Surgical Endoscopy And Other Interventional Techniques

, Volume 20, Issue 8, pp 1214–1220

Laparoscopic splenectomy for hematologic diseases: a preliminary analysis performed on the Italian Registry of Laparoscopic Surgery of the Spleen (IRLSS)

Authors

    • Advanced Laparoscopic Unit, Department of General Surgery and TransplantSan Martino University Hospital, University of Genoa
  • P. Torelli
    • Advanced Laparoscopic Unit, Department of General Surgery and TransplantSan Martino University Hospital, University of Genoa
  • S. Squarcia
    • Physic and Medical Statistics Laboratory, Department of PhysicUniversity of Genoa
  • M. P. Sormani
    • Clinical Epidemiology UnitNational Institute for Cancer Research
  • A. Savelli
    • Advanced Laparoscopic Unit, Department of General Surgery and TransplantSan Martino University Hospital, University of Genoa
  • B. Troilo
    • Advanced Laparoscopic Unit, Department of General Surgery and TransplantSan Martino University Hospital, University of Genoa
  • G. Santori
    • Advanced Laparoscopic Unit, Department of General Surgery and TransplantSan Martino University Hospital, University of Genoa
  • U. Valente
    • Advanced Laparoscopic Unit, Department of General Surgery and TransplantSan Martino University Hospital, University of Genoa
Article

DOI: 10.1007/s00464-005-0527-5

Cite this article as:
Casaccia, M., Torelli, P., Squarcia, S. et al. Surg Endosc (2006) 20: 1214. doi:10.1007/s00464-005-0527-5

Abstract

Background

The Italian Registry of Laparoscopic Surgery of the Spleen (IRLSS) was developed to provide at the national level an informative tool useful for performing multicenter studies in the field of spleen laparoscopic surgery. In this first study analyzing the IRLSS data, a cohort of patients with hematologic diseases was retrospectively investigated for potential predictive parameters that could affect the outcome of laparoscopic splenectomy.

Methods

A total of 309 patients who underwent laparoscopic splenectomy for hematologic diseases in 17 Italian centers (between February 1, 1993, and September 30, 2004) were entered in the IRLSS. Their records were analyzed retrospectively by the Student’s t-test, chi-square, and logistic regression.

Results

The mean operative time was 141 min (range, 30–420 min). Conversion was necessary in 21 cases (7%), and approximately 1 accessory spleen in 25 patients (9%) was found. The mean spleen weight was 1191 g (range, 85–4,500 g). Perioperative death occurred in two cases (0.6%). No complications were experienced by 253 patients (81.9%), who had a mean hospital stay of 5.4 days (range, 2–30 days). Overall morbidity occurred in 56 patients (18.1%), mainly associated with transient fever (n = 22), pleural effusion (n = 13), and actual or suspected hemorrhage (n = 12), requiring a reintervention for 7 patients. Multivariate analysis found that body mass index (p = 0.024) and clinical indication (p = 0.004) were independent predictors for surgical conversion. The clinical indication was almost significant as an independent predictor for the occurrence of postoperative complication (p = 0.05).

Conclusions

This first study analyzing the IRLSS data shows that laparoscopic splenectomy may represent the gold standard treatment for hematologic diseases with normal-size spleen. The low morbidity and mortality rate suggests that laparoscopic splenectomy can be successfully proposed also for splenomegaly in hematologic malignancies.

Keywords

Hematologic diseasesLaparoscopyMorbidityMortalitySplenectomy

Since 1991, when Delaître and Maignien [3] attempted the first splenectomy using the laparoscopic approach, laparoscopic splenectomy (LS) has gained worldwide popularity and acceptance as an effective and advantageous surgical option. Many studies suggest that LS represents the “gold standard” for the treatment of benign hematologic diseases, with or without splenomegaly [6, 12, 30].

