Exaggerated Protein Loss Complicating Sterile Peritonitis Results from Increased Diffusion

To assess the mechanism of protein loss during chemical peritonitis, dialyses with sodium desoxycholate added to the instilled solution were compared to controls in rabbits. Sodium desoxycholate caused an inflammatory reaction characterized by increased protein loss, leukocyte exudation, decreased ultrafiltration and increased clearance of solutes including neutral dextrans. It is suggested that increased protein loss accompanying peritonitis which is due to increased diffusion through capillaries unrelated to effects of membrane anionic charge.

Exaggerated protein loss is a recognized sequel of the peritonitis that complicates peritoneal dialysis (I, 2). Such excessive protein loss can contribute to malnutrition, impeding recovery from infection.

It has been assumed that the augmented protein loss is due to increased diffusion but this has not been verified. Secretion from inflammed cells adjacent to the peritoneum could account for increased protein loss. Enhanced transfer of plasma proteins across the capillary wall may be due to an overall increase in permeability following capillary dilation which stretches pores, or alternatively it might represent a loss of the anionic charges, which line the pores and retard transport of anionic macromolecules.

Previously the transfer rate of polydispersed dextrans from plasma to peritoneal dialysate has been used to assess peritoneal porosity, presumably reflecting capillary-wall permeability (3). The transfer of neutral dextrans should not be inhibited by the fixed anionic charges of the capillary wall interstices.

Moreover, these carbohydrates would not be secreted by injured cells; their distribution is extracellular. An increased rate of transfer from plasma to peritoneal dialysate should indicate enhanced diffusion, when there is no increase in convective transport accompanying ultrafiltration.

Sodium desoxycholate, the solvent used with amphotericin B, is an irritant that has been shown to induce a sterile peritonitis characterized by dialysate leukocytosis, increased diffusion of small solutes such as urea, and decreased ultrafiltration, owing to the loss of the glucose-induced osmotic gradient (4).

These studies were designed to determine whether sodium desoxycholate would increase the peritoneal transfer rate of neutral dextrans from plasma to dialysate. The higher rate of dextran transfer observed with this agent is interpreted to support the hypothesis that proteins, lost excessively during peritonitis, derive from plasma.

Cardiovascular Disease and Risk Factors in Patients Treated by Continuous Ambulatory Peritoneal Dialysis

The opinions and assertions contained herein are private and are not to be construed as official or representing those of the Uniformed Services University of the Health Sciences or the Department of Defense, U.S.A.

Augmentation of Peritoneal Hydraulic Permeability by Amphotericin B: Locus of Action

In normal rabbits, intraperitoneal instillation of amphotericin B increased peritoneal ultrafiltration rates without changing the osmotic gradient. Solute clearances were unaffected. When given intravenously the drug was ineffective, suggesting that amphotericin B acts on the serosal side of the vasculature or mesothelium, presumably creating channels through the membrane, thereby increasing hydraulic permeability.

Peritoneal transport rates are much lower than those of hemodialysis (I). To circumvent this deficiency, continuous peritoneal dialysis is now employed in many patients (2). Nevertheless. a segment of patients undergoing continuous ambulatory peritoneal dialysis have, or develop, inadequate transport rates for solutes, or more often for water (3). Hence, their dialyses are not efficient enough for maintenance therapy.

It is naive to consider that the peritoileum is inert like hemodialysis membranes. Accordingly, it has been observed that transport rates decrease with systemic vascular disease, or after vasoconstrictor drugs (4, 5) and increase when vasodilators are administered locally or have selective effects on the splanchnic vasculature (5 7). Also small number of studies have demonstrated that drugs, which affect membranes directly modulate peritoneal transport rates (8–10).

A preliminary evaluation of the effects of amphotericin B on peritoneal transport rates showed that osmotically induced water flux increases when the drug was given intraperitoneally (11). No direct effect on solute transport was detected and, over the range of 0.5 to 25 mg/kg, we noted no dose effect. Although amphotericin B is so large a solute that its own transmembrane transport would be slow, the site of action could not be determined with confidence from these studies. Accordingly, we undertook to compare the effects of intraperitoneal instillation with those of intravenous administration of amphotericin B in the same experimental animals.

Requirement for glutamine in the expression of alpha-actin and type IV collagen in mesangial cells

The importance of glutamine for the synthesis of proteins of the extracellular matrix was investigated in cultured mesangial cells. Glutamine at 2 mM elicited an increase in smooth muscle cell alpha-actin, alpha(1)-type IV collagen and fibronectin transcripts (19.0-, 16.7-, and 4.3-fold, respectively) accompanied by an increase in alpha-actin stress fibres compared to cells grown in absence of glutamine. The specificity for the glutamine requirement is suggested by the fact that mRNA levels of tenascin were not altered by addition of glutamine. This suggests that glutamine is required for the expression of important proteins of the extracellular matrix in cultured mesangial cells.

