Clinical Infectious Diseases MAJOR ARTICLE

Rapid Clearance and Frequent Reinfection With Enteric Pathogens Among Children With Acute Diarrhea in Zanzibar Maria Eva Andersson,1 Kristina Elfving,1,2 Deler Shakely,3,4 Staffan Nilsson,5 Mwinyi Msellem,6 Birger Trollfors,2 Andreas Mårtensson,7 Anders Björkman,4 and Magnus Lindh1 Departments of 1Infectious Diseases and 2Pediatrics, University of Gothenburg, 3Department of Medicine, Kungälv Hospital, 4Malaria Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, and 5Chalmers University of Technology, Gothenburg, Sweden; 6Zanzibar Malaria Elimination Programme, Ministry of Health, Tanzania; and 7Department of Women’s and Children’s Health, International Maternal and Child Health, Uppsala University, Sweden

Background.  Acute infectious gastroenteritis is an important cause of illness and death among children in low-income countries. In addition to rotavirus vaccination, actions to improve nutrition status, sanitation, and water quality are important to reduce enteric infections, which are frequent also among asymptomatic children. The aim of this study was to investigate if the high prevalence of these infections reflects that they often are not cleared properly by the immune response or rather is due to frequent pathogen exposure. Methods.  Rectal swabs were collected at time of acute diarrhea and 14 days later from 127 children, aged 2–59 months and living in rural Zanzibar, and were analyzed by real-time polymerase chain reaction targeting multiple pathogens. Results.  At baseline, detection rates >20% were found for each of enterotoxigenic Escherichia coli, Shigella, Campylobacter, Cryptosporidium, norovirus GII, and adenovirus. At follow-up, a large proportion of the infections had become cleared (34–100%), or the pathogen load reduced, and this was observed also for agents that were presumably unrelated to diarrhea. Still, the detection frequencies at follow-up were for most agents as high as at baseline, because new infections had been acquired. Neither clearance nor reinfection was associated with moderate malnutrition, which was present in 21% of the children. Conclusions.  Children residing in poor socioeconomic conditions, as in Zanzibar, are heavily exposed to enteric pathogens, but capable of rapidly clearing causative and coinfecting pathogens. Keywords.  gastroenteritis; diarrhea; etiology; PCR; follow-up.

Acute diarrhea is an important cause of illness and death worldwide in children 3.3 cycles, which corresponds to at least 1 log10 decline of the microbial concentration (Figure 1 and Table 2). Clinical Outcome

The clinical status normalized in the vast majority of children. Diarrhea was recorded in 5 children at follow-up: 1 had acquired blood in stool and a new Shigella infection, 1 had a new astrovirus infection, 1 had Shigella and ETEC-eltB at both time points, another had norovirus GII at both time points, and 1 had no pathogen detected at follow-up.

Table 1.  Changes in Detection Frequency Between Baseline and Follow-up

Pathogen

Positive at BL (n = 127)

Positive at FU (n = 127)

Cleared at FU (% of BL Positive)

New at FU (% of 127)

11 (44%)

19 (15%)

+32%

>.20

7 (6%)

+75%

>.20

Change BL to FU

P Valuea

Adenovirus

25

20%

33

26%

Astrovirus

4

3%

7

6%

26

20%

24

19%

13 (50%)

11 (9%)

–8%

>.20

7

6%

7

6%

4 (57%)

4 (3%)

0%

>.20

Norovirus GII Rotavirus Sapovirus Cryptosporidium

4 (100%)

8

6%

8

6%

6 (75%)

6 (5%)

0%

>.20

35

28%

34

27%

12 (34%)

11 (9%)

–3%

>.20

Campylobacter

43

34%

38

30%

18 (42%)

13 (10%)

–12%

>.20

ETEC-eltB

57

45%

65

51%

27 (47%)

35 (28%)

+14%

>.20

ETEC-estA

42

33%

25

20%

34 (81%)

17 (13%)

–40%

.024

Shigella

42

33%

30

24%

21 (50%)

9 (7%)

–29%

.043

Data are presented as No. (%) unless otherwise indicated. Abbreviations: BL, baseline; ETEC, enterotoxigenic Escherichia coli; FU, follow-up. a

P value by McNemar exact sign test analyzing whether there was a significant net change in detection of each pathogen.

