Diagnostic criteria for the Zappella variant of Rett syndrome (the preserved speech variant)

Transcription

1 Brain & Development 31 (2009) Original article Diagnostic criteria for the Zappella variant of Rett syndrome (the preserved speech variant) A. Renieri a, *, F. Mari a, M.A. Mencarelli a, E. Scala a, F. Ariani a, I. Longo a, I. Meloni a, G. Cevenini b, G. Pini c, G. Hayek d, M. Zappella c,d a Medical Genetics, Molecular Biology Department, University of Siena, Viale Bracci, 2, Siena 53100, Italy b Department of Surgery and Bioengineering, University of Siena, Siena, Italy c Child Neuropsychiatry, Versilia Hospital, Viareggio, Italy d Child Neuropsychiatry, Azienda Ospedaliera Senese, Siena, Italy Received 7 January 2008; received in revised form 18 April 2008; accepted 22 April Abstract The preserved speech variant is the milder form of Rett syndrome: affected girls show the same stages of this condition and by the second half of the first decade are making slow progress in manual and verbal abilities. They walk without help, and may be able to make simple drawings and write a few words. Most of them can speak in sentences. Autistic behavior can often be observed. We previously described several cases in the pre-molecular era and subsequently reported a survey of 12 cases with MECP2 mutations. Seventeen new patients with the preserved speech variant and a proven MECP2 mutation have been clinically evaluated. Additional clinical data of our previously described cases are reported. These 29 preserved speech variant cases were compared with 129 classic Rett patients using a clinical severity score system including 22 different signs. There was both statistical and clinical evidence of the existence of this variant. On the basis of their abilities these girls can be distinguished as low-, intermediate- and high-functioning. Girls of the last two groups show a greater homogeneity: they speak in sentences, use their hands more easily, have normal somatic features, mild neurovegetative abnormalities, with autistic behavior in 76%, epilepsy in 30%, while girls of the first group are closer to classic Rett syndrome. The majority of patients carries either missense mutations (especially the p.r133c change) or late truncating mutations in the MECP2 gene. These results confirm the existence of this variant of Rett syndrome (Zappella variant), a clear example of progress of manual and verbal abilities, and not of a preserved speech and suggest corresponding diagnostic criteria. Ó 2008 Elsevier B.V. All rights reserved. Keywords: Rett syndrome; Preserved speech variant; Zappella variant; Diagnostic criteria; Statistical analysis; MECP2 1. Introduction In the first consensus paper on RTT [1] the symptoms of this disorder included complete alalia, although it was not until the beginning of the 1990s that the first description was provided of the preserved speech variant in 3 girls, one of whom was the sister of a non-ambulant classic RTT [2]. A limited number of other articles on * Corresponding author. Tel.: ; fax: address: renieri@unisi.it (A. Renieri). this subject were presented in the following years [3 6] until it was possible to assess the presence of a MECP2 mutation in 12 girls affected by this variant with a more precise definition of clinical symptoms and genetic characterization [7 10]. The classic form of RTT is recognized worldwide. On the contrary, making a diagnosis of preserved speech variant appears to be a more difficult task. In the last few years, girls clearly affected by the preserved speech variant have been reported with various diagnoses, from autism to mental retardation, even in the presence of MECP2 mutations [11,12]. A need was therefore felt /$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi: /j.braindev

2 A. Renieri et al. / Brain & Development 31 (2009) of establishing clear criteria for diagnosis of this variant, based on a clinical study of patients showing MECP2 mutations. In order to achieve this goal and to increase medical knowledge about this variant, we describe here 29 preserved speech variant patients with MECP2 mutations and compare their phenotype with that of 129 classic RTT cases. We propose to refer this variant by the eponym Zappella variant of RTT (Z-RTT). 