Miriam E. Tucker

BETHESDA, Md. — The overall cost burden of antimicrobial resistance--as high as $38 billion in one 2009 hospital estimate--has shifted dramatically from Medicare to private payers over the last decade.

Medicare still pays the majority of the costs for excess length of stay, increased use of more expensive drugs, and poorer health attributable to treatment-resistant infections. However, the rise in infections caused by methicillin-resistant Staphylococcus aureus (MRSA), which largely affects younger, healthier individuals, has meant that the overall cost per patient has declined but more is being borne by private HMOs and PPOs, Susan D. Foster, Ph.D., said at the 2010 Conference on Antimicrobial Resistance sponsored by the National Foundation for Infectious Diseases.

"There's been a major shift in who's actually paying. I don't think the insurance companies are quite aware of this," said Dr. Foster, professor of international health at Boston University and director of public policy and education for the Alliance for the Prudent Use of Antibiotics, based at Tufts University, Boston.

Supported in part by an unrestricted educational grant from bioMérieux Inc., Dr. Foster analyzed data from three studies. The first study reviewed Massachusetts hospital discharge data from 2000 to 2007 to look for ICD9 "VO9" codes, which are specific for drug-resistant infections. Although these codes are complex and difficult to use and therefore represent a study limitation, they do allow for analysis of trends over time, she explained.

Overall, the number of hospital discharges reporting antibiotic resistance in Massachusetts increased from 3,861 in 2000 to 11,218 in 2007. The inflation-adjusted total cost more than doubled over the 7 years, from $135 to $285 million. However, the length of stay (LOS) per patient for drug-resistant infections dropped by 4.5 days, and the cost per patient fell by nearly $10,000.

In contrast, the length of stay for drug-susceptible infections didn't change during the study period (just under 5 days), while the cost per patient with susceptible infections rose only slightly. 

The drop in LOS and cost per patient with drug-resistant infections is largely explained by the dramatic shift in patient age, particularly among 19- to 64-year-olds: In 2000, that age group accounted for 30% of drug-resistant infection discharges, whereas in 2007 the proportion had risen to 45.5%. At the same time, the 65- to 80-year-old group dropped from 38% to 25%. While the proportion of infections due to drug-resistant organisms rose in all age groups, the greatest rise was among working-age adults, Dr. Foster noted.

Not surprisingly, then, was the concurrent payer shift: Medicare's proportion of the cost dropped from 73% in 2000 to 58% in 2007, while Medicaid's rose from 6% to 15%. The proportion paid by "Other," including private insurance, rose from 20.5% to 28%, she said.

Also not surprising--though perhaps not considered previously--were the declines in inpatient mortality due to drug-resistant infections (from 11% to 5%) and in patients being discharged to nursing homes (32% to 28%), with a concomitant increase in patients returning home from the hospital (33% to 48%).

The second study, from the Chicago Antimicrobial Resistance Project, analyzed discharge data at Cook County Hospital for a random sample of 1,391 high-risk (more than five ICD9 codes, excluding trauma, burn, or obstetric care) adult patients, of whom 13.5% (188) had an antibiotic-resistant infection (ARI). Societal costs for the study year 2000 were estimated at $10.7-$15 million, which extrapolated to $13.35 million in 2008 dollars (Clin. Infect. Dis. 2009;49:1175-84).

In that study, conducted by Dr. Rebecca Roberts and her associates, LOS was three times longer for the patients with ARIs (24 vs. 8 days) and mortality 6 times higher (18% vs. 3%). Total inpatient costs were $58,029, compared with $13,210 for non-ARI patients. Even the daily cost was $517 greater for the ARI group, Dr. Foster noted.

The most common type of ARI was MRSA (43%), followed by vancomycin-resistant enterococci (VRE, 31%), Escherichia coli/Klebsiella species (16%), and multiple infections (6%). By cost, however, VRE accounted for the greatest proportion (36%), followed by MRSA (34%), and multiple infections (16%).

Dr. Foster and Dr. Roberts extrapolated the Chicago data to the entire United States: In 2000, there were 900,000 admissions with the same criteria the study used. Applying the costs found in that study gives $16.6-$26 billion in additional health care costs (the range reflects different types of inflation adjustments). Updating the figure to 2009 costs using the Consumer Price Index gives an estimated $21-$34 billion, while using medical inflation rates boosts the figures to as high as $24-$38 billion, "a substantial burden," she commented.

The third study, also from Dr. Foster's group, was an Internet-based survey of more than 300 respondents recruited from MRSA chatrooms, listservs, and Google Adwords. Acknowledging the limitations of such surveys--particularly the bias toward those who are younger, healthier, and more likely to Internet access as well as to have strong opinions--she noted that there were "some heart-rending responses," including one from a 52-year-old woman who felt completely isolated from friends and family, a teacher who was fired when her MRSA diagnosis became known, and parents who felt they had to send their children away to prevent transmission.

Respondents reported a mean out-of-pocket expenditure of $2,251, including co-pays for office visits, prescription drugs, and hospital stays. Nearly 70% reported having private insurance (HMO or PPO), and 14% said they were uninsured, which approximately reflects the national average, Dr. Foster noted.
"Individuals and households affected by drug resistance bear a large uncompensated burden in terms of out-of-pocket expenses and lost wages," she concluded.

Dr. Foster declared that she had no relevant financial relationships other than bioMérieux's funding of her data analysis.

BETHESDA, Md. — The overall cost burden of antimicrobial resistance--as high as $38 billion in one 2009 hospital estimate--has shifted dramatically from Medicare to private payers over the last decade.

Medicare still pays the majority of the costs for excess length of stay, increased use of more expensive drugs, and poorer health attributable to treatment-resistant infections. However, the rise in infections caused by methicillin-resistant Staphylococcus aureus (MRSA), which largely affects younger, healthier individuals, has meant that the overall cost per patient has declined but more is being borne by private HMOs and PPOs, Susan D. Foster, Ph.D., said at the 2010 Conference on Antimicrobial Resistance sponsored by the National Foundation for Infectious Diseases.

"There's been a major shift in who's actually paying. I don't think the insurance companies are quite aware of this," said Dr. Foster, professor of international health at Boston University and director of public policy and education for the Alliance for the Prudent Use of Antibiotics, based at Tufts University, Boston.

Supported in part by an unrestricted educational grant from bioMérieux Inc., Dr. Foster analyzed data from three studies. The first study reviewed Massachusetts hospital discharge data from 2000 to 2007 to look for ICD9 "VO9" codes, which are specific for drug-resistant infections. Although these codes are complex and difficult to use and therefore represent a study limitation, they do allow for analysis of trends over time, she explained.

Overall, the number of hospital discharges reporting antibiotic resistance in Massachusetts increased from 3,861 in 2000 to 11,218 in 2007. The inflation-adjusted total cost more than doubled over the 7 years, from $135 to $285 million. However, the length of stay (LOS) per patient for drug-resistant infections dropped by 4.5 days, and the cost per patient fell by nearly $10,000.

In contrast, the length of stay for drug-susceptible infections didn't change during the study period (just under 5 days), while the cost per patient with susceptible infections rose only slightly. 

The drop in LOS and cost per patient with drug-resistant infections is largely explained by the dramatic shift in patient age, particularly among 19- to 64-year-olds: In 2000, that age group accounted for 30% of drug-resistant infection discharges, whereas in 2007 the proportion had risen to 45.5%. At the same time, the 65- to 80-year-old group dropped from 38% to 25%. While the proportion of infections due to drug-resistant organisms rose in all age groups, the greatest rise was among working-age adults, Dr. Foster noted.

Not surprisingly, then, was the concurrent payer shift: Medicare's proportion of the cost dropped from 73% in 2000 to 58% in 2007, while Medicaid's rose from 6% to 15%. The proportion paid by "Other," including private insurance, rose from 20.5% to 28%, she said.

Also not surprising--though perhaps not considered previously--were the declines in inpatient mortality due to drug-resistant infections (from 11% to 5%) and in patients being discharged to nursing homes (32% to 28%), with a concomitant increase in patients returning home from the hospital (33% to 48%).

The second study, from the Chicago Antimicrobial Resistance Project, analyzed discharge data at Cook County Hospital for a random sample of 1,391 high-risk (more than five ICD9 codes, excluding trauma, burn, or obstetric care) adult patients, of whom 13.5% (188) had an antibiotic-resistant infection (ARI). Societal costs for the study year 2000 were estimated at $10.7-$15 million, which extrapolated to $13.35 million in 2008 dollars (Clin. Infect. Dis. 2009;49:1175-84).

In that study, conducted by Dr. Rebecca Roberts and her associates, LOS was three times longer for the patients with ARIs (24 vs. 8 days) and mortality 6 times higher (18% vs. 3%). Total inpatient costs were $58,029, compared with $13,210 for non-ARI patients. Even the daily cost was $517 greater for the ARI group, Dr. Foster noted.

The most common type of ARI was MRSA (43%), followed by vancomycin-resistant enterococci (VRE, 31%), Escherichia coli/Klebsiella species (16%), and multiple infections (6%). By cost, however, VRE accounted for the greatest proportion (36%), followed by MRSA (34%), and multiple infections (16%).

Dr. Foster and Dr. Roberts extrapolated the Chicago data to the entire United States: In 2000, there were 900,000 admissions with the same criteria the study used. Applying the costs found in that study gives $16.6-$26 billion in additional health care costs (the range reflects different types of inflation adjustments). Updating the figure to 2009 costs using the Consumer Price Index gives an estimated $21-$34 billion, while using medical inflation rates boosts the figures to as high as $24-$38 billion, "a substantial burden," she commented.

The third study, also from Dr. Foster's group, was an Internet-based survey of more than 300 respondents recruited from MRSA chatrooms, listservs, and Google Adwords. Acknowledging the limitations of such surveys--particularly the bias toward those who are younger, healthier, and more likely to Internet access as well as to have strong opinions--she noted that there were "some heart-rending responses," including one from a 52-year-old woman who felt completely isolated from friends and family, a teacher who was fired when her MRSA diagnosis became known, and parents who felt they had to send their children away to prevent transmission.

Respondents reported a mean out-of-pocket expenditure of $2,251, including co-pays for office visits, prescription drugs, and hospital stays. Nearly 70% reported having private insurance (HMO or PPO), and 14% said they were uninsured, which approximately reflects the national average, Dr. Foster noted.
"Individuals and households affected by drug resistance bear a large uncompensated burden in terms of out-of-pocket expenses and lost wages," she concluded.

Dr. Foster declared that she had no relevant financial relationships other than bioMérieux's funding of her data analysis.

BETHESDA, Md. — The overall cost burden of antimicrobial resistance--as high as $38 billion in one 2009 hospital estimate--has shifted dramatically from Medicare to private payers over the last decade.

Medicare still pays the majority of the costs for excess length of stay, increased use of more expensive drugs, and poorer health attributable to treatment-resistant infections. However, the rise in infections caused by methicillin-resistant Staphylococcus aureus (MRSA), which largely affects younger, healthier individuals, has meant that the overall cost per patient has declined but more is being borne by private HMOs and PPOs, Susan D. Foster, Ph.D., said at the 2010 Conference on Antimicrobial Resistance sponsored by the National Foundation for Infectious Diseases.

"There's been a major shift in who's actually paying. I don't think the insurance companies are quite aware of this," said Dr. Foster, professor of international health at Boston University and director of public policy and education for the Alliance for the Prudent Use of Antibiotics, based at Tufts University, Boston.

Supported in part by an unrestricted educational grant from bioMérieux Inc., Dr. Foster analyzed data from three studies. The first study reviewed Massachusetts hospital discharge data from 2000 to 2007 to look for ICD9 "VO9" codes, which are specific for drug-resistant infections. Although these codes are complex and difficult to use and therefore represent a study limitation, they do allow for analysis of trends over time, she explained.

Overall, the number of hospital discharges reporting antibiotic resistance in Massachusetts increased from 3,861 in 2000 to 11,218 in 2007. The inflation-adjusted total cost more than doubled over the 7 years, from $135 to $285 million. However, the length of stay (LOS) per patient for drug-resistant infections dropped by 4.5 days, and the cost per patient fell by nearly $10,000.

In contrast, the length of stay for drug-susceptible infections didn't change during the study period (just under 5 days), while the cost per patient with susceptible infections rose only slightly. 

The drop in LOS and cost per patient with drug-resistant infections is largely explained by the dramatic shift in patient age, particularly among 19- to 64-year-olds: In 2000, that age group accounted for 30% of drug-resistant infection discharges, whereas in 2007 the proportion had risen to 45.5%. At the same time, the 65- to 80-year-old group dropped from 38% to 25%. While the proportion of infections due to drug-resistant organisms rose in all age groups, the greatest rise was among working-age adults, Dr. Foster noted.

