Abstract
Background
The measurement of fecal inflammatory biomarkers among individuals presenting to care with diarrhea could improve the identification of bacterial diarrheal episodes that would benefit from antibiotic therapy. We reviewed prior literature in this area and describe our proposed methods to evaluate 4 biomarkers in the Enterics for Global Health (EFGH) Shigella surveillance study.
Methods
We systematically reviewed studies since 1970 from PubMed and Embase that assessed the diagnostic characteristics of inflammatory biomarkers to identify bacterial diarrhea episodes. We extracted sensitivity and specificity and summarized the evidence by biomarker and diarrhea etiology. In EFGH, we propose using commercial enzyme-linked immunosorbent assays to test for myeloperoxidase, calprotectin, lipocalin-2, and hemoglobin in stored whole stool samples collected within 24 hours of enrollment from participants in the Bangladesh, Kenya, Malawi, Pakistan, Peru, and The Gambia sites. We will develop clinical prediction scores that incorporate the inflammatory biomarkers and evaluate their ability to identify Shigella and other bacterial etiologies of diarrhea as determined by quantitative polymerase chain reaction (qPCR).
Results
Forty-nine studies that assessed fecal leukocytes (n = 39), red blood cells (n = 26), lactoferrin (n = 13), calprotectin (n = 8), and myeloperoxidase (n = 1) were included in the systematic review. Sensitivities were high for identifying Shigella, moderate for identifying any bacteria, and comparable across biomarkers. Specificities varied depending on the outcomes assessed. Prior studies were generally small, identified red and white blood cells by microscopy, and used insensitive gold standard diagnostics, such as conventional bacteriological culture for pathogen detection.
Conclusions
Our evaluation of inflammatory biomarkers to distinguish diarrhea etiologies as determined by qPCR will provide an important addition to the prior literature, which was likely biased by the limited sensitivity of the gold standard diagnostics used. We will determine whether point-of-care biomarker tests could be a viable strategy to inform treatment decision making and increase appropriate targeting of antibiotic treatment to bacterial diarrhea episodes.
Keywords: diagnostic, diarrhea, inflammatory biomarker, Shigella, systematic review
Fecal inflammatory biomarkers may accurately identify bacterial diarrheal episodes, which could inform antibiotic treatment decisions. Prior research was limited by using insensitive diagnostics. We will evaluate 4 inflammatory biomarkers in stool from children with diarrhea in the EFGH study.
Bacterial pathogens, such as Shigella and diarrheagenic Escherichia coli, are leading causes of diarrhea among children <5 years of age in low-resource settings. Appropriate antibiotic treatment of bacterial diarrhea episodes can limit morbidity and mortality [1, 2]. In a large, multicountry trial of azithromycin for children with watery diarrhea and dehydration, severe stunting, or moderate wasting, the benefit of azithromycin was observed primarily in children with a bacterial cause of diarrhea, namely Campylobacter, typical enteropathogenic E coli (EPEC), heat-stable enterotoxigenic E coli (ST-ETEC), Salmonella, Shigella, and Vibrio cholerae [3]. Furthermore, antibiotic treatment of Shigella-attributed moderate-to-severe diarrhea (MSD) was associated with improved short-term linear growth in the Global Enteric Multicenter Study (GEMS) [4], and antibiotic treatment of MSD was associated with lower risk of persistent diarrhea in the Vaccine Impact on Diarrhea in Africa (VIDA) study [5].
Targeting antibiotics to children with bacterial diarrhea is needed to limit antibiotic overuse and development of antimicrobial resistance. Current treatment guidelines take a syndromic approach, recommending antibiotic treatment for dysentery or presumed cholera [6], which comprise a small proportion of all bacterial diarrhea episodes. In the Etiology, Risk Factors, and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health and Development (MAL-ED) and VIDA studies, caregivers reported blood in stool for only 14.5% [7] and 43.8% [8] of shigellosis episodes, respectively. Of all bacterial diarrhea episodes in MAL-ED, caregivers reported blood in 10.4%. This suggests that most cases of shigellosis and other bacterial diarrhea episodes are missed according to current guidelines. Furthermore, younger children, who are most likely to die or be hospitalized from diarrhea and could significantly benefit from treatment, are less likely to present with dysentery [8–10]. Watery bacterial episodes are difficult to distinguish clinically, and prediction scores for specific etiologies that have been developed tend to be driven more heavily by epidemiologic characteristics (including age and season) than the presence of symptoms. For example, a clinical prediction score for Shigella developed in MAL-ED classified that nearly all episodes among children >18 months should be treated, and identified only a few episodes from 6–18 months that should be treated depending primarily on the presence of blood in addition to other symptoms [7]. While it is an important improvement over using the presence of blood alone to identify shigellosis, this score still only identified half of Shigella-attributed episodes [7].
In the absence of readily available point-of-care (POC) diagnostics, measurement of inflammatory biomarkers (ie, those indicative of leukocytes and/or erythrocytes in stool) could substantially improve clinical prediction scores to identify the subset of watery diarrhea episodes that would benefit from antibiotic therapy. Shigella secretes virulence factors that have enterotoxic activity and allow Shigella to invade the colonic epithelium, inducing an inflammatory response [11]. Campylobacter, Salmonella, enteroaggregative E coli, EPEC, and enteroinvasive E coli are also inflammatory and can cause invasive disease [12]. Several studies beginning in the 1970s found that the presence of red blood cells (RBCs) and white blood cells (WBCs) on stool microscopy was more common in shigellosis and other bacterial diarrhea episodes compared to viral episodes [13–18], and immunoassays for biomarkers of leukocytes such as lactoferrin [19–22] and calprotectin [23–27] have also been assessed to distinguish diarrhea etiology. If these markers prove to be sufficiently predictive of watery bacterial diarrhea episodes, inflammatory biomarker stool tests could be adapted into lateral flow assays, which would be readily deployable at the POC to inform antibiotic treatment, with limited demand for staff training or laboratory infrastructure.
The Enterics for Global Health (EFGH) Shigella surveillance study offers an ideal platform to further investigate novel strategies to identify bacterial diarrhea given its rich dataset and sample archive among a geographically diverse sample of children with diarrhea. In this article, we systematically review the literature of studies that assessed the sensitivity and specificity of fecal inflammatory biomarkers to identify bacterial diarrhea and describe our proposed methods for characterizing the performance of inflammatory biomarker tests to identify watery shigellosis and other bacterial diarrhea episodes in EFGH.
METHODS
We searched PubMed and Embase databases for studies published after 1 January 1970, using a combination of search terms (Supplementary Appendix) to capture bacterial diarrhea, fecal inflammatory biomarkers, and diagnostic studies. Included studies were published in English and conducted in individuals of all ages with bacterial diarrhea in any setting. Studies of natural history, blood biomarkers, diarrheal illnesses related to chronic diseases, asymptomatic infections, and animals were excluded. Review articles, case reports, and studies without sufficient data for extraction were also excluded.
Study selection was conducted using Covidence software [28]. Screening of titles and abstracts and the full text review was performed independently by 2 reviewers (C. B., S. Q., H. B., W. V. S. L., or E. T. R. M.). Disagreements were resolved by a third reviewer. Enrollment dates, study location, number of participants with diarrhea and specific pathogens, diagnostic gold standard used, biomarkers assessed, and the associated sensitivities and specificities for the identification of Shigella and a combined bacterial diarrhea outcome if available (eg, all bacteria, invasive bacteria, or a group of specific pathogens; Supplementary Table 1), were extracted from included studies by 1 of the above authors and checked by a second author (C. B. or E. T. R. M.). This review is registered with PROSPERO (CRD42023409479).
If a combined bacterial diarrhea outcome was not reported, we summed counts of individual bacterial outcomes and calculated sensitivity for the group of bacteria. When specificity was not directly provided, we calculated specificity as 1-sensitivity for the detection of negative outcomes such as no pathogen detected, viruses or parasites detected, or noninflammatory or noninvasive bacteria detected (Supplementary Table 1). We categorized studies by setting and age group. The quality of studies was assessed using Quality Assessment of Diagnostic Accuracy Studies (QUADAS) criteria [29] (Supplementary Table 2).
RESULTS
After screening 3875 titles and abstracts, and reviewing 93 full texts, 49 studies met inclusion criteria (Supplementary Figure 1). The enrollment period of included studies ranged from 1975 to 2020 (Supplementary Table 1). Thirty-four (69%) studies included children, 12 studies (25%) included adults only, and age was unknown for 3 (6%) studies. Twenty-four (49%) studies were conducted in low- and middle-income countries (LMICs) compared to 25 (51%) in high-income countries. Four studies were of travelers or the military, and 1 study included experimentally infected healthy volunteers. Of the 49 studies included, 39 (80%) assessed fecal leukocytes, 26 (53%) assessed RBCs/occult blood, 13 (27%) assessed lactoferrin, 8 (16%) assessed calprotectin, and 1 (2%) assessed myeloperoxidase (MPO). Shigella was assessed in 21 (43%) studies, and other bacterial outcomes were assessed in 40 (82%) studies. Eighteen (36%) studies included children in LMICs, of which 17 (35%) examined leukocytes, 10 (20%) examined RBCs/occult blood, and 5 (10%) examined other biomarkers. Most studies (n = 43 [88%]) used traditional bacterial culture alone as the gold standard diagnostic to identify bacterial causes of diarrhea. Three studies (6%) used culture and polymerase chain reaction (PCR), and 3 studies (6%) used PCR alone.
