Gynaecomastia and Galactorrhoea
It is important to note that there are no standardised methods for measuring galact-orrhoea. Beumont et al.  describe massaging the breast for 3 min and then measuring the volume of milk produced. Others use a milk pump to extract milk and or just simply ask the patient. Prior to the introduction of antipsychotic medication in the 1950s neither galactorrhoea nor gynaecomastia had been noted as important complaints from patients with schizophrenia. However, Robinson  found an incidence rate for galactorrhoea of 10%, while Plante and Roy  reported an incidence rate of 57%. Tolis et al.  reported 65 cases of galactorrhoea and found that there was no correlation between PRL levels and the frequency and severity of galactor-rhoea. Beumont et al.  pointed out that the correlation between raised PRL and lactation is not a direct one as other factors were involved in the initiation and maintenance of galactorrhoea.
More recent studies suggest that the incidence of galactorrhoea in clinical practice is lower than earlier reports. Wesselmann and Windgassen  completed one of the first prospective studies of galactorrhoea in schizophrenia in 150 patients. They reported that an incidence rate of 14% and prevalence rate of 19% and galac-torrhoea occurred between the 7th and 75th days after initiating treatment. Only 28% of their patients spontaneously disclosed these side effects to their psychiatrists as they found this embarrassing. Underreporting and under recognition of symptoms could be one of the underlying factors in the great variety of reported prevalence and incidence of this adverse effect. Wesselmann and Windgassen  stated that more than half of his female patients who experienced galactorrhoea closely linked this to their femininity. They had 1 patient who misinterpreted this as pseudo-pregnancy. Galactorrhoea is reported much less commonly in males with rates around 1%. However, Halbreich and Kahn  reported that up to 33% of men with HPRL actually had galactorrhoea on examination. It is clear that galactorrhoea should not be left to self-reporting and be included in the physical examination of the patient.
Menstrual dysfunction associated with psychoses was recognised well before the introduction of antipsychotics . These early observations supported the view that illness and possibly life style factors also played a role in menstrual abnormalities, separate from the effects of antipsychotics on low oestrogen via hypogonad-ism. The rates of amenorrhoea vary widely between studies, ranging from very low figures to as high 78% as reported by Smith et al. , and averaging around the 30% level. This wide variation in rates may be a consequence of the definition of amenorrhoea. It has been defined as the temporary or permanent absence of menstruation for more than 6 months. Wong and Seeman  defined amenorrhoea as 3 consecutive missed periods. Hence, many pharmacological studies in schizophrenia designed for 12 weeks or less would miss clinical signs of adverse events of the menstrual cycle. Wong and Seeman  highlighted that 90% of the women linked normal menstruation with good health and subjects with irregular menses had significantly lower self-esteem than those with normal menses (p < 0.01). In the same study, patients with amenorrhoea had higher PRL levels than those without (57 vs. 27 ng/ml).
Smith et al.  followed up patients who had been stabilised on typical antipsychotics for at least 2 years measuring both PRL levels as well as gonadal hormones. Although the average dose of antipsychotics in chlorpromazine equivalents was only 384 mg, 75% of the females had PRL levels greater than the upper limit of normal, 36% had amenorrhoea, 32% oligomenorrhoea and 36% had apparent normal menstruation; however, only 37% of these women were actually ovulating. Furthermore, this study showed that HPRL in women was correlated with the degree of suppression of the HPG axis. Howes et al.  found similarly high levels of intermittent anovulatory cycles (92%) in a population of premenopausal females with a median exposure to antipsychotic of 3.3 years.
In clinical practice, menstrual disturbance is the most obvious feature of HPRL and has been traditionally used as a clinical sign for raised PRL. Smith et al.  and Howes et al.  demonstrated that many women with apparently regular menstruation would not be ovulating, indicating a hypogonadal state.
With the appearance of antipsychotics with a lower propensity for raised PRL levels and amenorrhoea, clinicians should also recognise that switching to these medications may initiate ovulation and put the patient at risk of unplanned pregnancy.
Prior to the introduction of antipsychotics, there were not many reports of fractures or osteopenia/osteoporosis in patients with schizophrenia, except for a few case reports of hip fractures in patients undergoing unmodified electroconvulsive therapy.
A great deal of our current knowledge and understanding about the risks of developing osteopenia/osteoporosis or fractures associated with HPRL comes from animal studies and reports of patients with pituitary tumours.