Although the most common indication for LS is idiopathic thrombocytopenic purpura, many other benign or malignant hematologic diseases derive benefit from this procedure. However, the role of laparoscopy for many hematologic malignancies (HM) is complex, and the benefits should be carefully evaluated against the risks [9]. In fact, patients with malignant disease tend to have larger spleens than patients with benign disorders. At the beginning of the laparoscopic experience, patients with enlarged spleens presented the greatest challenge for surgeons [17, 25, 27].

Currently, technical development and improved skills have produced an extension of the LS indications in HM cases, independently by the associated splenomegaly [1, 8, 28, 3236]. However, these conditions still are under discussion, and their treatment requires further studies [21, 31]. Questions have already been raised about patient selection and operative technique [26].

In view of these problems, the Italian Registry of Laparoscopic Surgery of the Spleen (IRLSS) was developed to provide at the national level an informative tool for performing multicenter studies in the field of spleen laparoscopic surgery [37]. In this first study analyzing IRLSS data, a cohort of patients with hematologic disease was retrospectively investigated for potential predictive parameters that could affect the LS outcome.

Patients and methods

Data collection

In December 2000, the IRLSS was formally launched under the auspices of the Italian Society for Endoscopic Surgery and New Technologies (SICE) [37]. A total of 17 Italian laparoscopic centers from all over the country asked to adhere to the Registry. Data collection, initially carried out by questionnaires. Subsequently, data collection was performed by use of a specific database developed in the MS Access (Microsoft Corporation, Redmond, WA, USA) environment. This database can be downloaded from the SICE Web site (http://www.siceitalia.com/db_on-line_milza.htm).

Patient characteristics

The first group entered in the IRLSS comprised 309 patients (mean age, 43 ± 21 years; range, 5–84 years; male/female: 153/156) treated using LS for different hematologic diseases (Table 1) between February 1, 1993, and September 30, 2004. The patients were classified according to American Society of Anesthesiologists (ASA) scoring as follows: ASA 1 (n = 61; 19.7%), ASA 2 (n = 158, 51.1%), and ASA 3 (n = 52, 16.8%). A total of 27 patients (8.7%) had a body mass index (BMI) greater than 30 kg/m2. Previous abdominal surgery before LS had been performed for 76 patients (24.6%).
Table 1

Overall characteristics of patients who underwent laparoscopic splenectomy and the results of their stratification for hematologic disease

 

Overall

HM

TP

HA

OP

Patient (n)

309

117

112

65

15

Age

     

  Mean ± SD (years)

43 ± 21

56.1 ± 13.4

40.1 ± 19.9

23.2 ± 17.6

46.3 ± 19.5

Range (years)

5–84

17–84

6–80

5–77

15–76

Gender

     

  Male: n (%)

156 (50.5)

75 (64)

46 (41)

27 (42)

1 (25)

  Female: n (%)

153 (49.5)

42 (36)

66 (59)

38 (58)

3 (75)

BMI

     

  Mean ± SD (kg/m2)

24.1 ± 4.7

24.8 ± 4.2

25.1 ± 4.8

20.4 ± 3.8

24.5 ± 1.9

Range (kg/m2)

12.8–45.3

14.7–38.2

17.1–45.3

12.8–29.1

21.5–27.3

Missing data: n (%)

38 (12.3)

14 (12)

11 (9.8)

11 (17)

2 (13)

ASA score

     

  Mean ± SD

2 ± 0.7

2.2 ± 0.6

1.8 ± 0.6

1.7 ± 0.5

2 ± 0.4

  Range

1–4

1–4

1–3

1–3

1–3

  Missing data: n (%)

38 (12.3)

16 (13.6)

13 (11.6)

7 (10.7)

2 (13)

HA, hemolytic anemia; OP, other pathologies; HM, hematologic malignancy; TP, thrombocytopenic purpura; BMI, body mass index; ASA, American Society of Anesthesiologists

The indications for splenectomy are detailed in Table 2. Administration of pneumococcal, meningococcal, and antihemophilus vaccine to 242 patients (78.3%) was reported.
Table 2

