In addition to presenting the effects of hypoprolactinemia on depression and sexual functions, the present study has demonstrated cut-off levels for basal and stimulated PRL levels for the diagnosis of hypoprolactinemia in different genders. A peak PRL response to TRH of > 18.3 ng/ml in men and > 41.6 ng/ml in women can be accepted as sufficient responses to the test. Basal PRL levels of ≤ 5.7 ng/ml in males and ≤ 7.1 ng/ml in females could predict inadequate responses to the TRH stimulation test with 100% specificity and 70% sensitivity in males and 100% specificity and 80% sensitivity in females. Basal PRL levels of ≥ 8.5 ng/ml in males and ≥ 15.2 ng/ml in females could predict adequate PRL responses to TRH with 100% specificity and 76% sensitivity in males and 96% specifity and 66% sensitivity in females. So, in majority of the patients basal PRL can accurately define the sufficiency of PRL.

As mentioned above, PRL is a neglected hormone in the diagnosis of hypopituitarism. So, although some cut-off levels for basal PRL and PRL response to TRH are defined, not all of them are properly determined. Basal PRL levels below the lower limit of the assay used or some cut-off levels such as 1.8, 2.35, 3, 4 or 5 ng/ml have been suggested [10,11,12,13,14]. TRH stimulation test can also be used for the diagnosis of PRL deficiency [11, 15, 16]. But some of the studies investigated TRH stimulation test in other disorders such as hyperprolactinemia, other pituitary disorders or sellar mass lesions [17,18,19] or as a combined test for hypopituitarism [15]. The interpretation of TRH stimulation test can be done by calculating R value which is equal to peak PRL minus basal PRL divided by basal PRL. An R value of < 1 (doubling of basal PRL) was suggested to be as insufficient and > 2 as sufficient [19]. If we had used the R value in the present study as a diagnostic criteria, 4 healthy volunteers, 2 patients with sufficient PRL response to TRH and 3 patients with insufficient response to PRL would be included in the indeterminate group (R value:1–2), 2 patients would have a false positive test result and one false negative test result (Online Resource 1). As can be seen from the table (Online Resource 1), a peak PRL level of 2.5 ng/ml can be in the indeterminate zone, on the other hand, a peak PRL level of 31.8 ng/ml can be diagnosed as an insufficient response according to R value. So incremental responses during dynamic stimulation tests can be affected from basal values and sometimes blunted responses can be seen. It doesn’t seem reasonable to consider every blunted PRL response to the TRH test as hypoprolactinemia without taking into account its absolute value.

In an analysis of patients with Sheehan syndrome, we suggested a basal cut-off level of < 4 ng/ml for hypoprolactinemia and > 7.8 ng/ml for normal PRL levels [11]. However, in contrast to the present study, all patients had a diagnosis of Sheehan syndrome and postpartum lactation could be questioned in each of them. On the other hand, since healthy people were not included in the study, reference values ​​for basal PRL were determined according to the history of postpartum lactation and TRH stimulation test, if available [11]. In the present study, we determined the cut-off levels according to TRH stimulation test results in healthy people. Use of basal PRL instead of TRH stimulation test in the diagnosis of hypoprolactinemia is important considering the limited availability of TRH ampul. This can substantially limit the necessity of TRH stimulation test.

Metoclopramide is another pharmacological agent that can increase serum PRL levels. It has been used for differential diagnosis of hyperprolactinemic disorders such as polycystic ovary syndrome or pituitary adenomas [20,21,22]. However, although metoclopramide may increase PRL levels, it is not a standardized method to evaluate PRL reserve in patients with hypopituitarism.

