You are on page 1of 4

Biol. Pharm. Bull.

40, 1775–1778 (2017)

Vol. 40, No. 101775


Psychiatric Patients with Antipsychotic Drug-Induced Hyperprolactinemia

and Menstruation Disorders
Kenshi Takechi,*,a,b Yurika Yoshioka,c Hitoshi Kawazoe,b Mamoru Tanaka,b Shingo Takatori,d
Miwako Kobayashi,c Ichiro Matsuoka,c Hiroaki Yanagawa,a Yoshito Zamami,e,f Masaki Imanishi, f
Keisuke Ishizawa,e,f Akihiro Tanaka,*,b and Hiroaki Arakib
 Clinical Trial Center for Developmental Therapeutics, Tokushima University Hospital; Kuramoto, Tokushima
770–8503, Japan: b Division of Pharmacy, Ehime University Hospital; Shitsukawa, Toon, Ehime 791–0295,
Japan: c Laboratory of Physiological Chemistry, College of Pharmaceutical Sciences, Matsuyama University;
Matsuyama, Ehime 790–8578, Japan: d Laboratory of Pharmaceutical Information, College of Pharmaceutical
Sciences, Matsuyama University; Matsuyama, Ehime 790–8578, Japan: e Department of Clinical Pharmacology and
Therapeutics, Institute of Biomedical Sciences, Tokushima University Graduate School; Kuramoto, Tokushima 770–
8503, Japan: and f Department of Pharmacy, Tokushima University Hospital; Kuramoto, Tokushima 770–8503, Japan.
Received January 16, 2017; accepted July 1, 2017

Treatment with antipsychotic drugs has been associated with hyperprolactinemia. The same antipsy-
chotic drugs have also been associated with side effects such as menstruation disorders. The aim of this study
was to evaluate the prevalence of hyperprolactinemia and menstruation disorders in women undergoing anti-
psychotic treatment. We performed a retrospective chart review study of psychiatric patients who underwent
laboratory testing for serum prolactin (PRL) level between March 2011 and March 2015 in Ehime University
Hospital. Patients presenting with and without menstruation disorders were evaluated to determine if they
presented concomitant hyperprolactinemia. Patients with menstrual disorders had a significant increase in
serum PRL level with a mean of approximately 90 ng/mL. Those with menstrual disorders presented in-
creased PRL levels by 2-fold that of patients without menstrual disorder. However, there was no significant
difference in the equivalent dose of chlorpromazine between these two groups. Additionally, about 70%
of patients with menstrual disorders received risperidone treatment. The receiver operating characteristic
curve showed that the optimal cutoff point of serum PRL level associated with the development of menstrual
disorders was 60 ng/mL. Based on these results, we concluded that patients with menstrual disorders pre-
sented increased serum PRL, and that most of them underwent treatment with risperidone.
Key words prolactin; antipsychotic drug; menstruation

Atypical antipsychotics are recommended as first-line treat- 20 ng/mL for men.9) However, if patients receiving antipsy-

ments for individuals with schizophrenia. These drugs not chotic therapy develop hyperprolactinemia (>25 ng/mL), it
only improve symptoms, such as auditory or visual hallucina- is  very  difficult  to  discontinue  medication  at  once,  because 
tions, but compared with typical antipsychotics, they are asso- of  therapeutic  efficacy.  It  remains  unclear  what  cutoff  level 
ciated with fewer side effects, such as sleepiness, weight gain, of PRL concentration causes menstrual disorders. Therefore,
and tremors.1) However, antipsychotic drugs are also known to the relationship between serum PRL and adverse side effects,
influence  the  hypothalamo-hypophyseal  axis  and  induce  vari- such as menstrual irregularities, needs further investigation.
able degrees of hyperprolactinemia.2,3) Hyperprolactinemia is The purpose of this study was to evaluate the prevalence
a recognized adverse side effect of antipsychotic drugs. It can of hyperprolactinemia in women with psychiatric disorders
be  acute  or  chronic,  and  it  can  be  associated  with  sexual  dys- receiving antipsychotic treatment.
function, menstrual irregularities, amenorrhea, galactorrhea
and osteoporosis.4) The proportion of premenopausal women PATIENTS AND METHODS
who develop hyperprolactinemia during treatment with anti-
psychotic drugs can be as high as 48 to 93%.5) Despite such Ethical Approval of the Study Protocol This study was
high prevalence, in most cases, hyperprolactinemia is asymp- conducted in accordance with the guidelines for the care of
tomatic. However, some patients present clinical symptoms human study participants adopted by the Ethics Committee of
along with severe disease.6) Additionally, the condition and its Ehime University Hospital (approval number: 1602009), the
management strategies are scarcely mentioned in psychiatric Ethical Guidelines for Medical and Health Research Involving
nursing  journals  or  textbooks.7) Therefore, hyperprolactinemia Human Subjects, and the principles of the Helsinki Declaration.
induced by antipsychotic drugs may be less successfully man- Research Data and Patient Characteristics This retro-
aged. This could be potentially harmful to patients and af- spective study was carried out at Ehime University Hospital
fect their QOL.6,8) In clinical practice, serum prolactin (PRL) using data from electronic medical records dated between
levels are helpful in the early detection of adverse side effects March  2011  and  March  2015.  We  excluded  the  medical  re-
induced by antipsychotic drugs. Hyperprolactinemia is the cords of pregnant women, nursing women, as well as women
presence of abnormally high levels of PRL in serum. Normal with hypothyroidism or hypophyseal adenoma. Patients who
PRL levels are less than 25 ng/mL for women and less than underwent serum PRL monitoring after administration of psy-

