Tuberculosis (TB) is a significant clinical and public health problem for patients with chronic kidney disease (CKD), particularly those requiring hemodialysis (HD). This population is particularly vulnerable due to the profound alteration of the immune response characteristic of CKD, which significantly increases the risk of reactivation of latent TB.
The Mantoux test, for example, is not suitable for diagnosis due to the low sensitivity of these patients and the high rate of false positives due to vaccination. In contrast, IGRAs (interferon-gamma release assays) show superior sensitivity and specificity for diagnosing latent TB. Similarly, the high frequency of extrapulmonary manifestations of TB in dialysis patients must be taken into account, as this can lead to a delay in diagnosis and therefore favor nosocomial transmission. This is why it is essential to maintain a high level of clinical suspicion and make appropriate use of the available complementary tests in order to establish an early diagnosis and initiate targeted treatment to reduce morbidity and mortality.
Regarding treatment in the general population, short, oral regimens are recommended for both drug-susceptible and drug-resistant TB in eligible individuals; however, there are no formal recommendations on the efficacy and/or safety of these regimens for the treatment of pulmonary tuberculosis in hemodialysis patients.
La tuberculosis (TB) representa un problema clínico y de salud pública relevante en pacientes con enfermedad renal crónica (ERC), especialmente en aquellos que precisan tratamiento con hemodiálisis (HD). Esta población es particularmente vulnerable debido a la profunda alteración de la respuesta inmunitaria propia de la ERC, lo que aumenta significativamente el riesgo de reactivación de la TB latente.
Para el diagnóstico, las pruebas de distribución universal como el test de Mantoux, muestran una sensibilidad muy baja debido a la anergia de estos pacientes, así como una alta tasa de falsos positivos debido a la vacunación; mientras que los ensayos de liberación de interferón-gamma (IGRAs) han demostrado una sensibilidad y especificidad superiores para el diagnóstico de TB latente. Además, es importante tener en cuenta la elevada frecuencia de manifestaciones extrapulmonares de la TB en pacientes en diálisis, lo que conlleva un retraso en el diagnóstico, y favorece por tanto la transmisión nosocomial. Es fundamental tener una elevada sospecha clínica, haciendo uso adecuado de las pruebas complementarias disponibles, para así poder establecer un diagnóstico precoz y poder iniciar un tratamiento dirigido que permita reducir la morbi-mortalidad.
En cuanto al tratamiento en la población general se recomiendan regímenes cortos y orales tanto para la TB sensible como la resistente a fármacos en individuos elegibles, sin embargo, no hay recomendaciones formales sobre la eficacia y/o seguridad de estos esquemas para el tratamiento de TB pulmonar en pacientes en HD.
Patients with chronic kidney disease (CKD), particularly those on hemodialysis (HD), exhibit a significant impairment of the immune response, primarily associated with uremia, which renders them more susceptible to the development of both acute and chronic infectious processes.1 Tuberculosis (TB) is a disease with increasing incidence and mortality, especially in certain vulnerable populations and in developing countries, where it often goes undetected, generating a major public health problem.2 The global incidence of TB in HD patients is estimated to be 10–20 times higher than in the general population (between 5000 and 6000 cases per 100,000 inhabitants/year),3 with a risk of developing active forms up to seven times greater and a mortality ranging from 17% to 75%.2
Latent TB occurs when the bacillus (Mycobacterium tuberculosis) is present without clinical, radiological, and/or microbiological evidence of active disease; therefore, it is not a contagious form of the disease. Conversely, active TB refers to evidence of infection by the bacillus, associated with clinical symptoms secondary to its multiplication and organ damage, which is in turn apparent through compatible radiological/microbiological findings.4
In this population, the diagnosis of TB poses a challenge due to the nonspecific nature of the associated symptoms,2 the low sensitivity of traditional diagnostic tests, and the high frequency of extrapulmonary manifestations (between 60% and 80% of cases),5 leading to a delay in diagnosis and, consequently, in appropriate treatment.6,7
The aim of this article is to review the available scientific literature from a practical standpoint, delving into discrepancies in the evidence. To this end, a real clinical case will be used as the guiding thread of this narrative review, and a search of the published evidence will be conducted.
