Diagnosis

The approach to evaluating erectile dysfunction impotence) continues to be both controversial and varied, with emphasis on identifying etiologies in patients who might benefit from specific therapies. The multifactorial etiology for most cases of impotence makes it unlikely that individual indices of erectile dysfunction will establish an etiologic diagnosis, and an accurate diagnosis will require a comprehensive multidisciplinary assessment of these patients (39,40). The evaluation of impotence is handicapped by the absence of universally accepted and validated objective measures of erectile function that correlate with sexual ability, and appropriate testing for diagnosis remains controversial (3). Nevertheless, evaluation procedures have evolved from clinical experience and include physical and psychological examinations, endocrine assessment by hormonal assays, evaluation of the competence of penile circulation, electrophysiological testing of neurologic competence, and nocturnal penile tumescence (NPT) testing (3,11,38,41).

Monitoring of nocturnal penile tumescence

It has been widely accepted in the past that patients with predominantly psychogenic impotence will regularly attain normal erections during rapid-eye-movement (REM) sleep, while patients with organic impotence will not demonstrate significant sleep-associated normal erections (42). The normal pattern of erections during sleep (between ages 3 and 79) are episodes occurring approximately every 85 minutes and lasting approximately 25 minutes. The total duration of erection time diminishes and the latency period between erections increases with aging (43). Until 1980, NPT testing was regarded as the only objective technique for the differential diagnosis of organic versus psychogenic impotence (43,44). Its major disadvantages are that it does not quantify penile rigidity, it depends on adequate REM sleep, and it lacks standardization among laboratories (20,43).

During NPT testing, mercury strain gauges placed on the shaft of the penis can detect and measure changes in penile circumference during sleep. An increased circumferential expansion of the penis does not necessarily correlate with adequate rigidity of sufficient duration required for normal intercourse (41,45). However, adequate and sufficient in this context has not been satisfactorily defined (42). During episodes of maximal tumescence (measured on an electroencephalographic tracing), penile rigidity can be measured by a rigidometer. Rigidity is a consequence of sufficient intracorporeal pressure rather than functionally related to changes in penile circumference (38). A commercial rigidometer is available for continuous and simultaneous monitoring of penile rigidity and circumferential expansion (45). The device consists of recording loops applied to the base and tip of the penis that are connected to a sensor able to measure and record data on rigidity and circumferential change for periods of up to 10 hours (45). Testing is commonly performed during three consecutive nights.

A normal result of NPT, measured with strain gauges, is two or more periods of sustained erections lasting more than 20 minutes and associated with at least a 25-mm increase in penile circumference (3). Normal NPT in a person with sexual dysfunction implies intact neurovascular structures and suggests psychogenic impotence (3). An abnormal recording only suggests organic impotence, which will then require further testing for a more definitive diagnosis (38). Nocturnal penile tumescence testing has an accuracy of 85-90 percent in measuring erectile capacity but can have a high incidence (15-20) of false-positive results (5,47). False-negative findings (not as common) can be seen with sensory neuropathies, pelvic steal, and corporeal leakage syndromes (17). The accuracy of NPT testing can be increased when it is combined with both visual inspection and rigidity determinations (48). In a study of 105 patients with impotence, Kaneko and Bradley demonstrated that rigidity measurements enabled a change of their initial diagnosis from psychogenic to organic in 3 of 11 cases (27), and from organic to psychogenic in 7 of 94 cases (7.4) (46). Some patients have been shown to achieve normal erections during their waking hours but remain flaccid during sleep. Awake erections have been regarded as a better reflection of sexual function than NPT testing (22,49).

Approximately 20 percent of subjects with impaired NPT have no demonstrable organic disease (16,37). Another 20 percent of patients with normal NPT have insufficient rigidity (11). These facts suggest subclinical physiologic deficits and/or psychogenic factors related to impotence. Nocturnal penile tumescence testing performed in a sleep laboratory with polysomnography (which measures the quality and quantity of REM sleep and other physiological variables) offers a noninvasive approach to the differential diagnosis of impotence. However, the value of NPT measurements in sleep laboratories has not been validated in large groups of patients (11,38). In addition, the criteria that will provide the maximum diagnostic accuracy of the NPT testing variables used to differentiate organic from psychogenic impotence have not been established (50). There is a need to continue the process of validating the sensitivity, specificity, and efficiency of NPT testing (51,52).

A common problem affecting the interpretation of NPT test data is the great variation in circumferential increase among normal subjects that makes it difficult to determine the significance of the maximum circumferential increase in individual patients. In an attempt to obviate this problem, Morales et al have devised a procedure, the corporeal calibration test (used as a complement to conventional NPT tests), that measures the circumferential size required to produce a rigid penis (4). The test involves the production of an artificial erection by the infusion of saline into the penile corpora and recording the increases in circumference with a mercury strain gauge placed behind the penile corona (at the same site as during NPT testing) and connected to a plethysmograph. Although NPT testing, especially with a rigidometer, can provide objective data, some investigators suggest that direct observation to confirm the quality of erections is necessary for the diagnosis of impotence (16,53). In one study, 37 percent of normal subjects were unable to achieve a full erection under laboratory conditions (54). In addition to the use of mercury strain gauges or a rigidometer, NPT testing has been monitored by other techniques including the postage stamp test, Poten Test, and Snap-Gauge. These techniques have been devised as simple, inexpensive screening devices for inhome self-testing. However, these devices provide no satisfactory indication of rigidity and duration of tumescence, regarded by many investigators as a necessary component in the evaluation of impotence.

