NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Holzheimer RG, Mannick JA, editors. Surgical Treatment: Evidence-Based and Problem-Oriented. Munich: Zuckschwerdt; 2001.

Cover of Surgical Treatment

Surgical Treatment: Evidence-Based and Problem-Oriented.

Show details

Laparoscopic splenectomy for benign and malignant hematologic diseases

, M.D. and , M.D.

Author Information

Since the first laparoscopic splenectomy (LS) performed by Delaître in 1991, numerous case series have shown the feasibility of LS for a large variety of benign and malignant hematologic diseases. Although there has not yet been a single prospective randomized study to support any advantage of laparoscopic over open splenectomy (OS), the laparoscopic approach has rapidly become very popular and has even been proposed as “gold standard” for splenectomy in hematologic diseases with normal size or moderately enlarged spleens. Nevertheless all available data and all recommendations for LS at this time derive exclusively from grade C (level III to V) studies.

Indications for LS

Splenectomies may be performed for diagnosis, staging or treatment of a variety of benign or malignant hematologic diseases. Indications for LS are the same than for open splenectomy. Proposed specific contraindications for LS are portal hypertension, the presence of an splenic hilum thickened by massive lymphadenopathies and massive splenomegaly (spleens of > 3.5 kg or largest diameter > 30 cm) leaving insufficient space for laparoscopic handling of the spleen.

A Medline review including all original series with a minimum of five cases published in the english-spoken literature until January 1999 found 40 series of LS in 908 adult patients and 12 pediatric series with a total of 140 cases. Immune thrombocytopenic purpura (ITP) has been by far the most frequent indication (61% of all cases), followed by hereditary spherocytosis (7%), autoimmune hemolytic anemia (6%), HIV associated thrombocytopenia (3%), thombocytopenic thrombotic purpura (3%). Seventy-nine of these LS (9%) have been performed for various hematologic malignancies with or without splenomegaly. Thirty of them were laparoscopic staging procedures for Hodgkin's disease (HD) whereas chronic lymphocytic leukemia, non-Hodkin lymphomas and a variety of other lympho- or myeloproliferative diseases have only sporadically been approached laparoscopically.

Operative techniques

Several technical strategies for LS have been described such as the hanging spleen technique by Delaître, the lateral technique by Parks and Gagner or the posterior approach to the splenic pedicle by Gossot. The differences among these techniques consist mainly in patient positioning (ranging from supine to full right lateral decubitus), approach to the splenic hilum (anterior, lateral or posterior access to the splenic pedicle) and in the sequence of performance of the different operative steps. Trocar positioning depends largely on the size of the specimen to be removed. After an initial controversy between proponents of the anterior and the lateral approach, the lateral approach, either using four or five trocars, nowadays is favored by most for LS of normal size spleens. Criticism against the lateral approach has been raised, however, for patients with ITP (Gigot 1996) because the intraoperative search for accessory spleens might be more difficult with the lateral than with the anterior approach and could therefore lead to recurrence due to inadequate accessory spleen detection. The anterior approach is preferred when associate procedures are required such as cholecystectomy (e.g. in hereditary spherocytosis), lymph node, liver or other biopsies and it may also be preferable in cases with massive splenomegaly because of easier access to the uppermost short gastric vessels and faster conversion if required.

Various technical devices have been proposed to facilitate LS. The benefit of using of preoperative splenic embolization (Poulin 1993), ultrasonic scalpels (Hashizume 1993) or endoscopic staplers (Miles 1996), however, remains controversial as LS performed without these devices has the same outcome.

Splenic embolization presently is recommended only for specimens of more than 20 cm diameter but entails additional costs and a 5% morbidity rate (severe pain, partial liver necrosis or lung emboli). For spleens with a modular vascular pattern embolization might advantageously be replaced by ligature of the splenic artery as the initial step of LS. Stapling devices may malfunction and have been reported to cause hemorrhage and pancreatic fistulas in case of en-bloc stapling of the splenic hilum.

Results of laparoscopic splenectomy

As for open surgery, the outcome of LS in terms of operative morbidity and therapeutic success rates may depend largely on the indication for splenectomy and the patient population which is concerned. Comparisons between the two approaches therefore should only be made within the same indication.

