Creatine Deficiency Syndromes: Clinical Spectrum, Neuroimaging Features and Treatment Response

ABSTRACT

Background

Creatine deficiency syndromes (CDS) are rare inborn errors of creatine biosynthesis or transport, predominantly affecting the central nervous system. This study aimed to evaluate the clinical features, neuroimaging findings, cardiac involvement and treatment outcomes of patients with CDS, while also increasing awareness of CDS in the differential diagnosis of autism spectrum disorder and developmental delay.

Methods

Patients diagnosed with CDS and followed at the Pediatric Metabolism Departments of Çukurova University and Adana City Hospital between 2014 and 2024 were retrospectively analysed. Demographic data, age at symptom onset and diagnosis, clinical findings, laboratory results, brain magnetic resonance imaging, magnetic resonance spectroscopy, genetic analyses, cardiological evaluations and treatment outcomes were recorded.

Results

Eight patients were included: two with arginine–glycine amidinotransferase (AGAT) deficiency, four with guanidinoacetate methyltransferase (GAMT) deficiency and two with creatine transporter deficiency (CTD). Developmental and speech delay were present in all patients. Seizures were observed in six patients and were controlled with antiepileptic therapy. Behavioural disorders, including autistic features, were detected in five patients. Brain MRS revealed reduced cerebral creatine peaks in evaluated patients. Cardiac evaluations showed no abnormalities in any patient. Follow‐up MRS performed after treatment initiation in six patients demonstrated a marked increase in cerebral creatine peaks in three patients.

Conclusion

CDS should be considered in patients with unexplained neurodevelopmental delay, epilepsy and autistic features. Early diagnosis and timely treatment are associated with improved outcomes.

This case series demonstrates the diverse clinical presentations, characteristic MR spectroscopy findings and treatment responses in creatine deficiency syndromes, emphasizing the importance of early diagnosis and disease‐specific therapy.

1. Introduction

Creatine plays a critical role in cerebral energy homeostasis and is essential for neuromodulation (Almeida et al. 2006). Creatine deficiency syndromes (CDS) are inborn errors of creatine biosynthesis and transport characterized predominantly by neurological manifestations, including developmental delay, intellectual disability, speech delay, seizures, movement disorders and behavioural abnormalities such as hyperactivity, aggression and autistic behaviours. CDS are classified into two categories: creatine biosynthesis defects (arginine–glycine amidinotransferase [AGAT] deficiency and guanidinoacetate methyltransferase [GAMT] deficiency) and creatine transporter deficiency (CTD). GAMT and AGAT deficiencies exhibit autosomal recessive inheritance, whereas creatine transport deficiency has an X‐linked trait (Mercimek‐Andrews and Salomons 2022). Although AGAT deficiency, GAMT deficiency and CTD share overlapping neurological manifestations, they differ in their biochemical profiles, neuroimaging findings, inheritance patterns and treatment strategies. Evaluating these disorders together provides a comprehensive overview of cerebral CDS.

The first case of GAMT deficiency was reported in 1994, in an infant presenting with hypotonia and movement disorders (Stöckler et al. 1994). There are approximately 150 cases of GAMT deficiency, 200 cases of CTD and fewer than 20 cases of AGAT deficiency reported globally (Mercimek‐Andrews and Salomons 2022). Low plasma creatinine levels may be a warning sign of AGAT and GAMT deficiencies but are typically normal in CTD. Measurement of urine guanidinoacetate (GAA) levels is useful in the differential diagnosis of AGAT and GAMT deficiencies. AGAT deficiency is characterized by low concentrations of GAA in urine and plasma, whereas GAMT deficiency is characterized by elevated GAA concentrations. Magnetic resonance spectroscopy (MRS) is a non‐invasive imaging technique that enables in vivo assessment of brain metabolites. The area of the creatine peak is proportional to the creatine concentration within the sampled brain tissue. Therefore, a decreased or absent creatine peak is a characteristic finding in CDS. Early treatment is particularly beneficial in AGAT and GAMT deficiencies, whereas in CTD, treatment outcomes are not encouraging.

