Performance of Routine C4d and C3d Immunostaining in Asymptomatic Heart Transplant EMB

Performance of routine C4d and C3d immunostaining on protocol EMBs in a prospective and unselected cohort of heart transplant patients

A Loupy, A Cazes, R Guillemain, M Tible, C Amrein, V Pezzella, JN Fabiani, C Suberbielle, GS Hill, D Nochy, P Bruneval and JP Duong-van-Huyen

Introduction: In heart transplants, subclinical forms of antibody mediated rejection (AMR) are increasingly recognized in protocol EMBs as an emerging cause of late transplant failure. In this setting, the clinical relevance of systematic C4d/C3d evaluation in unselected protocol EMBs remains to be determined.

Methods: This prospective study included all the protocol EMBs performed in our institution from October 1st 2009 to September 30th 2010. A strategy based on systematic immunoperoxidase C4d/C3d staining was applied in addition to the histological detection of microvascular inflammation (MI+) defined by intravascular macrophages. DSA were tested systematically according to the protocol DSA testing program in our institution and also in cases of biopsies suspicious for AMR.

Results: 260 EMBs performed in 100 consenting patients were included. 170/260 EMBs (65.4%) showed morphological evidence of cellular rejection (ISHLT 2005: grade 1R=61.9%, 2R=2.7%, 3R=0.8%). The prevalence of positive C4d in the entire cohort was 3.1%. A total of 38/260 EMBs (14.6%) showed histological evidence of microvascular inflammation (MI+) in which 8.3% were positive for C4d and 55.6% for C3d staining, respectively (sensitivity for C4d vs C3d: p<0.01). Among the MI+ biopsies, 42.2% did not show any C4d or C3d deposition. Finally, the prevalence of positive DSA in the MI+ group without C4d/C3d deposition was 73.3% (mean MFImaxDSA= 2906±1043) as compared to 77.3% in the MI+-C4d/C3d-positive group (mean MFImaxDSA= 2880±739, p=NS).

Conclusion: C4d staining has a high specificity but very low sensitivity in protocol EMBs performed in clinically stable patients. C3d has a significantly higher sensitivity than C4d alone to predict concomitant microvascular inflammation. Importantly, microvascular inflammation, even without C4d/C3d deposition is associated with circulating DSA. These data suggest that C4d/C3d positivity may not be sufficient for the diagnosis of AMR, excluding patients from this specific diagnosis.