The median sensitivity of fecal leukocytes for identifying Shigella was 78% (interquartile range [IQR], 68%–94%) and 45% (IQR, 33%–69%) for identifying combined bacterial outcomes (Table 1). Specificity for Shigella was assessed in only 3 studies and ranged from 61% to 74%. Specificity for combined bacterial outcomes varied depending on the nonbacterial outcomes considered, with a median specificity of 85% (IQR, 75%–90%). For RBCs/occult blood, median sensitivity for Shigella was 70% (IQR, 51%–84%) and for combined bacterial outcomes was 48% (IQR, 37%–74%; Table 2). Specificity for Shigella was estimated in 1 study at 61% [30], and median specificity for combined bacterial outcomes was 69% (IQR, 63%–85%). Sensitivity of lactoferrin for Shigella was estimated in 4 studies and ranged from 61% to 100% (Table 3). Median sensitivity and specificity of lactoferrin for combined bacterial outcomes was 88% (IQR, 74%–94%) and 51% (IQR, 26%–69%), respectively. For calprotectin, sensitivity for Shigella was 78% in 1 study. Median sensitivity and specificity of calprotectin for combined bacterial outcomes was 86% (IQR, 76%–93%) and 69% (IQR, 41%–87%), respectively. The 1 study that assessed MPO estimated sensitivity of 78% and specificity of 57% for combined bacterial outcomes [31]. In the 2 studies that assessed both lactoferrin and calprotectin, sensitivity for combined bacterial outcomes was higher for calprotectin [26, 31]. Specificity was not consistently higher or lower between the 2 biomarkers (Table 3) [26, 31].
Table 1.
Setting and Study | Study Location | WBC Cutoff (per HPFa) | No. With Diarrhea | No. With Shigella | Sensitivity for Shigella, % | Specificity for Shigella, % | No. With Combined Bacterial Outcome | Sensitivity for Combined Bacterial Outcome, % | Specificity for Combined Bacterial Outcomeb, % |
---|---|---|---|---|---|---|---|---|---|
LMIC children (n = 17) | |||||||||
Bardhan 2000 [43] | Bangladesh | >20 | 1008 | 205 | 63 | … | … | … | 86–96 |
Beltinger 1997 [44] | Bangladesh | >20 | 304 | 38 | 76 | … | 54 | 67 | 87 |
Bodhidatta 2002 [45] | Thailand | >10 | 623 | 56 | 83 | … | … | … | … |
Chang 2017 [46] | China | ≥5 | 680 | 1 | 100 | … | 215 | 23 | 94 |
Huicho 1993 [16] | Peru | >5 | 446 | 10 | 70 | … | 58 | 36 | 82–88 |
Huicho 1997 [21] | Peru | NS | 125 | 6 | … | … | 29 | 69 | 60 |
Ismail 1994 [30] | Indonesia | >5 | 619 | 44 | 68 | 61 | 60 | 52 | 60 |
Jindal 1991 [47] | India | >10 | 400 | 6 | 67 | … | 115 | 28 | 88 |
Khan 2006 [17] | Bangladesh | >20 | 843 | 454 | 71 | 62 | … | … | … |
Korzeniowski 1979 [14] | USA, Brazil | >10/LPF | 101 | 29 | 95 | … | … | … | 85 |
McNeely 1996 [48] | Mexico | >5 | 1040 | 143 | … | … | 173 | 42 | 76 |
Mercado 2011 [18] | Peru | >10 | 935 | … | … | … | 257 | 14 | 90 |
Mshana 2010 [49] | Uganda | NS | 226 | … | … | … | 14 | 90 | 47 |
Nordlander 1990 [50] | Cambodia | >5 | 50 | 2 | 100 | … | … | … | 70–100 |
Patwari 1993 [51] | India | >0 | 533 | 17 | … | … | 60 | 27 | 90 |
Sebodo 1978 [52] | Indonesia | NS | 92 | 6 | 50 | … | … | … | 75 |
Venkataraman 2003 [53] | India | >3 | 262 | 27 | … | … | 42 | 45 | 79 |
LMIC adults only or unknown (n = 3) | |||||||||
Hossain 1991 [15] | Bangladesh | >25 | 11358 | 3895 | 80 | 74 | … | … | … |
Pender 2022 [54] | Thailand, Nepal | NS | 453 | 34c | … | … | 565c | 85 | 24 |
Wang 2014 [55] | China | NS | 424 | 90c | … | … | 176c | 99 | 97 |
HIC children (n = 11) | |||||||||
Alvarado 1983 [56] | UK | ≥2 | 376 | 34 | 91 | … | … | … | 82–100 |
Alzaher 2022 [57] | Saudi Arabia | >0 | 1985 | 84 | … | … | 1766 | 34 | 78 |
Ascher 1991 [58] | USA | >5 | 180 | 8 | … | … | 24 | 83 | 96 |
Caprioli 1996 [59] | Italy | NS | 618 | 2 | … | … | 168 | 57–58 | 73–88 |
Denno 2005 [60] | USA | >0 | 226 | 2 | … | … | 12 | 45 | 96 |
DuBois 1988 [61] | USA | >2 | 69 | 12 | 83 | … | 25 | 82 | 83 |
Fan 1993 [62] | USA | NS | 1031 | 20 | 50 | … | 55 | 29 | 93 |
Koplan 1980 [63] | USA | NS | 27 | 13 | … | … | 52 | 36 | 69 |
McIver 2001 [64] | Australia | >0 | 412 | … | … | … | 30 | 43 | 87 |
Paccagnini 1987 [13] | Italy | >5 | 337 | … | … | … | 62 | 32–54 | 90 |
Park 2019 [31] | South Korea | >0 | 62 | … | … | … | 19c | 44 | 84 |
HIC adults only (n = 8) | |||||||||
Bouckenooghe 2000 [65] | USA | NS | 227 | … | … | … | 56 | 27 | 85 |
Lai 2016 [66] | Taiwan | >1 | 627 | 3c | … | … | 187c | 45 | 73 |
Lever 2021 [67] | UK | NS | 1450 | 25 | … | … | 269 | 32 | 91 |
Loosli 1985 [68] | Switzerland | NS | 119 | 3 | … | … | 35 | 86 | 67 |
Miller 1994 [69] | USA | >0 | 55 | 9 | 95 | … | … | … | … |
Scerpella 1994 [70] | USA | NS | 92 | 32 | 72 | … | 36 | 69 | 89 |
Siegel 1987 [71] | USA | ≥3 | 113 | 19 | 95 | … | 54 | 89 | 68 |
Tribble 2008 [72] | USA | >0 | 182 | … | … | … | 177 | 33 | 85 |
Abbreviations: HIC, high-income country; HPF, high-power field; LMIC, low- and middle-income country; LPF, low-power field; NS, not specified; UK, United Kingdom; USA, United States; WBC, white blood cell count.
aUnless otherwise specified.
bRange provided if specificity was calculated based on multiple negative outcome definitions (see Supplementary Table 1 for negative outcome definitions).
cIncludes detection by polymerase chain reaction; otherwise, detection by culture (see Supplementary Table 1 for details).
Table 2.
