1 Introduction

Inherited metabolic disorders (IMDs) encompass rare conditions of intermediary metabolism and metabolic cell signaling that predispose individuals to hypoglycemia [1]. Among intermediary metabolism disorders, the most common are Glycogen Storage Disorders (GSDs), Hereditary Fructose Metabolism Disorders (HFMDs), and disorders of fatty acid and ketone body metabolism, while congenital hyperinsulinism (CH) is the most frequent among metabolic cell signaling disorders. GSDs, are characterized by monogenic defects affecting glycogen synthesis or breakdown, leading to abundant glycogen in liver and/or muscle [2]. With an estimated incidence of 1 case per 20,000–43,000 live births and a wide spectrum of clinical presentation [3], individuals with GSDs necessitate dietary management to maintain glucose homeostasis, with organ transplantation being considered in some cases [2, 3].

HFMDs, including Hereditary Fructose Intolerance (HFI) and fructose-1,6-bisphosphatase (FBPase) deficiency, can lead to low glucose levels. HFI typically manifests after the ingestion of fructose, and treatment involves eliminating all sources of fructose from the diet [4]. FBPase deficiency is a defect in gluconeogenesis, where hypoglycemia typically occurs when liver glycogen is depleted, and maintenance of glucose homeostasis becomes dependent on gluconeogenesis. Dietary management include frequent carbohydrate-containing meals and avoiding prolonged intervals between feeds (e.g. during the overnight period and when individuals are sick and cannot drink and eat normally) [4].

Congenital Hyperinsulinism (CH) is a common cause of persistent hypoglycemia in children, caused by mutations in key genes leading to excessive insulin secretion [5, 6]. Long-term management entails pharmacological and surgical interventions to regulate insulin secretion and prevent hypoglycemia-associated brain injury [5,6,7].

These IMDs require meticulous blood glucose monitoring, which can be achieved through self-monitoring blood glucose (SMBG) or Continuous Glucose Monitoring (CGM). CGM has proven particularly valuable in Type 1 Diabetes Mellitus (T1DM) for improving glycemic control and detecting hypoglycemia [8,9,10,11]. Moreover, CGM has been increasingly adopted to enhance glucose and metabolic control in the management of various chronic diseases, including among diabetes subjects undergoing dialysis [10, 12]. Over the past few decades, there have been significant advancements in diabetes technology, particularly in glucose monitoring and insulin delivery systems. This has led to improved glycemic control, reduced diabetes-related complications, and higher quality of life for many patients [10, 11].

CGM devices, including Real-time CGM (Rt-CGM) and Flash Glucose Monitoring (FGM), offer real-time glucose data, supporting glycemic pattern analysis and hypoglycemia prevention [10, 13]. CGM has also been used to retrospectively analyze glucose patterns and trends for diagnosis and patient education.

The role and outcomes of CGM in individuals with IMDs at risk for hypoglycemia are still debated by experts in the field [13,14,15,16]. The current IMDs guidelines are cautious about the recommendations on the use of CGM and its clinical practice application in nutritional improvements or modifications [17,18,19,20]. Rossi et al. have recently provided the first prospective CGM data for individuals with GSDIa, which can enhance patient monitoring and support precision medicine in both clinical care and research [16]. Peeks et al. also indicated outcome parameters for hepatic GSDs [15]. In contrast, no recommendations are available for FGM use in these patients.

Worth et al. recently analyzed the state of art of the evidence for the use of CGM in non-diabetic children with hypoglycemia [13]. Some suggestions for an appropriate CGM use have been provided [13], although its use in nutritional therapy still needs to be further evaluated.

The systematic evaluation of Rt-CGM and FGM in IMDs during nutritional therapy is crucial, given the specific dietary treatments of these individuals. Therefore, this systematic review aims to explore the potential of glucose monitoring devices in managing hypoglycemia in individuals with IMDs, with a focus on dietary interventions and nutritional care.

