Pathogenesis of extra efferent vessel development in diabetic glomeruli

The development of extra efferent vessels (EEV) is a little-known feature of diabetic glomerulopathy. The only previous large study [Min W, Yamanaka N. Three-dimensional analysis of increased vasculature around the glomerular vascular pole in diabetic nephropathy. Virchows Archiv A Pathol Anat 1993; 423:201-7] known to us found that up to 5 EEV per glomerulus (glom) each drained a separate lobule. Most EEV connected to the second- and third-order branches of the afferent arteriole (AA), and drained into peritubular capillaries. Although not so stated, the illustrations suggested that some EEV could be shunts, and thus detrimental to glom function, and possibly glom health. There was no correlation between the unquantitated presence of increased EEV at the vascular pole (VP) and the severity of the major diabetic glomerular (glom) lesions. The authors opined that efferent arteriole (EA) stenosis by insudative lesions (IL) stimulated the formation of EEV. To confirm and extend these findings, we have repeated the study in 18 diabetic cases with mild to severe, but not end-stage, diffuse and nodular lesions (DL and NL), 8 controls, and the 2 normal traumatic nephrectomy cases. Up to 18 EEV per glom were found in diabetic cases along with occasional EEV in controls. EEV contained muscle and were almost identical to the EA in structure. Nearly all EEV connected with efferent glom capillaries at the VP, where they exited the glom through apparently preexisting gaps in the Bowman's capsule and/or glomerular capillary basement membranes (BCBM/GCBM). The EA exited through a similar gap, so the exit of EEV was accomplished without altering the anatomical relationships between the exiting vessels and the components of the VP thought to be important in the control of glom outflow. The largest number of EEV occurred in long-standing T2DM cases with mild to moderate DL and NL. Complete photographic glom reconstructions revealed numerous anastomoses among efferent glom capillaries in normal and diabetic gloms with mild to moderate DL and NL. No disproportionately dilated EEV were seen. The findings just cited confirm that EEV are common and surprisingly numerous in diabetic gloms. They suggest that EEV formation served to preserve glom function, and that EEV could neither shunt nor restrict glom outflow locally. In our opinion, the formation of EEV represents a significant, possibly hemodynamically induced, remodeling of the glom that should be added to the list of changes that occur in diabetes. It is hypothesized that EEV develop because of increased glom inflow, and that the latter may be attributable to AA muscle damage that impairs its contractile ability.