The postage stamp test involves the use of unperforated postage stamps measuring 1 by 1.5 inches that are wrapped around and pasted on the flaccid penis at bed-time. If upon awakening the stamps are broken along the perforation, a good erection is presumed to have occurred (38). Although no breakage can suggest organic dysfunction, breakage cannot lead to a diagnostic conclusion, and the test, therefore, has limited diagnostic value (55). Despite the fact that the stamp test frequently yields unreliable results relative to NPT, its low cost and simplicity has led to its use by some investigators as a screening device (56-58). The Poten Test is a standardized, commercially available stamp test that has essentially the same advantages and disadvantages of postage stamp testing (59).

The Snap-Gauge was introduced by Ek, Bradley, and Krane as a screening device to detect nocturnal erections (60). The device was designed to measure both circumferential expansion and rigidity (radial stiffness) (57). Some investigators reserve the Snap-Gauge for patients with an equivocal or positive stamp test result (24). The gauge consists of three pliofilm elements, each a different color, embedded in a band encircling the flaccid penile shaft (60). The elements have different release constants set at 26 g, 340 g, and 453 g of force (61). The band is adjustable and fastened around the penis using a self-adhesive strip. During tumescence, one, two, or three of the pliofilm elements may break, depending on the degree of pressure from the engorged penis, which thus provides a quantitative estimate of rigidity (60). If two or three bands break, nonorganic impotence is suspected. If none or only one breaks, the problem is more likely to be organic (61). However, the relationship between the breakage of particular elements and a specific diagnosis has not been established, and in general, the value of any diagnostic technique based solely on circumferential measurements must be questioned (57). In a study of 41 patients consecutively evaluated with both Snap-Gauge and a rigidometer, Bradley demonstrated equal agreement on the diagnosis of impotence with either device; however, a more precise definition of the abnormality was described by using the rigidometer (62).

Despite their shortcomings, inhome monitoring devices designed to measure tumescence and/or rigidity have been regarded by some investigators as reasonable screening procedures, especially when sleep laboratory facilities are not readily available or feasible (63). In an evaluation of the reliability of NPT monitoring, Condra et al found discrepancies and inconsistencies between penile brachial index (PBI) testing and other diagnostic tests for impotence and little agreement between NPT records and patient-partner self-reports (64). These investigators concluded that NPT was not a reliable index for evaluating impotence.

Penile vascular assessment

Because vasculogenic factors play such a prominent role in the pathogenesis of impotence and to a large extent are regarded as amenable to treatment interventions, penile vascular assessments have occupied a significant portion of the time and attention of investigators. In a series of 353 patients screened for neurologic, arterial, and venous components of impotence, vascular pathology was found to be the major cause (32). According to the National Institutes of Health (NIH), in a communication to the Office of Health Technology Assessment (OHTA), disruption of penile arterial flow and venous leakage from the corpora are the most frequent causes of organic impotence, affecting approximately 5 million men in the United States.

In addition to NPT testing, which is a nonspecific indicator of vascular competence (26). , evaluation of the functional status of the penile vasculature has been done by arteriography, ultrasonography, cavernosography, cavernosometry, and spectroscopy, as well as through evaluation of radioactive washout, determination of the PBI, response to intracorporeal injections of vasoactive drugs, temperature measurement, and the results of the hyperemic stress test (11,26,38,65).

Arteriography is an established technique used as a definitive diagnostic procedure for arterial spasm or occlusion as a cause for impotence when other noninvasive diagnostic measures have failed. The procedure involves the injection of contrast material into the internal iliac or internal pudendal arteries, or directly into the penile corpora. The procedure is painful and requires general or regional anesthesia and is associated with a considerable risk of morbidity. Generally, arteriography is reserved for those patients suspected of having isolated arterial disease or considered for surgical revascularization (11,20). In a prospective controlled trial of arteriography in 40 impotent subjects, both positive and negative results clearly agreed with findings from PBI (described below) during exercise (66).

Penile blood pressure measurements by Doppler ultrasound pulse-wave analysis or less commonly by strain-gauge plethysmography are the most frequently used techniques in evaluating vasculogenic impotence (20,45). Doppler ultrasound measurements are confined to a single vessel and are difficult to perform in the distal portion of the penis. They also give no information about cavernous arterial flow (47,66). On the other hand, plethysmography, involving complex equipment and numerous practical problems, measures blood flow in all the vessels in a cross-section of the penis, with results regarded as superior to those derived from Doppler measurements (20,67). In a study by Doyle and Yu comparing plethysmography with Doppler ultrasound in 61 patients, plethysmography was found to be more rapid and reliable and had a sensitivity, accuracy, and specificity of 93, 95, and 100 percent respectively (68). However, the relative ease, accuracy, and noninvasive nature of Doppler pulse-wave analysis have prompted many investigators to prefer its use over both impedance and mercury strain-gauge plethysmography for the evaluation of penile circulation (69-71).

In a double-blind controlled clinical trial comparing direct penile ultrasound (to detect intracavernous fibrosis) with NPT testing and penile blood pressure measurements in 88 impotent patients and 30 normal controls, ultrasound was found to correlate significantly with plethysmography when performed and evaluated by a skilled physician (71). These investigators advocated the use of ultrasound as a rapid, inexpensive, noninvasive screening method for classifying vasculogenic versus psychogenic or neurogenic impotence. A spectroscopic method of measuring systolic blood pressure adapted to the penis was described by Gaskell in 1971 but apparently has been abandoned (65).

The penile blood pressure by Doppler compared with the brachial blood pressure both at rest and after exercise yields the PBI. A ratio of less than 0.75 is suggestive of, and a ratio of less than 0.6 is considered diagnostic of, vasculogenic impotence (26,35). Patients with a steal syndrome exhibit a decrease in penile blood pressure by a diversion of the penile blood supply during and immediately after exercise (12,72). In a study of 216 impotent men aged 40-79, Kaiser et al found that 23 percent of the abnormal PBI findings were seen only after exercise (26). In a prospective study of 503 patients with organic impotence. Chiu et al found the predictive power of ultrasonically derived PBI was best demonstrated in patients with peripheral vascular disease (73). Penile brachial index findings have also been correlated with arteriographic findings in studies of impotent men (66).

