Incidence, Risk Factors, and Mortality Associated with Tuberculosis in Solid Organ Transplant Recipients in Taiwan
Article information
Abstract
Background
Solid organ transplantation (SOT) recipients are at increased risk of post-transplant tuberculosis (TB). However, the effect of this risk on mortality remains unclear. We examined the incidence and risk factors of posttransplant TB, and its effect on mortality in SOT recipients in Taiwan.
Methods
We collected data on 8,205 patients who received their first transplants from 2009 to 2018 from the National Health Insurance Research Database, and identified 201 new TB cases. Transplants were identified and verified by the medical procedure codes. A Cox proportional-hazards model was used to identify the determinants of TB infection.
Results
For the 7,685 recipients, with 34,412 person-years (PYs), 1,630 deaths (393.41/1,000 PYs) were reported. Male sex was associated with a 44 % increase in the risk of TB (hazard ratio [HR], 1.44; 95% confidence interval [CI], 1.05 to 1.98). In addition, age over 65 years was associated with a 4-fold increase in the risk of TB (HR, 4.04; 95% CI, 2.04 to 8.00). The mortality rates in the population varied by transplantation organ type (lungs, 187.75/1,000 PYs; heart, 81.11/1,000 PYs; liver:, 58.47/1,000 PYs; pancreas, 42.36/1,000 PYs; and kidneys, 23.76/1,000 PYs). Recipients with posttransplant TB had a 2.53-fold increased risk of mortality (HR, 2.53; 95% CI, 1.94 to 3.29).
Conclusion
Posttransplant TB is associated with an increased risk of mortality in SOT recipients. Preventing TB can mitigate this risk, which underscores the importance of monitoring and managing TB in this population.
Introduction
Tuberculosis (TB) is one of the most common infectious diseases worldwide. According to the World Health Organization, approximately 10.6 million cases of TB and 1.6 million TB-related deaths were reported in 2021 [1]. Taiwan is regarded as an endemic area for TB; it reported 6,576 cases of TB (28 cases per 100,000 individuals) and 477 TB-related deaths (2 cases per 100,000 individuals) in 2022 [2].
In solid organ transplantation (SOT) recipients, the risk of TB is estimated to be 20 to 74 times higher than that in the general population [3]. For example, the prevalence of TB in kidney transplant recipients was reported to be 59.2 times higher than that in the general population of Taiwan [4]. The variability in the incidence of posttransplant TB is influenced by the local endemicity of TB. In regions with low endemicity, the incidence of posttransplant TB in SOT recipients ranges 0.5% to 6.4%, whereas in regions with high endemicity, it can be as high as 15% [5].
Because of their long-term immunocompromised status, SOT recipients are at increased risk of posttransplant TB; the underlying disease processes leading to transplantation are also associated with an increased risk of TB. As well, the long-term requirement for immunosuppressive medication after transplantation is associated with an increased risk of graft failure and death [6-8]. Posttransplant TB predominantly occurs through the reactivation of latent TB infection, with other, rarer cases occurring as donor-derived TB through donor allograft or de novo infection [6]. Diagnosing TB in transplant recipients is often challenging and delayed. In addition, the drug–drug interaction that occurs between anti-TB treatment and immunosuppressive drugs may be problematic, because of its association with graft rejection [5]. Despite the importance of posttransplant TB, its mortality risk in SOT recipients has not yet been well defined, with a limited number of studies exploring the association between posttransplant TB and mortality in these patients. Therefore, in this study, we examined the epidemiology and risk factors of posttransplant TB in SOT recipients in Taiwan, and explored the association between posttransplant TB and mortality in these patients.
Materials and Methods
1. Study population
Data on 8,205 patients who received their first transplants in the period 2009−2018 were collected from the National Health Insurance Research Database. Patients meeting the following criteria were excluded from the study: having a history of TB infection before 2009 (n=176), being younger than 18 years, or having received a gut transplant (n=4). We excluded patients who had received a gut transplant because a gut transplant may alter a patient’s susceptibility to TB infection, and the number of such cases is relatively small. Ultimately, 7,685 transplant recipients were included in the study (Figure 1). This study was approved by the Institutional Review Board (IRB) of Taipei Medical University (IRB number: N202312064). This study was conducted using data from an encrypted research database in which all personal identifiers were removed prior to analysis. Because the study involved only secondary analysis of de-identified data and posed minimal risk to participants, individual informed consent was not required, in accordance with the approval of the IRB.
