Higher risk of death seen with oral steroids in RA interstitial lung disease

The use of prednisone for 3 or more months at a time was associated with a significantly elevated risk of death in patients with rheumatoid arthritis and interstitial lung disease in a retrospective cohort study.

Interstitial lung disease is present in about 5% of patients with rheumatoid arthritis. For years, oral steroids were commonly used in patients with the disease, but today’s rheumatologists "no longer view oral steroids as optimal treatment in RA-ILD [rheumatoid arthritis–associated interstitial lung disease], and our data now confirm that," said Dr. Clive Kelly of Queen Elizabeth Hospital in Gateshead, England, senior investigator of the study. He added that clinicians should avoid long-term treatment with steroids in this patient group whenever possible.

Dr. Kelly led the British Rheumatoid Interstitial Lung (BRILL) Network’s cohort study of 260 patients with RA-ILD diagnosed over a 25-year period. The BRILL study compared patients with RA-ILD and an equal number of RA controls without lung involvement who were matched for age, sex, and time of diagnosis.

At the annual European Congress of Rheumatology, Dr. Kelly reported that steroid-treated RA-ILD patients, who represented nearly 60% of the cohort, had an elevated relative risk of all-cause death, compared with those who had never been treated with steroids (RR, 1.65; 95% confidence interval, 1.2-2.3; P = .002). Although the relative risk of respiratory death was significantly increased for RA-ILD patients, regardless of treatment, when compared with RA patients without ILD, the risk was higher in those who had been on steroids (RR, 2.75; 95% CI, 1.6-4.7; P = .0002) than in those who had not received steroids (RR, 2.06; 95% CI, 1.1-3.8; P = .02), the investigators found.

The comparison also revealed other important findings related to RA-ILD. Patients with RA-ILD had significantly higher mortality than did those with RA alone. Over the course of the 25-year study period, however, mortality progressively improved among the RA-ILD patients, with median age at death rising from 63 to 76 years. This steady improvement, Dr. Kelly said, is partly the result of better and earlier diagnosis of lung involvement.

"It’s one of my many missions in life to get rheumatologists to listen to the lungs when they examine the joints in patients with rheumatoid arthritis," he said. "I think we are getting better. We’ve persuaded the British Society for Rheumatology to incorporate lung function testing and clinical examination of the chest into their basic assessment of a rheumatoid patient."

Also likely affecting the improved mortality seen over the cohort’s study period is a change in therapeutic approach. While RA-ILD patients diagnosed in the first half of the study period were likely to have been treated with only prednisone and azathioprine, in the latter half they were more likely to have received cyclophosphamide and methylprednisolone or mycophenolate. Over the last 12 years, more were treated with biologics, and in the final 6 years of the cohort, patients requiring biologics tended to be treated with rituximab, a B-cell inhibitor, rather than anti–tumor necrosis factor (anti-TNF) agents, the BRILL investigators found.

Mortality was lower among RA-ILD patients treated with mycophenolate than in those treated with other immunosuppressive agents. Among biologic agents used in the cohort, rituximab treatment was associated with improved mortality, but anti-TNF inhibitors were seen to be associated with elevated risk of death.

About 95% of RA-ILD patients are anticyclic citrullinated peptide (anti-CCP) antibody positive, compared with 55%-60% of the RA population as a whole, Dr. Kelly said, "so there’s a strong statistical association of seropositivity, and in those who are seropositive, rituximab works well."

The finding that rituximab was associated with improved survival in the cohort not only has implications for RA-ILD, he said, but also, potentially, for people with idiopathic pulmonary fibrosis (IPF) who are anti-CCP antibody positive. "What [rheumatologists] have, and chest physicians traditionally don’t, is access to rituximab and mycophenolate. But these might be worth trying in IPF as well," he said.

Dr. Kelly noted that prospective trials in RA-ILD are beginning to enroll patients with progressive disease to compare azathioprine and mycophenolate, allowing for the use of oral steroids, as well as patients with active RA and ILD to compare anti-TNF inhibitors against rituximab, also allowing oral steroids.

Dr. Kelly reported that he had no conflicts of interest related to his findings and that none of his fellow BRILL investigators had conflicts.

