A Pilot Study on the Combination of Applied Behavior Analysis and Bumetanide Treatment for Children with Autism

Abstract

Objective:

The purpose of this study was to investigate the therapeutic effects of combined bumetanide and applied behavior analysis (ABA) treatment in children with autism.

Methods:

Sixty children diagnosed with autism according to the International Classification of Diseases, Tenth Revision (ICD-10) criteria (mean age of 4.5 years) were randomly divided into two groups: A single treatment group (n=28) and a combined treatment group (n=32). The combined treatment group received ABA training combined with oral bumetanide (0.5 mg twice a day). The single treatment group received ABA training only. Autism symptoms were evaluated with the Autism Behavior Checklist (ABC) and the Childhood Autism Rating Scale (CARS), whereas severity of disease (SI) and global improvement (GI) were measured with the Clinical Global Impressions (CGI). Assessment of ABC, CARS, and CGI was performed immediately before and 3 months after initiation of the treatment(s).

Results:

Prior to intervention(s) no statistically significant differences in scores on the ABC, CARS, SI, or GI were found between the two groups. Total scores of the ABC, CARS, and SI were decreased in both groups after 3 months (p<0.05) compared with the scores prior to treatment. The total scores of the ABC and the CGI were significantly (p<0.05) lower in the combined treatment group than in the single treatment group. Although the total and item scores of the CARS in the combined treatment group were lower than in the single treatment group after a 3 month intervention, they did not reach statistical significance. No adverse effects of bumetanide were observed.

Conclusions:

Treatment with bumetanide combined with ABA training may result in a better outcome in children with autism than ABA training alone.

Introduction

Autism spectrum disorders (ASD) are neurodevelopmental disorders that are characterised by deficits in social interactions and communication, and stereotyped and repetitive behavior (Stanković et al. 2012). The current prevalence for ASD is estimated to be ∼30–100 per 10,000 (Rutter et al. 2008). The burden caused by this disease is high, not only to those families affected, but also to society. Costs associated with ASD are estimated to be $3,200,000 per individual in the United States (Ganz 2007). Because the etiology of autism is still undetermined, current treatment for autism relies mostly on different forms of treatment that are based on the principles of learning theory, and on treatment of comorbid disorders such as epilepsy. High intensive therapies (>20 hours per week), based on behavioral psychological theories such as applied behavior analysis (ABA), can bring about meaningful and positive changes in behavior. ABA has been documented in numerous studies as an effective method for reducing interfering maladaptive behaviors in children with autism (Myers and Johnson 2007; Walsh 2011). However, treatment effects of ABA vary significantly among individuals, with some cases showing limited response.

There is also no clear evidence for the effectiveness of medication treatment for autism. Second generation antipsychotic medications such as risperidone and aripiprazole are the only United States Food and Drug Administration (FDA)-approved medications for the treatment of target symptoms (e.g., irritability, rigid behaviour) in autism. Both agents have serious side effects such as weight gain, drowsiness, prolactinemia, and tremors. Therefore, they are only indicated for moderate to severe behavioral problems associated with autism, as there is no clear benefit in treating the core symptoms of autism (Jesner et al. 2007; Sharma and Shaw 2012). This has led to the development of new types of drugs, targeting different receptors in the brain. New agents targeting γ-aminobutyric acid (GABA) are considered to be very promising for the treatment of ASD. GABA is one of the major brain neurotransmitters, in addition to glutamate, which maintains normal functioning of the brain. The GABA system has been suggested to play an important role in autism (Cellot and Cherubini 2014).

The excitatory GABA neuron that has been found in patients with ASD because of the high level [Cl^-]_I can be reversed by NKCC1 inhibitor. Bumetanide is one of the selective blockers of the NKCC1 chloride importer. Lemonnier and Ben-Ari reported that bumetanide may have decreased autistic behaviour without serious adverse effects in five children with autism (Lemonnier and Ben-Ari 2010). Moreover, in a follow-up double blind controlled trial in 60 children with ASD, Lemonnier et al. found significant improvements in children treated with bumetanide (Lemonnier et al. 2012). Because this is the first report showing the efficacy of bumetanide treatment in autism, replication, especially in a non-Caucasian sample, is warranted.

