Clinical

HONOLULU – Administering glyceryl trinitrate (GTN) early after onset of ischemic stroke or transient ischemic attack (TIA) does not improve outcomes, according to data presented at the International Stroke Conference sponsored by the American Heart Association. Results suggest that GTN causes adverse effects in patients with intracerebral hemorrhage (ICH), but this observation is not definitive, according to the researchers. Study results were published online ahead of print Feb. 6 in the Lancet.

Nitric oxide is a regulatory molecule that has vasoactive effects and promotes blood pressure reduction. Vascular levels of nitric oxide are low in stroke, which suggests that the molecule may be a target for stroke treatment. GTN, a nitric oxide donor, lowered blood pressure and improved functional outcome among patients with acute stroke in the phase 2 Rapid Intervention with GTN in Hypertensive Stroke Trial (RIGHT).

Philip Bath, MD, Stroke Association Professor of Stroke Medicine at the University of Nottingham (England), and colleagues conducted the RIGHT-2 study to evaluate the safety and efficacy of GTN when administered early after onset of suspected stroke. Paramedics randomized patients in equal groups to a GTN patch or a sham patch in the ambulance. Three more patches were administered in the hospital on the following days. Active and sham patches looked similar and had no writing on them, thus ensuring effective blinding upon administration. Investigators followed up patients by telephone at 90 days to assess the modified Rankin Scale score and markers of disability, mood, cognition, and quality of life.

Eligible participants were adults who had dialed emergency services, independently or with assistance, because of a possible stroke. They had a Face, Arm, Speech, Time (FAST) score of 2 or 3, were within 4 hours of onset, and had a systolic blood pressure greater than 120 mm Hg. Patients from nursing homes, those with hypoglycemia, those who were unconscious, and those with a witnessed seizure were excluded.

Dr. Bath and colleagues planned to enroll 850 patients from five ambulance services in 30 hospitals across the United Kingdom. Data were to be examined through an intention-to-treat analysis. During the trial, however, the investigators observed that the rate of stroke mimics was 26%, rather than the 12% that they had anticipated. To ensure the proper power for the study, the investigators increased the sample size to 1,149 patients. They also changed the planned data analysis from intention-to-treat to hierarchical analysis. Specifically, the researchers planned to perform the primary analysis in patients with stroke or TIA. If the results were positive, then they would perform a standard intention-to-treat analysis.

More than 99% of patients received the first patch. Approximately 57% of the population received the first two patches. One reason for this decrease in adherence was that many patients were discharged from the hospital with a TIA or a stroke mimic. Participants’ average age was 72. The median time from onset to randomization was 71 minutes, and the median time to treatment was 73 minutes. Participants’ mean systolic blood pressure was 162 mm Hg. Approximately 60% of the patients had a FAST score of 3. About 50% of participants had ischemic stroke, 13% had ICH, 10% had TIA, and 26% had stroke mimics.

At 1 hour after treatment initiation, systolic blood pressure decreased by 6.2 mm Hg and diastolic blood pressure decreased by 2.7 mm Hg among patients who received GTN, compared with controls. At one day, the differences were 5.2 mm Hg and 2.5 mm Hg, respectively, in treated patients, compared with controls. Blood pressure became similar between groups thereafter, “in part because of the tachyphylaxis that we know happens with GTN,” said Dr. Bath.

The researchers found no evidence of an effect of GTN on functional outcome at 90 days in participants with stroke or transient ischemic attack. The adjusted common odds ratio of poor outcome was 1.25 in the GTN group, compared with the control group (95 % confidence interval, 0.97-1.60; P = .083). “We were close to getting a negative trial,” said Dr. Bath.

Subgroup analyses revealed differences in outcome according to the time to randomization. GTN had a negative effect in patients treated within 1 hour of onset. Results were neutral, but tended to be negative, in patients treated between 1 and 2 hours of onset. Results were neutral, but tended to be positive, among patients treated at more than 2 hours after onset. There was no difference between groups in the rate of mortality.

One of the study’s limitations was its single-blind design. In addition, the trial was conducted in a single country, and the investigators changed the protocol after it was initiated. “We had a higher-than-expected [stroke] mimic rate, although I’m reassured by most experts that ... this is probably about right,” said Dr. Bath.

A potential reason for the neutral results is the negative effect that GTN had among patients with ICH, said Dr. Bath. “In that very early first hour, we are of course breaking a law that we learned in medical school, which is that the first part of hemostasis is spasm. We gave an antispasmodic: a vasodilator,” he added. “That is speculation.”

The trial was funded by the British Heart Foundation. Dr. Bath declared a modest ownership interest in Platelet Solutions and consultant or advisory board positions with Moleac, DiaMedica, Phagenesis, Nestle, and ReNeuron. The other investigators declared no conflicts of interest.

[email protected]

SOURCE: Bath PM et al. ISC 2019, Abstract LB2.

HONOLULU – Administering glyceryl trinitrate (GTN) early after onset of ischemic stroke or transient ischemic attack (TIA) does not improve outcomes, according to data presented at the International Stroke Conference sponsored by the American Heart Association. Results suggest that GTN causes adverse effects in patients with intracerebral hemorrhage (ICH), but this observation is not definitive, according to the researchers. Study results were published online ahead of print Feb. 6 in the Lancet.

Nitric oxide is a regulatory molecule that has vasoactive effects and promotes blood pressure reduction. Vascular levels of nitric oxide are low in stroke, which suggests that the molecule may be a target for stroke treatment. GTN, a nitric oxide donor, lowered blood pressure and improved functional outcome among patients with acute stroke in the phase 2 Rapid Intervention with GTN in Hypertensive Stroke Trial (RIGHT).

Philip Bath, MD, Stroke Association Professor of Stroke Medicine at the University of Nottingham (England), and colleagues conducted the RIGHT-2 study to evaluate the safety and efficacy of GTN when administered early after onset of suspected stroke. Paramedics randomized patients in equal groups to a GTN patch or a sham patch in the ambulance. Three more patches were administered in the hospital on the following days. Active and sham patches looked similar and had no writing on them, thus ensuring effective blinding upon administration. Investigators followed up patients by telephone at 90 days to assess the modified Rankin Scale score and markers of disability, mood, cognition, and quality of life.

Eligible participants were adults who had dialed emergency services, independently or with assistance, because of a possible stroke. They had a Face, Arm, Speech, Time (FAST) score of 2 or 3, were within 4 hours of onset, and had a systolic blood pressure greater than 120 mm Hg. Patients from nursing homes, those with hypoglycemia, those who were unconscious, and those with a witnessed seizure were excluded.

Dr. Bath and colleagues planned to enroll 850 patients from five ambulance services in 30 hospitals across the United Kingdom. Data were to be examined through an intention-to-treat analysis. During the trial, however, the investigators observed that the rate of stroke mimics was 26%, rather than the 12% that they had anticipated. To ensure the proper power for the study, the investigators increased the sample size to 1,149 patients. They also changed the planned data analysis from intention-to-treat to hierarchical analysis. Specifically, the researchers planned to perform the primary analysis in patients with stroke or TIA. If the results were positive, then they would perform a standard intention-to-treat analysis.

More than 99% of patients received the first patch. Approximately 57% of the population received the first two patches. One reason for this decrease in adherence was that many patients were discharged from the hospital with a TIA or a stroke mimic. Participants’ average age was 72. The median time from onset to randomization was 71 minutes, and the median time to treatment was 73 minutes. Participants’ mean systolic blood pressure was 162 mm Hg. Approximately 60% of the patients had a FAST score of 3. About 50% of participants had ischemic stroke, 13% had ICH, 10% had TIA, and 26% had stroke mimics.

At 1 hour after treatment initiation, systolic blood pressure decreased by 6.2 mm Hg and diastolic blood pressure decreased by 2.7 mm Hg among patients who received GTN, compared with controls. At one day, the differences were 5.2 mm Hg and 2.5 mm Hg, respectively, in treated patients, compared with controls. Blood pressure became similar between groups thereafter, “in part because of the tachyphylaxis that we know happens with GTN,” said Dr. Bath.

The researchers found no evidence of an effect of GTN on functional outcome at 90 days in participants with stroke or transient ischemic attack. The adjusted common odds ratio of poor outcome was 1.25 in the GTN group, compared with the control group (95 % confidence interval, 0.97-1.60; P = .083). “We were close to getting a negative trial,” said Dr. Bath.

Subgroup analyses revealed differences in outcome according to the time to randomization. GTN had a negative effect in patients treated within 1 hour of onset. Results were neutral, but tended to be negative, in patients treated between 1 and 2 hours of onset. Results were neutral, but tended to be positive, among patients treated at more than 2 hours after onset. There was no difference between groups in the rate of mortality.

One of the study’s limitations was its single-blind design. In addition, the trial was conducted in a single country, and the investigators changed the protocol after it was initiated. “We had a higher-than-expected [stroke] mimic rate, although I’m reassured by most experts that ... this is probably about right,” said Dr. Bath.

A potential reason for the neutral results is the negative effect that GTN had among patients with ICH, said Dr. Bath. “In that very early first hour, we are of course breaking a law that we learned in medical school, which is that the first part of hemostasis is spasm. We gave an antispasmodic: a vasodilator,” he added. “That is speculation.”

The trial was funded by the British Heart Foundation. Dr. Bath declared a modest ownership interest in Platelet Solutions and consultant or advisory board positions with Moleac, DiaMedica, Phagenesis, Nestle, and ReNeuron. The other investigators declared no conflicts of interest.

[email protected]

SOURCE: Bath PM et al. ISC 2019, Abstract LB2.

