Metabolic profile in Prader-Willi syndrome patients followed at a single expert center of rare endocrine diseases

Abstract

Introduction

Prader-Willi syndrome (PWS) is a rare imprinting disorder characterized by typical dysmorphic features, lack of satiety, infantile hypotonia, and later morbid obesity with complications, short stature, hypogonadotropic hypogonadism, skeletal and psychiatric problems. From the literature, it is well known that patients with PWS have a more favorable metabolic pattern than healthy controls.

Aim

The aim of the study is to assess the metabolic profile of PWS patients followed at an Expert Center for Rare Endocrine Diseases compared with healthy controls and to look for relations between components of the metabolic syndrome (MetS), adipokines, and the compartments of body composition (BC—lean and fat mass).

Patients and methods

The current study is a cross-sectional evaluation of 25 patients with Prader–Willi syndrome (mean age 11.3 ± 8.2 years), with a total of 183.6 patient-years of regular follow-up (from the first visit to the center to the data collection cutoff date), compared with 24 age-, sex-, and BMI-matched healthy controls (mean age 11.3 ± 3.9 years). Each participant underwent anthropometric measurements, physical examination, biochemical and hormonal blood sampling, and whole-body DXA scan. Statistical analysis (SPSS 15.0 statistical package, Chicago, IL, USA) was performed to assess the relations between the metrics in the PWS group compared with controls.

Results

Patients with PWS showed a better profile of glucose homeostasis with significantly lower serum insulin concentration and calculated HOMA-IR index compared with the controls (p < 0.05). Taking into consideration age, sex, and body mass index (BMI) in the PWS group, the analysis showed strong positive correlations between waist circumference (WC) and systolic blood pressure (SBP) (r = 0.864, p < 0.001), and WC and diastolic blood pressure (DBP) (r = 0.534, p = 0.033). Partial correlation analysis with respect to age, sex, and pubertal development found significant positive WC correlations with insulin (r = 0.796, p = 0.006), HOMA-IR (r = 0.697, p = 0.025), LDL-cholesterol (r = 0.735, p = 0.002), uric acid (r = 0.735, p = 0.002), CRP (r = 0.600, p = 0.023), and leptin (r = 0.730, p = 0.005). Strong negative correlations existed between WC and SHBG (r = -0.772, p = 0.002) and HMW adiponectin (r = -0.998, p = 0.044). Additionally, a negative correlation of HMW adiponectin and SBP was demonstrated. 88% of the patients were treated with recombinant human growth hormone (rhGH). Bone mineral density adjusted for height (BMD/height) was significantly lower in patients with PWS (p < 0.05) compared with healthy controls. The analysis did not reveal significant relationships between BC compartments and metabolic and auxological parameters in the PWS group.

Conclusion

Our study confirms that patients with PWS have a favorable metabolic profile compared with healthy controls matched by age, sex, and BMI. Syndromic participants who manifest greater accumulation of abdominal adipose tissue have a higher risk of hemodynamic changes and metabolic disturbances predictive of the development of cardiovascular diseases (CVD) in adulthood. WC could serve as a predictive marker for detecting higher metabolic risk in this syndromic group of patients, and both WC and HMW adiponectin for hypertension. In the future, on this basis, we could possibly implement both of these metrics in clinical practice.

Background

Prader-Willi syndrome is a genetic condition with a frequency from 1:8000 up to 1:30,000, which is associated with deletions of chromosome 15 (region 15q11.2), maternal uniparental disomy, or other rare imprinting defects. In the early postnatal period, it is characterized by muscle hypotonia, failure to thrive, and the necessity of tube feeding. In infancy, the patients increase their interest in food and become obese. Other characteristics are short stature and specific dysmorphic features—narrow bifrontal diameter, almond-shaped palpebral fissures, narrow nasal bridge, down-turned mouth, lighter skin than parents, thin and long fingers, and cryptorchidism. In adulthood, those patients present with hypogonadism, behavioral problems progressing into psychiatric disorders, and early metabolic complications as a consequence of severe morbid obesity.