The use of prednisone for 3 or more months at a time was associated with a significantly elevated risk of death in patients with rheumatoid arthritis and interstitial lung disease in a retrospective cohort study.

Interstitial lung disease is present in about 5% of patients with rheumatoid arthritis. For years, oral steroids were commonly used in patients with the disease, but today’s rheumatologists "no longer view oral steroids as optimal treatment in RA-ILD [rheumatoid arthritis–associated interstitial lung disease], and our data now confirm that," said Dr. Clive Kelly of Queen Elizabeth Hospital in Gateshead, England, senior investigator of the study. He added that clinicians should avoid long-term treatment with steroids in this patient group whenever possible.

Dr. Kelly led the British Rheumatoid Interstitial Lung (BRILL) Network’s cohort study of 260 patients with RA-ILD diagnosed over a 25-year period. The BRILL study compared patients with RA-ILD and an equal number of RA controls without lung involvement who were matched for age, sex, and time of diagnosis.

At the annual European Congress of Rheumatology, Dr. Kelly reported that steroid-treated RA-ILD patients, who represented nearly 60% of the cohort, had an elevated relative risk of all-cause death, compared with those who had never been treated with steroids (RR, 1.65; 95% confidence interval, 1.2-2.3; P = .002). Although the relative risk of respiratory death was significantly increased for RA-ILD patients, regardless of treatment, when compared with RA patients without ILD, the risk was higher in those who had been on steroids (RR, 2.75; 95% CI, 1.6-4.7; P = .0002) than in those who had not received steroids (RR, 2.06; 95% CI, 1.1-3.8; P = .02), the investigators found.

The comparison also revealed other important findings related to RA-ILD. Patients with RA-ILD had significantly higher mortality than did those with RA alone. Over the course of the 25-year study period, however, mortality progressively improved among the RA-ILD patients, with median age at death rising from 63 to 76 years. This steady improvement, Dr. Kelly said, is partly the result of better and earlier diagnosis of lung involvement.

"It’s one of my many missions in life to get rheumatologists to listen to the lungs when they examine the joints in patients with rheumatoid arthritis," he said. "I think we are getting better. We’ve persuaded the British Society for Rheumatology to incorporate lung function testing and clinical examination of the chest into their basic assessment of a rheumatoid patient."

Also likely affecting the improved mortality seen over the cohort’s study period is a change in therapeutic approach. While RA-ILD patients diagnosed in the first half of the study period were likely to have been treated with only prednisone and azathioprine, in the latter half they were more likely to have received cyclophosphamide and methylprednisolone or mycophenolate. Over the last 12 years, more were treated with biologics, and in the final 6 years of the cohort, patients requiring biologics tended to be treated with rituximab, a B-cell inhibitor, rather than anti–tumor necrosis factor (anti-TNF) agents, the BRILL investigators found.

Mortality was lower among RA-ILD patients treated with mycophenolate than in those treated with other immunosuppressive agents. Among biologic agents used in the cohort, rituximab treatment was associated with improved mortality, but anti-TNF inhibitors were seen to be associated with elevated risk of death.

About 95% of RA-ILD patients are anticyclic citrullinated peptide (anti-CCP) antibody positive, compared with 55%-60% of the RA population as a whole, Dr. Kelly said, "so there’s a strong statistical association of seropositivity, and in those who are seropositive, rituximab works well."

The finding that rituximab was associated with improved survival in the cohort not only has implications for RA-ILD, he said, but also, potentially, for people with idiopathic pulmonary fibrosis (IPF) who are anti-CCP antibody positive. "What [rheumatologists] have, and chest physicians traditionally don’t, is access to rituximab and mycophenolate. But these might be worth trying in IPF as well," he said.

Dr. Kelly noted that prospective trials in RA-ILD are beginning to enroll patients with progressive disease to compare azathioprine and mycophenolate, allowing for the use of oral steroids, as well as patients with active RA and ILD to compare anti-TNF inhibitors against rituximab, also allowing oral steroids.

Dr. Kelly reported that he had no conflicts of interest related to his findings and that none of his fellow BRILL investigators had conflicts.

