Dengue Literature Review

Oishi2003 - PAIgG and Thrombocytopenia in Secondary Dengue

6 min read

Oishi2003 - PAIgG and Thrombocytopenia in Secondary Dengue

Full citation: Oishi K, Inoue S, Cinco MTDD, Dimaano EM, Alera MTP, Alfon JAR, Abanes F, Cruz DJM, Matias RR, Matsuura H, Hasebe F, Tanimura S, Kumatori A, Morita K, Natividad FF, Nagatake T. (2003). Correlation between increased platelet-associated IgG and thrombocytopenia in secondary dengue virus infections. Journal of Medical Virology, 71(2), 259–264. https://doi.org/10.1002/jmv.10478

Raw file: [[raw/oishi2003.pdf]]

Summary

This prospective hospital-based study investigated the relationship between platelet-associated IgG (PAIgG) and platelet count in 53 patients in the acute phase of secondary dengue infection, enrolled at San Lazaro Hospital, Manila, Philippines (June–November 2001). PAIgG levels were measured by competitive ELISA; circulating anti-platelet IgG autoantibodies were assessed by mixed passive hemagglutination; and IgG was eluted from platelets of a subset of patients and tested by indirect ELISA against a four-serotype dengue antigen mixture to determine whether bound IgG was anti-dengue rather than anti-platelet in specificity.

The study found a significant inverse correlation between PAIgG levels and platelet counts (r = −0.570, P < 0.0001), which was absent in 43 healthy volunteers (r = −0.116, P = 0.461). Anti-platelet IgG autoantibody was detected in the plasma of only 1/53 patients — nearly absent. In contrast, anti-dengue virus IgG was demonstrably present in platelet eluates from all 8 secondary-infection patients tested (OD 0.339 ± 0.114 vs. 0.089 ± 0.005 in healthy controls). A 12-patient longitudinal subset confirmed that PAIgG dropped and platelet counts recovered in tandem from acute to convalescent phase (4 days later).

The authors propose that immune complexes of dengue virus antigen + anti-dengue IgG adhere to platelets via direct dengue-platelet binding (not via FcγRII), and that these complexes trigger platelet clearance by macrophages and/or complement-mediated lysis — producing the transient thrombocytopenia of acute secondary infection.

Study Design

  • Type: Prospective hospital-based cohort
  • Sample size: 53 patients (acute secondary dengue); 12 patients (longitudinal acute-to-convalescent subset, 4-day interval); 8 patients (platelet IgG elution subset); 43 healthy volunteers (control)
  • Setting: San Lazaro Hospital (Blood Borne Diseases Ward) and St. Luke’s Medical Center, Philippines, June–November 2001
  • Population: Mean age 16.1 ± 7.6 years; 60.4% male; all confirmed secondary dengue by HI titer ≥1:2,560; mean days from symptom onset 5.9 ± 1.0

Key Findings

  • Inverse correlation between PAIgG and platelet count in secondary dengue: r = −0.570, P < 0.0001 (n=53 patients). No correlation in healthy volunteers (r = −0.116, P = 0.461).
  • PAIgG significantly elevated in thrombocytopenic patients (platelet <100 × 10³/μl): 0.86 ± 0.83 ng/10⁷ platelets vs. 0.27 ± 0.20 (non-thrombocytopenic patients) and 0.19 ± 0.13 (healthy controls), both P < 0.001.
  • Anti-platelet IgG autoantibody essentially absent in plasma: only 1/53 patients (1.9%) positive; 0/43 healthy controls.
  • Anti-dengue virus IgG confirmed in platelet eluates: OD 0.339 ± 0.114 in 8/8 acute secondary patients vs. OD 0.089 ± 0.005 in 6 healthy controls — demonstrating that the IgG deposited on platelets is directed against dengue virus antigen, not against platelet self-antigens.
  • Longitudinal recovery (n=12): platelet counts rose from 43 ± 36 to 307 ± 129 × 10³/μl (P = 0.0022); PAIgG fell from 1.49 ± 1.23 to 0.31 ± 0.11 ng/10⁷ platelets (P = 0.0037) — tracking in tandem.
  • No correlation between PAIgG and plasma HI titre (r = −0.130, P = 0.361), suggesting circulating anti-dengue IgG titre alone is insufficient to drive PAIgG formation — viral antigen on the platelet surface appears required.
  • 26.2% of thrombocytopenic patients (11/42) had PAIgG within the normal range, indicating other mechanisms (bone marrow suppression, DIC) contribute alongside PAIgG.
  • FcγRII not required for dengue IgG–platelet interaction: the authors cite Wang et al. (1995) showing dengue-2 virus binds platelets in the presence of a virus-specific antibody and that anti-FcγRII monoclonal antibody does not inhibit this binding.

