Delayed Mortality Seen After Pneumonia : Pneumonia patients discharged with high blood cytokine levels showed greater 90-day mortality.

SAN DIEGO — Community-acquired pneumonia patients are significantly more likely to die within 3 months of leaving the hospital if they have high concentrations of inflammatory cytokines before discharge, Dr. Sachin Yende reported at the international conference of the American Thoracic Society.

Above-normal levels of interleukin-6 (IL-6) and interleukin-10 (IL-10) were strongly associated with 90-day mortality in a large, ongoing study presented by Dr. Yende of the University of Pittsburgh Medical Center.

Predischarge IL-6 concentrations were nearly twice as high in discharged patients who died within 90 days: 10.6 pg/mL vs. 5.9 pg/mL in survivors. IL-10 levels were three times as high in the patients who died: 3.1 pg/mL vs. 1 pg/mL.

Concentrations of both cytokines were slightly higher in patients who had had sepsis while hospitalized but not significantly higher than the levels measured near discharge in those who had not had sepsis.

“People are sent home with fairly high cytokine concentrations,” Dr. Yende said in an interview.

“If physicians were aware, they would not have wanted to send these patients home.” The physicians were not aware, he said, because cytokine levels are not routinely ordered and the study was blinded.

The investigation follows an earlier finding that recovered community-acquired pneumonia patients have higher mortality 5 years after hospitalization than would be predicted by their age, gender, or other variables. “The increased mortality we have seen is not because these people are older or have more chronic health conditions. It is because of this hospitalization,” Dr. Yende said, adding that he and his colleagues were trying to find out why.

The cohort study collected blood daily for 1 week and once weekly subsequently, starting with 2,320 patients who presented with pneumonia in the emergency departments at 28 hospitals. Of 1,895 patients who were admitted, 87 (5%) died in the hospital.

Among the remaining patients, 1,799 were alive at discharge and included in the cohort followed by investigators, but 126 (7%) of these patients died within 90 days. Dr. Yende reported 1,636 were still alive, and outcomes were not known for 37 (2%) of the discharged patients.

The IL-6 and IL-10 data were based on 1,452 patients in whom concentrations were measured 48 hours before discharge. Overall, the median IL-6 and IL-10 levels as measured between the 10th and 90th percentiles were 6.2 pg/mL and 5 pg/mL, respectively. In comparison, Dr. Yende said, median IL-6 levels typically range from 1.8 pg/mL in healthy persons to 98.7 pg/mL in patients with severe sepsis.

While Dr. Yende emphasized that more work needs to be done before the investigators can establish causality, he speculated that an acute illness such as pneumonia might be a “feed-forward phenomenon” that leads to increased incidence of pneumonia or cardiovascular events.

“An acute event like pneumonia or sepsis may alter subsequent life trajectory even after hospital discharge,” he said.

The investigators obtained 1-year data on the discharged pneumonia patients shortly before Dr. Yende's presentation. Though they had not had time for a full analysis, he reported mortality had reached 16%. The five leading causes of death at 1 year were chronic ischemic heart disease, malignant neoplasms of the respiratory system, pneumonia, cerebrovascular disease, and acute myocardial infarction.

“These are the same diseases that have been shown in cardiovascular literature to be associated with increased risk when you have elevated C-reactive protein and other inflammatory markers,” Dr. Yende said.

At this point, he said, a recommendation for routine measurement of inflammatory markers would be premature. In the future, he predicted, this could become standard along with interventions and a faster follow-up visit in the physician's office.

“Normally you say [to] come back after 6 weeks,” he said. “Maybe there is a role to call them back earlier and remeasure and see how things go—but right now, there isn't evidence to support that.”

Asked whether pressure to send pneumonia patients home faster could be a factor, Dr. Yende said the vast majority of discharged patients are ready to go home. “They are on the road to recovery,” he said. “And that is the interesting point. Even if you think they are ready to go home, perhaps there is something going on that may be completely bad.”

