Relationship between change in skin score and disease outcome in diffuse cutaneous systemic sclerosis: application of a latent linear trajectory model.

OBJECTIVE: To explore the relationship between changes in the severity of skin disease and morbidity and mortality in patients with diffuse cutaneous systemic sclerosis (dcSSc).
METHODS: From a large single-center cohort, we identified 225 patients with dcSSc for whom serial clinical information was available from within 24 months of the onset of the first non-Raynaud's phenomenon manifestation of SSc. The end points analyzed included death and heart, lung, kidney, and gastrointestinal tract involvement. Latent linear trajectory modeling (LTM) was applied to identify patients with a similar trajectory of modified Rodnan skin thickness score (MRSS) changes over the first 3 years of followup. Clinical outcomes were compared between 3 different LTM subgroups.
RESULTS: LTM permitted classification of 131 patients (58%) into 1 of 3 subgroups with different skin score trajectories. Survival was lowest in the subgroup of patients who had a high baseline skin score and experienced little improvement during followup (P = 0.003). However, the frequency of clinical end points was similar in the subgroup with the most favorable trajectory (i.e., a low initial MRSS and subsequent improvement) and the subgroup with a high baseline MRSS and no improvement. Interestingly, the end point frequency was greatest in the subgroup with a high initial MRSS and subsequent improvement, suggesting that sustained severe skin disease does not necessarily predict the number of visceral complications, and that the relationship between the skin score and internal organ involvement in dcSSc is more complex than previously thought.
CONCLUSION: Although mortality was highest among patients with the worst skin-related outcomes, no simple relationship between burden of disease and change in skin score was observed.