Little is known about the putative role of PRL on bone cells and bone formation in humans . Adler et al.  investigated the effects of PRL excess and estrogen deficiency on bone in rats and concluded that HPRL with osteoporosis is likely to be due to PRL-induced hypogonadism rather than a direct effect of PRL on calcium homeostasis. Amenorrhoea associated with HPRL was the first recognised model of functional hypogonadal osteoporosis  showing an association with decrease cortical and trabecular bone density. Other recognised associations are smoking, lack of exercise and lithium-induced hyperparathyroidism . Klibanski et al.  were the first to highlight that young women were vulnerable to developing osteopenia/osteo-porosis. There are several studies of patients diagnosed with pituitary tumours who report increased rates of bone loss. Biller et al.  found that bone density decreased significantly in HPRL women who were amenorrheic for more than 20 months. This suggests that studies in treatment pathways in schizophrenia should continue for at least 2 years to pick up these phenomena. Further, it does confirm the clinical importance of amenorrhoea as a risk factor for osteoporosis and osteopenia. Greenspan et al.  confirmed that osteoporosis developed in men with hyperprolactinaemic hypogonadism.
O'Keane  points out that prior to 2000, there were no studies in schizophrenia which examined the relationship between hypogonadism, secondary to HPRL and bone density. The clinical trials in this area have been well described and reviewed by several authors including Leucht et al.  and Byerly et al. . Overall, there is a consistency of findings amongst these reviewers. Albeit limited by small cohorts and short duration, these studies show that patients with schizophrenia, on antipsy-chotics over a prolonged period of time, report increased rates of osteoporosis and osteopenia.
Kishimoto et al.  reports a study in 74 male patients (mean age 58.9) with schizophrenia, investigating the possible causes of reduced bone mineral density. They report that 37% had osteopenia and 27% osteoporosis, 87% of the subjects had HPRL and vitamin D levels were normal. Exercise and vitamin D did not protect these patients from lower bone mineral density. The high PRL group showed low levels of FSH and LH, and significantly low gonadal hormones. The results of this study support the hypothesis that HPRL impacts on the HPG axis which contributes to bone loss and the greatest bone loss correlated with the increased duration of HPRL. Vekemans and Robyn  highlighted that in men PRL levels rise with age, whereas in women PRL levels decline with age and by the time men are 60, their normal PRL levels are higher than women of similar ages. This observation raises the important question of increased risks in older men.
Howard et al.  completed the first large- scale population study to investigate patients with a history of schizophrenia and risk of hip fracture. The method used was a case-controlled study comparing hip fractures in the general practice research database (n = 16,341) with matched controls (n = 29,889). The key finding was that hip fractures were significantly associated with a diagnosis of schizophrenia (OR 1.73; 95% CI 1.32-2.28) and independently associated with PRL elevating antipsychotics (OR = 2.6; 95% CI 2.43-2.78). This is one of the main studies to show that schizophrenia has to be recognized as a significant risk factor for hip fractures.
Leucht et al. , in a systematic review of physical illness in schizophrenia, found that population studies which examined the relationship between schizophrenia and breast cancer were inconclusive. They also examined the question of whether high PRL levels promote breast cancer. Their findings supported the view that the evidence for this was discordant. This chapter explores new findings which were not included in the review by Leucht et al. .
Catts et al.  completed one of the first meta-analysis of cancer incidence rates in schizophrenia which included 100,000 patients and 70,000 parents and 73,000 siblings. This meta-analysis showed that the incidence of breast cancer was significantly increased in female patients SIR = 1.12 (CI 1.02-1.23; p = 0.02). In parents of patients with schizophrenia or their siblings, the pooled data for all cancers were shown to be significantly reduced in parents: SIR = 0.90 (CI 0.88-0.93) and siblings SIR = 0.89 (CI 0.84-0.94). These findings are consistent with genes associated with schizophrenia, being protective against cancers.
Hippisley-Cox et al.  report an epidemiological study using the QRESEARCH database (UK primary care clinical records). The population consisted of roughly 4 million patients with nearly 18.7 million person year's observation. The study showed an increased risk of breast cancer in patients with schizophrenia (adjusted OR 1.52 for deprivation, smoking, obesity and use of other medications, 95% CI 1.10-2.11).
