Mortality in a cohort of Egyptian systemic lupus erythematosus patients: retrospective two-center study

Systemic lupus erythematosus is a debilitating autoimmune disease with major contribution to the worldwide morbidity and mortality. This study aimed to investigate the causes of mortality in systemic lupus erythematosus (SLE) patients and the relation between clinical activity, disease-associated end-organ damage, laboratory markers and mortality. Among the 771 patients who were successfully followed up, 34 patients (4.4%) died. The leading causes of death were infectious causes (35.29%), cardiopulmonary causes (26.48%), renal causes (14.7%), unknown causes (14.7%), neuropsychiatric causes (5.88%), and lastly gastrointestinal causes (2.94%). Subjects who died had lower complement 3 level, more anemia, lymphopenia, neutropenia, leukocytosis, thrombocytopenia, decreased glomerular filtration rate, higher incidence of infection, end-stage renal disease, and cardiopulmonary complications. Higher glucocorticoid dosage with more immunosuppressant (mofetil and cyclophosphamide) treatment was observed in patients who died. SLE disease Activity Index and Systemic Lupus International Collaborating Clinics damage index were both significantly higher in deceased persons. Multivariable hazards regression analysis revealed that lymphopenia (p = 0.017), decreased glomerular filtration rate < 50% (p = 0.002) with end-stage renal disease (p = 0.001), and high steroid daily use of > 40 mg (p = 0.016) were independent risk factors for the mortality of SLE patients. Infections and cardiopulmonary complications are the leading causes of death in two centers caring for Egyptian SLE patients. Lymphopenia, end-stage renal failure, and high steroid daily use were associated with poor outcomes.


Background
Systemic lupus erythematosus (SLE) is a multisystem autoimmune disease, occurring most frequently in women during reproductive period of life with a female to male ratio of 10:1. It is a debilitating autoimmune disease with major contribution to the worldwide morbidity and mortality. This chronic, multisystem disease is characterized by widespread affections of different body organs in a progressive pattern with recurrent exacerbating attacks [1,2].
The risk of mortality in patients with SLE has previously been documented to range from twofold to fivefold relative to the general population [3][4][5][6]. Men are more likely to have higher mortality rates and a higher rate of fulminant illness than women with SLE [7]. In contrast to earlier controls, survival has improved dramatically [8,9]. Mortality rates tend to differ across race, ethnicity, and region, so when looking at the data, it is very important to keep all these factors in mind. Lupus was viewed as a rapidly lethal disease in the early to the mid-20th century, as medical options were limited and deficient. The predicted 5-year survival of SLE patients was almost 50% during the 1950s. Diagnostic methods and interventions were not so advanced; thus, it took longer to diagnose and thoroughly treat patients [10]. Studies of Afro-Caribbean women in Barbados [11] have shown a very high incidence of SLE and lupus nephritis (LN), as well as a high mortality rate and low 5-year survival (79.9%). Survival for SLE patients in North America showed marvelous improvement in the second half of the century. From 1975 to 1990, the 10-year survival improved obviously from 64 to 87%. Even more improvement has taken place from 1990 to 2004, when the 20year survival increased to 78% [10].
Nevertheless, as shown by Jorge et al. [12], this fabulous decrease in mortality seems to have plateaued after the 2000s. In contrast to the general population, early mortality remains substantially higher in SLE patients. The causes of this increased mortality have been investigated in numerous studies, which suggest that patients with SLE lose 2 decades of life on average [13]. Numerous causes of multiple organ damage are of greatest importance; however, increased activity of SLE disease is also considered an important related factor [14,15]. Several studies have shown a double peak in early lupus patients with higher mortality rates due to infection or active disease and later in the course of the disease (due to cardiovascular, cerebrovascular, infections, and disease flares). Increased mortality can also be attributed to the disease itself, disease complications, treatment complications, or even unrelated events [16][17][18].
It is important to note that despite the increase in survival of SLE patients, they continue to experience considerable morbidity and mortality.

Aim, study design and patients
This study aimed to investigate the causes of mortality in SLE patients and the relation between clinical activity, disease-associated end-organ damage, laboratory markers and mortality.
In this retrospective study, the medical records of all SLE patients under follow-up at the outpatient and inpatient clinics of Cairo and Kafr Elsheikh University hospitals through the period from November 2014 to August 2019 were included. Adult patients were included if they fulfilled the 2012 criteria of Systemic Lupus International Collaborating Clinics (SLICC) [19].

Data collection
The following data were retrieved from the records of eligible patients: demographics characteristics: age and duration of disease, comorbidities, complete blood count, immunological profile, SLE Disease Activity Index (SLEDAI) [20], SLICC Damage Index score [21], drug therapy, and cause of mortality in deceased persons.

