Bloodstream Infections Spiked With Loss of Short-term Glucose Control

CHICAGO – Hospitalized patients who maintained short-term glucose levels less than 110mg/dL were less likely than those with higher short-term levels to develop bloodstream infections, according to the findings of a retrospective study presented at the annual Interscience Conference on Antimicrobial Agents and Chemotherapy.

By contrast, long-term glucose measures – hemoglobin A_1c – did not appear to be associated with the rates of bloodstream infections, said the study’s lead investigator Christie Y. Jeon, Sc.D., of Columbia University, New York.

"Given that people with diabetes are at increased risk of infections, monitoring glucose control may reduce the burden of bloodstream infections," she said at the conference, sponsored by the American Society for Microbiology.

There is currently a dearth of information on how short-term and long-term glucose control impact health care–associated bloodstream infections and what level of glucose control is optimal for reducing the risk of infection. Yet "health care–associated bloodstream infections are estimated to cause over 30,000 deaths in the United States annually," she said in an interview.

In the study, Dr. Jeon and associates analyzed 18,736 records from diabetes patients and 60,778 records from patients without diabetes admitted to one of three hospitals in New York City to examine the relationship between both short- and long-term glucose control on bloodstream infection rates.

For short-term glucose control, the researchers assessed glucose levels during three different periods leading up to the index date of infection: period 1, the index date and 2 days prior (day 0 to day -2); period 2, day -1 to day -3; and period 3, day -2 to day -4. Then they classified them into six categories: less than 80 mg/dL, 80-109 mg/dL, 110-139 mg/dL, 140-179 mg/dL, 180-199 mg/dL, and 200 mg/dL or higher.

Long-term glucose control was determined by hemoglobin A_1c.

The researchers controlled for confounding variables, including age, gender, severity of illness, malignancy, use of immunosuppressive medications, renal failure, prior hospitalization, use of mechanical ventilation, feeding tubes, and central venous and urinary catheterizations. The median age of patients was 53 years, and 54% were female.

Short-term glucose levels of 110 mg/dL or higher during period 1 were associated with an increase in bloodstream infections in people with diabetes (hazard ratio from 2.18 to 3.32) and in those without diabetes (HR from 1.62 to 1.80). Glucose levels measured more remotely in period 3 were not associated with bloodstream infections. Hypoglycemia and hemoglobin A_1c levels were not associated with bloodstream infections.

"The fact that long-term glucose control was not associated with bloodstream infection was surprising and informative to us, as it is an indication that for acute infections such as bloodstream infections, long-term glucose control is limited in its impact," Dr. Jeon said. "Further, the fact that glucose measured closest to the index date of infection was most predictive of bloodstream infections suggests that associations between glucose levels and infections may be due to hyperglycemia as an effect, and not a cause, of health care–associated infections."

The study was limited by its observational design. In addition, "there is a possibility that ... confounders may have influenced glucose levels as well as infection risk."

The study was funded by the National Institutes of Health. Dr. Jeon was supported by a National Research Service Award from the National Institute of Allergy and Infectious Diseases. Dr. Jeon’s coathors on the study were Dr. Emily Y. Furuya, Mandar Apte, and Elaine L. Larson, Ph.D.

Dr. Jeon said that she had no relevant financial conflicts of interest.

CHICAGO – Hospitalized patients who maintained short-term glucose levels less than 110mg/dL were less likely than those with higher short-term levels to develop bloodstream infections, according to the findings of a retrospective study presented at the annual Interscience Conference on Antimicrobial Agents and Chemotherapy.

By contrast, long-term glucose measures – hemoglobin A_1c – did not appear to be associated with the rates of bloodstream infections, said the study’s lead investigator Christie Y. Jeon, Sc.D., of Columbia University, New York.

"Given that people with diabetes are at increased risk of infections, monitoring glucose control may reduce the burden of bloodstream infections," she said at the conference, sponsored by the American Society for Microbiology.

There is currently a dearth of information on how short-term and long-term glucose control impact health care–associated bloodstream infections and what level of glucose control is optimal for reducing the risk of infection. Yet "health care–associated bloodstream infections are estimated to cause over 30,000 deaths in the United States annually," she said in an interview.

In the study, Dr. Jeon and associates analyzed 18,736 records from diabetes patients and 60,778 records from patients without diabetes admitted to one of three hospitals in New York City to examine the relationship between both short- and long-term glucose control on bloodstream infection rates.

For short-term glucose control, the researchers assessed glucose levels during three different periods leading up to the index date of infection: period 1, the index date and 2 days prior (day 0 to day -2); period 2, day -1 to day -3; and period 3, day -2 to day -4. Then they classified them into six categories: less than 80 mg/dL, 80-109 mg/dL, 110-139 mg/dL, 140-179 mg/dL, 180-199 mg/dL, and 200 mg/dL or higher.

