Systemic lupus erythematosus (SLE) is a chronic, inflammatory, autoimmune disease of unpredictable course, alternating with remissions and exacerbations, in which multiple organs are affected with variable severity. It predominantly affects women of reproductive age, with a prevalence of approximately 1 in 1000, depending on race and socioeconomic status, 9 times higher incidence in women than in men, more frequent in non-Caucasians (Hispanics and African-Americans) and occupies an important place in the differential diagnosis of patients with multisystem diseases1,2.

The extreme variability of SLE in its clinical expression between individuals, as well as in different ethnic and racial groups, and its traditionally unpredictable course with unpredictable reactivations and remissions, have made it necessary to know the overall disease activity over time, as the severity and duration of the active inflammatory process in a particular affected organ is an important predictor of organ damage and mortality. Disease activity and cumulative damage are important in assessing prognosis3.

The Mexican SLE activity index (MEX-SLEDAI), suitable for assessing activity in SLE patients, is a simplification of the SLEDAI, in which some manifestations such as fatigue are added and headache, visual disturbances and immunological tests such as complement and anti-DNA antibodies are excluded4,5. The cumulative damage index, developed by international clinical collaborators of the American College of Rheumatology (SLICC-ACR-DI), is a validated instrument to measure damage, regardless of cause6,7.

Cross-sectional and retrospective studies have assessed variables that are predictive or associated with harm. The Latin American Group for the Study of Lupus (GLADEL) and the Lupus in Minorities Registry (LUMINA) have reported that disease manifestations and severity may vary in different racial/ethnic populations and socioeconomic status. Latin American, mestizo (mixed Amerindian and European descent) and African-American patients may develop the disease at a younger age, with more severe clinical manifestations, higher activity, cumulative damage and higher activity-related mortality and infections8,9.

Ethnicity plays an important role in the development of SLE, and people of Amerindian origin are more susceptible to developing the disease. In Mexico, women with SLE appear to have more severe disease, a lower age of onset than their European counterparts, as well as a higher frequency of disease activity flares. Significant ethnic variability has been reported among Mexican subpopulations, resulting from the combination of Amerindian, European and, to a lesser extent, African parental groups10,11. The Yucatan subpopulation has a distinctive Amerindian (Mayan) contribution, which tends to decrease towards the north of our country, and in the same sense the European increases proportionally, and they represent an ethnic group geographically distant from other Mexican Amerindian groups12,13. On the other hand, studies conducted in women with SLE from the Mayan subpopulation in Yucatan have shown the involvement of immune mechanisms and genetic factors associated with the development of the disease14–18.

Our objective was to determine the clinical evolution of activity and accumulated organ damage over the years in a dynamic cohort of patients with SLE in the Yucatan peninsula, as well as the prevalence and incidence of the disease.

Follow-up began in January 1995 and concluded in December 2016. The dynamic cohort consisted of 185 female and 15 male patients from the 3 states that make up the Yucatan peninsula, most of them from the state of Yucatan, followed by Campeche and Quintana Roo. The high school level was most frequently associated with teaching, and the medium and low levels with housework. During follow-up, there were 19 deaths and 25 paediatric patients with an age of onset between 10–16 years (Table 1).

Based on data provided by the yearbook of the Statistics Department of the ISSSTE Merida Regional Hospital, considering an average annual population corrected for ages 10–69 years of 250,788 and 105,582 beneficiaries for the peninsula and the State of Yucatan, respectively, an average prevalence of 48.43 (95% CI: 46.26–50.60) and 69.56 (95% CI: 64.20–74.93) per 100,000 beneficiaries was estimated for the period from January 1999 to December 2016. The average estimated incidence for the same period was 2.86 (95% CI: 2.18–3.54) and 4.31 (95% CI: 3.27–5.36) per 100,000 beneficiaries for the Peninsula and the state of Yucatan, respectively.

In relation to the disease, 62% of patients had mild, moderate and severe persistent activity. In those who presented exacerbations, these were mostly mild (Fig. 1).

Figure 1.

Clinical course of SLE activity as determined by the MEX-SLEDAI during the follow-up of the 200-patient cohort, as described in “Patients and methods”.

(0.21MB).

In the female patients we observed a decrease in activity during follow-up, with a tendency towards persistent, mild and moderate activity, and progression of accumulated damage that seems to stabilise from the ninth year of follow-up (Fig. 2A). In men, reactivations were frequent and severe, with an increase in cumulative damage (Fig. 2B).

Figure 2.

