Epidemiological and antibiotic susceptibility profiles of infectious bacterial diarrhoea in Juba, South Sudan

Abstract

Background: Diarrhoeal diseases have remained a major health problem in South Sudan where they accounted 45% prevalence in under five-year olds. Between 2006 and 2007, the country reported a morbidity of 8,337 cases and 176 deaths due to diarrhoeal outbreaks.

Methodology: We investigated causative agents of diarrhoeal diseases and their antibiogram in persons presenting with diarrhoea to selected health facilities in Juba.

Results: Bacterial agents were prevalent in 20(6.9%) of the 286 patients with 5.7%(4/70) in under five-year olds alone. S. dysenteriae 50% (10/20) accounted for the majority of the identified pathogens followed S. flexneri 25% (5/20) and S. typh 25% (5/20). Antibiotic testing showed that S. flexneri (5/5) and S. typhi (5/5) were all 100% sensitive to ceftriaxone, and gentamicin while S. dysenteriae had varying sensitivity to ciprofloxacin (70%), nalidixic acid (90%), and ceftriaxone(100%).  These pathogens had 100% resistance to amoxicillin, ampicillin, tetracycline and cotrimoxazole. No difference existed in isolation rates among different age groups, educational status, gender, water drank, use of chlorine, toilet use, exposure at home to diarrhoea patient, hand washing with soap and location of residence.  However, diarrhoeagenic bacteria isolation was higher for participants with no source of income (OR=6.08, p<0.05).

Conclusion: With emerging menace of resistance to commonly used antibiotics in South Sudan we recommend antibiotic resistance monitoring and regulation of antibiotic use.

Introduction

Diarrhoeal diseases are a major health problem in developing countries [1] and accounts for an approximate global mortality of two million people annually [2]. Globally, 88% of diarrhoea cases are attributable to unsafe drinking water, inadequate sanitation or insufficient hygiene.  In Africa, an average morbidity rate of 912.9 million diarrhoeal episodes per year in children has been reported with four out of 10 deaths annually caused by diarrhoeal disease [3].  Diarrhoeal diseases are also common among travelers to tropical areas [4]. Escherichia coli, Rotavirus, Salmonella spp., Shigella spp., Campylobacter jejuni, Entamoeba histolytica, and Giardia lamblia are the common diarrheogenic pathogens [5, 6, 7]. Ample fluid and electrolyte replacement and maintenance is the cornerstone for management of diarrheal illness [8]. Suitable antimicrobial treatment shortens bacterial excretion and clinical periods, but the incidence of multidrug-resistance is rising [9].

            South Sudan, with a reported 45% prevalence of diarrhoea in children aged under five years of age, has experienced outbreaks in the recent past. In 2009 and 2010 there were 68,983 and 147,071 diarrhoeal cases with 0.09 and 0.08 case fatality rates (CFRs) respectively. In April 2006 alone, there were 5,108 diarrhoea cases and 98 deaths (CFR: 1.92%) in Juba City [10]. From January-June 2007, there were 3,157 diarrhoeal cases with 74 deaths (CFR: 2.34%) [11]. These outbreaks could be due to the influx of people from neighboring countries following the comprehensive peace agreement (CPA). Movement of people is associated with spread of disease and antibiotic resistance [12]. To our knowledge, no diarrhoea-causing bacterial pathogens were identified in these outbreaks due to facilities for bacterial culture and isolation being unavailable. This study determined the prevalence, distribution and antibacterial susceptibility profiles of enteropathogenic bacteria causing diarrhoea in patients attending selected health facilities in Juba City.

Materials and methods

Study design, site, target population and sample size

 A cross-sectional study was conducted in Juba Teaching Hospital (JTH), Al Sabah children’s hospital, and Malakia, Kator, Munuki and Nyakuron primary health centres (PHC). JTH is the main teaching and referral health facility in South Sudan with 516 beds. A total of 286 stool specimens or rectal swabs from randomly selected adults and children with diarrhoea respectively were collected between September and December 2012 , inoculated into Cary Blair media (Oxoid®, UK) and transported in a triple package container at 2⁰C-8⁰C for laboratory analysis. Diarrhoea was defined as the passage of at least three loose or liquid stool in 24 hours. Patients who could not consent or had known underlying causes other than bacteria were left out.

