Journal of Neurology, Neurosurgery, and Psychiatry 1992;55:275-281

Neurological risk profile in organic erectileimpotence

E Kunesch, K Reiners, V Muller-Mattheis, T Strohmeyer, R Ackermann, H-J Freund

AbstractThirty men who presented with erectileimpotence to the urological departmentunderwent a thorough urological, angio-logical, and neurological examinationwith complementary neurophysiologicaltests of somatosensory and sympatheticand parasympathetic function. Most hadvascular and neurological abnormalities.Clinical findings and electrophysiologicaltests for autonomic dysfunction had thehighest yield of abnormal results. Nerveconduction studies and pudendal nervesomatosensory evoked potentials were farless informative. The lack of correlationbetween vascular and general neurolog-ical abnormalities emphasises thatpatients must be screened for both vas-cular and neurological dysfunction to pre-vent unrewarding vascular operation inimpotent men.

Heinrich-HeineUniversitit,MoorenstraBle S

W-4000 Dusseldorf,FRGNeurology ClinicE KuneschK ReinersH-J FreundUrology ClinicV Muller-MattheisT StrohmeyerR AckermannCorrespondence toDr Kunesch

Received 28 March 1991and in revised form 24 July1991.

Accepted 30 July 1991

IntroductionErectile impotence is defined as the persistentinability of the patient to obtain or maintain anerection suitable for vaginal penetration andthe subsequent coital act.' Erectile impotenceis commonly attributed to three main causes:vascular disease, neurological disturbances,and psychological factors. Erection requires a

sufficient arterial blood supply of the corporacavernosa arising from terminal branches ofthe internal pudendal artery2 and intact arter-iolar sphincter mechanisms, which either fillthe corpora cavernosa or shunt blood into theveins. By inflow of arterial blood into the penilesinoids these get distended and compress thedraining veins against the tunica albuginea.Thus venous outflow is reduced. In the cat,Semans and Langworthy3 were able to repro-duce erection, emission, and ejaculation in anormal sequence by stimulating parasympa-thetic, sympathetic, and somatic nerve fibressupplying the male sexual organs. Erection ismediated by efferent parasympathetic nervierigentes arising from the S2, S3, and S4segments of the spinal cord. Afferent somaticfibres of the pudendal nerve (S2, S3, and S4segments) may trigger erections on local stim-ulation. Stimulation of sympathetic nerves

(D12, LI, and L2 segments) is required foremission of seminal fluid and later detumes-cence. For expulsion of seminal fluid from theurethra, however, stimulation of somatic andautonomic nerve fibres of the pudendal nerveis necessary. In the absence of local stimula-tion, erections after cerebral stimulation may

be initiated via sympathetic pathways.4 Thesefindings indicate that normal male sexualfunction is critically dependent on the dynamicinterplay of the parasympathetic, sympathetic,and somatic nervous system. Diagnostic uro-logical evaluation of the erectile system hasbeen considerably improved by the introduc-tion of objective tests such as Doppler meas-urements of penile blood pressure and bloodflow,56 pharmacodynamic cavernosography,8pudendal arteriography,2 and the nocturnalpenile tumescence test.9 1oOnly during recent years have neurological

test methods for the assessment of erectilenerve dysfunction become available, com-plementing clinical neurological investigation.These include the pudendal somatosensoryevoked potentials" 12 and the electricallyinduced bulbocavernosus reflex"3-5 to screenfor the sensorimotor pathways and screeningmethods of autonomic dysfunction-sympa-thetic skin response,'6 30/15 test,'7 and respi-ratory heart rate variation.'8 19

In our study 30 consecutive patients whohad received a thorough urological investiga-tion, including vascular function tests wereinvestigated both clinically and by using anextended electrophysiological neurological testbattery. We assessed the prevalence of auto-nomic neuropathy or other neurological dis-ease and correlated these results to urovascularpathology.

