Latent Tuberculosis Infection Increases in Kidney Transplantation Recipients Compared With Transplantation Candidates: A Neglected Perspective in Tuberculosis Control

Abstract

Background

The prevalence and incidence of latent tuberculosis infection (LTBI) in patients with kidney transplantation remain unclear.

Methods

In this prospective study, we enrolled kidney transplantation candidates (KTCs) and recipients (KTRs) from 2014 to 2018. We defined LTBI as a positive result of QuantiFERON-TB Gold In-tube (QFT). We analyzed the predictors for LTBI acquisition and followed up on QFT assay test for 2 years among those initially without LTBI.

Results

Of 425 patients enrolled, 305 (71.8%) patients belonged to the KTC group and 120 (28.2%) to the KTR group. The initial QFT showed positive results in 32 (10.5%) and 24 (20.0%) patients in the KTC and KTR groups, respectively (P = .009). The QFT response value in patients with LTBI was higher in the KTR group than in the KTC group (1.85 vs 1.06 IU/mL, P = .046). Multivariate logistic regression showed that old age, absence of bacillus Calmette–Guérin (BCG) scar, presence of donor-specific antibody, and KTR group were independent factors for positive LTBI. For participants with initial negative QFT, positive QFT conversion within a 2-year follow-up was higher after kidney transplantation (20%) than in KTCs (5.5%) (P = .034).

Conclusions

This study is the first cohort to follow up LTBI status in patients with kidney transplantation and shows its higher prevalence and incidence in KTRs. It indicates that surveillance of LTBI after renal transplantation is important. In addition to status of kidney transplantation, old age, no BCG vaccination, and positive donor-specific antibody are also positive predictors for LTBI.

In 2017, an estimated 10.0 million people had active tuberculosis (TB), and 1.6 million TB-related deaths were recorded worldwide [1]. Tuberculosis remains the most common infectious disease in the world [2]. In the last decade, TB prevention has followed the Global Plan To Stop TB 2006–2015, which aims to reduce TB incidence and deaths [3]. After the goal was almost reached in 2015, the World Health Organization (WHO) suggested the END TB Strategy for the post-2015 era, including controlling latent TB infection (LTBI) towards TB elimination [4]. Among the strategies, we need to screen high-risk groups for LTBI treatment to prevent TB occurrence and transmission [5]. According to the WHO, 2 such high-risk groups are kidney transplantation candidates (KTCs) and kidney transplantation recipients (KTRs) [5].

In fact, patients with solid organ transplantation have been reported to have TB incidence as high as 506–512 per 100 000 person-years in China and Spain, respectively, which is appproximately 7–27 times higher than that of the general population [6, 7]. In particular, kidney transplantation has an odds ratio (OR) of 4.59 for developing TB [8]. In addition, preoperative LTBI has been identified as a risk factor for developing TB [8]. Once TB develops in patients with kidney transplantation, morbidity is high, approximately 6.1–8.9% [6, 7, 9, 10], and TB may cause loss of the renal allograft [11] because the manifestations of TB are atypical and delayed diagnosis ensues [9, 10]. Therefore, LTBI and TB are clinical concerns for patients belonging to the KTC and KTR groups.

Active TB can develop in patients after kidney transplantation from different pathogenesis, including reactivation from LTBI before or a new infection after transplantation [12], which increases due to immunosuppressive therapy. The WHO suggests that clinicians screen and manage LTBI before transplantation to reduce the possibility of reactivation [13]. Under the LTBI intervention strategy, the problem that remains after we have treated pre-existing LTBI is new infection or the reactivation of immune-unrecognized Mycobacterium tuberculosis after kidney transplantation. However, the prevalence of LTBI and its incidence trend after kidney transplantation remain unclear. Only a few cross-sectional studies have evaluated LTBI prevalence in KTRs [14, 15], and we have limited knowledge of new M. tuberculosis infection in this targeted population. Therefore, we conducted this study to evaluate the status of LTBI in patients after kidney transplantation in order to provide information for future TB-control consensus.