2. Materials and methods 2.1. Patients collection Girls with MECP2 mutation and the ability to say at least 10 words were included in this study as Z-RTT. Twelve were previously reported while 17 are described here for the first time [7,8,10]. The majority of the girls have been seen in the Clinics of Child Neuropsychiatry and Medical Genetics of the General Hospital of Siena. They were 29 out of a total of 158 girls with RTT and a pathogenic MECP2 mutation. All patients included in the study are inserted in the Italian Rett Biobank and Database ( [13]). These patients had all been visited by one of us (MZ) and by another child neuropsychiatrist who confirmed the diagnosis and all the other information collected. The following four aspects of the disease have received particular attention: (i) Soma (acquired microcephaly and microsomia); (ii) Central Nervous System (hand washing, reduced hand skills, ataxic gait, etc.); (iii) Development and Staging system; (iv) Neurovegetative System (respiratory abnormalities, constipation, gastro-esophageal reflux, etc.). Additional attention has been focused on the possible presence of autistic features and more extensively on behavioral peculiarities in these girls. Patients and their families received genetic counseling by one of us (AR or FM) Phenotypic scoring For each patient included in the present study, a phenotypic score has been calculated using a severity score system modified from previously reported ones [14 17] (Table 1). A score of 0 (mild), 1 (intermediate) or 2 (severe) has been assigned for 22 different clinical signs. Each sign contributed to the total severity score that ranges from 0 to 44. Given the difficulty of furnishing some specific clinical features, a complete total severity score was obtained for 26/29 patients (89.6%) Statistical analysis A non-hierarchical cluster analysis was performed, using the k-means method for identifying statistically different phenotypic groups of possible clinical interest. Two separated cluster sessions were carried out choosing, respectively, two and three groups to be formed. The statistical role of each clinical feature in group distinction was also evaluated by a stepwise multivariate discriminant analysis. This allowed the importance of each phenotypic characteristic in distinguishing among patient groups of potential clinical interest to be investigated. 3. Results 3.1. Clinical features of Z-RTT Among the 158 patients we evaluated, 29 patients with an age ranging from 5 to 37 years appeared to have a milder phenotype, especially regarding language abilities (Fig. 1 and Table 2). The group of 129 patients was either unable to speak or able to say less than 10 words, while the group of 29 presented a wide range of language abilities. In 12 patients language was limited to single words (white in Table 2), while the remaining 17 girls recovered the ability to speak using phrases at the mean age of 6 years and 3 months (ranging from 4 to 13 years). Among the latter group, 7 patients were able to speak with simple phrases (light grey in Table 2), while 10 girls had more complex language abilities even if they often used third person and were echolalic (dark grey in Table 2). The overall clinical evaluation demonstrated that the group of 29 patients differed also for several other features including physical anomalies, hand abilities, epilepsy and neurovegetative system dysfunction (Table 2) Somatic features Only 8 out of 29 patients (27.5%) showed postnatal microcephaly. Among the remaining girls, 6 showed a deceleration of head growth, 13 patients had a normal head circumference and head growth, and 2 girls were macrocephalic (head circumference above the 97 centile). Regarding the height, only 9/28 patients (32%) had short stature (height below the 5 centile). Among the remaining cases, 3 patients showed tall stature. Half of the patients were of normal weight; 7/28 girls had a weight below the 3 centile (25%), while 8 patients were overweight (28.6%) (Fig. 1). Fourteen patients had kyphosis (14/27; 52%), while 15 patients showed scoliosis (15/28; 53%) but only 2 patients required surgery Central nervous system, development and staging system Nine patients had manual abilities which allowed them to make simple drawings and write simple sentences (9/29; 31%) (Fig. 1). However, their hand use was below their general abilities. Among the remaining 20 cases, only 6 patients lacked voluntary hand use. Hands stereotypies were constant in 11, intermittent in