EpidemiologyWe present the case of a 62-year-old male from the Philippines with a history of CKD stage G5 on HD who begins a study protocol for inclusion on the kidney transplant waiting list.
It is estimated that approximately one-quarter of the world’s population is infected with latent TB, of whom 10% to 15% will develop active disease during their lifetime.8 However, the prevalence of the disease is higher in resource-limited regions where population density and socioeconomic factors favor its spread, reaching figures of up to 500 cases per 100,000 inhabitants in Africa and much of Asia (mainly India and China), with prevalence also being notable in Latin American countries such as Brazil, Peru, and Haiti. In Spain, TB has a significantly lower incidence than in endemic countries, of approximately 100 per 100,000 inhabitants; however, prevalence is influenced by migratory dynamics from countries with a high disease burden.
Clinical manifestationsAt the time of initiating HD in our unit, the patient was asymptomatic from a respiratory standpoint and reported symptoms of asthenia and hyporexia of several months’ duration.
In 80% of cases, active TB occurs through reactivation of latent TB, presenting asymptomatically. The risk of reactivation increases as CKD progresses, being highest in patients requiring kidney replacement therapy (KRT).3
The clinical presentation of active TB in HD patients tends to be insidious and atypical. Symptoms such as cough or hemoptysis are less frequent, with other symptoms predominating that may even mimic the underlying uremic state, such as fatigue, malaise, anorexia, weight loss, and low-grade fever.9 In part, this occurs due to the inherent involvement of CKD itself and also because of the high prevalence of extrapulmonary involvement of this disease (either in isolation or coexisting with pulmonary TB), primarily in the form of lymphadenitis, intestinal TB, pleural/pericardial effusion, genitourinary involvement, bone involvement, and miliary TB.10,11 Furthermore, TB itself may be the etiology of CKD and on numerous occasions goes unnoticed, driving its progression and even constituting a cause of active TB at the time of KRT initiation.12
Diagnostic considerationsAvailable diagnostic methodsAs part of the study for inclusion on the kidney transplant waiting list, abdominal ultrasound revealed hyperechoic nodular lesions in the liver, prompting further evaluation with a thoraco-abdomino-pelvic computed tomography scan (Fig. 1), which showed similar lesions in the pulmonary parenchyma suggestive of a neoplastic versus inflammatory process.
Early and accurate diagnosis of TB is of vital importance in patients with CKD on HD. This not only improves individual patient outcomes by allowing timely treatment, but is also essential for preventing disease transmission in healthcare settings, as well as reducing the overall disease burden in this vulnerable population.10,13 The combination of a high index of clinical suspicion, risk factor assessment, and the strategic use of the most sensitive available diagnostic tests are crucial aspects for effective TB management in patients with CKD.14
Added to the diagnostic challenge posed by the nonspecific nature of the symptoms is the limited sensitivity of traditional tests (such as the Mantoux test) due to the high prevalence of anergy in these patients (up to 50% false negatives)15 and the occurrence of false positives in those vaccinated with Bacillus Calmette-Guérin (BCG)10,16 (Table 1).
Sensitivity and specificity of the different available tests in hemodialysis patients.
