Ten years' survey of dialysis-associated tuberculosis

Renaissance of Tuberculosis in the 19905: Lessons for the Nephrologist

Objective

To call attention to the worldwide increase in tuberculosis and to review the disease in peritoneal dialysis patients.

Data Sources

Recent epidemiological publications.

Data Extraction

Epidemiological data summarized in tables and diagnostic and therapeutic recommendations reviewed.

Conclusions

Tuberculosis is on the rise worldwide. Multiple drug-resistant strains are emerging, causing therapeutic problems. The role of atypical mycobacterial infections in continuous ambulatory peritoneal dialysis (CAPD) is reviewed. Early diagnosis of cases and thorough chemotherapy are advocated.

Tuberculosis in haemodialysis patients: A single centre experience

We prospectively followed-up new patients of tuberculosis while on maintenance hemodialysis at a State Government-run tertiary care institute. Between 2000 and 2010, 1237 new patients were initiated on maintainence hemodialysis. The number of patients diagnosed with tuberculosis after initiation of hemodialysis was 131 (10.5% of 1237). The age was 46.4 ± 10.4 (range 8-85) years and there were 90 (68.7%) males. The number of patients diagnosed with tuberculosis on the basis of organ involvement were: Pulmonary-60, pleural effusion-31, lymph node-21, meningitis-8, pericardial effusion-7, peritoneum-2, latent tuberculosis-2. The incidence of tuberculosis in hemodialysis was found to be 105.9 per 1000 patient years. Male gender, diabetes mellitus, past history of tuberculosis, mining as an occupation, low serum albumin, and duration of hemodialysis more than 24 months, and unemployment were found to be significant risk-factors on univariate analysis.

Central nervous system infections among individuals with and without end-stage renal disease

OBJECTIVES

Individuals with immunodeficiencies are at increased risk of central nervous system (CNS) infections. Data are limited on the risk of CNS infections in patients with end-stage renal disease (ESRD).

METHODS

We conducted a population-based, nationwide cohort study among ESRD patients from 1990 to 2009. Data on ESRD patients was obtained from the Danish Nephrology Registry database. Each ESRD patient was matched with up to 19 population controls. Using the National Hospital Registry we identified hospital admissions with CNS infection as primary discharge diagnosis.

RESULTS

The study included 13,374 ESRD patients and 245,397 population controls, providing 67,012 person-years and 2,237,237 person-years of observation, respectively. The overall incidence rate (IR) of first-time hospitalisation for CNS infection was 149 per 100,000 person-years (95%-confidence interval [CI], 123-181) among ESRD patients. The IR of CNS infection was comparable among dialysis and transplant patients. Compared to population controls, the incidence rate-ratio of first CNS infection was 5.58 (95%-CI, 4.47-6.91) for ESRD patients in general. The 30-day mortality following hospitalisation for CNS infections was 21% (95%-CI, 14-30) among ESRD patients and 13% (95%-CI, 10-16) among population controls.

CONCLUSIONS

ESRD patients have considerable excess risk and mortality from CNS infections, which is of great clinical and public health concern.

Mycobacterium tuberculous peritonitis in CAPD patients: a report of 11 patients and review of literature

BACKGROUND

The aims of the present report were to document our experience of the prevalence of tuberculous peritonitis in continuous ambulatory peritoneal dialysis (CAPD) patients, mode of presentation, diagnosis and outcome and to discuss the current published data about catheter removal.

METHODS

A retrospective study of CAPD patients with tuberculous peritonitis was done. A minimum of three specimens of peritoneal fluid were examined for acid-fast bacilli smears. The BACTEC 9000 Blood Culture Series of instruments were used for the culture of Mycobacterium tuberculosis. After 2005, patients were treated with anti-tuberculous treatment, and catheter retention was started in our patients.

RESULTS

There were eleven patients (2.6 %) with tuberculous peritonitis among 414 CAPD patients. M. tuberculosis accounted for 4.47 % of all peritonitis episodes. The incidence of tuberculous peritonitis was 1/794 months. There were eight males and three females. The mean age was 49 years. Intestinal obstruction was reported in two patients, and two patients were treated for antecedent peritonitis. One of them had a simultaneous fungal peritonitis. One patient each developed a peritoneo-cutaneous fistula and ultrafiltration failure. Three were successfully treated without the removal of catheter.

CONCLUSION

Based on the analysis of all published reports of tuberculous peritonitis, there was no significant difference in patient survival between patients in whom CAPD catheter was removed or retained. Tuberculous peritonitis should be considered in patients with neutrophilic 'sterile' peritonitis with no response to antibacterial medications, predominance of lymphocytic peritonitis and in bacterial peritonitis not responding to antibiotics. After an early diagnosis, with close monitoring, an effort to retain the catheter after 5 days of anti-tuberculous therapy may be attempted.

