Postoperative outcomes after open splenectomy versus laparoscopic splenectomy in cirrhotic patients: a meta-analysis

Khaled Al-raimi and Shu-Sen ZhengHangzhou, China

?

Postoperative outcomes after open splenectomy versus laparoscopic splenectomy in cirrhotic patients: a meta-analysis

Khaled Al-raimi and Shu-Sen Zheng
Hangzhou, China

Author Affiliations: Division of Hepatobiliary Pancreatic Surgery, First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China (Al-raimi K and Zheng SS)

? 2016, Hepatobiliary Pancreat Dis Int. All rights reserved.

Published online January 7, 2016.

BACKGROUND: Laparoscopic splenectomy is considered the gold standard for resecting normal-to-moderately bigger spleens in benign conditions, and in addition could be tried for patients with malignant splenic disorders. However, the safety of laparoscopic splenectomy in patients with hypersplenism is not well-known. This study aimed to investigate the efficacy and safety of laparoscopic splenectomy for patients with hypersplenism secondary to liver cirrhosis by comparing with the open splenectomy.

DATA SOURCES: Several databases were searched to identify comparative studies fulfilling the predefined selection criteria from January 2000 to June 2015. The subsequent key words were utilized for browsing “l(fā)aparoscopy” or “l(fā)aparoscopic”, “open”, “splenectomy”, and “l(fā)iver cirrhosis”. Studies evaluating laparoscopic and open splenectomy for patients with liver cirrhosis were incorporated. Two evaluators personally strained the title and abstract of each publication. Citations with contemplated compliance within our eligibility criteria underwent compressed review. Meta-analysis was carried out according to the recommendations of the Cochrane Collaboration software (review manager 5.1).

RESULTS: Seven studies containing 509 patients were included. Compared with the open splenectomy group, patients in the laparoscopic splenectomy group had significantly less intraoperative blood loss (MD=210.30; 95% CI: 11.28-409.32; P=0.04), longer operative time (MD=-31.58; 95% CI: -53.34--9.82; P=0.004), shorter duration of postoperative hospital stay (MD=3.41; 95% CI: 2.39-4.43; P<0.01), lower incidence of postoperative complications (RR=1.34; 95% CI: 0.88-2.01; P=0.17), and decreased liver damage [ALT (MD=8.52; 95% CI: 0.19- 16.85; P=0.05) and total bilirubin (MD=5.12; 95% CI: 0.37 -9.87; P=0.03)].

CONCLUSION: Hypersplenism secondary to cirrhosis and portal hypertension should not be a contraindication for laparoscopic splenectomy.

(Hepatobiliary Pancreat Dis Int 2016;15:14-20)

KEY WORDS:open splenectomy; laparoscopic splenectomy; cirrhosis

Introduction

Since 1950, surgical splenectomy is common in practice for splenomegaly and hypersplenism.[1]However, open splenectomy (OS) is exceedingly intrusive, has more chance for intraoperative hemorrhage and postoperative pain, and cannot be performed if the patient has poor liver function.[1]Furthermore, OS has a substantial mortality in patients with poor liver function and related hypersplenism.[2]

Laparoscopic splenectomy (LS) was first used in 1997 and has been well acknowledged in patients with splenomegaly due to its promising clinical benefits.[3]Currently, LS is among the most de facto standard for removing typical to somewhat bigger spleens in benign conditions, and in addition could possibly be tried for patients with malignant splenic disorders. On the other hand, the security of LS in patients with hypersplenism is not wellknown. Patel et al[4]investigated the feasibility of LS on patients with massive splenomegaly and found that although LS is feasible in patients with giant spleen, it is associated with longer median operating time, high con-version rate, postoperative morbidity and median postoperative stay.[3]As a result of the collective experiences of laparoscopic surgeries and up to date advancements in operating equipments, specifically vessel sealing systems, laparoscopic approaches are progressively suggested for a variety of conditions which include patients with liver cirrhosis and portal hypertension.[5, 6]Tomikawa and colleagues[1]compared LS to partial splenic embolization and found that patients who had undergone LS had no significant complications, better liver function preservations, and higher platelet count, and LS did not impact the undergoing treatment. Reso and coworkers[7]demonstrated that after splenic artery embolization, handassisted LS or LS was safe, and there was no conversion to OS.

