1 Introduction

Traumatic brain injury (TBI), defined as a blunt or penetrating trauma causing structural and/or functional damage to the brain, affects 47.3 to 849 per 100,000 population per year across Europe [1, 2]. The prevalence widely varies depending on the studied populations' countries, genders, and ages [2]. Traumatic brain injury may cause transient or irreversible neurologic disabilities as well as disruptions in neuroendocrinological functions due to direct or indirect damage to the hypothalamus and pituitary gland [3].

The prevalence of pituitary dysfunction during long-term follow-up after TBI has been reported in a wide range from 16 to 70% in prospective studies [4]. The rate of pooled prevalence was 15% during the chronic phase (> 3 months) when studies with confirmatory testing were analyzed [5]. The most common hormonal disruption across studies was growth hormone (GH) deficiency, followed by central adrenal insufficiency, central hypogonadism, and central hypothyroidism [5]. However, the rates of dysfunctional axes may vary across studies. Pituitary dysfunction may occur during the acute or chronic phase following TBI and may have a dynamic course. Screening algorithms have been developed, as the symptomatology of pituitary dysfunction after TBI might be subtle and overlooked but may have detrimental effects on quality of life and may even carry mortality risk [5].

Prolactin is a hormone mainly known for its role during lactation. Its secretion is mainly controlled by the tonic inhibition of the hypothalamic secretion of dopamine. High prolactin levels might be caused by lesions that disrupt dopamine flow to the anterior pituitary, such as pituitary adenomas. Many other etiologies may lead to stalk disruption and hyperprolactinemia, such as inflammations, infections, injuries, extra-pituitary tumors, and congenital etiologies [6]. Prolactin-secreting tumors of the pituitary are another etiologic factor for high prolactin levels. The well-known clinical manifestations of hyperprolactinemia are hypogonadism symptoms, galactorrhea, and infertility. On the other hand, low serum prolactin might result from genetic etiologies, destroyed lactotrophs due to anterior pituitary damage, and medications [6]. The main clinical manifestation of hypoprolactinemia is alactogenesis in women, however, recent literature points out to some further associations such as metabolic syndrome and type 2 diabetes mellitus [7].

The prevalence of serum prolactin disturbances was reported in a wide range from 0% to 85% following TBI, either as hyper- or hypoprolactinemia [8, 9]. Stress response, fluctuations in dopaminergic activity and other mediators, and hypothalamic and/or pituitary gland injuries may cause derangements in serum prolactin levels after TBI [6, 10]. Prolactin measurements provide insight into the functionality of the hypothalamus and pituitary gland in patients with TBI. Some authors associated it with TBI's severity and long-term outcome [10,11,12]. However, pathophysiologic mechanisms, prognostic significance, and association with other pituitary hormone dysfunctions are incompletely understood.

Here, we review the literature regarding the prevalence, pathophysiology, and clinical implications of serum prolactin disturbances during the acute and chronic phases following TBI. We re-evaluate the data of our patients with TBI regarding prolactin levels during prospective long-term follow-up and discuss the results in light of current literature.

2 Prolactin disturbances following TBI

2.1 Factors that might confound with prolactin measurements in patients with TBI

It might be challenging to investigate the direct effects of TBI on serum prolactin levels due to many potential confounding factors in these patients. Medications regularly used by the patients, such as tricyclic antidepressants, antipsychotics, opioids, and some SSRIs (fluoxetine) or those given after admissions to the hospital after TBI, such as dopamine, glucocorticoids, and metoclopramide are among the confounding factors that might increase or decrease prolactin levels [6, 13, 14]. Chest wall injuries, renal and hepatic dysfunctions, hypothyroidism, coinciding hypothalamic or pituitary tumors, presence of macroprolactinemia, states of pregnancy, and breastfeeding are among the other factors that should be excluded while evaluating the associations between TBI and serum prolactin levels [14].