Not surprisingly, then, was the concurrent payer shift: Medicare's proportion of the cost dropped from 73% in 2000 to 58% in 2007, while Medicaid's rose from 6% to 15%. The proportion paid by "Other," including private insurance, rose from 20.5% to 28%, she said.

Also not surprising--though perhaps not considered previously--were the declines in inpatient mortality due to drug-resistant infections (from 11% to 5%) and in patients being discharged to nursing homes (32% to 28%), with a concomitant increase in patients returning home from the hospital (33% to 48%).

The second study, from the Chicago Antimicrobial Resistance Project, analyzed discharge data at Cook County Hospital for a random sample of 1,391 high-risk (more than five ICD9 codes, excluding trauma, burn, or obstetric care) adult patients, of whom 13.5% (188) had an antibiotic-resistant infection (ARI). Societal costs for the study year 2000 were estimated at $10.7-$15 million, which extrapolated to $13.35 million in 2008 dollars (Clin. Infect. Dis. 2009;49:1175-84).

In that study, conducted by Dr. Rebecca Roberts and her associates, LOS was three times longer for the patients with ARIs (24 vs. 8 days) and mortality 6 times higher (18% vs. 3%). Total inpatient costs were $58,029, compared with $13,210 for non-ARI patients. Even the daily cost was $517 greater for the ARI group, Dr. Foster noted.

The most common type of ARI was MRSA (43%), followed by vancomycin-resistant enterococci (VRE, 31%), Escherichia coli/Klebsiella species (16%), and multiple infections (6%). By cost, however, VRE accounted for the greatest proportion (36%), followed by MRSA (34%), and multiple infections (16%).

Dr. Foster and Dr. Roberts extrapolated the Chicago data to the entire United States: In 2000, there were 900,000 admissions with the same criteria the study used. Applying the costs found in that study gives $16.6-$26 billion in additional health care costs (the range reflects different types of inflation adjustments). Updating the figure to 2009 costs using the Consumer Price Index gives an estimated $21-$34 billion, while using medical inflation rates boosts the figures to as high as $24-$38 billion, "a substantial burden," she commented.

The third study, also from Dr. Foster's group, was an Internet-based survey of more than 300 respondents recruited from MRSA chatrooms, listservs, and Google Adwords. Acknowledging the limitations of such surveys--particularly the bias toward those who are younger, healthier, and more likely to Internet access as well as to have strong opinions--she noted that there were "some heart-rending responses," including one from a 52-year-old woman who felt completely isolated from friends and family, a teacher who was fired when her MRSA diagnosis became known, and parents who felt they had to send their children away to prevent transmission.

Respondents reported a mean out-of-pocket expenditure of $2,251, including co-pays for office visits, prescription drugs, and hospital stays. Nearly 70% reported having private insurance (HMO or PPO), and 14% said they were uninsured, which approximately reflects the national average, Dr. Foster noted.
"Individuals and households affected by drug resistance bear a large uncompensated burden in terms of out-of-pocket expenses and lost wages," she concluded.

Dr. Foster declared that she had no relevant financial relationships other than bioMérieux's funding of her data analysis.

Major Finding: HbA_1c values below 7.5% were associated with increased all-cause mortality and cardiovascular events.

Data Source: A general practice database analysis of 47,970 type 2 diabetes patients with recently intensified glucose-lowering therapy.

Disclosures: Funding by Eli Lilly & Co. The investigators disclosed ties to numerous manufacturers of diabetes treatments, including Eli Lilly.

Hemoglobin A_1c values below 7.5% were associated with increased all-cause mortality and cardiovascular events in patients with type 2 diabetes in an analysis of nearly 48,000 patients in a U.K. general practice database.

If confirmed, the findings suggest that diabetes guidelines might need revision to include a definition of a minimum HbA_1c value, Dr. Craig J. Currie of Cardiff (Wales) University and his associates said (Lancet 2010 Jan. 27 [doi:10.1016/S0140-6736(09)61969-3

The study also showed that insulin therapy was associated with higher mortality than combination oral therapy. Unadjusted mortality rates were 16.2 deaths per 1,000 person-years of follow-up for the oral combination therapy group and 27.2/1,000 for the insulin group. After exclusion of patients with high cardiovascular risk or renal impairment, insulin-based therapy remained associated with significantly greater all-cause mortality (hazard ratio 1.46) than did combination oral agents, the investigators reported.

The study used data from November 1986 to November 2008 in the U.K. General Practice Research Database. Two cohorts of patients aged 50 years and older with type 2 diabetes were assessed: 27,965 whose treatment regimen had been intensified from oral glucose-lowering monotherapy to a combination oral regimen with a sulfonylurea plus metformin, and 20,005 on oral hypoglycemic agents alone who were initiated on insulin with or without concomitant oral agents.

All-cause mortality was the primary outcome. The secondary outcome was the occurrence of a major cardiovascular event among those who had no record of cardiovascular disease before the index date. The mean follow-up was 4.5 years in the oral medication group and 5.2 years in the insulin group.

Patients were divided by HbA_1c decile, with the lowest decile (1) having a median HbA_1c of 6.4% and the highest decile (10) at 10.5%. Mortality varied by HbA_1c decile in both treatment groups, with increased mortality seen in both the highest and lowest deciles. Patients in decile 4, who had a median HbA_1c of 7.5%, had the lowest mortality across deciles.

In the combination oral therapy group, only those in deciles 1 and 10 had significantly increased mortality, compared with patients in HbA_1c decile 4. However, in the insulin-treated cohort, deciles 1, 2, 3, 9, and 10 all had significantly greater mortality, compared with decile 4.

Progression to large-vessel disease events occurred in 8.2% of 20,817 patients who did not have large-vessel disease at baseline in the oral medication group, and in 11.9% of 13,475 patients in the insulin group. After adjustment for covariates, the risk of progression to large-vessel disease in both groups was higher for decile 1 (HR 1.54) and decile 10 (HR 1.36).

Compared with combination oral therapy, insulin treatment also was associated with an increased likelihood of progression to a first large-vessel disease event (HR 1.31).

The data suggest that for patients on oral combination therapy, a wide HbA_1c range is safe with respect to all-cause mortality and large-vessel events, but a narrower range may be desirable for patients taking insulin, the investigators said.

In an editorial, Dr. Beverley Balkau and Dr. Dominique Simon noted that although this study lends support to earlier studies, epidemiologic studies cannot show causal relationships. Moreover, observational databases cannot provide the detailed information available in a randomized clinical trial, such as the actual frequency of hypoglycemia.

This study has the advantage of real-world observation, added Dr. Balkau and Dr. Simon of the CESP Centre for Research in Epidemiology and Population Health, Villejuif, France (Lancet 2010 Jan. 27 [doi:10.1016/S0140-6736(09)62192-9

Dr. Balkau and Dr. Simon said priority should be given to treatment with insulin sensitizers for as long as possible in patients with type 2 diabetes, because these drugs allow a low HbA_1c to be achieved without risk of hypoglycemia. For patients with type 2 diabetes using insulin secretagogues or insulin itself, this study provides a rationale for an HbA_1c threshold of 7.5%, which corresponds to the lowest threshold of death and lowest event rate for large-vessel disease, they said.

Disclosures: Dr. Currie has financial ties to Eli Lilly and other drug companies. Four study coauthors are employed by Eli Lilly. Dr. Balkau and Dr. Simon have financial ties to several drug companies.

Diabetes guidelines might need revision to include a definition of a minimum HbA_1c value.

Source DR. CURRIE

Major Finding: HbA_1c values below 7.5% were associated with increased all-cause mortality and cardiovascular events.

Data Source: A general practice database analysis of 47,970 type 2 diabetes patients with recently intensified glucose-lowering therapy.

Disclosures: Funding by Eli Lilly & Co. The investigators disclosed ties to numerous manufacturers of diabetes treatments, including Eli Lilly.

Hemoglobin A_1c values below 7.5% were associated with increased all-cause mortality and cardiovascular events in patients with type 2 diabetes in an analysis of nearly 48,000 patients in a U.K. general practice database.

If confirmed, the findings suggest that diabetes guidelines might need revision to include a definition of a minimum HbA_1c value, Dr. Craig J. Currie of Cardiff (Wales) University and his associates said (Lancet 2010 Jan. 27 [doi:10.1016/S0140-6736(09)61969-3

The study also showed that insulin therapy was associated with higher mortality than combination oral therapy. Unadjusted mortality rates were 16.2 deaths per 1,000 person-years of follow-up for the oral combination therapy group and 27.2/1,000 for the insulin group. After exclusion of patients with high cardiovascular risk or renal impairment, insulin-based therapy remained associated with significantly greater all-cause mortality (hazard ratio 1.46) than did combination oral agents, the investigators reported.

The study used data from November 1986 to November 2008 in the U.K. General Practice Research Database. Two cohorts of patients aged 50 years and older with type 2 diabetes were assessed: 27,965 whose treatment regimen had been intensified from oral glucose-lowering monotherapy to a combination oral regimen with a sulfonylurea plus metformin, and 20,005 on oral hypoglycemic agents alone who were initiated on insulin with or without concomitant oral agents.

All-cause mortality was the primary outcome. The secondary outcome was the occurrence of a major cardiovascular event among those who had no record of cardiovascular disease before the index date. The mean follow-up was 4.5 years in the oral medication group and 5.2 years in the insulin group.

Patients were divided by HbA_1c decile, with the lowest decile (1) having a median HbA_1c of 6.4% and the highest decile (10) at 10.5%. Mortality varied by HbA_1c decile in both treatment groups, with increased mortality seen in both the highest and lowest deciles. Patients in decile 4, who had a median HbA_1c of 7.5%, had the lowest mortality across deciles.

In the combination oral therapy group, only those in deciles 1 and 10 had significantly increased mortality, compared with patients in HbA_1c decile 4. However, in the insulin-treated cohort, deciles 1, 2, 3, 9, and 10 all had significantly greater mortality, compared with decile 4.

Progression to large-vessel disease events occurred in 8.2% of 20,817 patients who did not have large-vessel disease at baseline in the oral medication group, and in 11.9% of 13,475 patients in the insulin group. After adjustment for covariates, the risk of progression to large-vessel disease in both groups was higher for decile 1 (HR 1.54) and decile 10 (HR 1.36).

Compared with combination oral therapy, insulin treatment also was associated with an increased likelihood of progression to a first large-vessel disease event (HR 1.31).

The data suggest that for patients on oral combination therapy, a wide HbA_1c range is safe with respect to all-cause mortality and large-vessel events, but a narrower range may be desirable for patients taking insulin, the investigators said.

In an editorial, Dr. Beverley Balkau and Dr. Dominique Simon noted that although this study lends support to earlier studies, epidemiologic studies cannot show causal relationships. Moreover, observational databases cannot provide the detailed information available in a randomized clinical trial, such as the actual frequency of hypoglycemia.

This study has the advantage of real-world observation, added Dr. Balkau and Dr. Simon of the CESP Centre for Research in Epidemiology and Population Health, Villejuif, France (Lancet 2010 Jan. 27 [doi:10.1016/S0140-6736(09)62192-9

Dr. Balkau and Dr. Simon said priority should be given to treatment with insulin sensitizers for as long as possible in patients with type 2 diabetes, because these drugs allow a low HbA_1c to be achieved without risk of hypoglycemia. For patients with type 2 diabetes using insulin secretagogues or insulin itself, this study provides a rationale for an HbA_1c threshold of 7.5%, which corresponds to the lowest threshold of death and lowest event rate for large-vessel disease, they said.

Disclosures: Dr. Currie has financial ties to Eli Lilly and other drug companies. Four study coauthors are employed by Eli Lilly. Dr. Balkau and Dr. Simon have financial ties to several drug companies.

Diabetes guidelines might need revision to include a definition of a minimum HbA_1c value.

Source DR. CURRIE

Major Finding: HbA_1c values below 7.5% were associated with increased all-cause mortality and cardiovascular events.

Data Source: A general practice database analysis of 47,970 type 2 diabetes patients with recently intensified glucose-lowering therapy.

Disclosures: Funding by Eli Lilly & Co. The investigators disclosed ties to numerous manufacturers of diabetes treatments, including Eli Lilly.

Hemoglobin A_1c values below 7.5% were associated with increased all-cause mortality and cardiovascular events in patients with type 2 diabetes in an analysis of nearly 48,000 patients in a U.K. general practice database.