3 210 A. Renieri et al. / Brain & Development 31 (2009) Table 1 Clinical score Head 2 Postnatal microcephaly 1 Deceleration of head growth 0 No deceleration of head growth at age of 5 Weight 2 Below 3rd percentile 1 3rd 25th percentile 0 Above 25th percentile Height 2 Below 5th percentile 1 5th 25th percentile 0 Above 25th percentile Age of regression 2 Before 18 months 1 Between 18 months and 3 years 0 After 3 years Hand stereotypy 2 Dominating or constant or intermittent 0 None Voluntary hand use 2 None 1 Reduced or poor 0 Quite good hand use Sitting 2 Never learned to sit 1 Loss of ability to sit 0 Sitting unsupported at age of 5 Walking 2 Never learned to walk 1 Loss of ability to walk 0 Walking unsupported at age of 5 Age of walk 2 Never 1 After and equal to 18 months 0 Before 18 months Speech 2 Never spoken 1 Loss of ability to speak 0 More than 10 words at age of 5 Age of increasing words 2 Never 1 After 6 years 0 Before 6 years Level of speech 2 Absent 1 Single words 0 Phrases Level of phrases 2 Absent 1 Simple phrases 0 Complex phrases Epilepsy 2 Barely or not controlled by therapy 1 Controlled by therapy 0 No epilepsy at age of 5 GI disturbances 2 Severe Breathing disorders 2 Severe Cold extremities 2 Severe Sphincter control 2 Absent 1 Partial 0 Complete Table 1 (continued) Genu valgu/pes planus 2 Severe Kyphosis 2 Severe Scoliosis 2 Severe Intellectual disability 2 Apparently profound IQ < 20 1 Apparently severe IQ: IQ >40 16 and absent in 2 girls. The majority of patients (16) were able to walk at the expected age, while 12 patients showed a delay. All patients, except one, were still able to walk independently (median age of 14 years and 6 months at assessment). Sixteen girls suffered a regression after 18 months of age (4 of them after 3 years of age), while in the remaining 13 this occurred before 18 months. In 10 patients we were able to assess their intellectual disability measuring the IQ at or above 40. More than 50% of patients (17) have never shown epilepsy. In 6 out the 12 remaining patients epilepsy was controlled by therapy. Seventeen patients showed genu valgu/pes planus (17/27; 63%) Neurovegetative system Respiratory abnormalities, constipation and gastroesophageal reflux were only present in a minority of cases (24%). Cold extremities were present in half of the cases, while the same percentage had acquired sphincter control Statistical support in favor of the existence of the Z- RTT group In order to better define the clinical differences between the Z-RTT phenotype and classic RTT we performed a cluster analysis including the majority of Z-RTT patients described in this study (26 out of 29) and 83 out of 129 classic RTT patients for whom all clinical data were available. This type of analysis groups cases who share certain properties. It identifies a set of groups which both minimize within-group variation and maximize between-group variation. A nonhierarchical cluster analysis, choosing 2 groups to be formed, generated a group of 78 and another of 31 (Fig. 2A). These two groups differed for 18 out of 22 analyzed clinical signs, exceptions being hand stereotypies, genu valgu/pes planus, kyphosis and scoliosis (p = 0.206; p = 0.88; p = and p = 0.35). The group of 31 included our Z-RTT selected patients. This group has the lower clinical score (p ranging from 0 to

4 A. Renieri et al. / Brain & Development 31 (2009) Fig. 1. Pictures of Z-RTT and classic RTT patients. First raw: three high-functioning Z-RTT girls while writing or playing with cubes (A C). (D and E) and (G and H): Two pairs of sisters with discordant phenotype. (D) Classic RTT at the age of 27 years, sister of (E), high-functioning Z-RTT at the age of 33 years. (G) Classic RTT at the age of 12, sister of (H), Z-RTT at the age of 21. Note the impressive different physical features between classic and Z-RTT girls. Z-RTT girls are overweight and are able to stand up and write, while their RTT classic sisters are non-ambulant and very slim, even if both of them are younger than their Z-RTT sisters. On the right margin two additional Z-RTT are represented (F and I). Note the overweight of (F) and the good manual abilities of (I), being able to hold a fruit juice and drink it by herself. (A, B, G, H) reprinted with permission of Wiley Liss, Inc., a subsidiary of John Wiley & Sons, Inc., from Zappella et al. [10,8]. A = #386, B = #362, C = #169, E = #896, F = #1425, H = #139, I = # ) with a mean score of 13.8 ± 5.9. The group of 78 including classic RTT has a mean score of 27.5 ± 5.3. According to this analysis a threshold of 20 divides RTT from Z-RTT. A similar non-hierarchical cluster analysis, with three groups, divided the total of 109 patients into three groups of 37, 54 and 18 cases (dotted lines in Fig. 2A). These groups differed for 20 of the 22 clinical signs analyzed, the group of 37 having the highest and the group of 18 having the lowest clinical score (p ranging from 0 to 0.036). Exceptions were genu valgu/pes planus and kyphosis (p = 0.24 and p = 0.431). In the group of 18, 16 were Z-RTT patients and belonged to intermediate and high-functioning subgroups (see below).

