The diuretic bumetanide decreases autistic behaviour in five infants treated during 3 months with no side effects
The inhibitory transmitter GABA has been suggested to play an important role in infantile autistic syndrome (IAS), and extensive investigations suggest that excitatory actions of GABA in neurological disorders are because of a persistent increase of [Cl(-) ](I) . To test the effects of the chloride co-transporter NKCC1 diuretic compound Bumetanide that reduces [Cl(-) ](I) on IAS. Bumetanide was administered daily (1mg daily) during a 3-month period and clinical and biological tests made. We used 5 standard IAS severity tests - Childhood Autism Rating Scale, Aberrant Behaviour Checklist, Clinical Global Impressions; Repetitive and Restrictive Behaviour and the Regulation Disorder Evaluation Grid. We report a significant improvement in IAS with no side effects. Bumetanide decreases autistic behaviour with no side effects suggesting that diuretic agents may exert beneficial effects on IAS and that alterations of the actions of GABA may be efficient in IAS treatment calling for large scale randomized trials.
The diuretic bumetanide decreases autistic behaviour in ﬁve infants treated
during 3 months with no side effects
Eric Lemonnier (firstname.lastname@example.org)
, Yehezkel Ben-Ari
1.Centre de Resssources autisme de Bretagne, CHRU Brest Hopital Bohars, route de Ploudalmezeau, Bohars, France
2.INMED, INSERM U901, Universite
de la Me
e, Campus scientiﬁque de Luminy, Marseilles, France
Autism, Bumetanide, Chloride transport, Therapeutic
actions of a diuretic
Eric Lemonnier, Centre de Ressources Autisme de
Bretagne, CHRU Brest Hopital Bohars, route de Plo-
udalmezeau, Bohars, France.
Tel: 0033298015206 |
Fax: 0033298015233 |
27 April 2010; revised 1 June 2010;
accepted 23 June 2010.
The inhibitory transmitter GABA has been suggested to play an important role in infantile
autistic syndrome (IAS), and extensive investigations suggest that excitatory actions of
GABA in neurological disorders are because of a persistent increase of [Cl
Aims: To test the effects of the chloride co-transporter NKCC1 diuretic compound
Bumetanide that reduces [Cl
Methods: Bumetanide was administered daily (1 mg daily) during a 3 -month per-
iod and clinical and biological tests made. We used 5 standard IAS severity tests - Child-
hood Autism Rating Scale, Aberrant Behaviour Checklist, Clinical Global Impressions;
Repetitive and Restrictive Behaviour and the Regulation Disorder Evaluation Grid.
Results: We report a signiﬁcant improvement in IAS with no side effects.
Conclusion: Bumetanide decreases autistic behaviour with no side effects suggesting that
diuretic agents may exert beneﬁcial effects on IAS and that alterations of the actions of GABA may
be efﬁcient in IAS treatment calling for large scale randomized trials.
Infantile autistic syndromes (IAS) is a pervasive develop-
mental disorder characterized by impaired social interac-
tion, deﬁcits in verbal and nonverbal communication and
stereotyped interests and behaviours (1) as well as limited
interest in the surrounding environment associated with ste-
reotyped movements and repetitive plays (2). Research to
date indicates that a genetic predisposition may play a role
but one or more environmental factors must be in place for
symptoms to occur (3,4). It is suggested that genetic and
environmental hazards will alter developmental pro-
grammes leading to cortical and ⁄ or sub-cortical malforma-
tions and the formation of misplaced ⁄ misconnected
neuronal ensembles. The ﬁrst symptoms occur before
3 years of age with most likely an earlier origin. There is at
present no efﬁcient biological ⁄ pharmaceutical treatment to
Brain maturation is associated with a developmental
sequential expression of voltage gated, receptor synapse–
driven channels and brain patterns (5,6). The develop-
mental shifts of the actions of the inhibitory transmitter
GABA is but one example of these changes. Immature
neurons have a higher (Cl
than adults leading to para-
doxical excitatory actions of GABA (6). This is because
of an early expression of the co-transporter NKCC1 that
imports chloride and a late operation of KCC2 that
export chloride form neurons (7). In addition, the regula-
tion of (Cl
is altered by a variety of insults, lesions,
seizures and neurological disorders thereby converting
the actions of GABA from inhibitory to excitatory (8,9).