Flowchart of study population selection. Some of the numbers listed under the exclusions may be duplicated, which can lead to discrepancies when subtracting them directly compared to the final total.
2. SOT
Transplants were identified and verified by the medical procedure codes from January 1, 2009, to December 31, 2018, on the National Health Insurance Research Database. The following medical recorders were used: 68035A or 68035B for heart transplant surgery; 68037A, 68037B, or 68047B for lung transplant surgery; 75020A or 75020B for liver transplant surgery; 76020A or 76020B for kidney transplant surgery, respectively.
3. TB
TB was defined in accordance with the International Classification of Diseases, Ninth Edition (ICD-9) codes (010 to 018), or the International Classification of Diseases, Tenth Edition (ICD-10) codes (A15 to A19). All cases of TB were new infections and had no medical history with TB medical orders from January 1, 2009 through December 31, 2018. According to our defined criteria for SOT recipients, we ensured that they had not been infected with TB within the past 2 years, and had not undergone any other organ transplantation within the past 1 year.
4. Comorbidities
Comorbidities were defined in accordance with ICD-9 and ICD-10 codes, with at least two outpatient records or one hospitalization record. These comorbidities included stroke, coronary artery disease, congestive heart failure, hypertension, diabetes mellitus, dyslipidemia, chronic obstructive pulmonary disease (COPD), liver cirrhosis, and other types of hepatitis. Supplementary Table S1 lists the medical codes of these comorbidities. Calcineurin inhibitor (Anatomical Therapeutic Chemical [ATC]-7 codes: L04AD01 and L04AD02), antimetabolic agents (ATC-7 codes: L04AA06 and L04AX01), mammalian target of rapamycin inhibitors (mTORIs, ATC-7 codes: L04AA10 and L04AA18), and corticosteroids (ATC-7 codes: H02AB04, H02AB06, and H02AB07) were used to control the immune system. Patients who were prescribed these drugs within 90 days after transplant surgery were classified as drug users.
5. Statistical analysis
The transplant characteristics are expressed as the mean±standard deviation for continuous variables, such as age; and as proportions for categorical variables, such as comorbidities and sex. Survival was used to evaluate the factors that contribute to TB infection among transplant recipients. The index date was set as the date of transplant surgery. The date of TB infection, death, or the end of the study period was set as the end of follow-up. A Cox proportional-hazards model was used to identify the determinants of TB infection. A Kaplan–Meier survival analysis was conducted to estimate the survival function, while a log-rank test was used to test the homogeneity of the survival curves. TB was used as a time-varying variable in a multivariable model [9]. A stratified analysis was conducted to explore the association between TB and mortality among different sex, age group, and organ transplant recipients.
We further designed an age–sex–transplant duration-matched sub-cohort (n=760) with TB infection status (TB-infected [n=179] vs. non-infected [n=537]) among all transplants, with the index date being set as the date of TB infection. The index date of the control group (transplantation without TB infection) was identical to that of their TB-infected counterparts. The date of death or the end of the study period was set as the end of the follow-up period. The association between TB and mortality was examined using a multivariable Cox proportional-hazards model adjusted for comorbidities such as coronary artery disease, heart failure, hypertension, diabetes mellitus, dyslipidemia, COPD, hepatitis, and cirrhosis.
Results
1. Epidemiology of transplants
The average age of our SOT cohort was 50.3 years, with two-thirds of the cohort being men. Hypertension (56.58%) and cirrhosis (51.59%) were identified as the most common comorbidities in this cohort. Antimetabolic agents (86.64%) and mTORIs (28.71%) were used to address self-immune reactions (Table 1). Almost all patients were prescribed calcineurin inhibitor (97.49%, n=7,492) or corticosteroids (99.23%, n=7,626).