Patients and methodsSUBJECTS AND CLINICAL INVESTIGATIONWe studied 30 men aged 21-82 years (mean 45years; interquartile ranges: 21-38, 39-46,47-54, and 55-82 years) complaining of com-plete erectile impotence for 12 months to fiveyears. All were referred outpatients from theurological department. Each patient had athorough medical history taken including vege-tative function and drug history. Physicalexamination included an assessment of neuro-logical, urogenital, and vascular abnormalities.Neurological examination placed specialemphasis on possible symptoms arising fromdysfumction of the spinal cord, including analand cremaster reflexes, and the peripheralsensorimotor and autonomic nervous system.Laboratory tests included blood cell count,blood glucose and lipids (cholesterol, tri-glycerids), liver and kidney fimction, and sexhormones. The table gives further details. Theresults in each category were tested for linearcorrelation with all other categories with Stat-View II statistical software (Abacus Concepts

275 on A

pril 21, 2021 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.55.4.275 on 1 April 1992. D

ownloaded from

Kunesch, Reiners, Miiller-Mattheis, Strohmeyer, Ackermann, Freund

Table Categories and tests evaluated in 30 men with erectile impotence

Category Test items Criteria for abnormal test result Criteria for abnormal category

Neuropathic risk Alcohol abuse > 70g Ethyl-alcohol/day for > 6factors months before examination

Diabetes Known for > 6 months One or more out of three itemsUraemia Known for > 6 months abnormal

Sensorimotor Clinical examination Depressed tendon reflexes, distalneuropathy sensory loss, distal muscle wasting

Motor nerve conduction (tibial Below mean (2SD) of 43 m/s (for Two or more out of three itemsnerve) 20 years) to 35 m/s (for 80 years) abnormal

at 330CSensory nerve conduction (sural Below mean (2SD) of 48 m/s (fornerve) 20 years) to 45 m/s (for 80 years)

at 350CAutonomic Sympathetic skin response Absent from foot (all patients withdysfunction absent responses from hand also

had no responses from foot)"6Respiratory heart rate variation Below age related normal Two or more out of three items

range:'9 2 for example, for 30 abnormalyears, 20/min; for 80 years, 5/min

30/15 Test < 1-03'7Pudendal SEP Latency Absent or latency above normal Considered abnormal only if

range (36 ms for 170 cm, 39 ms peripheral nerve conductionfor 185 cm) studies were normal

Bulbocavemosus reflex Latency Absent or latency above 37-2 ms Considered abnormal only ifperipheral nerve conductionstudies were normal

Vascular disease Doppler ultrasound of penileblood flowPenile cavernosogram One or more out of three testsArteriogram of pelvic arteries abnormal

Vascular risk factors Arterial hypertensionNicotine abuse > 10 Cigarettes/day for > 5 yearsDiabetes mellitus Two or more out of five itemsElevated blood lipids abnormalHistory of coronary heart disease

Serum levels of sex Estradiol > 80 pg/mlhormones Testosterone < 2-7 ng/ml One or more tests abnormal

FSH < I or> 14 U/mlProlactine > 14 5 ng/ml

Inc, Berkeley, GA, USA). All investigationswere carried out with informed consent andaccording to the Declaration of Helsinki(1975).

ELECTROPHYSIOLOGICAL INVESTIGATIONSThe 30/15 test was carried out with a conven-tional ECG apparatus. All other tests wereperformed with a Medelec MS 20 Electro-physiological System (Mystro: Medelec, Wok-ing, UK). If electrophysiological data werecompared with normal data established in theliterature, care was taken to adhere to theprocedures as reported.

Nerve conduction studies-Investigation ofnerveconduction velocities of the sural (antidromic)and tibial nerves was performed in the conven-tional way with bipolar surface electrodes. Thefilter band pass was 20 Hz-2 kHz for sensorynerve conduction measurements and 3 Hz-10 kHz for motor conduction measure-ments.The skin surface temperature from thedorsum of the foot was measured by means ofa thermoelement. The normal values of nerveconduction velocities given by Ludin20 weregenerally accepted and adjusted with respect toskin temperature (- 2 m/s for 1°C below thereference temperature indicated in the table).

Bulbocavernosus reflex (BCR)-The pudendalnerve was stimulated on the penis shaft by twosurface ring electrodes with the anode placeddistally at 80-1 50V (stimulus duration100-500 ps) as described by Tackmann andPorst.21 The response was recorded by meansof bipolar surface electrodes (Toennies) mid-

way between the scrotum and the anus. Theanode was placed at the anterior superioriliacal spine. The sweep speed was set to 10 ms/div and the filter band pass to 10 Hz-3 kHz: 32sweeps were averaged. The latency was deter-mined from the first negative deflection of theresponse (see figure 1). The response wasconsidered to be abnormal if the onset latencywas longer than 37-2 ms. Normal latencies arenot consistently age dependent so no correc-tion for age was necessary.

