Abstract
Systemic lupus erythematosus (SLE) is a systemic autoimmune disease characterised by abnormal autoantibody production and clearance. This immunological background has been suggested to play a role in the susceptibility of SLE patients to infection. Moreover, drugs (most of them immunosuppressive or immunomodulating agents) used in the treatment of moderate and severe lupus give rise to a tendency for infections, including opportunistic ones. Infections may mimic the exacerbations of SLE, leading to confusion over the diagnosis and appropriate treatment. Despite increased awareness of this problem, infections remain a major source of morbidity and mortality in SLE. There are various strategies which can be applied to try and reduce the risk of infection in SLE patients. Options include vaccinations, antibiotic/antiviral prophylaxis and intravenous immunoglobulins.
Abbreviations
ANA
anti-nuclear antibody
C1q
complement factor 1q
C3
complement factor 3
CMV
cytomegalovirus
CRP
C-reactive protein
CVID
common variable immunodeficiency
DNA
deoxyribonucleic acid
ESR
erythrocyte sedimentation rate
EULAR
European League against Rheumatism
H1N1
haemagluttinin 1, neuraminidase 1
HIV
human immunodeficiency virus
HPV
human papilloma virus
iC3b
inactive C3b
IVIg
intravenous immunoglobulin
MBL
mannose binding lectin
MMR
mumps, measles, rubella
PCR
polymerase chain reaction
PJP
Pneumocystis jiroveci pneumonia
SLE
systemic lupus erythematosus
Sm
Smith
SLEDAI
Systemic Lupus Erythematosus Disease Activity Index
Introduction
Systemic lupus erythematosus (SLE) is a systemic autoimmune disease characterised by abnormal autoantibody production and clearance. This immunological background has been suggested to play a role in the susceptibility of SLE patients to infections . Moreover, drugs (most of them immunosuppressive or immunomodulating agents) used in the treatment of moderate and severe lupus give rise to a tendency for infections, including opportunistic ones.
Infections may mimic the exacerbations of SLE, leading to confusion over the diagnosis and appropriate treatment. Despite increased awareness of this problem, infections remain a major source of morbidity and mortality in SLE. Infectious diseases are still the most frequent causes of death in SLE in the first year after onset . The increase in SLE deaths associated with infection, especially pneumonia and septicaemia, is worrisome, especially in some areas of the globe . Indeed, survival rates for SLE patients in developing countries are comparatively lower than those reported in industrialised countries, with early death from infection and active disease .
In this review, we aim to review infections in SLE and appropriate prevention strategies.
Characteristics of infection in SLE
Infections are known to be a major cause of morbidity and mortality in SLE . Several studies evaluated the characteristics of major infections in SLE patients requiring hospitalisation . According to these studies, infections that SLE patients developed were attributed to the same pathogens as in the general population and included community-acquired pneumonia, urinary tract infection and vaginal infection, and some patients may develop tuberculosis. However, despite the pathogens often being the same as in the general population, the clinical manifestations of the infections can be atypical, due to an abnormal immunological response or due to ongoing treatment. Careful inspection and monitoring and timely collection of the specimens for bacterial culture are warranted to avoid misdiagnosis.
Some patients also develop viral, fungal and protozoan infections. Rarely, multiple organisms can be found .
Usually, in an outpatient setting, infections are non-life-threatening ones. Zonana-Nacach et al. reported that in this scenario, infections are associated with disease activity only, independently of socio-demographic and therapeutic factors . However, it is noteworthy that infection in SLE can require hospitalisation, especially when concomitant with a flare (mainly involving the kidney or central nervous system) or when therapy with steroids or cyclophosphamide is ongoing .
Predictors of major infections in SLE
Several studies have analysed the prevalence and associated clinical and laboratory features of infection in SLE . Common themes are medications, including use of steroids and/or cyclophosphamide or high-disease activity as measured by the SLE Disease Activity Index .