The purpose of this study was to describe the clinical presentation, radiological findings, cardiac involvement and treatment outcomes in patients with CDS, while emphasizing the importance of considering CDS in the differential diagnosis of autism spectrum disorder and developmental delay.

2. Materials and Methods

2.1. Study Design and Patients

Patients with a molecular diagnosis of CDS who were followed at the Pediatric Metabolism Department of Çukurova University between 2014 and 2024 and at the Pediatric Metabolism Department of Health Sciences University Adana City Training and Research Hospital were included in the study. This study was approved by the Non‐Interventional Clinical Studies Ethics Committee of Cukurova University Faculty of Medicine. This study was conducted in accordance with the principles of the Declaration of Helsinki. Informed consent was obtained from all the patients and their parents. Patients were referred to the paediatric metabolism clinics because of unexplained developmental delay, speech delay, intellectual disability, seizures, autism spectrum disorder or behavioural abnormalities. The initial diagnostic evaluation included detailed clinical assessment, neurological examination, biochemical evaluation, plasma ammonia measurement, tandem mass spectrometry, plasma amino acid analysis, urine organic acid analysis, brain magnetic resonance imaging (MRI) and MRS. Cerebral CDS were suspected based on the combination of compatible clinical features and reduced or absent creatine peak on MRS. The diagnosis was subsequently confirmed by molecular genetic analysis. Sociodemographic data, age at symptom onset and diagnosis, physical examination findings, neuropsychiatric evaluation, laboratory and radiological findings and genetic analysis findings were retrospectively recorded. All patients underwent electrocardiography and echocardiography to assess potential cardiac involvement associated with CDS. Developmental status was assessed using age‐appropriate standardized tests, including the Denver Developmental Screening Test II (DDST‐II) or Wechsler Intelligence Scale for Children—Fourth Edition (WISC‐IV), whenever applicable. In three patients, standardized developmental assessment could not be performed because of severe behavioural impairment or poor cooperation; therefore, developmental status was evaluated based on clinical assessment, developmental history and neurological examination. Similar challenges have been reported in patients with CTD (Miller et al. 2026). Speech delay was assessed through clinical evaluation, developmental history documented in the medical records and language domain performance on the DDST‐II.

2.2. DNA Extraction and Variant Analysis

Two millilitres of peripheral blood samples was collected from all patients and were isolated using an automated DNA isolation system (QIAcube, Qiagen, Germany) and the Qiagen DNA Blood Mini Kit (Qiagen, Germany). Concentrations were measured using a fluorometer (Qubit 3.0). All samples were subjected to fragmentation and sample‐specific molecular barcodes were added by adapter ligation. Barcoded samples were enriched for both genes (all exons and exon‐intron junctions) using a custom‐designed kit containing gene‐specific primers. Amplicons were labelled with a universal adapter prior to next‐generation sequencing, and sample libraries were generated. Next‐generation sequencing was performed using the Illumina MiSeq platform with a minimum coverage of 100×. Quality control and bioinformatic analyses were performed using QCI‐Analyse and QCI‐Interpret.

2.3. Radiological Evaluation

Brain MRI and MRS examinations were performed using a 3.0‐T MRI (Architect GE 3.0 T; GE Healthcare, USA). All patients were sedated with chloral hydrate. Brain MRI scans were performed to rule out any brain lesions. MRS was performed using a two‐dimensional multivoxel point‐resolved spectroscopy (PRESS) sequence. Acquisition parameters included a repetition time (TR) of 2000 ms, an echo time (TE) of 288 ms, an acquisition voxel size of 10 × 10 × 15 mm, a reconstructed voxel size of 9.58 × 9.58 × 15 mm and an acquisition matrix of 23 × 19. Spectroscopic data were acquired in the transverse plane with water suppression. The volume of interest (VOI) was consistently positioned at the level of the bilateral basal ganglia on axial MR images using the same anatomical landmarks. During follow‐up examinations, voxel placement was reproduced as closely as possible to the baseline examination to ensure evaluation of the same anatomical region over time. MRS evaluation was performed by experienced radiologists, who qualitatively assessed the presence and relative prominence of the creatine peak, the principal metabolic abnormality in CDS. For patients who underwent follow‐up MRS, baseline and follow‐up spectra obtained from the same voxel location within the bilateral basal ganglia were visually compared to assess changes in the creatine peak after treatment.

