NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Penson DF, Krishnaswami S, Jules A, et al. Evaluation and Treatment of Cryptorchidism [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2012 Dec. (Comparative Effectiveness Reviews, No. 88.)
This publication is provided for historical reference only and the information may be out of date.
The association of undescended testicles with later adverse health outcomes, including testicular dysfunction, infertility and cancer compelled early identification and management of this condition. Establishing a treatment plan requires knowing whether a testicle exists, where it is, and to what degree it is likely to be functional. While surgical approaches can provide definitive answers, the use of either or both imaging and hormonal stimulation has the potential to provide clinicians and families with information in a non-invasive way to guide treatment. This includes the possibility of avoiding surgery altogether if no testicle is present. Imaging may also serve to provide clinicians with certainty about the location of the testicle when it is not palpable, ensuring that when surgery is performed, it can be done in a site-specific rather than exploratory way. If it is established that surgery is appropriate—that there is a testicle present in some form—the choice is what type of surgery to pursue, and whether to use an open or laparoscopic approach.
State of the Literature
We identified 3,448 nonduplicate titles or abstracts through the search process with potential relevance, with 830 proceeding to full text review (Figure 2). Sixty-four were included in the review, representing 60 distinct studies: 16 randomized controlled trials (RCTs) (2 good quality,16, 17 2 fair quality,18, 19, 12 poor quality20–34), 5 prospective cohort studies (1 good quality,35 2 fair quality,36, 37 2 poor quality38, 39), 21 retrospective cohort studies (4 good quality,40–44 17 poor quality45–61), 16 prospective case series (1 good quality,62 3 fair quality,63–65 12 poor quality66–77), and 2 retrospective case series (1 fair78 and 1 poor quality79). Eighteen studies pertain to Key Question (KQ) 1a, 2 studies to KQ1b, 14 studies to KQ2, 26 studies to KQ3, 23 studies to KQ4, and 11 studies to KQ5.
Principal Findings and Considerations
Discussion of Approaches to Treatment Planning
Reliable techniques are needed to differentiate intra-abdominal from inguinal testicles and discriminate testicles from other structures including lymph nodes and vessels in order to create appropriate plans once a diagnosis of cryptorchidism is made. We identified studies on a range of imaging techniques, including ultrasonography (US), magnetic resonance imaging (MRI, (conventional and diffusion-weighted), computed tomography (CT) scan, magnetic resonance venography (MRV), magnetic resonance angiography (MRA), magnetic resonance arteriography (MRAr), and combinations thereof. Overall, MRA and MRV demonstrated greater accuracy based on one fair65 and two poor66, 77 quality studies (Table 29). These last two imaging studies, however, usually require anesthesia or sedation.
Table 29
Performance of imaging techniques.
Specifically, US (one good quality study62) had a sensitivity of 15 percent to 80 percent and a specificity of 100 percent for identifying testicles in most studies (Table 29). The false positive rate was very low, with a false negative rate of 39 percent. Overall accuracy ranged from 21 percent to 76 percent. Though US identified 92 percent of the testicles in the inguinal region (Table 28), it failed to identify two thirds of the intra-abdominal and more than 80 percent of the atrophied testicles. US also misidentified three abdominal testicles as inguinal. The true performance of US may be underestimated here due to higher number of abdominal testicles in these studies and due to the poor quality of most of the studies.
Studies using MRI (two fair63, 78 and eight poor quality65, 68–70, 72, 74, 76, 77) reported a sensitivity of 33 percent to 91 percent and specificity of 56 percent to 100 percent (Table 29). Again, there were fewer false positive findings with MRI including one testicular nubbin identified as intra-abdominal and another absent testicle as inguinal. MRI was able to correctly locate a higher proportion (71 percent vs. 34 percent) of intra-abdominal testicles than US and was comparable in locating inguino-scrotal testicles (83 percent vs. 92 percent, Table 28). MRI failed to locate more than two thirds of atrophied testicles. The diagnostic performance was superior when conventional MRI was used with DWI for an overall accuracy rate of more than 85 percent.78
There is a suggestion that the diagnostic performance of imaging techniques improves when a combination of techniques are used.63, 68, 72 Overall accuracy increased to 95 percent when US and MRI were used sequentially in one poor quality case series.72
The two studies, one of fair65 and one of poor quality,77 that evaluated the efficacy of MRA in identifying and locating nonpalpable testicles reported a perfect accuracy rate (100 percent), with correct identification of both normal and atrophied testicles as well as correct locations. MRI when used with MRAr/V66 located 80 percent of intra-abdominal and 40 percent of testicular nubbins correctly in a poor quality study.