Clinical indications for laparoscopic splenectomy

 

N

Patients (n)

309

HM (n)

117

  Non-Hodgkin’s lymphoma (n)

70

  Hodgkin’s lymphoma (n)

18

  Idiopathic myelofibrosis (n)

12

  Chronic lymphatic leukemia (n)

7

  Hairy-cell leukemia (n)

6

  Other HM (n)

4

TP (n)

112

  Idiopathic thrombocytopenic purpura (n)

98

  Thrombotic thrombocytopenic purpura (n)

6

  HIV-related thrombocytopenia (n)

2

  Other thrombocytopenia (n)

4

  Unknown (n)

2

HA (n)

65

  Hereditary spherocytosis (n)

34

  Major beta-thalassemia (n)

20

  Autoimmune hemolytic anemia (n)

9

  Other HA (n)

2

OP (n)

15

HM, hematologic malignancies; TP, thrombocytopenic purpura; HA, hemolytic anemia; HIV, human immunodeficiency virus; OP, other pathologies

Surgical procedures

Mean spleen longitudinal diameter (SLD), measured by preoperative ultrasonography or computed tomography (CT) scan, was obtained for 247 patients, whereas a measure of the transversal and sagittal axis was rarely performed. The fully laparoscopic approach was the preferred technique, with the hand-assistance approach used for only 24 patients (8%). The right hemilateral decubitus position was used for 207 patients (67%), with the surgeon and the assistant on the right side of the patient.

In general, four trocars and a 30° optic were necessary. After exploration of the abdominal cavity (liver and lymph nodes) and exclusion of an accessory spleen, the left colic angle was lowered, and the inferior pole of the spleen was mobilized. For this dissection, the most commonly used tool was the ultrasonic dissector (57%). Next, the short gastric vessels were cut after successful achievement of hemostasis with clips. The hilar splenic vessels were freed, and hemostasis was achieved with a stapler. The spleen was put in a plastic bag and extracted by fragmentation and passage through the orifice of the 15-mm trocar. For cases involving larger spleens and those requiring pathologic examination of the surgical specimen for documentation of the hematologic disease, a service minilaparotomy was performed through a left subcostal incision or through an upper midline/Pfannenstiel incision. A drain was left in situ in 242 patients. The spleen was retained for histopathologic examination. The different techniques used by the various surgical teams are reported in Table 3.
Table 3

Approach, techniques, and instruments applied during laparoscopic splenectomy

Approach (n)

295

  Fully laparoscopic: n (%)

271 (87.7)

  Hand-assisted: n (%)

24 (7.8)

  Missing data: n (%)

14 (4.5)

Decubitus (n)

294

  Hemi-lateral: n (%)

197 (63.75)

  Lateral: n (%)

71 (23)

  Supine: n (%)

26 (8.4)

  Missing data: n (%)

15 (4.85)

Trocar (n)

298

  4: n (%)

216 (70)

  3: n (%)

47 (15.2)

  5: n (%)

35 (11.3)

  Missing data: n (%)

11 (3.5)

Optic (n)

297

  30°: n (%)

281 (91)

  0°: n (%)

11 (3.5)

  45°: n (%)

5 (1.6)

  Missing data: n (%)

12 (3.9)

Dissection technique (337 dissection instruments/309 patients): n (%)

 

  Ultrasonic dissector n (%)

189 (56)

  Monopolar coagulation n (%)

106 (31.5)

  Bipolar coagulation n (%)

42 (12.5)

Splenic vessels division (391 techniques/309 patients): n (%)

 

  Vascular stapler, n (%)

222 (57)

  Clips, n (%)

81 (21)

  Ligature, n (%)

53 (13)

  Radiofrequency instruments, n (%)

35 (9)

Service minilaparotomy, (n)

296

  No: n (%)

197 (63.75)

  Yes: n (%)

99 (32)

    Left subcostal incision (n)

76

    Pfannenstiel incision (n)