* To whom correspondence should be addressed.  e-mail:;
© 2017 The Pharmaceutical Society of Japan
1776 Biol. Pharm. Bull. Vol. 40, No. 10 (2017)

Table 1. Background of Patients with and without Menstrual Disorders Table 2. The Proportion of Prescribed Atypical Antipsychotics between
Patients with and without Menstrual Disorders
No menstrual Menstrual
disorder disorder No menstrual disorder Menstrual disorder
(n=65) (n=18)
N 65 18
Age 37 [15–44] 32 [15–44]a) Aripiprazole (%) 16 (25) 0 (0)a)
Combination use (%) 8 (12) 4 (22) Risperidone (%) 15 (23) 12 (67)b)
Chlorpromazine equivalent dose (mg/d) 495.8±419.7 401.6±387.6 Olanzapine (%) 19 (29) 4 (22)
Quetiapine (%) 11 (17) 5 (28)
Rate of prescribed atypical antipsychotic following patients with menstrual disor-
der versus with no menstrual disorder: a) p<0.05. Combination use show the number Rate of prescribed atypical antipsychotic following patients with menstrual disor-
of patients with over two drugs induced by hyperprolactinemia such as chlorproma- der versus with no menstrual disorder: a) p<0.05; b) p<0.01.
zine, haloperidol, levomepromazine, risperidone, paliperidone, quetiapine, olanzapine,
blonanserin, aripiprazole and clozapine.

chiatric drugs were enrolled in this study. Patient character-

istics  (age  [under  55  years],  sex  [only  women])  and  necessary 
research data (serum PRL and prescription history) were col-
lected from the medical records of 83 patients with psychiatric
disorders (Table 1). The antipsychotic drugs prescribed in this
study were chlorpromazine, haloperidol, levomepromazine,
risperidone, paliperidone, quetiapine, olanzapine, blonanserin,
aripiprazole and clozapine. We used the equation described by
Woods10) to calculate the equivalent dose of chlorpromazine
for the dose of the other psychiatric drugs. The antipsychotic
drug dosages used were then compared by chlorpromazine
equivalent dosages.
Statistical Analyses The data are shown as means and
standard  deviation.  The  statistical  significance  of  the  differ-
ences resulting from comparisons of patients with or without
menstrual disorders was evaluated using the Mann–Whitney
U  test  or  Fisher’s  exact  test  after  checking  for  normal  distri-
bution using the Kolmogorov–Smirnov test. Receiver oper-
ating characteristic (ROC) curves were calculated, and the
area under the curve (AUC) was calculated to evaluate the
optimal cutoff point of serum PRL level to induce menstrual
disorders. The optimal cutoff value was calculated using the
following formula: (1−sensitivity)2+(1−specificity)2 with the
ROC curve. A value of p<0.05  was  considered  significant. 
Statistical  analyses  were  performed  using  Ekuseru-Toukei 
2012 (Social Survey Research Information Co., Ltd., Tokyo,


