Table 1.
Characteristics of included studies on diagnostic test effectiveness for schistosomiasis and strongyloidiasis, January 1993–February 2017.
| Study | Quality | Design | Population | Intervention/Outcomes | Results |
|---|---|---|---|---|---|
| Included systematic reviews of diagnostic tests to detect schistosomiasis | |||||
| Danso Appiah et al., 2016 [53] | AMSTAR: 11/11 GRADE: low to moderate-quality evidence |
Systematic review and meta-analysis | Preschool children and infants, school-aged children or adults from high-/low-prevalence locations | Intervention: POC CCA for Sc. mansoni Outcomes: detection of egg-positive urine—sensitivity/specificity (95% CI:) |
Sensitivity/specificity (95% CI) POC CCA (single standard) 90% (84–94)/56% (39–71); POC CCA (duplicate standard) 85% (80–88)/66% (53–76); POC CCA (triplicate standard) 91% (84–95)/56% (39–72) |
| Yang et al., 2015 [50] | AMSTAR: 11/11 GRADE: low to moderate-quality evidence |
Meta-analysis | Patients infected with schistosomiasis in endemic areas; mainly school children, Africa and China | Intervention: questionnaire screening for Schistosoma species. Outcomes: sensitivity/specificity (95% CI:) |
Sensitivity/specificity (95%CI:) Sc. haematobium 85% (84–86)/94% (94–94); Sc. mansoni 46% (45–47)/81% (80–82); Sc. japonicum 82% (79–85)/59% (57–60) |
| Ochodo et al., 2015 [44] | AMSTAR: 11/11 GRADE: very low to low-quality evidence |
Systematic review and meta-analysis of RCTs | Individuals with active infection with S. haematobium | Intervention: urine reagent strip tests; circulating antigen tests in urine/serum Outcomes: sensitivity/specificity (95% CI:) |
Sensitivity/specificity (95% CI) Sc. haematobuim: microhaematuria 75% (71–79)/87% (84–90); proteinuria 61% (53–68)/82% (77–88); leukocyturia 58% (44–71)/61% (34–88); Sc. mansoni (CCA test) 89% (86–92)/55% (46–65) |
| King and Bertsch, 2013 [45] | AMSTAR: 11/11 GRADE: low-quality evidence |
Systematic review and meta-analysis of surveys | Schools, communities with high/low prevalence, low intensity groups in Africa | Intervention: dipstick test Sc. haematobium. Outcomes: sensitivity and specificity (95% CI:), diagnostic odds ratio (DOR) |
Sensitivity/specificity (95% CI) Detection of egg-positive urine 81% (79–83)/89% (87–92). In high-prevalence settings 80% (78–83)/86% (82–90); lower in treated population 72% (61–78)/87% (81–94); in lower intensity population subgroups 65% (58–72)/82% (76–90) |
| Wang, et al., 2012 [46] | AMSTAR: 7/11 GRADE: very low- to low-quality evidence |
Systematic review and meta-analysis of RCTs, retro-/pro-observational studies | Infected patients with schistosomiasis in control programmes in China | Intervention: IHA and ELISA. Outcomes: true positive rates, sensitivity/specificity (95% CI:), DOR |
Sensitivity/specificity (95% CI) IHA 75.6% (74–77)/73% (72–74) ELISA 84.9% (83–87)/50.4% (49.2–51.6) The DOR of IHA was 9.41 (95% CI: 5–18), and ELISA 4.78 (95% CI: 3.21–7.13) |
| Included primary studies of diagnostic tests to detect schistosomiasis | |||||
| Espirito-Santo et al., 2015 [57] | QUADAS-2-11/14 GRADE: very low- to low-quality evidence |
Cross-sectional epidemiological survey in areas of low prevalence of Sc. Mansoni | The estimated sample size required was 650 individuals; Barra Mansa City, Rio de Janeiro State, Brazil |
Intervention: diagnostic assays: ELISA-IgG/ELISA-IgM/IFT-IgM/qPCR in faeces. Outcomes: sensitivity/specificity (95% CI:) |
Sensitivity/specificity (95% CI) KK 13.8% (4–32)/99.8% (99.0–100); ELISA-IgG 66.7% (48–82)/91.5% (89–94); ELISA-IgM 81.8% (64–93)/82% (79–85); IFT-IgM 78.8% (61–91)/87.7% (84.8–90); qPCR in faeces 51.7% (32–71)/92.6% (90–95); qPCR in serum 12.1% (3–28)/99.1% (98–99) |
| Espirito-Santo et al., 2014a [60] | QUADAS-2-12/14 GRADE: very low- to low-quality evidence |
Cross-sectional study | City of Barra Mansa, Rio de Janeiro State, Brazil, with an estimated prevalence of 1% | Intervention: diagnostic assays: ELISA-IgG and ELISA-IgM. Outcomes: sensitivity/specificity (95% CI:); PPV, NPV |