Value of the tuberculin skin testing and of an interferon-gamma release assay in haemodialysis patients after exposure to M. tuberculosis

Background

Patients with end-stage renal disease (ESRD) and Mycobacterium tuberculosis infection pose a high risk of developing active TB disease. It is therefore important to detect latent TB infection (LTBI) to be able to offer treatment and prevent progression to TB disease. We assessed the value of the tuberculin skin test (TST) and of an interferon-gamma release assay (Quantiferon®-TB Gold in-Tube, QFT) for diagnosing LTBI in ESRD patients, after prolonged exposure to a highly contagious TB case in a haemodialysis unit. As a high number of patients presented erythema without induration in the TST response, this type of reaction was also analysed.

Method

The TST and QFT were simultaneously performed twelve weeks after the last possible exposure to a bacilliferous TB patient. If the first TST (TST-1) was negative, a second TST (TST-2) was performed 15 days later to detect a booster response. A comparison was made between the TST responses (including those cases with erythema without induration) and those for the QFT. The correlation with risk of infection and the concordance between tests were both analysed.

Results

A total of 52 patients fulfilled the inclusion criteria. Overall, 11 patients (21.2%) had a positive TST response: 3 for TST-1 and 8 for TST-2, and 18 patients (34.6%) showed a positive QFT response (p = 0.065). Erythema without induration was found in 3 patients at TST-1 and in a further 9 patients at TST-2. The three patients with erythema without induration in TST-1 had a positive TST-2 response. Concordance between TST and QFT was weak for TST-1 (κ = 0.21); it was moderate for overall TST (κ = 0.49); and it was strong if both induration and erythema (κ = 0.67) were considered.

Conclusions

In patients with ESRD, erythema without induration in the TST response could potentially be an indicator of M. tuberculosis infection. The QFT shows better accuracy for LTBI diagnosis than the TST.

Tuberculin skin test for the diagnosis of latent tuberculosis during renal replacement therapy in an endemic area: A single center study

Patients on renal replacement therapy (RRT) are at-risk for developing tuberculosis (TB). There is limited information on tuberculin skin test (TST) and its predictability for development of TB. In this prospective cohort study, patients taken for RRT were included. Patients with active TB were excluded. TST was done with 5-tuberculin unit. In addition to TST, age, sex, diabetes as basic disease, number of dialysis and blood transfusion (BT), pre-transplant TB, hepatitis B and C infections and type of immunosuppression were correlated with the development of TB. Of the 200 patients included, TST was positive in 21 and negative in 179. In TST negative group, 20 (11.1%) and in TST positive group 5 (23.8%) patients developed TB. TB free survival in two groups was similar (P = 0.08). On multivariate Cox regression analysis, hazard of development of TB by TST was 2.7 [P = 0.11, confidence interval (CI) 0.78-9.7]. There was no difference between TST non-responsive and TST negative patients (P = 0.18). Sensitivity and specificity of TST for predicting TB was only 20 and 9%, respectively. Our study shows that TST in patients on dialysis is an insensitive and nonspecific test to predict development of active TB.

Tuberculosis of the breast in a patient undergoing hemodialysis

A 61-year-old female patient with diabetes undergoing maintenance hemodialysis presented with fever and swelling of her right breast. She had been unresponsive to nonspecific antimicrobial therapies. Breast tuberculosis and tuberculous lymphadenitis were diagnosed by excision biopsy and tissue culture. A combination antituberculous treatment including isoniazid, rifampin, ethambutol, and pyrazinamide was started and resulted in the remission of the lesions by the end of the first month of therapy.

Extrapulmonary Tuberculosis in Chronic Hemodialysis Patients

BACKGROUND

The incidence of extrapulmonary tuberculosis is higher in dialysis than general population. The aim of the study was to characterize clinical picture in dialysis patients, who developed extrapulmonary tuberculosis.

METHODS

We retrospectively investigated the hemodialysis patients with extrapulmonary tuberculosis. 2208 hemodialysis patients were reviewed for extrapulmonary tuberculosis from October 1986 to January 2001.

RESULTS

Seventeen patients (10 male, 7 female) were enrolled. The mean age was 57.4 +/- 12.4 years. The sites for extrapulmonary tuberculosis were peritoneum (35.3%, 6/17), cervical lymph node (17.6%. 3/17), bone marrow (5.9%, 1/17), spine (5.9%, 1/17), knee (5.9%, 1/17), brain (5.9%, 1/17), pericardium (5.9%, 1/17), cutaneous tissue (5.9%, 1/17) and genitourinary system (5.9%, 1/17). Fourteen of 15 tissue-biopsy specimens from suspicious sites revealed granulomatous inflammation. There were low yield in mycobacteria culture (11.1%, 1/9) and PCR (33.3%, 2/6). Three patients died during the treatment of the disease.

CONCLUSION

Extrapulmonary tuberculosis constitutes a major part of tuberculosis in dialysis patients. Tissue biopsy with invasive procedures, such as laparoscopy or laparotomy, may be necessary if clinical presentations are suspicious.

Active tuberculosis in patients undergoing hemodialysis for end-stage renal disease: a 9-year retrospective analysis in a single center

OBJECTIVE

Tuberculosis (TB) in patients undergoing hemodialysis (HD) for end-stage renal disease (ESRD) is commonly thought to be associated with a very poor prognosis. Moreover, it is difficult to diagnose. This report was designed to describe this condition and to determine the mortality rate and risk factors associated with mortality. In addition, the study evaluated the usefulness of QuantiFERON TB-2G((R)) (QFT-2G).