2 Materials and methods

2.1 Protocol and registration

This systematic review was developed according to preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement [21]. The protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO), registration number [CRD42024497744] (https://www.crd.york.ac.uk/prospero/).

2.2 Search strategy

We searched for studies about CGM in individuals with IMDs at risk for Hypoglycemia published from January 01, 2003, up to October 15, 2023 in the electronic database Pubmed (Supplemental Table 1).

2.3 Study selection, data extraction and management

We conducted a systematic literature research according to the PICOS model (Population, Intervention, Comparison, Results, Study design) (Supplemental Table 2).

Two investigators (G.G. and E.M.) independently screened the titles and abstracts of the identified studies and, after abstract selection, conducted the complete paper analysis (Supplemental Table 2). One investigator (G.G.) extracted data and the other (E.M.) checked the extracted data. Discrepancies in study selection and data extraction were resolved by consensus or, when consensus could not be reached, by involving a third investigator (N.V.). For each selected study, the first author, year of publication, country of origin, study design, sample characteristics, methods and outcomes were recorded.

The recording process relied on the utilization of Excel software, documenting all described procedures and ensuring traceability of decisions throughout the phases of study selection and data extraction.

2.4 Assessment of risk of bias in individual studies

National Institutes of Health (NIH) Quality Assessment Tools [22] were used to assess the risk of bias of individual studies. Two authors (G.G. and E.M.) independently scored articles based on the fundamental criteria of this tool, according to the type of the analyzed study. Reviewers were blinded to each other’s assessment, and disagreements were resolved by reaching consensus or by involving a third investigator (N.V.). Based on the ratings, an overall judgement was made regarding the quality of each study: (1) “good quality” if the study had a minimal risk of bias, (2) “fair quality” if the study was susceptible to some bias but was not deemed sufficient to invalidate its results, and (3) “poor quality” if the study raised substantial concerns.

3 Results

A total of 124 titles and abstracts were screened following the search, of which 44 were considered for full-text screening. Finally, 24 articles were selected and included in this review (Fig. 1).

Fig. 1
figure 1

Flowchart of literature screening and selection process for studies included in the review, adapted from PRISMA 2020 flow diagram

Full size image

Thirteen studies reported data on glucose monitoring in GSDs, ten studies in CH, and one in HFMDs.

Among studies reporting data on subjects with GSDs, ten studies described the use of only Rt-CGM (total of 139 GSD study participants) [15, 16, 23,24,25,26,27,28,29,30], two studies focused only on FGM in 2 individuals [31, 32], and one study reported data on 14 subjects monitored by both FGM and Rt-CGM [33]. Among studies reporting data on research participants with CH, one investigated FGM in 11 children [34], and nine evaluated the use of Rt-CGM (total of 159 individuals) [35,36,37,38,39,40,41,42,43]. The only study reporting data on HFMDs was a single case report study using FGM [44].

Seven studies employed blinded CGM [23, 24, 26, 29, 30, 36, 41], while nine utilized non-blinded CGM [25, 27, 28, 31,32,33,34,35, 43]. Additionally, five studies utilized both blinded and unblinded monitoring [15, 33, 37,38,39], and three studies did not specify the monitoring method used [16, 40, 42].

Key findings about Rt-CGM or FGM in IMDs prone to hypoglycemia are consolidated and summarized in Table 1.

Table 1 Main findings and highlights resulted by the systematic review
Full size table

4 Glucose monitoring in Glycogen Storage disorders

4.1 Rt-CGM in GSDs

Population features and study designs of the selected studies regarding CGM in GSDs are summarized in Table 2.

Table 2 Rt-CGM/FGM in GSDs: population features, devices and study design
Full size table

Most studies included both children and adults [15, 24, 25, 28,29,30]; and three studies specifically investigated adults [16, 26, 27].