Setting and Study | Study Location | RBC Test | No. With Diarrhea | No. With Shigella | Sensitivity for Shigella, % | Specificity for Shigella, % | No. With Combined Bacterial Outcome | Sensitivity for Combined Bacterial Outcome, % | Specificity for Combined Bacterial Outcomea, % |
---|---|---|---|---|---|---|---|---|---|
LMIC children (n = 10) | |||||||||
Ashraf 2007 [73] | Bangladesh | FOBT | 594 | 18 | 56 | … | 73 | 55 | 53 |
Bardhan 2000 [74] | Bangladesh | FOBT | 1008 | 205 | 87 | … | … | … | 18–80 |
Beltinger 1997 [44] | Bangladesh | FOBT | 304 | 38 | 82 | … | 54 | 69 | 66 |
Chang 2017 [46] | China | Micro | 680 | 1 | … | … | 215 | 13 | … |
Huicho 1993 [16] | Peru | FOBT | 446 | 10 | 70 | … | 58 | 43 | 62–69 |
Huicho 1997 [21] | Peru | FOBT | 125 | 6 | … | … | 29 | 79 | 50 |
Ismail 1994 [30] | Indonesia | Micro | 619 | 44 | 68 | 61 | 60 | 52 | … |
Korzeniowski 1979 [14] | USA, Brazil | FOBT | 101 | 29 | 85 | … | … | … | 89 |
McNeely 1996 [48] | Mexico | FOBT | 1040 | 143 | … | … | 173 | 79 | 64 |
Patwari 1993 [51] | India | Micro | 533 | 17 | … | … | 60 | 32 | 89 |
LMIC adults only or unknown (n = 4) | |||||||||
Pender 2022 [54] | Thailand, Nepal | Micro | 453 | 34b | … | … | 565b | 37 | 85 |
Wang 2014 [55] | China | Micro | 424 | 90b | … | … | 176b | 96 | 97 |
Aly 2005 [75] | Egypt | FOBT | 40 | 20 | 45 | … | … | … | 30 |
HIC children (n = 5) | |||||||||
Alzaher 2022 [57] | Saudi Arabia | FOBT | 1985 | 84 | … | … | 1766 | 48 | 84 |
Ascher 1991 [58] | USA | FOBT | 180 | 8 | … | … | 24 | 37 | 67 |
Denno 2005 [60] | USA | FOBT or Micro | 226 | 2 | … | … | 12 | 40 | 96 |
Paccagnini 1987 [13] | Italy | FOBT | 337 | … | … | … | 62 | 74–83 | 67 |
Park 2019 [31] | South Korea | FOBT | 62 | … | … | … | 19b | 61 | 82 |
HIC adults only (n = 7) | |||||||||
Bouckenooghe 2000 [65] | USA | FOBT | 227 | … | … | … | 56 | 30 | 71 |
Lai 2016 [66] | Taiwan | FOBT | 627 | 3b | … | … | 187b | 53 | 63 |
Lever 2021 [67] | UK | Micro | 1450 | 25 | … | … | 269 | 16 | 94 |
Loosli 1985 [68] | Switzerland | FOBT | 119 | 3 | … | … | 35 | 80 | 66 |
Scerpella 1994 [70] | USA | FOBT | 92 | 32 | 37 | … | 36 | 37 | 82 |
Shastri 2008 [26] | Germany | FOBT | 200 | 2 | … | … | 107 | 38 | 85 |
Siegel 1987 [71] | USA | FOBT | 113 | 19 | 84 | … | 54 | 87 | 58 |
Tribble 2008 [72] | USA | FOBT | 182 | … | … | … | 177 | 42 | 84 |
Abbreviations: FOBT, fecal occult blood test; HIC, high-income country; LMIC, low- and middle-income country; Micro, microscopy; RBC, red blood cell count; UK, United Kingdom; USA, United States.
aRange provided if specificity was calculated based on multiple negative outcome definitions (see Supplementary Table 1 for negative outcome definitions).
bIncludes detection by polymerase chain reaction; otherwise, detection by culture (see Supplementary Table 1 for details).
Table 3.
Setting and Study | Study Location | Biomarker Cutoff | No. With Diarrhea | No. With Shigella | Sensitivity for Shigella, % | Specificity for Shigella, % | No. With Combined Bacterial Outcome | Sensitivity for Combined Bacterial Outcome, % | Specificity for Combined Bacterial Outcome, % |
---|---|---|---|---|---|---|---|---|---|
Lactoferrin (n = 13) | |||||||||
Aly 2005 [75] | Egypt | 1:50 | 40 | 20 | 80 | … | … | … | 25 |
Ashraf 2007 [73] | Bangladesh | 1:50 | 594 | 18 | 61 | … | 73 | 52 | 68 |
Bouckenooghe 2000 [65] | USA | 1:50 | 227 | … | … | … | 56 | 27 | 79 |
Choi 1996 [20] | USA | 1:50 | 46 | 3 | … | … | 28 | 93 | 83 |
Huicho 1997 [21] | Peru | 1:50 | 125 | 6 | … | … | 29 | 97 | 15 |
McIver 2001 [64] | Australia | NS | 412 | … | … | … | 30 | 95 | 40 |
Mercado 2011 [18] | Peru | NS | 935 | … | … | … | 200 | 95 | … |
Miller 1994 [69] | USA | 1:50 | 55 | 9 | 100 | … | 43 | 60 | 8 |
Park 2019 [31] | South Korea | 22.8 µg/mL | 62 | … | … | … | 19a | 78 | 71 |
Scerpella 1994 [70] | USA | 1:50 | 92 | 32 | 94 | … | 36 | 94 | 47 |
Shastri 2008 [26] | Germany | 1:400 | 200 | 2 | … | … | 107 | 78 | 54 |
Tribble 2008 [72] | USA | 1:50 | 182 | … | … | … | 177 | 93 | 56 |
Venkataraman 2003 [53] | India | 1:50 | 262 | 27 | … | … | 42 | 83 | 28 |
Calprotectin (n = 8) | |||||||||
Ahn 2020 [76] | South Korea | 388 mg/kg | 400 | 7a | … | … | 197a | 71 | 61 |
Berger 2010 [77] | Unknown | >50 µg/mL | 168 | … | … | … | 108 | 80 | 89 |
Czub 2014 [78] | Poland | 15 µg/mL | 50 | … | … | … | 21 | 100 | 45 |
Duman 2015 [23] | Turkey | 710 mg/L | 84 | 7 | 78 | … | 9 | 89 | 76 |
Kim 2022 [79] | South Korea | 815 µg/g | 80 | … | … | … | 16a | 75b | 40 |
Park 2019 [31] | South Korea | 74.0 µg/g | 62 | … | … | … | 19a | 94 | 39 |
Shastri 2008 [26] | Germany | 14.9 mg/L | 200 | 2 | … | … | 107 | 83 | 87 |
Sýkora 2010 [27] | Czech Republic | 103.9 µg/g | 66 | … | … | … | 31 | 93 | 88 |
Myeloperoxidase (n = 1) | |||||||||
Park 2019 [31] | South Korea | 4.14 ng/mL | 62 | … | … | … | 19a | 78 | 57 |
Abbreviations: NS, not specified; USA, United States.
aIncludes detection by polymerase chain reaction; otherwise, detection by culture (see Supplementary Table 1 for details).
bEstimated from manuscript figure.
Different definitions of positive for WBC by microscopy (ranging from >0 to >20 cells per high-power field) and different assays and/or cutoffs for the other biomarkers made it difficult to compare results across studies. Similarly, heterogeneity resulted from differences in the combined bacterial outcome considered and the negative outcome used to calculate specificity. Most studies (n = 43 [88%]) were considered low quality (Supplementary Table 2) due to using bacterial culture, which is an insensitive gold standard diagnostic, particularly for Shigella [7]. Only 12 studies (24%) were conducted among children in LMICs (ie, the target population for EFGH), none of which used molecular diagnostics and only 2 of which assessed either lactoferrin or calprotectin. Only 6 (12%) studies included >1000 individuals with diarrhea.
METHODS IN THE EFGH INFLAMMATORY BIOMARKER SUBSTUDY
We will conduct an inflammatory biomarker substudy in 6 EFGH sites: Bangladesh, Kenya, Malawi, Pakistan, Peru, and The Gambia. The objective of this substudy is to evaluate whether inflammatory biomarkers measured in whole stool can identify the bacterial subset of diarrhea episodes, and shigellosis specifically. The primary EFGH study design is described elsewhere [32].
Sample Collection
Whole stool samples will be collected as soon as possible after enrollment from all enrolled children aged 6–35 months presenting with diarrhea at selected study health facilities. Samples will be collected if they are produced at any time while the participant is present at the enrolling facility or within 24 hours of leaving the enrolling facility. This strategy will increase the yield of whole stool collections since children may not produce stool during the enrollment visit. Study staff will conduct home visits to collect stools produced after leaving the facility. Caregivers will also have the option of returning the whole stool sample to the enrollment facility. In both cases, caregivers of participants who do not produce whole stool at the enrollment visit will be provided with a whole stool collection kit and will be instructed to collect the participant's first stool produced after leaving the enrollment facility. Home visits will only occur during routine working hours. Once retrieved by study personnel (and within 18 hours of stool production), whole stool samples will be placed into a cool box (2°C–8°C) for transportation to the laboratory. The following will be verified during accession: labeling, stool volume, and transport conditions, which include packaging and temperature monitoring using WarmMark (after collection by staff only). Whole stool will be aliquoted into up to five 2-mL cryotubes (up to 1 g per cryotube), and frozen at −80°C until testing.
Inflammatory Biomarker Testing
All whole stool samples will be tested for 4 biomarkers: MPO, calprotectin, lipocalin-2 (NGAL), and hemoglobin. These 4 were chosen to capture markers of both leukocytes and erythrocytes since both showed evidence of diagnostic ability in the systematic review. While most prior studies used microscopy to measure fecal WBC and RBC, microscopy would be impractical as a diagnostic in many settings; therefore, we selected protein biomarkers of leukocytes and erythrocytes that either had prior evidence of diagnostic ability (calprotectin, hemoglobin), had strong preliminary data from our prior work (MPO), or represented a novel component of the host immune response, which would limit collinearity between markers (NGAL). Lactoferrin was considered but ultimately rejected given its similarity with calprotectin and the better sensitivity for calprotectin over lactoferrin in studies from the systematic review that assessed both [26, 31]. We will not evaluate systemic biomarkers given the likely infeasibility and unacceptability of collecting blood samples at the POC in most clinical settings.