5 212 A. Renieri et al. / Brain & Development 31 (2009) Table 2 Clinical data of Z-RTT patients In dark grey high-functioning patients; in light grey, intermediate-functioning patients; in white low-functioning patients. The asterisk indicates: head column, patients showing macrocephaly; weight column overweight patients; and height column patients with tall stature. NA, not available; y, years; m, months MECP2 mutations Twenty-eight out of the 29 patients enrolled in this study carried a missense or late-truncating mutation in the MECP2 gene. Among these 29 mutations, 13 were missense (7 p.r133c, 2 p.r306c, 2 p.t158m, 1 p.t158a and 1 p.r106w) while 14 were late truncating (13 being C-terminal hot-spot deletions of bp, 1 p.r453x). One patient belonging to the low-functioning group carried the early truncating c.695delg, and another patient, sister of a classic RTT girl, carried an intragenic MECP2 deletion comprising exon 3 and most part of exon 4, probably to be considered as an early truncating mutation [18]. In these two latter patients we examined the X-inactivation status on blood and this result showed a random X-inactivation for both patients. Statistical analysis performed using the v 2 test, using the two phenotypic groups described above (31 = Z- RTT and 78 = RTT) versus the three kinds of mutations (late truncating, missense and early truncating), revealed a significant association (p = 0.000). Late truncating mutations were over-expressed (14) in the Z-RTT, followed by missense (11) and early truncating (6) (Fig. 2B). In the RTT group early truncating mutations were over-expressed (44) followed by missense (25) and late truncating (9) (Fig. 2B) Further clinical dissection of Z-RTT group The group of 29 Z-RTT patients could be further divided into three groups on the basis of their overall clinical features, with particular reference to their language abilities: low, intermediate and high-functioning (Table 2). In particular, among these 29 Z-RTT patients, 12 have been grouped in the low-functioning cohort, being closer to the classic RTT girls except for their ability to say a few words, 7 have been considered intermediate-functioning and 10 high-functioning, having a higher degree of language, manual abilities and better physical features. The somatic appearance of girls included in the intermediate and high-functioning groups was different from

6 A. Renieri et al. / Brain & Development 31 (2009) Fig. 2. The hypothetical model representing the clinical variability of RTT and the underlying genetic determinants. (A) Phenotypic distribution of RTT and Z-RTT. The vertical dark line separates the two groups generated by a non-hierarchical cluster analysis, choosing two groups to be formed. Vertical dotted lines separate the three groups generated by the same cluster analysis, choosing three groups to be formed. The dark grey area, on the left side of the curve, represents the classic RTT phenotype, while the light grey area represents the Z-RTT phenotype. The area with oblique grey lines represents the group of patients with a phenotype in between classic RTT and Z-RTT. This model underlines the continuous spectrum of RTT phenotype. (B) A scheme representing the higher percentage of late truncating mutations in the Z-RTT phenotype respect to the early truncating mutations. The opposite happens in classic RTT. For missense mutations there seems to be a similar percentage in classic RTT and Z-RTT. However, if we consider the p.r133c change alone, it is much more represented in Z-RTT than in classic RTT (grey area). most girls with classic RTT. In the high-functioning group only one girl out of 10 showed microcephaly, and 2 of the remaining were even macrocephalic. In the intermediate group 2/7 were microcephalic. In the low-functioning group 5/11 were microcephalic and the remaining 6 had a normal head circumference. In the high-functioning group 2/10 girls were small and 3 girls showed tall stature, with one of them above 2SD and a second at 2SD, while the third one s height was above 2SD at 8 years but at the follow up (12 years) it was at the 75 centile. Three out of 7 patients in the intermediate group and 4/10 in the low-functioning group were small. Regarding weight, the mean BMI of the girls in the intermediate and high-functioning groups was at a higher level than normal weighted people (mean BMI = 24), with 7 patients showing overweight, while the mean BMI of girls in the low-functioning group was at the lower level (mean BMI = 19.3). Slight scoliosis was present in 6/10 