Recently, Ruiz-Irastorza and co-workers in a nested case–control study design used within the prospective Lupus-Cruces cohort analysing 249 patients found that the risk of major infections in patients with SLE is mostly influenced by treatment. Prednisone treatment, even at moderate doses, increases the risk, whilst antimalarials have a protective effect . In detail, it was shown that the prednisone dose at the time of the event had a facilitating effect on infections, in agreement with previous studies . Moreover, it is worthy of note that the median dose of patients with major infections was only 7.5 mg day −1 , with an 11-times higher risk of suffering a serious infection for each increase of 10-mg day −1 prednisone.
SLE, infections and mortality
When the frequency, characteristics of the main causes and prognostic significance for morbidity and mortality in 1000 patients with SLE during a 5-year period were analysed, the most frequent causes of death were active SLE, infections and thromboses . In the long-term follow-up of the same study, infection was the present in 36% of the 1000 patients and most frequent causes of death were confirmed to be similarly divided among active SLE (26.5%), thromboses (26.5%) and infections (25%). A survival probability of 92% at 10 years was found .
In a retrospective study performed to describe the characteristics associated with a poor outcome in SLE patients admitted to hospital during a 1-year period, infection was found to be the second cause of hospitalisation after clinical flare of SLE .
Infections according to nature of microbe
Bacterial infections
A wide variety of infectious pathogens have been recognised in SLE. The most frequent types of infections are respiratory, urinary tract and soft-tissue infections . Most infections are caused by common pathogens and include Staphylococcus aureus , Streptococcus pneumoniae , Escherichia coli and Pseudomonas aeruginosa . Indeed, common pathogens that often behave more aggressively than in the healthy population. An increased incidence of Salmonella infection and pneumococcal sepsis is also observed .
S. pneumoniae
S. pneumoniae is a Gram-positive bacterium, which is a common cause of community-acquired pneumonia, meningitis and septicaemia. Several studies indicate that patients with SLE have an increased frequency and severity of S. pneumoniae infections, accounting for 6–18% of all bacterial infections in these patients . It has been suggested that some defective mechanisms in SLE patients can underpin this increased susceptibility to S. pneumoniae infections. Recently, Goldblatt et al. reported that opsonisation of S. pneumoniae with complement factor 3b/inactive C3b (C3b/iC3b) was significantly reduced in serum from patients with SLE compared with patients with non-SLE rheumatic disease and healthy controls, suggesting that a failure to appropriately activate the immune system via complement may contribute to the increased susceptibility of SLE subjects to infections and may correlate with a risk of pneumonia in a subgroup of SLE patients.
Salmonella
Infection with Salmonella species is recognised to be more common in SLE patients than the normal population and may be due to splenic dysfunction or to a defect in opsonisation as previously described . The risk factors of mortality for Salmonella infection have been recently analysed in a cohort of SLE patients hospitalised in a medical centre in Taiwan . Patients with Salmonella infection associated with lupus flare or re-infection with Salmonella species have been found to be the strongest factor associated with a higher risk of mortality . In another study, retrospectively reviewing 50 SLE patients diagnosed with bacteriologically proven non-typhoidal salmonellosis over a 20-year period, it was found that most episodes were bacteraemic without a localising focus, and some patients were afebrile. Mortality in general occurred from concomitant septic shock and major organ failure from active lupus (mainly renal) . It is noteworthy that osteomyelitis of the long bones due to salmonella can occur in SLE patients . Active SLE or co-existent underlying systemic disease, chronic renal failure, and immunosuppressive agents were shown as main predisposing factors . Recently, Navarra et al. described the spectrum of Salmonella infections among Filipino patients with SLE, with typhoid fever and septic arthritis as the most common presentation. Atypical involvement included soft-tissue abscess and meningitis, with the worst prognosis noted in those with sepsis syndrome.
Klebsiella
Some studies showed that Klebsiella was one of the leading causes of Gram-negative bacteraemia in the general population . SLE patients infected with Klebsiella were found to have lower probabilities of 14-day survival in a study analysing the short-term survival of patients with SLE after bacteraemia episodes.
It is noteworthy that sera from patients with Klebsiella pneumoniae were found to contain high titres of the common anti-DNA idiotype . However, the presence of autoantibodies in the serum of patients with Klebsiella infections may be the result of non-specific stimulation due to bacterial polyclonal activation. However, there might also be a specific stimulus triggered by idiotypic cross-reaction between autoantibodies and anti- Klebsiella antibodies .
