Hepatic resection for metastatic colorectal cancer results in cure for some patients.

OBJECTIVES: To determine the long-term disease-free and overall survivals for patients undergoing hepatic resection for colorectal cancer metastases and to define significant predictors of improved patients survival. DESIGN: Retrospective review. SETTING: Single tertiary care center. PATIENTS: Two hundred eighty consecutive patients underwent hepatic resection for colorectal cancer metastases at the Mayo Clinic from 1960 to 1987. Fifty patients alive at the completion of the study had a mean follow-up of 11.3 years (median, 121 months). MAIN OUTCOME MEASURES: Disease-free interval following initial hepatic resection and death. RESULTS: The overall 5-year survival of the 280 patients was 27%. Twenty-eight patients were alive at 10 years from the time of hepatic resection, and the 10-year actuarial survival was 20%. Only 2 patients alive and free of disease at 5 years had recurrent disease. For all other patients who were free of disease more than 5 years after hepatic resection and died, the cause of death was not cancer related. No patients characteristics or features of the primary tumor affected survival. Clinical presentation of metastatic disease, configuration of hepatic lesions, the presence of extrahepatic lymph node involvement, and the existence of resectable extrahepatic disease significantly affected long-term patient survival. Need for perioperative blood product transfusion was associated with a lower probability of long-term survival. CONCLUSIONS: Disease-free patient survival beyond 5 years from surgical resection of colorectal cancer metastases to the liver represents patient cure in nearly all instances.

Long-term consequences of intraoperative spillage of bile and gallstones during laparoscopic cholecystectomy.

Laparoscopic cholecystectomy is associated with a higher incidence of iatrogenic perforation of the gallbladder than open cholecystectomy. The long-term consequences of spilled bile and gallstones are unknown. Data were collected prospectively from 1059 consecutive patients undergoing laparoscopic cholecystectomy over a 3-year period. Details of the operative procedures and postoperative course of patients in whom gallbladder perforation occurred were reviewed. Long-term follow-up (range 24 to 59 months) was available for 92% of patients. Intraoperative perforation of the gallbladder occurred in 306 patients (29%); it was more common in men and was associated with increasing age, body weight, and the presence of omental adhesions (each P < 0.001). There was no increased risk in patients with acute cholecystitis (P = 0.13). Postoperatively pyrexia was more common in patients with spillage of gallbladder contents (18% vs. 9%; P < 0.001). Of the patients with long-term follow-up, intra- abdominal abscess developed in 1 (0.6%) of 177 with spillage of only bile, and in 3 (2.9%) of 103 patients with spillage of both bile and gallstones, whereas no intra- abdominal abscesses occurred in the 697 patients in whom the gallbladder was removed intact ( P < 0.001). Intraperitoneal spillage of gallbladder contents during laparoscopic cholecystectomy is associated with an increased risk of intra-abdominal abscess. Attempts should be made to irrigate the operative field to evacuate spilled bile and to retrieve all gallstones spilled during the operative procedure.

Transport of sodium and urea in outer medullary descending vasa recta.

We dissected and perfused outer medullary vasa recta (OMVR) from vascular bundles in the rat. Permeabilities of sodium (PNa) and urea (Pu) were simultaneously determined from the lumen-to-bath efflux of 22Na and [14C]urea. PNa and Pu were also measured by in vivo microperfusion of descending (DVR) and ascending vasa recta (AVR) at the papillary tip of Munich-Wistar rats. In some OMVR PNa was indistinguishable from zero. The mean +/- SE of PNa (x 10(-5), cm/s) in OMVR was 76 +/- 9. Pu in OMVR was always very high (x 10(-5), cm/s), 360 +/- 14. There was no correlation between OMVR PNa and Pu. Inner medullary AVR and DVR had PNa of 115 +/- 10 and 75 +/- 10, respectively, and Pu of 121 +/- 10 and 76 +/- 11, respectively. PNa and Pu in papillary vasa recta were always nearly identical and highly correlated. Transport of [14C] urea in OMVR was reversibly inhibited by addition of unlabeled urea or phloretin to the bath and lumen, providing evidence for carrier-mediated transport. These data suggest that sodium and urea might traverse the wall of inner medullary vasa recta by a paracellular pathway while urea also crosses by a transcellular route in OMVR. Electron microscopic examination of seven in vitro perfused OMVR revealed no fenestrations and exposure of these vessels to 10 microM calcium ionophore A23187 or 1 nM angiotensin II resulted in reversible contraction, suggesting that in vitro perfused OMVR are DVR only.