Only one study (of poor quality) assessed the use of CT scanning and reported a sensitivity of 57 percent and an overall accuracy rate of 60 percent, with correct identification of only 55 percent of the abdominal, inguinal and absent testicles.67 The low accuracy rate may be explained by the large number of children younger than 5 years of age in this study (~50 percent) as CT scan image quality suffers in this age group because of the diminished intraabdominal fat planes in young children and consequent difficulty in reading the images.
Other considerations are likely important in making a clinical decision about what practice to follow. Ultrasonography is readily available, relatively inexpensive, and requires no sedation. It does, however, require the patient’s cooperation. Also, the ability to read and interpret US results is operator dependent. MRI is more expensive than either US or CT scan, and it requires a long scanning time with sedation in young patients. It may, however, be clinically preferable to US as it allows global, multiplanar depiction of the anatomy of the structures and can distinguish testicles from lymph nodes by using specific orientation and sequences in axial or coronal plane films. Diffusion-weighted MRI is another non-invasive technique that can identify highly cellular intra-abdominal testicles. MRV71 shows great promise in correctly identifying and locating all nonpalpable testicles, but it is invasive, requires general anesthesia, and is difficult to perform.
The other pretreatment approach with potential to affect care is the use of hormonal stimulation testing to ascertain whether a viable testicle exists for surgical repair. Two studies (one fair37 and one poor quality39) of a total of 44 boys examined the utility of hormonal stimulation testing in reducing the need for surgery in prepubescent males with bilateral impalpable testicles. Both were cohort studies in which human chorionic gonadotropin (hCG) was used to stimulate testosterone secretion to diagnose impairment in testicular endocrine function, in order to predict anorchia or testicular aplasia. Both used a similar study design in which the participant was first given hCG to stimulate testosterone production and then underwent surgical exploration to confirm the absence of presence of viable testicular tissue.
Both studies reported 100 percent sensitivity, suggesting that hormonal stimulation may have potential for identifying patients in whom surgery would be inappropriate. However, the studies are small, and lack a proper comparison of test characteristics between the two thresholds discussed (a greater than two-fold increase in serum testosterone levels or a total testosterone of > 5ng/mmol after stimulation).
Discussion of Approaches to Treatment
Treatment options for cryptorchidism may include an initial trial of hormone therapy to elicit testicular descent, or surgical repair. Fourteen studies (3 of good, 16, 35, 43, 44 2 of fair,18, 19 and 9 of poor quality 20–30, 38) assessed hormonal therapy in treatment. Individual studies often included multiple arms. Six studies compared luteinizing hormone-releasing hormone (LHRH) with placebo; one compared hCG with placebo; four compared LHRH with hCG; and six compared various doses or regimens. Five studies were of poor19, 23–25, 27, 29, 30 and one was of fair18 quality, Four of five studies concluded that LHRH was more effective than placebo in inducing testicular descent with variable reported effect sizes across studies (Table 15).18, 19, 27, 29, 30 Human chorionic gonadotropin is administered with multiple injections and was compared to placebo in one fair-quality study.18 In that three-arm study (LHRH vs. hCG vs. placebo), hCG was superior to placebo, but with only one study of fair quality, the strength of evidence is insufficient. Four studies provided data on LHRH compared with hCG, with neither better than the other. The studies that compared doses and dosing schedules within hormone type were of poor quality and so heterogeneous as to preclude synthesis.
A small number of studies report that LHRH and hCG are somewhat more effective than placebo. Initial location of the testicle may influence success rates, and although no study was adequately powered to assess this possibility, studies that do report stratified data consistently report that lower testicles have higher response rates to hormone treatment. Some studies have reported temporary virilizing side effects, including increased penile length, erections and testicular enlargement associated with hormonal treatment. All side effects were transitory.