METHODS

Participant Enrollment

This prospective study was conducted in a tertiary referral medical center in northern Taiwan. Under the approval of the Research Ethics Committee of the study hospital (no. 201309056RINC), we recruited patients aged 20 years or older who were KTCs or KTRs in transplantation clinics from January 2014 to December 2018. All of the final enrolled participants provided signed informed-consent forms. We excluded patients with active TB, prior history of TB, liver cirrhosis, active cancer, and human immunodeficiency virus infection (Figure 1).

Tuberculosis and Latent Tuberculosis Infection Definitions

Each participant’s clinical history, symptoms, and chest radiograph were reviewed to exclude active TB disease [16]. Mycobacterial study of 3 sputum samples and computerized tomography were arranged if active TB was suspected [17]. Active TB was defined as microbiology positive for M. tuberculosis or typical pathology or radiographic findings if available. After excluding cases of active TB, we sampled each participant’s peripheral blood and examined a QuantiFERON-TB Gold In-Tube assay (QFT) (Cellestis, Australia) using a 3-tube kit [18]. Interferon-γ level was measured in the reaction supernatants, and the results were interpreted as positive, negative, or indeterminate according to the manufacturer’s recommendation [19, 20]. In this study, LTBI was defined as a positive QFT result. We then compared the prevalence of LTBI status between the KTR and KTC groups.

For patients with initial negative QFT, we followed up the QFT status every 6 months for 2 years. We categorized the positive conversion of QFT as new LTBI and stopped further follow-up. If a KTC with negative QFT received transplantation, he or she was transferred from the KTC group to the new KTR group for follow-up. For those without a pretransplantation test of QFT, we tested their LTBI status within 3 months after kidney transplantation and followed up with them if the QFT result was negative. The overall incidence rates of new LTBI were calculated for the KTC group, overall KTR group, and new KTR group.

For patients with positive QFT, we suggested preventive therapy if there were no contraindications. The acceptance rate of the therapy was not 100% because it was not a mandatory component of standard care in Taiwan. We followed up all participants for 2 years and checked for the occurrence of active TB in the Taiwan TB Registry.

Data Collection

The demographic and clinical data, including age, gender, underlying comorbidities, details of dialysis or transplantation, prior TB history, laboratory results, and immunosuppressants, were reviewed from the hospital’s electronic records. Trained assistants and investigators interviewed the participants for respiratory and constitutional symptoms, smoking status, and history of TB exposure with a questionnaire. Body mass index (BMI) and scar number for bacillus Calmette–Guérin (BCG) vaccination were checked by trained assistants.

For the KTR group, we recorded the data of the transplants, including living or deceased donor, the presence of donor-specific antibody (DSA), ABO blood type-incompatible (ABO incompatible) status of the donor, usage of induction therapy, and maintenance immunosuppressant drugs. We recorded induction therapy and the kinds of maintenance immunosuppressant drugs, especially during the examination of baseline QFT. The usual maintenance immunosuppressive regimen comprised tacrolimus with mycophenolate mofetil and steroids. We used rituximab plus plasmapheresis as the induction therapy for kidney transplantation with positive DSA (DSA+) [21] or ABO-incompatible donors [22]. For recipients with positive DSA, intravenous immunoglobulin (IVIG) infusion after plasmapheresis was given for desensitization [21].

Chest radiography and computerized tomography were interpreted independently by a pulmonologist and a radiologist [23]. If the interpretations by the 2 readers differed, we consulted a senior pulmonologist to discuss the final coding. The classifications of radiographic findings are detailed in the Supplementary File.

Statistical Analysis

Intergroup differences were analyzed using the Student’s t test for numerical variables, as appropriate, and chi-square test for categorical variables. Multivariate logistic regression analysis was used to identify factors associated with baseline LTBI using the stepwise method. (The details of factor selection are provided in the Supplementary File). A 2-sided P < .05 was considered significant. All analyses were performed in SPSS (version 19.0; IBM Corporation).

