Asian Journal of Psychiatry 44 (2019) 172–174

Contents lists available at ScienceDirect

Asian Journal of Psychiatry

journal homepage: [Link]/locate/ajp

Short communication

Lithium prophylaxis in early-onset Bipolar disorder: a descriptive study T

a b a b
Sandhiya Selvarajan , Harshini Manohar , Saibal Das , Priyanka Sakkarabani ,
Preeti Kandasamyb,
⁎

a
JIPMER, Department of Clinical Pharmacology, Dhanvantri Nagar, Puducherry, 605006, India
b
JIPMER, Department of Psychiatry, Dhanvantri Nagar, Puducherry, 605006, India

ARTICLE INFO ABSTRACT

Keywords: Subjects fulfilling DSM 5 criteria of Bipolar I disorder with onset before 18 years on Lithium prophylaxis were
Bipolar included. A total of 575 subjects with Bipolar Disorder were screened, 141 had early-onset Bipolar disorder, 72
Early-onset patients were on Lithium and 52 provided informed consent and entered the study. Thirty-four were in the
Adolescent prospective group, and 18 were in the retrospective arm. Mean age at onset was 16.13 (1.40) years. About 31%
Treatment
(n = 16) were initiated on Lithium following first episode. Mean age at initiating lithium was 19.73 (3.82) years.
Lithium
Clinical profile and treatment response of early-onset bipolar disorder in the Indian context needs further study.

1. Introduction 2. Methods

The global prevalence of the Bipolar disorder is 2−5%, and the peak The study was approved by the Institute Ethics Committee, JIPMER
age of onset falls during mid to late adolescence (Merinkangas and Tohen, and was registered with the clinical trials registry, India CTRI/2017/
2011). Bipolar disorder with very early-onset (< 13years) and early-onset 09/009836. Written informed consent was obtained from the parent
(< 18years), have a more malignant course with poor functional outcome and the patient (currently above 18 years). Assent was obtained from
(Perlis et al., 2004). Nearly, 50–80% of them experience one or more re- patients less than 18 years.
currences in the first five years, with the recurrence risk being high in the
first two years following the index episode (Pravin et al., 2005). This re- 2.1. Setting
sults in high rates of hospitalization, health service utilization, un-
employment, legal problems, and poor socio-academic functioning (Geller Patients attending Affective Disorder clinic of the Department of
et al., 2008; Jairam et al., 2008). Psychiatry, JIPMER, Puducherry.
Lithium, which is among the first-line maintenance treatments for
Bipolar disorder has shown to reduce recurrence, suicide and has 2.2. Sample
neuroprotective effects and is generally well-tolerated in pediatric
Mania (Dufy and Grof, 2018; Fountoulakis et al., 2012). There is recent The study has a prospective and retrospective group. The retro-
evidence that early initiation of lithium increases the probability of spective arm was planned as we expected a smaller cohort in the pro-
response (Kessing et al., 2014). Recent study report that nearly 63% of spective arm and possible attrition over two years follow-up.
Indian patients show a good response to treatment with lithium (Kapur For the prospective group patients with the diagnosis of Bipolar I
et al., 2019). However, upto 30–37% turn out to be lithium non-re- disorder as per DSM 5 diagnostic criteria with age at onset of the first
sponders in the Indian setting (Pravin et al., 2005; Kapur et al., 2019). episode before 18 years who were started on lithium as part of treat-
There is an existing gap in evidence regarding efficacy, tolerability, and ment-as-usual and who were willing for six-monthly follow up for a
acceptability of lithium in early-onset bipolar (Dufy and Grof, 2018). minimum of 2 years after stabilization of treatment were included. For
The current study examines the clinical profile of subjects with the retrospective group inclusion criteria: Patients with the diagnosis of
early-onset Bipolar disorder on Lithium Prophylaxis as part of treat- Bipolar I disorder as per DSM 5 diagnostic criteria, with age at onset of
ment-as-usual. This is part of an ongoing study on genomic biomarkers the first episode before 18 years who are on regular treatment with
in predicting prophylactic response to treatment in pediatric-onset bi- lithium and on follow up for a minimum of 2 years.
polar disorder.

⁎
Corresponding author.
E-mail address: preeti.k@[Link] (P. Kandasamy).