The mediators that play a role in prolactin secretion other than dopamine may also complicate the interpretation of serum prolactin imbalances following TBI. Prolactin is known to increase in response to stress [15], and even anxiety might provoke the increase [16]. Mediators such as dopamine, cytokines, and oxytocin are thought to play a role in prolactin increase during stress. However, the underlying mechanism still needs to be fully elucidated [17, 18]. An increase in prolactin was detected following exercise in athletes without any head trauma, explained to be secondary to increased central serotoninergic activity [19, 20]. Further studies reported increased prolactin levels following exercise, explaining that the increase was caused by metabolic mediators such as lactate and some cytokines [18, 21, 22]. It is reasonable to expect that these mediators might contribute to serum prolactin imbalances in patients with severe trauma who might suffer from conditions such as tissue hypoperfusion or sepsis besides TBI.

2.2 Hypothalamo-pituitary injuries causing serum prolactin derangements following TBI: Pathophysiology

Prolactin is synthesized and secreted from lactotrophs located in the lateral regions of the anterior pituitary, and the vascular supply mainly depends on the portal venous system [23, 24]. The release of prolactin from the pituitary is mainly regulated by the inhibitory effects of hypothalamic dopamine secretion [6]. Prolactin secretion is also modulated by some other stimulatory mediators such as TRH, serotonin, oxytocin, and estrogen, and suppressive neurotransmitters such as GABA [6, 14, 25].

TBI may cause damage to the hypothalamus and/or the pituitary gland through direct or indirect mechanisms. Direct injuries may be caused by so called ‘primary insults’ such as skull base fractures, shearing injuries, or compression by a hematoma or brain edema [5, 26,27,28]. Autopsy studies reported microhemorrhages, ischemic lesions, or both in the hypothalamus, more commonly among those with middle fossae fractures [29]. Anterior or posterior lobe hemorrhages, necrosis, and infarcts of varying extent were observed in the pituitary gland, and complete or partial transection of the stalk [30,31,32,33]. The distal transection of the stalk may lead to anterior pituitary infarcts due to impaired portal vascular supply [27]. Apart from these direct injuries, secondary insults such as increased intracranial pressure, hypotension, anemia, and hypoxia that occur following trauma may also cause damage to the hypothalamo-pituitary region [5, 34]. Autoimmunity is another mechanism proposed to cause pituitary dysfunction in the long term after TBI [35]. Genetic polymorphisms might be associated with the hypopituitarism risk after TBI [36].

Hyperprolactinemia following TBI has been recognized in some early reports [3]. Benvenga et al. reported a prevalence of 47.7% in their systematic review, including studies since 1970 [3]. Patients with varying evaluation time intervals from less than 1 year to more than 20 years after trauma were analyzed. The authors stated that this rate was in alignment with the prevalence of stalk and hypothalamic injuries from autopsy studies [3]. Hyperprolactinemia following TBI was generally associated with hypothalamic injuries disrupting dopaminergic inhibition. At the same time, hypoprolactinemia might result from lactotroph damage secondary to anterior pituitary damage such as infarcts, ischemia or necrosis [37, 38]. The pathologic findings of hypothalamo-pituitary damage after TBI mentioned above were generally derived from autopsy series, and there are no studies directly reporting associations between the lesions and serum prolactin levels or other pituitary hormones. Moreover, these might represent the most severe cases as the data were recruited from the mortal cases.

On the other hand, TBI may cause serum prolactin derangements on a functional basis [39]. Prolactin release might be affected by dopamine signal disruptions and other mediators. Baseline hormone levels and stimulation tests were used to evaluate functional disorders. As prolactin is under tonic inhibition of hypothalamic dopamine, hypothalamic injuries disrupting dopamine flow would result in high prolactin levels. In contrast, lactotroph damage would result in low prolactin and non-responsiveness to TRH stimulation. Tolis et al. analyzed functional prolactin secretion in patients with hypothalamo-pituitary dysfunction secondary to various etiologies [40]. Patients with panhypopituitarism had persistently low prolactin levels and, blunted response to TRH and chlorpromazine stimulation, and L-Dopa suppression. The patients with hypothalamic damage had elevated basal prolactin levels with a responsive TRH stimulation test but unresponsive chlorpromazine stimulation and L-Dopa tests, while those with intact hypothalamic functions had normal serum prolactin levels and were responsive to all tests [40].