The main types of GSDs represented were Glycogen Storage Disease type Ia (GSDIa) (95/130 = 45,4%), Glycogen Storage Disease type Ib (GSDIb) (6/130 = 6.2%), Glycogen Storage Disease type III (GSDIII) (10/130 = 10%), Glycogen Storage Disease type IX (GSDIX) (9/130 = 6.2%), and Glycogen Storage Disease type 0 (GSD0) (2/130 = 1.5%). Excluding the case report [31], seven studies focused exclusively on subjects with GSDI [15, 16, 23, 25,26,27,28].

The duration of glucose monitoring by Rt-CGM varied from 48 h to 7–10 days (Table 2).

All studies reported the use of medical foods as recommended by guidelines for GSD (such as UCCS or modified corn-starch (MCS) or glucose polymer) (Table 3), and most reported that research participants were already following an appropriate diet for GSD. All studies reported the use of UCCS, and eight studies provided the precise amount of UCCS (either in grams/day or grams per kg of body weight/day) [16, 24,25,26,27, 29, 30]. MCS use was mentioned in only five papers [15, 16, 26,27,28] with the precise amount provided (in grams/day or grams per kg of body weight/day) only in four papers [16, 26,27,28]. Peeks et al. [15] reported only a load test without the dose prescription. Glucose polymer use was used by White et al. [30], while Fukuda et al. [25] used a low-fat, lactose/fructose-free formula.

Table 3 Studies reporting dietary interventions, metabolic outcomes and reliability in patients with GSD using Rt-CGM
Full size table

Most studies described dietary changes during Rt-CGM [15, 16, 23, 24, 26,27,28, 30] (Table 3). Two studies [26, 29] analyzed relevant nutritional parameters and dietetic therapy during Rt-CGM without making any changes in diet. Maran et al. [29] concluded that Rt-CGM could provide a rationale for changes in dietary treatment because CGM revealed unrecognized hypoglycemia, while Fukuda et al. [25]. assessed nutritional parameters and observed different dietary intake between ages in GSD homozygous group. Three studies assessed the effects on metabolic control of specific dietary interventions, in particular the use of UCCS vs. MCS [15, 27, 28] or the use of three different meals with an equal content of carbohydrates (CHO) consumed before bedtime [27]. In particular, Peeks et al. evaluated modification of the amount of nocturnal UCCS and the change from UCCS to MCS during nocturnal CGM [15].

Rossi et al. focused on CGM metrics in patients with GSDIa and provided only an observational description of different dietary treatments, such as UCCS, MCS, or continuous nocturnal gastric drip-feeding [16].

4.2 Rt-CGM in GSDs: outcomes

No studies have directly compared dietary interventions during Rt-CGM with blood/capillary glucose monitoring (Table 3). In subjects treated with MCS or other carbohydrate sources vs. UCCS at bedtime [15, 27, 33], Hsu et al. reported improvements in sleep time and fasting duration in individuals treated with MCS [28], and Hochuli et al. mentioned achieving stable nocturnal glucose control with the “pasta al dente” meal, and prolonged fasting times in some enrollees treated with MCS [27]; Peeks et al. reported less glycemic variations with MCS [15].

Rt-CGM combined with dietary modifications led to improvements in glucose values that were associated with better pre-set target in range (70–150 mg/dL) [24], pre-set target below range [15], glycemic variation/glucose variability [15, 26], morning glucose levels [28], and stability of nocturnal glucose [27].

The only case-control study documented lower pre-set target in range and higher pre-set time above range in GSDs vs. healthy subjects [16]. In this study, dietary modifications during Rt-CGM were found to be useful in preventing severe low glucose levels associated with mental or physical functioning impairment (level 3 hypoglycemia) [16].

Some authors reported improvement of additional biochemical parameters such as aspartate aminotransferase (AST), alanine aminotransferase (ALT) [23, 28], lactate [23, 30], triglycerides [23] and ketones [30] (Table 3).

Peeks et al. reported improved gastrointestinal absorption (better macronutrients and UCCS uptake) in GSDIb with empagliflozin [15]. In an observational study, Fukuda et al. underscored the prevalence of asymptomatic hypoglycemia in individuals with hepatic GSDs, with total cholesterol and lactate independently linked to hypoglycemia [25].

