INTRODUCTION

Kidney transplant patients exhibit high risk for primary infection as well as to reactivating the latent infections induced with mycobacterium tuberculosis. It has been reported that tuberculosis incidence in transplant patients highly depends on region and ranges from 0.5% in North America and Western Europe to 11.8% in endemic areas [1-2].

The global incidence of tuberculosis in Serbia has varied in the last two decades with a falling trend until 1993 (61/100000 in 1980, 36/100000 in 1993), and slightly rising until 1996 (40/1000,000), to fall again thereafter (32/100000 in 1999) [[3]] (data obtained from National Referential Center for Tuberculosis). In recent years, geographical and ethnic differences in tuberculosis incidence have been registered. An important rise of tuberculous infection was registered in zones near Bosnia, Slavonia, and Romania, and particularly in Kosovo, while in the central part of Serbia the incidence of tuberculosis has remained unchanged [[3]].

During 1997, we noticed the remarkable increase in tuberculosis incidence following kidney transplantation. For that reason retrospective analyses were undertaken with the aim of finding out the prevalence, the clinical course of tuberculosis, and its impact on the long term outcome of transplant patients. In addition, we confirmed that the frequency of tuberculosis in kidney transplant patients correlated with the trend of tuberculosis in general population.

PATIENTS AND METHODS

All transplanted patients regularly followed at our outpatient department were included in the retrospective study. The population comprised the patients who received kidney transplants in our department in from 1980 to December 1998 (456 patients), 30 refugees transplanted at the centers in Bosnia and Croatia, and 25 patients transplanted in India. The followup duration was 1–14 years.

Routine immunosuppressive therapy consisted of azathioprine and prednisone until 1985, when cyclosporine was introduced. Since 1987, the majority of the patients have been treated with triple immunosuppressives (cyclosporine, azathioprine, and prednisone). The immunologically high risk group was treated initially with antilymphocyte globulin that was applied during the first two weeks after operation, followed by triple maintenance immunosuppressives given in standard doses. From 1980 to 1988 steroids were initiated in the dose of 500 mg, gradually reduced to 0.20 mg/kg b.w. up to the 60th day. Since 1989 the maintenance dose of steroids has changed (0.15 mg/kg b.w.). Azathioprine was administered the first four days in a dose of 3 mg/kg b.w., and then reduced to 2 mg/kg b.w. We have changed initial cyclosporin A therapy: from 1983 to 1988 the initial CyA dose was 14 mg/kg b.w., and the target CyA level was 200–400 ng/mL (RIA), and thereafter 8 mg/kg b.w. with the target CyA level of 150–200 ng/mL (Tdx Abbot) in the first posttransplant months. Acute rejection was treated with 500 mg of methyl-prednisolone over 3 to 5 successive days. In cases when this therapy failed to produce satisfactory response, either ALG or OKT3 was administered. [[4]].

We had been looking for tuberculosis in all transplant patients with fevers of unknown origin. The diagnosis of tuberculosis was established in cases where acid-fast bacilli were isolated from any clinical samples (sputum, urine, and stool). Disseminated tuberculosis was considered to be present when mycobacterium tuberculosis was isolated from two or more noncontiguous organs or from blood. Sample processing and organism identification was carried out by standard methods. The samples were cultured in Lowenstein media. Antimicrobial susceptibility testing of the mycobacterium tuberculosis isolates was routinely done.

All of the patients with tuberculous infection were treated with rifampicin, isoniasid, and ethambutol for the first two months. The therapy was continued at least 12 months. Liver function tests were performed weekly in the first month and monthly thereafter. Antituberculous (AT) prophylaxis was administered to patients transplanted after 1995 who had a history of tuberculosis, and/or tuberculosis residue on the chest X-ray, and who lived in regions known for the high prevalence of tuberculosis. In these cases, isoniazid was given at the dose of 300 mg/day for one year. Mortality was considered to be related to tuberculosis when the death incidence occurred during the treatment, and obtained microbiological evidence confirmed active tuberculosis at the time of death.