Some early studies investigated the aminergic turnover and concentrations of neurotransmitter metabolites in patients after severe TBI. Porta et al. reported increased concentrations of homovanillic acid (HVA) and 5-hydroxyindole-acetic acid (5HIAA) (metabolites of dopamine and serotonin, respectively) in ventricular cerebrospinal fluid (CSF) of the patients during acute-phase following severe TBI [41]. Markianos et al. confirmed high levels of HVA and 5HIAA in CSF in comatose patients after severe TBI [42]. In addition, HVA and 5HIAA prospectively decreased in the patients who improved but remained high in those with no improvement [42]. These data provide evidence of a dynamic change in brain amines following TBI. However, it is not known to what extent the alteration of these metabolites affects pituitary prolactin secretion. DeMarinis et al. reported a surprisingly increased response of prolactin to GHRH stimulation in comatose patients after severe TBI [43]. The stimulatory response increased from day 2 to day 16, correlating with improved clinical status. They interpreted the results to reflect altered central aminergic and peptidergic activities, such as decreased central dopaminergic tone [43]. A correlation between prolactin response to GHRH stimulation and outcome was reported by the same group in another study [44].

2.3 The role of prolactin in brain injuries

Recent studies revealed that prolactin is among the mediators that play a role during neurogenesis, neuroplasticity, and neuroprotection after neural injury [45]. Prolactin receptors are expressed widely in various brain regions such as the cerebral cortex, hippocampus, choroid plexus, thalamus, basal ganglia, and cerebellum, and various mechanisms have been proposed for their neural effects [45]. Prolactin has been reported as a factor that induces neurogenesis in mouse hippocampal cells both in vitro and in vivo [46]. Vermani et al. observed the protective role of prolactin during ischemic brain injury in rats; administration of prolactin decreased the neurotransmitter levels and reduced tissue damage [47]. On the other hand, it has been reported that prolactin has modulatory roles in seizure activity [48].

Excitotoxicity is a process caused by the excessive release of excitatory neurotransmitters that trigger post-synaptic neuron hyperactivation, leading to neuronal cell death. Excitotoxicity is one of the pathophysiologic mechanisms leading to neuronal apoptosis following TBI [49]. Vanoye et al. observed for the first time that the hippocampal area of lactating rats was protected against neuronal excitotoxic damage induced by kainic acid compared to rats in the diestrus phase [50, 51]. The authors hypothesized that lactation had a neuroprotective effect due to the actions of prolactin [51]. They treated rats with daily prolactin before induction of excitotoxicity by kainic acid and observed that prolactin-treated rats were protected from neural damage [52]. Prolactin exerts its neuroprotective effects against excitotoxicity via prolactin receptors [53]. Rivero-Segura et al. reported that the underlying mechanism of the neuroprotective role of prolactin is the attenuation of intracellular calcium overload, maintenance of normal mitochondrial activity, and NF-KB activation [54]. Another mechanism is that prolactin has antioxidant effects in neural tissue by decreasing the levels of lipid peroxidation products and preventing mitochondrial dysfunction [55].

Prolactin receptors are also expressed on glial cells besides neurons [56], and prolactin plays a role in glial responses after cerebral injury [57]. Prolactin has a protective role from oxidative insult in astrocytes in vitro [58]. Incubation of astrocyte cultures with prolactin caused increased activity of antioxidant enzymes, decreased accumulation of reactive oxygen species, and reduced cellular death [58]. It has been reported that prolactin also has anti-inflammatory effects in microglial cells due to the down-regulation of pro-inflammatory substances [59].

Most data regarding the effects of prolactin on neural tissue have been derived either from in vitro or in vivo animal studies. The effects of prolactin on human brain is a largely unexplored area. In a recent study, Tani et al. investigated the response of human neurons obtained from brain tissues from autopsy cases [60]. They observed that cells exposed to prolactin had increased neurite formation and that prolactin had protective effects against hypoxia-induced cell death [60].

