You are on page 1of 5

Original Article

Prevalence of Cholera in Pediatric Patients with Acute


Dehydrating Diarrhea
Vijesh S. Kuttiat, Rakesh Lodha, Bimal Das and Utkarsh Kohli

Department of Pediatrics, All-India Institute of Medical Sciences, New Delhi, India

ABSTRACT
Objective. To estimate the prevalence of culture-confirmed cholera in patients with acute dehydrating diarrhea, at a tertiary
care center in north India, during a 6-month period from March to August, 2006.

Methods. We studied 145 children, who presented to the pediatric emergency services of a tertiary care teaching hospital
in north India with acute dehydrating diarrhea. Each patient had his/her stool sample collected for Vibrio cholerae culture and
hanging drop preparation for darting motility. The stool specimen for hanging drop analysis was immediately transported to
the emergency laboratory, where a trained technician prepared the slides and examined them for darting motility characteristic
of Vibrio cholerae.

Results. V. cholerae was isolated in 36 (24.8%) patients. Forty-nine (33.7%) patients had a positive hanging drop examination.
Hanging drop examination had a sensitivity and specificity of 85.8% and 81.7%, respectively. Severe dehydration (OR 4.3;
P<0.01) and hanging drop positivity (OR 12.42; P<0.001) were associated with higher odds of cholera after adjustment for
other risk factors.

Conclusion. Cholera is an important cause of acute watery diarrhea in pediatric patients in urban north India and should be
ruled out in all children presenting with acute dehydrating diarrhea, particularly those with severe dehydration. Hanging drop
test is useful for diagnosis in the emergency setting. [Indian J Pediatr 2010; 77 (1) : 67-71] E-mail: utkarshkohli @hotmail.
com

Key words : Cholera; Pediatric; Acute dehydrating diarrhea

Diarrhea is a major cause of morbidity and mortality in is one of the few causes of childhood diarrhea where
the developing world. Cholera is an important contributor antibiotics are indicated and are known to improve
to the burden of diarrheal diseases, and its incidence outcome.6
varies across the developing world ranging from 0.5/
The present study was carried out to estimate the
1000/year in Indonesia to 4/1000/year in Mozambique.1
prevalence of culture-confirmed cholera in patients with
Incidence of cholera in India has been estimated to be 1.6/
acute dehydrating diarrhea at a tertiary care centre in
1000 population/year1 or 40/1000 cases of acute diarrhea.2
north India, during a 6-month period from March to
Cholera is a significant contributor to morbidity in the
August, 2006. We also evaluated were stool hanging drop
pediatric population in India and other developing
preparation slides from these patients for a secondary
countries. 1-3 Recently, concern has been raised about
analysis to determine the sensitivity and specificity of this
widespread under-reporting of cholera cases in the
rapid and inexpensive diagnostic test to accurately
developing world, particularly in south Asia, due to social
diagnose cholera.
and economic considerations. 4, 5 More importantly,
routine screening for cholera is not incorporated in our
national diarrheal diseases control program, leading to MATERIAL AND METHODS
under-diagnosis of this important condition.4 Implications
of under-diagnosis of cholera are potentially serious as it
We enrolled 145 patients, who presented to the pediatric
emergency services of a tertiary care teaching hospital in
Correspondence and Reprint requests : Dr Utkarsh Kohli, MD. north India with acute dehydrating diarrhea between
Research Fellow, Division of Clinical Pharmacology, Department of March and August, 2006. Children presenting during the
Medicine, Vanderbilt University Medical Center, Nashville, duty hours of the participating physician team were
Tennessee-37203, USA.
considered for enrollment. All children presenting with
[Received March 22, 2009; Accepted September 23, 2009]