If confirmed, the findings suggest that diabetes guidelines might need revision to include a definition of a minimum HbA_1c value, Dr. Craig J. Currie of Cardiff (Wales) University and his associates said (Lancet 2010 Jan. 27 [doi:10.1016/S0140-6736(09)61969-3

The study also showed that insulin therapy was associated with higher mortality than combination oral therapy. Unadjusted mortality rates were 16.2 deaths per 1,000 person-years of follow-up for the oral combination therapy group and 27.2/1,000 for the insulin group. After exclusion of patients with high cardiovascular risk or renal impairment, insulin-based therapy remained associated with significantly greater all-cause mortality (hazard ratio 1.46) than did combination oral agents, the investigators reported.

The study used data from November 1986 to November 2008 in the U.K. General Practice Research Database. Two cohorts of patients aged 50 years and older with type 2 diabetes were assessed: 27,965 whose treatment regimen had been intensified from oral glucose-lowering monotherapy to a combination oral regimen with a sulfonylurea plus metformin, and 20,005 on oral hypoglycemic agents alone who were initiated on insulin with or without concomitant oral agents.

All-cause mortality was the primary outcome. The secondary outcome was the occurrence of a major cardiovascular event among those who had no record of cardiovascular disease before the index date. The mean follow-up was 4.5 years in the oral medication group and 5.2 years in the insulin group.

Patients were divided by HbA_1c decile, with the lowest decile (1) having a median HbA_1c of 6.4% and the highest decile (10) at 10.5%. Mortality varied by HbA_1c decile in both treatment groups, with increased mortality seen in both the highest and lowest deciles. Patients in decile 4, who had a median HbA_1c of 7.5%, had the lowest mortality across deciles.

In the combination oral therapy group, only those in deciles 1 and 10 had significantly increased mortality, compared with patients in HbA_1c decile 4. However, in the insulin-treated cohort, deciles 1, 2, 3, 9, and 10 all had significantly greater mortality, compared with decile 4.

Progression to large-vessel disease events occurred in 8.2% of 20,817 patients who did not have large-vessel disease at baseline in the oral medication group, and in 11.9% of 13,475 patients in the insulin group. After adjustment for covariates, the risk of progression to large-vessel disease in both groups was higher for decile 1 (HR 1.54) and decile 10 (HR 1.36).

Compared with combination oral therapy, insulin treatment also was associated with an increased likelihood of progression to a first large-vessel disease event (HR 1.31).

The data suggest that for patients on oral combination therapy, a wide HbA_1c range is safe with respect to all-cause mortality and large-vessel events, but a narrower range may be desirable for patients taking insulin, the investigators said.

In an editorial, Dr. Beverley Balkau and Dr. Dominique Simon noted that although this study lends support to earlier studies, epidemiologic studies cannot show causal relationships. Moreover, observational databases cannot provide the detailed information available in a randomized clinical trial, such as the actual frequency of hypoglycemia.

This study has the advantage of real-world observation, added Dr. Balkau and Dr. Simon of the CESP Centre for Research in Epidemiology and Population Health, Villejuif, France (Lancet 2010 Jan. 27 [doi:10.1016/S0140-6736(09)62192-9

Dr. Balkau and Dr. Simon said priority should be given to treatment with insulin sensitizers for as long as possible in patients with type 2 diabetes, because these drugs allow a low HbA_1c to be achieved without risk of hypoglycemia. For patients with type 2 diabetes using insulin secretagogues or insulin itself, this study provides a rationale for an HbA_1c threshold of 7.5%, which corresponds to the lowest threshold of death and lowest event rate for large-vessel disease, they said.

Disclosures: Dr. Currie has financial ties to Eli Lilly and other drug companies. Four study coauthors are employed by Eli Lilly. Dr. Balkau and Dr. Simon have financial ties to several drug companies.

Diabetes guidelines might need revision to include a definition of a minimum HbA_1c value.

Source DR. CURRIE

NATIONAL HARBOR, MD. — Bone mineralization was not significantly altered among 17 children receiving chronic proton pump inhibitor therapy, including 12 who were also using inhaled steroids.

Proton pump inhibitors (PPIs) are commonly prescribed for acid suppression in children with gastroesophageal reflux, sometimes for long periods.

A significantly increased risk of bone fracture has been reported in adult patients receiving long-term PPI therapy (JAMA 2006;296:2947-53), and chronic acid suppression has also been shown to impair calcium absorption, thereby promoting bone resorption (Am. J. Med. 2005;118:778-81).

However, this pilot study is believed to be the first to look at bone mineralization or fractures in children on PPIs, Dr. Stephanie Willot said in a poster presentation at the annual meeting of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition.

The 17 patients (12 boys) had a mean age of 7.8 years (range 0.8-16.7 years). All had severe gastroesophageal reflux secondary to esophageal atresia and had received PPI therapy at a mean dosage of 2.0 mg/kg daily (1.0-3.2 mg/kg) for a mean of 2.6 years (0.6-11.3 years). Twelve of the children were also receiving chronic inhaled steroid therapy for pulmonary disease.

Lumbar spine areal bone mineral density (BMD) was assessed by using dual x-ray absorptiometry and was compared with normative data. Volumetric BMD, a parameter that more accurately assesses BMD in patients with short stature, was also calculated in order to account for differences in bone size, said Dr. Willot of the division of pediatric gastroenterology at Sainte-Justine Hospital, University of Montreal, who conducted the study with colleagues from the division of pediatric endocrinology.

No patient had a history of traumatic fracture. Five patients (29%) had a statural growth delay of less than −2 standard deviations for age. Among the 14 children older than 2 years, 5 (35%) had a body mass index less than the 10th percentile.

No patient had a significantly low BMD, defined in the study as a z score less than −2 standard deviations for age. Although six patients (35%) had a z score BMD of less than −1 standard deviation for age, they all had normal volumetric BMD (ranging from −0.8 to 0.6 standard deviation), as did the other seven children who were older than 4 years of age, Dr. Willot and her associates reported.

In a follow-up interview, Dr. Willot said that, given the small sample size of the study and its cross-sectional nature, its implications are limited to the finding that BMD is not low in children on chronic PPI therapy. “We cannot conclude about the association between PPI and fracture risk. In the future, we would like to follow our cohort to assess BMD in a longitudinal way to establish if z score could decrease during the course of PPI treatment.”

Dr. Willot stated that she had no personal financial disclosures.

NATIONAL HARBOR, MD. — Bone mineralization was not significantly altered among 17 children receiving chronic proton pump inhibitor therapy, including 12 who were also using inhaled steroids.

Proton pump inhibitors (PPIs) are commonly prescribed for acid suppression in children with gastroesophageal reflux, sometimes for long periods.

A significantly increased risk of bone fracture has been reported in adult patients receiving long-term PPI therapy (JAMA 2006;296:2947-53), and chronic acid suppression has also been shown to impair calcium absorption, thereby promoting bone resorption (Am. J. Med. 2005;118:778-81).

However, this pilot study is believed to be the first to look at bone mineralization or fractures in children on PPIs, Dr. Stephanie Willot said in a poster presentation at the annual meeting of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition.

The 17 patients (12 boys) had a mean age of 7.8 years (range 0.8-16.7 years). All had severe gastroesophageal reflux secondary to esophageal atresia and had received PPI therapy at a mean dosage of 2.0 mg/kg daily (1.0-3.2 mg/kg) for a mean of 2.6 years (0.6-11.3 years). Twelve of the children were also receiving chronic inhaled steroid therapy for pulmonary disease.

Lumbar spine areal bone mineral density (BMD) was assessed by using dual x-ray absorptiometry and was compared with normative data. Volumetric BMD, a parameter that more accurately assesses BMD in patients with short stature, was also calculated in order to account for differences in bone size, said Dr. Willot of the division of pediatric gastroenterology at Sainte-Justine Hospital, University of Montreal, who conducted the study with colleagues from the division of pediatric endocrinology.

No patient had a history of traumatic fracture. Five patients (29%) had a statural growth delay of less than −2 standard deviations for age. Among the 14 children older than 2 years, 5 (35%) had a body mass index less than the 10th percentile.

No patient had a significantly low BMD, defined in the study as a z score less than −2 standard deviations for age. Although six patients (35%) had a z score BMD of less than −1 standard deviation for age, they all had normal volumetric BMD (ranging from −0.8 to 0.6 standard deviation), as did the other seven children who were older than 4 years of age, Dr. Willot and her associates reported.

In a follow-up interview, Dr. Willot said that, given the small sample size of the study and its cross-sectional nature, its implications are limited to the finding that BMD is not low in children on chronic PPI therapy. “We cannot conclude about the association between PPI and fracture risk. In the future, we would like to follow our cohort to assess BMD in a longitudinal way to establish if z score could decrease during the course of PPI treatment.”

Dr. Willot stated that she had no personal financial disclosures.

NATIONAL HARBOR, MD. — Bone mineralization was not significantly altered among 17 children receiving chronic proton pump inhibitor therapy, including 12 who were also using inhaled steroids.

Proton pump inhibitors (PPIs) are commonly prescribed for acid suppression in children with gastroesophageal reflux, sometimes for long periods.

A significantly increased risk of bone fracture has been reported in adult patients receiving long-term PPI therapy (JAMA 2006;296:2947-53), and chronic acid suppression has also been shown to impair calcium absorption, thereby promoting bone resorption (Am. J. Med. 2005;118:778-81).

However, this pilot study is believed to be the first to look at bone mineralization or fractures in children on PPIs, Dr. Stephanie Willot said in a poster presentation at the annual meeting of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition.

The 17 patients (12 boys) had a mean age of 7.8 years (range 0.8-16.7 years). All had severe gastroesophageal reflux secondary to esophageal atresia and had received PPI therapy at a mean dosage of 2.0 mg/kg daily (1.0-3.2 mg/kg) for a mean of 2.6 years (0.6-11.3 years). Twelve of the children were also receiving chronic inhaled steroid therapy for pulmonary disease.

Lumbar spine areal bone mineral density (BMD) was assessed by using dual x-ray absorptiometry and was compared with normative data. Volumetric BMD, a parameter that more accurately assesses BMD in patients with short stature, was also calculated in order to account for differences in bone size, said Dr. Willot of the division of pediatric gastroenterology at Sainte-Justine Hospital, University of Montreal, who conducted the study with colleagues from the division of pediatric endocrinology.

No patient had a history of traumatic fracture. Five patients (29%) had a statural growth delay of less than −2 standard deviations for age. Among the 14 children older than 2 years, 5 (35%) had a body mass index less than the 10th percentile.

No patient had a significantly low BMD, defined in the study as a z score less than −2 standard deviations for age. Although six patients (35%) had a z score BMD of less than −1 standard deviation for age, they all had normal volumetric BMD (ranging from −0.8 to 0.6 standard deviation), as did the other seven children who were older than 4 years of age, Dr. Willot and her associates reported.

In a follow-up interview, Dr. Willot said that, given the small sample size of the study and its cross-sectional nature, its implications are limited to the finding that BMD is not low in children on chronic PPI therapy. “We cannot conclude about the association between PPI and fracture risk. In the future, we would like to follow our cohort to assess BMD in a longitudinal way to establish if z score could decrease during the course of PPI treatment.”

Dr. Willot stated that she had no personal financial disclosures.

Major Finding: Of the six pregnant and two postpartum patients who died, six had underlying medical conditions. None had received antiviral medication within 48 hours after symptom onset.

Data Source: Data from 94 pregnant women, 8 postpartum women, and 137 nonpregnant women of reproductive age who were hospitalized with or died of 2009 H1N1 influenza.

Disclosures: None reported.

Maternal mortality from 2009 influenza H1N1 was estimated at 4.3/100,000 live births in California, based on the results of statewide surveillance.

The maternal mortality ratio—the number of maternal deaths per 100,000 live births—from any cause was 19.3 in California in 2005 and 13.3 in the United States in 2006. More than two-thirds of maternal deaths in the United States are directly related to obstetrical factors, and deaths from influenza had been rare prior to the 2009 influenza H1N1 outbreak, Dr. Janice K. Louieof the California Department of Public Health, Richmond, and her associates wrote (N. Engl. J. Med. 2010;362:27-35).

Now, the high maternal death rate attributed to this flu has the potential to notably increase the overall maternal mortality rate in the United States for 2009, Dr. Louie and her associates said.

From April 23 through Aug. 11, 2009, data were reported for 94 pregnant women, 8 postpartum women, and 137 nonpregnant women of reproductive age who were hospitalized with or died of 2009 H1N1 influenza.

Of the 78 pregnant women whose race/ethnicity was known, 43 were Hispanic, 15 were white, 9 were Asian or Pacific Islander, 6 were non-Hispanic black, and 5 were “other.”