Isolation and identification of bacterial pathogens

Stool specimens were emulsified in peptone water, selenite F broth, Alkaline Peptone Water (APW) and incubated at 35-37⁰C for 6-8 hours. Peptone water-emulsified specimens were inoculated into Karmali medium, Sorbitol MacConkey (SMAC) and Deoxycholate Citrate agar (DCA) and Hektoen enteric agar (HE) for Campylobacter, E. coli O157 and Shigella spp isolation respectively, using sterile plastic loops. Selenite F broth emulsified specimens were inoculated into DCA and HE for Salmonella isolation. APW-emulsified specimens were inoculated into Thiosulfate Citrate Bile Salts sucrose agar for Vibrio cholerae isolation and sub-cultured in Heart infusion agar (HIA). Inoculated plates were incubated at 35-37⁰C except Karmali plates that were incubated at 42⁰C in 5% carbon dioxide atmosphere for 18- 24 hours. The plates were examined for growth and pure isolates with colonial characteristics of the target diarrhoeagenic bacteria were sub-cultured and identified using specific agglutinating antisera (Remel® UK) and API20E® (Biomerierux, UK).

Antibacterial susceptibility testing

Antibacterial susceptibility was determined by Kirby Bauer disc diffusion method using Mueller–Hinton (MH) agar and results interpreted according to Clinical Laboratory Standards Institute (CLSI, 2011). Commonly used antibacterials as per South Sudan treatment guidelines were selected for sensitivity testing.

Ethical consideration, data collection and analysis

Ethical approval and specimen shipment clearance was obtained from the Ministry of Health. Data were collected on standard structured questionnaire, entered, cleaned and analyzed by Epi Info® 7 (CDC, Atlanta, USA). Generated data were kept in a password protected computer. Personal identifiers were removed to ensure confidentiality.

Results

Demography and characteristics of the participants

We recruited 286 participants, 167 (58.4%) females and 119 (41.6%) males, mean age 19.81±14.8; range 2-71 years, between August and November 2011 from selected health facilities (HFs). The majority of the participants were from Munuki 143/286 (50.0%), followed by Kator and Juba (11.2%). Only 12 (4.2%) participants reached higher education with certificates or higher, and 70 (24.5%) were employed – see Table 1.  The participants presented with either diarrhoea with blood or other diarrhoea.

 

Table 1. Distribution and bacterial culture results for diarrhoeagenic bacterial among study participant attending selected health facilities in three payams of Juba City, South Sudan.
Payam (District)	Participants		Bacterial Isolation
	n (%)		(-)	(+)
			n (%)	n (%)
Juba	32(11.18)		29	3
Kator	111(38.81)		106	5
Munuki	143(50)		131	12
Education level and employment
Higher certificate and above	12(4.2)		12(4.5)	0(0.0)
Secondary	57(19.93)		50(18.9)	7(35.0)
Primary	84(29.37)		81(30.6)	3(15.0)
Uneducated	132(49.15)		122(46.0)	10(50.0)
Unemployed	216(75.5)		198(74.5)	18(90.0)
Employed	70(24.5)		68(25.)	2(10.0)
Health facilities		Sex
Al sabah	47(16.43)	female	26	2
		male	18	1
JTH	55(19.23)	female	32	1
		male	20	2
Kator	61(21.33)	female	28	4
		male	29	0
Melikia	26(9.09)	female	13	2
		male	11	0
Munuki	58(20.28)	female	28	3
		male	27	0
Nyakuron	39(13.64)	female	26	2
		male	8	3
All participants (286)		Female	153(53.49%)	14(4.89%)
		Male	113(39.52%)	6(2.09%)
(+)=culture positive and (-) = culture negative for diarrhoeagenic bacteria.

 

Bacterial isolates and isolation rates

A total of 286 stool specimens were analyzed and diarrhoeagenic pathogens isolated from 6.99% (20/286) of the specimens. The prevalence in under five-year olds was 5.7% (4/70) and adults 7.4% (16/216). The pathogen specific prevalence were S. typhi (1.7%), S. flexneri (1.7%) and S. dysenteriae (3.5%). Nonpathogenic gut flora grew in 93.01%(266/286) of total specimens. E. coli 18% (48/266) and Pseudomonas ssp 82% (216/266) were isolated (see Figure 1 No E. coli O157 was isolated from SMaC and other E. coli isolates were not analyzed further for presence of other pathogenic strains. The isolates of enteric bacteria pathogens had similar distribution trends across all ages as shown in Figure 2.

