Efficacy and safety of varicocelectomies: A meta-analysis

ABSTRACT

This study reviewed the efficacy and safety of the three surgical approaches for varicocele (microsurgical, laparoscopic, and open varicocelectomy). A systematic review of the relevant randomized clinical trials was performed. Trials were identified from specialized trials register of the Cochrane UGDP Group, the Cochrane library, additional electronic searches (mainly MEDLINE, EMBSAE, SCI, CBM), and handsearching. Clinical trials comparing microsurgical, laparoscopic and open varicocelectomies were included. Statistical analysis was managed using Review Manager 5.3. Seven clinical trials of 1,781 patients were included. The meta-analysis indicated that compared with open varicocelectomy, microsurgery had a higher pregnancy rate (p=0.002), while there was nonsignificant difference between microsurgical and laparoscopic varicocelectomies or between laparoscopic and open varicocelectomies. Both microsurgical and laparoscopic varicocelectomies had a greater increase in postoperative sperm concentration than open varicocelectomy (p=0.008 and p=0.001, respectively). Microsurgical varicocelectomy also showed better improvement in postoperative sperm motility (p=0.02). Compared with the other two, microsurgical varicocelectomy had the longest operative time (p=0.01 and p=0.0004 respectively). A nonsignificant difference was found in the hospital stay between the three approaches, whereas microsurgical and laparoscopic varicocelectomies had a shorter time to return to work. Moreover, microsurgical varicocelectomy had a lower incidence of postoperative complications and recurrence compared with the others. Analysis of current evidence shows that microsurgical varicocelectomy has a longer operative time, lower incidence of postoperative complications, and recurrence than laparoscopic and open varicocelectomies, and shows a higher pregnancy rate, with a greater increase in postoperative sperm concentration, better improvement in postoperative sperm motility, and shorter time to return to work than open varicocelectomy.

KEYWORDS:

• Laparoscopy
• meta-analysis
• microsurgery
• varicocele
• varicocelectomy

Introduction

The incidence of varicocele is 15-20% in the normal male population and 30-40% in patients with male infertility [Jarow 2001]. It is defined as dilated and tortuous veins within the pampiniform plexus of scrotal veins. Varicoceles have been recognized as the most prevalent cause of male infertility [Dubin and Amelar 1971; WHO 1992]. Recent studies on the physiopathology of varicocele-related infertility have shown that, due to blood stasis and accumulation of reactive oxygen species (ROS), nitric oxide, and other toxins, spermatogenesis is impaired, which eventually leads to dysfunction and morphological abnormalities of sperm, affecting male fertility [Miyaoka and Esteves 2012].

Among the several common treatments for varicocele, varicocelectomy-microsurgical, laparoscopic, and open techniques are used. Different surgical methods have their advantages and disadvantages with controversies regarding fertility improvement. The findings of Jin et al. [2009] showed that, compared with traditional Palomo surgery, microsurgical ligation had better postoperative semen parameters, a higher natural pregnancy rate, lower incidence of recurrence and hydrocele, and longer operative time. Long-term efficacy requires further verification by high-quality randomized controlled trials (RCTs) which consist of large samples. The meta-analysis of Ding et al. [2012] found that microsurgical varicocelectomy was the most effective and least morbid method among the three techniques for treating varicocele in infertile men. Our study aimed to focus on the efficacy and safety of three different treatments of varicocele, by means of evaluation of pregnancy rate, postoperative sperm concentration and motility, operation time, time to return to work, hospital stay, and incidence of postoperative complications and recurrence, for the purpose of providing evidence-based clinical treatment.