The causes of panhypopituitarism in the present study were surgery in 31% and surgery + radiotherapy in 10% and Sheehan syndrome in 29%. All of the patients who received combined surgery and radiotherapy or had Sheehan syndrome were PRL deficient. Sheehan syndrome, and combined surgery and radiotherapy were found to be more common in PRL deficient patients. The prevalence of hypoprolactinemia was 87.5% among patients with panhypopituitarism. The number of subjects with sufficient PRL levels (12.5%) was limited in the present study. It is important to note that the study exclusively focused on patients with panhypopituitarism, excluding individuals with acromegaly, prolactinoma and active Cushing disease. These selective inclusion criteria, hence the low number of patients with normal PRL levels might influence the distribution of hypopituitarism etiologies. The prevalence of hypoprolactinemia in patients with a lower number of deficient pituitary hormones is expected to be lower, certainly. In a retrospective study, hypoprolactinemia was found to be 27% among patients with hypopituitarism (13 mild PRL deficiency: 3–5 ng/ml and 14 severe PRL deficiency < 3 ng/ml) [23]. The frequency of hypoprolactinemia is not known well, since there is no clinical need for the diagnosis hypoprolactinemia in daily practice. One of the most common causes of hypoprolactinemia is Sheehan syndrome [4]. Younger age, etiologies such as pituitary apoplexy, craniopharyngioma, and idiopathic hypogonadotropic hypogonadism were suggested to be associated with PRL deficiency [23].

One interesting finding in the present study was the presence of PRL deficiency in all patients with diabetes insipidus. The etiologies of hypopituitarism in these five patients were surgery followed by radiotherapy in all of them (2 for non-functioning pituitary adenoma, 2 for craniopharyngioma and 1 for Cushing disease). Radiotherapy was probably associated with more profound anterior pituitary destruction hence hypoprolactinemia in these patients.

Acquired severe PRL deficiency was suggested to be a marker for more severe hypopituitarism, particularly GH deficiency [13, 23]. This may probably explain the high prevalence of PRL deficiency in the present study, since all included patients had concomitant GH, TSH, ACTH and gonadotropin deficiency. Furthermore, positive correlations of basal and stimulated PRL levels with IGF-1 support this situation. In severe GH deficiency, PRL deficiency was found to be independently associated with low IGF-1 levels. It was suggested that serum PRL might contribute to IGF-1 release in the absence of GH which may explain lower IGF-1 levels in both GH and PRL-deficient patients [14]. However, the role of PRL in IGF-1 generation in healthy people is not known. In the present study, unlike patients with hypopituitarism, no correlations of basal and stimulated PRL levels with IGF-1 were found in the healthy control group.

The effects of true hypoprolactinemia due to destruction of lactotrophs is not known well. Low-normal PRL levels, distinct from hypopituitarism, were found to be associated with type 2 diabetes, metabolic disease and in children with obesity which is reversed after weight loss [12, 24]. Large population-based studies demonstrated the association of low PRL levels with increased prevalence of type 2 DM particularly in women [25, 26]. Presence of PRL receptor is shown to be important for beta cell expansion during pregnancy [27]. Women with gestational diabetes with accompanying low PRL levels during pregnancy were found to have an increased risk of developing type 2 diabetes in their later life [28].

Kyrsiak et al. has shown that cabergoline induced hypoprolactinemia (< 5 ng/ml) was associated with higher levels of 2-h postchallenge glucose, HbA1c level, triglyceride, hsCRP, and lower HDL-cholesterol than patients with normal PRL levels and healthy individuals and these effects were reversible after dose reduction of cabergoline [6]. On the other hand, bromocriptine treatment, which is a dopamine agonist that lowers PRL levels, improves obesity, metabolic syndrome and cardiovascular parameters [29]. The effects of bromocriptine seems to be paradoxical since it both lowers PRL and improves metabolic parameters. So the negative feedback regulation of the dopamine system by PRL may partially explain metabolic abnormalities seen in hypoprolactinemia [30]. In other words, increased PRL leading to increased dopamine, decreased PRL leading to decreased dopamine. In the present study, neither fasting glucose nor lipid levels showed a difference from the control group or in between subgroups of PRL sufficient and insufficient hypopituitary subjects. However, the effects of other pituitary hormone deficiencies and their replacement therapies can also have some confounding effects on glucose metabolism.

