Attenuated response to PPD in the diagnosis of latent tuberculosis infection in patients with rheumatoid arthritis

Abstracts

INTRODUÇÃO: A utilização dos anti-TNF's na prática reumatológica tornou obrigatória a identificação de casos de infecção tuberculosa latente (ITBL) antes do início do tratamento, utilizando o PPD, RX de tórax e história de contato com tuberculose. Os pacientes com artrite reumatoide (AR) apresentam uma anormalidade da função celular imune, caracterizada por diminuição da responsividade de células mononucleares periféricas (linfócitos T Reg), o que acarreta prejuízo para a hipersensibilidade cutânea tardia, o que é fundamental para o reconhecimento de antígenos, como é o caso do PPD. OBJETIVOS: Avaliar a resposta ao PPD em pacientes com AR, comparado com pessoas saudáveis em uma área endêmica de tuberculose, como é o estado de Pernambuco. METODOLOGIA: Foram estudados 96 pacientes, 48 com AR e 48 indivíduos saudáveis, sendo a maioria do sexo feminino; foi realizada a inoculação de 0,1 mL do PPD RT-23, por via intradérmica e leitura da induração 72 horas após. RESULTADOS: No grupo AR o tempo médio de diagnóstico foi de 10,2 anos, a dose média de metotrexate foi de 15,5 mg/semana, dose média de prednisona foi de 12,7 mg/dia e a atividade média da doença medida pelo CDAI foi 30,4. Houve maior positividade do PPD no grupo de comparação (33,3%) quando comparado com o grupo AR (14,6%), com diferença estatisticamente significante (p = 0,034). CONCLUSÕES: Os autores alertam para o fraco desempenho do PPD para diagnóstico de ITBL em pacientes com AR e da necessidade de se fazer uma triagem ainda mais cuidadosa antes do início do tratamento com anti-TNF.

artrite reumatoide; PPD; tuberculose latente

ORIGINAL ARTICLE

^IEscola Pernambucana de Saúde - FBV/IMIP

^IIHospital das Clínicas - Universidade Federal de Pernambuco

^IIIPh.D student at Centro de Pesquisas Aggeu Magalhães

^IVResearchers at Centro de Pesquisas Aggeu Magalhães

Correspondence to

ABSTRACT

INTRODUCTION: With the introduction of Tumor Necrosis Factor Inhibitors (anti-TNFs) into rheumatological practice, it has become obligatory to identify cases of latent tuberculosis infection (LTBI) prior to the start of treatment, using PPD, chest radiography and clinical history of tuberculosis contact. Patients with Rheumatoid Arthritis (RA) have an abnormality of the cellular immune function, characterized by decreasing responsiveness of peripheral mononuclear cells (T Reg lymphocytes), leading to a loss in delayed hypersensitivity, which is fundamental for the recognition of antigens, such as PPD.

OBJECTIVES: The purpose of our study was to evaluate the response to PPD in patients with RA, compared with healthy people, in an area where tuberculosis is endemic, as is the state of Pernambuco.

METHODOLOGY: We studied 96 patients, 48 with RA and 48 healthy subjects, most of them females. All patients were given an interdermic injection of 0.1 mL PPD RT-23. The reading of the PPD result was carried out 72 hours after application, by way of palpation of maximum transverse diameter of induration, and the result was expressed in millimeters.

RESULTS:In the RA group, the average time of diagnosis was 10.2 years, the average dosage of methotrexate was 15.5 mg / week, the average dosage of prednisone 12.7 mg / day and the average activity of the disease, measured using CDAI, was 30.4. In the healthy subjects group there was a greater number of positive PPD results (33.3%) when compared with the results for the RA group (14.6%), with a statistically significant difference (p = 0.034).

CONCLUSION:The performance of PPB in LTBI diagnosis is poor in patients with RA. These results suggest that more careful screening needs to be undertaken before treatment with an anti-TNF drug.