21

    Upper midline incision (n)

2

  Missing data: n (%)

13 (4.2)

Statistical analysis

Only variables with no differences concerning distribution of missing data between laparoscopic centers were included for statistical analysis. The results are expressed as mean ± standard deviation. All comparisons between groups were performed using the Student’s t-test or the Mann–Whitney U test. The chi-square test was used to assess the relationships between categorical variables. A logistic regression model was used to identify variables affecting the probability of a conversion and of postoperative complication in univariate analysis. All parameters with a p value of 0.10 or less in univariate analysis were included in a multivariate model by use of a backward selection procedure to evaluate potential independent predictors for conversion and postoperative complications. Statistical analysis was performed using the software package SPSS 12.0.1 (SPSS Inc, Chicago, IL, USA).

Results

The main anatomic and surgical parameters of the patients who underwent LS and the results of their stratification for the most frequent hematologic diseases are shown in Table 4. The overall operative time was 141 ± 57 min, and the findings showed a significant difference between HM (152.4 ± 58.1 min) and non-HM (133.8 ± 54.7 min; p = 0.007) patients. The operative time for entirely laparoscopic operations (140 ± 55 min) was shorter than for cases that required an open conversion (154 ± 74 min), without reaching statistical significance (p = 0.29).
Table 4

Anatomic and surgical parameters for the patients undergoing laparoscopic splenectomy and the results of their stratification for hematologic disease

 

Overall

HM

TP

HA

OP

Patients (n)

309

117

112

65

15

SLD (cm)

     

  Mean ± SD (cm)

16.1 ± 5.5

19.2 ± 6.2

12.4 ± 2.9

16.3 ± 3.4

14.7 ± 3.3

  Range (cm)

8–35

8–35

8–26

9–25

10–20

  Missing data (n)

62

15

26

13

8

Spleen weight (g)

     

  Mean ± SD

1,191 ± 1,046

1,826 ± 1,178

493 ± 346

946 ± 750

980 ± 148

  Range (g)

85–4,500

150–4,500

85–1,600

300–3150

800–1,200

  Missing data (n)

231

82

86

53

10

Accessory spleen: n (%)

25 (8)

11 (9.4)

11 (9.8)

3 (4.6)

0

  Missing data: n (%)

18 (5.8)

13 (1.1)

4 (3.5)

0

1 (6.6)

Estimated blood loss

     

  <100 ml: n (%)

183 (59.2)

54 (46)

68 (60.71)

49 (75.38)

12 (80)

  100–500 ml: n (%)

81 (26.2)

40 (34)

29 (25.90)

10 (15.9)

2 (13.3)

  >500 ml: n (%)

26 (8.4)

18 (16)

7 (6.25)

1 (1.53)

0

  Missing data: n (%)

19 (6.1)

5 (4)

8 (7.14)

5 (7.7)

1 (6.7)

Blood transfusion: n (%)

83 (27)

40 (34)

26 (23)

15 (23)

2 (13.3)

  Missing data: n (%)

42 (13.6)

15 (12.8)

9 (8)

13 (20)

5 (33.3)

Operating time (min)

     

  Mean ± SD

141 ± 57

152 ± 58

129 ± 52

142 ± 57

134 ± 54

  Range (min)

25–420

25–420

55–420

60–300

45–220

  Missing data: n (%)

25 (81)

10 (8.5)

7 (6)

6 (9)

2 (13.3)

Conversion: n (%)

21 (7)

15 (12.8)

4 (3.5)

1 (1.5)

1 (6.7)

  Missing data: n (%)

2 (0.006)

0

1 (0.008)

0

1 (6.7)

Hospital stay (days)

     

  Mean ± SD

5.4 ± 3.1

6.1 ± 3.7

4.7 ± 2.2

4.8 ± 2.1

6.6 ± 5.7

  Range (days)

2–30

2–30

2–17

3–16

3–26

  Missing data: n (%)

13 (4.2)

8 (6.8)