2.4. Treatment and Follow‐Up

Two patients with AGAT deficiency received creatine monohydrate supplementation. All four patients with GAMT deficiency were treated with creatine monohydrate, L‐ornithine supplementation and a low‐protein diet (0.6–1.5‐g/kg/day natural protein). Both patients with CTD received combination therapy consisting of creatine monohydrate, L‐arginine and L‐glycine. In addition, Patient 6 received S‐adenosylmethionine (SAMe). Urine GAA levels could not be measured before or during the treatment because of technical limitations. Treatment outcomes were evaluated clinically and radiologically by MRS. Clinical examinations were performed monthly following the initiation of treatment and subsequently at 3‐month intervals. MRS examinations were scheduled annually after treatment initiation as part of routine clinical care. For the present study, the baseline examination was compared with the first posttreatment MRS. Posttreatment MRS data were available for six patients and were obtained approximately 1 year after treatment initiation. Changes over time were determined by comparing baseline and posttreatment spectra, focusing on changes in the creatine peak.

2.5. Statistical Analysis

Statistical analyses were performed using IBM SPSS Statistics for Windows (version 23.0; IBM Corp., Armonk, NY, USA). Continuous variables were presented as means ± standard deviation or medians (minimum [min]–maximum [max]) according to the distribution of data, and categorical variables as number and percentage (%).

3. Results

Eight patients diagnosed with CDS were included. Two patients were diagnosed with AGAT deficiency, four with GAMT deficiency and two with CTD. Seventy five percent of patients had parental consanguinity. Patients 1a and 1b and also 4a and 4b were siblings. The median age at diagnosis was 47 months (range, 24–156 months). Median current age was 72 months (range, 36–216 months). All patients had developmental delays, manifesting as delayed attainment of age‐appropriate developmental milestones in one or more developmental domains. This finding was present as a presenting symptom in three patients. The initial findings of four patients diagnosed with GAMT deficiency and one patient diagnosed with AGAT deficiency were seizures, whereas the other patient diagnosed with AGAT deficiency had self‐mutilation and autistic behaviours. Speech delay was also present in all patients, with expressive language skills below the expected level for age and delayed onset of speech compared with their peers. Six patients had seizures, predominantly of the generalized tonic–clonic type, whereas one patient presented with myoclonic seizures (Patient 1b), and one had atonic seizures (Patient 4a). Seizures in all patients were controlled with antiepileptic drugs. Five patients had behavioural disorders. Patient 1a was diagnosed with AGAT deficiency, and Patients 5 and 6, diagnosed with CTD, exhibited autistic behaviours. None of the patients had movement disorders. Patient 1a also had muscle weakness without elevated creatine kinase levels. None of the patients had significant gastrointestinal manifestations, including feeding difficulties, chronic constipation, gastroesophageal reflux or recurrent vomiting. Failure to thrive was observed in two patients with GAMT deficiency (Patients 2 and 3), both of whom had body mass index (BMI) below −2 SDS. Their weight SDS values were −3.05 and −3.10, and their height SDS values were −1.59 and −0.88, respectively. Height and weight SDS values were within the normal range in all other patients.

All patients, except Patient 5 with CTD, had low plasma creatinine levels. Seven patients underwent brain MRI and MRS, and MRS revealed a marked reduction in cerebral creatine levels in these patients. The pretreatment MRS of Patient 1b obtained from a voxel within the bilateral basal ganglia is shown in Figure 1. The electrocardiograms and echocardiograms of all patients did not reveal any abnormalities.

FIGURE 1.

Magnetic resonance spectroscopy of Patient 1b before treatment, obtained from a voxel within the bilateral basal ganglia, demonstrating a markedly reduced cerebral creatine peak.