HONOLULU – Administering glyceryl trinitrate (GTN) early after onset of ischemic stroke or transient ischemic attack (TIA) does not improve outcomes, according to data presented at the International Stroke Conference sponsored by the American Heart Association. Results suggest that GTN causes adverse effects in patients with intracerebral hemorrhage (ICH), but this observation is not definitive, according to the researchers. Study results were published online ahead of print Feb. 6 in the Lancet.

Nitric oxide is a regulatory molecule that has vasoactive effects and promotes blood pressure reduction. Vascular levels of nitric oxide are low in stroke, which suggests that the molecule may be a target for stroke treatment. GTN, a nitric oxide donor, lowered blood pressure and improved functional outcome among patients with acute stroke in the phase 2 Rapid Intervention with GTN in Hypertensive Stroke Trial (RIGHT).

Philip Bath, MD, Stroke Association Professor of Stroke Medicine at the University of Nottingham (England), and colleagues conducted the RIGHT-2 study to evaluate the safety and efficacy of GTN when administered early after onset of suspected stroke. Paramedics randomized patients in equal groups to a GTN patch or a sham patch in the ambulance. Three more patches were administered in the hospital on the following days. Active and sham patches looked similar and had no writing on them, thus ensuring effective blinding upon administration. Investigators followed up patients by telephone at 90 days to assess the modified Rankin Scale score and markers of disability, mood, cognition, and quality of life.

Eligible participants were adults who had dialed emergency services, independently or with assistance, because of a possible stroke. They had a Face, Arm, Speech, Time (FAST) score of 2 or 3, were within 4 hours of onset, and had a systolic blood pressure greater than 120 mm Hg. Patients from nursing homes, those with hypoglycemia, those who were unconscious, and those with a witnessed seizure were excluded.

Dr. Bath and colleagues planned to enroll 850 patients from five ambulance services in 30 hospitals across the United Kingdom. Data were to be examined through an intention-to-treat analysis. During the trial, however, the investigators observed that the rate of stroke mimics was 26%, rather than the 12% that they had anticipated. To ensure the proper power for the study, the investigators increased the sample size to 1,149 patients. They also changed the planned data analysis from intention-to-treat to hierarchical analysis. Specifically, the researchers planned to perform the primary analysis in patients with stroke or TIA. If the results were positive, then they would perform a standard intention-to-treat analysis.

More than 99% of patients received the first patch. Approximately 57% of the population received the first two patches. One reason for this decrease in adherence was that many patients were discharged from the hospital with a TIA or a stroke mimic. Participants’ average age was 72. The median time from onset to randomization was 71 minutes, and the median time to treatment was 73 minutes. Participants’ mean systolic blood pressure was 162 mm Hg. Approximately 60% of the patients had a FAST score of 3. About 50% of participants had ischemic stroke, 13% had ICH, 10% had TIA, and 26% had stroke mimics.

At 1 hour after treatment initiation, systolic blood pressure decreased by 6.2 mm Hg and diastolic blood pressure decreased by 2.7 mm Hg among patients who received GTN, compared with controls. At one day, the differences were 5.2 mm Hg and 2.5 mm Hg, respectively, in treated patients, compared with controls. Blood pressure became similar between groups thereafter, “in part because of the tachyphylaxis that we know happens with GTN,” said Dr. Bath.

The researchers found no evidence of an effect of GTN on functional outcome at 90 days in participants with stroke or transient ischemic attack. The adjusted common odds ratio of poor outcome was 1.25 in the GTN group, compared with the control group (95 % confidence interval, 0.97-1.60; P = .083). “We were close to getting a negative trial,” said Dr. Bath.

Subgroup analyses revealed differences in outcome according to the time to randomization. GTN had a negative effect in patients treated within 1 hour of onset. Results were neutral, but tended to be negative, in patients treated between 1 and 2 hours of onset. Results were neutral, but tended to be positive, among patients treated at more than 2 hours after onset. There was no difference between groups in the rate of mortality.

One of the study’s limitations was its single-blind design. In addition, the trial was conducted in a single country, and the investigators changed the protocol after it was initiated. “We had a higher-than-expected [stroke] mimic rate, although I’m reassured by most experts that ... this is probably about right,” said Dr. Bath.

A potential reason for the neutral results is the negative effect that GTN had among patients with ICH, said Dr. Bath. “In that very early first hour, we are of course breaking a law that we learned in medical school, which is that the first part of hemostasis is spasm. We gave an antispasmodic: a vasodilator,” he added. “That is speculation.”

The trial was funded by the British Heart Foundation. Dr. Bath declared a modest ownership interest in Platelet Solutions and consultant or advisory board positions with Moleac, DiaMedica, Phagenesis, Nestle, and ReNeuron. The other investigators declared no conflicts of interest.

[email protected]

SOURCE: Bath PM et al. ISC 2019, Abstract LB2.

SAN DIEGO – Powerful antibiotics come first to mind when hospitalized patients have sepsis, but a critical care pulmonology specialist urged colleagues to consider another treatment – heavy intravenous doses of vitamin C.

“There is evidence supporting benefit, and ample evidence supporting safety,” Michael H. Hooper, MD, who practices in Norfolk, Va., said in a pro-and-con debate over the use of vitamin C in sepsis at the Critical Care Congress sponsored by the Society of Critical Care Medicine.

Dr. Hooper’s debate opponent countered by noting the lack of quality research into vitamin C in sepsis and declared that its time has not yet come. “We need more data to know the safety of this drug,” said Andre Kalil, MD, professor of internal medicine and director of Transplant Infectious Diseases at the University of Nebraska Medical Center, Omaha.

Dr. Hooper was part of a member of a team led by Paul E. Marik, MD, FCCP, of Eastern Virginia Medical School, Norfolk, that made waves in 2017 with a study in Chest suggesting IV vitamin C has tremendous potential as a treatment for sepsis (Chest. 2017 Jun;151[6]:1229-38).

The retrospective study compared two groups of 47 patients with sepsis – a control group and a group that received treatment with intravenous vitamin C, hydrocortisone, and thiamine. Remarkably, the team found that 9% (4 of 47) of those in the treatment group died in the hospital, compared with 40% (19 of 47) in the control group (P less than .001).

The findings make sense, Dr. Hooper said, in light of the fact that “our patients are remarkably deficient” in vitamin C. He pointed to a 2017 study that found nearly 40% of 24 patients with septic shock were deficient in vitamin C – despite getting recommended enteral nutrition, parenteral nutrition or both – compared with 25% of patients who were not septic. The study authors believe the difference is probably due to “increased metabolism due to the enhanced inflammatory response observed in septic shock” (Crit Care. 2017 Dec 11;21[1]:300).

“We’re dealing with a population of patients who need some sort of repletion of this vitamin,” Dr. Hooper said.

Why not try oral administration of vitamin C? “Oral administration at regular doses doesn’t work,” he said. “If you have normal volunteers who are made deficient, then you administer the recommended allowance, it takes days or weeks to return levels to normal.”

Dr. Hooper added that the goal of vitamin C therapy isn’t simply to restore proper levels in plasma. In addition, he said, “we’re trying to restore levels in crucial organs.”

He said the cost of treatment with IV vitamin C is low, and no serious adverse events have been seen in studies of the vitamin’s use in critical care.

In his comments at the debate, Dr. Kalil pointed to several weaknesses in the 2017 study of vitamin C in sepsis. According to him, it had many problems, including a sample size that lacked statistical power and imbalances in the two groups. He raised concerns about the study in a 2017 letter published in Chest titled “Vitamin C Is Not Ready for Prime Time in Sepsis but a Solution Is Close,” noting that the control group was sicker and none of those patients had their vitamin C levels measured (Chest. 2017 Sep;152[3]:676).

Randy Dotinga/MDedge News

He added that “acute renal failure is associated with high doses of vitamin C.”

As of July 2018, several clinical trials into vitamin C, hydrocortisone, and thiamine for the treatment of septic shock were underway or planned, according to a report that described the current randomized, placebo-controlled, multicenter Ascorbic Acid, Corticosteroids, and Thiamine in Sepsis (ACTS) trial in the United States. The report notes that “robust evidence” for this approach is lacking, although “the potential effectiveness of this medication combination is rooted in biologic plausibility and supported by small clinical trials of the various individual components.” (Crit Care. 2018;22:283)

Dr. Hooper is an executive committee member and principal investigator with the Vitamin C, Thiamine And Steroids in Sepsis (VICTAS) study. Dr. Kalil reports no relevant disclosures.

SAN DIEGO – Powerful antibiotics come first to mind when hospitalized patients have sepsis, but a critical care pulmonology specialist urged colleagues to consider another treatment – heavy intravenous doses of vitamin C.

“There is evidence supporting benefit, and ample evidence supporting safety,” Michael H. Hooper, MD, who practices in Norfolk, Va., said in a pro-and-con debate over the use of vitamin C in sepsis at the Critical Care Congress sponsored by the Society of Critical Care Medicine.

Dr. Hooper’s debate opponent countered by noting the lack of quality research into vitamin C in sepsis and declared that its time has not yet come. “We need more data to know the safety of this drug,” said Andre Kalil, MD, professor of internal medicine and director of Transplant Infectious Diseases at the University of Nebraska Medical Center, Omaha.

Dr. Hooper was part of a member of a team led by Paul E. Marik, MD, FCCP, of Eastern Virginia Medical School, Norfolk, that made waves in 2017 with a study in Chest suggesting IV vitamin C has tremendous potential as a treatment for sepsis (Chest. 2017 Jun;151[6]:1229-38).

The retrospective study compared two groups of 47 patients with sepsis – a control group and a group that received treatment with intravenous vitamin C, hydrocortisone, and thiamine. Remarkably, the team found that 9% (4 of 47) of those in the treatment group died in the hospital, compared with 40% (19 of 47) in the control group (P less than .001).