RESULTS

In the course of the studied period, tuberculous infection was diagnosed in 16 patients (13 patients transplanted in our department and three refugees), which represented a prevalence of 3.13%. The first case was discovered in 1986, and five in 1997 (Tab. 1). The annual tuberculosis incidence varied from 2.2% in 1989 to 12.2% in 1997. In comparison to annual tuberculosis incidence in Serbia, it seems that tuberculosis in kidney transplant patients follows the trend of tuberculosis in the general population (Fig. 1).

DOES TUBERCULOSIS AFTER KIDNEY TRANSPLANTATION FOLLOW THE TREND OF TUBERCULOSIS IN GENERAL POPULATION?

All authors

V. Lezaic, Visnja Lezaic M.D., Ph.D., R. Radivojevic, G. Radosavljevic, R. Blagojevic, Lj. Djukanovic, S. Simic & T. Cvok

https://doi.org/10.1081/JDI-100001289

Published online:

07 July 2009

Figure 1. Annual Incidence of Tuberculosis in Kidney Transplant Patient Population and in General Population;——Number of transplant patients with active tuberculosis/all transplant patients followed up in our department per year (%);……Annual incidence of active tuberculosis in general population (Number/100,000)

Display full size

The data on patients with tuberculosis are presented in Table 1. There were nine males and seven females, average age 34.9, on regular hemodialysis for 7.5 to 108 months before transplantation. Living related donor transplantations were performed in eight patients, and cadaveric transplantations in the remaining eight patients. Eight patients were hepatitis B or C positive, none was HIV positive (data not presented). Fourteen patients had one to three episodes of acute rejections within one month to two years before tuberculosis was established, and were treated with additional immunosuppressives. At the time that tuberculous infection was diagnosed, eight patients exhibited normal graft function and eight suffered from chronic graft failure (s-creatinine between 210 μmol/L and 700 μmol/L).

Five patients had residual tuberculosis in preoperative chest X-rays, and three had prior histories of pulmonary (one patient) and renal tuberculosis (two patients). The latter with completed a full course of treatment with antituberculosis drugs more than 10 years before hemodialysis had been started. The medical records for 13 patients transplanted in our department were carefully reexamined, but they failed to reveal any signs or symptoms of active infection before kidney transplantation. Having in mind the geographical distribution of tuberculosis in our country, further analysis showed that seven patients lived in high-prevalence areas: two patients in Kosovo, four from Bosnia and Slavonia, and one lived in the border zone (Table 1). Additional epidemiological data showed that none of the patients' family members had active tuberculosis. Two patients with previous tuberculosis (one lived in a high-prevalence area) received AT chemoprophylaxis for one year following kidney transplantation.

Tuberculous infection was diagnosed after an average period of 40.5 months after transplantation (range from 1.5 months to 10 years). Three patients who had had prior tuberculosis developed infection first, that is, within the first three postoperative months, and three patients immigrating from Bosnia when the war began developed tuberculosis six, eight, and nine years respectively after kidney transplantation (Tab. 1).

The form of tuberculosis infection is presented in Tab. 2. It was pulmonary in 12 patients (75%) and urinary in two patients. Hematogenous dissemination of infection was diagnosed in remaining two patients: pulmonary and intestinal (one patient) and pleural; ascite, pericardial friction rub, pancytopenia (one patient). In three patients no typical signs of pulmonary tuberculosis were revealed, while in 13 patients the following typical pattern was found: multinodular infiltration in one or both of the upper lobes of the lung in 12 patients (with cavitation in one patient), pleural effusion in four, and hilar adenopathy in six patients. One patient experienced concomitant involvement of the throat. All patients with urinary tuberculosis were asymptotic, and had only sterile pyuria in the preceding six to eight months before positive urinary cultures for tuberculosis were found. The most common general symptoms were intermittent fever (the majority of patients had high fever for several months), progressive fatigue, loss of appetite, and weight loss. Other symptoms were related to the organ localization of the disease: dyspnea, chest pain, cough, hoarseness, diarrhea. Concomitant opportunistic infections were found in five patients (candidiasis, CMV, and HSV). Apart from progressive anemia and high erythrocyte sedimentation rate, the rest of the laboratory findings were not helpful in the majority of patients. One patient with disseminated tuberculosis had pancytopenia, and diagnosis was based on bone biopsy.