Worldwide medical studies emphasize the better metabolic profile of patients with PWS compared with non-syndromic obese patients or with obese controls [1]. Haqq et al. [2] examined 14 children with PWS, 14 matched for sex, age, and BMI obese controls and 14 lean matched for sex and age non-obese children. The authors found that fasting insulin levels and HOMA-IR in children with PWS were lower than those in obese controls and similar to lean controls. Comparing the same three groups, obese controls had 38% lower HDL-cholesterol and twofold higher triglyceride levels than lean controls, but this pattern was not valid for PWS patients. An Argentinian study including 75 children with PWS compared with obese controls showed lower basal insulin levels and HOMA-IR in the first group [3]. A publication including 14 adult women with PWS revealed a similar pattern—patients had reduced insulin resistance and milder hypertriglyceridemia compared with obese controls, after matching for age and general obesity [4]. Talebizadeh and Butler [5] confirmed these results in a heterogeneous group of PWS patients (n = 37, 10.4–44 years). In adults with PWS, Hirsch et al. demonstrated preserved insulin sensitivity and a favourable lipid profile despite severe obesity, suggesting a distinct metabolic phenotype possibly related to altered adipokine signalling [6]. More recently, Lazzer et al. reported that adults with PWS exhibit lower insulin resistance and a more favourable metabolic profile compared with BMI-matched obese controls, despite reduced basal metabolic rate and altered body composition [7]. Marzullo et al. [8] reported significantly lower uric acid in adult PWS patients(n = 89) compared to obese controls. Most studies confirming lower insulin levels, milder insulin resistance and favorable glucose metabolism in PWS patients are summarized in an extensive review by Qian et al. [9] confirming the still unclear underlying causes.

It is well known that patients with PWS have a greater percentage of fat mass and a lower percentage of muscle mass with a peculiar distribution compared with healthy controls, with the trend valid at any age [4, 10–16]. Reus et al. [17] summarized these results in a systematic review, including the benefits of rhGH treatment for reconfiguring body composition.

Many authors are interested in the levels of adipokines (leptin and adiponectin) in patients with PWS. Positive correlations between leptin levels and BMI, visceral and subcutaneous adipose tissue were reported previously [12], as well as numerous studies finding similar leptin levels in patients with PWS and matched controls [2, 15, 18, 19]. The established correlations between leptin levels and the body composition (BC) compartments are various and further investigations are needed [14, 20].

Previous adiponectin studies proved that its level is higher in obese PWS patients than in BMI matched controls [2, 15, 21–23]. The same pattern is described for HMW adiponectin [21]. Kennedy et al. [15] and Tanaka et al. [12] agree with the statement that accumulation of excessive visceral, rather than subcutaneous adipose tissue is associated with decreased production of adiponectin. Studies confirmed that PWS patients have more subcutaneous than visceral fat mass accumulation than BMI matched controls [2, 12].

The aim of the current study is to assess the metabolic profile of young patients with PWS compared with healthy controls and to revisit relations between components of metabolic syndrome (MetS), adipokines, and the compartments of BC (lean and fat mass).

Patients and methods

The study was conducted over a two-year period (September 2021–September 2023), with consecutive patient enrolment during this timе. All patients with PWS followed at the Expert Center for Rare Endocrine Diseases, UMHAT “Sveta Marina,” Varna, were invited to participate, as were non-syndromic control subjects visiting the clinic for other reasons (e.g., short stature, obesity, regular health check, etc.). A total of 49 patients and healthy controls (27 boys, 55%) with a mean age of 11.3 ± 6.4 (0.3–33.0) years participated in the study.

