Spectrum of Magnetic Resonance Abnormalities in Leigh Syndrome with Emphasis on Correlation of Diffusion-Weighted Imaging Findings with Clinical Presentation

Abstract

Background:

Leigh syndrome (LS) is a progressive neurodegenerative disorder of infancy/early childhood secondary to mitochondrial dysfunction. Imaging plays a pivotal role in the diagnosis of LS with certain typical magnetic resonance imaging (MRI) findings considered as a part of diagnostic criteria. We appraised various MRI findings on conventional MRI sequences and also assessed potential correlation between diffusion abnormalities and patient's clinical presentation.

Aims:

Our aim was to describe various patterns of central nervous system involvement in LS and to assess the correlation of diffusion-weighted imaging abnormalities with clinical presentation.

Settings and Design:

The design of the study was retrospective comprising 8 children with LS who had MRI between years 2014 and 2019.

Subjects and Methods:

Eight children between the age group of 4 months 8 years with LS based on clinical presentation, elevated lactate levels in CSF/Blood, and typical MRI findings were included in the study.

Results and Conclusions:

Brainstem was involved all (100%) patients while basal ganglia was affected in 5 (62.5%) children. Cerebral white matter involvement was present in 3 (37.5%) children, cerebellar in 2 (25%) children while spinal, corpus callosum, and thalamic involvement were observed in one (12.5%) patient each. Diffusion restriction was observed in 6 children, all of them presented with altered sensorium. Conventional MRI serves as an excellent tool for the diagnosis of LS in children with clinical suspicion. Acute encephalopathy frequently presents with diffusion restriction corresponding to active lesions. Hence, diffusion restriction on MRI predicts the activity of lesions in patients with LS.

INTRODUCTION

Leigh syndrome (LS), also known as subacute necrotizing encephalomyelopathy, is a form of progressive mitochondrial neurodegenerative disorder of childhood typically affecting children <2 years old. Pathologically, the disease is characterized by symmetrical spongiform lesions with a vacuolation of the neuropil and relative preservation of the neurons with capillary proliferation. There is specific involvement of brainstem and basal ganglia whilst any site in brain can be affected.

Pathologically, the disease is characterized by symmetrical spongiform lesions with a specific involvement of brainstem and basal ganglia. Whilst, any site in brain can be affected as a result of a vacuolation of the neuropil and a relative preservation of the neurons with capillary proliferation.[1] It is invariably fatal and clinically manifests as motor disturbances with regression of milestones, ophthalmoplegia, and lower cranial nerve palsies. The diagnosis of LS is based on clinical features of neurodegeneration, laboratory parameters suggestive of mitochondrial dysfunction, and symmetrical lesions on cranial magnetic resonance imaging (MRI).

Aims and objectives

• To study the MRI appearance of the brain and spinal cord in children with LS

• To establish any relationship between diffusion abnormalities in brain MRI with clinical presentation of the children.

SUBJECTS AND METHODS

The study was a retrospective analysis of children with suspected LS undergoing MRI at the Department of Radiodiagnosis at Dayanand Medical College, Ludhiana, between years 2014 and 2019. All children had a thorough clinical and neurological evaluation by an experienced pediatric neurologist and all the pertinent clinical findings as well as the laboratory parameters were recorded. All patients were followed up in outdoor patient department and telephonically.

The diagnostic criteria used were:

1. Progressive neurological disorder with motor and cognitive developmental delay

2. Signs and symptoms of brain-stem and/or basal ganglia disease

3. Increased lactate levels in blood and/or cerebrospinal fluid

4. Characteristic symmetric signal abnormalities in the basal ganglia and/or brainstem on MRI.

To avoid inclusion of non- LS children, the neurologist performed a clinical chart review to confirm that each subject fulfilled the criteria for LS.

The spectrum of MRI findings on conventional sequences as well as diffusion-weighted imaging (DWI) was also tabulated. The imaging was performed on MAGNETOM Avanto 18 Channel 1.5 Tesla MR equipment by Siemens healthcare. Protocol consisted of localizers in coronal, axial, and sagittal plane after proper positioning of the patient. The sequences in the axial plane were Turbo spin echo (SE) T2W sequence (repetition time [TR]/echo time [TE]/number of excitations n = 4050 ms/101 ms/3), SET1W sequence (TR/TE/n = 652 ms/17 ms/1), fluidattenuated inversion recovery sequence (TR/TE/n = 9000 ms/90 ms/1; inversion time, 2500 ms), and gradientecho sequence (TR/TE = 761 ms/26 ms).