 

Figure 1. Proportionate Isolation rate of bacteria among the study participants attending selected health facilities in Juba City.

 

Figure 2. Age group specific prevalence of diarrhoeagenic bacterial among study participant attending public health facilities in Juba City.

 

Antibiotic phenotypic susceptibility profiles

Antibiotic resistance in the three (3) different species of pathogenic bacteria of 30-100% was found to amoxicillin, ampicillin, chloramphenicol, tetracycline and co-trimoxazole. Although all isolates were however ceftriaxone susceptible, quinolone resistance was identified with intermediate susceptibility to ciprofloxacin in the majority of Shigella isolates and nalidixic acid resistance in all of the S. typhi isolates. 

Table 2. Antibiotic and bacterial strain specific distribution of   resistance in bacteria isolated from patients attending selected health facilities in Juba City, South Sudan
	S. Dysenteriae			S. Flexneri			S. Typhi
Antibiotics	S	I	R	S	I	R	S	I	R
CIP	7	3	-	-	5	-	5	-	-
NAL	9	1	-	-	5	-	-	-	5
AMP	-	-	10	-	-	5	-	1	4
CRO	10	-	-	5	-	-	5	-	-
CHL	2	1	7	-	-	5	2	-	3
GEN	3	4	3	5	-	-	5	-	-
TET	-	2	8	-	-	5	-	-	5
SXT	-	1	9	-	-	5	-	-	5
Age groups
0-5	3			0			1
6-15	2			2			1
16-25	1			1			1
26-35	1			2			1
34-45	2			0			1
46+	1			0			0
R=Resistant, S=sensitive, I=intermediate CIP=ciprofloxacin, NAL=nalidixic acid, AMP=ampicillin, CRO=ceftriaxone, CHL=chloramphenicol, GEN=gentamicine, TET=tetracycline, SXT=co-tremoxazole

Bivariate analysis

A bivariate analysis for association of the most plausible factors with stool culture results was done but did not yield any independent significant association - see Table 3. Reporting either having any other diarrhoea or diarrhoea with blood by the participants did not have any association with their education level, employment status and sex. Having a source of income was significantly associated (OR=6.08, p<0.05) with stool culture results - see Table 3.

 

Table 3. Bivariate analysis of factors for significance of association among the participants     attending selected health facilities in Juba City, South Sudan
Variables (exposures)		Culture Results		OR (95% CI)	P-Value
		(+v)	(-v)
Payam of Residence
Juba	(+)	3	152	0.71(0.19, 2.57)	0.84
	(-)	17	116
Kator	(+)	5	106	0.50(0.17, 2.57)	0.28
	(-)	15	160
Munuki	(+)	12	131	1.55(0.61, 3.90)	0.48
	(-)	8	135
Female Male		14	153	1.95(0.72, 5.24)	0.39
		6	113
Toilet facility	(+)	9	82	1.84(0.73, 4.60)	0.22
	(-)	11	184
Drink tank water	(+)	17	199	1.91(0.54, 6.71)	0.42
	(-)	3	67
Use Chlorine	(+)	10	123	1.16(0.47, 2.89)	0.82
	(-)	10	143
Use soap & water	(+)	2	18	1.53(0.33, 7.21)	0.92
	(-)	18	248
Employment	(+)	2	68	0.32(0.07, 7.21)	0.19
	(-)	18	198
No Income some income		16	152	6.08(1.56, 39,77)	<0.05
		2	116
Education	(+)	10	143	1.17(0.47, 2.90)	0.91
	(-)	10	122
Household exposure	(+)	4	51	0.98(0.32, 3.07)	0.79
	(-)	16	207
Female Male		55	112	0.84(0.51, 1.41)	0.61
		84	35
(+) Exposure to the variable, (-) No exposure to the variable ,(+v) culture positive, (-v) culture negative

 

Discussion

Infectious diarrhoea is a major public health problem in South Sudan. This study established the prevalence and antimicrobial resistance profiles of bacterial pathogens isolated from stool of persons with diarrhoea. S. typhi (1.7%), S. flexneri (1.7%) and S. dysenteriae (3.5%) were the enteric pathogens isolated with an overall prevalence of 6.99%. This prevalence was lower than that of Ifeanyi [13] in Nigeria probably due to difference in socioeconomic indices and the exclusion of E. coli for determination of our prevalence. The prevalence of diarrhoeagenic bacteria in under five-year olds was 5.7% (4/70) and in adults 7.4% (16/216), demonstrating that everyone is at risk. This concurred with the findings of Samie et al [14] in South Africa and Sabrina et al [15] in Tanzania.