Pudendal SEP-Stimulation ring electrodeswere placed at the shaft of the penis (anodedistally) as described above. The somatosen-sory evoked potentials were recorded from thescalp with the active needle electrode (imped-ance 3Q) placed subcutaneously 2 cm behindCz. The reference needle electrode was inser-ted at Fz; 300 to 800 sweeps were averaged,and each test was repeated up to three times inorder to assess the reproducibility of theevoked responses. The sweep speed was set to10 ms/div and the filter band pass from 20 Hz-2 kHz. The latency of the response was meas-ured from the P40 wave (figure 1) according toHaldeman et al" 12 and Tackmann andPorst.2' Response onset latencies longer thanthe body height dependent control limits givenby Tackmann and Porst2" were consideredabnormal.

Respiratory heart rate variation (HRV)-Themethod has been described in detail in aprevious article.'9 Briefly, the built-in singlefibre EMG-software program of the MedelecMS 20 electrophysiological system was used to

276 on A

pril 21, 2021 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.55.4.275 on 1 April 1992. D

ownloaded from

Neurological risk profile in organic erectile impotence

record 50 RR intervals (ECG) during consec-utive deep inspiration and expiration cycles(6/min). Conventional ECG plate electrodeswere placed on the right foot and on the left orright arm and connected to the positive andnegative input sockets of the MS 20 differentialpreamplifier. The sweep was triggered by theQRS complex, and the ECG activity wasrecorded with a gain of 200 or 500 mV/div, afilter band pass from 1-50 Hz, and a sweepduration of 1-2s depending on the heart rateso that QRS pairs consisting of the triggeringand the subsequent QRS complex could bemonitored on the same sweep. The variation ofthe time interval between consecutive QRScomplexes was assessed for 50 QRS pairs.Variation was expressed as the differencebetween the frequencies calculated from theshortest and the longest RR interval foundwithin an analysis period of approximately 45s.In our experience HRV is almost linearly agedependent to the same extent as described byLow et al.22 The table shows our lower normallimits for the extreme ages of our patients.

Sympathetic skin response (SSR)-The biphasicskin DC-potential shift in response to electricalshock stimuli was evaluated on the right handand foot as described by Shahani et al."6Toennies surface electrodes were attached tothe volar and dorsal side of the hand and theplantar and dorsal side of the foot. The patientwas instructed to relax in order to avoid artifactcontamination by tonic muscle activation.Supramaximal electrical shock stimuli wereapplied randomly to the median or tibial nerveof the contralateral side. Care was taken toapply the next stimulus no earlier than 20safter the previous one in order to avoidhabituation of the response. Sweep speed wasset to 500 ms/div and amplitude gain to 100,200, or 500 pV/div as suitable. The filter bandpass was set to 3 Hz-3 kHz. The response wasconsidered to be present if clear biphasicresponses could be obtained at latencies of1-3s during several trials. If necessary, the footwas warmed up to skin surface temperaturesabove 28°C either by bathing the foot in warmwater or by applying infrared light. The testwas considered to be abnormal if responseswere consistently absent."6

30/15 Test-Heart rate variation in response toorthostasis was assessed as described by Ewinget al.17 The patient lay supine and relaxed andwas instructed to take an upright standingposition as fast as possible. The RR intervalbetween the 15th to 16th and between the 30thto 31st QRS complex after attaining theupright position was measured and the ratio(30th/31st divided by 15th/16th interval) wascalculated. According to Ewing"7 the normalratio is above 1 03. There is a tendency forolder patients to have lower values than young-er ones, but in our experience even olderhealthy people retain a ratio above 1 03.

UROLOGICAL INVESTIGATIONThis included a thorough clinical urologicalinvestigation, invasive angiography of the main

pelvic arteries, and cavernosography of penileveins as well as an assessment of the penilearterial blood flow by means of Dopplerultrasound techniques and of sex hormoneserum levels.

ResultsCLINICAL NEUROLOGICAL INVESTIGATIONThe whole test battery was performed in all 30patients. Two patients were normal in all testsbut had major problems that made a psycho-logical disease the most likely factor and theywere excluded from further evaluation. Clin-ical neurological findings were abnormal in 15of the 28 remaining patients. Most showedsigns of sensorimotor as well as autonomicneuropathy. No patient had signs of a cerebrallesion or of pituitary or hypothalamic dysfunc-tion. One patient had a history of a spinetrauma without residual deficits, and theoccurrence of impotence was temporally unre-lated to the trauma.

