L-acetylcarnitine for treating fragile X syndrome

  • Review
  • Intervention

Authors

  • José-Ramón Rueda,

    Corresponding author
    1. University of the Basque Country, Department of Preventive Medicine and Public Health, Leioa, Bizkaia, Spain
    • José-Ramón Rueda, Department of Preventive Medicine and Public Health, University of the Basque Country, Barrio Sarriena, S.N., Leioa, Bizkaia, E-48080, Spain. joseramon.rueda@ehu.es.

  • Virginia Guillén,

    1. University of the Basque Country, Department of Neuroscience, Leioa, Spain
  • Javier Ballesteros,

    1. University of the Basque Country, Department of Neuroscience, Leioa, Spain
    2. University of the Basque Country, Centre for Biomedical Research Network on Mental Health, Leioa, Spain
  • Maria-Isabel Tejada,

    1. Hospital Universitario Cruces/BioCruces Health Research Institute, Genetics Service, Barakaldo, Bizkaia, Spain
  • Ivan Solà

    1. CIBER Epidemiología y Salud Pública (CIBERESP), Iberoamerican Cochrane Centre, Biomedical Research Institute Sant Pau (IIB Sant Pau), Barcelona, Catalunya, Spain

Abstract

Background

People with fragile X syndrome (FXS) have an intellectual dysfunction that can range from very mild to severe. Symptoms can include speech and language delays and behavioural difficulties such as aggression or self injurious behaviours, emotional lability, and anxiety-related problems (for example obsessive-compulsive symptoms and perseverative behaviours). In some cases, affected people may have an additional diagnosis of attention deficit hyperactivity disorder or an autism spectrum disorder.

Objectives

To review the efficacy and safety of L-acetylcarnitine in improving the psychological, intellectual, and social performance of people with FXS.

Search methods

In May 2015 we searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Embase, PsycINFO, Web of Science, and two other databases. We also searched three trials registers, four theses databases, and the reference lists of relevant studies and reviews.

Selection criteria

Randomised controlled trials (RCTs) that assessed the efficacy of L-acetylcarnitine, at any dose, in people of any age diagnosed with FXS compared with placebo.

Data collection and analysis

For each trial, two review authors independently extracted data on the children included and interventions compared, and assessed the risk of bias of the studies across the following domains: randomisation sequence generation, allocation concealment, blinding (of participants, personnel, and outcome assessors), incomplete outcome data, selective outcome reporting, and other potential sources of bias.

Main results

We found only two RCTs that compared oral L-acetylcarnitine (LAC) with oral placebo in children with FXS. The studies included a total of 83 participants, all of them male, who were treated and followed for one year. The age of participants at the start of treatment ranged from 6 to 13 years, with a mean age of 9 years. Neither study provided information on randomisation, allocation concealment procedures, or blinding of outcome assessment, and we received no responses from the authors we emailed for clarification. We therefore rated studies as being at unclear risk of bias on these domains. We judged both studies to be at low risk of bias for blinding of participants and personnel, incomplete outcome data, and selective reporting, but to be at high risk of other bias, as at least one study was funded by a drug company, and in both studies people working for the company were part of the research team.

We used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach to rate the quality of the available evidence. Overall, the quality of the evidence was low due to the imprecision of results and high risk of other bias.

Regarding the primary outcome of psychological and learning capabilities, both studies assessed the effect of interventions on children's verbal and non-verbal intellectual functioning using the Wechsler Intelligence Scale for Children - Revised. The authors did not provide detailed data on those results but said that they found no important differences between treatment and placebo.

Both studies evaluated the impact of the treatment on hyperactive behaviour using the Conners' Abbreviated Parent-Teacher Questionnaire. In one study, teachers' assessments of the children found no clear evidence of a difference (mean difference (MD) 0.50, 95% confidence interval (CI) -5.08 to 6.08, n = 51; low-quality evidence). The other study stated that there were no differences between treated and untreated participants, but did not provide detailed data for inclusion in the meta-analysis.

Parents' assessments favoured LAC in one study (MD -0.57, 95% CI -0.94 to -0.19, n = 17; low-quality evidence), but not in the other (MD -2.80, 95% CI -7.61 to 2.01, n = 51; low-quality evidence), though changes were not large enough to be considered clinically relevant.

Regarding social skills, one study reported no clear evidence of a difference in Vineland Adaptive Behavior composite scores (MD 8.20, 95% CI -0.02 to 16.42, n = 51; low-quality evidence), yet results in the socialisation domain favoured LAC (MD 11.30, 95% CI 2.52 to 20.08, n = 51; low-quality evidence).

Both studies assessed the safety of the active treatment and recorded no side effects. Neither of the included studies assessed the secondary outcome of caregiver burden.

Authors' conclusions

Low-quality evidence from two small trials showed that when compared to placebo, LAC may not improve intellectual functioning or hyperactive behaviour in children with FXS.

Résumé scientifique

Acétyl-L-carnitine dans le traitement du syndrome de l'X fragile

Contexte

Les personnes atteintes du syndrome de l'X fragile (SXF) présentent un dysfonctionnement intellectuel pouvant aller de très léger à sévère. Les symptômes peuvent inclure des retards orthophoniques et des difficultés comportementales telles que de l'agressivité ou des comportements d'automutilation, une instabilité émotionnelle ainsi que des problèmes liés à l'anxiété (par exemple, des symptômes obsessionnels-compulsifs ou des comportements de persévération). Dans certains cas, les personnes touchées peuvent avoir également un diagnostic de trouble du déficit de l'attention avec hyperactivité ou de trouble du spectre autistique.

Objectifs

Examiner l'efficacité et l'innocuité de l'acétyl-L-carnitine dans l'amélioration des performances psychologiques, intellectuelles et sociales des personnes atteintes de SXF.

Stratégie de recherche documentaire

En mai 2015, nous avons effectué des recherches dans le registre Cochrane des essais contrôlés (CENTRAL), MEDLINE, Embase, PsycINFO, Web of Science et deux autres bases de données. Nous avons également consulté trois registres d'essais, quatre bases de données des thèses et les références bibliographiques des études et revues pertinentes.

Critères de sélection

Essais contrôlés randomisés (ECR) évaluant l'efficacité de l'acétyl-L-carnitine, à n'importe quelle dose, chez des personnes de tout âge avec un diagnostic de SXF, par rapport à un placebo.

Recueil et analyse des données

Pour chaque essai, deux auteurs de la revue ont indépendamment extrait les données sur les enfants inclus et les interventions comparées, et ont évalué le risque de biais des études dans les domaines suivants : génération de la séquence de randomisation, assignation secrète, mise en aveugle (des participants, du personnel et des évaluateurs), données de résultats incomplètes, compte-rendu sélectif et autres sources potentielles de biais.

Résultats principaux

Nous n'avons trouvé que deux ECR comparant l'acétyl-L-carnitine (ALC) par voie orale avec un placebo par voie orale chez des enfants atteints de SXF. Les études incluaient un total de 83 participants, tous de sexe masculin, qui ont été traités et suivis pendant un an. L'âge des participants au début du traitement allait de 6 à 13 ans, l'âge moyen étant de 9 ans. Aucune de ces deux études n'a fourni d'informations sur la randomisation, les procédures d'assignation secrète ou la mise en aveugle de l'évaluation des résultats, et nous n'avons pas reçu de réponse des auteurs à qui nous avions demandé des éclaircissements par courrier électronique. C'est pourquoi nous avons jugé les études comme étant à risque de biais imprécis dans ces domaines. Nous avons estimé que les deux études étaient à faible risque de biais pour la mise en aveugle des participants et du personnel, les données de résultat incomplètes et la notification sélective, mais à risque élevé d'autres biais, car au moins une étude a été financée par un laboratoire pharmaceutique, et dans les deux études, des personnes travaillant pour la société faisaient partie de l'équipe de recherche.

Nous avons utilisé l'approche GRADE (Grading of Recommendations Assessment, Development and Evaluation) pour évaluer la qualité des preuves disponibles. Dans l'ensemble, la qualité des preuves était faible en raison de l'imprécision des résultats et du risque élevé d'autres biais.

Sur le principal critère de jugement, les aptitudes psychologiques et d'apprentissage, les deux études ont évalué l'effet des interventions sur le fonctionnement intellectuel verbal et non verbal des enfants à l'aide de l'échelle révisée de Wechsler pour l'évaluation de l'intelligence chez l'enfant. Les auteurs n'ont pas fourni de données détaillées sur ces résultats mais ont indiqué n'avoir relevé aucune différence notable entre le traitement et le placebo.

Les deux études évaluaient l'impact du traitement sur le comportement hyperactif à l'aide du questionnaire abrégé de Conners à destination des parents et des enseignants. Dans une étude, l'évaluation des enfants par les enseignants n'a mis en évidence aucune preuve claire de différence (différence moyenne (DM) 0,50 ; intervalle de confiance (IC) à 95 % de -5,08 à 6,08 ; n = 51, preuves de faible qualité). L'autre étude a indiqué qu'il n'y avait pas de différence entre les participants traités et non traités, mais n'a pas fourni de données détaillées pour l'inclusion dans la méta-analyse.

Les évaluations des parents étaient en faveur de l'ALC dans une étude (DM -0,57 ; IC à 95 % de -0,94 à -0,19 ; n = 17, preuves de faible qualité), mais pas dans l'autre (DM -2,80 ; IC à 95 % de -7,61 à 2,01 ; n = 51, preuves de faible qualité), bien que les changements aient été trop petits pour être considérés comme cliniquement pertinents.

Pour ce qui est des aptitudes sociales, une étude ne rapportait aucune preuve claire de différence dans les scores composites sur l'échelle Vineland Adaptive Behavior (DM 8,20 ; IC à 95 % de -0,02 à 16,42 ; n = 51, preuves de faible qualité), mais les résultats dans le domaine de la socialisation étaient en faveur de l'ALC (DM 11,30 ; IC à 95 % de 2,52 à 20,08 ; n = 51, preuves de faible qualité).

