Gross motor development of children with Hurler syndrome after umbilical cord blood transplantation.

Hurler syndrome is the most severe form of mucopolysaccharidosis type I, which results from an inborn error of metabolism and has an incidence of approximately 1.19 per 100,000 live births. (1,2) Children diagnosed with Hurler syndrome have a deficiency of a single enzyme, alpha-L-iduronidase, which is responsible for breaking down the glycosaminoglycans heparan sulfate and dermatan sulfate. The resulting accumulation of glycosaminoglycans throughout the body causes somatic, central nervous system, and musculoskeletal system impairments. (2)

Children diagnosed with Hurler syndrome typically have multiple clinical signs that may include retarded growth, coarse facial features, enlarged tongues, dysostosis multiplex, limitations in joint range of motion, thickening of cardiac valves, inguinal or umbilical hernias, deafness, liver and spleen enlargement, corneal clouding, and abnormal hair growth. The life expectancy for children with Hurler syndrome is typically less than 10 years, and mortality is usually a result of airway obstruction, respiratory infection, or cardiac complications. (2,3) Gross motor abilities are reported to be limited in children with Hurler syndrome who have not received medical intervention and in some who have undergone a bone marrow transplant. (4-6) Dusing et al (6) described the gross motor abilities and joint range of motion of 4 children (aged 9.5-16 months) with Hurler syndrome, 2 of whom were included in the sample for the current study. All 4 children had limitations in joint range of motion and had gross motor abilities that were below average on the Peabody Developmental Motor Scales, second edition (PDMS-2), (7) prior to any form of transplantation.

Umbilical cord blood transplantation (UCBT) is a relatively new treatment option for children with Hurler syndrome. (8) It is an alternative to bone marrow transplantation that frequently is used when a related bone marrow donor cannot be quickly identified. Umbilical cord blood transplantation utilizes stem cells from banked umbilical cord blood and requires a lower percent human leukocyte antigen match than bone marrow transplantation. Umbilical cord blood can be collected from the placenta after a baby is born and donated to an umbilical cord blood bank. This publicly banked cord blood is readily available for quick transplantation in unrelated patients. Recipients of UCBT undergo chemotherapy, but do not require radiation in preparation for transplantation. (8) Successful engraftment and survival were reported in 17 out of 20 children who received UCBT at a median age of 16 months to treat Hurler syndrome. (8) In these children, longitudinal growth velocity normalized and liver and spleen size decreased, and most children with cognitive delays gained abilities at a rate only slightly slower than their same-age peers.

Although UCBT has been documented to arrest the progression and reverse the effects of somatic impairments associated with Hurler syndrome, there is a paucity of evidence about the development of gross motor abilities in children after UCBT. (4,8) Dusing et al (4) described 2 children who had very delayed motor abilities on the Bayley Scales of Infant Development, second edition, (9) after UCBT. Both children were able to complete a higher percentage of fine motor items than gross motor items, suggesting a discrepancy between these motor domains. Hugh-Jones (5) described the psychomotor development of 5 children after bone marrow transplantation using Ruth Griffith's series of assessments. Although the author indicated that all the children "improved physically," only 2 of the 5 children had increasing scores on the locomotor subtest over time. Dusing et al (10) described the gait of 18 children, 17 of whom also were included in the current study. The children with Hurler syndrome had less mature gait at 24 and 36 months after UCBT than children of the same age who were developing typically. However, the gait of these children with Hurler syndrome after UCBT matured quickly and was similar to that of children who were developing typically by 48 months of age.

Gross motor abilities are related to a child's ability to participate in active, unstructured, and independent play, which, in turn, has been related to improved social, emotional, and cognitive development. (11) As more children with Hurler syndrome receive lifesaving treatments such as UCBT, physical therapy services will be needed to address potential gross motor difficulties and, inevitably, activity and participation limitations. Rehabilitation professionals require a more in-depth understanding of the motor abilities and deficits of children with Hurler syndrome after UCBT in order to select appropriate and valid assessment tools, to develop appropriate plans of care, and to implement efficacious therapeutic interventions. Baseline data describing the gross motor development of children with Hurler syndrome after UCBT, which differs from the natural history of Hurler syndrome, will facilitate future research on a variety of orthopedic and therapeutic intervention strategies.