Long-term glucose control was determined by hemoglobin A_1c.

The researchers controlled for confounding variables, including age, gender, severity of illness, malignancy, use of immunosuppressive medications, renal failure, prior hospitalization, use of mechanical ventilation, feeding tubes, and central venous and urinary catheterizations. The median age of patients was 53 years, and 54% were female.

Short-term glucose levels of 110 mg/dL or higher during period 1 were associated with an increase in bloodstream infections in people with diabetes (hazard ratio from 2.18 to 3.32) and in those without diabetes (HR from 1.62 to 1.80). Glucose levels measured more remotely in period 3 were not associated with bloodstream infections. Hypoglycemia and hemoglobin A_1c levels were not associated with bloodstream infections.

"The fact that long-term glucose control was not associated with bloodstream infection was surprising and informative to us, as it is an indication that for acute infections such as bloodstream infections, long-term glucose control is limited in its impact," Dr. Jeon said. "Further, the fact that glucose measured closest to the index date of infection was most predictive of bloodstream infections suggests that associations between glucose levels and infections may be due to hyperglycemia as an effect, and not a cause, of health care–associated infections."

The study was limited by its observational design. In addition, "there is a possibility that ... confounders may have influenced glucose levels as well as infection risk."

The study was funded by the National Institutes of Health. Dr. Jeon was supported by a National Research Service Award from the National Institute of Allergy and Infectious Diseases. Dr. Jeon’s coathors on the study were Dr. Emily Y. Furuya, Mandar Apte, and Elaine L. Larson, Ph.D.

Dr. Jeon said that she had no relevant financial conflicts of interest.

CHICAGO – Hospitalized patients who maintained short-term glucose levels less than 110mg/dL were less likely than those with higher short-term levels to develop bloodstream infections, according to the findings of a retrospective study presented at the annual Interscience Conference on Antimicrobial Agents and Chemotherapy.

By contrast, long-term glucose measures – hemoglobin A_1c – did not appear to be associated with the rates of bloodstream infections, said the study’s lead investigator Christie Y. Jeon, Sc.D., of Columbia University, New York.

"Given that people with diabetes are at increased risk of infections, monitoring glucose control may reduce the burden of bloodstream infections," she said at the conference, sponsored by the American Society for Microbiology.

There is currently a dearth of information on how short-term and long-term glucose control impact health care–associated bloodstream infections and what level of glucose control is optimal for reducing the risk of infection. Yet "health care–associated bloodstream infections are estimated to cause over 30,000 deaths in the United States annually," she said in an interview.

In the study, Dr. Jeon and associates analyzed 18,736 records from diabetes patients and 60,778 records from patients without diabetes admitted to one of three hospitals in New York City to examine the relationship between both short- and long-term glucose control on bloodstream infection rates.

For short-term glucose control, the researchers assessed glucose levels during three different periods leading up to the index date of infection: period 1, the index date and 2 days prior (day 0 to day -2); period 2, day -1 to day -3; and period 3, day -2 to day -4. Then they classified them into six categories: less than 80 mg/dL, 80-109 mg/dL, 110-139 mg/dL, 140-179 mg/dL, 180-199 mg/dL, and 200 mg/dL or higher.

Long-term glucose control was determined by hemoglobin A_1c.

The researchers controlled for confounding variables, including age, gender, severity of illness, malignancy, use of immunosuppressive medications, renal failure, prior hospitalization, use of mechanical ventilation, feeding tubes, and central venous and urinary catheterizations. The median age of patients was 53 years, and 54% were female.

Short-term glucose levels of 110 mg/dL or higher during period 1 were associated with an increase in bloodstream infections in people with diabetes (hazard ratio from 2.18 to 3.32) and in those without diabetes (HR from 1.62 to 1.80). Glucose levels measured more remotely in period 3 were not associated with bloodstream infections. Hypoglycemia and hemoglobin A_1c levels were not associated with bloodstream infections.

"The fact that long-term glucose control was not associated with bloodstream infection was surprising and informative to us, as it is an indication that for acute infections such as bloodstream infections, long-term glucose control is limited in its impact," Dr. Jeon said. "Further, the fact that glucose measured closest to the index date of infection was most predictive of bloodstream infections suggests that associations between glucose levels and infections may be due to hyperglycemia as an effect, and not a cause, of health care–associated infections."

The study was limited by its observational design. In addition, "there is a possibility that ... confounders may have influenced glucose levels as well as infection risk."

The study was funded by the National Institutes of Health. Dr. Jeon was supported by a National Research Service Award from the National Institute of Allergy and Infectious Diseases. Dr. Jeon’s coathors on the study were Dr. Emily Y. Furuya, Mandar Apte, and Elaine L. Larson, Ph.D.

Dr. Jeon said that she had no relevant financial conflicts of interest.