|CASE BASED REVIEW
|Ahead of print publication
Infections mimicking difficult-to-treat systemic lupus erythematosus
Niruj Rheumatology Clinic, Ahmedabad, Gujarat, India
|Date of Submission||22-Jun-2020|
|Date of Acceptance||26-Aug-2020|
Niruj Rheumatology Clinic, 209 Rajvi Complex, Rambaug, Ahmedabad - 380 008, Gujarat
Source of Support: None, Conflict of Interest: None
Infections are closely associated with systemic lupus erythematosus (SLE). SLE patients require aggressive immunosuppression making them vulnerable to unusual or atypical presentations of common infections. They have an inherent immune abnormality predisposing them to infection. On the other hand, infection can act as a trigger for the immune system and lead to a newonset SLE or flare. Here, we present three-patients with SLE and uncommon infection. First case is of a chronic cytomegalovirus-infection that mimicked a 3-year long SLE-disease. Detection of infection and treating it led to withdrawal of all immunosuppressive therapy. There was no further SLE-activity on follow-up. Similarly, in the second case, tuberculosis presented as diffuse lower lobe pneumonitis, which triggered a mild-quiescent undiagnosed SLE (ANA positive with low platelet count) into a full-blown SLE (nephritis and autoimmune hemolytic anemia). Initially, a diagnosis of lupus pneumonitis was made but later complications of cavitation and nonhealing bronchopleural fistula led to a diagnosis of tuberculosis. Again, in this case, treating the infection led to complete resolution of SLE. The third case is of herpes-simplex virus esophagitis that was confused with steroid withdrawal symptoms. It led to a refractory SLE-flare that settled after treatment of infection. Thus, here, we have discussed the complex interplay of these infections with the diagnosis and management of SLE along with an intriguing phenomenon of treating the infection leading to near-cure of an aggressive systemic autoimmune disease.
Keywords: Cytomegalovirus, herpes-simplex virus, lupus pneumonitis, systemic lupus erythematosus, tuberculosis
| Introduction|| |
Systemic lupus erythematosus (SLE) is a prototype of systemic autoimmune diseases with inflammation involving multiple organ-systems and can lead to permanent organ-damage. Aggressive immunosuppressive-therapy is often required to control disease-activity and prevent damage. Primary immune-defects, for example, dysfunction of neutrophils, decreased production of interleukin-8 and 12, complement defects, dysfunction of T, B, and NK-cells have been reported in SLE-patients., Thus, primary immune-dysfunction added to secondary immunosuppression makes SLE-patients susceptible to bacterial, viral and rarely fungal infections.
Infections can act as an environmental-trigger for onset of SLE in a genetically susceptible individual or can lead to flares in a known case of SLE. This can be explained by molecular mimicry, bystander phenomenon, epitope spreading or pathogen acting as a superantigen. Thus, leading to a disruption of self-tolerance and expansion of autoreactive immune cells. Infections and SLE-flares are difficult to differentiate. Both can have similar clinical presentations. The current biomarkers, C-reactive protein (CRP) and pro-calcitonin (PCT), cannot specifically differentiate between the two. Infections and SLE-flares can at times coexist, for the rheumatologist to treat both parallelly. Moreover, the treatment of SLE can cause suppressed inflammatory response to common pathogens resulting in an atypical presentation and adding to the confusion, for example, tuberculosis can present as diffuse bilateral lower lobe pneumonitis instead of unilateral apical involvement. Thus, clinical judgment between SLE-flare and infection can be easily mistaken in retrospect.,
There are instances when infection can be a real trigger converting mild-quiescent SLE into difficult-to-treat SLE. In such cases, treating the infection leads to spontaneous resolution of SLE-activity and immunosuppression can be reduced to minimum or stopped. Here, we discuss three cases of SLE with infections. Written and informed consents were taken from all the patients for sharing and publication of their clinical information without disclosing their identity.