Sensitivity/specificity (95%CI) ELISA-IgG 60.0% (15–95) /89.1% (86.2–91.5); ELISA-IgM 60.0(15–95)/79.2% (75.6–82.5) PPV/NPV (95%CI): ELISA-IgG 4.6% (1–13) /99.6% (98–100); ELISA-IgM 2.5% (0.5–7); NPV 99.6% (98.4–100.0) |
| Espirito-Santo et al., 2014b [56] | QUADAS-2-13/14 GRADE: very low- to low-quality evidence |
Cross-sectional epidemiological survey | 7000 inhabitants located in the outskirts of Barra Mansa, Rio de Janeiro, Brazil | Intervention: qPCR in serum or faeces. Outcomes: sensitivity/specificity (95% CI:); PPV, NPV |
Sensitivity/specificity (95% CI) qPCR in faeces 80.0% (28–99)/92.4% (90–94); qPCR in serum 20.0% (0.5–71.6)/98.8 (97–99) PPV/NPV (95% CI:): qPCR in faeces 8.0% (2–19)/99.8% (99–100); qPCR in serum 12.5% (0.3–52.7)/99.3% (98.2–99.8) |
| Lodh et al., 2013 [55] | QUADAS-2-12/14 GRADE: very low- to low-quality evidence |
Cross-sectional case study | Filtered urine specimens from infected and not-infected patients in Zambia | Intervention: qPCR ELISA IgG in serum or faeces; filtered Urine PCR. Outcomes: sensitivity/specificity (95% CI:); PPV, NPV |
Sensitivity/specificity (95%CI) KK test 57% (45–68)/100% (69–100); CCA rapid test 65% (56–77)/60% (26–88); PCR 100% (95–100)/100% (69–100) PPV/NPV: KK test 100%/23%; CCA rapid test 93%/19%; PCR 100%/100%. |
| Kinkel et al., 2012 [54] | QUADAS-2-12/14 GRADE: very low- to low-quality evidence |
Retrospective comparative diagnostic study: performance of 8 serological tests for Schistosoma spp | Serum specimens from infected patients and those without the infection in low-prevalence locations or non-endemic settings (Germany) | Intervention: serological assays: IFAT, ELISA-CA, ELISA-AWA, ELISA-SEA, IHA, ELISA-NovaTec, ELISA-DRG and ELISA-Viramed. Outcomes: sensitivity and specificity (95% CI:) |
Sensitivity/specificity-(95% CI): IFAT 75.7% (58–98)/98.1% (92–99); ELISA-CA 40.5% (25–59)/95.2% (89–98); ELISA-AWA 54.1% (37–70)/100% (95.6–100); ELISA-SEA-75.7% (58–98)/97.1% (91–99); IHA 73.0% (55.6–85.6)/99.0% (94.0–100); ELISA-NovaTec 64.9% (47–79)/99 (94–100); ELISA-DRG 78.3% (61.3–89.6)/88.4 (80–94); ELISA-Viramed 67.6% (50–81)/76.9% (67–84). |
| De Frotas et al., 2011 [58] | QUADAS-2-12/14 GRADE: very low- to low-quality evidence |
Cross-sectional survey | Stool and serum specimens from infected and not infected patients, low-endemic setting in Brazil |
Intervention: serological assays, ELISA IgG Outcomes: sensitivity and specificity (95%CI) |
Sensitivity/specificity (95% CI): ELISA-IgG 100% (68–100)/72.9% (67–78). PPV/NPV (95% CI): ELISA-IgG 26.0% (18–36) /100% (97–100). |
| Silveira et al., 2016 [59] | QUADAS-2-12/14 GRADE: very low- to low-quality evidence |
Evaluation of the CCA test to diagnose Sc. mansoni in Minas Gerais State, Brazil. | Infected individuals in regions with moderate to high prevalence | Intervention: CCA-immuno-chromatographic test. Outcomes: sensitivity/specificity (95% CI:) |
Sensitivity/specificity (95% CI): CCA-ICT 68.7% (54–81)/97.6% (87–99) |
| Beltrame et al., 2017 [61] | QUADAS-2-12/14 GRADE: very low- to low-quality evidence |
Accuracy of parasitological and immunological tests for the screening of human schistosomiasis in immigrants and refugees from African countries | Frozen serum specimens from recent African asylum seekers that were routinely screened for schistosomiasis in Italy | Intervention: urine CCA; Bordier-ELISA, Western Blot IgG, ICT IgG-IgM, microscopy compared with composite reference standard. Outcomes: sensitivity/specificity (95% CI:) |
Sensitivity/specificity (95% CI): Urine CCA 29% (22–37)/95% (91–97); Bordier-ELISA 71% (63–78)/99.6% (98–100); Western blot IgG 92% (86–96)/94% (90–97); ICT IgG-IgM 96% (91–99)/83% (77–87); microscopy 45% (37–54)/100% |
| Included systematic reviews for diagnostic effectiveness for strongyloidiasis | |||||
| Campo Polanco et al., 2014 [51] | AMSTAR: 11/11 GRADE: moderate-quality evidence |
Systematic review and meta-analysis | Individuals with active/chronic infection | Intervention: Baermann method, agar plate, direct faecal smear examination and formol-ether concentration technique. Outcome: sensitivity and specificity (95% CI:) |