Les deux études ont évalué l'innocuité du traitement actif et n'ont relevé aucun effet secondaire. Aucune des deux études incluses n'a évalué le critère de jugement secondaire, le fardeau du proche aidant.

Conclusions des auteurs

Des preuves de faible qualité issues de deux essais de petite taille montrent que, par rapport à un placebo, l'acétyl-L-carnitine pourrait ne pas améliorer le fonctionnement intellectuel ou le comportement hyperactif chez les enfants atteints de syndrome de l'X fragile.

Resumen

L-acetilcarnitina para el tratamiento del síndrome del X frágil

Antecedentes

Los pacientes con síndrome del cromosoma X frágil (SXF) presentan alteraciones intelectuales que pueden variar de muy leves a graves. Los síntomas pueden incluir retrasos en el habla y el lenguaje y dificultades conductuales como agresión o comportamientos autodestructivos, labilidad emocional y problemas relacionados con la ansiedad (por ejemplo, síntomas obsesivo-compulsivos y comportamientos perseverativos). En algunos casos los pacientes afectados pueden presentar un diagnóstico adicional de trastorno de hiperactividad y déficit de atención o un trastorno del espectro autista.

Objetivos

Examinar la eficacia y la seguridad de la L-acetilcarnitina para mejorar el rendimiento psicológico, intelectual y social de los pacientes con SFX.

Métodos de búsqueda

En mayo 2015, se hicieron búsquedas en el Registro Cochrane Central de Ensayos Controlados (Cochrane Central Register of Controlled Trials) (CENTRAL), MEDLINE, Embase, PsycINFO, Web of Science y en otras dos bases de datos. También se buscó en tres registros de ensayos, cuatro bases de datos de tesis, y en las listas de referencias de estudios y revisiones relevantes.

Criterios de selección

Ensayos controlados aleatorios (ECA) que evaluaran la eficacia de la L-acetilcarnitina, en cualquier dosis, en pacientes de cualquier edad diagnosticados con SFX en comparación con placebo.

Obtención y análisis de los datos

De cada ensayo, dos revisores de forma independiente extrajeron los datos sobre los niños incluidos y las intervenciones comparadas, y evaluaron el riesgo de sesgo de los estudios a través de los siguientes dominios: generación de la secuencia de asignación al azar, ocultación de la asignación, cegamiento (de los participantes, el personal y los evaluadores de resultados), datos de resultado incompletos, informe selectivo de resultados y otras fuentes posibles de sesgo.

Resultados principales

Solamente se encontraron dos ECA que compararon L-acetilcarnitina (LAC) oral con placebo oral en niños con SFX. Los estudios incluyeron un total de 83 participantes, todos ellos varones, que fueron tratados y seguidos durante un año. La edad de los participantes al inicio del tratamiento varió de seis a 13 años, con una media de edad de nueve años. Ningún estudio proporcionó información sobre la asignación al azar, los procedimientos de ocultación de la asignación ni el cegamiento de la evaluación de resultados, y no se recibieron respuestas de los autores a los que se les envió un correo electrónico para obtener aclaraciones. Por lo tanto, los estudios se calificaron con riesgo incierto de sesgo en estos dominios. Ambos estudios se consideraron en riesgo bajo de sesgo en cuanto al cegamiento de los participantes y el personal, los datos de resultado incompletos y el informe selectivo, aunque se consideraron en riesgo alto de otros sesgos debido a que al menos un estudio fue financiado por una compañía farmacéutica y en ambos estudios hubo personas que trabajaban para la empresa y formaron parte del equipo de investigación.

Se utilizó el sistema Grading of Recommendations Assessment, Development, and Evaluation (GRADE) para evaluar la calidad de las pruebas disponibles. En general, la calidad de las pruebas fue baja debido a la imprecisión de los resultados y al alto riesgo de otros sesgos.

Con respecto al resultado primario de las capacidades psicológicas y de aprendizaje, ambos estudios evaluaron el efecto de las intervenciones sobre el funcionamiento intelectual verbal y no verbal de los niños mediante la Wechsler Intelligence Scale for Children - Revised. Los autores no proporcionan datos detallados sobre esos resultados, pero dijeron que no encontraron diferencias importantes entre el tratamiento y el placebo.

Ambos estudios evaluaron la repercusión del tratamiento sobre el comportamiento hiperactivo mediante el Conners' Abbreviated Parent-Teacher Questionnaire. En un estudio, no se encontraron pruebas claras de una diferencia en las evaluaciones de los niños por los profesores (diferencia de medias [DM] 0,50; intervalo de confianza [IC] del 95%: -5,08 a 6,08; n = 51; pruebas de baja calidad). El otro estudio señaló que no hubo diferencias entre los participantes que recibieron tratamiento y los que no, aunque no proporcionó datos detallados para la inclusión en el metanálisis.

Las evaluaciones de los padres estuvieron a favor de la LAC en un estudio (DM -0,57; IC del 95%: -0,94 a -0,19; n = 17; pruebas de baja calidad), pero no en el otro (DM -2,80; IC del 95%: -7,61 a 2,01; n = 51; pruebas de baja calidad), aunque los cambios no fueron suficientemente grandes para ser considerados clínicamente relevantes.

Con respecto a las aptitudes sociales, un estudio no informó pruebas claras de una diferencia en las puntuaciones compuestas del Vineland Adaptive Behavior (DM 8,20; IC del 95%: -0,02 a 16,42; n = 51; pruebas de baja calidad), no obstante, los resultados del dominio de socialización estuvieron a favor de la LAC (DM 11,30; IC del 95%: 2,52 a 20,08; n = 51; pruebas de baja calidad).

Ambos estudios evaluaron la seguridad del tratamiento activo y no registraron efectos secundarios. Ninguno de los estudios incluidos evaluó el resultado secundario de la carga de los cuidadores.

Conclusiones de los autores

Pruebas de baja calidad de dos ensayos pequeños indicaron que, en comparación con placebo, la LAC puede no mejorar el funcionamiento intelectual o el comportamiento hiperactivo en los niños con SFX.

アブストラクト

脆弱X症候群の治療のためのL-アセチルカルニチン

背景

脆弱X症候群(FXS)患者には、ごく軽度から重度の知的障害が認められる。その症状には、言語発達遅滞および攻撃または自傷行動などの行動障害、情緒不安定、そして不安に関連する問題(例えば強迫症状および固執行動)などがある。一部の例では、この疾患の患者はさらに注意欠陥多動障害または自閉症スペクトラム障害と診断される場合がある。

目的

FXS患者の、心理的、知的、および社会的能力の改善に対するL-アセチルカルニチンの有効性および安全性についてレビューすること。

検索戦略

2015年5月に、Cochrane Central Register of Controlled Trials(CENTRAL)、MEDLINE、Embase、PsycINFO、Web of Science、およびその他の2つのデータベースを検索した。また、3つの試験登録簿、4つの論文データベース、そして関連性のある試験およびレビューの引用文献の一覧表も検索した。

選択基準

あらゆる年齢のFXS患者に対する、あらゆる用量のL-アセチルカルニチンの有効性を、プラセボと比較し評価したランダム化比較試験(RCT)。

データ収集と分析

小児を対象とし介入を比較した試験に対し、2名のレビュー著者が、それぞれデータを抽出し、次の領域における試験のバイアスのリスクについて評価した。ランダム化シーケンス生成、割りつけの隠蔽化、盲検化(参加者、職員、アウトカムの評価者について)、不完全なアウトカムデータ、選択的なアウトカム報告、およびその他のバイアスの可能性のあるソース。

主な結果

小児のFXS患者を対象に経口L-アセチルカルニチン(LAC)と経口のプラセボとを比較したRCTを、わずかに2件見出した。これらの試験は計83名の参加者を対象とし、全員が男性で、1年間治療を受け経過観察された。治療開始時の参加者の年齢は6~13歳で、年齢の中央値は9歳であった。どちらの試験からも、ランダム化、割りつけの隠蔽化方法、またはアウトカム評価の盲検化に関する情報は得られず、説明を求めて電子メールを送った著者らからの返答はなかった。そのため、これらの領域のバイアスのリスクが不明な試験であると格付けした。どちらの試験も、参加者および職員の盲検化、不完全なアウトカムデータ、および選択的報告に対するバイアスのリスクは低いと判断したが、1件以上の試験が製薬会社から資金提供を受け、またどちらの試験でもその会社の社員が研究チームの一員であったため、その他のバイアスのリスクは高いと判断した。

利用可能なエビデンスの質を格付けするため、Grading of Recommendations Assessment, Development and Evaluation(GRADE)法を用いた。全体として、結果が正確ではなく、その他のバイアスのリスクが高かったため、エビデンスの質は低かった。

どちらの試験でも、心理的能力および学習能力の主要アウトカムに関して、ウェクスラー児童知能検査(改訂版)を用いて小児の言語および非言語の知的機能に関する介入の効果が評価された。著者らは、これらの結果に関する詳細なデータを示さなかったが、治療とプラセボとの間に重要な差は認められなかったと述べた。

どちらの試験でも、Conners' Abbreviated Parent-Teacher Questionnaireを用いて、運動亢進に対する治療の影響が評価された。1件の試験では、小児に関する教師の評価に、差に関する明らかなエビデンスは認められなかった(平均差(MD)0.50、95%信頼区間(CI)-5.08~6.08、51名;低い質のエビデンス)。もう一方の試験では、治療を受けた参加者と無治療の参加者との間に差は認められないことが示されたが、メタアナリシスに組み入れるための詳細なデータは得られなかった。

保護者の評価は、1件の試験ではLACが有意であった(MD -0.57、95%CI -0.94~-0.19、17名;低い質のエビデンス)が、もう一方の試験では有意でなかった(MD -2.80、95%CI -7.61~2.01、51名;低い質のエビデンス)。しかし、変化は、臨床的関連性を考慮するのに十分な大きさではなかった。

社会能力に関して、1件の試験ではVineland Adaptive Behavior composite scoreにおける差に関する明らかなエビデンスは報告されなかった(MD8.20、95%CI -0.02~16.42、51名;低い質のエビデンス)が、社会化領域における結果は、LACが有意であった。(MD11.30、95%CI 2.52~20.08、51名;低い質のエビデンス)。

どちらの試験でも、実薬治療の安全性が評価され、副作用は記録されなかった。選択した試験のどちらにおいても、介護者の負担に関する副次的アウトカムは評価されなかった。

著者の結論

2件の小規模試験から得られた低い質のエビデンスにより、LACは、プラセボと比較した場合、小児のFXS患者の知的機能または運動亢進を改善しない可能性があることが示された。

Plain language summary

Is L-acetylcarnitine an effective treatment for fragile X syndrome?