The primary purpose of this study was to describe the gross motor development of children with Hurler syndrome after UCBT. A secondary purpose was to document variations in gross motor development between specific gross motor domains for children with Hurler syndrome after UCBT using the PDMS-2. Based on our previous research (4,6,10) and our clinical observations, we hypothesized that children with Hurler syndrome after UCBT would have below-average gross motor abilities but would gain new abilities in all gross motor domains at a rate similar to their peers who were developing typically.

Method Participants

Children diagnosed with Hurler syndrome who received UCBT at Duke University Medical Center were referred to the Program for Neurodevelopmental Function in Rare Disorders (NFRD Program) at the Center for the Study of Development and Learning at the University of North Carolina at Chapel Hill for an interdisciplinary assessment. The assessment team included members from neurodevelopmental pediatrics, physical therapy, speech-language therapy, audiology, psychology, and nursing. Children with Hurler syndrome were generally assessed at 3 or 6 months after UCBT and yearly thereafter. The exact frequency of assessments varied as a result of variability in length of hospitalization, residence in the local area after hospital discharge, and the family's ability to return to the local area after being discharged to their homes around the country.

The study sample comprised 21 children with Hurler syndrome with a mean age ([+ or -] SD) of 32.2 [+ or -] 16.0 months (range=8-68) at the time of their first assessment (Tab. 1). Of these 21 children, 15 were the same children included in a previous study investigating the medical outcomes of UCBT in this population. (8) Medical conditions, including hydrocephalus or shunt placement (33.3%), corneal clouding (85.7%), heating impairments (66.7%), cardiac abnormalities (42.9%), and genu valgum (81%), were documented in this sample of children. The mean standard score ([+ or -] SD) for cognition on the Mullen Scales of Early Learning (12) or the Differential Abilities Scales (13) was 70.68 [+ or -] 19.55 during the interdisciplinary assessments that occurred on the same day as the gross motor assessments.

Nineteen children were diagnosed with Hurler syndrome after clinical symptoms were present. Two children were diagnosed and received UCBT without clinical symptoms. Both children had a family history of Hurler syndrome, prompting the unusually early diagnosis. The mean age ([+ or -] SD) at the time of UCBT was 17.1 [+ or -] 8.5 months (Tab. 1). The mean time ([+ or -] SD) between UCBT and the first assessment was 16.2 [+ or -] 16.5 months (range = 2.9 - 60.9), with a mean ([+ or -] SD) of 2.6 [+ or -] 1.2 assessments per child. Seventeen (81.0%) of the children were receiving physical therapy care at the time of one or more of their assessments. Most (79.0%) of those children receiving physical therapy care were being treated at least weekly.

Each child participated in 1 to 6 assessments during the study period between September 2002 and May 2005, yielding a total of 54 completed assessments. Data from the normative sample of the PDMS-2 (7) were considered to represent the gross motor performance of children with typical development.

Procedure

As one portion of the interdisciplinary assessment with the NFRD Program, the gross motor abilities of each child less than 71 months of age were assessed by 1 of 3 physical therapists trained in the use of the PDMS-2. The physical therapists were aware of the child's diagnosis and medical history, as these assessments were completed as part of a clinical protocol.

The PDMS-2 is a norm-referenced and standardized clinical assessment tool that is frequently used by physical therapists and occupational therapists to evaluate a child's motor abilities in comparison with the normative sample. (7) The normative sample for the PDMS-2 consisted of 2,003 children who were developing typically, from 0 to 71 months of age, and from 46 states and one Canadian province. Children in the normative sample represented the US population of children less than 5 years of age in geographic region, rural or urban residence, race, sex, ethnicity, family income, parental education, and presence of disabilities. The gross motor quotient (GMQ) on the PDMS-2 is a standard score with a mean of 100 and a standard deviation of 15. (7)

The PDMS-2 has 4 subtests that measure gross motor abilities: reflex, stationary, locomotion, and object manipulation. The standard score for each subtest has a mean of 10 and a standard deviation of 3. The object manipulation subtest is administered only to children who are 12 months of age and older. As a result, the study sample for the object manipulation subtest included only 52 observations, because one child was too young to complete this subtest at 2 of her visits. Based on testing guidelines and the children's ages, only 2 assessments included the reflex subtest. Consequently, comparison between the reflex subtest and other subtests was not included in the analyses; however, the reflex subtest was used to calculate the GMQ when required.