Nutritional Status and Infections

Median z scores were –1.15 for weight for age, –1.22 for height for age, –0.88 for weight for height, and –1.05 for upper arm circumference for age. The z scores indicated moderate wasting (z score for weight for height  below  –2) in 21% (26/124), and moderate stunting (z score for height for age below –2) in 31% (37/124). There was no statistically significant association between moderate wasting (z scores below –2) and detection at baseline or with clearance, persistence, or new infection of any of the pathogens, as shown in detail for weight for height in Table 3. At baseline, the 26 children with moderate wasting had a total of 66 detected pathogens (2.54 per child), among which 67% had been cleared at follow-up, compared with 216 detected pathogens in 98 children without wasting (2.20 per child), among which 48% had been cleared at follow-up. New infections were also acquired at follow-up with the same rate (1.15 vs 1.01 agents per child). Antibiotics

Prescription of antibiotics at baseline was frequent, and was given as cotrimoxazole (n = 45), ampicillin/amoxicillin (n = 19), crystalline penicillin (n = 17), nalidixic acid (n = 21), erythromycin (n  =  17), or metronidazole (n  =  2), often because the child also had respiratory symptoms. Details of antibiotics prescribed to patients with different pathogens are given in Table 4, which shows that there was no significant association between antibiotic treatment and clearance of the pathogen. Sequencing

Sequencing was applied on selected samples to explore if the same strain of rotavirus, norovirus GII, Campylobacter, or Shigella was present at baseline and follow-up (Table 5). Of 11 paired samples with norovirus GII, 10 had the same type (GII.4 or GII.16 on both occasions, with Ct values increasing by on average 5.0 cycles), suggesting persistent infections with declining viral loads. One case showed a shift from GII.4 to GII.16.

The 3 pairs with rotavirus showed the same subtype (G8P10) on both occasions, with Ct values increasing by 9.2 to 13.9 cycles, indicating persistent infections with declining viral loads. Two of the 8 paired samples with Campylobacter had 1–3 nucleotide differences, and the other 6 had 23–102 nucleotide differences, suggesting that they represented a combination of clearance and new infection. Three of the 6 paired samples with Shigella had 17–27 nucleotide differences, suggesting that they represented a combination of clearance and new infection. DISCUSSION

Several previous studies in low-income countries have shown that infections with enteric pathogens are frequent among children with or without diarrhea [1–4, 14]. Most of these studies are cross-sectional and the rare longitudinal studies focus on only 1 or a few pathogens [15–18]. Therefore, it is not known how fast these enteric infections are cleared, or if clearance rates differ between the pathogens or between causative and coinfecting agents. Even less is known about the risk of becoming infected with new diarrheagenic agents after recovery from acute diarrhea. The present study, which investigates the shortterm evolution of enteric infections using a real-time PCR assay that targets a wide range of pathogens, provides novel data about clearance and acquisition of these infections, as well as about changes in pathogen concentration. At baseline, we found high detection frequencies for norovirus GII, Cryptosporidium, ETEC-estA, Shigella, Campylobacter, and ETEC-eltB, as previously reported [1]. At follow-up, a large proportion of pathogens detected at baseline had been cleared, with eradication rates ranging between 34% and 100%. Pathogen reduction was even greater if Ct values were taken into account, because a significant decrease of pathogen load was observed for almost all agents that were detected on both time points in the same patient. However, as a result of the high rates of reinfection, significant reductions of detection frequencies Short-term Changes of Enteric Infections  • CID 2017:00 (XX XXXX) • 3

Figure 1.  Real-time polymerase chain reaction cycle threshold (Ct) values at baseline and follow-up for all pathogens. Each patient is represented by one line: a green line for a cleared infection and a red line for a new infection on follow-up. Agents detected at both baseline and follow-up (putative persistent infections) are shown as blue lines. The y-axis scale is in reverse order because high Ct values represent low pathogen concentrations. Negative results (undetected pathogen) are shown as a Ct value of 40. Abbreviations: BL, baseline; Ct, cycle threshold; ETEC, enterotoxigenic Escherichia coli; FU, follow-up; NoV, norovirus.

at follow-up were achieved only for ETEC-estA and Shigella. Considering that ETEC-estA and Shigella have been strongly associated with symptomatic infection in previous studies [2, 3, 14], this finding suggests that enteric agents that cause diarrhea are cleared more effectively than other agents. Polymicrobial infections were frequent, and several pathogens were probably not associated with symptoms, because in our previous comparison they were not detected significantly more often than in healthy controls, as has also been reported by others [3]. Analysis of follow-up samples showed that such coinfecting pathogens, which in most cases likely were not causing symptoms, were also rapidly eradicated. This finding indicates that asymptomatic coinfections are not frequent because 4 • CID 2017:00 (XX XXXX) •  Andersson et al

they are long lasting. Instead, the high rates of enteric infections among healthy children are probably a result of frequent transient infections, most of which are cleared by effective immune responses. Within 2 weeks, 28% of the children had acquired a new ETEC-eltB infection, and new infections were relatively frequent also for ETEC-estA (13%), Campylobacter (10%), norovirus GII (9%), Cryptosporidium (9%), and Shigella (7%). These rates demonstrate the heavy exposure to enteric pathogens that children encounter in low-income settings such as Zanzibar. The findings suggest that improving sanitary conditions and safe water supply are essential for reducing diarrheal disease among children in low-income parts of the world [8].