Review question We wanted to know the efficacy and safety of L-acetylcarnitine (LAC) as compared to placebo in people with fragile X syndrome.

Background People with fragile X syndrome have intellectual impairments that can range from very mild to severe. Symptoms can include speech and language delays, and in some cases, affected people can present with behavioural problems associated with attention deficit hyperactivity disorder or autism spectrum disorder.

Study characteristics We searched the scientific literature for all randomised trials published up to May 2015 and found only two trials to include in the review. The trials recruited a total of 83 boys (6 to 13 years of age) who were treated for a maximum of one year.

Key results We found no clear evidence of important differences in verbal and non-verbal intellectual functioning between treatment with LAC and placebo.

Regarding hyperactive behaviour, teachers' assessments found no clear evidence of differences between treatment with LAC and placebo. Parents' assessments showed some differences between treatments favouring LAC, but changes were not large enough to be considered clinically relevant.

Only one study assessed social skills, and it reported no clear evidence of a difference between LAC and placebo in adaptive behaviour, though results in the socialisation domain favoured LAC.

No side effects were reported, and no study reported on the secondary outcome of caregiver burden.

Quality of the evidence The available evidence is of low quality. Risk of bias for these studies is unclear regarding the procedures used to randomise participants to LAC or placebo, to conceal treatment allocation, and to blind assessors to the results of the treatments.

Funding of the studies At least one of the studies was funded by a drug manufacturer with a commercial interest in the results. One of the studies was also funded by charitable monies.

Résumé simplifié

L'acétyl-L-carnitine est-il un traitement efficace dans le syndrome de l'X fragile ?

Problématique de la revue Nous voulions déterminer l'efficacité et l'innocuité de l'acétyl-L-carnitine (ALC) comparativement à un placebo chez les personnes atteintes du syndrome de l'X fragile.

Contexte Les personnes atteintes du syndrome de l'X fragile présentent une déficience intellectuelle qui peut aller de très légère à sévère. Les symptômes peuvent inclure des retards orthophoniques et, dans certains cas, les personnes touchées peuvent présenter des problèmes comportementaux associés à un trouble déficitaire de l'attention avec hyperactivité ou à un trouble du spectre autistique.

Caractéristiques des études Nous avons recherché dans la littérature scientifique tous les essais randomisés publiés jusqu'en mai 2015 et n'avons trouvé que deux essais à inclure dans la revue. Ces essais portaient sur un total de 83 garçons (âgés de 6 à 13 ans) qui étaient traités pendant un maximum de un an.

Principaux résultats Nous n'avons trouvé aucune preuve claire de différence importante dans le fonctionnement intellectuel verbal et non verbal entre le traitement par l'ALC et un placebo.

Pour ce qui est du comportement hyperactif, les évaluations des enseignants n'ont mis en évidence aucune preuve claire de différence entre le traitement par l'ALC et un placebo. Les évaluations des parents ont mis en évidence quelques différences entre les traitements en faveur de l'ALC, mais les changements étaient trop petits pour être considérés comme étant cliniquement pertinents.

Seule une étude a évalué les aptitudes sociales, et n'a observé aucune preuve claire de différence entre l'ALC et le placebo dans le comportement adaptatif, bien que les résultats dans le domaine de la socialisation aient favorisé l'ALC.

Aucun effet secondaire n'a été signalé, et aucune étude n'a rapporté de données sur le critère de jugement secondaire, le fardeau du proche aidant.

La qualité des preuves Les preuves disponibles sont de faible qualité. Le risque de biais de ces études n'est pas clair pour ce qui est des procédures utilisées pour assigner au hasard les participants à l'ALC ou un placebo, pour masquer l'assignation et pour la mise en aveugle des évaluateurs vis-à-vis des résultats des traitements.

Financement des étudesAu moins une des études a été financée par un fabricant de médicament ayant un intérêt commercial dans les résultats. L'une des études a également été financée par des fonds de charité.

Notes de traduction

Traduction réalisée par Cochrane France

Laički sažetak

L-acetilkarnitin za liječenje sindroma fragilnog X kromosoma

Istraživačko pitanje Autori istraživanja su istražili učinkovitost i sigurnost L-acetilkarnitina (LAC) u liječenju sindroma fragilnog X kromosoma, u usporedbi s placebom.

Dosadašnje spoznaje Osobe sa sindromom fragilnog X kromosoma imaju intelektualne nedostatke koji mogu biti vrlo blagi, ali i veoma ozbiljni. Simptomi mogu uključivati zaostajanje u govoru i jeziku, a u nekim slučajevima u osoba koje pate od tog sindroma mogu postojati problemi u ponašanju povezani s poremećajem pozornosti koji je povezan s poremećajem nedostatka pozornosti, hiperaktivnim poremećajem (ADHD) ili autizmom.

Značajke studija Pretražena je znanstvena literatura kako bi se našla sva randomizirana istraživanja objavljena do svibnja 2015. godine i pronađena su samo dva istraživanja, koja su uključena u sustavni pregled. Istraživanja su u uključila ukupno 83 dječaka (dobi od 6 do 13 godina) koji su bili liječeni najviše godinu dana.

Ključni rezultati Nisu pronađeni dokazi o značajnim razlikama u verbalnom i neverbalnom funkcioniranju između dječaka koji su liječeni LAC-om i onih koji su primali placebo.

U pogledu hiperaktivnog ponašanja dječaka, temeljem procjene njihovih učitelja nema dokaza da je ponašanje onih koji su bili podvrgnuti terapiji LAC-om bilo različito od onih koji su primali placebo. Procjena njihovih roditelja je pokazala neke razlike između jednih i drugih, u korist onih koji su bili liječeni LAC-om, ali te promjene nisu bile dovoljno velike da bi se smatrale klinički značajnima.

Samo je jedno istraživanje procjenjivalo i društvene vještine, i nije bilo jasnih dokaza da su dječaci liječeni LAC-om bili prilagodljivijeg ponašanja od onih koji su primali placebo, iako rezultati u procesu uklapanja u društvo idu u korist LAC-a.

Nuspojave nisu zabilježene, i ni jedno istraživanje nije pokazalo sekundarne rezultate kao što je opterećenje osoba koje se brinu za te dječake.

Kvaliteta dokaza Dostupni dokazi su loše kvalitete. Rizik od pristranosti u tim istraživanja je nejasan vezano za postupke koji su korišteni da bi se dobio slučajni odabir sudionika koji će primati LAC ili placebo, da bi se prikrilo kome je dodijeljeno liječenje kojom tvari i da se zaslijepe procjenjivači rezultata liječenja (da ne znaju koju terapiju ispitanici primaju).

Financiranje studija Barem jedno istraživanje je financirao proizvođač lijeka s komerijcalnim interesima vezanim za rezultate istraživanja. Jedno od istraživanja je financirala neprofitna udruga.

Bilješke prijevoda

Hrvatski Cochrane
Prevela: Barbara Pelcl
Ovaj sažetak preveden je u okviru volonterskog projekta prevođenja Cochrane sažetaka. Uključite se u projekt i pomozite nam u prevođenju brojnih preostalih Cochrane sažetaka koji su još uvijek dostupni samo na engleskom jeziku. Kontakt: cochrane_croatia@mefst.hr

Resumen en términos sencillos

¿La L-acetilcarnitina es un tratamiento efectivo para el síndrome del cromosoma X frágil?

Pregunta de la revisión Se deseaba conocer la eficacia y la seguridad de la L-acetilcarnitina (LAC) en comparación con placebo en pacientes con síndrome del cromosoma X frágil.

Antecedentes Los pacientes con síndrome del cromosoma X frágil presentan déficits que pueden variar de muy leves a graves. Los síntomas pueden incluir retrasos en el habla y el lenguaje y, en algunos casos, los pacientes afectados pueden presentar problemas conductuales asociados con el trastorno de hiperactividad y déficit de atención o un trastorno del espectro autista.

Características de los estudiosSe buscaron en la literatura científica todos los ensayos aleatorios publicados hasta mayo de 2015 y se encontraron sólo dos ensayos para incluir en la revisión. Los ensayos incluyeron a 83 niños (seis a 13 años de edad) que fueron tratados durante un período máximo de un año.

Resultados clave No se encontraron pruebas claras de diferencias importantes en el funcionamiento intelectual verbal y no verbal entre el tratamiento con LAC y placebo.

Con respecto al comportamiento hiperactivo, las evaluaciones de los profesores no encontraron pruebas claras de diferencias entre el tratamiento con LAC y placebo. Las evaluaciones de los padres mostraron algunas diferencias entre los tratamientos que favorecieron a la LAC, aunque los cambios no fueron suficientemente grandes como para ser considerados clínicamente relevantes.

Solamente un estudio evaluó las aptitudes sociales y no informó pruebas claras de una diferencia entre LAC y placebo en el comportamiento adaptativo, aunque los resultados en el dominio de la socialización estuvieron a favor de la LAC.

No se informaron efectos secundarios y ningún estudio informó el resultado secundario de la carga de los cuidadores.

Calidad de la evidenciaLas pruebas disponibles son de baja calidad. El riesgo de sesgo de estos estudios es incierto con respecto a los procedimientos utilizados para asignar al azar a los participantes a LAC o placebo, ocultar la asignación al tratamiento y cegar a los evaluadores de resultado a los hallazgos de los tratamientos.