We assessed the strength of evidence for our primary outcome of testicular descent. There is moderate strength of evidence for a superior effect of LHRH over placebo, low strength of evidence for the superior effect of hCG over placebo and low strength of evidence for equivalence between LHRH and hCG (Table 30). It is worth noting that followup in these studies of hormonal therapy was relatively short and likely did not capture cases of long-term re-ascent/failure, which may be a real concern in these patients.
Table 30
Strength of evidence of hormonal treatments for cryptorchidism.
No studies provided cancer or fertility outcomes for the comparisons listed so the strength of evidence is insufficient for these outcomes.
Twenty-four studies provided outcomes of various surgical interventions for the treatment of cryptorchidism. Four studies were judged to be of good quality,17, 40–42 one of fair quality,36 and the remainder of poor quality.31, 32, 45–61 Eleven studies compared outcomes following either one-stage Fowler-Stephens (FS) orchiopexy, two-stage FS orchiopexy or primary orchiopexy.40, 47, 48, 50, 52, 54, 56, 57, 59–61 Five studies primarily compared the same procedure performed through either a laparoscopic or open approach.17, 31, 36, 49, 55 Four studies compared minor surgical variations of primary orchiopexy to one another.32, 41, 45, 58 Three studies compared long-term fertility outcomes in men who underwent various surgical procedures for cryptorchidism in childhood,46, 51, 53 while one compared endocrine function in children with surgically treated or untreated cryptorchidism.42
Surgical treatment for cryptorchidism was associated with success rates of testicular descent that ranged in studies from 33 percent to 100 percent. No studies compare hormonal therapy alone to surgery. The three types of surgical repair are one-stage FS, two-stage FS and orchiopexy. If the testicular vessels are long enough to reach into the scrotum, then these structures should be spared and a primary orchiopexy is typically performed. The FS methods are used when the vessels are too short to allow mobilization of the testicle into the scrotum. In this approach, the gonadal vessels are divided and the testicular blood supply comes through collaterals, including the artery of the vas deferens. In the one-stage approach, the gonadal vessels are ligated and the testicle is immediately moved down into the scrotum with great care taken not to injure any collateral circulation. In the two-stage approach, after ligation of the gonadal vessels, the first procedure is ended. The patient is then followed for 3 to 6 months, allowing presumably better collateral circulation to develop. At the end of this waiting period, a second procedure is performed to move the testicle into the scrotum. These FS approaches are usually reserved for patients with arguably more severe conditions. None of the studies in our review appropriately controlled for initial location of the testicle; thus it is unsurprising that studies report better outcomes among patients undergoing primary orchiopexy rather than FS surgery,40, 47, 48, 50, 52, 54, 56, 60, 61 and the proportions achieving testicular descent should be looked at individually by surgery and not compared across surgical types.
The weighted success rate for all three approaches exceeds 75 percent, with an overall reported rate of 78.7 percent for one-stage FS, 86 percent for two-stage FS and 96.4 percent for primary orchiopexy. Each surgical approach was assessed independently for ability to achieve testicular descent because, as described in the report, each approach is used under different clinical circumstances, and thus it is inappropriate to compare them to one another. We assessed the strength of evidence as our confidence in the weighted average of successful testicular descent associated with each surgical approach separately (Table 31). Although retrospective studies typically had high risk of bias because of lack of a control group, in grading the overall strength of the evidence, we used an implicit comparator group given the known natural history of disease. Given the low rate of spontaneous testicular descent, despite the high risk of bias of retrospective studies, the strength of the evidence might be considered high because of the high magnitude of effect when compared with an implicit control.
Table 31
Strength of evidence of surgical treatments for cryptorchidism, testicular descent.
For the outcome of testicular descent, strength of evidence was moderate for one and two-stage orchiopexy and high for primary orchiopexy. All studies were retrospective cohort studies, and thus had high risk of bias, but we deemed these to be an appropriate study design for the question of ability of orchiopexy to achieve testicular descent and considered the relative challenges of this design to be outweighed by the magnitude of effect. Primary orchiopexy had higher strength of evidence than one-stage and two-stage based on the higher number of testicles (outcomes) reported in the literature.