girls of the high-functioning group while in 2 of the remaining cases severe scoliosis, required surgical treatment. Scoliosis was also present in 9/16 patients in the other two groups. Kyphosis occurred in about half of the cases, pes planus and genu valgu in most cases, irrespective of subgroups. Stereotypic hand activities were found in all except 2 low-functioning girls and were of similar severity in the different groups. Voluntary hand use increasingly improved in some girls of the high-functioning group. They were able to walk, except in two cases: one, lowfunctioning, had never acquired this ability, while an intermediate-functioning, presently aged 30, lost the ability to walk around 15 years when she entered stage 4. In the majority of patients of the high and intermediate-functioning groups the use of language increased before 6 years, usually around 4 5 years, while it remained constantly limited to single words in most of the low-functioning girls. Epilepsy was present only in 2/10 girls (20%) in the high-functioning group while 10/19 (53%) in the other two groups were affected by epilepsy of varying severity. Respiratory abnormalities and gastro-intestinal disorders were absent in all except 1 of the 10 high-functioning patients and present in a minority of the other cases (5/18). Cold extremities occurred in 2 of the high-functioning girls and in 10 of the remaining 16 patients. Autistic behavior was evaluated with DSM IV R criteria and in some cases also with additional scales such as the Autistic Behavior Checklist in patients belonging to the intermediate and to the high-functioning groups with the exception of the low-functioning girls, excluded from this evaluation due to their low values of mental age. In the high-functioning group 8 out of 10 (80%) were clearly autistic. Two girls, one aged 9 and another 34, did not show autistic behavior on the above quoted criteria: they were able to look at other persons with an

7 214 A. Renieri et al. / Brain & Development 31 (2009) active interest, their expression and reciprocal interaction were adequate for their mental age, they were also able to share emotions and interests with others, and spoke directly with an adequate use of pronouns. Five out of seven in the intermediate-functioning group showed autistic behavior also. Considering both groups together, the number of girls with clear autistic behavior was 13 out of 17, i.e. 76%. These affected girls had a similar profile with DSM IV R: (1) a+b+c+d+ (2)a+b+c+d+(3)c+. They displayed little facial expression, reciprocal interaction was very poor and the ability to share emotions and interests usually non-existent. The use of body gestures was absent as was the ability to recall events. Phrases were often echolalic, expressed in the third person while symbolic play was absent or minimal compared to mental age. Some of these girls were affectionate to their parents and rather pleasant in their interaction giving unexpected answers or statements, others had more detached relationships. Age was another factor in their ability to interact: when aged above 30 they interacted less than when they were younger, spoke less and looked at others with more limited interest Statistical support in favor of the existence of the Z- RTT subgroups A non-hierarchical cluster analysis, choosing three groups to be formed, was performed on patients with the score = 0 in the speech field ( more than 10 words at the age of 5 ). The analysis generated three groups of 12 (group A), 8 (group B) and 6 (group C) cases. These groups differed for 10 out of 22 analyzed clinical signs: head, height, voluntary hand use, level of speech, level of phrases, epilepsy, gastrointestinal disturbances, cold extremities, sphincter control and intellectual disabilities (p ranging from 0 to 0.032). The group with the lowest score, group A, included all the high-functioning patients while the highest score group, group C, comprised most of the low-functioning patients. Finally the majority of the intermediate patients were in the group with the intermediate score, group B, with the exception of 2 that were in group A. This statistical analysis demonstrates that actually among the Z-RTT patients three different groups can be recognized and they correlate very well with the a priori clinical separation in high, intermediate and low-functioning patients (Fig. 2A). 