Pediatric laparoscopic splenectomy

Children requiring splenectomy for hematologic conditions differ from adults by a greater susceptibility to post-splenectomy sepsis, a higher prevalence of accessory spleens and a higher spontaneous remission rate of “idiopathic” thrombocytopenia which may occur secondary to viral infections. Data from grade C studies comparing the outcome of pediatric LS with historic (Beanes 1995, Janu 1996) or concurrent groups of OS (Rescorla 1998, Esposito 1997, Farah 1997, Waldhausen 1997) suggest that LS may achieve similar outcome to OS in terms of morbidity and successful control of the hematologic disorder. Operative times, however, were constantly longer for laparoscopically treated patients. Several series found LS to require less postoperative analgesics and to allow shorter hospital stays and earlier return to full activities than OS, while one non-randomized comparative study found no difference between open and laparoscopic approach. Series of OS further suggest that by using a left upper abdominal muscle splitting incision (Geiger 1998), OS may be performed in children with minimal morbidity or transfusions and with duration's before tolerance of regular diet and hospital stays comparable or shorter than for most laparoscopic series. Data concerning hospital charges are controversial as some studies indicated lesser hospital costs for LS (Rescorla 1998), while others found either no difference (Farah 1997) or higher hospital charges after LS (Janu 1996, Waldhausen 1997) (grade C).

Results of LS in adult patients

According to a review of all studies addressing these issues, LS entailed a 0.94% mortality (8 of 851 patients), a 12% complication rate (104 of 888 cases) and conversion to laparotomy occurred in 8.5% of all LS (76 of 891 cases). Altogether blood transfusions were required in 13% (100 of 766 cases) and accessory spleens (AS) were detected during 14.3% of all LS (104 of 728 cases). Most series of LS report overall results for various types of hematologic diseases. As for OS, the results of LS however may be expected to vary from one indication to another.

Immune thrombocytopenic purpura

According to a literature review (Akwari 1987) open splenectomy may achieve remission of ITP in 72% of 872 collected cases and remission rates in more recent series ranged from 55 and 90%, depending on the definition of remission and the duration of follow-up (Chirletti 1992, Wilhelm 1988, Schwartz 1985). Splenectomy is the only definitive therapeutic procedure available for chronic ITP as corticosteroids achieve remission in some 30% of patients only. Nevertheless thrombocytopenia will relapse during the first years after splenectomy in up to 20% of initial responders (Rudowski 1985). Although many questions remain to be answered, such as the volume of AS needed to cause a recurrence, it seems that accessory spleens are a main cause of recurrent ITP. Surgical data as well as autopsy studies estimate the frequency of accessory spleens in adults from 11 to 18% and much higher in children (50% during the first decade of life with progressive diminution thereafter). AS may be identified by imaging techniques or radionuclide scans with heat damaged red blood cells in a proportion of patients with mostly late recurrences. Accessory splenectomy may then produce return to normal platelet levels in 73–80% of these patients.

Not every patient with ITP, however, responds to splenectomy and neither platelet antibody titers nor the pattern of presplenectomy platelet sequestration studies currently allow to reliably predict the response to splenectomy in an individual patient. Some further believe that splenectomy might be beneficial even in those who do not increase their platelet count after splenectomy as these patients only rarely have recurrent episodes of bleeding or petechias. Conventional surgery for ITP achieves high cure rates with very low needs for transfusion and complication rates between 7 and 20%. Operative mortality of OS for ITP was 6% in 872 collected cases with specialized centers reporting mortality rates as low as 2%.

Laparoscopic splenectomy for ITP might be superior to conventional surgery if it can offer similar or better control of thrombocytopenia with similar or less morbidity and improved comfort for the patient. No grade A or B data are available in 1999 and comparisons of results from many grade C series are unreliable as most series have used criteria for inclusion and response assessment which are far less restrictive than what has been recommended.

Several retrospective studies including small numbers of patients (10 to 28 per group) have compared LS with OS for ITP using either concurrent (Delaître 1997) or historic control groups (Hashizume 1996, Watson 1997, Lozano-Salazar 1998, Marassi 1999). Short-term response rates of thrombocytopenia were similar to what is expected for OS. All these studies found operative times to be longer than for open surgery. Morbidity and accessory splenectomy rates were not significantly different, but several of these studies showed a reduced blood loss for LS (Hashizume 1996), earlier return of bowel sounds (Delaître 1997, Marassi 1999), less analgesic consumption (Brunt 1996), faster return to normal diet (Hashizume 1996, Brunt 1996) and to normal activity (Diaz 1997). Transfusion requirements among both approaches were similar in five of these studies while six studies found hospital stays to be significantly shorter for LS.