METHODS

Retrospective study

PATIENTS

Patients with confirmed TB admitted between January 2001 and May 2009 were retrospectively identified and enrolled. The clinical, radiological, and bacteriological data at the time of admission were recorded. A multivariate analysis was performed to identify the predictive factors for mortality.

RESULTS

A total 19 TB patients (6 females; median age, 73 years) were included. TB occurred in most cases within 1.3 years from the initiation of dialysis. Most patients presented with fever (84.2%) and extrapulmonary TB (57.9%). The mortality rate within 24 weeks of the initiation of TB treatment was 36.8%. The factors associated with mortality were: a short duration of dialysis (HR 8.86, 95% CI 1.03-75.7, p=0.04), and underweight (HR 10.88, 95% CI 1.28-92.6, p=0.02). The sensitivity of QFT-2G, acid-fast smear, and polymerase chain reaction was 50, 80, and 88.2% respectively.

CONCLUSION

These data indicate a high incidence of TB in the early stages of HD and a high mortality rate among these patients. The clinical utility of QFT-2G was found to be limited. Hypoalbuminemia might therefore be related to either indeterminate or negative results of QFT-2G.

Male-female differences in the risk of tuberculosis in dialysis patients

AIM

To define any gender-related differences in the prevalence and risk for tuberculosis (TB) in hemodialysis (HD) patients.

METHODS

All active TB cases were recorded during a 36-month follow-up of 272 (193 male and 79 female) HD patients. Entering the study, HD patients were tested with tuberculin and 2,4-dinitrochlorobenzene, and a cell-mediated immunity (CMI) index was estimated. Relative risks (RR) for TB were calculated considering subjects from the background general population as a reference group. The independent effect of age, BMI and tuberculin sensitivity was determined using Cox's proportional hazard model.

RESULTS

Female HD patients presented significantly lower CMI indices and rates of positive Mantoux tests, but higher rates of DM, as compared to males. The male:female ratio in TB for the general and HD patients population was 1.8 and 0.6, respectively. There was a significantly lower TB prevalence in male as compared to female HD patients (7.7% vs. 11.3%), and a subsequent female predominance in risk for TB in those HD patients aged <49 and 50-69 years (M:F adjusted relative risk 0.67 and 0.53) was recorded.

CONCLUSIONS

In contrast to the general population, there is a female predominance among dialysis TB patients younger than 70 years associated with the coexistence of DM. Female gender should always be considered as a risk factor when evaluating diabetic HD patients for active TB.

Interferon-gamma release assays for diagnosing mycobacterium tuberculosis infection in renal dialysis patients

BACKGROUND AND OBJECTIVES

End-stage renal disease (ESRD) patients are at high risk for tuberculosis (TB). IFN-gamma release assays that assess immune responses to specific TB antigens offer potential advantages over tuberculin skin testing (TST) in screening such patients for Mycobacterium tuberculosis infection. This study sought to determine whether IFN-gamma release assay results are more closely associated with recent TB exposure than TST results.

DESIGN, SETTING, PARTICIPANTS, AND MEASURES

Prospective cohort investigation of patients at a hemodialysis center with a smear-positive case of TB. Patients without a history of TB underwent initial and repeat testing with TST, and with the IFN-gamma assays QuantiFERON-TB Gold (QFT-G) and ELISPOT test. Outcome measures included the prevalence of positive test results, identification of factors associated with positive results, and test result discordance.

RESULTS

A total of 100 (47% foreign born; median age, 55 yr; age range, 18 to 83 yr) of 124 eligible patients were enrolled. Twenty-six persons had positive TST results, 21 had positive QFT-G results, and 27 had positive ELISPOT results. Patients with TB case contact were likely to have a positive QFT-G result (P = 0.02) and ELISPOT results (P = 0.04), whereas TB case contact was not associated with positive TST results (P = 0.7). Positive TST results were associated with foreign birth (P = 0.04) and having had a TST in the previous year (P = 0.04).

CONCLUSIONS

Positive IFN-gamma assay results were more closely associated with recent TB exposure than were positive TST results. QFT-G and ELISPOT might offer a better method for detecting TB infection in ESRD patients.