About one-third (32) of the 93 pregnant women for whom the data were available had underlying medical conditions that placed them at increased risk for influenza complications, as did a fourth (2) of the 8 postpartum women and two-thirds (82) of the 137 nonpregnant women. The most common condition was asthma, affecting 16% of the pregnant and 28% of the nonpregnant women.

The most commonly reported symptoms among pregnant patients were cough (93%), fever (91%), sore throat (41%), shortness of breath (41%), muscle aches (41%), and nausea or vomiting (33%).

Eighteen (19%) of the pregnant patients were admitted to intensive care, as were 4 of the 8 (50%) postpartum patients and 41 (30%) of the nonpregnant patients.

In some of the cases, reliance on rapid influenza tests appears to have contributed to treatment delays. Rapid influenza tests were falsely negative in 38% of the total 153 who were tested. Of those 58 patients, 28 (48%) were pregnant. Only 7 of the 25 (28%) pregnant women with falsely negative results for whom information was available received antiviral treatment within the recommended 48 hours after symptom onset. Five of the eight patients (63%) who died had false-negative rapid test results, Dr. Louie and her associates noted.

In all, while 81% of both pregnant and nonpregnant women received antiviral treatment, only half of the pregnant women and a third of the nonpregnant women received it within the recommended 48-hour time frame, the investigators reported.

Of the six pregnant and two postpartum patients who died, six had underlying medical conditions, including hypothyroidism in two, gestational diabetes in one, and a history of Hodgkin's disease in one. All eight required intensive care, and none had received antiviral medication within 48 hours after symptom onset.

The maternal mortality ratio was based on an estimated 188,383 births in the state of California from April 3 through Aug. 5. The eight deaths caused by 2009 H1N1 during that time resulted in a cause-specific maternal mortality ratio of 4.3, they said.

Major Finding: Of the six pregnant and two postpartum patients who died, six had underlying medical conditions. None had received antiviral medication within 48 hours after symptom onset.

Data Source: Data from 94 pregnant women, 8 postpartum women, and 137 nonpregnant women of reproductive age who were hospitalized with or died of 2009 H1N1 influenza.

Disclosures: None reported.

Maternal mortality from 2009 influenza H1N1 was estimated at 4.3/100,000 live births in California, based on the results of statewide surveillance.

The maternal mortality ratio—the number of maternal deaths per 100,000 live births—from any cause was 19.3 in California in 2005 and 13.3 in the United States in 2006. More than two-thirds of maternal deaths in the United States are directly related to obstetrical factors, and deaths from influenza had been rare prior to the 2009 influenza H1N1 outbreak, Dr. Janice K. Louieof the California Department of Public Health, Richmond, and her associates wrote (N. Engl. J. Med. 2010;362:27-35).

Now, the high maternal death rate attributed to this flu has the potential to notably increase the overall maternal mortality rate in the United States for 2009, Dr. Louie and her associates said.

From April 23 through Aug. 11, 2009, data were reported for 94 pregnant women, 8 postpartum women, and 137 nonpregnant women of reproductive age who were hospitalized with or died of 2009 H1N1 influenza.

Of the 78 pregnant women whose race/ethnicity was known, 43 were Hispanic, 15 were white, 9 were Asian or Pacific Islander, 6 were non-Hispanic black, and 5 were “other.”

About one-third (32) of the 93 pregnant women for whom the data were available had underlying medical conditions that placed them at increased risk for influenza complications, as did a fourth (2) of the 8 postpartum women and two-thirds (82) of the 137 nonpregnant women. The most common condition was asthma, affecting 16% of the pregnant and 28% of the nonpregnant women.

The most commonly reported symptoms among pregnant patients were cough (93%), fever (91%), sore throat (41%), shortness of breath (41%), muscle aches (41%), and nausea or vomiting (33%).

Eighteen (19%) of the pregnant patients were admitted to intensive care, as were 4 of the 8 (50%) postpartum patients and 41 (30%) of the nonpregnant patients.

In some of the cases, reliance on rapid influenza tests appears to have contributed to treatment delays. Rapid influenza tests were falsely negative in 38% of the total 153 who were tested. Of those 58 patients, 28 (48%) were pregnant. Only 7 of the 25 (28%) pregnant women with falsely negative results for whom information was available received antiviral treatment within the recommended 48 hours after symptom onset. Five of the eight patients (63%) who died had false-negative rapid test results, Dr. Louie and her associates noted.

In all, while 81% of both pregnant and nonpregnant women received antiviral treatment, only half of the pregnant women and a third of the nonpregnant women received it within the recommended 48-hour time frame, the investigators reported.

Of the six pregnant and two postpartum patients who died, six had underlying medical conditions, including hypothyroidism in two, gestational diabetes in one, and a history of Hodgkin's disease in one. All eight required intensive care, and none had received antiviral medication within 48 hours after symptom onset.

The maternal mortality ratio was based on an estimated 188,383 births in the state of California from April 3 through Aug. 5. The eight deaths caused by 2009 H1N1 during that time resulted in a cause-specific maternal mortality ratio of 4.3, they said.

Major Finding: Of the six pregnant and two postpartum patients who died, six had underlying medical conditions. None had received antiviral medication within 48 hours after symptom onset.

Data Source: Data from 94 pregnant women, 8 postpartum women, and 137 nonpregnant women of reproductive age who were hospitalized with or died of 2009 H1N1 influenza.

Disclosures: None reported.

Maternal mortality from 2009 influenza H1N1 was estimated at 4.3/100,000 live births in California, based on the results of statewide surveillance.

The maternal mortality ratio—the number of maternal deaths per 100,000 live births—from any cause was 19.3 in California in 2005 and 13.3 in the United States in 2006. More than two-thirds of maternal deaths in the United States are directly related to obstetrical factors, and deaths from influenza had been rare prior to the 2009 influenza H1N1 outbreak, Dr. Janice K. Louieof the California Department of Public Health, Richmond, and her associates wrote (N. Engl. J. Med. 2010;362:27-35).

Now, the high maternal death rate attributed to this flu has the potential to notably increase the overall maternal mortality rate in the United States for 2009, Dr. Louie and her associates said.

From April 23 through Aug. 11, 2009, data were reported for 94 pregnant women, 8 postpartum women, and 137 nonpregnant women of reproductive age who were hospitalized with or died of 2009 H1N1 influenza.

Of the 78 pregnant women whose race/ethnicity was known, 43 were Hispanic, 15 were white, 9 were Asian or Pacific Islander, 6 were non-Hispanic black, and 5 were “other.”

About one-third (32) of the 93 pregnant women for whom the data were available had underlying medical conditions that placed them at increased risk for influenza complications, as did a fourth (2) of the 8 postpartum women and two-thirds (82) of the 137 nonpregnant women. The most common condition was asthma, affecting 16% of the pregnant and 28% of the nonpregnant women.

The most commonly reported symptoms among pregnant patients were cough (93%), fever (91%), sore throat (41%), shortness of breath (41%), muscle aches (41%), and nausea or vomiting (33%).

Eighteen (19%) of the pregnant patients were admitted to intensive care, as were 4 of the 8 (50%) postpartum patients and 41 (30%) of the nonpregnant patients.

In some of the cases, reliance on rapid influenza tests appears to have contributed to treatment delays. Rapid influenza tests were falsely negative in 38% of the total 153 who were tested. Of those 58 patients, 28 (48%) were pregnant. Only 7 of the 25 (28%) pregnant women with falsely negative results for whom information was available received antiviral treatment within the recommended 48 hours after symptom onset. Five of the eight patients (63%) who died had false-negative rapid test results, Dr. Louie and her associates noted.

In all, while 81% of both pregnant and nonpregnant women received antiviral treatment, only half of the pregnant women and a third of the nonpregnant women received it within the recommended 48-hour time frame, the investigators reported.

Of the six pregnant and two postpartum patients who died, six had underlying medical conditions, including hypothyroidism in two, gestational diabetes in one, and a history of Hodgkin's disease in one. All eight required intensive care, and none had received antiviral medication within 48 hours after symptom onset.

The maternal mortality ratio was based on an estimated 188,383 births in the state of California from April 3 through Aug. 5. The eight deaths caused by 2009 H1N1 during that time resulted in a cause-specific maternal mortality ratio of 4.3, they said.

The National Vaccine Advisory Committee has endorsed a working group report concluding that no safety signals have been identified so far with the 2009 H1N1 influenza vaccine.

In a public teleconference, working group chair Dr. Marie McCormick said the group reviewed data for a total of 74 million doses of inactivated (injected) vaccine and 19 million doses of live attenuated (intranasal) H1N1 vaccine distributed as of around Christmastime.

They concluded that the data do not suggest a safety signal between the outcomes examined and the vaccine, defining “signal” as an outcome temporally occurring more than anticipated by chance alone.

No serious increases in adverse events have been observed to date in any of the pandemic H1N1 vaccine clinical trials, and a comparison of serious events reported to the Vaccine Adverse Events Reporting System have shown similar levels between the H1N1vaccine and the seasonal influenza vaccine, as well as other vaccines.

In addition, active surveillance systems that are using rapid-cycle analysis of prespecified outcomes have also been within expected values, said Dr. McCormick, the Sumner and Esther Feldberg professor of Maternal and Child Health at Harvard University, Boston.

However, she cautioned, the size of the population captured under active surveillance is still somewhat limited, and some of the prespecified event analyses have involved small numbers.

A new federal project called Post-Licensure Rapid Immunization Safety Monitoring (PRISM), designed to monitor H1N1 vaccine safety in real-time using data from large health plans covering approximately 10% of the U.S. population, was set to begin in the next few days.

“As more data are available through active surveillance, conclusions will be based on a larger accumulation of data. Larger samples may be needed to detect rare adverse events,” she said.

The NVAC members voted to endorse the report during the teleconference. It now goes to the U.S. assistant secretary for health, who will review and consider it for formal implementation.

The National Vaccine Advisory Committee has endorsed a working group report concluding that no safety signals have been identified so far with the 2009 H1N1 influenza vaccine.

In a public teleconference, working group chair Dr. Marie McCormick said the group reviewed data for a total of 74 million doses of inactivated (injected) vaccine and 19 million doses of live attenuated (intranasal) H1N1 vaccine distributed as of around Christmastime.

They concluded that the data do not suggest a safety signal between the outcomes examined and the vaccine, defining “signal” as an outcome temporally occurring more than anticipated by chance alone.

No serious increases in adverse events have been observed to date in any of the pandemic H1N1 vaccine clinical trials, and a comparison of serious events reported to the Vaccine Adverse Events Reporting System have shown similar levels between the H1N1vaccine and the seasonal influenza vaccine, as well as other vaccines.

In addition, active surveillance systems that are using rapid-cycle analysis of prespecified outcomes have also been within expected values, said Dr. McCormick, the Sumner and Esther Feldberg professor of Maternal and Child Health at Harvard University, Boston.

However, she cautioned, the size of the population captured under active surveillance is still somewhat limited, and some of the prespecified event analyses have involved small numbers.

A new federal project called Post-Licensure Rapid Immunization Safety Monitoring (PRISM), designed to monitor H1N1 vaccine safety in real-time using data from large health plans covering approximately 10% of the U.S. population, was set to begin in the next few days.

“As more data are available through active surveillance, conclusions will be based on a larger accumulation of data. Larger samples may be needed to detect rare adverse events,” she said.

The NVAC members voted to endorse the report during the teleconference. It now goes to the U.S. assistant secretary for health, who will review and consider it for formal implementation.

The National Vaccine Advisory Committee has endorsed a working group report concluding that no safety signals have been identified so far with the 2009 H1N1 influenza vaccine.

In a public teleconference, working group chair Dr. Marie McCormick said the group reviewed data for a total of 74 million doses of inactivated (injected) vaccine and 19 million doses of live attenuated (intranasal) H1N1 vaccine distributed as of around Christmastime.

They concluded that the data do not suggest a safety signal between the outcomes examined and the vaccine, defining “signal” as an outcome temporally occurring more than anticipated by chance alone.

No serious increases in adverse events have been observed to date in any of the pandemic H1N1 vaccine clinical trials, and a comparison of serious events reported to the Vaccine Adverse Events Reporting System have shown similar levels between the H1N1vaccine and the seasonal influenza vaccine, as well as other vaccines.

In addition, active surveillance systems that are using rapid-cycle analysis of prespecified outcomes have also been within expected values, said Dr. McCormick, the Sumner and Esther Feldberg professor of Maternal and Child Health at Harvard University, Boston.

However, she cautioned, the size of the population captured under active surveillance is still somewhat limited, and some of the prespecified event analyses have involved small numbers.

A new federal project called Post-Licensure Rapid Immunization Safety Monitoring (PRISM), designed to monitor H1N1 vaccine safety in real-time using data from large health plans covering approximately 10% of the U.S. population, was set to begin in the next few days.

