Intranasally Administered Insulin as Neuromodulating Factor and Medication in Treatment of Neuropsychiatric Disorders—Current Findings from Clinical Trials

Abstract

As a metabolism-controlling peptide, insulin affects activity of almost all tissues in human organisms, including the ones located in the central nervous system. By modifying glucose uptake and processing, as well as inducing anabolic effects, insulin alters functions of various nerve centers. Data from numerous clinical trials prove that such actions can have positive influence on cognitive processes or might be utilized as measures to control appetite, mood, and blood flow, or to prevent unfavorable mental states associated with diminished ability to maintain homeostasis. The intranasal route of administration provides an efficient and targeted delivery method, allowing insulin to be applied directly to different brain regions via the nasal mucosa. Such an approach can also reduce the risk of potential adverse effects associated with this medication, including drops in plasma glucose levels. This review gathers clinical studies’ findings on intranasal insulin’s neuromodulatory properties and its efficacy as additional treatment measure in several neuropsychiatric disease entities.

1. Introduction

Insulin is a peptide hormone secreted by beta cells located in Langerhans’ islets of the pancreas in response to rising glucose levels in plasma. It serves as one of the most important factors responsible for regulation of glucose metabolism [1]. Biological outcomes of insulin activity are exerted through its receptor, which takes form of heterotetrameric tyrosine kinase (two extracellular α-subunits with a ligand binding site and two transmembrane β-subunits anchoring the protein in a cellular membrane) and occurs in two types—insulin receptor A or B (IR-A and IR-B) [1,2]. The main effector organs of insulin signaling include liver, muscle and adipose tissue [1,2]. Metabolic outcomes of this hormone’s action primarily encompass glucose uptake from blood and its storage in the form of glycogen, as well as stimulation of lipogenesis [2,3]. Lack or insufficient secretion of insulin disturbs carbohydrate processing in the organism and leads to a state known as hyperglycemia—a condition of chronically upregulated glucose level, which is the characteristic feature of diabetes mellitus (DM). Diabetic patients exhibit much higher risk of unfavorable health outcomes regarding functionality of the cardiovascular, renal and peripheral nervous systems [2,3,4]. Since insulin’s discovery in 1921, its subcutaneous and intravenous forms and analogs have established an undeniably crucial position in treatment of DM and improvement of both lifespan and quality of life in individuals with impaired glucose metabolism [4,5]. Nowadays more attention is being paid to insulin’s other functions and applications as well as states of tissues’ disturbed reactivity to this hormone. Of special interest for this review are the reports describing insulin’s action exerted towards structures of the central nervous system (CNS) and subsequent alterations for human health. Apart from being secreted from the pancreas and reaching the brain through the blood–brain barrier (BBB), small amounts of insulin are also produced by neurons and neural progenitor cells and expressed in different regions of the CNS, including the cerebellum, cerebral cortex, anterior olfactory bulb, cerebellar Purkinje neurons and the hippocampus [6,7,8]. Such brain-derived insulin acts more as an auto- or paracrine regulator molecule. Similarly to other organs, in the CNS the insulin signaling works mainly on the basis of IR-A, expressed by both astrocytes and neurons, and IR-B, expressed only by astrocytes. The first one is predominant in adult life and is characterized by higher affinity for both insulin and insulin-like growth factor (IGF) [1,2]. This feature results from alternative splicing of the insulin receptor gene’s exon 11, where 12 amino acids located at the carboxyl terminus of the α-subunit are spliced off. Moreover, IR-A does not display the phenomenon of negative cooperativity, suggesting that it depends on other forms of functional regulation. Several studies point to insulin-dependent generation of mitochondrial H₂O₂ in neurons as the key mechanism adjusting IR-A sensitivity to intensity of synaptic activity [9,10]. The mutual interactions between insulin signaling and mitochondrial function may also be the link connecting impaired insulin action in the brain and different neurological illnesses. Numerous reports highlight the potential association of disturbed synaptic function, lower sensitivity to insulin and deteriorated mitochondrial functionality in pathologies of the CNS, especially neurodegenerative disorders. Several in vitro and in vivo evaluations have proved that this hormone exerts important beneficial effects for proper functionality of nervous and glial tissues and as a result contributes to maintenance of neuropsychiatric health [6,7,8]. Insulin promotes neurodifferentiation, formation of the neurofilaments and counteracts the inflammatory state. Such modifications find their reflection in improved cognitive performance, as insulin has been shown to advantageously impact memory and mental well-being among subjects suffering from Alzheimer’s disease (AD), mild cognitive impairment (MCI), postoperative delirium (POD), cerebral ischemia or inborn mental retardation [8,11,12]. To increase insulin’s efficacy in reaching CNS tissues, several studies adopt the intranasal route of administration to participants. This approach allows the mentioned medication to reach cerebrospinal fluid (CSF) in just 10 min by bypassing the BBB via the nasal mucosa and olfactory bulb, and achieve peak concentration in 30 min. This route involves two possible pathways of passage: an intraneuronal and an extraneuronal. The first one includes direct internalization of insulin by neurons of the olfactory and trigeminal nerves. Further transport occurs in their axons, where some of the absorbed particles may be subjected to proteolysis. The second pathway is more efficient and involves passing through intercellular clefts of the olfactory epithelium to subarachnoid space [13]. This nerve-dependent transport results in rostral and caudal brain areas displaying the highest concentration of insulin or other peptides administered via intranasal route, as these regions remain in the immediate vicinity of the sites where the mentioned nerves enter the CNS [14,15]. Intranasal administration of insulin also limits possible side effects associated with reaction of other peripheral tissues, like modulation of plasma glucose and insulin levels [11,12]. However, the devices utilized for delivering insulin to the nasal mucosa differ in terms of several properties like generated droplet size or spray angle values, which may affect the efficacy of such treatment [16]. As the amounts of data regarding insulin’s signaling role in the maintenance of the CNS’s homeostasis and the intranasal route of its administration as an alternative therapeutic approach rise every year, the main focus of this review was to gather currently available findings from clinical trials referring to intranasal insulin’s effects on CNS functionality and potential treatment of complex diseases that are the focus of interest for neurologists and psychiatrists.

2. Intranasally Administered Insulin’s Effects on Memory, Cognitive Functions and Neurodegeneration

2.1. Memory and Cognitive Functions in Patients Without Confirmed Neuropsychiatric Disorders

Analyses performed on healthy and diabetic patients show that administration of insulin through the nasal cavity can cause intensification of resting-state connectivity between hippocampal regions and brain areas linked to integrative higher cognitive functions [17]. As a result, a long-term therapy with this form of medication may induce amelioration of verbal memory and executive functions [18]. In studies carried out by Benedict et al., the 8-week-lasting intranasal administration of insulin to healthy subjects proved to raise the ability of recalling words in a declarative memory assessing word-list learning task when compared with placebo. However, the effect was notable only in evaluations taking part at least one week after introducing the word list to the participants, without influence on immediate recall of words [19,20]. Immediate effect towards memory-based abilities was observed in another study by Reger et al., where insulin was capable of improving story recalling and selective word reminding. Nonetheless, such influence was specific to subjects suspected of MCI and, moreover, only those who did not bear apolipoprotein E-ε4 (APOE-ε4) alleles—the main risk factor of AD development [21]. This and the fact that the mean age difference between examined groups greatly differed in referenced studies might explain the observed distinctions in insulin‘s efficacy. In a trial evaluating the combined effectiveness of single-dose, intranasal insulin and preceding 3-day-long application of transdermal estradiol, the insulin lacked additional beneficial effects regarding mental condition. Participants pretreated with estradiol and later dosed with nasal spray containing insulin or placebo achieved similar results in various tests assessing cognitive functions, except for verbal recognition ability measured in Stroop test, where administration of insulin significantly improved subjects’ performance [22]. Similarly to the mentioned results, single application of insulin through nasal cavity was not capable of improving declarative or spatial memory performance in directly following assessments, based on 3D-simulated mazes with olfactory and visual cues [23]. An explanation for this lack of positive outcomes might come from the observation that single doses of insulin affect memory rather by impairing the acquisition of new data that interfere with and disrupt the already-established memory traces, instead of improving learning processes [24].