Tuberculosis in end-stage renal disease patients on hemodialysis

Recent studies have shown that there is an increase in the incidence of mycobacterium tuberculosis (MBT). This is more prevalent among immune compromised patients (those on dialysis) and recipients of organ transplants. Furthermore, extra-pulmonary presentation appears to be more common and difficult to diagnose. We aimed in this study to assess and evaluate the presentation of MBT in a retrospective study conducted among 256 hemodialysis (HD) patients where 18 of them were diagnosed and managed for tuberculosis over a 10-year period between 1990 and 2000. The mean age of the patients was 38 years (21-75 years). The mean interval between the onset of HD and the time of diagnosis was about 24 months (1-120 months). The diagnosis of tuberculosis was made either by isolation of acid-fast bacilli (AFB), the typical caseating granuloma on biopsy, or by recovery of tubercle bacilli from the culture of the biopsy material. Extra-pulmonary tuberculosis was more common (77.8%) than pulmonary tuberculosis (22.2%). The various extra-pulmonary tuberculosis sites noted were cervical lymphadenitis (16.7%), gastrointestinal (16.7%), genitourinary (11.1%), peritonitis (11.1%), pleural effusion (5.6%), pericardial effusion (5.6%), miliary tuberculosis (5.6%), and pyrexia of unknown origin (5.6%). None of the patients with extra-pulmonary tuberculosis had evidence of pulmonary tuberculosis. The atypical presentation with insidious onset was quite common. Anergy to tuberculin skin test was noticed in 56% of cases. All of our patients received modified antituberculosis treatment for 1 year with adequate response, and without undue side effects. We conclude that a high index of suspicion is required especially in the diagnosis of extra-pulmonary tuberculosis, and when there is a high percentage of anergy to tuberculin skin test. Tissue biopsy both for characteristic histology and demonstration of MTB, either by staining or culture, remains the main criteria for the diagnosis of extra-pulmonary tuberculosis.

Tuberculosis in dialysis patients, single centre experience from an endemic area

Because of immunity defect, patients with end-stage renal disease are at increased risk of developing infections, tuberculosis (TB) in particular. The incidence of TB is higher in dialysis patients than in general population. We retrospectively reviewed the charts of dialysis patients with TB in our facility. A total of 287 dialysis patients (153 male, 134 female, 223 haemodialysis (HD), 64 continuous ambulatory peritoneal dialysis (CAPD) patients, mean age 46 +/- 15) were reviewed from October 1997 to January 2002. TB developed in 30 patients (17 male, 13 female, 24 HD and six CAPD). Thirteen patients with TB presented with fever of unknown origin (FUO) and four of them subsequently developed military lesions on chest X-ray. Nine patients had pulmonary TB (four with pleural effusions), five patients had TB lymphadenits, two patients had TB peritonitis and one patient had vertebral TB. TB was presented mostly as FUO among dialysis patients in a region under poor socio-economic conditions. In such areas with endemic TB, dialysis patients who present with FUO should be carefully evaluated for the presence of TB, and test therapy for TB should be performed in otherwise unexplained FUO.

Ethambutol-induced optic neuritis in patients with end stage renal disease on hemodialysis: two case reports and literature review

Ethambutol, a synthetic bacteriostatic agent, is a first line agent against Mycobacterium tuberculosis. Although optic neuritis is the most serious adverse effect of ethambutol, most cases in the literature are reversible. Renal failure prolongs the half-life of ethambutol and increases the risk of ethambutol-induced optic neuritis. We present two patients with end stage renal disease (ESRD), who were on maintenance dialysis and suffering ethambutol-induced optic neuritis. The first woman had been suffering ESRD on hemodialysis for 2 years. After tuberculosis was diagnosed, she was prescribed three-combined anti-tuberculosis medications, including ethambutol 800 mg/day. Bilateral blurred vision suddenly occurred 4 months after the start of treatment, and she became totally blind despite discontinuing ethambutol. The second woman had been on hemodialysis for 5 months. Tuberculosis was diagnosed by lung biopsy. After 3 weeks of three-combined anti-tuberculosis medications including ethambutol (1,200 mg/day), reduced visual acuity and color vision defects occurred. One year after the discontinuation of ethambutol, visual acuity remained little improved. Physicians should be aware of ethambutol-induced optic neuritis and ethambutol should be used cautiously in patients with renal failure.

Tuberculosis and chronic renal disease

There is an increased risk (6.9- to 52.5-fold) of tuberculosis (TB) in patients with chronic renal failure and on dialysis as compared to the general population. The symptomatology in renal patients is often insidious and nonspecific, mimicking uremic symptoms, whereas the localization is often extrapulmonary (most frequently tuberculous lymphadenitis and peritonitis). Tuberculous peritonitis makes up a large part (37%) of the total number of TB cases in continuous ambulatory peritoneal dialysis (CAPD) patients. The prognosis is very much dependent on early diagnosis and treatment. Renal physicians should be aware of the unusual presentation and localization, and include TB in the differential diagnosis of any patient having nonspecific symptoms like anorexia, fever, and weight loss. All efforts should then be made (including invasive investigations) to reach an early diagnosis, a major determinant of the outcome. However, if this is not possible or the result is negative and the diagnosis remains strongly suspected, an empirical trial with anti-TB medication is justified, especially in endemic areas. In view of the increased prevalence of the disease in the dialysis population, TB prophylaxis is recommended in those patients with a positive tuberculin (Mantoux) skin test and radiographs suggestive of old TB.