Consequently, diuretic agents that reduce (Cl
tute novel antiepileptic and neuro-protective agents
(10,11) and are currently being tested in large clinical tri-
als in infantile epilepsies (Nemo FP7 EU program and
GABA signalling is profoundly altered in IAS (12,13) and
see Discussion). This is reﬂected by the paradoxical effects
of GABA acting benzodiazepines on patients with IAS sug-
gesting dysfunction of the GABA signalling and excessive
chloride accumulations in neurons. The speciﬁc NKCC1
antagonist Bumetanide (Bum) (14) is a classical diuretic
that reduces (Cl
(15) and shifts GABA from excitation to
inhibition. Bum has been extensively utilized in adults since
1975 and in children since 1986 and its pharmacokinetic in
adults and children and its side effects are well known (16).
Bum is used in acute (oedema following head trauma) and
long-term conditions including broncho-pulmonary dyspla-
sia, nephritic syndromes or heart congestions (16) and in
neonatal rodent hippocampi (11). The use of Bum is safe
provided that it is accompanied with clinical and biological
surveillance in children.
We have now tested the effects of Bumetanide in ﬁve ran-
domly selected autistic infants with ongoing clinical and
biological surveillance. The diuretic was administered
(1 mg ⁄ 24 h – 0.5 mg twice a day) and the treatment contin-
ued for 3 months, a minimal duration considered to be sufﬁ-
cient for an evaluation of the effects on IAS. We report a
signiﬁcant improvement of the IAS manifestations in the
Acta Pædiatrica ISSN 0803–5253
ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica 2010 99, pp. 1885–1888 1885
children. These observations call for wide-range screening
of the use of Bum in IAS.
In June 2009, we obtained the authorization from the ad
hoc committee of the hospital of Brest to proceed with a
large double blind (60 children) randomized trial in IAS
using the diuretic bumetanide starting from December
2009. The trial received the agreement of the CPP (4th of
June 2009, number CPP west 6-570) and the national ad
hoc committee AFFSAPs (A90936-66 the 4th December
2009). With the consent of the parents, we initiated a pilot
study in ﬁve IAS children. Because of the striking results
obtained, we decided to present them, in order to encourage
similar large scale tests by other investigators.
MATERIAL AND METHODS
The material and methods are described in detail in supple-
mental material (supl. Acta Pediat Scand). In brief, children
were diagnosed by experienced clinical psychiatrist using
strict ICD-10 criteria for autistic disorder. These children
had no history of neurological disease (normal EEG)
Genetic tests systematically performed were negative indi-
cating no identiﬁable mutation (Caryotype and fragile X).
The ADI-R (17) was collected for all participants to conﬁrm
the diagnoses. A clinical and biological examinations
showed that none of the infants had a counter indication to
bum (including blood ionogram, transaminases, alkaline
phosphatases, uraemia, creatinemia, creatinine clearance,
cGT, glycemia notably). Because hypokalemia can induce
wave burst arrhythmia, an ECG was performed to ensure
that none of the patients had a lengthening of the QT
because they have a higher propensity to generate arrhyth-
mia. A clinical weekly surveillance was performed during
the ﬁrst, second and third month after treatment onset
including blood sodium and potassium 1 week and
2 months after treatment onset. None of the infants had
associated neurological disorders, and none was under
other treatment since at least 3 months.