Activity (MEX-SLEDAI) and cumulative damage (SLICC-ACR-DI) indices in patients A) women and B) men, during the years of follow-up.

(0.37MB).

The mean MEX-SLEDAI in all patients during follow-up was 4.63 ± 2.85, and the clinical manifestations most frequently related to activity were mucocutaneous, immunological and joint manifestations, followed by fatigue/fever, haematological, renal, neurological, vasculitis, serositis and myositis. Regarding immunological markers, 97.5% presented ANA and anti-DNA, at titres greater than 1:80, with a homogeneous diffuse and mottled pattern, and 16.4% had antiphospholipid, anticardiolipin and lupus anticoagulant antibodies. With regard to SLICC-ACR-DI 66% of patients had values ≥1, with an overall mean of 1.10 ± 1.29, and the organs with the most frequent damage were musculoskeletal, neurological and gonadal failure, followed by renal, cutaneous, ocular, cardiovascular, diabetes mellitus, pulmonary, gastrointestinal and neoplasms. Hyperlipidaemia and antiphospholipid syndrome, followed by Rhupus, were the most frequent manifestations associated with SLE (Table 2).

The means of the MEX-SLEDAI index were compared with the different variables (Table 3), and we observed that men had higher activity than women. Patients with evolution and follow-up times of less than 10 years had significantly higher activity. Patients with renal, neurological, vasculitis, fatigue/fever and myositis activity were significantly associated with higher disease activity, as were those with more than 3 reactivations during follow-up. No differences were observed in the means of the activity index in relation to school level and age at disease onset. On the other hand, patients with higher mean MEX-SLEDAI received higher doses of prednisone, cyclophosphamide pulses, methylprednisolone, immunomodulators, biologic therapy and were significantly associated with deaths.

When analysing the average SLICC-ACR-DI values with the same variables (Table 4), it can be seen that men show higher damage indices than women, but not significantly, and those with evolution and follow-up of more than 10 years and with more than 3 relapses showed greater organ damage. The most affected organs were musculoskeletal, renal, gonadal, neurological, ocular and cardiac, among others; all with significant differences with respect to those who did not present damage to said organ. The SLICC-ACR-DI averages were significantly higher in those who died and in those who received higher doses of prednisone, immunomodulators, cyclophosphamide pulses, methylprednisolone pulses and biological therapy, compared to those with less activity and accumulated damage.

During the follow-up period, the 19 deaths were mainly due to systemic vascular complications, sepsis, and renal and hepatic failure, and involved severe persistent activity. The mean MEX-SLEDAI was significantly higher in those who died than in those who remained alive (7.81 ± 3.82 vs. 4.25 ± 2.48, p = .00079), as was the mean SLICC-ACR-DI index (2.77 ± 1.55 vs. .93 ± 1.13, p = .00006) (Table 5).

The cumulative individual survival probability over the time of evolution was calculated (Fig. 3), and shows a survival of 98.9% (95% CI 97.3–100 %) at 5 years, 94.4% (95% CI 90.7–98.3 %) at 10 years and 87.8% (95% CI 81.8–94.3 %) at 20 years. At the end of follow-up survival was 43.7% (95% CI 20.5–93.1 %) in those patients with 30 years of evolution.

Figure 3.

Kaplan-Meier curve of survival time (solid line) in relation to time to progression, with 95% confidence interval (dotted lines) of the 200 SLE patients.

(0.07MB).

This is the first study on the follow-up of clinical activity and cumulative damage in a dynamic cohort of entitled patients with a diagnosis of SLE, who regularly attended the rheumatology service of a tertiary level public hospital that receives patients from the 3 states of the Yucatan peninsula in Mexico: Yucatan, Campeche and Quintana Roo. Patients from this region are a little-studied mestizo ethnic/racial group of Mayan Amerindian descent, sharing sociodemographic, cultural, geographic and racial characteristics, different from others in Latin America11,12.