Each marker, their underlying mechanism of action, and rationale are outlined in Table 4. We will use commercially available enzyme-linked immunosorbent assays (ELISAs) for each biomarker according to their manufacturers’ instruction manuals [33–36]. Biomarker concentrations per gram of stool will be calculated from the raw optical density data using a 4-parameter curve fit to the standards, which will be run in duplicate on every plate. Each plate will also include a high and low concentration control run in duplicate. The analysis of raw optical densities will be centralized using a custom-built R-based Shiny application accessible on the web. The app will allow for monitoring of standards and controls with immediate feedback to the laboratories at each site and real-time quality control (QC) monitoring by a central coordinating team.
Table 4.
Biomarker | Target | Description | Rationale |
---|---|---|---|
Myeloperoxidase | Neutrophils | A peroxidase enzyme belonging to the heme-containing proteins, produced largely in neutrophils and in smaller quantities in monocytes [80, 81]. MPO is considered a biomarker for neutrophils, inflammatory activity in the gastrointestinal tract, and neutrophil damage [81]. Fecal MPO is a biomarker for IBD [82]. | Shigella detections in MAL-ED were associated with increases in MPO, and the association depended on Shigella quantity, such that MPO levels were more highly elevated as the quantity of Shigella increased [10]. Diarrheagenic Escherichia coli, specifically EAEC, EPEC, and ETEC, was also associated with elevated MPO, though to a lesser extent [83]. |
Calprotectin | Neutrophils | A protein biomarker of leukocytes and neutrophil damage during intestinal inflammation [23–25]. Fecal calprotectin helps to distinguish between IBD and noninflammatory bowel conditions and monitor IBD activity [84]. | Increased levels were observed among those with bacterial compared to viral infections during diarrhea [24]. Calprotectin was elevated in shigellosis at levels higher than other bacterial diarrheas (Clostridioides difficile, Salmonella, Campylobacter, and EIEC) [85]. |
Lipocalin-2 (NGAL) | Neutrophils; enterocyte damage | A circulatory protein commonly referred to as neutrophil gelatinase-associated lipocalin (NGAL). NGAL is responsible for the delivery of molecules including steroids, free fatty acids, and hormones to body organs [86]. It is an indicator of innate immunity [87], found in a variety of cells including neutrophils, and possesses antibacterial and anti-inflammatory functions, in addition to providing protection against cell and tissue stress [86]. NGAL is an indicator of enterocyte damage and acute and chronic renal injury [86, 88]. It is also a biomarker for intestinal inflammation and is associated with IBD [87]. | Studies in a Shigella murine model demonstrate that sensitivity for Shigella may be higher compared to MPO [89]. Lipocalin-2 decreases rapidly following inflammation [90]. |
Hemoglobin | Fecal occult blood | The iron-containing protein present in RBCs responsible for transporting oxygen to organs and tissues [91]. Hemoglobin is a marker of RBCs and its presence in stool indicates the presence of blood. Fecal hemoglobin helps to identify IBD patients with active inflammation [92]. | Shigella is the main cause of dysentery among children globally [7]. Presence of RBCs was predictive of shigellosis and other bacterial diarrhea in the systematic review. |
Abbreviations: EAEC, enteroaggregative Escherichia coli; EIEC, enteroinvasive Escherichia coli; EPEC, enteropathogenic Escherichia coli; ETEC, enterotoxigenic Escherichia coli; IBD, inflammatory bowel disease; MAL-ED, Etiology, Risk Factors, and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health and Development; MPO, myeloperoxidase; NGAL, neutrophil gelatinase-associated lipocalin; RBCs, red blood cells.
Detection of Enteric Pathogens
Total nucleic acid will be extracted from rectal swabs from each enrolled participant and tested for enteric pathogens by quantitative PCR (qPCR) using the TaqMan Array Card platform, as described elsewhere in this supplement [37]. Attribution of diarrhea episodes to specific infectious etiologies will be based on the quantities of pathogens detected, and assigned etiologies by qPCR will be considered the “gold standard” diagnostic against which we will compare the inflammatory biomarkers. Specifically, attribution will be assigned if a pathogen is detected at a cycle threshold value below the EFGH and pathogen-specific cutoff, described elsewhere [37]. The definition of etiology will not be dependent on the detection of other pathogens, such that multiple etiologies may be identified for each episode. The presence of co-etiologies will not be considered in the primary analysis (eg, if rotavirus is considered etiologic in addition to Shigella), but we will exclude episodes with multiple etiologies in a sensitivity analysis. Shigella will also be identified by bacterial culture in the main EFGH study, which will be considered in a separate sensitivity analysis.
Data Analysis
We will compare the diagnostic characteristics (eg, area under the curve, sensitivity, specificity) of the candidate biomarkers to identify watery diarrhea episodes that are attributed to Shigella, specific Shigella species, and other causes of bacterial diarrhea (eg, typical EPEC, ETEC, Campylobacter, Salmonella, and V cholerae) by qPCR. The correlation between markers will also be assessed to determine which may be complementary and/or redundant. Clinical prediction scores will be derived that incorporate the best-performing inflammatory biomarkers to identify an optimal diagnostic tool for watery bacterial diarrhea, and Shigella-attributed cases in particular. We will use SuperLearner, an algorithm that uses cross-validation to create an “ensemble” prediction model, which is an optimal weighted average of multiple machine learning models. Clinical and epidemiologic characteristics will include fever, duration of diarrhea, dehydration, vomiting, stool frequency, child age, season, length-for-age z score (LAZ, if <24 months of age) or height-for-age z score (HAZ, if ≥24 months of age), and breastfeeding at enrollment. Based on the variables included in the ensemble prediction model, we will create a more parsimonious prediction score based on scaled coefficients from a logistic regression model for bacterial etiology. This score could be practically applied in clinical settings; each characteristic included will be assigned points that would then be summed into a total score. Acknowledging the need to prioritize specificity to limit antibiotic overuse, we will derive cutoffs that maximize sensitivity at a minimum level of specificity of 80% to identify episodes that should be treated according to our algorithm. We will also report categories of confidence (eg, “most likely bacterial”) and/or percentage confidence based on the optimal machine learning algorithm. Finally, based on antibiotic treatments received and antibiotic susceptibility data [38], we will estimate the impact of appropriate antibiotic treatment on duration of diarrhea, hospitalization, 90-day mortality, and change in HAZ/LAZ in the subset of diarrhea episodes meeting the threshold for treatment based on the optimal treatment algorithm. We will compare the effects of treatment in the algorithm-defined subset with those among all episodes, in the subset of episodes that would be treated according to World Health Organization (WHO) guidelines (ie, dysentery only), and in etiology-specific subsets based on pathogen quantity detected by qPCR.
Preliminary analyses of MAL-ED data suggest that the biomarkers will be successful in identifying bacterial diarrhea. We added MPO concentrations that were de-trended for age to a clinical prediction score for Shigella previously developed in MAL-ED [7] to assess improvements in predictive ability of the score. The best improvements were achieved when MPO concentration was included with 5 categories and was weighed similarly heavily as child age. The clinical score alone achieved 40% sensitivity for identifying Shigella-attributable diarrhea episodes with 80% specificity (AUC = 0.74) in the subset of episodes that were also tested for MPO (n = 281). The addition of MPO increased sensitivity to 70% with 80% specificity (AUC = 0.79). This is a substantial improvement that may be more striking with the addition of multiple candidate biomarkers in a larger dataset.
POTENTIAL CHALLENGES AND LIMITATIONS
Our approach has some noteworthy limitations. Foremost, the algorithm will likely identify at least some episodes that should not be treated with antibiotics. Specifically, it could lead to unnecessary treatment of children with viral or parasitic diarrhea, which would facilitate antibiotic overuse and have implications for the development of antimicrobial resistance. However, use of even an imperfectly specific algorithm to inform treatment decisions would likely improve on current clinical practice given the often extreme overuse of antibiotics for diarrhea among children in low-resource settings [39]. The algorithm could also in theory cause harm by identifying bacterial diarrhea episodes for which antibiotics are contraindicated, for example, for children with Shiga toxin–producing E coli (STEC) [40]. However, the typical STEC clinical syndrome is bloody diarrhea without fever, whereas this algorithm would be primarily relevant for watery diarrhea since the WHO guidelines already recommend treatment for dysentery [6]. Furthermore, there was weak association between STEC and diarrhea in GEMS and MAL-ED, such that the role of STEC in children in resource-limited settings may be limited [7, 41]. Another key limitation is that the biomarker tests will not provide antibiotic susceptibility results, such that the algorithm will not be able to determine which antibiotic is likely to be effective in cases in which antibiotics are indicated. Clinicians will have to continue to rely on any available local susceptibility data.