Predischarge IL-6 concentrations were nearly twice as high in discharged patients who died within 90 days. DR. YENDE

SAN DIEGO — Community-acquired pneumonia patients are significantly more likely to die within 3 months of leaving the hospital if they have high concentrations of inflammatory cytokines before discharge, Dr. Sachin Yende reported at the international conference of the American Thoracic Society.

Above-normal levels of interleukin-6 (IL-6) and interleukin-10 (IL-10) were strongly associated with 90-day mortality in a large, ongoing study presented by Dr. Yende of the University of Pittsburgh Medical Center.

Predischarge IL-6 concentrations were nearly twice as high in discharged patients who died within 90 days: 10.6 pg/mL vs. 5.9 pg/mL in survivors. IL-10 levels were three times as high in the patients who died: 3.1 pg/mL vs. 1 pg/mL.

Concentrations of both cytokines were slightly higher in patients who had had sepsis while hospitalized but not significantly higher than the levels measured near discharge in those who had not had sepsis.

“People are sent home with fairly high cytokine concentrations,” Dr. Yende said in an interview.

“If physicians were aware, they would not have wanted to send these patients home.” The physicians were not aware, he said, because cytokine levels are not routinely ordered and the study was blinded.

The investigation follows an earlier finding that recovered community-acquired pneumonia patients have higher mortality 5 years after hospitalization than would be predicted by their age, gender, or other variables. “The increased mortality we have seen is not because these people are older or have more chronic health conditions. It is because of this hospitalization,” Dr. Yende said, adding that he and his colleagues were trying to find out why.

The cohort study collected blood daily for 1 week and once weekly subsequently, starting with 2,320 patients who presented with pneumonia in the emergency departments at 28 hospitals. Of 1,895 patients who were admitted, 87 (5%) died in the hospital.

Among the remaining patients, 1,799 were alive at discharge and included in the cohort followed by investigators, but 126 (7%) of these patients died within 90 days. Dr. Yende reported 1,636 were still alive, and outcomes were not known for 37 (2%) of the discharged patients.

The IL-6 and IL-10 data were based on 1,452 patients in whom concentrations were measured 48 hours before discharge. Overall, the median IL-6 and IL-10 levels as measured between the 10th and 90th percentiles were 6.2 pg/mL and 5 pg/mL, respectively. In comparison, Dr. Yende said, median IL-6 levels typically range from 1.8 pg/mL in healthy persons to 98.7 pg/mL in patients with severe sepsis.

While Dr. Yende emphasized that more work needs to be done before the investigators can establish causality, he speculated that an acute illness such as pneumonia might be a “feed-forward phenomenon” that leads to increased incidence of pneumonia or cardiovascular events.

“An acute event like pneumonia or sepsis may alter subsequent life trajectory even after hospital discharge,” he said.

The investigators obtained 1-year data on the discharged pneumonia patients shortly before Dr. Yende's presentation. Though they had not had time for a full analysis, he reported mortality had reached 16%. The five leading causes of death at 1 year were chronic ischemic heart disease, malignant neoplasms of the respiratory system, pneumonia, cerebrovascular disease, and acute myocardial infarction.

“These are the same diseases that have been shown in cardiovascular literature to be associated with increased risk when you have elevated C-reactive protein and other inflammatory markers,” Dr. Yende said.

At this point, he said, a recommendation for routine measurement of inflammatory markers would be premature. In the future, he predicted, this could become standard along with interventions and a faster follow-up visit in the physician's office.

“Normally you say [to] come back after 6 weeks,” he said. “Maybe there is a role to call them back earlier and remeasure and see how things go—but right now, there isn't evidence to support that.”

Asked whether pressure to send pneumonia patients home faster could be a factor, Dr. Yende said the vast majority of discharged patients are ready to go home. “They are on the road to recovery,” he said. “And that is the interesting point. Even if you think they are ready to go home, perhaps there is something going on that may be completely bad.”