Results

Studies evaluated

A total of 217 indexes were retrieved that included seven RCTs comprising 1,781 patients in the analysis [Cayan et al. 2000; Ghanem et al. 2004; Al-Kandari et al. 2007; Al-Said et al. 2008; Abdel-Maguid et al. 2010; Xie et al. 2011; Qin et al. 2015]. The literature search strategy that was employed is shown in Figure 1. The majority of excluded literature were not RCTs comparing the three different treatments. No additional documents from references within this literature qualified for inclusion in this analysis. Quality assessment of the seven trials were performed by an RCT quality evaluation form. The objectives of all seven studies were varicocele, without significant difference in baseline, such as age, varicocele severity, and varicocele side. All confounding factors of each study were properly identified with appropriate statistical analysis. Due to the particularity, the seven RCTs were not a blinded study (preoperative surgical consent form signed by the patient had informed the operation mode). The baseline and quality of RCTs met the inclusion criteria are summarized in Table 1.

Table 1. Characteristics and quality assessment of the studies included in the meta-analysis.

Studies	Treatment	Sample size (n)	Mean age (years)	Varicocele grade (I/II/III)	Varicocele side (left/both)	Jadad score
Cayan et al. 2000	Microsurgical	236	29.56	136/143/62	128/108	3
	open	232	28.89	115/148/59	142/90
Ghanem et al. 2004	Microsurgical	304	36.7	96/106/102	210/94	3
	open	109	35.8	37/42/30	77/32
Al-Kandari et al. 2007	Microsurgical	40	29	*	35/5	2
	laparoscopic	40	30		30/10
	open	40	28.5		28/12
Al-Said et al. 2008	Microsurgical	112	34	*	69/43	3
	laparoscopic	94	33		40/54
	open	92	34		41/51
Abdel-Maguid and Othman 2010	Microsurgical	82	34	25/52/38	49/33	3
	open	80	33.7	34/47/23	46/34
Xie et al. 2011	Microsurgical	70	25.5	0/39/31	58/12	3
	laparoscopic	70	24.2	0/41/29	51/19
Qin et al. 2015	Microsurgical	60	*	*	*	2
	laparoscopic	48
	open	72

*No specific figures but without significant difference.

Figure 1. Flow diagram of literature search. 217 indexes were retrieved from all databases (the Cochrane Library, PubMed, Embase, ISI–SCI, CBMdisc). 16 were left after articles screened on basis of title and abstract. A total of seven randomized controlled trials (RCTs) comprising 1,781 patients were included in the analysis. Main reasons of exlusion were: trials were not comparison of three surgical approaches, not RCTs or did not have similar endpoints.

Pregnancy rate and sperm quality

Based on heterogeneity analysis using a p > 0.05, fixed effects model analysis (Figure 2) showed that at the end of follow-up, pregnancy rate of microsurgical varicocelectomy was higher than that of the open group (RR = 1.37, 95% CI:1.12-1.69, p = 0.002), while a nonsignificant difference was found between the microsurgical group and the laparoscopic group (RR = 1.14, 95% CI: 0.91-1.42, p = 0 25) or between the laparoscopic group and the open group (RR = 1.09, 95% CI: 0.76-1.56, p = 0.66).

Figure 2. Pooled results of pregnancy rates after microsurgical, laparoscopic, and open varicocelectomies. Pregnancy rate of microsurgical varicocelectomy was higher than that of the open group (RR = 1.37, 95% CI: 1.12-1.69, p = 0.002), while nonsignificant difference was found between the microsurgical group and the laparoscopic group (RR = 1.14, 95% CI: 0.91-1.42, p = 0 25) or between the laparoscopic group and the open group (RR = 1.09, 95% CI:. 0.76-1.56, p = 0.66).

Heterogeneity of postoperative sperm concentration (million mL⁻¹) displayed p < 0.05. After using a random effects model analysis, the results showed that the microsurgical and laparoscopic groups postoperative sperm concentration was elevated compared to the open group (mean difference, MD = 5.59, 95% CI: 1.46-9.72, p = 0.008 and MD = 5.10, 95% CI: 1.29-5.13, p = 0.001, respectively). There was nonsignificant difference between the microsurgical and laparoscopic groups (MD = 3.21, 95% CI): -0.78 to 10.98, p = 0.09) (Figure 3).