In individuals who are not clearly impotent and whose PBI ratios range between 0.7 and 0.8, a hyperemic stress test has been devised for determining whether or not these ratios are indicative of vasculogenic impotence (74). This test is based on the principal of postocclusive hyperemia used in testing peripheral arterial vasculature. Systolic occlusion pressure of the dorsal penile artery is measured by means of a 1-inch sphygmomanometer cuff with a 9. 4 mm Hg Doppler probe. The penile cuff is inflated 10-15 percent above the systolic pressure, maintained for 5 minutes, and then rapidly deflated. Potent patients responded to this stress with an elevation of penile pressure by a factor of 20 percent, while impotent patients demonstrated no increase or decrease in pressure. Patients with increased flow demonstrate erythema of the penile skin. This postocclusive hyperemia is an expected response in patients with normal arterial flow and occurs within 60 seconds of releasing the occlusion. Patients with compromised arterial flow exhibit an absence or delay of the hyperemia.

Penile temperature measurement using either an electrothermometer or thermography is another noninvasive way to determine arterial flow (17,75). However, temperature measurement is subject to significant error and has not had its reliability as a screening test confirmed.

Since 1982, the use of vasoactive drugs injected into the penile corpora has been widely used as a simple screening technique for the differential diagnosis of vasculogenic impotence (45,76). Prior to the advent of this technique, the differentiation of organic from psychogenic impotence involved extensive time-consuming testing. Six groups of drugs having smooth muscle relaxing properties have been demonstrated to induce erections when injected intracorporally (76). These are: alpha-adrenergic blockers, beta agonists, smooth muscle relaxants, calcium channel blockers, antidepressants, and peptides.

The nonspecific smooth muscle relaxant, papaverine, is the most widely used agent for inducing erections. Although its precise mechanism of action is unknown, its alpha-adrenergic blocking effect induces arterial dilation, sinusoidal relaxation, and partial venous occlusion when injected into the penile corpora in patients with intact arterial and venous functions (77,78). The relaxation of the smooth muscle lining of the sinusoidal channels of the corpora permits engorgement of the erectile bodies with blood and the consequent erection. These hemodynamic changes are similar to those induced by neurostimulation causing spontaneous erections (76). Papaverine, 30-60 mg, injected into the penile corpora will generally induce a firm erection within l5-10 minutes, lasting 30-45 minutes (79,80). A full erection following the use of an intracorporeal injection excludes the presence of severe arterial insufficiency or substantial cavernous leakage (81). If a papaverine test is positive in an impotent patient, further testing, probably of NPT, may be required to differentiate neurogenic from psychogenic impotence. The failure to respond with an erection to a test dose of papaverine suggests significant vascular disease (76,79). An incomplete erection or rapid detumescence following a firm erection is suggestive of venous incompetence (67,79). However, a definite diagnosis of cavernosal leakage requires cavernosography, usually with cavernosometry (20).

In a current, ongoing trial of intracavernosal injections in the differential diagnosis of impotence (115 patients to date) it has been possible to differentiate nonvascular, arterial, and venous etiologies (80). A response to the smallest dose indicated nonvascular impotence. No response to the highest dose indicated venous insufficiency. However, in an earlier trial of 144 impotent patients with various etiologies (psychogenic and organic), Robinette and Moffat concluded that the high response rate (97) to the intracorporeal injection of a papaverine-phentolamine combination precluded its usefulness in differentiating between types of impotence (82).

Strachan and Pryor used intracavernosal papaverine in 129 patients in an attempt to differentiate vasculogenic from psychogenic impotence (83). They noted an age-related drop in response rate ranging from 86 percent at ages 18-29 to 15 percent at ages 60-71. They also reported that results of intracavernosal papaverine testing agreed with the results of NPT testing in 68 percent of cases. Abber et al compared papaverine injections with PBI and NPT testing and concluded that intracavernosal papaverine was superior to those other methods for the diagnosis of vasculogenic impotence (84).

Mellinger et al examined the correlation between Doppler analyses and papaverine injections in 100 impotent men (79). They found normal Doppler analyses failed to predict a lack of response to papaverine in 54 percent of subjects. However, abnormal Doppler analyses were highly predictive of a vascular abnormality. In the evaluation of vasculogenic impotence many investigators combine Doppler ultrasound with intracorporeal injections of papaverine to increase the diagnostic accuracy (67,79). Finally, studies of organic impotence using induced erections in aging men (up to age 85) have found a leakage factor in the corporal cavernosa to be the most frequent cause of erectile failure (77,79,85).

Cavernosography and cavernosometry are essential techniques to specifically identify the type of cavernosal leakage (86,87). Conventional cavernosography is the isolated x-ray visualization of the cavernous bodies and their venous drainage by means of contrast material. This procedure can detect scar tissue, abnormal vascularity, and specific venous leaks or shunts. Dynamic cavernosography consists of a cine-taped cavernosogram in conjunction with intracavernous pressure measurements (cavernosometry) at three different saline infusion rates into a single corpus to achieve an erection. An accurate determination is made of the flow rate and pressure within the cavernous bodies (87). A normal response is an erection at a pressure of approximately 20 mm Hg. No erection or tumescence at any flow rate is regarded as diagnostic of a venous leakage problem. Higher flow rates are seen in patients with significant venous drainage abnormalities. In patients with normal erections, no outflow of contrast material into penile or pelvic veins can be seen on x-ray.