Indian Journal of Pediatrics, Volume 77January, 2010 67


Vijesh S. Kuttiat et al

diarrheal symptoms were independently assessed by two 2 loose stools/hr and >2 loose stools/hr. Analyses were
physicians for signs of dehydration as per World Health performed with the statistical software STATA10
Organization (WHO) criteria7 and the patients were only (StataCorp, College Station, TX).
included if the findings of both the physicians were in
agreement.
RESULTS
Each patient with acute dehydrating diarrhea had his/
her stool sample collected for Vibrio cholerae culture and
hanging drop preparation for darting motility. The stool One hundred and forty-five patients (median age: 2 year;
samples were mixed with Venkataraman-Ramakrishnan interquartile range: 1/6-12 year; 80 males) were enrolled
(VR) transport medium and transported to the in the present study. The median duration of diarrhea for
bacteriology laboratory within 8 hours of collection as per at the time of presentation was 24 (2-120) hr, and the
WHO recommendations.8 Vibrio cholerae remains viable in median stool frequency (stools per hr) in the 8 hours prior
this medium for upto 6 weeks after stool collection and to presentation was 0.73 (0.12-3) loose stools/hr. Thirty-six
the medium can be stored at room temperature after (24.8%) patients had culture positivity and 49 (33.7%)
collection.8 The stool specimen for hanging drop analysis patients had a positive hanging drop examination.
was immediately transported to the emergency Hanging drop examination had a sensitivity and
laboratory, where a trained technician prepared the slides specificity of 85.8% and 81.7%, respectively, for detecting
and examined them for darting motility characteristic of a patient with positive Vibrio cholerae culture (Table 1).
Vibrio cholerae. All enrolled subjects were managed as per All the subjects had watery stools; 10.5% had typical
standard WHO case management guidelines.7 rice watery stools as assessed by the enrolling physicians.
The stool specimens were cultured directly on TABLE 1.. Demographic Characteristics
thiosulphate citrate bile salt sucrose (TCBS) agar and
Parameter (n=145) Median (IQR)
incubated at 37o Celcius for 16-20 hours. Pre-enrichment of
stool samples in alkaline peptone broth was also carried Age (yrs) 2 (1/6-12)
out with subsequent subculture on thiosulphate citrate Sex Male/Female 80/65
bile salt sucrose (TCBS) agar. Sucrose fermenting yellow Duration (hrs) 24 (2-120)
Frequency (stools/hr) 0.73 (0.12-3)
colored colonies were selected and further identified as V. Consistency Watery/Rice Watery* 129/15
cholerae by biochemical tests. Final confirmation was done Vomiting Yes/No* 125/19
using a slide agglutination test with specific anti-sera Fever Yes/No** 36/107
against V. cholerae (Denka Seiken, Tokyo, Japan). Dehydration Some/Severe 76/69
Antibiotic sensitivity testing was carried out on Mueller Hanging Drop Positive/Negative 49/96
Stool Culture Positive/Negative 36/109
Hinton agar (BecktonDickinson, USA) by Kirby-Bauer
disc diffusion method using Clinical and Laboratory * Information could not be recorded for 1 child
Standards Institute (CLSI) guidelines.9 Antibiotics used in ** Information could not be recorded for 2 children
the sensitivity testing were ampicillin (10 g), tetracycline
(30 g), septran (trimethoprim- sulfamethoxazole) (30 g),
nalidixic acid (30 g) and chloramphenicol (30 g).
Eschericia coli ATCC 25922 strain was used as quality
control strain. Antibiotic sensitivity profile of each isolate
was recorded as resistant, intermediate and sensitive
based on the zone diameter according to the guidelines of
CLSI.9
Statistical analysis: Data is expressed as median and
interquartile range from all normally distributed data are
expressed as mean S.D. Unadjusted and adjusted odds
ratios for having cholera, as defined by a positive stool
culture for Vibrio cholerae, were calculated using logistic
regression models. Covariates included severity of Fig. 1. Month wise break-up of dehydrated and Vibrio cholerae
culture positive patients.
dehydration, stool hanging drop positivity, stool
consistency, and stool frequency. Continuous variables Proportion of both culture positive and hanging drop
like stool frequency and stool duration were recoded as positive patients was significantly higher in patients with
categorical variables. Stool frequency per hour in 8 hours rice watery than watery stools (46.7% vs 21.7% [P=0.033]
prior to presentation to the pediatric emergency was for stool culture and 60% vs 30.2% [P= 0.021] for hanging
recoded as a categorical variable and patients were drop examination, respectively). Approximately half
divided into 3 categories: those with < 1 loose stool/hr, 1- (47.5%) of the patients had severe dehydration at the time

68 Indian Journal of Pediatrics, Volume 77January, 2010


Prevalence of Cholera in Pediatric Patients with Acute Dehydrating Diarrhea

TABLE 2. Risk Factors for Culture-Confirmed Cholera with Their Respective Odds Ratios.