Figure 3. Pooled results of sperm concentration (million mL⁻¹) after microsurgical, laparoscopic and open varicocelectomies. The microsurgical and laparoscopic groups both had more increases in postoperative sperm concentration than the open group (mean difference, MD = 5.59, 95% CI: 1.46-9.72, p = 0.008 and MD = 5.10, 95% CI: 1.29-5.13, p = 0.001, respectively). There was nonsignificant difference between the microsurgical and laparoscopic groups (MD = 3.21, 95% CI): -0.78 to 10.98, p = 0.09).

Based on postoperative sperm motility (%) heterogeneity analysis p < 0.05, a random effects model revealed that the microsurgical group had a better improvement in postoperative sperm motility than the open group (MD = 5.56, 95% CI: 0.77-10.36, p = 0.02), and a nonsignificant difference was found between the microsurgical group and laparoscopic group (MD = 2.09, 95% CI: -3.78 to 7.97, p = 0.49) and between the laparoscopic group and open group (MD = 1.89, 95% CI: -2.02 to 5.81, p = 0.34) (Figure 4).

Figure 4. Pooled results of sperm motility (%) after microsurgical, laparoscopic, and open varicocelectomies. Microsurgical group had better improvement in postoperative sperm motility than the open group (MD = 5.56, 95% CI: 0.77-10.36, p = 0.02), and nonsignificant difference was found between the microsurgical group and laparoscopic group (MD = 2.09, 95% CI: -3.78 to 7.97, p = 0.49) and between the laparoscopic group and open group (MD = 1.89, 95% CI: -2.02 to 5.81, p = 0.34).

Operative time

There were six studies comparing the operative time. Based on heterogeneity analysis of the six studies p < 0.05 a random effects model was employed. The duration of microsurgical varicocelectomy was longer than that of laparoscopic and open varicocelectomy (MD = 23.47, 95% CI: 4.81-42.13, p = 0.01 and MD = 25.18, 95% CI: 11.13-39.22, p = 0.0004, respectively). The difference between the laparoscopic group and open group was not statistically significant (MD = -0.62, 95% CI: -15.56 to 14.33, p = 0.94) (Figure 5).

Figure 5. Pooled results of operative time (minutes) after microsurgical, laparoscopic, and open varicocelectomies. The duration of microsurgical varicocelectomy was longer than that of laparoscopic and open varicocelectomy (MD = 23.47, 95% CI: 4.81-42.13, p = 0.01 and MD = 25.18, 95% CI: 11.13-39.22, p = 0.0004, respectively). The difference between the laparoscopic group and open group was not statistically significant (MD = -0.62, 95% CI: -15.56 to 14.33, p = 0.94).

Hospital stay and time to return to work

Heterogeneity analysis of three studies that compared the hospital stay was p < 0.05 and the random effects model indicated that there was a nonsignificant difference between the three groups (microsurgical vs. laparoscopic: MD = -0.41, 95% CI: -0.94 to 0.12, p = 0.13, microsurgical vs. open: MD = -1.01, 95% CI: -2.79 to 0.76, p = 0.26 and laparoscopic vs. open: MD = -0.63, 95% CI: -1.44 to 0.19, p = 0.13) (Figure 6). However, there were differences in time to return to work. A fixed effects model was used to analyze time to return to work since heterogeneity analysis revealed a p > 0.05. Results of the analysis displayed that the time to return to work of both the microsurgical group and the laparoscopic group was significantly shorter than that of the open group (MD = -1.96, 95% CI: -2.08 to -1.84, p <0.00001 and MD = -0.97, 95% CI: -1.60 to -0.34, p = 0.002, respectively). While the average time to return to work of the microsurgical group was shorter than that of the laparoscopic group, there was no statistical difference (MD = -0.21, 95% CI: -0.88 to 0.45, p = 0.53) (Figure 7).

