German Journal of Psychiatry

ISSN 1433-1055


Are Serum Melatonin Levels a Marker for the Differential Diagnosis of Dementia?

Magda Tsolaki1, Michael Karamouzis2, Konstantinos N. Fountoulakis3, Vasiliki Iakobidou1, Aristides Kazis1

13rd Department of Neurology, Aristotle University of Thessaloniki, General Hospital ‘G. Papanikolaou’, Exohi, Thessaloniki, Greece

2Laboratory of Biochemistry Aristotle University of Thessaloniki, Thessaloniki, Greece

33rd Department of Psychiatry, Aristotle University of Thessaloniki, University Hospital AHEPA, Thessaloniki, Greece 

Corresponding Author: KN Fountoulakis MD PhD, 53 Chrysostomou Smyrnis Street, 55132 Aretsou Thessaloniki, GREECE, Tel 30 31 435702, Fax 30 31 266570, cell phone 30 31 95776935, e-mail: kfount@med.auth.gr


 Background: Melatonin levels tend to decrease with age. Platelet levels of monoamine oxidase increase with age and demented patients are reported to have higher levels than age-matched controls. Material and Methods: One hundred and two subjects took part in the study. Twenty-one were control subjects (C), 33 were outpatients with Alzheimer's disease (AD), 28 were outpatients with vascular dementia (VD) and 20 were outpatients suffering from other types of dementia (OD). AD was diagnosed according to DSM-IV and NINCDS-ADRDA criteria. OD and VD were diagnosed according to the DSM-IV and NINDS-ARIEN criteria. Statistical Analysis: Two-Way ANOVA, Pearson Product Moment Correlation Coefficient and Discriminant Function Analysis were used. Results: VD patients had significantly higher melatonin values in comparison to all other groups. Five AD patients and 4 OD patients with high melatonin levels manifested factors predisposing to vascular lesions. Conclusion: Our results suggest that there are no significant differences in melatonin levels between patients with AD and those of non-demented age-matched controls and this is in accord with the international literature. Significantly higher levels of melatonin in patients with VD suggest a new, simple and inexpensive diagnostic method for the differentiation of AD from VD. They also suggest also that melatonin levels could constitute a routine examination in the investigation of dementia (German J Psychiatry 2000;3:1-11).


Key Words: Alzheimer’s Disease, Vascular Dementia, Melatonin


 The diet of the elderly is to a significant extent deficient in calories, calcium, iron, thiamin, niacin and proteins (Vatassery, 1980). Since human beings seem to have difficulty in obtaining adequate dietary tryptophan (Maurizi, 1984), it is possible that elderly people have significant tryptophan deficiencies, resulting in deficiencies of serotonin and melatonin. Melatonin may act as a monoamine oxidase inhibitor and elevate brain monoamines (Maurizi, 1984), however this report has not been confirmed. Its serum levels tend to decrease with age (Beck-Friis et al, 1984). Platelet levels of monoamine oxidase increase with age and demented patients have higher levels than age-matched controls (Adofsson et al, 1978). In Alzheimer's disease patients, the hippocampus has significantly decreased serotonin, decreased norepinephrine and increased monoamine oxidase activity, when compared to age-matched controls (Winblad et al, 1985) .

The physiologies of the pineal gland and its hormone melatonin have been studied extensively in mammals by Lerner et al (1958) who also discovered melatonin. Both nocturnal and diurnal animals synthesize and secrete melatonin almost exclusively during night-time darkness (Binkley et al, 1975). The pineal hormone is secreted in a circadian manner associated with low blood levels during the daytime and high levels at night (Waldhauser et al, 1986).The physiological role of melatonin in man has not been defined. However, the characteristic circadian periodicity of serum melatonin has been determined in humans, with a prominent peak during the night (Wurtman et al, 1977). The nocturnal rise of melatonin in human serum is the result of endogenously released norepinephrine acting upon beta-adrenergic receptors of the pineal gland (Frazer et al, 1986). The mechanisms controlling melatonin production involve indoleamines and catecholamines, adrenergic receptors, cyclic adenosine monophosphate (AMP), methylation, light sensitivity, and central rhythm-generating systems (Arendt, 1989). The rhythm of melatonin production is endogenous (i.e. internally generated) and arises from the suprachiasmatic nucleus in the hypothalamus - a so-called central "clock" generating many circadian rhythms (Moore et al, 1974).