This analysis looked into the efficiency and safe practices of LS for patients with hypersplenism secondary to liver cirrhosis by evaluating together with the OS outcomes. The results of hematologic change, impairment of liver function, and perioperative data were discussed.

Methods

Literature search strategies and selection

An internet-based search along with a manual search was utilized to acquire suitable studies. Five electronic data sources were scrutinized for this analysis (PubMed, Springer, Cochrane Reviews, Ovid and Embase) from January 2000 to June 2015. The following key words were used: “l(fā)aparoscopy” or “l(fā)aparoscopic”, “open”, “splenectomy”, and “l(fā)iver cirrhosis”. In case any meta-analysis or review was identified, a manual search of reference lists from these retrieved publications was performed.

Inclusion and exclusion criteria

Studies comparing laparoscopic and open splenectomy for patients with liver cirrhosis were included no matter whether they were randomized controlled trials or retrospective studies. The exclusion criteria were as follows: (1) it absolutely was unattainable to acquire the proper information through the published articles; (2) there were considerable convergence among authors, institutes, or patients in the literature; (3) the assessed outcomes were not obviously shown in the literature; and (4) long-term follow-up was not documented in articles as it had not been cited in the studies.

Study eligibility assessment

Two reviewers manually strained the publication title and summary for each publication. Details with considered conformity in the eligibility criteria went through compacted evaluation. If two reviewers identified a citation being most likely pertinent, we acquired the full-text article to get a complete evaluation. The two reviewers made a decision on the eligibility of incorporated publication’s citations for a full-text evaluation in the filtering process. The dissimilarities were resolved by the authors. Once this failed to provide a firm conclusion, the senior author developed a very last judgment on the eligibility of the study.

Outcome of interest

The primary outcomes included duration of operative time in the two groups, intraoperative blood loss, spleen weight after the resection, and duration of hospital stay. Additionally, the outcome measures were overall postoperative complications, and the changes in biochemical parameters after the procedure.

Data analysis and statistical processing

Meta-analysis was performed in line with recommendations from the Cochrane Collaboration software (review manager 5.1). Statistical power analysis was performed by G*Power (Version 3.1). Heterogeneity with the research was evaluated at first employing a randomeffects model. A P<0.05 was considered statistically significant. Continuous variables were determined by using mean difference (MD) as the summary statistics by the inverse-variance method. Odds ratio (OR) for dichotomous variables was analyzed by the Mantel-Haenszel method and equally was documented with 95% CI. MD and OR were regarded as P<0.05 if the 95% CI did not incorporate the value “1”. Treatment group was described as the odds of an undesirable event transpiring in the LS group in comparison with the OS group, but MDs symbolized the variations in continuous variables between the two groups. Finally, publication bias was assessed by Begg’s test and Egger’s test.

Results

Literature selection

Fig. 1 illustrated the search process and results. A total of 231 articles matched our search key words, and another seventy articles were searched manually. A total of 261 articles remained after duplicates were removed. After carefully reading titles and abstracts of the rest, 65 fulltext articles were included. Finally, 58 of the 65 papers were excluded because they lacked extractable data, had no proper arms to compare, were not suitable types of articles, or were not written in English. Only 7 studies met our inclusion criteria in the end.

Baseline characteristics of included studies

Fig. 1. Flow diagram of literature selection.

The baseline characteristics of the included studies are summarized in Table. A total of 7 studies containing 509 patients were included.[8-14]Patients were divided into two groups according to surgical procedures, 232 were in the LS group and 277 in the OS group. Gender, mean age, Child-Pugh class and spleen weight were compared between the two groups in each study except for the studies of Watanabe et al[8]and Ando et al,[9]in which some basic characteristic data could not be obtained. In our studies, liver cirrhosis was the main cause for hypersplenism or splenomegaly.

Meta-analysis results

Estimated intraoperative blood loss

Estimated intraoperative blood loss in OS vs LS was mentioned in seven studies.[8-14]Heterogeneity among these studies was significant (I2=97%; P<0.00001). Results of random-effect model meta-analysis showed that the amount of intraoperative blood loss of the OS group was significantly more than that of the LS group (MD=210.30; 95% CI: 11.28-409.32; P=0.04) (Fig. 2). The statistical power of the test was high (1-β=0.9999). Begg’s test and Egger’s test showed no publication bias (Z=0.30, P=0.764; t=-1.03, P=0.351).

Fig. 2. Estimated intraoperative blood loss in OS vs LS.