Two studies reported Rt-CGM as a safe, efficacious, and reliable system to identify asymptomatic hypoglycemia (Table 3) [23, 29].

Three studies compared Rt-CGM and blood/capillary glucose values: Herbert et al. and Rossi et al. reported good correlation between capillary glucose values and CGM [16, 24], even when glucose levels were < 70 mg/dL [24]; Hochuli et al. showed Rt-CGM was a reliable tool in outpatients, according to capillary blood measurements [27]. Herbert et al. stated that Rt-CGM is a useful tool both for patients to improve compliance with dietary treatment as well as for healthcare providers to facilitate home monitoring [24].

No studies evaluated individuals’ or parents’ perceptions of Rt-CGM in GSDs.

4.3 FGM in GSDs

Two case reports investigated FGM in two adults with GSDIa [31] and GSD III [32]. One study used both Rt-CGM and FGM, including 14 GSDIa and 2 GSDIb subjects, with a mean age of 21 years [33]. The duration of glucose monitoring with FGM varied from 6 days to 4 months (Table 2).

All FGM studies included the consumption of UCCS (Table 4). Marcalo et al. and Massimino et al. [31, 32] reported precise amounts in individual case reports (grams per day or grams per kilogram of body weight per day), while Kaiser et al. [33] did not describe UCCS doses. Massimino et al. also reported the use of glucose polymer [32]. Kaiser et al. showed 44% of GSD individuals were fed using a gastric tube as an alternative to UCCS or MCS, not specifying if the feeding source was glucose polymer [33].

Table 4 Studies reporting dietary interventions, metabolic outcomes and reliability in patients with GSD using Rt-FGM
Full size table

The two case reports described relevant dietetic modifications. By reducing UCCS dose, Massimino et al. gradually decreased the amount of CHO and increased fats [32]. Marcalo et al. revised the prescribed diet with stricter indications [31].

4.4 FGM in GSDs: outcomes

Massimino et al. showed a reduced number and duration of hypoglycemic episodes, and increased time in pre-set target range (70–140 mg/dL) [32] (Table 4).

Kaiser et al. found (micro)albuminuria in 59% of study participants, with more frequent events of low blood glucose and a trend for a higher AUC of low glucose in CGM measurements in patients with microalbuminuria (Table 4) [33]. The mean-total-serum-calcium was lower in GSDIb compared to GSDIa (statistically significant), while mean PTH concentration was slightly lower. Z-scores of bone density correlated negatively with serum lactate and positively with body mass index (BMI) [33].

Massimino et al. showed also a greater exercise capacity during dietary modifications and FGM [32]. The selected studies did not evaluate the concordance between the FGM and capillary or blood glucose monitoring nor individuals’ or parents’ perceptions.

4.5 Glucose monitoring in congenital hyperinsulinism

4.5.1 Rt-CGM in CH

Ten studies evaluated Rt-CGM in CH, involving a total of 159 subjects. The main characteristics of the selected studies are shown in Table 5.

Table 5 Rt-CGM/FGM in CH: population features, devices and study design
Full size table

Most studies included only children, with one study focusing solely on neonates [35], and only one investigation included both pediatric and adult subjects [43]. While most studies focused on individuals with CH, one study included CH treated with pancreatectomy [43], and another considered some uncharacterized hyperinsulinemic disorders [42]. The duration of glucose monitoring varied among the selected studies, ranging from 2 to 12 weeks. Two studies analyzed different periods of blinded or unblinded monitoring [37, 38], and two studies used only blinded Rt-CGM [38, 43].