The natriuretic peptides and their receptors.

Atrial natriuretic factor (ANF) is released from the cardiac atrium in response to stretch and acts through receptors to cause an increase in urinary flow and sodium excretion, vasodilatation, and a reduction in blood volume. Recently, two new natriuretic peptides, brain natriuretic peptide (BNP) and C-type natriuretic peptide (C-typeNP), have been isolated, and three different natriuretic peptide receptors have been identified. Two of the receptors, ANP-RGC(A) and ANP-RGC(B), mediate biologic actions. The natural ligand of ANP-RGC(A) is ANF, whereas that of ANP-RGC(B) is C-typeNP. In view of clear differences in ligand specificity and tissue distribution of these receptors, it has been proposed that ANF and its receptor, ANP-RGC(A), and C-typeNP and its receptor, ANP-RGC(B), represent two distinct natriuretic peptide regulatory systems. Whether a separate system exists that incorporates BNP awaits clarification of its natural receptor that mediates a biologic action. The third receptor, ANP-Rc, binds all three natriuretic peptides. Its messenger RNA lacks the guanylyl cyclase sequence present in the mRNA of the other natriuretic peptide receptors, suggesting that the principal function of ANP-Rc is to remove natriuretic peptides from the circulation, that is, to regulate plasma levels of the natriuretic peptides. However, ANP-Rc may also mediate a biologic effect. These findings raise several intriguing questions about the functional role of this family of natriuretic peptides.

Regulation of platelet clearance receptors for atrial natriuretic peptide in diabetic nephropathy.

The findings that circulating levels of atrial natriuretic peptide (ANP) are elevated in diabetic nephropathy and that the magnitude of the urinary excretion rate of cGMP in response to hypervolemia-induced ANP release is blunted have recently been reported. The purpose of this study was to determine whether these abnormalities are associated with the down-regulation of ANP receptors. Because biologically active (A) ANP receptors in the kidney are inaccessible, we have examined the binding of (125I alpha)ANP to clearance (C) receptors on platelets obtained from patients with diabetic nephropathy. Scatchard analysis revealed a reduction in such binding sites compared with those in healthy controls: 12 +/- 2 versus 19 +/- 2 per platelet, respectively (P less than 0.001). The dissociation constant, Kd, was higher: 66.7 +/- 33.1 versus 38.5 +/- 11 pM, respectively (P less than 0.02). The reduced number of receptors could reflect the down-regulation of ANP C receptors in response to an elevation of plasma levels of ANP, the median value of which was 10.6 versus 7.1 pmol/L in controls (P less than 0.05). Alternatively, the findings could represent a primary adaptation by C receptors to elevate plasma ANP levels and increase the availability of the peptide to biologically active renal receptors. The latter adaptation would serve to mitigate the sodium retention that attends diabetic nephropathy.

Identification of "B" receptor for natriuretic peptide in human kidney.

Three distinct receptor types for natriuretic peptides (NP) have been identified in human tissue. "A" and "B" receptors initiate biological actions, whereas the "C" receptor has a clearance function. It has been proposed that the natural ligand for the B receptor is c-type natriuretic peptide (CNP), rather than atrial natriuretic peptide (ANP), and that the B receptor is only found in the central nervous system (CNS) and is responsible for all NP-mediated effects in the CNS. Contrary to this hypothesis, we have identified, by means of the polymerase chain reaction (PCR), the B receptor in human kidney tissue. To detect A and C receptors, the PCR reaction was performed with primers which yielded predicted 600 and 378 base pair (bp) products, respectively. For the B receptor, 3 different primer sets were used, resulting in the expected 785, 453 and 228 bp fragments. Restriction mapping of the latter two products with Rsa I yielded the expected fragment numbers and sizes, indicating the PCR products were from B receptor mRNA. These results indicate that the human kidney has B as well as A and C receptors. Thus CNP may have a renal as well as a CNS site of action.