Financiación de los estudiosAl menos uno de los estudios fue financiado por un fabricante de fármacos con interés comercial en los resultados. Uno de los estudios también fue financiado con fondos provenientes de la beneficencia.

Notas de traducción

La traducción y edición de las revisiones Cochrane han sido realizadas bajo la responsabilidad del Centro Cochrane Iberoamericano, gracias a la suscripción efectuada por el Ministerio de Sanidad, Servicios Sociales e Igualdad del Gobierno español. Si detecta algún problema con la traducción, por favor, contacte con Infoglobal Suport, cochrane@infoglobal-suport.com.

平易な要約

L-アセチルカルニチンは脆弱X症候群の治療に有効なのか?

レビューの論点 脆弱X症候群患者に対するL-アセチルカルニチン(LAC)の有効性および安全性を、プラセボと比較し明らかにすることを目的とした。

背景 脆弱X症候群患者には、ごく軽度から重度の知的障害が認められる。その症状には言語発達遅滞などがあり、一部の例では、患者に、注意欠陥多動障害または自閉症スペクトラム障害に関連する行動障害が認められる場合がある。

試験の特性 2015年5月までに発表されたすべてのランダム化試験の科学文献を検索し、本レビューに組み入れるためにわずかに2件の試験を見出した。これらの試験では計83名の少年(6~13歳)を募集し、最大で1年間治療を受けた。

主な結果 LAC治療とプラセボとの間に、言語および非言語の知的機能の重要な差に関する明らかなエビデンスは認められなかった。

教師の評価では、運動亢進に関して、LAC治療とプラセボとの間に、差に関する明らかなエビデンスは認められなかった。保護者の評価では、各治療の間にいくつかの差が認められ、LACが有意であったが、変化は、臨床的関連性を考慮するのに十分な大きさではなかった。

社会能力が評価されたのは1件の試験のみであり、適応行動におけるLACとプラセボとの間の差に関する明らかなエビデンスは報告されなかったが、社会化領域における結果では、LACが有意であった。

副作用は報告されず、また介護者の負担に関する副次的アウトカムが報告された試験はなかった。

エビデンスの質 利用可能なエビデンスの質は低かった。LACまたはプラセボに参加者をランダム化するために用いた方法、治療割り付けの隠蔽化方法、および治療の結果に対する評価者の盲検化の方法に関するこれらの試験のバイアスのリスクは不明である。

これらの試験への資金提供 これらの試験のうち1件以上が、本結果に商業的関心のある製薬業者から資金提供を受けた。これらの試験のうちの1件では、慈善資金による資金提供も受けた。

訳注

《実施組織》厚生労働省「「統合医療」に係る情報発信等推進事業」(eJIM:http://www.ejim.ncgg.go.jp/)[2018.3.14]
《注意》この日本語訳は、臨床医、疫学研究者などによる翻訳のチェックを受けて公開していますが、訳語の間違いなどお気づきの点がございましたら、eJIM事務局までご連絡ください。なお、2013年6月からコクラン・ライブラリーのNew review, Updated reviewとも日単位で更新されています。eJIMでは最新版の日本語訳を掲載するよう努めておりますが、タイム・ラグが生じている場合もあります。ご利用に際しては、最新版(英語版)の内容をご確認ください。 
CD010012 Pub2

Summary of findings(Explanation)

Summary of findings for the main comparison. L-acetylcarnitine compared to placebo for treating fragile X syndrome
  1. 1 High risk of bias because the studies were funded by a drug company and the employees of that company were part of the research team. No information on randomisation procedures, allocation concealment, or blinding of outcome assessment.
    2 Studies only included males.

L-acetylcarnitine compared to placebo for treating fragile X syndrome
Patient or population: people being treated for fragile X syndrome
Settings: outpatients
Intervention: L-acetylcarnitine
Comparison: placebo

Outcomes:

Behaviour or social performance

Illustrative comparative risks* (95% CI)Relative effect
(95% CI)
Number of participants
(studies)
Quality of evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Placebo L-acetylcarnitine

Conners' Parent Rating Scale

(raw scores)

 The mean score in the intervention group was 0.57 lower (0.94 to 0.19 lower) 17
(1 study)
⊕⊕⊝⊝
low 1,2
Average basal levels not available

Conners' Parent Rating Scale

(standardised t-test scores)

 The mean score in the intervention group was 2.80 lower (-7.61 lower to 2.01 higher) 51
(1 study)
⊕⊕⊝⊝
low 1,2
Average basal levels: 69.4 (95% CI 66.1 to 72.8)

Conners' Teacher Rating Scale

(standardised t-test scores)

 The mean score in the intervention group was 0.5 higher (5.08 lower to 6.08 higher) 51
(1 study)
⊕⊕⊝⊝
low 1,2
Average basal levels: 72.6 (95% CI 69.2 to 76.9)

Vineland Adaptive Behaviour Scale: adaptive behaviour composite

(standardised scores)

 The mean score in the intervention group was 8.2 higher (0.02 lower to 16.42 higher) 51
(1 study)
⊕⊕⊝⊝
low 1,2
Average basal levels: 41.9 (95% CI 37.9 to 45.9)

Vineland Adaptive Behaviour Scale: socialisation domain

(standardised scores)

 The mean score in the intervention group was 11.3 higher (2.52 to 20.08 higher) 51
(1 study)
⊕⊕⊝⊝
low 1,2
Average basal levels: 57.1 (95% CI 52.6 to 61.6)
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Background

Description of the condition

Fragile X is a family of disorders (Chonchaiya 2009). It includes fragile X syndrome (FXS), the most common form of inherited intellectual disability, which is estimated to affect 1 in 4000 males and 1 in 8000 females (Crawford 2001). These estimates are based on population projections from cohorts of children with special educational needs and, given that some affected individuals may have IQs in the normal or borderline range, they could underestimate the real prevalence of the disease (Song 2003; Hagerman 2008).

FXS is caused by the lack or a deficiency of a protein called FMRP (fragile X mental retardation protein), produced by the FMR1 gene. This gene is on the X chromosome (Xq27.3) and can undergo changes affecting its pattern of DNA (deoxyribonucleic acid). Specifically, there is a trinucleotide (CGG) sequence within the promoter region that varies in normal people from 6 to 54 copies, or 'repeats'. The number of CGG repeats drives the consequences associated with FXS. When the number of CGG repeats exceeds 200 (called full mutation), there is transcription silencing (Verkerk 1991), where the FMR1 gene does not produce the FMRP protein and the FXS clinical syndrome appears. If the number of CGG repeats is between 55 to 200 (called premutation), the FMR1 gene is active, the FMRP protein is produced, and no clinical features of FXS appear (Fu 1991). Women with premutation have a high risk of increasing the number of the CGG repeats and of transmitting the full mutation to the next generation, which explains why FXS is so prevalent.

Most children with FXS are not diagnosed until three to five years of age, and the first diagnosis often comes from their behavioural phenotype and cognitive attributes.

In general, people with FXS have intellectual dysfunction that can range from very mild to severe, with IQs in the range of 35 to 50. Symptoms can also include speech and language delays and behavioural difficulties such as aggression or self injurious behaviour, emotional lability, and anxiety-related problems (for example obsessive-compulsive symptoms and perseverative behaviours). In some cases, affected people may have an additional diagnosis of attention deficit hyperactivity disorder (ADHD) or an autism spectrum disorder (ASD) (Clifford 2007; Harris 2008).

Regarding physical features, most affected males have large, prominent ears, long face, prominent chin, hyperextensible finger joints, soft skin, flat feet, mitral valve prolapse, pectus excavatum, and macroorchidism; these dysmorphic features usually appear at adult age. Epilepsy also occurs in about 20% of affected males. In females affected with FXS, the phenotypic and cognitive characteristics are usually less severe than in males, although the extent of these characteristics is very variable. Among their behavioural problems, shyness and social isolation are particularly common. About one-third of females with the full mutation are as intellectually impaired as males; these females also commonly show some of the facial features, though about half of them have only borderline features (Benneto 2002).

It was initially thought that individuals with the premutation are completely asymptomatic, but it was later found that they are not. Premutation alleles are associated with a significant elevation of FMR1 mRNA levels (Tassone 2000), and it has been shown that because of this, these individuals are at risk of developing fragile X-associated tremor, also known as ataxia syndrome (FXTAS). This late-onset neurodegenerative disorder usually affects males over 50 years of age and, to a lesser extent, females (Jacquemont 2003). Female premutation carriers are also at risk of fragile X-associated primary ovarian insufficiency (FXPOI), a disorder that involves ovarian insufficiency, infertility, and early menopause (Wittenberger 2007). Fibromyalgia, psychiatric disorders, and thyroid problems have also been observed in these women (Rodriguez-Revenga 2013). Neither of these disorders (FXPOI and FXTAS) are witnessed in people affected with FXS.

Description of the intervention

L-acetylcarnitine (LAC) is a pharmacological compound that is administered orally. LAC is a drug agonist of mitochondrial function, a neuronal growth factor, and can exert an antioxidant effect on neurons in the central nervous system (Hudson 2003).

LAC, the acetylated form of carnitine, is the most abundant carnitine ester in the cerebral tissue. LAC is a small water-soluble molecule belonging to the ubiquitous carnitine system, which includes carnitine, its esters (acylcarnitines), and the enzymes responsible for their metabolism. The enzymes involved in the synthesis of LAC are the carnitine acyl-transferases and, in particular, carnitine acetyl-transferase (Pettegrew 2000).

The carnitine system in the brain is fundamental to the transport of short- and medium-chain acyl groups outside the mitochondria, and regulates fatty acids and glucose utilisation. LAC also increases levels of dopamine, glycine, serine, threonine, alanine, and glutamine in selected brain regions and reduces glutamate dehydrogenase activity in neuronal mitochondria, suggesting an effect on neurotransmitter metabolism at synapses (Toth 1993; Gorini 1999).