Relation of Serum PRL and Menstrual Disorders The

changes in serum PRL level in the antipsychotic drug treat-
ment group were compared between patients with or with-
Fig. 1. Comparison of Serum Prolactin Levels (A) and Chlorpromazine
out menstrual disorders (Fig. 1). A total of 83 women were
Equivalent Dose (B) between Patients with and without Menstrual Dis-
included in this study. Patients with and without menstrual orders
disorder had a mean age (range) of 32 (15–44) and 37 (15–44) ** p<0.01.
years, respectively. The mean age of women with menstrual
disorder  was  significantly  less  than  that  of  patients  without 
menstrual disorders (p<0.05). The mean PRL serum level psychotic drugs in patients receiving antipsychotic drugs to
was  significantly  higher  in  patients  with  menstrual  disorders  identify a relationship between menstrual disorder and type
compared with those without menstrual disorder (89.3±36.9 of drug. The proportion of atypical psychotic drugs used is
vs. 42.5±54.5;  95%  confidence  interval  (CI):  46.8  [19.3–74.3],  shown in Table 2. The proportion of patients with menstrual
p<0.001) (Fig. 1A). However, patients with menstrual dis- disorders  that  received  risperidone  treatment  was  significantly 
order  showed  no  significant  difference  in  chlorpromazine  higher than that in those without menstrual disorders, while
equivalent dose (mg/d) compared with patients without men- the proportion of patients without menstrual disorders that re-
strual disorder (401.6±387.6 vs. 495.8±419.7; 95% CI: 94.2 ceived aripiprazole treatment was higher than that of patients
[−124.7–313.2], p=0.36) (Fig. 1B). with menstrual disorders.
Next,  we  focused  on  the  rate  of  prescribed  atypical  anti- Association of Serum PRL Level with Menstrual Disor-
Biol. Pharm. Bull.
Vol. 40, No. 10 (2017)1777

Zhao et al. reported that adjunctive aripiprazole treatment in

patients  with  schizophrenia  who  develop  risperidone-induced 
hyperprolactinemia  results  in  a  significant  reduction  of  serum 
PRL level.15) In a previous study, it was reported that at week
52, the median PRL had increased by 177% in the haloperi-
dol group and decreased by 40% in the aripiprazole group.16)
Interestingly, our results showed that the proportion of pa-
tients without menstrual disorders that received aripiprazole
treatment  was  significantly  higher  than  that  of  patients  with 
menstrual disorders. Thus, our results seem to be consistent
with  those  of  previous  research.  These  findings  indicate  that 
the increase in PRL induced by antipsychotics can be attribut-
able to the drug type rather than the dose of antipsychotics.
Bostwick et al. reported that if the effects of PRL are evident,
the drug can be changed to another agent that is less likely to
affect PRL levels.17)
Further research is needed to clarify the appropriate method
Fig. 2. ROC Curves between Serum Prolactin and Menstrual Disorders for PRL monitoring. In the present study, using ROC curve
The  maximum  area  under  the  curve  was  0.82.  The  optimal  cutoff  value  was  analysis, we estimated that the cutoff value of PRL that can
60 ng/mL (sensitivity: 0.83; specificity: 0.76).
lead to adverse side effects induced by antipsychotic drugs,
such as menstrual disorders, was 60 ng/mL. Additionally, the
ders   Finally, using the ROC curve, we examined the relation  AUC (0.82) of the ROC curve had an acceptable accuracy.
between hyperprolactinemia and adverse side effects, such as The limitations of this study and details that we should
menstrual disorders (Fig. 2). In the ROC curve, the AUC for consider in the future are as follows. This was a single-center 
serum PRL level was 0.82. The optimal cutoff point of serum study with a relatively small sample size. This study was ret-
PRL level for the detection of antipsychotic-induced hyperpro- rospective in nature and we did not include a control group.
lactinemia was found to be 60 ng/mL. Additionally, the previous treatments received by the patients
were  various  and  heterogeneous.  Therefore,  a  large-scale 
DISCUSSION prospective cohort validation study, including other factors
associated with hyperprolactinemia is necessary. Future large-
The present study results showed that serum PRL level in scale prospective validation studies may determine whether
women with psychiatric disorders without menstrual disorders the cutoff value of 60 ng/mL in serum PRL may be a useful
was about 40 ng/mL. The normal range of serum PRL level indicator of menstrual disorders induced by antipsychotic
is between 10 and 20 ng/mL in men and 10 and 25 ng/mL in treatment. Although further studies are required to clarify
women.11) This result suggests that in clinical practice, the the involvement of hyperprolactinemia in the adverse effects
therapeutic effect of antipsychotic drugs is prioritized over induced  by  psychiatric  treatment,  our  findings  indicate  that 
the risks associated with the use of these drugs. The group clinicians should monitor PRL in these patients based on the
with  menstrual  disorders  showed  a  significantly  higher  PRL  cutoff value of 60 ng/mL to improve early detection of adverse
serum value compared with the group without menstrual dis- effects such as menstrual disorders.
orders. We found that the mean age of patients with menstrual
disorders  was  significantly  less  than  that  of  patients  without  Conflict of Interest The  authors  declare  no  conflict  of 
menstrual disorder. However, in a previous study, there was interest.
no  significant  correlation  between  age  and  PRL  values  in  54 
patients with menstrual disorders, such as uterine bleeding.12) REFERENCES
Additionally, the dose of medication that was calculated based
on  the  chlorpromazine  dose  was  not  significantly  different  in    1)  Khanna  P,  Suo  T,  Komossa  K,  Ma  H,  Rummel-Kluge  C,  El-Sayeh 
either group (with or without menstrual disorders), and the HG, Leucht S, Xia J. Aripiprazole versus other atypical antipsychotics
patients receiving risperidone had menstrual disorders. In a for schizophrenia. Cochrane Database Syst. Rev., 2014, Cd006569
previous study, abnormal menstrual cycles were reported in
2) Howes OD, Wheeler MJ, Pilowsky LS, Landau S, Murray RM,
88% of women treated with risperidone compared with 48%
Smith  S.  Sexual  function  and  gonadal  hormones  in  patients  taking 
of  women  taking  first-generation  antipsychotics.5) Addition- antipsychotic treatment for schizophrenia or schizoaffective disor-
ally,  antipsychotic-induced  hyperprolactinemia  is  related  to  der. J. Clin. Psychiatry, 68, 361–367 (2007).
differential  dopamine  D2  receptor  affinity,  5-hydroxytrypta- 3) Konarzewska B, Wolczynski S, Szulc A, Galinska B, Poplawska
mine  (5-HT)  receptor  affinity,  and  blood–brain  disposition  of  R, Waszkiewicz N. Effect of risperidone and olanzapine on re-
antipsychotics.13) The introduction of atypical antipsychotics productive  hormones,  psychopathology  and  sexual  functioning  in 
represented a major advance in clinical therapy, and it was male patients with schizophrenia. Psychoneuroendocrinology, 34,
associated with lower adverse side effects compared with 129–139 (2009).
typical antipsychotics. However, risperidone has a strong 4) Holt RI, Peveler RC. Antipsychotics and hyperprolactinaemia:
mechanisms, consequences and management. Clin. Endocrinol.
D2-blocking  effect  and  a  simultaneous  effect  on  the  5-HT2
(Oxf.), 74, 141–147 (2011).
receptors.14) Thus, patients with hyperprolactinemia were
5) Kinon BJ, Gilmore JA, Liu H, Halbreich UM. Prevalence of hyper-
more likely to receive risperidone in our study. In contrast,
1778 Biol. Pharm. Bull. Vol. 40, No. 10 (2017)