Risk factor Unadjusted analysis Adjusted analysis


Odds ratio P Odds ratio P

Rice water diarrhea 3.15 0.04 1.55 0.58


Severe dehydration 4.97 <0.001 4.3 0.008
Hanging drop positive 18.43 <0.001 12.42 <0.001
Fever 0.40 0.08 0.77 0.71
Sex 0.55 0.12 0.89 0.83
Frequency/hr 1-2$ 1.07 0.89 0.91 0.89
Frequency/hr >2$ 6.46 0.04 4.3 0.25
Duration 24-48 hrs* 0.63 0.38 0.58 0.43
Duration >48 hrs* 0.24 0.03 0.36 0.25

*Baseline group: Duration less than 24 hr


$
Baseline group: Frequency/hr < 1

of enrollment. Proportion of both culture positive and as defined by culture positivity, was found in pediatric
hanging drop positive patients was significantly higher in patients with acute dehydrating diarrhea during the
patients with severe dehydration than some dehydration period from March to August, 2006. While current
(39.1% vs 11.8% [P=0.0001] for stool culture and 43.4% vs national guidelines do not recommend routine screening
23.6% [P=0.007] for hanging drop, respectively). A third for cholera in patients with acute dehydrating diarrhea,
of the patients (33.7%) had a positive stool hanging drop the present study data presents a strong case in favor of
preparation. Stool culture positivity was significantly routine screening of children with acute dehydrating
higher in patients with positive hanging drop preparation diarrhea for cholera presenting to the hospital.
(P<0.0001), and hanging drop positivity was the most
Hanging drop examination had a sensitivity of 85.8%
important indicator of culture confirmed cholera
and a specificity of 81.7% in the patients. It is inexpensive
(Adjusted odds ratio=12.4; P<0.001).
and can be rapidly performed in an emergency setting,
The unadjusted odds ratio (OR) for having cholera was and hence, can play an important role in rapid
6.5 times higher in subjects with >2 loose stools/hr than identification of cholera patients for appropriate therapy.
those with <1 loose stool/hr and 1-2 loose stools/hr Stool cultures reveal the pattern of Vibrio cholerae strains
(P=0.04). Stool frequency did not have a significant effect prevalent in the community and their antibiotic
on odds for having cholera after correction for other susceptibility and thus help in guiding appropriate
important risk factors like dehydration (P=0.25). antimicrobial therapy. The majority of the Vibrio cholerae
isolates were resistant to commonly prescribed antibiotics
Those who presented with diarrheal duration of more
like trimethoprim-sulfamethoxazole, nalidixic acid and
than 48 hrs had significantly lower odds in favor of
ampicillin. Interestingly, almost half of the cultured Vibrio
cholera than those who presented within 48 hrs of onset of
cholerae were sensitive to chloramphenicol and
diarrhea (OR=0.24; P=0.03). This effect was lost after
tetracycline (Table 3). These results indicate a change in
adjustment for other risk factors like hanging drop
antibiotic susceptibility pattern of circulating Vibrio
positivity, severity of dehydration and stool frequency
cholerae strains over the last decade with increased
(P=0.25) (Table 2). Also analyzed was the sensitivity
resistance to commonly prescribed antibiotics like
pattern for the isolates of Vibrio cholerae (Table 3). The
nalidixic acid.10
majority of the isolates were resistant to trimethoprim-
sulfamethoxazole, nalidixic acid, and ampicillin. While the hanging drop examination was able to detect
approximately 50% of culture positive cholera in earlier
DISCUSSION studies10, 11 it had high sensitivity and specificity in the
present study patients. This could be attributed to
A very high prevalence (24.8%) of culture positive cholera, immediate transportation of stool samples to the
emergency laboratory and proper training of the
TABLE 3. Antibiotic Sensitivity Pattern of Vibrio cholerae Isolates.
technicians. Prevalence of cholera in pediatric patients
Antibiotic Sensitive (%) Partially Resistant (%) seeking emergency care for acute watery diarrhea has
Sensitive (%) been estimated to be 10% in an earlier hospital-based
study from the same city.11 The proportion of dehydrated
Ampicillin 6 (17.1) 1 (2.9) 28 (80)
Tetracycline 19 (54.3) - 16 (45.7) patients in this study was not reported. 11 Since patients
Septran 4 (11.4) - 31 (88.6) with acute dehydrating diarrhea, were studied the
Nalidixic acid 1 (3) - 33 (97) prevalence of Vibrio cholerae is higher in the present study.
Chloramphenicol 23 (67.7) 1 (3) 10 (29.4) The present results further support the notion that cholera