RESULTS

Participant Recruitment and Demographics

During the study period, 489 participants were recruited. Among them, 17 were excluded due to prior TB and 47 withdrew their consent. Finally, 425 patients received QFT examinations. Among them, 305 patients belonged to the KTC group and the other 120 patients were in the KTR group (Figure 1).

With regard to demographic characteristics (Table 1), the average age between the 2 groups was similar (47.5 vs 48.9 years, P = .217), but the proportion of males was higher (60% vs 48%, P = .034) in the KTC group than in the KTR group. In addition, the presence of underlying diabetes mellitus and congestive heart failure was higher in the KTC group, whereas hepatitis B virus infection and rheumatoid arthritis were higher in the KTR group. Laboratory data showed lower blood hemoglobin and total bilirubin but higher blood leukocyte counts in the KTC group than in the KTR group. There were no differences regarding other TB-associated clinical factors, such as smoking, scar of BCG vaccination, BMI, TB exposure, and dialysis history. The presence of any respiratory symptoms and radiographic lesions as well as the values of the initial QFT response were similar between the 2 groups.

Details of Initial Latent Tuberculosis Infection Status

In the initial QFT examination, positive results were reported in 32 (10.5%) patients in the KTC group and 24 (20.0%) patients in the KTR group (Figure 1). The positive rate of QFT was significantly higher in the KTR group (P = .009 by chi-square test) than in the KTC group. Among the patients with positive initial QFT, 13 (54%) in the KTR group and 20 (63%) in the KTC group received preventive therapy for LTBI (P = .530). The remaining patients with LTBI requested observation and declined preventive LTBI treatment. The QFT response value in LTBI treatment recipients was 1.20 ± 1.10 IU/mL, which was similar to the response value of 1.89 ± 2.19 in those without treatment (P = .174). Although indeterminate results were also higher in the KTR group than in the KTC group, the difference was not significant (10% vs 5.9%, P = .138). With regard to the QFT response value, the response was higher in the KTR group with positive initial QFT results than in the KTC group (1.85 ± 2.01 vs 1.06 ± 0.82, P = .046). For those with initial negative or indeterminate results, the values were similar in the 2 groups.

Factors Associated With Baseline Latent Tuberculosis Infection

To further differentiate factors associated with baseline LTBI in all of the study participants, we performed logistic regression. Increasing age, coronary artery disease (CAD), hepatitis B virus infection, scar of BCG vaccination, positive DSA, induction therapy, and kidney transplantation were significant factors in the univariate analysis (Table 2). Multivariate logistic regression by a stepwise method was performed by using all significant factors in univariate analysis and other relevant factors, such as gender and TB exposure. The results showed that the positive independent factors for LTBI included increasing age (OR, 1.027; 95% confidence interval [CI], 1.000–1.055, per 1-year increment; P = .050), positive DSA (OR, 8.242; 95% CI, 1.249–54.399; P = .028), and receiving kidney transplantation (OR, 1.904; 95% CI, 1.031–3.516; P = .040), whereas the negative factor was BCG vaccination (OR [95% CI], 0.199 [.065–.608] [P = .005] for 1 scar and 0.175 [.051–.598] [P = .005] for 2 scars compared with no vaccination scar).

Within the KTR group (Table 3), positive DSA was also significantly associated with LTBI status. Underlying CAD, autoimmune disease, hepatitis C virus infection, prior kidney transplantation history, and presence of respiratory symptoms were associated with LTBI status. Immunosuppressants were not significantly associated with LTBI status.

