin appears to be a transformed lymphocyte. tation and both are alive (at 1 and 3 years, respec- Median survival after diagnosis is 3-4 years, but tively). Many reports attributing value to splenec- many patients live for 8-10 years or longer. Re- tomy for this condition have been anecdotal. How- sponses to intensive chemotherapy or to cortico- ever, several recent nonrandomized comparisons of steroids have been poor. Splenectomy is indicated operation versus no operation early in the chronic if the patient develops a hypersplenic complication phase of CGL have shown that splenectomy does such as severe anemia, thrombocytopenia, or not delay progression from the chronic to the blastic granulocytopenia. In our experience with 17 pa- phase and does not prolong survival [9, 10]. In a tients followed at our institution during the past 10 recent prospective trial, similar conclusions were years [1], 16 have had a splenectomy for one of reached [11]. these indications. There have been no operative deaths and only 1 postoperative superficial wound infection. Nine of the 16 splenectomized patients Stage IV are still alive with a median postoperative survival of more than 42 months. Only 2 patients have On rare occasions, patients with far-advanced required postoperative chemotherapy with low- Hodgkin's disease or non-Hodgkin's lymphoma will dose chlorambucil. Thirteen of 16 have achieved a develop severe hypersplenism which is refractory satisfactory hematologic response (hematocrit: 35 to chemotherapy, or a cytopenia interferes with vol % or above; granulocytes: 1,000/mm 3 or more; administration of adequate doses of chemothera- platelets: 100,000/mm 3 or greater) after operation. peutic agents. In our experience with 5 such pa- In a recent analysis of 391 cases [2], patients who tients, 1 died 3 weeks postoperatively with pneu- had a splenectomy survived longer than patients monia. The other 4 achieved a very satisfactory who did not undergo surgery; operative mortality cytologic response with survival from 5 months to rate was 2%. About 80% of patients still achieve an more than 8 years. The entire spectrum of "acceptable hematologic response" (as previously cytopenias were encountered: thrombocytopenia in defined) [2]. In patients who fail to achieve an 1 patient, hemolytic anemia in 1, hemolytic anemia adequate response to splenectomy, low-dose and thrombocytopenia in another, and pancyto- chlorambucil (4 mg/day) has been of value [3]. penia in the fourth. In 3 patients a sustained remis- Recently, bone marrow transplantation [4] or daily sion of hypersplenism was achieved until death administration of alpha interferon [5] have been from lymphoma. The fourth and longest surviving proposed as additional therapeutic modalities. patient had a relapse of a Coombs'-positive hemo- While many patients with chronic lymphocytic lytic anemia 2 years after operation which re- leukemia live for 5-10 years with little or no treat- sponded to steroid therapy and then remained in ment, a small fraction may develop hypersplenic remission until a terminal relapse 81/2 years after complications or massive symptomatic splenomeg- splenectomy. aly, usually relatively late in the course of this disease. Of our 15 patients operated on for one of Sarcoidosis these indications [1], 2 died after operation from persistent thrombocytopenic bleeding. Four pa- Although 6% of patients with sarcoidosis are said to tients with hemolytic anemia, 1 with pancytopenia, have splenomegaly [12], the need for splenectomy and 4 with symptomatic splenomegaly had excellent is extremely uncommon. In the largest reported responses to operation with a mean survival of series, splenectomy was performed for sympto- greater than 4 years. Of 6 patients with thrombo- matic splenomegaly in 7 patients, for infarction or cytopenia, only 2 achieved a satisfactory rise in rupture in 2, for hypersplenism in 5 (all of whom count. Others have reported similar results responded), and for other indications in 2 [13]. In [6, 7]. our personal experience with 8 patients, diagnosis