Sleep-Related Nocturnal Erections and Erections During Midazolam-Induced Sedation in Healthy Young Men

ORIGINAL ARTICLE Sleep-related nocturnal erections and erections during midazolam-induced sedation in healthy young men

YS Song1, ES Song2, KH Lee1, YH Park1, WC Shin3 and JH Ku4

1Department of Urology, Soonchunhyang School of Medicine, Seoul, Korea; 2Department of Obstetrics and Gynecology, Inha School of Medicine, Incheon, Korea; 3Department of Neurology, Kyunghee School of Medicine, Incheon, Korea and 4Department of Urology, Seoul Veterans Hospital, Seoul, Korea

This study was performed to evaluate the characteristics of penile erection during midazolam- induced sedation after nocturnal sleep deprivation (NSD) and to determine the effect of NSD on erectile episodes in healthy, sexually functional young men. This procedure might possibly prove to be a brief office-based method of assessing whether erectile dsyfunction is psychogenic or biogenic. Nineteen volunteers between the ages of 20 and 29 years participated in this study. We measured the morning penile erection after midazolam (3–5 mg) administration intravenously and all subjects completed 42 tests. Of 42 test, 28 tests revealed erectile episodes, wereas no erectile episodes were observed in 14 tests. Nocturnal sleep deprivation rate was significantly higher in tests with erectile episodes than in tests without erectile episode (P ¼ 0.030). Test order or duration of test was not different between two test results. Number of erectile episodes (r ¼ 0.374, P ¼ 0.015), tip radial rigidity (r ¼ 0.412, P ¼ 0.007), base radial rigidity (r ¼ 0.366, P ¼ 0.017) and tip tumescence (r ¼ 0.447, P ¼ 0.003) correlated with the degree of NSD. When we determined whether NSD was discriminative with regard to erectile episodes, the area under the receiver operating characteristic curve was calculated at 0.705 (95% confidence interval, 0.527–0.883; P ¼ 0.032) for the possibility of erectile episodes. Nocturnal sleep deprivation might recover the inhibited rapid eye movement sleep during midazolam-induced sedation. Our findings suggest that erection monitoring during midazolam-induced sedation after NSD may be convenient. However, validation of midazolam- induced morning penile tumescence monitoring with a large population is mandatory. International Journal of Impotence Research (2006) 18, 522–526. doi:10.1038/sj.ijir.3901463; published online 16 March 2006

Keywords: penis; penile erection; nocturnal penile tumescence; sleep-related erections; sedation