There is currently a good deal of interest in the pineal hormone melatonin and its correlation with elderly subjects (Touitou et al, 1984), with patients with senile dementia (Souetre et al, 1989), with affective disease (Maurizi, 1990) and disorders of cerebral circulation (Brovina et al, 1988).

The aim of the current study was to establish whether serum melatonin levels vary between patients with Alzheimer’s disease, vascular, and other types of dementia. If it does, then serum melatonin determination could be of help in the diagnosis or the differential diagnosis of dementia.




 One hundred and two subjects took part in the study. Four groups were formed: 21 healthy elderly age-matched control subjects (C), 50-84 years old, 33 outpatients with Alzheimer's disease (AD), 56-86 years old, 28 outpatients with vascular dementia (VD), 51-91 years old and 20 outpatients with other dementia (OD), 40-79 years old.

AD was diagnosed on the basis of a history of progressive dementia with an insidious onset according to the diagnostic criteria of DSM IV (American Psychiatric Association, 1994) and NINCDS-ADRDA (Report of the NINCDS ADRDA Work Group, 1984). OD and VD were diagnosed according to the diagnostic criteria of DSM-IV and NINDS-ARIEN. Diagnosis was achieved on the basis of the consensus of two of the authors. Cases with mixed type of dementia (Hachinski Ischaemia Score-HIS score 5-7) were excluded. Brain Computerized Tomography (CT) scan was performed on all patients. Brain Magnetic Resonance Imaging (MRI) was performed on some patients with a vascular history and normal CT scan. Blood and biochemical tests were normal. Vitamin B12 and folate were examined in all patients.

The Mini-Mental State Examination (MMSE) (Folstein et al, 1975) was used to assess the cognitive function of all subjects, which was recently validated in Greece (Fountoulakis et al, 1994). The Hachinski Ishaemia Score (HIS) (Hachinski et al, 1975) was used to assist the evaluation of the vascular status of the patients. The Functional Rating Scale for Symptoms of Dementia (FRSSD) (Hutton, 1990) was used to assess everyday function and the Hamilton Depression Rating Scale (HDRS) (Hamilton, 1960) was used to assess depressive symptomatology.

A Radio-Immuno Assay (RΙΑ) for melatonin by Nichols Institute was used in order to determine melatonin levels. The melatonin levels are expressed in pico-grams. The performance characteristics of the method were:

1.     Precision: the coefficient of variation was a. intra-assay CV%=6.7% and inter-assay CV%=7.5%

2.     Sensitivity: was calculated to be 0.3 pg/ml (1.3 pmole/lt)

3.     Specificity: the specific antiserum used, bound melatonin 100% and other similar substances <0.05%

According to the Nickols Institute, normal melatonin values are 0-15 pg/ml.

Serum melatonin measurement was repeated in 18 patients with AD, 7 patients with VD, 6 patients with OD and 7 controls, in order to examine the reliability and second the possibility that a repetition could help us to increase the validity of the discrimination between demented patients.

All blood samples for melatonin determination were collected between 11:00 and 14:00. There were no significant lighting history differences between subjects.

 Statistical analysis

 Two-Way ANOVA and discriminant function analyses were used for the statistical analysis (Altman, 1991). Discriminant Function Analysis produces two functions at each analysis, one for each group. The subject is classified in that group in whose function he or she obtains a higher score. This means that if {AD function}-{ VD function}>0 then the subject suffers from AD.


 The current study included 33 AD, 28 VD, 20 OD patients and 21 controls. The mean melatonin levels were 9.24±6.21 (range 2.7-31.9) for AD patients, 23.43±13.69 (range 6.8-60) for VD patients, 9±6.37 (range 1.7-25) for OD patients and 9.39±6.42 (range 1.4-20.3) for controls. Means and Standard deviations of patients' age, education, MMSE and Hachinski scores are shown in Table 1.