Survival of patients with pulmonary tuberculosis: clinical and molecular epidemiologic factors

Using restriction fragment-length polymorphism data, we conducted a retrospective cohort study of 139 adult patients with pulmonary tuberculosis to investigate the clinical impact of Mycobacterium tuberculosis infection with a clustered isolate. The cumulative all-cause mortality rate during treatment was 21%. Patients with clustered DNA fingerprint patterns had a reduced risk of death, compared with patients with unique patterns (hazard ratio [HR], 0.5; 95% confidence interval [CI], 0.2-1.1), but this finding was confounded by age (adjusted HR, 0.8; 95% CI, 0.4-1.8). After adjustment for age, the strongest predictors of death were such underlying illnesses as diabetes mellitus, renal failure, chronic obstructive pulmonary disease, and human immunodeficiency virus infection. We conclude that comorbidity and immunosuppression are important predictors of survival for patients with pulmonary tuberculosis in an inner-city cohort. Recently transmitted infection, as determined by use of DNA fingerprinting to classify patients' isolates as being either clustered or unique, was not independently associated with death.

Infections in patients undergoing peritoneal dialysis

Considering experience acquired in the past years, it seems as though physicians have reached a plateau in the frequency of peritonitis. A peritonitis rate of 1 every 2 patient years may be acceptable. Further reduction of this peritonitis rate will require inordinately large efforts on all fronts. One will have to consider what are the acceptable costs and risks of peritonitis in patients on peritoneal dialysis. New developments in catheter technology, improved connections, better understanding of patient selection and training programs, improved diagnostic and therapeutic methods in the management of peritonitis, and understanding of the infectious and immune processes are eagerly awaited developments.

Mycobacterial infections in patients with chronic renal disease

In this article, the authors have provided a comprehensive review of TB and MOTT infections in patients on renal dialysis and receiving kidney transplants. Because most published series are small retrospective studies or case reports, there are several uncertainties still involved in the diagnosis and treatment of such patients. Unanswered questions include selection of optimal dosage and duration of therapeutic agents; the best tests for screening and diagnosis, especially in high prevalence areas; and the best management of MOTT infections because of unavailability of highly effective therapy.

Extrapulmonary tuberculosis in patients with end stage renal disease--two case reports and a brief review

Chronic renal failure is a risk factor for tuberculosis. In the past five years we have identified two cases of tuberculosis in our dialysis population. The first patient, showed chronic failure to thrive on hemodialysis. An enlarged cervical lymph node was biopsied and, although no acid fast bacilli (AFB) were seen, a culture grew Mycobacterium tuberculosum (TB). Her chest X-ray did not show evidence of past tuberculosis. The second patient was a long time smoker who presented with an enlarged cervical node, which was biopsied. AFB were not seen on her biopsy, but her culture grew TB. Extrapulmonary TB is common in patients with ESRD, and lymph node involvement is the most common extrapulmonary presentation. Screening with the purified protein derivative (PPD) is not helpful in ESRD patients, since defects in cell mediated immunity are common. A high index of suspicion for TB is warranted in patients with ESRD.

Peritoneal tuberculosis in patients receiving continuous ambulatory peritoneal dialysis

BACKGROUND

Patients with chronic renal failure have an increased risk of tuberculosis (TB). This occurs with much higher frequency within the first 12 months of initiating dialysis and is usually extrapulmonary in nature. Patients most at risk are those from susceptible ethnic groups, especially the Indian subcontinent. Peritoneal TB, otherwise relatively uncommon, has emerged as an important form of TB in patients undergoing continuous ambulatory peritoneal dialysis (CAPD).

METHODS

All cases of peritoneal TB occurring at our institution in patients undergoing CAPD over a 13 year period were identified and analysed.

RESULTS

Eight cases were identified, of which seven were non-Caucasian. These patients' characteristics and outcomes are presented. All were undergoing CAPD and most developed TB within 12 months of initiating dialysis. All presented with fever, but symptoms and signs were indistinguishable from bacterial peritonitis. Six were culture-positive, mainly from peritoneal dialysis fluid, but only two cases proved smear-positive. All were treated with standard anti-tuberculous chemotherapy. Three went on to permanent haemodialysis as a result of peritonitis and three have died, one of these as a result of TB.

CONCLUSIONS

Peritoneal TB, whilst otherwise relatively uncommon, is an important manifestation of TB in CAPD patients and usually develops soon after commencing dialysis. The reasons for this are unknown and require further research.

Tuberculosis as a cause of recurrent fever of unknown origin

Recurrent fever constitutes a diagnostic challenge for clinicians, due mainly to the intermittent nature of the fever that results in incomplete investigations. We describe three patients with recurrent fever thought to be due to tuberculosis, and review the 14 previously reported cases who fulfil the criteria of recurrent fever for at least 1 month's duration. The median duration of symptoms before diagnosis was 5 months, and the duration of the febrile bouts ranged from a few hours to 1 week. The most common complaints were constitutional symptoms and abdominal pain, and most patients had significant underlying conditions. The mortality rate was 31%, and was limited to the earlier cases. Routine laboratory studies are not very helpful for the diagnosis of this condition, and chest radiographs showed some alteration in half the cases at the time of diagnosis, although in some cases represented old, healed lesions. PPD testing was positive in most cases, particularly in those without underlying conditions. Empirical antituberculous therapy should be considered in cases of recurrent fever, especially in areas of high prevalence or in patients with predisposing conditions.