The utilization of splenectomy in the manage- was the indication in 4 individuals who had no prior ment of chronic granulocytic leukemia (CGL) has diagnosis and manifestations compatible with pos- been much more controversial. In our hospital, the sible lymphoma (fever of undetermined origin, W.W. Coon: Splenectomy for Splenomegaly 439 splenomegaly, para-aortic lymphadenopathy, etc.). selection of patients and intensive preoperative One patient had symptomatic splenomegaly. In the preparation are essential. Our experience with a other 3 patients, surgery was performed for hemo- very short postoperative survival in subjects with lytic anemia or neutropenia and all achieved remis- secondary AMM indicates that surgery in this sub- sion after splenectomy. group is of questionable benefit. In patients with chronic anemia requiring frequent transfusion, a source of continued gastrointestinal blood toss Agnogenic Myeloid Metaplasia (AMM) should be eliminated. Determination of 51Cr- Selected patients with this myeloproliferative dis- erythrocyte survival may help separate those pa- order may obtain palliative benefit from splenec- tients with hypersplenic pooling and sequestration tomy, but operative morbidity and mortality rates from those with marrow production failure. Five to are appreciable and it is doubtful whether removal ten percent of patients with myeloid metaplasia will of the spleen affects prognosis [14]. Diagnosis is have secondary portal hypertension [15], but an based on the presence of splenomegaly and im- adjunctive splenorenal shunt is seldom indicated; mature myeloid and erythroid cells in peripheral Schwartz [16] has reported that in 6 patients with blood, anisocytosis and poikilocytosis, and variable AMM and increased portal pressures, splenectomy degrees of fibrosis in bone marrow. For determina- resulted in a decrease in portal pressures to normal tion of prognosis and decisions regarding operative or near normal in all. Preoperative assessment of benefit, patients with "primary" AMM should be coagulation profile and platelet function is impor- distinguished from those in whom AMM evolves tant because much of the immediate morbidity is following treatment for polycythemia rubra vera or related to thrombotic and hemorrhagic complica- essential . tions. Qualitative platelet abnormalities are fre- Our operative experience with this condition in- quently demonstrable in patients with myelopro- cludes 34 patients between the ages of 40 and 76 liferative disorders, and, if present, platelet trans- years, 27 of whom had primary AMM. Splenectomy fusion during operation may be helpful. In addition, was performed for hypersplenism in 20 individuals, Silverstein and Remine [15] have found laboratory for symptomatic splenomegaly in 11, in association evidence of subclinical disseminated intravascular with other operations in 2, and for traumatic rupture coagulation (decreased platelets and factors V and in 1 patient. The postoperative hematologic re- VIII and increased fibrin degradation products) in sponse could not be evaluated in 6 patients with 12% of these patients and advise against surgery in hypersptenism because of postoperative death, this circumstance. persistent chronic gastrointestinal bleeding, or We now restrict splenectomy in this condition to myelosuppressive chemotherapy. In 10 patients patients with primary myeloid metaplasia and with chronic anemia and decreased erythrocyte hypersplenic anemia or thrombocytopenia or symp- survival, 8 had a postoperative increase in the tomatic splenomegaly whose preoperative assess- hematocrit reading of 5 vol % or more which was ment indicates an acceptable operative risk. maintained for periods of 6 months to more than 8 years. A sustained increase in platelet count (for Felty's Syndrome more than 1 year) was observed in 3 of 4 thrombo- cytopenic patients. This entity is characterized by splenic neutropenia Symptomatic splenomegaly is manifested by (total count less than 3,500/ram 3 chronic left upper quadrant abdominal and back and neutrophil