Table 1. Age, school education in years, Mini-Mental State Examination (MMSE) and Hachinski score in all diagnostic groups. Some of the patients had no school eduction (0 years). Some patients in the late stage of the disease received an MMSE score of 0.  





(in years)




(in years)





Ischaemia Score



Alzheimer’s Disease










Vascular Dementia










Other types of dementia



















 A visual image of the distribution of melatonin levels in each group is shown in Figure 1. VD patients manifested higher levels of melatonin in comparison to all other diagnostic groups. This difference is significant: comparison of AD-VD, p<0.001; AD-OD, p=0.89; VD-OD, p<0.001; AD-C, p=0.93; VD-C, p<0.001; OD-C, p=0.84. The use of Discriminant Function Analysis gives a function with moderate power in discriminating between AD and VD (Table 2). According to the solved function, melatonin levels above 18 are indicative of VD.

Table 2. Discriminant Function Analysis; Alzheimer’s dementia (AD) vs. vascular dementia (VD). OD patients were not included in the analysis. The solution of the AD-VD function suggests that if the value (2.34-0.13) * [melatonin value] is greater than zero (i.e., if the melatonin value is greater than 18 pg/ml) then it is probable that the patient is suffering from VD and not from AD. 

Classification of Cases

Predicted classifications





Observed classifications




















Classification Functions








Melatonin Values








  Five out of 33 patients with AD had high melatonin levels, but all had a positive history of vascular disease, hypertension, diabetes mellitus or hyperlipidemia. Although HIS score was low (<4) (Table 1) and CT scan showed the presence only of atrophy, MRI was not undertaken for reasons of bureaucracy.

Nine out of 28 patients with VD (CT or MRI multiple infarcts) had normal melatonin levels. These patients did not manifest a stroke, not even a transient one recently, and therefore their physiology was not, at the time of examination, under an acute stress that could confound results. Perhaps a second measurement could provide additional information.

Four out of 20 patients with OD also had high melatonin levels, but they also had a history of vascular disease.

The melatonin levels of control subjects ranged from 1.4 to 22.4 pg/ml. Five out of 21 control subjects had high melatonin levels. They were not demented but they were suffering from a mild memory disorder (Age Associated Memory Impairment) which may be due to decreased Cerebral Blood Flow and atheromatosis.

A second assessment of serum melatonin levels in 22 patients showed marked consistency for subjects with a first measurement result in accord with their diagnostic group and a consistent second measurement result for those with a first one not in accord. The Pearson Product moment correlation coefficient for this test-retest was higher for normal subjects (R=0.96) and surprisingly low for AD patients (R=-0.63) with OD in the middle (R=0.87). This stresses the necessity for retest of AD patients in order to exclude false negative results.


 Alzheimer's disease (AD) is a very common brain disorder (Tsolaki et al, 1999), characterized by a progressive dementia that occurs in middle or late life. There are several characteristics which distinguish AD from other types of dementia, like a possible better response to antidepressant therapy when depression is also present (Fountoulakis et al, 1999), or the pupil response to light (Fotiou et al, in press), but it is evident that from a clinical point of view, it is very difficult to differentiate between different causes of dementia (Fountoulakis et al, 1998). There are criteria for clinical diagnosis of probable, possible and definite AD. Unfortunately the criteria for the diagnosis of definite AD are the clinical criteria for probable Alzheimer's disease plus histopathologic evidence obtained only post-mortem. The need to combine clinical diagnostic criteria with laboratory testing has been emphasized, because 20% or more of cases with the clinical diagnosis of AD are finally diagnosed at autopsy to suffer from other diseases and not AD. This diagnostic difficulty cannot help general in vivo research, and therapeutic trials are meaningless if there is no definite diagnosis.

