Introduction
Pineal cysts (PC) are benign, often incidentally discovered lesions of the pineal gland with a reported prevalence of up to 37.5% in MRI studies [24]. Many PCs remain asymptomatic, but PC patients also complain about headache, sleep disturbances, dizziness, and nausea [1, 5, 7]. Patients with PCs are about twice as likely to suffer from headaches compared to patients in a control group without PC [18]. The pathophysiological relationship between these symptoms and the PC has not yet been conclusively clarified. In some cases, PC can cause hydrocephalus or Parinaud syndrome; here, surgical treatment is the gold standard. However, the available data show that even patients with symptomatic PC who do not suffer from hydrocephalus benefit from cyst resection and cyst fenestration as well [6, 11]. In addition to mechanical explanations such as compression of the internal cerebral veins or intermittend obstruction of the aqueduct [2], a disturbed melatonin (ML) release is also being discussed. In animal experiments, there were no differences in the sleepâwake cycle before and after pinealectomy [8]. Slawik et al. conducted a prospective study on patients with pineal tumors, in which ML production in saliva was examined preoperatively and postoperatively. Postoperatively, a significant decrease in ML was found in the nighttime saliva of all patients with no significant effect on the patients' sleepâwake cycle [22]. Concerning non-hydrocephalic patients with symptomatic PC, data of ML levels is lacking.
ML (N-acetyl-5-methoxytryptamine) is synthesized in the pineal gland, which is a central neuroendocrine organ that plays a key role in regulating the circadian rhythm. The production of this hormone is subject to a circadian rhythm, with secretion typically peaking during the dark phase of the night. Pineal ML secretion is regulated by the hypothalamic nuclei and by the action of light via a multisynaptic pathway starting from retinal ganglion cells [12, 14]. Although ML can also be produced in other tissues, the systemically measurable proportion in saliva and blood mainly originates from the pineal gland [4]. However, human ML production is subject to interindividual variability. The greatest increase in ML concentration in serum has been observed in previous studies between 8 p.m. and 2 a.m. [3]. The maximum ML concentration in serum was reached between 2:00 a.m. and 6:00 a.m., and in urine between 6:00 a.m. and 8:00 a.m. [3]. ML peak is reached slightly earlier in saliva than in blood samples (between midnight and 3 a.m.) [16]. Previous studies have shown the highest measurement accuracy when measuring ML in serum every 20â30 minutes [3]. However, this requires a venous catheter, which is why we have chosen to use saliva samples for this study as part of the riskâbenefit assessment. Lewy already described in 1983 that this type of ML measurement is least prone to error when performed in dim light [13]. Previous studies on ML measurement in saliva were conducted with a sampling interval of 30 min [3], with the authors recommending frequent measurements during the period when the greatest change is expected. Intraindividual ML measurements are considered stable [3] while interindividual differences make the establishment of reference values difficult.
The aim of this study is to quantify the preoperative ML levels of patients with symptomatic PC and to correlate them with symptoms.
Methods
We conducted a monocentric prospective study on non-hydrocephalic patients with symptomatic PC between March 2023 and May 2024. Ethical approval was obtained by the ethics committee of the University Hospital Göttingen (No. 16/1/23).
Patients
We included patients who were over 18Â years of age with evidence of a symptomatic PC on MRI scan without evidence of hydrocephalus or a pineal tumor. We defined typical symptoms as headaches, difficulty falling asleep and staying asleep, and visual disturbances, after other underlying pathologies had been ruled out by MRI, neurological and, where necessary, ophthalmological examinations. Previous illnesses and current medication were recorded with a particular focus on migraine, hormonal disorders, and any prior neurosurgical procedures. ML intake during the last 2Â weeks was an exclusion criterion. In addition, the duration of symptoms in years and the year of initial diagnosis were recorded.