count less than 2,000/ram 3) develop- pain and early satiety (from gastric displacement). ing in a patient with rheumatoid arthritis and usually Splenic weight in this group ranged from 980 to accompanied by splenomegaly and anemia and oc- 4,200 g. All survivors obtained significant sympto- casionally by cutaneous ulcers of the leg. There has matic benefit. been considerable debate in recent years concern- Four postoperative deaths resulted from leuke- ing the value of splenectomy in preventing episodes mic conversion, gastrointestinal bleeding, acute he- of infection. Since only approximately 1 in 300 patic necrosis, or , and morbidity from bleed- patients with rheumatoid arthritis develop this syn- ing or sepsis was encountered in 12 patients. Mean drome [17], no prospective trials have been con- duration of survival following splenectomy was 22 ducted. Many physicians have tried to manage months for men and 46 months for women with problems of sepsis with antibiotic therapy alone or primary AMM but only 8 months for secondary in combination with administration of lithium in an AMM. The majority of late deaths were due to attempt to stimulate granulopoietic precursors in leukemic conversion or to gastrointestinal bleeding. the bone marrow. In our recent nonrandomized Since morbidity and mortality are considerable assessment of 20 evaluable patients receiving and survival is limited after splenectomy, careful nonoperative management and 20 splenectomized 440 World J. Surg. Vol. 9, No. 3, June 1985 patients [18], we have attempted to refine possible the spectrum is Banti's syndrome in a patient with criteria for operation. frank cirrhosis, portal hypertension, and spleno- Among the 20 subjects not undergoing surgery, megaly with or without clinically significant 16 presented with an infection and/or a leg ulcer. hypersplenism. At the other end is the individual There have been 2 deaths from sepsis and 1 episode with idiopathic splenomegaly and/or hypersplenism of severe . Leg ulcers persisted in 5 of 6 in whom, after splenectomy, the pathologist de- patients, healing in 1 after a 7-year period of obser- scribes "congestion" of the spleen but no obvious vation. Of 5 patients given a trial of lithium in doses liver disease or extrahepatic portal or splenic ve- of 900 to 1,800 mg/day, only 1 achieved a satisfac- nous obstruction is demonstrable. Between these tory granulocytic response. extremes are many patients with parasitic, "infiltra- Of the 20 splenectomized patients, 4 were lost to tive" (sarcoidosis, amyloidosis), and myeloprolif- follow-up after hospital discharge, 2 having erative disorders in whom variable degrees of achieved a satisfactory neutrophil response to intrahepatic or extrahepatic portal venous obstruc- greater than 2,000/ram 3. There was no operative tion may play a pathogenic role in the development mortality. In the remaining 16 patients who have of splenomegaly and hypersplenism. been observed for 6 months to 20 years (mean: 4 Splenectomy is the treatment of choice in pa- years), 13 have had a prompt and sustained tients with splenic vein thrombosis, "left-sided por- leukocyte (above 3,500) and neutrophil response tal hypertension," and bleeding esophageal varices. (above 2,000). None have had further infections. Of The indication for operation is almost always re- the 3 patients with follow-up who were splenectomy lated to the presence of esophageal varices rather "failures" (unsatisfactory cytologic response), 1 than clinically significant hypersplenism. Although died of sepsis 5 years later and a second has had an chronic pancreatitis is the most frequent underlying episode of pharyngitis responsive to antibiotics. Of cause of isolated splenic vein thrombosis, one 5 patients with leg ulcers, 3 have healed; 1 patient should be aware that previously undiagnosed pan- was lost to follow-up. No clinical or laboratory creatic carcinoma or lymphoma can also produce criteria have been found that can predict response this condition. to splenectomy. In patients who had a satisfactory