Figure 6. Pooled results of hospital stay (days) after microsurgical, laparoscopic, and open varicocelectomies. There was nonsignificant difference between the three groups (microsurgical vs. laparoscopic: MD = -0.41, 95% CI: -0.94 to 0.12,; = 0.13, microsurgical vs. open: MD = -1.01, 95% CI: -2.79 to 0.76, p = 0.26, and laparoscopic vs. open: MD = -0.63, 95% CI: -1.44 to 0.19, p = 0.13).

Figure 7. Pooled results of time to return to work (days) after microsurgical, laparoscopic, and open varicocelectomies. The time to return to work of both microsurgical group and laparoscopic group was significantly shorter than that of the open group (MD = -1.96, 95% CI: -2.08 to -1.84, p <0.00001 and MD = -0.97, 95% CI: -1.60 to -0.34, p = 0.002 respectively). While the average time to return to work of the microsurgical group was shorter than that of the laparoscopic group, there was no statistical difference (MD = -0.21, 95% CI: -0.88 to 0.45, p = 0.53).

The incidence of postoperative complications

Heterogeneity of studies showed small differences, so a fixed effects model analysis was used. The results showed that the microsurgical varicocelectomy had a significantly lower incidence of postoperative complications than the laparoscopic group and open group (OR = 0.22, 95% CI: 0.11-0.46, p <0.0001 and OR = 0.21, 95% CI: 0.12-0.37, p <0.00001, respectively). Between the laparoscopic group and open group, no significant difference was found (OR = 1.17, 95% CI: 0.68-2.03, P = 0.57) (Figure 8).

Figure 8. Comparison of incidence of postoperative complications between the three groups. The microsurgical varicocelectomy had significantly lower incidence of postoperative complications than the laparoscopic group and open group (OR = 0.22, 95% CI: 0.11-0.46, p <0.0001 and OR = 0.21, 95% CI: 0.12-0.37, p <0.00001, respectively). Between the laparoscopic group and open group, nonsignificant difference was found (OR = 1.17, 95% CI: 0.68-2.03, p = 0.57).

The recurrence rate

A total of seven studies comparing recurrence rate differences of the three surgical techniques had shown their heterogeneity was small (p > 0.05). After the fixed effects model analysis, recurrence rate of microsurgical group was significantly lower than that of the laparoscopic group and open group (OR = 0.17, 95% CI: 0.07-0.37, p < 0.00001 and OR = 0.11, 95% CI: 0.06-0.21, p < 0.00001, respectively). There was no statistical difference between the laparoscopic group and open group (OR = 1.39, 95% CI: 0.82-2.37, p = 0.22) (Figure 9).

Figure 9. Comparison of recurrence rate between the three groups. Recurrence rate of microsurgical group was significantly lower than that of the laparoscopic group and open group (OR = 0.17, 95% CI: 0.07-0.37, p <0.00001 and OR = 0.11, 95% CI: 0.06-0.21, p <0.00001, respectively). There was no statistical difference between the laparoscopic group and open group (OR = 1.39, 95% CI: 0.82-2.37, p = 0.22).

Discussion

The goal of varicocele treatment is to eliminate varicocele and reverse damage of venous stasis to spermatogenesis, while minimizing recurrence and complications. With the application of endoscopic techniques and the development of minimally invasive urological surgery, the treatment of varicocele and clinical efficacy have been further improved. A series of evidence-based systematic reviews have shown that laparoscopic varicocele ligation had obvious advantages in reducing postoperative complications, recurrence, and improving the pregnancy rate over the traditional open surgeries [Borruto et al. 2010]. In recent years, with the increase of aesthetic requirements and the rapid development of microsurgery, comparative studies on microsurgical and laparoscopic surgeries have gradually increased. Cayan et al. concluded in their systematic review that the microsurgical varicocelectomy technique had higher spontaneous pregnancy rates and lower postoperative recurrence and hydrocele formation than conventional varicocelectomy techniques in infertile men [Cayan et al. 2009]. The influence of varicocele on the patient primarily reduces testicular spermatogenesis further affecting reproductive function. We found that microsurgical varicocelectomy had a higher pregnancy rate than the open group. There was no statistical difference between the microsurgical group and laparoscopic group, nor between the laparoscopic group and open group. In comparison of postoperative sperm concentration, the microsurgical group showed an increase more than the open group along with the laparoscopic group. There was no significant difference between the microsurgical group and the laparoscopic group. Analysis of postoperative sperm motility displayed a better improvement of the microsurgical group than the open group and no significant difference between the microsurgical group and the laparoscopic group or between the laparoscopic group and the open group was observed. These results must be considered in light of the small amount of literature available to evaluate. A bias of archive, publication, and inclusion criteria affecting the results may exist.