The findings make sense, Dr. Hooper said, in light of the fact that “our patients are remarkably deficient” in vitamin C. He pointed to a 2017 study that found nearly 40% of 24 patients with septic shock were deficient in vitamin C – despite getting recommended enteral nutrition, parenteral nutrition or both – compared with 25% of patients who were not septic. The study authors believe the difference is probably due to “increased metabolism due to the enhanced inflammatory response observed in septic shock” (Crit Care. 2017 Dec 11;21[1]:300).

“We’re dealing with a population of patients who need some sort of repletion of this vitamin,” Dr. Hooper said.

Why not try oral administration of vitamin C? “Oral administration at regular doses doesn’t work,” he said. “If you have normal volunteers who are made deficient, then you administer the recommended allowance, it takes days or weeks to return levels to normal.”

Dr. Hooper added that the goal of vitamin C therapy isn’t simply to restore proper levels in plasma. In addition, he said, “we’re trying to restore levels in crucial organs.”

He said the cost of treatment with IV vitamin C is low, and no serious adverse events have been seen in studies of the vitamin’s use in critical care.

In his comments at the debate, Dr. Kalil pointed to several weaknesses in the 2017 study of vitamin C in sepsis. According to him, it had many problems, including a sample size that lacked statistical power and imbalances in the two groups. He raised concerns about the study in a 2017 letter published in Chest titled “Vitamin C Is Not Ready for Prime Time in Sepsis but a Solution Is Close,” noting that the control group was sicker and none of those patients had their vitamin C levels measured (Chest. 2017 Sep;152[3]:676).

Randy Dotinga/MDedge News

He added that “acute renal failure is associated with high doses of vitamin C.”

As of July 2018, several clinical trials into vitamin C, hydrocortisone, and thiamine for the treatment of septic shock were underway or planned, according to a report that described the current randomized, placebo-controlled, multicenter Ascorbic Acid, Corticosteroids, and Thiamine in Sepsis (ACTS) trial in the United States. The report notes that “robust evidence” for this approach is lacking, although “the potential effectiveness of this medication combination is rooted in biologic plausibility and supported by small clinical trials of the various individual components.” (Crit Care. 2018;22:283)

Dr. Hooper is an executive committee member and principal investigator with the Vitamin C, Thiamine And Steroids in Sepsis (VICTAS) study. Dr. Kalil reports no relevant disclosures.

SAN DIEGO – Powerful antibiotics come first to mind when hospitalized patients have sepsis, but a critical care pulmonology specialist urged colleagues to consider another treatment – heavy intravenous doses of vitamin C.

“There is evidence supporting benefit, and ample evidence supporting safety,” Michael H. Hooper, MD, who practices in Norfolk, Va., said in a pro-and-con debate over the use of vitamin C in sepsis at the Critical Care Congress sponsored by the Society of Critical Care Medicine.

Dr. Hooper’s debate opponent countered by noting the lack of quality research into vitamin C in sepsis and declared that its time has not yet come. “We need more data to know the safety of this drug,” said Andre Kalil, MD, professor of internal medicine and director of Transplant Infectious Diseases at the University of Nebraska Medical Center, Omaha.

Dr. Hooper was part of a member of a team led by Paul E. Marik, MD, FCCP, of Eastern Virginia Medical School, Norfolk, that made waves in 2017 with a study in Chest suggesting IV vitamin C has tremendous potential as a treatment for sepsis (Chest. 2017 Jun;151[6]:1229-38).

The retrospective study compared two groups of 47 patients with sepsis – a control group and a group that received treatment with intravenous vitamin C, hydrocortisone, and thiamine. Remarkably, the team found that 9% (4 of 47) of those in the treatment group died in the hospital, compared with 40% (19 of 47) in the control group (P less than .001).

The findings make sense, Dr. Hooper said, in light of the fact that “our patients are remarkably deficient” in vitamin C. He pointed to a 2017 study that found nearly 40% of 24 patients with septic shock were deficient in vitamin C – despite getting recommended enteral nutrition, parenteral nutrition or both – compared with 25% of patients who were not septic. The study authors believe the difference is probably due to “increased metabolism due to the enhanced inflammatory response observed in septic shock” (Crit Care. 2017 Dec 11;21[1]:300).

“We’re dealing with a population of patients who need some sort of repletion of this vitamin,” Dr. Hooper said.

Why not try oral administration of vitamin C? “Oral administration at regular doses doesn’t work,” he said. “If you have normal volunteers who are made deficient, then you administer the recommended allowance, it takes days or weeks to return levels to normal.”

Dr. Hooper added that the goal of vitamin C therapy isn’t simply to restore proper levels in plasma. In addition, he said, “we’re trying to restore levels in crucial organs.”

He said the cost of treatment with IV vitamin C is low, and no serious adverse events have been seen in studies of the vitamin’s use in critical care.

In his comments at the debate, Dr. Kalil pointed to several weaknesses in the 2017 study of vitamin C in sepsis. According to him, it had many problems, including a sample size that lacked statistical power and imbalances in the two groups. He raised concerns about the study in a 2017 letter published in Chest titled “Vitamin C Is Not Ready for Prime Time in Sepsis but a Solution Is Close,” noting that the control group was sicker and none of those patients had their vitamin C levels measured (Chest. 2017 Sep;152[3]:676).

Randy Dotinga/MDedge News

He added that “acute renal failure is associated with high doses of vitamin C.”

As of July 2018, several clinical trials into vitamin C, hydrocortisone, and thiamine for the treatment of septic shock were underway or planned, according to a report that described the current randomized, placebo-controlled, multicenter Ascorbic Acid, Corticosteroids, and Thiamine in Sepsis (ACTS) trial in the United States. The report notes that “robust evidence” for this approach is lacking, although “the potential effectiveness of this medication combination is rooted in biologic plausibility and supported by small clinical trials of the various individual components.” (Crit Care. 2018;22:283)

Dr. Hooper is an executive committee member and principal investigator with the Vitamin C, Thiamine And Steroids in Sepsis (VICTAS) study. Dr. Kalil reports no relevant disclosures.

SNOWMASS, COLO. – Infective endocarditis in 2019 is very different from the disease most physicians encountered in training, both in terms of epidemiology and clinical presentation, Patrick T. O’Gara, MD, observed at the Annual Cardiovascular Conference at Snowmass sponsored by the American College of Cardiology.

Bruce Jancin/MDedge News

The classic description of infective endocarditis provided by Sir William Osler, MD, was of a subacute bacterial infection characterized by a long latent phase of low-grade fever, back pain, weight loss, and night sweats. It was mainly a right-heart disease of younger individuals with an infected native valve, and the predominant pathogens were streptococci, Dr. O’Gara said.

“I think in the current era endocarditis is more often characterized by an acute illness with toxic features in the context of adults with a high burden of degenerative diseases – for example, patients with rheumatoid arthritis or psoriatic arthritis on immunosuppressive therapy, or diabetes, end-stage renal disease, and risk factors for hospital-acquired infection. Injectable drug use is through the roof, there’s a wider prevalence of cardiac implanted electronic devices, which are a wonderful place for bacteria to hide, and Staphylococcus aureus has certainly become the leading pathogen with regard to endocarditis in the United States, especially MRSA, often multidrug resistant,” said Dr. O’Gara, professor of medicine at Harvard Medical School, Boston.

“Also, no talk about endocarditis is sufficient without paying some attention to the opioid crisis in which we find ourselves. It’s one of the top three causes of death among young men in the United States, along with accidents and gun violence. No region of the country is spared. This has completely inundated our ER and hospitalist services and our inpatient cardiology services with folks who are often repeat offenders when it comes to the difficulty in being able to give up an injectable drug use habit. They have multiple infections and hospitalizations, tricuspid valve involvement, and depending upon the aggressiveness of the Staphylococcus organism, typically they have left-sided disease with multiple complications, including aortic regurgitation and heart failure,” the cardiologist continued.

This description underscored one of Dr. O’Gara’s major points about the challenges posed by infective endocarditis in contemporary practice: “Expect the unexpected,” he advised. “When you’ve seen one case of infective endocarditis, you’ve seen one case of infective endocarditis.”
 

Outcomes are ‘sobering’

In the current era, outcomes are “sobering,” the cardiologist noted. Infective endocarditis carries a 6-month mortality rate of 20%-25% despite early surgery being performed during the index hospitalization in up to 60% of patients, with a relatively high perioperative mortality rate of about 10%. However, the risk of reinfection occurring in a newly implanted cardiac valve is impressively low at about 2%.

Refer early for multimodality imaging and surgical consultation

Transesophageal echocardiography is valuable in assessment of the infected valve. However, when extravalvular extension of the infection is suspected and the echo assessment is nondiagnostic or indeterminate, it’s time to quickly move on to advanced imaging, such as PET-CT.

The ACC/American Heart Association class I recommendations for early surgery in infected native valves haven’t changed substantially in over a decade. Based largely on observational data, there is an association between early surgery and lower in-hospital mortality (Lancet. 2012 Mar 10;379[9819]:965-975).

Class IIa recommendations for native valve surgery include recurrent emboli and a persistent vegetation despite appropriate antibiotic therapy. A “very controversial” class IIb recommendation for surgery because of weak supporting data is the identification of a mobile vegetation larger than 10 mm, particularly if it’s located on an anterior mitral valve leaflet, he said.

If the decision is made to forgo early surgery, be sure to repeat transesophageal echocardiography on day 7-10 to reassess the size of the patient’s vegetation.

“There is an association between size of vegetation and 1-year mortality, with a cut point of greater than 15 mm. Some would argue this constitutes a reasonable indication for early surgery,” Dr. O’Gara noted.

The embolization rate in patients with infective endocarditis is highest during the day before presentation, the day of presentation, and through the first 2 days afterward. The rate drops precipitously within 2 weeks after initiation of appropriate antibiotic therapy. Thus, to utilize early surgery to maximum effect in order to decrease the risk of embolization, it makes sense to operate within the first several days following presentation, before antibiotics have had sufficient time to catch up with the evolving disease process.
 