To determine the possible therapeutic index efﬁcacy, we
relied on ﬁve classical behavioural determination of IAS
severity (details in supplemental material) including:
• The Childhood Autism Rating Scale (CARS) is a 15-
item rating scale that is used as a screening instrument
and to assess the changes in symptoms of autism over
time.(18) and can be used to determine alterations pro-
duced by a treatment (19) (DiLalla and Rogers, 1994)
and see the French version (20). The notation was
obtained during a session when the children were placed
in a game and animated discussion with the parents con-
cerning the behaviour of the child during the last week.
• The ABC (Aberrant Behaviour Checklist) is a ques-
tionnaire ﬁlled by the treating doctor during a discussion
with the parents (21).A French version has been used in
this study (22).
• The Clinical Global Impressions (CGI) is widely used
in the majority of clinical trials to examine disease sever-
ity and also on novel generation psychotic agents with
little side effects (23).
• RDEG the regulation disorder Evaluation grid is a
French scale of activity (96) that enables to detect the
level of dys-regulation, and the slowness of response of
the infants (2).
• The Repetitive and Restricted Behaviour (RRB) scale
(24) is a 35-item standardized checklist that allows item
rating on a 5-point scale from 0—the behaviour is never
expressed by the person—to —the behaviour is severely
expressed and characteristic of the person). Factorial
analysis produces four clinical meaningful factors, i.e.
sensori-motor stereotypes (F1), reaction to change (F2),
restricted behaviours (F3) and modulation insufﬁciency
Our patients were selected with no a priori from a large
group of IAS children placed in institutions or at home. A
summary of the patients is shown in Table 1. Three boys
used a functional language, whereas the remaining boy and
girl did not. The scores of ADI-R are above the threshold
conﬁrming the clinical diagnosis. Childs 1, 3, 5 follow a tra-
ditional school accompanied by an auxiliary person. Child 1
and child 2 are followed by a psychologist using the ABA
approach (once a week for child 1 and 3 times a week for
child 2). Child 2 has also two weekly orthophonic sessions
relying on picture exchange communication system. Child 3
has an orthophonic treatment weekly, and child 5 has no
treatment. Child 4 is treated in a medical institution specia-
lised in mentally retarded children. The test of bum was
made during the summer vacation, when behavioural ther-
apy and school were interrupted.
The scores of the different scales used before and after
3 months treatment are shown in the Supporting Table S1
(Supporting information). Before the treatment, four chil-
dren (1–4) had a CARS score above 36 indicating a
Table 1 Summary of patients included in the study. Four girls and one boy
aged between 3 years and 8 months to 11 years and 5 months with classical
autistic signs (F84.0 of ICD 10)
Age Sex Diagno ADI1 ADI2 ADI3 ADI4
1 8 years 11 months M F84.00 18 (10) 13 (8) 5 (3) 5 (1)
2 3 years 8 months F F84.00 20 (10) 8 (7) 4 (3) 5 (1)
3 8 years 7 months M F84.00 21 (10) 8 (7) 5 (3) 4 (1)
4 11 years 5 months M F84.00 24 (10) 14 (7) 4 (3) 5 (1)
5 10 years 1 months M F84.00 17 (10) 20 (8) 4 (3) 3 (1)
Bumetanide in autism Lemonnier and Ben-Ari
1886 ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica 2010 99, pp. 1885–1888
severe IAS. Child 5 showed a medium degree of autism.
Results show an improvement of the total scores of
CARS, ABC, RDEG and RRB for all children 3 months
after the treatment. CGI1 was not signiﬁcantly altered,
but this test concerns the severity of the disease that at
this stage does not reveal signiﬁcant changes. We also
observed a small global amelioration of CGI2 for the ﬁve
children. Patients 1, 2 and 5 had an index of 3 in CGI3
indicating a moderate action with no side effects. Patients
3 and 4 had an index of 2 indicating a minimal action
with no side effects. The number of items equal or above
3 with CARS was reduced by the treatment in the ﬁve
children. The sub-score of ABC5, was not altered by the
treatment. In the ﬁve children, ABC2, ABC3, ABC4,
RDEG dys-regulation, RDEG slowness, RRBF4, RRB F1
were ameliorated to a variable degree. In contrast, the
results of ABC1, RRB F2, RRB F3 are heterogeneous.