2.2. Memory and Cognitive Functions in Patients Affected by Alzheimer’s Disease or Mild Cognitive Impairment

The very idea of utilizing intranasally administered insulin in patients suffering from AD or MCI-associated dementia is dictated by the observations of impaired insulin action in the CNS and its reduced CSF to plasma ratio noted among such subjects—the characteristics of the state described as brain insulin resistance. As the examinations of this medication’s effectiveness show, even a 3-week-long therapy can cause significant upswing of selective attention and general functional status of patients who developed this neurodegenerative disorder or its prodromal state—MCI. Furthermore, application of intranasal insulin leads to upregulation of the Aβ40/42 ratio in peripheral blood. Higher values of this index are considered markers of milder pathology in AD [25,26]. Another abnormality observed among patients with AD or MCI is the progression of metabolic inhibition detectable in specific brain parts, including parietal, temporal, frontal and occipital cortices as well as precuneus or cuneus regions. It is a sign of gradual neurodegeneration, but appropriate intranasal application of insulin may mitigate the following changes [27]. The severity of AD or MCI-caused dementia also correlates with the presence of the white matter hyperintensities (WMH) detected by magnetic resonance imaging (MRI) in the CNS. Such lesions are postulated to reflect progression of gliosis, demyelination, axonal loss and arteriosclerosis. Intranasal insulin administration does not prevent formation of this sort of pathologies, yet it significantly counteracts their further growth and progression [28]. Insulin also positively modulates the immunological profile of cytokines in CSF, as it upregulates interferon-γ (IFN-γ), eotaxin and interleukin 2 (IL-2) levels while simultaneously downregulating IL-6 concentration. Such alterations are postulated to alleviate AD progression [29]. The clinical response to the described medication might depend on the patients’ sex and genetic predisposition. In a study by Claxton et al., the outcomes of 4-month-long administration of intranasal insulin preparation were associated with participants’ gender. Females exhibited less severe functional deterioration when compared with males treated with the same dose of insulin (10 or 20 IU twice a day). On the other hand, in terms of story-recalling ability, the higher utilized dose (20 IU) was only successful among males and the effect was characteristic for APOE-ε4 negative participants [30]. The undesirable influence of APOE-ε4 alleles’ presence towards the response to intranasal insulin was also indicated by Rosenbloom et al., who noted that it unfavorably affects potential improvement of visuospatial skills. The number of patients enrolled in the mentioned study was, however, relatively small when compared with other trials concerning this problem, which limits the significance of the observed results [31]. Interestingly, the interplay between insulin and APOE-ε4-positive genotype can depend on the type of applied medication. In all the previously referenced studies, the administered drug was a regular human insulin or its short-acting analog. Trial involving detemir, which is a long-acting analog of insulin, shows that doses of 20 IU, applied twice a day for three weeks, may improve verbal memory among APOE-ε4-bearing subjects. Moreover, they also ameliorate performance in verbal working memory- and visuospatial working memory-assessing tests, regardless of patients‘ genotype [32]. Unfortunately, further study revealed that prolonging the therapy period to 4 months of detemir appliance results in decrease in its efficacy, whereas for regular human insulin the beneficial action towards memory functions is sustained [33]. Additionally, the regular insulin proves to be more efficient in counteracting loss of important CNS regions’ volumes among AD-suffering patients [33]. In search of potential biomarkers that would help to distinguish AD- or MCI-affected patients with a high versus low likelihood of benefitting from intranasal insulin therapy, Mustapic et al. performed an assessment of different compounds carried by plasma-derived neuronal extracellular vesicles. The presence of phosphorylated forms of insulin receptor substrate (IRS), in mentioned vesicles, correlated with positive outcomes of intranasal insulin application and improvement of cognitive functions in APOE-ε4 negative participants. However, the exact associations between specific levels of referenced markers and further results of treatment with intranasal insulin still need broader research and evaluation [34]. The vast majority of clinical trials referenced above report intranasal insulin’s efficacy in alleviating cognitive dysfunction among subjects suffering from AD or MCI, even if it is partial or relatively weak. Yet, there are still assessments that completely lack statistically significant, positive outcomes for mental condition among insulin-treated participants. In a trial assessing the combined impact of high-dose vitamin D and intranasal insulin, no beneficial modulation of participants’ cognitive functions was noted. However, the lack of favorable outcomes might be the result of a relatively short administration time, as the administration of insulin continued for only 48 h. The authors also point out that this discrepancy may be dictated by high severity of AD in the examined group where most interventions would prove insufficient, as a majority of the subjects were previously receiving additional dementia-alleviating drugs (rivastigmine, donepezil, galantamine, memantine) [35]. Much longer time of exposure was adopted in a recent study by Craft et al., where patients were dosed with regular human insulin for 12 months. The neuropsychological evaluations did not reveal any significant beneficial effects of such long-term therapy. Moreover, the participant cohort had an estimated 240 subjects, which makes it the largest clinical trial assessing intranasal insulin’s efficacy in treatment of AD or MCI-associated memory impairment. These results suggest that both too-short and too-long treatment with intranasal insulin appear to be inefficient in alleviating the severity of cognitive deficits. Possibly, a time range of few days is insufficient to allow insulin’s effects to fully develop, while in longer perspective CNS tissues start exhibiting some sort of resistance. Craft et al. also noted that the device used for intranasal insulin administration in their trial had not been previously validated for this specific application, which may have resulted in suboptimal or ineffective drug delivery [36].

2.3. Memory and Cognitive Functions in Patients Affected by Parkinson’s Disease or Multiple System Atrophy

The only available data regarding intranasal insulin’s effects in patients with Parkinson’s disease (PD) or multiple system atrophy (MSA) come from a small study conducted on fifteen participants. All subjects were diagnosed and treated for PD, with one participant also receiving treatment for possible MSA. Patients were randomized to receive either 40 IU of intranasal human insulin or placebo once daily for four weeks, and the baseline characteristics including disease severity and cognitive scores were similar between groups. After treatment, the group applied with insulin showed improved verbal fluency scores measured by FAS test, as well as better motor scores according to Unified Parkinson Disease Rating Scale III (UPDRS-III) and Hoehn and Yahr disease severity ratings. The sole participant with MSA was assigned to the insulin-treated group and remained stable without disease progression. No hypoglycemic episodes or serious adverse events were noted during the course of the study [37].