Details of Latent Tuberculosis Infection in the Follow-up Course

The participants with initial negative QFT status were invited to be followed up for QFT every 6 months. There were 161 in the KTC group, 51 in the overall KTR group, and 15 in the new KTR group (Table 4). The demographic characteristics among those with initial negative LTBI and follow-up were similar, except that the blood hemoglobin and total bilirubin levels were higher but leukocyte counts were lower in the overall KTR group than in the KTC group. With regard to induction therapy, 3 patients received induction therapy before kidney transplantation with incompatible ABO status in the KTR group. Among them, all 3 patients had no LTBI conversion (6.8% vs 0, P = .476). All of the cohort reported no TB exposure during the follow-up. The new positive QFT numbers in the 3 groups are shown in Figure 2A. Within the 2-year follow-up, 9 (5.6%), 7 (13.7%), and 3 (20%) patients had QFT positive conversion in the KTC, overall KTR, and new KTR groups, respectively. The positive conversion rate was higher in the new and overall KTR groups than in the KTC group (P = .034 and .055, respectively) (Figure 2B).

Among those who underwent kidney transplantation, 2 cases had active TB development. The first patient received kidney transplantation from a deceased donor. His maintenance immunosuppressants included tacrolimus, mycophenolate mofetil, and prednisolone. His initial QFT was positive, but no preventive therapy was given because the patient refused it. Active pulmonary TB was diagnosed by suspicious findings on radiography plus treatment response at approximately 11.5 months after kidney transplantation. He was treated with an anti-TB regimen with good graft function. The other patient, who had hepatitis C virus infection, had received a deceased-donor kidney transplantation 27 years before the initial negative QFT. The maintenance immunosuppressants were cyclosporine, mycophenolate mofetil, and prednisolone. During our follow-up, his kidney function declined, and he received a second renal transplantation from a deceased donor at the 22nd month of follow-up. He had chronic cough and QFT at the 24th month, indicating positive conversion. Pulmonary TB was diagnosed by positive sputum acid-fast smear and mycobacterial culture for M. tuberculosis. The treatment course for TB was uneventful, without major drug adverse effects. The incidence of TB was approximately 833 per 100 000 person-years (2 in 120 cases with kidney transplantation in 2 years). In contrast, no active TB cases developed in the KTC group during the 2-year follow-up.

DISCUSSION

In the present study, we report that KTRs had a higher LTBI prevalence and incidence than KTCs in a cross-sectional and cohort study design, respectively. For this population, old age, absence of BCG vaccination scar, and positive DSA, in addition to receiving a kidney transplantation, were significant factors associated with positive QFT-defined LTBI.

In the post-2015 era, the WHO aims to eliminate TB worldwide and has established many frameworks [4]. Among them, LTBI screening and treatment are vitally important issues [5] and have been recommended in high- or intermediate-income countries with a TB incidence of less than 100 per 100 000 person-years [13]. Of the LTBI-risk population, patients with kidney transplantation account for a major proportion of transplantation patients [24] and need to be targeted because of their immunocompromised status and high mortality once active TB develops. Under contemporary recommendations, pretransplantation LTBI screening has been implemented worldwide [7, 25]; however, no evidence of LTBI survey after transplantation has been reported before this prospective study.

The present cross-sectional study found a higher LTBI prevalence in the KTR group than in the KTC group, which was supported by adjustment of the multivariate analysis. In addition, the occurrence of new LTBI infection and active TB development within 2 years was also higher in the KTR group than in the KTC group. These facts emphasize the importance of LTBI/TB control after kidney transplantation. Among the factors associated with LTBI status after kidney transplantation, positive DSA was an independent factor. The process and medication given before transplantation with positive DSA included induction with rituximab and IVIG infusion after plasmapheresis for desensitization [21]. In comparison with incompatible ABO status, IVIG was the only additional medication in recipients with positive DSA. Therefore, we suggest that IVIG-related suppression of T cells [26] and innate immune cells [27] might be an important factor that leads to susceptibility to M. tuberculosis infection. In addition, the status of kidney transplantation was still one of the independent predictors for LTBI (Table 2). Kidney transplantation status might be a surrogate of many complex factors after transplantation, one which cannot be dissected with the scale of the current study and the collinearity between transplantation status and immunosuppressant medications. Although the procedure of induction therapy and the immunosuppressants might be important concerns related to LTBI status, no statistical significance was found in this study. Further large-scale studies focusing on the effects of medication are required.

