Predictive factors

When counseling men regarding the probability of finding sperm via any technique, multiple factors have been assessed to attempt to predict the success of sperm retrieval. Easily observed physical examination findings, such as testicular volumes, and frequently obtained laboratory values, including testosterone and FSH levels, have not demonstrated reliability in predicting success . There has been some interest in inhibin B, a glycoprotein produced by the Sertoli cells, but results using this marker also have been mixed. Some studies found elevated inhibin B levels correlated poorly with success at retrieval , but others found that inhibin B levels higher than 40 pg/mL had 90% sensitivity and 100% specificity for a successful retrieval using testicular sperm extraction (TESE) with a diagnostic accuracy of 94.1% . In addition to using each value alone, some investigators have combined clinical parameters of FSH, total testosterone, and inhibin B levels in a prognostic equation with a sensitivity of 71% and a specificity of 71.4% for successful retrieval .

More reliable predictors of successful sperm acquisition include testicular histology and azoospermia factor (AZF) deletions. Based on histology, success rates range from 80% to 90% (in men who have hypospermatogenesis) to 15% to 20% (in men who have Sertoli cell–only syndrome), with intermediate success rates for men who have various degrees of maturation arrest . Table 1 compares the success rates for sperm retrieval based on testicular histology. It seems that the more advanced the level of spermatogenesis, the more likely is the successful retrieval of mature sperm, although even finding sperm on diagnostic biopsy does not guarantee a successful outcome. Furthermore, one of the studies used histology from the sperm retrieval procedures, rather than a diagnostic biopsy. Finally, most studies that discuss sperm retrieval rates did not correlate outcomes with histology, so the studies in Table 1 may not be comparable to all other studies. The ability to retrieve mature sperm increases as the level of spermatogenesis increases. The drawback of the use of histology is that it requires a testis biopsy for this information to be of use; nonetheless, it is available to assess further chance of success in men who have undergone prior sperm retrieval procedures or even prior diagnostic biopsies. In addition to the histologic information, the success or failure of prior biopsies offers insight into the success of subsequent additional procedures. Ramsamy reported that retrieval rates with microscopic TESE (microTESE) were lower (23%) in patients who had had three or four negative biopsies than in patients who had had no prior biopsies (56%) or one or two to prior negative biopsies (51%).

DNA material is obtained more easily, and the assessment of karyotype abnormalities or gene deletions has predictive value. It has been reported that no sperm is recovered from men who have complete deletions in AZFa and/or AZFb, whereas, recovery rates in men who have AZFc deletions are similar to those in patients without deletions . This finding holds true for men who have Klinefelter’s syndrome as well . Other less well-recognized factors have been assessed also. Protamine-2 is a gene expressed postmeiotically by round spermatids. Song and colleagues found spermatozoa in 85.7% (18/21) of men found to have protamine-2 transcripts on reverse-transcription polymerase chain reaction from testicular tissue.

Percutaneous aspiration

Whether identified as percutaneous testicular sperm aspiration (PTSA), testicular fine-needle aspiration (FNA), or TESA, sperm is aspirated with a 19- or 21-gauge butterfly needle that is inserted percutaneously into the testis using a local anesthetic. Using a Cameco piston syringe to generate suction, the needle is manipulated in an in-and-out fashion to release a series of testicular tubules. These tubules are grasped at the skin and collected into media for analysis and, if viable sperm are identified, storage .

Mercan and colleagues report their experience with this procedure in 452 men who had NOA. Their overall retrieval rate was 64.4% (291/452). Approximately 15% of the study group had a successful PTSA procedure; the remaining 228 went on to have success with TESE. Men who had successful PTSA retrievals were more likely to have hypospermatogenesis and were less likely to have germ cell aplasia or maturation arrest. In addition, couples with a successful PTSA had a significantly higher clinical pregnancy rate than those who underwent random biopsy (46% versus 29%), but this finding may have been influenced by the higher number of embryos transferred (4.38 versus 3.90). Findings reported by Vicari and colleagues in a sample of 55 men who had NOA and who underwent either PTSA or TESE were slightly better than the 11% to 14% success rate with PTSA reported by other authors . In Vicari’s population, 47.3% of all PTSA attempts were successful. Further stratification revealed that sperm retrieval with PTSA was successful in 100% of men who had diagnostic biopsies demonstrating hypospermatogenesis or maturation arrest with focal spermatogenesis, but the success rates with this technique were lower in complete maturation arrest (42.3%), Sertoli cell–only syndrome (14.3%), and Sertoli cell–only syndrome with focal spermatogenesis (0%) .

The main advantage of this technique is its minimally invasive nature, which may help limit the rate of complications. In a study of 32 men who had OA undergoing TESA with ultrasound follow-up at 1.5, 3, and 6 months, 6.3% of patients developed hypoechoic lesions consistent with an intratesticular hematoma. All had resolved by 3 months . Similar data exist in the NOA population; Carpi monitored 54 patients with ultrasound following a combination of FNA and large-needle aspiration biopsy and found hypoechoic lesions in 11.1% of patients undergoing biopsy up to 63 days following biopsy. The long-term or functional significance of these lesions has not been ascertained.