Characteristics of the study population
The patients with TB were older at 53.9 years versus 50.2 years age (p<0.0001), and more likely to be men at 73.13% versus 66.49% (p=0.0489), compared with their non-TB counterparts. Those with TB had a similar distribution of comorbidities to those without, with the exception of dyslipidemia. No significant differences were observed between those with and without TB in terms of the use of antimetabolic agents at 88.06% versus 86.60% (p=0.6006). However, the use of mTORIs was higher with TB than non-TB at 37.31% versus 28.47% (p=0.0072) (Table 1).
According to data on transplantation procedures performed in Taiwan during the period 2009−2018, the prevalence of medical conditions before transplant surgery among transplant recipients at that time was as follows: coronary artery disease, 20.87%; congestive heart failure, 15.32%; hypertension, 56.58%; diabetes, 33.04%; dyslipidemia, 31.97%; COPD, 18.61%; hepatitis, 49.25%; and cirrhosis, 51.96%. Of the patients who received heart transplants, more than 90% had congestive heart failure, and 70% had coronary artery disease (Table 1).
The organs involved in transplantation procedures were the heart (n=651, 8.47%), liver (n=4,267, 55.52%), kidneys (n=2,638, 34.33%), lungs (n=57, 0.74%), and pancreas (n=72, 0.94%). Heart (84.49%) and liver (72.72%) transplantation procedures were more common in male than in female patients (Supplementary Table S2). The comorbidities of organ transplant recipients are also provided (Supplementary Table S3).
2. Risk stratification for posttransplant TB after SOT
Age, sex, and comorbidities used during the first 90 days after SOT were included in the risk analysis (Table 2). Univariate analysis identified two risk factors, namely male sex, and age older than 65 years, and one possible protective factor, namely dyslipidemia (all p<0.05). Male sex and advanced age were identified as significant risk factors for incident TB. Male sex was associated with a 44% increase in the risk of TB (hazard ratio [HR], 1.44; 95% confidence interval [CI], 1.05 to 1.98). In addition, age older than 65 years was associated with a 4-fold increase in the risk of TB (HR, 4.04; 95% CI, 2.04 to 8.00). By contrast, dyslipidemia was associated with a 32% decrease in the risk of TB after SOT.
Determinants of tuberculosis in organ transplant recipients
Multivariate analysis revealed that male sex, advanced age, and dyslipidemia were significant independent determinants of TB after SOT. Male sex (HR, 1.42; 95% CI, 1.04 to 1.95; p=0.0268) and advanced age (HR, 4.02; 95% CI, 2.11 to 7.64; p<0.0001) served as independent risk factors for TB after SOT. By contrast, dyslipidemia served as a possible protective factor against TB after SOT (HR, 0.72; 95% CI, 0.52 to 0.99; p=0.0424) (Table 2).
3. Incidence and association of TB in SOT recipients
The median follow-up for SOT recipients was 5.18 years (interquartile range, 3.13 to 7.46). With a total of 33,853 person-years (PYs) of follow-up, 201 SOT recipients were identified as having newly diagnosed cases of TB, resulting in an overall incidence rate of 5.94/1,000 PYs. The incidence rate of TB was 8.06/1,000 PYs for heart transplants, 6.42/1,000 PYs for liver transplants, 4.84/1,000 PYs for kidney transplants, and 19.13/1,000 PYs for lung transplants, with an incidence rate of zero for pancreas transplants (Table 3).
Follow-up duration of TB among organ transplant recipients in Taiwan from 2009 to 2018 (n=7,685)
The incidence of TB decreased during the follow-up period. In patients who received heart transplants, the incidence rate of TB was 19.5/1,000 PYs (n=11) for the first year, 12.5/1,000 PYs (n=13) for the second year, and 11.83/1,000 PYs (n=17) for the third year. Similar trends of TB incidence were observed during the follow-up period among the recipients of various organs, except lung (Figure 2A).
Incidence of pulmonary (A) tuberculosis and (B) total mortality in organ transplant recipients. PY: person-years.
4. TB and mortality
With a total of 34,412.41 PYs of follow-up, 1,630 of 7,685 (21.21%) SOT recipients died, resulting in an overall mortality rate of 393.45/1,000 PYs. The mortality rates of those who received different organs of lung, heart, liver, pancreas, and kidney transplants were (187.75, 81.11, 58.47, 42.36, and 23.76)/1,000 PYs, respectively (Supplementary Table S4).
