Immunopathogenesis of chronic pyelonephritis.

Immunopathologic responses to urinary Tamm-Horsfall protein in the development of chronic pyelonephritis were examined by four different approaches. First, in a rabbit model, tubulointerstitial nephritis developed in 64 of 102 rabbits injected intravenously with urine or rabbit Tamm-Horsfall protein as compared with only one of 17 rabbits in two control groups. Circulating cytotoxic lymphocytes plus immunoglobulin G (IgG) antibodies against Tamm-Horsfall protein were found in 51 percent of challenged (urine or Tamm-Horsfall protein) rabbits with tubulointerstitial nephritis as compared with only 8 percent of those without it (p less than 0.001). Second, in a porcine model of reflux nephropathy, 16 of 21 pigs with pyelographic findings indicative of reflux had elevated serum titers of anti-Tamm-Horsfall protein antibody as compared with 0 of 13 with normal pyelograms. Five of 10 refluxing pigs tested also had circulating lymphocytes that were cytotoxic in the presence of Tamm-Horsfall protein as compared with 0 of 13 with normal pyelograms. Third, in human studies, 12 of 49 patients with recurrent nephrolithiasis demonstrated abnormal elevations in anti-Tamm-Horsfall protein antibody; 13 of 49 had an abnormality in one of two assays of cell-mediated immunity to Tamm-Horsfall protein as compared with 0 of the normal control subjects. These abnormalities were not associated with overt obstruction or bacteriuria, but appeared to be more common in patients with recent onset and active recurrent nephrolithiasis. Lastly, an inhibitor of the binding reaction between human Tamm-Horsfall protein and its IgG antibody was detected in extracts of three uropathic coliforms. The inhibitors were partially purified by chromatographic means. Preliminary immunoautoradiographic studies revealed three or less protein-containing subunits of Escherichia coli that cross-reacted with anti-Tamm-Horsfall protein antibody. These studies suggest that autoimmune responses to Tamm-Horsfall protein may occur after exposure to Tamm-Horsfall protein by intravenous challenge, urinary reflux, or recurrent nephrolithiasis. This autoimmune response to Tamm-Horsfall protein may be the pathogenetic mechanism by which these factors, including bacteriuria, contribute to chronic pyelonephritis.

Evolving aortic mass in a patient with sepsis and systemic embolization. Detection by transesophageal echocardiography.

An elderly patient with sepsis and systemic embolization is described. An intraluminal aortic mass was discovered by transesophageal echocardiography that appeared to be the source of infection in this patient. Transesophageal echocardiography can be a useful diagnostic test in patients with sepsis and systemic embolization of unknown etiology.

Successful medical therapy for endocarditis due to Candida parapsilosis. A clinical and epidemiologic study.

A heroin addict with asymmetric septal hypertrophy and persistent fungemia with Candida parapsilosis was treated with amphotericin B and flucytosine (5-fluorocytosine). The diagnosis of endocarditis was based on the subsequent development of a murmur of mitral regurgitation and echocardiographic evidence of prolapse of the posterior leaflet of the mitral valve. Cure was effected with antifungal therapy alone. Thus, when the diagnosis of fungal endocarditis is made early in its course, open-heart surgery may not be needed. To investigate the relative frequency of isolation of C parapsilosis from particular sites, a mycologic survey was conducted in our hospital. Among the isolates of yeasts, C parapsilosis represented 8.0, 17.1, and 26.7 percent of those from all cultured sites, from contaminated intravenous catheters, and from cultures of blood, respectively. Since this trend to cluster in cases of fungemia was not seen with other yeasts, C parapsilosis appears to be more invasive than other species of Candida.

Urinary tract antiseptics.

The urinary tract antiseptics discussed herein have specific advantages and disadvantages. All share a pharmacokinetic fate that makes them effective in treating acute, uncomplicated symptomatic bladder bacteriuria. Nitrofurantoin appears to be the most versatile because it is effective against upper tract infection, recurrent bacteriuria, and as a long-term suppressive agent in children and pregnant patients with only a low incidence of the development of resistance. Methenamine, when used with proper understanding of it pharmacokinetic behavior, is also effective in females with uncomplicated recurrent bacteriuria including those with multiply resistant pathogens, as well as a prophylactic agent in males with recurrent infection. There is little evidence that methenamine combined with mandelic or hippuric acid confers any pharmacologic or therapeutic advantage over the use of methanamine base alone. Nalidixic acid and oxolinic acid, in addition to effectiveness in treating uncomplicated acute lower urinary tract infections, may be effective in some patients with recurrent infections, but requires careful sensitivity monitoring of pathogens for the development of resistance. Finally, in a society whose economic pressures are such that it may not be cost-effective to use sulfamethoxazole-trimethoprim for urinary tract prophylaxis--unless two or more acute infections occur per year--the use of these urinary antiseptics may offer increasing advantages now and in the future.

Tubulointerstitial nephritis in rabbits challenged with homologous Tamm-Horsfall protein: the role of endotoxin.