Resistance of descending vasa recta to the transport of water.

The effect of varying intracapillary oncotic pressure on the rate of transcapillary volume flux in microperfused descending vasa recta (DVR) was studied during furosemide diuresis in the Munich-Wistar rat. At the papillary base, plasma protein concentration and hydraulic pressure were 5.7 +/- 0.1 g/dl and 11.7 +/- 0.7 mmHg in nonperfused DVR, respectively, and 5.6 +/- 0.1 g/dl and 9.4 +/- 0.4 mmHg in nonperfused ascending vasa recta (AVR), respectively. These results demonstrate that the papillary microcirculation does not remove water from the interstitium during furosemide diuresis and defines Starling forces in the pericapillary interstitium. Osmolality and urea concentration were 380 +/- 11 mosmol/kgH2O and 56 +/- 5 mM in DVR plasma at the papillary base, respectively, and 386 +/- 10 mosmol/kgH2O and 62 +/- 5 mM in DVR plasma at the tip, respectively. These results demonstrate abolition of corticomedullary small solute gradients. DVR were perfused at a rate of 10 nl/min with a buffer solution containing small-solute concentrations that matched those of plasma in nonperfused DVR. The buffer solution also contained 2 x 10(6) mol wt fluorescein isothiocyanate-labeled dextran (FITC-Dx, 5 mg/ml) and either 0.1 or 5.0 g/dl albumin. Microperfused DVR were punctured a second time downstream of the perfusion site for sample collection or servo-nulling pressure measurement. The rate of transmembrane volume flux, determined from the change in FITC-Dx concentration from perfusate to collectate, was 0.99 +/- 0.29 nl.min-1.mm-1 when perfusate contained 0.1 g/dl albumin and 0.00 +/- 0.23 nl.min-1.mm-1 with 5.0 g/dl albumin (P less than 0.01). Intracapillary hydraulic pressures were 21.7 and 20.4 mmHg during microperfusion of DVR with 0.1 and 5.0 g/dl albumin, respectively. These results demonstrate that transcapillary driving forces of 20 mmHg (5 g/dl albumin) influence transcapillary water movement across the DVR endothelium. For an average capillary diameter of 12.9 microns, DVR hydraulic conductivity is calculated to be greater than 1.4 x 10(-6) cm.s-1.mmHg-1.

Effect of small-solute gradients on transcapillary fluid movement in renal inner medulla.

Volume efflux from descending vasa recta (DVR) of hydropenic rats occurs despite a higher oncotic pressure than hydraulic pressure. To explain this, we previously proposed that transcapillary small-solute gradients exert an additional driving force for volume efflux. This hypothesis was tested by micropuncture of DVR at the base and tip of the exposed renal papilla of control hydropenic and furosemide-treated rats. The DVR plasma osmolality at the base, 573 +/- 40, rose to 1,011 +/- 118 mosmol/kg H2O at the tip in control animals but was 356 +/- 8 and 377 +/- 6, respectively in furosemide animals, demonstrating that the axial osmotic gradient was nearly eliminated. The DVR plasma protein concentration in control hydropenic rats was higher at the tip, 6.8 +/- 0.4 g/dl, than at the base, 5.2 g/dl (P less than 0.01), because of volume efflux from DVR between base and tip, but was 5.4 +/- 0.3 and 5.5 +/- 0.2 g/dl at the base and tip, respectively, in furosemide animals, indicating virtual abolition of volume efflux. In DVR at the tip in control animals, plasma protein concentration positively correlated with plasma osmolality (R = 0.72, P less than 0.02). These results support the hypothesis that transcapillary small-solute gradients influence transcapillary fluid movement in DVR.