| Case Reports|| |
In September' 15, a 67-year-old female visited a rheumatologist with complaints of fever, polyarthralgia, and pancytopenia. Based on these clinical features and initial investigations showing rise in creatinine and urinary abnormalities [Table 1], she was diagnosed to have SLE. Kidney biopsy was not done. She was treated with oral prednisolone (1 mg/kg) and mycophenolate mofetil (MMF, 2 g/day). Fever subsided, serum creatinine and urine abnormalities normalized. After 2 months, she developed MMF related chronic diarrhoea and was shifted to azathioprine (100 mg/day). However, she developed bone marrow suppression requiring hospitalization. She than preferred to continue mycophenolic acid and was able to tolerate it. By September 17, prednisolone was tapered and stopped but her symptoms fever and arthralgia recurred. Fever used to respond at higher doses of oral prednisolone but used to recur at doses below 7.5 mg. During fever episodes, her investigations [Table 1] showed fall in hemoglobin and platelet count with a rise in creatinine and active urine sediments. erythrocyte sedimentation rate (ESR) and CRP were high with normal PCT. Finally, to control the disease-activity tacrolimus (2 mg daily for 2 months followed by 4 mg) was added to MMF. It took few months to see the response of changes in therapy.
In July 18, she presented to us with chief complaint of fever up to 102°F not responding to the above treatment. The cause of prolonged-fever was not clear, may be due to disease activity or secondary infection. Her investigations were in favor of SLE-flare [Table 1]. Based on the clinical judgment and investigations, prolonged-fever was attributed to SLE-disease activity and was treated with rituximab 1 g each two-dose 14 days apart. Fever subsided after few weeks and she started feeling better. Prednisolone was tapered and stopped. Investigation showed improvement [Table 1]. However, fever recurred after 3½ months of rituximab therapy. The fever was spiking up to 104°F.
Positron emission tomography computed tomography scan and bone marrow aspiration, biopsy and culture were done to look for any missed malignant or infectious cause. Both investigations were normal. So again, SLE disease activity was believed to be the culprit and rituximab was repeated 500 mg single dose. One of her investigation sent at this point markedly changed the diagnosis and further course. This investigation was blood cytomegalovirus (CMV) DNA viral load of 5 lac IU/ml. She was treated with intravenous ganciclovir infusion followed oral valganciclovir (900 mg twice daily for 4-monts as CMV-DNA was negative). Her fever subsided and investigations improved over few weeks. Prednisolone was tapered and stopped in 3- months. She is now under follow-up for 2 years and off-treatment (hydroxychloroquine [HCQ] stopped) from the last 16-month without any symptoms of SLE. Her current investigation still shows immunological evidence of SLE [Table 1].
In December 15, a 49-year-old postmenopausal lady presented with complaints of proximal muscle weakness, recurrent oral ulcers, arthritis, and bluish discoloration of fingers (mild, no ulcers) from 1 month. She was a diagnosed case of SLE from 6 years but not on regular follow-up with the rheumatologist. The disease-activity was managed by self-medication using prednisolone 10 mg and HCQ200 mg irregularly. Here, she presented to us with myalgias and progressive proximal muscle weakness for a month. Her investigation showed lymphopenia, raised muscle enzymes, proteinuria with spot protein/creatinine ratio <1, pyuria 2–5 cells/hpf, hematuria 2–5/hpf, serum levels of complements (C3 and C4) within normal range and antibody to double standard DNA (anti-dsDNA) was negative. Serum creatinine was normal. Hence, she was diagnosed to have a flare in SLE-activity presenting as myositis. Methotrexate 15–20 mg/week, HCQ300 mg/daily and oral prednisolone 1 mg/kg/day were started to manage the disease. Her muscle weakness gradually improved over few months and prednisolone was tapered to 5 mg/day. She also had hypothyroidism and osteoporosis (postmenopausal plus steroid induced, treated with 6 monthly denosumab) as comorbidities.
By August 16, myositis had recovered but she had complaints of intermittent arthritis, arthralgia, fever, oral ulcers and Raynaud's phenomenon. Her investigations showed lymphopenia, raised ESR 58 mm at 1 h and urine examination showed proteinuria 0.5–0.8 g/day with few white blood cells (WBCs) and red blood cells (RBCs). Due to intermittent disease activity, she was advised MMF but had nausea and vomiting and could not tolerate. She preferred to continue methotrexate 20 mg/week and over months her complaints subsided. By November' 18, prednisolone was tapered to 2.5 mg alternate day.