Table 2.  Cycle Threshold Values at Baseline and Follow-up in Patients

Pathogen

Total No. Detected

Ct Value at BL, Median

Ct Value at FU, Median

No. persistent

Ct Changea Mean

P Valueb

1.48

.008

Adenovirus

90

34.4

36.2

14

Astrovirus

12

19.9

26.4

0

Norovirus GII

54

24.7

28

Rotavirus

17

23.6

31.9

Sapovirus

23

25.6

25.9

Cryptosporidium

83

31.7

34

23

+3.09

0.75

.0005

Campylobacter

135

30.3

27.3

25

–0.02

1.47

.98

ETEC-eltB

190

29.1

33.3

30

+3.34

1.36

.02

ETEC-estA

96

30.6

36.1

8

+3.84

1.91

.084

126

28.8

33.2

21

+4.78

0.97

40 or undetected pathogen were not included in this analysis.  Abbreviations: BL, baseline; Ct, cycle threshold; ETEC, enterotoxigenic Escherichia coli; FU, follow-up; SE, standard error. a

Shows Ct value difference when polymerase chain reaction was positive at both BL and FU. 

b

P value by paired t test.

Most of the children in this study were living under poor conditions and the nutritional status of the children was reduced, with weight and height being on average 1 standard deviation lower than normal for age, and z scores being below –2 in 27% (weightfor-height) and 31% (height-for-age) of the children. Diarrhea, enteric infections, and malnutrition are interrelated in a complex manner, in which diarrhea may contribute to malnutrition, and malnutrition leads to more frequent and prolonged enteric infections [5–7]. The majority of the infections were eradicated or strongly suppressed within 2 weeks, and there was no difference in pathogen clearance between children with z scores for weight or height below –2 as compared with above –2 in z score. In fact, children with moderate wasting cleared 67% of all pathogens, compared with 48% of children without wasting (weight-forheight z score above –2). This suggests that the high prevalence of

enteric infections, observed in this and several previous studies, is not due to poor immune clearance of these infections because of malnutrition, but more likely a result of heavy exposure to pathogens, from contaminated household water [8, 19] or other sources. Actions to improve sanitation and water supply are therefore of high priority, but such efforts may not prove effective [20], or might be of uncertain value [8]. To measure the impact of programs for improved sanitation and safe water supply on enteric infections, it is necessary to have suitable tools. The strategy to analyze paired samples collected 2 weeks apart allowed us to identify new infections and to calculate exposure, and should be useful for evaluating the impact of intervention. There are a few longitudinal studies to which our results can be at least partly related. In longitudinal studies from in Bangladesh [17], Egypt [16], and Guinea-Bissau [14, 18, 21],

Table 3.  Cleared and Acquired Pathogens at Follow-up in Children With or Without Moderate Wasting (Weight-for-Height z Scores Below –2) All (n = 124)

z Score Below –2 (n = 26) Cleareda

Newb

Cleareda

Newc

BL

FU

Adenovirus

25

31

6

2

33%

4

15%

19

9

47%

7

Astrovirus

3

7

2

2

100%

2

8%

1

1

100%

5

5%

25

23

5

4

80%

1

4%

20

9

45%

10

10%

Pathogen

Norovirus GII

BL

z Score Above –2 (n = 98) BL

13%

Rotavirus

7

7

1

1

100%

1

4%

6

3

50%

3

3%

Sapovirus

7

8

2

2

100%

2

8%

5

3

60%

4

4%

Cryptosporidium

34

33

6

4

67%

3

12%

28

8

29%

8

8%

Campylobacter

42

37

12

9

75%

3

12%

30

9

30%

10

10%

ETEC-eltB

55

63

11

7

64%

8

31%

44

20

45%

27

28%

ETEC-estA

41

24

10

9

90%

3

12%

31

25

81%

14

14%

Shigella

42

29

12

4

36%

3

12%

31

17

55%

5

5%

281

272

66

44

67%

30

115%

216

104

48%

93

101%

Total

Data are presented as No. unless otherwise indicated. Abbreviations: BL, baseline; ETEC, enterotoxigenic Escherichia coli; FU, follow-up.  a

Proportion as number cleared of pathogens detected at baseline. 

b

Proportion of all the 26 patients. 

c

Proportion of all the 98 patients.