Inner medullary blood flow in postischemic acute renal failure in the rat.

To study the effect of renal ischemia on the circulation in the inner medulla, blood flow in descending and ascending vasa recta was determined by fluorescence videomicroscopy in the exposed papilla of the uninephrectomized rat after clamping of the renal artery for 45 min. Total renal blood flow was determined in parallel studies with an electromagnetic flowmeter. Animals were studied 90 min (group 1E) and 24 h (group 2E) after right nephrectomy and release of the left renal artery clamp. Control rats were studied 90 min (group 1C) and 24 h (group 2C) after right nephrectomy alone. In groups 1E and 2E, total renal blood flow was reduced to 70 and 80% of that in their respective controls; renal vascular resistance increased by 50 and 73%, respectively. In striking contrast, blood flow was markedly elevated in descending and ascending vasa recta in groups 1E and 2E compared with the values in their respective uninephrectomized controls. These results indicate that the circulation in the inner medulla is rapidly restored after 45 min of total renal ischemia and that vasa recta blood flow rises above normal after 90 min and 24 h, despite a reduction in total renal blood flow and an increase in renal vascular resistance.

Effect of ureteral excision on inner medullary solute concentration in rats.

To elucidate the role of the ureter in urinary concentration we studied the effect of partial and complete ureteral excision on urinary osmolality and papillary interstitial osmolality and on sodium, potassium, and urea concentrations in the antidiuretic rat. Urine and descending vasa recta (DVR) plasma samples were obtained by micropuncture of the left renal papilla before (period 1) and 45 min after (period 2) complete (group 1, n = 10 rats) or partial (group 2, n = 10 rats) ureteral excision. Urine osmolality fell from 2,063 +/- 156 (mean +/- SE) to 736 +/- 116 mosmol/kgH2O after complete ureteral excision (P less than 0.01). After partial ureteral excision, the fall was less than half as great, from 2,038 +/- 167 to 1,551 +/- 162 mosmol/kgH2O (P less than 0.01). Vasa recta plasma osmolality decreased from 1,742 +/- 133 to 860 +/- 119 mosmol/kgH2O after complete excision (P less than 0.01) but only from 1,830 +/- 146 to 1,504 +/- 154 mosmol/kgH2O after partial excision (P less than 0.05). Mean DVR plasma sodium concentration declined from 339 +/- 25 to 211 +/- 25 meq/l (P less than 0.01) in group 1 but did not change in group 2 (348 +/- 21 to 347 +/- 28 meq/l). The fraction of DVR plasma osmolality accounted for by urea decreased significantly from 0.59 +/- 0.01 to 0.46 +/- 0.02 mM/(mosmol/kgH2O) in group 1 and from 0.59 +/- 0.02 to 0.49 +/- 0.03 mM/(mosmol/kgH2O) for group 2 (P less than 0.01, both groups). We interpret these findings to show that the remnant ureter moderates the fall in interstitial osmolality at least in part through preservation of the corticomedullary sodium chloride gradient.

Effects of intrarenal adenosine on renal function and medullary blood flow in the rat.

Adenosine is a potent vasodilator of the systemic circulation. Infusion of adenosine into the aorta causes water and sodium retention and a fall in glomerular filtration rate and renal blood flow. The effect of adenosine on medullary blood flow is unknown. Because systemic vasodilatory effects may confound its renal actions, adenosine was infused into the renal artery of anesthetized Munich-Wistar rats at doses of 2, 6, and 15 micrograms/min. A marked dose-dependent increase in urinary flow and sodium excretion was observed. Inulin and p-aminohippuric acid clearance did not change significantly. Blood flow in vasa recta in the exposed renal papilla, as determined by fluorescence videomicroscopy, increased significantly only with the highest dose of adenosine. In control animals infused with the vehicle only, there was no change in any of the above variables. These results indicate that direct intrarenal infusion of adenosine in the rat increases urinary flow and sodium excretion and at higher doses also increases vasa recta blood flow. The effects on urinary flow and sodium excretion were therefore mediated by a mechanism other than an increase in vasa recta blood flow.