TYPICAL RESULTS OF ELECTROPHYSIOLOGICALINVESTIGATIONSThe upper panel of figure 1 shows the patternof the bulbocavernosus reflex in three patients

BULBOCAVERNOSUS - REFLEX

PUDENDAL - SEPnoymAdII J5pV

I stim.II I I I I I I

0 10 20 30 40 50 60 70 80 90 100Ms

Figure 1 Examples of normal, delayed, and absentreponses for bulbocavernosus reflex measurements andpudendal SEP. For all cases tw trials with 32(bulbocavernosus reflex) and 300-500 averaged sweeps(pudendal SEP) are illustrated. Points at which latencieswere measured indicated by arrows. Negative is up.

277 on A

pril 21, 2021 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.55.4.275 on 1 April 1992. D

ownloaded from

Kunesch, Reiners, Muller-Mattheis, Strohmeyer, Ackermann, Freund

with erectile impotence. The response ofpatient 1 (two upper traces) was normal, thatof patient 2 was delayed (traces 3 and 4), andin patient 3 no response was obtained (traces 5and 6). Examples of pudendal SEP investiga-tions are shown in the lower panel in figure 1.The pudendal SEP study of patient 1 (twoupper traces) was normal, whereas in patient 3no cortical response could be recorded (twolower traces). For the assessment of autonomicnerve function the 30/15-test, the HRV and theSSR were used. Examples are illustrated infigure 2. The upper two traces of both theupper and lower panel (patient 4) show normalSSRs recorded from the right hand and foot.Response latencies after the electrical stimuluswere approximately 1 2-2s. The lower traces(patient 5) show examples of absent responses.The responses from hand and foot remainedabsent in several trials. Examples of HRVrecordings are illustrated in figure 3. Patient 4had a normal HRV of 29/min (left panel),whereas the result of patient 5 (right panel)was abnormal (7/min).

INCIDENCE OF ABNORMAL FINDINGS

Clinical neurological findings, SSR in the foot,and the 30/15 test were often abnormal (16 outof 30 (53%), 16 (53%), 17 (57%)) whereas thepudendal SEP was abnormal only in three(figure 4).Only one out of eight patients withabnormal bulbocavernosus reflexes had noother pathological neurological findings orneuropathic risk factors. Urological examina-tion showed that angiography was abnormal in21 patients (angiography of pelvic arteries ineight, arterial penile blood flow in 20) andcavernosography in eight, but serum hormonelevels were abnormal in only five (figure 4b).Polyneuropathy and vascular risk factors werefound in half (figure 4c and d). Pathological

SYMPATHETIC SKIN RESPONSEhandnormal 4 I2)1V

t stim.abnormal

tstim. J1ooV

foot

normal Iwo/IV

tstim.abnormal

tstim. 10011V

0 1 2 3 4 s 5

Figure 2 Examples of two patients with normal andabsent sympathetic skin responses recorded in right handand foot evoked by electrical stimuli appliedcontralaterally. For patient with absent responses only oneout of several sweeps iUlustrated. Stimulation andbeginning of responses indicated by arrows. Note long timebase.

findings in neurological and urovascular cate-gories were found in 19 patients. Sevenpatients were affected only urologically, andthree only neurologically. The upper panel offigure 5 illustrates the distribution of normaland abnormal findings among our patients forclinical neurological examination, autonomictests and nerve conduction velocity studies. Sixout of 13 patients with more than one abnor-mal autonomic test result were normal onclinical examination. Eleven out of 13 patientshad signs of both sympathetic and para-sympathetic dysfunction. The coincidence ofabnormal findings in both SSR and 30/15 testwas high (figure 5, lower panel). Autonomicdysfunction was more often observed thansensorimotor neuropathy (13 v nine patients).None of the patients with sensorimotor neu-ropathy would have been missed even if nerveconduction studies had not been performed asin all these patients clinical neurological exam-ination had already revealed abnormal find-ings. No significant correlations were foundwithin the 95% confidence limits between thecategories listed in the table. While most (24)patients were aged between 24 and 64 yearsonly three out of 30 were 65 years or older.The number of pathological findings in allcategories varied positively but not signifi-cantly (p > 0 05) with age.

