4.1 Introduction

Operative sperm retrieval from the epididymis or testis is performed to retrieve sperm for intracytoplasmic sperm injection (ICSI) in men who have no sperm in their ejaculate. It is indicated in men with obstructive azoospermia (OA) when the obstruction is not correctable, or if the couple prefer assisted reproduction over reconstructive surgery due to social, personal, or medical reasons. It is also performed in men with nonobstructive azoospermia (NOA), since some of these men will have focal spermatogenesis in the testes, though the number of sperm in the testes is too few to appear in the ejaculate. Testicular sperm may also be used in nonazoospermic men when there is total necrozoospermia, high DNA fragmentation, cryptozoospermia, or severe oligozoospermia with recurrent in vitro fertilization (IVF) failure [1,2].

Surgical sperm retrieval has a long history of innovative procedures. Way back in 1955, delivering the Hunterian Lecture at the Royal College of Surgeons, Hanley described tucking a loosely folded ball of amnion under the epididymal tunica to create an artificial spermatocele and was able to recover sperm in 6 of 11 cases [3].

In the 1970s, pioneering work in creating artificial spermatoceles for sperm recovery was done by Schoysman, Wagenknecht, and Kelâmi [4]. In 1980, Cruz reported the first birth following cervical insemination with sperm recovered from a synthetic spermatocele [5], followed by another success reported by Kelâmi in 1981 [6].

However, these were anecdotal successes, and it was only after the development of IVF, and specifically ICSI, that the use of epididymal and testicular sperm became highly successful and a variety of surgical techniques were developed to recover sperm in a range of clinical situations.

4.2 Epididymal Sperm Retrieval

The first successful use of epididymal sperm for IVF was reported by Temple-Smith et al. in 1985. The patient was a man who had two failed attempts at vasectomy reversal. Using microsurgical techniques, the epididymal tubule was opened at three locations and the effluxing fluid was aspirated with a Medicut micropipette. A final concentration of 4.28 million sperm with 61 percent motility was achieved, and one of five eggs fertilized after incubation, resulting in a viable pregnancy [7].

This was soon followed by a report by Silber et al. [8] on the use of microsurgical epididymal sperm aspiration (MESA) in men with congenital vas aplasia. In his initial report, the sperm were used for IVF and fertilization rates were low. It was only after the advent of ICSI that MESA-ICSI became a very successful treatment for OA and Silber et al. reported an increase of pregnancy rates in their group of CBAVD men, from 4.5 percent with conventional IVF to 47 percent with ICSI [9].

With the increasing success and popularity of MESA-ICSI in the treatment of OA, attempts were made to simplify epididymal sperm retrieval. In 1994, at the Tenth Annual Meeting of ESHRE, Craft and colleagues presented their experience with percutaneous epididymal sperm aspiration (PESA)[10]. The procedure was quick, simple, and did not need an operating microscope. Their subsequent papers [11,12] emphasized their high success rate with the technique and PESA gained popularity.

However, many surgeons continued to prefer MESA, claiming that many more motile sperm could be retrieved and cryopreserved for multiple cycles. Accordingly, Shah [13] proposed open non-microsurgical epididymal sperm aspiration (OESA), which was a modified version of MESA that was quick, simple, did not need an operating microscope, and allowed collection of a large number of sperm under vision by puncturing the exposed epididymis with a needle and aspirating the effluxing fluid.

4.3 Testicular Sperm Retrieval

In 1993, Craft [14] reported that even testicular sperm could fertilize oocytes by IVF, but the first viable pregnancy was achieved only by the use of ICSI by Schyosman et al. [15], who performed open testicular biopsies to retrieve sperm from men with OA in whom no sperm could be retrieved from the epididymis. They used these sperm for subzonal insemination (SUZI) and ICSI, and achieved one viable pregnancy. Devroey et al. [16] confirmed that testicular sperm from men with OA were highly successful in producing viable pregnancies by ICSI. Nagy et al. [17] showed that equivalent fertilization rates could be achieved whether fresh or cryopreserved testicular sperm were used. All these preliminary studies used open, conventional testicular biopsies.

Once the fertilizing capability of sperm from normal testes was established, clinicians tried to extract and use sperm from men with testicular failure. The presence of focal areas of sperm production in testicular failure was known from earlier histological studies [18]. In 1995, Devroey et al. published their initial work on men with testicular failure and reported finding sperm by testicular sperm extraction (TESE) in 13 of 15 men [19]. Fertilization rate was 47.8 percent, and there were three ongoing pregnancies from 12 transfers. Yemini et al. [20] reported sperm in a man with tubular atrophy, while Gil-Salom et al. [21] found sperm in a man with very high FSH and a testicular histology showing Sertoli-cell-only syndrome (SCOS). In all these initial cases, testicular sperm were retrieved by conventional open biopsy, and usually multiple biopsies were required. Correlating testicular histology with sperm retrieval, Tournaye et al. [22] reported that 2.8 ± 2.5 biopsies were required in men with SCOS, and 4.5 ± 4.2 biopsies were needed in cases with maturation arrest. Accordingly, in 2004, Shah presented the SST technique in which single seminiferous tubules were pulled out through puncture holes from all over the testicular surface, thus achieving extensive random biopsies without any testicular damage [23].