            The distribution and isolation rates observed in this study were lower for S. flexenri but similar for S. typhi and S. dysenteriae compared to a related study by Jafari et al [16] in Tehran. These could be due to seasonal differences and geography [13]. Little difference existed in pathogenic bacteria isolation rates among different age groups, educational status, gender, the water drank, use of chlorine, use of toilet, exposure at home to diarrhoea patient, use of water and soap for washing hands, and location of residence.  However, odds of diarrhoeagenic bacteria isolation was higher for participants with no source of income as compared to those with some income (OR=6.08, p<0.05).

            Evaluation of antibiotic susceptibility profiles revealed that S. flexneri 5(100%) were all sensitive to ceftriaxone, nalidixic acid, gentamicin and resistant to amoxicillin, ampicillin, tetracycline and co-trimoxazole. This is not unusual as Ruslan et al [17] also found out that S. flexneri was resistant to ampicillin, co-trimoxazole, chloramphenicol and tetracycline. S. dysenteriae isolates were sensitive to ciprofloxacin (70%), nalidixic acid (90%), ceftriaxone (100%), while resistance was observed to amoxicillin (100%), ampicillin (100%), co-trimoxazole (90%), chloramphenicol 7(70%) and tetracycline 8(80%).  A similar result was obtained in a related study in Kenya [18,19]. S. typhi 5(100%) were sensitive to ciprofloxacin, and ceftriaxone but 100-90% resistant to tetracycline, amoxicillin, ampicillin, cotrimoxazole, gentamycin nalidixic acid although Willie et al [18] reported a lower resistance to nalidixic acid at 44%.

Conclusion

This study has established enteropathogenic bacteria as a significant cause of diarrhoea among children and adults irrespective of their ages as evident by 6.99% prevalence. Although it is difficult to draw firm conclusions about antibiotic susceptibility patterns from the small numbers of pathogenic bacteria isolated in this study, profiles were generally consistent with other studies.  The high rates of resistance to ampicillin, co-trimoxazole, tetracycline and chloramphenicol indicate they should not be given as empirical treatment for these infections. Further study of antimicrobial resistance trends in the area, including during outbreak periods, would help to inform antibiotic choices. There is an emerging menace of resistance to commonly used antibiotics in South Sudan with ceftriaxone being the only antibiotic currently reliably active against the enteropathogenic bacteria isolated in this study. Antibiotic use needs to be monitored and regulated in South Sudan to prevent outbreak of multidrug resistance.

Limitation

In this study we could not analyze E. coli isolates for pathogenic strains due to financial constraints, however presence of E. coli O157 is ruled out since there was no growth on SMaC agar media. Other pathogens (virus) and physiological factors could be responsible for much of the uncharacterized diarrhoea in the study but which are difficult to identify by methods deployed in this study.

Acknowledgment

We acknowledge and appreciate the support of AMREF Kenya country office, Wilson Branch for allowing the testing of the research specimens and the World Health Organization, South Sudan for supporting the shipment of specimens.

References

 

