The glomerular dynamic correlates of failed filtration were studied in volume replete rats with established glycerol-induced acute renal failure (ARF). Over one-half of all nephrons formed virtually no filtrate, while the single nephron glomerular filtration rate (SNGFR) of fluid-filled nephrons, measured at the glomerulotubular junction to preclude the possibility of covert tubular leakage, averaged one-sixth of control (P < 0.001). Even that low mean value was elevated by a few nephrons with a near normal SNGFR. Renal failure thus reflected both total filtration failure in the majority of nephrons and massively reduced filtration in most of the remainder. Glomerular capillary pressure (Pg) averaged some 14 mmHg below control (P < 0.001), whereas the arterial colloid osmotic and Bowman's space pressures were not significantly altered. Renocortical and whole kidney blood flow were also unchanged. Marked internephron functional heterogeneity precluded estimates of the ultrafiltration coefficient. However, the fall in SNGFR correlated well with the markedly depressed Pg and afferent net filtration pressure (APnetA, P < 0.001), which in turn were caused by increased preglomerular resistance and a reciprocal fall in efferent arteriolar resistance. This complex change in intrarenal resistances was largely, if not entirely, responsible for failed filtration in this ARF model.

No abstract

As manifest by tubular collapse and the virtual absence of flow into the glomerulotubular junction (GTJ), filtration in most nephrons (SNGFR) of rats poisoned with 9 mg/kg body wt HgCl2 16 to 28 hours earlier was virtually absent. Arterial colloid osmotic pressure (COPA) and Bowman's space pressure (PBS) were modestly depressed (P less than 0.05 or below), and mean blood pressure was reduced from 115 +/- 2 mm Hg (SEM) to 97 +/- 1 mm Hg (P less than 0.001). Glomerular capillary hydraulic pressure (Pg), 25.6 +/- 1.3 mm Hg was some 24 mm Hg lower than control (P less than 0.001) and yielded a net afferent effective filtration pressure (Pnet) of 4.1 +/- 1.2 mm Hg. Excluding three rats with values greater than 10 mm Hg, Pnet averaged 2.0 +/- 0.9 mm Hg (N = 17 rats) versus 20.0 +/- 1.8 mm Hg in controls (N = 10, P less than 0.001), the former being statistically almost indistinguishable from 0 mm Hg and barely able to support any filtration. This decrease in Pg was caused by a major increase in preglomerular resistance (RA) and a reciprocal fall in efferent arteriolar resistance (RE), the RA/RE ratio of 7.2 +/- 0.8 being fourfold higher than control (P less than 0.001). Renocortical blood flow was not different from control (P greater than 0.2). A wide spread of Pg values in individual glomeruli and the absence of tubular flow despite the appearance of i.v. injected lissamine green in a quadrant of surface glomeruli suggested the possibility of a greatly increased, glomerular capillary resistance. It is concluded that reciprocal changes in RA and RE are the immediate cause of filtration failure in this form of ARF and that, in the virtual absence of filtration, tubular leakage can play no important role. Since PBS was depressed in both the developmental and established phases of ARF, tubular obstruction appears to play no direct role in the pathogenesis of this particular model of murine acute renal failure.

Quintuplicate determinations of the parameters measured in studies of glomerular dynamics revealed that the intra-animal coefficients of variation for Bowman's space and star vessel pressures, nephron filtration rate, and filtration fractions were 54 to 72% larger than the corresponding interanimal coefficients of variation; those for glomerular capillary pressure were more nearly equal. With a net efferent filtration pressure (delta PE) of 10.6 +/- SEM 1.9 mm Hg, the rats were far from filtration pressure equilibrium and the calculated ultrafiltration coefficient (Kf) of 2.1 +/- SEM 0.2 nl/min X mm Hg was lower than in many other studies. Statistical analysis revealed that the precision of estimates of both the measured and the derived parameters in glomerular dynamic studies is affected appreciably by ignoring the intra-animal effect. The importance of the intra-animal variance in glomerular dynamic studies is greatest when only one or two samples of each measured parameter are obtained in every rat (k = 1 or 2) and least when k is large. Triplicate sampling provides combined SEMs that are not greatly larger than those obtained with k = 5, however, and offers the greatest economy in studies of glomerular dynamics. The number of animals required to provide values with delta PE and Kf that are within +/- 20% of the "true" values is rather large.

The colloid osmotic pressure (COP) of efferent arteriolar plasma in glomerular dynamic studies generally is estimated from the measured protein concentration (CE) while the nephron filtration fraction (SNFF) is derived from CE and the systemic plasma protein concentration (CA) according to the equation SNFF = (1 - CA/CE). Estimates of both SNFF and COPE are quite sensitive to small errors in protein measurement, however, with a putative coefficient of variation of +/- 5% in protein measurement at a typical SNFF of 0.33, for example, providing an uncertainty (i.e., +/- SD) of +/- 14% in the SNFF estimate and +/- 2.4 mmHg in the estimated COPE value. In this study, we evaluated in vitro the precision with which the COP of plasma samples can be estimated after ultrafiltration by coupling direct oncometry of native plasma with isotopically measured filtration fractions derived employing nanoliter and microliter volumes and applying a modification of the equation of Ladegaard-Pedersen (Scand. J. Clin. Lab. Invest. 23: 153-158, 1969). The measured and estimated oncotic pressures were then compared. The mean differences between theoretic and measured COP values at filtration fractions of less than 0.1, 0.1-0.2, 0.2-0.3 and greater were: -0.4 +/- 0.8 (SE) (n = 22); 1.8 +/- 1.1; 3.9 +/- 1.0; and 6.0 +/- 1.7%, respectively. It is concluded that the coupling of direct oncometric measurement of arterial plasma colloid osmotic pressure with isotopically determined filtration fractions provides a satisfactory estimate of COPE that is suitable for studies of glomerular dynamics.