Mycobacterium tuberculosis
The reported prevalence of M. tuberculosis infection in SLE patients ranges widely from 5% to 30% . The higher prevalence of tuberculous infections in SLE is attributed to multiple immune abnormalities observed in these patients and to the immunosuppressant therapy . The clinical presentation of M. tuberculosis infection seems to be different in SLE patients when compared with the general population; in fact, more frequent extrapulmonary involvement as well as more extensive pulmonary involvement and a high relapse rate even if treated with prophylactic isoniazid have been reported . Mycobacterial infection and SLE may have a similar presentation and may mimic each other. In an individual patient, the differential diagnosis is crucial. In a retrospective analysis involving more than 3000 SLE patients, Hou et al. documented 19 lupus patients with 21 episodes, 10 of which were pulmonary while the other 11 episodes were extrapulmonary (joint, cutaneous or visceral-organ involvement). Fever and cough were found to be the most common manifestations of tuberculosis. However, Sayarlioglu et al. , comparing lupus patients without and with tuberculosis, found that arthritis and renal disease were significantly higher in the latter group, underling the importance of an accurate diagnostic approach.
Opportunistic infections
Increasing evidence indicates that opportunistic infections contribute to the infectious mortality in SLE, as stated before. The burden of opportunistic infections in SLE is complex; often, they are under-reported due to difficulties in diagnosis, as they can mimic or be superimposed upon active lupus. Listeriosis, nocardiosis, candidiasis, cryptococcal meningitis, Pneumocystis jiroveci pneumonia (PJP) and invasive aspergillosis are described in patients with SLE . Sometimes, even more rare infections are reported to occur in SLE, such as haemotrophic mycoplasma .
Cases of SLE with fungaemia or invasive fungal infection are rare but life-threatening conditions in SLE . Severe Candida infection is the most frequently identified opportunistic fungal infection in several SLE series, associated with steroid and cytotoxic drug therapy . Nocardial infections have been also described in steroid-treated SLE patients, and pneumonia and brain abscess are the most frequent clinical presentations . Taken together, all the studies stressed that active lupus disease (SLEDAI > 7) is probably the main risk factor for opportunistic infection. It is noteworthy to remember that low prednisolone doses before fungal infection or high prednisolone doses following fungal infection are associated with higher mortality .
Viral infections
The most commonly reported viral infections in patients who have SLE are parvovirus B19 (HPV-B19) (there are more than 30 reports of primary B19 infection reported as lupus-like syndrome) and cytomegalovirus (CMV) (predominantly presenting in severely immunosuppressed patients). It is not among the purposes of this review to analysis the causative role of virus in the pathogenesis of SLE. Herewith, we focus on the clinical settings when the two conditions, SLE and viral infection, co-exist.
Ramos-Casals et al. described the largest series of acute viral infections in SLE patients. Among 25 patients diagnosed with new-onset SLE, HPV-B-19, CMV, Epstein–Barr virus and hepatitis A were concomitantly detected. In patients already diagnosed with SLE, symptoms related to infection mimicked a lupus flare due to disseminated viral infection and a severe, multi-organ process similar to that described in catastrophic antiphospholipid syndrome was reported. Mortality was high, with 12 patients dying due to infection.
Parvovirus B19
The occurrence of HPV-B19 infection has been documented in patients with SLE, in particular in relation to disease onset. The main reported clinical manifestations were fever, articular involvement, cutaneous lesions, lymphadenopathy, hepato and/or splenomegaly, serositis, renal involvement and cerebral impairment. Cytopaenia was also frequently observed. Thus, the differential diagnosis between HPV-B19 infection and SLE flare is a real challenge, also because HPV-B19 infection may induce a serological profile mimicking a flare. Elevated titres of double-stranded DNA, Sm (Smith), nuclear ribonucleoprotein, Ro-SSA, La-SSB, cardiolipin and/or beta2-glycoprotein I antibodies were reported in concomitant to B19 infection . The B19 infection has usually a self-limiting course; nevertheless, in immunocompromised SLE patients symptoms may persist several months after the viral infection and induce severe clinical settings.