Two patients diagnosed with AGAT deficiency had a homozygous c.1238G>A missense variant in GATM. Three patients diagnosed with GAMT deficiency had the homozygous c.327G>A variant, and one had the homozygous IVS1+1G>A novel variant in GAMT. Hemizygous variants in SLC6A8 were identified in two patients (c.570_571delTG and c.1583C>T, respectively). The clinical, laboratory, radiological, electrophysiological and molecular findings of the patients are shown in Table 1.

TABLE 1.

Clinical, laboratory, radiological, electrophysiological, molecular findings and treatment characteristics of the patients.

#	1a	1b	2	3	4a	4b	5	6
Sex	M	F	M	F	M	F	M	M
Current age (months)	216	72	36	72	54	108	162	60
Parental consanguinity	+	+	−	+	+	+	−	+
Initial findings	Autistic behaviour Self‐mutilation	Seizure	Seizure	Seizure	Seizure	Seizure	Autistic behaviour	Speech delay
Age at initial findings (months)	12	16	3	12	24	24	42	36
Age at diagnosis (months)	156	36	30	24	36	90	150	58
Developmental delay	+	+	+	+	+	+	+	+
Speech delay	+	+	+	+	+	+	+	+
Seizures type	−	Myoclonic	GTCS	GTCS	Atonic	GTCS	GTCS	−
Behavioural disorders	+	−	−	−	+	+	+	+
Autism	+	−	−	−	−	−	+	+
Myopathy	+	−	−	−	−	−	−	−
Plasma creatine (mg/dL)	0.05	0.13	0.03	0.1	0.1	0.05	0.32	0.11
Urine GAA μmol/mmol creatinine (reference range: 10–150)	NA	NA	1181.27	548	NA	NA	NA	NA
EEG	NA	Frontotemporal spike and wave discharges	Frontocentrotemporal spike and wave discharges	Multifocal epileptiform activity involving cortical and subcortical regions	Epileptiform activity in posterior hemisphere	Epileptiform activity and generalized frontal sharp slow‐wave discharges	Frontocentrotemporal spike and wave discharges	Normal
Brain MR	NA	Diffuse signal hyperintensities in corpus callosum and periventricular white matter	Bilateral globus pallidus T2/FLAIR hyperintensity	Normal	Normal	Normal	Cerebral and cerebellar atrophy	Normal
Brain MRS	NA	Decreased creatine peak	Decreased creatine peak	Decreased creatine peak	Decreased creatine peak	Decreased creatine peak	Decreased creatine peak	Decreased creatine peak
Treatment	Creatine (400 mg/kg/day)	Creatine (400 mg/kg/day)	Creatine (400 mg/kg/day) L‐ornithine (400 mg/kg/day) Protein‐restricted diet a	Creatine (400 mg/kg/day) L‐ornithine (400 mg/kg/day) Protein‐restricted diet a	Creatine (400 mg/kg/day) L‐ornithine (400 mg/kg/day) Protein‐restricted diet a	Creatine (400 mg/kg/day) L‐ornithine (400 mg/kg/day) Protein‐restricted diet a	Creatine (200 mg/kg/day) L‐arginine (400 mg/kg/day) L‐glycine (150 mg/kg/day)	Creatine (130 mg/kg/day) L‐arginine (400 mg/kg/day) L‐glycine (150 mg/kg/day) S‐adenosylmethionine (1500 mg/day)
After treatment brain MRS	NA	Markedly increased creatine peak	Decreased creatine peak	Decreased creatine peak	Markedly increased creatine peak	Markedly increased creatine peak	Decreased creatine peak	—
Pathogenic variants	Homozygous c.1238G>A variant in GATM	Homozygous c.1238G>A variant in GATM	Homozygous c.327G>A variant in GAMT	Homozygous IVS1+1G>A variant in GAMT	Homozygous c.327G>A variant in GAMT	Homozygous c.327G>A variant in GAMT	Hemizygous c.570_571delTG variant in SLC6A8	Hemizygous c.1583C>T variant in SLC6A8

Note: Novel mutations are indicated as both.

Abbreviations: EEG, electroencephalography; F, female; GAA, guanidinoacetate; GTCS, generalized tonic–clonic seizure; M, male; MR, magnetic resonance imaging; MRS, magnetic resonance spectroscopy; NA, not available.