4. Discussion The present study provides clinical and statistical evidence that among the wide clinical variability of RTT, a milder form can be recognized. In this variant only 27.5% of patients (8/29) has microcephaly, 32% (9/28) has microsomia and 25% (7/28) has low weight, characteristics which are present in almost all cases of the classic form [19,20]. It is important to note that 7% (2/29) of cases are macrocephalic, some girls (8/28, 29%) are overweight and 11% (3/28) of patients shows tall stature. This variant presents all the four stages of the disease: (1) stagnation, (2) regression, (3) pseudostationary stage and (4) late deterioration stage. However, both the regression stage and the late deterioration stage are significantly delayed. The mean age of regression is close to 24 months, with respect to months of the classic form. Among the 8 patients with an age over 18 years, 6 were still in stage 3, while classic RTT girls often move to stage 4 during school age or early adolescence [20]. Hand stereotypies are present as in the classic form but may be less intense and less evident in some periods of development. Hand use is much better than in the classic form but some degrees of dypraxia are usually present. Epilepsy is significantly less frequent in the Z- RTT patients: 41% (12/29) versus 80% of the classic form. Language is always present in contrast to the classic form characterized by absence of speech. In some cases (41%; 12/29) that are more similar to the classic form it is limited to single words, while in others simple phrases (24%; 7/29) or complex phrases (35%, 10/29) are present. It must be noted that this improvement cannot be accounted for by a differential environmental treatment. The age when these girls are examined is an important factor for diagnosis and for assessing the degree of severity: before age 5 these girls seldom show clear features of this variant which can usually only be suspected. After this age follow up is necessary to evaluate along the years the degree of improvements in language, use of hands and other abilities which can be relatively slow or more striking. Educational and rehabilitation programs were the same as in classic RTT. This was particularly evident for two couples of sisters where one was a classic RTT and the other was a Z-RTT. Gastrointestinal disturbances, breathing irregularities and cold extremities were present in a minority of cases, compared with the classic form [20]. Autistic behavior was present in 13 out of 17 (76%) including the intermediate and high-functioning groups. It must be noted that in the many years of the Siena Clinic of Child Neuropsychiatry s activity where thousands of children with Autism have been observed, all girls with MECP2 mutation fell within the diagnosis of the RTT complex, i.e. either classic RTT or variants, including Z-RTT, and it may well be that girls with MECP2 mutation reported in the literature as only Autistic were in fact cases of Z- RTT [21]. The subject is made more complex by the fact that some authors have described girls similar to those reported in this article claiming that they are not autistic: unfortunately none of the criteria usually held in the study of subjects with Autism such as DSM IV R criteria or other tests were used by these authors [22].

8 A. Renieri et al. / Brain & Development 31 (2009) The Z-RTT is due to MECP2 mutations. However, the kind of mutation differs statistically from the mutation type present in classic RTT. Only two patients carry an early truncating mutation that, in turn, is present in about 60% of classic RTT cases [13,23]. In addition, it has been noted a significant presence of late-truncating mutations followed by missense mutations, especially the p.r133c, in the Z-RTT group in comparison with classic RTT, where the trend is completely the opposite, with the majority being early truncating mutations followed by missense and late-truncating mutations (Fig. 2B). The molecular data reported in this article confirm and expand previous observations about the presence of p.r133c missense mutation and C-terminal hot-spot deletions of MECP2 in milder phenotypes [7,10,24,25]. The few exceptions may be explained by the X-inactivation status or by other modifier genes. The analysis of the X-inactivation status in blood in our two patients does not confirm the role of the X-inactivation in modulating the phenotype. However, the analysis was performed on blood and we cannot exclude that in the brain tissue a skewed X-inactivation in favor of the normal allele would be present in these patients. The clarification of the modifier genes and the underlying mechanisms involved in the Z-RTT will allow to understand