Brain magnetic resonance imaging findings in patients with mucopolysaccharidosis VI

Mucopolysaccharidosis type VI (MPS VI) is a rare lysosomal storage disorder caused by the deficient activity of N‐acetylgalactosamine 4‐sulfatase. MPS VI is usually considered as not being associated with mental retardation.


Introduction
Mucopolysaccharidosis VI (MPS VI, Maroteaux-Lamy syndrome, MIM ID #253200) is a lysosomal storage disorder (LSD) characterized by the accumulation of glycosaminoglycans (GAG) in various organs and tissues due to the deficiency of the lysosomal enzyme N-acetylgalactosamine 4-sulfatase (arylsulfatase B, ARSB, EC 3.1.6.12) (Neufeld and Muenzer 2001). The incidence estimates of MPS VI range from about 1 in 43,000 to 1 in 1,500,000 live births, depending on the population and time of the study (Baehner et al 2005;Malm et al 2008;Valayannopoulos et al 2010). The disease is progressive and affects mainly the skeletal and cardiopulmonary systems, cornea, skin, liver, spleen, brain and meninges (Neufeld and Muenzer 2001). Treatment includes general supporting measures, transplant of hematopoietic cells in selected cases and enzyme replacement therapy (ERT) with Galsulfase (recombinant ARSB) (Valayannopoulos et al 2010;Giugliani et al 2007). ERT for MPS VI is associated with the reduction of levels of the urinary GAG, improvement of joint mobility (Harmatz et al 2005(Harmatz et al , 2006 and endurance of the patient, which was measured through a walk test (Harmatz et al 2005(Harmatz et al , 2006.
Regarding the neurological compromise, MPS VI patients in general have normal intelligence (Wicker et al 1991;Thorne et al 2001;Vougioukas et al 2001;Karageorgos et al 2007), but cognitive acquisition may be affected by both hearing and visual deficits as well as by other physical limitations presented by these patients (Neufeld and Muenzer 2001). We were able to find few descriptions of mental retardation in MPS VI in the literature: Vestermark et al (1987) described one patient and Azevedo et al (2004) described one patient, but IQ evaluation was not performed in any of these patients. Although mental development is usually normal, brain abnormalities such as white matter lesions and ventricular enlargement have been described in few patients ( The main objective of the present study was to describe brain MRI findings and their correlations with clinical and biochemical findings in MPS VI patients.