[Link]
Received 26 March 2019; Received in revised form 17 July 2019; Accepted 29 July 2019
1876-2018/ © 2019 Elsevier B.V. All rights reserved.
S. Selvarajan, et al. Asian Journal of Psychiatry 44 (2019) 172–174

2.3. Exclusion criteria Table 2

Frequency of clinical predictors of lithium response in early-onset bipolar dis-
Comorbid Intellectual disability or autism spectrum disorder or order.
patients with organic Bipolar disorder. In addition for the prospective S no Variables N (%)
group, women (> 18 years old at intake) who are planning pregnancy,
currently pregnant or lactating were excluded. 1 Manic polarity of first episode 36 (69.2)
2 Family H/o Bipolar 13 (25)
Mini-International Neuropsychiatric Interview (MINI) and Mini
3 Family H/o good response to Lithium 10 (19.2)
International Neuropsychiatric Interview for Children and Adolescents 4 Past H/o suicidal attempt 11 (21.2)
(MINI-KID) version 6.0 were used for structured evaluation to ascertain 5 Psychotic symptoms 12 (23.1)
the diagnosis of Bipolar I disorder in both arms. Non-response was 6 Mixed features 1 (1.9)
7 Rapid cycling 0 (0)
defined as recurrence in the 2 year follow up while on therapeutic dose
8 Comorbid substance use 0 (0)
of Lithium. For retrospective arm, the Retrospective criteria of long- 9 Comorbid anxiety 1 (1.9)
term treatment response in research subjects with Bipolar disorder was
employed (Schulze et al., 2010) to establish non-response. Serum Li-
thium, Thyroid function test, Blood Urea, and Creatinine were done at initiated following the first episode in about 30.77% (n = 16) of the
baseline and six-monthly thereafter as part of routine treatment care. sample and in 63.46% (n = 33) of participants lithium was started
during the second episode.
3. Results As the risk of recurrence is high in the first two years, 40 (76.9%)
subjects who had completed two years of treatment with lithium and on
A total of 575 subjects with Bipolar disorder were screened for regular follow-up were examined for recurrence of an episode. Eight
eligibility for participation. Of them, 141 had an onset of the first (20%) subjects had a recurrence. One out of 14, initiated on lithium
episode before 18 years. Sixty-four (45.39%) subjects were on other following first episode and 7 out of 26 started on lithium later during
mood stabilizers or antipsychotics. Among them, the most commonly the course had a recurrent episode. Among them non-response while on
prescribed was valproate (n = 39, 27.6%), followed by risperidone therapeutic dose was established in 4 subjects.
(n = 13, 9.2%), and carbamazepine (n = 12, 8.5%). Five patients were The profile of the non-responders although less in number was
excluded because of comorbid Intellectual disability. Of the 72 (51%) comparable to responders except that one subject had psychotic
patients with early onset bipolar on lithium, 52 provided informed symptoms, two others required antipsychotics during the maintenance
consent and entered the study. Thirty-four were in the prospective phase, two were started on lithium later during the course of illness.
group, and 18 were in the retrospective group.
Of the sample, 28 were male patients, and 24 were females. Thirty-
two subjects were students, n = 19 were school going, n = 5 in college, 4. Discussion
n = 8 were engaged in a semiskilled occupation; n = 6 were profes-
sionals and n = 6 were unemployed. Mean age at onset of bipolar The mean age in this sample was 16.13 years (Table 1) indicating
disorder and initiation of treatment is described in Table 1. Age at onset that majority of them had adolescent-onset as compared to western
ranged from 12 to 18 years with only one subject qualifying for very- literature where more than one-third of the patients are reported to
early onset Bipolar. The mean age at intake was 22.9 (SD = 7.12) years, have childhood-onset Bipolar (Perlis, 2009).
with a median duration of illness of 4 years. Mean years of schooling in First episode polarity was Mania in 69%, and this is similar to other
years was 11.83 (3). The prospective and retrospective arm was com- Indian studies that have reported predominant manic polarity in both
parable in terms of age at initiation of treatment, age at initiation of adult and juvenile-onset Bipolar disorder (Rangappa et al., 2016). The
lithium, the total number of episodes, and the number of episodes age at initiation of lithium was 3.6 years after the onset of illness, this
following lithium initiation (Table 1). needs to be critically reviewed, given the current evidence favoring
In 92.3% (n = 48) patients, the polarity of the first episode was early initiation of lithium (Kessing et al., 2016). Time to arrive at a
Mania, however only 6 had euphoric Mania and irritability was noted diagnosis of bipolarity may be delayed because of atypicality in clinical
as the predominant presentation. One participant had an episode with presentation in early-onset Bipolar. It is known that typical symptoms
mixed features, 12 had psychotic symptoms. With respect to number of increase with age and severity and initial episodes are often amorphous
episodes, 25% had only one episode, 38.6% had two episodes and in presentation and comorbidities complicate the clinical presentation
36.4% had more than two episodes. Family history and clinical pre- (Jairam et al., 2008; Kandasamy et al., 2016).
dictors of lithium response are listed in Table 2. Among those receiving diagnosis of Bipolar in the first episode, the
The mean dose of lithium was 845.19 (178.13), and Serum Lithium clinical predictors of response to lithium such as polarity of first epi-
was 0.796 (0.12). It was well tolerated with no adverse effects war- sode, presence of euphoria, family history of response, absence of
ranting a change of mood stabilizer. Renal and thyroid function test psychotic symptoms, substance abuse and rapid cycling (Ayano, 2016),
were within normal limits at baseline. During follow-up, one patient often employed in clinical setting may not favour administration of li-
was diagnosed to have hypothyroidism (TSH-14.2). thium in many patients with paediatric bipolar (Table 2). In this
During the maintenance phase, ten patients (19%) were on addi- sample, there were fewer participants if any with mixed features, rapid
tional antipsychotic, two were on benzodiazepines, and two were on a cycling or substance abuse, and about ten (19.2%) subjects had psy-
combination of antipsychotics and benzodiazepines. Lithium was chotic symptoms in contrast to western literature. Very few (n = 6;