In a study of these combined techniques in 122 impotent patients and 18 normal patients used as controls, Porst et al found that 54 percent of the impotent patients exhibited venous leakage (88). Forty-four percent of this group had isolated dorsal penile vein insufficiency and 32 percent had complex venous leakage. These investigators then compared the results of papaverine injections in 63 of these impotent patients to those of dynamic cavernosography and found papaverine injection provided a sensitivity of more than 90 percent in the diagnosis or exclusion of a hemodynamically relevant venous insufficiency. However, this was obtained only in cases having a clear-cut test result. Papaverine testing was not reliable in cases of incomplete erections, although it remained useful as a screening procedure.

The evaluation of both arterial and venous penile blood flow can also be determined by injecting radioactive (133)Xenon (Xe) into one corpus and measuring its washout (45,89). This technique is rapid, minimally invasive, reproducible, and a direct measure of penile blood flow (rather than indirectly via Doppler or plethysmography). It is regarded as potentially superior to other standard techniques as a screening test for vasculogenic impotence (1,20). However, conflicting results have been reported by other investigators using (133)Xe in evaluating penile blood flow in human volunteers viewing erotic films (76).

Neurological Assessment

In addition to NPT and penile vascular assessment methods, neurological assessments can assist in the evaluation of the degree and nature of organic impotence. Although the erectile process is primarily a vascular phenomenon, it is mediated by complex neural mechanisms involving the sacral reflexogenic erection center (parasympathetic) and the thoracolumbar psychogenic erection center (sympathetic) (90). Interruption of any of these pathways can result in impotence (32).

Nerve conduction studies including the bulbocavernous reflex latency time and the internal pudendal nerve-evoked potentials can detect abnormalities in the function of the pelvic nerves supplying the penis (11,20). Bulbocavernosal reflex testing has been shown to be useful in evaluating patients suspected of having impotence related to cord lesions at the sacral level or pudendal neuropathy (17,32). However, this reflex test is insensitive, being present in only 70-80 percent of potent men. The reflex is elicited by stimulation of the penile glans, and the response is contraction of both the bulbocavernosal muscles and the external anal sphincter. Measurement of pudendal nerve velocity can be used to quantify the reflex. In a study of 52 men with electrophysiologic testing of this reflex, excellent correlation was obtained between prolonged latency and clinical evidence of neuropathy (91). Of the 17 patients who were evaluated for impotence, 10 had normal physical examination findings and were believed to have psychogenic impotence. Of the remaining seven, four had diabetes mellitus, one had prior pelvic surgery, one had cauda equina syndrome, and one had a history of exposure to toxic chemicals. After a study of the bulbocavernosal reflex in 100 consecutive patients with erectile dysfunction, Wabrek concluded that reflex testing needs to be done when penile revascularization is considered as a therapeutic option (92). An abnormal latency would contraindicate revascularization.

An objective and direct neurophysiologic assessment of the entire pudendal nerve afferent pathway (including the dorsal penile nerve) may be obtained by evoked-potential testing and penile biothesiometry (32). The evoked-potential testing involves electrical stimulation of the dorsal penile nerve in a manner similar to that used in testing the bulbocavernosal reflex but recording the evoked-potential electroencephalographic waveforms overlying the sacral region of the spinal cord and cerebral cortex, rather than the electromyographic response seen in the bulbocavernosal test. This test is most useful as a measurement of the presence, location, and nature of penile sensory dysfunction.

As another sensory screening test, penile biothesiometry measures the vibration perception threshold of the skin of the penis using a handled electromagnetic vibration device having a fixed frequency but variable amplitude of vibration. The test is performed by asking the patient to respond to the first sensation of vibration as the device is placed against the lateral shaft and glans of the penis as the amplitude of vibration is slowly increased (17). In a study of 137 patients with organic impotence tested with both biothesiometry and subsequent sensory evoked potential testing, 80 percent of the 38 patients who had a normal biothesiometric examination had normal evoked-potential testing (32). This increased to 93 percent for those under age 60. Ninety-nine patients had an abnormal biothesiometric result, 47 percent of whom also had an abnormal evoked-potential test. The majority of patients with erectile dysfunction related to a sensory deficit were subsequently found to have an underlying diabetic or alcoholic neuropathy. Using this same study design with another group of more than 500 patients, abnormal biothesiometric values were seen in 23 percent, 50 percent of whom had an abnormal evoked-potential test that was highly correlated with neurologic disease. A normal evoked-potential test in patients with abnormal biothesiometry results suggests vascular insufficiency of the somatosensory receptor cells or collagen infiltration of the penile skin (93). These investigators concluded that penile sensation is most objectively measured by evoked-potential testing; however, it may be more simply assessed by biothesiometry.

In another recent study, skin potentials were recorded from the genitals and extremities of potent and impotent men (94). Fifty patients (10 psychogenically impotent, 32 impotent diabetics, and 8 potent diabetics) and a control group of normals were evaluated using an adapted skin potential recorder. More than 50 percent of the impotent diabetics had an abnormal recording. All others had normal recordings.

Since the detrusor mechanism of the bladder shares the same autonomic innervation as the erectile mechanisms of the penis, cystometric studies can also provide a measure of the integrity of the neural pathways (25).

Electrostimulation and electromyographic analyses have also been used in the diagnosis of impotence in patients with neurological problems (17). However, their use has not been studied in sufficient numbers of patients to be properly evaluated.

Endocrine assessment

Endocrine evaluation in impotent men includes the determination of serum levels of testosterone, prolactin, follicle stimulating hormone, and luteinizing hormone (15,20). Testosterone levels are frequently measured alone, and other hormonal levels are measured only if testosterone levels are reduced (22). However, some patients with impotence associated with hyperprolactinemia do not exhibit reduced testosterone levels (95).

Although hormone changes may contribute to diminished libido, reduced sexual function, and ejaculation dysfunction, they generally do not cause impotence (28,96). However, there continues to be a wide range in the reported incidence of impotence associated with endocrine abnormalities. In one study, a screening of 105 consecutive impotent patients found 37 (35) with previously unsuspected endocrine disorders involving the hypothalamic-pituitary-gonadal axis (13). Another study of 300 impotent men screened by testosterone and prolactin assays detected endrocrine dysfunction in only five patients (1.7) (29).