Hydrogen peroxide mediates FK506-induced cytotoxicity in renal cells

BACKGROUND

The nephrotoxicity induced by immunosuppressant FK506 remains a serious clinical problem, and the underlying mechanism has not been completely understood. The present study was undertaken to determine the role of hydrogen peroxide in FK506-mediated cytotoxicity in a porcine renal proximal tubular cell line, LLC-PK1 cells, and human embryonic kidney (HEK293) cells.

METHODS

Cytotoxicity was estimated by crystal violet and lactate dehydrogenase release assays. The activity of reactive oxygen species (ROS) was detected by flow cytometry. FK506-induced cell death was examined in the presence of the hydrogen peroxide scavenger, catalase, or a scavenger of hydroxyl radicals, sodium benzoate. As a control, FK506-induced cell death was also measured in the presence of superoxide anion inhibitor, 4,5-dihydroxy-1,2-benzene disulfonic acid (Tiron), TEMPO, or overexpressed human manganese superoxide dismutase (MnSOD). Catalase was also used in tumor necrosis factor-alpha (TNF-alpha)-induced cell injury to determine whether the enzyme specifically protected cells against FK506-mediated cytotoxicity.

RESULTS

FK506 induced cell death in a dose-dependent manner and coincided with a dose-dependent increase in ROS activity. Abrogation of FK506-mediated ROS by catalase and N-acetylcysteine blunted FK506-induced cell death. Furthermore, overexpression of catalase, sodium benzoate, and deferoxamine inhibited the cytotoxic effect of FK506. In contrast, Tiron, TEMPO, or overexpression of human MnSOD failed to show cytoprotection. In fact, TEMPO or expression of MnSOD enhanced the effect of FK506. Catalase did not significantly affect TNF-alpha-induced cell injury.

CONCLUSION

Catalase is uniquely required in cellular protection against FK506 cytotoxicity, which suggests an important role for hydrogen peroxide in the cellular actions of FK506.

Growth hormone increases low-density lipoprotein receptor and HMG-CoA reductase mRNA expression in mesangial cells

A dysregulation of the negative feedback mechanism of the low-density lipoprotein receptor (LDL-r) induced by hormones and cytokines may contribute to the development of glomerular injury and specifically could underlie growth hormone (GH)-induced glomerulosclerosis. The present study investigates the role of GH in the regulation of LDL-r and HMG-CoA reductase mRNA expression in mesangial cells. Mouse mesangial cells were equilibrated in a medium containing 5% lipoprotein-deprived serum (LPDS) for 48 h, prior to addition of GH (0.25 microM). Transcript levels of LDL-r, HMG-CoA reductase and GH-receptor (GH-r) were measured at days 2 and 4 and intracellular lipid content was evaluated by oil red-O staining. The addition of GH significantly increased both the LDL-r and HMG-CoA reductase transcript levels at day 2 compared to control. Oil red-O positive staining increased following the initial period of 48 h lipoprotein deprivation, but addition of GH in a subsequent 48-hour period did not alter lipid content to a measurable degree compared with control. The present study demonstrates that GH significantly increased HMG-CoA reductase and LDL-r transcript levels in mesangial cells deprived of lipoproteins suggesting that abnormal levels of GH may play a role in glomerular lipid accumulation.

Inhibition of Na,K-ATPase activates PI3 kinase and inhibits apoptosis in LLC-PK1 cells

In the present study we used LLC-PK1 cells, a porcine renal proximal tubular cell line, to investigate whether PI3 kinase activation was involved in the anti-apoptotic effect of ouabain, a specific inhibitor of Na,K-ATPase. Apoptosis was induced by actinomycin D (Act D, 5 microM) and assessed by appearance of hypodiploid nuclei and DNA fragmentation. Ouabain attenuated Act D-induced apoptotic response in a dose-dependent manner. Incubation in a low K(+) medium (0.1 mM) which is another way to decrease Na,K-ATPase activity also had anti-apoptotic effect. Both ouabain and low K(+) medium increased the PI3 kinase activity in p85 immunoprecipitates. Ouabain, as well as incubation in the low K(+) medium, also increased the phosphorylation of Akt. Inhibition of PI3 kinase by either wortmannin or LY294002 reversed the cytoprotective effect of ouabain. These data together indicate that inhibition of Na,K-ATPase activates PI3 kinase in LLC-PK1 cells which could then exert the cytoprotective effect.