The mission of the male gamete is not only to fertilize the oocyte but also to transfer its information to the oocyte. This is essential to early embryonic development and embryonic capacity for implantation [Dimitriadis et al 2009a; Dimitriadis et al 2009b; Tsounapi et al 2016]. It would be of clinical significance to compare the live birth rates additionally to the pregnancy rates among the three surgical approaches. However, this could not be carried out for the lack of original data .

Our study also found that operative time of microsurgical varicocelectomy was longer than that of the laparoscopic group and the open group. In addition, the time to return to work of the microsurgical group and the laparoscopic group was significantly shorter than the open group. While the average time to return to work of the microsurgical group was shorter than that of the laparoscopic group, no statistically significant difference was found between the two groups.

In our study some results (comparisons of sperm concentration, sperm motility, operative time, hospital stay) of meta-analyses were biased by high heterogeneity. The main reasons were: i) the long span of the included studies (2000-2015), there may be difference in experience (including experience of surgical techniques and usage of microscope), and levels among the surgeons; ii) surgical improvement, such as whether to retain the spermatic artery, may also affect the operative time. Also, hospital stay was different as a result of different discharge standards. Owing to great heterogeneity, we used the random effects model.

Sensitivity analyses for some comparisons (sperm concentration, sperm motility, operative time) was afforded by excluding data from trials with greatly different conclusions. This significantly reduced heterogeneity p > 0.05. However, sensitivity analysis and subgroup analysis could not be performed in the remaining comparisons, because the sample size was too small.

The incision of open varicocelectomy was partly subinguinal [Ghanem et al. 2004 and Abdel-Maguid and Othman 2010] and partly inguinal (the other trials) and that of microscopic varicocelectomy was subinguinal. Opening of the aponeurosis of the external oblique abdominal muscle might contribute to a longer hospital stay and longer time until returning to work. A separate analysis that excluded the data from trials of Ghanem et al. [2004] and Abdel-Maguid and Othman [2010] was undertaken and the result was the same. The microsurgical group had longer operative times.

Incidence of postoperative complications and recurrence is an important parameter of effect and provides an index to compare surgical efficacy. Baazeem et al. [2011] considered the best surgical approach should meet the lowest possible recurrence rate and complication rate. Sofikitisa et al. [2014] found that an advantage of the microsurgical approach was the maintenance of testicular lymph vessels. This alleviated the testicular extracellular edema, and subsequently improved testicular arterial nutrition. In our study the trials of Cayan et al., Ghanem et al. [2004], Al-Kandari et al. [2007], and Al-Said et al. [2008] compared only the incidence of hydrocele after surgical approaches, while the rest compared also scrotal edema, wound infection, and testicular atrophy. We found that among the included studies these two indicators had small differences and the results showed that the incidence of complications and recurrent varicocele was significantly lower after microsurgical than after laparoscopic or open varicocelectomy, which indicated microsurgery was relatively safer.

Laparoendoscopic single-site surgery (LESS), natural orifice transluminal endoscopic surgery (NOTES), robot-assisted laparoscopic surgery (RALS), and other new technologies are emerging. Lee et al.’s [2012] study showed that using LESS to treat varicocele could reduce postoperative pain scores and increase patient satisfaction. Hidalgo-Tamola et al. [2009] also reported RALS was feasible in varicocele treatment with fewer complications. But there is no comparison between the two surgical techniques and the three surgeries in our study.