The annual mortality rates decreased during the follow-up period. In patients who received heart transplants, the annual mortality rate was 45.7/1,000 PYs (n=119) for the first year, 12.5/1,000 PYs (n=32) for the second year, and 5.5/1,000 PYs (n=14) for the third year. Similar trends of annual mortality were observed during the follow-up period among recipients of various organs (Figure 2B).
Posttransplant TB was associated with an increased risk of mortality in SOT recipients. The HR of TB as a time-varying variable for all-cause mortality was 2.53 (95% CI, 1.94 to 3.29). Multivariable model analysis revealed that male sex, advanced age, stroke, heart failure, diabetes mellitus, and cirrhosis were positively associated with mortality, whereas hypertension and dyslipidemia were negatively associated with mortality (Table 4). Analysis of the TB cohort with matched age, sex, and transplantation duration (n=760) revealed that the cumulative mortality rate of patients with posttransplant TB was significantly higher than that of patients without TB (Figure 3). In the sub-cohort with matched age, sex, and transplantation duration, patients with TB were at significantly higher risk of mortality, compared with those without TB (n=760 pairs). The HR of TB infection for overall mortality was 3.15 (95% CI, 3.04 to 4.85) with statistical power >99.99%.
Association between TB and mortality among transplant recipients
Survival curves of tuberculosis (TB) for mortality. In the sub-cohort with matched age, sex and transplantation duration, patients with TB were at a significantly higher risk of mortality compared with those without TB (n=760 pairs). The hazard ratio of TB infection for overall mortality was 3.15 (95% confidence interval [CI], 3.04 to 4.85) with the statistical power >99.99%.
Discussion
In this longitudinal study, we examined the incidence of posttransplant TB in SOT recipients in Taiwan. Our results indicated that posttransplant TB often occurred during the first year after transplantation, with the incidence of TB decreasing during the follow-up period. The incidence rate of TB was higher in lung transplant recipients, compared to the recipients of other organs. We also discovered that the incidence of TB among SOT recipients was higher than that of the general Taiwanese population. In addition, we explored the risk factors for posttransplant TB in SOT recipients. We also identified an association between posttransplant TB and mortality among SOT recipients in Taiwan.
TB remains a key concern in Taiwan, which has an incidence rate of approximately 33 per 100,000 individuals. Although this rate is lower than the global average [1], it considerably exceeds that of the United States, which has an incidence rate of 2.9 per 100 000 individuals [10]. In this study, we discovered that TB affected 201 SOT recipients with a total of 33,853 PYs of follow-up, indicating a total incidence of 594 per 100,000 individuals, which is approximately 20 times higher than that of the general population. This incidence rate is similar to that reported among SOT recipients in California for the period 2010−2020 for whom the TB incidence rate was 56.0 per 100,000 PYs, representing a 19-fold increase [11]. Research on a Spanish Network of Infection in Transplantation cohort revealed a TB risk of approximately 512 per 100,000 individuals among SOT recipients [12], which is approximately 27 times higher than that of the general population. These findings indicate that the risk of TB in SOT recipients in Taiwan is similar to that reported in previous studies, which have reported a TB infection risk that is approximately 20 to 74 times higher in SOT recipients than in the general population [13,14].
Our results indicate that the incidence of TB after organ transplantation is generally higher in the first year than in the second year, and higher in the second year than in the third year. However, among lung transplant recipients, no cases of TB were observed during the first three years. In patients who received heart transplants, the incidence of TB was 19.5/1,000 PYs in the first year, 12.5/1,000 PYs in the second year, and 11.83/1,000 PYs in the third year. Similar trends of TB incidence were observed during the follow-up period among the recipients of other organs. These results are consistent with those reported in other countries [7,13]. This early incidence may be attributable to cytotoxic T-cells, a key host defense system against mycobacterial infections, being compromised by the administration of the high doses of immunosuppressant drugs that are administered after SOT [13,15,16].