Predischarge IL-6 concentrations were nearly twice as high in discharged patients who died within 90 days. DR. YENDE

SAN DIEGO — Community-acquired pneumonia patients are significantly more likely to die within 3 months of leaving the hospital if they have high concentrations of inflammatory cytokines before discharge, Dr. Sachin Yende reported at the international conference of the American Thoracic Society.

Above-normal levels of interleukin-6 (IL-6) and interleukin-10 (IL-10) were strongly associated with 90-day mortality in a large, ongoing study presented by Dr. Yende of the University of Pittsburgh Medical Center.

Predischarge IL-6 concentrations were nearly twice as high in discharged patients who died within 90 days: 10.6 pg/mL vs. 5.9 pg/mL in survivors. IL-10 levels were three times as high in the patients who died: 3.1 pg/mL vs. 1 pg/mL.

Concentrations of both cytokines were slightly higher in patients who had had sepsis while hospitalized but not significantly higher than the levels measured near discharge in those who had not had sepsis.

“People are sent home with fairly high cytokine concentrations,” Dr. Yende said in an interview.

“If physicians were aware, they would not have wanted to send these patients home.” The physicians were not aware, he said, because cytokine levels are not routinely ordered and the study was blinded.

The investigation follows an earlier finding that recovered community-acquired pneumonia patients have higher mortality 5 years after hospitalization than would be predicted by their age, gender, or other variables. “The increased mortality we have seen is not because these people are older or have more chronic health conditions. It is because of this hospitalization,” Dr. Yende said, adding that he and his colleagues were trying to find out why.

The cohort study collected blood daily for 1 week and once weekly subsequently, starting with 2,320 patients who presented with pneumonia in the emergency departments at 28 hospitals. Of 1,895 patients who were admitted, 87 (5%) died in the hospital.

Among the remaining patients, 1,799 were alive at discharge and included in the cohort followed by investigators, but 126 (7%) of these patients died within 90 days. Dr. Yende reported 1,636 were still alive, and outcomes were not known for 37 (2%) of the discharged patients.

The IL-6 and IL-10 data were based on 1,452 patients in whom concentrations were measured 48 hours before discharge. Overall, the median IL-6 and IL-10 levels as measured between the 10th and 90th percentiles were 6.2 pg/mL and 5 pg/mL, respectively. In comparison, Dr. Yende said, median IL-6 levels typically range from 1.8 pg/mL in healthy persons to 98.7 pg/mL in patients with severe sepsis.

While Dr. Yende emphasized that more work needs to be done before the investigators can establish causality, he speculated that an acute illness such as pneumonia might be a “feed-forward phenomenon” that leads to increased incidence of pneumonia or cardiovascular events.

“An acute event like pneumonia or sepsis may alter subsequent life trajectory even after hospital discharge,” he said.

The investigators obtained 1-year data on the discharged pneumonia patients shortly before Dr. Yende's presentation. Though they had not had time for a full analysis, he reported mortality had reached 16%. The five leading causes of death at 1 year were chronic ischemic heart disease, malignant neoplasms of the respiratory system, pneumonia, cerebrovascular disease, and acute myocardial infarction.

“These are the same diseases that have been shown in cardiovascular literature to be associated with increased risk when you have elevated C-reactive protein and other inflammatory markers,” Dr. Yende said.

At this point, he said, a recommendation for routine measurement of inflammatory markers would be premature. In the future, he predicted, this could become standard along with interventions and a faster follow-up visit in the physician's office.

“Normally you say [to] come back after 6 weeks,” he said. “Maybe there is a role to call them back earlier and remeasure and see how things go—but right now, there isn't evidence to support that.”

Asked whether pressure to send pneumonia patients home faster could be a factor, Dr. Yende said the vast majority of discharged patients are ready to go home. “They are on the road to recovery,” he said. “And that is the interesting point. Even if you think they are ready to go home, perhaps there is something going on that may be completely bad.”

Predischarge IL-6 concentrations were nearly twice as high in discharged patients who died within 90 days. DR. YENDE