We also assessed strength of evidence for the outcome of testicular atrophy, and on the same methodologic basis as was used for testicular descent, found the strength of evidence to be low for a 28.1 percent atrophy rate with one-stage FS, low for an 8.2 percent atrophy rate with two-stage FS and moderate for a 1.83 percent atrophy rate for primary orchiopexy (Table 32).
Table 32
Strength of evidence of surgical treatments for cryptorchidism, atrophy.
Laparoscopy in the treatment of cryptorchidism has two roles: (1) as an exploratory tool to locate a nonpalpable undescended testicle in the abdomen; and (2) as a minimally invasive method of orchiopexy. Two studies (one of good17 and one of poor55 quality) assessed laparoscopy for determining the location of the testicle, and reported that it was similar to open exploration. Success of the ensuing surgeries was also similar, regardless of exploratory approach. Neither study addressed postoperative pain or time to return to normal activity.
All but one of the studies in our review published in the past 5 years that included assessment of the abdomen for a nonpalpable testicle31, 36, 48–50, 57, 59, 60 used laparoscopy for this part of the procedure, even if they used an open technique to repair the cryptorchidism, suggesting that the results of the two studies above are applicable to current practice. We assessed the strength of evidence for equivalence of laparoscopic and open approaches for achieving testicular descent to be low with one RCT of poor quality31 and two cohort studies of poor49 or fair quality36 that provided consistent results (Table 33). While the strength of the evidence is low. the individual studies report that success rates are similar with both approaches. Studies reported similar clinical outcomes and less pain, shorter hospital stays and a quicker return to normal activity.31, 36, 49 No studies reported on the surgical learning curve, which is a potential modifier of effectiveness.
Table 33
Strength of evidence of surgical approach for orchiopexy, testicular descent.
Similarly, strength of evidence was low for the outcome of atrophy and the use of laparoscopic approach (Table 34).
Table 34
Strength of evidence of surgical approach for orchiopexy, atrophy.
Few studies compare the effectiveness of interventions on later fertility. Furthermore, in those studies (where the participants are obviously adults who had cryptorchidism in childhood), the primary outcome is usually semen analysis parameters, which are at best a proxy for fertility. One poor quality study examined paternity (ability to father children) and focused on the addition of hormonal therapy to surgery, finding no advantage to the combination of hormones and surgery compared to surgery alone.46 No studies compared paternity rates between surgery and hormonal therapy in isolation. To this end, it is difficult to comment on whether one approach is superior to another in terms of fertility outcomes, although the relationship of untreated cryptorchidism and later poor fertility outcomes has been reported.81, 82
Applicability
The degree to which the data presented in this report are applicable to clinical care depends on the degree to which the population included in the studies represents the patient population in clinical care, as well as the availability of the interventions and the degree to which the study settings mirror those in usual clinical practice. Across all KQs, there is no indication that study populations are different from those in standard clinical practice. Indeed, many of the studies included are reports of clinical practice, including chart reviews. Study populations included children with undescended testicles, as required by our inclusion criteria. Data are provided across the studies on children with a range of initial testicular location (e.g., intra-abdominal, inguinal, or scrotal) and unilateral or bilateral disease.
Applicability of the imaging results depends on the availability of the specific imaging technologies in a given clinical setting, including the availability of trained and experienced operators. Ultrasonography is readily available, relatively inexpensive, and requires no sedation. It does, however, require the patient’s cooperation. The ability to read and interpret US results is also operator dependent. MRI is more expensive than either US or CT scan and it requires a long scanning time with sedation in young patients. It does, however allow global, multiplanar depiction of the anatomy of the structures and can distinguish testicles from lymph nodes by using specific orientation and sequences in axial or coronal plane films.
The applicability of imaging may also depend on patient age, with technologies requiring patient cooperation potentially more challenging to use in infants. With improvements in imaging techniques, it may also be the case that early studies underestimated effectiveness of imaging technologies.
Hormonal approaches to treatment were assessed in children with both bilateral and unilateral cryptorchidism and at varying ages, providing data for the range of patients likely to be seen in clinical practice in the United States. Although most studies were done in Europe, the results should be applicable to a U.S. population as the hormonal agents studied are readily available in the United States.
