Relationships Between Microvascular Function and Capillary Structure in Diabetic and Nondiabetic Human Skin

  1. John A Gaines
  1. Research Service and Benjamin W. Zweifach Micro-circulation Laboratories, and Renal Section, Veterans Administration Medical Center, and Renal Section, Department of Anatomy, and Division of Computer Systems and Biostatistics, University of Arizona College of Medicine Tucson Division of Neuropathology, Barrow Neurological Institute Phoenix, Arizona
  1. Address correspondence and reprint requests to Dr. Murray A. Katz, Research Service (151), Veterans Administration Medical Center, Tucson, AZ 85723.

Abstract

Despite the commonly held view that abnormalities in capillary morphology, in particular thickening of the capillary basement membrane, are partly responsible for diabetic ischemia, few studies have correlated anatomic and hemodynamic variables in the same diabetic subjects. In a previous study of 24 type II (non-insulin-dependent) diabetic subjects and 24 agematched control subjects, we showed that a standard finger exercise vasodilated cutaneous forearm vessels nearly equally (51%), but the postarteriolar flow responded differently between groups. Nondiabetic subjects increased flow by recruitment of capillaries, whereas diabetic subjects did so by capillary flow augmentation. Moreover, resting permeability-surface area product (PS) to pentetic acid was 85% higher in diabetic than nondiabetic subjects. In this study, these same subjects had their forearm skin biopsied and examined morphometrically by electron microscopy for capillary radius, basement membrane thickness, endothelial cell density, and a folding index of luminal membrane reduplication. All morphological variables were correlated stepwise in a saturated, analysis of covariance model with the physiological results. The correlations were sparse and specifically excluded basement membrane thickness. The highest r2 value was .432 between resting PS and a ratio of capillary density to endothelial cell number per capillary. These studies show little evidence that diabetic microvascular physiological variables are tightly connected to morphometric changes except for minor permeability changes, which rise with capillary density and decrease with endothelial cell number. Because PS to pentetic acid is increased in diabetic subjects at any level of capillary density, it seems reasonable that permeability may be increased above that of nondiabetic subjects. However, such a conclusion is tentative because anatomic measurement of capillary density gives only the maximum estimate of capillaries in the tissue, which may be greater than the number of capillaries with flow at the time of physiological measurement.

  • Received March 6, 1989.
  • Revision received June 14, 1989.
  • Accepted June 14, 1989.
| Table of Contents