Direct operative charges were higher for LS than for OS in all studies addressing this issue, while total hospital charges were found either to be lower for LS (Hashizume 1996, Watson 1997), similar (Smith 1996) or higher for laparoscopic than for conventional splenectomy (Diaz 1997). Unusually long hospital stays (means of 10–20 days) after conventional surgery however might have biased the reported superiority of LS in terms of hospital stays and costs in some of these studies. Prospective randomized studies are clearly required to clarify these issues.

Splenectomy for malignant hematologic diseases

Splenectomy for malignant diseases mostly is performed for diagnosis or staging and only rarely aims at cure of the disease. In a small proportion of patients, palliation of secondary hypersplenism or symptomatic splenomegaly may be required, generally late during the evolution of these malignancies. Patients who undergo splenectomy for malignant diseases are older and in poorer general condition than those with benign disorders. Many of these patients have undergone multiple courses of chemotherapy prior to splenectomy, thus putting them at high risk for impaired wound healing and other complications. Open splenectomy for malignant diseases thus classically carries a high morbidity with reported mortality rates ranging from 5 to 13% and complication rates ranging from 14 to 61% for series including various hematological malignancies (Horowitz 1997).

On the other hand it has been shown that splenectomy in this setting may achieve lasting palliation of secondary thrombocytopenia or anemia thus allowing these patients to undergo more intensive chemotherapeutic regimens which would be contraindicated in the presence of hypersplenism.

Hodgkin's disease

Staging laparotomy for HD classically carries a low morbidity because these patients are usually young, take little or no immuno-suppressive drugs, have normal platelet counts and are in better general shape than those with other hematologic indications for splenectomy. Reported mortality rates vary from 0 to 0.7% with only exceptional need for blood transfusions, short hospital stays and complication rates ranging from 9 to 25%. Over the last decade fewer staging laparotomies are being performed mainly because of improvements of non-invasive diagnostic means and changes in the indications for chemotherapy with a trend towards chemotherapy in earlier stages of HD. Although the feasibility of laparoscopic staging procedures has been established, there is no evidence for superiority over open procedures. Due to recent progress in non-invasive staging techniques (MRI, *67Gallium scintigraphy and whole-body positron emission tomography), the need for surgical staging will further vanish.

Other lympho- or myeloproliferative disorders

Only a few series of LS have included various lymphoproliferative diseases other than Hodgkin's disease (non-Hodgkin lymphomas, chronic lymphocytic leukemia (CLL), hairy cell leukemia …) or myeloproliferative diseases such as myelofibrosis, acute or chronic myelogenous leukemia, polycythemia vera and other more rare diseases.

In the case of hairy cell leukemia (HCL) with cytopenia survival has been shown to be significantly longer after open splenectomy than with medical treatment alone (Jansen 1981), but only sporadic reports of LS for HCL have been reported so far. For all other conditions splenectomy is rarely required, usually as a palliative procedure for symptomatic hypersplenism.

Many patients with CLL live with little or no treatment for five to ten years before they develop the symptoms of hypersplenism for which they eventually may require splenectomy. For CLL excellent response rates to open splenectomy have been reported with mean survival after splenectomy of more than 4 years. Reported series of OS for LPD are small and mortality rates range from 0 to 9% and complication rates ranging from 17 to 65%.

Open splenectomy in myeloproliferative diseases (MPD) classically carries the highest morbidity as most of these patients requiring a splenectomy have progressed to an advanced stage of their disease and have little or no medullary reserve. Reported mortality rates of OS for MPD vary from 2 to 28% with complication rates from 39 to 58%. Only sporadic but mostly successful cases of LS have been reported in these indications.

When performed for the purpose of diagnosis, an intact specimen is required for pathology and the warranted utility laparotomy for organ retrieval might alter the benefits of a LS. Nevertheless as splenectomy generally is not performed for diagnosis or cure but merely for palliation of severe hypersplenism, considerations on potential risks of port site metastasis or the integrity of the specimen to be removed might be secondary to the aim of controlling hypersplenism to allow the patient to undergo further systemic therapies. Such indications, however, are very rare and only a few series of LS have reported more than 10 cases (Rhodes 1995, Decker 1998, Targarona 1998). Although their complication, conversion and transfusion rates were higher then when LS is performed for ITP, the poor outcome of open surgery for these indications may render the laparoscopic approach more beneficial for malignant than for benign hematologic diseases requiring a splenectomy.