There will also be several challenges in the research methods. Because not all participants will produce a stool specimen at the clinic during the enrollment visit, we will include stool samples produced within 24 hours of leaving the enrollment facility to achieve the sample size required to adequately power the study. This may confound the results in 2 ways. First, sampling will be performed at variable time points from the onset of illness, which may impact the levels of biomarkers within stool. Second, since antibiotics are likely to be administered before and/or at the enrollment visit, the biomarker levels in stools may have changed as a result of antibiotic action. We will adjust for the time between presentation to care and sample collection in the analysis to mitigate this concern. We will evaluate whether the performance of the algorithm differs by time since symptom onset and antibiotics received, since there will also be heterogeneity in these factors at the POC.
The inclusion of stool samples collected at participants’ homes may also result in variable time in which samples are outside of cold chain. While NGAL, calprotectin, and MPO are stable at room temperature [34–36], hemoglobin degrades up to 50% per day at room temperature [33]. In addition, to maximize the efficiency of sample testing, stool samples may be stored frozen for variable amounts of time and up to several months before being tested. While the test manufacturers stipulate that stool samples for the NGAL and calprotectin assays can be stored at −20°C for 1 year, samples for the hemoglobin and MPO assays should not be stored at −20°C for more than 1 and 2.5 months, respectively [33–36]. We will test all samples regardless and adjust for storage time in the analysis as necessary.
Next, the quantitative biomarker assays employed in this study require 15–50 mg of whole stool. The gold standard diagnostic comparator (pathogen detection by qPCR) will be evaluated in rectal swabs collected during the enrollment visit, rather than in the same whole stool sample in which the biomarkers will be tested. This aligns with the parent study protocol, which specifies using rectal swabs for qPCR to ensure etiology information is available for every enrolled case. Biomarker concentrations and/or pathogen detection may differ between the 2 samples due to differences in sample type and time of collection. While some patients and providers may prefer collection of whole stool rather than rectal swabs, whole stool may be an impractical clinical specimen on which to base a POC test for bacterial diarrhea since a rapid result would be required to guide clinical management. Should this study show that biomarkers are useful for predicting which children have bacterial diarrhea, future work will need to establish the validity of rectal swabs for POC biomarker testing.
A strength of our study is the inclusion of all medically attended diarrhea cases with etiology determined by qPCR among children from a diverse range of geographical locations. The inclusion of less-severe diarrhea is important since mortality is likely to be similar in both moderate-to-severe and less-severe cases [42]. The inclusion of 6 different study locations, however, presents a challenge when ensuring standardization of laboratory procedures between multiple sites. To ensure that the results are reproducible between sites, standardized ELISA training is being performed at all study sites and the centralized analysis platform will facilitate QC monitoring. In addition, there may be regional differences in the baseline levels of inflammatory biomarkers between populations driven by population genetics, the microbiota, subclinical infections, and/or differences in diet. If large enough, these differences may mean that generalizable biomarker concentration cutoff levels capable of guiding bacterial diagnosis cannot be established. We will evaluate and describe heterogeneity in the performance of the algorithm by study site.
DISCUSSION
Despite the limitations of the included studies in the systematic review, the sensitivity of all biomarkers to identify Shigella was high, and sensitivity to identify combined bacterial outcomes was moderate. Not surprisingly, specificity was lower for Shigella than for combined bacterial outcomes since multiple causes of diarrhea are inflammatory. The insensitivity of culture for Shigella may also have resulted in lower estimated specificities for Shigella. Performance was broadly comparable across biomarkers, justifying the assessment of multiple markers of leukocytes and erythrocytes in EFGH.
Our proposed inflammatory biomarker substudy will improve on prior efforts to use fecal biomarkers to identify bacterial diarrhea by applying ELISAs and molecular diagnostics in a large, geographically diverse study population. By combining inflammatory biomarker test results with clinical prediction scores to maximize predictive validity, we will determine whether POC biomarker tests would be a viable strategy to improve appropriate antibiotic treatment of watery bacterial diarrhea episodes. In the absence of readily available diagnostics for enteric pathogens, these tools could improve short- and long-term outcomes of diarrhea. If the inflammatory biomarkers are acceptably sensitive and specific, further development of low-cost POC biomarker tests would be warranted. Conversely, if these tools are not adequate, development of low-cost assays for the direct detection of enteropathogens in the clinical setting should be prioritized. Either type of POC test would have the dual benefit of increasing appropriate treatment of the episodes that are likely to respond, while also reducing inappropriate and overtreatment of viral and parasitic episodes. In settings where antibiotic overuse is common, the application of such tests may improve access to appropriate therapy while decreasing antibiotic use for diarrhea overall.
Supplementary Data
Supplementary materials are available at Open Forum 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.
Supplementary Material
Contributor Information
Courtney Babb, Department of Epidemiology, Emory University, Atlanta, Georgia, USA.
Henry Badji, Medical Research Council Unit The Gambia, London School of Hygiene and Tropical Medicine, Fajara, The Gambia.
Md Taufiqur Rahman Bhuiyan, Infectious Diseases Division, International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh.
Jennifer Cornick, Institute of Infection, Veterinary and Ecological Sciences, Department of Clinical Infection, Microbiology and Immunology, University of Liverpool, Liverpool, United Kingdom; Malawi Liverpool Wellcome Programme, Blantyre, Malawi.
Sonia Qureshi, Department of Pediatrics and Child Health, The Aga Khan University, Karachi, Pakistan.
Catherine Sonye, Center for Global Health Research, Kenya Medical Research Institute, Kisumu, Kenya.
Wagner V Shapiama Lopez, Asociación Benéfica PRISMA, Iquitos, Peru.
Mehreen Adnan, Department of Pediatrics and Child Health, The Aga Khan University, Karachi, Pakistan.
Hannah E Atlas, Department of Global Health, University of Washington, Seattle, Washington, USA.
Kehkashan Begum, Department of Pediatrics and Child Health, The Aga Khan University, Karachi, Pakistan.
Stephanie A Brennhofer, Division of Infectious Diseases and International Health, University of Virginia, Charlottesville, Virginia, USA.
Bubacarr E Ceesay, Medical Research Council Unit The Gambia, London School of Hygiene and Tropical Medicine, Fajara, The Gambia.
Abdoulie K Ceesay, Medical Research Council Unit The Gambia, London School of Hygiene and Tropical Medicine, Fajara, The Gambia.
Nigel A Cunliffe, Institute of Infection, Veterinary and Ecological Sciences, Department of Clinical Infection, Microbiology and Immunology, University of Liverpool, Liverpool, United Kingdom.
Paul F Garcia Bardales, Asociación Benéfica PRISMA, Iquitos, Peru.
Shahinur Haque, Infectious Diseases Division, International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh.
Bri’Anna Horne, Center for Vaccine Development and Global Health, University of Maryland School of Medicine, Baltimore, Maryland, USA; Department of Medicine, University of Maryland School of Medicine, Baltimore, Maryland, USA.
M Jahangir Hossain, Medical Research Council Unit The Gambia, London School of Hygiene and Tropical Medicine, Fajara, The Gambia.
Junaid Iqbal, Department of Pediatrics and Child Health, The Aga Khan University, Karachi, Pakistan.
Md Taufiqul Islam, Infectious Diseases Division, International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh.
Sadia Islam, Infectious Diseases Division, International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh.
Farhana Khanam, Infectious Diseases Division, International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh.
Karen L Kotloff, Center for Vaccine Development and Global Health, University of Maryland School of Medicine, Baltimore, Maryland, USA; Department of Medicine, University of Maryland School of Medicine, Baltimore, Maryland, USA.
Thandizo Malemia, Malawi Liverpool Wellcome Programme, Blantyre, Malawi.
Katia Manzanares Villanueva, Asociación Benéfica PRISMA, Iquitos, Peru.
Gertrude Malola Million, Malawi Liverpool Wellcome Programme, Blantyre, Malawi.
Vitumbiko Munthali, Malawi Liverpool Wellcome Programme, Blantyre, Malawi.
John Benjamin Ochieng, Center for Global Health Research, Kenya Medical Research Institute, Kisumu, Kenya.
Billy Ogwel, Center for Global Health Research, Kenya Medical Research Institute, Kisumu, Kenya.
Maribel Paredes Olortegui, Asociación Benéfica PRISMA, Iquitos, Peru.
Richard Omore, Center for Global Health Research, Kenya Medical Research Institute, Kisumu, Kenya.
Patricia B Pavlinac, Department of Global Health, University of Washington, Seattle, Washington, USA.
James A Platts-Mills, Division of Infectious Diseases and International Health, University of Virginia, Charlottesville, Virginia, USA.
Khandra T Sears, Center for Vaccine Development and Global Health, University of Maryland School of Medicine, Baltimore, Maryland, USA; Department of Medicine, University of Maryland School of Medicine, Baltimore, Maryland, USA.
Ousman Secka, Medical Research Council Unit The Gambia, London School of Hygiene and Tropical Medicine, Fajara, The Gambia.
Sharon M Tennant, Center for Vaccine Development and Global Health, University of Maryland School of Medicine, Baltimore, Maryland, USA; Department of Medicine, University of Maryland School of Medicine, Baltimore, Maryland, USA.
