Residual Splenic Function After Laparoscopic Splenectomy: Title and subTitle BreakA Clinical Concern FREE

LAPAROSCOPIC splenectomy (LS) has become a useful alternative to open splenectomy, offering the advantages (less pain and fewer scars) of laparoscopic surgery.^{1 - 2} However, there is no wide experience with this technically demanding operation; as with other laparoscopic procedures, new clinical situations appear after the application of laparoscopic techniques. Clinical success after splenectomy is related to the complete removal of splenic tissue. However, there are some theoretical risks during laparoscopy of inducing splenic tissue spillage or of leaving unidentified accessory spleens (ASs). It is well known that splenic tissue can implant and grow into the peritoneal cavity.^{3 - 5} However, manipulation and retrieval of solid organs during laparoscopy can be difficult; also, the intra-abdominal bag can break during spleen morcellation, with a risk of splenic tissue spillage.⁶ Accessory spleens are present in more than 15% of the population; if they are left or not identified, they can induce relapsing disease.^{7 - 12} Some previous reports have suggested a more difficult identification of ASs during LS.^{13 - 14} This article documents residual splenic function in 3 patients of a series of 48 patients in whom LS was performed.

Between February 1992 and October 1996, 48 consecutive patients with a wide range of hematologic diseases requiring splenectomy underwent a laparoscopic approach; the clinical data and technical details were prospectively recorded. Laparoscopic splenectomy was performed via an anterior approach in 13 patients and via a lateral approach in the remaining 35 patients. Both techniques have been previously described elsewhere.^{15 - 16} In all cases, ASs were sought routinely after opening the omental pouch. The spleen was retrieved in a bag or, in selected cases, through an accessory incision to obtain an intact organ or because of the size of the spleen. Early and late responses to splenectomy were assessed with clinical and analytic examinations. The conditions of patients in whom there was a late failure to improve after LS were reevaluated with technetium Tc 99m sodium pertechnetate heat-damaged red blood cell scintigraphy (LFOV 609 γ camera, high-resolution collimator, Elscint, Haifa, Israel) and computed tomographic (CT) scanning.

Early failure is defined as the inability to normalize the platelet count during the first week after LS. Late failure is defined as the decline of the platelet count below 100×10⁹/L (partial failure) or below 50×10⁹/L (complete failure) after a transient increase (>200×10⁹/L).

Between February 1992 and October 1996, 48 LSs were performed. The demographic and operative variables can be summarized as follows (all data are given as the number of patients unless otherwise specified):

The platelet count of the 32 patients in whom an LS was performed for the treatment of autoimmune thrombocytopenia ranged between 5×10⁹/L and 102×10⁹/L.

Laparoscopic splenectomy was accomplished in 45 (94%) of the 48 patients. Conversion was required in 2 patients because of the difficulty in handling the spleen and because of diffuse oozing (in one patient with idiopathic thrombocytopenic purpura and in one patient with human immunodeficiency virus–related thrombocytopenia); conversion was required in a third patient with massive splenomegaly (>3500 g) because of the difficulties in controlling the splenic vein. Accessory spleens were found in 6 (12.5%) of the 48 patients. Five of them were easily located close to the inferior pole of the spleen or next to the great curvature; however, one, corresponding to one of the converted cases (massive splenomegaly for spherocytosis), was found in the mesocolon, only felt by palpation, and completely hidden from view.

The spleen was extracted through a bag in 38 patients, and a laparotomy was performed in 7 patients. Accidental perforation of the bag occurred in 4 cases, but all the spleen pieces were recovered.

The platelet count normalized within 48 hours of the operation in 29 of the 32 patients with a low platelet count.

After a mean follow-up of 18 months, the hematologic status of 39 (81%) of the 48 patients improved; however, 9 patients had a low platelet count (1 patient with human immunodeficiency virus–related thrombocytopenia and 8 patients with idiopathic thrombocytopenic purpura). There were 6 total failures and 3 partial failures. All 9 patients were assessed with scintigraphy and CT scanning. Scintigraphy showed residual splenic function in 3 patients with idiopathic thrombocytopenic purpura (Figure 1, Figure 2, Figure 3 and Table 1). Two corresponded to a hot-spotted image that the CT scan confirmed as an AS. In one of these patients, another AS had been retrieved during a former LS. The residual spleen was located in a pararenal area. Scintigraphy showed multiple images in the splenic fossa in a third patient; a CT scan confirmed these images as implants. This patient was one of the patients converted to open surgery because of the difficulty in handling the spleen during LS. A previous rupture of the bag did not occur during spleen retrieval in any of these patients. The platelet counts in 2 of these patients were between 70×10⁹/L and 100×10⁹/L without associated treatment; an observation policy was established for these patients. The third patient refused further explorations or surgery.