2.4. Memory and Cognitive Functions in Patients Affected by Schizophrenia and Affective Disorders

In a trial assessing the effect of a single-dose intranasal insulin on cognitive functions of patients diagnosed with schizophrenia, no significant beneficial improvements were noted [38]. In accordance with these results, a long-term evaluation (8 weeks) of a similar cohort, applied with 4-times-higher doses of insulin (160 IU/day) presented a lack of any advantageous outcomes regarding memory potential or cognition [39]. Among subjects suffering from bipolar disorder treated with the same doses of insulin and for an identical amount of time, the only detectable improvement was better performance in examination that estimated the participants’ ability to switch attention [40]. Absence of favorable consequences towards cognitive functioning was also reported in a similar study concerning patients with treatment-resistant major depressive disorder (TRMDD) [41]. The results from clinical trials concerning intranasal insulin’s effects on cognitive function and memory-related disorders are summarized in

Table 1. Summarized results of clinical trials concerning intranasal insulin’s effects on cognitive function and memory-related disorders.

Participants	Males	Females	Age (Years)	Insulin	Applied Dose	Results	Ref.
14 healthy participants and 14 participants with type 2 of DM	11	17	61 (mean)	regular human insulin	40 IU (1 time/day for 2 days)	increased resting-state connectivity between the medial frontal cortex, right inferior parietal cortex, posterior cingulate gyrus, anterior cingulate cortex and hippocampal regions among subjects with diabetes increased resting-state connectivity in the medial frontal cortex, posterior cingulate gyrus and anterior cingulate cortex among healthy subjects	[17]
117 healthy participants and 106 participants with type 2 of DM	114	109	65 (mean)	regular human insulin	40 IU (1 time/day for 24 weeks)	improved executive function composite scores (paired associates learning and spatial working memory) on- and post-treatment and verbal memory composite scores (verbal immediate free recall and immediate and delayed verbal recognition memory) post-treatment in healthy subjects	[18]
38 healthy participants	24	14	18–34	regular human insulin	40 IU (4 times/day for 8 weeks)	improved delayed recall of words in declarative memory-assessing word list learning task	[19]
36 healthy participants	36	0	18–35	regular human insulin and aspart	40 IU (4 times/day for 8 weeks)	improved delayed recall of words in declarative memory-assessing word list learning task insulin aspart displayed stronger beneficial effect than regular human insulin	[20]
35 healthy participants and 26 participants with probable AD or MCI	28	33	73–77 (mean)	regular human insulin	20 IU or 40 IU (single dose)	both 20 IU and 40 IU improved story recalling ability among MCI-suspected participants without APOE-ε4 alleles 40 IU improved selective word reminding in Buschke Selective Reminding Test among MCI-suspected participants without APOE-ε4 alleles	[21]
32 healthy participants	32	0	18–35	regular human insulin	160 IU (single dose preceded by 3-day long administration of transdermal estradiol or placebo)	no beneficial effects towards divergent thinking, convergent thinking, immediate verbal recall and visuospatial memory recall insulin improved verbal recognition in groups pretreated with estradiol	[22]
11 healthy participants	11	0	25 (mean)	regular human insulin	40 IU (single dose)	no beneficial effects towards declarative and spatial memory performance lack of altered hippocampal activity	[23]
32 healthy participants	16	16	18–30	regular human insulin	160 IU (single dose)	impaired acquisition of new information that interferes with already fixed memory traces	[24]
24 participants with AD or MCI with amnestic features	gender structure not specified		77–79 (mean)	regular human insulin	20 IU (2 times/day for 3 weeks)	improved story recalling ability improved selective attention and performance speed in Stroop test improved functional status among participants with more severe baseline dementia rise in fasting Aβ40/42 ratio	[25]
104 participants with AD or MCI with amnestic features	59	45	70–75 (mean)	regular human insulin	10 IU or 20 IU (2 times/day for 4 months)	10 IU improved story recalling ability 10 IU and 20 IU alleviated decline in cognition measured by DSRS and ADAS-cog 10 IU and 20 IU alleviated functional deterioration measured by ADCS-ADL scale among participants with AD 10 IU alleviated progression of hypometabolism in bilateral frontal, right temporal, bilateral occipital and right precuneus and cuneus regions 20 IU alleviated progression of hypometabolism in bilateral frontal, bilateral occipital, left parietal cortex and right precuneus and cuneus regions	[27]
49 participants with AD or MCI with amnestic features	32	17	71 (mean)	regular human insulin	20 IU (2 times/day for 12 months)	less severe increase in white matter hyperintensities volume in frontal lobe and deep white matter regions	[28]
49 participants with AD or MCI with amnestic features	32	17	70–72 (mean)	regular human insulin	20 IU (2 times/day for 12 months)	upregulation of IFN-γ, eotaxin and IL-2 levels in CSF downregulation of IL-6 levels in CSF	[29]
104 participants with AD or MCI with amnestic features	59	45	67–74 (mean)	regular human insulin	10 IU or 20 IU (2 times/day for 4 months)	10 IU improved story recalling ability among both sexes 20 IU improved story recalling ability among male participants (effect exceptionally notable in APOE-ε4 negative subgroup) 10 IU and 20 IU alleviated functional deterioration measured by ADCS-ADL scale among female participants	[30]
12 participants with AD	9	3	65–85	glulisine	20 IU (single dose)	improved performance in RBANS line orientation task in APOE-ε4 negative subgroup improved performance in trails B test, assessing the ability to switch subject’s attention, regardless of APOE-ε4 allel presence	[31]
60 participants with AD or MCI with amnestic features	37	23	69–75	detemir	10 IU or 20 IU (2 times/day for 3 weeks)	20 IU improved verbal memory (summarized performance in immediate story recall, delayed story recall, immediate word list recall and delayed word list recall) in APOE-ε4 positive subgroup 20 IU improved verbal working memory (Dot Counting N-back) and visuospatial working memory (Benton Visual Retention Test) regardless of APOE-ε4 presence	[32]
36 participants with AD or MCI with amnestic features	17	19	67–71 (mean)	regular human insulin or detemir	20 IU (2 times/day for 4 months)	regular human insulin improved delayed verbal memory composite score (summarized performance in delayed story recall and delayed Selective Reminding Test) regular human insulin counteracted loss of left superior parietal cortex, right middle cingulum, left cuneus and right parahippocampal gyrus volume. detemir counteracted loss of left anterior and middle cingulum volume. regular human insulin lowered tau-P181 to Aβ-42 ratio in CSF	[33]
91 participants with AD or MCI with amnestic features	35	56	70–76 (mean)	regular human insulin	20 IU or 40 IU (2 times/day for 4 months)	positive correlation between plasma-derived, neuronal, extracellular vesicle contents of pS312-insulin receptor substrate-1 and pY-insulin receptor substrate-1 and cognitive performance measured by ADAS-cog scale among APOE-ε4 negative participants treated with 20 IU of intranasal insulin	[34]
32 participants with AD	15	17	69–80	regular human insulin	60 IU (4 times/day for 2 days) insulin application was preceded by 8-week-long administration of vitamin D (6000 U/day) or placebo	no significant improvement of cognitive functions	[35]
240 participants with AD or MCI with amnestic features	123	117	55–85	regular human insulin	20 IU (2 times/day for 12 months)	no significant improvement of cognitive functions	[36]
15 participants with PD or MSA	9	5	63 (mean)	regular human insulin	40 IU (1 time/day for 4 weeks)	increased verbal fluency measured by total number of words given in FAS test decreased scores in the Hoehn and Yahr scale reflecting severity of parkinsonism decreased motor impairment measured by UPDRS-III scale	[37]
30 participants with schizophrenia	10	20	18–65	regular human insulin	40 IU (single dose)	no significant effect on cognition	[38]
45 participants with schizophrenia	36	9	18–65	regular human insulin	40 IU (4 times/day for 8 weeks)	no significant effect on cognition	[39]
62 participants with bipolar disorder	33	29	40 (mean)	regular human insulin	40 IU (4 times/day for 8 weeks)	improved performance in trails B test, assessing the ability to switch subject’s attention	[40]
35 participants with TRMDD	13	22	18–65	regular human insulin	40 IU (4 times/day for 4 weeks)	no significant effect on cognition	[41]