All patients were treated with rifampicin and isoniazid, with the addition of ethambutol for the two first months. Two patients ceased AT therapy after 3 and 4.5 months due to noncompliance. In the remaining patients the treatment lasted 12 to 22 months. Microbiological response was evaluated in all patients and was favorable, that is, no relapse occurred during the median followup period of 3.7 years (range from 1 to 12 years) even in two patients who ceased their treatment. The immunosuppressive treatment during AT therapy had to be modified due to severe drug nephrotoxicity (temporary elevation of sCr in seven patients) and hepatotoxicity (in four patients). It was continued with triple immunosuppression, but cyclosporin doses had to be increased 1 to 3.5 times according to their whole blood levels. No acute rejection episodes were registered during the AT treatment.

The outcomes of patients with tuberculosis is presented in Table 2. Fourteen patients recovered completely. Two patients with disseminated tuberculosis died within the first six months. Accordingly, the mortality rate was 12.5%. In the followup period, graft function remained stable in all eight patients with normal graft function at the time of tuberculosis onset. Although the remaining six patients with chronic graft failure recovered, progressive graft failure developed, and hemodialysis was restarted in one patient with advanced graft failure two months after AT therapy onset and in two patients three years later. Four patients died 2 to 14 months after AT therapy withdrawal (Tab. 2). Causes of death included severe liver failure (two patients), cerebrovascular insult in one, and CMV virus infection in one patient.

DISCUSSION

The retrospective analysis in this study showed that the prevalence of tuberculous infection in our renal transplant patients (3.13%) is similar to referential studies [1-2], [5-6]. During the ten-year period prior to 1986, no tuberculous infection had been diagnosed, but thereafter 16 tuberculosis infections were evidenced. This discrepancy is partly due to the constantly increasing number of patients transplanted in our department after 1986. At the same time, we have noticed an annual rise of tuberculosis in the hemodialysis patient population regularly followed up at our department, particularly during 1997 [[7]]. In comparison to the data obtained from National Referential Center for Tuberculosis in Serbia, it seems that the incidence of tuberculosis in patients on renal replacement therapy follows the trend of tuberculosis in our general population.

The previous report emphasized that reactivation of old healed primary or postprimary lesions were common forms of tuberculosis in transplant patients [[2]], [[8]]. Despite the fact that medical records did not register any signs or symptoms of active infection before kidney transplantation, five of our patients had residual tuberculosis, or had had a tuberculous infection in their childhood. Also, we noted that four of them continued to live in the high prevalence region of our country after transplantation, which means they were at high risk for tuberculosis. Throughout the first posttransplant months, tuberculosis developed in only three of the latter, indicating that preselection of patients for kidney transplantation should include the screening tests that are likely to be more effective [[9]], particularly in our population where the incidence of the disease and latent infection is high. Furthermore, at the outbreak of the Bosnian war in 1992, a lot of refugees emigrated to Serbia. Caloric malnutrition and poor living conditions in combination with long term immunosuppression could facilitate the tuberculous infection in our three patients who developed tuberculosis six, eight, and nine years after kidney transplantation.

The role of maintenance immunosuppression in reactivation of tuberculosis has been the subject of many studies. Although the authors agree that variations in the protocol could not facilitate tuberculosis development [[2]], [[6]], Qunibi noted an increased incidence of tuberculosis in patients receiving more than 10 mg/day of prednisolone [[1]]. It is also possible that cyclosporine could have led to early recrudescence of tuberculosis in patients harboring dormant tubercle bacilli [[10]]. In addition to maintenance immunosuppression, 14 of our patients with tuberculosis were treated with methylprednisolone pulses, OKT 3 and ALG for rejection episodes, indicating that increased immunosuppression might be one of the risk factors for clinical tuberculosis, which is in agreement with others [[1]], [[8]]. These additional immunosuppressives probably predisposed five of them to other pathogens (viruses and mushrooms) at the same time, which complicated final diagnosis and treatment, as was shown in Hall's et al. study [[11]].

