Background
Vasectomy is performed in two distinct steps: delivering and exposing the vas deferens out of the scrotum (isolation), and occluding the vas. To isolate the vas, the use of the no-scalpel vasectomy (NSV) technique [
1] is increasing among physicians who perform vasectomy in the United States [
2] and in developing countries [
1,
3‐
6]. NSV proponents claim the technique leads to fewer hematomas, less bleeding, fewer infections, shorter operating times, less pain, and an enhanced acceptance of vasectomy [
1]. On the other hand, others believe that NSV does not reduce the risk of surgical complications over the standard incisional approach to expose the vas [
7].
Various surgical approaches to occlude the vas have been recommended over the years. Ligation with suture material and excision of a small vas segment is believed to be the most common method used world-wide [
8]. According to the most recently available data, about 18% of vasectomies performed in the United States are performed by ligation with suture or metal clips [
2,
8,
9]. Although it is the simplest method, it is considered to be the least effective [
10]. Other methods have been recommended to increase the effectiveness of occlusion such as cautery of the vas lumen, interposing fascial tissue between the segments of the severed vas (know as fascial interposition (FI)), folding back of one or both vas segments onto itself, excision of a long vas segment, or a combination of two or more of the preceding techniques. Cautery combined with FI, which is believed by some to be the most effective occlusion technique [
11,
12], has been adopted increasingly in recent years by physicians who perform vasectomy in the United States [
2]. Some authors have suggested leaving the testicular end of the vas unoccluded (known as open-end vasectomy) as a way to reduce the occurrence of post-vasectomy orchi-epidydimitis and painful granuloma [
13‐
15].
Claims of superiority of one occlusion technique over another have been challenged by some authors [
2,
16,
17]. They argue that the quality of data available precludes firm evidence-based conclusions about comparative effectiveness and safety of different vasectomy techniques. In view of both the apparent lack of high quality information and of existing controversies about the best vasectomy surgical techniques, we conducted a systematic review to appraise the available evidence.
The main objective of this systematic review was to assess if any surgical vasectomy technique is associated with better outcomes in terms of occlusive and contraceptive effectiveness, and complications. The following questions were considered:
1.
Is NSV associated with a lower risk of surgical complications compared with the standard incisional technique?
2.
Is any single occlusion method associated with a higher occlusive and/or contraceptive effectiveness compared with any other occlusion method?
3.
Is any single occlusion method associated with a lower risk of complications compared with any other occlusion method?
Methods
Search strategy
We first performed a MEDLINE search (PubMed, 1966-June 2003)[
18] using the terms « Vasectomy » [MESH], "Human", and "Male". We then added the terms "review" and "not review" to sort reviews and original research articles. The search strategy is described in Table
1. A similar search using the same strategy was then done with EMBASE (Excerpta Medica online, 1980-June 2003)[
19]. No language restriction was applied, as at least one member of the research team was fluent in English, French or Spanish. We also looked at additional references in urology textbooks, the authors' personal files, and the references of all original research articles and relevant review articles retrieved.
Table 1
Search strategy on MEDLINE and EMBASE
#1 | "vasectomy" [MESH] |
#2 | Human |
#3 | Male |
#4 | #1 AND #2 AND #3 NOT review |
#5 | #1 AND #2 AND #3 AND review |
Study selection criteria and procedures
Two independent members of the research team (CD and KSt-H) reviewed all titles and available abstracts identified through the MEDLINE search. We identified all articles potentially reporting: 1) a study comparing occlusive effectiveness based on semen analysis (SA), contraceptive effectiveness based on pregnancy, or post-vasectomy complications in two or more groups of men vasectomized with any different surgical techniques, or 2) two or more case series of vasectomized men published by the same author(s) using any different surgical techniques. We excluded: 1) non comparative studies (case series with no control group), 2) studies comparing a vasectomy surgical technique to a non-vasectomy vas occlusion method including intravasal devices or compounds, and 3) studies evaluating long-term post-vasectomy complications such as cancer, cardiovascular disease, and auto-immune disease or non-clinical physiopathological outcomes.