As with the hormonal treatment literature, the surgical literature applies to the range of patients likely to be seen in practice, and the surgical techniques assessed are those commonly used in U.S. clinical care. The most common outcome assessed was testicular position; this is typically the outcome targeted in clinical practice. In addition, most studies provided adequate followup data to assess later atrophy.
Tables describing the PICOS elements as they can be used to assess applicability are provided in Appendix F.
Implications for Clinical and Policy Decisionmaking
The goal of any intervention for cryptorchidism is to move the undescended testicle to a normal position in the scrotum, in as safe and as least invasive a way as possible. This report has reviewed the literature on treatment planning and therapeutic interventions to achieve these goals.
Studies do not provide support that imaging has been shown to be helpful in guiding treatment decisions. Knowing where the testicle is (high or low intra-abdominal, inguinal or scrotal) could be helpful for planning a surgical approach, although there is no evidence that it can affect outcomes. The imaging literature provides mixed results, with studies not pointing to a particular approach that provides complete accuracy at identifying atrophy or absence of the testicle. Studies of hormonal stimulation testing suggest that this approach may be able to identify viable testicular tissue, but with only two studies (one of fair37 and one of poor39 quality) available, the literature is limited and much more information is needed.
Both open and laparoscopic approaches to the surgical repair techniques are viable. The question of which to use may be more clinically determined by issues such as desire for shorter recovery, tempered by provider skill at the particular approach. Providers adept at laparoscopy may choose this approach, and the evidence suggests that outcomes are similar to an open approach. Providers adept at the open surgical approach may also elect to do the surgery in an open way. Again, the outcomes do not differ in the literature, but the strength of the evidence around this similarity is low.
Contextual Information Not Covered in the Review
Although the comparative literature does not include long term data on either fertility or cancer outcomes, there is a body of epidemiologic data and data from noncomparative studies to which we refer end users of this report. These studies did not meet the specific scope or inclusion criteria for our review, but they provide important information on long term outcomes for individuals with cryptorchidism. Because we did not review these studies systematically as part of the scope of this review, we do not suggest that the references included here are comprehensive; rather they are representative and provide additional context for decision makers. Of particular note is a series of studies by Peter Lee and colleagues that describe long term outcomes, including paternity rates, among men treated surgically for cryptorchidism who have attempted paternity.83–85 These studies suggest that while men who were treated for unilateral cryptorchidism as children do not have substantially lower paternity rates than control subjects, men treated for bilateral cryptorchidism experience substantially lower paternity than both those with previous unilateral cryptorchidism and controls.85 These studies do not include untreated individuals with cryptorchidism; rather the cryptorchid groups are “previously cryptorchid” having undergone surgery.
Methodologic Issues
The literature available to assess treatment for cryptorchidism is characterized by a lack of standardization of outcomes. Studies routinely use the term “success rates” but fail to define a successful outcome. In some cases, the authors report “success rates” as proper placement of the testicle in the scrotum in the early post-operative period and then report 6-month or greater atrophy rates separately. Given that the goal of the procedure is usually to place the testicle in the scrotum and to maximize long-term endocrine function and fertility, the definition of success should always reflect both of these important endpoints (testicular location and size), and we encourage researchers to report both.
Because the majority of reviewed studies were observational, the potential for confounding and effect measure modification in this literature to obscure true effects is significant. Studies intended to address comparative effectiveness of treatment in this condition, including one-stage versus two-stage FS orchiopexy for nonpalpable abdominal testicles should either use a randomized design or carefully control for covariates such as testicular location, size and appearance, ectopia and unilateral or bilateral disease. This is particularly important for initial testicular location, which may be both a modifier of effectiveness and a factor used to choose the surgical procedure. Finally, these studies must include follow-up for at least 6 to 12 months to observe for delayed atrophy of the testicle.
Research Gaps
Although a number of studies are available on US and magnetic resonance approaches, very few studies are available on MRA and MRV, which could be useful for confirming findings after negative US and MRI. A study comparing multiple imaging modalities, such as US or MR, in the same patient followed by diagnostic laparoscopy and appropriate treatment would be of tremendous value in assessing the equivalence or superiority of any imaging approach.