Indian Journal of Pediatrics, Volume 77January, 2010 69


Vijesh S. Kuttiat et al

is an important cause of morbidity in pediatric patients study, the proportion of Vibrio Cholerae culture positive
with acute diarrhea in urban Delhi, especially those who patients ranged from a low of approximately 20% in May,
have concomitant dehydration. June and August to a high of 38.2% in July. These findings
are highly significant as there was no ongoing cholera
More importantly, data from the present study and
epidemic in Delhi during the study period. However,
other studies suggest a significant increase in cholera
since the study was carried out in the peak cholera season
prevalence over last 15 years. Stool culture positivity for
it is possible that true prevalence of cholera for the year
cholera was found to be 0.35% in acute watery diarrhea
2006 could be lower.3, 17 Also, patients who had received
patients from the same city in 1995. The proportion of
antibiotic therapy prior to presentation to the study site
dehydrated patients was not reported in this study.11
and fulfilled other inclusion criteria were included in the
Due to lack of financial resources and trained study. Antibiotic therapy prior to obtaining stool culture
manpower, laboratory facilities for the accurate diagnosis may reduce the yield stool culture for Vibrio cholerae and
of cholera are not widely available in the developing bias the prevalence estimates towards a lower value.
world. Therefore, efforts have been made to identify Therefore, it is possible that some patients with cholera
clinical parameters that are predictive of cholera, in were not picked up on stool culture and the true
patients presenting with acute watery diarrhea, in order prevalence of culture-confirmed cholera in the present
to start them on appropriate fluid and antimicrobial study population was higher.
therapy. Earlier studies have identified risk factors such as
age more than 24 mth, history of hospitalization for
CONCLUSION
diarrhea, short duration of illness, rice water stools, high
purge rate (>12 in 24 hr), severe dehydration, and
vomiting (>4/hr), as clinical predictors of a higher cholera Cholera is an important cause of acute watery diarrhea in
risk in patients with acute watery diarrhea. 10, 12 While children in urban north India and should be ruled out in
clinical parameters like rice water diarrhea, severe all children presenting with acute dehydrating diarrhea,
dehydration, stool frequency >2/hr and duration of particularly those with severe dehydration. The hanging
illness < 48 hr were predictive of higher risk of cholera in drop test is useful for diagnosis in emergency setting.
the present study patients, only severe dehydration was
Contributions: RL, BD, VSK and UK were involved in planning the
associated with a higher cholera risk after adjustment for study. Data collection and data entry was done by VSK and UK. UK
other risk factors (including hanging drop positivity). analyzed the data and drafted the manuscript. BD carrid out the
microbiological studies.
Currently available guidelines do not recommend
microscopy of stool in children with acute diarrhea.13, 14 Conflict of Interest: None of the authors have a interest relevant to
There has been emphasis on the clinical presentation for the work presented.
diagnosis. However, the present study suggests that Role of Funding Source: No external funding.
clinical features have limited use. In contrast, a simple
laboratory test, stool hanging drop examination is helpful
in predicting cholera in children presenting with acute REFERENCES
dehydrating diarrhea, particularly those with severe
dehydration.
1. Deen JL, von Seidlein L, Sur D, Agtini M, Lucas ME, Lopez
Cholera is known to cause severe diarrhea that can be AL et al. The high burden of cholera in children: comparison
rapidly dehydrating, and these sick patients are more of incidence from endemic areas in Asia and Africa. PLoS Negl
Trop Dis 2008;2: e173.
likely to be referred to tertiary care centers like ours for
2. Sur D, Deen JL, Manna B, Niyogi SK, Deb AK, Kanungo S et
management. Therefore, these results should not be al. The burden of cholera in the slums of Kolkata, India: data
extrapolated to other settings including the outpatient from a prospective, community based study. Arch Dis Child
clinics. Also, we did not ascertain the socio-economic 2005; 90: 1175-1181.
status of the patients. Cholera is more prevalent in 3. Singh J, Sachdeva V, Bhatia R, Bora D, Jain DC, Sokhey J.
individuals of lower socio-economic status as these Endemic cholera in Delhi, 1995: analysis of data from a
sentinel centre. J Diarrhoeal Dis Res 1998;16:66-73.
groups have inadequate access to clean drinking water 4. Steffen R, Acar J, Walker E, Zuckerman J. Cholera: assessing
and proper sanitation. 15, 16 Patients from lower socio- the risk to travellers and identifying methods of protection.
economic strata are more likely to seek care from a facility Travel Med Infect Dis 2003;1: 80-88.
like the present one where medical care is provided at a 5 Zuckerman JN, Rombo L, Fisch A. The true burden and risk of
nominal cost. Thus, the prevalence of data obtained from cholera: implications for prevention and control. Lancet Infect
Dis 2007; 7: 521-530.
the cohort may not reflect the community prevalence of
6. Alam S, Bhatnagar S. Current status of anti-diarrheal and anti-
cholera. secretory drugs in the management of acute childhood
diarrhea. Indian J Pediatr 2006; 73: 693-696.
The present study was carried out in March to August
7. Diarrhoea. In WHO, ed. Pocket book of hospital care for children:
which is the peak cholera season in Delhi. In the present guidelines for the management of common illnesses with