MRI
Volumetric measurement of PC in T2
Assessment of specific symptoms
Headache and sleep disturbances were assessed over the course of a week at home in the form of a diary. Difficulties falling asleep, difficulties staying asleep, and the duration of sleep per day were recorded, as well as the frequency and severity of headaches using the Visual Analog Scale (VAS). In addition, visual disturbances and nausea were queried, and specific or nonspecific symptoms were recorded using free text.
Melatonin level measurement
Salivary ML levels were measured preoperatively on the first night following hospital admission (two days before scheduled surgery), with samples collected between 8:30 p.m. and 1:00 a.m.. To reduce physical and psychological stress, we decided against later measurement times. In order to rule out short-term fluctuations and measurement errors, measurements were taken every 30Â minutes, even if reference values were not available for each specific point in time. Patients were instructed not to brush their teeth, eat or drink two hours before and during the measurement period. 1â2Â hours before the measurement, the room was darkened and compliance with dim light exposure was randomly checked by the nursing staff. Patients were not allowed to use bedside lamps and were restricted in their use of cell phones; watching TV or reading was not permitted. Patients were instructed to lie down in bed and rest as if they were trying to fall asleep.
The lighting conditions only allowed them to orient themselves in the room in order to provide the samples. The saliva samples were collected using a straw without any stimulating measures. After receiving appropriate instructions, the patients independently provided saliva samples in a test tube for each measurement, and the samples were then stored in a dark box. The patient room was kept dimly lit throughout the measurement period. Sample collection was regularly supervised by a nurse. ML levels were determined in an external laboratory. The samples were sent in a dark container the morning after the measurement. ML levels were then determined using Enzyme-Linked Immunosorbent Assay (IBL).
Statistics
Statistical analysis was performed using SPSS (IBM, Version 30.0). We analyzed the relationship between symptoms and the ML level using Spearman's correlation and MannâWhitney-U test. To assess whether the measured salivary ML concentrations in the subjects significantly deviate from expected reference values, two statistical tests were applied. First, a binomial test was performed to evaluate whether the proportion of measurements falling within the normative reference range (assumed 95%) differs significantly from this expected proportion. The null hypothesis for this test states that the success probability (measurements within the reference range) equals 0.95. Second, to analyze the continuous concentration values, a Wilcoxon signed-rank test was conducted to determine if the median difference between the observed values and the reference values differs significantly from zero, without assuming normal data distribution. All tests used a significance level of 0.05.
Results
ML levels per patient throughout the measuring period. The progression of melatonin concentration per individual is shown in the respective colors. The red lines delimit the reference values
Discussion
This prospective study is one of the first to examine ML levels in non-hydrocephalic patients with symptomatic PC. To date, mainly three underlying pathologies have been discussed as possible causes of the typical symptoms experienced by patients with PC. Those are compression of the internal cerebral veins, intermittent obstruction of the aqueduct and a change in ML levels.
The main finding of this study is a noticeable lack of increase of ML during the night in most patients. While six patients were still within the normal range at 8:30 p.m. (p = 0.001), the number significantly dropped to two patients at 10:00 p.m. and to zero at 00:30 a.m., with 2 patients being above and seven patients being below the normal range. However, due to interindividual differences and limited data, the reference values must be interpreted with caution. The deviation at individual measurement points can only be assessed to a limited extent. However, the lack of an increase during the night is relevant. In other studies, the determination of ML in saliva in healthy subjects showed an increase in its course corresponding to that of blood samples. This cannot be observed in our cohort of PC patients. Benloucif et al. showed ML levels of approximately 5 pg/ml at 8 p.m. and approximately 20 pg/ml at 10 p.m. Analysis of individual patient factors, such as symptoms and cyst size, did not reveal any significant differences between the patients. This finding requires further investigation in a larger cohort.
