Seminal plasma magnesium and premature ejaculation: a case-control study
Urology Research Center of Sina Hospital, Tehran University of Medical Sciences and Health Services, Iran
Accepted for publication 12 April 2006

OBJECTIVE To determine the relationship between premature ejaculation (PE) and serum and seminal plasma magnesium levels, in a casecontrol study. PATIENTS AND METHODS Thirty-eight patients referred to the authors’ urology outpatient clinic were evaluated in two groups; cases comprised 19 men complaining of PE, defined using the Diagnostic and Statistical Manual of Mental Disorders IV criteria and an intravaginal

ejaculatory latency time (IELT) of <1 min, and a control group of 19 married men with a normal IELT. All men had a history taken, a systemic physical examination and laboratory studies. After organic and psychogenic disorders were excluded, the 19 patients were included in the study. Seminal plasma and serum magnesium levels were determined using atomic absorption spectrophotometry. RESULTS

significant relationship between seminal plasma magnesium, but not the plasma level, and PE (P < 0.001 and 0.597 respectively). CONCLUSION PE is significantly related with a lower level of seminal plasma magnesium. The pathological physiology of this relationship requires more investigation. KEYWORDS

The mean (SD) plasma magnesium level was 94.7 (10.9) mg/L in the cases and 116.7 (11.6) mg/L in the controls. There was a

premature ejaculation, seminal plasma magnesium, plasma magnesium

INTRODUCTION Premature ejaculation (PE) is the most prevalent sexual disorder in men [1]. Magnesium is one of the elements present in human semen; it is required for enzymes that act on phosphate-containing substrates. A decrease in magnesium level will result in an increase in thromboxane A2, which leads to a rise in endothelial intracellular calcium and a subsequent decline in nitric oxide (NO) [2,3]. As NO is a vascular smooth muscle-relaxing factor [4], cavernosal smooth muscle contraction from decreased NO might be a contributing factor to PE [5]. There have been few attempts to assess the possible relationship between semen magnesium level and PE. Thus our objective was to evaluate the relationship between seminal magnesium level, body mass index (BMI) and PE.

Mental Disorders IV criteria [6] and an intravaginal ejaculatory latency time (IELT) of <1 min, measured using a stopwatch. Patients were included if their history showed a duration of marriage of >6 months, PE for >6 months and no response to sex therapy; they were excluded if they had organic disorders (e.g. diabetes mellitus, hypertension, vascular disorders, endocrine disorders, renal failure and previous genitourinary surgery), PE for <6 months, intermittent PE, and an abnormal mental state or history of psychiatric disorder. The control group included 19 married men aged 20–50 years with a normal IELT and other normal variables, with no history of any organic or psychiatric disorder, and who were referred to our clinic for nephrolithiasis. Patients had their history taken and had a physical examination. The duration of marriage, smoking habits (pack-years), coital habits, level of education, history of psychiatric problems and drug abuse were also elicited. Blood pressure and peripheral pulses were routinely checked. The weight and height of the men were also recorded, and the BMI calculated. Special features such as gynaecomastia, genital abnormalities,

and secondary sexual characteristics were noted. Semen was analysed in all subjects according to the WHO guidelines [7]. After 3–5 days of sexual abstinence, the semen obtained by masturbation was collected into a sterile acidwashed container, using no lubricant jelly. Specimens were centrifuged at 110 g for 10 min at 4 °C within 30 min of sample collection. Aliquoted samples were stored at −80 °C until they were assayed. Blood samples were taken at the same time. The magnesium level was measured using atomic absorption spectrophotometry (model AA670, Shimadzu, Japan). The supernatant samples were liquefied at room temperature and diluted to 1 : 10 in deionized water. Phosphate ions were eliminated by lanthanum chloride. The magnesium stock standard was obtained from Sigma Chem. Co, St Louis, MO, USA. Semen samples contaminated with blood or pus, and those with a pH of <7 or >8 were discarded. Each variable (demographic data, and serum and seminal plasma magnesium levels) was assessed using a univariate analysis by Student’s t-test and the chi-square test (with Yates’ correction as needed) and Pearson

PATIENTS AND METHODS Thirty-eight patients referred to the authors’ urology outpatient clinic between January 2002 and December 2004 were enrolled in the study. The cases comprised 19 patients aged 20–50 years complaining of PE, defined using the Diagnostic and Statistical Manual of





2 0 0 6 B J U I N T E R N A T I O N A L | 9 8 , 4 0 2 – 4 0 4 | doi:10.1111/j.1464-410X.2006.06320.x