Low-protein diet suppresses serum insulin-like growth factor-1 and decelerates the progression of growth hormone-induced glomerulosclerosis

A low-protein (LP) diet has been associated with amelioration of renal function in glomerulosclerosis (GS). However, the mechanisms involved are still unclear. We have used a mouse transgenic for bovine growth hormone (GH), which develops progressive GS and exhibits consistently elevated levels of circulating GH and insulin-like growth factor (IGF)-1, to study the effect of dietary protein restriction. LP (6% protein) and normal-protein (NP, 20% protein) diets were maintained for 30 weeks in mice with established GS of mild/moderate degree. The degree of GS was markedly attenuated in LP compared to NP mice. Quantitative analysis revealed a significantly lower GS index (1.4 +/- 0.9 in LP vs. 2.8 +/- 0.8 in NP) and glomerular volume (0.8 x 10(6) +/- 0.1 x 10(6) microm(3) in LP vs. 1.2 x 10(6) +/- 0.1 x 10(6) microm(3) in NP) in mice with restricted protein intake. These morphologic changes were accompanied by a significant reduction in renal expression of alpha(1) type-IV collagen (2.4-fold) and tenascin (1.4-fold) in LP mice. Serum IGF-1 decreased by 40% and showed a significant correlation with alpha(1) type-IV collagen expression with the LP diet. The present finding supports the use of the LP diet to decelerate the progression of GS and furthermore suggests that one of the mechanisms involved in this process is the GH/IGF-1 regulation by protein intake.

Disulfiram inhibits TNF-alpha-induced cell death

Disulfiram, a clinically employed alcohol deterrent, was recently discovered to inhibit caspase-3 and DNA fragmentation. Using LLC-PK1 cells and murine liver as models, we examined if the drug inhibited TNF-alpha-induced cell death. Disulfiram produced dose-dependent inhibition of TNF-alpha-induced cell death as well as caspase-3-like activity. Disulfiram retained 80% of its effect when added 4 h after TNF-alpha. Disulfiram protected the cells from cytokine-induced death for at least 6 days. The cells rescued by the drug preserved the ability to proliferate. The cells died spontaneously after exposure to TNF-alpha for just 70 min. Co-administration of 15 microM disulfiram and TNF-alpha for 70 min prior to their removal abolished TNF-alpha-induced killing, and this was associated with restoration of mitochondrial membrane potential and suppression of reactive oxygen species. Treatment of mice with TNF-alpha and D-galactosamine for 5 h markedly increased hepatic DNA fragmentation and caspase-3-like activity. Disulfiram at 0.6 mmol/kg abolished these effects. We conclude that disulfiram is a potent inhibitor of TNF-alpha-induced cell death in vitro. The underlying mechanisms include stabilization of mitochondrial membrane potential, suppression of reactive oxygen species, and inhibition of caspase-3-like activity. We further conclude that disulfiram inhibits DNA fragmentation in vivo in association with the blockade of caspase-3-like activity.

Growth hormone increases inducible nitric oxide synthase expression in mesangial cells

Mice transgenic for bovine growth hormone (GH) develop progressive glomerulosclerosis. However, the proximal signaling events that lead to increased matrix deposition in this pathologic condition are still unclear. Components of the L-arginine metabolic pathway, especially inducible nitric oxide (NO) synthase (iNOS), ornithine aminotransferase (OAT), and ornithine decarboxylase (ODC), have been associated with glomerular scarring. In this study, mesangial cells were treated with GH, and the expression of iNOS, ODC, and OAT was determined using reverse transcription-PCR. In addition, nitrite accumulation in the conditioned media of mesangial cell cultures was measured in the presence or absence of GH. The findings revealed that GH increased iNOS transcript levels in a dose-dependent manner, with the highest levels being attained at GH concentrations of 20 to 50 ng/ml. The GH-induced increase in iNOS transcript levels was accompanied by a significant increase in nitrite concentrations in conditioned media, which was blocked by the addition of L-N(G)-monomethylarginine. The effect of GH (50 ng/ml) in eliciting nitrite production was as potent as that of bacterial lipopolysaccharide (10 microg/ml). The expression of OAT and ODC, in contrast, was not altered at any of the GH concentrations tested. GH receptor mRNA was also expressed by mesangial cells, independently of the GH concentration present in the cell culture medium. These data indicate that GH may interact with its receptor to regulate the L-arginine/NO pathway in mesangial cells, by directly modulating iNOS expression and NO production, without altering the arginase/OAT/ODC pathway.

Gliotoxin-induced cytotoxicity proceeds via apoptosis and is mediated by caspases and reactive oxygen species in LLC-PK1 cells