CMV
Sekigawa et al. , reporting SLE patients with spatially related SLE and CMV, emphasised the main features of the complex relationship between SLE and viral infection: a) CMV infection and SLE exacerbation may be impossible or difficult to distinguish, b) the development of SLE may be triggered by a CMV infection and 3) existing SLE may undergo an exacerbation following a CMV infection. Notably, CMV infection can be considered as an opportunistic infection, when affecting SLE patients on chronic steroids and/or immunosuppressive agents. Among the possible CMV-related manifestations, retinitis has to be kept in mind, especially when ongoing therapy includes both azathioprine and low-dose corticosteroid .
HIV
Interestingly, SLE has a lower incidence in the human immunodeficiency virus (HIV)-infected population when compared to the general population . It has also been suggested that SLE may be influenced by HIV infection. It has been suggested that the immunosuppression resulting from HIV infection can prevent the emergence of SLE, as the immunosuppressive effect of HIV may inhibit the development of autoimmune diathesis . To date, several cases of concomitant association between the two diseases have been reported, but the diagnosis was simultaneous in very few of those. Very recently, Carugati et al., reporting two cases and review of the literature, concluded that SLE could occur despite the loss of immunocompetence caused by HIV infection. Moreover, they stated that SLE and HIV infection might influence each other possibly through immunologic mechanisms determining awkward manifestations .
Serological markers of infection in SLE
Conventional biomarkers of lupus flares are hypocomplementaemia, anti-double-stranded-DNA antibodies and erythrocyte sedimentation rate. Their value to predict disease flares and efficacy of therapeutics has been proven; however, even taken together, they are not specific and lack diagnostic accuracy in differentiating between flares and infections . In SLE, unlike in other rheumatic diseases where increase in levels has been observed, changes in C-reactive protein (CRP) level have been less frequently observed during disease flares, proposing this marker as a valid tool to discriminate infection and lupus activity . Albeit with some limitations (such as the presence of arthritis and serositis), to date CRP levels and erythrocyte sedimentation rate/C-reactive protein (ESR/CRP) ratio seem to be the best marker to differentiate SLE activity from infection. Other biomarkers such as procalcitonin, and autoantibodies against complement fraction C1q, have been investigated to distinguish infections from other inflammatory processes but further studies are warranted.
Infections according to nature of microbe
Bacterial infections
A wide variety of infectious pathogens have been recognised in SLE. The most frequent types of infections are respiratory, urinary tract and soft-tissue infections . Most infections are caused by common pathogens and include Staphylococcus aureus , Streptococcus pneumoniae , Escherichia coli and Pseudomonas aeruginosa . Indeed, common pathogens that often behave more aggressively than in the healthy population. An increased incidence of Salmonella infection and pneumococcal sepsis is also observed .
S. pneumoniae
S. pneumoniae is a Gram-positive bacterium, which is a common cause of community-acquired pneumonia, meningitis and septicaemia. Several studies indicate that patients with SLE have an increased frequency and severity of S. pneumoniae infections, accounting for 6–18% of all bacterial infections in these patients . It has been suggested that some defective mechanisms in SLE patients can underpin this increased susceptibility to S. pneumoniae infections. Recently, Goldblatt et al. reported that opsonisation of S. pneumoniae with complement factor 3b/inactive C3b (C3b/iC3b) was significantly reduced in serum from patients with SLE compared with patients with non-SLE rheumatic disease and healthy controls, suggesting that a failure to appropriately activate the immune system via complement may contribute to the increased susceptibility of SLE subjects to infections and may correlate with a risk of pneumonia in a subgroup of SLE patients.
