Isotope labelled microspheres were used to study the capillary blood perfusion of the rabbit tracheal mucosa. Under resting conditions the perfusion was about 0.3 ml/min - g (i.e. about 60% of the relative cerebral blood flow). Irritation of the tracheal mucosa by an endotracheal tube caused a steep rise in blood flow, tenfold or more. This was probably due to relaxation of the arterioles caused by a release of histamine-like substances. When an endotracheal tube is equipped with a small cuff (small resting diameter, low residual volume), the part of the mucosa in contact with the cuff, i.e. the mucosa covering the surface and edges of the cartilages, will be ischaemic at a cuff to tracheal wall pressure (C-T pressure) of greater than 30 millimeters of mercury. This abrupt ischaemia threshold contributes to the risk of deep mucosal damage with subsequent tracheal scarring, possibly proceeding to stenosis. Our present studies indicate that the ideal large cuff, with properties resembling those of an air cushion, will allow the major part of the arterial pressure to be propagated as far down as the capillaries. Under these conditions the cuff would permit some of the capillary blood perfusion of the tracheal mucosa covering the cartilages also at C-T pressures exceeding 30 mmHg. Although this investigation supports the concept that the ideal thin-walled large cuff interferes much less with the mucosal blood perfusion than the small cuff, we recommend that the cuff pressure be monitored and kept below 20 mmHg.

The distribution of blood flow in the rat kidney after 60 minutes of renal ischemia was studied by single-fiber laser-Doppler flowmetry. Blood flow in superficial cortex and inner medulla was measured with a probe directed towards the kidney surface and exposed papilla, respectively. Outer medullary blood flow was measured with a probe introduced through the renal core. After ischemia the blood flow decreased to 60% of the preischemic value (P less than 0.01) in superficial cortex and to 16% (P less than 0.01) in outer medulla, while inner medullary blood flow increased paradoxically to 125% (P less than 0.01). There was extensive trapping of red blood cells (RBC) in the outer medulla, but not in the inner medulla or cortex. The fractional RBC volume as measured by radiolabeled RBCs was 21% in the inner stripe of the outer medulla, but 2% in this area in a normal kidney. To investigate the influence of RBC trapping on intrarenal distribution of blood flow after ischemia, the hematocrit was reduced from 46% to 31% by isovolemic hemodilution. When performed before ischemia, this maneuver almost completely abolished RBC trapping. In this group blood flow in both outer and inner medulla was almost unchanged after ischemia, while superficial cortical blood flow decreased to 66% (P less than 0.01) of the pre-ischemic value. It is concluded that RBC trapping in the outer medulla causes a large decrease in blood flow in this area and, at the same time, shunting of blood to the inner medulla. In the absence of RBC trapping, blood flow of both outer and inner medulla is well preserved after ischemia.

Acute renal failure was induced in heparinized rats by clamping the renal artery for 45 min. Ten minutes after recirculation the intrarenal blood flow distribution was measured. For this purpose labelled microspheres were injected together with 86-Rb chloride. The microspheres were used for determination of cardiac output, total renal and cortical blood flow, and 86-Rb for calculations of medullary blood flow. Total renal blood flow was reduced from 7.6 to 3.8 ml . min-1 . g-1 and cortical blood flow was reduced from 11.7 to 7.0 ml . min-1 . g-1. In the outer stripe of the medulla there was a reduction from 2.5 to 1.4 ml . min-1 . g-1. In the inner stripe there was a more pronounced reduction from 1.8 to 0.2 ml . min-1 . g-1 and in the inner zone from 0.8 to 0.1 ml . min-1 . g-1. The marked reduction in the blood flow to the renal medulla after recirculation is suggestive for a medullary ischemia, which might be responsible for the characteristic dysfunctions in acute renal failure.

The influence of neutrophils on peritubular capillary permeability and intravascular red blood cell (RBC) aggregation after renal ischemia was studied in anesthetized Sprague-Dawley rats. Intraperitoneal administration of antineutrophil serum (ANS) reduced the number of neutrophils in the blood to 3% of normal. The control group received an equal volume of inactive serum. Renal macromolecular capillary permeability was studied from 1) extravasation of albumin and 2) plasma to lymph transport of plasma proteins and of neutral and negatively charged lactate dehydrogenase (LDH). The net driving force (NDF) for fluid transfer over the peritubular capillary membrane was determined by the micropuncture technique. The intrarenal distributions of neutrophils and RBC were measured by a histochemical method and 51Cr-labeled RBC, respectively. Under preischemic control conditions neither macromolecular permeability nor renal clearance of inulin was affected by ANS. However, the steep increase in the macromolecular transport from plasma to lymph resulting from 45 min of ischemia and reperfusion was blunted by ANS, and preischemic control values were restored after 1 h of recirculation. In the control group the mass transport of plasma proteins increased twofold and that of both neutral and negatively charged LDH fourfold. NDF was equal in the two groups. In the ANS-treated animals the intrarenal neutrophil content was only 2% of the control. Neutrophils were found mainly in the cortex, whereas RBC aggregation was observed only in the renal medulla. It is concluded that neutrophils mediate postischemic capillary leakage. It is suggested that this leakage underlies RBC aggregation and incomplete return of blood flow in the renal medulla after ischemia.