Renal failure associated with aspergillosis is caused by pathogenic fungi. Gliotoxin is a toxic epipolythiodioxopiperazine metabolite produced by the pathogens. The present study investigated the cytotoxicity and underlying mechanisms induced by gliotoxin in LLC-PK1 cells, a porcine renal proximal tubular cell line. Gliotoxin at 100 ng/ml did not show a cytotoxic effect, but unmasked a dose-dependent cell death induced by TNF-alpha. TNF-alpha-induced cell death in the presence of gliotoxin was associated with hypodiploid nuclei and activation of caspase-3-like proteases. Blockade of caspases by boc-aspartyl (OMe)-fluoromethylketone and z-DEVD.fmk inhibited TNF-alpha-induced cell death. As the concentrations of gliotoxin were increased, gliotoxin killed the cells directly in a dose-dependent manner. Further analyses of DNA fragmentation, hypodiploid nuclei, mitochondrial membrane potential, and plasma membrane integrity revealed that cell death proceeded via apoptosis. Gliotoxin-induced apoptosis was associated with dose-dependent and time-dependent activation of caspase-3-like proteases. Boc-aspartyl (OMe)-fluoromethylketone attenuated the killing effect. Gliotoxin also increased the intracellular levels of reactive oxygen species as measured by flow cytometry. N-acetylcysteine, a well-known antioxidant, completely abolished the gliotoxin-induced caspase-3-like activity, cytotoxicity, and reactive oxygen species. In conclusion, (1) gliotoxin at 100 ng/ml unmasks the ability of TNF-alpha-induced apoptosis, and the effect of TNF-alpha is mediated by caspase-3-like proteases; and (2) at higher concentrations gliotoxin itself induces cell death, which is via apoptosis and dependent on caspase-3-like activity and reactive oxygen species.

Renal hemodynamic response to the creation of vascular access in patients with end-stage renal disease

To evaluate the possibility that the placement of arteriovenous anastomosis (a/v a) may lead to the attenuation of glomerular hyperfiltration, we studied 5 nondiabetic patients before and after creation of vascular access for hemodialysis. Patients received no EPO and antihypertensive therapy was discontinued 24 h before each study. Cardiac output (CO) and a/v a flow rates were measured by Doppler echo, and GFR and ERPF by plasma decay curves of Tc99m DTPA and 131I-hippuran, respectively. Other parameters were calculated by standard formulas. Augmentation of CO and decrease in systemic vascular resistance occurred in all patients (p = 0.05), yet renal findings were less predictable since only three patients showed a decrease in renal vascular resistance and filtration fraction post a/v a. Thus, there is a discordant pattern of renal hemodynamic response to the creation of a/v a in end-stage renal disease and further studies are needed to better define the subset of patients who are prone to renal vasodilation after the placement of a/v a.

Radiocontrast removal by dialysis membranes

To examine the effect of membrane characteristics on radiocontrast mass transfer, we studied in vitro clearances with cuprophane and polyacrylonitrile (PAN) dialyzers and with polysulfone hemofilter. A perfusate of saline at 37 degrees C, with 7.5 mmol/l (45 mg/dl) of urea and either Renografin (molecular weight 723) or Hexabrix (1405), was pumped through the blood path of dialyzers at 200 ml/min for 5 min. Each radiocontrast agent (RCA) and dialyzer was tested at 0, 150, and 250 or 300 mm Hg transmembrane pressure. In experiments with the use of hemofilter, clearances were tested at the perfusate flow of 50, 100 and 150 ml/min. RCAs were measured by fluorescent excitation analysis of iodine. Mean urea clearance was 16% higher in PAN than in cuprophane dialyzers. Clearance of RCAs was 1.5-3 times higher in PAN than cuprophane dialysers. With the latter, increases in transmembrane pressure resulted in a small amount of ultrafiltration (UF) and little increment in RCA clearance. With the former, increases in transmembrane pressure resulted in massive UF and remarkable increases in RCA clearance. Renografin clearance generally exceeded that of Hexabrix, which we attributed to Renografin's lower molecular weight. With the hemofilter, sieving coefficients were approximately 0.8 for each RCA. Yet, because of the lack of diffusive transport and a small surface area even at the highest perfusate flow rates, RCA clearance by the hemofilter was 20-50% less than that of cuprophane dialyzers. We conclude that PAN dialyzers are more efficient for RCA removal than cuprophane dialyzers or polysulfone hemofilters.

Splanchnic volume, not flow rate, determines peritoneal permeability

To distinguish the effects of splanchnic blood flow rate from those of splanchnic volume on peritoneal transfer rates, measurements were made in rabbits before and after intraperitoneal exposure to sodium chromate. The sodium chromate induced reversible hepatic sinusoidal block with consequent portal venous congestion and stasis, which was demonstrable on histologic sections. Concurrently the ultrafiltration rate, and ultrafiltration coefficient each doubled after chromium even though the dialysate reabsorptive rate increased and the absorptive diffusion of glucose was at least as high as in control dialyses. Chromium induced significant increases in mass transfer coefficients of urea, potassium and phosphate and in protein clearance. These data suggest that splanchnic volume is an important determinant of peritoneal transfer functions and that the hepatic capillaries may contribute appreciably to transfer ordinarily ascribed to peritoneal capillaries alone.

