Background
Tuberculous lymphadenitis is one of the most common manifestations of extrapulmonary tuberculosis (TB) [
1‐
4]. Although the total number of TB cases is declining in developed countries, the cases of extrapulmonary TB are not significantly decreasing, and such condition accounts for approximately 15–20% of all TB cases [
1]. Approximately 30–50% of all extrapulmonary TB cases involve the peripheral LNs, of which cervical LN involvement is most common [
2‐
5]. In Korea, 51.5 million individuals were newly diagnosed with TB in 2018 according to the declining trend. The proportion of extrapulmonary TB increased up to 21%, and tuberculous lymphadenitis accounted for 20% of extrapulmonary TB [
6].
Although the duration of treatment for tuberculous lymphadenitis has been controversial, a six-month treatment is recommended for drug-susceptible organisms based on studies showing no significant differences in treatment failure or complication [
7‐
9]. However, evidence supporting this recommendation for all cases of tuberculous lymphadenitis is limited because, unlike in pulmonary TB, there is no clear criteria for the assessment of extrapulmonary TB after treatment. The characteristics of extrapulmonary TB differ from those of pulmonary TB. First, the rates of drug susceptibility tests are relatively low because the rate of positive culture is low. Second, the response to antituberculous treatment is delayed, and paradoxical reactions occur more frequently in individuals with extrapulmonary TB than in those with pulmonary TB. Residual LNs may be observed in 12–30% of patients after antituberculous treatment [
5]. Therefore, clinicians often encounter residual LNs after the six-month treatment, and there are controversies in relation to the interpretation of residual LNs. Some studies have recommended that an additional therapeutic approach is required due to the possibility of recurrence or resistance, and some have argued that prolonged treatment is not necessary because paradoxical reactions may occur [
10‐
12]. Meanwhile, recent studies have shown that not all residual LNs indicate treatment failure. Additional treatment is recommended on the basis of microbiological testing results showing that LN is still found after short-term observation.
However, the evaluation of treatment failure in all residual LNs is challenging due to the limitations in terms of invasive tests and low diagnostic yield. Non-invasive modalities used for the evaluation of residual LNs after antituberculous chemotherapy are required. One study has indicated that computed tomography (CT) scan of the neck may be useful for the evaluation of treatment response [
13]. A comparison of neck CT scan results before and after therapy for cervical tuberculous lymphadenitis (CTBL) has shown that the incidence of central necrosis, perinodal infiltration, and peripheral rim enhancement was significantly lower in the treatment success group than in the treatment failed group.
We described the clinical, radiological, microbiological and pathologic characteristics of residual LNs after six-month of antituberculous therapy in human immunodeficiency virus (HIV)-negative individuals with CTBL.
Methods
Study design and population
A prospective study was conducted to identify the clinical course and characteristics of CTBL. We enrolled adult patients older than 18 years who were diagnosed with CTBL between July 2009 and December 2017 at the Korea University Ansan Hospital, a 780-bed tertiary teaching hospital in Ansan, Republic of Korea. We excluded patients with HIV infection. All patients with CTBL were diagnosed via fine-needle aspiration biopsy (FNAB); the patients with CTBL who presented with cervical LN involvement only after completing the first-line antituberculous treatment were re-assessed. Such treatment comprised isoniazid (5 mg/kg), rifampin (10 mg/kg), ethambutol (25 mg/kg), and pyrazinamide (15–20 mg/kg) daily for the first 2 months, followed by isoniazid, rifampicin, and a reduced dose of ethambutol (15 mg/kg) for the next 4 months. Follow-up CT scans were performed within 1 month after six-month of antituberculous treatment. Re-FNAB was performed in patients with residual LN on follow-up CT scans if possible. The treatment of patients who maintained antituberculous drugs due to residual LNs was discontinued after confirming that Mycobacterium tuberculosis was not isolated from the culture. We compared the characteristics of the patients with and without residual LNs. The study protocol was approved by the institutional review board of Korea University Ansan Hospital (no. 2009AS0050). Written informed consent was obtained from each study participant.
Definition
The diagnosis had been confirmed either by the detection of acid-fast bacilli (AFB) in smears from FNA or from biopsy and/or by positive mycobacterial culture of aspirates or biopsy and/or by histological evidence of TB. The histologic findings indicative of TB included necrotizing or caseating granuloma and necrosis without granuloma. When only cellular components were observed on cytological testing, TB was defined as the presence of histiocytes, multinucleated giant cells, and infiltrating lymphocytes [
14]. The CT findings indicative of TB include central low density and peripheral rim enhancement with thick and irregular patterns [
15]. The residual LN on CT scan after treatment was defined as LN > 10 mm in diameter with enhancement pattern of TB and central low attenuation with peripheral rim enhancement [
13,
16]. Treatment failure was defined as residual LN that is > 10 mm in diameter and culture positivity after treatment. Recurrence was defined as reappearance of an LN or the appearance of a new tuberculous node after the completion of antituberculous therapy with a period of initial clinical remission, as previously described [
9].
Microbiology
Aspirates and digested fresh tissues were used for microscopic examination (Ziehl–Neelsen stain), culture, and TB PCR. For TB culture, samples were cultured on liquid and solid culture media. M. tuberculosis was identified using a commercial DNA probe (AccueProbe Mycobacterium tuberculosis Complex Culture Identification Test; Gen-Prove, San Diego, CA, the USA). A commercially available PCR test kit (COBAS® AMPLICORMTB; Roche Diagnostics, Branchburg, NJ, the USA) was used to assay fresh and fixed samples.