Introduction across time. Although clinical sleep-related penile erection testing is less commonly applied today Sleep-related penile erections occur in all healthy, with the advent of minimally invasive medical potent men in close temporal association with rapid therapies for erectile dysfunction, as an objective eye movement sleep. Nocturnal penile tumescence measure of erectile function, sleep-related penile monitoring provides a physiologic, objective, diag- erection recording for research provides a precise nostically useful technique for evaluating erectile technique for examining the mechanisms of erection capacity because sleep-related penile erections are and is still conducted to resolve legal disputes. naturally occurring, physiologically normal, occur However, in the standard sleep-related penile erec- several times on any given night, and are stable tion evaluation, patients sleep in a hospital setting for two or three nights, reporting to the facility 1–2 h before their normal bedtime. Thus, practical drawbacks of sleep-related penile erection Correspondence: Dr JH Ku, Department of Urology, Seoul monitoring include the cost and inconvenience.1 Veterans Hospital, 6-2, Doonchon Dong, Kangdong Ku, To date, a wide assortment of alternative techniques Seoul 134-791, Republic of Korea. emerged,2–4 but these methods have met with only E-mail: [email protected] 5 Received 12 January 2006; revised 19 January 2006; limited success. accepted 6 February 2006; published online 16 March Midazolam has been most widely used as a 2006 sedative premedication because it has lots of Sedation-induced erections YS Song et al 523 advantages, such as a short half-life, a faster onset of Study design sedation and an excellent sedative hyponotic effect We measured the subjects’ sleep-related penile without any particular side effects, such as vascu- erection monitoring over a consecutive three nights litis. Several classes of drugs can adversely affect using Rigiscan-Plust (DacoMed, Minneapolis, MN, erectile function and sleep-related penile erections. USA). In addition, all night continuous polysomno- Some drugs act directly on sleep-related penile grams were recorded according to standard practice. erections while others act indirectly by altering On each night, technicians attached sensors to the rapid eye movement sleep. Antiandrogens decrease subjects to monitor penile base and tip circumfer- sleep-related penile erections without necessarily ence changes. We recorded data from five electro- affecting rapid eye movement sleep. The beta- encephalography channels, two electrooculography blocker propanolol can decrease sleep-related penile channels, one electromyography, three electroder- erections in some men.6 Tricyclic antidepressants, mal activity channels (EDA), one electrocardiogra- monoamine oxidase inhibitors, and selective sero- phy channel, and one respiratory airflow channel. tonin reuptake inhibitors all potentially suppress The EDA channels (medical phalanx of middle rapid eye movement sleep and are known to finger of left hand, medial phalanx of middle finger adversely affect sexual function in some indivi- of right hand, penis) monitored skin resistance duals. Since rapid eye movement sleep may be responses. inhibited during midazolam-induced sedation,7,8 The morning penile erection monitoring was midazolam act directly or indirectly on sleep-related performed up to third session until it showed penile erection. erectile episodes. Medications such as nicotine, Electroencephalography patterns of the morning alcohol, caffeinated beverages were avoided in all after nocturnal sleep deprivation (NSD) resembled volunteers for at least 12 h before to the morning sleep stages from the second half of the night, with penile erection monitoring in order to maximize the rapid eye movement sleep dominating.9 Nocturnal likelihood that they would able to sleep. To resume sleep deprivation resumed their previously inter- rapid eye movement sleep from the previous night’s rupted rapid eye movement cycles at the next sleep interrupted sleep, deprivation of their normal onset. Thus, it may be hypothesized that NSD may night’s sleep before the morning penile erection recover the inhibited rapid eye movement sleep monitoring were recommended to all subjects. during midazolam-induced sedation. This study Morning penile tumescence recording and poly- was performed to evaluate the characteristics of somnographic recording were started between 0700 penile erection during midazolam-induced seda- and 1000 hours and recording duration was 1–3 h. tion after NSD in healthy, sexually functional Subjects slept in private, electrically shielded, young men and to identify the effect of NSD on sound-attenuated, temperature-controlled bedrooms. penile erections during midazolam-induced seda- Midazolam was administered intravenously to the tion in this population. The purpose of this study subjects at doses of 3–5 mg. The initial intravenous was to assess the possibility of the erection dose was 1 mg (no more than 0.03 mg/kg) given monitoring during midazolam-induced sedation slowly over at least 2 min, titrating to the desired as an alternative to sleep-related penile erection level of sedation. An intravenous dose of 0.02– testing. 0.03 mg/kg was repeated at 2-min intervals while continually monitoring for the appropriate level of sedation. A total intravenous dose of more than 5 mg Patients and methods was not necessary for all examinations. After the administration of midazolam, the presence of com- Patients plications or side effects including apnea, oxygen From March to August of 2005, 19 sexually func- desaturation, autonomic movement, chest pain, tional and active young adult men between the ages arrythmia, injection in situ pain and phlebitis were of 20 and 29 years (mean: 25.970.7 years) partici- examined. The criterion of discharge was that pated in this study. A disorder-free medical and subjects have fully recovered orientation in time sexual history and normal erectile functions were and space with vital signs within the normal range. the only inclusion criteria. Subjects who had severe In this study, we did not use flumazenil, the cardiovascular disease, pulmonary disease, a antidote for midazolam for the rapid recovery of chronic alcohol history or drug abuse were not consciousness and orientation. included in the study. We also excluded subjects taking medications such as erythromycin, verapa- mil, diltiazem, itraconazole and ketoconazole, which are capable of drug interactions with the Statistical analysis benzodiazepine class of midazolam specifically. The survey responses were coded and analyzed Each subjects provided informed consent and using descriptive statistics, which are reported as studies were approved by institutional review the median with 25th to 75th percentiles (quantita- boards at our medical center. tive variables) or as the number and percentage