The data regarding prolactin's role in neural protection, neurogenesis, and neuroinflammation suggest that it might play a role during the recovery phase of TBI. However, no studies have investigated the effects of prolactin in patients with TBI. Future studies will provide more data on the role of prolactin after neural injuries, the underlying mechanisms, and, eventually, its therapeutic potential.

2.4 Prolactin and radiologic findings in patients with TBI

There have been many attempts to investigate if any predictive associations exist between radiologic findings and pituitary dysfunctions in patients with TBI. The presence of basal skull fractures, diffuse axonal injury, diffuse brain swelling, and subdural hemorrhage on CT were reported as risk factors for pituitary dysfunction [28, 32, 61, 62]. However, to the best of our knowledge, only one study reported the associations between imaging studies and prolactin levels in the literature. Zheng et al. compared apparent diffusion coefficients (ADC) between patients with TBI and healthy controls using diffusion-weighted imaging in MRI studies [63]. ADC values of pituitary were decreased in patients with pituitary dysfunction compared to those with intact pituitary functions 6 months following TBI. Interestingly, patients with higher prolactin levels had lower ADC values than those with normal prolactin levels. They also reported a correlation between ADC values and GOS, implying an association with injury severity [63].

2.5 Serum prolactin levels in sports related concussion

There is paucity of studies investigating prolactin levels in athletes and its possible use in the outcome or return to play. LaFountaine et al. hypothesized that concussion may cause altered dopaminergic tone reflecting dynamically on serum prolactin levels [64]. They prospectively analyzed serum prolactin 48 h, 7, and 14 days after concussion in 4 male athletes [64]. Prolactin levels remained in reference ranges but were at nadir at the 48th hour and showed an increase at first and second-week sampling. The authors speculated that the decreased prolactin levels during the acute phase were caused by a functional increase in dopaminergic activation of an intact hypothalamus and pituitary. They also reported that an increase in serum prolactin accompanied the symptom improvements and the proper timing of return-to-play [64]. Battista et al. compared the neuroendocrine hormone levels between athletes with and without a sports related concussion 2 to 7 days after injury [65]. They reported similar prolactin levels between the groups. However, serum prolactin levels were positively associated with time to medical clearance in the group with head injury. These results are contradicting the findings of LaFountaine et al. Moreover, prolactin levels were associated positively with cognitive disturbances [65]. Further studies eliminating confounders for prolactin may provide more data.

3 Prolactin during the acute phase following TBI

An increasing number of studies have investigated the pituitary functions following TBI. We searched the literature for studies reporting on prolactin levels after TBI. The keywords 'prolactin, hypoprolactinemia, hyperprolactinemia, traumatic brain injury, TBI, head trauma, pituitary, pituitary dysfunction' were used. The studies recognized during further search were also added, and the results are displayed in Table 1.

Table 1 Studies reporting prolactin levels in patients with TBI
Full size table

The acute phase following TBI is defined as 2 weeks after TBI in general [5]. Serum prolactin levels might be increased, stable, or decreased. The prevalence of hyperprolactinemia was in the range of 0% to 86%, and of hypoprolactinemia, 0% to 6% during the acute phase (Table 1). Only a few studies reported on hypoprolactinemia during this phase. The wide range of prolactin levels and other pituitary hormones across studies might be caused by different characteristics and severities of TBI, heterogeneous study methodologies with varying diagnostic tests and cut-offs, and diverse patient characteristics [5, 66].