To examine interrater reliability on the PDMS-2, each of the 3 physical therapists assessing children with the NFRD Program reviewed and scored videotaped assessments of 4 children (3 with mucopolysaccharidosis disorders and 1 who was developing typically). Interrater reliability of the therapists' scores ranged from fair to very good (intraclass correlation coefficient [ICC(3,1)] = .74 -.98) for the GMQ, raw scores, age equivalence, and all standard scores with the exception of the stationary subtest standard score, which had an ICC of .63. The lower reliability of the stationary subtest standard score may have been caused by the raters reporting slight differences in the number of seconds that 3 of the children could stand on one foot, a task that greatly affects a child's standard score on this subtest. The first author (SCD) reviewed the scoring of the PDMS-2 for each assessment and used the Peabody Developmental Motor Scales scoring and reporting system (PDMS-2 Software Kit, version 1.2) * to calculate all scores.

Data Analysis

Hierarchical linear models (HLMs) were fit using SAS Proc Mixed (version 9.01) ([dagger]) to analyze the development of gross motor abilities. Hierarchical linear models allow for variability in the number and timing of assessments while accounting for correlations in the data from repeated measures of individual subjects. (14)

Three separate HLMs were fit, each to answer a separate developmental question. The first HLM described both mean differences and the rate of development in gross motor abilities over time between children with and without Hurler syndrome using GMQ as the dependent variable and time between UCBT and the first assessment as the predictor. The second HLM was a multivariate model that was used to evaluate and compare longitudinal development, both with each gross motor subtest and between subtests. The outcome variables for this model were the standard scores on each of 3 gross motor subtests and included time after UCBT as both a linear and quadratic term and a categorical variable designating the gross motor subtest (stationary, locomotion, and object manipulation) as well as the 2-way interaction between these variables as predictors. A final HLM was used to verify that the children were gaining new abilities on each subtest. This HLM used the raw scores for the gross motor subtests (stationary, locomotion, and object manipulation) as the dependent variables and time between UCBT and the first assessment and age at the time of UCBT as the predictors.

Results

The children with Hurler syndrome had GMQs below that of the normative sample on the PDMS-2 between 0 and 48 months after UCBT (intercept=70.93, standard error [SE]=2.21, P<.001) (Fig. 1). There was no difference between the GMQs at 12 and 36 months after UCBT for children with Hurler syndrome (GMQ=70.93 and 70.80, respectively) (F=0.0, df = 19.2, P=.96). The rate of gross motor development was similar to that of the children in the PDMS-2 normative sample at the same age, as indicated by a stable GMQ. However, there was marked between-subject variability (slope= -0.006, SE=0.12, P=.96) (Fig. 1).

Over time, raw scores increased on all 3 subtests for the children with Hurler syndrome (Tab. 2). However, their standard scores on the stationary, locomotion, and object manipulation subtests were lower than those of the children in the PDMS-2 normative sample at the same age (Fig. 2). The changes in subtest standard scores were gradual but differed across subtests (F=10.85; df=2,131; P<.001) (Fig. 2). Shortly after the UCBT, the children were the most delayed on the locomotor subtest but gained abilities at the fastest rate on this subtest (slope= 0.04, SE=0.02, P=.09). Children with Hurler syndrome after UCBT gained skills on the object manipulation subtest at the same rate as children who were developing typically (slope=0.003, SE=0.02, P=.84). Stationary abilities improved at the slowest rate, resulting in increasing delay in this gross motor domain (slope=-0.05, SE=0.03, P=.07).

[FIGURE 1 OMITTED]

Discussion

Children with Hurler syndrome performed, on average, 28.33 points lower on the GMQs (mean GMQ= 71.67) than their same-age peers in the PDMS-2 normative sample (mean GMQ=100). These children had marked deficits in gross motor abilities, similar to children assessed prior to transplantation. (6) However, the description of gross motor development for this population was oversimplified by using the GMQ. The children with Hurler syndrome were gaining new abilities at different rates on each gross motor subtest (Fig. 2). The individual subtests of the PDMS-2 provide a more specific description of each child's abilities and should be used to describe gross motor development in this population.

