One in four lupus patients meets metabolic syndrome criteria

Design

Systematic review and random-effects meta-analysis of observational studies

Population

Patients with systemic lupus erythematosus, studies published to January 2026, pooled globally

Primary outcome

Prevalence of metabolic syndrome

Effect

27% overall (95% CI 24 to 30); 36% in men; hydroxychloroquine OR 0.66 (0.52 to 0.83), cyclophosphamide 1.39 (1.06 to 1.81)

Observational studies worldwide were pooled to estimate how common metabolic syndrome is in systemic lupus erythematosus. The answer is 27 per cent overall, 36 per cent in men, and varying by region from 21 per cent in North America to 36 per cent in Africa, with Asia at 27 per cent.

The drug associations are the part to read carefully. Hydroxychloroquine was associated with lower odds of metabolic syndrome, and cyclophosphamide, azathioprine and mycophenolate with higher odds. Corticosteroids — the drug everyone would name first — showed no significant association, with an interval that just crossed 1.0.

None of this is causal. Patients on cyclophosphamide have more severe disease, more cumulative steroid and more renal involvement than patients on hydroxychloroquine alone, and confounding by indication runs through every one of these estimates. What survives is the prevalence: a quarter of a lupus clinic has metabolic syndrome, and most of them are not being screened for it at the visit where their disease activity is assessed.

• Measure waist circumference, blood pressure, fasting glucose and lipids annually in every lupus patient.
• Use the disease activity visit to do it; these patients rarely see anyone else.
• Do not read the drug associations as a reason to change immunosuppression.
• In Indian practice, use ethnicity-specific waist thresholds — the standard criteria understate central obesity in South Asians.
• Hydroxychloroquine adherence has yet another argument behind it.

Why it matters

The comorbidity most likely to kill a lupus patient is the one not measured at the lupus visit.

Don't overread it

These are associations between drugs and metabolic syndrome, heavily confounded by the disease severity that determined the prescription.

The statistics, in plain English

A pooled prevalence of 0.27 with an interval of 0.24 to 0.30 is a precise estimate of a heterogeneous quantity: studies used different metabolic syndrome definitions and different populations, so the narrow interval reflects the number of studies more than the consistency of the underlying rate. The medication odds ratios are observational associations, confounded by disease severity, and the corticosteroid estimate (1.34, 0.97 to 1.83) is a good example of a result that is almost certainly real and not demonstrated here.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.