DiscussionEXPERIMENTAL LIMITATIONSThere are some limits in the interpretation ofour results. Firstly, our patients were referredfrom the urological outpatient department andmay, owing to a preselection bias for urologicalproblems, not be representative of the popula-tion of impotent men. Even in this urologicalselection of patients, however, a high incidenceof urological and neurological abnormalitieswas found. The number of abnormal neuro-logical findings would probably have been evenhigher in a sample of neurological patients-for example, multiple sclerosis is a commoncause of erectile impotence. A structuredpsychological interview and more specific psy-chological tests like the Minnesota multiphasicpersonality inventory were not used. Thereforepsychological problems which were not appar-ent in taking the case history but contributedto the erectile impotence may have beenmissed. In urological practice, a confirmativeanswer to the question ofwhether there are stillspontaneous erections in the morning is takento rule out an organic cause and therefore tosuggest a psychological background of erectileimpotence. This was the case in two of ourpatients, who, in addition to normal somatictest results, had positive evidence of majorpsychological problems. We could not deter-mine a simple cause and effect relationshipbetween the organic abnormalities found andthe presence of erectile impotence. Significantcorrelations among the categories of our testbattery could not be established. A set ofdifferent tests is therefore necessary to screenfor abnormalities unrelated to each other butadding up to erectile dysfunction.

278 on A

pril 21, 2021 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.55.4.275 on 1 April 1992. D

ownloaded from

Neurological risk profile in organic erectileimpotence27

normal

78/min

RESPIRATORY HEART RATE VARIATION

abowmals

49/mmn

- ISOOieumV -I 500 tV0msIn lOOms

Figure 3 Examples of normal (left panel) and abnormal (right panel) heart rate variation measurements. In top tracesSO sweeps are superimposed. Position of cursors indicate minimal and maximal RR intervals, from which heart rate/mmnwas cakculated. Nine consecutive sweeps illustrating normal variability of RR interval in left in contrast to right panelare shown below.

A

40

a1 30

a40

;~~~,~ ,vf*

B

a

D

IaU

Figure 4 Percentage of abnormal findings for neurological (A) and urovascular (B)

investigations and for risk factors for vascular disease (C) and polyneuropathy (D). Clin

Exam: clinical examination; NC Studies: nerve conduction studies; HRV~respiratoryheart rate variation; SSR: sympathetic skin response; ud SEP:- pudendal somawosensoryevoked potentials; BC Reflex: bulbocavernosus reflex; Diab mefll: diabetes mellitus; Cor

Heart Dis: coronary heart disease.

PREVALENCE OF NEUROLOGICAL ABNORMALITESOur patients varied widely in age so thequestion of whether ageing itself should beregarded as a risk factor for erectile impotenceis relevant. Previous studies from 194823 con-cluded that the incidence of impotence ingeneral is as low as 1-9% in 40 year old men'

~ -2 (psychogenIc cause)

28 PATIENTS

NMEUROLOGCA 15 aibnormal 13 normal

7

AUTONOMI v' 5nra

(FU 1~~131 abn.13 5nra

5BX4~~~ 11

EECTROPIIYOTL 9 abnormal 19 normal

30/5Test Newt Rate Vwarlsto

Symup. Skin Response

Figure S Numbers ofpatients with normal andabnormal findings in clinical neurological investigation,tes.ts for autonomic dysfuinction, and nerve conductionstudies illustrated by flow chart in upper panel. Lowerpanel shows coincidence of abnormal findings among threeautonomic tests (30/15 test, sympathetic skin response,heart rate variation).

279

oo,..te

on April 21, 2021 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.55.4.275 on 1 A

pril 1992. Dow

nloaded from

Kunesch, Reiners, Mililer-Mattheis, Strohmeyer, Ackermann, Freund

but 25% in 65 year old men. These figuresobviously relate to a sample which was notadequately tested for urological, vascular, orneurological dysfunction. In our series thenumber of pathological results in all categoriesof neurological and urovascular dysfunctionincreased with age, although the correlationdid not reach the 95% level of significance.Therefore, the higher proportion of impotenceamong older men probably results from ahigher incidence of risk factors and neuro-logical or urovascular dysfunction in older menrather than from ageing itself.