The ideal model to investigate this would be to perform a double-blind placebo- controlled study. This would require a very large study with a large investment.

We therefore set out to perform an open study comparing a treatment that has been proven to have positive effects, with treatment combined with bumetanide. We hypothesized that the combinational intervention of ABA and oral bumetanide would be more effective than ABA training alone in treating children with autism.

Materials and Methods

Participants

Sixty subjects (51 boys and 9 girls) who met International Classification of Diseases, Tenth Revision (ICD-10) criteria for ASD, from 2.5 to 6.5 years of age with a mean age of 4.5 years (95% confidence interval [CI]: 2.7–6.3) were included. None of the subjects had received any treatment at least 3 months prior to inclusion. Written informed consent was obtained from the parents of each participant. The research protocol was approved by the Ethics Committee of the First Hospital of Jilin University. Participants were randomly divided into two groups: A single treatment group (n=28), who received ABA, and a combined treatment group (n=32) who received both ABA and bumetanide treatment. Subjects received ABA training every day, for 30–40 minutes. Those receiving additional bumetanide treatment were administered bumetanide at a dose of 0.5 mg, twice a day.

Only subjects without a history of seizures, abnormal karyotype, Fragile X syndrome or congenital metabolic disorder were included. Children with any abnormal changes in the routine blood test, routine urine test, liver function test, renal function test, blood electrolytes, blood glucose test, or electrocardiogram (ECG) were excluded.

Assessment tools

Behavioral assessment included the Autism Behavior Checklist (ABC) (Marteleto and Pedromônico 2005), the Childhood Autism Rating Scale (CARS) (Breidbord and Croudace 2013) and the Clinical Global Impressions (CGI) (Busner and Targum 2007) scale. The CGI includes scores of severity of illness (SI) and of efficacy index (EI). The assessment of the ABC, CARS, and CGI was performed by well-trained physicians, with expertise in ASD, before and after 3 months of intervention in both groups. Raters were not blind to both groups.

Monitoring of adverse effects

Results of laboratory tests such as routine blood test, routine urine test, liver function test, renal function test, blood electrolytes, blood glucose, and ECG were examined before and after 3 months of intervention in the treatment group. Possible adverse effects were routinely checked in each session.

Statistical analysis

All data were analyzed with SPSS 13.0 software (SPSS, Chicago, IL) and expressed as mean±SD. Descriptive data were analyzed with the χ² test and the t test, depending upon the type of distribution. Multivariate analysis and paired t tests were used to compare the data concerning scores before and after treatment of both groups. A value of p<0.05 was considered to be statistically significant.

Results

Data of general characteristics of all participants

A t test for independent samples was used to compare possible group differences concerning age, age of onset, duration, maternal age, paternal age, and birth weight (Table 1). For the same purpose, the χ² test has been used for gender, positive family history, parity, delivery type, birth asphyxia, history of crawling, and language development. There was no statistically significance between the two groups (p>0.05) in general characteristics.

Table 1.

