The effects of a seminar method to improve the teaching of ward attending physicians were evaluated. Forty-six attending physicians from four institutions were randomly assigned to experimental and control groups. The method was evaluated to assess its effects on attending physicians' performances and attitudes, and impact on learners. Evaluation methods included ratings of videotapes of ward rounds, teachers' subjective assessments of both their teaching performances and their experiences in the study, and trainee ratings. Videotape ratings, the teachers' own assessments, and the trainees' assessments of the attending physicians' impact on learning were significantly different, favoring the experimental group (p less than 0.05). It is concluded that the seminar method can provide the basis for effective and feasible approaches for improving clinical teaching by attending physicians.

We evaluated the size-selective properties of the glomerular barrier in 30 patients in whom diabetic nephropathy was associated with urinary IgG losses. Neutral dextrans of graded size were used to characterize glomerular membrane-pore structure. A fractional IgG clearance (relative to freely permeable inulin) smaller or greater than 0.001 was used to distinguish patients with minor (group 1, N = 14) and major (group 2, N = 16) urinary IgG leakage, respectively. Fractional clearances of dextrans (theta D) of smaller size (radii 20-40 A) were similar, but those of larger dextrans (radii 42-60 A) were elevated in group 2 relative to group 1 patients. When plotted on log-normal probability coordinates, the correlation between theta D and radius in healthy subjects is linear, suggesting that glomerular pores form one population with a normal distribution. In diabetic nephropathy with urinary IgG leakage, however, theta D for large molecules was elevated and departed from linearity, suggesting a bimodal pore size distribution within the glomerular membrane. A pore model of solute transport revealed (1) the upper pore mode was highly permeable to large dextrans equivalent in size to IgG and (2) the fraction of glomerular filtrate permeating the large pores was greater in group 2 than in group 1 patients with diabetic nephropathy, 6% versus 3%, respectively. We conclude that urinary IgG leakage in diabetic nephropathy is determined by the development of a subpopulation of enlarged pores. The magnitude of urinary IgG losses appears to be a function of the membrane area-fraction occupied by the enlarged pores.

We describe in physiological terms the increasing glomerular capillary wall (GCW) dysfunction of 20 patients with diabetic glomerulopathy and heavy proteinuria. The clearances of uncharged polysaccharide markers of graded size were used to probe the glomerular filter on three occasions over a 24-mo period. The findings were analyzed with a theoretical model of solute transport that depicts most of the GCW as an isoporous membrane and the minor portion as a nondiscriminatory shunt pathway. Initially, the mean glomerular ultrafiltration coefficient Kf is computed to have been 3-5 times lower and mean pore radius of the major membrane component (r0) 2 A smaller than normal control values. In contrast, the model computes the fraction of filtrate volume permeating the nondiscriminatory shunt pathway (omega 2) to have been sixfold elevated above control values and to have correlated strongly in individual patients with the fractional clearances of albumin (r = .72) and of IgG (r = .73). Sequential studies after 12 and 24 mo revealed an invariable decline in glomerular filtration rate (GFR). Fractional clearances of albumin and IgG increased with time in most patients but declined in a few instances (20-25%). Change in omega 2 tended to occur in parallel with fractional protein clearance, regardless of its direction. We conclude that in progressive diabetic glomerulopathy GFR declines because of a loss by glomerular capillaries of ultrafiltration capacity, proteinuria is largely a consequence of increasingly impaired barrier-size selectivity, and the foregoing injuries reflect damage to different parts of the GCW and may become dissociated from one another with the passage of time.

No abstract