The significance level was set at 0. [5] levels of magnesium. After logistic regression modelling to adjust for these confounders. and a simultaneous increase in urinary excretion [9.SEMINAL PLASMA MAGNESIUM AND PREMATURE EJACULATION TABLE 1 The demographic characteristics of the subjects and the magnesium levels Variables.012). and reports suggest a possible relationship between age and PE [12–14]. to cause emission and ejaculation. There were normal serum and semen levels of all elements in the three groups.212 P 0.020) 0. Also. there is convincing evidence that a long duration of physical effort in men leads to a decrease in extracellular magnesium. pack-years BMI. years Duration of marriage. there remained a significant relationship between seminal plasma magnesium and PE.87) 4. in a study by Zavaczki et al. The maximum likelihood approach was used to estimate weights of the logistic parameters.66) 94.63) 5. The Kolmogorov–Smirnov test was used to check that the magnesium levels had a normal distribution. or PE. not only was there a significant relationship between PE and seminal plasma magnesium.10].83) 24.823 (0. years Smoking. but it was weaker in cases (r = 0. Hypomagnesaemia stimulates angiotensin-induced aldosterone synthesis and thromboxane-A2 overproduction by phospholipase-A2 activation. RESULTS The demographic characteristics of the study groups are shown in Table 1. SEM (β) = 9.93 0.001 In the study by Omu et al.110 −0.22 0. This will decrease cGMP.68 (11.53) 13.027 0.851. respectively.59) 23. DISCUSSION In the present study there was a significant relationship between seminal plasma magnesium and PE.075 0.3 on univariate analysis (seminal plasma magnesium.46. independent prognostic factors associated with PE were age and seminal plasma magnesium (Table 2). mean (SD) or n N Age.928) TABLE 2 Independent predictors of PE β = 29. Variables Age Seminal plasma magnesium B 0. Elevated cytosolic Ca2+ in endothelial cells promotes phosphodiesterases and decreases G-cyclase activity [3. It is difficult to understand how a small change in the magnesium level of seminal plasma between the cases and controls could result in such a considerable difference in the clinical situation.80) 116.26 (2.66) P 0.39) 14. The cases were younger than the controls.89 (26.0001.22) 1 20.597 <0. n Magnesium levels.027).001 <0. Therefore. but magnesium is an essential ion for enzyme activation in the body. This transient hypomagnesaemia might manifest as uncontrolled contractility of the male genital tract. due to a transient shift between extracellular and intracellular magnesium components.4]. Initially the confounding factors appeared to be responsible for this difference.10) 3 20. age and BMI were used as continuous variables with a constant odds ratio for each score) were assessed in a multivariate analysis with a logistical regression procedure and forward stepwise selection. statistical analysis showed that there was no significant difference in age between the groups but the BMI was significantly lower in cases than controls (P = 0.71 (0. goodness-of-fit = 3. df = 8. copper and selenium were evaluated in serum and seminal plasma in three groups of men.73 (2.10 (9.663–1.84) 3. P = 0.13 (3.890.809 (0.047) than in controls (r = 0. this shows that each 1-year increase in age and each unit increase in seminal plasma magnesium were associated with an 18% and 19% decrease in the incidence of PE. The dependent variable was coded as zero for absence and 1 for presence of PE.030. changes in seminal plasma magnesium levels were possibly related to the magnesium content of the diet.15 with a tolerance level at 0. P < 0. and another relationship is also possible. which in turn results in decreased NO production. magnesium in both groups showed a significant association. P = 0. or oligoasthenozoospermia. so that these two variables (age and BMI) could be considered as confounders.29 0.57.002 Odds ratio (95% CI) 0. mg/L Serum Seminal plasma Seminal/serum ratio Cases 19 31. In the present study. resulting in decreased NO production and its release from the endothelium [2]. kg/m2 History of drug abuse. Engagement of thromboxane-A2 results in Ca2+ influx [2. The seminal plasma magnesium level was significantly lower in cases than controls (Table 1. but a significantly lower seminal plasma magnesium level in men with PE. correlation. Variables that were significantly related to PE (P < 0. The predictive model was based on the three variables listed.73 (10.68 (0.78 (28. [15]. Also.92 0. zinc. P = 0. it seems that changes in seminal plasma magnesium secondary to this factor or any other probable factor could be the cause of PE in the present patients. but also a lower level of magnesium in semen with increasing incidence of PE. the BMI in the cases was also less than in controls.12 (2. Seminal plasma magnesium levels (>70 mg/L) are much higher than in serum (17–24 mg/L) [8].597) analysis of the correlation between the serum and seminal plasma © 2006 THE AUTHORS JOURNAL COMPILATION © 2006 BJU INTERNATIONAL 403 .81) 7.706–0. Although there was no significant difference in the serum magnesium levels of cases and controls (P = 0. There were no significant differences in serum magnesium levels between the present two groups either.37 (3. i. with normal sperm values.58) Controls 19 34.73 (2. decreased levels of NO consequently lead to vasoconstriction.618. As NO is a vascular smooth muscle relaxing factor [4].1 (8.05) or where P < 0.001).e.11]. After multivariate analysis and adjusting the model for BMI.

