Varicoceles develop as a result of dilatation and tortuosity of veins of the pampiniform plexus secondary to retrograde flow into the internal spermatic vein (ISV). A varicocele is a common abnormality, occurring in approximately 15% of men. Some patients may have scrotal pain and swelling, but more importantly, a varicocele is considered to be a potential cause of male infertility. [1, 2, 3]
This relationship is controversial, but improved fertility and sperm quality have been reported after treatment, including occlusive treatment for varicoceles. On physical examination, large varicoceles are easily identified as the classic "bag of worms" surrounding the testis. Ultrasonography, particularly Doppler ultrasonography, allows accurate diagnosis of varicoceles, even subclinical varicoceles. (See the images below.) [1, 4, 5, 6, 7, 8, 9]
The role of radionuclide studies and magnetic resonance imaging (MRI) in the investigation of varicoceles is limited; these offer no advantage over ultrasonography.
Venography is the most reliable modality for the detection of subclinical varicoceles because the findings demonstrate abnormal retrograde flow into the spermatic veins or pampiniform plexus. However, the procedure remains invasive and is usually reserved for patients undergoing sclerotherapy.
Limitations of techniques
Ultrasonography does not always show reflux into the spermatic veins or pampiniform plexus. In addition, an ultrasonographic diagnosis of varicocele does not always indicate that such a lesion is the cause of the patient's symptoms and/or infertility.
Venography is highly accurate, but the procedure is invasive and exposes the patient to ionizing radiation.
The ionizing radiation of computed tomography (CT) scanning limits its use.
CT scans can demonstrate varicoceles. In one study, 2 transverse scans were obtained with the patient in a supine position during quiet breathing and during Valsalva maneuver.  The transsectional area of the spermatic cord was 80-100 mm2 on the side without the varicocele and 100-200 mm2 on the affected side. Increased intra-abdominal pressure dilated the veins of the pampiniform plexus, increasing the transsectional area by 40-80% on the side without varicocele and by 100-200% on the side with the varicocele.  The authors noted that a spermatic cord area (measured at the root of the scrotum) of greater than 100 mm2 without an increased intra-abdominal pressure and an area greater than 200 mm2 with an increased intra-abdominal pressure is indicative of a varicocele.
Degree of confidence
CT scanning with increased intra-abdominal pressure can be used as a noninvasive method to detect a varicocele and to show proximal extension of the lesion into the inguinal canal. However, exposure to radiation is a disadvantage, and to the authors' knowledge, no comparative studies with ultrasonography have been performed. The sample of patients in whom the use of CT scanning has been reported is small, and no firm conclusions can be drawn from the above-described study. 
Magnetic Resonance Imaging
On MRIs, a varicocele is demonstrated as a mass of dilatated serpiginous vessels, usually adjacent to the epididymal head. The spermatic canal is widened, and the intrascrotal spermatic cord and/or pampiniform plexus are prominent.
The spermatic cord has a heterogeneous signal intensity. It contains serpiginous high signal intensity structures, which are presumably due to phase-shift artifact from slow blood flow.
Flow-related enhancement secondary to slow flow may result in an increased intraluminal signal intensity on images obtained with all MRI sequences.
Abdominal compression over the sacral promontory may exaggerate the serpiginous vessels.
Three-dimensional (3-D) phase-contrast magnetic resonance angiography (MRA) has been described in a series of 4 patients with recurrent varicoceles. The scrotal part of the varicocele was demonstrated in 3 patients, and the spermatic vein was shown in 2 patients. Some have suggested that this technique can provide an alternative to spermatic venography in the radiologic mapping of dilatated spermatic veins. 
Degree of confidence
The role of MRI in the diagnosis of varicoceles has not been established because an insufficient number of patients have been examined with MRI.
Ultrasonographic findings in varicoceles include tortuous anechoic tubular structures adjacent to the testis.
With the patient in an upright position, the diameter of the dominant vein at the inguinal canal measures more than 2.5 mm and is associated with an increase in diameter of at least 1 mm during a Valsalva maneuver.
Varicoceles can be found anywhere in the scrotum (ie, medial, lateral, anterior, posterior, or inferior to the testis). They can be small to very large, with some enlarged vessels as large as 8 mm in diameter. Intratesticular varicoceles may appear as a vague hypoechoic area in the testis. They appear tubular or oval shaped and usually lie near the mediastinum testis.
Color Doppler ultrasonographic imaging can be helpful in differentiating venous channels from epidermoid cysts or spermatoceles when doubt exists.
Bidirectional Doppler ultrasonography performed with the patient in the upright position with quiet respiration shows a shunt type of flow in 86% of patients in whom insufficient distal valves allow spontaneous and continuous reflux from the ISV into the cremasteric vein and the vein of the vas deferens.
Bidirectional Doppler ultrasonography performed with the patient in the upright position with quiet respiration shows a stop type of flow in 14% patients in whom intact valves allow only sporadic reflux from the spermatic vein into the pampiniform plexus with a Valsalva maneuver.
Doppler ultrasonography can be used to grade venous reflux as static (grade I), intermittent (grade II), or continuous (grade III).
Effective sclerotherapy is indicated when venous dilatation resolves and the overall diameter of the veins decreases.
The ultrasonographic characteristics of varicoceles are demonstrated in the images below.
Degree of confidence
Using the diameter as the criteria of venous channel dilatation, Hamm et al found that ultrasonography had a sensitivity of 92.2%, a specificity of 100%, and an accuracy of 92.7%.  Ultrasonography is capable of demonstrating both palpable and subclinical varicoceles.
Epidermoid cysts and spermatoceles may mimic varicoceles. If in doubt, color Doppler ultrasonographic findings are diagnostic. Intratesticular varicoceles may mimic tubular ectasia.
Technetium-99m (99m Tc)–labeled red blood cells are the radiopharmaceutical agent of choice for the evaluation of varicoceles, although other radionuclides have been used as well. Images are obtained with the patient in both the supine and erect positions. Static images reveal moderate to intense intrascrotal accumulation of the labeled red cells; this accumulation may be discrete or patchy. Palpable and nonpalpable varicoceles have been identified on blood-pool images, but cases of reflux without increased blood pool usually cannot be identified on static images. 
A Valsalva maneuver during image acquisition may be helpful in detecting subclinical cases. The importance of detecting subclinical cases is related to the proposed association of varicoceles to infertility.
The use of dynamic scanning is controversial. A reduction in early blood flow on the affected side has been observed. Dynamic images allow calculation of the difference in arrival time of radioactivity between the iliac artery and the pampiniform plexus; this time is believed to shorten with increasing grades of varicoceles.
Degree of confidence
Varicoceles have been identified during blood-pool imaging, but independent correlation in subclinical cases is difficult. As with most scrotal pathologies, bilateral disease is difficult to confirm on scrotal radionuclide imaging. The sensitivity for clinically apparent varicoceles has been reported to be 90%. The specificity is difficult to determine.
Venography is the most reliable modality for the detection of small or subclinical varicoceles because findings demonstrate abnormal venous blood reflux in a retrograde fashion into the ISVs and the pampiniform plexus. (See the images below.)
Degree of confidence
Because of the invasive nature of venography, the technique is usually reserved for use in patients undergoing occlusive therapy for mapping of the venous anatomy. Occasionally, it is used in symptomatic patients when the diagnosis is equivocal with other methods.
Testicular veins often spasm, and rarely, opacification of the vein with contrast medium may be difficult. Moreover, problems may be encountered in cannulating the right testicular vein.