There have been many attempts to develop a laboratory test that would help physicians to differentiate AD from other types of dementia. Increased α-chymothrypsine's inhibitory activity (Miller et al, 1993), increased neuronal thread protein levels (De la Monte et al, 1992), Alzheimer's disease-associated proteins (Miller et al, 1993), lower somatostatin levels (Bissette et al, 1992), Alzheimer's β-peptide (Seubert et al, 1992) and decreased amyloid β-protein precursor (Van Nostrand et al, 1992; Farlow et al, 1992) have all recently been reported as potential cerebrospinal fluid markers for AD. Although lumbar puncture is an easy examination, it is difficult for outpatients and cannot be used as a routine examination .

The secretion of melatonin is minimal in humans during the daylight hours, when the sympathetic innervation to the pineal gland is quiescent. However, with the onset of darkness, impulses become active, releasing noradrenaline into the pineal's parenchymal cells and thereby initiating melatonin synthesis and release. There is a direct relation between catecholamines and melatonin. The catecholamines act in combination with beta-noradrenergic receptors; this stimulates the synthesis of cyclic AMP, liberates 5-HT from its storage pool, and activates serotonine-N-acetyltransferase. As pineal melatonin levels rise, the lipid-soluble hormone enters the blood stream via passive diffusion across a concentration gradient (Wurtman et al, 1986).

Preliminary data suggest that melatonin secretion may be severely impaired in patients suffering from AD either before or after 5-methoxypsoralen administration (low serum levels) (Souette et al, 1989; Renfrew et al, 1987). However both studies deal with small numbers of demented patients . There is also a study in which low CSF melatonin levels were found in patients with AD, and a degenerative hypothesis of the pineal gland is thought to be the cause (Tohgi et al, 1992).

The results of the current study suggest that there were no significant differences between melatonin levels of patients with AD and those of non-demented age-matched controls, as other authors also have found (Touitou et al, 1984). The normal melatonin levels in patients with AD suggest that the pineal gland may not be impaired in these patients. So the medical hypothesis that, if a chronic melatonin deficiency could cause dementia, then some cases of dementia would be preventable (Maurizi, 1987), is not confirmed by this study. However, there is some evidence that daily variation in melatonin disappears in the older subjects and in AD patients (Skene et al, 1990; Mirmiran et al, 1992). This cannot be used for dementia differentiation because all patients with dementia are usually elderly. As far as our normal elderly subjects are concerned, our results showed that melatonin levels were not significantly lower, in contrast to the results reported from other studies (Touitou et al, 1984; Iguchi et al, 1982).

As far as we know, there are no reports on melatonin serum levels in patients with vascular or other types of dementia. The current study provides new information on the distribution of melatonin levels in demented patients. The high melatonin levels in patients with VD, with a possibly more marked difference in those patients with a recent vascular lesion or with a continuos progressive vascular degeneration, could help neurologists, psychiatrists and gerontologists, when they face diagnostic problems about the type of dementia.

It is not quite clear why there were high serum melatonin levels in patients with vascular dementia.

The first possible explanation is the increase of catecholamines in VD, because catecholamine excretion - measured by urinary norepinephrine plus epinephrine - is reported to be markedly elevated in patients with stroke and acute arterial hypertension (Feibel et al, 1981). In vascular dementia minor strokes may occur continuously and this may be the reason why melatonin levels sometimes rise and, when no stroke occurs, are sometimes lower.

The second explanation is that epiphysis dysfunction may be greater during the development and course of cerebrovascular disease than during the course of AD.

Three patients with AD had a higher second melatonin value, but all of them also suffered from hypertension. A possible explanation could be that hypertension was the reason for the increased level of melatonin. This cannot be verified, since few of our vascular patients in this study with high melatonin levels had hypertension.


 Our results suggest a new, simple and inexpensive diagnostic method for differentiating between VD and other types of dementia, and especially from AD. They also suggest that melatonin levels could constitute a routine examination in the investigation of dementia. Two high values with a three month interval between them may be a good marker for confirming the diagnosis of VD when case history, neurological examination and neuroimaging findings are insufficient to render a diagnosis.