Short-term Changes of Enteric Infections  • CID 2017:00 (XX XXXX) • 5

Table 4.  Antibiotic Treatment and Bacterial Detection at Baseline and Follow-up Campylobacter

ETEC-eltB

ETEC-estA

Shigella

Antibiotic

BL

Cleared at FU

BL

Cleared at FU

BL

Cleared at FU

BL

Ampicillina

9

22%

8

50%

5

80%

7

71%

Cotrimoxazole

7

29%

16

56%

13

92%

15

40%

Erythromycin

7

57%

7

43%

2

100%

6

33%

Metronidazole

1

0%

2

0%

1

100%

0

Nalidixic acid

Cleared at FU

3

67%

8

50%

6

50%

5

60%

Any antibioticb

27

37%

41

49%

28

81%

33

48%

No antibiotic

16

50%

14

50%

14

79%

9

56%

Data are presented as No. (%). Abbreviations: BL, baseline; ETEC, enterotoxigenic Escherichia coli; FU, follow-up.  a

Includes also those who received amoxicillin. 

b

Excluding the few patients that received other antibiotics than the 5 listed here.

relatively high rates of ETEC infections were observed in samples collected weekly or every other week, but the detection frequencies for ETEC as well as other agents were lower than in our study. Data on clearance were only presented in a study from Guinea-Bissau, which in agreement with our findings showed that 81% of ETEC infections cleared within 2 weeks. A longitudinal study from Cameroon comprising monthly sampling during 1  year observed rare protracted norovirus infections [15]. Similarly, in our study most norovirus GII infections were rapidly cleared, but almost a third remained after 2 weeks, including some with moderately reduced viral loads, suggesting that they might persist even longer. One limitation of this study is that the study period was only 3 months. This probably influenced the detection rates for agents that fluctuate over the year, such as rotavirus. Because seasonal variations are relatively slow, they probably had small impact on the short-term changes in pathogen detection that

Table  5.  Sequencing Results for Samples With Possible Persistent Infections (the Same Pathogen Detected at Baseline and Follow-up)

Pathogen

No.

Norovirus GII (capsid)

11

BL vs FU 7 with GII.4→GII.4

Interpretation (Same or Different Strain) Same

3 with GII.16→GII.16

Same

1 with GII.4→GII.16

Different

Rotavirus (VP4 and VP7)

3

3 with G10[P8]→G10[P8]

Same

Campylobacter (flaA)

8

2 with 1–3 nt differencesa

Same

6 with 50–102 nt differencesb Shigella (trpB)

6

3 with 2–5 nt differencesc 3 with 17–27 nt differencesd

Abbreviations: BL, baseline; FU, follow-up; nt, nucleotide.  a

Of 429–561 nt.

b

Of 259–565 nt.

c

Of 546–629 nt.

d

Of 598–629 nt.

6 • CID 2017:00 (XX XXXX) •  Andersson et al

Different Same Different

were the main objective of the study. Another limitation is that baseline and follow-up samples were not collected from healthy controls, and therefore we do not know the rate of pathogen clearance in children without diarrhea. Finally, it should be pointed out that anamnestic or measured fever was an inclusion criterion. Although mild in the majority of cases, it is possible that fever could have influenced the likelihood of clearance or the risk of acquiring a new infection. In summary, the results indicate that children residing in poor socioeconomic conditions like Zanzibar are heavily exposed to enteric pathogens, and that long-term presence of pathogenic microbes in the intestine is not due to host tolerance or weak immune responses; rather, most symptomatic and asymptomatic infections appear to be eradicated, repeatedly and efficaciously. Supplementary Data Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author. Notes Financial support.  This work was supported by the ACT Consortium through an award from the Bill & Melinda Gates Foundation (grant number ITGB5170-KI), and by governmental funds to the Sahlgrenska University Hospital (grant number ALFGBG-146611). Potential conflicts of interest.  All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed. References 1. Elfving K, Andersson M, Msellem MI, et al. Real-time PCR threshold cycle cutoffs help to identify agents causing acute childhood diarrhea in Zanzibar. J Clin Microbiol 2014; 52:916–23. 2. Kabayiza JC, Andersson ME, Nilsson S, et al. Diarrhoeagenic microbes by realtime PCR in Rwandan children under 5 years of age with acute gastroenteritis. Clin Microbiol Infect 2014; 20:O1128–35. 3. Kotloff KL, Nataro JP, Blackwelder WC, et al. Burden and aetiology of diarrhoeal disease in infants and young children in developing countries (the Global Enteric Multicenter Study, GEMS): a prospective, case-control study. Lancet 2013; 382:209–22.

