Know what up to definition

Current Diagnosis and Management of Peripheral Tuberculous Lymphadenitis

Abstract

Peripheral tuberculous lymphadenitis accounts for 10% of tuberculosis cases in the United States. Epidemiologic characteristics include a 1.4: 1 female-to-male ratio, a peak age range of 30–40 years, and dominant foreign birth, especially East Asian. Patients present with a 1-2 month history of painless swelling of a single group of cervical lymph nodes. Definitive diagnosis is by culture or nucleic amplification of M.ycobacterium tuberculosis; demonstration of acid fast bacilli and granulomatous inflammation may be helpful. Excisional biopsy has the highest sensitivity at 80%, but fine-needle aspiration is less invasive and may be useful, especially in immunocompromised hosts and in resource-limited settings. Antimycobacterial therapy remains the cornerstone of treatment, but response is slower than with pulmonary tuberculosis; persistent pain and swelling are common, and paradoxical upgrading reactions may occur in 20% of patients. The role of steroids is controversial. Initial excisional biopsy deserves consideration for both optimal diagnosis and management of the otherwise slow response to therapy.

Peripheral tuberculous lymphadenitis — previously termed “scrofula” —is a unique manifestation of disease due to organisms of the M.ycobacterium tuberculosis complex. Epidemiologic characteristics differ from those of pulmonary tuberculosis, clinical manifestations are variable, and diagnosis may be challenging. Of most importance for the clinician, response to therapy may be slow or paradoxical, with the frequent development of enlarging or new lymph nodes during and even after effective treatment in HIV-negative patients and of immune reconstitution inflammatory syndrome (IRIS) in HIV-positive patients. The optimal approach to management of such responses deserves reconsideration in light of newer studies in both HIV-negative and HIV-positive patients.

Recent studies have helped define the contemporary presentation of tuberculous lymphadenitis and its epidemiology. In addition, more extensive data are now available on both standard and novel diagnostic methods and the optimal management of complications during treatment. We searched Medline for English articles with use of the Medical Subject Heading term “Tuberculosis, Lymph Node” as a major topic from 1990 through 2011 to provide the clinician a contemporary perspective on these issues from both developed and developing countries.

EPIDEMIOLOGY

While overall rates of pulmonary tuberculosis have continued to decrease in the United States, the proportion of extrapulmonary cases, with their principal subset, lymphadenitis, has increased. Of the 12 904 cases of tuberculosis in the United States in 2008, 1103 (8.5%) represented lymphadenitis [1]. Epidemiologic characteristics from 14 studies on tuberculous lymphadenitis are shown in Table 1 and are separated into reports from countries where tuberculosis is endemic (> 40 cases / 100 000 population) and from countries where it is not endemic. In most series, tuberculous lymphadenitis is more common among women than among men (composite ratio, 1.4: 1) —a different pattern than for pulmonary tuberculosis, for which disease is more common among men [13]. Although formerly a disease of children, the peak age range in recent series has been 30-40 years. In countries where tuberculosis is not endemic, the majority of patients are foreign-born, with a pattern consistent with reactivation disease.

Table 1.

Epidemiology of Tuberculous Lymphadenitis

LocationDateNMean ageFemale%Foreign-born%HIV + (nPulmonary involved * (%)
Non-TB endemic
California [2]1992 40 38 52 82 11 28 
Washington DC [3]1995 30 62 N / A
Texas [4]2003 73 41 62 68 
California [5]2005 106 34 66 92 
Minneapolis [6]2006 124 25 57 100 
US [7]2009 19 107 38 58 61 2102 
Australia [8]1998 31 35 N / A87 
France [9]1999 59 38 52 69 
Germany [10]2002 60 41 68 70 
UK [11]2007 128 41 53 90 17 
UK [12]2010 97 14–89⊥ 59 90 N / A
TB endemic
Taiwan [13]1992 71 42 59 42 
Zambia [14]1997 28 24 54 32 
Taiwan [15]2008 79 37 58 
India [16]2009 893 20 58 18 
Qatar [17]2009 35 29 20 86 
LocationDateNMean ageFemale%Foreign-born%HIV + (nPulmonary involved * (%)
Non-TB endemic
California [2]1992 40 38 52 82 11 28 
Washington DC [3]1995 30 62 N / A
Texas [4]2003 73 41 62 68 
California [5]2005 106 34 66 92 
Minneapolis [6]2006 124 25 57 100 
US [7]2009 19 107 38 58 61 2102 
Australia [8]1998 31 35 N / A87 
France [9]1999 59 38 52 69 
Germany [10]2002 60 41 68 70 
UK [11]2007 128 41 53 90 17 
UK [12]2010 97 14–89⊥ 59 90 N / A
TB endemic
Taiwan [13]1992 71 42 59 42 
Zambia [14]1997 28 24 54 32 
Taiwan [15]2008 79 37 58 
India [16]2009 893 20 58 18 
Qatar [17]2009 35 29 20 86 
Table 1.