How the intervention might work

LAC has been shown to inhibit in vitro the cytogenetic expression of the FXS-associated fragile site (fra(X)(q27.3)) (Pomponi 1994), and LAC effectively acetylates histones in the FMR1 promoter gene, although this effect does not result in a substantial reactivation of the fully mutated FMR1 gene in vitro (Pascale 2003; Tabolacci 2005). LAC treatment has proven particularly effective in experimental animals (rats) with learning deficits and hyperactivity (Dell'Anna 1997).

The exogenous administration of LAC may affect brain activity in FXS by: 1) modulation of fuel partitioning for energy production, which, at the mitochondrial level, is associated with the Krebs cycle metabolic role in neurotransmitter synthesis; 2) remodelling of lipid membrane in terms of LAC actively determining the production of polyunsaturated fatty acids; and 3) preferential effect on the attention component of the cholinergic system, which relies on its peculiar modality of communication in the central nervous system (Calvani 2001).

In human studies, LAC improved brain measures of membrane phospholipid and high-energy phosphate metabolism, which corresponded with a delay in people's cognitive decline (Pettegrew 1995). A Cochrane systematic review assessed the effectiveness of LAC in the treatment of people with dementia and found evidence of a benefit of LAC on the Clinical Global Impression scale as a categorical measure, and on the Mini Mental State Examination at 24 weeks, but no evidence using objective assessments for any other outcome. It concluded that the evidence does not suggest that LAC is likely to prove to be an important therapeutic agent (Hudson 2003).

It is thus of interest to evaluate the impact of LAC on cognitive function, learning capabilities, and behaviour or social performance of people affected by FXS.

Why it is important to do this review

L-acetylcarnitine has been tested in the treatment of ADHD, a disorder common in people with FXS. Published studies on the efficacy of LAC in people without FXS have found contradictory results. A study of school-aged boys with ADHD showed that carnitine therapy has a beneficial effect on hyperactive-impulsive behaviour, without showing any adverse side effects (Van Oudheusden 2002). Another study found no effect on the overall ADHD population (especially the combined type) (Arnold 2007), but considered that it deserves further exploration for possible benefits, specifically in the inattentive type. Abbasi 2011b found no difference on the Conners' Parent and Teacher Rating Scales scores for application of LAC as an adjunctive therapy to methylphenidate in children and adolescents with ADHD.

To our knowledge, while several recently published articles include narrative reviews on pharmacological interventions in people with FXS (Hagerman 2009; Hall 2009; Rueda 2009), no comprehensive systematic review of the efficacy and safety of LAC has been published to date.

Objectives

To review the efficacy and safety of LAC in improving the psychological, intellectual, and social performance of people with FSX.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs).

Types of participants

Any person, of any age, diagnosed with FXS.

Types of interventions

LAC at any dose versus placebo.

Types of outcome measures

Primary outcomes
  1. Psychological and learning capabilities (assessed through validated tools such as Stanford-Binet, Leiter, or Wechsler intelligence (IQ) tests, or Peabody Picture Vocabulary Test - Revised).*

  2. Behaviour or social performance (assessed through validated questionnaires such as Autistic Descriptors Checklist, Autism Behaviour Checklist, Childhood Autism Rating Scale, or Conners’ Parent and Teaching Rating Scales, and activities of daily living (assessed through validated scales such as Katz Index of Independence in Activities of Daily Living or Lawton Instrumental Activities of Daily Living).*

  3. Adverse effects (i.e. safety)

    1. Number of participants reporting severe adverse side effects.*

    2. Number of participants leaving the study early due to side effects.

Secondary outcomes
  1. Caregiver burden (assessed using validated tools such as the Zarit Burden Interview).*

Outcomes relating to children could be measured by rating scales, questionnaires, or interviews conducted with children's parents, representatives, or caretakers, or their teachers, tutors, or other professionals.

In cases where data were available, we used those outcomes marked with an asterisk to populate the Summary of findings for the main comparison.

The primary time point to assess outcomes was the longest follow-up period available: at least 12 months of treatment.

Search methods for identification of studies

Electronic searches

We searched the following electronic databases on 12 May 2015. We applied no date or language restrictions.

  1. Cochrane Central Register of Controlled Trials (CENTRAL) 2015 (Issue 5), part of the Cochrane Library.

  2. Ovid MEDLINE(R) (1946 to May, Week 1, 2015).

  3. Embase (Ovid) (1980 to 2015, Week 19).

  4. PsycINFO (Ovid) (2002 to May, Week 1, 2015).

  5. Web of Science (1970 to current).

  6. CINAHL Plus (EBSCOhost) 1937 to current.

  7. LILACS (http://lilacs.bvsalud.org/es/) all available years.

  8. WorldCat (limited to theses) (www.worldcat.org/) all available years.

  9. DART-Europe E-theses Portal (www.dart-europe.eu/basic-search.php) all available years.

  10. Networked Digital Library of Theses and Dissertations (www.ndltd.org/) all available years.

  11. Trove (limited to theses) (http://trove.nla.gov.au/) all available years.

  12. ClinicalTrials.gov (www.clinicaltrials.gov/).

  13. metaRegister of Controlled Trials (www.controlled-trials.com/mrct/).

  14. WHO International Clinical Trials Registry (http://apps.who.int/trialsearch/).

We have provided the search strategies for each database in Appendix 1.

Searching other resources

We checked reference lists from relevant trials and reviews.

Data collection and analysis

Selection of studies

Two review authors (JRR and VG) independently screened titles and abstracts identified by the search strategy to assess whether they met the inclusion criteria. Next, they independently screened full-text papers or reports for trials that appeared relevant or for which more information was needed to determine relevance. JRR and VG resolved disagreements through discussion and by consulting the other review authors (JB, MIT, and IS). JRR and VG recorded reasons for excluding trials in the 'Characteristics of excluded studies' tables. Neither review author was blinded to the authors, institutions, or journals in which the articles were published.

Data extraction and management

For each trial, two review authors (JRR and VG) independently extracted data on: population, interventions, comparisons, randomisation methods, blinding, outcome measures, follow-up duration, attrition and handling of missing data, and methods of analysis. See 'Characteristics of included studies' tables for further details.

Assessment of risk of bias in included studies

Using The Cochrane Collaboration’s tool for assessing risk of bias (Higgins 2011), two review authors (JRR and VG) independently assessed the adequacy with which the included studies managed risk of bias across the following domains: sequence generation, allocation concealment, blinding (of participants, personnel, and outcome assessors), incomplete outcome data, selective outcome reporting, and other sources of bias (for example stopping the trial early, changing methods during the trial). Two review authors (JRR and VG) independently assigned ratings of 'low', 'high', and 'unclear' risk of bias to each domain for each included study. There were no disagreements among review authors on the judgements presented in the 'Risk of bias' tables. For both included studies, there was insufficient information to allow us to make a judgement on key domains such as sequence generation, allocation concealment, and blinding of outcome assessors. We emailed the authors for clarification but they have yet to respond. We reported the risk of bias as 'unclear'.

Measures of treatment effect

For continuous outcome data, we calculated the mean difference (MD) and 95% confidence intervals (CI) when outcomes were measured using the same scale.

Please see our protocol and Appendix 2 for further details on the 'measures of treatment effect' that we will use, if applicable, in future updates of this review (Rueda 2012).

Unit of analysis issues

We did not find any cluster or multiple intervention trials, and consequently had no specific 'unit of analysis issues' to report. Please see our protocol and Appendix 2 for further information on how we will resolve unit of analysis issues in future updates of this review (Rueda 2012). As stated in the protocol (Rueda 2012), we would not include cross-over trials in the review.

Dealing with missing data

One of the studies included 20 participants (Torrioli 1999), 3 of which were lost to follow-up. The other study included 63 participants (Torrioli 2008), but 3 participants in the LAC arm and 4 participants in the placebo arm were lost to follow-up. Outcome data was not provided for a further three participants in the LAC arm and two participants in the placebo arm. We emailed authors for clarification, but received no response. Consequently, the analyses that we report here do not correspond to intention-to-treat, but to the set of complete data reported in the publications.

Please see Appendix 2 for further details on methods for 'dealing with missing data', which have been archived for future updates of this review.

Assessment of heterogeneity

To examine statistical heterogeneity, we used the Chi² test, where a low P value (less than 0.10) indicates heterogeneity of treatment effects (Higgins 2002). We also used the I² statistic to quantify inconsistency and to determine the percentage of variability that is due to heterogeneity rather than sampling error or chance (Higgins 2003). In case heterogeneity was present, we assessed possible reasons underlying clinical heterogeneity such as participant characteristics (gender, age) and clinical features such as degree of intellectual disability or presence of autism spectrum disorder or ADHD.

Assessment of reporting biases

We found only two studies, an insufficient number to draw funnel plots (plotting trial effect against standard error) to investigate the likelihood of reporting biases.

Please see our protocol and Appendix 2 for further details on the 'assessment of reporting biases', which have been archived for future updates of this review (Rueda 2012).

Data synthesis

We conducted meta-analyses when event rates or means and standard deviations (SDs) were available or could be calculated and studies included similar doses and outcome measurements. We only combined data from studies that were clinically similar and when the Chi² test or the I² statistic did not show large deviations from homogeneity. We conducted all meta-analyses using a random-effects model.

When meta-analysis was inappropriate, we provided a narrative description of the individual study results.

Summary of findings

We have presented key information concerning the magnitude of the effect of the interventions examined and the sum of available data on the main outcomes in Summary of findings for the main comparison. This table also reports on the quality of the evidence for each individual outcome according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach, which considers within-study risk of bias, directness of evidence, heterogeneity, precision of effect estimates, and risks of publication bias (Grade Working Group 2004).

Subgroup analysis and investigation of heterogeneity

We did not conduct a subgroup analysis since there was an insufficient number of studies in the review to assess meaningful clinical differences among them. The two studies included in this review involved participants with similar characteristics in relation to gender and age, and doses and duration of the active treatment. Furthermore, in one study, all participants had a diagnosis of ADHD (Torrioli 2008); the other study provided no information on the ADHD status of participants (Torrioli 1999).