| Test | Sensitivity | Specificity | Remarks |
|---|---|---|---|
| Tuberculin skin test (TST)/Mantoux test | Very low/< 50%/31% due to anergy inherent to CKD | Low/63%: due to BCG vaccination and exposure to non-tuberculous mycobacteria. | Increasingly less used since the adoption of IGRA for TB diagnosis |
| Interferon-Gamma Release Assays (IGRAs) | |||
| T-SPOT.TB | 91.7% for active forms (HD) 73.1%–78.6% for latent forms | 64.7% for active forms; 61.2% for latent forms | Considered more sensitive than QFT-G in some general studies. |
| QuantiFERON-TB Gold (QFT-G/QFT-GIT) | Some studies report sensitivity of up to 100% in patients with active TB and between 46%–56% for latent TB | 89.7% for active TB; 75.5% for latent TB. | In HD, QFT-G is the most widely used and may be more sensitive in some patients. Rates of indeterminate results range from 5% to 40% |
| Sputum smear microscopy (AFB) | No conclusive data available | No conclusive data available | Rapid and inexpensive method. Requires at least one positive result for smear-positive pulmonary TB. Less sensitive than culture. |
| Mycobacterium tuberculosis culture | No conclusive data available | No conclusive data available | "Gold standard" for the diagnosis of active pulmonary TB. Results may take up to 12 weeks. |
| Nucleic Acid Amplification Tests (NAAT/PCR, Xpert MTB/RIF) | ∼80% (in AFB-negative sputum) In extrapulmonary specimens, variable: 83%–96% lymph node; 30%–46% pleural fluid | No conclusive data available | Rapid diagnosis with better accuracy than smear microscopy. Xpert MTB/RIF detects the Mycobacterium tuberculosis complex and rifampicin resistance in 2 h. |
TB: tuberculosis; CKD: chronic kidney disease; BCG: Bacillus Calmette-Guérin; HD: hemodialysis.
Interferon-gamma release assay (IGRA)-based tests have demonstrated superior sensitivity and specificity compared with the Mantoux test in this population.10 However, these assays are not infallible and may produce indeterminate results, particularly in patients with anemia, hypoalbuminemia, and those with longer time on dialysis.10 The introduction of rapid molecular tests, such as the Xpert MTB/RIF assay,17 which simultaneously detects M. tuberculosis and rifampicin resistance, has improved diagnostic accuracy.10 Definitive diagnosis is often based on the isolation of acid-fast bacilli (AFB) in a representative sample, the presence of caseating granulomas on biopsy, or growth of the mycobacterium in culture. Nevertheless, up to 20% of clinically diagnosed TB cases never achieve bacteriological/histological confirmation (Table 1).18
Diagnostic challenges in the hemodialysis patientLiver lesion biopsy was performed for pathological examination, which revealed nonspecific granulomas without features suggestive of malignancy; therefore, fiberoptic bronchoscopy was performed, with samples obtained from paratracheal lesions and a tracheal ulcer (Fig. 2) that were sent for microbiological and pathological analysis.
Absence of a "gold standard" for latent tuberculosis18
Indeterminate results of the Mantoux test and IGRA. Both tests are based on the host immune response to purified antigens, which poses a challenge in patients with CKD/on dialysis. However, between the two tests, the IGRA has shown greater sensitivity as it is an in vitro assay in which immune cells are separated from the uremic environment of the patient, thus potentially functioning more adequately.19
Atypical presentation and need for high clinical suspicionImpact of ethnicity/country of origin: Patients of Asian/British-Asian origin or from India, South America, and countries with limited access to healthcare services have a significantly higher risk of developing TB.8
Time on HD: The risk of active TB is significantly higher in patients with less than 12 months on HD (it tends to appear during the first months after initiating dialysis).14
Age and comorbidities: Advanced age (>65 years), diabetes mellitus, low body mass index (BMI), and primary TB lesions on chest radiograph are independent risk factors for the development of active forms of the disease in HD patients.14,20
Proposed algorithm for tuberculosis screeningGiven the elevated risk of developing active forms/reactivation of latent TB, the following aspects should be considered for ruling out this disease in patients with CKD (Fig. 3):
Proposed diagnostic algorithm for tuberculosis in hemodialysis patients.
TB: tuberculosis; CXR: chest radiograph; HD: hemodialysis; IGRA: interferon-gamma release assay.
* There is insufficient consensus regarding screening all patients initiating in-center hemodialysis; however, some centers have implemented this measure7 to prevent nosocomial transmission.
** These tests are available almost globally. In low-resource centers or those without access to them, the Mantoux test may be considered, bearing in mind its limitations and the need for confirmatory tests in case of a positive result.