Methods Used

  • IgM-IgG Serology ELISA (IgM-capture ELISA for dengue diagnosis, Bundo & Igarashi 1985)
  • Hemagglutination Inhibition Test (HI assay; titer ≥1:2,560 used to classify secondary infection; all four DENV serotype antigens tested)
  • Competitive ELISA for PAIgG (Mitsubishi Kagaku BCL kit; Kawaguchi 1992 method; results as ng IgG per 10⁷ platelets)
  • Mixed passive hemagglutination for plasma anti-platelet IgG (Olympus kit)
  • Platelet IgG elution + indirect ELISA for anti-dengue virus IgG specificity (four-serotype antigen mixture; sucrose-gradient-purified dengue virus from Aedes albopictus C6/36 cells)

Entities Mentioned

Concepts Addressed

Relevance & Notes

Critical relationship to Lin2001: The two papers together establish an infection-order-dependent bifurcation of thrombocytopenia mechanism in dengue:

  • Primary infection (Lin2001, DENV-3 Taiwan): Thrombocytopenia driven by IgM anti-platelet autoantibodies generated through NS1 molecular mimicry; complement-mediated platelet lysis; severity-correlated.
  • Secondary infection (Oishi2003, Philippines): Thrombocytopenia driven by anti-dengue virus IgG immune complexes deposited on platelets via direct dengue-platelet binding; FcγRII NOT required; essentially no anti-platelet autoantibodies.

The mechanistic shift from autoantibody-mediated (primary) to immune-complex-mediated (secondary) thrombocytopenia parallels the broader distinction in dengue immunopathology between primary and secondary infections. It also means that FcγRIIa genotype (H131/R131), which modulates IgG immune complex handling, may be relevant to PAIgG formation and platelet clearance in secondary infection — yet the paper explicitly states FcγRII is not the docking mechanism, creating an apparent paradox worth tracking.

Limitation: The clinical severity category (DF vs. DHF) was undefined for most patients due to insufficient hospitalisation time to assess vascular permeability. The study therefore cannot directly test whether PAIgG correlates with DHF/DSS severity within secondary infection — only that it correlates with thrombocytopenia. The authors acknowledge this gap explicitly.

Limitation: The longitudinal subset (n=12) and elution subset (n=8) are small; the acute-to-convalescent interval is only 4 days, which is sufficient for the acute-phase recovery but does not address longer-term dynamics.

Questions Raised

  • Does the PAIgG mechanism of secondary-infection thrombocytopenia correlate with DHF/DSS severity — or is it a near-universal feature of secondary infection that is severity-independent?
  • Does FcγRIIa genotype (HH/HR/RR) modulate PAIgG levels in secondary infection, even though the initial dengue-platelet binding step does not require FcγRII? (FcγRII could still influence downstream macrophage clearance of PAIgG-coated platelets.)
  • Do primary and secondary dengue patients have qualitatively distinct platelet destruction mechanisms (IgM autoAb-complement in primary vs. immune complex-macrophage/complement in secondary) — or do both operate simultaneously in secondary infection?
  • Is the absence of anti-platelet IgG autoantibodies in secondary infection (Oishi2003) a genuine mechanistic difference from primary infection, or does it reflect the dominance of the immune complex pathway obscuring a smaller autoantibody signal?

Graph View

Backlinks