In summary, analysis of the current evidence shows that microsurgical varicocelectomy has a longer operative time, lower incidence of postoperative complications, and recurrence than laparoscopic and open varicocelectomies. Microsurgery also shows a higher pregnancy rate, a greater increase in post-operative sperm concentration, better improvement in post-operative sperm motility, and shorter time to return to work than open varicocelectomy. Nevertheless, additional standardized and high-quality randomized controlled trials are required to confirm these results.

Materials and methods

Inclusion criteria

The inclusion criteria were fourfold. 1) Original studies on RCTs of surgical treatment of varicocele, with good study design, regardless of whether allocation concealment or blinding; 2) Diagnosed with left or bilateral varicocele by confirmation of clinical or ultrasound examination, subjects accepted varicocelectomies; 3) RCTs testing comparison between microsurgical, laparoscopic, and open varicocelectomies; and 4) Endpoints included pregnancy rate, improvement of postoperative sperm concentration and motility, operative time, time to return to work, hospital stay, incidence of postoperative complications, and recurrence.

Document retrieval

We searched the Cochrane Library (2016 issue 6), Medline database (PubMed) (1966-June 2016), Embase (1974-June 2016), the Institute for Scientific Information (ISI) – Science Citation Index (1955 –June 2016), and Chinese Biomedical Literature Database (CBMdisc) (1978-June 2016), with the search terms “varicocele” and “surgery” to identify all relevant RCTs on the subjects without language restriction. References of the retrieved documents were checked to expand search.

Data collection and analysis

Indexes were retrieved by two reviewers independently to collect information in accordance with the above policy. Then we filtered the data in strict accordance with the inclusion criteria. In case of inconsistency, re-checking and discussion or application for third-party arbitration were used. The acquisition of original literature was primarily through Sichuan University Library and downloaded from the Internet. Designed form was applied for data extraction.

The methodological quality of included RCTs were assessed using the Jadad quality scoring method [Jadad et al. 1996]. With scores ranging from 1 to 5 points, 3 points or more were considered high-quality research and l to 2 points were considered low-quality research.

Statistical analysis

The collected data were analyzed using qualitative and quantitative methods. Qualitative analysis used the describing method, to analyze the differences in baseline, stringency of research methods between studies. Quantitative analysis was completed by meta-analysis software RevMan5.3 version (downloaded from the Cochrane Library). First, the data were extracted for test of heterogeneity. Heterogeneity existing, random effects model was used and vice versa fixed effects model. Results of analysis were quantized by odds ratio (OR), risk ratio (RR), 95% confidence intervals (95% CI) and P values. P values ​​less than 0.05 were considered statistically significant.

Ethics committee statement

This study is a secondary analysis of the published literature, not related to specific patients and experimental animals, and has been approved by the Ethics Committee of Sichuan University.

Declaration of interest

The authors report no declarations of interest.

Acknowledgments

We acknowledge the work of the staff of Sichuan University Library.

Additional information

Notes on contributors

Renbin Yuan

Proposed the conception of the idea and guided the study: WQ; Performed literature search and review, meta-analysis, and interpretation of results: YRB, CDH; Drafted the manuscript: YRB. All authors approved the final version.

Hui Zhuo

Proposed the conception of the idea and guided the study: WQ; Performed literature search and review, meta-analysis, and interpretation of results: YRB, CDH; Drafted the manuscript: YRB. All authors approved the final version.

Dehong Cao

Proposed the conception of the idea and guided the study: WQ; Performed literature search and review, meta-analysis, and interpretation of results: YRB, CDH; Drafted the manuscript: YRB. All authors approved the final version.

Qiang Wei

Proposed the conception of the idea and guided the study: WQ; Performed literature search and review, meta-analysis, and interpretation of results: YRB, CDH; Drafted the manuscript: YRB. All authors approved the final version.