Tubulointerstitial nephritis developed in 25 of 34 (74%) rabbits challenged intravenously for 2-20 weeks with adjuvant and endotoxin free homologous Tamm-Horsfall protein (THP). Lesions were characterized by focal mononuclear cellular infiltrates and microscopic scarring localized to distal nephron segments identified as thick ascending limb of the loop of Henle. Interstitial deposits of THP were found in the kidneys of more severely affected rabbits and metabolic studies demonstrated transient polyuria and tubular dysfunction. Elevations in serum IgG antibody against THP were detected in seven of 34 challenged rabbits. Tubulointerstitial nephritis was found in six of the seven rabbits with elevated antibody as well as in 19 of 27 rabbits without elevated antibody. By contrast, peripheral lymphocytes from eight of 13 rabbits with tubulointerstitial nephritis were cytotoxic against target fibroblasts in the presence of THP as compared to none of eight age matched challenged or unchallenged rabbits with normal kidneys. The presence or absence of endotoxin in vitro did not influence determinations of antibody- or lymphocyte-mediated cytotoxicity. These observations suggest that the tubulointerstitial nephritis which develops in rabbits challenged with THP is primarily the result of cell-mediated immune responses directed against THP, and does not require the presence of endotoxin in the challenge solution, or serum IgG antibodies directed against THP.

Interaction of Tamm-horsfall protein with bacterial extracts.

Crude extracts of uropathic Escherichia coli have been reported to inhibit the binding of human Tamm-Horsfall protein (THP) to homologous and heterologous anti-THP antibody in immunoassays. This phenomenon was believed to be due to immunologic cross reactivity between THP and the bacterial antigens for the same antibody. Our attempts to further purify and characterize these "cross reactive" antigens with ion exchange and molecular sieve chromatography were unsuccessful. When purified anti-THP antibody was conjugated to sepharose beads forming an immunoadsorption column capable of isolating THP and cross reactive antigens from solution, the bacterial extracts did not react with the affinity column. However, binding between THP and the bacterial extracts and between THP and whole bacteria were demonstrated. These findings suggest that the cross reactivity seen in the immunoassays is caused by the interaction between the bacterial extracts and THP, and is not representative of true immunologic cross reaction for a common antibody.

Tubulointerstitial nephritis and immunologic responses to Tamm-Horsfall protein in rabbits challenged with homologous urine or Tamm-Horsfall protein.

Tubulointerstitial nephritis was produced in 19 of 23 New Zealand White rabbits challenged i.v. with adjuvant-free homologous urine for greater than or equal to 16 wk and 11 of 14 challenged with adjuvant-free rabbit Tamm-Horsfall protein for 2 to 24 wk. Lesions were identical in the two groups of rabbits and were characterized by focal mononuclear infiltrates and microscopic scarring localized to distal nephron segments identified as the thick ascending limb of the loop of Henle. Concomitant immunoglobulin deposition was not detected despite antecedent elevations in serum IgG antibody directed against Tamm-Horsfall protein in 17 of 19 and 10 of 11 affected rabbits, respectively. Peripheral lymphocytes from affected rabbits were found to be cytotoxic and underwent blast transformation in the presence of homologous urine or Tamm-Horsfall protein in vitro. These lymphocytes were shown to produce a soluble cytotoxic product upon exposure to Tamm-Horsfall protein. Neither tubulointerstitial nephritis nor this pattern of cellular and humoral immune response to Tamm-Horsfall protein was found in two age-matched control groups: one unchallenged, and the other challenged i.v. with urine that had been selectively depleted of Tamm-Horsfall protein by 95%. It is concluded that the tubulointerstitial nephritis produced in rabbits by injection of urine or Tamm-Horsfall protein is the result of a predominately cellular immune response directed against Tamm-Horsfall protein.

MR/K hemagglutination of Providencia stuartii correlates with adherence to catheters and with persistence in catheter-associated bacteriuria.

Providencia stuartii was the most prevalent bacterial species isolated, for one year, from weekly urine specimens from 51 long-term catheterized patients. Significantly more strains causing bacteriuric episodes of long duration expressed MR/K (mannose-resistant/Klebsiella-like) hemagglutination (74%) than did those causing episodes of short duration (26%; P = .004). Isolates expressing MR/K hemagglutinin bound in higher numbers to catheter material (P = .023) than did those not expressing this hemagglutinin. Significantly more strains causing bacteriuric episodes of short duration expressed the mannose-sensitive (MS) hemagglutinin (43%) than did those causing episodes of long duration (7%; P = .014). Isolates expressing MS hemagglutinin bound significantly more 125I-labeled Tamm-Horsfall protein (THP) than did isolates not expressing this hemagglutinin (P = .0001). Our results indicate that MR/K hemagglutinin plays an important role in the ability of P. stuartii to persist and suggest that MR/K adheres to the catheter. Conversely, MS hemagglutinin binds to THP and may prevent persistence of P. stuartii in the catheterized urinary tract.