This study differs from many of the earlier studies as it included many patients aged >50 years. Since breast cancer is more common in older women, it is likely that a greater powered study would be required to detect increased rates of breast cancer in younger women.
Hankinson et al.  investigated the risk of breast cancer associated with high PRL by doing a series of epidemiological studies now well known as the US Nurse Health Study. This study cohort was establishes in 1976 with 121,700 US female registered nurses (30-55 years) and followed up every 2 years. From 1989 through 1990 blood samples for different blood biochemistry including PRL results were collected from 32.826 women. Hankinson et al.  observed a significant association with observed PRL levels and postmenopausal breast cancer (highest versus lowest quartile multivariate relative risk of 2.02; 95% CI 1.24-3.31). Hankinson along with Tworoger have now published a series of papers which show increase risk of breast cancer both in premenopausal and postmenopausal women. This is probably the strongest clinical evidence supporting the view that elevated PRL is a risk factor for breast cancer.
Harvey et al.  believe that the balance of evidence from in vitro and animal studies now supports the view that raised PRL is associated with breast cancer. Furthermore, they challenge the long-held view that findings from animal data may not be applicable to humans. These findings are now being included in some of the summary of product characteristics in the US for some of the antipsychotics. There are some confounds with the PRL story, as there is laboratory evidence that trifluo-roperazine inhibits the development of cancer cells.
Halbreich et al.  identified high rates of breast cancer in patients with chronic schizophrenia who were screened with mammography. Wernicke et al.  completed one of the first studies to assess breast screening for patients with schizophrenia. Overall, they found psychiatric patients with a diagnosis of psychosis (OR = 0.33, CI 0.18-0.61; p < 0.01) were less like to attend for breast screening than the general population.
Both studies highlight the importance of breast screening in patients with schizophrenia.
In conclusion, findings from a meta-analysis suggests that schizophrenia maybe protective of breast cancer in parents and siblings, but breast cancer rates were significantly increased in female patients. Epidemiological studies by Hankinson and Tworoger provide strong evidence supporting the thesis that higher levels of PRL potentially put patients at increased risk of breast cancer.
Sexual dysfunction is associated with a wide variety of possible mechanisms. Gitlin  described the possible effects of a direct CNS effect on the neurotransmitter system, sedation secondary to histamine, a peripheral effect resulting in priaprism, and hormonal effects primarily through HPRL.
In patients with schizophrenia, it is extremely difficult to disentangle the sexual dysfunction caused by HPRL from the psychopathology of schizophrenia and the impact of antipsychotics and other medication on a variety of receptor systems such as serotonin and histamine. Weizman et al.  conducted a study to look at sexual dysfunction associated with HPRL patients with renal failure undergoing haemodi-alysis. They noted that patients with high levels of PRL (130 ng/ml) as compared to patients with low levels of PRL (36.3 ng/ml) were clinically impotent. On treatment with bromocriptine, 4 males and 1 female showed a restoration of sexual desire and potency. These findings support the hypothesis that HPRL can be a major cause for reversible sexual dysfunction. Smith et al.  published one of the first controlled studies investigating the frequency and underlying mechanisms of sexual dysfunction in people on antipsychotics. Sexual dysfunction occurred in 45% of patients taking antipsychotics and 17% of non-treated controls. They concluded that HPRL was the main cause of sexual dysfunction in females (with 75% of the women being HPRL) whereas in males (with 34% of the men being HPRL) autonomic side effects were a dominant cause of sexual dysfunction but in those men with HPRL this superseded other causes of sexual dysfunction.
Howes et al.  investigated the relationship between sexual function and gonadal hormones. Although this study showed high rates of sexual dysfunction and high rates of hypogonadism (92% women and 28% men) there was no correlation between sexual function and gonadal hormones which indicated that this is not the main aetiological factor in these cases. Costa et al.  reported no differences in sexual function between patients treated with the atypical antipsychotic olanzapine and conventional antipsychotics but they did show differences in the rate of normalisation of hormone levels. This illustrates that sexual function is a complex multi-factorial activity controlled by many different neurotransmitters and not just PRL.
Was this article helpful?
Are Menopause Symptoms Playing Havoc With Your Health and Relationships? Are you tired of the mood swings, dryness, hair loss and wrinkles that come with the change of life? Do you want to do something about it but are wary of taking the estrogen or antidepressants usually prescribed for menopause symptoms?