In July 19, she complained of nausea, vomiting, weight loss and marked fatigue from few months. Her serum cortisol random-level was low. Hence, her symptoms were attributed to steroid withdrawal and prednisolone was raised to 5 mg. She felt better for few months. However, in September 19, symptoms progressed and she developed dysphagia, fever and polyarthritis requiring admission. Her investigation showed increase in proteinuria 3+, increased pyuria, hematuria with ultrasonography showing right renal infarct. This time, serum complements level C3 and C4 were low and anti-dsDNA was raised. Antiphospholipid antibodies work-up was negative but oral anticoagulation was started due to renal infarct. With an impression of SLE flare, she was treated with methylprednisolone pulses 1 g/day for 3 days and cyclophosphamide pulse of 500 mg. Upper gastrointestinal endoscopy showed multiple deep mid-esophageal ulcers likely herpes esophagitis [Figure 1]. Histopathology showed kerato-acanthotic mucosal bits with ulceration and polymorphonuclear slough with granulation tissue. No viral inclusions or atypical cells were seen. She was treated with valacyclovir 1 g twice daily for 14 days and repeat endoscopy showed healed ulcers. The diagnosis of herpes-simplex virus (HSV) infection was validated based on clinical presentation and response to anti-HSV therapy. Oral prednisolone was gradually tapered from 0.5 mg/kg/day to 5 mg/day and methotrexate was restarted after 3-pulses of cyclophosphamide every 2 weeks. Over few months, prednisolone was further tapered to 2.5 mg/day while continuing methotrexate 20 mg/weekly and HCQ300 mg/day. Oral anticoagulation was stopped after 3 months. She is on regular follow-up. Currently, the disease is in both clinical and serological remission (normal complements and anti-dsDNA negative) while on treatment.
|Figure 1: Upper gastrointestinal endoscopy showing multiple deep ulcers noted at mid-oesophagus (4 in number and 12 cm size each), Grade-1 esophagitis. The findings likely to be herpes esophagitis|
Click here to view
In June 18, a 21-year-old boy presented with 2-year history of inflammatory back pain, dactylitis, episodic arthritis, and intermittent episodes of low platelet count (<1 lakh/mm3, unrelated to medications). Investigations showed haemoglobin-11.9 g/dl, leucocytes-5800/mm3, polymorphs 63%, lymphocytes 25%, platelet count 1.73 lakh/mm3, serum creatinine-0.6 mg/dl, urine routine examination normal, ESR-52 mm at 1 h, serum CRP–4.3 mg/dl, HLA-B27 (polymerase chain reaction method) – positive and ANA2+ speckled 1:80. Magnetic resonance imaging-pelvis showed bilateral sacroiliitis. Based on these findings, he was diagnosed to have ankylosing spondylitis and treated with sulfasalazine 2 g/day plus etoricoxib 90 mg/day. Later, methotrexate 15 mg/week was added to control arthritis as etoricoxib was tapered and stopped. In November 18, his investigation showed a fall in platelet count to 29,000/mm3. His symptoms of ankylosing spondylitis were controlled so he stopped all medicines. Over next 2 months, platelet count continued to fluctuate between 50,000 and 1 lakh/mm3.
In January 19, he presented with symptoms of progressive generalized edema, fever, malar rash, oral ulcers, dry cough, pleuritic chest pain, and breathing difficulty from 20 days. He required an admission in the emergency. His investigation showed fall in haemoglobin-6.5 g/dl, leukocytes-7400/mm3, polymorphs 70%, lymphocytes 24%, platelet count–43,000/mm3, ESR = 55 mm at 1 h, CRP = 15 mg/dl, PCT = 0.9 ng/ml, C3 = low, C4 = normal, anti-nucleosome positive, anti-dsDNA-240U/ml and serum creatinine–1.85 mg/dl. Direct and indirect Coomb's test was positive. Urine examination showed 4+ proteinuria with protein to creatinine ratio 4.3, WBC = 15–18/hpf, RBC = 50–55/hpf. Computer tomography of chest showed bilateral pleural effusion with lower lobe consolidations [Figure 2]a. Based on this evaluation, he was diagnosed to have SLE with pneumonitis, nephritis, and hemolytic anemia. Methylprednisolone pulse 1 g for 3 days plus cyclophosphamide 700 mg pulse was given to control SLE-activity along with other supportive measures. Oral prednisolone and HCQ were added.