*** There is insufficient consensus regarding the interval at which screening should be repeated; some authors suggest it should be performed semiannually, others support annual screening, and it has also been proposed that repetition is unnecessary unless the aforementioned criteria are met.
Diagnosis will only be cost-effective and ethical in those patients who are candidates for active treatment, taking into account the implications of prolonged treatment and the side effects in this patient profile.19
High-risk populations: It is important to focus screening and treatment efforts on CKD patients who present a higher risk of TB exposure: patients older than 65–70 years, from endemic areas, with a history of diabetes mellitus and/or prior TB, chronic malnutrition, and an additional immunosuppressive component.18,21
Kidney transplant candidates: Screening is imperative for patients who are candidates for kidney transplantation before surgery, primarily because of the risk associated with immunosuppression.2
Some guidelines suggest considering screening of young patients with a prolonged life expectancy.18
It has been observed that, on average, the median time from HD initiation to active TB diagnosis is 300 days.14
Diagnostic tools for latent tuberculosisAccording to the patient’s context and the availability of resources in the setting, the use of the tuberculin skin test (Mantoux) versus IGRA may be considered, bearing in mind, as previously mentioned, the sensitivity and specificity of both tests in HD patients; however, interferon-gamma release assays remain the test of choice in cases of suspicion and/or screening.22
Ruling out active tuberculosisComprehensive clinical evaluation: A positive IGRA result does not distinguish between latent and active TB.10
Methods for ruling out active tuberculosisSymptoms and signs: The presence of TB symptoms and signs should be investigated, despite how nonspecific they may be.2
Chest radiograph: This is a key component of the evaluation. Typical findings of active TB include focal infiltrates in the upper lobes, cavitation, and hilar lymphadenopathy, as well as fibrosis of variable distribution.18 However, the frequent extrapulmonary involvement in CKD patients must be taken into account, which may hinder diagnosis when chest radiography is used exclusively.10,23
Clinical specimens: If suspicion exists, sputum or other clinical specimens should be obtained for AFB smear and M. tuberculosis culture.14 Definitive confirmation of active TB requires isolation of the bacillus, the finding of caseating granulomas on biopsy, or a positive culture.10
TreatmentFinally, fiberoptic bronchoscopy revealed AFB in the specimen from the tracheal ulcer, as well as granulomas in the paratracheal nodules on histological examination. M. tuberculosis susceptible to isoniazid and rifampicin was isolated from bronchoalveolar lavage culture, thus confirming the case of active TB. The patient was jointly evaluated with the Infectious Diseases service, and a four-drug protocol was initiated for two months, followed by a maintenance phase of initially four months, with the possibility of extending treatment depending on response and tolerance.
Antituberculous treatment, both prophylactic and therapeutic, has undergone important updates. However, in HD patients, the situation entails considerations inherent to the immunosuppressed state and other associated factors.24 Preventive treatment in individuals with latent infection reduces the risk of developing active TB by approximately 90%.10,25
With regard to the most recent recommendations, shorter regimens have been proposed for the treatment of pulmonary TB; however, patients with CKD and/or on HD were excluded from these studies, so formal recommendations regarding shortening the standard regimen cannot be made.24
The treatment of TB and how it differs in HD patients is detailed below (Table 2):
Proposed treatment for active and latent tuberculosis in hemodialysis patients.31
| Active tuberculosis | Latent tuberculosis | |
|---|---|---|
| Standard treatment | New proposal (shortened regimen)12 | |
| INITIAL PHASE (2 months): | Only proposed for pulmonary tuberculosis; however, NOT APPLICABLE TO HEMODIALYSIS PATIENTS. The efficacy and/or safety of shortened regimens has not been studied in these patients. | Short-duration regimens: |
|
| |
| SECOND PHASE: isoniazid + rifampicin, same dose for: 4 months (6 months total): pulmonary, pericardial, lymph node, pleural, peritoneal, genitourinary, and disseminated TB. 7 months (9 months total): osteoarticular TB, BMI < 18.5 kg/m², and disseminated TB with insufficient initial response. 7–10 months (9–12 months total): tuberculous meningitis. | Long-duration regimens: Isoniazid (300 mg/day) + pyridoxine 25–50 mg: for 6–9 months. | |