Keywords: rheumatoid arthritis, PPD, latent tuberculosis.

INTRODUCTION

Due to an increasing incidence and severity of tuberculosis (TB) infections after the introduction of tumor necrosis factor inhibitors (anti-TNFs) into rheumatoid arthritis (RA) treatment, it has become obligatory to identify cases of latent tuberculosis infection (LTBI) prior to the start of treatment. Brazilian guidelines for the performance of LTBI or active disease screening recommend that assessment prior to starting anti-TNF treatment should include: the complete clinical history (previous treatment or chemoprophylaxis, domestic contact or institutional contact with TB), chest radiography and purified protein derivatives (PPD).^1-3 However, PPD use in RA patients presents a major complicating factor: an abnormality in the cellular immune function observed in such patients.⁴ There is a decreasead responsiveness of peripheral mononuclear cells, leading to a loss in delayed hypersensitivity, which is fundamental for the recognition of antigens, such as PPD.^4-6 The mechanism for this change is not exactly known, but it has been demonstrated that it may be caused by IL-2⁷ production deficiency or chronic TNF exposure.⁶ Regulatory T cells (T reg) (CD4⁺CD25⁺), which has a fundamental role in autoimmunity prevention, present a reduction in number and function in RA patients,⁸ and, as demonstrated in a recent Brazilian paper, the number of T CD4⁺ (T Reg) cells is directly related to PPD response magnitude.⁹

The purpose of our study was to evaluate the response to PPD in patients with RA, compared with healthy individuals, in an area where tuberculosis is endemic, as is the state of Pernambuco, Brazil.

MATERIAL AND METHODS

A transversal study was carried out, including 96 patients using nonprobability sampling, by convenience type, divided in two groups: 48 patients with RA according to American College of Rheumatology¹⁰ criteria (RA group) indicated for infliximab use and 48 healthy subjects, forming the comparison group (COMP group), from May to October, 2007, selected at the Rheumatology Clinic at Hospital das Clínicas at University of Pernambuco (HC-UFPE), in the state of Pernambuco.

In order to be part of this study, RA patients had to be, at least, 18 years old, with active disease and assigned for treatment with infliximab. The following items were considered as exclusion criteria: active TB; BCG vaccine less than 15 years before; previous treatment using biological agents (infliximab, etanercept or adalimumab); known diagnosis of other diseases considered risk factors for TB: AIDS, malnutrition, diabetes, kidney disease or liver disease and neoplasia; acute infections; inpatients; and pregnant women. The comparison group were patients from the Rheumatology Clinic who did not present autoimmune diseases, meeting the same exclusion criteria for RA group.

This study was approved by Aggeu Magalhães Research Center ethics committee and all patients read and signed the informed consent.

Then patients were given a specific questionnaire, created to establish the epidemic history for TB and RA activity, measured by CDAI (Clinical Disease Activity Index).¹¹ PPD was performed by intradermal injection of 0.1 mL (2UT) of PPD RT-23, in the left mid-arm (approximately 8cm under elbow) (Figure 1).

PPD result measurement was done 72 hours after inoculation, using palpation of maximum induration diameter, with results in millimeters (Figure 2). Measurement was done by only one examiner in all patients. PPD response was analyzed according to the following:

RA group: 0 to 4 mm, negative; > 5, positive

COMP group: 0 to 4 mm, negative; 5 to 10 mm, weak positive; more than 10 mm, strong positive.

RESULTS

In this study, 96 patients were included, divided into RA group (48) and COMP group (48). Most of them were female in both groups (89.6% in RA group and 72.9% in COMP group). In RA group, mean age was 49.71 years old (±12.41; minimum 19 and maximum 78 years old), while in COMP group it was 46.29 years old (±13.99; minimum 21 and maximum 75 years old).