|Figure 2: (a) Computed tomography scan showing bilateral lower lobe pneumonitis with bilateral pleural effusion (b) X-ray image showing right pneumothorax (c) X-ray image showing inserted inter-costal drainage tube on the involved side with re-expansion of the lung (d) computed tomography scan showing right residual pneumothorax, right lower lobe collapse and bilateral lower-lobe showing cavitary lesions (e) X-ray image post right lower-lobe lobectomy|
Click here to view
In March 19, he had an episode of pneumonia. He had fever and purulent expectoration with X-ray suggestive of the right lower-lobe consolidation. He was on tapering doses of oral prednisolone and monthly pulse of cyclophosphamide. Intravenous antibiotics ceftriaxone 2 gm/day and levofloxacin 750 mg/day was given for pneumonia, he responded well and recovered. By April' 19, 5-doses of cyclophosphamide were given. Generalized edema and fever subsided but patient continued to have dry cough and intermittent pleuritic chest pain. ESR and CRP were still raised and urine showed 3+protein, WBC-1–2/hpf, RBC-60–70/hpf. He was thought to have a refractory disease with pleurisy and nephritis persisting, so rituximab was added 1 g two doses 2 weeks apart.
After 20 days of rituximab therapy, the patient presented in emergency with right pneumothorax [Figure 2]b. Intercostal drainage tube was inserted [Figure 2]c. CT chest showed multiple cavitary lesions in right and left lower lobe [Figure 2]d. Brochoalveolar lavage and tube drainage fluid were negative for acid-fast-bacilli or fungus but lavage culture came positive for Mycobacterium tuberculosis. He developed nonhealing bronchopleural fistula and required right lower-lobe lobectomy [Figure 2]e. He improved on anti-tuberculosis treatment. Immunosuppression was gradually tapered to HCQ200 mg/day and steroid was stopped. He is under regular follow-up and has completed treatment for tuberculosis. Currently, he is in clinical and serological remission for >6 months and there are no symptoms of ankylosing spondylitis.
| Discussion|| |
Here, we reported three-patients of SLE and infection, two had herpes virus infections (CMV and HSV) and the third-one had M. tuberculosis infection. These infections closely mimicked SLE-flare or the entire SLE-disease. In the first case, chronic-CMV infection alone mimicking 3-year long disease course of SLE and its autoantibody profile is an intriguing possibility. While the case of M. tuberculosis infection may be an atypical presentation of pulmonary tuberculosis that flared-up a milder incomplete SLE-disease (ANA-positive and low-platelet) into a full-blown new-onset SLE with nephritis and hemolytic anemia. Based on the progression in radiological images [Figure 2] and evaluating the case retrospectively, the pneumonitis seen at presentation can be a tuberculosis infection to start with. Lupus pneumonitis although known, is relatively rare and not well-defined feature of SLE. In both cases, detection and treatment of infection almost cured SLE and the immunosuppressive treatment was withdrawn without any relapse over 1.5 years' follow-up in case-1 and 9 months' follow-up in case-3. This suggests that infection may be a primary culprit triggering the autoimmunity and driving the clinical features. Although the same cannot be conclusively stated. The second case was of a less common infection presenting as herpes simplex esophagitis in a diagnosed case of SLE, which can be easily missed. It triggered a refractory SLE-flare. Treating the infection again stabilized the activity and immunosuppression was reduced. Thus, SLE and infection are intertwined in a complex manner.
What was primary, infection or SLE? In case of CMV and tuberculosis the exact point of onset of infection is debatable. Although the duration of symptoms in CMV was long, chronic-CMV infection for years mimicking SLE can be a possibility. Persistent prolonged fever refractory to various immunomodulators and pancytopenia as chief presenting complaints with mild active urine sediments can be explained by chronic-CMV infection. Even the autoantibody positivity seen in the case can be induced by a chronic CMV-infection. Moreover, the common features of SLE for example mucocutaneous symptoms, malar rash, discoid lupus erythematosus and definite evidence of synovitis were missing in this case. The kidney biopsy would have added to the information but was not done. In addition, there is an another strong possibility of catching CMV infection during aggressive immunosuppressive treatment, which may seem more prudent to many. The persistence of autoantibodies related to SLE even after treating the infection during follow-up is a point in favor. However, the use of anti-viral converted a difficult-to-treat autoimmune disease into a disease with complete clinical remission without any immunomodulators for >16 months. This might make a case more in favor of a chronic-CMV infection alone.
