Survival status was collected up until the end of April 2014. Status was based on records in the various country cohort databases and telephone tracing of those who had no data entered for 24 or more months. Determination of lost to follow-up was when no data had been entered for 24 or more months and at least two attempts at telephone follow-up had been unsuccessful.
Early Deaths Not Found Using Prevalent Cohort
In the inception cohort, 140 deaths (13%) occurred during follow-up. A pooled standardized mortality ratio (SMR) of 4.06 was found (95% confidence interval [95% CI] 3.39–4.85).
In the prevalent cohort the pooled SMR was lower at 3.39 (95% CI 3.06–3.71). In the extra SMR analysis of the noninception cohort, the results were even lower at 3.20 (95% CI 2.86–4.85).
Overall survival in the first decade in the combined inception cohort was lower than in the combined prevalent group at Years 1, 3, 5 and 8. It was also lower than what was seen in the combined non-inception cohort. This difference was greater for men than for women.
“Our work confirmed that studies focusing on prevalent cohorts underestimate mortality,” says Dr. Nikpour. “The pooled SMR for the prevalent cohort was less than the SMR of 4 for the inception cohort. This means that many deaths in scleroderma occur early in the disease course and would be ‘missed’ if only patients with long standing disease (the ‘survivors’) are recruited into cohort studies.”
Cause-of-Death Differences Seen
Comparison of inception and prevalent cohort analyses shows some difference in causes of death in the early rather than later course of disease. Proportionately, in the prevalent cohort, fewer deaths (just over half; 55%) were due specifically to scleroderma. In the inception cohort, that figure was 62.1%.
The remaining one-third of deaths in scleroderma was due to cancers, infections, heart disease and stroke. All of these are thought to occur with increased frequency in scleroderma compared with the general population.
Pulmonary arterial hypertension was a major cause of SSc-related death, both early and later in the disease course (25.3% of deaths in the inception cohort). Interstitial lung disease (ILD) accounted for a higher proportion of deaths in the prevalent cohort than in the inception cohort, pointing to ILD as a risk factor for poor long-term survival opposed to early death.
Carefully Follow Patients
According to Dr. Nikpour, the study’s findings indicate that rheumatologists should follow patients carefully for the development of heart-lung and renal complications, particularly those with the diffuse disease subtype, and treat early and as aggressively as required. Rheumatologists must also be vigilant of cancers and infections, and consider assessing and treating atherosclerotic risk factors in scleroderma patients.