Tuberculous peritonitis in patients undergoing continuous ambulatory peritoneal dialysis: case report and review

A case of tuberculous peritonitis complicating continuous ambulatory peritoneal dialysis (CAPD) in a 37-year-old man who presented with fever, abdominal pain, and a malfunctioning Tenckhoff catheter is reported. The patient was initially treated for presumed bacterial peritonitis but remained febrile and had persistent abdominal pain and peritoneal fluid pleocytosis, despite broad-spectrum antibiotic therapy. Mycobacterium tuberculosis was isolated in a culture of peritoneal fluid, and the patient responded promptly to antituberculous therapy. More than 50 cases of tuberculous peritonitis complicating CAPD that have been reported in the English-language literature since the initial case was reported in 1980 are reviewed. The most common symptoms are fever (78%), abdominal pain (92%), and cloudy dialysate (90%); 76% of cases had a predominance of polymorphonuclear cells in peritoneal fluid. A smear for acid-fast bacilli or a culture was positive in 73% of cases. The peritoneal dialysis catheter was removed in 53% of cases, although this was rarely considered necessary for cure of tuberculosis. The attributable mortality rate is 15%, with the most significant factor being treatment delay (mean time from presentation to initiation of treatment, 6.74 weeks). We conclude that tuberculosis is an important diagnostic consideration for CAPD patients with peritonitis that is refractory to broad-spectrum antibiotics.

Reactivation of tuberculosis after conversion from azathioprine to mycophenolate mofetil 16 years after renal transplantation

The incidence of tuberculosis among transplant recipients is greater than in the general population. Mycophenolate mofetil (MMF) is a potent immunosuppressive agent that has become part of most standard immunosuppressive protocols after renal transplantation. We have recently shown that conversion from azathioprine (AZA) to MMF in patients with chronic allograft dysfunction may be beneficial. Here, we report a patient with a history of pulmonary tuberculosis during his childhood. This patient was converted from AZA to MMF therapy 16 years after allogenic renal transplantation because of chronic allograft dysfunction. Two months later, he developed axillary lymph node tuberculosis caused by Mycobacterium tuberculosis. Because he denied contact with infectious persons, we diagnosed reactivation of old dormant tuberculosis. After surgical extirpation, quadruple antituberculous therapy was administered for 3 months (isoniazid, rifampicin, ethambutol, and pyrazinamide), followed by dual therapy for 3 months (isoniazid and rifampicin), and monotherapy for another 3 months (isoniazid). In the follow-up period, he remained asymptomatic with stable graft function. We conclude that MMF therapy in renal allograft recipients may cause reactivation of old dormant tuberculosis, even in the very late posttransplantation period. In these patients, close monitoring and isoniazid prophylaxis may be useful.

The effect of hemodialysis on cycloserine, ethionamide, para-aminosalicylate, and clofazimine

STUDY OBJECTIVES

Determine hemodialysis clearances of the second-line antitubercular drugs cycloserine (CS), ethionamide (ETA), para-aminosalicylate (PAS), and clofazimine (CFZ).

DESIGN

Open-label, pharmacokinetic study

SETTING

Outpatient long-term hemodialysis unit

PARTICIPANTS

Eight long-term hemodialysis patients Interventions: Single oral doses of CS, 500 mg, ETA, 500 mg, PAS, 4,000 mg, and CFZ, 200 mg, were given 2 h (4 h for PAS) prior to hemodialysis (median blood flow rate, 400 mL/min; median dialysate flow rate, 600 mL/min; median hemodialysis time, 3.5 h).

MEASUREMENTS AND RESULTS

Arterial and venous serum samples were collected at the beginning and end of hemodialysis, and hourly during hemodialysis. Dialysate fluid was collected for the duration of hemodialysis. All samples were assayed for drug concentrations using validated high-performance liquid chromatography (for ETA and PAS), capillary electrophoresis (for CS), and colorimetry (for CFZ). Dialysate samples were analyzed for acetyl-PAS. Median recoveries of drug in dialysate were 56% (CS), 2.1% (ETA), 6.3% (PAS parent compound), and 0% (CFZ) of the doses administered. Acetyl-PAS was dialyzed to a greater extent than its parent compound. Median hemodialysis clearances calculated by dividing the amount recovered in dialysate by the serum area under the curve during dialysis were 189 (CS), 58 (ETA), 206 (PAS), and 0 (CFZ) mL/min.

CONCLUSIONS

ETA, CFZ, and PAS were not significantly dialyzed. CS is significantly removed by hemodialysis and should be dosed after hemodialysis.