Table. Baseline characteristics of included studies

Operative time

Operative time in open vs laparoscopic approach was reported in all included studies.[8-14]Heterogeneity was also detected (I2=84%; P<0.00001). The pooled outcome using random-effect model indicated that the operative time was much longer in the LS group than that in the OS group (MD=-31.58; 95% CI: -53.34--9.82, P=0.004) (Fig. 3). The statistical power of the test was high (1-β=1.0000). Begg’s test and Egger’s test revealed no publication bias (Z=1.20, P=0.230; t=-0.59, P=0.579).

Duration of postoperative hospital stay

There were six studies[9-14]on the difference of postoperative hospital stay between the two groups. The heterogeneity in these studies was significant (I2=73%; P=0.003), so with the random-effect model, we observed that the postoperative hospital stay of the OS group was longer than that of the LS group (MD=3.41; 95% CI: 2.39 -4.43; P<0.00001) (Fig. 4). The statistical power of the test was high (1-β=1.0000). No publication bias was found by Begg’s test and Egger’s test (Z=0.00, P=1.000; t=1.36, P=0.232).

Postoperative complications

The included articles all reported the postoperative complications.[8-14]The overall postoperative complication rate was 27.1% (75/277) in the OS group and 21.1% (49/232) in the LS group. Homogeneity was detected among studies (I2=18%; P=0.29), and fixed-effect model meta-analysis showed that there was a less postoperative complication rate in the LS group, but the difference was not significant (RR=1.34; 95% CI: 0.88-2.01; P=0.17) (Fig. 5). The statistical power of the test was high (1-β=0.8562). No publication bias was found by Begg’s test and Egger’s test (Z=1.88, P=0.060; t=1.86, P=0.136).

Changes in blood count and liver function

Three studies[11, 12, 14]focused on the changes in blood count and liver function. Blood count, hemoglobin level (MD=-2.02; 95% CI: -11.63-7.58; P=0.68; test of power: 1-β=0.1074) and white blood cell count (MD=0.95; 95% CI: -0.48-2.38; P=0.19; test of power: 1-β=0.3754) showed no significant difference between the OS and LS groups. Regarding the liver function, ALT (MD=8.52; 95% CI: 0.19-16.85; P=0.05; test of power: 1-β=0.9999) and total bilirubin (MD=5.12; 95% CI: 0.37-9.87; P=0.03; test of power: 1-β=0.9980) levels were significantly lower in the LS group than that in the OS group. However, AST level (MD=12.79; 95% CI: -6.59-32.18; P=0.20; test of power: 1-β=1.0000) was not different between the two groups (Fig. 6). No publication bias was detected in blood count and liver function by Begg’s test and Egger’s test (P>0.05).

Fig. 3. Operative time in OS vs LS.

Fig. 4. Postoperative hospital stay in OS vs LS.

Fig. 5. Overall postoperative complications in OS vs LS.

Fig. 6. Changes in blood count and liver function in OS vs LS. WBC: white blood cell; ALT: alanine aminotransferase; AST: aspartate aminotransferase; TBiL: total bilirubin.

Discussion

The massive bigger spleen is definitely an obstacle to handle while using OS or LS and its manipulation is quite difficult, and this can be traumatic and lead to hemorrhage or rupture during surgery. Various reports[15, 16]have suggested that splenomegaly, not including massive splenomegaly, seems not to modify the rate of conversion from LS to OS. Conflict remains over utilization of LS to treat hypersplenism secondary to liver cirrhosis as a result of some drawback and downsides related to laparoscopic surgery, which includes insufficient tactile sensation feedback, poor understanding of splenic hi-lum, deficit of instant access in the case of unpredicted hemorrhage, and difficulty in locating sufficient working room as a result of splenomegaly.[17]Cirrhotic patients have a high risk of hemorrhage due to the association of perisplenic collateral vessels, significantly less platelet counts, massive splenomegaly, and poor liver function.[18]LS can significantly alleviate hypersplenism and improve Child-Pugh scores.[3, 19, 20]The reason behind liver function betterment in liver cirrhosis after splenectomy continues to be unidentified and needs further research.[19]Intra-operative complications were documented both in LS and OS. Less blood loss, shorter hospital stay, and better liver function preservation were observed in LS compared with OS.[11]Platelet count was substantially higher in the LS group than in the partial splenic embolization group at one or two weeks after the intervention.[1]A study[21]found the feasibility of LS for hypersplenism secondary to liver cirrhosis and portal hypertension, demonstrating that LS is really an attainable, effective, and risk-free surgical treatment for patients who would undergo splenectomy for their disease. Hypersplenism secondary to cirrhosis and portal hypertension should not be regarded as a contraindication for LS. Compared with the OS group, patients with hypersplenism secondary to liver cirrhosis who underwent LS were older, had decreased preoperative hemoglobin levels and leukocyte counts, and poorer Child-Pugh class. The LS group needed additional operative time and experienced more blood loss. Moreover, there were no significant differences in conversion rate, transfusion rate, complication rate and postoperative course.