At this step, these a priori inclusion and exclusion criteria were loosely applied to maximize sensitivity over specificity. Reviewers agreed on potential relevance in 92% of the 1,575 titles and abstracts identified (Kappa 0.53, 95% confidence interval (CI) 0.48–0.56). Disagreement was resolved by discussion involving a third reviewer (ML) and we finally identified 201 potentially relevant articles. Among the 473 potentially relevant articles found in EMBASE, only 13 were not also found in MEDLINE. We also identified eight potentially relevant articles through hand-searching references of retrieved articles and two more in personal files.
Following the initial selection process, full reports of 219 of the 224 articles were retrieved; five articles from Chinese journals could not be obtained. Two articles in Danish were translated. Two independent reviewers, using the same inclusion and exclusion criteria reviewed all articles again. Reviewers agreed on the selection of 95% of the 219 articles (Kappa 0.80, 95% CI 0.67–0.83). Disagreement was resolved by discussion involving a third reviewer. We finally included a total of 37 research articles in our review, identified through MEDLINE (n = 29), EMBASE (n = 4), references of retrieved articles (n = 2), and personal files (n = 2).
For each study, two reviewers extracted independently the research design, eligibility criteria, sample size, setting, study period, type and number of surgeons, isolation method, occlusion method, length and method of follow-up, outcome measures (occlusive and contraceptive failure and/or complications), data collection process, and results using a data extraction grid. Discrepancies were discussed and resolved by consensus involving a third reviewer. Studies were unmasked for authors and journals. We attempted to contact the authors of some studies in order to clarify key methodological elements or results.
Quality assessment of studies
Quality assessment of studies was carried out based on study design, sample size, comparability of study groups, effectiveness and complication assessment methodology, and follow-up/compliance rate (Table 2,
Additional file 1). No global quality score was calculated but a level qualifying the methodology (five levels ranging from very low to very high) was attributed to each study based on the criteria described in Table 2. Two independent reviewers assessed the studies. Discrepancies were discussed and resolved by consensus.
Data synthesis
Studies were grouped in three broad categories: isolation, occlusion, and combined isolation and occlusion techniques. Within each category, studies were further divided into sub-categories according to specific surgical techniques, and evidence tables were created (Tables 3 to 30.
Additional file 2 to
8). Only qualitative synthesis was performed within each sub-category because of heterogeneity between the studies in study design, intervention and outcome assessment. Occlusive and contraceptive effectiveness are presented as the proportion (risk) of vasectomized men reported to have a failure either based on SA (occlusive failure) or pregnancy (contraceptive failure). Within each sub-category, qualitative sensitivity analyses were performed according to these variables, and quality assessment of the methodology was carried out.
Discussion
The aim of this systematic review was to identify the best vasectomy surgical techniques in terms of effectiveness and safety. We identified many studies comparing various vasectomy surgical techniques and we divided them into sub-categories based on the technique used. For many reasons, we were unable to perform a meta-analysis of the results and it was even difficult to compare the findings in a qualitative synthesis. Firstly, many sub-categories included only a few studies and six included only one study. Secondly, there was much heterogeneity among the studies in each sub-category with respect to setting and population studied. Most importantly, there was too much heterogeneity with regard to study design, specific surgical technique performed, and outcome assessment.
Although all studies analyzed were comparative, we did not limit our review to experimental prospective studies (trials) in order to provide a comprehensive overview of the literature on vasectomy surgical techniques. Furthermore, based on our prior knowledge of the literature, we suspected that there would be very few trials on vasectomy surgical techniques. Indeed, we found only four RCTs, with no more than two evaluating the same technique, namely NSV. In most sub-categories, there was much heterogeneity in study design with the majority of the reports retrieved being retrospective observational studies.