The effect of hemodialysis on isoniazid, rifampin, pyrazinamide, and ethambutol

This study examines hemodialysis clearances of isoniazid (INH), rifampin (RIF), pyrazinamide (PZA), and ethambutol (EMB). Seven chronic hemodialysis patients were studied. Six were given single oral doses (INH 300 mg, RIF 600 mg, PZA 1000 mg, and EMB 25 mg/kg) 2 h before hemodialysis (Cobe Centrysystem 3 hemodialysis machine; Fresenius F80B dialyzer; median blood flow rate 400 ml/min; dialysate flow rate 600 ml/min; median hemodialysis time 3.5 h). The seventh subject, being treated for tuberculosis (TB), was studied with his usual regimen. Arterial and venous serum samples were collected at the beginning and end of hemodialysis, and hourly during hemodialysis. Dialysate was collected for the duration of hemodialysis. All samples were assayed for drug concentrations using high-performance liquid chromatography (HPLC) (INH, RIF) and gas chromatography/mass spectrometry (GC/MS) (PZA, EMB) methods. Median recoveries of drug in dialysate were 9% (INH), 4% (RIF), 45% (PZA), and 2% (EMB) of the doses administered. Median hemodialysis clearances calculated by dividing the amount recovered in dialysate by the serum area under the curve during hemodialysis were 124 (INH), 40 (RIF), 270 (PZA), and 46 (EMB) ml/min. INH, RIF, and EMB were not significantly removed by hemodialysis. PZA is significantly dialyzed and should be dosed after hemodialysis.

Unusual presentations of extrapulmonary tuberculosis in three hemodialysis patients

We report three patients who received maintenance hemodialysis and suffered from extrapulmonary tuberculosis with unusual presentations. The first patient presented with fever of unknown origin. All studies showed negative findings except high erythrocyte sedimentation rate and high value of C-reactive protein. He failed to response to broad-spectrum antibiotics but showed a complete response to antituberculosis therapy. The second patient presented with right supraclavicular lymphadenopathy and weakness. Lymph node biopsy revealed caseating granuloma with positive acid-fast bacilli. The third patient presented with tumor mass of left sternoclavicular joint for which malignancy was suspected initially. Ultimately, tuberculosis was documented by histopathologic studies showing caseating granuloma. All three patients had normal findings of chest x-ray and did not have previous history of tuberculosis and diabetes mellitus.

Tuberculous osteomyelitis: an unusual case of tuberculous infection in a child undergoing continuous ambulatory peritoneal dialysis

A 10-year-old boy on continuous ambulatory peritoneal dialysis had low-grade fever, pain and tenderness of the right shoulder; he had no history of infection or exposure to tuberculosis. The underlying granulomatous infection was diagnosed by histological examination of bone and the polymerase chain reaction with primer sequences specific for Mycobacterium tuberculosis. Special stains and cultures were negative. The initiation of antituberculous therapy was followed by a sharp improvement in the fever, malaise and shoulder pain within several weeks. Extrapulmonary tuberculosis is common in patients with chronic renal failure on dialysis and the diagnosis is difficult. The clinician must recognise the high and early mortality rate in order to initiate an aggressive diagnostic approach and early therapy.

Mycobacterial infection is an important infective complication in British Asian dialysis patients

To define the extent and nature of mycobacterial infection in patients on an adult dialysis unit whose catchment population contains a large proportion of non-Caucasian subjects, a retrospective survey of all new patients accepted onto our maintenance dialysis programme between January 1987 and December 1989 was carried out. Twenty-six Asian, 13 Afro-Caribbean, two Oriental and 170 Caucasian patients were accepted onto the dialysis programme in the three-year recruitment period. Eight of the 26 Asian patients, but none of the others, had developed mycobacterial infection by the end of December 1990. One patient had a cerebral tuberculoma with miliary mottling on chest X-ray, one pulmonary tuberculosis, one tuberculous adenitis and 5 tuberculous peritonitis (four due to Mycobacterium tuberculosis and one Mycobacterium kansasii). All the patients had been living in the UK for an average of 15 (range 6-24) years, with no known recent exposure to tuberculosis. Five patients are now alive and well, one developed malabsorption following M. kansasii peritonitis, but two with tuberculous peritonitis died before treatment could be instituted. Mycobacterial infections were associated with a high level of mortality and morbidity. No Asian patient developed mycobacterial infection during post-transplant immunosuppressive therapy in the study period, probably because of the routine anti-tuberculous chemoprophylaxis employed in this group of patients. The diagnosis of mycobacterial infection should be suspected when an Asian dialysis patient develops a pyrexia of unknown origin. It is likely, though not proven, that anti-tuberculous chemoprophylaxis might reduce this high incidence of tuberculous infection in Asian dialysis patients.

Tuberculosis in patients on maintenance dialysis

Tuberculosis was diagnosed in 7 of 25 patients who underwent maintenance dialysis, an incidence of 28%, five of the patients (71.4%) were females. It was characterized clinically by an insidious onset, the main symptoms being anorexia, loss of weight, and a low-grade fever. There were extrapulmonary presentations in 5 of the 7 patients. Tuberculous lymphadenitis predominated in the extrapulmonary form (5 patients, 57%). Pulmonary tuberculosis was noted in one patient (21.7%) who presented with pleural effusion. One patient had a tuberculous abscess involving the seventh rib. Total white cell and differential count was normal in all patients examined. All of the patients were diagnosed at the beginning of their dialysis treatment and early therapy was initiated. Treatment led to recovery in all seven patients, and no recurrence was observed. Despite earlier reports of high mortality in similar groups of patients, it is suggested that awareness of the increased incidence of tuberculosis in dialysis patients, together with recognition of its unusual presentation and consequent early diagnosis, results in a good prognosis.