The use of prednisone for 3 or more months at a time was associated with a significantly elevated risk of death in patients with rheumatoid arthritis and interstitial lung disease in a retrospective cohort study.

Interstitial lung disease is present in about 5% of patients with rheumatoid arthritis. For years, oral steroids were commonly used in patients with the disease, but today’s rheumatologists "no longer view oral steroids as optimal treatment in RA-ILD [rheumatoid arthritis–associated interstitial lung disease], and our data now confirm that," said Dr. Clive Kelly of Queen Elizabeth Hospital in Gateshead, England, senior investigator of the study. He added that clinicians should avoid long-term treatment with steroids in this patient group whenever possible.

Dr. Kelly led the British Rheumatoid Interstitial Lung (BRILL) Network’s cohort study of 260 patients with RA-ILD diagnosed over a 25-year period. The BRILL study compared patients with RA-ILD and an equal number of RA controls without lung involvement who were matched for age, sex, and time of diagnosis.

At the annual European Congress of Rheumatology, Dr. Kelly reported that steroid-treated RA-ILD patients, who represented nearly 60% of the cohort, had an elevated relative risk of all-cause death, compared with those who had never been treated with steroids (RR, 1.65; 95% confidence interval, 1.2-2.3; P = .002). Although the relative risk of respiratory death was significantly increased for RA-ILD patients, regardless of treatment, when compared with RA patients without ILD, the risk was higher in those who had been on steroids (RR, 2.75; 95% CI, 1.6-4.7; P = .0002) than in those who had not received steroids (RR, 2.06; 95% CI, 1.1-3.8; P = .02), the investigators found.

The comparison also revealed other important findings related to RA-ILD. Patients with RA-ILD had significantly higher mortality than did those with RA alone. Over the course of the 25-year study period, however, mortality progressively improved among the RA-ILD patients, with median age at death rising from 63 to 76 years. This steady improvement, Dr. Kelly said, is partly the result of better and earlier diagnosis of lung involvement.

"It’s one of my many missions in life to get rheumatologists to listen to the lungs when they examine the joints in patients with rheumatoid arthritis," he said. "I think we are getting better. We’ve persuaded the British Society for Rheumatology to incorporate lung function testing and clinical examination of the chest into their basic assessment of a rheumatoid patient."

Also likely affecting the improved mortality seen over the cohort’s study period is a change in therapeutic approach. While RA-ILD patients diagnosed in the first half of the study period were likely to have been treated with only prednisone and azathioprine, in the latter half they were more likely to have received cyclophosphamide and methylprednisolone or mycophenolate. Over the last 12 years, more were treated with biologics, and in the final 6 years of the cohort, patients requiring biologics tended to be treated with rituximab, a B-cell inhibitor, rather than anti–tumor necrosis factor (anti-TNF) agents, the BRILL investigators found.

Mortality was lower among RA-ILD patients treated with mycophenolate than in those treated with other immunosuppressive agents. Among biologic agents used in the cohort, rituximab treatment was associated with improved mortality, but anti-TNF inhibitors were seen to be associated with elevated risk of death.

About 95% of RA-ILD patients are anticyclic citrullinated peptide (anti-CCP) antibody positive, compared with 55%-60% of the RA population as a whole, Dr. Kelly said, "so there’s a strong statistical association of seropositivity, and in those who are seropositive, rituximab works well."

The finding that rituximab was associated with improved survival in the cohort not only has implications for RA-ILD, he said, but also, potentially, for people with idiopathic pulmonary fibrosis (IPF) who are anti-CCP antibody positive. "What [rheumatologists] have, and chest physicians traditionally don’t, is access to rituximab and mycophenolate. But these might be worth trying in IPF as well," he said.

Dr. Kelly noted that prospective trials in RA-ILD are beginning to enroll patients with progressive disease to compare azathioprine and mycophenolate, allowing for the use of oral steroids, as well as patients with active RA and ILD to compare anti-TNF inhibitors against rituximab, also allowing oral steroids.

Dr. Kelly reported that he had no conflicts of interest related to his findings and that none of his fellow BRILL investigators had conflicts.