Sensitivity: Baermann method (72%) with LR+228 and LR−0.32; APC 89%, LR+341 and LR−0.11; stool microscopy 21%, LR + 67 and LR−0.67; formol-ether concentration 48%, LR + 110 and LR−0.59. Specificity: 100% in all four tests. APC and Baermann method are best. |
| Requena-Méndez et al., [19] | AMSTAR: 7/11 GRADE: low- to moderate-quality evidence |
Systematic review | Individuals with active/chronic infection | Intervention: Baermann method, agar plate, direct faecal smear examination and formol-ether concentration technique, serological techniques. Outcome: sensitivity and specificity (95% CI:) |
No meta-analysis was undertaken. Sensitivity and specificity of different techniques were individually reported. |
| Included primary studies for diagnostic effectiveness for strongyloidiasis | |||||
| Bisofi et al., 2014 [62] | QUADAS-2: 13/14 GRADE: low-quality evidence |
Retrospective comparative diagnostic study to evaluate the performance of 5 tests for St. stercoralis. | Serum specimens from subjects with St. stercoralis; healthy people and patients with previous exposure | Intervention: IFAT, NIE-LIPS NIE-ELISA, IVD-ELISA- and Bordier-ELISA Outcome: sensitivity and specificity (95% CI:) |
Sensitivity/specificity (95% CI): NIE-ELISA 75.4% (67–83)/94.8% (91–99); NIE-LIPS 85.1% (78–92)/100% (100–100); IFAT 93.9% (89–98)/92.2% (87–97); IVD-ELISA 91.2% (86–96)/99.1% (97.4–100.0); Bordier-ELISA 89.5% (84–95) 98.3% (96–100). |
| Rascoe et al., 2015 [63] | QUADAS-2: 10/14 GRADE: low-quality evidence |
Retrospective comparative diagnostic study of 5 tests for the follow-up of patients infected with St. stercoralis | Serum samples positive for St. stercoralis and negative samples from United States residents with no history of foreign travel | Intervention: Ss-NIE-1 ELISA, Ss-NIE-1 Luminex. Outcome: sensitivity and specificity (95% CI:) |
Sensitivity/specificity (95% CI): SS-NIE-1 ELISA 95% (92–97)/93% (90–96); Ss-NIE-1 Luminex 93% (88–96)/95% (93–97). The inter-assay coefficient of variation was determined to be 22% for the low-positive control serum and 10% for the medium-positive control serum. |
| Knopp et al., 2014 [64] | QUADAS-2: 11/14 GRADE: low-quality evidence |
International standard randomised controlled trial | Children and adults residing in rural villages in the Baga moyo District, Tanzania (endemic areas) | Intervention: Real-time PCR, FLOTAC technique, KK method. Outcome: sensitivity and specificity (95% CI:) |
Sensitivity/specificity (95% CI): PCR + pseudo-standard PCR 17.4 (8–31)/3.9 (89–97); Baermann + pseudo-standard 47 (23–72)/78.4 (72 -84); PCR + multiple gold standard 30.9 (19.1–44.8)/100 (100–100); Baermann + multiple gold standard 83.6 (71.2–92.2)/100 (100–100) |
AWA: adult worm antigen; AMSTAR: a tool for assessing the methodological quality of systematic reviews; APC: agar plate culture; CA: Cercarial antigen; CCA: circulatory cathodic antigen; CI: confidence interval; DOR: diagnostic odds ratio; GRADE: Grading of Recommendations, Assessment, Development and Evaluation; ELISA: enzyme-linked immunosorbent assay; FLOTAC: novel multivalent faecal egg count method; ICT: Immuno chromatographic test; IFAT: indirect fluorescent antibody technique; IHA: indirect haemagglutination: In Vitro Diagnostic kit; KK: Kato–Katz method; LIPS: luciferase immunoprecipitation system; LR+: positive likelihood ratio; LR−: negative likelihood ratio; NIE: a 31-kDa recombinant antigen; NovaTec: NovaTec Immundiagnostica, Dietzenbach, Germany; NPV: negative predictive value; POC: point-of-care; qPCR: quantitative PCR (real-time polymerase chain reaction); PPV: positive predictive value; RCT: randomised controlled trial; SEA: soluble egg antigen; Ss-NIE-1: a luciferase tagged recombinant protein of St. stercoralis for IgG and IgG4 specific antibodies; QUADAS-2: a tool for the quality assessment of diagnostic accuracy studies; Viramed®: Viramed Biotech, Planegg, Germany).