2 (1.8)

2 (3)

1 (6.7)

Morbidity: n (%)

56 (18)

26 (22)

20 (18)

5 (7.7)

5 (33.3)

Mortality: n (%)

2 (0.6)

2 (1.7)

0

0

0

HA, hemolytic anemia; OP, other pathology; HM, hematologic malignancies; TP, thrombocytopenic purpura; SLD, spleen longitudinal diameter

Obesity was not found to be a factor significantly affecting operating time (144 ± 53 min for patients with a BMI >30 vs 139 ± 59 min for patients with a BMI <30; p = 0.72). In the HM group, LS was associated with a lymph node biopsy (n = 7) or a liver biopsy (n = 10) for diagnostic or disease-staging reasons. The ASA score was significantly higher for HM patients (2.2 ± 0.6) than for non-HM patients (1.81 ± 0.6; p < 0.001), as was the BMI (24.8 ± 4.2 vs 23.6 ± 4.9, respectively; p = 0.024). A cholecystectomy was associated with the procedure for 25 hemolytic anemia patients and 6 thrombocytopenic purpura (TP) patients. One or more accessory spleens were found and removed for 25 patients (9%) (Table 4). The overall spleen weight was 1,191 ± 1,046 g (range, 85–4,500 g), whereas the SLD was 16.1 ± 5.5 cm (range, 8–35 cm). In HM patients, both spleen weight (1,826 ± 1,178 g) and SLD (19.2 ± 6.2 cm) reached significantly higher values than in non-HM patients (676 ± 522 g and 13.9 ± 3.7 cm, respectively), with a statistical significance for each comparison (p <0.001).

The estimated blood loss was less than 100 ml for 183 patients (63%), between 100 and 500 ml for 81 patients (28%), and more than 500 ml for 26 patients (9%) (Table 4). A total of 83 patients (31.1%) had a perioperative and/or a postoperative transfusion, and HM patients had a significantly higher transfusion rate (39.2%) than non-HM patients (24.6%, p = 0.029).

The accessory incision rate was affected by the need to examine the whole spleen or the presence of massive splenomegaly. For this reason, the accessory incision rate increased in patients grouped for malignant or enlarged spleen (51.8%) with respect to benign (30.8%) and normal-size spleen (17.6%) groups. A conversion from LS to open splenectomy was required for 21 patients (7%). Surgery conversion was because of hemorrhagic problems involving the hilar splenic vessels (12 cases), a complicated dissection of the hilum (3 cases), massive splenomegaly (2 cases), strong diaphragmatic adhesions (2 cases), and a locally advanced disease with nearby organ involvement. When conversion rate was considered according to hematologic diseases, it was found to be threefold higher in the malignant group (12.8%) than in the benign group (4%), with the difference reaching statistical significance (p = 0.003). The findings showed that 21.1% of the patients who experienced a conversion from LS to open splenectomy were obese, as compared with 9.2% of the patients who underwent completed laparoscopic procedures (p = 0.10).

Oral feeding started on postoperative day 2.29 ± 0.99. The postoperative hospital stay was 5.4 ± 3.1 days (Table 4). This parameter was significantly greater after surgery conversion (8.7 ± 6.6 days) than after laparoscopically completed splenectomy (5.1 ± 2.6 days, p < 0.001), and for HM patients (6.1 ± 3.7 days) than for non-HM patients (4.9 ± 2.7 days, p = 0.002). Oral feeding and postoperative hospital stay were not significantly different between obese patients and those with normal weight (data not shown).

The postoperative course was uncomplicated for 253 patients. Overall morbidity was 18.1% (66 episodes in 56 patients) and attributable to transient fever (22 cases), pleural effusion (13 cases), actual or suspected hemorrhage (12 cases) requiring reintervention (7 cases), subphrenic collections (6 cases), abdominal wall complications (6 cases), pneumopathology (5 cases), and pancreatitis (2 cases). Postoperative complications occurred for 7 of 21 converted patients (33.3%), as compared with 49 of 286 patients without surgery conversion (17.1%), but this difference failed to reach statistical significance (p = 0.078). Two patients (0.6%) experienced perioperative deaths. One patient affected by idiopathic myelofibrosis died 15 days after surgery as a consequence of secondary blast crisis. One perioperative death attributable to myocardial infarction occurred for an 80-year-old patient affected by non-Hodgkin’s lymphoma.

