Pre-Penile Arteries Are Dominant in the Regulation of Penile Vascular Resistance in the Rat

Pre-penile arteries are dominant in the regulation of penile vascular resistance in the rat

K Manabe1,2, JPW Heaton1*, A Morales1, H Kumon2 and MA Adams1*

1Departments of Pharmacology and Toxicology and Urology and the Human Sexuality Group, Queen's University, Kingston, Ontario, Canada; and 2Department of Urology, Okayama University Medical School, Okayama, Japan

The amount of blood ¯ow into the penis that will produce an erection is dependent on the sum of in¯ow resistance from the feeder arteries, arterioles and the intra-penile vasculature. In the present study, our objective was to determine quantitatively the contribution to in¯ow resistance of these different components of the rat penile vasculature. Using methods developed previously, we determined the resistance properties of the isolated perfused whole penis in situ, both in an intact system and after serial transactions of the vessels. These cuts eliminated progressively larger distal segments of the vascular bed. Perfusion pressures were recorded at different ¯ow rates (0.5 ± 3ml=min=kg body weight) under conditions of maximal dilatation and maximal vasoconstriction induced by methoxamine (MXA, 40 mg=ml). Regardless of the level of vascular tone, the pudendal artery contributes approximately 70% of the total resistance of the penile vasculature. In contrast, the vasculature within the penis (tip, shaft, crus) contributes only about one quarter of the resistance. Penile arterial in¯ow resistance properties both at maximal vasodilation and maximal a1-adrenergic constriction are dominated by the extra-penile vasculature in the rat. The implications of these ®ndings are that alterations in the pudendal-artery (eg vasodilation, vasoconstriction, stenosis) would have primary control of arterial in¯ow and suggest an important role for pharmacological agents which can promote a more generalized vasodilation (eg phosphodiesterase inhibitors) in contrast to selective corpus cavernosal agents. International Journal of Impotence Research (2000) 12, 183±189.