Effect of long-term, alternate day feeding on renal function in aging conscious rats.

To examine the renal effects of lifelong intermittent feeding, we performed clearance and pathologic studies in 86 week old, awake male Sprague-Dawley rats fed on alternate days (N = 9) or ad libitum (N = 8) since the age of four weeks. Alternate day-fed rats were studied on both feeding and fasting days, and the values averaged. In the alternate day group the clearances of inulin (Cinulin) and PAH (CPAH), factored by body wt, were higher by 23% and 27%, respectively (P less than 0.05); albumin excretion (UalbV) was two orders of magnitude lower (P less than 0.001) and the percentage of glomeruli with lesions was eightfold lower (P less than 0.02) than in the ad libitum-fed group. The fractional clearances of neutral dextrans ranging in radii from 20 A to 42 A did not differ between the two groups. Compared to a previously published study of 30 week old, alternate day-fed rats, values for Cinulin and CPAH were similar while UalbV was higher (P less than 0.025) in the 86 week old alternate day-fed rats. Cinulin, however, was lower (P less than 0.005) while UalbV was much higher (P less than 0.001) in 86 week old, ad libitum-fed rats than in 30 week old, ad libitum-fed rats. The results indicate that long-term alternate day feeding preserves glomerular filtration rate (GFR) and renal plasma flow (RPF), while glomerular permselectivity is not completely preserved, as evidenced by an increase in microalbuminuria in aging awake male rats. Conversely, ad libitum feeding results in a significant decline in GFR and probably in RPF, in association with massive albuminuria and segmental glomerular sclerosis.

Urea flux in the ureter.

It has been hypothesized that urea from the final urine is recycled into the renal papilla through the pelvic epithelium. To test this hypothesis, samples of urine were collected by micropuncture proximally and distally through the intact, contracting ureter of the anesthetized rat. In 12 rats, in which urine flow was 5.89 +/- 0.67 microliter/min (a moderate antidiuresis), the ratio of proximal-to-distal urea concentration, corrected for water movement, was 0.93 +/- 0.03 (P less than 0.01 compared with unity), indicating that approximately 7% of urea in the urine emerging from the terminal collecting duct was reabsorbed by the time it reached the distal ureter. To assess the possible contribution of urea reabsorption by the ureter, the ureter was cannulated proximally and distally and perfused with urine of known composition at 6.26 +/- 0.10 microliter/min. In nine rats, the ratio of urea concentration in the perfusate collected from the distal end of the ureter to that in the perfusate entering the proximal end was 0.93 +/- 0.02 (P less than 0.01 compared with unity), indicating 7% reabsorption. Movement of solute across the ureteral epithelium was not restricted to urea. Potassium and creatinine were also reabsorbed [3.4 +/- 0.9 (P less than 0.01) and 3.5 +/- 1.2% (P less than 0.05), respectively], whereas sodium was secreted [9.2 +/- 2.3% (P less than 0.01)].(ABSTRACT TRUNCATED AT 250 WORDS)

Course and pathogenesis of postischemic acute renal failure in the rat.

The effect of 45-min clamping of the renal artery was studied in the conscious uninephrectomized rat to reproduce the syndrome of hemodynamically mediated acute renal failure in humans after a single ischemic insult. Twenty-four hours after ischemia, creatinine clearance was reduced by 90%, whereas fractional excretion of sodium was markedly increased; over the subsequent 5 days, both values returned to normal. The animals were nonoliguric. Fractional clearances of graded sizes of neutral dextrans (radii 20-44 A), employed to detect transtubular backleak of inulin, were not significantly different 24 or 48 h postischemia from those in normal animals. The implication that the normal fractional dextran clearances excluded tubular backleak was tested directly by microinjecting [methoxy-3H]inulin into the proximal tubule. In most tubules injected, the recovery of radioactivity in the urine was markedly lower 24 and 48 h postischemia than that in normal rats; in a few injected tubules of postischemic kidneys, recovery was not different from that in normal animals. The low recovery of radioactive inulin was accounted for, at least in part, by transtubular backleak, as shown in experiments in which rats subjected to renal ischemia were cross-transfused with normal animals. These studies indicate that, despite the normal fractional dextran clearances, most tubules were severely injured as shown by tubule backleak of inulin.