Epidemiology of Tuberculous Lymphadenitis

LocationDateNMean ageFemale%Foreign-born%HIV + (nPulmonary involved * (%)
Non-TB endemic
California [2]1992 40 38 52 82 11 28 
Washington DC [3]1995 30 62 N / A
Texas [4]2003 73 41 62 68 
California [5]2005 106 34 66 92 
Minneapolis [6]2006 124 25 57 100 
US [7]2009 19 107 38 58 61 2102 
Australia [8]1998 31 35 N / A87 
France [9]1999 59 38 52 69 
Germany [10]2002 60 41 68 70 
UK [11]2007 128 41 53 90 17 
UK [12]2010 97 14–89⊥ 59 90 N / A
TB endemic
Taiwan [13]1992 71 42 59 42 
Zambia [14]1997 28 24 54 32 
Taiwan [15]2008 79 37 58 
India [16]2009 893 20 58 18 
Qatar [17]2009 35 29 20 86 
LocationDateNMean ageFemale%Foreign-born%HIV + (nPulmonary involved * (%)
Non-TB endemic
California [2]1992 40 38 52 82 11 28 
Washington DC [3]1995 30 62 N / A
Texas [4]2003 73 41 62 68 
California [5]2005 106 34 66 92 
Minneapolis [6]2006 124 25 57 100 
US [7]2009 19 107 38 58 61 2102 
Australia [8]1998 31 35 N / A87 
France [9]1999 59 38 52 69 
Germany [10]2002 60 41 68 70 
UK [11]2007 128 41 53 90 17 
UK [12]2010 97 14–89⊥ 59 90 N / A
TB endemic
Taiwan [13]1992 71 42 59 42 
Zambia [14]1997 28 24 54 32 
Taiwan [15]2008 79 37 58 
India [16]2009 893 20 58 18 
Qatar [17]2009 35 29 20 86 

A consistent observation in studies from nonendemic countries is that immigrants from Southeast Asia and India appear to have a special predilection for tuberculous lymphadenitis [4, 5, 8, 11]. In a study from Texas, the odds ratio (OR) was 11.3 for patients from Southeast Asia (P. <.01) and 12.7 for patients from India (P. <.01), compared with other ethnicities [4]. In a study involving HIV-negative Somalis in Minnesota [6], 30% of 407 patients with tuberculosis had lymphadenitis, which suggests that Africans may also have an increased risk of lymph node tuberculosis.

The basis for enhanced risk among women and Asians and, possibly, Africans is not known. Possible host factors include occupations or cultural practices favoring oropharyngeal exposures to M. tuberculosis complex (eg, exposure to M.ycobacterium bovis or M. tuberculosis from milking cows), genetically determined organ tropism, hormonal influences, effects related to bacillus Calmette-Guérin (BCG) immunization, and differences in health-seeking behavior.

In addition, genetic differences in the virulence of organ tropism of different strains of M. tuberculosis may play a role [18, 19]. Extrapulmonary tuberculosis, including lymphatic tuberculosis, is more common among immunocompromised patients, including those with HIV infection [20, 21]. Although diabetes mellitus is a risk factor for pulmonary tuberculosis, studies suggest that it may reduce the relative risk of tuberculous lymphadenitis [4, 5]. In a review of extrapulmonary tuberculosis in the United States, traditional risk factors for pulmonary tuberculosis, such as homelessness and excess alcohol use, were associated with a lower risk of disease [7].