Depression scores were found to be higher and male sexual function scores were found to be lower in patients with panhypopituitarism with accompanying PRL deficiency than in healthy control group. Depression scores were higher in patients with hypoprolactinemia than in the control group, but did not show a significant difference from patients with sufficient PRL. So hypoprolactinemia may be associated with higher depressive scores in patients with hypopituitarism. Kyrsiak et al. has previously shown that dopamine agonist-induced subnormal PRL levels were characterised by reduced well-being in both men and women [31, 32]. GH deficiency may also contribute to the depressive symptoms of the patients. To address the effects of other hormone deficiencies, patients underwent evaluation under replacement therapies. However, GH replacement therapy was commenced on to the patients who were willing to receive GH. Because daily requirement of an injectable therapy, concerns of the patients about the side effects of GH when they are informed or lack of clinical benefits after replacement in some patients resulted in discontinuation or not starting GH therapy. It would be perfect if we could reach the targeted IGF-1 levels in all patients to purify the effects of PRL which may be a limitation of the study. Nevertheless, despite being slightly lower in subjects deficient in PRL compared to those with sufficient PRL, IGF-1 levels were similar in both groups reducing the impact of GH on metabolic or other parameters.

Sexual functions were evaluated using valid questionnaires designed for different genders. In the entire hypopituitarism group, male sexual functions were lower compared to the control group. Specifically, in PRL-deficient subjects, male sexual functions were lower than in the control group but similar to PRL-sufficient subjects. Despite testosterone replacement in all male subjects, variations in testosterone levels during replacement and the presence of GH deficiency might still influence sexual functions. Previous findings have indicated that dopamine agonist-induced subnormal PRL levels were associated with impaired sexual functioning in men [31, 32]. Low PRL levels may indeed be associated with sexual dysfunction, and this association could be partially explained by increased dopaminergic tone and a low central serotonin tone [33]. It is important to note that the relationship between hormones, neurotransmitters, and sexual function is complex, and individual responses can vary. Clinical assessments and further research are usually necessary to better understand the specific mechanisms and implications of low PRL levels on sexual dysfunction.

Dopamine agonist-induced sexual dysfunction can also be seen in females [31]. In the present study, female sexual functions do not appear to be significantly affected by hypopituitarism, but since all female patients had PRL deficiency, the effects of hypoprolactinemia could not be evaluated. However, an interesting observation is made regarding the potential influence of societal attitudes on the sexual lives of unmarried female patients, suggesting that conservative mentalities may have inhibitory effects. This raises a question about the validity of the questionnaire in the female group, possibly indicating that social and cultural factors might play a role in shaping perceptions of female sexual functioning. It is crucial to recognize that assessing sexual function involves not only physiological aspects but also psychological and sociocultural factors. Cultural norms, societal expectations, and personal beliefs can influence an individual’s perception of sexual well-being. The validity of questionnaires in assessing female sexual function may indeed be influenced by such factors. Further research may be necessary for a better understanding of the interplay between hormonal factors, societal attitudes, and female sexual functions.

On the other hand, a family analysis, including 3 women with postpartum agalactia (2 sisters and one niece) with a heterozygous mutation identified in the exon 5 of PRL gene, revealed that PRL deficiency can be related to decreased fertility and early menopause, but normal sexual development in females [34].

Recombinant PRL has been used in mothers with lactation insufficiency with or without PRL deficiency and improved milk secretion and content [35]. Lactation is a complex procedure and lack of PRL secretion is seen only in a minority of the mothers. So, other pharmaceutical galactogogues such as domperidone, metoclopramide, GH, metformin and TRH have been tried in the treatment of lactation insufficiency [36]. Acute opioid administration was also shown to increase PRL levels possibly acting through the dopaminergic system [37, 38]. Studies are warranted for defining the therapeutic effects of recombinant PRL, other than lactation, in PRL deficiency.

In conclusion, PRL deficiency is prevalent among individuals with panhypopituitarism, with the potential to result in elevated depression scores in both sexes and impaired sexual functions in males. Notably, the observed lack of PRL effects on BMI, glucose, and lipid levels may be attributed to the limited number of patients with sufficient PRL in the study. A basal PRL level seems to be sufficient for the diagnosis of hypoprolactinemia in routine clinical practice and TRH stimulation test can be reserved for unequivocal results, if it will change our clinical approach, that may presumably occur in the future.