In 2013, MĂĄjovsky et al. investigated serum ML levels in five symptomatic and two asymptomatic patients with PC [15]. However, sleep disorders were not recorded, but mainly headaches and dizziness as well as neurological focal deficits. No significant difference was found between symptomatic and asymptomatic patients, although ML levels were slightly lower in symptomatic patients. They therefore propagated normal ML secretion in patients with PC. However, that study did not look at any deviation from reference values due to possible individual variations. Our study with symptomatic patients, however, revealed a different situation. We observed noticeably low ML values, especially after 10 p.m. There were even significant deviations for the times 8:30 p.m. and 01:00 a.m. Another difference between the studies is the measurement method using blood samples or saliva. For our study, patients had to be woken up every 30 min to provide saliva samples. For a hormone that is directly related to the sleep cycle, this must be discussed as a possible limitation. On the other hand, several studies have rated the measurement of ML in saliva as reliable [3, 17, 23]. Despite waking the test subjects, similar peaks were observed in healthy subjects as in blood samples, although the absolute values in saliva are lower. In our cohort, there was no statistically significant correlation between the documented symptoms and the deviation of ML levels below the norm. Since all patients were symptomatic, the variance in the cohort was probably too small to achieve significance.
Due to the small number of cases and the lack of a healthy comparison cohort, there are limitations regarding the statistical significance in our cohort. Nevertheless, our results show a very noticeable tendency toward a lack of increase in ML at night in non-hydrocephalic patients with symptomatic PC. Furthermore, we consider our results to be valid, as previous studies on the reproducibility of ML measurements showed no relevant intraindividual variation over a period of several weeks [19]. The fact that our cohort consists solely of women does not mitigate this through possible hormonal influences, since previous studies showed no fluctuation in ML concentration depending on the menstrual cycle [20]. Moreover, we did not include patients who had taken ML within the last two weeks, in order to allow sufficient time between the last oral ML substitution and both the period during which symptoms were recorded and the period during which ML was measured in saliva. Studies have shown an almost complete degradation of the substituted ML after 4 h, so we are definitely outside this time frame [21]. Nevertheless, depending on the study and measurement method, the reference values given vary slightly, but are always above the references we assume and thus above those of our patients with PC [16, 23]. While MĂĄjovsky et al. measured serum ML, we decided to measure saliva ML, since it is less invasive. Although the concentration of ML in saliva is lower than in serum, it is reproducibly subject to the same circadian rhythm [16, 23]. Therefore, we consider our results to be sufficiently valid to conclude that ML levels in patients with symptomatic PC are below normal.
It is also important to note that we focused on sleep disturbances among the symptoms, which were not explicitly mentioned in MĂĄjovsky's study. We therefore consider our cohort to be particularly significant, as ML plays a key role in the sleep cycle, even in healthy individuals, and because sleep disturbances are present in up to 21â27% of patients with symptomatic PC [9, 11].
Limitations
Although our prospective study provides important insights into ML levels in patients with symptomatic PC, it has limitations. The sample size, and consequently the statistical power, is low. Furthermore, no control group with asymptomatic PC was analyzed. Due to interindividual variations, the determination of ML reference values and thus the interpretation of deviations of individual values from these can only be assessed to a limited extent. The external MRI scans did not allow the use of sequences required to analyze thalamic edema as a differential diagnosis due to potential venous compression. On the other hand, other studies have shown that ADC is not an adequate diagnostic tool in patients with PC, and there was no evidence of thalamic edema as the cause [10]. As a next step, a multicenter study should investigate both symptomatic and asymptomatic PC patients, and the influence on postoperative outcome needs to be clarified. Nonetheless, we consider our results a significant contribution to PC research and an important basis for future studies. This includes investigation in a larger cohort, postoperative measurement of ML, and analysis of oral ML substitution as a treatment option.
Conclusion
The ML level in non-hydrocephalic patients with symptomatic PC shows a lack of increase at night, which may provide an explanation for the symptoms. As a next step more symptomatic and asymptomatic patients need to be analyzed, as well as the influence of ML levels on the postoperative outcome.
Supplementary Information
Below is the link to the electronic supplementary material. ESM 1 Supplementary Material 1 (DOCXÂ 16.5 KB)