Logistic regression analysis was performed to evaluate patient characteristics potentially able to affect conversion from LS to open surgery splenectomy. The parameters included in the model are shown in Table 5. Univariate analysis showed that pathology type (HM vs non-HM; odds ratio [OR], 4.51; p = 0.003), BMI (OR, 1.12; p = 0.014), and ASA score (OR, 2.21; p = 0.025) were significantly associated with the risk for surgical conversion. The type of pathology was coded as a binary variable (HM vs non-HM patients) because the risk of a conversion was similar for all non-HM patients. Patient age (OR, 1.02; p = 0.06) and sex (males vs females; OR, 2.57; p = 0.058) were correlated with the probability of conversion, whereas no significant correlation was found for SLD and spleen weight.
Table 5

Univariate and multivariate logistic regression of potential predictive parameters for surgical conversion from laparoscopic splenectomy to open surgery

 

Univariate analysis

Multivariate analysis

Parameter

OR

95% CI

p Value

OR

95% CI

p Value

Patient age (years)

1.02

0.99–1.05

0.06

0.99

0.96–1.02

0.52

Patient sex

      

Female

1

     

Male

2.57

0.87–6.82

0.058

1.68

0.55–5.19

0.37

Pathology

      

  HM

4.51

1.7–11.98

0.003

4.71

1.62–13.64

0.004

  Non-HM

1

     

ASA score

2.21

1.11–4.39

0.025

2.12

0.91–4.96

0.08

BMI

1.12

1.02–1.22

0.014

1.11

1.01–1.23

0.024

SLD

1.06

0.98–1.15

0.16

 

Not entered

 

Spleen weight

1.32

0.63–2.8

0.46

 

Not entered

 

OR, odds ratio; CI, confidence interval; HM, hematologic malignancies; ASA, American Society of Anesthesiologists; BMI, body mass index; SLD, spleen longitudinal diameter

All the variables with a p value less than 0.10 at univariate analysis were entered in a multivariate model. The final model retained pathology type (OR, 4.71 for HM vs non-HM; p = 0.004) and BMI (OR, 1.11 for each 1-point increase; p = 0.024). Univariate and multivariate analyses of parameters affecting the risk for postoperative complications are shown in Table 6. In the univariate analysis, only the pathology was significantly associated with the risk for a postoperative complication (p = 0.05), with hemolytic anemia patients showing a lower risk. The multivariate analysis also included the following variables with a p value lower than 0.10 in the univariate analysis: age (p = 0.097), BMI (p = 0.074), SLD (p = 0.08), and conversion (p = 0.071). None of these variables was retained in the final model.
Table 6

Univariate and multivariate logistic regression of potential predictive parameters for morbidity in laparoscopic splenectomy

 

Univariate analysis

Multivariate analysis

Parameter

OR

95% CI

P Value

OR

95% CI

p Value

Patient age

1.012

0.99–1.02

0.10

0.99

0.97–1.02

0.73

Patient sex

      

  Female

1

     

  Male

1.52

0.84–2.73

0.16

 

Not entered

 

Pathology

      

  HM

3.43

1.25–9.42

 

3.10

0.77–12.51

 

  HA

1

0.05

 

0.073

  TP

2.61

0.93–7.33

 

2.26

0.61–8.43

 

  Other

6

1.47–24.55

 

8.49

1.21–59.63

 

ASA score

1.14

0.71–1.82

0.59

 

Not entered

 