Blaivas et al24 found abnormalities of penileblood pressure in 67%, neurological abnor-malities in 41%, psychiatric disorders in 35%,and a combination of abnormalities in 48% oftheir patients. The occurrence of neurologicalabnormalities was thus comparable to ourresults, even though they did not investigateautonomic nerve function electrophysiolog-ically. In contrast, we were unable to confirmfindings of Tackmann and Porst,21 who moreoften observed abnormalities of the bulboca-vernosus reflex (BCR) (abnormal findings in62 out of 122 patients tested) and of thepudendal SEP (abnormal in 31 out of 108patients tested). For the BCR, similar resultsto ours were found by Wabrek.25 Our dataindicate that investigating BCR does not addgreatly to the detection of neurological deficitsand is therefore not necessary in the neuro-logical workup.Our study confirmed the previous finding

that diabetes mellitus is a major risk factor forerectile impotence.26 Whether the decisiveabnormality in impotent diabetics is vascularocclusive disease26 or diabetic neuropathy isstill debatable.27 The more common abnor-malities of the BCR in diabetics28 highlightspudendal neuropathy as the more importantfactor. From our data on a urological sample ofpatients with erectile impotence we can drawthe following conclusions. Firstly, in mostimpotent men both neurological and urovas-cular abnormalities contribute to erectile dys-function. In contrast CNS dysfunction is rarelya cause of erectile impotence. Secondly, withmore than 50% of pathological findings foreach category, history and clinical neurologicalevaluation of the peripheral sensorimotor andautonomic nervous system (NS) have thehighest yield of pathological neurologicalresults. The SSR in the foot was abnormal in asimilar proportion ofpatients but it is known tobe rather unspecifically abnormal in elderlypatients.22 Thirdly, nerve conduction studiesand BCR were abnormal in a third of thepatients while pudendal SEPs did not provideadditional evidence in identifying erectileimpotence patients with neurological prob-lems. All patients with abnormal nerve con-duction results had neuropathic findings onclinical examination. This suggests that if aneuropathy causes or contributes to erectileimpotence it must be clinically overt and notjust detectable electrophysiologically. Finally,autonomic tests (30/15 test and HRV forevaluation of the parasympathetic NS and SSRin the hand for the sympathetic NS) were

abnormal in 64% of our patients. In mostpatients with signs of autonomic dysfunctionabnormal results were found in both sym-pathetic and parasympathetic tests. More than20% of the patients show abnormal autonomictests in the absence of motor and somatosen-sory dysfunction.

DIAGNOSTIC STRATEGY AND THERAPEUTICAL IMPLI-

CATIONS

Our results indicate that it is essential to usetest screening for autonomic dysfunction. Assensorimotor polyneuropathy may mostly bedetected clinically, nerve conduction studiesare not always necessary. In addition, it is alsounnecessary to perform a pudendal SEP ineach patient. None of our patients with neuro-logical involvement would have been missed ifpudendal SEP had not been performed. Wetherefore suggest the following sequence ofinvestigations in a patient with erectile impo-tence: exclude CNS disorders by clinicalexamination; evaluate peripheral NS clinically;and test autonomic sympathetic and para-sympathetic NS. With erectile impotence beingcommonly (68%) associated with both uro-logical and neurological abnormalities treat-ment ofurovascular dysfunction-for example,by revascularisation-may not be sufficient ormay even not be indicated in patients withsevere neurological disturbances. Also, psycho-logical therapy alone will obviously fail. Neu-rological dysfunction will remain undetected ifappropriate tests of sensorimotor and auto-nomic fumction are not performed. Clearlymost patients with erectile dysfunction haveorganic deficits. Therefore it is not justified tobrand them as having a psychogenic disease.

This work was supported by the Deutsche Forschungsge-meinschaft SFB 194, (A2).

1 Krane RJ, Siroky MB. Clinical neuro-urology. Boston: LittleBrown and Company. 1979.

2 Ginestie JF, Romier A. Radiologic exploration of impotence.The Hague/Boston /London: Martinus Nijhoff MedicalDivision,1978.

3 Semans JH, Langworthy OR. Observations on the neu-rophysiology of sexual functions in the male cat. J Urol1938;40:836-46.

4 Root WS, Bard P. Mediation of feline erection throughsympathetic pathways with some remarks on sexualbehavior after deafferentation of genitalia. AmerJ Physiol1947;151:80-90.

5 Abelson D. Diagnostic value of the penile pulse and bloodpressure: a doppler study of impotence in diabetics. J Urol1975;113:636-9.