Statistical analysis
To compare the two groups, Pearson χ2 tests and Fisher’s exact tests were used for categorical variables, and student’s t-test and Mann–Whitney U tests were utilized for continuous variables if appropriate. All statistical tests were two-tailed, and p-values ≤0.05 were considered statistically significant.
Discussion
In this study, cervical tuberculous lymphadenopathy remains in almost a quarter of the patients on CT scans after 6 months of treatment, and none experienced treatment failure.
Lymphadenopathy usually disappears in 30–40% of patients after 3 months of antituberculous chemotherapy and in 80% after 6 months of treatment. However, LN that is > 5 mm in diameter may last for a long period of time in 20% of patients [
9]. Although all residual LNs do not have an unfavorable outcome, defined as treatment failure or relapse [
9], treatment might be prolonged or re-started in real world because unfavorable outcomes were not only on bacteriological or histological examination but also according to clinical findings. In contrast to smear-positive pulmonary TB, bacteriological treatment cannot be confirmed because it is challenging to obtain specimens from tuberculous lymphadenitis. In hospitals with poor healthcare facilities, decision-making in terms of treatment usually depends on clinical judgement when post-treatment lymphadenopathy occurs. In Korea where the prevalence of TB and the rate of drug-resistance is high, antituberculous treatment may be prolonged to 9–12 months when residual LNs remain [
11].
In the absence of a consensus regarding the interpretation of the post-treatment residual LNs, this study revealed the characteristics of residual LNs based on pathologic findings. Biopsy or culture of residual nodes showed granuloma formation and negative culture results with or without positive AFB stains and/or TB PCR previously described as post-treatment paradoxical response [
17,
18]. These features are consistent with hypersensitivity to the antigen of
M. tuberculosis, which may be poorly cleared from the disease site even after prolonged therapy [
1,
18]. Notably, not all residual LNs indicate treatment failure.
The risk factors of residual LNs must be identified because residual LNs, not treatment failure, can occur in some patients. However, the risk factors are not fully elucidated to date. Prior studies have indicated the following risk factors: younger age, male gender, size ≥3 cm, and local tenderness [
11,
12,
17]. In our study, younger age was significantly correlated to residual LNs, which is consistent with previous studies. Thus, residual LNs may appear at the end of the six-month antituberculous therapy in younger patients, and treatment failure must be confirmed through bacteriological examination after short-term observation rather than prolonged treatment or drug changes.
Non-invasive methods, such as CT or fluorodeoxyglucose (FDG)-positron emission tomography (PET), and the identification of clinical risk factors may be helpful in evaluating the treatment response for tuberculous lymphadenitis [
13,
16,
19]. Previous studies have shown that central necrosis, perinodal infiltration, and peripheral rim enhancement on CT scans were more frequently observed in patients with treatment failure [
13]. When a higher SUVmax value from FDG is added to this CT scan findings, distinguishing the treatment responder and non-responder was more helpful [
16]. Therefore, the size of the LN at the end of antituberculous treatment was not associated with treatment failure.
In the treatment period, the guidelines recommends six-month treatment for tuberculous lymphadenitis caused by drug-susceptible organisms [
1,
20,
21]. The six-month recommendation is supported by studies showing that no difference was observed between 6 and 9 months of treatment in terms of cure rates (89–94%) [
8,
22] or relapse rates (3%) [
9]. The Joint Tuberculosis Committee of the British Thoracic Society has stated that follow-up is not required after a successful treatment. However, patients should be re-referred if symptoms recur due to the limited number of relapse cases [
9,
23]. In this study, the duration of antituberculous treatment with residual LNs was shorter in the group whose microbiological characteristics were identified through FNAB. Considering the microbiological characteristics of residual LNs, this study may be supporting the six-month therapy in the preexisting guideline.
During the follow-up period with a median of 658 days, no differences were observed in terms of recurrence between the groups with and without residual LNs. However, recurrences are diagnosed without microbiological findings in patients with recurrence, which may reflect post-treatment paradoxical response rather than recurrence [
17]. The gold standard for confirming recurrence is mycobacterial culture. However, culture alone cannot diagnose recurrence because of low sensitivity. Prior studies have already pointed out these problems [
9,
17]. In one study, among patients with recurrence, three were diagnosed with recurrence according to mycobacterial culture results and based on histological findings, indicating that clinical judgement is still critical in identifying recurrence [
9]. The recurrence period is usually within 12 months after the end of treatment. However, recurrence may occur after 1 year. Therefore, further investigation must be conducted as it remains unclear how long the required follow-up period is for residual LNs after antituberculous treatment.
This study presents the clinical, radiologic, and pathologic findings at the end of six-month treatment for tuberculous lymphadenitis. The limitations of this study are as follows: FNAB was not performed in all patients with residual LNs, and some of the specimens were inappropriate. Most patients with residual LNs were treated with antituberculous drugs for more than 6 months, and CT scans were not obtained in further follow up course. The variables correlated to the recurrence of TB among the residual LN group must be identified next. Further follow-up studies with a higher number of patients must be conducted for statistical significance.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.