Patients and methods
This was an observational and prospective study approved by the local Ethics Committee. It included the 25 patients with clinical and biochemical diagnosis of MPS VI (activity of leukocyte ARSB<10 % of the normal lower limit, normal activity of other sulfatases, and increased levels of dermatan sulfate in urine) who were seen at the MPS Clinics of the Medical Genetics Service at Hospital de Clínicas Porto Alegre, Brazil (MGS/HCPA) from 2004 to 2006. MGS/ HCPA is a reference center for the diagnosis and treatment of the MPS in South America. After the informed consent was signed, for about a one-week period, the following evaluations were performed: -Clinical/physical evaluation: performed by the same physician. Information regarding the following variables were obtained: consanguinity, age of onset of symptoms, type of first symptom, age at diagnosis, present weight and weight deficit (difference between the present weight and weight at percentile 50 (p50) for the patient's age X 100 / weight at p50, based on the curves of National Center for Health Statistics -NCHS, USA), height and height deficit (difference between the present and height at p50 for the patient's age X 100 / height at p50 based on the curves of NCHS, USA). -Dosage of GAG in urine: a random urine sample was analyzed through the spectrophotometric test of 1.9dimethylmethylene blue described by Jong et al (1992). The values were analyzed in its original unit (ug/mg of creatinine) and in times over the highest normal limit for the age. When the patient was in ERT, the value obtained just before the onset of therapy was considered. -Brain MRI: Both qualitative and quantitative MRI were acquired with a 1.5 T scannner (Symphony, Siemens, Erlangen, Germany). The MRI protocol included axial slices using two pulse sequences: 1) fluid-attenuated inversion recovery (FLAIR) sequence with repetition time (TR) of 9.000 msec, echo time (TE) of 114 msec, inversion time (TI) of 2.500 msec; and 2) T2-weighted images with repetition time (TR) of 4.000 msec, echo time (TE) of 99 msec. The slice thickness was 5 mm, the field of wiew ranged from 180 to 230, and pixel size ranged from 0.45 to 0.55 cm. No anesthesia, sedation or paramagnetic agent was used. The following qualitative MRI variables were analyzed: presence and severity of white matter (WM) lesions, which were subjectively graded according to Vedolin et al (2007a) (mild0scattered and<5; moderate05; and severe0extensive, confluent, and>5); presence of hydrocephalus; presence and severity of dilatation of the perivascular spaces (DPVS), which were graded according to the numbers of DPVS (mild0<3; moderate03 to 10; and severe 0 >10); (presence of megacisterna magna); and presence of cortical atrophy. These variables were analyzed jointly at a workstation by two of the authors (L.V. and M.K), with 9 and 6 years experience in MRI posprocessing, respectively. Discrepancies were analyzed in a consensus mode by the observers. They were blinded to the age, type, and clinical status of the patients. The following quantitative variables were analyzed by a third appraiser with 5 years experience in segmentation and MRI post-processing: normalized white matter lesion load (NLL -defined as white matter lesion loads corrected for different skull sizes); normalized cerebrospinal fluid (CSF) volume (NCSFV -defined as CSF volumes corrected for different skull sizes); normalized ventricular volume (NVV -defined as ventricular volumes corrected for different ventricular sizes) and normalized cerebral volume (NCV -defined as cerebral volume corrected for different skull sizes). Data acquisition and imaging pos-processing were performed with the same methodology. Briefly, for preprocessing, segmentation and quantitative analysis, ImageJ® software (http://rsb.info.nih.gov/ij) was used. This software has both semiautomated and automated segmentation tools. Volumes were counted in voxels using the Voxel Counter plug-in of ImageJ® software. For normalization, the outer table of the skull was defined as the peripheral edge of the volume of interest. Normalized volumes, which are corrected for different skull sizes, were used for statistical analysis. -IQ evaluation: performed by the same psychologist with experience in the evaluation of MPS patients. The following tests were applied: WPPSI (Wechsler Preschool and Primary Scale of Intelligence) -for children between 4 and 6.5 years of age; WISC III (Wechsler Intelligence Scale for Children) -to evaluate children between 6 and 16 years of age; WAIS-III (Wechsler Adult Intelligence Scale) -for individuals over 16 years of age. WISC-III and WAIS-III have been validated in Brazil (Nascimento and Figueiredo 2002); WIPPSI has not been validated in Brazil yet. All the following scores were determined: total IQ, verbal IQ, performance IQ, and cognitive potential. IQ was classified according to Wechsler as follows: above average (≥120); within average (80-119); below average (70-79), and mental retardation (≤69). The parameters below were employed to quantify cognitive potential a c c o r d i n gt oW e c h s l e r: above average (≥14); within average (6-13), below average (4-5), and mental retardation (≤3).

Statistical analysis
The results were assessed using SPSS V14.0. Categorical variables were summarized using frequencies and percentages. Continuous variables with a normal distribution were summarized using mean, standard deviation (SD) and maximum and minimum values. Continuous variables with asymmetric distribution were summarized using median, interquartile range (25-75th percentile) and maximum and minimum values. The Student t-test and the Mann-Whitney U test were used for the analysis of continuous variables with normal and asymmetric distribution, respectively. Analysis of categories of variables was conducted using the χ 2 test or Fisher's exact test. Pearson's correlation coefficient (r) was used in correlations of symmetric distribution variables, and Spearman's correlation coefficient was used in the others, considering α00.05 and β080 %.

Results
The mean age at evaluation was 10.6±4.52 years (range: 4.4-23.4 years); 14 patients (56 %) were female, and only one patient had been previously shunted. At evaluation, ten patients had been receiving ERT for 69±33.6 weeks (range: 7-96 weeks). All patients had their IQ evaluated, but there were considerable difficulties since nine (36 %) patients presented severe visual deficit and two (8 %) patients also presented severe hearing deficit. The mean total IQ for the 16 patients who completed all subtests was 75.3±16.7 (range: 50-125); according to this parameter, 5/16 (31.2 %) patients presented with mental retardation. For the 25 patients, mean cognitive potential was 6.8±3.4 (range: 2-17); according to this parameter, 4/25 (16 %) patients presented with mental retardation, but if the patients with severe visual and hearing deficit are excluded, this number decreases to 2/23 (8.7 %). Four of 25 patients did not collaborate with the performance of the brain MRI. Only 2/21 (9.5 %) patients presented with normal MRI in the qualitative analysis ( Table 1). Most of the patients present WM lesions in the periventricular and deep WM of the frontal and parietal lobes. Temporal and occipital lobes, and posterior fossa, are less commonly affected. Figure 1 shows an example of white matter lesion and hydrocephalus in a MPS VI patient (10.6 yo). Figure 2 shows white matter lesions and DPVS in another MPS VI patient (23.4 yo). Quantitative analysis of the brain MRI findings are shown in Table 2. Correlations were found between: NLL and age (r00.46, p00.04); NLL and height deficit (r00.48, p00.04); NCV and age (r0−0.56, p00.01); NCV and weight deficit (r0−0.58, p00.01), and NCV and height deficit (r0 −0.55, p00.01). No correlations were found between the following: urinary GAG levels (measured in μg/mg of creatinine and in times number above the upper limit considered for the age) and the quantitative MRI variables, and between the IQ scores and the quantitative MRI variables