Micropuncture study of the effect of ANP on the papillary collecting duct in the rat.

Micropuncture collections were obtained from the terminal collecting duct (CD) at base and tip of the renal papilla of the rat. Group 1 was studied before and during infusion with atrial natriuretic peptide (ANP), group 2 was administered the vehicle only, and group 3 received acetazolamide to increase sodium delivery to the base to a similar extent as after ANP. ANP caused a decrease in blood pressure, a slight increase in GFR, natriuresis, and diuresis. Sodium delivery to the collecting duct at the base of the papilla increased. Between base and tip, sodium reabsorption was inhibited. Tubule fluid sodium concentration (TFNa) was increased at the base and remained high at the tip; in contrast TFNa fell between base and tip in control and acetazolamide groups. After acetazolamide, sodium reabsorption in the terminal CD was not inhibited. These results demonstrate that in vivo ANP 1) increases the delivery of sodium to the terminal CD and 2) inhibits sodium reabsorption in the terminal CD. The findings for chloride were similar to those for sodium. ANP also increased delivery of H2O, K, Ca, and Mg to the CD at the papillary base but did not significantly affect their transport by the terminal CD.

Effect of adrenalectomy on transport in the rat medullary thick ascending limb.

Previous studies in adrenalectomized (Adx) rats suggest that aldosterone may regulate ion transport in the ascending portion of Helen's loop. In order to examine directly the effect of adrenalectomy on transport, medullary thick ascending limb (Mtal) segments were isolated from Adx, Adx replaced with aldosterone (Adx + Ald, 0.5 micrograms X 100 g X body wt X d), and control Sprague-Dawley rats. Both net sodium and net chloride fluxes were significantly less in the Mtal segments from Adx rats compared with those in the control or Adx + Ald group. Physiologic levels of exogenous aldosterone increased net sodium chloride flux toward control values in the Adx + Ald group. Net potassium flux was not different among the three groups. We conclude that adrenalectomy impairs reabsorptive NaCl but not K transport in the Mtal, and that aldosterone restores this process. This reabsorptive defect may contribute to the urinary concentrating and diluting abnormality associated with adrenal insufficiency.

Effect of V2-receptor-mediated changes on inner medullary blood flow induced by AVP.

We have previously shown that arginine vasopressin (AVP) in physiological amounts reduces inner medullary blood flow and that the mechanism of this decrease is at least in part mediated by the vasopressor (V1-receptor) action of AVP. To determine whether the antidiuretic action of AVP (V2-receptor) also contributes to the reduction in inner medullary blood flow, we determined capillary blood flow (QVR) in individual descending vasa recta (DVR) and ascending vasa recta (AVR) using fluorescence videomicroscopy in the exposed renal papilla of the anesthetized rat. Three groups of chronically water-diuretic rats were studied in three consecutive periods: control (period 1), experimental (period 2), and recovery (period 3). Group I rats (designated the AVP group) received AVP, 45 ng X h-1 X kg body wt-1; group II (AVP + V2-inhibitor), AVP plus its specific antidiuretic antagonist d(CH2)5[D-Ile2,Thr4]AVP; and group III (V2-inhibitor), the antagonist alone, respectively, in the experimental period 2. Only group I rats concentrated their urine, urine osmolality (Uosmol) = 499 +/- 48 mosmol/kgH2O, whereas urine remained hypotonic throughout in groups II and III. In group I, QVR in DVR and AVR decreased in period 2; but in groups II and III, QVR tended to increase. These results suggest that the AVP-induced decrease in papillary vasa recta blood flow is in part mediated by its antidiuretic V2-receptor as well as by its vasopressor (V1-receptor). They also suggest that the rate of urinary flow in the medullary collecting ducts is a determinant of inner medullary blood flow.