BMI

1.06

0.99–1.13

0.07

1.03

0.96–1.12

0.42

SLD

1.05

0.99–1.11

0.08

1.02

0.95–1.09

0.62

Spleen weight

0.99

0.56–21.76

0.93

 

Not entered

 

Conversion

      

  No

1

     

  Yes

2.42

0.93–6.3

0.07

2.31

0.68–7.83

0.18

OR, odds ratio; CI, confidence interval; HM, hematologic malignancies; HA, hemolytic anemia; TP, thrombocytopenic purpura; ASA, American Society of Anesthesiologists; BMI, body mass index; SLD, spleen longitudinal diameter

Discussion

The IRLSS was developed with the technical support of SICE to provide at the national level an informatics tool useful for performing both retrospective and prospective multicenter studies investigating spleen laparoscopic surgery [37]. As an official member of the European Association for Endoscopic Surgery since 1996, SICE is the leading Italian scientific society for endoscopic surgery, representing more than 500 surgeons involved in practice and diffusion of new technologies in this field. Although adherence to the IRLSS is on a voluntary basis, the majority of the Italian laparoscopic centers have participated with data entry, making IRLSS highly representative of LS activity in Italy.

In this first study analyzing IRLSS data, a cohort of patients with hematologic diseases was retrospectively investigated for potentially predictive parameters that could affect LS outcome. Although blood parameters such as preoperative hemoglobin and platelet count may play a role in postoperative morbidity [13, 33], they were not entered for statistical analysis because a nonhomogeneous distribution of missing data between laparoscopic centers occurred. Notwithstanding the excellent outcome for non-HM according to elective LS reporting [6, 12, 22, 25], the role of the laparoscopic approach in HM is more complex. In fact, for defining the “perfect candidate” to undergo LS, spleen size is a fairly basic parameter, considering that several factors (e.g., patient age, sex, BMI, ASA score, presence of malignancy, associated treatment) may potentially affect hematologic disease outcome [33]. Thus, the IRLSS data also were analyzed for several clinical diagnosis-related parameters. In our series, HM was an important indication for LS, representing 38% of the patients overall, in contrast to previous reports [12, 20].

In the literature, there is no univocal definition for the term “splenomegaly.” Some authors use the term “spleen weight,” although differences emerge from one author to another [1, 5, 6, 19]. These data, largely missing in IRLSS series, are provided for only 78 patients. However, this parameter is only a postoperative finding, which cannot give a reliable indication for laparoscopic feasibility. Although spleen CT volumetric study may provide valuable information, it still is difficult to obtain and seems to be non–cost effective. The easiest parameter available is SLD measurement, taken by simple preoperative ultrasonography [1, 10, 12, 22, 28]. In the current series, an SLD measurement was obtained for the majority of the IRLSS patients (79.3%).

Concerning the surgical technique, we differed from other groups in our choice of the decubitus position. In our study, the hemilateral decubitus was the preferred position (67%), as compared with the lateral (24%) or supine decubitus (9%) position. This is partly attributable to a high rate of HM indications and massive splenomegaly. With the lateral approach, originally described as the “hanged spleen technique” [2], the patient is put in the right lateral decubitus position, with the surgeon on the right side of the patient. This approach is convenient for a spleen of normal size because the organ is suspended and dissection of the pedicle is easier, thanks to the fall of the stomach and omentum from the operative field.

In the case of massive splenomegaly, the lateral decubitus threatens to displace the huge spleen medially and to decrease the already reduced abdominal “working space.” For this reason, the hemilateral position seems to be more indicated, because the operative table can be tilted to a supine or complete lateral decubitus position if necessary, according to the different dissection steps. Moreover, in the case of an enlarged spleen, a supine position favors the ligature of the splenic artery as the first step of the procedure, as suggested by Samama et al. [27]. In addition, a supine position is indicated when a cholecystectomy is associated with the procedure, to avoid patient repositioning on the operative table, as observed with 25 hemolytic anemia and 6 TP patients in the IRLSS series.