Discussion
In the present study, 25 MPS VI patients were evaluated through brain MRI, and the data obtained were analyzed both qualitatively and quantitatively. This is the largest series of MPS VI patients studied until now regarding these variables.
The presence of DVPS was the most frequent abnormality found, followed by WM lesions. Because intraneuronal deposition of GAG and microglial activation (gliosis) in MPS models have been reported in the literature (Kurihara et al 1992;D u r k i ne ta l1998) and may be the cause of the WM lesions, we can speculate that loss of myelin can also be a potential cause of this finding. However, no significant correlation between NLL and the patients' IQ was found. Sá et al (2006) found a weak correlation between white matter signal alterations and mental retardation in patients with MPS I, II and III. Matheus et al (2004) did not find any correlation between brain MRI and clinical manifestations in patients with MPS I and II without mental retardation. On the other hand, in a study involving 19 patients with MPS II, Vedolin et al (2007b) found more lesions in the MRI in patients with neurological compromise. We also found positive correlations both between NLL and age and between NLL and height deficit, which suggest white matter lesions are progressive in   CSF cerebral spinal fluid, NCV normalized cerebral volume,N L L normalized white matter lesion load, NCSFV normalized CSF volume, NVV normalized ventricular volume a Variables with symmetric distribution are described by mean±SD. Variables with asymmetric distribution are described by median value ±interquartile range p25-75 MPS VI disease. NCSFV and NVV were not found to be correlated with age, weight deficit or height deficit. No correlation was found between urinary GAG levels and the quantitative MRI variables, neither between the IQ scores and the quantitative MRI variables. Cerebral atrophy was observed in 47.6 % patients, and NCV was shown to be negatively correlated with age, weight deficit and height deficit, which suggest that reduction of the cerebral volume is progressive. Hydrocephalus was also common, and in all cases it was of the communicant type. Our hypothesis is that hydrocephalus associated with MPS VI is secondary to high pressure at the venous system due to craniocervical dysplasia -this fact causes impairment to the CSF flow movement from the ventricle/ interstitial tissue to the venous sinus. However, we did not evaluate the cervical disease in this study.
Regarding total IQ, about 30 % of the sample presented total IQ≤69. Considering the estimated mental retardation prevalence of 3 % in developing countries (da Rocha et al 2006;Durkin et al 1998), the present results suggest that some degree of cognitive impairment may occur in patients with MPS VI. However, it is known that total IQ is calculated based on subtests that evaluate verbal IQ and performance IQ and that the presence of physical limitations and other factors such as fatigue or cultural deprivation can affect the final result. Consequently, in our analysis we employed the cognitive potential, a measurement that reduces the probabilities of emotional factors affecting productivity during test performance. Therefore, when considering cognitive potential, we found a smaller proportion of mental retardation (16 %) in our sample, and when excluding individuals with severe visual and hearing deficits this value was reduced to 8.7 %. No correlation was found between total IQ or cognitive potential results and other variables analyzed. Patients with and without altered IQ tests were not found to differ in relation to the other variables. This may suggest that the results of IQ tests do not have relation to disease severity or to the alterations presented in the brain MRI.
In conclusion, this study showed that brain abnormalities were frequent findings in MPS VI and that there were correlations between NLL and age, as well as between NCV and age in this disease. Although MPS VI patients may present mental retardation, it is our understanding that the results of the IQ tests were influenced by cultural deprivation and by the physical limitations associated with the disease. Additional studies are required to confirm our findings, perhaps using other neurological tests and IQ evaluations.