Adofsson R, Gottfries CG, Oreland L et al. Reduced levels of catecholamines in the brain and increased activity of monoamine oxidase in platelets in Alzheimer's disease:Therapeutic implications. Aging 1978;7:441-451.

Altman DG: Practical Statistics for Medical Research. Chapman & Hall, London, 1991

American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th edition. DSM IV, Washington DC, 1994.

Arendt J. Melatonin:A new probe in psychiatric investigation. Brit J Psych 1989;155:585-590.

Beck-Friis J, Von Rosen D, Kjellman BF et al. Melatonin in relation to body measures, sex age, season and the use of drugs in patients with major affective disorders and healthy subjects. Psychoneuroendocrinology 1984;9:261-277.

Binkley S, MacBridge SE, Klein DC, Ralph CL.Isolation of melatonin of pineal gland factor that lightens melanosites. Endocrinology 1975;96:848-856.

Bissette G, Myers B. Somatostatin in Alzheimer's disease and depression . Life Sci 1992;51:1389-410.

Brovina NN, Zadorozhnyl VV. Endocrine function of the epiphysis in disorders of cerebral circuletion (morphological study). Zh Nevropatol Psikhiatr 1988;88:43-49.

De la Monte S, Ghanbari HA>. Automated microparticle enzyme immunoassay for neural thread protein in cerebrospinal fluid from Alzheimer's disease patients.J Clin Lab Anal 1992;6:379-383.

Farlow M, Ghetti B, Benson MD, Farrow JS, Van Nostrand WE, Wagner SL.. Low cerebrospinal fluid concentrations of soluble amyloid beta protein precursor in hereditary Alzheimer's disease. Lancet 1992;340:453-454.

Feibel JH, Baldwin CA, Joynt RJ. Catecholamine-associated refractory hypertension following acute intracranial hemorrhage: control with propanolol. Ann Neurol 1981;9:340-343.

Folstein MF, Folstein SE, McHugh PR. MMSE: A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-198.

Fotiou F, Fountoulakis K, Tsolaki M, Goulas A, Palikaras A: Changes in the Pupil Reaction to Light in Alzheimer’s Disease Patients. Int J Psychophysiol, in press

Fountoulakis K, Tsolaki M, Chantzi E,Kazis A. Mini Mental State Examination. Validation Study in Demented Patients from the elderly Greek Population. Enkephalos (Greece), 1994;31:93-102

Fountoulakis KN, Tsolaki M, Kazis A: Does Cognitive Disorder Affect Antidepressant Pharmacotherapy in Geriatric Depression? Int J Geriatr Psychopharmacol, 1999;2:28-32

Fountoulakis KN, Tsolaki M, Mohs RC, Kazis A: Epidemiological Dementia Index: A Screening Instrument for Alzheimer’ s Disease and Other Types of Dementia Suitable for Use in Populations with Low Education Level. Dement Geriatr Cogn Dis, 1998;9:329-338.

Frazer A, Brown R, Kocsis J, Caroff S, Amsterdam J, Winokur A, Sweeney J, and Stokes P. Patterns of meletonin rhythms in depression. J Neural Transm 1986;Suppl. 21:269-290.

Hachinski VV, Iliff LD, Zihkla E et al. Cerebral blood flow in dementia. Arch Neurol 1975;32:632-637.

Hamilton M. A Rating Scale for depression . J Neurol Neurosurg and Psychiatr 1960:23-56.

Hutton JT. Alzheimer's Disease in Conn's Current Therary 1990;778-781.

Iguchi H, Kato K, Ibayashi H. Age- depended reduction in serum melatonin. Concentrations in healthy human subjects . J Clin Endocrinol Metab 1982;55:27-29.

Lerner B, Case JD, Takahashi T, Lee W, Mori W. Regulation of splenael rhynses in chicken: refractory period and nonvisual light perception. J Am Chem Soc 1958;80:2587-95.