Diffusion-weighted (DW) and apparent diffusion coefficient (ADC) imaging were performed using echo planar imaging sequence with TR/TE = 3500 ms/109 ms (minimum), field of view = 23 cm × 23 cm, number of excitations = 3, slice thickness = 5 mm, interslice gap = 1.5 mm, matrix size = 128 × 128. Diffusion sensitizing gradients were applied along the three orthogonal directions with diffusion sensitivity of b = 0, b = 500, and b = 1000 s/mm².

RESULTS

Clinical and laboratory findings

Demographics, clinical presentation, neurological and biochemical findings of the children have been illustrated in Table 1. All 8 children were males and between the age of 4 months and 3 years. Five children were in altered sensorium, 2 presented with developmental delay/neurological regression, and one with seizures. Moderate to profound hypotonia was present in six (75%) children while the tone was increased in two (25%) children. Systemic evaluation of these children was negative for cardiac, ophthalmic, or endocrinal involvement which could have suggested an alternate diagnosis. Four children (50%) expired shortly after their initial presentation, 2 were lost to follow-up while 2 are on follow-up with marked clinical disabilities.

Table 1.

Clinical details, neurological and biochemical findings with follow-up of patients

Case	Age at onset/sex	Clinical presentation	Neurological status at presentation	Developmental regression	Muscle tone and DTR's	Biochemical profile		Clinical follow-up
						Serum/CSF lactate mmol/L	Metabolic acidosis
1.	4 months/male sex	Fever, altered sensorium	GCS 7/13, unresponsive	Lack of social smile	Hypotonia+++ DTR's - Brisk Extensor plantars	3.5/8.0	Severe	Expired
2	6 months/male sex	Lethargy, poor feeding, altered sensorium	GCS 7/13, unresponsive	Normal till 3 months and then regression Loss of eye contact, social smile and stopped reaching for the subjects	Hypotonia+++ DTR's - Brisk Extensor plantars	3.8/6.3	Severe	Expired
3	6 months/male sex	Seizures, lethargy	GCS 10/13, alert but disoriented	Mild delay, inability to reach for objects	Hypotonia+++ DTR's - Brisk Flexor plantars	2.8/3.4	Nil	Alive at 3 years of age
4	10 months/male sex	Developmental delay	GCS 12/13, conscious and oriented	No vocalization, head holding present Decreased movements in all 4 limbs	Hypotonia+++ DTR's - Absent Extensor plantars Decreased movement in all extremities Head lag present	2.9/not done	Nil	Expired
5	1 year 6 months/male sex	Altered sensorium	GCS 7/13, unresponsive	Unstable gait, poor language skills	Hypotonia+++ DTR's-brisk Plantars mute	3.7/7.5	Severe	Lost to follow up
6	1 year 11 months/male sex	Altered sensorium, vomiting, respiratory distress	GCS 7/13, unresponsive	Global developmental delay, unstable gait, poor language skills	Hypotonia+++ DTR's - Not elicitable	4.1/not done	Severe	Alive at 3 years 4 months of age
7	3 years/male sex	Altered sensorium, abnormal Body movements	GCS 7/13, unresponsive	Regression of milestones, global development delay. delayed walking, poor language skills	Hypotonia++DTR's - Brisk Extensor plantars Tonic posturing	2.1/10.0	Nil	Lost to follow up
8	3 years/male sex	Generalized weakness, neurological regression	GCS 12/13, conscious and oriented	Global development delay, delayed walking, and poor language skills	Hypotonia+++ DTR's - Brisk Flexor plantars	2.6/3.8	Nil	Expired

Normal serum lactate levels=0.5–2.2 mmol/L, Normal CSF lactate levels=1.1–2.4 mmol/L. GCS=Glasgow coma scale, DTR's=Deep tendon reflexes, CSF=Cerebrospinal Fluid

Laboratory analysis revealed elevated serum lactate levels in seven (87.5%) children and normal levels in one (12.5%) child. The concentration of lactate in CSF was assessed in six children and all had elevated levels. In addition, four children had laboratory evidence of severe metabolic acidosis.

Magnetic resonance imaging findings

The distribution of lesions on T2W images is presented in Table 2. The lesions on T2W images were symmetrical [Figure 1]. Brainstem involvement was seen in all the children while basal ganglia was involved in 5 (62.5%) children. Cerebral white matter involvement was detected in 3 (37.5%) children, cerebellar in 2 (25%), and spinal, corpus callosum and thalamic in one (12.5%) patient each. Case 5 had a predominant white matter disease with extensive cavitation involving the white matter as well as the corpus callosum which is an unusual presentation, the brainstem involvement was restricted to substantia nigra in this patient [Figure 2]. Within the brainstem, the substantia nigra was involved in all patients followed by subthalamic involvement in 6 (75%) and periaqueductal in 5 (62.5%).