The average scores from grouped data for the four major
tests (total RRB, total ABC, total CARS and total RDEG)
are statistically signiﬁcant only in evolution score of total
RRB (55 ± 4.7 in control vs 33.4 ± 2.5 in bum treatment,
p < 0.05). However, all these scores are statistically signiﬁ-
cant in children 1, 2 and 3 [total CARS: 39 ± 0.8 in control
and 31.3 ± 2.4 in bum (p < 0.05); total ABC: 84 ± 2.9 in
control vs 56 ± 3.8 in bum (p < 0.05); total RRB: 61 ± 5.3
in control vs 29.7 ± 1.9 in bum (p < 0.01); total RDEG:
49.3 ± 6.1 in control vs 37.7 ± 5.8 in bum (p < 0.05)].
Therefore, bumetanide may be more effective in younger
patients (also see Discussion).
Starting 1 week after the treatment and once monthly, a
clinical surveillance was made including research of deshy-
dratation, orthostatic hypotension, hyper-sensitivity,
cramps, asthenia, diarrhoeas, myalgia, arthralgia, nausea,
dizziness. The levels of sodium and potassium remained sta-
ble (tests made a week and 2 months after the beginning of
the treatment). No adverse effect was found.
Present results suggest that bumetanide ameliorates
behavioural aspects of IAS, suggesting that the diuretic has
a global action. To the best of our knowledge, this is the ﬁrst
report raising the possibility of chloride alterations in aut-
Extensive observations suggest that GABA signalling play
a role in IAS including speciﬁc reduction of GABAergic
receptor systems, reduction of
[H] ﬂunitrazepam and
muscimol biding (ibid), reduced levels of glutamic acid
decarboxylase (reviewed in 25). Genetic forms of autism
include deletions of the proximal long arm of chromosome
15 where 3 GABA receptor subunit genes are located (26).
The dynamic alterations of (Cl
and shifts of GABA from
inhibitory to excitatory (10,11) after seizures and other
insults provide the conceptual basis for this study. Also,
patients treated with oxytocin—that alters (Cl
bit more appropriate social behaviour and affect (28).
The conclusions derived form our observations are
hampered by the lack of randomized double blind and
placebo investigations that are more prevalent in children
than adults (29). Clearly, wide-scale investigations are
needed to conﬁrm or inﬁrm the observations. Neverthe-
less, we were encouraged to present our observations
because bumetanide has no side effects, the dramatic
behavioural amelioration and the insistence of the par-
ents that their children are more present and their wish
to pursue the treatment. Interestingly, the same term of
presence was used by all parents.