Within most sub-categories, the surgery performed was a mix of various technical components resulting in significant variation among the studies. For example, folding back (FB) was performed with either clips or sutures on the prostatic or the testicular end. Cautery was performed either with or without fascial interposition (FI) on the prostatic or the testicular end, and various lengths of vas segment were excised or cauterized either by electro- or thermal cautery, in addition to leaving or not leaving the testicular end open. These variations make it difficult to isolate the contribution to overall effectiveness and safety of a specific component of the vasectomy technique.
Evaluation of vasectomy effectiveness was based on the results of semen analysis (SA) in all studies. Data on actual contraceptive effectiveness based on pregnancy results were very scarce. Although pregnancy data would be the best endpoint to evaluate vasectomy effectiveness, SA is a good surrogate measure for pregnancy risk. Pregnancies are very rarely reported when SA shows azoospermia or only rare non motile sperm [
50‐
53]. Very few studies specified the operational criteria used to define occlusive failure (or success) namely the number of SA, the sperm count cut-off, the motility status, and the time interval between vasectomy and last SA to establish failure (or success). The laboratory methodology and criteria used when performing SA were also rarely mentioned. It is clear from the few studies where operational definitions of occlusive effectiveness were included, that the criteria used differ from one study to the other. Nevertheless, results on occlusive effectiveness appear to be much more valid than those on safety outcomes. In most studies, assessment of post-vasectomy complications was based on unsystematic, self-referred, unblinded medical consultations with no objective criteria and no timing and length of follow-up specified. All the preceding pitfalls contribute to the overall low methodological quality of most currently available comparative studies on vasectomy occlusion techniques.
Our review has some limitations. Firstly, as mentioned earlier, most studies analysed were retrospective observational studies including case series published by the same author(s) using different techniques. Results based on case series compared with historical controls – either reported in the same or in different papers – must be interpreted with caution. In such studies, increased experience of surgeons and variations in patients' characteristics, data collection process, and outcome assessment over time are sources of potential bias. This limitation adds to the fact that in many studies the sample size was small and there was a lack of detailed information on several key methodological elements.
Secondly, it is possible that despite our rigorous process of searching and selecting articles, we have overlooked some relevant comparative studies. However, for many years, two of us (ML and MAB) have been independently surveying the literature for vasectomy studies and our own database reviewed after performing MEDLINE and EMBASE searches revealed only two studies [
36,
45], one not yet published at the time the computerized searches were performed. On the other hand, some potentially relevant articles from Chinese journals could not be retrieved despite many attempts, although none appeared to be a RCT based on the title or abstract in English when available. Although we did not formally ascertain publication bias, it is unlikely that a large RCT on a specific technique would have been unpublished.
Thirdly, the sets of criteria and scales we used to extract the data from the studies and to assess the quality of the methodology were not validated. However, our quality assessment criteria were largely based on the CONSORT guidelines for quality assessment of published RCT [
54], they were established before the review was initiated, and were applied by independent reviewers.
Our aim was to answer three questions based on the best evidence from published reports keeping in mind the methodological limitations of the available studies.
Is NSV associated with a lower risk of surgical complications compared with the standard incisional technique?
We found good evidence that NSV is associated with a clinically significant lower risk of surgical complications, namely hematoma/bleeding and infection, compared with approaches involving one or two sutured scrotal incisions. NSV should be the preferred approach to isolate and expose the vas when performing vasectomy.
Is any occlusion method more effective in terms of occlusion and contraception compared with any other occlusion method?
Among the various vas occlusion methods reviewed, there is no evidence that, when a small vas segment is excised, ligating the vas with metal clips or folding back and suturing a vas segment over itself results in higher occlusive effectiveness than simply ligating the vas with suture material. On the other hand, there is good evidence, mainly based on a single high quality RCT, that FI increases vasectomy occlusive effectiveness when ligation with suture material and excision of a small vas segment is performed.