Table 1
Clinical profile of early-onset bipolar disorder.
Clinical profile of EOBD Prospective (n = 34) Mean (SD)/n (%) Retrospective (n = 18) Mean (SD)/n (%) Total (n = 52) Mean (SD)/n (%)

Mean age at onset of Bipolar disorder (years) 16.26 (1.330 15.89 (1.53) 16.13 (1.40)
Mean age at initiation of medications for Bipolar (years) 17.29 (2.44) 18.11 (3.83) 17.58 (2.99)
Mean age at initiation of lithium (years) 19.91 (3.90) 19.39 (3.74) 19.73 (3.82)
Mean number of episodes (lifetime) 2.53 (1.9) 2.78 (2.18) 2.6 (1.98)
Number of episodes after starting Lithium 0.44 (1.73) 1.11 (2.4) 0.61 (2.02)

173
S. Selvarajan, et al. Asian Journal of Psychiatry 44 (2019) 172–174

11.5%) had classical euphoric Mania in contrast to adult onset Bipolar; Declaration of competing interest
however, the presence of irritable Mania did not determine non-re-
sponse to lithium. None.
It is however essential to clarify that this sample constitutes only
51% of those with early-onset Bipolar disorder as it included only Funding agency
subjects on lithium prophylaxis. Patients with atypical features could
have received Valproate or other mood stabilizers whose data is un- Funded by the Science and Engineering Research Board (SERB), a
available as part of this study. Future studies examining and comparing statutory body of Department of Science and Technology, Government
the clinical profile, response, and tolerability to various mood stabi- of India (File No. ECR/2016/001849).
lizers in early-onset Bipolar disorder may be indicated in our setting.
This will help address high rates of recurrence and poor functional Acknowledgement
outcome affecting the developmental and socio-academic trajectory of
these adolescents. As the peak incidence typically coincides with the We would like to acknowledge the contribution of (late) Prof Steven
secondary and higher secondary board exams in the Indian setting with A Dkhar, Professor of Pharmacology, and (late) Dr Surendiran Adhitan,
the negative long-term academic outcome, early prophylaxis may also Associate Professor of Pharmacology, JIPMER.
have unique relevance to the Indian setting.
One of the significant concerns in prescribing lithium is the varia- References
bility in response. Though the sample is small in this study, a subset
turned out to be partial responders and non-responders, and this is in Ayano, G., 2016. Bipolar disorders and lithium: pharmacokinetics, pharmacodynamics,
line with earlier studies which have reported that about 30–37% turn therapeutic effects and indications of lithium: review of articles. Austin J. Psychiatry
Behav. Sci. 3, 1053.
out to be non-responders (Pravin et al., 2005; Bellivier and Marie- Bellivier, F., Marie-Claire, C., 2017. Lithium response variability: new avenues and hy-
Claire, 2017; Kapur et al., 2019). Given the current lack of predict- potheses. In: Malhi, G.S., Masson, M., Bellivier, F. (Eds.), Science and Practice of
ability of response, it needs careful risk-benefit evaluation and facil- Lithium Therapy. Springer, pp. 157–178.
Dufy, A., Grof, P., 2018. Lithium treatment in children and adolescents. Review article.
itation of informed decision making by the parents (Kandasamy et al., Pharmacopsychiatry 51 (5), 189–193.
2016). Considering the ethical concerns in justifying a trial of lithium Fountoulakis, K.N., Kasper, S., Andreassen, O., Blier, P., Okasha, A., Severus, E., et al.,
for young patients with Bipolar disorder identifying predictive bio- 2012. Efficacy of pharmacotherapy in bipolar disorder: a report by the WPA section
on pharmacopsychiatry. Eur. Arch. Psychiatry Clin. Neurosci. 262, 1–48.