“As more data are available through active surveillance, conclusions will be based on a larger accumulation of data. Larger samples may be needed to detect rare adverse events,” she said.

The NVAC members voted to endorse the report during the teleconference. It now goes to the U.S. assistant secretary for health, who will review and consider it for formal implementation.

Major Finding: The proportion of 5- to 14-year-olds infected with 2009 H1N1 flu in high-risk regions of England increased from 3.7% at baseline to 45.7% in September 2009.

Data Source: Serologic survey using 1,403 serum samples pre–2009 H1N1 outbreak and 1,954 from August and September 2009.

Disclosures: Lead author has none; two others have ties to Novartis, Sanofi Pasteur, Baxter, and CSI Australia.

Serologic testing revealed that about one in every three children living in high-risk areas of England was infected with 2009 H1N1 influenza during the first wave of infections—10 times more than estimated based on clinical surveillance.

The finding comes from a serologic survey sponsored by the U.K.'s National Institute for Health Research Health Technology Assessment Programme.

The survey also showed that preexisting antibody in older individuals protected them from infection.

Serologic surveillance provides a more stable method for predicting future disease incidence than does clinical surveillance, which captures only individuals who present with symptoms, said Dr. Elizabeth Miller of the Health Protection Agency, London, and her associates. (Lancet 2010 Jan. 21[doi:10.1016/S0140-6736(09)62126-7

Of 1,403 serum samples taken in late 2008 and early 2009 prior to the H1N1 outbreak, the proportions with a hemagglutination inhibition titer of around 1:32—which is assumed to be protective—were 2.8% of the 359 children under 15 years of age and 23.3% of 549 adults aged 65 years and older, they reported.

The range varied from 1.8% in the 0- to 4-year-old age group to 31.3% in those aged over 80 years.

“Individuals born before 1957 might have been exposed to influenza H1 strains circulating in the first half of the 20th century, which are more closely related to current swine-origin 2009 pandemic H1N1 viruses,” the investigators said.

Once the outbreak began, the proportion of samples with titers of 1:32 or more increased from 11.6% of 43 within the first week of infection to 89.1% of 55 after 21 days, Dr. Miller and her associates reported.

Of a total of 1,954 samples tested during August and September 2009, there was significant geographical variation in the number with protective titers.

Infection rates and increases were highest in London and the West Midlands: Among children aged 0-4 years in those regions, immunity increased from 1.8% at baseline to 23.1% in September 2009.

For 5- to 14-year-olds, the increase was from 3.7% to 45.7%, and for 15- to 24-year-olds, from 17.5% to 38.1%, Dr. Miller and her associates found.

In contrast, there was no increase in the proportions with protective titers among older adults.

Overall, rates of infection in the first wave were greatest in children younger than 15 years of age, with an estimated 42% of schoolchildren aged 5-14 years infected in high-incidence regions.

“This finding is consistent with the high level of susceptibility in children and the increased potential for transmission that occurs within schools. We also showed substantial differences between regions in the extent of infection during the first wave,” the investigators commented.

Infection rates seen in this study are approximately 10 times greater than original clinically based estimates from the Health Protection Agency and have implications beyond England, Dr. Miller and her associates noted.

“This serological study shows the true extent of H1N1 infection in the initial wave of the pandemic in England in 2009. Its findings should be applicable to other countries that have experienced a similar first wave,” they concluded.

Major Finding: The proportion of 5- to 14-year-olds infected with 2009 H1N1 flu in high-risk regions of England increased from 3.7% at baseline to 45.7% in September 2009.

Data Source: Serologic survey using 1,403 serum samples pre–2009 H1N1 outbreak and 1,954 from August and September 2009.

Disclosures: Lead author has none; two others have ties to Novartis, Sanofi Pasteur, Baxter, and CSI Australia.

Serologic testing revealed that about one in every three children living in high-risk areas of England was infected with 2009 H1N1 influenza during the first wave of infections—10 times more than estimated based on clinical surveillance.

The finding comes from a serologic survey sponsored by the U.K.'s National Institute for Health Research Health Technology Assessment Programme.

The survey also showed that preexisting antibody in older individuals protected them from infection.

Serologic surveillance provides a more stable method for predicting future disease incidence than does clinical surveillance, which captures only individuals who present with symptoms, said Dr. Elizabeth Miller of the Health Protection Agency, London, and her associates. (Lancet 2010 Jan. 21[doi:10.1016/S0140-6736(09)62126-7

Of 1,403 serum samples taken in late 2008 and early 2009 prior to the H1N1 outbreak, the proportions with a hemagglutination inhibition titer of around 1:32—which is assumed to be protective—were 2.8% of the 359 children under 15 years of age and 23.3% of 549 adults aged 65 years and older, they reported.

The range varied from 1.8% in the 0- to 4-year-old age group to 31.3% in those aged over 80 years.

“Individuals born before 1957 might have been exposed to influenza H1 strains circulating in the first half of the 20th century, which are more closely related to current swine-origin 2009 pandemic H1N1 viruses,” the investigators said.

Once the outbreak began, the proportion of samples with titers of 1:32 or more increased from 11.6% of 43 within the first week of infection to 89.1% of 55 after 21 days, Dr. Miller and her associates reported.

Of a total of 1,954 samples tested during August and September 2009, there was significant geographical variation in the number with protective titers.

Infection rates and increases were highest in London and the West Midlands: Among children aged 0-4 years in those regions, immunity increased from 1.8% at baseline to 23.1% in September 2009.

For 5- to 14-year-olds, the increase was from 3.7% to 45.7%, and for 15- to 24-year-olds, from 17.5% to 38.1%, Dr. Miller and her associates found.

In contrast, there was no increase in the proportions with protective titers among older adults.

Overall, rates of infection in the first wave were greatest in children younger than 15 years of age, with an estimated 42% of schoolchildren aged 5-14 years infected in high-incidence regions.

“This finding is consistent with the high level of susceptibility in children and the increased potential for transmission that occurs within schools. We also showed substantial differences between regions in the extent of infection during the first wave,” the investigators commented.

Infection rates seen in this study are approximately 10 times greater than original clinically based estimates from the Health Protection Agency and have implications beyond England, Dr. Miller and her associates noted.

“This serological study shows the true extent of H1N1 infection in the initial wave of the pandemic in England in 2009. Its findings should be applicable to other countries that have experienced a similar first wave,” they concluded.

Major Finding: The proportion of 5- to 14-year-olds infected with 2009 H1N1 flu in high-risk regions of England increased from 3.7% at baseline to 45.7% in September 2009.

Data Source: Serologic survey using 1,403 serum samples pre–2009 H1N1 outbreak and 1,954 from August and September 2009.

Disclosures: Lead author has none; two others have ties to Novartis, Sanofi Pasteur, Baxter, and CSI Australia.

Serologic testing revealed that about one in every three children living in high-risk areas of England was infected with 2009 H1N1 influenza during the first wave of infections—10 times more than estimated based on clinical surveillance.

The finding comes from a serologic survey sponsored by the U.K.'s National Institute for Health Research Health Technology Assessment Programme.

The survey also showed that preexisting antibody in older individuals protected them from infection.

Serologic surveillance provides a more stable method for predicting future disease incidence than does clinical surveillance, which captures only individuals who present with symptoms, said Dr. Elizabeth Miller of the Health Protection Agency, London, and her associates. (Lancet 2010 Jan. 21[doi:10.1016/S0140-6736(09)62126-7

Of 1,403 serum samples taken in late 2008 and early 2009 prior to the H1N1 outbreak, the proportions with a hemagglutination inhibition titer of around 1:32—which is assumed to be protective—were 2.8% of the 359 children under 15 years of age and 23.3% of 549 adults aged 65 years and older, they reported.

The range varied from 1.8% in the 0- to 4-year-old age group to 31.3% in those aged over 80 years.

“Individuals born before 1957 might have been exposed to influenza H1 strains circulating in the first half of the 20th century, which are more closely related to current swine-origin 2009 pandemic H1N1 viruses,” the investigators said.

Once the outbreak began, the proportion of samples with titers of 1:32 or more increased from 11.6% of 43 within the first week of infection to 89.1% of 55 after 21 days, Dr. Miller and her associates reported.

Of a total of 1,954 samples tested during August and September 2009, there was significant geographical variation in the number with protective titers.

Infection rates and increases were highest in London and the West Midlands: Among children aged 0-4 years in those regions, immunity increased from 1.8% at baseline to 23.1% in September 2009.

For 5- to 14-year-olds, the increase was from 3.7% to 45.7%, and for 15- to 24-year-olds, from 17.5% to 38.1%, Dr. Miller and her associates found.

In contrast, there was no increase in the proportions with protective titers among older adults.

Overall, rates of infection in the first wave were greatest in children younger than 15 years of age, with an estimated 42% of schoolchildren aged 5-14 years infected in high-incidence regions.

“This finding is consistent with the high level of susceptibility in children and the increased potential for transmission that occurs within schools. We also showed substantial differences between regions in the extent of infection during the first wave,” the investigators commented.

Infection rates seen in this study are approximately 10 times greater than original clinically based estimates from the Health Protection Agency and have implications beyond England, Dr. Miller and her associates noted.

“This serological study shows the true extent of H1N1 infection in the initial wave of the pandemic in England in 2009. Its findings should be applicable to other countries that have experienced a similar first wave,” they concluded.

NATIONAL HARBOR, MD. — Bone mineralization was not significantly altered among 17 children receiving chronic proton pump inhibitor therapy, including 12 who were also using inhaled steroids.

The 17 patients (12 boys) had a mean age of 7.8 years (range 0.8–16.7 years). All had severe gastroesophageal reflux secondary to esophageal atresia and had received proton pump inhibitor (PPI) therapy at a mean dosage of 2.0 mg/kg daily (1.0–3.2 mg/kg) for a mean of 2.6 years (0.6–11.3 years). Twelve of the children were also receiving chronic inhaled steroid therapy for pulmonary disease, Dr. Stephanie Willot said in at the annual meeting of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition.

Lumbar spine areal bone mineral density (BMD) was assessed by using dual-energy x-ray absorptiometry and was compared with normative data. Volumetric BMD, a parameter that more accurately assesses BMD in patients with short stature, was also calculated in order to account for differences in bone size, said Dr. Willot of the division of pediatric gastroenterology at Sainte-Justine Hospital, University of Montreal, who conducted the study with colleagues from the division of pediatric endocrinology.

No patient had a history of traumatic fracture. Five patients (29%) had a statural growth delay of less than −2 standard deviations for age. Among the 14 children older than 2 years, 5 (35%) had a body mass index less than the 10th percentile.

No patient had a significantly low BMD, defined as a z score less than −2 standard deviations for age. Although six patients (35%) had a z score BMD of less than −1 standard deviation for age, they all had normal volumetric BMD (ranging from −0.8 to 0.6 standard deviation), as did the other seven children who were older than 4 years of age.

Given the small sample size of the study and its cross-sectional nature, “We cannot conclude about the association between PPI and fracture risk,” said Dr. Willot, who reported having no disclosures.

NATIONAL HARBOR, MD. — Bone mineralization was not significantly altered among 17 children receiving chronic proton pump inhibitor therapy, including 12 who were also using inhaled steroids.

The 17 patients (12 boys) had a mean age of 7.8 years (range 0.8–16.7 years). All had severe gastroesophageal reflux secondary to esophageal atresia and had received proton pump inhibitor (PPI) therapy at a mean dosage of 2.0 mg/kg daily (1.0–3.2 mg/kg) for a mean of 2.6 years (0.6–11.3 years). Twelve of the children were also receiving chronic inhaled steroid therapy for pulmonary disease, Dr. Stephanie Willot said in at the annual meeting of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition.

Lumbar spine areal bone mineral density (BMD) was assessed by using dual-energy x-ray absorptiometry and was compared with normative data. Volumetric BMD, a parameter that more accurately assesses BMD in patients with short stature, was also calculated in order to account for differences in bone size, said Dr. Willot of the division of pediatric gastroenterology at Sainte-Justine Hospital, University of Montreal, who conducted the study with colleagues from the division of pediatric endocrinology.

No patient had a history of traumatic fracture. Five patients (29%) had a statural growth delay of less than −2 standard deviations for age. Among the 14 children older than 2 years, 5 (35%) had a body mass index less than the 10th percentile.

No patient had a significantly low BMD, defined as a z score less than −2 standard deviations for age. Although six patients (35%) had a z score BMD of less than −1 standard deviation for age, they all had normal volumetric BMD (ranging from −0.8 to 0.6 standard deviation), as did the other seven children who were older than 4 years of age.