^a Protein‐restricted diet providing approximately 0.8 g/kg/day of natural protein.

Overall, the prescribed treatments were well tolerated, and no treatment‐related adverse effects requiring discontinuation were observed. Poor treatment adherence was noted in two patients during follow‐up (Patient 2 and 3). Treatment outcomes evaluated by MRS are shown in Table 1. Six patients underwent evaluation by MRS 12 months after treatment. Posttreatment MRS could not be performed in Patients 1a and 6. A desired increase in the creatine peak on MRS was observed in three patients (Patients 1b, 4a and 4b) following treatment. The posttreatment MRS of Patient 1b obtained from the same voxel location within the bilateral basal ganglia is shown in Figure 2.

FIGURE 2.

Magnetic resonance spectroscopy of Patient 1b after treatment, obtained from the same voxel location within the bilateral basal ganglia, demonstrating an increased cerebral creatine peak. Cho: choline; Cr: creatine; NAA: N‐acetylaspartate; Lac: lactate; mI: myo‐inositol; Glx: glutamate and glutamine.

4. Discussion

CDS are classified into two categories: creatine biosynthesis defects AGAT deficiency, GAMT deficiency and CTD. AGAT deficiency is an ultrarare disorder that affects the first step in creatine synthesis, leading to decreased GAA production (Item et al. 2001). GAMT deficiency impairs the methylation of GAA to creatine, resulting in creatine deficiency and accumulation of GAA (Mulik and Mercimek‐Andrews 2023). As both AGAT and GAMT deficiencies are treatable, early recognition and screening are important to enable timely diagnosis and treatment, thereby improving clinical outcomes.

GAMT deficiency was the first creatine deficiency disorder to be described, initially reported by Stöckler et al. (1994). In our case series, GAMT deficiency was the most common. The median age at onset initial findings of diagnosed with GAMT deficiency was 17.4 ± 9.6 months, and the mean age at diagnosis was 45.6 ± 26.4 months. In the study by Yıldız et al. (2024), the median ages at presentation and diagnosis were 11.5 months (8–12 months) and 63 months (18 months–11 years), respectively. A study conducted in 2015 reported less than 20 patients worldwide diagnosed with AGAT, and the clinical and biochemical features of 16 of these patients were reported (Stockler‐Ipsiroglu et al. 2015). The median age of patients at diagnosis was 8 years. Our patients were siblings, with one being diagnosed at 156 months and the other at 36 months. We had two patients with CTD whose ages at symptom onset and at diagnosis were later than those of our other patients. In a study conducted by Curie et al. (2024), the median age at diagnosis for 31 patients with CTD was 6.5 years. These findings in the literature and our study indicate a significant diagnostic delay in patients with CDS. The prevalence of CDS has been found to be 0.25% in a cohort of patients with unknown neurological disorders (Cheillan et al. 2012). Given the diagnostic delay and relatively high frequency, CDS should be considered in cases of unexplained neurodevelopmental findings.

The main clinical features of CDS are related to neurological involvement. In our study, developmental delay was observed in all patients. In a 2014 study by Stockler‐Ipsiroglu et al. (2014), developmental delay and speech delay were reported in approximately 92% of 48 patients with GAMT deficiency. Additionally, it was remarkable that all the patients in our study also had speech delay. GAMT deficiency may lead to a more severe clinical presentation. This is believed to be due to the accumulation of GAA (Rackayova et al. 2017). Epilepsy is one of the major neurological manifestations of GAMT deficiency and has been reported in approximately two‐thirds of affected individuals (Stockler‐Ipsiroglu et al. 2014). Although no specific seizure pattern has been consistently associated with GAMT deficiency, a wide range of seizure types may occur, including febrile seizures, generalized tonic–clonic seizures, absence seizures, myoclonic seizures and drop attacks (Biagiotti et al. 2025; Yıldız et al. 2024). In our cohort, one patient presented with atonic seizures, whereas the remaining patients experienced generalized tonic–clonic seizures. Notably, seizures were well controlled with levetiracetam monotherapy, and none of the patients had refractory epilepsy. Seizures have been reported in 10% of patients with AGAT deficiency and are typically nonrecurrent seizures (Mercimek‐Andrews and Salomons 2022). One of our patients experienced a single myoclonic seizure; however, the antiepileptic medication was discontinued during follow‐up, and no further seizures were observed. Seizures occurred in 59% of male patients diagnosed with CTD. The reported types of seizures included simple or complex partial seizures, generalized tonic–clonic seizures, absence seizures and myoclonic seizures (Van de Kamp et al. 2013). Patient 5 with CTD experienced generalized tonic–clonic seizures, which were well controlled with levetiracetam.