LS is usually a feasible, achievable, and efficient technique in patients with hypersplenism secondary to liver cirrhosis.[12]Standardization of the LS technique improves safe practice in liver cirrhosis patients with portal hypertension. Extensive isolation and sufficient elevation with the superior pole from the spleen ought to be carried out before the splenic hilar dissection. With all of the current technical standardization, LS becomes an attainable and risk-free technique in the setting of liver cirrhosis and portal hypertension.[18]Tomikawa and coworkers[1]compared LS and partial splenic embolization and found that LS is superior to partial splenic embolization in cirrhotic patients with hypersplenism, especially for those with peripheral cytopenia. The LS group took advantage of much less intraoperative loss of blood along with a reduced postoperative stay in the hospital vs the OS group. Liver function preservation was similar in both groups. This analysis demonstrated that LS can be viewed as a well-disposed procedure with good surgical outcomes weighed against OS.[14]The other study compared the level of immunity in cirrhotic patients with OS and LS, and found that LS induced less surgical stress and had long-term advantages compared with OS. The cellular immune response was less depressed after LS compared with OS.[12]

The small number of clinical trials and patients in comparing the outcomes of LS and OS secondary to liver cirrhosis is the main drawback of this meta-analysis. However, this analysis demonstrated clearly that LS approach favors the fewer postoperative complications and shorter hospitalization compared with OS. Due to the lack of substantial evidence, it is still difficult to make a clear consensus whether laparoscopic approach is better or not for splenectomy in cirrhotic patients. Multiple RCT is necessary to validate our results.

Furthermore, the actual result of LS can not be effectively examined until a randomized study or perhaps a prospective case-control study is carried out. In addition, significant researches are needed to figure out the everlasting outcomes of LS on perioperative inflammatory response and immunological objective of patients with portal hypertension-induced splenomegaly.

In conclusion, LS is a feasible and safe technique for hypersplenism secondary to liver cirrhosis. LS offers better liver function preservation, less blood loss and shorter hospital stay. LS could be chosen over incomplete splenic embolization for cirrhotic patients with hypersplenism. Hypersplenism secondary to cirrhosis and portal hypertension should not be a contraindication for LS.

Contributors: AK wrote the main body of the article and data extraction under the supervision of ZSS. ZSS provided advice on medical aspects. ZSS is the guarantor.

Funding: None.

Ethical approval: Not needed.

Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

References

1 Tomikawa M, Akahoshi T, Sugimachi K, Ikeda Y, Yoshida K, Tanabe Y, et al. Laparoscopic splenectomy may be a superior supportive intervention for cirrhotic patients with hypersplenism. J Gastroenterol Hepatol 2010;25:397-402.

2 Hama T, Takifuji K, Uchiyama K, Tani M, Kawai M, Yamaue H. Laparoscopic splenectomy is a safe and effective procedure for patients with splenomegaly due to portal hypertension. J Hepatobiliary Pancreat Surg 2008;15:304-309.

3 Shimada M, Hashizume M, Shirabe K, Takenaka K, Sugimachi K. A new surgical strategy for cirrhotic patients with hepatocellular carcinoma and hypersplenism. Performing a hepatectomy after a laparoscopic splenectomy. Surg Endosc 2000;14: 127-130.

4 Patel AG, Parker JE, Wallwork B, Kau KB, Donaldson N, Rhodes MR, et al. Massive splenomegaly is associated with sig-nificant morbidity after laparoscopic splenectomy. Ann Surg 2003;238:235-240.

5 Delaitre B. Laparoscopic splenectomy. The “hanged spleen”technique. Surg Endosc 1995;9:528-529.

6 Nicholson IA, Falk GL, Mulligan SC. Laparoscopically assisted massive splenectomy. A preliminary report of the technique of early hilar devascularization. Surg Endosc 1998;12:73-75.