Abbreviations: AD, Alzheimer’s disease; ADAS-cog, Alzheimer’s disease assessment scale–cognitive subscale; ADCS-ADL, Alzheimer’s disease cooperative study—activities of daily living; APOE-ε4, apolipoprotein E- ε4; Aβ40/42, amyloid β 40/42; CSF, cerebrospinal fluid; DM, diabetes mellitus; DSRS, dementia severity rating scale; IFN-γ, interferon γ; IL-2, interleukin 2; IL-6, interleukin 6; IU, international units; MCI, mild cognitive impairment; MSA, multiple system atrophy; PD, Parkinson’s disease; RBANS, Repeatable Battery for the Assessment of Neuropsychological Status; TRMDD, treatment resistant major depressive disorder; UPDRS-III, Unified Parkinson Disease Rating Scale.

.

3. Intranasally Administered Insulin’s Effects on the Sense of Smell

In a trial assessing the effect of single-dose intranasal insulin on olfactory sensitivity among normosmic subjects, it was shown that such treatment diminishes ability to detect the n-butanol odorant [42]. A similar study provided opposite results: an assessment performed with Sniffin’ Sticks test battery revealed a dose-dependent capacity of insulin to elevate odor sensitivity [43]. The exact reasons behind such vastly different outcomes are not clear, although since in both cases the examined cohorts were relatively small and the study that reported insulin’s beneficial action recruited only male subjects, the cause might lay in sociodemographic characteristics disparity. The evaluations focusing on people with an already-impaired sense of smell are definitely more unanimous. According to Schopf et al., single-dose insulin application into the nasal cavity led to improvement of smell sensitivity and the ability to discriminate individual odors in patients suffering from post-infectious olfactory loss. The capability to identify particular scents was, however, diminished [44]. Another study, enrolling participants with mild to severe hyposmia, also demonstrated intranasal insulin effectiveness in treatment of impaired smell, with positive outcomes being still detectable even 4 months after termination of the 4-week-long intervention [45]. A recent examination of CNS structures’ response to visual and olfactory stimuli provided a possible explanation of how exactly insulin affects reception of different scents in human subjects. The authors point out that this medication exerts stimulatory influence towards activity of medial superior focal gyrus, inferior focal gyrus and elements of the orbitofrontal cortex—structures associated with a modulatory role in multimodal processing of various stimuli [46]. Moreover, considering that the olfactory bulb presents significant content of insulin receptors, insulin may also directly stimulate its regeneration, thanks to its vasodilatory effects as well as general neuroprotective potential [47]. The anti-inflammatory capability of insulin might also be of significance for recovery of the sense of smell, as it was shown to counteract activation of the NLR family pyrin domain containing 3 (NLRP3) inflammasome and diminish production of nitric oxide (NO), reactive oxygen species (ROS) and tumor necrosis factor α (TNF-α) by microglia, providing a stable and regeneration-promoting cellular environment [48,49]. The results from clinical trials concerning intranasal insulin’s effects on the smell sense functioning are summarized in

Table 2. Summarized results of clinical trials concerning intranasal insulin’s effects on the functioning of the sense of smell.

Participants	Males	Females	Age (Years)	Insulin	Applied Dose	Results	Ref.
17 healthy participants	10	7	24.5 (mean)	regular human insulin	40 IU (single dose)	decreased sensitivity to detect the n-butanol odorant	[42]
36 healthy participants	36	0	25.5 (mean)	regular human insulin	40, 100 or 160 IU (single dose)	improved odor sensitivity measured by Sniffin’ Sticks test battery doses of 100 IU and 160 IU displayed stronger beneficial effect	[43]
10 participants with post-infectious olfactory loss	7	3	22–56	regular human insulin	40 IU (single dose)	improved odor sensitivity and discrimination ability measured by Sniffin’ Sticks test battery decreased odor identification ability measured by Sniffin’ Sticks test battery	[44]
38 participants with hyposmia	20	18	18–70	regular human insulin	40 IU (2 times/week for 4 weeks)	decreased severity of hyposmia measured by CCCRC test	[45]
26 healthy participants	26	0	19–31	regular human insulin	40 IU (single dose)	increased activation of mediodorsal thalamus, anterior cingulate cortex, postcentral gyrus, posterior cingulate gyrus, precentral gyrus, supramarginal gyrus, insula and caudate nucleus in response to visual and olfactory stimuli	[46]

Abbreviations: CCCRC, Connecticut Chemosensory Clinical Research Center; IU, international units.

.

4. Intranasally Administered Insulin’s Effects in Postoperative Delirium

Postoperative delirium (POD) is an acute neuropsychiatric state, typically occurring within 7 days after surgery. It develops more often among elderly patients as a result of impaired mechanisms of homeostasis maintenance in critical conditions. Some other risk factors of this syndrome include malnutrition, higher general severity of the patient’s illness and longer surgical procedure time. The depth of sedation might also be of significance, especially in patients without comorbidities [50]. POD incidence undermines postoperative convalescence and significantly increases the costs associated with postoperative care. Intranasally administered insulin, at dose of 20 IU, has been shown to exert neuroprotective capabilities and decrease frequency of POD among patients undergoing radical excision of gastrointestinal tumors. One of the possible explanations for such effect, proposed by the authors of the referenced trial, encompasses downregulation of serum IL-1β, IL-6 and tumor necrosis factor α (TNF-α)—strong proinflammatory cytokines that can surpass BBB and contribute to CNS disfunction after surgery [51]. A different study concerning the safety of intranasal insulin in cardiovascular surgeries did not note significant alterations in mentioned cytokines levels between placebo and insulin-treated groups, although this discrepancy might come from the lack of repetitive insulin dosage and much lower sample size. However, this study demonstrated that hypoglycemia, which is the main adverse state associated with insulin administration, occurred only when doses of 240 IU were applied, which displays that at lower dose ranges this medication can be considered as generally safe [52]. The available data concerning insulin dosing recommendations for prevention of this neuropsychiatric state are very limited, although most studies note ranges between 20 and 40 IU as most optimal [53]. Another potential mechanism to prevent development of POD by inserting insulin into the nasal cavity can be counteracting the hypometabolism of brain tissues that favor this neuropsychiatric condition. Yang et al. noted that patients undergoing joint replacement receiving intranasal insulin exhibited higher glucose levels in CSF on the day of surgery, implying an increased available energy source for the CNS. Moreover, the CSF obtained from insulin-treated participants also showed upregulated uncarboxylated and total osteocalcin (ucOC and tOC) as well as brain-derived neurotrophic factor (BDNF) contents—the neurotrophins crucial for maintaining proper cognitive functionality. All those alterations were accompanied by decreased POD incidence and severity measured by delirium rating scale (DRS)-98 [54]. Apart from reducing the risk of POD, intranasal insulin can also counteract perioperative cognitive deficits and ameliorate the sleep quality. Patients qualified for cardiopulmonary bypass formation surgery experienced less severe deterioration of mental functions measured by mini-mental state examination (MMSE) and displayed better outcomes regarding total sleep time and sleep efficiency according to Richards–Campbell Sleep Questionnaire (RSCQ) [55]. A congruous trial assessing subjects undergoing valve surgery provides even more data referring to participants’ sleep and shows that perioperative administration of intranasal insulin results in lower Pittsburgh Sleep Quality Index (PSOI) scores, light sleep ratio and number of awakenings. Moreover, evaluations performed by sleep monitoring watch suggest prolongation of the deep and rapid eye movement (REM) sleep phases, leading to better total sleep quality. As the sleep disorders are also referenced as another factor promoting the POD incidence, it is not surprising that the mentioned positive influence on sleep properties came with decline of delirium occurrence in referenced studies [55,56]. The results from clinical trials concerning intranasal insulin’s effects on prevention and treatment of POD are summarized in

