Juan Carlos Ayus

EXTRACORPOREAL IMMUNOADSORPTION: INITIAL EXPERIENCE IN HUMAN SYSTEMIC LUPUS ERYTHEMATOSUS

Extracorporeal immunoadsorption and filtration was used in treatment of a 29-year-old woman with severe lupus nephritis. For the previous 35 days single-stranded DNA (ssDNA) antibody and immune-complex levels had been continuously raised with only partial improvement on prednisone therapy. Levels of immune complexes and ssDNA antibodies were substantially reduced by extracorporeal plasma filtration and adsorption of DNA collodion charcoal. There were no major clinical complications. After perfusion, the reduction of ssDNA binding and immune complexes in serum was sustained, serum C3 became normal, and serum creatinine and proteinuria improved. 28 days after perfusion, subendothelial glomerular deposits were much reduced compared with those in a pre-perfusion biopsy specimen. Selective immunoadsorption is a promising new approach to persistent lupus nephritis refractory to drug therapy alone.

Renal papillary necrosis: An update

The clinical and diagnostic features of renal papillary necrosis (RPN) of 27 patients were studied. Diabetes mellitus was the most frequent (56%) condition associated with RPN. Analgesic abuse, sickle hemoglobinopathy and urinary tract obstruction were present in 4 patients each; in 6 of these 12 patients these conditions were present as a coexistent disease with diabetes mellitus. There was evidence of an acute or chronic infection of the urinary tract in 18 patients, as a coexistent condition with another underlying disease that itself can cause RPN in 14 patients and as the only cause of RPN in another 4. Thus, the presence of more than one diagnostic condition which might be implicated in the causation of RPN was present in 15 patients or 55% of the cases in this series. When infection was excluded, six patients or 22% of the cases had two coexisting diseases, each of which has been implicated as a cause of RPN. This observation underlines the multifactorial nature of this entity and might explain why RPN is not encountered more frequently in each of the various primary diseases with which it has been associated. The average age of the patients at the time of diagnosis was 53 years for women and 56 years for men. Only six of the patients were younger than 40 years, and three of these had sickle hemoglobinopathy. The diagnosis of RPN was based on x-ray findings in eight patients, on the histologic examination of papillary tissue in urine in one, and on autopsy findings in the rest. Papillary necrosis was bilateral in three-fourths of the cases. The clinical picture varied. Most of the patients (67%) presented with chills and fever. Flank pain and dysuria were present in 11 patients (41%). As a rule oliguria was rare and progressive uremia was uncommon. In cases diagnosed at post-mortem, the patients had succumbed to infection or to a primary severe extrarenal disorder with the possibility of RPN having been entertained clinically in only half these cases prior to autopsy.

Asymptomatic Pericardial Effusion in Uremic Patients: Effect of Long-Term Dialysis

The present study was undertaken to determine the prevalence of asymptomatic pericardial effusion in a population with end-stage renal failure just prior to the initiation of chronic dialysis, and to determine the effect of long-term dialysis on these effusions. We prospectively studied 50 uremic patients with M-mode echocardiograms prior to initiation of chronic dialysis and followed 33 of these patients after 10 months of intense dialytic therapy. Predialysis effusion was present in 18/50 (36%) patients. Only 3/50 patients had clinical evidence of pericarditis (none of these individuals had an effusion). The incidence of clinical congestion and radiological evidence of volume overload was significantly higher in the patients with an asymptomatic pericardial effusion. Of these, the effusion disappeared on improved in 6 (43%), remained unchanged in 6 (43%), and worsened in 2 (14%). No patients developed new pericardise effusions during chronic dialysis. Changes in effusion size were related to changes in body weight between dialysis treatments (r = 0.39; p less than 0.05). Our data show that asymptomatic pericardial effusions are frequent in uremic patients prior to initiation of dialysis, the etiology of asymptomatic pericardial effusions in these patients appears to be related to volume overload, only 43% of the patients improved their effusions with chronic dialysis.

Clotted arteriovenous grafts: a silent source of infection.