the molecular and the pathophysiological aspects that differentiate classic and Z-RTT patients. It is already known that a spectrum of RTT exists: the present data suggest however that some cases show specific progress in manual and verbal abilities which is not possible in most cases of the RTT series and these cases have a specific constellation of MECP2 mutations. The clinical variability of RTT may be represented in a hypothetical model as a Gaussian curve and we can image the Z-RTT group as the extreme of this curve which can be subdivided in three parts corresponding to the three subgroups: low, intermediate and high-functioning. The low-functioning group represents a grey zone in between the classic and the Z-RTT variant form. Considering the statistical data and the clinical description reported above, patients corresponding to the intermediate and high-functioning groups are more identifiable as Z-RTT, the low-functioning group representing a link with classic RTT. The primary aim of this work is to increase medical knowledge about the Z-RTT phenotype, in order to avoid missing the diagnosis in the future. Therefore, thanks to these data we propose clinical criteria which will be of practical value in favoring the diagnosis (Table 3). Making a correct diagnosis is very important in order to assess both the risk of recurrence and prognosis. If, for example, parents receive a diagnosis of autism for their daughter they may envisage a 2 5% risk of recurrence; but if instead the diagnosis is within the RTT complex the risk of recurrence is much lower and in Table 3 Clinical criteria for Z-RTT respect to classic RTT Main criteria 1. RTT disease profile with a disease regression after 3 years (mean age of regression = 1y 9m) and a prolonged stage 3 (delayed stage 4) 2. Hand stereotypies as in RTT 3. Milder reduction of hand skills allowing quite good hand use 4. Language is recovered at a mean age of 6 years, using single words or complex phrases 5. Normal head circumference, weight and height in most patients 6. Milder intellectual disabilities (up to IQ of 50) Supportive criteria 1. Rarer epilepsy 2. Rarer neurovegetative dysfunctions: gastrointestinal disturbances, breathing irregularities, cold extremities 3. Milder scoliosis and kyphosis 4. Presence of autistic behavior 5. Presence of late truncating deletions or missense (especially p.r133c) mutations in MECP2 addition it is possible to make use of well-defined prenatal diagnostic measures when a MECP2 mutation has been identified [26 29]. A notable difference may concern also the future of the child: the development of most girls with Autism is usually stable; in contrast several girls with the Z-RTT show a considerable improvement in the last part of the first decade of life but may subsequently show a considerable deterioration. Education in Autism is often based on visually structured environment and manual activities: this approach is not appropriate in girls with Z-RTT who love music, do not show a particular visual ability and, in spite of some progress in their manual activities, often keep elements of hand dyspraxia. Music with active participation, routine and educational guidance in the recovery of their abilities as in subjects with intellectual disability are more appropriate in these cases. In the original article describing the first cases, the name of preserved speech variant was initially suggested by reviewers [2]. Increasing experience has shown that language is not preserved but represents a later improvement as well as a subsequent advance in hand use: a clear example of progress in a manual and verbal abilities, in contrast to classic RTT where these symptoms do not improve. In addition, as has been shown in this article, this variant differs from the classic RTT not only as regards language abilities but also for other aspects including somatic features and cognitive abilities. Thus, the term preserved speech variant is in some way misleading and we propose to change it in favor of Zappella variant of RTT (Z-RTT). Acknowledgments We thank the families for their ongoing participation in our study. We also acknowledge the Telethon Foun-