Given the small sample size of the study and its cross-sectional nature, “We cannot conclude about the association between PPI and fracture risk,” said Dr. Willot, who reported having no disclosures.

NATIONAL HARBOR, MD. — Bone mineralization was not significantly altered among 17 children receiving chronic proton pump inhibitor therapy, including 12 who were also using inhaled steroids.

The 17 patients (12 boys) had a mean age of 7.8 years (range 0.8–16.7 years). All had severe gastroesophageal reflux secondary to esophageal atresia and had received proton pump inhibitor (PPI) therapy at a mean dosage of 2.0 mg/kg daily (1.0–3.2 mg/kg) for a mean of 2.6 years (0.6–11.3 years). Twelve of the children were also receiving chronic inhaled steroid therapy for pulmonary disease, Dr. Stephanie Willot said in at the annual meeting of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition.

Lumbar spine areal bone mineral density (BMD) was assessed by using dual-energy x-ray absorptiometry and was compared with normative data. Volumetric BMD, a parameter that more accurately assesses BMD in patients with short stature, was also calculated in order to account for differences in bone size, said Dr. Willot of the division of pediatric gastroenterology at Sainte-Justine Hospital, University of Montreal, who conducted the study with colleagues from the division of pediatric endocrinology.

No patient had a history of traumatic fracture. Five patients (29%) had a statural growth delay of less than −2 standard deviations for age. Among the 14 children older than 2 years, 5 (35%) had a body mass index less than the 10th percentile.

No patient had a significantly low BMD, defined as a z score less than −2 standard deviations for age. Although six patients (35%) had a z score BMD of less than −1 standard deviation for age, they all had normal volumetric BMD (ranging from −0.8 to 0.6 standard deviation), as did the other seven children who were older than 4 years of age.

Given the small sample size of the study and its cross-sectional nature, “We cannot conclude about the association between PPI and fracture risk,” said Dr. Willot, who reported having no disclosures.

NATIONAL HARBOR, MD. — Abdominal migraine may be responsible for up to 15% of all cases of idiopathic recurrent abdominal pain in children, according to an analysis of records from more than 400 children.

Abdominal migraine is an idiopathic disorder characterized by moderate to severe midline abdominal pain lasting 1–72 hours associated with vasomotor symptoms, nausea, and vomiting. It is recognized by the International Headache Society (IHS) as being among the “periodic syndromes of childhood that are commonly precursors of migraine” (Cephalagia 2004;24:suppl 1:9–160).

Most of the literature on the topic is from Europe, and the diagnosis is far more common there than it is in the United States, where it is largely underdiagnosed, Dr. Laura D. Carson and her associates said in a poster she presented at the annual meeting of the North American Society of Pediatric Gastroenterology, Hepatology, and Nutrition.

In a retrospective chart review of 600 children and young adults (ages 1–21 years, 59% female) who were referred to a pediatric gastroenterologist during 2006–2007 for recurrent abdominal pain, 23.5% (141) were excluded because of a preexisting diagnosis. Of 458 who met inclusion criteria, 4% (20) met the IHS diagnostic criteria for abdominal migraine (see box), while another 11% (50) were considered probable diagnoses of abdominal migraine with documentation lacking for at least one diagnostic criterion. The remaining 85% (388) did not meet the criteria, said Dr. Carson and her associates, of Eastern Virginia Medical School and Children's Hospital of the King's Daughters, both in Norfolk, Va.

No relationship has been identified among those with abdominal migraine and those who had family histories of either abdominal pain or headache. However, children who met the abdominal migraine criteria were four times more likely to have migraine headaches.

Despite its inclusion in both the IHS classification as well as the 2006 Rome III GI Criteria (Gastroenterology 2006;130:1527–37), abdominal migraine is infrequently considered in the differential diagnosis of recurrent abdominal pain in children. Expertise in migraine lies with neurologists, who are rarely called upon to evaluate abdominal pain.

Abdominal migraine occurs only in children, whereas in adulthood it presents to neurologists as classic migraine headache. Children with recurrent abdominal pain often are referred to gastroenterologists, who rule out other organic causes but might not consider migraine as an etiology, Dr. Carson said.

“Given the spectrum of treatment modalities now available for pediatric migraine, increased awareness of cardinal features of abdominal migraine by pediatricians and pediatric gastroenterologists may result in improved diagnostic accuracy and early institution of both acute and preventative migraine-specific treatments,” Dr. Carson and her associates said in their poster.

The study was funded by the Children's Specialty Group Chairman's Fund, based at Children's Hospital of the King's Daughters, Norfolk. Dr. Carson stated that she had no other disclosures.

Criteria in Brief

At least 5 attacks fulfilling criteria B-D.

Attacks of abdominal pain lasting 1–72 hours.

Abdominal pain has all of the following characteristics:

Midline, periumbilical, or poorly localized.

Dull, or “just sore” quality.

Moderate or severe intensity.

During abdominal pain, at least two of the following:

Anorexia.

Nausea.

Vomiting.

Pallor.

Not attributed to another disorder.

Note: ROME criteria are similar, but also specify intervening periods of usual health and include headache and photophobia among the possible symptoms during abdominal pain.

Source: 2004 International Classification of Headache Disorders

NATIONAL HARBOR, MD. — Abdominal migraine may be responsible for up to 15% of all cases of idiopathic recurrent abdominal pain in children, according to an analysis of records from more than 400 children.

Abdominal migraine is an idiopathic disorder characterized by moderate to severe midline abdominal pain lasting 1–72 hours associated with vasomotor symptoms, nausea, and vomiting. It is recognized by the International Headache Society (IHS) as being among the “periodic syndromes of childhood that are commonly precursors of migraine” (Cephalagia 2004;24:suppl 1:9–160).

Most of the literature on the topic is from Europe, and the diagnosis is far more common there than it is in the United States, where it is largely underdiagnosed, Dr. Laura D. Carson and her associates said in a poster she presented at the annual meeting of the North American Society of Pediatric Gastroenterology, Hepatology, and Nutrition.

In a retrospective chart review of 600 children and young adults (ages 1–21 years, 59% female) who were referred to a pediatric gastroenterologist during 2006–2007 for recurrent abdominal pain, 23.5% (141) were excluded because of a preexisting diagnosis. Of 458 who met inclusion criteria, 4% (20) met the IHS diagnostic criteria for abdominal migraine (see box), while another 11% (50) were considered probable diagnoses of abdominal migraine with documentation lacking for at least one diagnostic criterion. The remaining 85% (388) did not meet the criteria, said Dr. Carson and her associates, of Eastern Virginia Medical School and Children's Hospital of the King's Daughters, both in Norfolk, Va.

No relationship has been identified among those with abdominal migraine and those who had family histories of either abdominal pain or headache. However, children who met the abdominal migraine criteria were four times more likely to have migraine headaches.

Despite its inclusion in both the IHS classification as well as the 2006 Rome III GI Criteria (Gastroenterology 2006;130:1527–37), abdominal migraine is infrequently considered in the differential diagnosis of recurrent abdominal pain in children. Expertise in migraine lies with neurologists, who are rarely called upon to evaluate abdominal pain.

Abdominal migraine occurs only in children, whereas in adulthood it presents to neurologists as classic migraine headache. Children with recurrent abdominal pain often are referred to gastroenterologists, who rule out other organic causes but might not consider migraine as an etiology, Dr. Carson said.

“Given the spectrum of treatment modalities now available for pediatric migraine, increased awareness of cardinal features of abdominal migraine by pediatricians and pediatric gastroenterologists may result in improved diagnostic accuracy and early institution of both acute and preventative migraine-specific treatments,” Dr. Carson and her associates said in their poster.

The study was funded by the Children's Specialty Group Chairman's Fund, based at Children's Hospital of the King's Daughters, Norfolk. Dr. Carson stated that she had no other disclosures.

Criteria in Brief

At least 5 attacks fulfilling criteria B-D.

Attacks of abdominal pain lasting 1–72 hours.

Abdominal pain has all of the following characteristics:

Midline, periumbilical, or poorly localized.

Dull, or “just sore” quality.

Moderate or severe intensity.

During abdominal pain, at least two of the following:

Anorexia.

Nausea.

Vomiting.

Pallor.

Not attributed to another disorder.

Note: ROME criteria are similar, but also specify intervening periods of usual health and include headache and photophobia among the possible symptoms during abdominal pain.

Source: 2004 International Classification of Headache Disorders

NATIONAL HARBOR, MD. — Abdominal migraine may be responsible for up to 15% of all cases of idiopathic recurrent abdominal pain in children, according to an analysis of records from more than 400 children.

Abdominal migraine is an idiopathic disorder characterized by moderate to severe midline abdominal pain lasting 1–72 hours associated with vasomotor symptoms, nausea, and vomiting. It is recognized by the International Headache Society (IHS) as being among the “periodic syndromes of childhood that are commonly precursors of migraine” (Cephalagia 2004;24:suppl 1:9–160).

Most of the literature on the topic is from Europe, and the diagnosis is far more common there than it is in the United States, where it is largely underdiagnosed, Dr. Laura D. Carson and her associates said in a poster she presented at the annual meeting of the North American Society of Pediatric Gastroenterology, Hepatology, and Nutrition.

In a retrospective chart review of 600 children and young adults (ages 1–21 years, 59% female) who were referred to a pediatric gastroenterologist during 2006–2007 for recurrent abdominal pain, 23.5% (141) were excluded because of a preexisting diagnosis. Of 458 who met inclusion criteria, 4% (20) met the IHS diagnostic criteria for abdominal migraine (see box), while another 11% (50) were considered probable diagnoses of abdominal migraine with documentation lacking for at least one diagnostic criterion. The remaining 85% (388) did not meet the criteria, said Dr. Carson and her associates, of Eastern Virginia Medical School and Children's Hospital of the King's Daughters, both in Norfolk, Va.

No relationship has been identified among those with abdominal migraine and those who had family histories of either abdominal pain or headache. However, children who met the abdominal migraine criteria were four times more likely to have migraine headaches.

Despite its inclusion in both the IHS classification as well as the 2006 Rome III GI Criteria (Gastroenterology 2006;130:1527–37), abdominal migraine is infrequently considered in the differential diagnosis of recurrent abdominal pain in children. Expertise in migraine lies with neurologists, who are rarely called upon to evaluate abdominal pain.

Abdominal migraine occurs only in children, whereas in adulthood it presents to neurologists as classic migraine headache. Children with recurrent abdominal pain often are referred to gastroenterologists, who rule out other organic causes but might not consider migraine as an etiology, Dr. Carson said.

“Given the spectrum of treatment modalities now available for pediatric migraine, increased awareness of cardinal features of abdominal migraine by pediatricians and pediatric gastroenterologists may result in improved diagnostic accuracy and early institution of both acute and preventative migraine-specific treatments,” Dr. Carson and her associates said in their poster.

The study was funded by the Children's Specialty Group Chairman's Fund, based at Children's Hospital of the King's Daughters, Norfolk. Dr. Carson stated that she had no other disclosures.

Criteria in Brief

At least 5 attacks fulfilling criteria B-D.

Attacks of abdominal pain lasting 1–72 hours.

Abdominal pain has all of the following characteristics:

Midline, periumbilical, or poorly localized.

Dull, or “just sore” quality.

Moderate or severe intensity.

During abdominal pain, at least two of the following:

Anorexia.

Nausea.

Vomiting.

Pallor.

Not attributed to another disorder.

Note: ROME criteria are similar, but also specify intervening periods of usual health and include headache and photophobia among the possible symptoms during abdominal pain.

Source: 2004 International Classification of Headache Disorders

The American Diabetes Association has officially endorsed the use of hemoglobin A_1c as an option for diagnosing diabetes.

In its Standards of Medical Care in Diabetes, updated annually, the ADA for the first time in 2010 is officially endorsing the use of HbA_1c as one of four options for diagnosing diabetes, with a cut-point of 6.5% or greater. Recommendations for use of the three previous diagnostic criteria remain unchanged, including a fasting plasma glucose (FPG) of 126 mg/dL or above, a 2-hour plasma glucose of 200 mg/dL or greater following a 75-g oral glucose tolerance test, or a random plasma glucose of 200 mg/dL or greater in an individual with classic symptoms of hyperglycemia (Diabetes Care 2010[suppl 1]:S11–61 [doi:10.2337/dc10-S011

In June 2009, the use of HbA_1c for diabetes diagnosis was endorsed in a consensus statement by an expert panel comprising members of the American Diabetes Association, the European Association for the Study of Diabetes, and the International Diabetes Federation. However, that statement was not the official position of the respective organizations (Diabetes Care 2009;32:1327–34).