Average time of RA diagnosis was 10.2 years (±12.41). All patients were taken prednisone, with an average dosage of 12.7 mg/day (±6.7). Out of these 48 patients, only 30 were taking methotrexate (MTX), with an average dosage of 15.5 mg/week (±4.3). Clinical variables in RA group are summarized in Table 1.

Frequencies of other variables, as well as comparative analysis of equivalence in both groups, are demonstrated in Table 2. We can observe that these two groups were similar, and there was no statistically significant difference in gender, race, education level, agglomeration, tuberculosis infection history or domestic transmission tuberculosis contact. Statistically significant differences (p = 0.017 and p = 0.034 respectively) were observed between the two groups only in income and origin variables. It is important to highlight that in the comparison group health professionals who worked at rheumatology clinic at Hospital das Clínicas were included, due to the latent tuberculosis infection risk that this group presents. All patients in both groups presented BCG vaccination history during childhood, which could be confirmed by a scar in the right arm, and none of them presented alcoholism history. Table 3 shows the frequency of positive PPD in both groups.

No association was found among RA clinical variables, such as duration of use of prednisone or MTX dosage, duration of RA activity or disease measured by CDAI and PPD results.

DISCUSSION

Tuberculin skin test, which uses a standard preparation of purified protein derivative (PPD), is used since 1931 to screen people infected by M. tuberculosis. It contains a mixture of antigens that leads to a late hypersensitivity reaction and reflects the cellular immunity against bacillus, and, despite its known limitations in sensitivity and specificity, it continues to be used as a standard criterion for LTBI diagnosis. Although it is occasionally used in symptomatic infection diagnosis, its primary use is LTBI detection.¹² However, screening for TB using PPD is discouraging, due to the low specificity in the test, as both BCG vaccine and nonmycobacterium tuberculosis exposure lead to a response similar to the one induced by M. tuberculosis infection.¹³

Mantoux method evaluates in vivo the immune cellular response against purified protein derivative of M. tuberculosis, resulting in a classical reaction of delayed skin hypersensitivity, depending on the migration of INF³ CD4⁺ T cells to the antigen injection location. Patients with RA should present incapacity to produce a proper PPD response, even in subjects infected with M. tuberculosis, causing this test to be inadequate for recognizing latent forms of TB in these patients.⁷ Some authors recommended that patients with RA and a negative PPD, but who have a high clinical or epidemiological risk for tuberculosis infection, should be empirically treated as they had LTBI before start using a biological agent.¹²

Our study demonstrates that the frequency LTBI diagnosis by PPD in patients with RA is lower than in healthy individuals, with statistically significant difference (OR = 0.31; 0.11 - 0.84, p = 0.034), similar to results already described in the literature, which could provide a false judgment of RA being a protective factor for LTBI. However, this factor probably occurs due to low PPD responsiveness in RA or due to the number of false positive PPD results in healthy individuals (cross reaction with other mycobacterium or BCG vaccine). The fact that we have included subjects who work in the health system did not influence the number of positive results, once that out of 13 people in the 48-person group only 2 presented a positive PPD result, which did not influence the final average of the total positive number in this group.

In a study carried out in Peru,¹⁴ where TB is endemic, PPD was performed in one group with RA patients and another group of immunocompetent volunteers, matched by gender and age. A result higher or equal to 5 mm in RA group and higher or equal to 10 mm in the immunocompetent group was considered positive. A result lower than 5 mm after 72h was considered negative in both groups. A 71% PPD positivity was found in immunocompetent group against 29% in RA group. All patients in RA group were taking dosages lower than 7.5mg/day of corticosteroid, which does not abolish delayed skin hypersensitivity. This discrepancy in results has been associated to RA cellular immunity abnormality. In another study, carried out in Turkey, area where TB prevalence is relatively high, a low PPD positivity (29.8%) was observed in RA patients, while compared to patients with ankylosing spondylitis (65.9%), gouty arthritis (68.8%) and osteoarthritis (63%).¹⁵

Provenzano, Ferrante e Simon³ evaluated 69 Italian patients with chronic inflammatory joint disease, who would start anti-TNF treatment. During screening for LTBI, 2.9% patients were found with previous treated TB history, 8.7% positive PPD and radiographic changes compatible with TB scar in 20.3%, demonstrating that PPD failed in identifying all patients with LTBI, which indicated that a chest radiography procedure to be is mandatory.