Don’t use half measures when it comes to removal of cardiac implanted electronic devices

The guidelines are clear regarding infected pacemakers, implanted cardioverter-defibrillators, and cardiac resynchronization devices: “It all needs to come out,” Dr. O’Gara emphasized. That includes all leads and the generator in patients with documented infection of only one portion of the device system, as a class I, level of evidence B recommendation. Moreover, complete removal of a pacemaker or defibrillator system is deemed “reasonable” as a class IIa recommendation in all patients with valvular infection caused by S. aureus or fungi even in the absence of evidence of device infection.

“I think we as general cardiologists have become increasingly impressed about how sick and festering these kinds of patients can become, even when we’re not able to prove that the lead is infected. The lead looks okay on transesophageal echo or PET-CT, blood cultures are negative, the valvular heart disease is really not that advanced, but several days go by and the patient is just not responding. We should have a high index of suspicion that there’s an infection we cannot appreciate. But obviously, you make these difficult decisions in consultation with your electrophysiology colleagues,” he added.
 

Know when the cardiologist should say ‘no’ to early aggressive surgery

While an aggressive early surgical approach often pays off in terms of prevention of embolic sequelae and a reduction in heart failure, the timing of surgery in the 20%-40% of patients with infective endocarditis who present with stroke or other neurologic complications remains controversial. An international group of Canadian and French cardiac surgeons and neurologists developed a useful algorithm regarding the types of neurologic complications for which early cardiac surgery is a poor idea because of the high risk of neurologic exacerbation. For example, a mycotic neuroaneurysm is grounds for postponement of cardiac surgery for at least 4 weeks (Circulation. 2016 Oct 25;134[17]:1280-92).

Dr. O’Gara reported receiving funding from the National Heart, Lung and Blood Institute, the National Institute of Dental and Craniofacial Research, from Medtronic in conjunction with the ongoing pivotal APOLLO transcatheter mitral valve replacement trial, and from Edwards Lifesciences for the ongoing EARLY TAVR trial.

[email protected]

SNOWMASS, COLO. – Infective endocarditis in 2019 is very different from the disease most physicians encountered in training, both in terms of epidemiology and clinical presentation, Patrick T. O’Gara, MD, observed at the Annual Cardiovascular Conference at Snowmass sponsored by the American College of Cardiology.

Bruce Jancin/MDedge News

The classic description of infective endocarditis provided by Sir William Osler, MD, was of a subacute bacterial infection characterized by a long latent phase of low-grade fever, back pain, weight loss, and night sweats. It was mainly a right-heart disease of younger individuals with an infected native valve, and the predominant pathogens were streptococci, Dr. O’Gara said.

“I think in the current era endocarditis is more often characterized by an acute illness with toxic features in the context of adults with a high burden of degenerative diseases – for example, patients with rheumatoid arthritis or psoriatic arthritis on immunosuppressive therapy, or diabetes, end-stage renal disease, and risk factors for hospital-acquired infection. Injectable drug use is through the roof, there’s a wider prevalence of cardiac implanted electronic devices, which are a wonderful place for bacteria to hide, and Staphylococcus aureus has certainly become the leading pathogen with regard to endocarditis in the United States, especially MRSA, often multidrug resistant,” said Dr. O’Gara, professor of medicine at Harvard Medical School, Boston.

“Also, no talk about endocarditis is sufficient without paying some attention to the opioid crisis in which we find ourselves. It’s one of the top three causes of death among young men in the United States, along with accidents and gun violence. No region of the country is spared. This has completely inundated our ER and hospitalist services and our inpatient cardiology services with folks who are often repeat offenders when it comes to the difficulty in being able to give up an injectable drug use habit. They have multiple infections and hospitalizations, tricuspid valve involvement, and depending upon the aggressiveness of the Staphylococcus organism, typically they have left-sided disease with multiple complications, including aortic regurgitation and heart failure,” the cardiologist continued.

This description underscored one of Dr. O’Gara’s major points about the challenges posed by infective endocarditis in contemporary practice: “Expect the unexpected,” he advised. “When you’ve seen one case of infective endocarditis, you’ve seen one case of infective endocarditis.”
 

Outcomes are ‘sobering’

In the current era, outcomes are “sobering,” the cardiologist noted. Infective endocarditis carries a 6-month mortality rate of 20%-25% despite early surgery being performed during the index hospitalization in up to 60% of patients, with a relatively high perioperative mortality rate of about 10%. However, the risk of reinfection occurring in a newly implanted cardiac valve is impressively low at about 2%.

Refer early for multimodality imaging and surgical consultation

Transesophageal echocardiography is valuable in assessment of the infected valve. However, when extravalvular extension of the infection is suspected and the echo assessment is nondiagnostic or indeterminate, it’s time to quickly move on to advanced imaging, such as PET-CT.

The ACC/American Heart Association class I recommendations for early surgery in infected native valves haven’t changed substantially in over a decade. Based largely on observational data, there is an association between early surgery and lower in-hospital mortality (Lancet. 2012 Mar 10;379[9819]:965-975).

Class IIa recommendations for native valve surgery include recurrent emboli and a persistent vegetation despite appropriate antibiotic therapy. A “very controversial” class IIb recommendation for surgery because of weak supporting data is the identification of a mobile vegetation larger than 10 mm, particularly if it’s located on an anterior mitral valve leaflet, he said.

If the decision is made to forgo early surgery, be sure to repeat transesophageal echocardiography on day 7-10 to reassess the size of the patient’s vegetation.

“There is an association between size of vegetation and 1-year mortality, with a cut point of greater than 15 mm. Some would argue this constitutes a reasonable indication for early surgery,” Dr. O’Gara noted.

The embolization rate in patients with infective endocarditis is highest during the day before presentation, the day of presentation, and through the first 2 days afterward. The rate drops precipitously within 2 weeks after initiation of appropriate antibiotic therapy. Thus, to utilize early surgery to maximum effect in order to decrease the risk of embolization, it makes sense to operate within the first several days following presentation, before antibiotics have had sufficient time to catch up with the evolving disease process.
 

Don’t use half measures when it comes to removal of cardiac implanted electronic devices

The guidelines are clear regarding infected pacemakers, implanted cardioverter-defibrillators, and cardiac resynchronization devices: “It all needs to come out,” Dr. O’Gara emphasized. That includes all leads and the generator in patients with documented infection of only one portion of the device system, as a class I, level of evidence B recommendation. Moreover, complete removal of a pacemaker or defibrillator system is deemed “reasonable” as a class IIa recommendation in all patients with valvular infection caused by S. aureus or fungi even in the absence of evidence of device infection.

“I think we as general cardiologists have become increasingly impressed about how sick and festering these kinds of patients can become, even when we’re not able to prove that the lead is infected. The lead looks okay on transesophageal echo or PET-CT, blood cultures are negative, the valvular heart disease is really not that advanced, but several days go by and the patient is just not responding. We should have a high index of suspicion that there’s an infection we cannot appreciate. But obviously, you make these difficult decisions in consultation with your electrophysiology colleagues,” he added.
 

Know when the cardiologist should say ‘no’ to early aggressive surgery

While an aggressive early surgical approach often pays off in terms of prevention of embolic sequelae and a reduction in heart failure, the timing of surgery in the 20%-40% of patients with infective endocarditis who present with stroke or other neurologic complications remains controversial. An international group of Canadian and French cardiac surgeons and neurologists developed a useful algorithm regarding the types of neurologic complications for which early cardiac surgery is a poor idea because of the high risk of neurologic exacerbation. For example, a mycotic neuroaneurysm is grounds for postponement of cardiac surgery for at least 4 weeks (Circulation. 2016 Oct 25;134[17]:1280-92).

Dr. O’Gara reported receiving funding from the National Heart, Lung and Blood Institute, the National Institute of Dental and Craniofacial Research, from Medtronic in conjunction with the ongoing pivotal APOLLO transcatheter mitral valve replacement trial, and from Edwards Lifesciences for the ongoing EARLY TAVR trial.

[email protected]

SNOWMASS, COLO. – Infective endocarditis in 2019 is very different from the disease most physicians encountered in training, both in terms of epidemiology and clinical presentation, Patrick T. O’Gara, MD, observed at the Annual Cardiovascular Conference at Snowmass sponsored by the American College of Cardiology.

Bruce Jancin/MDedge News

The classic description of infective endocarditis provided by Sir William Osler, MD, was of a subacute bacterial infection characterized by a long latent phase of low-grade fever, back pain, weight loss, and night sweats. It was mainly a right-heart disease of younger individuals with an infected native valve, and the predominant pathogens were streptococci, Dr. O’Gara said.

“I think in the current era endocarditis is more often characterized by an acute illness with toxic features in the context of adults with a high burden of degenerative diseases – for example, patients with rheumatoid arthritis or psoriatic arthritis on immunosuppressive therapy, or diabetes, end-stage renal disease, and risk factors for hospital-acquired infection. Injectable drug use is through the roof, there’s a wider prevalence of cardiac implanted electronic devices, which are a wonderful place for bacteria to hide, and Staphylococcus aureus has certainly become the leading pathogen with regard to endocarditis in the United States, especially MRSA, often multidrug resistant,” said Dr. O’Gara, professor of medicine at Harvard Medical School, Boston.

“Also, no talk about endocarditis is sufficient without paying some attention to the opioid crisis in which we find ourselves. It’s one of the top three causes of death among young men in the United States, along with accidents and gun violence. No region of the country is spared. This has completely inundated our ER and hospitalist services and our inpatient cardiology services with folks who are often repeat offenders when it comes to the difficulty in being able to give up an injectable drug use habit. They have multiple infections and hospitalizations, tricuspid valve involvement, and depending upon the aggressiveness of the Staphylococcus organism, typically they have left-sided disease with multiple complications, including aortic regurgitation and heart failure,” the cardiologist continued.