International Journal of Impotence Research Sedation-induced erections YS Song et al 524 (qualitative variables). The statistical analysis was erectile episodes during midazolam-induced seda- carried out using Mann–Whitney U-test for contin- tion and all sessions were less than 3 h in duration. uous data and w2 test for categorical data. Spear- Aforementioned side effects of midazolam were not man’s correlation coefficients were determined to detected in the present study. Of 19 subjects, erectile evaluate the associations with the degree of sleep episodes were noted in 18 subjects. Of 42 tests, 28 deprivation and the data regarding erectile episodes. tests revealed erectile episodes, while no erectile We attempted to determine whether or not sleep episodes were observed in 14 tests. The median deprivation was discriminative for erectile episodes number of sessions needed to observe erectile during midazolam-induced sedation, using the area episodes was 1 (range 1–3). In test with erectile under the receiver operating characteristic (ROC) episodes, rapid eye movement sleep was observed curve. The area under this curve is generally and erections with good quality were present during considered to be an appropriate parameter for the midazolam-induced sedation. The results are shown summarization of the overall discriminative or in Table 1. diagnostic value of a model, and can range from Table 2 shows the characteristics according to 0.5 (flipping a coin, a useless model) to 1.0 (perfect erectile episodes during midazolam-induced seda- discrimination).10 The closer the area under the tion. Sleep deprivation rate was significantly higher ROC curve approached to 100% (i.e. the more the in tests with erectile episodes than in tests without ROC curve approached the upper left corner), erectile episode (66.0 versus 50.2%, P ¼ 0.030). Test the greater the predictive power of the model is order or duration of test was not different between considered to be. A value exceeding 0.7 is inter- two test results. preted as reasonable, and a value in excess of 0.8 is Spearman’s correlation coefficients were per- considered to be good.11 A 5% level of significance formed to evaluate the associations between the was used throughout, and all statistical tests were sleep deprivation and erection. Among parameters, two sided. The statistical analyses were performed number of erectile episodes (r ¼ 0.374, P ¼ 0.015), tip using a commercially available program, SPSS 10.0 radial rigidity (r ¼ 0.412, P ¼ 0.007), base radial (SPSS Inc., Chicago, IL, USA). rigidity (r ¼ 0.366, P ¼ 0.017) and tip tumescence (r ¼ 0.447, P ¼ 0.003) correlated with the degree of sleep deprivation. However, duration of erection and base tumescence did not correlate with the Results degree of sleep deprivation. Table 3 contains the correlation coefficients between the sleep depriva- Full erection was observed in sleep-related penile tion and the data of erectile episodes. erection monitoring in nocturnal penile tumescence When we determined whether sleep deprivation monitoring. All 19 subjects completed 42 tests for was discriminative with regard to erectile episodes,