Bahr et al.¹⁶ researched the HLA type in established RA and its relation with 4 types of skin tests for mycobacteria (tuberculin, leprosin A, scrofulin and vaccine), injected in forearm, in an Arabic population. Three groups of patients were analyzed: established RA (46), active tuberculosis infection (111) and healthy individuals (79). Responsiveness to mycobacterium skin tests was significantly lower in the RA patients group and it is related to haplotype DR7, which did not happen in the other two groups.

Although it is widely accepted, screening procedure before anti-TNF use has been criticized, because, in several situations, it is not able to identify LTBI. Among the problems described, there is the uncertainty of patient´s medical history or TB contact, lack of specific LTBI radiological signals, additionally to difficulties already described in PPD use. Thus, the use of these tests may lead to both lack of TB chemoprophylaxis, due to low sensitivity of skin test in RA, and unnecessary treatment (in case of false positive PPD).¹⁷

The use of specific antigen tests to detect INF³ has caused a radical change in LTBI diagnosis in countries with low TB prevalence, and has already shown a better specificity than PPD in immunocompetent people. In countries with a high TB prevalence, there is also an uncertainty about the possibility to help in latent infection diagnosis. The evaluation of these new tests performances is damaged due to a lack of golden standard to make it possible to differentiate true cross reaction between latent infection and BCG vaccine or mycobacterium infection other than M. tuberculosis.¹⁸

While diagnosing and treating LTBI in developed countries is an essential component for TB control, this is a less important strategy in developing countries. In these countries, due to the high risk of infection characterized by the high incidence of involvement of the pediatric population, the health system should prioritize treatment of active cases, reducing the risk of transmission. Additionally, a substantial part of the population in these countries has LTBI and, once the risk of these subjects to develop active infection is 5% and almost 90% of chemoprophylaxis is completely performed, it is unnecessary to add this cost (LTBI diagnosis) to TB control program in developing countries. Thus, the use of these tests in such locations should be restricted to situations that include: Epidemiologic surveillance, malnourished children and patients with tuberculosis, tuberculosis infection in patients with AIDS and RA.¹⁸ In clinical practice, we should know that the performance of PPD for LTBI screening is poor in patients with RA. These results suggest that more careful screening need to be undertaken before treatment with an anti-TNF drug, and a good clinical history and radiological changes should be added to the PPD result until we have more specific tests available, such as those based on INF³ production.