This description underscored one of Dr. O’Gara’s major points about the challenges posed by infective endocarditis in contemporary practice: “Expect the unexpected,” he advised. “When you’ve seen one case of infective endocarditis, you’ve seen one case of infective endocarditis.”
 

Outcomes are ‘sobering’

In the current era, outcomes are “sobering,” the cardiologist noted. Infective endocarditis carries a 6-month mortality rate of 20%-25% despite early surgery being performed during the index hospitalization in up to 60% of patients, with a relatively high perioperative mortality rate of about 10%. However, the risk of reinfection occurring in a newly implanted cardiac valve is impressively low at about 2%.

Refer early for multimodality imaging and surgical consultation

Transesophageal echocardiography is valuable in assessment of the infected valve. However, when extravalvular extension of the infection is suspected and the echo assessment is nondiagnostic or indeterminate, it’s time to quickly move on to advanced imaging, such as PET-CT.

The ACC/American Heart Association class I recommendations for early surgery in infected native valves haven’t changed substantially in over a decade. Based largely on observational data, there is an association between early surgery and lower in-hospital mortality (Lancet. 2012 Mar 10;379[9819]:965-975).

Class IIa recommendations for native valve surgery include recurrent emboli and a persistent vegetation despite appropriate antibiotic therapy. A “very controversial” class IIb recommendation for surgery because of weak supporting data is the identification of a mobile vegetation larger than 10 mm, particularly if it’s located on an anterior mitral valve leaflet, he said.

If the decision is made to forgo early surgery, be sure to repeat transesophageal echocardiography on day 7-10 to reassess the size of the patient’s vegetation.

“There is an association between size of vegetation and 1-year mortality, with a cut point of greater than 15 mm. Some would argue this constitutes a reasonable indication for early surgery,” Dr. O’Gara noted.

The embolization rate in patients with infective endocarditis is highest during the day before presentation, the day of presentation, and through the first 2 days afterward. The rate drops precipitously within 2 weeks after initiation of appropriate antibiotic therapy. Thus, to utilize early surgery to maximum effect in order to decrease the risk of embolization, it makes sense to operate within the first several days following presentation, before antibiotics have had sufficient time to catch up with the evolving disease process.
 

Don’t use half measures when it comes to removal of cardiac implanted electronic devices

The guidelines are clear regarding infected pacemakers, implanted cardioverter-defibrillators, and cardiac resynchronization devices: “It all needs to come out,” Dr. O’Gara emphasized. That includes all leads and the generator in patients with documented infection of only one portion of the device system, as a class I, level of evidence B recommendation. Moreover, complete removal of a pacemaker or defibrillator system is deemed “reasonable” as a class IIa recommendation in all patients with valvular infection caused by S. aureus or fungi even in the absence of evidence of device infection.

“I think we as general cardiologists have become increasingly impressed about how sick and festering these kinds of patients can become, even when we’re not able to prove that the lead is infected. The lead looks okay on transesophageal echo or PET-CT, blood cultures are negative, the valvular heart disease is really not that advanced, but several days go by and the patient is just not responding. We should have a high index of suspicion that there’s an infection we cannot appreciate. But obviously, you make these difficult decisions in consultation with your electrophysiology colleagues,” he added.
 

Know when the cardiologist should say ‘no’ to early aggressive surgery

While an aggressive early surgical approach often pays off in terms of prevention of embolic sequelae and a reduction in heart failure, the timing of surgery in the 20%-40% of patients with infective endocarditis who present with stroke or other neurologic complications remains controversial. An international group of Canadian and French cardiac surgeons and neurologists developed a useful algorithm regarding the types of neurologic complications for which early cardiac surgery is a poor idea because of the high risk of neurologic exacerbation. For example, a mycotic neuroaneurysm is grounds for postponement of cardiac surgery for at least 4 weeks (Circulation. 2016 Oct 25;134[17]:1280-92).

Dr. O’Gara reported receiving funding from the National Heart, Lung and Blood Institute, the National Institute of Dental and Craniofacial Research, from Medtronic in conjunction with the ongoing pivotal APOLLO transcatheter mitral valve replacement trial, and from Edwards Lifesciences for the ongoing EARLY TAVR trial.

[email protected]

Women hospitalized with cirrhosis are less likely to die in the hospital than are men, according to a retrospective analysis of more than half a million patients.

Although women more often had infections and comorbidities, men more often had liver decompensation, which contributed most significantly to their higher mortality rate, reported lead author Jessica Rubin, MD, of the University of California, San Francisco, and her colleagues.

Their findings add to an existing body of knowledge about sex-related differences in chronic liver disease. Women are less likely to develop chronic liver disease; however, when women do develop disease, it often follows a unique clinical course, with milder early disease followed by more severe end-stage disease, meaning many women are too sick for a transplant, or die on the waiting list.

“The reasons behind this ‘reversal’ in [sex] disparities is unknown,” the investigators wrote in Journal of Clinical Gastroenterology.

Considering recent findings that showed a correlation between hospitalization and mortality rates in chronic liver disease, the investigators believed that a comparison of hospital-related outcomes in men and women could explain why women apparently fare worse when dealing with end-stage disease.

The retrospective, cross-sectional study involved 553,017 patients (median age, 57 years) who were hospitalized for cirrhosis between 2009 and 2013. Data were drawn from the National Inpatient Sample (NIS). Inpatient mortality was the primary outcome.

In agreement with previous findings, the minority of patients were women (39%). Against expectations, however, women had a significantly lower mortality rate than that of men (5.7% vs. 6.4%; multivariable analysis odds ratio, 0.86). Better survival was associated with lower rates of decompensation (Baveno IV criteria; 34% vs. 38.8%) and other cirrhosis complications, such as hepatorenal syndrome, variceal bleeding, ascites, and spontaneous bacterial peritonitis. The only cirrhosis complication more common in women than men was hepatic encephalopathy (17.8% vs. 16.8%). Owing to fewer complications, fewer women required liver-related interventions, including transjugular intrahepatic portosystemic shunt (0.8% vs. 1.0%), upper endoscopy (12.8% vs. 13.0%), or paracentesis (17.6% vs. 20.6%).

While less frequent complications and a lower mortality rate might suggest that women were admitted with better overall clinical pictures, not all data supported this conclusion. For instance, women were more likely to have noncirrhosis comorbidities, including diabetes, hypertension, heart failure, stroke, and cancer. Furthermore, women had a higher rate of acute bacterial infection than that of men (34.9% vs. 28.2%), although this disparity should be considered in light of urinary tract infections (UTIs), which were significantly more common among women (18.8% vs. 8.0%).

“Interestingly, infections were a stronger predictor of inpatient mortality in women than men,” the investigators wrote. “Despite this, women in our cohort were less likely to die in the hospital than men.”

Additional analysis revealed etiological differences that may have contributed to differences in mortality rates. For instance, women less often had liver disease due to viral hepatitis (27.6% vs. 35.2%) or alcohol (24.1% vs. 38.7%). In contrast, women more often had autoimmune hepatitis (2.5% vs. 0.4%) or cirrhosis due to unspecified or miscellaneous reasons (45.7% vs. 25.7%).

“Our data suggest that differential rates of ongoing liver injury – including by cofactors such as active alcohol use – explain some but not all of the [sex] difference we observed in hepatic decompensation,” the investigators wrote, before redirecting focus to a clearer clinical finding. “The poor prognosis of decompensated cirrhosis ... provides a reasonable explanation for the higher rates of in-hospital mortality seen among men versus women,” they concluded.

Considering the surprising findings and previously known sex disparities, Dr. Rubin and her colleagues suggested that more research in this area is needed, along with efforts to deliver sex-appropriate care.

“The development of [sex]-specific cirrhosis management programs – focused on interventions to manage the interaction between cirrhosis and other common comorbidities, improving physical function both before and during hospitalization, and postacute discharge programs to facilitate resumption of independent living – would target differential needs of women and men living with cirrhosis, with the ultimate goal of improving long-term outcomes in these patients,” the investigators wrote.

The study was funded by a National Institute on Aging Paul B. Beeson Career Development Award in Aging and a National Institute of Diabetes and Digestive and Kidney Diseases National Research Service Award hepatology training grant. The investigators declared no conflicts of interest.

SOURCE: Rubin et al. J Clin Gastroenterol. 2019 Feb 22. doi: 10.1097/MCG.0000000000001192.

Women hospitalized with cirrhosis are less likely to die in the hospital than are men, according to a retrospective analysis of more than half a million patients.

Although women more often had infections and comorbidities, men more often had liver decompensation, which contributed most significantly to their higher mortality rate, reported lead author Jessica Rubin, MD, of the University of California, San Francisco, and her colleagues.

Their findings add to an existing body of knowledge about sex-related differences in chronic liver disease. Women are less likely to develop chronic liver disease; however, when women do develop disease, it often follows a unique clinical course, with milder early disease followed by more severe end-stage disease, meaning many women are too sick for a transplant, or die on the waiting list.

“The reasons behind this ‘reversal’ in [sex] disparities is unknown,” the investigators wrote in Journal of Clinical Gastroenterology.

Considering recent findings that showed a correlation between hospitalization and mortality rates in chronic liver disease, the investigators believed that a comparison of hospital-related outcomes in men and women could explain why women apparently fare worse when dealing with end-stage disease.

The retrospective, cross-sectional study involved 553,017 patients (median age, 57 years) who were hospitalized for cirrhosis between 2009 and 2013. Data were drawn from the National Inpatient Sample (NIS). Inpatient mortality was the primary outcome.