The operative time in the current series was similar to that reported in single-center series [1, 28, 32, 35]. In particular, a direct correlation between surgery time and spleen size was observed. This may be explained by the fact that during LS for splenomegaly, the space available in the abdominal cavity after creation of the pneumoperitoneum is diminished, making organ manipulation more difficult. In addition, extra time must be added for extraction of a large spleen. In our series, retrieval of the spleen had a major impact on the operating time. In an effort to minimize the time required for retrieval of the large spleens, an accessory incision was made in some cases.

In this first IRLSS dataset, the overall conversion rate reached 7%, consistent with previous LS series [1, 4, 10, 12, 2022]. The BMI values and ASA scores were higher for the patients who required conversion from LS to open splenectomy, confirming that BMI may increase the technical difficulty of laparoscopic procedures [24]. In our series, besides the BMI value and the ASA score, the most powerful predictor of conversion to open splenectomy was the type of pathology, considering that the conversion rate for HM patients rose to 12.8%. In any case, this result seems to represent a good outcome, as compared with other series, in which only data for HM patients were taken into account [15, 18, 21, 38].

An SLD value exceeding 22 to 23 cm was correlated with a higher risk of conversion to laparotomy [14, 34]. In the IRLSS series, SLD was found to be a poor predictor of conversion to open splenectomy, suggesting that other factors may affect the surgical outcome. In particular, abdominal “working space” may play a critical role, being strongly related to both body habitus and abdominal wall relaxation. Spleen size may be associated with abdominal dimensions for an approximate indication of the intraoperative “working space.” However, to date, no quantitative parameter or ratio of spleen size to patient body habitus is available.

Open splenectomy for massive splenomegaly is associated with high mortality and morbidity rates (20% to 60%) [7, 11, 29]. More recently, a mortality rate of 2% and a morbidity rate of 26% in a series of 47 patients with splenomegaly undergoing open splenectomy was reported [16], making these results comparable with those of the latest series of massive splenomegaly treated using laparoscopy [26, 38]. In previous series, a higher complication rate was observed in splenectomy performed for malignancy [10, 15, 38]. In the univariate analysis performed on the IRLSS dataset, postoperative complications occurred more frequently for patients with a diagnosis of HM. In the multivariate analysis performed by entering the variables that in univariate analysis reached a p value less than 0.1 [23], only pathology showed statistical significance, suggesting a major role for HM, in agreement with Targarona et al. [33]. The HM group presented a morbidity rate of 22.2%, according to previous series [10, 15, 38]. The TP patients presented a morbidity rate not very different from that of the HM group. The main postoperative concerns in the TP group were pulmonary and urinary complications and/or skin incision infections. These complications may be explained by long-term administration of steroids to TP patients. Both corticosteroid therapy and low platelet count also may be responsible for cases of hemoperitoneum or abdominal fluid collection [13]. Two deaths occurred for patients who had HM with massive splenomegaly, but the deaths had no relationship with the surgical procedure. The mortality rate in the IRLSS series was 0.6%, similar to that of the aforementioned series.

This first study analyzing the IRLSS data represents one of the largest multicenter analyses concerning LS for patients with hematologic diseases. The IRLSS confirms that LS may be considered the “gold standard” treatment for benign hematologic diseases involving spleens of normal size. Moreover, the laparoscopic approach also can be proposed successfully for HM with associated splenomegaly. Malignancy is the only preoperative clinical parameter found to be predictive for both conversion to laparotomy and a complicated postoperative course. However, the morbidity and mortality rates may remain at a low level in experienced hands and with the usage of up-to-date technological devices. This preliminary study suggests that malignancies and massive splenomegaly may be usefully considered for laparoscopic surgery. Further prospective studies involving the totality of the Italian laparoscopic centers should be performed to confirm these data.

Acknowledgments

This study was technically supported by the Italian Society for Endoscopic Surgery and New Technologies (SICE; http://www.siceitalia.com).

Copyright information

© Springer Science+Business Media, Inc. 2006