Antibody-based detection tests for the diagnosis of Helicobacter pylori infection in children: a meta-analysis

Abstract

Background: Numerous serologic tests are available for the diagnosis of H. pylori infection in children. Common designs of antibody-based detection tests are ELISA and Western Blot (WB). For developing countries with limited laboratory resources and access, ELISA would be the preferred method because of its simplicity, lower cost and speed. Although in adults ELISA has proven to be highly accurate in diagnosing H. pylori infection; in children, it has shown variable accuracy.

Methods/findings: We conducted a systematic review and meta-analysis to assess the accuracy of antibody-based detection tests for the diagnosis of H. pylori infection in children. Selection criteria included participation of at least 30 children and the use of a gold standard for H. pylori diagnosis. In a comprehensive search we identified 68 studies. Subgroup analyses were carried out by technique, immunoglobulin class, and source of test (commercial and in-house). The results demonstrated: 1) WB tests showed high overall performance, sensitivity 91.3% (95% CI, 88.9-93.3), specificity 89% (95% CI, 85.7-91.9), LR+ 8.2 (95% CI, 5.1-13.3), LR- 0.06 (95% CI, 0.02-0.16), DOR 158.8 (95% CI, 57.8-435.8); 2) ELISA-IgG assays showed low sensitivity 79.2% (95% CI, 77.3-81.0) and high specificity (92.4%, 95% CI, 91.6-93.3); 3) ELISA commercial tests varied widely in performance (test for heterogeneity p<0.0001); and 4) In-house ELISA with whole-cell antigen tests showed the highest overall performance: sensitivity 94% (95% CI, 90.2-96.7), specificity 96.4% (95% CI, 94.2-97.9), LR+ 19.9 (95% CI, 7.9-49.8), LR- 0.08 (95% CI, 0.04-0.15) DOR 292.8 (95% CI, 101.8-841.7).

Conclusions/significance: WB test and in-house ELISA with whole-cell antigen tests are the most reliable tests for the diagnosis of H. pylori infection in children. Antigens obtained from local strains of the community could partially explain the good overall accuracy of the in-house ELISA. Because of its cost and technical demands, in-house ELISA might be more suitable for use in developing countries.

Figure 1. Study selection process and reasons for exclusion.

Figure 2. Forest Plot for the Sensitivity and Specificity of the Serum IgG-ELISA Diagnostic Tests.

The squares and lines represent the point estimates and 95% CIs, respectively. The size of square indicates the study size. The pooled estimated is denoted by a diamond at the bottom. a = Enzygnost II, b = Pyloritest, c = MedMira; d = FlexSure, e = Eurospital, f = Eurospital-CagA, g = Cobas II, h = HM-CAP, i = In-house WC, j = In-house Urease, k = In-house CagA, l = GAP-Test, m = HEL-p II, n = Cobas I, o = InmunoLISA, p = Helico-G, q = Pyloriset, r = Helory, s = Platelia, t = Immulite, u = PyloriStat, v = GAP-Biomerica

Figure 3. Summary Receiver Operator Curve (SROC) for Serum IgG-ELISA Diagnostic Tests.

Each solid square represents an individual study in the meta-analysis. The curve is the regression line that summarizes the overall diagnostic accuracy. AUC = area under curve, SE (AUC) = standard error of AUC, Q* = index defined by point of the SROC curve where the sensitivity and specificity are equal; SE (Q*) = standard error of Q* index.

Figure 4. Forest Plot for the Sensitivity and Specificity of the Western Blot Diagnostic Tests.

The squares and lines represent the point estimates and 95%CIs, respectively. The size of square indicates the study size. The pooled estimated is denoted by a diamond at the bottom. a = HelicoBlot 2.0, b = HelicoBlot 2.1, c = HelicoBlot 2.1 CIM; d = In-house

Figure 5. Summary Receiver Operator Curve (SROC) for Western Blot Diagnostic Tests.

Each solid square represents an individual study in the meta-analysis. The curve is the regression line that summarizes the overall diagnostic accuracy. AUC = area under curve, SE (AUC) = standard error of AUC, Q* = index defined by point of the SROC curve where the sensitivity and specificity are equal; SE (Q*) = standard error of Q* index.

Figure 6. Summary Receiver Operator Curves (SROC) for commercial and in-house tests from Serum IgG-ELISA.

Each solid square represents an individual study in the meta-analysis. The curve is the regression line that summarizes the overall diagnostic accuracy. AUC = area under curve, SE (AUC) = standard error of AUC, Q* = index defined by point of the SROC curve where the sensitivity and specificity are equal; SE (Q*) = standard error of Q* index.