The higher rate of positive conversion among the new KTR group might have arisen because some of them had false-negative QFT prior to transplantation due to dormant status or a long interval since M. tuberculosis infection [28]. This kind of activation of dormant or immune-unrecognized M. tuberculosis might be the pathogenesis of positive conversion of LTBI soon after recent kidney transplantation with the use of immunosuppressants. On the other hand, a new M. tuberculosis infection could exist among the KTR or KTC group [29]. However, the susceptibility to new M. tuberculosis infection might be higher in the KTR group than in the KTC population (Figure 2B), possibly due to the process and immunosuppression therapy used for the transplantation [30, 31]. The QFT conversion rates in the KTR groups were similar, with 7–8.5% conversion at the 12-month follow-up [32, 33] and 13.6% conversion at the 24-month follow-up [34] among patients with rheumatoid disease receiving biologic agents. Therefore, this result indicates that, in addition to pretransplantation screening, surveillance of LTBI is also needed after transplantation. Although the duration of follow-up remains unclear and requires further investigation, we propose LTBI checks at least twice annually after transplantation because the highest risk of LTBI conversion occurs approximately 6–12 months after kidney transplantation, and active TB occurs more frequently within the first 2 years, according to the literature and our report [6, 35]. We have no suggestions for periods exceeding 2 years due to limited data. In addition, data may need to be collected in an area with less TB prevalence.

On the other hand, many clinical factors in addition to kidney transplantation favored positive LTBI status. Among them, age is a well-known factor that is positively correlated with the QFT-positive rate [36]. In contrast, a present BCG vaccination scar shows a negative correlation with LTBI, which is consistent with investigations in children [37] and adult prisoners [38]. With regard to TB exposure, the association with positive QFT is not significant and may be related to the information not being recent and having been collected by questionnaire. Therefore, the effect and significance might be less correlated than in a previous report on TB contact [39].

There are several limitations to the present study. First, the participants were voluntarily enrolled, so selection bias might exist. Second, the use of questionnaires for some data collection, such as TB exposure and smoking, might have incurred recall bias. Third, the case number for the 2-year follow-up was small. A validation study is needed. Fourth, the QFT statuses of the living donors were not checked. Last, the participants were enrolled in Taiwan, so generalization to other areas or ethnicities may not be applicable and will require further study.

In conclusion, LTBI was higher in the KTR group than in the KTC group. Old age, absent BCG scar, positive DSA, and status of kidney transplantation were LTBI-associated factors. In particular, the prevalence of positive LTBI conversion was approximately 20% within 2 years in the new KTR group. This finding indicated that a strategy for posttransplantation LTBI surveillance is needed, despite this not being included in current LTBI recommendations.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. C-C. S., C-Y. L., and M-K. T. designed and performed the study. C-C. S., C-Y. L., S-W. L., J-Y. W., C-J. Y., and M-K. T. were involved in data interpretation, analysis, critique, and manuscript preparation. C-Y. L. was responsible for coordinating the study.

Acknowledgments. The authors thank the second, third, seventh, and eighth core laboratories of the Department of Medical Research, National Taiwan University Hospital, for providing technical services and support with data and statistics. The Research Ethics Committee of National Taiwan University Hospital approved this study (Institutional Review Board No.: 201309056RINC).

Disclaimer. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Financial support. The work was supported in part by the Centers for Disease Control, Taiwan (grant number MOHW-107-CDC-C-114-000104; MOHW103-CDC-C-114-112302; MOHW104-CDC-C-114-122202; MOHW105-CDC-C-114-000103; and MOHW106-CDC-C-114-000103).

Potential conflicts of interest. The authors report no potential conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References