MICROBIOLOGY

Many studies of tuberculous lymphadenitis do not report speciation of the causative organism in the M. tuberculosis complex. M. bovis was historically a common cause of tuberculous lymphadenitis, but pasteurization and bovine tuberculosis programs have virtually eliminated this source of human infection in developed countries; risk remains with consumption of unpasteurized milk [22]. M. tuberculosis is the usual cause of tuberculous lymphadenitis [23]. Other infectious causes of chronic lymphadenitis include nontuberculous mycobacteria (including M.. scrofulaceum, M. avium, other M.. haemophilum), Toxoplasma species, Bartonella species, and fungi. Noninfectious causes include neoplasms, sarcoidosis, Castleman disease, drug reactions, and nonspecific reactive hyperplasia.

CLINICAL FEATURES

Tuberculous lymphadenitis usually presents as a slowly progressive, painless swelling of a single group of lymph nodes [3, 24]. The duration of symptoms at the time of presentation is typically 1–2 months, varying from 3 weeks to 8 months [3, 5, 24]. In a series of patients in India, the mean duration of symptoms was significantly longer in men than in women [24].

Median lymph node size is 3 cm, but nodes may be up to 8-10 cm in diameter [15]. Patients do not generally report significant pain at presentation, and node tenderness during examination is noted in only 10% -35% of cases [3, 15, 17]. A draining sinus may be present in 4–11% of cases [3, 17, 24]. Unilateral involvement of 1-3 nodes has been noted in 85% of cases [8]. Cervical chain involvement is most common and is reported in 45% -70% of cases, with 12% -26% in the supraclavicular region; ∼20% of cases are bilateral [2, 5, 15, 17]. In a study from Zambia, symmetrical adenopathy with nodes typically <3 cm was reported in 94% of patients with HIV-induced lymphadenopathy, compared with 29% of patients with HIV-associated tuberculous lymphadenitis. In contrast, symmetrical adenopathy was observed in only 11% of HIV-negative patients with tuberculosis lymphadenitis, and nodes in this group were typically> 3 cm [14].

Rates of systemic symptoms reported in different series vary depending in part on geographic origin and case selection (Table 2). In a series of 104 predominantly HIV-negative patients from California, fever was reported in 19% and weight loss in 16% [5]. In contrast, fever and weight loss were reported in 40% -60% of HIV-negative patients in series from Qatar and India [17, 24]. Systemic symptoms are reported more frequently in HIV-positive patients than in HIV-negative patients (76% of 21 vs 12% of 43 in a report from Taiwan) [15]. Concomitant pulmonary tuberculosis is reported in 18% –42% of patients (Table 1), with higher rates among HIV-positive patients than among HIV-negative patients (90% of 10 vs 28% of 25 in a study from Los Angeles) [ 2]. HIV-positive patients with tuberculous lymphadenitis typically have a higher rate of disseminated disease than do HIV-negative patients (38% vs 8%; P. <.001) [27].

Table 2.

Presenting Signs and Symptoms of Tuberculous Lymphadenitis

Nn
Fever
Cough
Cervical Involvement
Abscess formation (%)
Sinus drainage (%)
Location (Year)HIV (-)HIV (+)HIV (-)HIV (+)HIV (-)HIV (+)HIV (-)HIV (+)HIV (-)HIV (+)HIV (-)HIV (+)
Non-TB endemic
California (1992) [2] *40 29 11 18% 63% N / AN / A58% 36% N / AN / AN / AN / A
California (2005) [5]106 101 18% 80% 19% 100% 57% 13% 8% 
UK (1996) [25]23 N / A52% N / A40% 30% 23% 
UK (2007) [11]128 126 26% N / A87% 100% 21% N / A
TB endemic
Zambia (1997) [14]185 28 157 32% 44% N / AN / A96% 99% 0% 4% 7% 3% 
India (2007) [26]121 45 20 23% N / A69% N / A4% 
India (2009) [16]893 893 4% 0% 10% 0% 89% 0% 4% 0% 2% 0%