Effects of histamine and its receptor antagonists on peritoneal permeability

Peritoneal fluid and mass transfer rates were studied in rabbits undergoing control dialyses and dialyses with intraperitoneal histamine, or its receptor antagonists alone or in combination. These drugs had negligible effects on peritoneal ultrafiltration and small solute clearances. Histamine raised protein exudation from 1.6 to 2.9 mg/kg/min, an effect blocked by its antagonists which given alone did not lower protein loss. These data demonstrate the existence of histamine receptors in the peritoneal diffusion barrier and show that they do not control transport under baseline conditions, but can be blocked should abnormal histamine release occur. Increased peritoneal permeability with sterile peritonitis was unaffected by ranitidine, suggesting alternative mediators.

Urinary Charcot-Leyden crystals in the hypereosinophilic syndrome with acute renal failure

A 48-year-old man with idiopathic hypereosinophilic syndrome (IHS) developed blast crisis along with a fulminant autoimmune hemolytic anemia. Hemoglobinuria and anuric acute renal failure (ARF) ensued. Urinalysis revealed countless Charcot-Leyden crysals (CLC). This is the only known report of Charcot-Leyden crystalluria. The CLC protein (lysophospholipase) should normally undergo glomerular filtration and catabolism by the tubules during reabsorption. Its abundant presence in the urine of our patient may reflect impairment of tubular reabsorption, saturation of the tubular reabsorptive process by excessive CLC load through residual functioning glomeruli, or a combination thereof. The extreme degree of hypereosinophilia suggests a massive load of CLC protein and acute tubular necrosis implies impaired tubular function, so both mechanisms should have been operative. At the autopsy, no eosinophilic infiltrates were present in the kidneys, which points against a local spillage of CLC protein into the tubules.

Prolonged intraperitoneal dwell decreases ultrafiltration coefficient in rabbits

In rabbits undergoing peritoneal dialysis, hypertonic (6% dextrose) dialysis solution increased the net ultrafiltration rate (UF) from 233 to 462 microL/kg/min, which was not proportional to the increment in the osmotic gradient, so the ultrafiltration coefficient decreased. As intraperitoneal dwell of hypertonic dialysate was prolonged, the gross and net UFs and ultrafiltration coefficients decreased, and the UF per dextrose absorption declined. The decrement in UF was multifactorial, including a component of fluid and solute stagnation, increasing the distance over which osmotic forces must exert their effects. Excessively hypertonic dialysis fluid should be used only briefly to achieve ultrafiltration efficiently and to avoid the high dextrose loading.

The role of the capillary wall in restricting diffusion of macromolecules. A study of peritoneal clearance of dextrans

In 5 nephrectomized rabbits the peritoneal clearance of neutral dextrans from plasma to dialysate decreased from 7.8 to 3.3 microliters/kg/min as molecular mass increased from 17,000 to 43,000 daltons, and was relatively constant at 2.8 microliters/kg/min from 49,000 to 97,000 daltons in accord with prior studies. The clearance from dialysate to plasma was measured by determining the distribution volume, which averaged 72 ml/kg, and the plasma concentration 5 h after intraperitoneal instillation. Inward clearances ranged from 11.4 to 19.9 microliter/kg/min, did not correlate well with solute size and were significantly higher than outward clearances. The data suggest that while the capillary wall is the major barrier to macromolecule transfer, absorption can bypass vascular capillaries and occur via the lymphatics. It is suggested that lymphatic flow rate from the peritoneum exceeds 16 microliter/kg/min.

Contrasting effects of amphotericin B and the solvent sodium desoxycholate on peritoneal transport

To distinguish amphotericin B effects on peritoneal transport from those of the solvent, sodium desoxycholate, dialyses in intact rabbits with either substance added intraperitoneally were compared to controls. Powered amphotericin B added to instilled dialysis fluid increased peritoneal ultrafiltration from 0.31 to 0.44 ml/kg/min (p less than 0.02), but did not affect mass transport (e.g. urea clearance changed from 0.86 to 1.04 ml/kg/min). In contrast, 10 mg of desoxycholate induced peritoneal irritation and raised clearances of urea (0.76-1.34 ml/kg/min), potassium, phosphate and dextrose, but did not affect ultrafiltration. Intraperitoneally, 1 mg/kg of desoxycholate changed clearances inconsistently, but lowered the ultrafiltration rate from 0.33 to 0.21 ml/kg/min. The dialysate-plasma dextrose gradient dissipated faster with 10 mg/kg of desoxycholate. Amphotericin B tended to raise ultrafiltration per osmotic gradient and mass transport of sodium. Selective increase in fluid flux results from amphotericin B, not its solvent.