Movement disorders, such as dystonia, chorea, hemiballismus, ataxia and spasticity, were reported in 27.5% of patients, along with behavioural disorders, such as hyperactivity, aggression or autistic behaviours (Stockler‐Ipsiroglu et al. 2014). Movement disorders were not observed in any of our patients. However, our patients with AGAT deficiency had self‐mutilation and autistic behaviours, and Patient 5 with a CTD exhibited autistic behaviours. In a study by Stockler‐Ipsiroglu et al. (2015), autistic behaviours were reported in four patients with diagnosed AGAT deficiency. Patient 1 was misdiagnosed and followed up with autism for many years; therefore, CDS should be considered in the differential diagnosis of autism spectrum disorders. However, a prospective multicenter study conducted in Canada and the United States involving 443 children diagnosed with autism spectrum disorder was unable to identify any cases of CDS (Schulze et al. 2016). Another study reported behavioural issues such as cheerful attitudes and hyperactivity (Narayan et al. 2020). Language delays have been reported in most patients, and all of our patients also had language delay (Stockler‐Ipsiroglu et al. 2015).

AGAT deficiency is the only CDS that can present with myopathy. GAA depletion is likely the primary cause of myopathy in patients with an AGAT deficiency. It has been shown that GAA could serve as an alternative substrate to creatine kinase (CK) (Fitch and Chevli 1980). In GAMT and CTD, GAA‐phosphate acts as an alternative high‐energy reservoir to creatine phosphate, thereby preventing myopathy (Schütz and Stöckler‐Ipsiroglu 2007). However, in AGAT deficiency, impaired synthesis of GAA prevents the formation of GAA‐phosphate, leading to the complete depletion of muscular high‐energy phosphates. Patient 1a had muscle weakness but was unable to cooperate with electromyography (EMG) due to behavioural disturbances. Patient 1b had normal muscle strength, suggesting that early diagnosis, treatment and physical therapy may have prevented the development of muscle weakness. There was no muscle weakness in patients with GAMT deficiency or CTD.

In addition to neurological involvement, gastrointestinal manifestations, including feeding difficulties, chronic constipation, gastroesophageal reflux, recurrent vomiting and failure to thrive, have also been reported, particularly in patients with CTD (Uçar et al. 2023; Miller et al. 2019; Li and Xu 2023). However, none of our patients with CTD had failure to thrive; this finding was observed only in two patients with GAMT deficiency. Furthermore, none of the patients had feeding difficulties, chronic constipation, gastroesophageal reflux or recurrent vomiting.

Studies in AGAT–/– and GAMT–/– mice have shown that impaired cardiac function can occur. (Faller et al. 2018; Schneider et al. 2008; Ten Hove et al. 2005). Yet, these defects may not significantly affect survival, cardiac function or left ventricular remodelling, which may suggest that creatine deficiency does not worsen contractile dysfunction (Lygate et al. 2013). Data regarding cardiac involvement in patients with creatine deficiency disorders remain very limited. In contrast to the cardiac contractility defects observed in animal models, electrocardiographic and echocardiographic evaluations of all patients in our cohort, including those with AGAT deficiency, were normal. Nevertheless, further studies involving larger cohorts and comprehensive cardiac assessments are needed to clarify the clinical significance of cardiac involvement in creatine deficiency disorders.