Table 1 Test results during midazolam-induced sedation

Subject no. Test order

First Second Third

1 125 min/50%/erection 103 min/50%/no erection 114 min/50%/erection 2 126 min/50%/erection 110 min/50%/erection 3 136 min/25%/no erection 95 min/25%/no erection 107 min/50%/no erection 4 128 min/25%/erection 122 min/37%/no erection 5 75 min/50%/no erection 30 min/75%/no erection 106 min/63%/erection 6 68 min/90%/erection 7 175 min/90%/erection 144 min/83%/erection 8 129 min/88%/no erection 137 min/100%/no erection 136 min/78%/erection 9 168 min/33%/erection 162 min/33%/no erection 10 163 min/0%/no erection 191 min/20%/no erection 148 min/67%/erection 11 150 min/50%/erection 182 min/66%/erection 12 187 min/100%/erection 117 min/75%/erection 13 170 min/50%/erection 133 min/33%/erection 14 97 min/100%/erection 101 min/90%/erection 15 142 min/66%/erection 101 min/43%/erection 16 114 min/50%/no erection 123 min/75%/erection 17 124 min/86%/erection 123 min/57%/erection 18 149 min/67%/erection 145 min/50%/erection 19 137 min/63%/no erection 139 min/90%/erection

Data presented are test time/sleep deprivation/test results.

International Journal of Impotence Research Sedation-induced erections YS Song et al 525 Table 2 Characteristics according to erectile episodes during midazolam-induced sedation

Total Erectile episodes No of erectile episodes

Number of tests 42 28 14 Number of subjects 19 18 9 Sleep deprivation (%)a 53.8 (48.3–79.3) 66.0 (50.0–85.3) 50.0 (25.0–66.0)

Number of test order (%)b First 19 (100.0) 13 (68.4) 6 (31.6) Second 18 (100.0) 11 (61.1) 7 (38.9) Third 5 (100.0) 4 (80.0) 1 (20.1) Duration of test (min)c 128.5 (109.3–148.3) 130.5 (115.5–148.8) 125.5 (101.0–143.3) Number of test needed to induce erectile episodes — 1 (1–2) —

Number of erectile episodes (%) 0 14 (33.3) — 14 1 12 (28.6) 12 — 2 14 (33.3) 14 — 3 2 (4.8) 2 — Duration of erection (min) 15.8 (0.0–30.3) 23.8 (15.6–41.3) — Tip radial rigidity (%) 44 (0.0–63.3) 58.5 (43.5–65.8) — Base radial rigidity (%) 61 (0.0–77.5) 72.5 (60.5–81.8) — Tip tumescence (cm) 1.4 (0.0–1.9) 1.8 (1.4–2.0) — Base tumescence (cm) 2.2 (0.0–2.7) 2.5 (2.1–3.0) —

Data presented are median (25th–75th percentiles) or number (%). a P ¼ 0.030 by Mann–Whiney U-test. b P ¼ 0.874 by Armitage test. c P ¼ 0.348 by Mann–Whiney U-test.

Table 3 Correlation coefficients between the degree of sleep sleep-related penile erection characteristics of deprivation and the data regarding erectile episodes patients even in vegetative state are similar to those of normal individuals.14 Although reports on sleep- rP-value related penile erections in mammalian species different from humans are extremely scanty, the Number of erectile episodes 0.374 0.015 Duration of erection 0.199 0.206 occurrence of sleep-related penile erections during Tip radial rigidity (%) 0.412 0.007 rapid eye movement sleep has also been reported in Base radial rigidity (%) 0.366 0.017 rats.15 However, paradoxical sleep erectile events Tip tumescence (cm) 0.447 0.003 may be not a universal phenomenon of mammalian Base tumescence (cm) 0.228 0.147 physiology. Although microinjection of carbachol, a cholinergic agonist, into certain region of the pontine tegmentum in rats induced rapid eye the area under the ROC curve was calculated at movement sleep-like state,16 carbachol-induced 0.705 (95% confidence interval, 0.527–0.883; rapid eye movement sleep was not accompanied P ¼ 0.032) for the possibility of erectile episodes by penile erections in rats.17 In addition, penile (i.e., reasonable discrimination). erections were not observed during paradoxical sleep in the armadillo Chaetophractus villosus.18 Midazolam has a high affinity for the benzodiazepine receptor in the central nervous system, with Discussion in vitro data demonstrating approximately twice the affinity of diazepam.19,20 The amino-acid neuro- The comparative study of penile behavior during transmitter gamma-aminobutyric acid (GABA) must sleep in different species might be particularly be present for the benzodiazepine to elicit a rewarding for the understanding of the functional response21 and benzodiazepines enhance the inhi- organization of the brain in that physiologic state. bitory action of the GABA.22 The actions of Although the neuronal control of rapid eye move- benzodiazepines do not involve the synthesis, ment sleep-related penile erections remains un- release, or altered metabolism of GABA but rather known, it has been suggested that the areas in the potentiate the inhibitory actions of GABA by forebrain may play an important role in rapid eye augmenting the flow of chloride ions through ion movement sleep-related penile erections.12 In the channels. The increase flux of chloride ions into the forebrain, the lateral preoptic area has been cell decreases the ability of the cell to initiate an implicated in sleep-related penile erections.13 The action potential.23