In agreement with previous findings, the minority of patients were women (39%). Against expectations, however, women had a significantly lower mortality rate than that of men (5.7% vs. 6.4%; multivariable analysis odds ratio, 0.86). Better survival was associated with lower rates of decompensation (Baveno IV criteria; 34% vs. 38.8%) and other cirrhosis complications, such as hepatorenal syndrome, variceal bleeding, ascites, and spontaneous bacterial peritonitis. The only cirrhosis complication more common in women than men was hepatic encephalopathy (17.8% vs. 16.8%). Owing to fewer complications, fewer women required liver-related interventions, including transjugular intrahepatic portosystemic shunt (0.8% vs. 1.0%), upper endoscopy (12.8% vs. 13.0%), or paracentesis (17.6% vs. 20.6%).

While less frequent complications and a lower mortality rate might suggest that women were admitted with better overall clinical pictures, not all data supported this conclusion. For instance, women were more likely to have noncirrhosis comorbidities, including diabetes, hypertension, heart failure, stroke, and cancer. Furthermore, women had a higher rate of acute bacterial infection than that of men (34.9% vs. 28.2%), although this disparity should be considered in light of urinary tract infections (UTIs), which were significantly more common among women (18.8% vs. 8.0%).

“Interestingly, infections were a stronger predictor of inpatient mortality in women than men,” the investigators wrote. “Despite this, women in our cohort were less likely to die in the hospital than men.”

Additional analysis revealed etiological differences that may have contributed to differences in mortality rates. For instance, women less often had liver disease due to viral hepatitis (27.6% vs. 35.2%) or alcohol (24.1% vs. 38.7%). In contrast, women more often had autoimmune hepatitis (2.5% vs. 0.4%) or cirrhosis due to unspecified or miscellaneous reasons (45.7% vs. 25.7%).

“Our data suggest that differential rates of ongoing liver injury – including by cofactors such as active alcohol use – explain some but not all of the [sex] difference we observed in hepatic decompensation,” the investigators wrote, before redirecting focus to a clearer clinical finding. “The poor prognosis of decompensated cirrhosis ... provides a reasonable explanation for the higher rates of in-hospital mortality seen among men versus women,” they concluded.

Considering the surprising findings and previously known sex disparities, Dr. Rubin and her colleagues suggested that more research in this area is needed, along with efforts to deliver sex-appropriate care.

“The development of [sex]-specific cirrhosis management programs – focused on interventions to manage the interaction between cirrhosis and other common comorbidities, improving physical function both before and during hospitalization, and postacute discharge programs to facilitate resumption of independent living – would target differential needs of women and men living with cirrhosis, with the ultimate goal of improving long-term outcomes in these patients,” the investigators wrote.

The study was funded by a National Institute on Aging Paul B. Beeson Career Development Award in Aging and a National Institute of Diabetes and Digestive and Kidney Diseases National Research Service Award hepatology training grant. The investigators declared no conflicts of interest.

SOURCE: Rubin et al. J Clin Gastroenterol. 2019 Feb 22. doi: 10.1097/MCG.0000000000001192.

Women hospitalized with cirrhosis are less likely to die in the hospital than are men, according to a retrospective analysis of more than half a million patients.

Although women more often had infections and comorbidities, men more often had liver decompensation, which contributed most significantly to their higher mortality rate, reported lead author Jessica Rubin, MD, of the University of California, San Francisco, and her colleagues.

Their findings add to an existing body of knowledge about sex-related differences in chronic liver disease. Women are less likely to develop chronic liver disease; however, when women do develop disease, it often follows a unique clinical course, with milder early disease followed by more severe end-stage disease, meaning many women are too sick for a transplant, or die on the waiting list.

“The reasons behind this ‘reversal’ in [sex] disparities is unknown,” the investigators wrote in Journal of Clinical Gastroenterology.

Considering recent findings that showed a correlation between hospitalization and mortality rates in chronic liver disease, the investigators believed that a comparison of hospital-related outcomes in men and women could explain why women apparently fare worse when dealing with end-stage disease.

The retrospective, cross-sectional study involved 553,017 patients (median age, 57 years) who were hospitalized for cirrhosis between 2009 and 2013. Data were drawn from the National Inpatient Sample (NIS). Inpatient mortality was the primary outcome.

In agreement with previous findings, the minority of patients were women (39%). Against expectations, however, women had a significantly lower mortality rate than that of men (5.7% vs. 6.4%; multivariable analysis odds ratio, 0.86). Better survival was associated with lower rates of decompensation (Baveno IV criteria; 34% vs. 38.8%) and other cirrhosis complications, such as hepatorenal syndrome, variceal bleeding, ascites, and spontaneous bacterial peritonitis. The only cirrhosis complication more common in women than men was hepatic encephalopathy (17.8% vs. 16.8%). Owing to fewer complications, fewer women required liver-related interventions, including transjugular intrahepatic portosystemic shunt (0.8% vs. 1.0%), upper endoscopy (12.8% vs. 13.0%), or paracentesis (17.6% vs. 20.6%).

While less frequent complications and a lower mortality rate might suggest that women were admitted with better overall clinical pictures, not all data supported this conclusion. For instance, women were more likely to have noncirrhosis comorbidities, including diabetes, hypertension, heart failure, stroke, and cancer. Furthermore, women had a higher rate of acute bacterial infection than that of men (34.9% vs. 28.2%), although this disparity should be considered in light of urinary tract infections (UTIs), which were significantly more common among women (18.8% vs. 8.0%).

“Interestingly, infections were a stronger predictor of inpatient mortality in women than men,” the investigators wrote. “Despite this, women in our cohort were less likely to die in the hospital than men.”

Additional analysis revealed etiological differences that may have contributed to differences in mortality rates. For instance, women less often had liver disease due to viral hepatitis (27.6% vs. 35.2%) or alcohol (24.1% vs. 38.7%). In contrast, women more often had autoimmune hepatitis (2.5% vs. 0.4%) or cirrhosis due to unspecified or miscellaneous reasons (45.7% vs. 25.7%).

“Our data suggest that differential rates of ongoing liver injury – including by cofactors such as active alcohol use – explain some but not all of the [sex] difference we observed in hepatic decompensation,” the investigators wrote, before redirecting focus to a clearer clinical finding. “The poor prognosis of decompensated cirrhosis ... provides a reasonable explanation for the higher rates of in-hospital mortality seen among men versus women,” they concluded.

Considering the surprising findings and previously known sex disparities, Dr. Rubin and her colleagues suggested that more research in this area is needed, along with efforts to deliver sex-appropriate care.

“The development of [sex]-specific cirrhosis management programs – focused on interventions to manage the interaction between cirrhosis and other common comorbidities, improving physical function both before and during hospitalization, and postacute discharge programs to facilitate resumption of independent living – would target differential needs of women and men living with cirrhosis, with the ultimate goal of improving long-term outcomes in these patients,” the investigators wrote.

The study was funded by a National Institute on Aging Paul B. Beeson Career Development Award in Aging and a National Institute of Diabetes and Digestive and Kidney Diseases National Research Service Award hepatology training grant. The investigators declared no conflicts of interest.

SOURCE: Rubin et al. J Clin Gastroenterol. 2019 Feb 22. doi: 10.1097/MCG.0000000000001192.

SAN DIEGO – In the pediatric intensive care unit (PICU), rapid whole genome sequencing (rWGS) with targeted phenotype analysis often leads to specific changes in patient management, similar to what is seen when rWGS is applied to neonatal ICUs. The findings come from the first study to look at outcomes of rWGS in the PICU outside of infancy.

NaiyanaDonraman/Thinkstock

In the NICU, previous studies have shown an rWGS diagnostic rate ranging from 36% to 57%, and an estimated 49%-72% of these diagnoses result in changes to patient management; 38%-45% of those diagnoses had not been previously considered.

The researchers found similar benefits when the age of patients was extended to 18 years in the PICU. “We were happy to see we were able to make specific changes in ICU management, with three of those being medication changes based on the diagnosis, and one being in the transition to palliative care while the patient was still in the PICU,” Erica Sanford, MD, said in an interview.

Dr. Sanford is a pediatric clinical care fellow at the University of California, San Diego. She presented the study at the Critical Care Congress sponsored by the Society of Critical Care Medicine.

Changes as a result of rWGS included factor replacement for a factor XIII deficiency, avoidance of renal biopsy because the diagnosis was made genetically, and use of serial MRI and other imaging methods to identify disorder-related sequela. “We’re hopeful as the turnaround for genome sequencing decreases that we’ll be able to offer this as a test when appropriate when a patient is presenting with an illness that doesn’t have a clear etiologic diagnosis,” said Dr. Sanford.

Certainly not every patient in the PICU should undergo rWGS, given its expense. Although the study was too small to identify candidate patients, there were some trends – patients in cardiac arrest were more likely to be receiving a genetic diagnosis. “We weren’t able to answer the question of which patients in the pediatric ICU should get whole genome sequencing – our next step is to have a larger group of patients so we can determine specifically which patients might benefit,” said Dr. Sanford. The team also plans to do a cost analysis of rWGS in the PICU setting.

The researchers examined data from a PICU at a tertiary children’s hospital, including records from 38 patients who underwent rWGS between July 2016 and May 2018. Based on the initial sequencing results, 18 underwent diagnostic rWGS. The average age of children was 5.7 years (median 3 years), with patients ranging from 4 months to 18 years old.

The most common reasons for PICU admission were shock (16% of all patients, 17% of patients who received diagnostic rWGS), respiratory failure (18% and 17%), cardiac arrest (13% and 22%), and altered mental status (13% and 11%).

Eleven of 18 patients (61%) who received an rWGS diagnosis experienced a subacute, non-ICU change in management. Four diagnoses (22%) led to a change in ICU management, and three led to no changes in patient management.

The National Institutes of Health funded the study. Dr. Sanford had no relevant financial disclosures.