The new ADA endorsement is based in part on the fact that HbA_1c assays are now highly standardized, and “their results can be uniformly applied both temporally and across populations.” In addition, epidemiologic data show a relation between HbA1_c and the risk of retinopathy similar to that shown for corresponding FPG and 2-hour postprandial glucose thresholds. The HbA1_c is also more convenient since fasting is not required, and is likely to be more stable than glucose measurements, the statement said.

The ADA acknowledged that these advantages must be balanced by greater cost, limited availability of HbA_1c in some parts of the developing world, and incomplete correlation between HbA_1c and the average glucose in certain individuals. Also, the HbA_1c can be misleading in patients with certain forms of anemia and hemoglobinopathies. Indeed, unpublished data suggest that use of the HbA_1c with a cutoff of 6.5% or higher identifies one-third fewer cases of undiagnosed diabetes than does a FPG of 126 mg/dL or greater.

However, the ADA said, “in practice, a large portion of the diabetic population remains unaware of their condition. Thus, the lower sensitivity of A_1c at the designated cut-point may well be offset by the test's greater practicality, and wider application of a more convenient test (A_1c) may actually increase the number of diagnoses made.” (See sidebar for diagnostic criteria.)

Not everyone agrees. Dr. Zachary T. Bloomgarden of Mount Sinai School of Medicine, New York, said in an interview that while it may be appropriate to use HbA_1c as a screening tool to determine who would then be asked to return for an oral glucose tolerance test, using it for diagnosis is not appropriate because it could lead to overdiagnosis among people with high hemoglobin glycation, or “high glycators,” and underdiagnosis of “low glycators.”

The ADA's decision to endorse the HbA_1c as a diagnostic tool is “overall, not to my mind satisfactory,” said Dr. Bloomgarden, editor of the Journal of Diabetes.

But Dr. Mayer Davidson, who was part of the expert panel that endorsed HbA_1c for diagnosing diabetes last summer, is on the opposite end of the spectrum. He said the recommendation to use HbA_1c for diabetes diagnosis is long overdue.

“Unfortunately, the ADA kept the glucose criteria, which will lead to the confusing situation of people who have diabetes by one criterion but not by the other when both are measured, which is likely to occur frequently,” said Dr. Davidson, professor of medicine, Charles Drew University and David Geffen School of Medicine at the University of California, Los Angeles.

Based on the expert committee's deliberations, it's likely that the ADA and the other organizations will ultimately transition to use of HbA_1c alone for diagnosis, but it may take time. Until then, he advised that physicians who want to use repeat testing for diagnosis stick to the same test both times to avoid confusion. Bottom line: “One should not intermingle the glucose and A_1c criteria.”

The ADA document says that using the same test is “preferred” but provides specific guidance for both testing scenarios.

Along with the 6.5% cutoff for diabetes diagnosis, the ADA now categorizes patients with HbA_1c levels of 5.7%–6.4% under the new heading “Categories of Increased Risk for Diabetes,” replacing “Diagnosis of Pre-Diabetes.”

The 5.7% threshold was derived from unpublished data suggesting that it has the best combination of sensitivity (39%) and specificity (91%) to identify cases of impaired fasting glucose. Other analyses suggest that an HbA_1c of 5.7% is associated with a diabetes risk similar to that of the high-risk participants in the landmark Diabetes Prevention Program trial.

Other significant changes from the ADA's 2009 Standards of Medical Care include the following:

▸ The section “Antiplatelet agents” has been extensively revised to reflect recent trial data that call into question the benefit of aspirin for primary cardiovascular disease prevention in moderate- or low-risk patients.

▸ The section “Retinopathy screening and treatment” has been updated to include a recommendation on use of fundus photography as a screening strategy.

▸ The section “Diabetes care in the hospital” has been extensively revised to reflect new evidence calling into question very tight glycemic control goals in critically ill patients.

Both Dr. Bloomgarden and Dr. Davidson stated that they have no financial disclosures.

'Unfortunately, the ADA kept the glucose criteria,' so some people may be diabetic by one criterion but not the other.

Source DR. DAVIDSON

The Diagnostic Criteria in Brief

1. Hemoglobin A_1c 6.5% or greater.

OR

2. FPG 126 mg/dL or greater (fasting is defined as no caloric intake for at least 8 hours).

OR

3. Two-hour plasma glucose of 200 mg/dL or greater during an oral glucose tolerance test.

OR

4. In a patient with classic symptoms of hyperglycemia or hyperglycemic crisis, a random plasma glucose of 200 mg/dL or greater.

Source: Diabetes Care 2010 (

doi:10.2337/dc10-S011

The American Diabetes Association has officially endorsed the use of hemoglobin A_1c as an option for diagnosing diabetes.

In its Standards of Medical Care in Diabetes, updated annually, the ADA for the first time in 2010 is officially endorsing the use of HbA_1c as one of four options for diagnosing diabetes, with a cut-point of 6.5% or greater. Recommendations for use of the three previous diagnostic criteria remain unchanged, including a fasting plasma glucose (FPG) of 126 mg/dL or above, a 2-hour plasma glucose of 200 mg/dL or greater following a 75-g oral glucose tolerance test, or a random plasma glucose of 200 mg/dL or greater in an individual with classic symptoms of hyperglycemia (Diabetes Care 2010[suppl 1]:S11–61 [doi:10.2337/dc10-S011

In June 2009, the use of HbA_1c for diabetes diagnosis was endorsed in a consensus statement by an expert panel comprising members of the American Diabetes Association, the European Association for the Study of Diabetes, and the International Diabetes Federation. However, that statement was not the official position of the respective organizations (Diabetes Care 2009;32:1327–34).

The new ADA endorsement is based in part on the fact that HbA_1c assays are now highly standardized, and “their results can be uniformly applied both temporally and across populations.” In addition, epidemiologic data show a relation between HbA1_c and the risk of retinopathy similar to that shown for corresponding FPG and 2-hour postprandial glucose thresholds. The HbA1_c is also more convenient since fasting is not required, and is likely to be more stable than glucose measurements, the statement said.

The ADA acknowledged that these advantages must be balanced by greater cost, limited availability of HbA_1c in some parts of the developing world, and incomplete correlation between HbA_1c and the average glucose in certain individuals. Also, the HbA_1c can be misleading in patients with certain forms of anemia and hemoglobinopathies. Indeed, unpublished data suggest that use of the HbA_1c with a cutoff of 6.5% or higher identifies one-third fewer cases of undiagnosed diabetes than does a FPG of 126 mg/dL or greater.

However, the ADA said, “in practice, a large portion of the diabetic population remains unaware of their condition. Thus, the lower sensitivity of A_1c at the designated cut-point may well be offset by the test's greater practicality, and wider application of a more convenient test (A_1c) may actually increase the number of diagnoses made.” (See sidebar for diagnostic criteria.)

Not everyone agrees. Dr. Zachary T. Bloomgarden of Mount Sinai School of Medicine, New York, said in an interview that while it may be appropriate to use HbA_1c as a screening tool to determine who would then be asked to return for an oral glucose tolerance test, using it for diagnosis is not appropriate because it could lead to overdiagnosis among people with high hemoglobin glycation, or “high glycators,” and underdiagnosis of “low glycators.”

The ADA's decision to endorse the HbA_1c as a diagnostic tool is “overall, not to my mind satisfactory,” said Dr. Bloomgarden, editor of the Journal of Diabetes.

But Dr. Mayer Davidson, who was part of the expert panel that endorsed HbA_1c for diagnosing diabetes last summer, is on the opposite end of the spectrum. He said the recommendation to use HbA_1c for diabetes diagnosis is long overdue.

“Unfortunately, the ADA kept the glucose criteria, which will lead to the confusing situation of people who have diabetes by one criterion but not by the other when both are measured, which is likely to occur frequently,” said Dr. Davidson, professor of medicine, Charles Drew University and David Geffen School of Medicine at the University of California, Los Angeles.

Based on the expert committee's deliberations, it's likely that the ADA and the other organizations will ultimately transition to use of HbA_1c alone for diagnosis, but it may take time. Until then, he advised that physicians who want to use repeat testing for diagnosis stick to the same test both times to avoid confusion. Bottom line: “One should not intermingle the glucose and A_1c criteria.”

The ADA document says that using the same test is “preferred” but provides specific guidance for both testing scenarios.

Along with the 6.5% cutoff for diabetes diagnosis, the ADA now categorizes patients with HbA_1c levels of 5.7%–6.4% under the new heading “Categories of Increased Risk for Diabetes,” replacing “Diagnosis of Pre-Diabetes.”

The 5.7% threshold was derived from unpublished data suggesting that it has the best combination of sensitivity (39%) and specificity (91%) to identify cases of impaired fasting glucose. Other analyses suggest that an HbA_1c of 5.7% is associated with a diabetes risk similar to that of the high-risk participants in the landmark Diabetes Prevention Program trial.

Other significant changes from the ADA's 2009 Standards of Medical Care include the following:

▸ The section “Antiplatelet agents” has been extensively revised to reflect recent trial data that call into question the benefit of aspirin for primary cardiovascular disease prevention in moderate- or low-risk patients.

▸ The section “Retinopathy screening and treatment” has been updated to include a recommendation on use of fundus photography as a screening strategy.

▸ The section “Diabetes care in the hospital” has been extensively revised to reflect new evidence calling into question very tight glycemic control goals in critically ill patients.

Both Dr. Bloomgarden and Dr. Davidson stated that they have no financial disclosures.

'Unfortunately, the ADA kept the glucose criteria,' so some people may be diabetic by one criterion but not the other.

Source DR. DAVIDSON

The Diagnostic Criteria in Brief

1. Hemoglobin A_1c 6.5% or greater.

OR

2. FPG 126 mg/dL or greater (fasting is defined as no caloric intake for at least 8 hours).

OR

3. Two-hour plasma glucose of 200 mg/dL or greater during an oral glucose tolerance test.

OR

4. In a patient with classic symptoms of hyperglycemia or hyperglycemic crisis, a random plasma glucose of 200 mg/dL or greater.

Source: Diabetes Care 2010 (

doi:10.2337/dc10-S011

The American Diabetes Association has officially endorsed the use of hemoglobin A_1c as an option for diagnosing diabetes.

In its Standards of Medical Care in Diabetes, updated annually, the ADA for the first time in 2010 is officially endorsing the use of HbA_1c as one of four options for diagnosing diabetes, with a cut-point of 6.5% or greater. Recommendations for use of the three previous diagnostic criteria remain unchanged, including a fasting plasma glucose (FPG) of 126 mg/dL or above, a 2-hour plasma glucose of 200 mg/dL or greater following a 75-g oral glucose tolerance test, or a random plasma glucose of 200 mg/dL or greater in an individual with classic symptoms of hyperglycemia (Diabetes Care 2010[suppl 1]:S11–61 [doi:10.2337/dc10-S011

In June 2009, the use of HbA_1c for diabetes diagnosis was endorsed in a consensus statement by an expert panel comprising members of the American Diabetes Association, the European Association for the Study of Diabetes, and the International Diabetes Federation. However, that statement was not the official position of the respective organizations (Diabetes Care 2009;32:1327–34).

The new ADA endorsement is based in part on the fact that HbA_1c assays are now highly standardized, and “their results can be uniformly applied both temporally and across populations.” In addition, epidemiologic data show a relation between HbA1_c and the risk of retinopathy similar to that shown for corresponding FPG and 2-hour postprandial glucose thresholds. The HbA1_c is also more convenient since fasting is not required, and is likely to be more stable than glucose measurements, the statement said.

The ADA acknowledged that these advantages must be balanced by greater cost, limited availability of HbA_1c in some parts of the developing world, and incomplete correlation between HbA_1c and the average glucose in certain individuals. Also, the HbA_1c can be misleading in patients with certain forms of anemia and hemoglobinopathies. Indeed, unpublished data suggest that use of the HbA_1c with a cutoff of 6.5% or higher identifies one-third fewer cases of undiagnosed diabetes than does a FPG of 126 mg/dL or greater.

However, the ADA said, “in practice, a large portion of the diabetic population remains unaware of their condition. Thus, the lower sensitivity of A_1c at the designated cut-point may well be offset by the test's greater practicality, and wider application of a more convenient test (A_1c) may actually increase the number of diagnoses made.” (See sidebar for diagnostic criteria.)

Not everyone agrees. Dr. Zachary T. Bloomgarden of Mount Sinai School of Medicine, New York, said in an interview that while it may be appropriate to use HbA_1c as a screening tool to determine who would then be asked to return for an oral glucose tolerance test, using it for diagnosis is not appropriate because it could lead to overdiagnosis among people with high hemoglobin glycation, or “high glycators,” and underdiagnosis of “low glycators.”