In future updates, when possible and necessary, we will conduct further investigation of the causes of heterogeneity using subgroup analyses for age groups, additional diagnosis of autism spectrum disorder or ADHD, and different doses of LAC (see Rueda 2012 and Appendix 2).

Sensitivity analysis

We did not conduct a sensitivity analysis since there was an insufficient number of studies in the review and the only pooled analysis did not present heterogeneous findings (see Analysis 1.1).

Please see our protocol and Appendix 2 for further details on sensitivity analyses, which have been archived for future updates of this review (Rueda 2012).

Results

Description of studies

Results of the search

Database searching identified 83 records. After removing duplicates and irrelevant records, we identified five potentially eligible studies. We excluded three of these after reading the full text because the participants did not have a diagnosis of FXS (see Figure 1).

Figure 1.

Study flow diagram

Included studies

Only two studies met the inclusion criteria. Both were conducted in Europe by the same leading researcher. The first was a single-centre study, which took place in Italy and was reported as a 'Letter to the Editor' (Torrioli 1999). The second study involved several centres in three European countries: Italy, France, and Spain (Torrioli 2008).

The studies included 83 participants in total. All participants were male.

The age of participants at the start of treatment ranged from 6 to 13 years; the mean age was 9 years.

In one of the studies (Torrioli 2008), a FXS diagnosis of all participants was confirmed by Southern blot techniques, and all had a diagnosis of ADHD according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (APA 2000). In this study, all participants received extra care and stimulation, both at school and at home, regardless of whether they were taking LAC or placebo. Moreover, the regular visits by neuropsychiatrists prompted a revision of the support activities that the children received (speech therapy, physical therapy, occupational therapy, or other modality).

Both studies compared oral LAC with oral placebo, with the doses of the active branch being 50 mg/kg twice a day in Torrioli 1999 and 500 mg twice a day in Torrioli 2008. Duration of treatment was one year in both studies.

See Characteristics of included studies.

Excluded studies

We excluded three RCTs that analysed the effects of carnitine products on participants who were diagnosed with ADHD but not FXS (Van Oudheusden 2002; Arnold 2007; Abbasi 2011a) (see Characteristics of excluded studies).

Risk of bias in included studies

We considered risk of bias as 'unclear' for both included studies in relation to random sequence allocation, allocation concealment, and blinding of outcome assessors. Both studies were considered to be at low risk of bias for blinding of participants and personnel, incomplete outcome data, and selective outcome reporting, but at high risk of other bias. (See Characteristics of included studies and Figure 2).

Figure 2.

'Risk of bias' summary: review authors' judgements about each 'Risk of bias' item for each included study

Allocation

Neither study detailed information on random sequence generation and allocation concealment, and for this reason we rated both studies as being at unclear risk of bias for these domains.

Blinding

As both included studies were defined by the authors as double-blinded, we rated them as being at low risk of bias for blinding of participants and personnel. However, neither study provided information on whether outcome evaluators knew the treatment status of the participants, and so we rated this domain as being at unclear risk of bias.

Incomplete outcome data

In Torrioli 1999, 3 of the 20 initial participants stopped treatment after 1 month, but no more details were provided in the publication. However, this number is small, and we have no reason to believe that these losses were related to treatment with LAC or placebo.

In Torrioli 2008, out of the initial 63 participants enrolled, 7 dropped out (4 placebo and 3 LAC). Five dropped out because of family choice (three because of problems in attending the evaluation sessions, two because of an apparent increase of hyperactivity), one developed gynaecomastia, and one had to leave the study to start psychoactive drug treatment. The two participants who appeared to have increased hyperactive behaviour and the one who later developed gynaecomastia appeared to have received placebo. Fifty-six participants completed the study, but data from 5 of them (2 placebo and 3 LAC) were not included in the statistical analysis at 12-month follow-up.

Given that reasons for withdrawals seemed to be unrelated to either treatment, it is unlikely that the aforementioned situations could result in bias, and we therefore rated the studies as being at low risk of bias on this domain.

Selective reporting

Though neither of the studies had published protocols in advance of the trial, the authors commented on all expected outcomes in the reports. For this reason, we rated the studies as being at low risk of bias on this domain.

Other potential sources of bias

At least one study was funded by a drug company with a commercial interest in the results, and in both studies, employees of a company that produces the drug under investigation were part of the research team. For these reasons, we rated both studies to be at high risk of other bias.

Effects of interventions

See: Summary of findings for the main comparison L-acetylcarnitine compared to placebo for treating fragile X syndrome

We have presented here the results of interventions 12 months after the beginning of treatment.

Psychological and learning capabilities

Neither Torrioli 1999 nor Torrioli 2008 found the active treatment to impact on verbal or non-verbal scores as assessed by the Wechsler Intelligence Scale for Children - Revised (Wechsler 2004). Lack of detailed data prevented us from conducting a meta-analysis of the results of the two studies.

Also, Torrioli 1999 said that they did not find differences in visual-perceptive and graphic functions as assessed by the Bender-Gestalt test (Bender 1938), but did not provide detailed information.

Behaviour or social performance

Both studies assessed the impact of LAC on hyperactive behaviour; Torrioli 1999 using the Conners' Abbreviated Parent - Teacher Questionnaire and Torrioli 2008 using the Conners' Global Index - Parent Version and Conners' Global Index - Teacher Version scales (Conners 1990; Conners 1998). Neither study showed LAC to be effective in reducing hyperactive behaviour.

Parents' assessments showed a small difference favouring LAC in one study (Torrioli 1999) (MD -0.57, 95% CI -0.94 to -0.19, n = 17; low-quality evidence), but not in the other (MD -2.80, 95% CI -7.61 to 2.01, n = 51; low-quality evidence) (see Figure 3) (Torrioli 2008). We could not combine results from the two studies because of the different outcomes reported (Torrioli 1999 reported change since baseline scores, and Torrioli 2008 reported independent t-test scores).

Figure 3.

Forest plot of comparison: 1 L-acetylcarnitine versus placebo, outcome: 1.1 Conners' Parent Rating Scale

Regarding teachers' assessments, Torrioli 1999 stated that there were no significant differences between treated and untreated participants, but did not provide detailed data. Torrioli 2008 provided more detailed information: after one year of treatment, the mean scores were 67.5 (LAC) and 67.0 (placebo), indicative of a very small non-significant difference (see Figure 4).

Figure 4.

Forest plot of comparison: 1 L-acetylcarnitine versus placebo, outcome: 1.2 Conners' Teacher Rating Scale

Torrioli 2008 also used the Vineland Adaptive Behavior Scale to assess the effect of LAC compared to placebo on adaptive behaviour (Doll 1965), and provided detailed data for the composite and socialisation domains. After one year of treatment, the MD in the composite scores, calculated by us, was 8.20 (95% CI -0.02 to 16.42); there was no significant difference between LAC and placebo (see Figure 5). With respect to the socialisation domain, the mean score difference, calculated by us, was 11.30 (95% CI 2.52 to 20.08), in favour of LAC (see Figure 6).

Figure 5.

Forest plot of comparison: 1 L-acetylcarnitine versus placebo, outcome: 1.3 Vineland Adaptive Behaviour Scale: adaptive behaviour composite

Figure 6.

Forest plot of comparison: 1 L-acetylcarnitine versus placebo, outcome: 1.4 Vineland Adaptive Behaviour Scale: socialisation domain

Adverse effects (safety)

Both included studies assessed the safety of the active treatment. Parents were asked to report any side effects associated with the medication at each control visit. The list of side effects included headache, nausea, vomiting, and diarrhoea. No parents reported these symptoms.

Caregiver burden

Neither of the included studies assessed caregiver burden.

Discussion

Summary of main results

In this review, we assessed whether LAC could improve psychological and learning capabilities, behaviour, or social performance of children affected with FXS compared with placebo. We found only two RCTs, involving 83 participants, all of them boys aged 6 to 13 years, who were treated and followed up for one year. One of the studies was reported in a 'Letter to the Editor' (Torrioli 1999).

Both studies reported that LAC did not improve the psychological and learning capabilities of the participants as assessed by the Wechsler Intelligence Scale for Children - Revised, but did not provide detailed data on the results.

Regarding hyperactive behaviour as measured by Conners' questionnaires, we found no clear evidence of relevant differences between LAC and placebo when evaluated by teachers (MD 0.50, 95% CI -5.08 to 6.08, 1 study, n = 51; low-quality evidence) or by parents (two uncombined studies, Torrioli 1999: MD -0.57, 95% CI -0.94 to -0.19, n = 17; low-quality evidence; and Torrioli 2008: MD -2.80, 95% CI -7.61 to 2.01, n = 51; low-quality evidence). Even though parents' evaluations showed a difference in favour of LAC in one study, differences of half a point on that questionnaire are very small and cannot be considered clinically relevant, particularly if we take into account that the basal score of participants in the larger study was about 70 (Torrioli 2008).

One study measured the impact of treatment on participants' social skills using the Vineland Adaptive Behavior composite scale (Torrioli 2008). Although they found no clear evidence of a difference between LAC and placebo as measured by composite/global scores (MD 8.20, 95% CI -0.02 to 16.42, n = 51; low-quality evidence), they did find evidence in favour of LAC on the socialisation domain (MD 11.30, 95% CI 2.52 to 20.88, n = 51; low-quality evidence).

Both studies assessed the safety of the active treatment and reported no adverse side effects. Neither study reported on the secondary outcome of caregiver burden.

Overall completeness and applicability of evidence

We undertook a comprehensive search of the literature and identified any relevant research. We found only two trials that examined the effect of LAC in participants with FXS, all of them boys between 6 and 13 years of age. Consequently, it is unclear if these results will generalise to young females or older people with FXS.

Quality of the evidence

This systematic review found limited evidence of sufficient methodological rigour or adequate reporting to confidently assess the clinical effects of LAC for children with FXS. The quality of the evidence according to the GRADE approach is 'low'.