Keywords: penis; penile vascular resistance; pudendal artery

Introduction arteries which feed directly, not via capillaries, into the interconnecting cavernous sinuses. The unique It is well established that the generation of a circulation of the penis allows for both ®lling of physiological penile erection involves both corporal corporal sinuses with blood during an erection and smooth muscle relaxation and the vasodilatation of arterial-venous shunting in the ¯accid state. This penile arterial blood vessels to produce an adequate occurs through changes in the contractile state of level of arterial in¯ow into the corpus sinuses.1±4 the vascular smooth muscle of the arterio-venous Therefore, inadequate penile arterial in¯ow is system which includes all the pre-sinus and widely acknowledged to be one of the major pre-capillary arteries and arterioles in addition to etiologies of erectile dysfunction5 although fre- well recognized changes in helicine arteries and quently the precise `cause' of the abnormality in sinuses.6±8 the arterial in¯ow is dif®cult to elucidate and may Hemodynamic studies in dogs8,9 and in normal involve functional de®ciencies as well as anatomic young men10 have shown that an erectile stimulus problems. The arterial supply to the erectile bodies provokes a large and transient increase in intraca- in both humans and animals is normally derived vernosal arterial in¯ow. Other specialized vascular from bilateral cavernous arteries which are branches beds have been studied and a signi®cant role for of the internal pudendal arteries.3,4 These cavernous feeder vessels that respond directly to changes in arteries give rise to numerous, tortuous helicine metabolic demand has been identi®ed. In active skeletal muscles, for example, the increase in blood ¯ow can exceed resting values by many fold. In this vascular bed, unlike many other beds in which *Correspondence: MA Adams or JPW Heaton, Department of vessels smaller than 100 mm predominate,11 the Pharmacology and Toxicology Queen's University, Kingston, Ontario, Canada K7L 3N6. larger feeding artery to skeletal muscle can con- Received 23 March 1999; revised 18 May 1999; accepted tribute as much as 50 ± 60% of the total resistance 28 January 2000 in certain circumstances.13 Resistance control of the penile vasculature K Manabe et al 184 Studies assessing the control of the vasculature and the lower abdominal aorta was exposed through feeding the penis have focused more on neural or a midline incision. The isolation of the pudendal local vascular control mechanisms within the penis vasculature was accomplished by sequential liga- (ie helicine arteries and corporeal smooth muscle tion of all branches of the abdominal aorta not relaxation)3,4 rather than on the resistance proper- directly supplying the pudendal vasculature. As ties of the pre-penile vasculature. However, the illustrated in Figure 1, the following arteries were locus of arterial resistance in the penis may not carefully dissected free, ligated with 7-0 silk, and reside in these small arteries and arterioles (ie cauterized to ensure that only penile vessels are helicine arteries), but as with skeletal muscle, may perfused: ie iliaca, femoralis, poplitea, glutea cra- be located in the arteries of the pre-penile vascu- nial, umbilicalis, epigastria caudal, pudenda exter- lature. Accordingly, a major objective of the current nal, glutea caudal, obturatoria, circum¯exa femoris study, using an intact, isolated perfused rat penile lateral, circum¯exa femoris medial. After heparini- vascular preparation, was to determine quantita- zation (1000 IU=kg), the abdominal aorta was can- tively the relative contribution of multiple levels nulated with a smooth, blunted 21-gauge needle. of the relevant arterial tree to resistance in the The arterial catheter was placed caudad into the penile vasculature both under conditions of max- aorta, and the inferior vena cava was cut and vented imal vasodilatation and a1-adrenoceptor mediated freely through the wide open abdominal cavity. The maximal vasoconstriction. tip of the needle was advanced down the aorta to the iliac bifurcation and sutured into place with 5-0 silk. Flow of perfusate (1.0 ml=min=kg body weight) Methods through the abdominal cannula was started immediately after opening the thorax and transection of Animals the abdominal aorta, inferior vena cava, spinal cord and all tissue 1 cm lateral to the spinal cord between T3 and T7. The perfusion pressure was recorded by a Male Wistar rats (350 ± 450 g), obtained from Charles physiograph (E & M Instrument Co., Inc., Houston, River Laboratories (Montreal, Quebec, Canada), Texas) via a pressure transducer that was connected were housed individually under conditions of a to the catheter in the aorta. The perfusate was 12-h light=12-h dark cycle (temperature of 22 ± infused for 20 min to ¯ush the penile vasculature of 24C), and received free access to Purina1 rodent blood and obtain a stable resting perfusion pressure, chow and tap water for at least 2 days before any before the beginning of any experiment. experiment. The perfusate consisted of dextran (1.5%, average molecular weight 71 400 Sigma Chemical Co., St Louis, Missouri) in Tyrode's solution, which was Perfused penile vasculature acrated with 95% O2 and 5% CO2. The composition (mg=100 ml) of the Tyrode's solution was KCl (20), CaCl2Á2H2O (32.3), MgCl6ÁH2O (5.1), NaH2- We used the model of the isolated perfused penile PO4Á2H2O(6.2), NaHCO3 (100), glucose (100) and vasculature in male rats that we had previously NaCl (800). The perfusate was held in a reservoir, developed.14 The rats were anesthetized with so- and passed through a bubble trapping=mixing dium pentobarbital (60 mg=kg intraperitoneally), chamber and heating bath. An injection port was located distal to the bubble trap for the introduction of pharmacological agents. An in-line peristaltic pump was used to establish ¯ow at 1.0 ml=min=kg body weight (Minipuls-2, Gilson Medical Electro- nics, Inc., Middleton, Wisconsin). The a1-adrenoceptor agonist, methoxamine (MXA), was delivered by a syringe pump (Harvard Apparatus Infusion= Withdrawal Pump, Millis, Massachusetts). A servo- controlled heated chamber served to maintain the ambient temperature and the entire preparation at 37 ± 38C.

Quantitative determination of contribution resistance of penile vascular circulation

Figure 1 Schematic depiction of rat preparation of isolated perfused penile vasculature. Serial transection of penile vascu- The perfusion pressure (PP) was measured before lature was made from distal to proximal as shown. and after serial transection of penile vasculature.