Most surgeons would now agree that splenec- neutrophil response to splenectomy or who had a tomy alone is inappropriate in patients with hepatic remission during nonoperative management, ane- cirrhosis, portal hypertension, bleeding esophageal mia also improved. The infections that appeared varices, and signs of hypersplenism. Warren and after splenectomy occurred in patients with an associates [21] have recently reported that platelet inadequate neutrophil response. Other authors have counts less than 50,000/mm 3 were found in 8-10% of also reported correction of neutropenia and relief their patients presenting with variceal bleeding; from recurrent infection in 60-80% of patients un- thrombocytopenia was observed more frequently in dergoing splenectomy [17, 19, 20]. patients with nonalcoholic as compared to those Our data on the natural history of Felty's syn- with alcoholic liver disease. Boivin [22] observed a drome in patients not undergoing surgery lead us to lower frequency of thrombocytopenia in patients believe that, with rare exceptions, hematologic ab- with cirrhosis (7 of 185 cases with platelets less than normalities alone are not a sufficient indication for 50,000/mm3). In his experience the level of operation unless the patient has an associated se- thrombocytopenia present in these individuals did vere hemolytic anemia or thrombocytopenia. Al- not affect mortality or morbidity after operations for though neutropenia of fewer than 500 cells/mm 3 has portal decompression. While most of the currently been proposed as an indication for operation, we utilized shunt procedures for management of portal found no correlation between lowest neutrophil hypertension leave the spleen intact, the majority of count and the development, severity, or frequency patients with associated thrombocytopenia will re- of infection. On the other hand, after the appear- spond with an increase in platelet count postopera- ance of serious or recurrent infection, splenectomy tively, but in some, the rise may not be sustained appears to be of value regardless of the severity of [23] and the increase in platelets (or leukocytes) will the neutropenia. Although our sample is small, the not reach normal values [24]. Splenectomy should likelihood of healing of a leg ulcer seems improved be reserved for the very rare patient with severe after splenectomy. residual thrombocytopenia after shunt. In our experience with idiopathic splenomegaly and hypersplenism [25], we have encountered 3 "Congestive Splenomegaly" patients in whom the diagnosis of relatively mild cirrhosis was made only at the time of splenecto- This nonspecific term includes a spectrum of condi- my and liver biopsy. Thrombocytopenia (platelet tions of varying etiology and severity. At one end of counts: 38 to 68,000) improved after splenectomy, W.W. Coon: Splenectomy for Splenomegaly 441 and platelet counts have remained normal for peri- In the past, splenectomy has been performed in a ods of follow-up of 1 to 8 years. In 4 other patients few patients with Niemann-Pick disease, but with idiopathic neutropenia (2 with associated follow-up of these subjects with a very limited thrombocytopenia), the only histologic finding was prognosis has demonstrated no alteration in sur- "splenic congestion" without evidence of liver dis- vival, although anemia improved moderately in ease. All achieved a satisfactory cytologic response several [32]. to splenectomy. Portal hypertension secondary to either extrahepatic or intrahepatic portal obstruction in Tropical Splenomegaly Syndrome patients with hypersplenism associated with myeloproliferative disorders [26] may add to mor- This condition, recognized in Africa and New bidity and mortality. As mentioned in prior discus- Guinea for many years, is seen in patients with high sion, many of these patients achieve little or only titers of antimalarial antibodies and is associated very short-term benefit from an operation and, as with in the bone marrow, anemia Schwartz [16] has shown, adjunctive shunt proce- and occasionally mild