Maurizi CP. Demantia. The failure of hippocampal plasticity and dreams. Is there a preventative role for melatonin? Medical Hypotheses 1987;24:59-68.

Maurizi CP. Disorder of the pineal gland associated with depression peptic ulcers and sexual dysfunction. South Med J. 1984;77:1516-1517.

Maurizi CP. The therapeutic potential for tryptophan and meletonin possible roles in depression, sleep, Alzheimer's disease and abnormal ageing. Med Hypotheses 1990;31:233-242

Miller Dl, Papayannopoulos IA, Styles J, Bobin SA, Linn YY, Biemann K. Peptide compositions of the cerebrovascular and senile plaque core amyloid deposits of Alzheimer's disease. Arch Biochem Biophys !993;15:301:41-52.

Mirmiran M, Swaab DF, Kok JH, Hofman MA, Witting W, Van Gool WA. Prog Brain Res 1992;93:151-162

Moore RY and Klein DC. Visual pathways and the central neural control of a circadian rhythm in pineal serotonin N-acetyltransferase. Brain Res 1974;71:13-33.

Renfrew JW, May C, Tamarkin L et al. Circadian rhythms in Alzheimer disease. Neurosciences Abstracts 1987;1:322.

Report of the NINCDS-ADRDA Work Group: Clinical diagnosis of Alzheimer's disease. Neurology 1984;34:939-944.

Seubert P, Vigo-Pelfrey J, Swindlehurst C et al. Isolation and quantification of soluble Alzheimer's beta peptide from biological fluids.. Nature 1992;359:325-327.

Skene DJ, Vivien RB, Sparks DL, Hunsaker JC, Pevet P, Ravid D, Swaab DF. Daily variation in the concentration of melatonin and 5-methoxytryptophol in the human pineal gland: effect of age and Alzheimer's disease. Brain Res 1990;528:170-74.

Souetre E, Salvati E, Krebs B, Belougou JL, Darcourt G. Abnorl melatonin response to methoypsoralen in dementia. Am J Psychiatry 1989;146:1037-40.

Tohgi H, Abe T, Takahashi S, Kimura M, Takahashi J, Kikuchi T. Concentrations of serotonin and its related substances in teh cerebrospinal fluid in patients with Alzheimer type dementia. neurosci Lett 1992;141:9-12.

Touitou Y, Fevre M, Bogdan A, Reinber A, De prins J, Beck H, Touitou C. Patterns of plasma melatonin with ageing and mental condition: stability of nyctohemeral rhythms and differences in seasonal variations. Acta Endocrinol 1984;106:145-151.

Tsolaki M, Fountoulakis KN, Pavlopoulos I, Chatzi E, Kazis A: Prevalence and Incidence of Alzheimer’s Disease and Other Dementing Disorders in Pylea Greece. Am J Alzheimer’s Dis, 1999;14 (3): 138-149

Urry RL, Ellis LC. Monoamine oxidase activity of the hypoyhalamus and pituitary:Alterations after pinealectomy, changes in photoperiod or additions of melatonin in vitro. Experientia 1975;31:891-892.

Van Nostrand WE, Wagner SL, Haan J, Bakker E, roos RA.. Alzheimer's disease and hereditary cerebral hemorrhage with amyloidosis-Dutch type share a decrease in cerebrospinal fluid levels of amyloid beta-protein precursor. Ann Neurol 1992;32:215-218

Vatassery GT.Nutrition of the senescent brain. Adv. Neurogerontol 1980;1:79-88.

Waldhauser F and Steger H. Changes in meletonin secretion with age and pubescence. In Wurtman RJ and Waldhauser F: Melatonin in humans by Springer-Verlag 1986:188.

Winblad B, Hardy J, Backman L, Nilsson L. Memory function and brain biochemistry in normal aging and in senile dementia. Ann NY Acad Sci 1985;444:255-268.

Wurtman RJ and Waldhauser F. Melatonin in humans. J Neural Transmission1986;Suppl 21:2

Wurtman RJ, Moskowitz MA. The pineal organ. NEJM;1977:296:1329.