Disease incidence and prevalence have been reported to vary between populations in different geographic areas and racial/ethnic groups. In the USA, during the period 2002–2004, Lim et al. studied 1446 cases of SLE from 2 communities in Georgia (Fulton and DeKalb), and estimated an age-adjusted incidence and prevalence of 5.6/100,000 and 74.4/100,000, respectively, much higher in women of colour than in men25. Somers et al. studied 2139 cases from a sociodemographically diverse population in southeast Michigan, and reported a similar incidence and prevalence of 5.5/100,000 and 72.8/100,000, respectively, with women of colour being more affected26. On the other hand, in the period from 2007 to 2009, Dall'Era et al. analysed 1257 cases from San Francisco, including different ethnic groups such as Caucasians, Asians, African-Americans and Hispanics, and reported an incidence and prevalence in Hispanics of 5.6 and 110.5/100,000 inhabitants, respectively27. In the same period Izmirly et al. studied 1078 cases from New York and observed an incidence and prevalence in Hispanics of 4.1 and 84.6/100,000 respectively28. In Mexico, Peláez-Ballestas et al., studying non-traumatic musculoskeletal pain during the last 7 days in 5 regions of Mexico and the southeast of the country, including Yucatán, by COPCORD methodology report a national SLE prevalence of .06% and in Yucatán of .07% (95% CI: .01–.02)29. In our study, considering the number of patients analysed in the 18-year period (1999–2016), the estimated average annual incidence and prevalence of the disease in a tertiary public health institution in the Yucatan peninsula is lower compared to those mentioned above, which were conducted multicentrically and in an open population. However, the incidence and prevalence in the State of Yucatan tend to be similar, and seem to correlate with the prevalence reported by Pelaez-Ballestas et al.

In relation to disease, despite the sample size and long follow-up period, our data show a higher frequency in female patients (92.5%) compared to male patients (7.5%), and correlate with those reported in the GLADEL (89% vs. 11%) and LUMINA (93% vs. 7%) studies. The age of onset in our patients is associated with the documented reproductive stage and correlates with that reported in these same studies8,9.

During the follow-up of the clinical course, we observed that most of our patients had persistent activity, followed by reactivations/remissions and clinical remission. This clinical behaviour is similar to that reported by Barr et al. who found 58% persistent activity, 26% reactivations/remissions and 16% remissions with global medical assessment, and 40% chronic activity, 35% reactivations/remissions and 25% remissions with MEX-SLEDAI21. On the other hand, in our cohort we observed a delay in diagnosis, with a mean of 6 months from symptom onset, and the most frequent clinical course was persistent activity and mild to moderate reactivations, similar to that reported by Nikpour et al.30 The mean MEX-SLEDAI value in our cohort was 4.63 ± 2.85, lower than that of the GLADEL study of 8.6 ± 4.9, probably related to the fact that in our cohort of patients all had social security medical service and continuous follow-up every 3 months in the rheumatology area. During follow-up, male patients had significantly higher activity rates than female patients, comparable to the study by Molina et al. in male Latin American patients with more severe disease31.

In relation to clinical manifestations related to disease activity, we found a higher frequency of mucocutaneous and fatigue/fever compared to the GLADEL study (90.7% and 52.9%). However, we observed lower frequency of musculoskeletal, renal and haematological manifestations with those reported by LUMINA (93%, 62%, 90%) and GLADEL (92.5%, 58.3%, 74.3%), and neurological and pleuritis was similar to that of GLADEL mestizos (27.7% and 20.9%)8,9.

With regard to SLICC-ACR-DI, the mean value was 1.1070 ± 1.29, higher than that reported by GLADEL (.59 ± 1), probably due to the longer evolution and follow-up time of our cohort. On the other hand, the use of oral and intravenous steroiSD, antimalarials (hydroxychloroquine) and immunosuppressants (azathioprine and cyclophosphamide) in our cohort of patients were slightly higher than in the GLADEL study, which is related to the increase in organ damage and duration of follow-up. However, we observed a 5-year survival rate of 98.9%, higher than the GLADEL study of 95% at 4 years. Musculoskeletal organ damage and gonadal failure did not influence survival, while patients with higher activity and renal, neurological and cardiopulmonary damage received higher doses of steroiSD, immunosuppressants and biologic therapy, which is related to better control of activity and relapse and increased organ damage. 9.5% of deaths were associated with persistent high activity with cumulative organ damage related to systemic vasculitis, gastrointestinal sepsis, chronic renal and liver failure, and to a lesser extent neoplasms.

The course and clinical presentation during follow-up in the cohort of SLE patients from the Yucatan peninsula appears to be similar to that described in other Latin American populations. However, during follow-up we observed a predominance of persistent activity with mild to moderate reactivations, and more aggressive behaviour in men than in women; with longer evolution and follow-up time, we observed a decrease in clinical activity and an increase in cumulative damage, probably conditioned by treatment. The average MEX-SLEDAI in our study was lower than reported in other cohorts, while the average SLICC-ACR-DI increased during follow-up. The higher observed survival and lower renal involvement suggests a more benign course of the disease in the population of the Yucatan Peninsula, Mexico.

All the authors have no conflict of interests to declare related to this research.