The present study assessed the value of TST and QFT in haemodialysis patients after prolonged exposure to a highly contagious pulmonary TB patient. The results obtained showed three major findings. First, a single TST has low sensitivity in detecting LTBI and a two-step strategy must be carried out to gain sensitivity. Second, erythema alone (without induration) may be a valuable marker when interpreting TST in haemodialysis patients. And third, the QFT shows better sensitivity than the TST in detecting LTBI in haemodialysis patients.
TST has been shown to be unreliable in patients with advanced chronic kidney disease, and although a positive test may be useful for LTBI diagnosis, a negative one cannot be assumed to be a true negative [8
]. Uraemia is a well-known risk factor for impaired immune cellular response [9
], and this fact is decisive for both a lower sensitivity of the TST and a higher risk of progression from LTBI to TB disease [12
]. It has been shown that immune cells can be activated in ESRD patients, but the functionality of these cells is impaired. This is especially true for the subpopulation of T helper 1 (TH1) lymphocytes, which are the cells mainly responsible for cellular immunity and delayed hypersensitivity responses. Moreover, there is a decrease in the number of B cells in a pro-inflammatory environment (complement activation, inflammatory cytokines), together with an imbalance in the ratio of TH1/TH2 lymphocytes [20
]. The suppression of the immune TH1 status entails a higher rate of anergy, eventually leading to lower positive TST responses compared to the QFT results. Although limited by the absence of a gold-standard method to confirm LTBI, in our study QFT has shown a higher response than TST, which approaches statistical significance.
Two-step testing in TST has been shown to be crucial for LTBI diagnosis in patients with ESRD, in the absence of a known recent exposure to a contagious index case [21
]. In our study, 8 patients with an initially negative TST had a second positive TST 15
days later; thereby, highlighting the importance of the booster effect in ESRD patients. The booster effect is believed to result from recall of waned cell-mediated immunity, akin to the anamnestic response. A first tuberculin test may boost cell-mediated immunity in patients with otherwise impaired immune response, such as patients with ESRD [23
]. Several studies have addressed the booster phenomenon in ESRD patients without known recent exposure to a TB index case [21
]. Cengiz et al. detected a boosted response in a second TST, performed seven days after a first TST, in 24.3% of patients on haemodialysis [22
]. Similarly, Habesglu et al. found the booster phenomenon in 29.7% of patients undergoing haemodialysis treatment and with no epidemiologically recognised factors for LTBI [25
]. Our study has addressed the booster response in the context of a recent exposure to a highly contagious TB index case. Therefore, it could be difficult to distinguish between a booster effect from a real conversion after the tuberculin “window period”. Nevertheless, this is very unlikely, because both TST and QFT were performed 12
weeks after the last possible exposure to the Index Case, which would make very unlikely a conversion attributable to the “window period” [27
]. Our findings support the view that two-step TST must be performed in patients with ESRD, even in the context of a recent exposure to M. tuberculosis.
An unexpected finding of our study was the high number of patients with erythema, but without induration, in the TST responses. Erythema without induration is not generally considered in the interpretation of the TST in most International Guidelines. However, some Japanese guidelines suggest taking erythema without induration into consideration when the measurement is over 20
]. In a contact tracing study carried out with 566 BCG-vaccinated school students in Japan, the erythema measurement was correlated with both the induration measurement and the degree of exposure to the index case.
These results suggest that erythema (irrespective of the induration response) could have a significant value in the interpretation of the TST results. However, in this study, no concomitant IGRA testing was performed [29
]. In our study, joint assessment of induration and erythema has shown better correlation with the QFT results than when only induration was taken into account in the TST response. All 3 patients with erythema, but with a negative result in the initial TST-1, developed a positive TST response 15
days later. Moreover, of the 9 patients with erythema without induration in the second TST, six (66.7%) had a positive QFT result. Overall, our results suggest that in patients with impaired cell-mediated immune response, such as patients with ESRD, erythema without induration could be indicative of a weak immune response to M. tuberculosis
antigens, even though it did not yield skin induration. Our findings indicate that M. tuberculosis
infection could be considered in patients with ESRD and with no induration erythema shown in the TST response, particularly in situations of high epidemiological risk of infection. Nevertheless, more studies and a higher number of patients than those presented in our study are required to fully verify this statement.
As there is no gold-standard method to confirm diagnosis of LTBI, the accuracy of tests for evaluating LTBI relies on indirect evidence of infection. There is increasing evidence that IGRA results have a better correlation with known risk factors for LTBI than the TST responses [13
]. In addition, several studies have assessed this issue in patients with chronic kidney disease, indicating that for these groups of patients, IGRAs have a greater accuracy than TST in the diagnosis of LTBI [30
]. However, only a few studies have compared TST and IGRA results in haemodialysis patients after a recent exposure to a bacilliferous TB patient [16
]. Winthrop et al. found a better correlation with risk exposure using Quantiferon-TB-Gold® test (an older generation of QFT) than when using the TST after exposure to a pulmonary TB patient in a haemodialysis centre [16
]. Yet, they found that the rates of positive results using the TST and the IGRA were not significantly different, whereas we found a value close to statistical significance (p
0.065). The conclusion derived from the above cited study and from our current findings (using the new generation of QFT) is that IGRAs may offer greater accuracy for the diagnosis of LTBI in ESRD patients, after exposure to an infectious TB patient [14
Our study has several limitations that require further comments. First, due to the limited number of patients included in the study, our results must be interpreted with caution, and larger series of patients are necessary to confirm our findings. This is especially true for the value of erythema, as very few studies have previously assessed this item. Second, as there is no gold-standard method for diagnosing true LTBI, it cannot be ruled out that higher positivity rates for the QFT could include some false positive results. Nevertheless, there is increasing evidence about the value of an IGRA as a marker for LTBI, and even as a marker for the risk of progression to active TB [35
]. Finally, only one commercial IGRA (QFT) has been evaluated in our study. It has been suggested that the other commercially available IGRA (T-SPOT.TB®) may have an improved sensitivity in patients with immunodeficiency disorders [36
]. There are several previous studies that have compared different commercially available IGRAs in ESRD patients [30
]. In a Korean study comparing the performance of TST, QFT and T-SPOT.TB in patients undergoing haemodialysis, the positive rates were 23.5%, 45.9% and 60.4% respectively, suggesting a higher sensitivity for the T-SPOT.TB test. In addition, the frequency of indeterminate results was higher for the QFT compared with the T-SPOT.TB test [32
]. In contrast, two other studies failed to demonstrate significantly better sensitivities with the T-SPOT.TB test than with the QFT: 47% vs. 40%, respectively, using an older generation of QFT in a Taiwanese study [30
], or 22% vs. 46%, respectively, in 62 Swiss patients deemed to have LTBI [33