Agha et al. reported a rate of hyperprolactinemia of 52% among patients with moderate to severe TBI after a median of 12 days (7 to 20 days) [67]. The authors observed a negative association between serum prolactin levels and GCS [67]. Tanriverdi et al. prospectively evaluated 52 patients during the first 24 h and one year following TBI and reported hyperprolactinemia in 12% of the patients during the acute phase [73]. They confirmed the negative correlation between serum prolactin and GCS [73]. In another study, Tanriverdi et al. investigated 104 patients with TBI during the first 24 h after admission [38]. Hyperprolactinemia was detected in 18.3% of the patients, and high prolactin levels were associated with low GCS scores [38]. As prolactin is mainly under the inhibitory control of dopamine, it was suggested that hyperprolactinemia was caused by hypothalamic injury rather than pituitary. As the patients with mild TBI had various anterior pituitary hormone deficiencies but normal prolactin levels, the authors hypothesized that more severe trauma is required for stalk and hypothalamic injuries. In contrast, mild traumas may generally cause pituitary damage [38]. Lower rates of hyperprolactinemia in both studies performed by Tanriverdi et al. might be explained by the fact that nearly half of the patients had mild TBI.

In the study conducted by Matsuura et al., TRH stimulation was performed in patients with TBI immediately after admission before any medications were started [82]. They also observed that the patients with more severe TBI, evaluated through GCS, had higher baseline prolactin levels but also blunted responses after TRH stimulation. The authors discussed the possibility that the low response to TRH might either be associated with anterior pituitary damage or an effect of negative feedback caused by hyperprolactinemia [82].

Contrary to these results, some authors reported no associations between prolactin levels and GCS [9, 75], while others reported a positive correlation [11, 79] (Table 1). Chiolero et al. reported that the patients with the most severe TBI and the worst outcome had the lowest prolactin levels. The decreased prolactin level was associated with the severity of TBI, intracranial pressure, and outcome [11].

One of the reasons behind these conflicting results might be that the use of GCS score solely is not an ideal measure for the severity of head trauma. Besides the limitations that might affect the scoring, such as the timing of assessment, use of sedating medications, and inter-observer bias, GCS does not include some parameters such as transient loss of consciousness, amnesia, and location of the lesion [83, 84]. Considering these limitations, Tanriverdi et al. proposed a screening algorithm defining a subgroup of 'complicated mild TBI' which included factors such as hospitalization, ICU monitoring, and radiological changes [5].

Hypoprolactinemia was observed in some patients during acute-phase following TBI. It was generally less frequently reported than hyperprolactinemia [3] (Table 1). It might be hypothesized that alterations in neuroendocrine mediators, such as decreased dopaminergic tone, occur much more frequently following trauma than an injury to the anterior pituitary that will cause low prolactin levels. On the other hand, there might have been a reporting bias towards hyperprolactinemia as hypoprolactinemia has no notable clinical manifestations in patients with TBI and might have been neglected by the authors.

4 Prolactin during the chronic phase following TBI

Prospective studies have reported that prolactin levels can be high, normal, or low during the chronic phase after TBI [13, 71, 73, 76]. Persistently high prolactin levels might indicate hypothalamic injury. Case reports presenting patients with persistent hyperprolactinemia who were responsive to TRH stimulation but non-responsive to hypoglycemia and metyrapone during the chronic phase suggested an underlying hypothalamic injury or impaired portal circulation [85, 86].

The functioning of the pituitary gland following TBI is a dynamic process; novel hormonal deficiencies can develop, but improvements may also be observed [12, 87]. Daniel et al. reported vital cells surrounding the infarcted anterior lobe as a rim and neighboring the neural lobe. They also observed mitotic figures among these cells and discussed that these surviving cells might proliferate to regain function if the vascular supply is restored [27]. These pathologic findings explain pituitary dysfunction improving after months or even years following TBI [12]. On the other hand, pituitary functions may deteriorate over time after TBI. Autopsy reports performed years following TBI reported gliosis of the hypothalamus, pituitary atrophy, and atrophied target glands as chronic changes in patients with panhypopituitarism [88]. As fibrosis prevails in the pituitary gland during the long term, a radiologic image of empty sella may be observed [89, 90]. Like other anterior pituitary hormones, prolactin may have a dynamic course during long-term follow-up after TBI. It might be expected that prolactin levels decrease or improve over time. There is a need for further studies to investigate the rate of reversibility of hypoprolactinemia after TBI and its clinical implications.