1. Sea C, Alarcon M, Aragon JC, Beneit S, Quiñonez M, Guerra H, Gotuzzo E. Surveillance of Bacterial Pathogens Associated with Acute Diarrhoea in Lima, Peru. Int. J. Infect. Dis. 2000, 4, 96–99.
2. Kosek M, Bern C, Guerrant RL. The global burden of diarrhoeal disease, as estimated from studies published between 1992 and 2000. Bull World Health Organ 2003; 81: 197-204.
3. World Health Organization. Changing History. World Health Report 2004. Available: http://www.who.int/whr/2004/en/report04_en.pdf
4. Lima AA. Tropical diarrhoea: new developments in traveller's diarrhoea. Curr Opin Infect Dis 2001; 14:547-52
5. Gascon J, Vargas M, Schellenberg D, Urassa H, Casals C, Kahigwa E, Aponte J, Mshinda H, Vila J, 2000. Diarrhea in children under five years of age in Ifakara, Tanzania: Case-control study. J Clin Microbiol 38: 4459–4462.
6. Vargas M, Gascon J, Casals C, Schellenberg D, Urassa H, Kahigwa E, Ruiz J, Vila J, 2004. Etiology of diarrhea in children less than five years of age in Ifakara, Tanzania. Am J Trop Med Hyg 70: 536–539.
7. Okeke IN, Ojo O, Lamikanra A, Kapper JB, Etiology of acute diarrhea in adults southwestern Nigeria. J Clin Microbiol 2003. 41: 4525–4530.
8. Guerrant RL, Van Gilder T, Steiner TS, Thielman NM, Slutsker L, Tauve RV, Hennessy T, Griffin PM, DuPont H, Sack RB, Tharr P, Neill M, Nachamkin I, Reller RB, Osterholm MT, Bennish ML, Pickerring LK, Pratice guidelines for the managment of infectious diarrhea. Clin Infect Dis 2001. 32: 31–51.
9. Shapiro RL, Kumar L, Phillips-Howard P, Wells JG, Adcock P, Brooks J, Ackers ML, Ochieng JB, Mintz E, Wahlquist S, Waiyaki P, Slutsker L, Antimicrobial-resistant bacterial diarrhea in rural western Kenya. J Infect Dis 183: 2001. 1701–1704.
10. World Health Organization South Sudan. Acute Watery Diarrhoea outbreak in Southern Sudan, 2006.
11. Kur L, MS, Mounir C, Lagu J,  Muita M, Rumunu J,  Ochieng B, Weathers A,  Nsubuga P,  Maes E, Rolle I. Cholera Outbreak- Southern Sudan, 2007. Mmwr April 10, 2009 / 58(13);337-341. Available: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5813a3.htm
12. Joseph N. Eisenberg S, Goldstick J, Cevallos W, Trueba G, Levy K, Scott J, Percha B, Segovia R, Ponce K, Hubbard A, Marrs C, Foxman B, David LS, Trostle J. In-roads to the spread of antibiotic resistance: regional patterns of microbial transmission in northern coastal Ecuador. J. R. Soc. Interface published online 28 September 2011 doi: 10.1098/rsif.2011.0499
13. Ifeanyi C, Ifeanyi C.C, Isu R. N, Akpa A.C and Ikeneche N. F. Enteric Bacteria Pathogens AssociatedWith Diarrhoea of Children in the Federal Capital Territory Abuja, Nigeria. New York Science Journal, 2010;3(1).
14. Samie, R.L. Guerrant, L. Barrett, P.O. Bessong, E.O. Igumbor, and C.L. Obi. Prevalence of Intestinal Parasitic and Bacterial Pathogens in Diarrhoeal and Non-diarroeal Human Stools from Vhembe District, South Africa. J Health Popul Nutr. Dec 2009; 27(6): 739–745.
15. Moyo SJ, Gro N, Matee MI, Kitundu J, Myrmel H, Mylvaganam H, Maselle SY, Langeland N.  Age specific aetiological agents of diarrhoea in hospitalized children aged less than five years in Dar es Salaam, Tanzania. BMC Pediatrics 2011, 11:19 doi:10.1186/1471-2431-11-19
16. Madiyarov RS, Bektemirov AM, Ibadova GA, Abdukhalilova GK, Khodiev AV, Bodhidatta L, Sethabutr O, Mason CJ. Antimicrobial resistance patterns and prevalence of class 1 and 2 integrons in Shigella flexneri and Shigella sonneiisolated in Uzbekistan. Gut Pathogens 2010, 2:18 doi:10.1186/1757-4749-2-18
17. Sang WK, Oundo V, Schnabel D. Prevalence and antibiotic resistance of bacterial pathogens isolated from childhood diarrhoea in four provinces of Kenya. J Infect Dev Ctries 2012; 6(7):572-578.
18. Sang WK, Kariuki SM, Schnabel D, Boga HI, Waiyaki PG, Wamae C.N. Antibiotic susceptibility of Enteric pathogens from the Maasai community, Narok and Kajiado Districts, Kenya. African Journal of Health Sciences, Volume 19, Number 3-4 June-December 2011.

16.    Jafari F, Shokrzadeh L, Hamidian M, Salmanzadeh-Ahrabi S, Zali MR, 2008. Acute diarrhea due to enteropathogenic bacteria in patients at hospitals in Tehran. Jpn J Infect Dis 61(4):269-73.

More from the November 2015 issue