Table 2.

Localization of magnetic resonance imaging abnormalities on T2W images

Areas involved	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6	Case 7	Case 8
Basal ganglia	−	+	+	−	−	+	+	+
Thalamus	−	+	−	−	−	−	−	−
Brainstem
Subthalamic nucleus	+	+	+/−	+	−	−	+	+
Substantia nigra	+	+	+	+	+	+/−	+	+
Periaqueductal	−	+	+	+	+	+	−	−
Pons	+	−	+	−	−	−	−	−
Medulla	+	−	−	−	−	−	−	+
Cerebral white matter	−	−	−	−	+	+	−	+
Corpus callosum	−	−	−	−	+	−	−	−
Cerebellum	−	−	−	−	−	+	−	+
Spine	−	−	−	+	−	−	−	−

Figure 1.

Six-month-old child with altered sensorium (Case2): (a) Coronal T2W image reveals symmetrical hyperintense lesions involving the thalamus (white short arrow), subthalamic nuclei (white long arrows), and substantia nigra (dashed black arrows). (b and c) Diffusion-weighted images reveal restricted diffusion in substantia nigra (black dashed arrow), tegmentum and periaqueductal location (white dashed arrow), basal ganglia (short dashed black arrow) and thalamus (white short arrow). Child expired at the initial presentation

Figure 2.

Eighteen months old with encephalopathy (Case 5): (a) Axial T2W image reveals symmetrical lesions in the substantia nigra (arrows). (b) Axial T2W image reveals diffuse hyperintense signal in splenium of corpus callosum (short arrows) with symmetrical hyperintense signal involving the white matter (Black dotted arrows). (c and d) Corresponding diffusion-weighted imaging and apparent diffusion coefficient images reveal areas of restricted diffusion in white matter white dotted arrows) and corpus callosum (short arrows)

Diffusion restriction was observed in 6 (75%) children [Table 3], of which 5 were unresponsive at the time of presentation while one child was awake but disoriented. The ADC values in the areas corresponding to diffusion abnormalities were significantly lesser as compared to the reference values. DW images reveal hyperintense signal in case 4 who presented with neurological regression with normal ADC values. Case 8 had extensive involvement of basal ganglia, brainstem as well as the cerebellar white matter with cavitation and atrophy of the basal ganglia. ADC values were increased in this case suggestive of facilitated diffusion in the basal ganglia [Figure 3].

Table 3.

Signal on diffusion-weighted imaging and apparent diffusion coefficient values in patients

Case	ROI	Diffusion signal	ADC signal	ADC value	Reference value
1	Substantia nigra	Bright	Dark	576×10−6mm2/s	944×10−6mm2/s
2	Basal ganglia	Bright	Dark	610×10−6mm2/s	890×10−6mm2/s
3	Basal ganglia	Bright	Dark	518×10−6mm2/s	829×10−6mm2/s
4	Substantia nigra	Bright	Normal	868×10−6mm2/s	882×10−6mm2/s
5	White matter	Bright	Dark	413×10−6mm2/s	982×10−6mm2/s
6	Basal ganglia	Bright	Dark	557×10−6mm2/s	903×10−6mm2/s
7	Basal ganglia	Bright	Dark	589×10−6mm2/s	880×10−6mm2/s
8	Basal ganglia	Bright	Bright	1417×10−6mm2/s	935×10−6mm2/s

ADC=Apparent diffusion coefficient, ROI=Region of Interest

Figure 3.

Three years old with neural regression (Case 8): (a) Axial T2W images reveal cystic changes involving the basal ganglia (arrows) with hyperintense lesions in frontal white matter. (b and c) diffusion-weighted images reveal hyperintense signal, however, the apparent diffusion coefficient images also reveal a bright signal suggestive of facilitated diffusion

DISCUSSION

The pathophysiology of LS is mitochondrial dysfunction which results in extensive neurodegeneration pathologically characterized by spongiform necrosis, myelin degeneration, vascular proliferation, and gliosis in one or more areas of the central nervous system predominantly affecting the brainstem and basal ganglia while any site including spinal cord may be affected.[2] The dysfunction of the mitochondrial respiratory chain causes an impairment of oxidative phosphorylation and decreases adenosine triphosphate synthesis.[3] The diagnostic criteria of LS include typical symmetrical brain lesions on neuroimaging or histopathology (brainstem or cerebral deep gray nuclei), signs and symptoms consistent with mitochondrial disease and elevated lactate in the cerebrospinal fluid or brain.[4]