The mechanism of dextrose-enhanced peritoneal mass transport rates

The mechanism whereby hypertonic dextrose affects peritoneal transport was investigated in a short-term model of peritoneal dialysis using alert intact rabbits. During control (1.5% dextrose) dialyses osmotic ultrafiltration was 0.28 mg/kg/min, the clearance of potassium was 0.98, urea 0.54, phosphate 0.32, and dextrose (reverse) 0.21 ml/kg/min. With 4.25% dextrose, the ultrafiltration rate increased to 0.73 ml/kg/min (P less than 0.02), but solute transport did not increase despite the added convective flux. The posthypertonic exchanges did not differ from control despite the effect of residual dialysate contaminating this peritoneal lavage. By indicator dilution residual volume averaged 12% of total dialysate volume. Acute volume expansion by intravenous dextrose after desoxycorticosterone acetate (DOCA) pretreatment increased the ultrafiltration coefficient, potassium and urea clearances significantly, and DOCA alone was ineffective. It is suggested that in uremic humans hypertonic dextrose dialysis increases peritoneal mass transport rates because the absorbed dextrose causes extracellular volume expansion that cannot be eliminated promptly. No evidence of a direct effect of dextrose augmenting peritoneal permeability was detected.

Mercury-induced autoimmune glomerulonephritis in inbred rats. I. Kinetics and species specificity of autoimmune responses

The nephropathy observed in rats after administration of mercuric chloride can be used to clarify the mechanisms underlying renal autoimmunity induced by chemicals. As a necessary preliminary step in the study of this animal model, we have investigated the kinetics and species-specificity of autoimmune responses to renal antigens. By a recently developed enzyme-linked immunosorbent assay (ELISA), circulating autoantibodies to the glomerular basement membrane of the kidney (anti-GBM) have been detected within 8 days after the initiation of mercuric chloride treatment. Anti-GBM antibodies reach a peak by 15 days and then decrease rapidly in the following 2 weeks. Extensive cross-reactions between rat and human GBM antigens have been detected by ELISA, indicating a high degree of conservation of some renal autoantigens and suggesting certain similarities between the autoimmune response induced in rats by mercuric chloride and that observed in human glomerulonephritis caused by anti-GBM. Dose-response studies have been performed to ascertain whether anti-GBM responses are correlated with massive kidney damage and release of renal antigens. We have noted that a wide range of levels of mercuric chloride are capable of stimulating the production of anti-GBM and that animals receiving this chemical in as low a concentration as 0.02 mg/100 g body weight (i.e. a dose ten times lower than those causing massive nephrotoxic effects) still have anti-GBM specifically bound to their kidneys. Thus, it is possible that the administration of mercury compounds to BN rats results in kidney autoimmunity not only because of the release of renal autoantigens, but also through the activation of specific lymphocytes and/or disruption of regulatory networks. Finally, we have observed that both BN and MAXX rats produce anti-GBM after mercuric chloride treatment, while M520 rats do not. Since the MAXX strain was initially obtained from a cross of BN and Lewis rats and shares antigens of the major histocompatibility complex with the BN strain, our findings stress the importance of genetic factors in chemical-induced autoimmunity and suggest that a similar situation may occur in human subjects exposed to environmental chemicals.

Amphotericin selectively increases peritoneal ultrafiltration

Because amphotericin B is known to affect transport rates across biologic membranes, the effects of this agent on transport parameters in an animal model of peritoneal dialysis were investigated. When amphotericin B in doses ranging from 0.5 to 25 mg/kg was instilled intraperitoneally with commercial dialysis solution, diffusive clearances of phosphate and urea did not differ from control values measured in the same animals, and only a modest increase in potassium clearance was detected. Ultrafiltration due to the osmotic gradient induced by the dextrose content of the dialysis solution increased significantly to 0.31 mL/kg/min with amphotericin B, compared with control values of 0.18 mL/kg/min. The drug did not affect dextrose transport and the osmotic gradient did not differ in the two groups. Hence, the ultrafiltration coefficient was higher with amphotericin B (14 microL/kg/min/mosm), than during control dialyses (6 microL/kg/min/mosm). Increased water flux was detected at the lowest dose and there was no dose relationship over the range studied. Amphotericin B may be the type of agent that will be clinically useful in patients with reduced peritoneal ultrafiltration capacity, and safer analogues should be explored.

Acceleration of peritoneal solute transport by cytochalasin D

Because cytochalasin D affects intercellular junctions the effect of this agent on peritoneal transport was investigated in normal rabbits. Using commercially available dialysis solution, short-term control peritoneal dialyses were compared in the same animals to dialyses in which cytochalasin D was added intraperitoneally. A dose (325-920 micrograms/kg) dependent increase in peritoneal clearances of urea (49% increment at high dose) and of creatinine (67% increment) occurred when cytochalasin D was added. When solute transport was highest at the maximal dose, osmotically induced ultrafiltration decreased significantly to 33% of control values. Cytochalasin D induces aberrations in solute transport that resemble those accompanying and occasionally following peritonitis.