The slopes of the regression lines for each of the 3 gross motor subtests were not significantly different from zero, which implies that these children were gaining skills at a similar rate as children who are developing typically. However, multivariate analysis revealed differences in rates of development. Children with Hurler syndrome had the least difficulty with stationary balance tasks such as prone positioning and sitting while manipulating objects shortly after UCBT. However, as stationary balance tasks became increasingly demanding, requiring standing, standing on one foot, or standing with a smaller base of support, these children had increasing difficulty, as demonstrated by a decreasing standard score on the stationary subtest of the PDMS-2.

[FIGURE 2 OMITTED]

Locomotor abilities such as transitioning ha and out of sitting, creeping, and walking were the most delayed after UCBT. However, locomotor abilities increased at the fastest rate, reducing the degree of delay ha the 48 months after UCBT. Increased practice with independent mobility after hospital discharge, improved strength, reduced number of medications, and, ha some cases, the initiation of therapy may have contributed to faster improvements ha locomotor abilities than on other subtests. Dusing et al (10) reported that gait speed and step length were delayed at 2 years of age ha children with Hurler syndrome who received UCBT but were similar to those of age-matched peers by 4 years of age. Increasing gait speed and step length may have accounted for some increase ha the locomotor scores on the PDMS-2; however, the locomotor subtest includes items such as running, jumping, and skipping as well.

The rate of gross motor progression on the object manipulation subtest was the same as that of children who were developing typically, although a delay persisted through 48 months after UCBT. The children ha this study were generally delayed ha their ability to maintain standing balance while throwing or kicking a ball. They also had difficulty throwing a ball underhand secondary to limited forearm supination.

The 2 children who received UCBT prior to the presentation of clinical symptoms had a mean GMQ that was greater than the mean GMQ of children who were symptomatic prior to UCBT (93.80 [+ or -] 8.44 and 69.41 [+ or -] 6.97, respectively). In addition, these 2 children had UBCT at a younger age (mean age [[+ or -] SD]=2.3 [+ or -] 0.4 versus 18.6 [+ or -] 7.3 months). Although the outcomes of these 2 children are promising, further research and larger samples are needed to determine whether children without clinical signs of Hurler syndrome who receive UCBT will develop gross motor delays.

The cognitive abilities of the children included ha this study were lower than those of children who were developing typically. However, the small sample size limited our ability to control for cognitive abilities in the statistical models. Anticipating these cognitive delays, we chose the PDMS-2 because the items are described using standard directions, imitation, demonstration, and--ha some cases--facilitation through environmental manipulations, allowing for administration to children of varying ages and cognitive abilities. In addition, children are provided with more than one trial to demonstrate performance on items, minimizing the effect of cognitive impairments on a child's performance.

Although the gross motor abilities of the children in this study were lower than those of children who were developing typically, their gross motor abilities were not compared with those of children with Hurler syndrome who had not received UCBT. Description of the natural history of gross motor abilities in Hurler syndrome is limited, and prospective studies cannot ethically withhold treatments that are currently available. In addition, it is unclear what affect medical procedures such as UCBT may have on gross motor abilities in children without Hurler syndrome. It is possible that some of the motor deficits identified in this study were related to the UCBT rather than to Hurler syndrome.

Previous researchers (15,16) have suggested that the rate of gross motor development is variable in children who are developing typically when assessed using percentile rank scores on the Alberta Infant Motor Scales or Peabody Developmental Motor Scales, first edition. They recommended that multiple assessments be completed to ensure accurate measurement of a child's gross motor development. However, instability in motor development has not been investigated using the PDMS-2 or using standard scores that are better suited for rigorous statistical analysis. (17) Although multiple assessments were completed on most of the children with Hurler syndrome included in the current study, some of the variability observed could be related to typical variations in developmental progression that may have been observed regardless of the child's diagnosis or course of treatment.

Clinical Implications and Future Research

Physical therapists working with children or adolescents with Hurler syndrome should be aware of the high probability for ongoing gross motor delays with variable rates of development in different gross motor domains through at least 48 months after UCBT. Regularly scheduled quantitative assessment, qualitative assessment, and clinical impressions should be used to gauge progress. The findings of this study and previous research (4) support the use of standardized assessments, including distinct gross and fine motor subtests and measuring multiple domains, to elucidate each child's motor abilities. Motor abilities documented on the subtests should be used to determine eligibility for services and plan treatment rather than relying on summary measures such as the GMQ.