Pablo Peñataro Yori, Division of Infectious Diseases and International Health, University of Virginia, Charlottesville, Virginia, USA.
Mohammad Tahir Yousafzai, Department of Pediatrics and Child Health, The Aga Khan University, Karachi, Pakistan.
Khuzwayo C Jere, Institute of Infection, Veterinary and Ecological Sciences, Department of Clinical Infection, Microbiology and Immunology, University of Liverpool, Liverpool, United Kingdom; Malawi Liverpool Wellcome Programme, Blantyre, Malawi; Department of Medical Laboratory Sciences, Kamuzu University of Health Sciences, School of Life Sciences and Health Professions, Blantyre, Malawi.
Margaret N Kosek, Division of Infectious Diseases and International Health, University of Virginia, Charlottesville, Virginia, USA.
Stephen Munga, Center for Global Health Research, Kenya Medical Research Institute, Kisumu, Kenya.
Usman N Ikumapayi, Medical Research Council Unit The Gambia, London School of Hygiene and Tropical Medicine, Fajara, The Gambia.
Firdausi Qadri, Infectious Diseases Division, International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh.
Farah Naz Qamar, Department of Pediatrics and Child Health, The Aga Khan University, Karachi, Pakistan.
Elizabeth T Rogawski McQuade, Department of Epidemiology, Emory University, Atlanta, Georgia, USA.
Notes
Author contributions. C. B., H. B., T. R. B., J. C., S. Q., C. S., W. V. S. L., K. C. J., M. N. K., S. M., U. N. I., F. N. Q., F. Q., and E. T. R. M. actively participated in monthly working group meetings during which the conceptualization and outline were discussed and agreed upon. C. B., H. B., W. V. S. L., S. Q., and E. T. R. M. conducted the systematic review. C. B., H. B., T. R. B., J. C., S. Q., C. S., W. V. S. L., and E. T. R. M. wrote the first draft of the manuscript. K. C. J., M. N. K., S. M., U. N. I., F. N. Q., and F. Q. provided review, scientific input, and editing of the first draft of the manuscript. M. A., H. E. A., K. B., S. A. B., B. E. C., A. K. C., N. A. C., P. F. G. B., S. H., B. H., M. J. H., J. I., T. I., S. I., F. K., K. L. K., T. M., K. M. V., G. M. M., V. M., J. B. O., B. O., M. P. O., R. O., P. B. P., J. A. P.-M., K. T. S., O. S., S. M. T., P. P. Y., and M. T. Y. reviewed and edited the manuscript. All authors approved the content of the final manuscript.
Disclaimer. The views expressed are those of the author(s) and not necessarily those of the National Institute for Health and Care Research (NIHR), the Department of Health and Social Care, the UK government, or the UK Health Security Agency.
Financial support. This project is supported by the Bill & Melinda Gates Foundation (grant numbers INV-016650, INV-031791, INV-036891, INV-036892, INV-028721, INV-041730, INV-044311, INV-044317) and the US National Institutes of Health (grant number D43TW010913 to M. N. K. and M. P. O.). The Gambia team's work is also supported by the UK Research and Innovation Medical Research Council (program number MC_UU_00031—Disease Control and Elimination). N. A. C. is an NIHR Senior Investigator (NIHR203756). N. A. C., J. C., and K. C. J. are affiliated with the NIHR Global Health Research Group on Gastrointestinal Infections at the University of Liverpool, and with the NIHR Health Protection Research Unit in Gastrointestinal Infections at the University of Liverpool, a partnership with the UK Health Security Agency in collaboration with the University of Warwick.
Supplement sponsorship . This article appears as part of the supplement “Enterics for Global Health (EFGH) Shigella Surveillance Study-Rationale and Methods,” sponsored by the Bill & Melinda Gates Foundation.
References
- 1. Christopher PR, David KV, John SM, Sankarapandian V. Antibiotic therapy for Shigella dysentery. Cochrane Database Syst Rev 2010; 2010:CD006784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Mendizábal-Morris CA, Mata LJ, Gangarosa EJ, Guzmán G. Epidemic Shiga-bacillus dysentery in Central America. Derivation of the epidemic and its progression in Guatemala, 1968-69. Am J Trop Med Hyg 1971; 20:927–33. [DOI] [PubMed] [Google Scholar]
- 3. Pavlinac PB, Platts-Mills J, Liu J, et al. Azithromycin for bacterial watery diarrhea: a reanalysis of the AntiBiotics for Children with severe Diarrhea (ABCD) trial incorporating molecular diagnostics [manuscript published online ahead of print 5 July 2023]. J Infect Dis 2023. doi: 10.1093/infdis/jiad252 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Nasrin D, Blackwelder WC, Sommerfelt H, et al. Pathogens associated with linear growth faltering in children with diarrhea and impact of antibiotic treatment: the Global Enteric Multicenter Study. J Infect Dis 2021; 224:S848–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Buchwald AG, Verani JR, Keita AM, et al. Etiology, presentation, and risk factors for diarrheal syndromes in 3 sub-Saharan African countries after the introduction of rotavirus vaccines from the Vaccine Impact on Diarrhea in Africa (VIDA) study. Clin Infect Dis 2023; 76:S12–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. World Health Organization . The treatment of diarrhoea: a manual for physicians and other senior health workers. 2005. Available at: https://www.who.int/publications/i/item/9241593180. Accessed 12 June 2023.
- 7. Platts-Mills JA, Liu J, Rogawski ET, et al. Use of quantitative molecular diagnostic methods to assess the aetiology, burden, and clinical characteristics of diarrhoea in children in low-resource settings: a reanalysis of the MAL-ED cohort study. Lancet Glob Health 2018; 6:e1309–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Kasumba IN, Badji H, Powell H, et al. Shigella in Africa: new insights from the Vaccine Impact on Diarrhea in Africa (VIDA) study. Clin Infect Dis 2023; 76:S66–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Pavlinac PB, Platts-Mills JA, Tickell KD, et al. The clinical presentation of culture-positive and culture-negative, quantitative polymerase chain reaction (qPCR)–attributable shigellosis in the Global Enteric Multicenter Study and derivation of a Shigella severity score: implications for pediatric Shigella vaccine trials. Clin Infect Dis 2021; 73:e569–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. McQuade ETR, Shaheen F, Kabir F, et al. Epidemiology of Shigella infections and diarrhea in the first two years of life using culture-independent diagnostics in 8 low-resource settings. PLoS Negl Trop Dis 2020; 14:e0008536. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Mattock E, Blocker AJ. How do the virulence factors of Shigella work together to cause disease? Front Cell Infect Microbiol 2017; 7:64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Kosek M, Ahmed T, Bhutta Z, et al. Causal pathways from enteropathogens to environmental enteropathy: findings from the MAL-ED birth cohort study. EBioMedicine 2017; 18:109–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Paccagnini S, Fontana M, Ceriani R, et al. Occult blood and faecal leucocyte tests in acute infectious diarrhoea in children. Lancet 1987; 329:442. [DOI] [PubMed] [Google Scholar]
- 14. Korzeniowski OM, Barada FA, Rouse JD, Guerrant RL. Value of examination for fecal leukocytes in the early diagnosis of shigellosis. Am J Trop Med Hyg 1979; 28:1031–35. [DOI] [PubMed] [Google Scholar]
- 15. Hossain MA, Albert MJ. Effect of duration of diarrhoea and predictive values of stool leucocytes and red blood cells in the isolation of different serogroups or serotypes of Shigella. Trans R Soc Trop Med Hyg 1991; 85:664–66. [DOI] [PubMed] [Google Scholar]
- 16. Huicho L, Sanchez D, Contreras M, et al. Occult blood and fecal leukocytes as screening tests in childhood infectious diarrhea: an old problem revisited. Pediatr Infect Dis J 1993; 12:474–77. [DOI] [PubMed] [Google Scholar]
- 17. Khan AI, Huq S, Malek MA, et al. Analysis of fecal leukocytes and erythrocytes in Shigella infections in urban Bangladesh. Southeast Asian J Trop Med Public Health 2006; 37:747–54. [PubMed] [Google Scholar]
- 18. Mercado EH, Ochoa TJ, Ecker L, et al. Fecal leukocytes in children infected with diarrheagenic Escherichia coli. J Clin Microbiol 2011; 49:1376–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Lee HM, Lee S, Lee B-I, et al. Clinical significance of fecal lactoferrin and multiplex polymerase chain reaction in patients with acute diarrhea. Gut Liver 2015; 9:636–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Choi SW, Park CH, Silva TM, Zaenker EI, Guerrant RL. To culture or not to culture: fecal lactoferrin screening for inflammatory bacterial diarrhea. J Clin Microbiol 1996; 34:928–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Huicho L, Garaycochea V, Uchima N, Zerpa R, Guerrant RL. Fecal lactoferrin, fecal leukocytes and occult blood in the diagnostic approach to childhood invasive diarrhea. Pediatr Infect Dis J 1997; 16:644–47. [DOI] [PubMed] [Google Scholar]
- 22. Ruiz-Peláez JG, Mattar S. Accuracy of fecal lactoferrin and other stool tests for diagnosis of invasive diarrhea at a Colombian pediatric hospital. Pediatr Infect Dis J 1999; 18:342–6. [DOI] [PubMed] [Google Scholar]
- 23. Duman M, Gencpinar P, Biçmen M, et al. Fecal calprotectin: can be used to distinguish between bacterial and viral gastroenteritis in children? Am J Emerg Med 2015; 33:1436–9. [DOI] [PubMed] [Google Scholar]
- 24. Chen C-C, Huang J-L, Chang C-J, Kong M-S. Fecal calprotectin as a correlative marker in clinical severity of infectious diarrhea and usefulness in evaluating bacterial or viral pathogens in children. J Pediatr Gastroenterol Nutr 2012; 55:541–7. [DOI] [PubMed] [Google Scholar]
- 25. Lepe-Balsalobre E, Rubio-Sánchez R, Úbeda C, Úbeda Ontiveros JM. Diagnostic utility of fecal calprotectin in chronic diarrhea of bacterial etiology in pediatric patients. Enferm Infecc Microbiol Clin (Engl Ed) 2021; 39:307–8. [DOI] [PubMed] [Google Scholar]
- 26. Shastri YM, Bergis D, Povse N, et al. Prospective multicenter study evaluating fecal calprotectin in adult acute bacterial diarrhea. Am J Med 2008; 121:1099–106. [DOI] [PubMed] [Google Scholar]
- 27. Sýkora J, Siala K, Huml M, Varvařovská J, Schwarz J, Pomahačová R. Evaluation of faecal calprotectin as a valuable non-invasive marker in distinguishing gut pathogens in young children with acute gastroenteritis. Acta Paediatr 2010; 99:1389–95. [DOI] [PubMed] [Google Scholar]
- 28. Covidence . Systematic review tool. Available at: www.covidence.org. Accessed 12 June 2023.
