The results of our study highlight that in patients with permanent AVP deficiency treated with the sublingual desmopressin formulation, age is the main predictor of the drug dose required to achieve good clinical and biochemical control. In particular, our findings clearly confirm that there is a negative correlation between age and dDAVP dose requirement.

The data currently available in the literature on the variability of desmopressin dose requirements in patients with AVP deficiency are limited and are primarily represented for the oral or intranasal formulations [10,11,12], while evidence regarding desmopressin dose in patients using the sublingual formulation is scanty.

The sublingual formulation, similar to the nasal one, has an increased bioavailability compared to tablets, since orally delivered dDAVP is partly degraded by gastric acid [16].

In this regard only one Japanese paper has been recently published which identified sex, age and eGFR as significant factors associated with desmopressin dose requirements [13].

In our study, age was the only factor confirmed to be associated with the daily dose of dDAVP. The presence of age-related dysfunction of the hypothalamic-neurohypophyseal-renal axis, indeed, has been known for more than 60 years now [17, 18]. Several mechanisms have been proposed, including changes in hypothalamic-pituitary regulation of thirst and AVP secretion, as well as hypersensitivity of the osmoregulatory system [18, 19].

Older subjects often present with decreased thirst, likely as a consequence of a higher osmolar set point for thirst sensation [20]. In our cohort, indeed, subjects older than the median age of the population reported a lower fluid intake compared to younger subjects. Moreover, most of the literature regarding water homeostasis has demonstrated that older individuals exhibit a greater increase in AVP secretion per unit change in plasma osmolality than younger subjects [18, 19, 21], supporting the theory of an increase in osmoreceptor sensitivity with aging.

The underlying cause of the age-related increase in sensitivity of the osmoregulatory system has not been fully elucidated yet. It may partially be due to increased pituitary reserves of readily releasable AVP, as histological studies have shown an accumulation of neurosecretory material in the posterior pituitary of elderly individuals [19]. Alternatively, it could represent a compensatory response to the diminished baroreceptor-mediated control of AVP secretion in response to hemodynamic changes, which has been consistently documented in the elderly [18, 19, 22].

Finally, older individuals often exhibit an impaired capacity to excrete free water in the urine, as well as reduced renal function; in this regard, the eGFR decreases by approximately 1 mL/min/1.73m2 per year starting at the age of 40, with a more pronounced acceleration beginning at age 65 [18, 23]. In our population, in fact, patients with even slightly reduced renal function (eGFR < 90 ml/min/1.73m2) were significantly older than other subjects.

While eGFR showed a positive association with the desmopressin dose in the bivariate analysis, this significance was lost in the multivariate analysis. This modest correlation is not an unexpected finding: the pharmacokinetics of dDAVP involve a major proportion of excretion through the kidneys, and it has been demonstrated that the drug's half-life is longer in patients with impaired renal function [24, 25]. In our cohort indeed, patients with even mild reduction in eGFR required significant lower doses than those with normal renal function.

Taking all factors into account, it is plausible that both reduced renal function, with consequent diminished desmopressin clearance, and lower fluid intake observed in older subjects could contribute to the negative correlation with age evident in our study, which was also confirmed in the multivariate analysis. It is worth noting that although renal function specifically lost statistical significance in the multivariate analysis, the result remained borderline significant.

In contrast to the study by Hoshino et al., however, no difference in the daily desmopressin dose between males and females was observed. Several studies in the past have reported a greater susceptibility of women to desmopressin, with a risk up to 5 times higher of developing dilutional hyponatremia compared to males [26,27,28]. Nonetheless, these findings stem from studies conducted on both healthy subjects and individuals with nocturnal enuresis, while data in patients with AVP deficiency are currently lacking. In this regard, it is therefore unknown whether the increased susceptibility of females to desmopressin is also present in a condition of chronic hormonal deficiency.

Moreover, an important difference between our study and that of Hoshino et al. concerns the characteristics of the patients included in the analysis. While in their work, patients were recruited immediately post-pituitary neurosurgery and were evaluated for up to 12 months [13], in our study only patients with a disease duration of more than 12 months were considered.