Further research is needed to investigate the efficacy of various types of physical therapy, orthotic, and orthopedic interventions for this population. Additionally, research is needed to determine whether a relationship exists between gross motor abilities and other impairments frequently seen in children with Hurler syndrome such as genu valgum, hip dysplasia, and impaired strength, balance, and cognition.

Conclusion

Children with Hurler syndrome who are symptomatic prior to UCBT generally have gross motor delays through 48 months after UCBT. However, these children gain new gross motor abilities in all domains following UCBT. The rate of gross motor progression varies across gross motor domains, warranting detailed assessments rather than reliance on gross motor screening tools or summary scores such as the GMQ to report changes in gross motor abilities or qualify children for services.

Dr Dusing, Dr Thorpe, Dr Rosenberg, and Dr Mercer provided concept/idea/research design. Dr Dusing and Dr Mercer provided writing. Dr Dusing provided data collection, and Dr Dusing and Dr Poe provided data analysis. Dr Dusing and Dr Escolar provided project management. Dr Dusing, Dr Rosenberg, and Dr Escolar provided fund procurement. Dr Thorpe, Dr Rosenberg, and Dr Escolar provided institutional liaisons. Dr Escolar provided subjects and clerical support. Dr Thorpe, Dr Poe, Dr Rosenberg, Dr Mercer, and Dr Escolar provided consultation (including review of manuscript before submission). The authors thank Matthew Gribbin for technical assistance with SAS programming.

This article was a portion of Dr Dusing's doctoral dissertation.

This study was approved by the Biomedical Institutional Review Board at University of North Carolina at Chapel Hill.

A poster presentation of this work was given at the Combined Sections Meeting of the American Physical Therapy Association; February 1-5, 2006; San Diego, Calif.

This work was funded by ion for the awards to following the first author (SCD): Promotion of Doctoral Studies Award, Foundation for Physical Therapy; Smith Research Grant, Graduate School of the University of North Carolina at Chapel Hill; and Leadership Education in Neurodevelopmental and Related Disabilities Fellowship, The Maternal and Child Health Bureau, North Carolina.

This article was submitted January 10, 2007, and was accepted July 9, 2007.

DOI: 10.2522/ptj.20070017

References

(1) Poorthuis BJ, Wevers RA, Kleijer WJ, et al. The frequency of lysosomal storage diseases in the Netherlands. Hum Genet. 1999;105:151-156.

(2) Neufeld EF, Muenzer J. The mucopolysaccharidoses. In: Scriver CR, Beaudet AM, Sly WS, Valle D, eds. The Metabolic and Molecular Basis of Inherited Disease. Vol 3. 8th ed. New York, NY: McGraw-Hill; 2001:3421-3452.

(3) Muenzer J. Mucopolysaccharidoses. Adv Pediatr. 1986;33:269-302.

(4) Dusing SC, Rosenberg A, Heimenz JR, et al. Gross and fine motor skills of children with Hurler syndrome (MPS-IH) post umbilical cord blood transplant: a case series report. Pediatr Phys Ther. 2005;17: 264-267.

(5) Hugh-Jones K. Psychomotor development of children with mucopolysaccharidosis type 1-H following bone marrow transplantation. Birth Defects Orig Artic Ser. 1986;22:25-29.

(6) Dusing SC, Thorpe DE, Rosenberg A, et al. Gross motor abilities in children with Hurler syndrome. Dev Med Child Neurol. 2006;48:927-930.

(7) Folio RM, Fewell RR. Peabody Developmental Motor Scales, Second Edition: Examiner's Manual. Austin, Tex: PRO-ED Inc; 2000. 87

(8) Staba SL, Escolar ML, Poe M, et al. Cord-blood transplants from unrelated donors in patients with Hurler's syndrome. N Engl J Med. 2004;350:1960-1969.

(9) Bayley N. Bayley Scales of Infant Development: Second Edition. San Antonio, Tex: The Psychological Corporation; 1993.

(10) Dusing SC, Thorpe DE, Mercer VS, et al. Temporal and spatial gait characteristics of children with Hurler syndrome after umbilical cord blood transplantation. Phys Ther. 2007:87:978-985.

(11) Burdette HL, Whitaker RC. Resurrecting free play in young children: looking beyond fitness and fatness to attention, affiliation, and affect. Arch Pediatr Adolesc Med. 2005;159:46-50.