Regulation by calcium of the nitric oxide/cyclic GMP system in cerebellar granule cells and astroglia in culture. myaloosh@yahoo. Croft PR. Kiss SA et al. e-mail: md_aloosh@hotmail. Current Medical Diagnosis and Treatment. Chapter 19. Oriowo MA. Magnesium homeostasis during high-intensity anaerobic exercise in men. Guezennec CY. J Appl Physiol 1987. CONFLICT OF INTEREST None declared. The association between low seminal magnesium levels and PE in the present study is clinically significant and accordingly three hypotheses are suggested: (i) decreases in seminal magnesium could be a consequence of a defect in the active transport system that transports magnesium from blood to semen. 1993: 5–23 8 Bartis CA. More trials into the role of magnesium in the physiology of the male reproductive tract. nitric oxide. (iii) previous hypomagnesaemia caused by low consumption of magnesium might contribute to the decline in seminal plasma magnesium levels. 1995: 742–67 Correspondence: Mehdi Aloosh. Fluid and electrolyte disorders. 15: 519–24 Dunn KM. Kyle SB. Kaplan and Sadock’s Comprehensive Textbook of Psychiatry. 49: 333–41 5 Omu AE. Szollosi J. Dolev E. Al-Bader AA. Mechanism of vasoconstriction induced by 9. J Neurosci Res 1997. Requested vs routine. Papadakis MA eds. 5th edn. Hao ZY. Shi HQ.com Abbreviations: PE. Fam Pract 1998. Sexual problems: a study of the prevalence and need for health care in the general population. Association of sexual problems with social. 2001: 805 9 Deuster PA. Tehran University of Medical Sciences and Health Services. as shown by Zavaczki et al.A L O O S H ET AL. 3rd edn. Iran. Kuriyama H. 60: 402–9 4 Baltrons MA. body mass index. Frequency of hypomagnesemia and hypermagnesemia. 3rd edn. Hackett GI. Sadock BJ. Wang KX. Epidemiological studies report that the amount of magnesium consumed by most individuals during prolonged periods is 20–30% less than the recommended dietary allowance [16]. interventional studies with magnesium supplements might be helpful. Chapter 38. Philadelphia: WB Saunders. Curr Psychiatry Rep 2000. 16: 131–6 Papadakis MA. 53: 144–8 Zavaczki Z.12 methanothromboxane A2 in the rabbit coronary artery. 152: 145–8 3 Kanmura Y. Philadelphia: Lippincott. Vol. Saadoun S. J Epidemiol Community Health 1999. Durlach J. Arch Androl 2001. Magnesium-orotate supplementation for idiopathic infertile male patients: a randomized. (ii) there might be a magnesium-diminishing factor like chelating agents in the semen of the patients. 404 © JOURNAL COMPILATION 2006 THE AUTHORS © 2006 BJU INTERNATIONAL . 2: 189–95 Ryzen E. [15]. psychological. and physical problems in men and women: a cross sectional population survey. WHO Laboratory Manual for the Examination of Human Semen and Semen-Cervical Mucus Interaction. Tietz Fundamentals of Clinical Chemistry. Mephee SJ. Ashwood ER. West J Med 1990.Hill.com. Prevalence and risk factors of sexual dysfunction in men and women. premature ejaculation. In Sadock VA ed. New York: McGraw. BJU Int 2004. Agullo L. placebo-controlled clinical pilot study. Magnes Res 1990. Normal human sexuality and sexual and gender identity disorders. New York: Cambridge University Press. are advocated. In Enders DB. and especially its association with PE. 3: 93–102 Whang R. Zhang XJ. JAMA 1990. 62: 545–50 10 Rayssiguier Y. Circ Res 1987. 46: 59–66 3 Sadock VA. Mineral and bone metabolism. REFERENCES 1 Rosen RC. Prevalence of sexual dysfunction in Chinese men with chronic prostatitis. Schoomaker EB. 93: 568–70 Dunn KM. In Tierney LM. 1. Croft PR. Urology Research Center of Sina Hospital. it is probable that the consumption of more magnesium in the diet leads to an increase in seminal Mg2+ levels. Hackett GI. Hence. BMI. Itoh T. IELT. Rude RK.11-epithio-11.. 2000: 1592–3 7 World Health Organization. Magnesium in human semen: possible role in premature ejaculation. Magnes Res 2003. NO. Anderson RA. Rude RK eds. 263: 3063–4 Liang CZ. Garcia A. New experimental and clinical data on the 11 12 13 14 15 16 relationship between magnesium and sport. Dashti H. Williams & Wilkins. Ryder KW. 7th edn. Low intracellular magnesium in patients with acute 2 pancreatitis and hypocalcemia. Thus. intravaginal ejaculatory latency time.