SOURCE: Sanford E et al. CCC48, Abstract 373.

SAN DIEGO – In the pediatric intensive care unit (PICU), rapid whole genome sequencing (rWGS) with targeted phenotype analysis often leads to specific changes in patient management, similar to what is seen when rWGS is applied to neonatal ICUs. The findings come from the first study to look at outcomes of rWGS in the PICU outside of infancy.

NaiyanaDonraman/Thinkstock

In the NICU, previous studies have shown an rWGS diagnostic rate ranging from 36% to 57%, and an estimated 49%-72% of these diagnoses result in changes to patient management; 38%-45% of those diagnoses had not been previously considered.

The researchers found similar benefits when the age of patients was extended to 18 years in the PICU. “We were happy to see we were able to make specific changes in ICU management, with three of those being medication changes based on the diagnosis, and one being in the transition to palliative care while the patient was still in the PICU,” Erica Sanford, MD, said in an interview.

Dr. Sanford is a pediatric clinical care fellow at the University of California, San Diego. She presented the study at the Critical Care Congress sponsored by the Society of Critical Care Medicine.

Changes as a result of rWGS included factor replacement for a factor XIII deficiency, avoidance of renal biopsy because the diagnosis was made genetically, and use of serial MRI and other imaging methods to identify disorder-related sequela. “We’re hopeful as the turnaround for genome sequencing decreases that we’ll be able to offer this as a test when appropriate when a patient is presenting with an illness that doesn’t have a clear etiologic diagnosis,” said Dr. Sanford.

Certainly not every patient in the PICU should undergo rWGS, given its expense. Although the study was too small to identify candidate patients, there were some trends – patients in cardiac arrest were more likely to be receiving a genetic diagnosis. “We weren’t able to answer the question of which patients in the pediatric ICU should get whole genome sequencing – our next step is to have a larger group of patients so we can determine specifically which patients might benefit,” said Dr. Sanford. The team also plans to do a cost analysis of rWGS in the PICU setting.

The researchers examined data from a PICU at a tertiary children’s hospital, including records from 38 patients who underwent rWGS between July 2016 and May 2018. Based on the initial sequencing results, 18 underwent diagnostic rWGS. The average age of children was 5.7 years (median 3 years), with patients ranging from 4 months to 18 years old.

The most common reasons for PICU admission were shock (16% of all patients, 17% of patients who received diagnostic rWGS), respiratory failure (18% and 17%), cardiac arrest (13% and 22%), and altered mental status (13% and 11%).

Eleven of 18 patients (61%) who received an rWGS diagnosis experienced a subacute, non-ICU change in management. Four diagnoses (22%) led to a change in ICU management, and three led to no changes in patient management.

The National Institutes of Health funded the study. Dr. Sanford had no relevant financial disclosures.

SOURCE: Sanford E et al. CCC48, Abstract 373.

SAN DIEGO – In the pediatric intensive care unit (PICU), rapid whole genome sequencing (rWGS) with targeted phenotype analysis often leads to specific changes in patient management, similar to what is seen when rWGS is applied to neonatal ICUs. The findings come from the first study to look at outcomes of rWGS in the PICU outside of infancy.

NaiyanaDonraman/Thinkstock

In the NICU, previous studies have shown an rWGS diagnostic rate ranging from 36% to 57%, and an estimated 49%-72% of these diagnoses result in changes to patient management; 38%-45% of those diagnoses had not been previously considered.

The researchers found similar benefits when the age of patients was extended to 18 years in the PICU. “We were happy to see we were able to make specific changes in ICU management, with three of those being medication changes based on the diagnosis, and one being in the transition to palliative care while the patient was still in the PICU,” Erica Sanford, MD, said in an interview.

Dr. Sanford is a pediatric clinical care fellow at the University of California, San Diego. She presented the study at the Critical Care Congress sponsored by the Society of Critical Care Medicine.

Changes as a result of rWGS included factor replacement for a factor XIII deficiency, avoidance of renal biopsy because the diagnosis was made genetically, and use of serial MRI and other imaging methods to identify disorder-related sequela. “We’re hopeful as the turnaround for genome sequencing decreases that we’ll be able to offer this as a test when appropriate when a patient is presenting with an illness that doesn’t have a clear etiologic diagnosis,” said Dr. Sanford.

Certainly not every patient in the PICU should undergo rWGS, given its expense. Although the study was too small to identify candidate patients, there were some trends – patients in cardiac arrest were more likely to be receiving a genetic diagnosis. “We weren’t able to answer the question of which patients in the pediatric ICU should get whole genome sequencing – our next step is to have a larger group of patients so we can determine specifically which patients might benefit,” said Dr. Sanford. The team also plans to do a cost analysis of rWGS in the PICU setting.

The researchers examined data from a PICU at a tertiary children’s hospital, including records from 38 patients who underwent rWGS between July 2016 and May 2018. Based on the initial sequencing results, 18 underwent diagnostic rWGS. The average age of children was 5.7 years (median 3 years), with patients ranging from 4 months to 18 years old.

The most common reasons for PICU admission were shock (16% of all patients, 17% of patients who received diagnostic rWGS), respiratory failure (18% and 17%), cardiac arrest (13% and 22%), and altered mental status (13% and 11%).

Eleven of 18 patients (61%) who received an rWGS diagnosis experienced a subacute, non-ICU change in management. Four diagnoses (22%) led to a change in ICU management, and three led to no changes in patient management.

The National Institutes of Health funded the study. Dr. Sanford had no relevant financial disclosures.

SOURCE: Sanford E et al. CCC48, Abstract 373.

Serious infections from Staphylococcus aureus caused nearly 20,000 deaths and 119,000 illnesses in the United States in 2017, according to data from a Vital Signs report issued by the Centers for Disease Control and Prevention. The data include both methicillin-resistant S. aureus (MRSA) and methicillin-susceptible S. aureus (MSSA).

Although MRSA infections in health care settings declined by approximately 17% during 2005-2012, rates plateaued during 2012-2017, Anne Schuchat, MD, principal deputy director of the CDC, said in a teleconference March 5 to present the findings. The report emphasizes the potential for serious illness and death with any staph infection and the need for ongoing vigilance on the part of clinicians, she said.

In addition, community-onset MSSA infections increased by 3.9%/year during 2012-2017. Data from previous studies suggest that this increase may be connected to the opioid epidemic, said Dr. Schuchat.

“People who inject drugs are 16% more likely to develop a staph infection” than are those who don’t inject drugs, she said.

Community-onset MRSA declined by 6.9% during 2001-2016, attributed to declines in health care–associated infections, according to Vital Signs author Athena P. Kourtis, MD, of the CDC’s National Center for Emerging and Zoonotic Infectious Diseases, and her colleagues. Rates of hospital-associated MSSA infection remained essentially unchanged (P = .11). The overall unadjusted in-hospital mortality among patients with S. aureus bloodstream infections over the study period was 18%.

The data for the report were collected from electronic health records at more than 400 acute care hospitals, as well as population-based surveillance data from the CDC’s Emerging Infections Program.

Most people carry staph on their skin with no ill effects, but the bacteria become dangerous when they enter the bloodstream, Dr. Schuchat emphasized. “We hope the new data today will refocus the nation’s efforts to protect patients from staph infections,” she said.

SOURCE: Kourtis AP et al. MMWR. 2019 Mar 5; 68:1-6.

Serious infections from Staphylococcus aureus caused nearly 20,000 deaths and 119,000 illnesses in the United States in 2017, according to data from a Vital Signs report issued by the Centers for Disease Control and Prevention. The data include both methicillin-resistant S. aureus (MRSA) and methicillin-susceptible S. aureus (MSSA).

Although MRSA infections in health care settings declined by approximately 17% during 2005-2012, rates plateaued during 2012-2017, Anne Schuchat, MD, principal deputy director of the CDC, said in a teleconference March 5 to present the findings. The report emphasizes the potential for serious illness and death with any staph infection and the need for ongoing vigilance on the part of clinicians, she said.

In addition, community-onset MSSA infections increased by 3.9%/year during 2012-2017. Data from previous studies suggest that this increase may be connected to the opioid epidemic, said Dr. Schuchat.

“People who inject drugs are 16% more likely to develop a staph infection” than are those who don’t inject drugs, she said.

Community-onset MRSA declined by 6.9% during 2001-2016, attributed to declines in health care–associated infections, according to Vital Signs author Athena P. Kourtis, MD, of the CDC’s National Center for Emerging and Zoonotic Infectious Diseases, and her colleagues. Rates of hospital-associated MSSA infection remained essentially unchanged (P = .11). The overall unadjusted in-hospital mortality among patients with S. aureus bloodstream infections over the study period was 18%.

The data for the report were collected from electronic health records at more than 400 acute care hospitals, as well as population-based surveillance data from the CDC’s Emerging Infections Program.

Most people carry staph on their skin with no ill effects, but the bacteria become dangerous when they enter the bloodstream, Dr. Schuchat emphasized. “We hope the new data today will refocus the nation’s efforts to protect patients from staph infections,” she said.

SOURCE: Kourtis AP et al. MMWR. 2019 Mar 5; 68:1-6.

Serious infections from Staphylococcus aureus caused nearly 20,000 deaths and 119,000 illnesses in the United States in 2017, according to data from a Vital Signs report issued by the Centers for Disease Control and Prevention. The data include both methicillin-resistant S. aureus (MRSA) and methicillin-susceptible S. aureus (MSSA).

Although MRSA infections in health care settings declined by approximately 17% during 2005-2012, rates plateaued during 2012-2017, Anne Schuchat, MD, principal deputy director of the CDC, said in a teleconference March 5 to present the findings. The report emphasizes the potential for serious illness and death with any staph infection and the need for ongoing vigilance on the part of clinicians, she said.