Data of General Characteristics of Two Groups

Items	CG(n=28)	TG(n=32)	p value
Age (year)	4.50±1.67	4.60±1.90	0.881
Age of onset (month)	27.0±10.2	26.9±8.50	0.271
Duration (month)	16.5±14.3	16.6±11.5	0.194
Maternal age (year)	27.2±5.82	27.8±5.13	0.675
Paternal age (year)	28.3±5.83	29.6±5.26	0.876
Birth weight (kg)	3.50±0.40	3.60±0.50	0.528
Gender (%)
Male	24 (85.7)	27 (84.4)	0.917
Female	4 (14.3)	5 (15.6)
Positive family history (%)
No	22 (78.6)	25 (78.1)	0.193
Yes	6 (21.4)	7 (21.9)
Parity (%)
G1P1	18 (64.3)	21 (65.6)	0.389
Others	10 (35.7)	11 (34.4)
Delivery type (%)
Natural childbirth	8 (28.6)	9 (28.1)	0.563
Cesarean delivery	20 (71.4)	23 (71.9)
Birth asphyxia (%)
Yes	6 (21.4)	8 (25.0)	0.499
No	22 (78.6)	24 (75.0)
History of crawling (%)
Normal	22 (78.6)	25(78.1)	0.335
Abnormal	6(21.4)	7(21.9)
History of language development (%)
Delayed type	20 (71.4)	22 (68.8)	0.433
Regressive type	8 (28.6)	10 (31.2)

Postdate expressed as mean±SD.

TG, treatment group; CG, control group.

Data of assessment tools (ABC, CARS, CGI)

After treatment, the total and item scores of the ABC, CARS and CGI showed general improvement that reached statistical significance in both groups (Table 2). The combination treatment group showed statistically significant (p<0.05) better treatment scores on the ABC and CGI, and a nonsignificant better treatment outcome on the CARS.

Table 2.

Scores of ABC, CARS, and CGI Before and After Treatment

	CG(n=26)		TG(n=29)
	Before treatment	After treatment	Before treatment	After treatment	p _a-value	p _b-value
Scores for ABC
Sensory stimuli sensorial	11.48±2.80	9.24±2.36^D	11.47±3.88	8.38±3.18^D	0.993	0.261
Interaction	21.84±3.37	18.30±3.14^D	20.36±3.58	16.03±3.01^D ^*	0.110	0.006
Body and object use	12.08±3.92	10.60±3.53^D	13.25±5.86	9.41±4.86	0.388	0.305
Social self-care	10.60±3.44	8.56±2.63^D	11.22±3.77	8.50±3.44^D	0.514	0.942
Language	18.68±4.23	16.12±3.59^D	19.31±5.75	15.15±4.40^D	0.645	0.370
Total score	74.68±4.47	62.88±4.96^D	75.61±11.38	57.47±7.19^D ^*	0.700	0.003
Scores for CARS
Personal relationships	2.84±0.68	2.28±0.45^D	2.78±0.51	2.21±0.41^D	0.707	0.517
Imitation	2.92±0.57	2.12±0.44^D	2.92±0.69	2.26±0.44^D	0.984	0.222
Emotional response	2.48±0.51	2.16±0.37^D	2.47±0.61	2.09±0.45^D	0.958	0.520
Body motor ability	2.72±0.54	2.20±0.40^D	2.61±0.55	2.09±0.38^D	0.447	0.283
Relation to inanimate objects	2.52±0.51	2.12±0.44^D	2.36±0.63	2.06±0.48^D	0.305	0.622
Adaptation to environmental changes	2.28±0.84	1.88±0.33^D	2.08±0.55	1.68±0.59^D	0.275	0.126
Visual response	2.40±0.50	2.04±0.20^D	2.36±0.59	2.12±0.32^D	0.790	0.298
Auditory response	2.32±0.55	2.08±0.49^○	2.44±0.73	1.94±0.34	0.477	0.207
Near sensory responses	2.08±0.81	1.84±0.62^○	2.11±0.85	1.82±0.57^D	0.887	0.917
Anxiety	1.88±0.60	1.86±0.52	1.83±0.56	1.68±0.53^D	0.757	0.151
Verbal communication	2.72±0.54	2.20±0.40^D	2.72±0.45	2.18±0.37^D	0.986	0.822
Non verbal communication	2.32±0.47	2.08±0.27^○	2.17±0.44	2.06±0.34	0.205	0.801
Hyperactivity	2.60±0.50	2.16±0.37^D	2.61±0.55	2.15±0.35^D	0.936	0.894
Intellectual function	2.36±0.70	2.24±0.59	2.28±0.74	2.24±0.74	0.665	0.979
Overall impression	2.88±0.52	2.36±0.49^D	2.83±0.50	2.24±0.43^D	0.729	0.304
Total score	37.28±3.53	31.64±1.72^D	36.39±3.85	30.79±2.45^D	0.362	0.304
Scores for CGI
SI	4.48±0.82	3.40±0.50^D	4.47±1.05	3.17±0.38^D#	0.975	0.042
GI	-	2.08±0.27	-	1.89±0.39^#	-	0.043
EI	-	2.88±0.33	-	3.11±0.39^#	-	0.020

Postdate expressed as mean±SD between two groups.

