Evaluation of Capillary Perfusion

The following is the abstract of the article discussed in the subsequent letter:

	ABSTRACT

Bentzer Peter, Lis Kongstad, and Per-Olof Grände. Capillary filtration coefficient is independent of number of perfused capillaries in cat skeletal muscle. Am J Physiol Heart Circ Physiol 280: H2697-H2706, 2001.The capillary filtration coefficient (CFC) is assumed to reflect both microvascular hydraulic conductivity and the number of perfused capillaries at a given moment (precapillary sphincter activity). Estimation of hydraulic conductivity in vivo with the CFC method has therefore been performed under conditions of unchanged vascular tone and metabolic influence. There are studies, however, that did not show any change in CFC after changes in vascular tone and metabolic influence, and these studies indicate that CFC may not be influenced by alteration in the number of perfused capillaries. The present study reexamined to what extent CFC in a pressure-controlled preparation depends on the vascular tone and number of perfused capillaries by analyzing how CFC is influenced by 1) vasoconstriction, 2) increase in metabolic influence by decrease in arterial blood pressure, and 3) occlusion of precapillary microvessels by arterial infusion of microspheres. CFC was calculated from the filtration rate induced by a fixed decrease in tissue pressure. Vascular tone was increased in two steps by norepinephrine (n = 7) or angiotensin II (n = 6), causing a blood flow reduction from 7.2 ± 0.8 to at most 2.7 ± 0.2 ml · min¹ · 100 g¹ (P < 0.05). The decrease in arterial pressure reduced blood flow from 4.8 ± 0.4 to 1.40 ± 0.1 ml · min¹ · 100 g¹ (n = 6). Vascular resistance increased to 990 ± 260% of control after the infusion of microspheres (n = 6). CFC was not significantly altered from control after any of the experimental interventions. We conclude that CFC under these conditions is independent of the vascular tone and number of perfused capillaries and that variation in CFC reflects variation in microvascular hydraulic conductivity.

	LETTER

To the Editor: The paper by Bentzler et al. (Am J Physiol Heart Circ Physiol 280: H2697-H2706, 2001) is an interesting and potentially important paper, but its title is misleading. No direct measurements of open capillary number (N) or surface area (S) were made by the authors. They measured hydrodynamic resistances (R) and inferred that the number of perfused capillaries was inversely proportional. They altered R by 1) norepinephrine, 2) angiotensin, 3) lowering perfusion pressure, and 4) by intra-arterial infusion of 15-mm microspheres. R increased two- to threefold with norepinephrine and angiotensin, changed little with lowering perfusion pressure, and increased 10 times with by intra-arterial infusion of 15-mm microspheres. In all cases, they found no changes in the capillary filtration coefficient (CFC).

The discussion of their results shows that they recognized the problem of defining "open capillary" or "perfused capillary surface area." To explain the lack of change in CFC after blocking arterioles with microspheres, they suggested that retrograde refilling of capillaries from the remaining perfused venules could maintain functional capillary surface. This is possible, but it seems more likely to me that the entire capillary bed could be perfused in the normal direction from the few unblocked arterioles via connectives on the arterial side of the capillary network. CFC would be unaffected if capillary blood flow were sufficient to prevent a substantial increase in plasma colloid osmotic pressure. The same argument applies to the other experimental procedures described.

It seems significant, in this connection, that other observers have reported variable results of norepinephrine, angiotensin, lowering perfusion pressure, or comparable procedures on CFC. (To my knowledge, results of intra-arterial infusion of 15-mm microspheres have not been reported before.) Maintaining constant CFC appears to depend on the balance of factors that redistribute capillary flow. Thus the summary contention of the authors, namely that measurements of CFC might be used to evaluate capillary hydraulic conductivity (CFC/S) without independent measurement of S (or N) is not valid unless balanced redistribution of capillary flow is achieved.

	REPLY

To the Editor: We truly appreciate that Dr. Renkin has paid attention to our study (1), and we find his comments of great relevance. Two major concerns were raised about the study. First, that the title is misleading by using the term "number of perfused capillaries," and second, that we infer that the CFC method may be used to study fluid permeability changes without an independent measure of surface area.

"Number of perfused capillaries" as defined in our study (1) represents the degree of precapillary sphincter activity and microocclusion and thus the functional surface area available for diffusion exchange. In previous studies, changes in functional surface area available for diffusion exchange have been estimated by measurement of tissue uptake of a tracer substance at constant flow for calculation of the permeability surface area product (2, 5, 6). However, to study CFC during conditions as close to normal physiology as possible, we used an autoperfused and pressure-controlled preparation, which means that blood flow varies during the experiment. To make tissue uptake independent of flow in such a preparation, the permeability of the tracer has to be low, which makes tissue uptake measurements prone to error. Although the number of perfused capillaries as defined above was not measured in our study (1), many studies have shown that variation in vascular resistance or microocclusion influence this parameter, as estimated in preparations perfused at constant flow (2, 5-7). It is therefore highly unlikely that the number of perfused capillaries remains unchanged following the rather extreme experimental interventions performed in our study, and therefore we think the title is acceptable.