CMV infection is well-known in SLE and can present in various ways. It is known that CMV might be involved in the etiopathogenesis of SLE, can trigger a new-onset SLE, may closely mimic SLE with common clinical features and can lead to refractory-flares presenting as macrophage activation-syndrome and death., CMV in immunocompetent hosts can be asymptomatic, may not be diagnosed but can remain detectable in blood more so in monocytes for years. Chronic CMV infection, its clinical features, effect on immune system and general health is less studied and reported in literature. Chronic presence of CMV DNA in blood monocytes of elderly woman over 12-year follow-up (1995 and 2007) leads to persistently raised serum levels of interleukin-6 and CRP. The impact of chronic CMV infection on the health of individuals is not well known and so is often not looked for and ignored. Thus, chronic CMV infection particularly in elderly (age of presentation in current case was 67 years) can lead to a state of chronic inflammation and mimic a systemic autoimmune disease. Here, treating the patient with immunosuppressive would have further aggravated the infection resulting into high spiking fever that finally lead to the identification of CMV and the diagnosis.
The presentation of SLE as acute lupus pneumonitis with high spiking fever is reported in the literature but at the best empirical and can be argued. Moreover, as shown in the figure, the lung regions initially involved got gradually cavitated resulting in bronchopleural fistula that finally cultured M. tuberculosis. Thus, raising the possibility of atypical presentation of tuberculosis infection as diffuse bilateral lower lobe pneumonitis with high grade spiking fever, which was attributed to lupus-pneumonitis. Tuberculosis presentation can be atypical in SLE. They can present as acute pneumonitis, with extensive lower-lobe involvement and bilateral pleural effusion., Lupus pneumonitis can be seen in 1%–4% of SLE patients, presents as emergency with high mortality. However, the presence of term can be argued and is controversial. An autopsy study of 120 SLE-patients done to determine the lung parenchymal changes found that many of such changes were probably secondary to infection, cardiac or renal failure and oxygen toxicity. However, the patient in addition to pneumonitis had nephritis and autoimmune hemolytic anemia. The tuberculosis pneumonitis might have triggered an ANA-positive low-platelet incomplete SLE-phenotype to evolve into a full-blown SLE. Treating the infection removed the trigger and again the autoimmune disease was quiescent. Kidney biopsy was not possible in this case due to abrupt presentation with multiple critical organ involvements. Nephritis subsided with other features. The after-effect of cyclophosphamide and rituximab in the remission of disease (particularly nephritis) cannot be totally ruled out. Moreover, it can also be argued that tuberculosis was just a secondary infection to aggressive immunosuppression used.
Herpesviruses, for example, Epstein–Barr virus, CMV, and HSV has been linked to the etiopathogenesis of SLE. In a literature reviewing, 88 cases with viral infection in SLE-patients, 28% (n = 25) were associated with new-onset-SLE. Of 25-patients, 8-showed spontaneous disappearance of symptoms within 1 and 6-month of acute viral-infection and were classified as primary viral-infection mimicking SLE (parvovirus B19 7 cases and one patient with hepatitis-A virus), 11-patients required use of steroid to control symptoms, 4-had altered immunological markers at the end of follow-up and 2-cases of SLE recurred in spite of therapy. Common clinical manifestations were fever, arthralgia, malaise, hematologic, and cutaneous rash including malar rash. ANA was positive in 23-patients and antibodies to ds-DNA, Smith and antiphospholipid were also found. CMV was diagnosed in 6/25 patients, 4-were treated with ganciclovir and showed clinical improvement.
Out of total 88 patients, 63-viral infections occurred in known cases of SLE. The authors divided these patients in three categories; 18 patients with viral infection mimicking SLE-flare, 36 patients organ-specific viral infections, and 10 patients disseminated multiorgan failure disease. In organ specific category, gastrointestinal (n = 20) was most common. Esophagitis was seen in five-patients, 3HSV, 1 CMV, and 1 varicella-zoster infection. Other manifestations related to HSV-infection were hepatitis, pharyngitis, laryngitis, bilateral pneumonitis, retinitis, cervicitis, and encephalitis. Four patients of HSV with SLE died and diagnosis was disclosed on autopsy as disseminated herpetic infection. In a study from Taiwan, the incidence rate of severe HSV-infection in SLE-cohort was 5.36/10,000 person-years and was significantly higher than the control-cohort with an incidence-rate-ratio of 3.9. They concluded that older patients (>18 years) with previous history of herpes infection and use of intravenous or oral steroid (>7.5 mg) were the risk-factors for severe HSV-infection in SLE-patients. Thus, HSV infection although less-known in SLE is well documented in literature and if missed can lead to mortality., Here, the patient with HSV had symptoms of esophagitis which were confused with steroid withdrawal. Due to persistent symptoms, endoscopy was advised but could be done only when SLE-symptoms flared-up. Treating the infection stabilized the patient on his previous immunosuppressive medicines. Thus, HSV and SLE can have varied presentation and awareness about the same can help the rheumatologist to diagnose it early.