Patients on chronic hemodialysis often exhibit impotence, which is thought to be related to increase blood levels of prolactin (97). In addition, it has been demonstrated that zinc deficiency, commonly associated with uremia, is a major cause of abnormalities in testosterone synthesis and metabolism (98).

Treatment

The therapeutic modalities that have been applied to the management of impotence include psychotherapy, vascular surgery, prosthetic devices (external or implanted), intracavernosal injections, and hormonal and other drug therapies. Psychotherapy, in the case of psychogenic impotence, and the insertion of penile implants for organic impotence have until recently been primary treatments (11,99). However, as mentioned earlier, review of these treatments are beyond the scope of this assessment and will not be addressed here. Because the vast majority of organic impotence is of vascular or neurogenic etiology, surgery (revascularization or prosthetic) has been the most widely used form of therapy. More recently, the use of intracavernosal injections of vasoactive drugs has garnered the most interest and attention in the treatment of impotence.

Revascularization

Vascular surgery for the treatment of impotence involves either surgical or microsurgical arterial bypass procedures or surgery involving the venous drainage of the penis. These procedures have been shown to restore potency in selected patients with vasculogenic impotence (18,100,101).

A variety of surgical techniques have been proposed to treat vasculogenic impotence. Aortoiliac reconstruction, endarterectomy, or arterial dilators have been successful in treating proximal occlusive lesions (Leriche syndrome). In a series of 38 patients with Leriche syndrome, 27 (71) were functionally impotent preoperatively (102). Postoperatively, an increased penile pressure was noted in 14 patients (37), a decrease was seen in 8 patients (21), and no change occurred in 16 patients (42). In the 27 patients who were preoperatively impotent, increased penile pressure was associated with restored potency in 8 of 11 patients. Only 1 of the 10 patients with unchanged penile pressure regained potency. For distal lesions, femoral or inferior epigastric blood can be directed to the dorsal penile arteries or vein or directly to the corpora (103). However, patients having these procedures have generally had short followup, and bypass has been frequently followed by occlusion.

In 1982 Virag reported a success rate of nearly 50 percent with a series of cases involving arterialization of the deep dorsal vein (104). This procedure entails anastomosis of the inferior epigastric artery to the deep dorsal penile vein (105). Six variations of the procedure are described. In 1988 Bennett reported the results of 50 venous arterializations, in which 60 percent of the patients were able to resume sexual activity (106). Patients were followed from 1-3 years. Major complications occurred in three patients. One case of priapism and two cases of hypervascularization of the glans required ligation of the anastomosis resulting in failure of the procedure. Balko et al reported inconsistent results in patients followed for up to 20 months after deep penile vein arterialization for arterial impotence (107). Their procedure consisted of femoral artery to deep penile vein saphenous bypass.

Successful penile revascularization was first reported in 1973 and has had only limited success since that time (11,16). Success rates have ranged from 30-80 percent (20). Goldstein has reported success in approximately 70 percent of young patients with pelvic injuries treated by anastomosis of the inferior epigastric artery to the dorsal penile artery (108). Crespo et al reported their experience in the treatment of vasculogenic impotence using microvascular techniques involving the interposition of an autologous vein segment between the femoral artery and cavernous and/or dorsal arteries of the penis (103). They performed 95 bypasses in 45 impotent patients. Of their 22 patients who were seen for 6 months after surgery, 17 were cured, 3 improved, and 2 were unchanged. Pearl and McGhee reported a 60 percent long-term success rate using microsurgical arterial revascularization in the treatment of vasculogenic impotence (109). McDougal and Jeffrey described revascularization in eight impotent patients using bilateral anastomosis of the inferior epigastric arteries to the dorsal penile arteries (111). In a review of the status of penile artery revascularization, Shaw and Zorgniotti concluded that the best results were seen in younger patients. Older patients or those with diabetes or hypertention or patients who have had coronary bypass surgery do not appear to have good results (100).

Dynamic infusion cavernosometry and cavernosography studies have indicated that corporeal leakage exists in more than 80 percent of patients with organic impotence (111,112). Leakage is generally associated with an increase in both the number and size of the veins draining the corpora or abnormal shunting between the corpora and the glans (113). More than 50 percent of patients with venous leakage as a cause of impotence have insufficiency of the dorsal vein or ectopic penile veins, both of which are amenable to surgery (87).

The common surgical procedures for the treatment of venous leakage include dorsal vein resection either alone or combined with arterialization with or without pericavernous ligation (88). Arterialization procedures are best reserved for patients with mixed etiology, for example, arterial disease in addition to venous leakage (113). Ligation of the deep dorsal vein has resulted in improved sexual function in 50-75 percent of patients with venous incompetence (20,112). Wespes and Schulman reported correction of venous leakage via the deep dorsal vein by simple ligation of this vein (113). Sixteen of their 20 patients exhibited improved erections satisfactory for intercourse.

A recent review by Lewis of the results of venous dissection surgery in 549 selected patients with cavernous venous leakage indicated good to excellent results in the range of 19-73 percent reported at followup from 3-26 months (114). Those patients with poor surgical results were able to obtain satisfactory pharmacologically induced erections where this was not possible prior to surgery. In a study of eight patients who underwent deep penile vein arterialization combined with ligation of the superficial veins at the base of the penis, Balko et al reported that nocturnal tumescene was excellent in four patients with venous leakage and one of four patients with mixed factors (107). A novel treatment of venous leakage has been described by Courtheroux et al using detachable balloons and coils to embolize the deep dorsal vein of the penis (115). They treated 31 patients having venous leakage with the implantation of balloons and coils. Twenty-six of these patients recovered normal sexual function.