International Journal of Impotence Research Resistance control of the penile vasculature K Manabe et al 185 The transection sites (distal-to-proximal) were: distal cavernosum, base of the penile shaft, at the point of in¯ow to the cavernosa, proximal end of the pudendal artery, internal iliac artery immediately proximal to the pudendal bifurcation and abdominal aorta immediately distal to the cannula (Figure 1). Each transection involved a full cut through the entire penile shaft thereby eliminating all segments of the vasculature distal to the cut from contributing to resistance. At each transection, the perfusion pressure was determined at a range of ¯ow rates (0.5 ± 3.0 ml=min=kg body weight) for the fully dilated state. At maximal vasoconstriction induced by MXA (40 mg=ml Sigma Chemical Co.), we used only one ¯ow rate (1.0 ml=min=kg body weight) and made the cuts as quickly as possible in order to protect the tissue from potential damage at sustained high pressures ( > 200 mmHg).15

Data analysis

At the end of each experiment, the pressure resulting from cannula resistance and ¯ow was determined and subtracted from the raw pressure recordings. The PP changes between the cuts were determined and were used to calculate of the relative proportion (%) of PP of the entire vascular Figure 2 Relationship between ¯ow rate (ml=min=kg body bed. The Student's t-test was used to compare each weight) and the perfusion pressures (mmHg) in the isolated group mean and the Bonferroni correction was used penile vasculature (n 7). Flow rates from 0.5 to 3 ml=min=kg body weight were used (R2 0.978, slope 9.17). Points are the when multiple comparisons were done. mean, and vertical bars indicate s.d.

Results 1ml=min=kg body weight for the pudendal cut alone was more than 25 mmHg out of the total 38.6 mmHg We have previously established that this prepara- for the intact vascular bed. Similarly, at maximum tion reaches maximal dilatation in the resting state constriction, most of the decrease in PP was found because additions of supra-maximal concentration when the proximal site of pudendal artery was of sodium nitroprusside do not alter perfusion transected, the PP falling from 118.8 mmHg to pressure (PP). In addition, we have previously 2 mmHg out of a total PP of 163.2 mmHg in the established the integrity of complete pudendal intact rat (Figure 3b). Thus, although the magnitude artery isolation using an infusion of ink.14 Perfusion of the decrease in PP in the two conditions was pressure was found to correlate linearly when markedly different, the similar pattern of the de- increasing ¯ow rate from 0.5 to 3 ml=min=kg body crease in PP suggests a proximal location for the weight (R2 0.978, slope 9.17, P < 0.001) (Figure primary locus of resistance. 2). The infusion of the a1-adrenoceptor agonist, To quantitate the contribution of the different methoxamine (40 mg=ml) at the relatively low ¯ow penile vascular segments to total penile vascular rate of 1.0 ml=min=kg body weight induced a steady- resistance, we calculated the relative contribution state of constriction at a PP of approximately of each segment as a proportion (%) of total 163 mmHg. This value represents a PP which is less resistance. The calculations revealed that the pu- than 60% of the physiological maximum of the dendal artery contributes more than 65% of the total penile vasculature, which we have found to be resistance of the penile vasculature (Figure 4a). This approximately 300 mmHg. major contribution of the pudendal artery was also Assessment of changes in PP at maximal vasodi- found to be equivalent under both conditions of lation demonstrated that at all ¯ow rates, the total maximum vasodilatation and maximum a1-adreno- decrease in perfusion pressure was less than 30% ceptor mediated vasoconstriction (Figure 4b). On for all of the cuts-up until transection of the the other hand, the distal segments of the penile pudendal artery was performed (Figure 3a). For vasculature (tip, shaft, crus) together contributed example, the decrease in perfusion pressure at only about one quarter of the resistance with the

International Journal of Impotence Research Resistance control of the penile vasculature K Manabe et al 186

Figure 3 (a) Perfusion pressure changes after serial transections of penile vasculature in maximal vasodilation in seven rats. Flow Figure 4 The relative proportion (%) of each vascular segment to rate range from 0.5 to 3 ml=min=kg body weight were used. Note total resistance in maximal vasodilation (n 7) (a), and maximal the greatest decrease in perfusion pressure when the proximal vasoconstriction induced by MXA, a1-adrenoceptor agonist end of pudendal artery was cut. (b) Perfusion pressure changes (n 5) (b). Vertical bars indicate the s.d. after serial transections of penile vasculature in maximal vasoconstriction induced by a1-adrenoceptor agonist in ®ve rats. Flow rate at 1 ml=min=kg body weight was used. Points are the Table 1 Calculated values indicating relative changes in arterial mean, and vertical bars indicate the s.d. in¯ow at extremes of vascular tone; note that blood ¯ow when both pre- and intra-penile vessels are constricted is ®xed at 1