7 Reso A, Brar MS, Church N, Mitchell P, Dixon E, Debru E. Outcome of laparoscopic splenectomy with preoperative splenic artery embolization for massive splenomegaly. Surg Endosc 2010;24:2008-2012.

8 Watanabe Y, Horiuchi A, Yoshida M, Yamamoto Y, Sugishita H, Kumagi T, et al. Significance of laparoscopic splenectomy in patients with hypersplenism. World J Surg 2007;31:549-555.

9 Ando K, Kurokawa T, Nagata H, Arikawa T, Yasuda A, Ito N, et al. Laparoscopic surgery in the management of hypersplenism and esophagogastric varices: our initial experiences. Surg Innov 2012;19:421-427.

10 Wang B, Wu HS, Wang GB, Tao KX. Laparoscopy splenectomy for massive splenomegaly. J Invest Surg 2013;26:154- 157.

11 Zhu JH, Wang YD, Ye ZY, Zhao T, Zhu YW, Xie ZJ, et al. Laparoscopic versus open splenectomy for hypersplenism secondary to liver cirrhosis. Surg Laparosc Endosc Percutan Tech 2009;19:258-262.

12 Cai YQ, Zhou J, Chen XD, Wang YC, Wu Z, Peng B. Laparoscopic splenectomy is an effective and safe intervention for hypersplenism secondary to liver cirrhosis. Surg Endosc 2011;25:3791-3797.

13 Wu Z, Zhou J, Pankaj P, Peng B. Laparoscopic and open splenectomy for splenomegaly secondary to liver cirrhosis: an evaluation of immunity. Surg Endosc 2012;26:3557-3564.

14 Zhou J, Wu Z, Pankaj P, Peng B. Long-term postoperative outcomes of hypersplenism: laparoscopic versus open splenectomy secondary to liver cirrhosis. Surg Endosc 2012;26:3391-3400.

15 Heniford BT, Park A, Walsh RM, Kercher KW, Matthews BD, Frenette G, et al. Laparoscopic splenectomy in patients with normal-sized spleens versus splenomegaly: does size matter? Am Surg 2001;67:854-858.

16 Targarona EM, Espert JJ, Cerdán G, Balagué C, Piulachs J, Sugra?es G, et al. Effect of spleen size on splenectomy outcome. A comparison of open and laparoscopic surgery. Surg Endosc 1999;13:559-562.

17 Wang YD, Ye H, Ye ZY, Zhu YW, Xie ZJ, Zhu JH, et al. Laparoscopic splenectomy and azygoportal disconnection for bleeding varices with hypersplenism. J Laparoendosc Adv Surg Tech A 2008;18:37-41.

18 Kawanaka H, Akahoshi T, Kinjo N, Konishi K, Yoshida D, Anegawa G, et al. Technical standardization of laparoscopic splenectomy harmonized with hand-assisted laparoscopic surgery for patients with liver cirrhosis and hypersplenism. J Hepatobiliary Pancreat Surg 2009;16:749-757.

19 Anegawa G, Kawanaka H, Uehara H, Akahoshi T, Konishi K, Yoshida D, et al. Effect of laparoscopic splenectomy on portal hypertensive gastropathy in cirrhotic patients with portal hypertension. J Gastroenterol Hepatol 2009;24:1554-1558.

20 Elsebae MM, Abu-Zekri NB. A study of the effect of splenectomy on hepatic functional reserve and structural damage in patients with chronic hepatitis C virus infection by non-invasive serum markers. A prospective study. Int J Surg 2008;6:362-366.

21 Wang Y, Zhan X, Zhu Y, Xie Z, Zhu J, Ye Z. Laparoscopic splenectomy in portal hypertension: a single-surgeon 13-year experience. Surg Endosc 2010;24:1164-1169.

Received October 9, 2015

Accepted after revision December 8, 2015

Reviw Article

doi:10.1016/S1499-3872(16)60053-X

Corresponding Author:Shu-Sen Zheng, MD, PhD, FACS, Division of Hepatobiliary Pancreatic Surgery, First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou 310003, China (Tel: +86-571-87236738; Fax: +86-571-87236739; Email: shusenzheng@zju.edu.cn)

Meta-analysis