Lupus nephritis (LN) is a major cause of morbidity and mortality in systemic lupus erythematosus (SLE).1 As previously described, LN occurs more frequently in Hispanic and African-American people in the multiethnic US LUMINA (Lupus in Minorities: Nature vs Nurture) cohort.2 With more patients (one-third) and years of observation (2949.3 patient-years from enrolment or baseline), we have now examined whether LN occurs earlier in the disease course in these patients relative to Caucasians. Four hundred and forty-nine patients (American College of Rheumatology (ACR) criteria, age ≥16 years, disease duration ≤5 years, predominantly women) were studied. Time-to-LN (ACR criterion (persistent proteinuria >0.5 g/day or ≥3 if quantitation not performed or cellular casts)) from baseline (incident cases) was examined by multivariable Cox proportional hazards regression adjusting for pertinent baseline clinical and socioeconomic-demographic variables. The results (table 1) are expressed as HRs; a HR >1 represents a shorter time to the event and a HR <1 represents a longer time. Age, gender and ethnicity were entered in all models. Ethnicity remained a risk factor for LN; Hispanic (HR 2.32, 95% CI 1.03 to 5.25; p=0.043) and African-American ethnicities (HR 2.57, 95% CI 1.26 to 5.26; p=0.001) were associated with a shorter time-to-LN occurrence compared with Caucasians, as were photosensitivity (HR 2.10, 95% CI 1.09 to 4.01; p=0.025), serositis (pleurisy/pericarditis) (HR 2.00, 95% CI 1.09 to 3.67; p=0.025) and the immunological criterion (HR 5.56, 95% CI 2.12 to 14.60; p=0.001). Otherwise, age was weakly associated with a longer time-to-LN (HR 0.97, 95% CI 0.94 to 0.99, p=0.033). When poverty was entered into the model, however, only the association with African-American ethnicity remained, albeit weaker (data not shown).

Several studies have reported a higher frequency of LN in Hispanic and African-American people compared with Caucasians.2 3 Higher disease activity, relapse rates, chronic renal failure and early mortality have also been reported in these patients.3,–,6 In addition, ethnicity may influence the treatment response in LN7 and, along with poverty, is a factor in the progression of LN to renal damage.8,–,10 However, these studies have suggested—but have not conclusively shown—that LN is an early event in the course of SLE (compared with Caucasians). Interestingly, >80% of our prevalent LN cases occurred in non-Caucasians.

LN is known to be associated with anti-DNA antibodies and activity in other organ systems, so the association with the immunological criterion, photosensitivity and serositis (pleurisy and pericarditis) was not unexpected.

Our inability to disentangle the biological versus the environmental (socioeconomic levels, poverty included) components of ethnicity as a risk factor for early occurrence of LN is the main limiting factor of our study. We have previously shown that these patients have higher poverty levels than Caucasians, which may explain the loss of significance for Hispanics when poverty was included in the model. Our data should therefore be interpreted with caution owing to their observational nature, possible changes in poverty levels after disease onset and the self-reported definition of ethnicity.

In summary, US patients of African and Hispanic ancestry should be carefully monitored so that, when LN supervenes, they can be aggressively treated to prevent the occurrence of renal damage and its devastating consequences.