Four studies reported the use of special medical foods in CH individuals during Rt-CGM. Skae et al. utilized dietary supplements with polyunsaturated fatty acids (PUFA) [41], Sivasubramanianan et al. described oral feeds or gastric tube without describing the precise amount [42], and Rayannavar et al. reported the use of enteral dextrose overnight [36] (Table 6). Among the studies describing relevant dietetic modifications during Rt-CGM in CH, Skae et al. showed PUFA supplementation reduced glycemic variability in nine children with CH responsive to diazoxide; however, glucose levels didn’t significantly increase overall, and safety parameters remained normal [41]. Rayannavar et al. compared carbohydrate interventions during CGM in CH individuals using a bi-hormonal bionic pancreas (BHBP) versus usual care, finding fewer interventions needed during the BHBP period [43](Table 6).

Table 6 Studies reporting dietary interventions, metabolic outcomes and reliability in patients with CH using Rt-CGM
Full size table

4.6 Rt-CGM in CH: outcomes

All selected studies investigated metabolic or glycemic control in CH subjects during Rt-CGM (Table 6). Five studies focused on the identification of hypoglycemia and its patterns during continuous monitoring [35,36,37, 39, 40]. Rayannavar et al. reported CGM as an effective adjunctive tool for managing persistent hypoglycemia, decreasing the risk of undetected hypoglycemia by alerts [36]. Worth et al. reported large variations in hypoglycemia and the risk for hypoglycemia during the day in a national cohort [40]. By practicing proactive behavioral changes, a reduction of pre-set TBR from 7.1 to 5.4% has been documented [37].

The pilot pre-post intervention trial conducted by Skae et al. included an examination of serum parameters related to lipid and metabolic profiles during Rt-CGM (fasting glucose, insulin, HOMA-IR index, ALT, albumin levels, triglycerides, total cholesterol, and LDL cholesterol) (Table 6) [41]. The study found no significant changes in insulin resistance or liver function. Elevated serum cholesterol in one child post-PUFA treatment was reported as an adverse event, while LDL cholesterol remained within normal ranges. Overall, the study suggested stable metabolic parameters during Rt-CGM, emphasizing its potential use as a monitoring tool in congenital hyperinsulinism.

Four studies analyzed the concordance between Rt-CGM and capillary/blood glucose monitoring in CH [35, 36, 39, 42] (Table 6). Accuracy of Rt-CGM was considered acceptable in two studies, in which the Mean Absolute Relative Difference (MARD) was 11% in neonatal CH (sensitivity 0.59, specificity 0.94) [35] and 13% in subjects with hypoglycemia, including one-third of individuals with unknown hyperinsulinism [42].

MARD was not considered acceptable in two studies: 17.5% in 14 CH children aged 6 months-10 years (sensitivity 0.86, specificity 0.81) [36]; 19.3% based on 1441 paired values of Rt-CGM and capillary/blood glucose monitoring [39].

Worth et al. developed a Hypoglycemia Error Grid (HEG) for assessing Rt-CGM accuracy in CH individuals, finding it insufficient for diagnostic purposes and, unlike those designed for diabetes, HEG assigned lower risk to missed hyperglycemia and higher risk to small errors around missed hypoglycemia [39].

Two studies revealed positive individual and caregiver perceptions of Rt-CGM, despite some disliking alarms and receiver proximity [38, 42].

4.7 FGM in CH

In CH, only one study considered FGM [34]: Alsaffar et al. conducted a prospective observational study investigating 11 children monitored with FGM for two weeks (Table 5). The selected study did not evaluate the use of medical food, dietary interventions, glycemic or metabolic control (Table 7).

Table 7 Studies reporting dietary interventions, metabolic outcomes and reliability in patients with CH using FGM
Full size table

4.8 FGM in CH: outcomes

Alsaffar et al. observed a MARD of 17.9%, with a mean variation between FGM and Capillary Blood Glucose (CBG) of 0.29 mmol/L and a positive correlation between the two methods (r = 0.7) (Table 7) [34].

Despite this, interviewed parents reported that FGM was a very easy and convenient method, especially during sleep. However, 77% felt FGM was a non-reliable method [34].