Evidence from the comparative studies available on cautery is not conclusive but data suggest strongly that cautery combined with FI provides the highest level of occlusive effectiveness. Even with the large heterogeneity between studies, the risks of occlusive failure reported with cautery are systematically much lower than those reported with any other occlusion methods. The higher failure risk with cautery observed in the single moderate-quality study [
32] may result from putting silk sutures on the cauterized vas segments. Suturing the vas over a cauterized segment may cause necrosis and sloughing off of the segment distal to the suture and shortening of the length of the intraluminal scar. There are insufficient data to draw conclusions about the use of thermal cautery versus electro-cautery. Apart from the single comparative study included in this review, which did not show a statistically significant difference between the techniques [
32], we are aware of only one other small study comparing both cautery methods [
55]. The authors found better sealing of the vas after thermal cautery based on histologic studies showing fewer cases of vasitis nodasa and spermatic granuloma afterwards.
Other recent study results are consistent with our findings on cautery. In a very large non-comparative case series of 45,000 vasectomies performed with electro-cautery without FI at the Marie Stopes vasectomy centers in the UK, the observed occlusive failure risk was 0.6%, a figure comparable to that reported in most of the major comparative studies available, and the risk of contraceptive failure was 0.03% [
56]. Family Health International, NC, USA and EngenderHealth have just completed a prospective observational study of vasectomy performed with cautery (D Sokal, personal communication). When occlusive failure was defined as >10 million sperm/mL at 12 weeks the risk of early failures was 4/389 (1.0%). Applying the same definition of failure to the RCT on FI conducted by the same organizations [
45], early occlusive failure risks with ligation and excision were 4.9% and 12.5% in the groups with and without FI, respectively.
The available data are insufficient to draw conclusions about the actual contraceptive effectiveness of any occlusion method over another. Two relevant large descriptive retrospective observational studies have been published recently. A study involving 1,052 men in Nepal showed that within three years after vasectomy 4.2% had been responsible for an unplanned pregnancy; most vasectomies were performed with ligation with suture material and excision of a small vas segment [
57]. In a similar study conducted in China among 1,555 couples using vasectomy as a contraceptive method, the risk of an unplanned pregnancy varied between 4.2% after one year to 9.5% after five years [
58]. There were no details on the occlusion technique used in this study.
Is any occlusion method associated with a lower risk of complications compared with any other occlusion method?
The evidence is lacking to support any one occlusion technique over others in terms of decreased risk of complications. The incidence of surgery related complications such as bleeding and/or infections does not appear to be influenced by any specific occlusion technique. Leaving the testicular end open has been hypothesized to be associated with less increase in the epidydimal pressure that may eventually lead to vas blow-outs and secondary obstruction of the epididymis, reducing the chances of vaso-vasostomy success [
59,
60]. Furthermore this approach was promoted to decrease the occurrence of post-vasectomy pain syndrome namely congestive orchi-epididymitis and painful granuloma [
13‐
15]. Our review of comparative studies revealed insufficient evidence to prove that this is the case. However, occlusive effectiveness does not seems to be negatively affected by leaving the testicular end open when cautery and FI are combined to occlude the prostatic end of the vas.
Acknowledgements
We wish to thank Mrs Lynn Dunikowski for her invaluable help in searching the medical literature. We also thank Drs David Sokal, Eric Schaff, Eric Smith, and Max Yarowsky who kindly reviewed the article. Finally we thank all the authors of the retrieved papers who kindly shared information with us. Partial support for this study was provided by EngenderHealth, New York, NY, with funds from the US Agency for International Development (USAID), Cooperative Agreement # HRN-A-00-98-00042-00 although the views expressed in this article are those of the authors and do not necessarily reflect those of EngenderHealth or USAID.
Competing interests
Two of the authors (ML and MB) are also authors of some of the articles in this review.
Authors' contributions
ML conceived and designed the study, participated in data collection and analysis, and drafted the manuscript. CD designed the study, coordinated data collection and analysis, and drafted the manuscript. MB conceived of the study, participated in data analysis and drafted the manuscript. KSt-H participated in data collection and analysis. All authors read and approved the final manuscript.