In lupus, recent cumulative steroid dose predicts organ damage

Design

Nationwide cohort study with weighted-cumulative-exposure modelling of register data

Population

3,749 patients with newly diagnosed systemic lupus erythematosus (2006–2022)

Primary outcome

First organ damage overall and by domain, versus glucocorticoid dosing window

Effect

Past-2-year exposure best predictive (adjusted hazard ratio 1.42 per 5 mg/day); 3-year window for cardiovascular damage

Glucocorticoids drive much of the long-term damage in systemic lupus erythematosus, but because the dose changes constantly it has been unclear which exposure window matters. This nationwide cohort of 3,749 newly-diagnosed patients used weighted-cumulative-exposure modelling of pharmacy records to find out.

The risk of organ damage was best captured by glucocorticoid exposure over the preceding two years, and fairly well by the past year — with a measurable excess even around 5 mg/day of prednisolone (adjusted hazard ratio about 1.42 per 5 mg/day over two years). For cardiovascular damage specifically, the relevant window stretched to three years.

The practice point is how you assess steroid risk. A single current dose understates it; what matters is cumulative exposure over the last one to three years. When weighing damage risk, deciding on a taper, or counselling a patient, look back across that window — and treat even 'low-dose' maintenance steroids as contributing to accrual rather than as risk-free.

• Organ-damage risk was best predicted by glucocorticoid exposure over the past 2 years (adjusted hazard ratio 1.42 per 5 mg/day, 95% CI 1.31–1.53).
• The past 1 year captured much of the risk too (adjusted hazard ratio 1.34 per 5 mg/day).
• For cardiovascular damage, the relevant exposure window extended to 3 years.
• Assess cumulative steroid exposure over the last 1–3 years, not just the current dose, when judging damage risk.

Why it matters

It shifts steroid-risk assessment from today's dose to the running total over years, and marks 'low-dose' maintenance as not harmless.

Don't overread it

Observational register data — it shows association between cumulative exposure and damage, not that a specific tapering strategy prevents it.

The statistics, in plain English

A hazard ratio of 1.42 per 5 mg/day means each sustained 5 mg of daily prednisolone over the window raised organ-damage risk by about 42%, with a tight interval making it a firm association. The modelling identifies which past period best predicts damage rather than a single snapshot.

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.