Medical history was provided for all patients with PWS and subjects in the control group, and a general physical examination was performed. The frequency of some of the risk factors, which are part of the diagnostic algorithm for patients with PWS, was analyzed—48.0% of the PWS patients and 12.5% of the matched controls (p < 0.001) were born SGA. The participants with PWS were born with significantly lower birth weight than the patients from the control group (p = 0.001). Weight was measured to the nearest 0.1 kg with a digital scale SECA 861 (SECA Ltd, Hamburg, Germany). Height was measured to the nearest 1 mm with a standard stadiometer (HARPENDEN), with the subject upright and the head in the Frankfurt plane. WC was measured to the nearest 1 mm with a flexible, non-elastic tape midway between the tenth rib and the iliac crest during exhalation. The circumference measurement was made over the naked skin and the mean value of two measurements was taken for the analysis. BMI (kg/m2) was calculated by the standard formula. Overweight was defined as BMI ≥ 85th and < 95th percentile, and obesity as BMI ≥ 95th percentile, according to Cole’s international age- and sex-specific references for children aged ≥ 2 years [24] and the corresponding references for adults [25]. Infants under 2 years of age (n = 3) were not classified as overweight or obese, as BMI-based definitions are not validated for this age group, and were therefore excluded from BMI-based analyses. WC was assessed according to the Bulgarian references of WC for children and adolescents (over 6 and up to 18 years of age) [26] or the IDF criteria for participants over 18 years of age [25]. Participants under 6 years of age (n = 7) were excluded from the analysis. Central obesity was found in 62.7% of the patients with PWS (n = 14) and in 66.7% (n = 12) of their matched controls. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured in the supine position after 5 min rest, using a standard mercury sphygmomanometer with an appropriately sized cuff (method of Korotkoff). The average of two separate measurements was used. Hypertension was defined by applying the European Society of Hypertension guidelines for the management of high blood pressure in children and adolescents (2016) [27]. Tanner stage was defined in all the subjects. IDF criteria [28] for metabolic syndrome (MetS) and for dyslipidemia were used to determine reference ranges of different parameters (waist circumference, triglyceride levels, HDL cholesterol levels, blood pressure, and fasting plasma glucose). In our cohort, only one patient was under 2 years of age. We performed an analysis excluding this patient, because the natural history of the disease suggests that the onset of obesity, and the expected metabolic changes, generally occurs after the age of 2 years. This exclusion did not result in significant changes in the overall analysis or in the correlations between metabolic parameters, with a single exception (correlation between LDL-cholesterol and waist circumference lost statistical significance, described in section Result and taken into account in the Discussion). Taking in consideration the lack of impact on the overall findings, we considered to keep the full cohort in the analysis.

Blood sampling was done after an overnight fast. Venipuncture was performed for determination of blood glucose level (BGL), insulin, LDL-cholesterol, uric acid, CRP, SHBG, leptin, and HMW adiponectin. Oral glucose tolerance test was performed to the patients who gave consent. The investigated parameters included blood glucose (BG) and fasting serum insulin levels, with glucose load of 1.75 g/kg body weight, with a maximum dose of 75 g, with values recorded at 0, 30, and 120 min. Enzymatic and immunoturbidimetric methods (analyzer ADVIA 1800) were used for measurement of the biochemical parameters. Serum insulin levels were measured by electrochemiluminescence immunoassay (analyzer Roche Cobas 6000) and SHBG by chemiluminescence immunoassay (analyzer Immulite 2000). Insulin resistance was measured by homeostasis model assessment (HOMA-index), calculated as [insulin (µIU/mL) × BGL (mmol/L) / 22.5]. Leptin was measured by ELISA Sandwich assay, Tecan. Intra-assay variability 6.3–7.0%, inter-assay variability 7.88–19.21%; sensitivity < 0.25 ng/mL, reference values 0.25–100 µg/L. HMW adiponectin was measured by ELISA Sandwich assay, Fine Test. Sensitivity up to 0.469 ng/mL, reference values 0.8–7.6 µg/mL.