leukopenia and thrombocy- dures are seldom indicated. However, awareness of topenia, reactive macroglobulinemia, increased the possible presence of portal hypertension, and IgM levels, and reduced serum complement (C3) also platelet and coagulative abnormalities [15] in [33]. A causal relationship to has not been patients with these conditions is important because established. Splenectomy has been shown to im- some of these individuals will require urgent opera- prove the cytopenias [34, 35]. Many patients will tions for other problems. As Wasserman and respond to prolonged antimalarial therapy [36]; Gilbert [27] have reported for patients with polycy- splenectomy is now reserved for those patients in themia vera (and it is equally true in other whom anemia fails to respond to antimalarial drugs myeloproliferative conditions), any abdominal op- and those with symptomatic splenomegaly (chronic eration is associated with an appreciable mortality ) [34]. and morbidity from hemorrhage or thrombosis.

Gaucher's Disease and Other Reticuloendothelioses Idiopathic Splenomegaly and Hypersplenism The benefit of splenectomy in these conditions is A mrlange of "idiopathic" conditions have been chiefly in selected patients with Gaucher's disease. described over the past 40 years. Wiseman and This hereditofamilial deficiency of acid-beta-gluco- Doan [37] classified a group of patients with sidase results in progressive accumulation of gluco- splenomegaly and neutropenia without specific sylceramide in the spleen, bringing about massive splenic histology as "primary splenic neutropenia." splenomegaly and frequently secondary hypersplen- More recently, Dacie and associates [38, 39] have ism. Since this is a disease of the Jewish population, followed these patients with "non-tropical idio- primarily Ashkenazi Jews, the greatest experience pathic splenomegaly" and hypersplenism in whom has been accumulated in Israel and in Jewish hos- no diagnosis was made after splenectomy; 4 have pitals in other countries. Matoth and Fried [28] have subsequently developed lymphoma. reported 16 splenectomies without mortality in pa- Our patients with idiopathic splenomegaly [25] tients of ages 5 to 47 years; splenectomy was have differed from those of Dacie and of other performed principally for thrombocytopenia with a investigators [40, 41] but are similar to those of prompt increase in platelets which is usually main- Goonewardene and co-workers [42]. Although 18 of tained for years. The effect on anemia is less 28 patients with splenomegaly of undetermined ori- dramatic. Similar results have been obtained in gin had some degree of hypersplenism (hematocrit several other small series of patients [29, 30]. Our < 30 vol %, leukocytes < 3,500/ram 3 or neutrophils limited experience with 3 patients is similar. All had < 2,000/mm 3 or platelets < 100,000/mm3), the a prompt and sustained remission of thrombocy- cytopenias were usually mild; splenectomy was topenia. An associated anemia in a 4-year-old fe- performed for detection of possible lymphoma. No male also improved. Although there has been some lymphoma was found at the time of operation and concern that splenectomy may worsen the clinical only 1 patient has subsequently had this diagnosis course and accelerate the progression of orthopedic made. Only 7 had a specific diagnosis made as a problems, others feel that this impression is errone- result of the operation (cirrhosis, splenic cyst, ous because the more severely affected young pa- sarcoidosis). tients are the ones who have early splenectomy and This experience with a group of patients who are more likely to develop later orthopedic prob- were relatively young (mean age: 28 years) has led lems [31]. us to be more conservative with respect to recom- 442 World J. Surg. Vol. 9, No. 3, June 1985 mending splenectomy, provided that no other mani- 2. Jansen, J., Hermans, J.: Splenectomy in hairy cell festations compatible with the diagnosis of leukemia. A retrospective multicenter analysis. Can- lymphoma are present. Many who are young, cer 46:2066,1981 3. Golomb, H.M.: Progress report on chlorambucil asymptomatic, and with a history of recent infec- therapy in postsplenectomy patients with progressive tion can be closely watched and a decision regard- hairy cell leukemia. Blood 57:464, 1981 ing splenectomy deferred. 