Measurement of GAA levels is important for the differential diagnosis of CDS. While GAMT deficiency is characterized by increased GAA levels, AGAT deficiency is characterized by decreased GAA levels. In CTD, the GAA level is normal but the urine creatine/creatinine ratio is increased. Therefore, although the three CDS subtypes share overlapping clinical manifestations, including developmental delay, epilepsy and autistic features, their distinct biochemical profiles facilitate an accurate differential diagnosis and appropriate treatment. Additionally, functional studies such as the measurement of AGAT and GAMT enzyme activities in fibroblasts and lymphoblasts, and creatine uptake assessment in cultured fibroblasts are available in highly specialized centres (Salomons et al. 2001). Urine/plasma GAA level measurements were not available at our centre.

In a study evaluating 16 patients diagnosed with AGAT deficiency worldwide, brain MRI findings were normal in 12 patients in whom MRI could be performed (Stockler‐Ipsiroglu et al. 2015). Brain MRI in patients with GAMT deficiency may be normal or reveal elevated T2 signal intensity in the basal ganglia, dorsal pons or signs of cortical atrophy (Morris et al. 2007; Khaikin et al. 2018). Stockler‐Ipsiroglu et al. (2014) reported that T2 signal hyperintensities in the globus pallidus were present in 47% (18/38) of brain MRI scans at the time of diagnosis in GAMT deficiency. In our cohort, brain MRI was normal in the majority of patients with GAMT deficiency (3/4), with bilateral globus pallidus T2/FLAIR hyperintensity observed in one patient. In our patient with AGAT deficiency, diffuse signal hyperintensities were observed in the corpus callosum and periventricular white matter. Brain MRI findings in CTD may be normal or reveal subtle abnormalities, such as T2‐weighted hyperintensities, slightly delayed myelination, mild enlargement of the ventricles and extracerebral spaces, and mild cerebral and cerebellar atrophy (Van de Kamp et al. 2013). In our cohort, cerebral and cerebellar atrophy was observed in Patient 5 with CTD. There are no pathognomonic MRI findings for CDS. Moreover, since a normal MRI in some patients does not exclude CDS, MRS should be performed even if the MRI appears normal.

In AGAT, GAMT deficiency and CTD, decreased or absent creatine peaks were observed in the brain MRS. However, it has been reported that creatine peaks may be normal to mildly decreased in affected female patients with CTD (Mulik and Mercimek‐Andrews 2023; Bruun et al. 2018). The creatine peak with MRS was decreased in all of our patients. Peak creatine levels in MRS or GAA levels are not helpful in predicting the severity of the disease phenotype in patients with mild, moderate and severe phenotypes at the time of diagnosis (Mercimek‐Mahmutoglu et al. 2014). However, change in creatine peaks over time may indicate treatment outcomes. In a study by Stockler‐Ipsiroglu et al. (2014), it was shown that in 79% of patients (30/38) with follow‐up MRS during treatment, brain creatine levels were significantly higher than the baseline levels. In our study, brain MRS could not be performed in one patient, and one patient had MRS only at diagnosis because the treatment had just been initiated. Among the six patients who underwent follow‐up MRS, three (50%) showed a marked improvement in the creatine peak after treatment. In contrast, no improvement in the creatine peak was observed in the two patients with poor treatment adherence.

Over 80 distinct variants have been identified in GAMT, and the most common variant is c.327G>A (p.Lys109Lys), which was found in three of our patients (Mulik and Mercimek‐Andrews 2023; Stockler‐Ipsiroglu et al. 2014). A novel pathogenic variant of the GAMT gene was detected in Patient 3. Homozygous missense c.1238G>A variants were detected in our patients with AGAT deficiency. Our patients with CTD had hemizygous c.570_571delTG mutation and c.1583C>Tin SLC6A8 gene.