The time of tuberculosis onset following kidney transplantation varies, but in the majority of patients, tuberculosis developed after the second posttransplant year, as was indicated in other studies [[5]], [[12]]. Pulmonary tuberculosis was a more common form of infection in our renal transplant patients, comprising 75% of all cases. In this study, the prevalence of disseminated disease was less common than reported elsewhere [[1]]. It is possible that some of our patients with apparently localized disease forms had subclinical involvement of multiple sites.

Antituberculous therapy duration ranged from 7 to 22 months in all but four patients (two died and two were noncompliant). All patients had favorable response to AT treatment, and no relapse occurred in the course of the incoming period, even in two patients who ceased their treatment due to noncompliance. Thus, the issue of necessary duration of therapy remains. There is no consensus regarding the protocol to be applied in these patients. Several authors have suggested an effective course from six to nine months, at least in patients with more localized disease forms [[5]], [[8]], [[11]]. However, other authors recommend a longer course [1-2]. It is well recognized that toxicity due to AT drugs is higher in transplant patients because of well known interaction with immunosuppressive drugs. The most serious problem associated with AT treatment was acute rejection or graft loss associated with withdrawal of immunosuppression [[2]], [[5]], [[11]]. Two-thirds of our patients developed different drug toxicity during treatment. However, no one lost their grafts. Hepatotoxicity was the most severe adverse reaction, and in spite of AT withdrawal in two patients, previous chronic hepatitis was fatal. Other authors [[5]], [[12]] have made similar observation.

Prophylaxis with isoniazid after kidney transplantation was not used routinely in our transplant unit before 1995, even in patients with previous tuberculosis. Chronic hepatitis was an additional risk in eight (50%) of our patients studied. All transplant patients with prolonged fever were carefully monitored for tuberculosis in our department. Only two patients with residual tuberculosis in preoperative chest X-ray (one lived in high prevalence area) received the prophylaxis with isoniazid during the first year after transplantation, but developed tuberculosis one year later. Prophylaxis with isoniazid in the transplant population is a controversial issue due to a high risk of hepatotoxicity and infection with isoniazid-resistant tubercle bacilli. The American Thoracic Society and others recommend Mantoux testing–based prophylaxis [[13]]. However an anergy to tuberculin is well known to occur in up to 70% of recipients due to an impaired immunity. Whereas selective administration of chemoprophylaxis may be used in countries with relatively low incidence of tuberculosis, our experience indicates that it should be used regularly in developing countries. Authors from other developing countries [[1]], [[6]], [[12]] stressed the same.

Tuberculosis and AT therapy have important implications in the outcomes of our transplant patients. The mortality rate directly related to tuberculosis (12.5%) is similar to those in other studies [[1]], [[5]]. In a recently published paper, Jha et al. have proposed an aggressive search for tubercle bacilli using bronchoscopy and examination of BAL fluid in febrile transplant patients with respiratory infection. It enabled them to start prompt specific therapy before the bacillary load became high, and consequently, improved the outcomes of their transplant patients with tuberculosis [[14]]. Forty percent of our patients died during the whole followup period of 1 to 14 years, but only in two patients with disseminated tuberculosis was death directly related to tuberculosis, which could be the consequence of delayed diagnosis and adequate treatment.

According to the results obtained here, it could be concluded that the appearance of active tuberculosis after kidney transplantation follows the tuberculosis trend in the general population. High incidence of tuberculosis in our general population requires high suspicion and careful monitoring for tuberculosis among our kidney transplant recipients. Additional precautions should be taken with those patients with previous tuberculosis and those on high dose steroids, as well as with those patients who immigrate from high prevalence regions. In these cases, chemoprophylaxis could be recommended.