4. Cheever, M.A., Fefer, A., Greenberg, P.E., Ap- pelbaum F., Armitage, J.O., Buckner C.D., Sale, R6um6 G.E., Storb, R., Witherspoon, R.P., Thomas, E.D.: Treatment of hairy-cell leukemia with chemoradi- La spl6nom6galie avec hyperspl6nisme secondaire otherapy and identical-twin bone marrow transplan- rel6ve de multiples causes: infection aigue ou tation. N. Engl. J. Med. 30:479, 1982 chronique, 6tats autoimmunologiques, hyperten- 5. Quesada, J.R., Reuben, J., Manning, J.T., Hersh, sion portale, thrombose de la veine spl6nique, E.M., Gutterman, J.U.: Alpha interferon for induc- 16sions tumorales spl6niques. L'exp6rience de tion of remission in hairy cell leukemia. N. Engl. J. Med. 310:15,1984 l'auteur qui rejoint celle de nombreux coll6gues lui 6. Christensen, B.E., Hansen, L.K., Kristensen, J.K., permet d'affirmer que les indications de la spl6- Videboek, A.: Splenectomy in haematology: Indica- nectomie doivent ~tre bien d6finies et sont stricte- tions, results and complications in 41 cases. Scand. J. ment limit6es. A l'exception de la spl6nom6galie Haematol. 7:247, 1970 idiopathique, l'existence et l'intensit6 de 7. Holt, J.M., Witts, L.J.: Splenectomy in leukaemia l'hyperspl6nisme, l'importance des symptomes and the reticuloses. Q. J. Med. 35:369, 1966 8. Karanas, A., Silver, R.T.: Characteristics of the provoqu6s par la spl6nom6galie doivent 6tre terminal phase of chronic granulocytic leukemia. aprr6ci6es avec pr6cision. Dans chaque cas le Blood 32:445, 1968 potentiel de la r6mission de l'affection et la dur6e de 9. Ihde, D.C., Canellos, G.P., Schwartz, J.H., DeVita, la r6mission doivent 6tre pris en consid6ration en V.T.: Splenectomy in the chronic phase of chronic fonction de l'6ventuelle morbidit6 et de l'6ventuelle granulocytic leukemia. Ann. Intern. Med. 84:17, 1976 mortalit6 de la spl6nectomie (par comparaison avec 10. Baccarani, M., CorbeUi, G., Tuca, S., and the Italian la spl6nectomie pour hyperspl6nisme primaire). Cooperative Study Group on Chronic Myeloid Leu- kemia: Early splenectomy and polychemotherapy versus polychemotherapy alone in chronic myeloid Resumen leukemia. Leuk. Res. 5:149, 1981 Eplenomegalia e hiperesplenismo secundario 11. Medical Research Council's Working Party for Ther- pueden estar asociados con infecciones agudas y apeutic Trials in Leukemia: Randomized trial of splenectomy in Phi-positive chronic granulocytic cr6nicas, estados autoinmunes (sfndrome de Felty, leukaemia, including an analysis of prognostic fea- lupus eritematoso sist6mico), "esplenomegalia con- tures. Br. J. Haematol. 54:415, 1983 gestiva" por hipertensi6n portal o trombosis de la 12. James, D.G., Neville, E., Siltzbach, L.E.: A world- vena espl6nica y con una variedad de entidades de wide review of sarcoidosis. Ann. N.Y. Acad. Sci. tipo infiltrativo y neoplfisico que afectan al bazo 278:321, 1976 (sarcoidosis, enfermedad de Gaucher, varios 13. Webb, A.K., Mitchell, D.N., Bradstreet, C.M.P., des6rdenes mieloproliferativos y linfomas). Salsbury, A.J.: Splenomegaly and splenectomy in sarcoidosis. J. Clin. Pathol. 32:1050, 1979 Nuestra experiencia, y aquella de otros autores, 14. Coon, W.W., Liepman, M.K.: Splenectomy for con tales condiciones demuestra que la decisi6n de agnogenic myeloid metaplasia. Surg. Gynecol. realizar esplenectomfa debe estar fundamentada en Obstet. 