Intracavernous injections

In addition to their use as a screening technique for the differential diagnosis of impotence, vasoactive drugs have been extensively applied as a therapeutic modality for impotence. Currently, the intracavernous injection of the nonspecific smooth muscle relaxant, papaverine, has become the most widely used pharmacologic agent in the treatment of impotence (15,116).

The concept of intracavernous injections of vasoactive substances was prompted independently by Virag in 1982 and Brindley in 1983. Virag used papaverine to induce erections and noted that this was followed in some cases by the ability to have spontaneous erections for variable lengths of time until a repeat injection was required (117). Brindley used the alpha-adrenergic blocker phenoxybenzamine (not available in the United States) to produce erections that were of unpredictable duration lasting between 30 minutes and 30 hours (81,118).

Papaverine, either alone or combined with the alpha blocker, phentolamine, has evolved as the favored agent worldwide for pharmacologically induced erections (15,119-121). However, a number of other agents including theophylline, prostaglandin E(1) (PGE(1)), nitroglycerine, and vasoactive intestinal polypeptide (VIP) can also induce erections (39).

Intracavernous injections of papaverine at doses of 5-120 mg commonly produce erections within 10 minutes (usually in less than 5 minutes) lasting from 30 minutes to 6 hours (20,119). Phentolamine, injected in doses of 2-4 mg, commonly produces tumescence but not full rigidity (as seen with papaverine), in the presence of organic impotence (33). The combination of papaverine plus phentolamine permits the use of reduced doses of both drugs and achieves a higher response rate with lessened morbidity (119).

The ability of vascoactive drugs injected intracavernously to induce erections satisfactory for intercourse has been confirmed in a large number of published reports. However, these drugs have no effect on the ability to achieve ejaculation or orgasm (39). The results are generally better in patients with psychogenic, hormonal, or neurogenic impotence and worse with vasculogenic impotence (20,122). However, patients with severe psychogenic problems may have a poor response because of increased blood levels of adrenaline (19). The effect of these injections may persist for up to 4 weeks, a phenomenon that has been difficult to explain (123).

Although the precise mechanism of action for injectable vasoactive drugs producing erections is unknown, the effects have been referred to as a pharmacologic angioplasty in patients with mild penile arterial insufficiency and also as a chemical prosthesis (18,20). It has been suggested by Sidi that the injection of vasoactive drugs mimics the action of endogenous neurotransmitters to induced penile erections (39).

Orvis and Lue, referring to a report of a double-blind crossover trial in which 29 percent of patients with organic impotence reported improved erections after saline injections, have suggested that psychological factors may play a role in the response of intracavernosal papaverine (18). Further trials using placebo injections were recommended.

A recent report has made the point that intracavernous injections do not result in erections but rather priapisms with the typical sign of the nonerected glans (124). Priapism is defined as an abnormal and persistent erection that does not subside after intercourse or masturbation. Typically it affects only the corpora cavernosa, not the corpus spongiosum or the glans (125).

In a 1986 review of the published experience with papaverine alone in 491 patients, and papaverine combined with phentolamine in 175 patients, response rates were seen in the range of 19-60 percent with papaverine alone and 81-95 percent with the combination (126). Papaverine alone yielded responses of 1-2 hours' duration. The combination required less dosage and produced longer-lasting erections. Proper responses were seen in patients with a long history of impotence, diabetes, hypertension, heart disease, or peripheral vascular disease. Adverse reactions included dizziness (7 cases); prolonged erections over 4 hours (83 cases); and local pain, ecchymosis, and hematoma (32 cases). In a series of 66 patients with impotence (as a consequence of spinal cord injury) treated with papaverine or papaverine/phentolamine, 52 achieved a good response, and 71 percent of this group continue to practice self-injection at home with relatively minor adverse reactions (127).

In a study of 100 consecutive patients treated with papaverine as a single agent, Lue and Tanagho reported 60 patients achieved improved sexual function for periods of 1-3 weeks before returning to the preinjection impotent state (128). Complications included transient local paresthesia, hematoma, infection, ecchymosis, priapism, and fibrosis in the penile shaft. The authors concluded that intracavernous injections were a useful alternative to implants but expressed concern over priapism and the unknown potential for adverse effects with long-term usage. Injections were not recommended for patients with sickle-cell disease, coagulation defects, or those experiencing adverse reactions to a test dose of drug.

Nellans et al reported on their experience with papaverine/phentolamine in 69 patients with organic or psychogenic impotence (129). Sixty-seven percent of patients achieved a response satisfactory for intercourse; 74 percent of these responders practiced self-injections at home, and 50 percent of this group reported improvement in their ability to achieve erections without continued injections. Althof et al described a study of 82 patients with impotence of varied etiologies involved in a self-injection program using papaverine/phentolamine (130). Ninety-six percent of this group achieved satisfactory erection. The dropout rate was 35 percent. Morbidity included bruising, 26 percent; pain at injection site, 5 percent; abnormal liver function tests, 7 percent; and nodule development in penile shaft, 21 percent. The authors concluded that this technique was reasonable, reliable, and relatively safe.

In a large clinical series involving 336 patients with mixed etiologies of impotence, Sidi and Chen reported excellent results using papaverine alone or in combination with phentolamine (131). Patients with neurologic or psychogenic impotence had a 97-100 percent success rate. Patients with vasculogenic impotence had an 82 percent success rate. Of the original group, 273 elected to continue with treatment at home, and tolerance to the drug (lack of efficacy) developed in 9 percent of these patients. Six percent of patients had adverse reactions, and prolonged erections were noted in 3 percent of patients. After 1 year, 63 percent of patients continued self-injections at a frequency ranging from twice a week to once a month.