Advances in laparoscopic surgical techniques have favored a laparoscopic approach for splenectomy.^{13 - 26} However, surgical laparoscopy entails different features from open surgery. After the wide application of laparoscopy in other intra-abdominal surgical diseases, which include the treatment of known (colon) or unsuspected (gallbladder) tumors, a higher incidence of early recurrence has been observed at the site ports or at the site of abdominal dissemination.²⁷ Some recent experimental studies suggest that the pneumoperitoneum is able to disseminate free abdominal malignant cells.^{28 - 29} It is well known from experimental and clinical studies that spleen tissue possesses a great facility to implant, even when it is removed after trauma.^{3 - 5} Laparoscopic splenectomy is a technically demanding procedure associated with a notably longer operative time than open splenectomy. The spleen parenchyma is frail; during LS, the spleen is mobilized and tears or parenchymal bleeding can occur. There may be a theoretical high risk of splenosis. One of the patients (patient 2) described in this series had several implants in the splenic fossa. These implants may have been secondary to rupture of the spleen capsula, although the patient was converted to open surgery and major breaches were not observed. Another situation in which spleen tissue can be spilled is during morcellation and retrieval if the bag is broken; however, none of the 4 patients in whom the bag was broken had residual splenic tissue.

Accessory spleens are frequently observed (10%-30%), and the incidence in open series is higher than that reported in laparoscopic series (Table 2).⁷ Accessory spleens can be located in many sites (splenic hilum to scrotum) and are sometimes surrounded by fatty tissue that impairs their visualization. When left in situ, they can prevent the resolution of hematologic disease, mainly in patients with autoimmune thrombocytopenia.^{9 - 12} Gigot et al¹³ described the increased difficulty in locating such accessory organs after the impossibility of finding an AS previously demonstrated by CT scanning. The experience in our series shows that ASs were not identified in 3 cases, although they were looked for. In one case, another AS had been removed previously during LS; in another case, the AS was found during conversion, totally surrounded by fatty tissue. It is difficult to say if ASs are more difficult to visualize by laparoscopy or open surgery, because comparative studies are lacking.

The laparoscopic technique for splenectomy should include, as an important step, the search for ASs. Because of the variable situations, an exploratory laparoscopy should be performed as an initial step. The splenic flexure of the colon and the omentum should be palpated to exclude the existence of ASs at this level. The next step is to explore the large curvature of the stomach. Then, the omental pouch should be opened and the medial face of the spleen and splenic hilum should be inspected. The technique that LS proposes with the initial mobilization of the posterior face of the spleen, which is then excised with several applications of the endostapler without opening the omental pouch, has the drawback that it does not permit the existence of ASs to be identified.²³ The use of perioperative laparoscopic ultrasonography has been proposed, but its real value is unknown. Another point from the practical point of view could be preoperative identification by a CT scan.

The excision of residual ASs is followed by a clinical response of 20% to 60%; laparoscopic identification and removal has been reported.³⁰ In our patients, because of the small size of the residual splenic tissue and the clinical improvement of 2 patients (platelet counts ranging between 70×10⁹/L and 100×10⁹/L), an observation policy was followed. A third patient refused further surgery.

Laparoscopic splenectomy has a promising role in the management of hematologic diseases requiring splenectomy, but it requires exquisite care to avoid parenchymal rupture and cell spillage and to avoid leaving ASs, which can induce the failure of surgical treatment.

This study was supported by grant 97/760 from the Fondo de Investigaciones Sanitarias, Madrid, Spain.

Reprints: Manuel Trias, MD, Service of General and Digestive Surgery, Hospital de la Santa Creu i Sant Pau, Avda Pare Claret 167, 08025 Barcelona, Spain.