The ADA's decision to endorse the HbA_1c as a diagnostic tool is “overall, not to my mind satisfactory,” said Dr. Bloomgarden, editor of the Journal of Diabetes.

But Dr. Mayer Davidson, who was part of the expert panel that endorsed HbA_1c for diagnosing diabetes last summer, is on the opposite end of the spectrum. He said the recommendation to use HbA_1c for diabetes diagnosis is long overdue.

“Unfortunately, the ADA kept the glucose criteria, which will lead to the confusing situation of people who have diabetes by one criterion but not by the other when both are measured, which is likely to occur frequently,” said Dr. Davidson, professor of medicine, Charles Drew University and David Geffen School of Medicine at the University of California, Los Angeles.

Based on the expert committee's deliberations, it's likely that the ADA and the other organizations will ultimately transition to use of HbA_1c alone for diagnosis, but it may take time. Until then, he advised that physicians who want to use repeat testing for diagnosis stick to the same test both times to avoid confusion. Bottom line: “One should not intermingle the glucose and A_1c criteria.”

The ADA document says that using the same test is “preferred” but provides specific guidance for both testing scenarios.

Along with the 6.5% cutoff for diabetes diagnosis, the ADA now categorizes patients with HbA_1c levels of 5.7%–6.4% under the new heading “Categories of Increased Risk for Diabetes,” replacing “Diagnosis of Pre-Diabetes.”

The 5.7% threshold was derived from unpublished data suggesting that it has the best combination of sensitivity (39%) and specificity (91%) to identify cases of impaired fasting glucose. Other analyses suggest that an HbA_1c of 5.7% is associated with a diabetes risk similar to that of the high-risk participants in the landmark Diabetes Prevention Program trial.

Other significant changes from the ADA's 2009 Standards of Medical Care include the following:

▸ The section “Antiplatelet agents” has been extensively revised to reflect recent trial data that call into question the benefit of aspirin for primary cardiovascular disease prevention in moderate- or low-risk patients.

▸ The section “Retinopathy screening and treatment” has been updated to include a recommendation on use of fundus photography as a screening strategy.

▸ The section “Diabetes care in the hospital” has been extensively revised to reflect new evidence calling into question very tight glycemic control goals in critically ill patients.

Both Dr. Bloomgarden and Dr. Davidson stated that they have no financial disclosures.

'Unfortunately, the ADA kept the glucose criteria,' so some people may be diabetic by one criterion but not the other.

Source DR. DAVIDSON

The Diagnostic Criteria in Brief

1. Hemoglobin A_1c 6.5% or greater.

OR

2. FPG 126 mg/dL or greater (fasting is defined as no caloric intake for at least 8 hours).

OR

3. Two-hour plasma glucose of 200 mg/dL or greater during an oral glucose tolerance test.

OR

4. In a patient with classic symptoms of hyperglycemia or hyperglycemic crisis, a random plasma glucose of 200 mg/dL or greater.

Source: Diabetes Care 2010 (

doi:10.2337/dc10-S011

NATIONAL HARBOR, MD. — Hospitalized children and adolescents with inflammatory bowel disease had more than twice the risk for thrombotic events, compared with other hospitalized youth, in a retrospective cohort study that utilized a nationwide inpatient database.

Compared with the entire discharge population, the odds ratio for any thrombotic event among children and adolescents with inflammatory bowel disease (IBD) was 2.13. For those with ulcerative colitis, it was 1.72, and for Crohn's disease, it was 2.22. Among the individual events in all youth with IBD, the highest odds ratios were for Budd-Chiari syndrome (3.21), portal vein thrombosis (2.79), thrombophlebitis (2.75), and deep vein thrombosis (2.44).

IBD is associated with an increased risk for venous and arterial thrombosis in adults, but the risk of thrombosis in children with IBD has not been established previously, Dr. Cade M. Nylund and Dr. Lee A. Denson said in a poster at the annual meeting of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition.

Data were obtained from the Healthcare Cost and Utilization Project Kids' Inpatient Database—the first pediatric-specific inpatient database in the country—for the combined years 1997, 2000, 2003, and 2006 among children and teens aged 6–20 years. From a total weighted sample of 8,162,120 discharges, there were 34,298 thrombotic events and 49,280 children with IBD. Of the latter, 44,039 had Crohn's disease and 5,241 had ulcerative colitis, said Dr. Nylund and Dr. Denson of Cincinnati Children's Hospital Medical Center.

Thrombotic events were reported in 553 of the young people with IBD. They had a median age of 17 years (range 15–19), and a median length of stay of 8 days (5–15 days). More than half (55%) were male, 70% were white, and 17% were black. About two-thirds (68%) had peripherally inserted central catheter (PICC) lines, and 20% underwent surgery.

The overall absolute risk for any thrombotic event was 112.4/10,000 hospitalized IBD patients, compared with 44.5/10,000 for discharges overall. The rate of all thrombotic events was 87.8/10,000 for ulcerative colitis patients and 115.4/10,000 for Crohn's disease patients. IBD patients had significantly increased rates of deep vein thrombosis, thrombophlebitis, pulmonary embolism, portal vein thrombus, and Budd-Chiari syndrome, but not of cerebral vascular disease or arterial thrombus.

Demographic risk factors that significantly increased the risk for thrombotic events were the presence of a PICC line (3.6), age greater than 15 years (1.8), black vs. white race (1.4), and other races vs. white (1.3). Female gender was protective (0.49) among those with IBD, while surgery was not a risk factor, Dr. Nylund and Dr. Denson reported.

Dr. Nylund noted in an interview that the findings do not justify universal pharmacologic prophylaxis in hospitalized children and adolescents with IBD who have rectal bleeding, given the low absolute risk of thrombotic events in young people.

“However, in those more severe IBD children—such as those with prolonged hospitalization length of stay, and those requiring PICC lines—physicians should be aware of this increased risk of thrombotic events and seriously consider initiation of pharmacologic prophylaxis anticoagulation therapy,” he said. “Despite lacking evidence in the pediatric population for thromboembolism prevention techniques, conservative thrombotic prophylaxis such as frequent mobilization and pneumatic compression should at least be considered in all hospitalized children with IBD.”

Disclosures: Dr. Nylund and Dr. Denson had no conflicts of interest to report.

NATIONAL HARBOR, MD. — Hospitalized children and adolescents with inflammatory bowel disease had more than twice the risk for thrombotic events, compared with other hospitalized youth, in a retrospective cohort study that utilized a nationwide inpatient database.

Compared with the entire discharge population, the odds ratio for any thrombotic event among children and adolescents with inflammatory bowel disease (IBD) was 2.13. For those with ulcerative colitis, it was 1.72, and for Crohn's disease, it was 2.22. Among the individual events in all youth with IBD, the highest odds ratios were for Budd-Chiari syndrome (3.21), portal vein thrombosis (2.79), thrombophlebitis (2.75), and deep vein thrombosis (2.44).

IBD is associated with an increased risk for venous and arterial thrombosis in adults, but the risk of thrombosis in children with IBD has not been established previously, Dr. Cade M. Nylund and Dr. Lee A. Denson said in a poster at the annual meeting of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition.

Data were obtained from the Healthcare Cost and Utilization Project Kids' Inpatient Database—the first pediatric-specific inpatient database in the country—for the combined years 1997, 2000, 2003, and 2006 among children and teens aged 6–20 years. From a total weighted sample of 8,162,120 discharges, there were 34,298 thrombotic events and 49,280 children with IBD. Of the latter, 44,039 had Crohn's disease and 5,241 had ulcerative colitis, said Dr. Nylund and Dr. Denson of Cincinnati Children's Hospital Medical Center.

Thrombotic events were reported in 553 of the young people with IBD. They had a median age of 17 years (range 15–19), and a median length of stay of 8 days (5–15 days). More than half (55%) were male, 70% were white, and 17% were black. About two-thirds (68%) had peripherally inserted central catheter (PICC) lines, and 20% underwent surgery.

The overall absolute risk for any thrombotic event was 112.4/10,000 hospitalized IBD patients, compared with 44.5/10,000 for discharges overall. The rate of all thrombotic events was 87.8/10,000 for ulcerative colitis patients and 115.4/10,000 for Crohn's disease patients. IBD patients had significantly increased rates of deep vein thrombosis, thrombophlebitis, pulmonary embolism, portal vein thrombus, and Budd-Chiari syndrome, but not of cerebral vascular disease or arterial thrombus.

Demographic risk factors that significantly increased the risk for thrombotic events were the presence of a PICC line (3.6), age greater than 15 years (1.8), black vs. white race (1.4), and other races vs. white (1.3). Female gender was protective (0.49) among those with IBD, while surgery was not a risk factor, Dr. Nylund and Dr. Denson reported.

Dr. Nylund noted in an interview that the findings do not justify universal pharmacologic prophylaxis in hospitalized children and adolescents with IBD who have rectal bleeding, given the low absolute risk of thrombotic events in young people.

“However, in those more severe IBD children—such as those with prolonged hospitalization length of stay, and those requiring PICC lines—physicians should be aware of this increased risk of thrombotic events and seriously consider initiation of pharmacologic prophylaxis anticoagulation therapy,” he said. “Despite lacking evidence in the pediatric population for thromboembolism prevention techniques, conservative thrombotic prophylaxis such as frequent mobilization and pneumatic compression should at least be considered in all hospitalized children with IBD.”

Disclosures: Dr. Nylund and Dr. Denson had no conflicts of interest to report.

NATIONAL HARBOR, MD. — Hospitalized children and adolescents with inflammatory bowel disease had more than twice the risk for thrombotic events, compared with other hospitalized youth, in a retrospective cohort study that utilized a nationwide inpatient database.

Compared with the entire discharge population, the odds ratio for any thrombotic event among children and adolescents with inflammatory bowel disease (IBD) was 2.13. For those with ulcerative colitis, it was 1.72, and for Crohn's disease, it was 2.22. Among the individual events in all youth with IBD, the highest odds ratios were for Budd-Chiari syndrome (3.21), portal vein thrombosis (2.79), thrombophlebitis (2.75), and deep vein thrombosis (2.44).

IBD is associated with an increased risk for venous and arterial thrombosis in adults, but the risk of thrombosis in children with IBD has not been established previously, Dr. Cade M. Nylund and Dr. Lee A. Denson said in a poster at the annual meeting of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition.

Data were obtained from the Healthcare Cost and Utilization Project Kids' Inpatient Database—the first pediatric-specific inpatient database in the country—for the combined years 1997, 2000, 2003, and 2006 among children and teens aged 6–20 years. From a total weighted sample of 8,162,120 discharges, there were 34,298 thrombotic events and 49,280 children with IBD. Of the latter, 44,039 had Crohn's disease and 5,241 had ulcerative colitis, said Dr. Nylund and Dr. Denson of Cincinnati Children's Hospital Medical Center.

Thrombotic events were reported in 553 of the young people with IBD. They had a median age of 17 years (range 15–19), and a median length of stay of 8 days (5–15 days). More than half (55%) were male, 70% were white, and 17% were black. About two-thirds (68%) had peripherally inserted central catheter (PICC) lines, and 20% underwent surgery.

The overall absolute risk for any thrombotic event was 112.4/10,000 hospitalized IBD patients, compared with 44.5/10,000 for discharges overall. The rate of all thrombotic events was 87.8/10,000 for ulcerative colitis patients and 115.4/10,000 for Crohn's disease patients. IBD patients had significantly increased rates of deep vein thrombosis, thrombophlebitis, pulmonary embolism, portal vein thrombus, and Budd-Chiari syndrome, but not of cerebral vascular disease or arterial thrombus.

Demographic risk factors that significantly increased the risk for thrombotic events were the presence of a PICC line (3.6), age greater than 15 years (1.8), black vs. white race (1.4), and other races vs. white (1.3). Female gender was protective (0.49) among those with IBD, while surgery was not a risk factor, Dr. Nylund and Dr. Denson reported.

Dr. Nylund noted in an interview that the findings do not justify universal pharmacologic prophylaxis in hospitalized children and adolescents with IBD who have rectal bleeding, given the low absolute risk of thrombotic events in young people.

“However, in those more severe IBD children—such as those with prolonged hospitalization length of stay, and those requiring PICC lines—physicians should be aware of this increased risk of thrombotic events and seriously consider initiation of pharmacologic prophylaxis anticoagulation therapy,” he said. “Despite lacking evidence in the pediatric population for thromboembolism prevention techniques, conservative thrombotic prophylaxis such as frequent mobilization and pneumatic compression should at least be considered in all hospitalized children with IBD.”

Disclosures: Dr. Nylund and Dr. Denson had no conflicts of interest to report.