154:561, 1982 indicaciones bien definidas y estrictamente limita- 15. Silverstein, M.N., Remine, W.H.: Splenectomy in das. Excepto en casos de esplenomegalia idio- myeloid metaplasia. Blood 53:515, 1979 p~tica, la presencia y severidad del hiperesplenismo 16. Schwartz, S.I.: Myeloproliferative disorders. Ann. secundario o de esplenomegalia severamente sinto- Surg. 182:464, 1975 m~itica debe ser bien documentada. En cada caso 17. Green, R.A., Fromke, V.L.: Splenectomy in Felty's syndrome. Ann. Intern. Med. 64:1265, 1966 debe determinarse el potencial de paliaci6n y la 18. Coon, W.W.: Felty's syndrome: When is splenec- duraci6n de la respuesta que se espera obtener tomy indicated? Am. J. Surg. 149:272, 1985 frente a la incrementada morbilidad y mortalidad de 19. Moore, R.A., Brunner, C.M., Sanduskey, W.R., la esplenectomia (en comparaci6n con la operaci6n Byrd, S.L.: Felty's syndrome. Long-term follow-up que se realiza por hiperesplenismo "primario"). after splenectomy. Ann. Intern. Med. 75:381, 1971 20. Laszlo, J., Jones, R., Silverman, H.R., Banks, P.R.: Splenectomy for Felty's syndrome. Arch. Intern. References Med. 138:597, 1978 21. Warren, W.D., Millikan, W.J., Jr., Henderson, J.M., 1. Coon, W.W.: The limited role of splenectomy in Rasheed, M.E., Solam, A.A.: Selective variceal de- patients with leukemia. Surg. Gynecol. Obstet. (in compression after sptenectomy or splenic vein press) thrombosis. Ann. Surg. 199:694, 1984 W.W. Coon: Splenectomy for Splenomegaly 443

22. Boivin, P.: Les thrombop6nies des cirrhoses, sont- 32. Crocker, A.C., Barber, S.: Niemann-Pick disease: A elles une indication de spl6nectomie? Acta review of eighteen cases. Medicine 437:1, 1958 Gastroenterol. Belg. 44:203, 1981 33. Ziegler, J.L., Stuiver, P.C.: Tropical splenomegaly in 23. Morris, P.W., Patton, T.B., Bolint, J.A., Hirshowitz, a Rwandon kindred in Uganda. Br. Med. J. 3:79, 1972 B.I.: Portal hypertension, congestive splenomegaly 34. Hamilton, P.J.S., Richmond, J., Donaldson, and portacaval shunt. Gastroenterology 42:555, 1962 G.W.K., Williams, R., Hurt, M.S.R., Lugumba, V.: 24. Hutson, D.G., Zeppa, R., Levi, J.U., Schiff, E.R., Splenectomy in "Big Spleen Disease." Br. Med. J. Livingstone, A.S., Fink, P.L.: The effect of the distal 3:823, 1967 splenorenal shunt on hypersplenism. Ann. Surg. 35. Pryor, D.S.: Splenectomy in tropical splenomegaly. 185:605, 1977 Br. Med. J. 3:825, 1967 25. Knudson, P., Coon, W., Schnitzer, R., Liepman, M.: 36. Watson-Williams, E.J., Allan, N.C.: Idiopathic tropi- Splenomegaly without an apparent cause. Surg. cal splenomegaly syndrome in Ibadan. Br. Med. J. Gynecol. Obstet. 155:705, 1982 4:793, 1968 26. Shaldon, S., Sherlock, S.: Portal hypertension in the 37. Wiseman, B.K., Doan, C.A.: Primary splenic neutro- myeloproliferative syndrome and the reticuloses. penia. Ann. Intern. Med. 16:1097, 1942 Am. J. Med. 32:758, 1962 38. Dacie, J.V., Brain, M.C., Harrison, C.V., Lewis, 27. Wasserman, L.R., Gilbert, H.S.: Surgery in polycy- S.M., Worlledge, S.M.: Nontropical idiopathic themia vera. N. Engl. J. Med. 269:1226, 1963 splenomegaly ("primary hypersplenism"); a review 28. Matoth, Y., Fried, K.: Chronic Gaucher's disease: of ten cases and their relationship to malignant Clinical observations on 34 patients. Isr. J. Med. Sci. lymphoma. Br. J. Haematol. 17:317, 1969 1:521, 1965 39. Dacie, J.V., Galton, D.A.G., Cordon-Smith, E.C., 29. Medoff, A.S., Bayrd, E.D.: Gaucher's disease in 29 Harrision, C.V.: Nontropical "idiopathic splenomeg- cases: Hematologic complications and effect of sple- aly"; a follow-up study of ten patients described in nectomy. Ann. Intern. Med. 40:481, 1954 30. Aufses, A.H., Jr., Salky, B.M.: The surgical manage- 1969. Br. J. Haematol. 38:185, 1978 ment of Gaucher's disease. In Gaucher's Disease: A 40. Hermann, R.E., DeHaven, K.E., Hawk, W.A.: Sple- Century of Delineation and Research, R.J. Desnick, nectomy for the diagnosis of splenomegaly. Ann. S. Gott, G.A. Grabowski, editors. New York, Alan Surg. 168:896, 1968 R. Liss, Inc., 1982, pp. 107-129 41. Long, J.C., Aisenberg, A.A.: Malignant lymphoma 31. Beighton, P., Goldblatt, J., Sachs, S.: Bone involve- diagnosed at splenectomy and idiopathic splenomeg- ment in Gaucher's disease. In Gaucher's Disease: A aly. Cancer 33:1054, 1974 Century of Delineation and Research, R.J. Desnick, 42. Goonewardene, A., Bourke, J.B., Ferguson, R., S. Gott, G.A. Grabowski, editors. New York, Alan Toghill, P.J.: Splenectomy for undiagnosed R. Liss, Inc., 1982, pp. 603-616 splenomegaly. Br. J. Surg. 66:62, 1979