Intra-penile vessels large proximal arteries (iliac and aorta) contributing less than 5%. Maximal Maximal To further characterize the impact that these dilation constriction resistance properties would have on the in¯ow Maximal dilation 4.22 2.25 blood into the penis, we calculated the ¯ow that Pre-penile vessels would occur at 100 mmHg under different condi- Maximal constriction 1.26 1 tions using Ohm's law; R 1333 Â P=Q 1 constriction of pre-penile vessels. Under conditions in which Q (ml=sec) is perfusion ¯ow rate, P of constant perfusion pressure, the calculated ¯ow at (mmHg) is mean arterial pressure ( perfusion pres- maximal dilation of all vessels was found to be more sure), R (dyne sec=cm5) is the resistance of blood than four times that which would be found at vessels and 1333 is a unit based constant. Thus, we maximum constriction (Table 1). calculated the in¯ow which would occur under four We calculated the contribution of resistance from different conditions (Table 1): (1) maximum dilation the segment of the vasculature to the resistance of of both intra-and pre-penile vessels; (2) maximum the in¯ow penile vasculature from the perfusion constriction of both intra-and pre-penile vessels; (3) pressure dropped by serial transections. This ap- maximum constriction of intra-penile and maximum proach was modi®ed from that of Longhurst et al15 dilation of pre-penile vessels; and, conversely, (4) who showed that in the rat the basal perfusion maximum dilation of intra-penile and maximum pressure of the intact mesenteric vasculature is

International Journal of Impotence Research Resistance control of the penile vasculature K Manabe et al 187 much greater than when the small vessels surround- penile vascular bed. Our model further reveals that ing the intestine are removed. To do this, they `cut- by combining dilation of both intra-and pre-penile away' the distal small arterioles (internal diameter vessels, the arterial in¯ow would increase more than less than 200 mm) to determine if they substantially four fold over the in¯ow under conditions of contributed to the resistance.16 maximum constriction. Thus, it is essential that From our data, we calculated the average dia- concomitant vasodilation in both the upstream and meter of pudendal artery in maximalvasodilation downstream segments should occur if maximal and constriction by the following formula. perfusion ¯ow is to be obtained. Interestingly, this 4 is equivalent to the situation in exercising muscles R 8ZL=pr 2 in that pre-muscular vessels contribute 50 ± 60% of R (dyne sec=cm5) is resistance of blood vessels, Z is the total in¯ow resistance.13 viscosity, L (cm) is length of blood vessels and r (cm) Taken together, our data suggest that, for there to is radius. At a ¯ow rate of 1 ml=min=kg body weight, be suf®cient ¯ow to permit a rapid erection to occur, assuming Z 1 and length of the pudendal artery (L) at least in the rat, dilation of both the pre- and intra- is 2.5 cm, the radius of the pudendal artery (r)is penile vessels is essential. This is emphasized from calculated from equations (1) and (2). observations that a full erectile response in rats The result is that the calculated average diameter occurs within seconds. In previous studies, differ- of the pudendal artery is predicted to be approxi- ences in the amplitude and time course of erections mately 350 ± 500 mm which is not signi®cantly in rats, induced by electrical nerve stimulation as different from that previous reported.17 differentiated from those obtained by intracavernous injection, might be explained by this concept. That is, electrical nerve stimulation produces an increase Discussion