Versus prior to treatment in each group (paired t test) p<0.01.

○

Versus prior to treatment in each group (paired t test) p<0.05.

Versus after treatment between two groups (one way ANOVA) p<0.01.

Versus after treatment between two groups (one way ANOVA) p<0.05.

ABC, Autism Behavior Checklist; CARS, Childhood Autism Rating Scale; CGI, Clinical Global Impressions; TG, treatment group; CG, control group; p _a value, p value before treatment; p _b value, p value after treatment; SI, severity of illness; GI, global improvement; EI, efficacy index.

Follow-up and monitoring side effects

During the 3 month follow-up, five subjects choose no longer to participate in the study. For two subjects, this was related to socioeconomic problems. Two other subjects discontinued because of lack of treatment efficacy. One subject discontinued because of polyuria. No laboratory abnormality was found in the treatment group, in the routine blood test, routine urine test, liver function test, renal function test, blood electrolytes, blood glucose test, or ECG.

Discussion

The present study investigated the therapeutic effects of combined bumetanide and ABA treatment in children with autism. By comparing treatment with ABA alone with the combined treatment condition, it was found that combined therapy may be more effective than single therapy with ABA. At the same time, both types of therapy showed positive effects. No serious adverse effects of bumetanide treatment occurred during the 3 month follow-up.

Although there are a large number of studies that show the effectiveness of ABA in autism treatment, this is the first study to report on ABA treatment in combination with bumetanide treatment.

Bumetanide is an agent that only recently has been studied as potential form of treatment in ASD. Bumetanide is a chloride-importer antagonist that can reduce intracellular chloride to reinforce GABAergic inhibition. Tyzio et al. have shown in their study that treatment with bumetanide for pregnant VPA rats and Fragile X mice shortly before delivery inhibits the excitatory action of GABA and prevents autistic-like behavior in offspring in both animal models (Tyzio et al. 2014). Also, the NKCC1 blocker bumetanide could reduce the excitatory action of GABA in epileptic neurons, which also indicates that bumetanide should be useful in the treatment of autism (Dzhala et al. 2005; Nardou et al. 2009). Both features are in accordance with Lemonnier and Ben-Ari's study in five autistic children (Lemonnier and Ben-Ari 2010). Lemonnier's further research shows that chronic bumetanide treatment significantly improved accuracy in facial emotional labelling and increased brain activation in areas involved in social and emotional perception. Side effects were restricted to an occasional mild hypokalemia that was treated with supplemental potassium (Lemonnier et al. 2012).

A possible explanation for the improved results of combined therapy of bumetanide and ABA could be that bumetanide provides a biological stability (e.g., GABAergic inhibition) at the cellular level that allows the brain to become more susceptible to ABA. It is important to mention that the GABAergic system develops during a person's lifetime; therefore, the effects of aging may play a role (Lehmann et al. 2012). We suspect that without ABA treatment, the effects of bumetanide will be less pronounced.

Limitations

Ideally, a double-blind placebo-controlled study with four arms consisting of a control group, a group receiving ABA, a group receiving bumetanine, and a group receiving ABA and bumetanine should be used, with a long-term follow-up. Before such an enterprise is undertaken, we set out to gather more evidence for bumetanide treatment in a combination therapy. As such, limitations are the relative short duration of follow-up, the limited sample size, and the fact that an open treatment design was used.