A B S T R A C T Distribution of intrarenal blood flow was studied in 12 dogs anesthetized with Nembutal. Medullary blood flow was estimated by local clearance of hydrogen gas from the outer medulla measured polarographically with needleshaped platinum electrodes, and by local clearance of 85Kr and mean transit time of 32P-labeled erythrocytes measured with a small semiconductor detector placed in the outer medulla. Cortical blood flow was estimated from cortical red cell transit time and from total renal blood flow measured by electromagnetic flowmeter.Bleeding to a mean arterial pressure of 50-65 mm Hg in the course of 8-20 min reduced cortical and medullary blood flow on the average to the same extent. In half of the experiments both cortical and medullary blood flow were reduced proportionately less than mean arterial pressure during the first half hour of bleeding. Maintenance of mean arterial pressure at 50-65 mm Hg in all cases led to progressive reduction of both cortical and medullary blood flow, out of proportion to the reduction of arterial pressure. A two step bleeding procedure used in two experiments also led to uniform reduction of renal blood flow. Reinfusion of blood after 2-3 hr of hypotension increased total renal blood flow to an average of 82% and outer medullary hydrogen clearance to an average of 92%o of control values. All dogs survived the experiment without evidence of renal failure.

The influence of the hematocrit (Hct) on the trapping of red blood cells (RBC) in the renal microvasculature and its effect on the long-term outcome following unilateral ischemia were investigated in the rat. The results showed that an increase in the duration of ischemia increased the RBC trapping, as measured by 51Cr-labeled erythrocytes, in a dose-dependent manner. At normal Hct (46%) the period of ischemia producing half-maximum RBC trapping was 45 minutes, whereas after hemodilution (Hct = 31%) or hemoconcentration (Hct = 60%) the corresponding periods were 80 and 25 minutes, respectively. Regarding the long-term outcome, 45 minutes of ischemia with a normal Hct was associated with a marked decrease in kidney weight, GFR and urine osmolarity after four weeks of recovery, which could be prevented to a large extent by hemodilution. Conversely, with hemoconcentration there was severe damage after only 25 minutes of ischemia. It is suggested that these long-term effects are attributable to RBC trapping in the microvasculature of the outer medulla, which may cause added ischemia in this area of the kidney. It is also suggested that cortical atrophy is secondary to the medullary injury, and is brought about to avoid extensive water and salt losses.

The pressure conditions at the distal end of the interlobular arteries and in the interlobular veins were investigated from the pressures obtained in superficial small arteries and veins, accidentally found on the kidney surface, during the subsequent blockade of the blood stream in the down-stream and up-stream direction, respectively. The results suggested a hydrostatic pressure in the distal end of the interlobular arteries of about 85 mm Hg under normotensive conditions-a pressure which remained fairly constant when the perfusion pressure in the renal artery was decreased within the autoregulation range. The results indicate a considerable pressure drop of about 40 mm Hg along the interlobular arteries. During hypotension this pressure drop decreased, implying a decreased resistance in the interlobular arteries, i.e. a typical autoregulative response. The pressure in the interlobular veins amounted to about 5 mm Hg, which is a few mm Hg higher than that in the renal vein and about 7 mm lower than that in the peritubular capillary network. The results suggest a flow resistance located somewhere between the peritubular capillaries and the intrarenal veins. This resistance is not influenced by vasoactive substances but it is decreased when the systemic venous pressure is raised above 10 mm Hg. The resistance seems to act in the direction of protecting the peritubular capillaries from minor changes in the central venous pressure.

The validity of creatinine as a marker for the glomerular filtration rate was studied in 8 healthy volunteers in different stages of hydration and during large variations in urinary flow rates. The urine flow was 8.4 ml/min in a dehydrated state (due to furosemide infusion; 8 mg/h) and raised to 23.2 ml/min after rapid rehydration. The creatinine to inulin clearance ratio changed considerably from 1.47 in rehydrated state, indicating a substantial tubular secretion of creatinine, to 1.05 in dehydrated state, indicating a reabsorption of creatinine almost equal to secretion. Thus, substantial tubular secretion and reabsorption of creatinine, changing in relative importance in relation to the degree of hydration, make creatinine clearance an unreliable marker for the glomerular filtration rate.