Massive splenomegaly

Regardless the type of disease, massive splenomegaly may become symptomatic for mechanical-physical reasons or due to hypersplenism or splenic infarction. Series of open splenectomy for massive splenomegaly (more than 1500 or 2000 gr) have reported mortality rates of 8 to 15% and complication rates of 25 to 63% (Horowitz 1996, Goldstone 1987, Coon 1989, McRae 1992), with postoperative hemorrhage, dehiscence of midline incisions and left pleural effusion being the most frequent complications. Average blood loss of OS for massive splenomegaly usually exceeds 1 liter and although initial ligation of the splenic artery has been shown to significantly reduce blood loss (Goldstone 1987), intraoperative transfusions remain necessary in more than half of the patients.

Numerous case reports have shown the feasibility of LS in conditions with massive splenomegaly. Recently some small series have shown the feasibility of LS with splenomegaly. Although some series included specimens of only 500 gr (ref. 10) others have included only spleens of more than 1000 (ref. 9), respectively 1500 gr (ref. 5). No mortality was reported and complications rates ranged from 12 to 30%, transfusion rates from 12 to 30% and conversion rates from 0 to 30%.

Operative times for LS were significantly longer for splenomegaly than for spleens of normal size (less than 500 gr) but blood loss, conversion rates, complication rates and postoperative hospital stays were not significantly different with or without splenomegaly (grade C).

Terrosu et al. further compared LS with OS with splenomegaly and found similar complication rates in both groups and longer operative times for LS but less intraoperative blood transfusions, shorter delays to liquid diet and shorter hospital stays in the laparoscopic group (grade C).


No grade A or B evidence is currently available to support any superiority of LS over OS in terms of morbidity, comfort to the patient or costs, while the efficiency of LS for issues like detection of accessory spleens has been questioned and long-term results on thrombocytopenia still remain unknown. As LS now is increasingly performed in many centers, prospective randomized studies are urgently required to establish the utility of LS for benign hematologic diseases.

Malignant hematologic conditions only rarely require a splenectomy and randomized trials therefore will hardly ever be achievable. Nevertheless available grade C data indicates that a laparoscopic approach might improve the outcome of splenectomy in these frequently high risk patients.


Poulin E C, Mamazza J, Schlachta C M. Splenic artery embolization before laparoscopic splenectomy. Surg Endosc. (1998);12:870–875. [PubMed: 9602009]
Akwari O E, Itani K M F, Coleman R E, Rosse W F. Splenectomy for primary or recurrent immune thrombocytopenic purpura. Current criteria for patient selection and results. Ann Surg. (1987);206:529. [PMC free article: PMC1493222] [PubMed: 3662662]
Delaître B, Pitre J. Laparoscopic splenectomy versus open splenectomy: A comparative study. Hepato Gastroenterol. (1997);44:45–49. [PubMed: 9058117]
Kathkouda N, Hurwitz M B, Rivera R T, Chandra M, Waldrep D J, Gugenheim J, Mouiel J. Laparoscopic splenectomy. Outcome and efficacy in 103 consecutive patients. Ann Surg. (1998);228:568–578. [PMC free article: PMC1191537] [PubMed: 9790346]
Decker G, Millat B, Guillon F, Atger J, Linon M. Laparoscopic splenectomy for benign and malignant hematologic diseases: 35 consecutive cases. World J Surg. (1998);22:62–68. [PubMed: 9465763]
Rhodes M, Rudd M, O’Rourke N, Nathanson L, Fielding G. Laparoscopic splenectomy and lymph node biopsy for hematologic disorders. Ann Surg. (1995);222:43–46. [PMC free article: PMC1234753] [PubMed: 7618967]
Brunt L M, Langer J C, Quasebarth M A, Whitman E D. Comparative analysis of laparoscopic versus open splenectomy. Am J Surg. (1996);172:596–601. [PubMed: 8942570]
Glasgow R E, Yee L F, Mulvihill S J. Laparoscopic splenectomy. The emerging standard. Surg Endosc. (1997);11:108–112. [PubMed: 9069137]
Targarona E M, Espert J J, Balagué C, Piulachs J, Artigas V, Trias M. Splenomegaly should not be considered a contraindication for laparoscopic splenectomy. Ann Surg. (1998);228:35–39. [PMC free article: PMC1191425] [PubMed: 9671064]
Terrosu G, Donini U, Baccarani U, Vianello V, Anania G, Zala F, Pasgualucci A, Bresadola F. Laparoscopic versus open splenectomy in the management of splenomegaly: Our preliminary experience. Surgery. (1998);124:839–843. [PubMed: 9823396]
Copyright © 2001, W. Zuckschwerdt Verlag GmbH.
Bookshelf ID: NBK6981


  • PubReader
  • Print View
  • Cite this Page

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...