Mesangial proliferative glomerulonephritis with IgM deposits. Clinicopathologic analysis and evidence for morphologic transitions

To determine the natural history of mesangial proliferative glomerulonephritis (MesPGN) with IgM deposits and its relationship to minimal change disease (MC) and focal segmental glomerulosclerosis (FGS), we studied the clinical characteristics and outcome in 20 patients with MesPGN, 8 with MC, and 10 with FGS. IgM deposits were present in glomeruli of all MesPGN patients. Progression to FGS was documented in 2 patients with MesPGN, 1 of whom developed renal failure. Transition from MC to MesPGN occurred in 1 patient. 2 MC patients developed FGS, with decline in renal function in 1 of them. These data suggest the possibility of histologic transition from MC to FGS directly or through the stage of MesPGN.

Glycosylated hemoglobin in patients treated by chronic dialysis

A glycosylated hemoglobin (HbA1) test was used to evaluate the role of dialysate glucose in the development of carbohydrate intolerance and hyperlipidemia in chronic hemodialysis patients and chronic peritoneal dialysis patients. HbA1 levels were significantly elevated in all groups of patients. HbA1 levels were not ameliorated with 8 weeks of glucose-free hemodialysis. There was no correlation between HbA1 and serum glucose, triglycerides, or cholesterol. Thus, HbA1 elevation cannot be explained solely by glucose reabsorption from dialysate. This test is helpful in the detection of carbohydrate intolerance, but its usefulness in evaluation of hyperlipidemia of dialysis patients is uncertain.

Peritoneal dialysis in the treatment of severe hypercalcemia

A 56-year-old man with multiple myeloma and compromised renal function underwent peritoneal dialysis for the treatment of severe hypercalcemia. During dialysis, peritoneal clearances of total calcium, unbound calcium, urea, and creatinine were assessed. Clearances of total calcium (4.8 +/- 0.4 ml/min) and unbound calcium (7.8 +/- 0.5 ml/min) were shown to vary directly with the clearances of urea (15.5 +/- 1.3 ml/min) and creatinine (8.5 +/- 0.8 ml/min). Despite relatively low clearances of all these solutes, during the period of 42 hours, 1,638 mg of calcium was removed in the dialysate and total serum calcium decreased from 17.6 mg/dl to 10.2 mg/dl. Our data indicates that peritoneal dialysis is an effective adjunct in controlling severe hypercalcemia and should be considered when other forms of therapy are inadequate.

Augmentation of peritoneal mass transport by dopamine: comparison with norepinephrine and evaluation of pharmacologic mechanisms

The effect of catecholamines on transport during peritoneal dialysis was studied in unanesthetized rabbits. Intravenous I-norepinephrine consistently decreased peritoneal clearances of urea and creatinine to 84% of control values or less but did not affect osmotically induced water flux. Comparable pressor doses of dopamine increased clearances of urea and creatinine to 145% of control values, whereas osmotic fluid flux increased only slightly. Dopamine also increased urea transport when administered intraperitoneally. The augmentation of solute transport by dopamine was unaffected by simultaneous administration of propranalol, was decreased by phentolamine, and was abolished by haloperidol. Dopamine may be preferable toI-norepinephrine when vasopressor therapy is required during peritoneal dialysis. The augmented transport with dopamine appears to depend on the action of dopamine receptors causing mesenteric vasodilation and in part on alpha-adrenergic receptors simultaneously increasing blood pressure while mesenteric blood flow is maintained.

Prediction of solute transport during peritoneal dialysis

Solute transport, predominantly diffusion, across the peritoneum correlates inversely with molecular weight. Provided that the solute is water soluble, not protein bound, not of unusual density, not ionized, does not have a large hydration shell, and is transported from plasma to dialysate, the peritoneal clearance is predictable over the molecular weight range from 60 to 11,000 daltons. Transport reates that deviate from the predicted can be explained by known physical properties of particular solutes.

Effects of gastrointestinal hormones on transport by peritoneal dialysis

Because the gastrointestinal hormones are known to dilate the splanchnic vasculature, their effects on transport of water and solutes during peritoneal dialysis were studied in an experimental model, the rabbit. In unanesthetized rabbits, dialysate volume was calculated by isotope dilution, and clearances were estimated by dialysate/plasma concentration ratio factored by minute volume. With isotonic dialysis solution, the mean increment in dialysate volume per minute of intraperitoneal dwell was 0.19 ml/kg/min, and mean clearances of creatinine and urea were 0.71 and 0.90 ml/kg/min, respectively. When administered intravenously, secretin significantly augmented osmotically induced water flux, but not when given intraperitoneally. Neither glucagon nor cholecystokinin affected dialysate volume. Intravenously, but not intraperitoneally, glucagon increased peritoneal clearances of creatinine and urea to more than 150% of control values. Neither cholecystokinin nor secretin augmented significantly peritoneal mass transport when given by either route. The data suggest that the site of acton is the endothelial surface of the membrane, that the mechanisms of augmenting transport involve increased permeability and/or surface area, and that agents which combine an increase in mass transport and capillary filtration coefficient may be clinically useful.