Table 3. Summarized results of clinical trials concerning intranasal insulin’s effects on prevention and treatment of POD.

Participants	Males	Females	Age (Years)	Insulin Type	Applied Dose	Results	Ref.
80 participants undergoing laparoscopic radical excision of gastrointestinal tumor	46	24	65–70	regular human insulin	20 IU (2 times/day for 2 days preceding the surgery and 1 final dose applied 10 min before anesthesia induction)	decreased incidence of POD within 5 days after surgery decreased serum levels of TNF-α, IL-6 and IL-1β	[51]
27 participants undergoing cardiac or major vascular surgery	20	7	62–75 (mean)	regular human insulin	40, 80, 160 or 240 IU (single dose during anesthesia)	2 cases of hypoglycemia among subjects receiving 240 IU of intranasal insulin lack of statistically significant alterations in serum TNF-α, IL-6 and IL-1β levels	[52]
130 participants undergoing unilateral hip arthroplasty or closed reduction and intramedullary nailing	37	93	72–87	regular human insulin	20 or 40 IU (3 doses in total, first—the day before surgery, second—50 min before anesthesia, third—in the evening after surgery)	decreased incidence of POD within 3 days after surgery (stronger but not statistically significant tendency in 20 IU-treated group) increased glucose levels in CSF in 40 IU-treated group) lack of statistically significant alterations in lactate levels in CSF and plasma glucose	[53]
195 participants undergoing elective joint replacement surgery	50	145	73 (mean)	detemir	40 IU (2 times/day for 3 days preceding the surgery and 5 days after surgery)	decreased incidence of POD within 5 days after surgery lower severity of POD occurring in the insulin-treated group (measured by DRS-98 scale) increased ucOC, tOC, BDNF and glucose levels in CSF increased tOC levels in plasma	[54]
76 participants undergoing elective CBP surgery	29	42	45–65	regular human insulin	20 IU (1 time/day for 3 days, first dose—1 h before surgery, second and third dose—postoperative days 1 and 2)	decreased incidence of POD within 4 days after surgery decreased deterioration of perioperative MMSE test results improved sleep efficiency and total sleep time on postoperative day 1	[55]
76 participants undergoing valve surgery with cardiopulmonary bypass for rheumatic heart disease	26	50	16–65	regular human insulin	20 IU (2 times/day for 2 days preceding the surgery and 1 final dose applied 10 min before anesthesia induction)	decreased incidence of POD and lower scores in CAM-CR within 3 days after surgery decreased PSOI scores, light sleep ratio and number of awakenings increased higher deep sleep ratio, REM sleep ratio, deep sleep continuity score, and total sleep quality score decreased serum cortisol level before surgery and anesthesia	[56]

Abbreviations: BDNF, brain-derived neurotrophic factor; CAM-CR, Core Confusion Assessment Method; CBP, cardiopulmonary bypass; CSF, cerebrospinal fluid; DRS-98, delirium rating scale 98; IL-1β, interleukin 1β; IL-6, interleukin 6; IU, international units; MMSE, mini-mental state examination; POD, postoperative delirium; PSOI, Pittsburgh Sleep Quality Index; REM, rapid eye movement; TNF-α, tumor necrosis factor α; tOC, total osteocalcin; ucOC, uncarboxylated osteocalcin.

.

5. Intranasally Administered Insulin’s Effects on Mood

Clinical observations conducted for 8 weeks on 38 healthy subjects, without any already-diagnosed affective disorders and undergoing intranasal insulin, showed that such treatment had rapidly improved subjective personal welfare and self-confidence and resulted in lower levels of perceived anger when compared to the placebo group [19]. In another study concerning population of obese women, similar outcomes could be noted as the participants displayed mood enhancement, assessed utilizing the positive and negative affect schedule (PANAS) questionnaire. The beneficial effects of insulin administration were detectable after receiving just one dose (160 IU) of this treatment. This might be attributable to potential higher sensitivity of obese humans to intranasal insulin’s influence exerted on food reward processes and insulin-dependent activation of the insula—a key brain region responsible for integration of bodily sensations with cognitive and emotional information [57]. The data regarding efficacy and results of intranasally administered insulin in mood modulation among patients with affective disorders instead of mentally healthy subjects are limited to one study examining a TRMDD-affected population. Despite utilizing a similar total daily dose of insulin (160 IU/day) as evaluations conducted on healthy patients, the examination of this trials’ participants did not reveal any beneficial outcomes in terms of experienced emotions and general mood alteration. The investigators mark that the adopted intervention time of 4 weeks might have been insufficient to exert positive influence on subjects’ mental condition. Most of the participants in the referenced study experienced their first major depressive episode at the age of about 18 years old. As the mean age of the examined sample was much higher, a majority of the subjects were suffering from repetitive episodes of depressed mood through the years. Possibly, such a course of illness significantly affected their cognitive appraisal and permanently altered synaptic activity of several brain regions, impairing response to different interventions. Therefore, pleiotropic mechanisms of insulin in the CNS, responsible for mood improvement observed in both healthy and obese individuals, were most likely significantly limited in the depression-affected population [41]. The currently available literature concerning this topic lacks any other human trials that would investigate intranasal insulin’s effects in patients diagnosed with affective disorders. Possibly, further evaluations, performed in groups of participants with milder forms of mood abnormalities or sustaining the insulin supply for longer period, could still show insulin’s efficacy in this field. The results from clinical trials concerning intranasal insulin’s effects on mood and treatment of affective disorders are summarized in

Table 4. Summarized reports of clinical trials concerning intranasal insulin’s effects on mood and in treatment of affective disorders.

Participants	Males	Females	Age (Years)	Insulin Type	Applied Dose	Results	Ref.
38 healthy participants	24	14	18–34	regular human insulin	40 IU (4 times/day for 8 weeks)	improved emotional state and self-assurance reduced self-reported anger levels	[19]
35 participants with TRMDD	13	22	18–65	regular human insulin	40 IU (4 times/day for 4 weeks)	no significant impact on mood	[41]
17 obese participants	0	17	26 (mean)	regular human insulin	160 IU (single dose)	mood improvement revealed by scores of PANAS questionnaire	[57]

Abbreviations: IU, international units; PANAS, Positive and Negative Affect Schedule; TRMDD, treatment-resistant major depressive disorder.