(12) Mullen E. Mullen Scales of Early Learning. Circle Pines, Minn: American Guidance Service Inc; 1995.

(13) Elliott C. Differential Abilities Scales. San Antonio, Tex: The Psychological Corporation; 1990.

(14) Singer J, Willett J. Applied Longitudinal Data Analysis: Modeling Change and Event Occurrence. New York, NY: Oxford University Press; 2003.

(15) Darrah J, Hodge M, Magill-Evans J, Kembhavi G. Stability of serial assessments of motor and communication abilities in typically developing infants: implications for screening. Early Hum Dev. 2003;72: 97-110.

(16) Darrah J, Redfern L, Maguire TO, et al. Intra-individual stability of rate of gross motor development in full-term infants. Early Hum Dev. 1998;52:169-179.

(17) Kamphaus R. Clinical Assessment of Children's Intelligence. Needham, Mass: Allyn & Bacon, Div of Simon & Schuster Inc; 1993:90-124.

* PRO-ED Inc, 8700 Shoal Creek Blvd, Austin, TX 78757-6897.

([dagger]) SAS Institute Inc, 100 SAS Campus Dr, Cary, NC 27513-2414.

SC Dusing, PT, PhD, is Assistant Professor, Department of Physical Therapy, School of Allied Health Professions, Virginia Commonwealth University, 1200 E Broad St, PO Box 980224, Richmond, VA 23298-0224 (USA). Address all correspondence to Dr Dusing at: [e-mail].

DE Thorpe, PT, PhD, PCS, is Associate Professor, Center for Human Movement Science, Division of Physical Therapy, University of North Carolina, Chapel Hill, NC.

MD Poe, PhD, is Investigator, Frank Porter Graham Child Development Institute, University of North Carolina.

AE Rosenberg, PT, DrPH, is Assistant Professor, Center for Human Movement Science, Division of Physical Therapy, University of North Carolina, and Training Director, Center for the Study of Development and Learning, University of North Carolina.

VS Mercer, PT, PhD, is Associate Professor, Center for Human Movement Science, Division of Physical Therapy, University of North Carolina.

ML Escolar, MD, is Director, Program for Neurodevelopmental Function in Rare Disorders, Center for the Study of Development and Learning, University of North Carolina.

[Dusing SC, Thorpe DE, Poe MD, et al. Gross motor development of children with Hurler syndrome after umbilical cord blood transplantation. Phys Ther. 2007;87:1433-1440.]

Table 1.
Subject Demographics ([bar.X] [+ or -] SD)

               No. of         Height               Weight
               Assessments    (cm)                 (kg)

Total sample   54             89.0 [+ or -] 11.2   15.1 [+ or -] 4.0
  (N = 21)
Symptomatic    49             90.0 [+ or -] 10.9   15.5 [+ or -] 4.0
  (n = 19)
Asymptomatic   5              79.5 [+ or -] 10.0   10.9 [+ or -] 1.7
  (n = 2)

               No. of         Age                  Age (mo)
               Female         (mo) at              at Time of
               Participants   Time of              First
                              UCBT (a)             Assessment

Total sample   13             17.1 [+ or -] 8.5    32.2 [+ or -] 16.0
  (N = 21)
Symptomatic    11             18.6 [+ or -] 7.3    32.6 [+ or -] 15.3
  (n = 19)
Asymptomatic   2              2-3 [+ or -] 0.4     28.5 [+ or -] 29.0
  (n = 2)
               Time (mo)
               Between UCBT
               and First
               Assessment

Total sample   16.2 [+ or -] 16.5
  (N = 21)
Symptomatic    15.1 [+ or -] 15.6
  (n = 19)
Asymptomatic   26.4 [+ or -] 28.6
  (n = 2)

(a) UCBT = umbilical cord blood transplantation.

Table 2.
Mean Raw Scores on the Subtests of the Peabody Developmental Motor
Scales, Second Edition, Derived From the Hierarchical Linear Models
at 12 and 36 Months After Umbilical Cord Blood Transplantation (UCBT)

Subtest Raw           12 mo        36 mo
Score                 After UCBT   After UCBT

Stationary            35.01        40.72
Locomotion            55.83        95.78
Object manipulation   4.17         18.90