Divergent effects of catecholamines on peritoneal mass transport

In rabbits, intravenous vasopressor doses of dopamine augmented peritoneal clearances of creatinine and urea, suggesting increased mesenteric blood flow and possibly augmented permeability. Intraperitoneal dopamine also accelerated peritoneal transport of urea. Solute transport across the peritoneum was decreased by intravenous infusion of 1-norepinephrine. Intraperitoneal administration of the alpha-adrenergic antagonist phentolamine partially abolished the augmentation of peritoneal clearances induced by intravenous dopamine. The results suggest that in patients undergoing peritoneal dialysis who require vasopressor therapy, dopamine should be preferred to norepinephrine.

Augmented peritoneal mass transport with intraperitoneal nitroprusside

Lightly restrained, alert New Zealand white rabbits underwent peritoneal dialysis by percutaneous instillation of standard dialysis solution with or without intraperitoneal nitroprusside. Corrected to a mean intraperitoneal dwell time of 36 minutes, mean clearances of creatinine and urea were 0.74 and 0.90 ml/kg/min in six rabbits. With intraperitoneal nitroprusside, 1.13 mg/kg clearances increased to 1.13 and 1.30 ml/kg/min (p less than 0.01) respectively. The 53% increment in creatinine clearance maintained the ratio clearance larger/smaller solute suggesting increased peritoneal permeability and/or area. Lower nitroprusside doses were less effective and not significantly above control. Nitroprusside also increased clearances during hypertonic peritoneal dialysis, but had no effect on osmotically induced water flux. Lavage studies demonstrated a persistent effect of nitroprusside after a single exposure and a sustained effect with repeated use.

Uremic autonomic neuropathy: evaluation of ephedrine sulphate therapy for hemodialysis-induced hypotension

The hemodynamic response to ephedrine sulphate were studied in a patient on maintenance hemodialysis therapy with chronic renal failure due to renal amyloidosis. The evaluation (including cardiac catheterization studies) and estimation of responses to Valsalva maneuvers before and after ephedrine administration documented the diagnosis of autonomic insufficiency. Oral ephedrine failed to influence the episodes of severe dialysis-induced hypotension. Also the patient did not benefit from the infusion of Aramine. These studies suggest that catecholamine stores of adrenergic nerves may be depleted in uremic patients with clinical signs of autonomic neuropathy.

Peritoneal clearances in scleroderma and diabetes mellitus: effects of intraperitioneal isoproterenol

Reduced peritoneal clearances of creatinine and urate were demonstrated repeatedly in a patient with scleroderma and a patient with diabetes mellitus. Urea clearances were not significantly different from usual values. The findings suggest decreased peritoneal membrane permeability and/or area (if urea clearance is flow limited). Clearances increased to usual values with intraperitioneal isoproterenol in the patient with diabetes. There was no effect of isoproterenol in the patient with scleroderma.

Effect of hemodialysis on factors influencing oxygen transport

Ten patients underwent 4 study hemodialyses, one with standard dialysis conditions, one with an isophosphate dialysate, one with simultaneous ammonium chloride loading, and other, after pretreatment, with sodium bicarbonate. Measurement of hemoglobin oxygen affinity (P-50), erythrocyte 2,3-DPG, blood-gasses, and serum chemistries revealed biochemically effective hemodialyses and slight changes in oxygen transport parameters. The P-50 (in vivo) values decreased slightly but significantly (p greater than 0.05) with dialysis. When corrected to pH 7.4, eliminating the Bohr effect, P-50 increased (p greater than 0.05). With unmodified dialysis elevated values of 2,3-DPG (in comparison to normal) decreased, a change that did not correlate with delta-p-50, delta-serum phosphate, or delta-serum creatinine. With standard and isophosphate dialyses Po-2 decreased significantly. The decrease correlated with delta-hydrogen ion concentration and did not occur with dialyses designed to maintain pH constant. Thus, hemodialysis influences many factors that affect oxygen transport in different and counterbalancing directions. These changes are not totally explained by alterations in 2,3-DPG, pH or serum phosphate. Maintenance of acidosis or hyperphosphatemia during dialysis is not recommended.

Influence of peritoneal dialysis on factors affecting oxygen transport

To determine the effect of changing concentrations of uremic metabolites on factors affecting oxygen transport, without the effects of extracorporeal blood pumping, we studied five patients before, during and after peritoneal dialysis. Significant decreases in serum urea, creatinine and phosphate and increase in serum bicarbonate were not associated with changes in P50, a reflection of hemoglobin-oxygen affinity. High erythrocyte 2,3-DPG concentrations decreased only slightly. Arterial pO2 increased slightly as negative fluid balance was achieved. The slight changes in oxygen transport parameters with dialysis suggest an interplay of compensatory factors and do not warrant modifying dialysis to limit the correction on acidosis or hyperphosphatemia. Effects on hemoglobin and pO2 resulting from fluid loss can be the dominant influence of peritoneal dialysis on tissue oxygenation.