The patients included in this study are quite heteroge-
neous in age, in the severity of the disease and age of detec-
tion. It is interesting that patients 4–5 were the least
responder to the treatment. These are older children, and
patient 5 goes to conventional school with clearly less
severe IAS, whereas patient 4 is severely affected and hospi-
talized in a specialized institution. Therefore, positive
effects of the diuretic may prevail in certain types of IAS but
this will require large trails such as the one presently being
It is not possible at present to determine whether bumeta-
nide exerts a preferential action on one aspect of the symp-
tomatology. The lack of effects of bum on
ABC5—inappropriate, excessive speech out of context—is
expected because amelioration in 3 months of speech is
unlikely to occur. In contrast, almost all scores were amelio-
rated to variable degrees stressing the general action of the
diuretic. Bumetanide provided a better cognitive regulation
in keeping with the improved presence reported by the par-
ents. The results of the subscales of ABC suggest an amelio-
ration of states of vigilance and social interactions,
stereotypic movements and hyperactivity again in keeping
with the notion of cognitive regulation. A wide range of
experimental investigations suggest that bum reduces sei-
zure severity (11). Bumetanide is currently being investi-
gated as a novel treatment for neonatal seizures (EU FP7
Our observations are compatible with the concept that
neurons who fail to respect the developmental pro-
gramme keep immature features, including possibly high
and other electrical and architectural properties
In conclusion, an emerging series of studies suggest that
chloride accumulates during brain maturation in relation to
various developmental malformations. Present observations
suggest that a conventional diuretic that reduces this accu-
mulation and acts to reinstate the inhibitory actions of
GABA exerts beneﬁcial actions in autism calling for more
detailed experimental and clinical studies on the links
between GABA ⁄ (Cl
The authors thank the ARAPI (Association pour la Recher-
che sur l’Autisme et la Pre
vention des Inadaptations) which
by organizing the autumn universities have been instrumen-
tal in helping this project take root. We are indebted to Prof
Barthelemy for criticism and to R Tyzio for his help and
Lemonnier and Ben-Ari Bumetanide in autism
ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica 2010 99, pp. 1885–1888 1887
1. World Health Organisation. Diagnostic Criteria for Research
(10th edition). Geneva, 1994.
2. Adrien JL, Rossignol-Deletang N, Martineau J, Couturier G,
Barthelemy C. Regulation of cognitive activity and early com-
munication development in young autistic, mentally retarded,
and young normal children. Dev Psychobiol 2001; 39: 124–36.
3. Patterson PH. Maternal infection: window on neuroimmune
interactions in fetal brain development and mental illness. Curr
Opin Neurobiol 2002; 12: 115–8.
4. Persico AM, Bourgeron T. Searching for ways out of the autism
maze: genetic, epigenetic and environmental clues. Trends
Neurosci 2006; 29: 349–58.
5. Ben-Ari Y, Gaiarsa JL, Tyzio R, Khazipov R. GABA: a pioneer
transmitter that excites immature neurons and generates primi-
tive oscillations. Physiol Rev 2007; 87: 1215–84.
6. Spitzer NC, Gu X, Olson E. Action potentials, calcium tran-
sients and the control of differentiation of excitable cells. Curr
Opin Neurobiol 1994; 4: 70–7.
7. Rivera C, Voipio J, Payne JA, Ruusuvuori E, Lahtinen H, Lamsa
K, et al. The K+ ⁄ Cl) co-transporter KCC2 renders GABA hy-
perpolarizing during neuronal maturation. Nature 1999; 397:
8. Cohen I, Navarro V, Clemenceau S, Baulac M, Miles R. On the
origin of interictal activity in human temporal lobe epilepsy in
vitro. Science 2002; 298: 1418–21.
9. Khalilov I, Le Van QM, Gozlan H, Ben Ari Y. Epileptogenic
actions of GABA and fast oscillations in the developing hippo-
campus. Neuron 2005; 48: 787–96.
10. Dzhala VI, Talos DM, Sdrulla DA, Brumback AC, Mathews
GC, Benke TA, et al. NKCC1 transporter facilitates seizures in
the developing brain. Nat Med 2005; 11: 1205–13.
11. Nardou R, Ben-Ari Y, Khalilov I. Bumetanide, an NKCC1
antagonist, does not prevent formation of epileptogenic focus
but blocks epileptic focus seizures in immature rat hippocam-
pus. J Neurophysiol 2009; 101: 2878–88.
12. Dhossche D, Applegate H, Abraham A, Maertens P, Bland L,
Bencsath A, et al. Elevated plasma gamma-aminobutyric acid
(GABA) levels in autistic youngsters: stimulus for a GABA
hypothesis of autism. Med Sci Monit 2002; 8: R1–6.
13. Garreau B, Herry D, Zilbovicius M, Samson Y, Guerin P,
Lelord G. Theoretical aspects of the study of benzodiazepine
receptors in infantile autism. Acta Paedopsychiatr 1993; 56:
14. Cohen M. Pharmacology of bumetanide. J Clin Pharmacol
1981; 21: 537–42.