in intracavernous pressure to near maximal levels within seconds18 compared with the much slower, The major ®nding in this study is that a major locus blunted pro®le over 20 ± 25 sec that occurs with of resistance to blood ¯ow in the penile vascular direct corporal injection.19 In addition, Garban et al bed, approximately 70%, comes from pre-penile observed that the maximum intracavernous pressure vessels and not from the small cavernosal and pre- elicited by direct injection into the corpus caverno- cavernosal arterioles within the penis. Further, we sum is 25% lower than that caused by electrical determined that predominance of pre-penile resis- nerve stimulation.20 A likely explanation for this tance control is equivalent under both conditions of type of result is that whereas corpus cavernosal maximal constriction and maximal dilitation. Con- injection activates only that region's vasorelaxant cepts widely accepted in vascular physiology sug- response, electrical stimulation of penile nerves will gest that small arteries with diameters smaller than globally activate ®bers innervating both the large 100 mm are normally considered to be the major site and small feeding arteries as well as the corporal of resistance.11 Applying that concept to the penis smooth muscle. The studies of Holmquist and suggests that the helicine arterioles, whose diameter Olin21 provide direct evidence of this response in is 50 ± 300 mm,3,6 is a major contributor to vascular that they electrically stimulated the pelvic nerves resistance. In fact, our results demonstrate that the and induced marked internal pudendal artery contribution of the small arteries within the penis is relaxation as part of the overall erectile response. less than 20% of the total resistance. The ®ndings of the present studies emphasize a Based on our data, we calculated the impact of predominant role for pre-penile vascular resistance. differential changes in pre-penile and intra-penile Thus, an understanding of the innervation, receptor resistance as indicated in Table 1. Modeling differ- density as well as types and quantities of neuro- ential changes in pre- versus intra-penile vessels transmitters and endothelial-derived vasoactive represent abnormal circumstances of vascular con- factors controlling the large `feeder' vessels, parti- trol but serve to reveal the impact of incomplete, cularly the internal pudendal artery, is critical to regionalized vasorelaxation. For example, when fully elucidate the mechanisms regulating penile maximal dilation of only intra-penile vessels (mini- erection. The dilation induced by intracavernous mum intra-penile vascular resistance) is achieved injection will be more localized to the corpus with pre-penile vessels remaining constricted, the cavernosum and would not initially or rapidly arterial in¯ow would increase only 1.2 fold, an involve changes in the tone of pre-penile vessels. amount not suf®cient to promote a full or rapid Thus, these data emphasize the importance erection. Conversely, when the pre-penile vessels of understanding vascular control mechanisms only are maximally dilated with the intra-penile not only in the corpus cavernosum but also in the vessels remaining constricted, arterial in¯ow in- pre-penile arteries. creased 2.3 fold. The greater ¯ow generated arising There are many studies contributing to the from proximal vasorelaxation only occurs precisely understanding of the control mechanisms of erec- because this portion of the circulation contributes a tion. Most investigators agree that regulation of greater amount to the overall resistance of the entire arterial in¯ow, venous occlusion, as well as smooth