Stief et al reported on more than 4,800 self-injections using papaverine/phentolamine in 186 patients over a 2-year period. Less than 1 percent had prolonged erections, and no patient developed infection or fibrosis (132). In two other studies using similar treatments in 203 patients with a followup of 6-12 months there were infrequent complications, and results were regarded as excellent in the treatment of both organic and psychologenic impotence (133,134).

Goldstein reported on his experience with several thousand injections in 100 patients, 83 of whom were followed for more than 6 months (135). Two patients developed curvature of the penile erection with a Peyronie's plaque, and two other patients developed penile fibromas. Only four patients experience prolonged erections (longer than 4 hours). Hu et al have also reported the development of penile curvature deformities and fibrosis of the tunica albuginea after long-term (6-12 months) self-injection using papaverine/phentolamine. There has been a case report of a fatal pulmonary embolism associated with papaverine-induced priapism and thromboses in the penile dorsal vein and corpora (136).

A recent review of the worldwide use of tens of thousands of injections of papaverine and phentolamine in over 4,000 patients documents its acceptance in both the diagnosis and treatment of impotence (118). Reported morbidity included hematomas, 5-10 percent; pain at injection site; 2 percent; and urethral damage, 2.5 percent. Rarely encountered morbidity included infections, penile fibrosis, hypotension, dizziness, and hepatotoxicity. Prolonged erections were noted in 5-15 percent of cases and were generally treated by local injections of the sympathomimetic drug metaraminol. Approximately 60 percent of patients continued self-injections after 1-2 years of use, attesting to patient acceptability.

Prolonged erections and the development of penile fibrosis appear to be the complications of the use of papaverine/phentolamine arousing the greatest concern. Fibrosis is generally reversible following discontinuation of the injections (137). However, the presence of fibrosis may compromise the success of future implant surgery (138).

Ideally, the goal of intracavernous injections is to produce a satisfactory erection lasting 30-45 minutes, with a maximum frequency of 1-2 injections per week to minimize morbidity (39). Prolonged erections (over 4-6 hours) predispose to irreversible smooth muscle necrosis, ischemic damage of the cavernosal tissues with irreversible scarring, and deterioration of the tightness of the tunica albuginea, all of which can result in further erectile dysfunction secondary to treatment (36,111,139). Prolonged erections are generally manageable by treatment with aspiration of blood from the corpora (5-70 mL) and/or irrigation of the tissues with diluted solution of phenylephrine, dopamine, epinephrine, or metaraminol (140,141). Surgical shunting procedures have occasionally been required but are rarely indicated (78,142).

There has been some preliminary experience with an implantable drug delivery system designed to overcome some of the morbidity associated with direct intracavernosal injections (143). Experience with the injection of drugs other than papaverine and phentolamine has been quite limited. Intestinal polypeptide has been demonstrated to be only partially effective (144). The use of PGE(1) appears to exhibit a mechanism of action similar to that of papaverine in inducing erections; however, it is metabolized much more rapidly, which may account for the lack of associated systemic reactions or prolonged erections (45,116). Reports of the efficacy of PGE(1) have been somewhat contradictory. Stackl et al report a 68 percent positive response to PGE(1) (143 of 210 patients) versus an 83 percent response to papaverine (116). Porst reported on 61 patients given PGE1. He found it to be twice as effective as a papaverine/phentolamine combination (145).

Endocrine treatment

The precise contribution of androgen, luteinizing hormone, and prolactin to erectile physiology is not completely understood, making endocrine treatment somewhat empiric (118). In general, hormonal therapy represents a specific treatment for a known endocrine abnormality, and such treatment can restore potency to most patients with endocrine-associated dysfunction (13). Indications for the use of testosterone is limited to patients with proven hypogonadism, and testosterone replacement therapy has a significant dose-dependent effect on libido, coital frequency, total number of erections, and ejaculation (28,118). However, in some patients, only libido is improved with little or no effect on potency (146).

Elevated levels of prolactin are known to be associated with impotence. The prolactin-lowering effects to dopamine agonists such as bromocriptine and apomorphine as well as dihydroxyphenylalanine have achieved modest success in the treatment of impotence in trials involving limited numbers of patients (147,148). Apomorphine has been reported to induce erections in both potent and impotent men.

In a study of eight impotent subjects tested with apomorphine or placebo injected subcutaneously, four of the eight patients developed full erections within 5-15 minutes (149). Three of the four responders had elevated basal prolactin levels. It was suggested that apomorphine may be used to identify a subgroup of patients with impairment of central dopamine agonist function who might respond to treatment with bromocriptine. However, bromocriptine is of little value unless hyperprolactinemia is a component of the impotence (150). Bromocriptine has also been shown to be useful in restoring sexual function to patients whose impotence is related to long-term hemodialysis (151).

Other therapeutic agents

A recent controlled clinical trial using oral phentolamine has demonstrated its effectiveness in producing tumescene and enhanced responses to sexual stimulation in 11 of 16 patients with nonspecific impotence (152). The use of nitroglycerine paste (as a smooth muscle relaxant) applied to the shaft of the penis has also been shown to produce satisfactory erections in various clinical trials (15,22,153). However, the limited data are insufficient for a proper evaluation of its safety and efficacy.

Yohimbine has had a long history as a reputed aphrodisiac, but this has not been well documented by clinical studies (118). Yohimbine is an indolalkylamine alkaloid derived from Rubacea and related trees. Its action is that of blocking presynaptic alpha-2 adrenergic receptors resulting in increased parasympathetic and decreased sympathetic activity (24). This alpha-adrenoceptor blocking effect is the same as that of the commonly used intracavernosal agents. Although the mechanism of action of yohimbine in treating impotence is unclear (including possible central nervous system effects), it has been speculated that its use might increase nonadrenergic activity at the alpha-1 adrenoceptor level in the penile sympathetic plexus, constrict shunt arterioles connecting the corpus cavernosa with the corpus spongiosum, and lead to an erection (18,99). However, it has been shown that alpha-1 adrenoceptor stimulation decreases penile blood flow, which appears to confound this hypothesis (154).