9 216 A. Renieri et al. / Brain & Development 31 (2009) dation for GGP05005 and GGP06170A Grants to A.R. This work was also supported by the Emma and Ernesto Rulfo Foundation, by MIUR (PRIN 2005) and by the University of Siena (PAR 2004) to A.R. References [1] Trevathan E, Moser HW. Diagnostic criteria for Rett syndrome. Ann Neurol 1988;23: [2] Zappella M. The Rett girls with preserved speech. Brain Dev 1992;14: [3] Zappella M. Rett syndrome-like hand washing, developmental arrest and autistic symptoms in two Italian girls. Eur Child Adolesc Psychiatry 1994;3:52 6. [4] Skjeldal O, von Tetzchner S, Jacobsen K, Smith L, Heiberg A. Rett syndrome-distribution of phenotypes with special attention to the preserved speech variant. Neuropediatrics 1995;26:87. [5] Zappella M. The preserved speech variant of the Rett complex: a report of 8 cases. Eur Child Adolesc Psychiatry 1997;6:23 5. [6] Zappella M, Gillberg C, Ehlers S. The preserved speech variant: a subgroup of the Rett Complex: a clinical report of 30 cases. J Autism Dev Disord 1998;28: [7] De Bona C, Zappella M, Hayek G, Meloni I, Vitelli F, Bruttini M, et al. Preserved speech variant is allelic of classic Rett syndrome. Eur J Hum Genet 2000;8: [8] Zappella M, Meloni I, Longo I, Hayek G, Renieri A. Preserved speech variants of the Rett syndrome: molecular and clinical analysis. Am J Med Genet 2001;104: [9] Yamashita Y, Kondo I, Fukuda T, Morishima R, Kusaga A, Iwanaga R, et al. Mutation analysis of the methyl-cpg-binding protein 2 gene (MECP2) in Rett patients with preserved speech. Brain Dev 2001;23(Suppl. 1):S [10] Zappella M, Meloni I, Longo I, Canitano R, Hayek G, Rosaia L, et al. Study of MECP2 gene in Rett syndrome variants and autistic girls. Am J Med Genet B Neuropsychiatr Genet 2003;119B: [11] Oexle K, Thamm-Mucke B, Mayer T, Tinschert S. Macrocephalic mental retardation associated with a novel C-terminal MECP2 frameshift deletion. Eur J Pediatr 2005;164: [12] Zappella M, Mari F, Renieri A. Should a syndrome be called by its correct name? The example of the preserved speech variant of Rett syndrome. Eur J Pediatr 2005;164:710, [author reply ]. [13] Sampieri K, Meloni I, Scala E, Ariani F, Caselli R, Pescucci C, et al. Italian Rett database and biobank. Hum Mutat 2007;28: [14] Kerr AM, Nomura Y, Armstrong D, Anvret M, Belichenko PV, Budden S, et al. Guidelines for reporting clinical features in cases with MECP2 mutations. Brain Dev 2001;23: [15] Huppke P, Held M, Hanefeld F, Engel W, Laccone F. Influence of mutation type and location on phenotype in 123 patients with Rett syndrome. Neuropediatrics 2002;33: [16] Ronnett GV, Leopold D, Cai X, Hoffbuhr KC, Moses L, Hoffman EP, et al. Olfactory biopsies demonstrate a defect in neuronal development in Rett s syndrome. Ann Neurol 2003;54: [17] Charman T, Neilson TC, Mash V, Archer H, Gardiner MT, Knudsen GP, et al. Dimensional phenotypic analysis and functional categorisation of mutations reveal novel genotype phenotype associations in Rett syndrome. Eur J Hum Genet 2005;13: [18] Scala E, Longo I, Ottimo F, Speciale C, Sampieri K, Katzaki E, et al. MECP2 deletions and genotype phenotype correlation in Rett syndrome. Am J Med Genet A [19] Hagberg BA. Rett syndrome clinical & biological aspects. London: McKeith Press; [20] Kerr A, Witt-Engerstrom I. Rett disorder and the developing brain. New York: Oxford University Press; [21] Young DJ, Bebbington A, Anderson A, Ravine D, Ellaway C, Kulkarni A, et al. The diagnosis of autism in a female: could it be Rett syndrome? Eur J Pediatr [22] Kerr AM, Archer HL, Evans JC, Prescott RJ, Gibbon F. People with MECP2 mutation-positive Rett disorder who converse. J Intellect Disabil Res 2006;50: [23] Philippe C, Villard L, De Roux N, Raynaud M, Bonnefond JP, Pasquier L, et al. Spectrum and distribution of MECP2 mutations in 424 Rett syndrome patients: a molecular update. Eur J Med Genet 2006;49:9 18. [24] Leonard H, Colvin L, Christodoulou J, Schiavello T, Williamson S, Davis M, et al. Patients with the R133C mutation: is their phenotype different from patients with Rett syndrome with other mutations? J Med Genet 2003;40:e52. [25] Smeets E, Terhal P, Casaer P, Peters A, Midro A, Schollen E, et al. Rett syndrome in females with CTS hot spot deletions: a disorder profile. Am J Med Genet A 2005;132: [26] Mari F, Caselli R, Russo S, Cogliati F, Ariani F, Longo I, et al. Germiline mosaicism in Rett syndrome identified by prenatal diagnosis. Clin Genet 2004;67: [27] Yaron Y, Ben Zeev B, Shomrat R, Bercovich D, Naiman T, Orr- Urtreger A. MECP2 mutation in Israel: implications for molecular analysis, genetic counseling, and prenatal diagnosis in Rett syndrome. Hum Mutat 2002;20: [28] Gill H, Cheadle JP, Maynard J, Fleming N, Whatley S, Cranston T, et al. Mutation analysis in the MECP2 gene and genetic counselling for Rett syndrome. J Med Genet 2003;40: [29] Muhle R, Trentacoste SV, Rapin I. The genetics of autism. Pediatrics 2004;113:e