Dual-energy X-ray absorptiometry (DXA) is an X-ray imaging technique that uses two energy levels to differentiate between lean mass, fat mass, and bone mineral content based on differential attenuation. In our study, measurements were performed using a Lunar Prodigy DXA system (GE Healthcare), with paediatric-specific software applied to account for age-related differences in body composition. DXA was performed in 32 of the participants (14 PWS patients (age range 7.83–22.75) and 18 obese controls (age range 6-16.75)) and BC and bone mineral density (BMD) were measured. Not all patients signed informed consent for DXA, and infants and toddlers were not able to stay still for 5 min as per the study protocol.

The study was approved by the Research Ethics Committee at the Medical University of Varna (№104/01.07.2021). Written informed consent was obtained from the patients and their parents or caregivers. Statistically, the anthropometric indexes were described as means and standard deviations by sex within each age group. The relationships between the parameters were examined by age-, sex-, and Tanner stage-adjusted partial correlation analysis. The data were analyzed with the SPSS 15.0 statistical package (Chicago, IL, USA). We used a p-value < 0.05 to determine statistical significance.

Results

A total of 49 patients/healthy controls (27 boys) with a mean age of 11.3 ± 6.4 years (0.3–33.0 years) participated in this study. Twenty-five of them had genetically confirmed PWS (mean age 11.3 ± 8.2, 0.25–33 years, 64% male) and 24 were healthy subjects, matched for age, gender, and BMI (mean age 11.3 ± 3.9 years, 3–16.5, 46% male). Treatment with rhGH was applied in 88% of the patients. No significant difference for gender and age distribution was found between the participants from the two groups. Genetic testing to confirm the diagnosis of PWS was performed at a mean age of 3.9 ± 5.0 years (0.1–15.0 years). The distribution of the genetic variants is presented in Fig. 1.

Fig. 1.

Distribution of the genetic variance among PWS patients (%)

In the current study, one-fifth of patients with PWS are overweight and almost half of them presented with obesity (Fig. 2). Differences in generalized and abdominal obesity indicators (BMI and WC) did not reach statistical significance (p > 0.05), although the median values in the PWS cases were lower than those of the controls.

Fig. 2.

Prevalence of overweight and obesity in the patients’ and in the control group (%)

WC showed a correlation with SBP in all participants (r = 0.463, p = 0.004). When taking into consideration age, sex, and BMI only in the PWS group, the analysis found stronger positive correlations between WC and SBP (r = 0.864, p < 0.001) and between WC and DBP (r = 0.534, p = 0.033).

According to our study, systolic hypertension was diagnosed in 4.0% of the patients with PWS and in 29.2% of the control group. Diastolic hypertension was found in 16.0% of participants with syndromic disease and in 20.8% of matched controls.

Patients with PWS showed a better profile of glucose homeostasis with significantly lower serum insulin concentration (16.0 ± 10.9 mIU/L vs. 23.9 ± 11.9 mIU/L, p = 0.025) and calculated HOMA-IR index compared with the controls (3.2 ± 2.5 vs. 5.4 ± 3.1, p = 0.012) (Table 1, attached at the end of the document).

Table 1.

Metabolic parameters

Parameter	PWS	Control group	р-value
BGL 0 min (mmol/l)	4.8 ± 0.8	4.8 ± 0.8	NS
BGL 120 min (mmol/l)	6.9 ± 1.5	6.6 ± 1.0	NS
Insulin (mIU/L)	16.0 ± 10.9	23.9 ± 11.9	0.025
HOMA-IR	3.2 ± 2.5	5.4 ± 3.1	0.012
HbA1C (%)	5.6 ± 0.4	5.4 ± 0.4	NS
Total cholesterol (mmol/l)	4.6 ± 0.7	4.3 ± 0.8	NS
TG (mmol/l)	1.1 ± 0.5	1.1 ± 0.6	NS
HDL-С (mmol/l)	1.3 ± 0.3	1.2 ± 0.2	NS
LDL-С (mmol/l)	2.8 ± 0.7	2.6 ± 0.7	NS
Uric acid (µmol/l)	279.6 ± 112.8	344.0 ± 71.4	0.030
CRP (mg/l)	1.5 ± 1.6	4.4 ± 3.7	0.020

Eight participants (16.3%) with fasting hyperglycemia (BGL ≥ 5.6 mmol/L) were found—four with PWS (16.0%) and four from the control group (16.7%) (p > 0.05). No significant difference in the mean BGL between the two groups (4.8 ± 0.8 vs. 4.8 ± 0.8 mmol/L, p > 0.05) was found (Table 1).