- 29. Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med Res Methodol 2003; 3:25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Ismail . Indicators for antibiotic therapy in invasive bacterial diarrhoea—record details—Embase. 1994. Available at: https://www.embase.com/records?subaction=viewrecord&rid=2&page=1&id=L25048116. Accessed 10 July 2023.
- 31. Park Y, Son M, Jekarl DW, Choi HY, Kim SY, Lee S. Clinical significance of inflammatory biomarkers in acute pediatric diarrhea. Pediatr Gastroenterol Hepatol Nutr 2019; 22:369–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Atlas HE, Conteh B, Islam MT, et al. Diarrhea case surveillance in the Enterics for Global Health Shigella surveillance study: epidemiologic methods. Open Forum Infect Dis 2024; 11(Suppl 1):S6–16. [DOI] [PMC free article] [PubMed]
- 33. Immundiagnostik AG. IDK hemoglobin ELISA: manual. 2022. Available at: https://www.immundiagnostik.com/media/pages/testkits/k-7816d/8a2f137078-1679409720/k7816d_2022-02-22_haemoglobin.pdf. Accessed 14 July 2023.
- 34. Immundiagnostik AG. IDK calprotectin ELISA: manual. 2022. Available at: https://www.immundiagnostik.com/media/pages/testkits/k-6927/ef2d689aad-1679409720/k6927_2022-02-15_idk_calprotectin_stuhl_1h.pdf. Accessed 14 July 2023.
- 35. Immundiagnostik AG. IDK MPO ELISA: manual. 2022. Available at: https://www.immundiagnostik.com/media/pages/testkits/k-6630/10684925dd-1679409720/k6630_2022-01-26_mpo.pdf. Accessed 14 July 2023.
- 36. Eagle Biosciences . NGAL (stool) ELISA assay kit: package insert. Available at: https://eaglebio.com/wp-content/uploads/2014/06/NGL35-K01_NGAL_Stool_ELISA_Assay_Kit_Package_Insert_v7.1_07.2023-1.pdf. Accessed 14 July 2023.
- 37. Liu J, Garcia Bardales PF, Islam K. Shigella detection and molecular serotyping with a customized TaqMan Array Card in the Enterics for Global Health (EFGH): Shigella surveillance study. Open Forum Infect Dis 2024; 11(Suppl 1):S34–40. [DOI] [PMC free article] [PubMed]
- 38. Horne B, Badji H, Bhuiyan TR, et al. Microbiological methods used in the Enterics for Global Health Shigella surveillance study. Open Forum Infect Dis 2024; 11(Suppl 1):S25–33. [DOI] [PMC free article] [PubMed]
- 39. Rogawski ET, Platts-Mills JA, Seidman JC, et al. Use of antibiotics in children younger than two years in eight countries: a prospective cohort study. Bull World Health Organ 2017; 95:49–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Freedman SB, Xie J, Neufeld MS, et al. Shiga toxin–producing Escherichia coli infection, antibiotics, and risk of developing hemolytic uremic syndrome: a meta-analysis. Clin Infect Dis 2016; 62:1251–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Liu J, Platts-Mills JA, Juma J, et al. Use of quantitative molecular diagnostic methods to identify causes of diarrhoea in children: a reanalysis of the GEMS case-control study. Lancet 2016; 388:1291–301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Levine MM, Nasrin D, Acácio S, et al. Diarrhoeal disease and subsequent risk of death in infants and children residing in low-income and middle-income countries: analysis of the GEMS case-control study and 12-month GEMS-1A follow-on study. Lancet Glob Health 2020; 8:e204–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Bardhan PK, Beltinger J, Beltinger RW, Hossain A, Mahalanabis D, Gyr K. Screening of patients with acute infectious diarrhoea: evaluation of clinical features, faecal microscopy, and faecal occult blood testing. Scand J Gastroenterol 2000; 35:54–60. [DOI] [PubMed] [Google Scholar]
- 44. Beltinger J, Walther R, Bardhan P, Mahalanabis D, Gyr K. Immunological testing for occult blood in patients with acute infectious diarrhea (can it improve the specificity of the guaiac test?). Dig Dis Sci 1997; 42:366–71. [DOI] [PubMed] [Google Scholar]
- 45. Bodhidatta L, Vjthayasai N, Eimpokalarp B, Pitarangsi C, Serichantalergs O, Isenbarger D. Bacterial enteric pathogens in children with acute dysentery in Thailand: increasing importance of quinolone-resistant Campylobacter. Southeast Asian J Trop Med Public Health 2002; 33:752–7. [PubMed] [Google Scholar]
- 46. Chang H, Zhang L, Ge Y, et al. A hospital-based case-control study of diarrhea in children in Shanghai. Pediatr Infect Dis J 2017; 36:1057–63. [DOI] [PubMed] [Google Scholar]
- 47. Jindal N, Arora S. Role of faecal leucocytes in the diagnostic evaluation of acute diarrhoea. Indian J Med Sci 1991; 45:261–64. [PubMed] [Google Scholar]
- 48. McNeely WS, Dupont HL, Mathewson JJ, Oberhelman RA, Ericsson CD. Occult blood versus fecal leukocytes in the diagnosis of bacterial diarrhea: a study of U.S. travelers to Mexico and Mexican children. Am J Trop Med Hyg 1996; 55:430–33. [DOI] [PubMed] [Google Scholar]
- 49. Mshana SE, Joloba ML, Kakooza A, Kaddu-Mulindwa D. Role of microscopic examination of stool specimens in the diagnosis of Campylobacter infection from children with acute diarrhoea in Kampala, Uganda. Tanzan J Health Res 2010; 12:100–3. [DOI] [PubMed] [Google Scholar]
- 50. Nordlander E, Phuphaisan S, Bodhidatta L, Arthur J, Echeverria P. Microscopic examination of stools and a latex slide agglutination test for the rapid identification of bacterial enteric infections in Khmer children. Diagn Microbiol Infect Dis 1990; 13:273–6. [DOI] [PubMed] [Google Scholar]
- 51. Patwari AK, Deb M, Dudeja M, Jayasheela M, Agarwal A, Singh P. Clinical and laboratory predictors of invasive diarrhoea in children less than five years old. J Diarrhoeal Dis Res 1993; 11:211–6. [PubMed] [Google Scholar]
- 52. Sebodo T, Soetarjo T, Sadjimin T, Soenarto Y, Sanborn WR. Study on the etiology of diarrhea. J Trop Pediatr Environ Child Health 1978; 24:107–9. [DOI] [PubMed] [Google Scholar]
- 53. Venkataraman S, Ramakrishna BS, Kang G, Rajan DP, Mathan VI. Faecal lactoferrin as a predictor of positive faecal culture in south Indian children with acute diarrhoea. Ann Trop Paediatr 2003; 23:9–13. [DOI] [PubMed] [Google Scholar]
- 54. Pender MA, Smith T, Brintz BJ, et al. Weather variables as important clinical predictors of bacterial diarrhoea among international travellers. J Travel Med 2022; 29:taac012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55. Wang Y, Zhang TP, Xiao HL, Qi HY, Yin CH. Formulation of an early warning infectivity score system for adult patients with acute bacterial diarrhea. Biomed Environ Sci 2014; 27:65–9. [DOI] [PubMed] [Google Scholar]
- 56. Alvarado T. Faecal leucocytes in patients with infectious diarrhoea. Trans R Soc Trop Med Hyg 1983; 77:316–20. [DOI] [PubMed] [Google Scholar]
- 57. Alzaher MZ, Almugahwi AA, Almulla AA, Almeer HH, Alshammasi MM, El-Badry AA. Diagnostic yield of stool culture and probable predictive factors, a single-center experience. Acta Biomed 2022; 93:e2022302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58. Ascher DP, Ednsada-Corpus R. Clinical and laboratory predictors of bacterial diarrhea in a tropical environment. Mil Med 1991; 156:74–6. [PubMed] [Google Scholar]
- 59. Caprioli A, Pezzella C, Morelli R, et al. Enteropathogens associated with childhood diarrhea in Italy. Pediatr Infect Dis J 1996; 15:876–83. [DOI] [PubMed] [Google Scholar]