The diagnosis of permanent postoperative AVP deficiency is generally made based on the persistence of the hydro-electrolytic disorder and the need to continue dDAVP therapy for more than 12 months after surgery, as the recovery of antidiuretic function after one year is considered exceptional [3]. In contrast, in the immediate post-surgery period, various disorders of AVP secretion can be present, such as non-pathological polyuria and transient AVP deficiency [3, 4, 29, 30].

Based on our results, it cannot be excluded that the previously reported greater susceptibility to dDAVP in women, if also present in patients with AVP deficiency, may reduce over time, resulting in a similar drug dose between the two sexes.

In our cohort, moreover, males had undergone neurosurgery more frequently than females and presented a higher incidence of anterior pituitary deficiency. However, none of these variables were associated with the dose of dDAVP required to achieve good clinical and biochemical control.

In previous studies, a positive association between BMI and desmopressin dose has also been observed, more evident for the tablets [10, 12], although a weak association has also been reported in patients using the orally lyophilized formulation [13]. In our study, a modest association was observed between drug dose and weight while a clear correlation with BMI was not evident; anyway, obese patients required higher doses compared to other subjects.

In this regard, the greatest evidence is for the oral drug, and this could depend on differences in the pharmacokinetics of parenteral formulations compared to tablets. Indeed, the latter must be absorbed in the digestive system before entering the bloodstream, while nasal and sublingual formulations are absorbed directly into the bloodstream through the mucosa [16].

Moreover, Pedersen et al. also observed a higher dose requirement in patients with AVP deficiency after removal of craniopharyngioma compared to patients undergoing neurosurgery for other pituitary tumors [10], while no difference was observed in our population.

Patients with craniopharyngioma are at very high risk of postoperative AVP deficiency because surgery is usually more invasive, often involving the pituitary stalk, and can sometimes result in complete loss of thirst sensation (adipsic AVP deficiency) [3, 31, 32].

In the aforementioned study, information regarding thirst sensation was not provided, and approximately 30% of patients had experienced at least one episode of hyponatremia in the previous 12 months [10]. It is therefore possible that the evidence of a higher dose of dDAVP in patients with craniopharyngioma was simultaneously affected by a higher prevalence of hyponatremia in this group of patients. It is important to note that previous studies have only considered the current dose of dDAVP at the last follow-up, which however does not always reflect the patient's actual needs.

In this regard, the definition of good control for AVP deficiency during desmopressin therapy is not straightforward, but in an attempt to be as homogeneous as possible, we included in our study only patients with a disease duration of at least 12 months and in good clinical and biochemical control without the need for dosage adjustments in the last year of therapy.

Finally, in contrast with the evidence from the studies by Pedersen et al. and by Almutlaq et al. [10, 11], we did not observe any difference in the dDAVP dose among patients with congenital or acquired etiology. However, only a very small proportion of patients were affected by the congenital form in our cohort. Of note, those patients were significantly more likely to be treated with nasal desmopressin and, as expected, presented a longer duration of disease.

Regarding nasal desmopressin formulation, several studies in the past have reported a higher risk of hyponatremia compared to the oral formulation [33, 34]. At the same time, this drug exhibits non-constant absorption, as inflammatory processes at the level of the nasal mucosa can interfere significantly [35].

Then, it should not be surprising that the majority of patients in our study used the sublingual formulation, and mostly patients with a very long disease duration were on chronic therapy with nasal desmopressin, given that sublingual desmopressin has only been available since 2005 [16, 33].

Our study presents some strengths and limitations. One major strength is the rigorous inclusion of only patients with permanent AVP deficiency with intact thirst sensation and in good biochemical and clinical control under stable dose of substitution therapy for at least 12 months. We have thus reasonably avoided the inclusion of transient postoperative forms and have used clinically and biochemically recognized parameters as fundamental elements to define the appropriate drug dose for each patient.

The main limitations of the study include its retrospective design and the fact that a confirmation test was available only for a smaller proportion of patients, considering that the etiology of the disorder was mostly post-neurosurgical. Anyway as suggested by several experts, patients were correctly instructed to periodically omit or delay desmopressin intake until the reoccurrence of polyuria, which occurred in all cases [7].