International Journal of Impotence Research Resistance control of the penile vasculature K Manabe et al 188 muscle cells both in trabecular and penile arterial occur following intracavernous injection.26 First, blood vessels (ie cavernous and helicine arteries) are local injection of vasoactive drugs causes smooth of critical importance in determining erectile func- muscle relaxation immediately around the injection.1±5 A large number of in vitro studies have been tion site. Second, the initial small increase in local performed using methodology that includes elec- blood in¯ow activates ¯ow-induced endothelium- trical stimulation of nerves in situ and in vitro, mediated relaxing factors. Third, the zone of smooth histochemical and immunohistochemical localiza- muscle relaxation enlarges via gap junction com- tion of neurotransmitters and enzymes, and in vitro munication,27 further increasing the in¯ow of blood. pharmacological assessment of cavernous smooth As indicated earlier, one of the differences between muscle vasoconstrictor and vasodilator responses. normal physiological erections and those induced Based on these studies investigators have suggested by intracavernous injection is the slower time roles for a number of different neurotransmitters course. In the human male, the development of an including nitric oxide (NO), vasoactive intestinal erection induced by intracavernous injection can polypeptide (VIP), prostaglandins, calcitonin-gene take 10 min or longer28 in contrast to a normal related peptide (CGRP) and others.2 ± 4,22 ± 25 In erection, which can take less than a minute. Taken particular, it has been suggested that NO, presum- together with previous research, our ®ndings em- ably from non-cholinergic non-adrenergic (NANC) phasize that in the human penis, as in rats, these neurons and vascular endothelium, is a preeminent time differences can be attributed to penile arterial mediator of smooth muscle relaxation.3,4 However, in¯ow changes which, in turn, are critically depen- most of the studies in which the role of these dent on the degree of pudendal arterial dilation. substances has been explored have limited the Further, these results are useful in shedding some assessment to penile tissue only. This is based on light on the varying results of penile revasculariza- the `penocentric' idea that the intrinsic mechanisms tion surgery. For example, the most benign vascular that control the corpus cavernosum are the most surgery attempts to reconstitute normal vascular important in erectile function. Our data demonstrate anatomy in an individual with a discrete vascular that mechanisms limited to the penis are not enough lesion (eg trauma) and otherwise normal vascula- to explain the rapid arterial in¯ow required for ture. This procedure would be expected to work as normal endogenous erection and that pre-penile long as there are no substantive interruptions in control mechanisms play a key role. We are not neuro-vascular control. In contrast, revasculariza- aware of any studies that have characterized the tion would not be expected to be effective in a neural control of pre-penile vessels. situation in which the intra-penile circulation was Expanding our concept, it may be that the severely compromised. That is, despite having the mechanism of intracavernous injection of vasoactive potential for increased penile arterial in-¯ow, the agents is the clinical demonstration of this concept. capacity for tumescence would be markedly attenu- The full mechanistic details involved in erections ated thereby compromising the erectile response. In induced by intracavernous agents remain to be addition, the impact of revascularization on neuro- elucidated. Our data suggest that the slow progres- vascular control is critical in the assessment of sion of an erection induced by intracavernous response. injection likely results from the selective dilation With regard to pharmacotherapy, the revelation of the corpus cavernosum and immediately prox- that pre-penile vasculature is a major determinant of imal vessels without signi®cantly altering pre- in¯ow resistance returns some focus to agents that penile vessel tone. An alternative explanation to have a non-speci®c vasodilator action. If the puden- the slow time course of an erection following dal bed is dilated, as by adrenoceptor antagonists (eg intracavernous injection may be that there is slow phentolamine), phosphodiesterase inhibitors (eg transmission of the vasodilator signal via cell-to-cell sildena®l) or NO donors (nitroglycerin), while communication. In fact, intracavernous injection systemic blood pressure is maintained, then the may, in certain circumstances, not involve any resulting small decrease in penile vascular signi®cant pre-penile vessel relaxation. A similar resistance will result in an increased blood ¯ow process has been shown to be possible in exercising which might be further ampli®ed by a subsequent muscle, whereby a vasodilator response ascends erectogenic stimulus. retrogradely from terminal arteries to proximal In summary, the present study demonstrates that arteries. Two mechanisms have been proposed pre-penile arteries are the major contributors to to explain this phenomenon. One mechanism in¯ow resistance to the penis. Relaxation of corpus involves conduction of the dilatory signal along cavernosal arterioles only does not appear to be the vessel walls via gap junctions between endothe- suf®cient to cause a rapid erection whereas com- lial cells and=or smooth muscle cells, while a bined pre-penile relaxation will result in an ampli- second process involves endothelium-mediated ®ed in¯ow potential. Our ®ndings suggest that a ¯ow-induced vasodilation.12,13 detailed study of the functional innervation and the A speci®c example of this principle was proposed pharmacologic responsiveness of the pre-penile by Goldstein et al suggesting that a cascade of events vasculature would greatly enhance our overall