70 Indian Journal of Pediatrics, Volume 77January, 2010


Prevalence of Cholera in Pediatric Patients with Acute Dehydrating Diarrhea

limited resources. Geneva; WHO, 2005; 109-130. Task Force. Consensus Statement fo IAP National Task Force:
8. WHO. WHO SEARO, 2006. http://www.searo.who.int/en/ status report on management of acute diarrhea. Indian Pediatr
Section10/Section17/Section53/Section482_1792.htm (27 April 2004; 41: 335-348.
2006, date last accessed). 14. Bhatnagar S, Lodha R, Choudhury P, Sachdev HP, Shah N,
9. Wikler MA. Clinical and Laboratory Standards. In Narayan S et al. IAP Guidelines 2006 on management of acute
Performance standards for antimicrobial susceptibility testing: diarrhea. Indian Pediatr 2007; 44: 380-389.
sixteenth informational supplement. Wayne, Pa; Clincial and 15. Olago D, Marshall M, Wandiga SO, Opondo M, Yanda PZ,
Laboratory Standards Institute, 2005;26: M100-S15. Kanalawe R et al. Climatic, socio-economic, and health factors
10. Amin V, Patwari AK, Kumar G, Anand VK, Diwan N, Peshin affecting human vulnerability to cholera in the Lake Victoria
S. Clinical profile of cholera in young childrena hospital basin, East Africa. Ambio 2007; 36: 350-358.
based report. Indian Pediatr 1995; 32: 755-761. 16. Mahalanabis D, Faruque AS, Albert MJ, Salam MA, Hoque SS.
11. Gupta S, Faridi MM. Cholera pattern in children of Delhi. An epidemic of cholera due to Vibrio cholerae O139 in Dhaka,
Indian Pediatr 2005; 42: 90-91. Bangladesh: clinical and epidemiological features. Epidemiol
12. Fukuda JM, Yi A, Chaparro L, Campos M, Chea E. Clinical Infect 1994; 112: 463-471.
characteristics and risk factors for Vibrio cholerae infection in 17. Sharma NC, Mandal PK, Dhillon R, Jain M. Changing profile
children. J Pediatr 1995; 126: 882-886. of Vibrio cholerae O1, O139 in Delhi and its periphery (2003-
13. Bhatnagar S, Bhandari N, Mouli UC, Bhan MK. IAP National 2005). Indian J Med Res 2007; 125: 633-640.

Indian Journal of Pediatrics, Volume 77January, 2010 71