The evidence comes from two RCTs involving a total of 83 participants, all of them boys, with a follow-up period of only one year. The age range of participants was wide (6 to 13 years), particularly if we take into account that FXS produces an array of developmental disabilities, each with its own trajectory. In that context it is important to identify whether the intervention is applicable to a specific age group, yet the trials neither conducted that analysis nor provided data to allow for it. Moreover, the sample sizes of the included studies were so small that it was not feasible to do any subgroup analyses.

We did not find any registered trials, and with the results of only two studies it was not possible to assess risk of publication bias using funnel plots. Also, there was insufficient information in the reports with regards to randomisation procedures, allocation concealment, and blinding of outcome evaluators. Consequently, we were unable to conduct a thorough assessment of the risk of bias of the two RCTs. However, at least one of the studies was funded by a drug company with a commercial interest in the results, and in both studies, employees of a company that produces the drug were part of the research team; these factors put the studies at a high risk of bias.

Regarding the other domains of the GRADE system, the studies showed no relevant inconsistency, though the results were probably imprecise, due to the small number of participants in the sample.

Potential biases in the review process

We did not identify any potential biases in the review process. Our search strategy was comprehensive and included searches of the most important clinical trial registers. Moreover, at least two review authors independently extracted and analysed the data.

Agreements and disagreements with other studies or reviews

We did not find any other studies or specific reviews on LAC for people with FXS.

Two Cochrane reviews have assessed the efficacy of pharmacological treatments for ADHD in people with intellectual disabilities. The review on risperidone did not find any RCTs (Thomson 2009a). The review on amphetamine included one cross-over study of 15 children with ADHD, intellectual disabilities, and FXS (Thomson 2009b). However, the comparison in that study was with placebo, not LAC; the duration of treatment was only one week; and the study reported no significant difference between amphetamine and placebo for any of the ADHD measures, while reporting significantly more side effects (mainly mood lability and irritability) in those taking amphetamine.

A RCT evaluated the effect of LAC as an adjunctive therapy to methylphenidate in children and adolescents with ADHD (Abbasi 2011a), and found no differences in Conners' Parent and Teacher Rating Scales (CPTRS) scores between participants receiving LAC or placebo. Another study comparing LAC and placebo in children with ADHD did not find significant differences of effect between treatments on behaviour or social performance as measured by the CPTRS (Arnold 2007).

Torrioli et al considered that a given child's hyperactive behaviour might be minimal or absent in a strict one-to-one relationship with a support teacher, such as that provided by the Italian school legislation to intellectually disabled students. Those support teachers are thus unlikely to have seen substantial modifications in the children's behaviour during the trial (Torrioli 1999). However, others contend that teachers' assessments are useful for detecting change during treatment of children and young people with ADHD (Van Oudheusden 2002; Biederman 2004).

Torrioli 2008 concluded that LAC can be recommended as an effective treatment for ADHD in children with FXS, especially in those who do not tolerate stimulant drugs. However, we do not agree, since positive changes detected by parents in children taking LAC were too small to be considered clinically relevant.

Authors' conclusions

Implications for practice

There is low-quality evidence showing that LAC has small or uncertain effects on psychological and learning capabilities and behaviour performance outcomes of children with FXS. Results for social performance were inconsistent. Our findings did not support LAC as an effective treatment for children with FXS.

Implications for research

Results of the RCTs included in this review did not support LAC pharmacological treatment as a research priority area for treatment of people with FXS. People with FXS willing to participate in RCTs are scarce and should be recruited for studies that evaluate more promising drugs or interventions.

Given that intellectual, behavioural, emotional, and learning performance in people with FXS is strongly influenced by different social factors, future studies should also include the evaluation of non-pharmacological interventions, such as modifications in the home environment, tailored behavioural interventions and classroom environments, or language and occupational therapy. They should also include affected female children and conduct separate analyses for different age groups.

Acknowledgements

We acknowledge the Cochrane Developmental, Psychosocial and Learning Problems Group for their assistance and guidance throughout the whole process.

Data and analyses

Download statistical data

Comparison 1. L-acetylcarnitine versus placebo
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Conners' Parent Rating Scales2 Mean Difference (IV, Random, 95% CI)Totals not selected
1.1 Change scores in Conners' Abbreviated Parent - Teacher Questionnaire (APTQ)1 Mean Difference (IV, Random, 95% CI)0.0 [0.0, 0.0]
1.2 T-scores from standardized profiles in Conners’ Global Index - Parent Version1 Mean Difference (IV, Random, 95% CI)0.0 [0.0, 0.0]
2 Conners' Teachers Rating Scale1 Mean Difference (IV, Random, 95% CI)Totals not selected
3 Vineland Adaptive Behaviour Scale: adaptive behaviour composite1 Mean Difference (IV, Random, 95% CI)Totals not selected
4 Vineland Adaptive Behaviour Scale: socialization domain1 Mean Difference (IV, Random, 95% CI)Totals not selected
Analysis 1.1.

Analysis 1.1.

Comparison 1 L-acetylcarnitine versus placebo, Outcome 1 Conners' Parent Rating Scales.

Analysis 1.2.

Analysis 1.2.

Comparison 1 L-acetylcarnitine versus placebo, Outcome 2 Conners' Teachers Rating Scale.

Analysis 1.3.

Analysis 1.3.

Comparison 1 L-acetylcarnitine versus placebo, Outcome 3 Vineland Adaptive Behaviour Scale: adaptive behaviour composite.

Analysis 1.4.

Analysis 1.4.

Comparison 1 L-acetylcarnitine versus placebo, Outcome 4 Vineland Adaptive Behaviour Scale: socialization domain.

Appendices

Appendix 1. Search strategies

Cochrane Central Register of Controlled Trials (CENTRAL; Cochrane Library)

Issue 5 of 12, May 2015. Search date: 12 May 2015 (N = 2 records)

  1. Fragile X Syndrome

  2. acetylcarnitine

  3. #1 AND #2

Ovid MEDLINE(R)

1946 to May, Week 1, 2015. Search date: 12 May 2015 (N = 9 records)

  1. Fragile X Syndrome/

  2. Mental Retardation, X-linked/

  3. (fragile-X or FRAXE or FRAXA or Martin Bell$ or FXS).tw.

  4. 1 or 2 or 3

  5. acetylcarnitine/

  6. (carnitine- acetyl or acetyl-carnitine or medosan or acetyl l-carnitine or acetylcarnitine or levocarnitine acetyl or branigen or alcar or ALC).mp.

  7. 5 or 6

  8. randomized controlled trial.pt.

  9. controlled clinical trial.pt.

  10. randomi#ed.ab.

  11. placebo$.ab.

  12. drug therapy.fs.

  13. randomly.ab.

  14. trial.ab.

  15. groups.ab.

  16. or/8-15

  17. 4 and 7 and 16

Embase (Ovid)

1980 to 2015, Week 19. Search date: 12 May 2015 (N = 15 records)

  1. fragile X syndrome/

  2. X linked mental retardation/

  3. (fragile-X or FRAXE or FRAXA or Martin Bell$ or FXS).tw.

  4. acetylcarnitine/

  5. (carnitine-acetyl or acetyl-carnitine or medosan or acetyl l-carnitine or acetylcarnitine or levocarnitine acetyl or branigen or alcar or ALC).mp.

  6. 1 or 2 or 3

  7. 4 or 5

  8. 6 and 7

PsycINFO (Ovid)

1987 to May, Week 1, 2015. Search date: 12 May 2015 (N = 2 records)

  1. fragile X syndrome/

  2. X linked mental retardation/

  3. (fragile-X or FRAXE or FRAXA or Martin Bell$ or FXS).tw.

  4. acetylcarnitine/

  5. (carnitine-acetyl or acetyl-carnitine or medosan or acetyl l-carnitine or acetylcarnitine or levocarnitine acetyl or branigen or alcar or ALC).mp.

  6. 1 or 2 or 3

  7. 4 or 5

  8. 6 and 7

Web of Science

(Includes Science Citation Index Expanded (SCI-EXPANDED), Social Sciences Citation Index (SSCI), Conference Proceedings Citation Index-Science (CPCI-S), Conference Proceedings Citation Index-Social Science & Humanities (CPCI-SSH), Conference Proceedings Citation Index-SS&H). Search date: 12 May 2015 (N = 13 records)

fragile X syndrome and acetylcarnitine

CINAHL Plus (EBSCOhost)

Search date: 12 May 2015 (N = 9 records)

  1. fragile X syndrome AND acetylcarnitine

LILACS (http://bases.bireme.br/cgi-bin/wxislind.exe/iah/online/?IsisScript=iah/iah.xis&base=LILACS&lang=e)

Search date: 12 May 2015 (N = 0 records)

fragile X syndrome AND acetylcarnitine

WorldCat (www.worldcat.org/)

Search date: 12 May 2015 (N = 28 records)

fragile X syndrome and acetylcarnitine

DART-Europe E-theses Portal (www.dart-europe.eu/basic-search.php)

Search date: 12 May 2015 (N = 0 records)

  1. fragile X syndrome

  2. acetylcarnitine

  3. fragile X syndrome and acetylcarnitine

Networked Digital Library of Theses and Dissertations (NDLTD; www.ndltd.org/)

Search date: 12 May 2015 (N = 0 records)

  1. fragile X syndrome

  2. acetylcarnitine

  3. fragile X syndrome and acetylcarnitine

Trove (http://trove.nla.gov.au/)

Search date: 12 May 2015 (N = 5 records)

fragile X syndrome and acetylcarnitine

ClinicalTrials.gov (www.clinicaltrials.gov/)

Search date: 12 May 2015 (N = 0)

fragile X syndrome and acetylcarnitine

metaRegister of Controlled Trials mRCT (www.controlled-trials.com/mrct)

Search date: 12 May 2015 (N = 0 records)

fragile X syndrome AND acetylcarnitine

WHO International Clinical Trials Registry (ICTRP) (www.who.int/ictrp/search/en/)

Search date: 12 May 2015 (N = 0 records)

fragile X syndrome AND acetylcarnitine

Appendix 2. Methods to be used in future updates of this review

Measures of treatment effect

We did not find outcomes measured using different scales or dichotomous outcomes. However, for future updates, if different studies use different scales for the same continuous outcome, we will calculate the standardised mean difference using Hedges' g, and if studies report dichotomous outcomes, we will calculate the risk ratio with 95% confidence intervals.