High incidence of tuberculosis in renal patients

The incidence and clinical characteristics of mycobacteriosis in patients attending the nephrology department in a hospital in Madrid, Spain, during a 30-month period were analysed retrospectively. Twenty-two new cases of tuberculosis were detected among 525 patients studied. No cases of clinically significant atypical mycobacteriosis were found. The estimated overall annual incidence of tuberculosis in these patients was 259 cases per 100,000 population, which is much higher than the national annual incidence in Spain of 35 per 100,000 population. Most cases were asymptomatic when diagnosed and extrapulmonary involvement was the rule (86% of all patients). In areas with a high prevalence of tuberculosis, renal patients in high-risk groups (renal transplant recipients, haemodialysis and CAPD patients) should be examined periodically to exclude silent infection.

Tuberculosis in patients with chronic renal failure

Nine cases of tuberculosis (TB) were diagnosed among 800 uremic patients, followed-up during 11 years, a prevalence of 1125%, 2.5 times higher than that in the general population. Six patients (66.7%) had lymph node involvement (4 cervical and 2 mediastinal). Three patients (33.3%) had pulmonary involvement (2 pleuro-pulmonary and 1 bilateral apical pulmonary). Eight patients were undergoing dialysis and 1 was pre-dialytic. The duration of dialysis ranged from 1 to 60 months. Three patients had previously received immunosuppressive drugs for unsuccessful renal transplantation. Daily fever was present in all but one patient; he was asymptomatic and TB was suspected after routine chest radiography. Biopsy was the diagnostic procedure in 7 patients (77.8%), four by direct cervical lymph node biopsy, 2 by mediastinal, performed by mediastinoscopy and 1 by pleural biopsy. In 2 other patients TB was confirmed by the presence of tubercle bacilli; in sputum (1 patient) and in a bronchial flushing specimen (the other patient). Triple therapy was used in all patients (isoniazid and ethambutol in all), plus rifampicin in 8 and streptomycin in 1. One patient had jaundice and another had optical neuritis. Five patients were cured. The other four died during treatment of causes unrelated to TB or its treatment.

Tuberculosis and renal transplantation--observations from an endemic area of tuberculosis

Ninety-five renal transplant recipients from an endemic area of tuberculosis were investigated to find out the prevalence and course of tuberculosis in pre- and post-transplant periods. Eleven patients had tuberculosis in the pre-transplant period - pulmonary (2), pleural (2), miliary (1), abdominal (2), lymph node (5) and pericardial (1). They were transplanted after antituberculous therapy of 3 to 6 months with satisfactory results. The anti-tuberculous treatment was usually continued for 2 years. Only one of the above 11 patients had evidence of tuberculosis in the post-transplant period. Nine patients developed tuberculosis for the first time in the post-transplant period - pulmonary (4), pleural (1), miliary (1), lymph node (4) and pericardial (1). There was no mortality due to tuberculosis. Thorough search for tuberculosis is mandatory both during pre-transplant assessment and post-transplant follow-up in areas of endemic tuberculosis.

Tuberculous peritonitis in chronic renal failure managed by continuous ambulatory peritoneal dialysis

Continuous ambulatory peritoneal dialysis (CAPD) is being increasingly used to treat chronic renal failure in New Zealand. Peritonitis due in particular to gram positive organisms remains the major complication. Three of 92 CAPD patients trained in the Wellington Renal Unit had tuberculous peritonitis, a previously rarely reported complication. Gram positive or Gram negative bacterial infections preceded or followed isolation of Mycobacterium tuberculosis. Differential peritoneal fluid leucocyte counts were not predictive of tuberculous infection and total leucocyte counts remained elevated in tuberculous patients treated for other concurrent bacterial peritonitides. Systemic toxicity was not encountered in these patients, symptoms being confined almost entirely to the peritoneum. CAPD was continued during treatment with anti-tuberculous therapy, in all three patients. However, peritoneal pain on dialysis fluid in-flow necessitated temporary hemodialysis management in two. Anti-tuberculous chemoprophylaxis may be prudent in the at-risk Polynesian patient with chronic renal failure who is being considered for CAPD management.

Tuberculosis in patients with systemic lupus erythematosus

Tuberculosis associated with systemic lupus erythematosus (SLE) was studied in a cohort of 311 patients seen between 1963 to 1979. There were 16 such patients, giving rise to a prevalence rate of 5%. The characteristics of SLE-associated tuberculosis include a high incidence of miliary and far-advanced pulmonary disease, delay in establishing diagnosis, especially the extrapulmonary form, and tendency to attribute symptoms like fever, malaise, and weight loss to the lupus process. Treatment was successful in 9 patients. Of the 7 death 5 were attributed directly to the mycobacterial infection and 2 to complications of SLE.