4. Eibach D, Krumkamp R, Hahn A, et al. Application of a multiplex PCR assay for the detection of gastrointestinal pathogens in a rural African setting. BMC Infect Dis 2016; 16:150. 5. Guerrant RL, Schorling JB, McAuliffe JF, de Souza MA. Diarrhea as a cause and an effect of malnutrition: diarrhea prevents catch-up growth and malnutrition increases diarrhea frequency and duration. Am J Trop Med Hyg 1992; 47:28–35. 6. Rytter MJ, Kolte L, Briend A, Friis H, Christensen VB. The immune system in children with malnutrition—a systematic review. PLoS One 2014; 9:e105017. 7. Schaible UE, Kaufmann SH. Malnutrition and infection: complex mechanisms and global impacts. PLoS Med 2007; 4:e115. 8. Wolf J, Prüss-Ustün A, Cumming O, et al. Assessing the impact of drinking water and sanitation on diarrhoeal disease in low- and middle-income settings: systematic review and meta-regression. Trop Med Int Health 2014; 19:928–42. 9. Freeman MC, Stocks ME, Cumming O, et  al. Hygiene and health: systematic review of handwashing practices worldwide and update of health effects. Trop Med Int Health 2014; 19:906–16. 10. Greene LE, Freeman MC, Akoko D, Saboori S, Moe C, Rheingans R. Impact of a school-based hygiene promotion and sanitation intervention on pupil hand contamination in Western Kenya: a cluster randomized trial. Am J Trop Med Hyg 2012; 87:385–93. 11. Elfving K, Shakely D, Andersson M, et al. Acute uncomplicated febrile illness in children aged 2-59  months in Zanzibar—aetiologies, antibiotic treatment and outcome. PLoS One 2016; 11:e0146054. 12. Petersen L, Newell DG. The ability of Fla-typing schemes to discriminate between strains of Campylobacter jejuni. J Appl Microbiol 2001; 91:217–24.

13. Peng J, Yang J, Jin Q. The molecular evolutionary history of Shigella spp. and enteroinvasive Escherichia coli. Infect Genet Evol 2009; 9:147–52. 14. Kabayiza JC, Andersson ME, Nilsson S, Bergström T, Muhirwa G, Lindh M. Realtime PCR identification of agents causing diarrhea in Rwandan children less than 5 years of age. Pediatr Infect Dis J 2014; 33:1037–42. 15. Ayukekbong JA, Fobisong C, Tah F, Lindh M, Nkuo-Akenji T, Bergström T. Pattern of circulation of norovirus GII strains during natural infection. J Clin Microbiol 2014; 52:4253–9. 16. Mansour A, Shaheen HI, Amine M, et al. Diarrhea burden due to natural infection with enterotoxigenic Escherichia coli in a birth cohort in a rural Egyptian community. J Clin Microbiol 2014; 52:2595–603. 17. Qadri F, Saha A, Ahmed T, Al Tarique A, Begum YA, Svennerholm AM. Disease burden due to enterotoxigenic Escherichia coli in the first 2  years of life in an urban community in Bangladesh. Infect Immun 2007; 75:3961–8. 18. Valentiner-Branth P, Steinsland H, Fischer TK, et  al. Cohort study of Guinean children: incidence, pathogenicity, conferred protection, and attributable risk for enteropathogens during the first 2 years of life. J Clin Microbiol 2003; 41:4238–45. 19. Hassan KE, Mansour A, Shaheen H, et al. The impact of household hygiene on the risk of bacterial diarrhea among Egyptian children in rural areas, 2004-2007. J Infect Dev Ctries 2014; 8:1541–51. 20. Clasen T, Boisson S, Routray P, et al. Effectiveness of a rural sanitation programme on diarrhoea, soil-transmitted helminth infection, and child malnutrition in Odisha, India: a cluster-randomised trial. Lancet Glob Health 2014; 2:e645–53. 21. Steinsland H, Valentiner-Branth P, Perch M, et al. Enterotoxigenic Escherichia coli infections and diarrhea in a cohort of young children in Guinea-Bissau. J Infect Dis 2002; 186:1740–7.

Short-term Changes of Enteric Infections  • CID 2017:00 (XX XXXX) • 7

Rapid Clearance and Frequent Reinfection With Enteric Pathogens Among Children With Acute Diarrhea in Zanzibar.

Acute infectious gastroenteritis is an important cause of illness and death among children in low-income countries. In addition to rotavirus vaccinati...
862KB Sizes 0 Downloads 7 Views