Of the patients who gave consent (15 PWS patients and 21 of the control group) to standard oral glucose tolerance test (OGTT), 16.7% (6/36 participants) presented with impaired glucose tolerance (IGT)—BGL at 120 min 7.8–11.1 mmol/L. Although not significantly different, patients with PWS and IGT demonstrated lower levels of preprandial BGL, serum insulin, and calculated HOMA-IR index.

Six participants (12.2%) had elevated TGs—12.0% (n = 3) from the PWS patients and 12.5% (n = 3) from the controls. Although no significant difference was found in their distribution of sex, age, and adiposity indicators, PWS patients with elevated TGs demonstrated significantly higher levels of cardioprotective HDL-cholesterol compared with the corresponding control participants (p = 0.041). Nine participants (18.4%) presented with HDL-cholesterol below 1.03/1.1 mmol/L (according to age)—12.0% from the patients with PWS and 25.0% from the controls, which is statistically significant (p = 0.041) (Table 1).

The correlations between the adiposity indicators, WC in particular, and other components of MetS were studied in detail among the PWS patients, after taking into consideration the influence of gender, age, and the stage of pubertal development. According to partial correlation analysis, positive correlations were established between WC and insulin (r = 0.796, p = 0.006), HOMA-IR (r = 0.697, p = 0.025), LDL-cholesterol (r = 0.735, p = 0.002), uric acid (r = 0.735, p = 0.002), CRP (r = 0.600, p = 0.023), and leptin (r = 0.730, p = 0.005). Strong negative correlations between WC and SHBG (r = -0.772, p = 0.002) and HMW adiponectin (r = -0.998, p = 0.044) were also found. After exclusion of the patient under 2 years of age from the analysis, the correlation between LDL-cholesterol and waist circumference lost statistical significance. Patients with PWS presented with significantly lower levels of CRP and uric acid, and significantly higher levels of SHBG and adiponectin. Adiponectin demonstrated a significant negative correlation with fasting BGL and SBP (Table 1).

The participants with PWS had an overall history of 147.4 rhGH treatment years, while none of the control subjects was ever rhGH treated. Despite the rhGH treatment, bone mineral density adjusted for height (BMD/height) was significantly lower in patients with PWS (p < 0.001) compared with healthy controls. They exhibited lower content of lean body mass (LBM) adjusted for height (LBM/height) (p < 0.05). Partial correlation analysis in the PWS group, after controlling for sex, age, and stage of pubertal development, demonstrated that LBM/height correlated significantly with BMD z-score (r = 0.816, p = 0.025), while fat mass (FM) and %FM, as expected, were significantly associated with leptin levels (r = 0.985, p = 0.002 and r = 0.965, p = 0.008, respectively), but with no other metabolic risk markers.

The analysis did not reveal further significant relationships between BC compartments and metabolic and auxological parameters alone in the PWS group or in comparison with controls.

Discussion

This is the first study in Bulgaria describing metabolic profile and body composition in patients with molecularly confirmed PWS compared with age-, gender-, and partially BMI-matched controls. Our results confirm the favourable glucose and lipid metabolism in the PWS group as well as their peculiar body composition and found some correlations that could be useful in clinical practice.

The genetic variants of the PWS group showed the highest percentage of deletions of the 15th chromosome (60%), followed by mUPD15 (28%). Similar genetic patterns were reported in the literature from recent studies [29]. In most publications before 2015, the percentage of deletions was considered to be close to 75%, possibly due to the more limited opportunities for extensive genetic analysis to demonstrate uniparental disomy or the even rarer variants that require targeted analysis [30].