- 60. Denno DM, Stapp JR, Boster DR, et al. Etiology of diarrhea in pediatric outpatient settings. Pediatr Infect Dis J 2005; 24:142–48. [DOI] [PubMed] [Google Scholar]
- 61. DuBois D, Binder L, Nelson B. Usefulness of the stool Wight's stain in the emergency department. J Emerg Med 1988; 6:483–6. [DOI] [PubMed] [Google Scholar]
- 62. Fan K, Morris AJ, Reller LB. Application of rejection criteria for stool cultures for bacterial enteric pathogens. J Clin Microbiol 1993; 31:2233–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63. Koplan JP, Fineberg HV, Ferraro MJ, Rosenberg ML. Value of stool cultures. Lancet 1980; 316:413–6. [DOI] [PubMed] [Google Scholar]
- 64. McIver CJ, Hansman G, White P, Doultree JC, Catton M, Rawlinson WD. Diagnosis of enteric pathogens in children with gastroenteritis. Pathology 2001; 33:353–8. [PubMed] [Google Scholar]
- 65. Bouckenooghe AR, Dupont HL, Jiang ZD, et al. Markers of enteric inflammation in enteroaggregative Escherichia coli diarrhea in travelers. Am J Trop Med Hyg 2000; 62:711–3. [DOI] [PubMed] [Google Scholar]
- 66. Lai C-C, Ji D-D, Wu F-T, et al. Etiology and risk factors of acute gastroenteritis in a Taipei emergency department: clinical features for bacterial gastroenteritis. J Epidemiol 2016; 26:216–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67. Lever RA, Tapper L, Skarbek S, Chiodini PL, Armstrong M, Bailey RL. Predictors of aetiology and outcomes of acute gastrointestinal illness in returning travellers: a retrospective cohort analysis. BMC Infect Dis 2021; 21:599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68. Loosli J, Gyr K, Stalder H, et al. Etiology of acute infectious diarrhea in a highly industrialized area of Switzerland. Gastroenterology 1985; 88:75–9. [DOI] [PubMed] [Google Scholar]
- 69. Miller JR, Barrett LJ, Kotloff K, Guerrant RL. A rapid test for infectious and inflammatory enteritis. Arch Intern Med 1994; 154:2660–4. [DOI] [PubMed] [Google Scholar]
- 70. Scerpella EG, Okhuysen PC, Mathewson JJ, et al. Evaluation of a new latex agglutination test for fecal lactoferrin in travelers’ diarrhea. J Travel Med 1994; 1:68–71. [DOI] [PubMed] [Google Scholar]
- 71. Siegel D, Cohen P, Neighbor M, et al. Predictive value of stool examination in acute diarrhea. Arch Pathol Lab Med 1987; 111:715–8. [PubMed] [Google Scholar]
- 72. Tribble DR, Baqar S, Pang LW, et al. Diagnostic approach to acute diarrheal illness in a military population on training exercises in Thailand, a region of Campylobacter hyperendemicity. J Clin Microbiol 2008; 46:1418–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 73. Ashraf H, Beltinger J, Alam NH, et al. Evaluation of faecal occult blood test and lactoferrin latex agglutination test in screening hospitalized patients for diagnosing inflammatory and non-inflammatory diarrhoea in Dhaka, Bangladesh. Gastroenterologia 2007; 76:256–61. [DOI] [PubMed] [Google Scholar]
- 74. Bardhan PK, Beltinger J, Beltinger RW, Hossain A, Mahalanabis D, Gyr K. Screening of patients with acute infectious diarrhoea: evaluation of clinical features, faecal microscopy, and faecal occult blood testing. Scand J Gastroenterol 2000; 35:54–60. [DOI] [PubMed] [Google Scholar]
- 75. Aly SM, El-Zawawy LA, Said DE, Fathy FM, Mohamed ON. The utility of lactoferrin in differentiating parasitic from bacterial infections. J Egypt Soc Parasitol 2005; 35:1149–62. [PubMed] [Google Scholar]
- 76. Ahn JS, Seo SI, Kim J, et al. Efficacy of stool multiplex polymerase chain reaction assay in adult patients with acute infectious diarrhea. World J Clin Cases 2020; 8:3708–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 77. Berger C, Loitsch SM, Hartmann F, Stein J. 657 comparative evaluation of fecal calprotectin and S100A12 as non-invasive markers in predicting microbiological diagnosis for acute bacterial diarrhea: prospective multicenter study. Gastroenterology 2010; 5:S88. [Google Scholar]
- 78. Czub E, Nowak JK, Moczko J, et al. Comparison of fecal pyruvate kinase isoform M2 and calprotectin in acute diarrhea in hospitalized children. Sci Rep 2014; 4:4769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 79. Kim HJ. Efficacy of fecal calprotectin combined with stool hemoglobin in differentiating bacterial origin in acute gastroenteritis. Pediatr Emerg Care 2022; 38:e670–3. [DOI] [PubMed] [Google Scholar]
- 80. Aratani Y. Myeloperoxidase: its role for host defense, inflammation, and neutrophil function. Arch Biochem Biophys 2018; 640:47–52. [DOI] [PubMed] [Google Scholar]
- 81. Khan AA, Alsahli MA, Rahmani AH. Myeloperoxidase as an active disease biomarker: recent biochemical and pathological perspectives. Med Sci (Basel) 2018; 6:33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 82. Hansberry DR, Shah K, Agarwal P, et al. Fecal myeloperoxidase as a biomarker for inflammatory bowel disease. Cureus 2017; 9:e1004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 83. Guerrant RL, Bolick DT, Swann JR. Modeling enteropathy or diarrhea with the top bacterial and protozoal pathogens: differential determinants of outcomes. ACS Infect Dis 2021; 7:1020–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 84. Sharbatdaran M, Holaku A, Kashifard M, et al. Fecal calprotectin level in patients with IBD and noninflammatory disease of colon: a study in Babol, northern Iran. Caspian J Intern Med 2018; 9:60–4. [PMC free article] [PubMed] [Google Scholar]
- 85. Shastri YM, Bergis D, Povse N, et al. Prospective multicenter study evaluating fecal calprotectin in adult acute bacterial diarrhea. Am J Med 2008; 121:1099–106. [DOI] [PubMed] [Google Scholar]
- 86. Jaberi SA, Cohen A, D'Souza C, et al. Lipocalin-2: structure, function, distribution and role in metabolic disorders. Biomed Pharmacother 2021; 142:112002. [DOI] [PubMed] [Google Scholar]
- 87. Moschen AR, Adolph TE, Gerner RR, Wieser V, Tilg H. Lipocalin-2: a master mediator of intestinal and metabolic inflammation. Trends Endocrinol Metab 2017; 28:388–97. [DOI] [PubMed] [Google Scholar]
- 88. Li D, Yan Sun W, Fu B, Xu A, Wang Y. Lipocalin-2—the myth of its expression and function. Basic Clin Pharmacol Toxicol 2020; 127:142–51. [DOI] [PubMed] [Google Scholar]
- 89. Q S Medeiros PH, Ledwaba SE, Bolick DT, et al. A murine model of diarrhea, growth impairment and metabolic disturbances with Shigella flexneri infection and the role of zinc deficiency. Gut Microbes 2019; 10:615–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 90. Chassaing B, Srinivasan G, Delgado MA, Young AN, Gewirtz AT, Vijay-Kumar M. Fecal lipocalin 2, a sensitive and broadly dynamic non-invasive biomarker for intestinal inflammation. PLoS One 2012; 7:e44328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 91. Walker HK, Hall WD, Hurst JW, eds. Clinical methods: the history, physical, and laboratory examinations. 3rd ed. Boston, MA: Butterworths; 1990. [PubMed] [Google Scholar]
- 92. Mooiweer E, Fidder HH, Siersema PD, Laheij RJF, Oldenburg B. Fecal hemoglobin and calprotectin are equally effective in identifying patients with inflammatory bowel disease with active endoscopic inflammation. Inflamm Bowel Dis 2014; 20:307–14. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.