.

6. Intranasally Administered Insulin’s Effects on Blood Flow in Cerebral Vascular Bed

Insulin has been attributed with an ability to affect blood flow through direct stimulation of endothelial cells as well as other elements of blood vessels’ walls. Most studies point at its potential to act as a vasodilator, although in higher concentrations it is also capable of inducing vasoconstrictive responses [58]. A medical measure to modulate cerebral blood flow (CBF), both at the main vessels and at local circulation levels, could prove critical for improving treatment of various neuropsychiatric diseases. By raising the influx of nutrients and oxygen to neurons and glial cells this could possibly positively impact their activity and regeneration. Such properties would potentially help in counteracting the development of neurodegenerative disorders and ameliorate rehabilitation after physical damage or ischemia of CNS structures. Current research on this topic, concerning alteration of CBF by intranasal insulin among human subjects, seems promising, although there are some inconsistencies. Available clinical trials report that the administration of insulin through the nasal cavity proves to increase blood flow rate in areas of occipital grey matter, thalamus, insula, temporal lobe and prefrontal cortex. Detectable modulation can be achieved with just a single dose of 40 IU of insulin. However, individual assessments significantly differ regarding affected brain regions, age of the participants and presence of comorbidities. Visible distinctions in notified outcomes may also result from the relatively small numbers of participants, as the whole examined population has not exceeded 50 subjects per study in available trials [59,60,61]. Studies assessing the utilization of a combined approach, consisting of intranasal insulin and other substances, showed that cortisol does not present any ability to affect CBF alterations induced by insulin while supply of nitrates ameliorates the insulin-induced rise in blood flow in numerous brain regions, including putamen, amygdala, accumbens, pallidum and parietal lobes [61,62]. Some clinical trials report that insulin can impair CBF as well; however, it should be noted that they concern higher doses of the mentioned medication—160 IU. Such dosage was shown to significantly decrease blood flow in the hippocampus, hypothalamus, prefrontal cortex, insula and temporal or frontal lobes. The outcomes vary to some degree, depending on the obesity grade of the subjects, which may reflect systemic insulin resistance [63,64]. All the referenced studies do not provide an accurate and detailed explanation of the mechanism in which the insulin modifies CBF. Although there is evidence for direct impact of insulin on the vessels’ structural elements, most authors tend to favor the hypothesis that metabolic effects exerted on neurons and glial cells are of greater significance in matter of regulating blood flow in the cerebral vascular bed. By altering the mentioned cells metabolism, insulin increases their activity and indirectly induces them to stimulate blood flow in areas of higher synaptic transmission [58,59,60]. This would explain region-specific insulin action in CBF regulation and possible differences in its potential among insulin-resistant subjects or those treated with doses higher than physiological [62,63,64]. To clarify those suspicions, more trials are necessary involving larger participant groups, longer dosage periods, simultaneous biochemical and histopathological assessments of vascular function and recruitment of subjects with different states of systemic resistance to insulin. The results from clinical trials concerning intranasal insulin’s effects on blood flow within CNS structures are summarized in

Table 5. Summarized reports of clinical trials concerning intranasal insulin’s effects on blood flow within CNS structures.

Participants	Males	Females	Age (Years)	Insulin	Applied Dose	Results	Ref.
19 healthy participants	19	0	20–69	regular human insulin	40 IU (single dose)	increased blood flow through occipital grey matter and thalamus area among older participants (≥60 years old)	[59]
14 healthy participants and 15 participants with DM	12	17	60 (mean)	regular human insulin	40 IU (single dose)	increased blood flow in the right insula in all subjects	[60]
48 healthy participants	48	0	23 (mean)	regular human insulin	40 IU (single dose) + 10 mg of cortisol (1 dose every 15 min up to 3 doses total)	increased blood flow in the insula region lack of significant distinctions between groups receiving sole insulin or insulin + cortisol	[61]
18 abdominally obese participants	18	0	18–60	regular human insulin	160 IU (single dose administered 165–195 min after application of 625 mg of potassium nitrate)	increased blood flow in putamen, amygdala, accumbens, pallidum and temporal, frontal and parietal lobes in subjects receiving insulin + potassium nitrate (comparison with subjects receiving sole insulin) increased blood flow in putamen, caudate, thalamus and occipital and parietal lobe in subjects receiving insulin decreased blood flow in temporal and frontal lobe	[62]
12 healthy participants and 14 overweight participants	26	0	26 (mean)	regular human insulin	160 IU (single dose)	decreased blood flow in the hippocampus, insula and cerebral cortex among overweight participants	[63]
25 healthy participants and 23 overweight/obese participants	27	21	26 (mean)	regular human insulin	160 IU (single dose)	decreased blood flow in the hypothalamus in both groups of subjects decreased blood flow in the prefrontal cortex in slim individuals	[64]

Abbreviations: DM, diabetes mellitus; IU, international units.

.

7. Intranasally Administered Insulin’s Effects in Neurodevelopmental Disorders

Intranasal insulin has emerged as a promising therapeutic intervention for multiple neurodevelopmental conditions through brain signaling modulation. Various assessments of rodent brain samples, conducted with techniques of immunohistochemistry and enzyme-linked immunosorbent assays, revealed that insulin counteracts tau phosphorylation and reduces amyloid burden—the main characteristic features of Alzheimer’s-like pathology, which also develop in Down syndrome-affected individuals by the ages of 35–40 years [65,66]. Rosenbloom et al. conducted a double-blind, placebo-controlled pilot study with twelve Down syndrome-affected adults, aged 35–53 years, using single-dose intranasal glulisine (20 IU), demonstrating its excellent safety and an ability to induce a non-significant trend toward improved memory retention [67]. Schmidt et al. performed an exploratory study in six children with Phelan–McDermid syndrome (22q13 deletion), a rare genetic condition causing developmental delays and cognitive impairments due to ProSAP2/Shank3 gene loss. The one-year trial with intranasally administered insulin showed significant improvements in motor skills, cognitive function, and non-verbal communication within six weeks that sustained through twelve months. The positive outcomes were accompanied by good tolerability and no serious adverse events or blood glucose-level alterations [68]. Zwanenburg et al. validated these findings using a stepped-wedge design with 25 pediatric participants carrying 22q13.3 deletions, encompassing the SHANK3 gene over 18 months. While results concerning overall developmental improvements lacked statistical significance, children aged >3 years showed significant enhancements in cognitive function and social competencies, particularly notable given the characteristic developmental regression in population affected by this illness [69]. These studies demonstrate intranasal insulin as a non-invasive, safe therapeutic approach enhancing cognitive function through improved cerebral glucose metabolism and synaptic plasticity across neurodevelopmental populations, though future research based on larger trials remains necessary [67,68,69]. The results from clinical trials concerning intranasal insulin’s effects in neurodevelopmental disorders are summarized in

Table 6. Summarized results of clinical trials concerning intranasal insulin’s effects in neurodevelopmental disorders.