4.9 Rt-CGM/FGM in Hereditary Fructose Metabolism diseases

In HFMDs, only one study evaluated Rt-CGM/FGM (Table 8). Morales-Alvarez et al. described the use of FGM in an undiagnosed HFI adult aged 33 years [44]. The authors reported a dietetic change after FGM and diagnosis: from a low-glycemic index diet with six meals a day (once every 3 h) to a fructose and sucrose-free diet (Table 8) [44]. The sole use of interstitial glucose monitoring in HFI revealed FGM as a potential adjunctive tool in the investigation of postprandial hypoglycemia leading to suspicion of HFI diagnosis. Indeed, as reported, a correct dietary restriction of fructose, saccharose, sorbitol, and their derivatives prevented recurrent hypoglycemia [44]. The study did not analyze the accuracy and concordance of FGM with capillary/blood glucose monitoring.

Table 8 Rt-CGM/FGM in HFMDs: population features, devices and study design; dietary interventions, metabolic outcomes and reliability in patients with HFMDs using FGM
Full size table

4.10 Assessment of Risk of Bias

Supplemental Tables 3–7 present a summary of the risk of bias in individual studies. Overall, only two studies (8%) were classified as good quality, nine studies (34%) as poor quality, and 15 studies (57%) as fair quality. A subset from Peeks et al.‘s study, identified as a Controlled Intervention Study, was deemed of poor quality [15] (Supplemental Table 3). Rossi et al. conducted a case-control study [16] which was assessed as fair quality (Supplemental Table 4). Among the pre-post studies without a control group [15, 23, 28, 41, 43] three studies exhibited fair quality and two poor (Supplemental Table 5). 50% of the studies employed an observational cohort or cross-sectional design [24,25,26,27, 33,34,35,36,37,38,39,40, 42]. Among these, five studies were assessed as poor quality, six as fair quality, and two as good quality (Supplemental Table 6). Additionally, there were six case series, including one subset from the study by Peeks et al. [29,30,31,32, 44], with two being categorized as poor quality and four as fair quality (Supplemental Table 7).

When assessing CGM in GSDs, 60% of the studies were classified as fair quality, while the remaining 40% were deemed poor quality (Table 2). All of these studies were non-randomized intervention studies, except for a subset by Peeks et al., which was a randomized, double-blind crossover trial involving eleven GSD participants [15]. Rossi et al. included matched voluntary healthy controls but did not employ randomization [16]. The majority of studies were prospective or retrospective observational or cohort studies [24,25,26,27, 33], three were pre-post intervention studies without control group [15, 23, 28], and five were case series or case reports [15, 29,30,31,32] (Table 2). In the context of CH, 30% of studies were classified as poor quality, 50% as fair quality, and 20% as good quality. None of these studies was a randomized intervention study or included matched voluntary healthy controls (Table 5). Two studies were pre-post without a control group [41, 43], with only one being a pilot trial (open label, Phase 2) [41]. Most CH studies were observational or crossover studies [34,35,36,37,38,39,40, 42] (Table 5).

5 Discussion

This review aimed to systematically examine the role of glucose monitoring devices in individuals with IMDs at risk for hypoglycemia, focusing on their effectiveness in preventing low glucose levels and guiding nutritional management.

A notable finding is the lack of high-quality studies providing strong evidence on the nutritional interventions during Rt-CGM or FGM in IMDs at risk for hypoglycemia. The studies included in this review varied in design and outcomes evaluation, despite reporting the monitoring of over 300 subjects using these devices. Most studies were deemed to be of fair or poor quality, with only one clinical trial and one case-control study available. The majority employed observational, crossover (retrospective or prospective) or pre-post intervention designs, lacking control groups.

In GSDs, most studies focused on glycemic values or fasting duration during dietary interventions in individuals previously monitored by Rt-CGM/FGM. Various dietary modifications were reported, including revised dietary instructions [23, 31], changes in CHO sources (type and/or amount) and meal intervals [30, 32], substitution in the type of nocturnal source of CHO [15], the use of MCS vs. UCCS [15, 26, 28] or with three different meals iso-CHO before bedtime [27]. Only a few studies evaluated aspects like sleep quality or exercise capacity [28, 32].