Thus, CMV-infection can closely mimic SLE-disease and it can be at times difficult to differentiate between the two. Tuberculosis can have an atypical presentation in patients of SLE and can mimic lupus pneumonitis. HSV esophagitis is a less known infection in SLE and should be considered in cases complaining of upper gastrointestinal symptoms. Infections should be kept as an important differential diagnosis for difficult-to-treat SLE. This is particularly when the SLE-disease not responding to the usual immunosuppressive therapies. Early diagnosis of such infections in SLE can prevent exposure to aggressive immunosuppression and can be lifesaving.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
We would like to acknowledge the team of doctors involved in managing these patients; Dr Ashutosh Shah (MD General Medicine, Shraddha Hospital, Ahmedabad), Dr Vipul Shah (MD General Medicine, Infectious Disease Specialist), Dr Aniket Shah (MD, Fellowship Infectious disease specialist), Dr Rahul K Jalan (MD DNB Pulmonologist).
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Doaty S, Agrawal H, Bauer E, Furst DE. Infection and lupus: Which causes which? Curr Rheumatol Rep 2016;18:13.
Jung JY, Suh CH. Infection in systemic lupus erythematosus, similarities, and differences with lupus flare. Korean J Intern Med 2017;32:429-38.
Illescas-Montes R, Corona-Castro CC, Melguizo-Rodríguez L, Ruiz C, Costela-Ruiz VJ. Infectious processes and systemic lupus erythematosus. Immunology 2019;158:153-60.
Ospina FE, Echeverri A, Zambrano D, Suso JP, Martínez-Blanco J, Cañas CA, et al
. Distinguishing infections vs. flares in patients with systemic lupus erythematosus. Rheumatology (Oxford) 2017;56:i46-54.
Ramos-Casals M, Cuadrado MJ, Alba P, Sanna G, Brito-Zerón P, Bertolaccini L, et al
. Acute viral infections in patients with systemic lupus erythematosus: Description of 23 cases and review of the literature. Medicine (Baltimore) 2008;87:311-8.
Li TH, Lai CC, Wang WH, Chen WS, Tsao YP, Tsai CY, et al
. Risk of severe herpes simplex virus infection in systemic lupus erythematosus: Analysis of epidemiology and risk factors analysis in Taiwan. Ann Rheum Dis 2019;78:941-6.
Berman N, Belmont HM. Disseminated cytomegalovirus infection complicating active treatment of systemic lupus erythematosus: An emerging problem. Lupus 2017;26:431-4.
Zandman-Goddard G, Shoenfeld Y. SLE and infections. Clin Rev Allergy Immunol 2003;25:29-40.
Li H, Weng P, Najarro K, Xue QL, Semba RD, Margolick JB, et al
. Chronic CMV infection in older women: Longitudinal comparisons of CMV DNA in peripheral monocytes, anti-CMV IgG titers, serum IL-6 levels, and CMV pp65 (NLV)-specific CD8(+) T-cell frequencies with twelve year follow-up. Exp Gerontol 2014;54:84-9.
Keane MP, Lynch JP 3rd
. Pleuropulmonary manifestations of systemic lupus erythematosus. Thorax 2000;55:159-66.
Haupt HM, Moore GW, Hutchins GM. The lung in systemic lupus erythematosus. Analysis of the pathologic changes in 120 patients. Am J Med 1981;71:791-8
Balbi GG, Machado-Ribeiro F, Marques CDL, Signorelli F, Levy RA. The interplay between tuberculosis and systemic lupus erythematosus. Curr Opin Rheumatol 2018;30:395-402.
Yun JE, Lee SW, Kim TH, Jun JB, Jung S, Bae SC, et al
. The incidence and clinical characteristics of Mycobacterium tuberculosis
infection among systemic lupus erythematosus and rheumatoid arthritis patients in Korea. Clin Exp Rheumatol 2002;20:127-32.
[Figure 1], [Figure 2]