In our study, 75% children presented acutely, with altered sensorium and seizures. Delayed milestones and neurological regression were the presenting complaints in the remainder. Moderate to profound hypotonia was the most common motor symptom.[5] Children with acute symptoms, deteriorated and died shortly after presentation, while rest of them showed progressive disease and deterioration.[2,4,6]

Brainstem involvement was the most consistent imaging finding in our study, the substantia nigra and subthalamic nuclei were more frequently involved than lower brainstem in periaqueductal region, pons, or medulla. Our findings are in coherence with prior studies, which reported consistent involvement of the brainstem in all patients suffering from LS with SURF 1 gene mutations and COX deficiency.[7,8]

In cases with non-SURF 1 gene mutations, the involvement of the brainstem is relatively milder and less frequent. Brainstem lesions are associated with respiratory failure in patients with LS. The involvement of the brainstem has been documented by other authors also at the time of presentation or subsequently during the disease course.[1,2]

Symmetrical basal ganglia involvement with predominant affection of the putamen is one of the most characteristic lesions in Leigh disease, and, in our study, majority of the children had lesions of the basal ganglia. Putaminal lesions were detected in all cases of Leigh disease with basal ganglia involvement in our study.[9,10] In addition, we observed caudate nuclei involvement in two cases which is in coherence with other studies.[2] It has been described in literature that basal ganglia involvement is a more frequent feature in children with non-SURF 1 mutation.[8] Thalamic involvement was observed in one case and its involvement is infrequent in other studies also.[6]

We observed cerebellar involvement in 25% children which was centered around the dentate nuclei. Cerebellar involvement is a well-documented feature in LS although the frequency is much less than brainstem or basal ganglia involvement.[6,9]

Cerebral white matter was involved in 3children. Cerebral involvement is less frequent and has been found to be associated with more extensive disease at other characteristic sites.[6]

Case 5 in our study had extensive supratentorial white matter involvement with symmetrical involvement of substantia nigra. There was corpus callosum involvement as well as white matter cavitation in this child. Atypical findings like involvement of gray matter and leukodystrophy have been reported in LS, however, these findings make the diagnosis difficult.[1,2] The presence of white matter lesions with developmental regression is not considered a predominant feature of Leigh disease.[9] The presence of cysts/cavitation within abnormal white matter points toward the diagnosis of other mitochondrial disease. The concomitant presence of Leigh-like gray matter lesions and raised lactate levels suggest the diagnosis in such patients.[11]

Cord involvement is a rare feature in LS and has been reported infrequently in literature.[1] The involvement has been described in concurrence with characteristic lesions elsewhere in brain which help to frame the diagnosis. The cord involvement has been recently attributed to some mutations of the mitochondrial aspartyl-tRNA synthetase gene and is a hallmark of a specific subtype of mitochondrial disease.[12]

Abnormalities on DW imaging have been described in preclinical LS disease. According to literature, true diffusion restriction is seen in affected areas in children who present acutely either at the onset of the disease or during the disease.[6]

Similarly, we found that children with acute presentation had diffusion abnormalities on MRI. In our study, 50% children initially presented with severe acute disease and had active lesions showing diffusion restriction on MRI.

One child presented in well oriented and awake state and had few foci of diffusion restriction.

Rest of the cases had widespread involvement with features of chronic disease in the form of cerebral atrophy and cystic changes in the involved areas.

It has been documented that DWI appearance of the lesions can be subcategorized into homogenous restricted diffusion, seen in acute edematous lesions (Type A), heterogeneous restriction with a “target-sign” appearance (Type B), and homogenous free diffusion (Type C) reflecting scar tissue. Lesions with a “target-sign” appearance have been considered classical of LS and probably suggest subacute lesions mirroring a mix of biologically viable brain and necrotic tissue.[13] Target sign appearance helps to differentiate Leighs from other entities affecting the basal ganglia.

In our case series, all children with acute presentation had diffusion restriction with reduced ADC values-suggestive of active disease. There is a direct correlation between the severity of presentation and the imaging appearance on DW imaging.[14,15,16] There are few studies in literature where the presentation of disease is compared with the neuroimaging spectrum, especially ADC and DWI. Most authors opined that diffusion restriction in lesions suggests active disease, our study also confirmed this observation. This sign can help the clinician to know the extent of acute disease and to predict the course of illness and hospital stay for the patients.

Limitations

Main limitation of this study is its retrospective design, with insufficient specific data from the hospital record, such as previous admissions for acute exacerbations and time and nature of treatment taken.

Another limitation is the small number of children comprising study data.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.