In this study, the average age of the SOT recipients ranged 45 to 55 years. Except for those who received heart transplants (approximately 85% men) and liver transplants (approximately 73% men), the sex ratio among the transplant recipients was relatively balanced. However, we discovered that TB was associated with varying levels of risk in the SOT recipients of different organs. Those who received lung transplants had the highest risk of TB (19.1 per 100,000), followed by those who received heart transplants (8.1 per 100,000) and kidney transplants (4.8 per 100,000). These TB risks correlated to some extent with the all-cause mortality rates of these recipients, with lung transplant recipients having the highest rate of mortality (187.8 per 100,000), followed by heart transplant recipients (81.1 per 100,000) and kidney transplant recipients (23.8 per 100,000). Taken together, these findings indicate that the risks of TB infection and TB-related mortality significantly vary across the recipients of different organs. We found that lung transplant recipients had the highest overall incidence and mortality with TB among SOT patients, distinct from the other transplanted organs. In other study, lung transplant recipients have shown a greater risk of active TB compared to other transplanted organs, with a 5.6-fold increased risk seen in a large Spanish cohort [12]. Therefore, special attention must be given to the risks of TB infection in the recipients of certain organ transplants, particularly those receiving lung transplants [17]. Continual monitoring for TB infection after transplantation is also essential in these populations.
Our results indicated that SOT recipients with posttransplant TB were at a 2.53 times higher risk of mortality compared to those without (HR, 2.53; 95% CI, 1.94 to 3.29; p<0.0001). This finding is consistent with Katrak et al. [11], who reported that SOT recipients with posttransplant TB in California had an adjusted HR of 2.8 for mortality (95% CI, 2.0 to 4.1). In our study, analysis of the cohort with matched age, sex, and transplantation duration confirmed that the cumulative mortality rate of patients with posttransplant TB was significantly higher than that of patients without. This finding may be attributable to the immunosuppressive therapy administered to SOT recipients, or the interactions between anti-TB therapy and other medications [11]. Taken together, these findings underscore the strong effect of TB on the survival of transplant recipients, emphasizing the need for effective monitoring and management for TB in this vulnerable population.
Among the risk factors for active TB in the general population are age [18,19], male sex [19,20], undernutrition [19,21], diabetes mellitus [22,23], smoking [24], and alcohol abuse [25]. In SOT recipients, diabetes mellitus, chronic kidney disease, chronic liver disease, and concomitant infections serve as independent risk factors for posttransplant TB [5,26]. Overall, our findings are consistent with those reported in other countries. In the general population, TB is more commonly diagnosed in men, with a global male-to-female ratio of 1.66:1 [1]. Advanced age is a known risk factor for TB, partly due to age-related comorbidities, and progressive decline in pulmonary and immune function. Structural lung changes and immunosenescence—marked by reduced naïve T-cell output and impaired immune memory—further increase susceptibility to Mycobacterium tuberculosis infection [18,27,28]. Consistent with prior findings, our study demonstrated that older age was associated with a higher risk of posttransplant TB among SOT recipients.
In this study, we discovered that dyslipidemia was negatively associated with posttransplant TB in SOT recipients. This negative association between dyslipidemia and the risk of TB could be explained by the key role of cholesterol in immune function, and its correlation with a positive nutritional and socioeconomic status. Low total cholesterol levels are significantly associated with an increased risk of TB, particularly in individuals under 65 years of age who do not have diabetes or obesity, and do not use statins. Cholesterol is a crucial component of cell membranes that is essential to maintain membrane fluidity and support various cellular functions [29,30]. However, previous studies have shown that among kidney transplant patients, those with TB exhibited higher rates of dyslipidemia, compared to those without TB [31]. This finding is somewhat inconsistent with the results of our study. Possible reasons for this discrepancy include our transplant patient cohort encompassing all types of organ transplant recipients, such as lung transplant patients, and the different time period from which our data were drawn. Therefore, further research is warranted to confirm the association between dyslipidemia and TB risk in organ transplant patients.