The fact that 20% of the PWS patients are overweight and almost half are obese (Fig. 2) demonstrates that, despite their systematic follow-up at an ECRED, controlling weight is still difficult. Meade et al. [31] studied a small group of patients’ parents (n = 44) and concluded that PWS significantly affects the quality of life of both the syndromic children and their parents. They found that, according to parents, obesity and aging have a negative impact not only on their child’s psychosocial health and social functioning, but also significantly affect family functioning and relationships between members.

The current study found positive correlations between WC and SBP, and between WC and DBP. Therefore, greater accumulation of abdominal adipose tissue in our group of patients with syndromic obesity is associated with hemodynamic changes that are an independent predictor of CVD in adulthood. We could conclude that enlargement of WC could serve as a predictive marker for hypertension in these patients.

The analysis of the data showed significantly lower values of SBP and DBP in the patients with PWS compared with their matched controls (p < 0.05), while the difference in heart rate did not reach significance. These results are confirmatory to those described in the literature so far. Brambilla et al. [32] compared obese and non-obese children with PWS (n = 109) to obese controls and found significantly lower SBP and DBP in non-obese PWS than obese PWS and obese non-syndromic controls. The authors concluded that BMI is a predictor of metabolic complications in both patient groups. They determined the incidence of arterial hypertension (BP ≥ 95th percentile for the corresponding age, sex, and height), which was significantly lower in non-obese children with PWS (12%) compared with 32% and 35%, respectively, for the other two groups. The lower incidence of hypertension in these studies confirms the more favourable CV profile in young PWS patients. In young adults with PWS (approximately 18–30 years old), this trend appears to reverse. Data from a Korean PWS cohort (n = 68) demonstrated a significantly higher prevalence of hypertension in young adults with PWS compared with age-matched controls (30.8% vs. 16.1%) [33]. This finding suggests that the more favourable cardiovascular profile observed in childhood may not persist into adulthood, limiting the applicability of this protective pattern beyond the pediatric age range.

As shown above, patients with PWS showed a better profile of glucose homeostasis with significantly lower serum insulin concentration and calculated HOMA-IR index compared with the controls. So far, only several studies have investigated glucose metabolism in PWS in children [2, 3], in a mixed group [5], as well as in adult patients [4]. The current study results are in agreement.

In our group, four PWS patients (16%) presented with fasting hyperglycemia (BGL ≥ 5.6 mmol/L). Contrary to our observation, Brambilla et al. [32] compared patients with PWS with normal weight, PWS patients with obesity, and obese controls (n = 96) and reported lower levels of preprandial blood glucose in all the syndromic patients (n = 109). The reason for the lack of difference in our study could be non-compliance with standard preparation the evening before performing the blood tests, as well as the age diversity of the group, including patients over 18 years of age. According to our results, patients with PWS and IGT demonstrated lower levels of preprandial BGL, serum insulin, and calculated HOMA-IR index, although no significant difference was found between the two groups. These results are confirmatory to the reports mentioned above [2, 3, 32].

As in our patients with elevated TG, higher levels of cardioprotective HDL-cholesterol were also reported by Haqq et al. [2] in 2011. 12% of the patients with PWS from our study group presented with low levels of HDL-cholesterol. Those with low levels of “good” cholesterol were found to have significantly lower serum TG concentrations compared with the control group, despite the lack of significant difference in BMI and WC. These two outcomes concerning lipid metabolism once again confirm the more favorable metabolic pattern of the PWS population.