| In cases where pyrazinamide cannot be used (hepatotoxicity, documented resistance): | Always administer post-dialysis | |
| ||
| TB: tuberculosis; BMI: body mass index. | ||
| ||
Dosing and drug adjustment: The treatment of active TB in HD patients generally follows standard regimens, but doses must be adjusted for drugs that are eliminated via the renal/HD route, such as ethambutol and pyrazinamide.26 Isoniazid and rifampicin can generally be administered at unadjusted doses (hepatic elimination). All antituberculous drugs should be administered after HD to prevent their premature removal.24
Some authors suggest that ethambutol should only be administered in these patients if drug levels can be monitored.24
Rifampicin may have pharmacological interactions with other medications, such as antihypertensives, anticoagulants, and immunosuppressants. In general, it should be noted that, as a cytochrome P450 inducer, it may lead to the rapid metabolism of other concomitantly used drugs.2
Side effects: HD patients are more prone to experiencing side effects with antituberculous medications.27
Isoniazid is most frequently associated with neuropsychiatric symptoms and ototoxicity, and the risk of hepatitis/hepatotoxicity increases with age.2
Pyrazinamide may accumulate metabolites and cause hyperuricemia and gout; however, greater safety is achieved by decreasing the frequency of administration, and periodic measurement of plasma levels is suggested if toxicity is suspected.28 Nevertheless, regimens without pyrazinamide have been shown to be less efficacious; therefore, unless contraindicated or hepatotoxicity is suspected, it should be included in the therapeutic arsenal.28
Ethambutol may cause ocular toxicity,29 and is therefore administered at lower doses. However, according to the most recent recommendations, this drug is no longer an essential component of standard regimens.24
Although there are limited data on the pharmacokinetics of rifapentine in patients with CKD, some single-center studies have successfully used this regimen in HD patients awaiting kidney transplantation.19
Treatment duration: In some cases, if there is a delayed response to treatment, extension of the regimen to 39 weeks may be considered.12,30
Conclusions and future directionsHD patients have a 10- to 20-fold higher risk of TB compared with the general population, with significantly elevated mortality. Diagnosis is challenging owing to the atypical presentation, with a high frequency of extrapulmonary forms, and the low reliability of the Mantoux test due to anergy in the uremic patient. Modern tests such as IGRAs and molecular assays (Xpert MTB/RIF) are superior for the detection of both latent and active TB in this population.
Key risk factors have been identified, including advanced age, diabetes, low BMI, and previous radiographic lesions, as well as the first 12 months after HD initiation. The frequent diagnostic delay increases the risk of nosocomial transmission in HD units, which justifies selective screening for latent infection in the highest-risk groups to prevent reactivation. Treatment requires dose adjustments and strict monitoring, with standardized guidelines for this vulnerable population being crucial.
For these reasons, it would be of interest to consider the following aspects for future studies:
Optimize diagnostic methods for latent TB and increase evidence on which CKD patient subgroups would benefit from latent TB screening.
Most protocols suggest performing chest imaging to rule out active forms of the disease; however, given the high prevalence of extrapulmonary forms, it would be pertinent to begin performing other tests aimed at ruling out TB at sites beyond the pulmonary system.
Evaluate the efficacy of new treatment regimens for extrapulmonary and disseminated forms of the disease.
Prospectively validate a protocol combining clinical risk assessment with IGRA results to improve predictive value.
Assess the effect of CKD stage on the performance of screening tests.
Determine the impact of nosocomial transmission compared with endogenous reactivation.
FundingMAB and BQ research are funded by Instituto de Salud Carlos III (ISCIII): RICORS program to RICORS2040-Renal (RD24/0004/0028) co-funded by European Union (FEDER funds) and public research grants (PI25/00413).
The authors declare that they have no conflicts of interest.