Statistical analysis
Continuous data were expressed as mean ± SD. An independent-sample t test was used for comparing the variables between the two studied groups with group I representing the deceased patients and group II the survived ones. Non-normally distributed data were analyzed using the Mann-Whitney U test. Categorical data were stated as numbers or percentages. Either the chi-square test or the Fisher's exact test was used to analyze the relation between categorical variables. The risk factors for all-cause mortality in patients with SLE were determined using univariate analysis Cox proportional hazard model. Multivariate Cox regression analysis was done for the variables with statistical significance from the univariate analysis. The stepwise method was used for the multivariate analysis, and the results were expressed as 95% confidential intervals (95% CI). Two-tailed tests were used, and p < 0.05 was considered statistically significant. Data were analyzed using SPSS 13.0 (SPSS Inc., Chicago, IL, USA).
Regarding the demographic data, the sex, age and disease duration did not differ in both groups. Subjects who died had lower complement 3 (C3) level, more anemia, lymphopenia, neutropenia, leukocytosis, thrombocytopenia, and decreased glomerular filtration rate (GFR < 50%). Subjects who died had a significantly higher incidence of infection, end-stage renal disease (ESRD), and cardiopulmonary complications. There were also significant differences between the two groups in glucocorticoid dosage with more immunosuppressants (mofetil and cyclophosphamide) treatment observed in patients who died. SLEDAI and SLICC damage index were both significantly higher in deceased persons reflecting higher disease activity and end-organ damage (Tables 2, 3 and 4) Univariate cox regression model stated that lymphopenia (p < 0.001), thrombocytopenia (p < 0.001), low C3 levels (p < 0.007), lower GFR (p = 0.003), ESRD (p < 0.001), SLICC damage index (p = 0.001), immunosuppressant drugs and methylprednisone pulse therapy, high steroid daily use (p = 0.001), and infections (p < 0.001) are risk factors for mortality. While in the multivariate analysis, lymphopenia (p = 0.017), lower GFR (p = 0.002), ESRD (p = 0.001), and high steroid daily dose (p = 0.016) were independent risk factors for mortality (Table 5).