In an uncontrolled study of yohimbine in 23 patients with organic impotence of varied etiologies, an oral dose of 6 mg, three times a day, for a minimum of 10 weeks, produced full erections and resumption of satisfactory sexual performance in six patients (26) (99). Thirteen patients (56) reported no improved erectile function. Four patients (17) reported improved but unsatisfactory erections. The drug was well tolerated. Side effects included nausea, dizziness, and nervousness noted among three patients. A recent placebo-controlled, double-blind, 10-week trial of yohimbine in 48 patients with psychogenic impotence demonstrated improved sexual function in 62 percent of patients versus only 16 percent with the placebo (155). The effect in responders was noted in 2-3 weeks; however, there was no method of determining which patients were likely to respond to the drug.

In another randomized controlled trial by Morales et al, this time using yohimbine in 100 patients with organic impotence, the response rate was thought to be, at best, marginal (156). The responses were 42.6 percent for yohimbine versus 27.6 percent for the placebo (p <0.5). There were no serious side effects of the drug. It has been reported that yohimbine combined with isoxsuprine (a beta-adrenergic stimulant) is sometimes effective in patients refractory to either agent alone (47).

External devices

On the basis of observations that an erection-like state (tumescene) sufficient for intercourse can be achieved when a flaccid penis is placed in a vacuum, and that this state can be maintained by impeding venous outflow using a constriction band around the base of the penis, vacuum devices have been used in patients with organic or psychogenic impotence for the past 70 years (12,61,157). Currently, there are five such devices being marketed that are regarded as noninvasive, safe, effective, and relatively inexpensive methods for achieving sexual potency, especially for patients who are poor surgical candidates or object to other therapies (12,158). As a prerequisite for the use of these devices, patients must have manual dexterity and an adequate penile blood supply.

Vacuum/constriction devices represent a reversible therapeutic modality that can theoretically be used by almost any patient needing erectile enhancement (159). These devices are most effective in patients with partial impotence and are contraindicated for patients having blood dyscrasias or those taking anticoagulants. The device may also be used as a trial to determine if a satisfactory erection can be achieved prior to consideration of more invasive therapies.

The external vacuum devices are basically similar and consist of a hollow plastic cylinder, vacuum pump, connector tubing, and elastic constriction bands (12,159). Their use involves the application of a water-soluble lubricant to the entire flaccid penis and placing the cylinder over the penis to create an airtight seal. Plastic tubing fitted to the cylinder is attached to a vacuum pump that is activated to draw blood into the penis to produce an erection or augment an ineffective erection. When adequate tumescence is achieved, usually within 2-7 minutes, a constriction band is pushed down from the base of the cylinder around the base of the penis to trap blood within it. The vacuum is then released and the cylinder is removed. Tumescence is achieved only distal to the constriction band; therefore, the proximal portion of the penis remains flaccid (159). The erection is maintained for a sufficient time to permit intercourse. Normal orgasm, but no emission, may be achieved while the constriction band is in place (61). The recommended maximum use of the constriction band is 30 minutes since it produces both partial venous and arterial obstruction, distension of the superficial veins, and cyanosis of the penis. Petechiae and ecchymosis commonly occur with the use of these devices but are not clinically significant. Mild penile pain and/or numbness has been reported by some patients; however, in a small trial (nine-person) reported by Cooper, all participants expressed enthusiasm for continued use of the device (61).

Korenman et al tested a vacuum/constriction device in eight patients aged 49-69 with organic impotence of varied etiologies (160). Twenty-eight percent of patients had a history of no erections, and 72 percent reported inadequate erections. Three patients had failed to respond to intracavernosal papaverine injections. Eighty-four percent of these patients achieved satisfactory success using this device. Erections lasted a mean of 16 minutes and the device was preferred over intracavernosal injections or the consideration of an implant. These investigators concluded that this device was safe and Nadig et al used a vacuum device in 35 patients with organic impotence who had refused the option of a penile implant (157). Thirty-two of the group achieved penile rigidity sufficient for intercourse, and 24 patients continued regular use of the device with satisfactory results. Three failures in the initial group were associated with either a penile deformity due to prior surgery (one case) or problems associated with proper fit of the cylinder (two cases). Eight patients experienced numerous petechiae, three demonstrated ecchymosis, and two had painless swelling of the penis. All of these side effects were reversible.

An evaluation of vacuum devices by the Testing and Evaluation Program of the Veterans Administration (VA) Marketing Center found that in patients randomly recruited by referrals to VA urology services with either organic or psychogenic impotence who were using the devices, most were either satisfied with or enthusiastic about their performance in permitting sexual intercourse (161). These devices have now achieved widespread use throughout the VA hospital system, and during the year ending December 1988 more than 1,000 devices were distributed.

Another device that is not used extensively by the VA system but is nonetheless quite popular is the external splint. The splint is a semirigid, silicone rubber, condom-shaped apparatus that can be fitted over the penis (159). A small tube connected to the device permits air to be evacuated from its interior. A thin collar at the proximal open end of the device acts as a valve to form a seal around the penis following the evacuation of air. After lubricating both the device and the penis with water soluble jelly, the penis is placed into the device and a vacuum is created by mouth suction on the attached tubing that draws the flaccid penis into the device while simultaneously producing tumescence. The proximal collar is then unfolded toward the base of the penis to reduce the possibility of vacuum loss. The tubing is occluded by a valve and loosely wrapped around ridges on the base of the device. A lubricant is then applied to the exterior of the device to facilitate intercourse while the device is worn. Tumescence is retained as long as the device is Release of the vacuum lock-valve permits removal of the device. The primary adverse reaction reported was penile and vaginal irritation.