International Journal of Impotence Research Sedation-induced erections YS Song et al 526 In contrast to full night sleep-related penile 5 Allen R, Brendler CB. Snap-gauge compared to a full nocturnal erection studies, where a first night effect has been penile tumescence study for evaluation of patients with observed, we found that volunteers fell asleep erectile impotence. J Urol 1990; 143: 51–54. 6 Rosen RC, Kostis JB, Jekelis AW. Beta-blocker effects on quickly and slept well in this study and that NSD sexual function in normal males. Arch Sex Behav 1988; 17: might recover the inhibited rapid eye movement 241–255. sleep during midazolam-induced sedation. Further- 7 Lancel M, Cronlein TA, Muller-Preuss P, Holsboer F. Pregne- more, the short period of morning penile tumes- nolone enhances EEG delta activity during non-rapid eye movement sleep in the rat, in contrast to midazolam. Brain Res cence recording faithfully reflects the changes noted 1994; 646: 85–94. during the 8 or more hours of sleep-related penile 8 Wisor JP, DeLorey TM, Homanics GE, Edgar DM. Sleep states erection tracings. Comparison of tracings of penile and sleep electroencephalographic spectral power in mice activity at night and during the daytime periods lacking the b3 subunit of the GABAA receptor. Brain Res 2002; shows a remarkable internal consistency. It is 955: 221–228. 9 Webb WB, Agnew HW, Sternthal H. Sleep during the early reported that electroencephalographic patterns of morning. Psychon Sci 1966; 6: 277–278. the nap in the early morning resembled sleep stages 10 Hanley JA, McNeil BJ. The meaning and use of the area under from the second half of the night, with rapid eye a receiver operating characteristic (ROC) curve. Radiology movement sleep dominating. Morning nap resumed 1982; 143: 29–36. 11 Moons KG, van Es GA, Michel BC, Buller HR, Habbema JD, their previously interrupted rapid eye movement Grobbee DE. Redundancy of single diagnostic test evaluation. cycles at the next sleep onset, with rapid eye Epidemiology 1999; 10: 276–281. movement onset frequently occurring within 12 Schmidt MH, Sakai K, Valatx JL, Jouvet M. The effects of 10 min of nap onset.24 Nocturnal sleep deprivation spinal or mesencephalic transections on sleep-related erec- might resume their previously interrupted rapid eye tions and ex-copula penile reflexes in the rat. Sleep 1999; 22: 409–418. movement cycles at the next sleep onset. 13 Schmidt MH, Valatx JL, Sakai K, Fort P, Jouvet M. Role of As diurnal penile tumescence may be a summary the lateral preoptic area in sleep-related erectile mechanisms reflection of sleep-related penile erections, tumes- and sleep generation in the rat. J Neurosci 2000; 20: cence monitoring during naps may be useful as an 6640–6647. inexpensive and convenient screening approach in 14 Oksenberg A, Arons E, Sazbon L, Mizrahi A, Radwan H. 25–27 Sleep-related erections in vegetative state patients. Sleep the assessment of male erectile dysfunction. For 2000; 23: 953–957. the clinician, a major attraction of daytime penile 15 Schmidt MH, Valatx JL, Schmidt HS, Wauquier A, Jouvet M. tumescence monitoring is the opportunity to evalu- Experimental evidence of penile erections during paradoxical ate sequentially several patients at the same time sleep in the rat. Neuroreport 1994; 5: 561–564. 