This study has several strengths. First, given that the pool of organ transplant recipients is limited, extensive and long-term databases are essential to provide the analytical data required for investigations involving such recipients. Since 1996, the National Health Insurance Research Database of Taiwan has maintained comprehensive records on these patients [32]. Therefore, the database provides information with a sufficient follow-up duration, which aided in the present evaluation of the incidence of TB and its effect on mortality in SOT recipients. Second, because the incidence of TB varies over time for each organ transplant recipient, we employed a time-varying variable in our multivariable model. By incorporating data on the incidence of TB at various time points, we were able to more comprehensively evaluate the effects of TB [9]. Third, this study is one of the few studies that have revealed the effect of TB on the risk of mortality in SOT recipients. We conducted various analyses after matching for age and sex to provide a clear view of the actual risks of mortality associated with TB. Despite the strengths we have mentioned, this study has some limitations. First, we did not include data regarding socioeconomic status or detailed medical record; social history, such as smoking, alcohol consumption, or illicit drug use; whether living or deceased donor transplantation was conducted; the number of episodes of rejection; the grade of rejection; or the level of human leukocyte antigen matching. Second, we were unable to differentiate between the reactivation of latent TB, de novo infection, and donor-derived TB infection through donor allograft based on our data sets. Third, we excluded patients with TB who received their diagnoses 2 years before transplantation; our 10-year data sets did not allow for the identification of TB incidence beyond our study period. Fourth, organ transplant candidates usually undergo screening for latent tuberculosis infection before transplantation; however, after transplantation, the TB treatment outcomes among the patients included in this study may not be available. Fifth, some comorbidities, such as chronic kidney disease, may not be available in this cohort.
In conclusion, patients with posttransplant TB are at a 2.53 times higher risk of mortality, compared with those without. Therefore, physicians should be more vigilant during the first year after transplantation, because TB infections often occur during this period. Optimal use of immunosuppressive agents to minimize acute rejection and monitoring of high-risk recipients are recommended to manage TB infections in endemic areas, such as Taiwan. In cases of clinical or radiological suspicion of TB infection in a SOT recipient, prompt diagnosis and appropriate anti-TB treatment are essential.
Notes
Authors’ Contributions
Conceptualization: Chuang SY, Tang CF, Lin KH, Lai CH, Yin YT, Tsai MK, Wu MS, Wu MY. Methodology: Chuang SY, Tang CF, Lin KH, Lai CH, Yin YT, Tsai MK, Wu MS, Wu MY. Formal analysis: Chuang SY, Lai CH. Data curation: Chuang SY. Funding acquisition: Wu MS, Wu MY. Project administration: Wu MS, Wu MY. Visualization: Lin KH, Yin YT, Wu MS, Wu MY. Software: Chuang SY, Lai CH. Validation: Chuang SY, Lai CH. Investigation: Lin KH, Yin YT, Wu MS, Wu MY. Writing - original draft preparation: Chuang SY, Tang CF, Tsai MK, Wu MY. Writing - review and editing: Chuang SY, Tang CF, Lin KH, Lai CH, Yin YT, Tsai MK, Wu MS, Wu MY. Approval of final manuscript: Chuang SY, Tang CF, Lin KH, Lai CH, Yin YT, Tsai MK, Wu MS, Wu MY.
Conflicts of Interest
No potential conflict of interest relevant to this article was reported.
Acknowledgments
Data are from the National Health Insurance Research Database (NHIRD), published by the Taiwan National Health Insurance (NHI) Bureau. However, due to legal restrictions under Taiwan’s Personal Information Protection Act, the data cannot be publicly disclosed.
Funding
This work was partly funded by National Science and Technology Council in Taiwan (NSTC) (grant no. 109-2314-B-038-106-MY3 and 113-2314-B-038-030). Study sponsors had no role in study design; collection, analysis, and interpretation of data; writing the report; and the decision to submit the report for publication.
Supplementary Material
Supplementary material can be found in the journal homepage (http://www.e-trd.org).
Definitions and medical codes for comorbidities.
Age and sex distribution of organ transplant recipients in Taiwan from 2009 to 2018 (n=7,685).
Distribution of comorbidities of organ transplant recipients in Taiwan from 2009 to 2018 (n=7,685).
Follow-up duration until mortality among organ transplant recipients in Taiwan from 2009 to 2018 (n=7,685).