REFERENCES

• ¹
  Mangini C, Melo FAF. Artrite reumatoide, terapia imunossupressora e tuberculose. Rev Bras Reumatol 2003;43:XI-XV.
• ²
  Ormerod LP, Milburn HJ, Gillespie S, Ledingham J, Rampton D. British Thoracic Society (BTS) Standards of Care Committee. BTS recommendations for assessing risk and managing Mycobacterium tuberculosis infection and disease in patients due to start anti-TNF-α treatment. Thorax 2005;60(10):800-5.
• ³
  Provenzano G, Ferrante MC, Simon G. TB screening and anti-TNF-α treatment. Thorax 2005;60(10):612-5.
• ⁴
  Panayi G, Corrigal V, Pitzalis C. Pathogenesis of rheumatoid arthritis. The role of cells and others beasts. Rheum Dis Clin North Am 2001;161:221-47.
• ⁵
  Emery P, Panayi GS, Symmons D, Brown G. Mechanisms of depressed delayed-type hypersensitivity in rheumatoid arthritis: the role of protein energy malnutrition. Ann Rheum Dis 1984;43(3):430-4.
• ⁶
  Corrigal VM, Garyfallos G, Panayi GS. The relative proportions of secreted interleukin-2 and interleukin-10 determine the magnitude of rheumatoid arthritis T-cell proliferation to the recall antigen tuberculin purified protein derivative. Rheumatology 1999;38:1203-7.
• ⁷
  Kingsley GH, Pitzalis C, Panayi GS. Abnormal lymphocyte reactivity to self-major histocompatibility antigens in rheumatoid arthritis. J Rheumatol 1987;14(4):667-73.
• ⁸
  Ehrenstein MR, Evans JG, Singh A, Moore S, Warnes G, Isenberg DA, Mauri C. Compromised Function of Regulatory T Cells in Rheumatoid Arthritis and Reversal by Anti-TNF-α Therapy. J Exp Med 2004;200:277-85.
• ⁹
  Martins M, Lima M, Duppre N, Matos H, Spencer J, Brennan P et al The level of PPD-specific INF- γ-producing CD4⁺ T cells in the blood predicts the in vivo response to PPD. Tuberculosis 2006;87(3):202-11.
• ¹⁰
  Arnett FC, Edworthy SM, Bloch DA, Mcshane DJ, Fries JF, Cooper NS et al The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31(3):315-24.
• ¹¹
  Aletaha D, Smolen J. The Simplified Disease Activity Index (SDAI) and the Clinical Disease Activity Index (CDAI): A review of their usefulness and validity in rheumatoid arthritis. Clin Exp Rheum 2005;23(S39):S100-8.
• ¹²
  Furin JJ, Johnson JL. Recent advances in the diagnosis and management of tuberculosis. Curr Opin Pulm Med 2005;11(3):189-94.
• ¹³
  Andersen P, Munk ME, Pollock JM, Doherty TM. Specific immune-based diagnosis of tuberculosis. Lancet 2000;356(9235):1099-104.
• ¹⁴
  Ponce de León D, Acevedo-Vazquez E, Sánchez-Torres A, Cucho M, Alfaro J, Perich R et al Attenuated response to purified protein derivate in patients with rheumatoid artritis: study in a population with high prevalence of tuberculosis. Ann Rheum Dis 2005;64:1360-1.
• ¹⁵
  Köker IH, Pamuk ÖN, Karlikaya N, Tunçbilek N, Cakir N. A low prevalence of purified protein derivative test positivity in Turkish patients with rheumatoid arthritis. Association with clinical features and HRCT findings. Clin Exp Rheumatol 2007;25(1): 54-9.
• ¹⁶
  Bahr GM, Sattar MA, Stanford JL, Shaaban MA, Al Shimali B, Siddiqui Z et al HLA-DR and tuberculin tests in rheumatoid arthritis and tuberculosis. Ann Rheum Dis 1989;48(1):63-8.
• ¹⁷
  Dinser R, Fousse M, Sester U, Albrecht K, Singh M, Köhler H et al Evaluation of latent tuberculosis infection in patients with inflammatory arthropathies before treatment with TNF-α blocking drugs using a novel flow-cytometric interferon-γ release assay. Rheumatology 2008;47(2):212-218.
• ¹⁸
  Dheda K, Udwadia Z, Huggett JF, Johnson MA, Rook GA et al Utility of the antigen-specific interferon-γ assay for the management of tuberculosis. Curr Opin Pulm Med 2005;11(3):195-202.

Attenuated response to PPD in the diagnosis of latent tuberculosis infection in patients with rheumatoid arthritis

Cláudia Diniz Lopes Marques^I; Ângela Luzia Branco Pinto Duarte^II; Virginia Maria Barros de Lorena^III; Joelma Rodrigues de Souza^III; Wayner Souza^IV; Yara de Miranda Gomes^IV; Eduardo Maia Freese de Carvalho^IV

Publication Dates

• Publication in this collection
  13 Apr 2009
• Date of issue
  Apr 2009

History

• Accepted
  30 Sept 2008
• Received
  05 May 2008