Electron microscopy in determining the etiology of kidney allograft dysfunction.

BACKGROUND

Biopsy of the transplanted kidney serves a definitive role in elucidating the possible causes of allograft dysfunction. At our transplant clinic we have observed that electron microscopy (EM) does not usually refine the findings initially disclosed by light microscopy with direct immunofluorescence (LM).

METHODS

We studied whether EM results differ from or add to LM results. We compared the reports of 65 allograft biopsies performed on 60 patients over 82 consecutive weeks. We classified biopsy interpretations by 15 possible diagnoses and categorically by glomerular versus nonglomerular disease. We determined agreement between LM and EM reports by Cohen's kappa statistic, and applied the McNemar test to determine whether EM interpretation yielded significantly more glomerular diagnoses on the same biopsy samples.

RESULTS

There was strong agreement (kappa, 0.94: 95% confidence interval [CI], 0.88-1.00), between the EM- and LM-based interpretations. EM did not detect more glomerular disease than LM (discordance rate, 4.6%: 95% CI, -1.92% to 4.62%: P = .25). EM did not add to the diagnosis of rejection. EM described 3 additional cases of transplant glomerulopathy, but did not lead in a change in management of kidney allograft dysfunction.

CONCLUSION

Electron microscopy, used routinely, does not add to light microscopy in the evaluation of kidney transplant dysfunction.