Salmonella
Infection with Salmonella species is recognised to be more common in SLE patients than the normal population and may be due to splenic dysfunction or to a defect in opsonisation as previously described . The risk factors of mortality for Salmonella infection have been recently analysed in a cohort of SLE patients hospitalised in a medical centre in Taiwan . Patients with Salmonella infection associated with lupus flare or re-infection with Salmonella species have been found to be the strongest factor associated with a higher risk of mortality . In another study, retrospectively reviewing 50 SLE patients diagnosed with bacteriologically proven non-typhoidal salmonellosis over a 20-year period, it was found that most episodes were bacteraemic without a localising focus, and some patients were afebrile. Mortality in general occurred from concomitant septic shock and major organ failure from active lupus (mainly renal) . It is noteworthy that osteomyelitis of the long bones due to salmonella can occur in SLE patients . Active SLE or co-existent underlying systemic disease, chronic renal failure, and immunosuppressive agents were shown as main predisposing factors . Recently, Navarra et al. described the spectrum of Salmonella infections among Filipino patients with SLE, with typhoid fever and septic arthritis as the most common presentation. Atypical involvement included soft-tissue abscess and meningitis, with the worst prognosis noted in those with sepsis syndrome.
Klebsiella
Some studies showed that Klebsiella was one of the leading causes of Gram-negative bacteraemia in the general population . SLE patients infected with Klebsiella were found to have lower probabilities of 14-day survival in a study analysing the short-term survival of patients with SLE after bacteraemia episodes.
It is noteworthy that sera from patients with Klebsiella pneumoniae were found to contain high titres of the common anti-DNA idiotype . However, the presence of autoantibodies in the serum of patients with Klebsiella infections may be the result of non-specific stimulation due to bacterial polyclonal activation. However, there might also be a specific stimulus triggered by idiotypic cross-reaction between autoantibodies and anti- Klebsiella antibodies .
Mycobacterium tuberculosis
The reported prevalence of M. tuberculosis infection in SLE patients ranges widely from 5% to 30% . The higher prevalence of tuberculous infections in SLE is attributed to multiple immune abnormalities observed in these patients and to the immunosuppressant therapy . The clinical presentation of M. tuberculosis infection seems to be different in SLE patients when compared with the general population; in fact, more frequent extrapulmonary involvement as well as more extensive pulmonary involvement and a high relapse rate even if treated with prophylactic isoniazid have been reported . Mycobacterial infection and SLE may have a similar presentation and may mimic each other. In an individual patient, the differential diagnosis is crucial. In a retrospective analysis involving more than 3000 SLE patients, Hou et al. documented 19 lupus patients with 21 episodes, 10 of which were pulmonary while the other 11 episodes were extrapulmonary (joint, cutaneous or visceral-organ involvement). Fever and cough were found to be the most common manifestations of tuberculosis. However, Sayarlioglu et al. , comparing lupus patients without and with tuberculosis, found that arthritis and renal disease were significantly higher in the latter group, underling the importance of an accurate diagnostic approach.
Opportunistic infections
Increasing evidence indicates that opportunistic infections contribute to the infectious mortality in SLE, as stated before. The burden of opportunistic infections in SLE is complex; often, they are under-reported due to difficulties in diagnosis, as they can mimic or be superimposed upon active lupus. Listeriosis, nocardiosis, candidiasis, cryptococcal meningitis, Pneumocystis jiroveci pneumonia (PJP) and invasive aspergillosis are described in patients with SLE . Sometimes, even more rare infections are reported to occur in SLE, such as haemotrophic mycoplasma .
Cases of SLE with fungaemia or invasive fungal infection are rare but life-threatening conditions in SLE . Severe Candida infection is the most frequently identified opportunistic fungal infection in several SLE series, associated with steroid and cytotoxic drug therapy . Nocardial infections have been also described in steroid-treated SLE patients, and pneumonia and brain abscess are the most frequent clinical presentations . Taken together, all the studies stressed that active lupus disease (SLEDAI > 7) is probably the main risk factor for opportunistic infection. It is noteworthy to remember that low prednisolone doses before fungal infection or high prednisolone doses following fungal infection are associated with higher mortality .
Viral infections
The most commonly reported viral infections in patients who have SLE are parvovirus B19 (HPV-B19) (there are more than 30 reports of primary B19 infection reported as lupus-like syndrome) and cytomegalovirus (CMV) (predominantly presenting in severely immunosuppressed patients). It is not among the purposes of this review to analysis the causative role of virus in the pathogenesis of SLE. Herewith, we focus on the clinical settings when the two conditions, SLE and viral infection, co-exist.