markers is necessary to improve the prophylactic management of early- Geller, B., Tillman, R., Bolhofner, K., Zimerman, B., 2008. Child bipolar I disorder: pro-
onset bipolar (Bellivier and Marie-Claire, 2017). While this is still un- spective continuity with adult bipolar I disorder; characteristics of second and third
derway, the decision of early initiation is an option that clinician should episodes; predictors of 8-year outcome. Arch. Gen. Psychiatry 65, 1125–1133.
Jairam, R., Hanstock, T., Cahill, C., Hazell, P., Walter, G., Malhi, G., 2008. The changing
provide families given the current evidence and recurrence risk.
face of bipolar disorder: adolescence to adulthood. Minerva Pediatr. 60, 59–68.
Kandasamy, P., Jairam, R., Srinath, S., 2016. Affective disorders in children and ado-
5. Limitations lescents: current status and controversies. In: In: Malhotra, S., Paramala, S. (Eds.),
Child and Adolescent Psychiatry- Asian Perspectives 1. Springer, pp. 51–62.
Kapur, V., Nadella, R.K., Sathur Raghuraman, B., Saraf, G., Mishra, S., Srinivasmurthy, N.,
The study has a relatively small sample size and included both a Jain, S., Del Zompo, M., Viswanath, B., 2019. Clinical factors associated with lithium
prospective and retrospective sample. As subjects on other mood sta- treatment response in bipolar disorder patients from India. Asian J. Psychiatr. 39
bilizers were excluded, there were limitations in comparing clinical (January), 165–168.
Kessing, L.V., Vradi, E., Andersen, P.K., 2014. Starting lithium prophylaxis early v. late in
profile and the response of patients on different mood stabilizers. The bipolar disorder . Brit. J. Psychiatry. 205, 214–220.
study employed MINI/MINI KID version 6. Merinkangas, K.R., Tohen, M., 2011. Epidemiology of bipolar disorder in adults and
children. In: Tsuang, M.T., Tohen, M.T., Jones, P.B. (Eds.), Textbook in Psychiatric
Epidemiology. John Wiley and Sons, Chichester, pp. 329–342.
6. Conclusion and future directions Perlis, R.H., Miyahara, S., Marangell, L.B., Wisniewski, S.R., Ostacher, M., DelBello, M.P.,
Bowden, C.L., Sachs, G.S., Nierenberg, A.A., STEP-BD Investigators, 2004. Long-term
The clinical profile of early-onset Bipolar disorder in the Indian implications of early onset in bipolar disorder: data from the first 1000 participants in
the systematic treatment enhancement program for bipolar disorder (STEP-BD). Biol.
setting may differ from western literature in terms of clinical pre-
Psychiatry 55 (9), 875–881.
sentation, comorbid patterns and therefore, treatment response. Pravin, D., Rajkumar, R.P., Prabhuswamy, M.Y., Srinath, S., Girimaji, S., Seshadri, S.P.,
Currently, there are limitations in predicting non-response to lithium 2005. Course and outcome of bipolar affective disorder in children. J. Indian Assoc.
Child Adolesc. Mental Health 1, 5.
clinically. Identifying genomic biomarkers may help predict treatment
Rangappa, S.B., Munivenkatappa, S., Narayanaswamy, J.C., Jain, S., Reddy, Y.C., 2016.
response to lithium and plan personalized care for young people with Predominant mania course in Indian patients with bipolar I disorder. Asian J.
this disorder in the early course of illness, given its unique properties. Psychiatr. 22, 22–27.
More research from Indian context examining the safety, tolerability Schulze, T.G., Alda, M., Adli, M., Akula, N., Ardau, R., Bui, E.T., et al., 2010. The
International Consortium on Lithium genetics (ConLiGen): an initiative by the NIMH
including neuro-cognitive effects of mood stabilizers among youth with and IGSLI to study the genetic basis of response to lithium treatment.
early-onset Bipolar disorder is warranted. Neuropsychobiology 62, 72–78.

174