Infectious complications are frequent causes of morbidity and mortality among hemodialysis patients. Of particular concern to this population are infections related to the hemodialysis vascular access. The incidence of bacteremic and nonbacteremic infections is highest when the dialysis access is a central venous catheter and lowest when it is a native arteriovenous fistula (1, 2). Unfortunately the prosthetic arteriovenous graft (PAVG), which has become an acceptable alternative to the native dialysis fistula when the latter is not surgically feasible, is often plagued by infection (3, 4). The risk of PAVG infection starts at the time of surgical placement. Zibari et al. (5) reported a 30-day graft infection rate of 6% in 208 patients undergoing PAVG placement. While the incidence of postoperative PAVG infection can be reduced with prophylactic antimicrobial therapy, a persistent risk of PAVG infection is posed by the need for repetitive cannulation of the graft for dialytic purposes (6). Difficulty in cannulation of the PAVG, perigraft hematoma formation, and a break in the sterile technique further increase the likelihood of PAVG infection. The risk of PAVG infection does not end when the graft is no longer in use. Unused grafts are typically nonfunctioning with a thrombosed lumen. It is common practice to leave these clotted grafts in place, and thus many hemodialysis patients have one or more old clotted grafts in their extremities. Even though these old clotted grafts tend to be considered innocuous by most health care practitioners, it has recently been recognized that they may harbor occult bacterial infection that can lead to serious infectious complications (7). It is difficult to diagnose infection in old clotted grafts largely due to the fact that such infections tend to be silent. Clinical recognition of PAVG-related infection is easy when there is tenderness, erythema, warmth, induration, or local drainage around the graft. In patent grafts, these localizing signs of infection are typically noted at the sites of previous graft punctures or surgical manipulation. In old clotted grafts, these signs are frequently absent. Thus the diagnosis of clinically silent graft infection requires a high index of suspicion and is frequently missed (7, 8). To evaluate the prevalence and clinical relevance of silent infection in clotted grafts, we studied a series of 20 hemodialysis patients with old clotted PAVGs who presented with fever (15 patients) or fever and clinical signs of sepsis (5 patients) in whom the source of infection was not immediately localized to any organ system. Comparison was made with 21 asymptomatic hemodialysis patients with clotted PAVGs who served as control subj cts. Both febrile patients and control subjects were evaluated with indium scans and then subjected to surgical removal of the graft. Bacterial cultures of the recovered surgical material and blood were done simultaneously in all study participants. Blood cultures were positive for bacterial pathogens in 15 of the 20 febrile patients, indicating that serious illness was present. In contrast, all of the asymptomatic control subjects had negative blood cultures. Indium uptake in or around the clotted grafts was present on scanning in all 20 patients and in 15 of the control subjects. The importance of the indium scan findings was verified when purulent infected material was recovered from graft material in all 20 patients and in 13 of 15 indium scan-positive control subjects. The pathogens recovered from blood culture were identical to those cultured from the graft material in all patients, strongly indicating a causal relationship. By far the most frequent pathogen recovered from the graft material was Staphylococcus aureus; this was followed byS. epidermidisand less frequently, Escherichia coli, Serratia marcescens , and Streptococcus pneumonia . The finding of staphylococcal species as the most frequent pathogens in infected graft material is consistent with the high prevalence of staphylococcal bacteremia in emodialysis patients. Bacterial infections, particularly those caused by S. aureus , may result in life-threatening complications, especially in an immune-compromised host such as the chronic hemodialysis patient (9). Mortality from S. aureusbacteremia is about 20–25%. Late serious complications of S. aureusbacteremia may be as high as 43% and include recurrent bacteremia, endocarditis, purulent pericarditis, septic arthritis, epidural abscess, and osteomyelitis (10). These complications are more common with short (2 weeks or less) courses of antimicrobial therapy, and several authorities advocate at least 4 weeks of treatment. Such prolonged treatment will increase the cost and the likelihood of antimicrobialrelated side effects. Considering the high cost of treating bacteremic episodes and the serious complications of bacteremia, infected old clotted grafts should be recognized as a serious hazard. Indium scanning is a well-established and accepted method for localization of infection in general (11, 12). This technique requires administration of autologous indium-labeled leukocytes, and subsequent (24 h later) total body scanning with a gamma camera (7). It has been shown to be useful in vascular graft infection, with a satisfactory overall level of sensitivity and specificity (13, 14). A variety of lesions other than foci of Address correspondence to: Juan Carlos Ayus, MD, Baylor College of Medicine, 4 Brompton Ct., Houston, TX 77024. Seminars in Dialysis—Vol 13, No 1 (January–February) 2000 pp. 1–3

Brown Tumor in End-Stage Renal Disease

Figure 1. A 62-year-old man with end-stage renal disease from hypertensive nephrosclerosis that had been treated by hemodialysis for 14 years presented to the emergency room with a productive cough and fever, from which he recovered. A chest film showed right hilar prominence and pleural-based masses (arrows in Panel A), with destructive changes in the adjacent ribs on the left. The cardiac silhouette, pulmonary vasculature, and position of a LeVeen shunt placed for recurrent dialysis-associated ascites were unchanged from a previous chest film. Computed tomography of the chest showed mild mediastinal adenopathy and multiple expansile lytic lesions of the ribs .xa0.xa0.

Effect of Unilateral Nephrectomy and its Accompanying Hemodynamic Changes on Calcium and Phosphate Excretion in the Dog

Acute unilateral nephrectomy (AUN) results in a prompt increase in electrolyte excretion by the controlateral kidney in the dog (1). Humphreys and Ayus have shown that this increased electrolyte excretion by the remaining kidney consists mainly of Na+ and K+ and is accompanied by striking systemic hemodynamic changes (2). The latter consists of a drop in cardiac output; an increase in diastolic pressure and total peripheral resistance; but a constant glomerular filtration rate. These findings resemble those observed following closure of a systemic arteriovenous fistula (3, 4). Of particular interest in this regard is the fact that when an artificial arteriovenous fistula is created at the time of uninephrectomy, all of the changes seen following AUN are abolished only to reappear again when the fistula is closed. Thus, it is evident that AUN induces hemodynamic changes resembling those seen following closure of a systemic arteriovenous fistula and that the observed increase in electrolyte excretion following AUN is not the result of the removal of the kidney itself, but rather secondary to the hemodynamic changes induced by the AUN.

Renal Handling of Divalent Cations and Phosphate During Water Immersion in the Awake Dog

Previous studies in the human have failed to show any effects of head-out water immersion on serum parathyroid hormone (PTH) concentration, serum ionized calcium concentration (Ca++) or urinary phosphate excretion (UPO4V) (2). No studies have been reported on the effects of water immersion on the renal handling of phosphate and divalent cations in animals. Thus, the present experiments were designed to characterize the effects of head-out water immersion on the urinary excretion of phosphate (UPO4V), calcium (UCaV), and magnesium (UMgv) in the awake dog.