Partial correlations were established between WC and other components of the metabolic syndrome. The positive correlations between WC (abdominal obesity) and levels of insulin and HOMA-IR in our PWS group justify the interpretation of WC as a more appropriate marker for metabolic risk assessment [26]. The same analysis showed a significant positive correlation between WC and LDL-C. In concordance with this result are the conclusions of a retrospective Italian study [34] with 45 adult patients. The authors measured FM percentage and BMI six years before and after participation in a multidisciplinary rehabilitation program for weight loss. They found a clinically relevant reduction in LDL-C at the same time as the reported weight reduction. Similar results were reported for a mixed age group of patients (n = 37, age range 10.4 to 49 years) by Talebizadeh and Butler [5]. After excluding the patient under 2 years of age from our cohort, the correlation between LDL-cholesterol and waist circumference was no longer statistically significant. This result does not fully confirm the findings of the above-mentioned studies and may be explained by the small smple size and age heterogeneity of our group. In the current study, a positive correlation was established between WC and CRP. These results are close to those reported by Haqq et al. [2]. Publications investigating the levels of uric acid in a PWS cohort are scarce, Marzullo et al. [8] In our study, we found the above relation and, for the first time, a correlation between WC and uric acid is reported.

In the study, strong negative correlations between WC and SHBG and HMW adiponectin were found. As could be expected, the lower the BMI and the smaller the WC—reflecting lower abdominal obesity—the higher the levels of these two protective markers of metabolic status. HMW adiponectin has a significant negative correlation with fasting BGL and SBP. A Brazilian study by Cunha et al. [35] including 353 children, reported that HMW adiponectin could be used as a metabolic marker for predicting hypertension in obese non-syndromic children. A negative correlation of adiponectin was demonstrated only with HOMA-IR, insulin, and leptin in previous publications of patients with PWS, but not with SBP [2, 21]. On this basis, we could possibly offer the use of this adipokine as a predictive marker for hypertension in patients with PWS in the future and implement it in clinical practice after proper validation.

In our PWS cohort, bone mineral density adjusted for height (BMD/height) and LBM adjusted for height (LBM/height) were significantly lower in comparison with healthy controls. The PWS patients’ group was significantly shorter compared with matched controls (p = 0.017), so adjustment for height was needed. 88% of our patients were treated with rhGH. Numerous studies report the presence of reduced BMD in adults and in children with PWS, and the effect of rhGH therapy on BMD and LBM is disputable [36–38]. Despite treatment with rhGH, patients with PWS reach subnormal or low-normal muscle mass, which has been confirmed in the medical literature as well as in our patient group.

When sex, age, and pubertal development were controlled for, the analysis indicated that LBM adjusted for height was positively linked to bone mineral density and, as expected, fat mass parameters (FM and FM%) were related to leptin concentrations. This finding underscores the close interplay between muscle mass and bone metabolism, which is well recognized in both the general population and in syndromic obesity [30]. In PWS, reduced muscle mass and physical inactivity are likely to impair mechanical loading on bone, contributing to decreased bone mineral density despite rhGH therapy. These results suggest that bone health in PWS depends not only on hormonal therapy but also on improvements in lean body mass and physical activity.

Future studies should consider reporting the metabolic load–capacity index (LCI), which reflects the ratio between adipose and skeletal muscle tissue–containing compartments. Examining the associations between the LCI and metabolic parameters may provide additional insights into cardiometabolic risk beyond traditional body composition indices.

For the first time, Brambilla et al. [32] determined the frequency of the metabolic syndrome in a representative sample of patients with PWS—7.3%. The authors focused on three groups and concluded that none of the non-obese PWS patients presented with MetS, in contrast to obese PWS patients and obese non-syndromic controls. They hypothesized that obesity could play a major role in the determination of individual metabolic risk in the PWS population. Although 62.7% of our PWS patients met the criteria for central obesity based on waist circumference, previous evidence indicates that adiposity in PWS is mainly subcutaneous rather than visceral [2, 5, 12, 15, 21]. Advanced imaging studies using magnetic resonance imaging (MRI) have demonstrated that youth with PWS exhibit greater abdominal adiposity mainly due to increased subcutaneous and intermuscular adipose tissue, while the ratio of visceral to subcutaneous fat is significantly lower compared with non-PWS controls [39]. This relative predominance of subcutaneous over visceral adipose tissue may partly explain the more favorable metabolic profile observed in PWS despite increased abdominal fat accumulation. Therefore, the central fat distribution observed in our cohort likely reflects subcutaneous accumulation in the abdominal region, which may contribute to their relatively favorable metabolic profile compared with non-syndromic obese individuals.