Participants	Males	Females	Age (Years)	Insulin Type	Applied Dose	Results	Ref.
12 participants with Down syndrome	6	6	35–53	glulisine	20 IU (single dose)	non-significant trend toward improved memory retention	[67]
6 participants with 22q13 deletion syndrome	2	4	1.5–9.5	regular human insulin	1 dose/day for 12 months (dosage increased gradually at 3-day intervals starting at 2 IU/day and reaching final values of 0.5–1.5 IU/kg/day)	after six weeks: 4 children demonstrated significant progress in motor control, attention span, and behavioral regulation after 12 months: 4 children showed sustained improvements in motor functions, speech understanding, communication, attention span, and daily autonomy	[68]
25 participants with 22q13 deletion syndrome	6	19	1–16	regular human insulin	20 IU (children 1–3 years old) 30 IU (children 3–9 years old) 40 IU (children 9–18 years old) complete daily dose was divided into 2 smaller doses/day; treatment period lasted 18 months	provided additional developmental gains of cognition, receptive language, and fine motor skills children > 36 months showed more pronounced improvements in cognition and social skills, while the control group showed developmental decline	[69]

Abbrevations: IU, international units.

.

8. Intranasally Administered Insulin’s Effects on Appetite and Modulation of Food Intake

Intranasal insulin administration modulates appetite and food intake through direct CNS targeting, with effects varying substantially by dose, timing and individual metabolic characteristics dictated by endocrine and inflammatory activity of peripheral fat tissue and sensitivity to insulin. At higher doses (160 IU), intranasal insulin reduces brain activation in response to food images. Affected brain regions include the fusiform gyrus, hippocampus and frontal cortex, potentially signaling postprandial satiety [70]. Insulin-induced enhancement of resting-state functional connectivity between prefrontal regions and the hippocampus–hypothalamus network correlates with hunger reduction at 120 min post-administration [71]. When administered postprandially, this dose of 160 IU selectively decreases consumption of highly palatable foods, particularly chocolate chip cookies, by approximately 32% and without affecting less-rewarding snacks. This suggests a specific modulation of hedonic rather than homeostatic feeding pathways [72]. Timing-dependent effect appears crucial, as the same dose applied during fasting shows no impact on appetite or food intake [72]. Moreover, lower doses (40 IU) fail to alter virtual supermarket shopping behavior and cookie consumption in healthy young men [73]. The mechanism underlying appetite-suppressing effects involves rapid enhancement of brain energy metabolism, with intranasal insulin increasing cerebral ATP and phosphocreatine levels within 10 min. Consequently, these elevated energy levels strongly correlate with reduced food consumption, supporting the hypothesis that the brain monitors its own energetic status to regulate feeding behavior [74]. Neuroimaging reveals decreased activity in the hypothalamus and orbitofrontal cortex—key areas for homeostatic and reward-related food processing—with individual responses varying according to body mass index (BMI), as prefrontal and anterior cingulate cortex activity shows positive correlations with body weight [75]. During sweet taste anticipation, normal-weight individuals demonstrate greater insulin-induced activity than overweight subjects in reward-processing regions including the anterior cingulate cortex, ventromedial prefrontal cortex and nucleus accumbens, with nucleus accumbens activity inversely correlating with BMI [76]. The relationship between insulin activity and BMI is bidirectional. As mentioned above, some of the effects exerted by insulin are BMI-dependent, but insulin can also affect body weight and consequently BMI. By acting on insulin receptors present on hypothalamic arcuate neurons, insulin stimulates anorexigenic pro-opiomelanocortin (POMC)/cocaine- and amphetamine-regulated transcript (CART) cells and simultaneously inhibits activity of orexigenic agouti-related peptide (AgRP)/neuropeptide Y (NPY) neurons. Moreover, insulin affects functionality of mesolimbic dopaminergic circuits involved in food reward, such as the ventral tegmental area (VTA) and nucleus accumbens (NAc). These mechanisms combined lead to inhibition of food intake and therefore limit weight gain [77]. Individual characteristics fundamentally shape insulin’s neural and behavioral effects, with women showing increased and men showing decreased dorsolateral prefrontal cortex activity for highly desired foods, while the insula displays complex patterns where normal-weight men and overweight women exhibit increased activation [78]. Paradoxically, women with obesity demonstrate greater sensitivity to intranasal insulin’s appetite-suppressing effects compared to lean women who show more pronounced reductions in cookie intake, appetite ratings, and food-associated feeling of reward [57]. Behaviorally, the 160 IU dose reduces hunger particularly in normal-weight men and overweight women while increasing preference for low-caloric foods. The effects are mediated by enhanced hippocampal connectivity, especially in individuals with high peripheral insulin sensitivity [71,78]. Despite promising short-acting outcomes, clinical translation remains uncertain, as a large, 24-week-lasting randomized controlled trial utilizing 40 IU of daily intranasal insulin in older adults found no effects on food intake, appetite or body weight, regardless of gender, BMI or type 2 DM status [79]. This discrepancy likely reflects differences in dosing protocols, with short-term studies using higher single doses of insulin, showing effects that may not translate to chronic lower dose administration [57,72,74,79]. The collective evidence demonstrates that intranasal insulin modulates the entire food processing pipeline from visual perception through anticipation to consumption via enhanced brain energy metabolism, altered reward processing and improved satiety signaling, with effects moderated by peripheral insulin sensitivity, cognitive restraint and body weight status, suggesting therapeutic potential for appetite regulation requires careful optimization of dose, timing and consideration of individual metabolic profiles [64,75,76,78]. The results from clinical trials concerning intranasal insulin’s effects on appetite and food intake modulation are summarized in

Table 7. Summarized results of clinical trials concerning intranasal insulin’s effects on appetite and modulation of food intake.