The success of treatment was affected by the age at which it was started, and it has been shown that cognitive functions are improved in patients treated before the age of 2 years (Stockler‐Ipsiroglu et al. 2015). The standard approach for treating AGAT deficiency involves 400–800 mg/kg/day of oral creatine supplementation, which leads to cognitive improvements in certain patients and improves myopathy in all individuals (Mulik and Mercimek‐Andrews 2023). Unfortunately, patients who began treatment after 10 years of age did not show improvements in cognitive function. When treated in the first months of life, before the onset of symptoms, patients achieved normal neurodevelopment (Stockler‐Ipsiroglu et al. 2015). In patients with moderate intellectual disability who started treatment late, a positive trend was observed; however, no improvement in cognitive function was observed in those with severe ID. Two siblings with AGAT deficiency responded differently to treatment. Patient 1b, who started treatment at 18 months of age with mild intellectual disability, showed near‐normal development after 3 years of treatment. Additionally, she has remained seizure‐free without medication. In contrast, her elder brother, who had severe intellectual disability, began treatment at 156 months of age and, despite the same treatment duration, showed only a reduction in behavioural problems.

The treatment of GAMT deficiency includes creatine (400–800 mg/kg/day), L‐ornithine (400–800 mg/kg/day), natural protein or an arginine‐restricted diet with essential amino acid supplements (Fernandes‐Pires and Braissant 2022; Stockler‐Ipsiroglu et al. 2014). Reduced intake of arginine combined with ornithine supplementation is believed to decrease neurotoxic GAA levels and positively influence basal ganglia dysfunction and movement‐related disorders (Rackayova et al. 2017; Stockler‐Ipsiroglu et al. 2014). Ornithine has been demonstrated to decrease GAA production and protect the brain from damage, particularly in early childhood (Khaikin et al. 2018; Stockler‐Ipsiroglu et al. 2014). Improvement was observed in the developmental tests of patients with GAMT deficiency, and an increase in the creatine peak on MRS scans was seen in Patients 1b, 4a and 4b following treatment.

Management of CTD is not promising as that of GAMT and AGAT deficiencies. Oral creatine supplementation has been found to be ineffective, as CTD is the sole transporter responsible for facilitating creatine passage across the blood–brain barrier (Braissant 2012). Treatment with creatine (100–200 mg/kg/day, divided into three doses), L‐arginine (400 mg/kg/day, divided into three doses), and L‐glycine (150 mg/kg/day, divided into three doses) may have slowed disease progression (Mercimek‐Mahmutoglu et al. 2010; Van de Kamp et al. 2012). A systematic review of treatment outcomes in 59 patients with CTD showed that 19 patients (32.2%) responded to combined treatment. In Patient 5, no significant improvement was observed clinically or radiologically. Patient 6 could not be evaluated as treatment had only recently been initiated, and assessment will be performed during a subsequent follow‐up period.

This study has several limitations. Biochemical markers such as plasma and urine GAA levels were not available in all patients. In addition, standardized neurodevelopmental assessments could not be performed in all patients during follow‐up because of severe behavioural impairment or poor cooperation, limiting the objective evaluation of treatment‐related cognitive outcomes. MRS findings were assessed qualitatively as part of routine clinical practice, and future prospective studies incorporating quantitative MRS analyses may provide a more objective assessment of treatment response.

5. Conclusion

CDS should be suspected in patients with developmental delay, epilepsy and autistic behaviours. Early diagnosis is associated with better treatment response, which highlights the critical importance of awareness and early diagnosis in CDS, a treatable group of disorders. Cerebral MR imaging may result in non‐specific or completely normal; therefore, CDS should be considered regardless of radiological findings. On the other hand, imaging with MRS is highly beneficial in the diagnosis and follow‐up of these patients. Evaluation for cardiac involvement did not reveal any abnormalities, in contrast to findings from animal studies.

Author Contributions

E.B. and N.Ö.M. designed and managed the study. M.Y.Ç., E.K., B.K., İ.K., N.T.G. and G.G.M. collected the data. F.D.B. and D.K. contributed significantly to data analysis. Ö.K. handled image processing. E.B. wrote the manuscript. N.Ö.M. reviewed and edited the manuscript. All authors have read, edited and approved the final version of the manuscript.

Funding

The authors have nothing to report.

Ethics Statement

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of University Cukurova University (5 December 2025/161th meeting). Patients and their parents gave their approval to be enrolled in this study in an informed written consent.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.