6 Engel G, Burnham SJ, Carter MF. Penile blood pressure inthe evaluation of erectile impotence. Fertl Steril 1978;30:687-90.

7 Lue TF, Hricak H, Schmidt RA, Tanagho EA. Functionalevaluation of penile veins by cavernosography in papaver-ine-induced erection. J Urol 1986;135:479-82.

8 Juenemann KP, Lue TF, Fournier Jr GR, Tanagho EA.Hemodynamics of papaverine- and phentolarnine-induced penile erection. J Urol 1968;136:158-63.

9 Karacan I. Clinical value of noctumal erection in theprogress and diagnosis of impotence. Med Aspects HumSex 1970;4:27-34.

10 Karacan I, Salis PJ, Ware C, et al. Nocturnal peniletumescence and diagnosis in diabetic impotence. Amer JPsychiatry 1978;135:191-7.

11 Haldeman S, Bradley WE, Bhatia NN, Johnson BK.Pudendal evoked responses. Arch Neurol 1982;39:28R-3.

12 Haldeman S, Bradley WE, Bhatia NN. Evoked responsesfrom the pudendal nerve. J Urol 1982;128:974-80.

13 Ertekin C, Reel F. Bulbocavernosus reflex in normal menand in patients with neurogenic bladder and/or impo-tence. J Neurol Sci 1976;28:1-15.

14 Rushworth G. Diagnostic value of the electromyographicstudy of reflex activity in man. Electroenceph Clin Neu-rophysiol (Suppl) 1967;25:65-73.

280 on A

pril 21, 2021 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.55.4.275 on 1 April 1992. D

ownloaded from

Neurological risk profile in organic erectile impotence

15 Dick HC, Bradley WE, Scott FB, TImm GW. Pudendalsexual reflexes. Electrophysiologic investigations. Urology1974;3:376-9.

16 Shahani BT, Halperin JJ, Boulu P, Cohen J. Sympatheticskin response: a method of assessing unmyelinated axondysfunction in peripheral neuropathies. Neurol Neuro-surg Psychiatry 1984;47:536-42.

17 Ewing DJ, Campbell IW, Murray A, Neilson JMM, ClarkeBF. Immediate heart-rate response to standing: simpletest for autonomic neuropathy in diabetes. Brit Med1978;1: 145-7.

18 Murray A, Ewing DJ, Campbell IW, Neilson JMM, ClarkeBF. RR interval variation in young male diabetics. BrHeart 1975;37:882-5.

19 Kunesch E, Reiners K A simple method for the routineassessment of heart-rate variation in autonomic neuro-pathy. Electromyogr Clin Neurophysiol 1989;29:293-7.

20 Ludin HP. Praktische Elektromyographie. Stuttgart: Enke-Verlag, 1988.

21 Tackmann W, Porst H. Diagnostik neurogener Potenzstor-ungen mit Hilfe des Bulbocavernosusreflexes und soma-tosensorisch evozierter Potentiale nach Stimulation desNervus pudendus. Nervenarzt 1987;58:292-9.

22 Low PA, Opfer-Gehrking TL, Proper CJ, Zimmerman I.

The effect of aging on cardiac autonomic and post-ganglionic sudomotor function. Muscle Nerve1990;13:152-7.

23 Kinsey AC, PomeroyW, Martin CE. Age and sexual outlet.In: Kinsey AC, Pomeroy WB, Martin CE, eds. Sexualbehavior in the human male. Philadelphia: Saunders,1948. 218-62.

24 Blaivas JG, O'Donnell TF, Gottlieb P, Labib KB. Compre-hensive laboratory evaluation of impotent men. Urol1980;124:201-4.

25 Wabrek AJ. Bulbocavernosus reflex testing in 100consecutive cases of erectile dysfunction. Urology1985;25:495-8.

26 Lehmann TP, Jacobs JA. Etiology of diabetic impotence.Urol 1983;129:291-4.

27 Faerman I, Glocer L, Celener D, Jadzinsky M, Fox D,Rapaport M. Impotence and diabetes. Diabetes1974;23:971-6.

28 Ertekin C, Akyiirekli 0, Nurettin A, Turgut H. The value ofsomatosensory-evoked potentials and bulbocavemosusreflex in patients with impotence. Acta Neurol Scand1985;71:48-53.

281

on April 21, 2021 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.55.4.275 on 1 A

pril 1992. Dow

nloaded from