The clinical significance of hyperprolactinemia is better defined than hypoprolactinemia. High prolactin, besides being an indicator for hypothalamic injuries, may cause its well-known symptoms of menstrual irregularities, galactorrhea, and fertility problems during long term follow-up after TBI. On the other hand, prolactin deficiency is generally a neglected diagnosis, most probably due to the lack of clinical signs and symptoms, except agalactia, and no clinical management strategies [91]. However, recent data evidences multiple regulatory roles of prolactin on behavior, stress response, bone homeostasis, and metabolism [6]. There have been an increasing number of studies pointing out to the associations between hypoprolactinemia and metabolic syndrome and type 2 diabetes mellitus [7, 92, 93].

Moreover, diagnosing hypoprolactinemia is not only important regarding clinical grounds, but also it might be associated with deficiencies of other pituitary hormones indicating a more severe pituitary damage [94]. In a recent review, it was proposed that patients with panhypopituitarism should be further classified based on the presence or absence of prolactin deficiency [91]. It might be an interesting area of research to detect metabolic risks in patients with hypoprolactinemia during the chronic phase following TBI.

Some studies investigated whether serum prolactin can be a predictive factor for a favorable long-term outcome after TBI. Zhong et al. reported no associations between prolactin levels during the acute phase and regain of consciousness at 6 months [95]. However, Marina et al. reported that an increase in stress hormone levels, including prolactin, in the third month following TBI was associated with a worse outcome in terms of functioning level in the first year [10]. The authors stated that the results reflect a worse outcome in patients with prolonged stress response rather than hypothalamopituitary dysfunction [10].

There is a paucity of studies investigating serum prolactin levels long-term after TBI. We re-evaluated the data of our previous study, reporting 5-year prospective results of patients with TBI. [12]. In the next section, we present the results.

4.1 Dynamic alteration of serum prolactin following TBI: Our experience

Tanriverdi et al. prospectively investigated the pituitary functions of 25 patients in the first and 5th year following TBI; 17 patients were also analyzed in 3rd year [12]. The authors analyzed the rates of improvement and deterioration of pituitary functions and associated factors. Serum samples were obtained for baseline hormonal evaluation, and relevant stimulation tests were performed. GH deficiency was the most common hormone deficiency, and the authors concluded that the patients with mild and moderate TBI were more likely to improve pituitary functions during follow-up than those with severe TBI [12]. However, the dynamic changes in prolactin levels were not reported then. Here, we present the prospective data on serum prolactin levels of these 25 patients. Serum levels of hormones obtained within 24 h of admission were also included in the analysis, besides 1st and 5th-year controls (Table 2). The 3rd year evaluation was omitted due to lack of data.

Table 2 Serum prolactin values of the patients with TBI during 5-year prospective follow-up
Full size table

Three patients (12%) had prolactin levels above the reference range during the acute phase; one developed prolactin deficiency in 5th year of control (Table 2). The median (IQR) values of serum prolactin were 8.34 (6.41–12.41) ng/mL during the acute phase, 9.56 (5.40–13.56) ng/mL at 1st-year control, and 5.75 (4.37–8.43) ng/mL at 5th-year control. The median values significantly decreased in the 5th year compared to baseline and 1st-year values (Fig. 1). However, there was no change between the baseline and 1st year prolactin values. A similar result regarding one-year follow-up was obtained in another prospective study performed at our center [73]. The serum prolactin levels of 52 patients during the first 24 h and one year following TBI were similar [73]. A significant decrease in prolactin levels at 5th-year control might reflect pituitary damage occurring over the years. It might be hypothesized that decreasing prolactin levels might be associated with pituitary volume. It is a known phenomenon that empty sella, in association with hormone deficiencies, may develop after TBI [89, 90, 96]. The gradual decrease in prolactin levels presented here might be paralleled with decreasing pituitary volume over time. However, the data of imaging studies is lacking. This assumption should be tested with further investigations.