International Journal of Impotence Research Resistance control of the penile vasculature K Manabe et al 189 understanding of the factors controlling erectile 13 Segal SS, Duling BR. Communication between feed arteries function as well as the potential therapeutic targets. and microvesseels in hamster striated muscle: segmental vascular responses are functionally coordinated. Circ Res 1986; 59: 283 ± 290. 14 Banting JD et al. Isolation and perfusion of the pudendal Acknowledgements vasculature in male rats. J Urol 1995; 154: 587 ± 590. 15 Longhurst PA, Stitzel RE, Head RJ. Perfusion of the intact and Grant supports from the Medical Research Council partially isolated rat mesenteric vascular bed: application to vessels from hypertensive and normotensive rats. Blood of Canada and Kidney Foundation. MAA and JPWH Vessels, 1986; 23: 288 ± 296. are awardees of PMAC=HRF-MRC Career Awards in 16 Christensen KL, Mulvany MJ. Mesenteric arcade arteries Medicine. contribute substantially to vascular resistance in conscious rats. J Vasc Res 1993; 30: 73 ± 79. 17 Fernandez E, Dail WG, Walton G, Martinez G. The vasculature of the rat penis: a scanning electron microscopic and References histologic study. Am J Anat 1991; 192: 307 ± 318. 18 Chen KK et al. Intracavernous pressure as an experimental index in a rat model for the evaluation of penile erection. J 1 Lerner SE, Emlman A, Christ GJ. A review of erectile Urol 1992; 147: 1124 ± 1128. dysfunction: new insights and more questions. J Urol 1993; 19 Finberg JP, Levy S, Vardi Y. Inhibition of nerve stimulation- 149: 1246 ± 1255. induced vasodilatation in corpora cavernosa of the pithed rat 2 Krane RJ, Goldstein I, Saenz de Tejada I. Impotence. New Engl by blockade of nitric oxide synthase. Br J Pharmacol 1993; 108: J Med 1989; 321: 1648 ± 1659. 1038 ± 1042. 3 Andersson KE, Wagner G. Physiology of penile erection. 20 Garban H et al. Effect of aging on nitric oxide-mediated penile Physiol Rev 1995; 75: 191 ± 236. erection in rats. Am J Physiol 1995; 268:, H467 ± H475. 4 Anderson KE. Pharmacology of lower urinary tract smooth 21 Holmquist B, Olin T. Angiography of the internal pudendal muscles and penile erectile tissues. Pharmacol Rev 1993; 45: artery at electrical stimulation of the pelvic newes and at 253 ± 308. injection of posterior pituitary hormones. Scand J Urol 5 Aboseif SR et al. Erectile response to acute and chronic Nephrol 1969; 3: 291 ± 296. occlusion of the internal pudendal and penile artery. J Urol 22 Ignarro LJ et al. Nitric oxide and cyclic GMP formation upon 1989; 141: 398 ± 402. electrical ®eld stimulation cause relaxation of corpus caver- 6 Banya Y et al. Two circulatory routes within the human nousum smooth muscle. Biochem Biophys Res Commun 1990; corpus cavernosum penis: a scanning electron microscopic 170: 843 ± 850. study of corrosion casts. J Urol 1989; 142: 879 ± 883. 23 Hedlund H, Andersson KE. Contraction and relaxation 7 Fugleholm K, Schmalbrush H, Wagner G. The vascular induced by some prostanoids in isolated human penile anatomy of the cavernous body of green monkeys. J Urol erectile tissue and cavernous artery. J Urol 1985; 134: 1989; 142: 181 ± 188. 1245 ± 1250. 8 Carati CJ, Creed KE, Keogh EJ. Vascular changes during penile 24 Stief CG et al. A possible role for calcitonin-gene-related erection in the dog. J. Physiol (Lond) 1988; 400: 75 ± 88. peptide in the regulation of the smooth muscle tone of the 9 Azadzoi KM, Vlachiotis J, Pontari M, Siroky MB. Hemody- bladder and penis. J Urol 1990; 143: 392 ± 397. namics of penile erection: iii. Measurement of deep intraca- 25 Andersson PO, Bloom SR, Mellander S. Hemodynamics of vernosal and subtunical blood ¯ow and oxygen tension. J Urol pelvic nerve-induced penile erection in the dog: possible 1995; 153: 521 ± 526. mediation by vasoactive intestinal polypeptide. J Physiol 10 Shirai M, Ishii N, Mitsukawa S, Nakamura M. Hemodynamics (Lond) 1984; 350: 209 ± 224. mechanism of erection in the human penis. Arch Androl 1978; 26 Goldstein I. Comment on pharmacotherapy of Impotence. In: 1: 345 ± 349. Lue TF (ed). World Book of Impotence, Chap. 13. Smith- 11 Zweifach BW, Lipowsky HH. Pressure-¯ow relations in blood Gordon: London, 1992, pp 194 ± 196. and lymph microcirculation. In: Renkin EM, Michel CC (eds). 27 Christ GJ, Brink PR, Melman A, Spray DC. The role of gap Handbook of Physiology, Section 2: The Cardiovascular junctions and ion channels in the modulation of electrical and System. Microcirculation, part 1, chap. 7, vol. 4. American chemical signals in human corpus cavernosum smooth Physiological Society: Bethesda, 1984, pp 251 ± 307. muscle. Int J Impot Res 1993; 5: 77 ± 96. 12 Segal SS. Cell to cell communication coordinates blood ¯ow 28 Fallon B. Intracavernous injection therapy for male erectile control. Hypertension 1994; 23: 1113 ± 1120. dysfunction. Urol Clin North Am 1995 22: 833 ± 845.

International Journal of Impotence Research