In GSD individuals, dietary interventions generally led to metabolic improvements, such as better metabolic control [23, 28, 30], reduced time spent in hypoglycemia [15, 32], decreased glycemic variability [15], less severe hypoglycemia [16], and improved stability of nocturnal glucose levels [27]. However, only one study linked dietary overtreatment to preset glycemic targets [15].

Reliability was assessed only for Rt-CGM, which was found to be a safe, effective, and reliable system for detecting asymptomatic hypoglycemia. Unfortunately, no studies evaluated the perceptions of individuals or parents regarding Rt-CGM/FGM use.

In CH, two studies reported relevant dietetic modifications during Rt-CGM: PUFA supplementation reduced glycemic variability in one study [41], while another showed that a bi-hormonal bionic pancreas (BHBP) required fewer carbohydrate interventions for hypoglycemia compared to usual care [43]. Rayannavar et al. found CGM to be an effective adjunctive tool for persistent hypoglycemia, reducing the risk of undetected episodes through alerts [36]. The concordance between Rt-CGM and capillary/blood glucose monitoring in CH was deemed acceptable. Additionally, two studies highlighted that both individuals and caregivers had a positive perception of CGM, noting its beneficial impact on quality of life [38, 42].

On the other hand, evidence for HFMDs is poor and very limited.

The devices most commonly used across the studies were Dexcom (G4, G5, G6), Medtronic iPro 2, Minimed, Enlite, and Guardian, with only a few studies using FGM (Freestyle Libre). FGM, widely used in diabetes, is cheaper than Rt-CGM but may be less reliable and accurate, with threshold-based alarms rather than predictive ones. The varied use of blinding or unblinding in studies complicates nutritional modifications and outcome assessments.

Technological advancements continue to improve RT-CGM/FGM devices, yet the lack of compatibility and connectivity between different technologies limits their combined use. Moreover, the observation periods varied greatly, likely contributing to differences in study outcomes. High MARD values, particularly in CH, may reflect decreased accuracy during hypoglycemia, a common issue in individuals with frequent hypoglycemic episodes [45]. Lack of consensus on glycemic targets or CGM metrics was found among the analyzed studies, and the use of older versions of sensors [23, 24, 27, 39] may have negatively impacted the quality of life for individuals and families, possibly deterring the adoption of newer monitoring technologies.

CGM offers multiple applications, including targeting the nutritional therapy. It can be used intermittently before clinical reviews to retrospectively analyze glycemic patterns and optimize nutritional therapy or in real-time to prevent hypoglycemia. However, the high cost and limited evidence regarding accuracy currently restricts its broader use.

Notably, no studies have reported on the reimbursement of CGM in individuals with IMDs at risk for hypoglycemia. Despite the high costs, targeted treatment aimed at improving long-term outcomes could be beneficial, as suggested by some studies in GSD subjects [13, 46, 47].

The lack of guidelines also limits the application of CGM innovations in patient care [13].

The main limitation of this review is the absence of studies comparing metabolic outcomes during dietetic interventions in individuals monitored by Rt-CGM/FGM versus capillary/blood glucose monitoring. This lack of comparison means that the metabolic improvements observed across the studies may not be directly attributable to use of these devices. Nevertheless, dietetic therapy remains crucial in subjects with IMDs at risk for hypoglycemia and can be significantly enhanced with Rt-CGM or FGM.

The scarcity of high-quality evidence can be attributed to the rarity of the diseases and the challenges in standardizing personalized nutritional therapy, which complicates the measurement of consistent and replicable outcomes. Multicenter studies could offer more insights into the use of CGM in these contexts.

In addition to its role in real-time hypoglycemia prevention, CGM should be explored for optimizing nutritional status. It could also enhance personalized nutritional guidance, aiding the management of glycemic patterns for individuals undergoing long-term dietary treatment.