Studies of hormonal treatment of cryptorchidism have focused primarily on LHRH and its agonists because it is easily administered intranasally. A wide range of success rates is seen across studies, possibly due to heterogeneity in the study populations or potentially due to variability in drug absorption through the intranasal route. Some literature suggests that differences may be due to initial location of the testicle, but this is an area warranting more attention, including conducting additional studies in which patients are carefully selected to assess efficacy by testicle location, or analyses carefully controlled for this effect. Long term data on effects of hormones are missing from the literature and more complete data on harms should be gathered.
Current literature lacks assessment of whether a one-stage or two-stage surgical method is superior, or if they are similar in terms of outcomes after controlling for testicular location. An RCT of one-stage versus two-stage FS in children with nonpalpable presumably intra-abdominal testicles who are determined by the surgeon not to be candidates for primary orchiopexy is needed. The primary outcome should be successful placement of a normal sized testicle in the dependent portion of the scrotum. The trial would also need to control for location within the abdomen at the time of presentation in order to avoid the shortfalls of the current observational literature. Ideally, outcomes would be measured both immediately and in the longer term (6 to 12 months later) to assess late complications, recurrences and atrophy. Such a trial would be ethical, as primary orchiopexy would not be the focus of the study and would not be denied to patients in whom it could be performed.
The appropriate age for treatment remains unknown, with very little data available on the modifying effect of age on outcomes. Across all approaches to treatment planning and therapy, this important question has yet to be answered.
Studies of long-term outcomes of treatment, both fertility and cancer, are notably missing from this literature. Certainly, these are difficult studies to complete, but development of a long-term cohort or a registry could provide broader and longer-term data, and warrants consideration. When studies are published on fertility outcomes, the specific measures are inadequate, focusing understandably on the easier measures of semen analysis (normal: > 15 sperm per millimeter, >50 percent having normal motility and >4 percent normal morphology).86 More appropriate measures should be included such as the ability to achieve paternity when desired. Additional outcomes that warrant study are good measures of endocrine function, which is assessed by serum testosterone levels (normal values vary by patient age and laboratory but are generally 10–44 nmol/l).
Conclusions
No specific imaging technique is adequate to identify anorchia or complete descent of the testicles and thus eliminate the need for further surgical evaluation. The lack of good quality studies affects our degree of confidence in establishing a rate of prediction of anorchia, but results do not seem to be directly related to study quality. Accuracy varies by location of the testicles, with less invasive methods (nine studies of ultrasound including one good,62 two of fair,63, 64 and six poor quality;68, 72, 73, 75, 77, 79 ten studies of MRI, including three of fair63, 65, 78 and seven of poor quality68–70, 72, 74, 76, 77) demonstrating poor accuracy for abdominally located testicles and those that are atrophied. Hormonal stimulation testing may predict anorchia, but more research is needed with only two studies (one fair37 and one poor39 quality) of fewer than 50 participants. Hormonal treatment is marginally effective relative to placebo, with moderate strength of evidence, but is successful in some children and with minimal side effects, suggesting that it may be an appropriate trial of care for some patients. If successful, these patients should continue to be monitored for late re-ascent, as most of the studies on this issue did not include long-term followup. Surgical options appear effective, with rates of normal post-operative scrotal position above 75 percent. Our ability to draw definitive conclusions regarding the comparative effectiveness of the surgical approaches is limited by confounding by indication in the individual studies, which also affects the quality of the literature. The strength of the evidence for the effects of either one-stage or two-stage FS procedures on testicular descent is moderate (low for atrophy) and high for primary orchiopexy (moderate for atrophy). Comparable outcomes have been seen with laparoscopic and open approaches to surgical repair (low strength of evidence for testicular descent and atrophy in studies comparing these approaches).
- Discussion - Evaluation and Treatment of CryptorchidismDiscussion - Evaluation and Treatment of Cryptorchidism
- Results - Evaluation and Treatment of CryptorchidismResults - Evaluation and Treatment of Cryptorchidism
- Technical Expert Panel - Evaluation and Treatment of CryptorchidismTechnical Expert Panel - Evaluation and Treatment of Cryptorchidism
Your browsing activity is empty.
Activity recording is turned off.
See more...