Fig. 1
figure 1

The dynamic change in serum prolactin levels during prospective 5-year follow-up after TBI

Full size image

We also analyzed if there were any patients who developed prolactin deficiency during 5-year follow-up after TBI. However, there is no universally accepted cut-off level for prolactin deficiency, and different cut-offs have been used in the literature [92, 94, 97,98,99,100]. We adopted a cut-off level of 4 ng/mL proposed by Diri et al. [97]. They analyzed the patients with panhypopituitarism secondary to Sheehan syndrome in the study. They reported that none of the patients with baseline prolactin below 4 ng/mL had a sufficient response to the TRH stimulation test [97]. One may discuss that this cut-off is not appropriate as all the patients were females diagnosed with Sheehan syndrome [97], while the majority of the patients in Tanriverdi et al. [12] were male and had TBI. However, a newly published study conducted by our group included patients prediagnosed with panhypopituitarism to investigate the cut-off level for baseline serum prolactin to diagnose prolactin deficiency [100]. A baseline prolactin level of ≤ 5.7 ng/ml in males and 7.11 ng/ml in females was 100% specific and 80% and 70% sensitive, respectively, to predict an insufficient response to TRH stimulation [100]. We used an even lower cut-off of 4 ng/mL to prevent over-diagnosis. Moreover, both studies [12, 97] were performed at the same center at a similar time using the same laboratory facilities and analytical methods enabling a direct comparison of the values.

When the diagnostic cut-off of 4 ng/mL for baseline prolactin was used, 5 patients had prolactin deficiency in the 5th year of control. Four of them also had other pituitary hormone deficiencies: GH deficiency in 3, and GH deficiency and central hypogonadism in one during the 5th year (Table 2). GH deficiency, diagnosed by stimulation tests, was significantly more common among the patients who had prolactin deficiency (< 4 ng/mL) compared to those who were prolactin sufficient (≥ 4 ng/ml) (p = 0.012). Moreover, patients with low baseline prolactin (< 4 ng/mL) had significantly lower serum IGF-1 levels than those with higher prolactin levels (≥ 4 ng/ml) (median (IQR) 167 ng/mL (129.0–208.97) and 278 ng/mL (219.66–455.23), respectively) (p = 0.006).

According to these results, GH deficiency seems to be the most frequent hormone deficiency in patients with low prolactin following TBI. These results align with the previous reports associating low prolactin levels with GH deficiency and other pituitary hormone deficiencies [94, 98]. It is not known if the neighboring localization of somatotrophs and lactotrophs in the lateral wings of the gland might be a cause for this result. Another mechanism underlying the association between prolactin deficiency and low IGF-1 was a probable stimulatory role of prolactin over IGF-1 [99]. Furthermore, it might be interesting to investigate if low prolactin levels have a predictive role for GH deficiency in patients with TBI and if low prolactin could be used as an indication to proceed with further testing in patients with normal IGF-1 levels but suspected GH deficiency. Further studies need to verify these preliminary data of a limited number of patients.

Isolated prolactin deficiency (IPRLD) is a rare phenomenon that might be idiopathic or occur secondary to iatrogenic causes and more rarely due to genetic and autoimmune etiologies [101,102,103]. We are unaware of studies or case reports reporting isolated prolactin deficiency following TBI. One of the patients in our cohort (No.16) had low prolactin levels during the acute phase that remained low in the 5th year of control, while GH and ACTH deficiencies improved (Table 2). As she was at postmenopausal stage no symptoms such as alactogenesis could be observed. It is reasonable to expect IPRLD following TBI. As new data emerge and awareness increases about the effects of prolactin deficiency, further evidence regarding IPRLD following TBI might be obtained.

In conclusion, from the current literature and our re-evaluated data, elevated prolactin levels serve as a marker of TBI severity during the acute phase after TBI. In contrast, in the chronic phase, hypoprolactinemia may function as an indirect indicator of pituitary dysfunction and decreased pituitary volume. There is a lack of evidence to suggest routine prolactin screening following TBI during clinical practice, and further investigations are needed to elucidate the pathophysiologic mechanisms underlying prolactin trend following TBI, its predictive role, and associations with other pituitary hormone dysfunctions.