Limitations

This study has several limitations. First, although the control group was matched for age and sex, BMI matching was only partial. This reflects the specific body composition of patients with Prader–Willi syndrome and the difficulty of finding fully BMI-matched controls in a rare disease.

Second, the sample size was relatively small, which may have limited the ability to detect weaker associations and increased the effect of individual data points on the results. A sensitivity analysis excluding the patient under 2 years of age did not change the overall findings, excluding a single correlation between waist circumference (WC) and LDL-cholesterol, which was no longer statistically significant. This suggests that this result should be interpreted with caution, as it may be influenced by age differences or by a single observation.

Third, the cross-sectional design does not allow conclusions about cause and effect. Longitudinal studies with detailed body composition analysis are needed to better understand the relationship between fat mass, lean mass, and metabolic risk in this group.

Finally, growth hormone treatment in the PWS group was heterogeneous, with differences in start time, duration, dose, and continuity. The small number of untreated patients and the variability among treated patients did not allow meaningful analysis of treatment effects or dose–response relationships.

Conclusion

The present study confirms that patients with PWS have a favorable metabolic profile compared with healthy controls matched by age, sex, and BMI. PWS participants who manifest with greater accumulation of abdominal adipose tissue have a higher risk of hemodynamic changes and metabolic disturbances that are predictive of the development of CVD in adulthood. WC and HMW adiponectin could serve as prognostic markers for hypertension and higher metabolic risk in this syndromic group of patients. These findings highlight the importance of routine metabolic and body composition assessment in the follow-up of patients with PWS and should be validated in larger, multicenter cohorts to better define predictive thresholds and clinical applicability.

Abbreviations

BGL

Blood Glucose Level

BMI

Body Mass Index

BMD

Bone Mineral Density

BP

Blood Pressure

CRP

C-Reactive Protein

CVD

Cardiovascular Diseases

DBP

Diastolic Blood Pressure

DXA

Dual-Energy X-ray Absorptiometry

ECRED

Expert Center for Rare Endocrine Diseases

ELISA

Enzyme-Linked Immunosorbent Assay

FM

Fat Mass

HDL-C

High-Density Lipoprotein Cholesterol

HMW

High Molecular Weight

HOMA-IR

Homeostatic Model Assessment for Insulin Resistance

hrGH

Human Recombinant Growth Hormone

LDL-C

Low-Density Lipoprotein Cholesterol

MetS

Metabolic Syndrome

OGTT

Oral Glucose Tolerance Test

PWS

Prader-Willi Syndrome

SBP

Systolic Blood Pressure

SDS

Standard Deviation Score (Z-Score)

SGA

Small for Gestational Age

TC

Total Cholesterol

TG

Triglycerides

UA

Uric Acid

UPD

Maternal Uniparental Disomy

WC

Waist Circumference

Author contributions

All authors participated in the conception and writing of the manuscript. The final version of the manuscript was read and approved by all authors.

Funding

This study was supported by the Varna Pediatric Endocrine Society. The funding body had no role in the design of the study, data collection, analysis, interpretation, or in writing the manuscript.

Data availability

The datasets generated and/or analysed during the current study are not publicly available due to patient privacy and ethical restrictions but are available from the corresponding author onreasonable request.

Declarations

Ethics approval

The study was approved by the Research Ethics Committee at the Medical University of Varna. (Approval No. 104/01.07.2021). Written informed consent was obtained from all patients and their parents or caregivers. The study was conducted in accordance with the ethical standards of the institutional researchcommittee and with the 1964 Declaration of Helsinki and its later amendments.

Consent to participate

Not applicable.

Consent to publish

Not applicable.

Competing interests

The authors declare no competing interests.