Discussion
In this study of Egyptian SLE cohort, we reported mortality rate of 4.4% (with infection and cardiopulmonary issues as the leading causes of death). Of course, if this is compared to the mortality rate of the Egyptian population which ranged between 5.8 and 6 per 1000 people in the same duration of our study, the lupus burden will be very obvious. Overall, our cohort mortality is comparable to the findings of many researches from other countries [22][23][24][25][26][27][28], and while the leading cause of death is not compatible with that of many Western and American nations [16,17,22,23], which stated cardiopulmonary causes as the leading cause of death, it goes in accordance with many studies with a recent one published in 2018 that stated infection as the leading cause of death. Of course, race and more aggressive treatment are contributing factors [24][25][26][27][28].
A wide range of demographic, clinical, and laboratory variables have been postulated as risk factors for infections in SLE. These include low socioeconomic status, nephritis, high disease activity, low complement level, and the degree of immunosuppression therapy [28][29][30][31].
Generally, cardiovascular disease, infection, and active disease are the main specific causes of death in SLE populations. Typically, in the first few years of illness, the primary causes of death are serious infection due to immunosuppression or death from complications of active disease such as lupus nephritis or neuropsychiatric lupus, whereas long-term damage such as ESRD, treatment complications, and, most importantly, cardiovascular disease are primary causes of late death [32][33][34].
Of utmost importance is the cardiovascular risk in SLE specially accelerated atherosclerosis. Premature atherosclerosis is an apparent leading cause of SLE morbidity and mortality. The likelihood of death due to cardiovascular disease remains largely unchanged despite the decrease in all-cause mortality in SLE over the past 20 years [3,35,36].
There was no statistical difference regarding the demographic data among died and survived groups. This is not in accordance with many studies which reported sex and disease duration as risk factor for mortality, but the most consistent finding in these studies was that the overall standardized mortality rate in men was comparable to that in women. Although many studies reported a difference in demographic data, they were always in controversy that may be explained by different environmental factors, ethnic group, and even accessibility of medical care [7,[29][30][31][33][34][35][36][37][38]. Nevertheless, 21 deceased patients were of less than 33 years with disease duration of less than 5 years which represent a high early mortality in relation to remaining 13 patients.
When we tried to figure out the specific-cause mortality, we noticed that patients with some important organ involvement had significantly higher mortality as end-stage renal failure and cardiopulmonary complications. Also, patients who died had lower C3 level, more anemia, lymphopenia, neutropenia, leukocytosis, thrombocytopenia, and lower glomerular filtration rate < 50%. There was also a statistical significant difference between the two groups in glucocorticoid and immunosuppressant (mofetil and cyclophosphamide) therapy. The link between mortality and the intake of corticosteroids and immunosuppressants could not be discussed without knowing the accurate doses and treatment duration of these drugs. SLEDAI and SLICC damage index were both significantly higher in deceased persons reflecting higher disease activity and end-organ damage. Most of these items reflect high disease activity and cumulative damage, and this emphasizes the utmost importance of disease control and standardized drug therapy.
Similarly, the prevalence of the various causes of death in SLE was investigated in a meta-analysis by Yurkovitch et al. [37]. Twelve trials, including 27,123 SLE patients with 4993 deaths, were chosen. The highest mortality rate was observed for SLE patients with renal disease, and the risk of death due to infection, renal disease, and cardiovascular disease was significantly increased overall.  We confirmed that lupus-related organ involvements had the strongest impact on the prognosis of SLE patients by the univariate and multivariate hazard regression analysis. Lymphopenia, low GFR, ESRD, and high steroid daily were the strongest independent risks of mortality by multivariate hazard regression analysis. This reflects the effect of cumulative renal disease (GFR and ESRD), risk of infection (lymphopenia), and higher disease activity (high steroid daily dose). This was similar to the reports from western countries including the UK, France, and Finland [16,17,22]. Several other studies from developing countries such as Thailand, Malaysia, and Brazil described similar results where the renal burden of the disease was very obvious. Unfortunately, data from eastern African countries is not available [3,[24][25][26][27][28]. Our findings are not in accordance with a Chinese study which found that renal affection was not identified as a risk factor for mortality by the hazard analyses [28].
Lymphopenia is a common finding in active SLE patients that was reported in 90% of lupus patients throughout the disease course. Other factors may cause lymphopenia including infections and medications such as corticosteroids and cytotoxic agents. Lymphopenia has been reported as a risk factor for opportunistic and serious infections [39][40][41]. Although not reported in many studies as a risk factor for mortality [22][23][24][25][26][27][28], it was raised up in our study, giving great importance for hematological changes in SLE patients.
This research has some limitations. Despite considerable efforts to determine the cause of death in each subject, in 5 patients who were reported as an unknown cause of death, we were unable to confirm this, and more than one cause was also apparent in a few cases.
Unfortunately, accurate doses and treatment duration of the immunosuppressants intake were not available in this retrospective study. Definitely, the small number of cases is a limitation of the study so a multicenter, prospective longitudinal study with larger subject number will more accurately quantify mortality and evaluate the impact of treatment on mortality in Egyptian SLE patients.

Conclusion
Infections and cardiopulmonary complications are the leading causes of death in two centers caring for Egyptian SLE patients. Lymphopenia, end-stage renal failure, and high steroid daily use were associated with poor outcomes.
Tight strict control of the disease activity should be targeted and regular assessment of the response to immunosuppressant with strict adherence to the management recommendations is mandatory for SLE patients. With the advent of earlier diagnosis and recognition of disease, as well as the introduction of newer less toxic therapeutic measures, it would be expected that mortality in SLE patients would decline but increased mortality in comparison to general populations remains part of the natural history of lupus.
Future research, with particular emphasis on infection risk factors, cardiopulmonary diseases and lupus nephritis, should concentrate on the development of novel therapies and the treatment of comorbid conditions of SLE. As always, screening and prevention are the keys to preventing illness and therapy complications.
The importance of this research lies in being one of the earliest studies addressing mortality problems in African SLE patients, which may stimulate further reports

Funding
This study has no funding sources.

Availability of data and materials Available
Ethics approval and consent to participate We confirm none of the present study's procedures had violated the principles stated by the latest version of the Declaration of Helsinki. The protocol of the present study was registered by the local ethics committee of Kafr Elsheikh Faculty of Medicine with approval code MKSU4-11-2020. We confirm that the manuscript has been read and approved by all the authors, that the requirements for authorship as stated earlier in this document have been met, and that each author believes that the manuscript represents honest work.

Consent for publication
Not applicable. ESRD End-stage renal disease, GFR Glomerular filtration rate, APL Antiphospholipid antibodies, SLICC Systemic Lupus International Collaborating Clinics Damage Index score, SLEDAI SLE Disease Activity Index, Anti-dsDNA Anti-double-stranded DNA antibodies, C3 Complement 3, HR Hazard ratio, CI Confidence interval All variables with p < 0.05 was included in the multivariate *Statistically significant at p ≤ 0.05