15. Delpire E, Mount DB. Human and murine phenotypes associ-
ated with defects in cation-chloride cotransport. Annu Rev
Physiol 2002; 64: 803–43.
16. Sullivan JE, Witte MK, Yamashita TS, Myers CM, Blumer JL.
Pharmacokinetics of bumetanide in critically ill infants. Clin
Pharmacol Ther 1996; 60: 405–13.
17. Lord C, Rutter M, Le CA. Autism Diagnostic Interview-
Revised: a revised version of a diagnostic interview for caregiv-
ers of individuals with possible pervasive developmental disor-
ders. J Autism Dev Disord 1994; 24: 659–85.
18. Rogers SJ, Ozonoff S, Maslin-Cole C. Developmental aspects of
attachment behavior in young children with pervasive develop-
mental disorders. J Am Acad Child Adolesc Psychiatry 1993;
19. DiLalla DL, Rogers SJ. Domains of the Childhood Autism Rat-
ing Scale: relevance for diagnosis and treatment. J Autism Dev
Disord 1994; 24: 115–28.
B. Echelle d’e
valuation de l’autisme infantile-version
traduite (CARS-T). EAP ⁄ ECPA. 1989; Ref Type: Generic
21. Aman MG, Singh NN, Stewart AW, Field CJ. The aberrant
behavior checklist: a behavior rating scale for the assessment of
treatment effects. Am J Ment Deﬁc 1985; 89: 485–91.
22. Bouvard M. Liste des comportements aberrants- version tradu-
ite. EAP ⁄ ECPA. 2000; Ref Type: Generic
23. Guy W. Assessment Manual for psychopharmacology. Early
Clinical Drug Evaluation Unit. 2000. National institute of
Mental Health. Ref Type: Generic
24. Bourreau Y, Roux S, Gomot M, Bonnet-Brilhault F, Barthele-
my C. Validation of the repetitive and restricted behaviour
scale in autism spectrum disorders. Eur Child Adolesc
Psychiatry 2009; 18: 675–82.
25. Dossche DM. GABA in autism. international review of neuro-
biology 71, 1-481. 2005. Amsterdam: Academic Press. Ref
26. Schroer RJ, Phelan MC, Michaelis RC, Crawford EC, Skinner
SA, Cuccaro M, et al. Autism and maternally derived aberra-
tions of chromosome 15q. Am J Med Genet 1998; 76: 327–36.
27. Tyzio R, Cossart R, Khalilov I, Minlebaev M, Hubner CA,
Represa A, et al. Maternal oxytocin triggers a transient
inhibitory switch in GABA signaling in the fetal brain during
delivery. Science 2006; 314: 1788–92.
28. Andari E, Duhamel JR, Zalla T, Herbrecht E, Leboyer M, Sirigu
A. Promoting social behavior with oxytocin in high-functioning
autism spectrum disorders. Proc Natl Acad Sci U S A 2010;
29. Rheims S, Cucherat M, Arzimanoglou A, Ryvlin P. Greater
response to placebo in children than in adults: a systematic
review and meta-analysis in drug-resistant partial epilepsy.
PLoS Med 2008; 5: e166.
30. Ben-Ari Y. Neuro-archaeology: pre-symptomatic architecture
and signature of neurological disorders. Trends Neurosci 2008;
Additional Supporting Information may be found in the
online version of this article.
Table S1 Summary scores of the effects of bumetanide in
the ﬁve patients (C = before and Bum = 3 months after
Please note: Wiley-Blackwell is not responsible for the con-
tent or functionality of any supporting materials supplied by
the authors. Any queries (other than missing material)
should be directed to the corresponding author for the
Bumetanide in autism Lemonnier and Ben-Ari
1888 ª2010 The Author(s)/Acta Pædiatrica ª2010 Foundation Acta Pædiatrica 2010 99, pp. 1885–1888