Conclusions

This is the first study to show enhanced effects of combined ABA and bumetanide treatment in a non-Caucasian ASD sample. Importantly, bearing in mind the limitations of this study, no serious side effects of bumetanide treatment were observed.

Clinical Significance

Our results, together with previous results on bumetanide treatment, are promising. Given the burden of ASD, double-blind placebo-controlled trials need to be performed as soon as possible.

Footnotes

Disclosures

No competing financial interests exist.

References

Breidbord

, Croudace

: Reliability generalization for childhood autism rating scale. J Autism Dev Disord, 43:2855–2865, 2013.

Busner

, Targum

: The clinical global impressions scale: Applying a research tool in clinical practice. Psychiatry (Edgmont), 4:28–37, 2007.

Cellot

, Cherubini

. GABAergic signaling as therapeutic target for autism spectrum disorders. Front Pediatr, 2:70, 2014.

Dzhala

, Talos

, Sdrulla

, Brumback

, Mathews

, Benke

, Delpire

, Jensen

, Staley

: NKCC1 transporter facilitates seizures in the developing brain. Nat Med, 11:1205–1213, 2005.

Ganz

: The lifetime distribution of the incremental societal costs of autism. Arch Pediatr Adolesc Med, 161:343–349, 2007.

Jesner

, Aref–Adib

, Coren

:Risperidone for autism spectrum disorder. Cochrane Database Syst Rev, 24:CD005040, 2007.

Lehmann

, Steinecke

, Bolz

. GABA through the ages: regulation of cortical function and plasticity by inhibitory interneurons. Neural Plast, 2012:892784, 2012.

Lemonnier

, Ben-Ari

: The diuretic bumetanide decreases autistic behaviour in five infants treated during 3 months with no side effects. Acta Paediatr, 99:1885–1888, 2010.

Lemonnier

, Degrez

, Phelep

, Tyzio

, Josse

, Grandgeorge

, Hadjikhani

, Ben-Ari

: A randomized controlled trial of bumetanide in the treatment of autism in children. Transl Psychiatry, 2:e202, 2012.

10.

Marteleto

, Pedromônico

: Validity of Autism Behavior Checklist (ABC): Preliminary study. Rev Bras Psiquiatr, 27:295–301, 2005.

11.

Myers

, Johnson

: American Academy of Pediatrics Council on Children with Disabilities: Management of children with autism spectrum disorders. Pediatrics, 120:1162–1182, 2007.

12.

Nardou

, Ben-Ari

, Khalilov

: Bumetanide, an NKCC1 antagonist, does not prevent form,ation of epileptogenic focus but blocks epileptic focus seizures in immature rat hippocampus. J Neurophysiol, 101:2878–2888, 2009.

13.

Rutter

, Bishop

, Pine

, Scott , Scott

, Stevenson

, Taylor

, Thapar

: Rutter's Child and Adolescent Psychiatry, 5th ed. Part IV: Clinical Syndromes, 46 Autism Spectrum Disorders. Hoboken, NJ: John Wiley & Sons, 2011.

14.

Sharma

, Shaw

: Efficacy of risperidone in managing maladaptive behaviors for children with autistic spectrum disorder: A meta-analysis. J Pediatr Health Care, 26:291–299, 2012.

15.

Stanković

, Lakić

, Ilić

: Autism and autistic spectrum disorders in the context of new DSM-V classification, and clinical and epidemiological data. Srp Arh Celok Lek, 140:236–243, 2012.

16.

Tyzio

, Nardou

, Ferrari

, Tsintsadze

, Shahrokhi

, Eftekhari

, Khalilov

, Tsintsadze

, Brouchoud

, Chazal

, Lemonnier

, Lozovaya

, Burnashev

, Ben-Ari

: Oxytocin-mediated GABA inhibition during delivery attenuates autism patho- genesis in rodent offspring. Science, 343:675–679, 2014.

17.

Walsh

: The top 10 reasons children with autism deserve ABA. Behav Anal Pract, 4:72–79, 2011.