16 Bourgin P, Escourrou P, Gaultier C, Adrien J. Induction of facility and at regular intervals during office hours. rapid eye movement sleep by carbachol infusion into the Our findings suggest that erection monitoring dur- pontine reticular formation in the rat. Neuroreport 1995; 6: ing midazolam-induced sedation after NSD may be 532–536. convenient inasmuch as it requires less time of both 17 Gulia KK, Mallick HN, Kumar VM. Sleep-related penile erections do not occur in rats during carbachol-induced rapid the practitioner and the patients, and it can be eye movement sleep. Behav Brain Res 2004; 154: 585–587. conducted in an outpatient setting. However, In the 18 Affanni JM, Cervino CO, Marcos HJ. Absence of penile present study, the sample size was small, partici- erections during paradoxical sleep. Peculiar penile events pants in this study were young, sexually healthy during wakefulness and slow wave sleep in the armadillo. men, and no consensus of the criteria about normal J Sleep Res 2001; 10: 219–228. 19 Reves JG, Fragen RJ, Vinik HR, Greenblatt DJ. Midazolam: morning penile tumescence findings was made. pharmacology and uses. Anesthesiology 1985; 62: 310–324. Thus, it is possible that older or dysfunctional 20 Kanto JH. Midazolam: the first water-soluble benzodiazepine. men might respond differently. To complete the Pharmacology, pharmacokinetics and efficacy in insomnia process of validation, large groups of normal sub- and anesthesia. Pharmacotherapy 1985; 5: 138–155. 21 Amrein R, Hetzel W. Pharmacology of drugs frequently used in jects and impotent patients with a defined etiology ICUs: midazolam and flumazenil. Intensive Care Med 1991; must be studies to establish norms for morning 17(Suppl 1): S1–S10. penile tumescence recording. 22 Greenblatt DJ, Shader RI, Abernethy DR. Drug therapy. Current status of benzodiazepines. N Eng J Med 1983; 309: 410–416. 23 Study RE, Barker JL. Cellular mechanisms of benzodiazepine action. JAMA 1982; 247: 2147–2151. 24 Moses JM, Hord DJ, Lubin A, Johnson LC, Naitoh P. Dynamics References of nap sleep during a 40 h period. Electroencephalogr Clin Neurophysiol 1975; 39: 627–633. 1 Wincze JP, Carey MP. Sexual Dysfunction: A Guide for 25 Morales A, Condra M, Heaton JP, Johnston B, Fenemore J. Assessment and Treatment. Guilford: New York, 1991. Diurnal penile tumescence recording in the etiological 2 Condra M, Morales A, Harris C, Heaton JP. A call for caution: diagnosis of erectile dysfunction. J Urol 1994; 152: 1111–1114. the use of visual sexual stimulation as a tool for the etiological 26 Gordon CM, Carey MP. Penile tumescence monitoring during diagnosis of impotence. Int J Impot Res 1990; 2: 203–208. morning naps: a pilot investigation of a cost-effective alter- 3 Marshall P, Earls C, Morales A, Surridge D. Nocturnal penile native to full night sleep studies in the assessment of male tumescence recording with stamps: a validity study. J Urol erectile disorder. Behav Res Ther 1993; 31: 503–506. 1982; 128: 946–947. 27 Gordon CM, Carey MP. Penile tumescence monitoring during 4 Condra M, Fenemore J, Reid K, Philips P, Morales A, Owen J morning naps to assess male erectile functioning: an initial et al. Screening assessment of penile tumescence and rigidity. study of healthy men of varied ages. Arch Sex Behav 1995; 24: Clinical test of snap-gauge. Urology 1987; 29: 254–257. 291–307.

International Journal of Impotence Research