prolactinemia in schizophrenic patients treated with conventional 11)  Peuskens J, Pani L, Detraux J, De Hert M. The effects of novel and 

antipsychotic medications or risperidone. Psychoneuroendocrinol- newly approved antipsychotics on serum prolactin levels: a compre-
ogy, 28 (Suppl. 2), 55–68 (2003). hensive review. CNS Drugs, 28, 421–453 (2014).
6) Johnsen E, Kroken RA, Abaza M, Olberg H, Jorgensen HA. Anti- 12) Ratchev E, Dokumov S. Premenopausal bleeding associated with
psychotic-induced  hyperprolactinemia:  a  cross-sectional  survey.  J. hyperprolactinaemia. Maturitas, 21, 197–200 (1995).
Clin. Psychopharmacol., 28, 686–690 (2008). 13) Fitzgerald P, Dinan TG. Prolactin and dopamine: what is the con-
  7)  Wong-Anuchit  C.  Clinical  management  of  antipsychotic-induced  nection? A review article. J. Psychopharmacol., 22 (Suppl.), 12–19
hyperprolactinemia. Perspect. Psychiatr. Care, 52, 145–152 (2016). (2008).
8) Lu ML, Shen WW, Chen CH. Time course of the changes in 14)  Just  MJ.  The  influence  of  atypical  antipsychotic  drugs  on  sexual 
antipsychotic-induced  hyperprolactinemia  following  the  switch  to  function. Neuropsychiatr. Dis. Treat., 11, 1655–1661 (2015).
aripiprazole. Prog. Neuropsychopharmacol. Biol. Psychiatry, 32, 15) Zhao J, Song X, Ai X, Gu X, Huang G, Li X, Pang L, Ding M,
1978–1981 (2008). Ding  S,  Lv  L.  Adjunctive  Aripiprazole  Treatment  for  Risperidone-
9) Halbreich U, Kinon BJ, Gilmore JA, Kahn LS. Elevated prolactin Induced  Hyperprolactinemia:  An  8-week  randomized,  open-label, 
levels in patients with schizophrenia: mechanisms and related comparative clinical trial. PLOS ONE, 10, e0139717 (2015).
adverse effects. Psychoneuroendocrinology, 28 (Suppl. 1), 53–67 16) Argo TR, Carnahan RM, Perry PJ. Aripiprazole, a novel atypical
(2003). antipsychotic drug. Pharmacotherapy, 24, 212–228 (2004).
10) Woods SW. Chlorpromazine equivalent doses for the newer atypical 17)  Bostwick  JR,  Guthrie  SK,  Ellingrod  VL.  Antipsychotic-induced 
antipsychotics. J. Clin. Psychiatry, 64, 663–667 (2003). hyperprolactinemia. Pharmacotherapy, 29, 64–73 (2009).