Ramos-Casals et al. described the largest series of acute viral infections in SLE patients. Among 25 patients diagnosed with new-onset SLE, HPV-B-19, CMV, Epstein–Barr virus and hepatitis A were concomitantly detected. In patients already diagnosed with SLE, symptoms related to infection mimicked a lupus flare due to disseminated viral infection and a severe, multi-organ process similar to that described in catastrophic antiphospholipid syndrome was reported. Mortality was high, with 12 patients dying due to infection.
Parvovirus B19
The occurrence of HPV-B19 infection has been documented in patients with SLE, in particular in relation to disease onset. The main reported clinical manifestations were fever, articular involvement, cutaneous lesions, lymphadenopathy, hepato and/or splenomegaly, serositis, renal involvement and cerebral impairment. Cytopaenia was also frequently observed. Thus, the differential diagnosis between HPV-B19 infection and SLE flare is a real challenge, also because HPV-B19 infection may induce a serological profile mimicking a flare. Elevated titres of double-stranded DNA, Sm (Smith), nuclear ribonucleoprotein, Ro-SSA, La-SSB, cardiolipin and/or beta2-glycoprotein I antibodies were reported in concomitant to B19 infection . The B19 infection has usually a self-limiting course; nevertheless, in immunocompromised SLE patients symptoms may persist several months after the viral infection and induce severe clinical settings.
CMV
Sekigawa et al. , reporting SLE patients with spatially related SLE and CMV, emphasised the main features of the complex relationship between SLE and viral infection: a) CMV infection and SLE exacerbation may be impossible or difficult to distinguish, b) the development of SLE may be triggered by a CMV infection and 3) existing SLE may undergo an exacerbation following a CMV infection. Notably, CMV infection can be considered as an opportunistic infection, when affecting SLE patients on chronic steroids and/or immunosuppressive agents. Among the possible CMV-related manifestations, retinitis has to be kept in mind, especially when ongoing therapy includes both azathioprine and low-dose corticosteroid .
HIV
Interestingly, SLE has a lower incidence in the human immunodeficiency virus (HIV)-infected population when compared to the general population . It has also been suggested that SLE may be influenced by HIV infection. It has been suggested that the immunosuppression resulting from HIV infection can prevent the emergence of SLE, as the immunosuppressive effect of HIV may inhibit the development of autoimmune diathesis . To date, several cases of concomitant association between the two diseases have been reported, but the diagnosis was simultaneous in very few of those. Very recently, Carugati et al., reporting two cases and review of the literature, concluded that SLE could occur despite the loss of immunocompetence caused by HIV infection. Moreover, they stated that SLE and HIV infection might influence each other possibly through immunologic mechanisms determining awkward manifestations .
Serological markers of infection in SLE
Conventional biomarkers of lupus flares are hypocomplementaemia, anti-double-stranded-DNA antibodies and erythrocyte sedimentation rate. Their value to predict disease flares and efficacy of therapeutics has been proven; however, even taken together, they are not specific and lack diagnostic accuracy in differentiating between flares and infections . In SLE, unlike in other rheumatic diseases where increase in levels has been observed, changes in C-reactive protein (CRP) level have been less frequently observed during disease flares, proposing this marker as a valid tool to discriminate infection and lupus activity . Albeit with some limitations (such as the presence of arthritis and serositis), to date CRP levels and erythrocyte sedimentation rate/C-reactive protein (ESR/CRP) ratio seem to be the best marker to differentiate SLE activity from infection. Other biomarkers such as procalcitonin, and autoantibodies against complement fraction C1q, have been investigated to distinguish infections from other inflammatory processes but further studies are warranted.
Related posts:
Adherence to treatment in systemic lupus erythematosus patients
Paediatric-onset systemic lupus erythematosus
Biologics in SLE: Towards new approaches
Fatigue, health-related quality of life and other patient-reported outcomes in systemic lupus erythematosus
Pulmonary hypertension in systemic lupus erythematosus
Managing lupus patients during pregnancy
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