Once it became clear that even men with testicular failure often had some sperm in their testes, but that this sperm production could be very localized and difficult to find, a number of different techniques for testicular sperm retrieval in NOA were developed.

Fine-needle aspiration of the testis has been in use as a diagnostic tool for over 30 years [24]. Craft et al. [12] suggested that a similar process of testicular sperm aspiration (TESA) could be used to retrieve sperm from men with OA. In 1995, Lewin et al. [25] coined the term TEFNA (testicular fine-needle aspiration) and used a 20 ml syringe mounted on an aspiration handle to perform an extensive aspiration in men with NOA. In 1999, they published their experience with 111 TEFNA procedures in 85 patients [26]. The procedure involved 15 punctures with multi-directional aspirations with a 23 G needle from each testis. They reported 48 percent sperm recovery in SCOS, 46 percent in maturation arrest, and 66 percent (four of six patients) in non-mosaic Klinefelter syndrome. Due to its ease and simplicity, TEFNA soon became a widely used alternative to open biopsies.

However, the utility of TEFNA or TESA was disputed by other clinicians. In a direct comparison of TEFNA with open biopsy in men with NOA, Friedler et al. [27] performed TEFNA (six punctures on each testis with a 21 G needle) followed by open biopsies (up to three on each side) in the same patients. TEFNA retrieved sperm in 11 percent, while sperm were found in 43 percent of the open biopsies. Even in men with OA, Tournaye et al. failed to retrieve enough sperm by TESA in 2 of 53 men.

In an attempt to improve the success of TESA, Foresta et al. [28] used a color Doppler to compare the intra-testicular blood flow patterns in men with OA and NOA. They found that spermatogenesis correlated with better blood flow and they were able to aspirate sperm in 12 of 16 testis (with no sperm on random aspiration) when the aspiration was performed in a region where blood flow was detected. Similar findings were reported by subsequent researchers [29,30].

As an alternative to fine-needle cytological aspiration, clinicians also used larger bore needles to take testicular biopsies percutaneously, obtaining tissue equivalent to an open biopsy. In 1993, Morey et al. described the use of a Biopty gun to obtain testis biopsies [31]. In 1994, Mallidis and Baker experimented with a variety of needles and concluded that the modified Menghini and Turner biopsy needles gave the best biopsy samples [32]. Other authors [13,33] used a 19 or 18 G butterfly needle to aspirate a core of seminiferous tubules which were then pulled out of the testis to give a large percutaneous biopsy.

In 1997, Schlegel and Su reviewed their experience with conventional TESE done by multiple open biopsies or a single long biopsy and showed that these methods could cause significant testicular damage, demonstrable up to six months post-procedure [34]. Hence, they proposed microdissection TESE (mTESE) as a technique that would cause less testicular damage by taking very selective biopsies only from favorable areas that were visually identified. In 1999, Schlegel presented results in 27 men who had undergone both conventional open biopsies and mTESE biopsies on the same testes at the same session [35]. In 11/27 men (41 percent) he was able to find sperm by conventional biopsies. In an additional six men, sperm were found only in the mTESE samples, thus giving a total retrieval rate of 63 percent. Interestingly, in one case sperm were found only in the random conventional biopsies, but not in the mTESE samples from the same testis, which underscores the difficulty in visually identifying sperm-containing tubules in cases where the pattern is uniform.

These findings have been replicated by many subsequent studies, which confirm that mTESE is the most efficient method for finding sperm in NOA [36]. However, they also suggest that mTESE has an advantage only in certain testicular histologies: SCOS, testicular atrophy, and small testes with high FSH [37,38]. In cases with a visually uniform pattern of testicular parenchyma, mTESE may not be superior to random, multiple biopsies. Other studies showed that mTESE was not innocuous; Everaert et al. [39] found that 16 percent of 48 men undergoing mTESE developed de novo androgen deficiency. A recent meta-analysis showed that the fall in testosterone is greatest in men with Klinefelter syndrome, is maximum at 6 months, and recovers by 24 months in most, though not all, men [40].

Hence, staged sperm retrieval has been proposed [41]. During the same surgical session, progressively more invasive methods are used until sperm are found, starting with needle biopsies, progressing to therapeutic mapping, and leading up to mTESE. This helps avoid unnecessarily invasive procedures when sperm can be found by simpler methods, while ensuring that those who need mTESE are not deprived of the opportunity. While it has been suggested that such an approach may “reduce the costs, time and efforts involved in surgery” [42], this is yet to be validated in controlled studies.

An alternative approach was suggested by Shin and Turek [43], who proposed that diagnostic testicular mapping by fine-needle aspiration could be used prior to therapeutic retrieval to identify which patients would be candidates for sperm retrieval surgery, and from which areas of the testis sperm were likely to be retrieved.

Currently, microdissection TESE is the gold standard for sperm retrieval in men with NOA, against which all future techniques and modifications will have to be compared.