CFC is supposed to represent the product of hydraulic conductivity (L_p) and the functional surface area available for fluid exchange, and it is often assumed that this surface area equals the surface area available for filtration during the CFC procedure, here denoted as S (3, 4). Dr. Renkin argues that we cannot conclude for sure that S is unaltered during variation in vascular tone and number of occluded capillaries because S was not measured independently. The difficulties to evaluate S in a whole organ preparation and to describe this parameter in adequate terms have plagued most studies in this field of research including ours. Naturally, an independent measure of S would have been of value, but how should such a measurement be performed? As mentioned above, measurement of changes in the permeability surface area product has been used in previous studies as a measure of changes in the functional surface area available for diffusion (2, 5-7). However, functional surface area for diffusion (number of perfused capillaries) is not only difficult to measure during variation in blood flow, but it is also unlikely that this surface area equals the surface area contributing to fluid filtration during the CFC procedure (S) because diffusion is more dependent on flow rate than on filtration (3, 7). To our knowledge, there is no method available to independently evaluate changes in S.

Our conclusion that S (surface area contributing to CFC) is constant is based on the observation that CFC did not change when vascular resistance or number of occluded microvessels were altered. If there had been variation in S under these conditions, such a change must be counteracted by a corresponding change in capillary flow distribution acting in the opposite direction to maintain a constant CFC in all four situations analyzed. We think it is highly unlikely that such a balanced change in both surface area and capillary flow distribution occurs following the different four experimental interventions. Apparently, the capillary blood flow is sufficient to prevent substantial increase in plasma colloid osmotic pressure during the CFC-induced filtration also during relatively extreme conditions of arteriolar occlusion. Therefore, even though we did not measure S, the only reasonable explanation to our results is that S is independent of vascular tone and precapillary sphincter activity and independent of the number of perfused capillaries. If so, our conclusion that the CFC method may be used to evaluate variation in hydraulic conductivity without an independent measurement of S is valid.

Dr. Renkin raises an important issue when suggesting that the entire capillary bed may be perfused from the open arterioles rather than from the venous side. It follows that our results may be dependent on a rich microvascular network with many intercapillary connections to "redistribute capillary flow."

	REFERENCES

1.   Bentzer, P, Kongstad L, and Grände PO. Capillary filtration coefficient is independent of variation in vascular tone and number of perfused capillaries in cat skeletal muscle. Am J Physiol Heart Circ Physiol 280: H2697-H2706, 2001[Abstract/Free Full Text].

2.   Haraldsson, B. Effects of noradrenaline on the transcapillary passage of albumin, fluid and CrEDTA in the perfused rat hindlimb. Acta Physiol Scand 125: 561-571, 1985[Web of Science][Medline].

3.   Mellander, S, and Johansson B. Control of resistance, exchange, and capacitance functions in the peripheral circulation. Pharmacol Rev 20: 117-196, 1968[Free Full Text].

4.   Renkin, EM. Control of microcirculation and blood-tissue exchange. Handbook of Physiology. The Cardiovascular System. Microcirculation. Bethesda, MD: Am. Physiol. Soc., 1984, sect. 2, vol. IV, pt. 2, chapt. 14, p. 627-687.

5.   Renkin, EM, Hudlická O, and Sheehan RM. Influence of metabolic vasodilation on blood tissue diffusion in skeletal muscle. Am J Physiol 211: 87-98, 1966.

6.   Rippe, B, and Folkow B. Simultaneous measurements of capillary filtration and diffusion capacities during graded infusions of noradrenaline (NA) and 5-hydroxytryptamine (5-HT) into the rat hindquarter vascular bed. Acta Physiol Scand 109: 265-273, 1980[Web of Science][Medline].

7.   Rippe, B, Kamiya A, and Folkow B. Simultaneous measurements of capillary diffusion and filtration exchange during shifts in filtration-absorption at graded alterations in the capillary permeability surface area product. Acta Physiol Scand 104: 318-336, 1978[Web of Science][Medline].

Peter Bentzer, Per-Olof Grände, Department of Physiological Sciences University of Lund, SE 22184 Lund, Sweden E-mail: per-olof.grande{at}mphy.lu.se