The relatively small samples of tissue obtained with TESA limit the amount of sperm that can be recovered, particularly in men who have severe spermatogenic failure. Therefore success rates are not as high as with open sperm retrieval in all populations. This method tends to work better in men who have more advanced degrees of spermatogenesis. As a result, most authors conclude that most patients who have NOA require open testicular extraction to obtain sperm adequate for IVF .

Open sperm extraction

Open TESE can be accomplished using various methods, including multiple random biopsies, multiple directed biopsies, and microTESE. Numerous studies have demonstrated the success of open TESE in the acquisition of sperm in men who have NOA. These techniques vary in their invasiveness, the amount of tissue removed, and the success rate in retrieving usable sperm.

Random biopsies

Random biopsies have been used to sample multiple areas within the testis. These biopsies typically have yielded large amounts of tissue, and the success rates have varied. Avascular regions near the midportion of the medial, lateral, or anterior surface of the tunica are incised, avoiding the capsular and testicular vessels. The amount of tissue taken varies from 150 mg to 400 to 500 mg . One early series noted a 62% successful retrieval rate, with a live delivery rate of 25% . In a later series of 628 men who had NOA undergoing 784 TESE procedures, the overall retrieval rate was 48%; 41.6% were successful on the first attempt . Most practitioners agree that multiple random biopsies are likely to yield poorer results than the more directed biopsy (as discussed in the following section) and remove more tissue . In addition, the risk of side effects may be amplified after random biopsies. Using ultrasound, Harrington demonstrated a 29% rate of intratesticular hematoma, which can lead to significant loss of testicular parenchyma secondary to interruption of the blood supply or even to atrophy. In another study, 64% of men developed hypoechoic areas consistent with linear scars after open biopsy . Because of the increased risks of complications and decreased rate of sperm acquisition, random biopsies are not the first choice for retrieval in patients who have NOA, although some patients who have hypospermatogenesis may have adequate outcomes.

Multiple directed biopsies

Various techniques have been investigated in attempts to improve efficiency and minimize the theoretic risks of multiple random biopsies, which include testicular hematoma, further reduction in spermatogenesis, hypogonadism, or even complete devascularization. Multiple directed biopsies using various imaging modalities such as power Doppler have been described as a way to take advantage of architectural differences in sperm-producing areas of the testis . These imaging modalities detect a difference in blood flow, with areas of increased perfusion being the most likely to contain more advanced spermatogenesis, thereby directing the surgeon to the most likely sites of sperm production. Other studies investigated whether specific areas within the testis (eg, near the hilum) are more or less likely to contain sperm and found that there is no area that is more or less likely to yield mature sperm .

Testicular mapping

To identify sites of successful sperm acquisition and to minimize the emotional and financial costs of cancelled ICSI cycles, testicular mapping has been developed to guide the surgeon intraoperatively to areas most likely to yield mature sperm. The technique of testicular mapping is based on the concept that spermatogenesis often is sporadic and focal throughout the testis . FNA systematically assesses the testicular quadrants for the presence or absence of retrievable sperm. Initial analysis revealed that FNA mapping was more sensitive than testis biopsy and was as specific for detecting sperm . In follow-up studies, Turek found that FNA mapping identified sperm in 47% of men who had NOA (27/57) and that sufficient sperm was obtained for all oocytes in 95% of cycles (20/21); the clinical pregnancy rate was 48%. This less invasive approach can be done in the office with a cord block, but it does require the patient to undergo two procedures. It also requires expertise in cytologic technique, because it is important that the specimens be fixed and read correctly for optimal outcomes: a prospective blinded review of 113 testis biopsies found a concordance rate of only 54% between the first pathologic diagnosis and the second. More importantly, in the cases in which differences in histology between the reviewers were detected, the differences in 30 cases had the potential to change management (ie, from pure Sertoli cell–only syndrome to a mixed pattern that would raise the potential for sperm retrieval) . Because this study looked at intraobserver concordance in biopsies, not cytologies, it is pertinent to note that there does seem to be better correlation between FNA cytology and biopsy cytology; reported concordance rates range from 86% to 97% . The possibility that a misreading of histology might alter the treatment course is one risk specific to mapping. The remaining procedural risks include those noted with other percutaneous testicular biopsy procedures.

Microdissection testicular sperm extraction

Expanding on the concept of directed biopsy, microTESE has been described by several authors and is thought by some to be the most accurate method of obtaining sperm in NOA . This technique involves delivery and bivalving of the testis. The operating microscope then is used to examine individual seminiferous tubules. It has been reported that tubules containing germ cells are significantly larger and more opaque than those without germ cells. These areas are then sampled and examined by embryologists in the operating room, who confirm the presence or absence of mature sperm. This technique allows greater accuracy in finding sperm with significantly less tissue sampled. In one comparison study, microTESE retrieved 160,000 spermatozoa per 9.4 mg of testicular tissue, whereas conventional TESE obtained 64,000 spermatozoa per 720 mg of tissue obtained .

Numerous authors have reported fairly consistent sperm retrieval rates of approximately 50% in unselected patients. In patients who have had previous diagnostic biopsies, however, the positive identification of sperm on microdissection correlates with the most advanced degree of spermatogenesis on those biopsies . Complications are rare but include the possibility of intratesticular hematomas (12% at 1 month; 2.5% at 6 months) decreased testicular volume (2.5%) , temporary declines in serum testosterone values , and, in theory, the development of long-term hormonal insufficiency. These are inherent potential complications of all sperm acquisition techniques.