Participants	Males	Females	Age (Years)	Insulin Type	Applied Dose	Results	Ref.
52 healthy female participants (35 lean, 17 obese)	0	52	22–28	regular human insulin	160 IU (16 doses of 10 IU in 30 s intervals)	reduced appetite and cookie intake in obese women increased left insula activation for food vs. non-food pictures in both lean and obese participants	[57]
48 healthy participants (25 lean, 23 overweight or obese)	27	21	26 (mean)	regular human insulin	160 IU (single dose)	reduced desire for sweet foods in lean men	[64]
9 healthy participants	5	4	25 (mean)	regular human insulin	160 IU (4 doses of 40 IU administered within 5 min)	reduced brain activity in response to food pictures (but not nonfood pictures) in several regions: right and left fusiform gyrus, right hippocampus, right temporal superior cortex, right frontal middle cortex, and left postcentral cortex	[70]
47 healthy participants (25 lean, 10 overweight, 12 obese)	26	21	26 (mean)	regular human insulin	160 IU (single dose)	improved functional connectivity between prefrontal regions of the default-mode network and both hippocampus and hypothalamus improved connectivity between: -dorsal/anterior medial prefrontal cortex and right hippocampus -anterior medial prefrontal cortex and hypothalamus (only in peripherally insulin-sensitive individuals	[71]
30 female participants taking oral contraceptives	0	30	22–23 (mean)	regular human insulin	160 IU (16 doses of 10 IU administered at 60 s intervals)	postprandial intranasal insulin administration reduced appetite ratings and decreased intake of chocolate chip cookies (the most palatable snack) by approximately 32% compared to placebo the reduction was selective for highly palatable foods, with no effect on less rewarding snacks (spritz or coconut cookies)	[72]
30 healthy male participants	30	0	26 (mean)	regular human insulin	40 IU (single dose)	no significant effect on food ratings, calorie content of purchased food products, and cookie consumption	[73]
15 healthy male participants	15	0	25 (mean)	regular human insulin	40 IU (4 doses of 10 IU administered at 60 s intervals)	increased brain energy metabolism and ATP levels reduced food consumption and calorie intake (particularly from carbohydrates and protein) strong inverse relationship between brain energy levels and food consumption	[74]
17 healthy lean female participants	0	17	24.5 (mean)	regular human insulin	160 IU (4 doses of 40 IU administered within 5 min)	decreased activity and fALFF in hypothalamus and orbitofrontal cortex increased fALFF induction (positively correlating with BMI values) in prefrontal cortex at 30 min after insulin application and anterior cingulate cortex at 90 min after application	[75]
24 healthy male participants (10 lean, 14 overweight)	24	0	26 (mean)	regular human insulin	160 IU (4 doses of 40 IU administered within 5 min)	increased activity of anterior cingulate cortex, ventromedial prefrontal cortex and nucleus accumbens in lean participants for sweet cue anticipation lower activation of amygdala during stevia receipt in lean participants	[76]
60 healthy participants (37 lean, 23 overweight or obese)	30	30	21–69	regular human insulin	160 IU (4 doses of 40 IU administered within 4 min)	BOLD activation of amygdala to food cues increased activity of insula in lean men and overweight women (insula response correlated with peripheral insulin sensitivity and cognitive restraint) increased activity in women and decreased activity in men of dorsolateral prefrontal cortex for highly desired food cues reduced hunger (especially for lean mean and overweight women) increased desire for low-caloric foods	[78]
89 participants (51 healthy, 38 with type 2 of DM)	47	42	50–85	regular human insulin	40 IU (1 dose/day for 24 weeks)	no significant effect on appetite, hunger, food intake or any of the secondary outcomes (anthropometric measures including body weight and body composition)	[79]

Abbreviations: ATP, adenosine triphosphate; BMI, body mass index; BOLD, blood-oxygen-level-dependent; DM, diabetes mellitus; fALFF, fractional amplitude of low-frequency fluctuations; IU, international units.

.

9. Other Potential Neuropsychiatric Applications of Intranasal Insulin

Among reports from clinical trials concerning intranasally administered insulin’s applications in treatment of neuropsychiatric disease entities, there are also individual relations from studies and case reports regarding fear and stress response, nicotine addiction and poisoning with carbon oxide (CO).

Intranasal insulin has emerged as a promising modulator of stress response and cognitive function. Administration of insulin via the nasal cavity (40–160 IU) effectively attenuated the hypothalamic–pituitary–adrenal (HPA) axis response to psychosocial stress, resulting in downregulation of cortisol levels by 49–68% following acute stress exposure [80,81]. This effect occurred without altering baseline cortisol, glucose or peripheral insulin levels, confirming that the mechanism is centrally mediated through direct access to the cerebrospinal fluid via the olfactory route [80,82].

Beyond modulation of stress hormone, intranasal insulin demonstrates significant effects on both appetitive and aversive learning processes. In the context of addiction, insulin reduced nicotine cravings and normalized the typically blunted cortisol response to stress seen in smokers—both factors that independently predict smoking relapse [81]. For fear-related disorders, intranasal insulin facilitated fear extinction learning, with participants showing enhanced reduction in fear-potentiated startle responses and improved extinction recall. The noted effect was particularly visible in women who demonstrated greater decrease in skin conductance responses during both early extinction and re-extinction phases [82].

The therapeutic potential of intranasal insulin extends to severe neurological conditions, as demonstrated in a case of delayed neuropsychiatric syndrome, following CO poisoning, where combined therapy encompassing intranasal insulin (50 IU twice daily for 10 days per month), umbilical cord blood stem cells and nicotine resulted in complete neurological recovery with MRI-confirmed lesion reduction [83]. The mechanisms underlying these diverse effects likely involve insulin’s action on multiple brain regions with high receptor density, including the hippocampus, amygdala, and hypothalamic nuclei, where it modulates glutamatergic and cholinergic signaling, influences dopamine release in reward circuitry and affects memory consolidation processes [80,81,82]. The results from clinical trials concerning intranasal insulin’s effects on fear response are summarized in

Table 8. Summarized results of clinical trials concerning intranasal insulin’s effects on fear response.

Participants	Males	Females	Age (Years)	Insulin Type	Applied Dose	Results	Ref.
26 healthy male participants	26	0	20–31	regular human insulin	40 IU (single dose)	decreased blood plasma and saliva cortisol levels	[80]
37 healthy smokers	27	10	18–65	regular human insulin	60 IU (single dose) following overnight abstinence from nicotine	reduced nicotine cravings over time, effect lasted through psychosocial stress period (100 min post-administration) stress response normalization: significantly increased peak cortisol levels following stress normalized the typically blunted cortisol response to stress seen in smokers no effect on cardiovascular stress measures (HR, BP)	[81]
123 healthy participants	60	63	18–35	regular human insulin	160 IU (single dose)	decreased fear-potentiated startle decreased SCR to fear in women participants significant over-time decay of US expectancy	[82]

Abbreviations: BP, blood pressure; HR, heart rate; IU, international units; SCR, skin conductance response; US, unconditioned stimulus.

.

10. Conclusions

Intranasally administered insulin emerges as valuable candidate to expand therapeutic schemes in various disorders that originate from the CNS or impair its functionality. Although the actual state of knowledge in this matter is based on a relatively limited set of clinical studies, the available reports allow us to notice the most promising directions of future research. Particularly reasonable seem further evaluations of its cognitive-boosting potential, as most of the trials concerning patients affected by AD, MCI or inborn forms of mental and developmental disability present significant and positive outcomes for insulin-treated participants. Metabolism-altering capabilities of insulin in brain structures, that may also affect regional blood flow rates, should not remain unnoticed as they may possibly expand perspectives for neurorehabilitation and regeneration of damaged regions. Moreover, such metabolic modulations appear to affect central processing of appetite and mechanisms associated with food intake. Further examinations of this phenomenon may possibly allow us to find applications for intranasally administered insulin in the treatment of eating disorders or obesity. The results of studies concerning POD show insulin as a very effective measure to prevent occurrence of this emergency state and consequently significantly ameliorate perioperative care. Of course, those are only promising prospects and a good basis for continuing research, as many issues associated with intranasal insulin’s clinical translation remain unanswered. The main flaw and limitation of the vast majority of the trials referenced in this review is their small number of participants or short exposition time. Future analyses should focus on recruiting more subjects as well as assessing outcomes in longer supply periods. Insulin’s action in human organisms is inextricably connected with tissues’ insulin resistance and activity of other metabolism-altering hormones or medications. Further evaluations of intranasal insulin’s efficacy, in treatment of neuropsychiatric diseases, may expand current findings by putting more emphasis on estimating if factors like age, sex, obesity state, markers of systemic insulin resistance or the presence of other comorbidities affect the response to insulin in the CNS. In addition, it also seems important to verify safety of long-term applications of insulin to nasal mucosa and determine if it does not exert any negative morphological changes, similar to the hypertrophy that develops in skin parts subjected to repetitive insulin injections. As most studies referenced in this review utilize preparations of regular human insulin, an assessment of insulin’s analogs could significantly expand actual findings and possibly reveal increased efficacy of those alternative forms.