Dealing with missing data

If there are new RCTs that report missing data, we will present, investigate, and report the reasons, numbers, and characteristics of dropouts, exclusions from the analysis, or missing data.

We will use the following strategies to deal with missing data:

  • Contact original investigators to request missing data;

  • Assume that participants did not experience a change in their clinical outcomes; and

  • Perform sensitivity analyses, excluding studies with missing data.

If necessary, we will discuss the potential impact of missing data on the findings of the review in the Discussion section.

Assessment of reporting biases

If at least 10 studies are included for meta-analysis for a particular outcome measure, we will draw funnel plots (plotting trial effect against standard error) to investigate the likelihood of reporting biases.

Subgroup analysis and investigation of heterogeneity

In future updates of this review, when possible and necessary, we will conduct further investigation of the causes of heterogeneity using subgroup analyses for age groups, additional diagnosis of autism spectrum disorder or attention deficit hyperactivity disorder, and different doses of L-acetylcarnitine.

Sensitivity analysis

When data permit, we will conduct sensitivity analyses to assess the robustness of the results; that is, we will remove studies with particular characteristics and re-analyse the remaining studies to determine whether specific factors affect the results. We will conduct these analyses to examine the impact of:

  1. including only trials in which intention-to-treat analysis was done (i.e. all participants randomised to treatment arms are included in the outcome analysis (no per-protocol analysis trials));

  2. including only studies with complete data;

  3. removing studies with inconsistencies in the definition, measurement, or reporting of results (e.g. if the number of participants varies in the report or if measures were not taken at consistent time points for all participants); and

  4. regarding individual missing data, assuming that these people had not experienced a change in their clinical outcomes.

Contributions of authors

All review authors contributed to the development and writing of this review. JRR and MIT drafted the protocol. JRR, VG, and IS devised the search strategy and assessed the adequacy of the management of the risk of bias in the included studies. JB performed the statistical analyses of the effect of the interventions.

Declarations of interest

José-Ramón Rueda - none known.
Virginia Guillén - none known.
Javier Ballesteros - none known.
Maria-Isabel Tejada - none known.
Ivan Solà - none known.

Sources of support

Internal sources

  • University of the Basque Country, Spain.

  • Iberoamerican Cochrane Centre, Spain.

External sources

  • No sources of support supplied

Differences between protocol and review

The protocol stated that this review would present outcomes at several follow-up time points: 6 months, 7 to 12 months, and 13 to 24 months. However, since one of the trials only reported results at 12-months follow-up (Torrioli 1999), and it was not possible to obtain results for the other time points, we decided to present results at the last time point recorded for both trials only (that is 12 months).

We changed the last lines of the search strategy for Ovid MEDLINE(R) in order to combine three sets of terms (condition AND intervention AND RCTs).

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Torrioli 1999

Methods

RCT

Single centre: Italy

Study dates: Unknown

Participants

N: 20 boys with FXS

Ages: mean 9.2 years; range 6 to 13 years

Inclusion criteria: Boys with FXS

Exclusion criteria: None explicit

Interventions

Intervention: LAC at a dose of 50 mg/kg twice a day for a 1-year period

Control: placebo twice a day for a 1-year period

Outcomes
  1. IQ changes: measured with the WISC-R

  2. Visual-perceptive and graphics functions related to the extent of intellectual disability: measured with the Bender-Gestalt test

  3. Hyperactive behaviour: measured with the APTQ

A psychologist performed neuropsychological testing before treatment (TO), and at 1 month (T30), 6 months (T180), and 12 months (T365) after the beginning of treatment

Notes

Information from 'Letter to the Editor' section

Trial registration number: None available

Funding: None declared. Probably supported financially by Sigma-Tau Pharmaceuticals, which produced the drug under investigation. Employees of that company were part of the research team

Attempted to get additional information from authors of the study via email (21 January 2013; 14 May 2013), but received no response

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskQuote: "Twenty fragile-X boys...were assigned randomly" (p 366). Procedure not detailed in the report
Allocation concealment (selection bias)Unclear riskNothing mentioned in the report
Blinding of participants and personnel (performance bias)
All outcomes
Low riskQuote: "double-blind" (p 366)
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNothing mentioned in the report
Incomplete outcome data (attrition bias)
All outcomes
Low risk3 participants stopped the treatment after 1 month. 17 completed the study (8 of these were treated with LAC and 9 with placebo)
Selective reporting (reporting bias)Low riskProtocol of the study was not available, but assessed outcomes presented in the methods section
Other biasHigh riskEmployees of a company that produced the drug under investigation were part of the research team

Torrioli 2008

  1. a

    APTQ: Abbreviated Parent-Teacher Questionnaire
    CGI-P: Conners' Global Index - Parent Version
    CGI-T: Conners' Global Index - Teacher Version
    DSM-IV: Diagnostic and Statistical Manual of Mental Disorders - Fourth Edition
    FXS: fragile X syndrome
    IQ: intelligence quotient
    LAC: L-acetylcarnitine
    RCT: randomised controlled trial
    VABS-SF: Vineland Adaptive Behavior Scales - Survey Form
    WISC-R: Wechsler Intelligence Scale for Children - Revised

Methods

RCT

Multicentre: 8 centres in Italy, France, and Spain

Study dates: Unknown

Participants

N: 63 boys with FXS, recruited through genetic clinics

Ages: mean 9.18 years; range 6 to 13 years

Inclusion criteria:

  1. FXS diagnosis was confirmed by Southern blot

  2. Attention deficit hyperactivity disorder according to the DSM-IV

  3. Aged between 6 and 13 years

  4. Signed informed consent was obtained from a legal representative (children were also informed about aims and methodology of the research)

Exclusion criteria:

  1. Disorders of the central nervous system other than FXS, excluding epilepsy

  2. Had a malignancy or suffered from cardiovascular, renal, or other medical conditions that would compromise their safety or ability to comply with the protocol

  3. Had head trauma within the previous 12 months

  4. Suffered from malabsorption

  5. Physical or mental disability was so severe as to compromise the administration and evaluation of the tests

  6. Had known hypersensitivity to, or had been treated with, carnitine or its derivatives during the last 3 months

  7. Were being treated with drugs active on the central nervous system (stimulants, tranquillizers, hypnotics), excluding antiepileptic drugs

Interventions

Intervention: LAC at a dose of 500 mg twice a day for a 1-year period

Control: placebo twice a day for a 1-year period

Quote: "All patients participating in this study received extra care and stimulation both at school and at home, regardless of their taking LAC or placebo, which was unknown to parents and teachers. Moreover, the regular visits by neuropsychiatrists prompted a revision of the support activities that children received (speech therapy, physical therapy, occupational therapy, or other modality)" (p 809)

Outcomes

Cognitive and behavioural assessment:

  1. Hyperactivity and attention problems: measured with CGI-P and CGI-T, administered at T0 (baseline), T1 (1 month), T6 (6 months), and T12 (12 months). This index is a 10-item rating scale. High scores indicate hyperactivity and attention problems at home (CGI-P) or at school (CGI-T)

  2. Adaptative behaviour: measured with the Vineland Adaptive Behavior Scales Survey Form (VABS-SF). This instrument has an interview format that concerns 4 domains: communication, daily living skills, socialisation, and motor skills

  3. Verbal and non-verbal intellectual functioning: measured with the WISC-R

An interdisciplinary team of developmental specialists (child neuropsychiatrists and psychologists) evaluated alI enrolled participants on 4 occasions: at the start of the study (T0), after 1 month (T1), after 6 months (T6), and after 12 months (T12) of treatment

Notes

Quote: "The study was a Phase II study" (p 805)

Trial registration number: None available

Funding: Financially supported by Sigma-Tau Pharmaceuticals. Grants from Telethon. Employees of Sigma-Tau, which produced the drug under investigation, were part of the research team

Attempted to get additional information from authors of the study via email (21 January 2013; 14 May 2013), but received no response

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskQuote: "randomized" (p 805). Procedure not detailed in the report
Allocation concealment (selection bias)Unclear riskNothing mentioned in the report
Blinding of participants and personnel (performance bias)
All outcomes
Low riskQuote: "double-blind"; "their taking LAC or placebo,...was unknown to parents and teachers" (p 805; 809)
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNothing mentioned in the report
Incomplete outcome data (attrition bias)
All outcomes
Low riskQuote: "Out of the initial 63 patients enrolled in the study seven dropped out (four placebo and three LAC treated): five because of family choice (three because of problems in attending the evaluation sessions, two because of apparent increase of hyperactivity), one developed gynecomastia, and one had to start a psychoactive drug treatment and had to leave the study. It is important to note that the two patients who appeared to have increased hyperactive behavior and the one who developed gynecomastia later appeared to have received placebo. Fifty-six patients completed the study but the data of five of them (two placebo and three LAC treated) were not included in the statistical analysis because some time points were missing. Of the remaining 51 patients, we found that 24 had been treated with LAC and 27 with placebo once the randomization list became available" (p 806)
Selective reporting (reporting bias)Low riskProtocol of the study was not available, but assessed outcomes presented in the methods section
Other biasHigh riskEmployees of a company that produced the drug under investigation were part of the research team, and the study received funds from that company

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    ADHD: attention deficit hyperactivity disorder
    FXS: fragile X syndrome

Abbasi 2011aParticipants with ADHD but without a diagnosis of FXS. Compared treatments: methylphenidate plus placebo versus methylphenidate plus L-acetylcarnitine
Arnold 2007Participants with ADHD but without a diagnosis of FXS. Compared treatments: placebo versus L-acetylcarnitine
Van Oudheusden 2002Double-cross-over trial. Participants without FXS. Compared treatments: carnitine versus placebo