In addition, community-onset MSSA infections increased by 3.9%/year during 2012-2017. Data from previous studies suggest that this increase may be connected to the opioid epidemic, said Dr. Schuchat.

“People who inject drugs are 16% more likely to develop a staph infection” than are those who don’t inject drugs, she said.

Community-onset MRSA declined by 6.9% during 2001-2016, attributed to declines in health care–associated infections, according to Vital Signs author Athena P. Kourtis, MD, of the CDC’s National Center for Emerging and Zoonotic Infectious Diseases, and her colleagues. Rates of hospital-associated MSSA infection remained essentially unchanged (P = .11). The overall unadjusted in-hospital mortality among patients with S. aureus bloodstream infections over the study period was 18%.

The data for the report were collected from electronic health records at more than 400 acute care hospitals, as well as population-based surveillance data from the CDC’s Emerging Infections Program.

Most people carry staph on their skin with no ill effects, but the bacteria become dangerous when they enter the bloodstream, Dr. Schuchat emphasized. “We hope the new data today will refocus the nation’s efforts to protect patients from staph infections,” she said.

SOURCE: Kourtis AP et al. MMWR. 2019 Mar 5; 68:1-6.

SAN DIEGO – A simplified version of the Sequential Organ Failure Assessment (SOFA) criteria, known as eSOFA, has the potential to make it easier for hospitals to benchmark sepsis outcomes and quality of care, and could propel new sepsis research. The new method replaces some of SOFA’s more subjective criteria with objective measures.

eSOFA relies on electronic health records to reduce reliance on administrative records, which suffer from cross-hospital variability in diagnosis and coding practices, as well as changes in these practices over time. The diagnosis of sepsis itself is also highly subjective. Instead, eSOFA determines dysfunction in six organ systems, indicated by use of vasopressors and mechanical ventilation, and the presence of abnormal laboratory values.

“The SOFA score includes measures like the Glasgow Coma Scale, which undoubtedly at the bedside is a very important clinical sign, but when trying to implement something that is objective for purposes of retrospective case counting and standardization, it can be problematic. The measures we chose [for eSOFA] are concrete, important maneuvers that were initiated by clinicians,” Chanu Rhee, MD, said in an interview.

Dr. Rhee is assistant professor of population medicine at Harvard Medical School and Brigham and Women’s Hospital, Boston. He presented the results of the study at the Critical Care Congress sponsored by the Society of Critical Care Medicine, and the work was simultaneously published online in Critical Care Medicine.

Key elements of SOFA that pose challenges for administrative data include: PaO₂/FiO₂, which are not routinely measured, and can be difficult to assign to arterial or venous samples; inconsistency in blood pressure and transient increases in vasopressor dose; the subjectivity of the Glasgow Coma Scale, which is also difficult to assess in sedated patients; and inconsistent urine output.

eSOFA introduced new measures for various organ functions, including cardiovascular (vasopressor initiation), pulmonary (mechanical ventilation initiation), renal (doubling of creatinine levels or a 50% or greater decrease in estimated glomerular filtration rate, compared with baseline), hepatic (bilirubin levels greater than or equal to 2.0 mg/dL and at least doubled from baseline), coagulation (platelet count less than 100 cells/mcL and at least a 50% decrease from a baseline of at least 100 cells/mcL), and neurological (lactate greater than or equal to 2.0 mmol/L).

“[eSOFA] opens a window into inter-facility comparisons that has not been possible to do. It’s really critical to ask, ‘How am I doing compared to my peer institutions?’ If you’re doing worse, you can look at the whole spectrum of things to try to drive improvements in care,” said Dr. Rhee.

The new tool isn’t just limited to quality improvement research. Shaeesta Khan, MD, assistant professor of critical care medicine at Geisinger Medical Center,Danville, Pa., has found eSOFA to be useful in her research into how genetic polymorphisms play a role in sepsis outcomes. Geisinger has a large population of patients with completed whole genome sequencing, and Dr. Khan began by trying to glean sepsis outcomes from administrative data.

“I explained SOFA scores to our data broker, and he pulled up 3,000 patients and gave everybody a SOFA score based on the algorithm he created, and it was all over the chart. Once I started doing chart review and phenotype verification, it was just a nightmare,” Dr. Khan said in an interview.

After struggling with the project, one of her mentors put her in touch with one of Dr. Rhee’s colleagues, and she asked the data broker to modify the eSOFA algorithm to fit her specific criteria. “It was a blessing,” she said.

Now, she has data from 5,000 patients with sepsis and sequenced DNA, and can begin comparing outcomes and genetic variants. About 20 candidate genes for sepsis outcomes have been identified to date, but she has a particular interest in PCSK9, which is an innate immune system regulator. She hopes to present results at CCC49 in 2020.
 

Validating mortality prediction

The researchers compared eSOFA and SOFA in a sample from 111 U.S. acute care hospitals to see if eSOFA had a comparable predictive validity for mortality. The analysis included 942,360 adults seen between 2013 and 2015. A total of 11.1% (104,903) had a presumed serious infection based on a blood culture order and at least 4 consecutive days of antibiotic use.

The analysis showed that 6.1% of those with infections had a sepsis event based on at least a 2-point increase in SOFA score from baseline (Sepsis-3 criteria), compared with 4.4% identified by at least a 1-point increase in eSOFA score. A total of 34,174 patients (3.6%) overlapped between SOFA and eSOFA, which represented good agreement (Cronbach’s alpha, 0.81). Compared with SOFA/Sepsis-3, eSOFA had a sensitivity of 60%, and a positive predictive value of 82%.

Patients identified by eSOFA were slightly more ill, with more requiring ICU admission (41% vs. 35%), and a greater frequency of in-hospital mortality (17% vs. 14%). Those patients who were identified by SOFA/Sepsis-3, but missed by eSOFA, had an overall lower mortality (6%).

There was a similar risk of mortality across deciles between SOFA- and eSOFA-identified sepsis patients. In an independent analysis of four hospitals from the Emory system, the area under the receiver operating characteristics was 0.77 for eSOFA and 0.76 for SOFA (P less than .001).

The Centers for Disease Control and Prevention and the Agency for Healthcare Research and Quality funded the study. Dr. Rhee and Dr. Khan have no relevant financial conflicts.
 

SOURCE: Rhee C et al. Crit Care Med. 2019;47(3):307-14.

SAN DIEGO – A simplified version of the Sequential Organ Failure Assessment (SOFA) criteria, known as eSOFA, has the potential to make it easier for hospitals to benchmark sepsis outcomes and quality of care, and could propel new sepsis research. The new method replaces some of SOFA’s more subjective criteria with objective measures.

eSOFA relies on electronic health records to reduce reliance on administrative records, which suffer from cross-hospital variability in diagnosis and coding practices, as well as changes in these practices over time. The diagnosis of sepsis itself is also highly subjective. Instead, eSOFA determines dysfunction in six organ systems, indicated by use of vasopressors and mechanical ventilation, and the presence of abnormal laboratory values.

“The SOFA score includes measures like the Glasgow Coma Scale, which undoubtedly at the bedside is a very important clinical sign, but when trying to implement something that is objective for purposes of retrospective case counting and standardization, it can be problematic. The measures we chose [for eSOFA] are concrete, important maneuvers that were initiated by clinicians,” Chanu Rhee, MD, said in an interview.

Dr. Rhee is assistant professor of population medicine at Harvard Medical School and Brigham and Women’s Hospital, Boston. He presented the results of the study at the Critical Care Congress sponsored by the Society of Critical Care Medicine, and the work was simultaneously published online in Critical Care Medicine.

Key elements of SOFA that pose challenges for administrative data include: PaO₂/FiO₂, which are not routinely measured, and can be difficult to assign to arterial or venous samples; inconsistency in blood pressure and transient increases in vasopressor dose; the subjectivity of the Glasgow Coma Scale, which is also difficult to assess in sedated patients; and inconsistent urine output.

eSOFA introduced new measures for various organ functions, including cardiovascular (vasopressor initiation), pulmonary (mechanical ventilation initiation), renal (doubling of creatinine levels or a 50% or greater decrease in estimated glomerular filtration rate, compared with baseline), hepatic (bilirubin levels greater than or equal to 2.0 mg/dL and at least doubled from baseline), coagulation (platelet count less than 100 cells/mcL and at least a 50% decrease from a baseline of at least 100 cells/mcL), and neurological (lactate greater than or equal to 2.0 mmol/L).

“[eSOFA] opens a window into inter-facility comparisons that has not been possible to do. It’s really critical to ask, ‘How am I doing compared to my peer institutions?’ If you’re doing worse, you can look at the whole spectrum of things to try to drive improvements in care,” said Dr. Rhee.

The new tool isn’t just limited to quality improvement research. Shaeesta Khan, MD, assistant professor of critical care medicine at Geisinger Medical Center,Danville, Pa., has found eSOFA to be useful in her research into how genetic polymorphisms play a role in sepsis outcomes. Geisinger has a large population of patients with completed whole genome sequencing, and Dr. Khan began by trying to glean sepsis outcomes from administrative data.

“I explained SOFA scores to our data broker, and he pulled up 3,000 patients and gave everybody a SOFA score based on the algorithm he created, and it was all over the chart. Once I started doing chart review and phenotype verification, it was just a nightmare,” Dr. Khan said in an interview.

After struggling with the project, one of her mentors put her in touch with one of Dr. Rhee’s colleagues, and she asked the data broker to modify the eSOFA algorithm to fit her specific criteria. “It was a blessing,” she said.

Now, she has data from 5,000 patients with sepsis and sequenced DNA, and can begin comparing outcomes and genetic variants. About 20 candidate genes for sepsis outcomes have been identified to date, but she has a particular interest in PCSK9, which is an innate immune system regulator. She hopes to present results at CCC49 in 2020.
 

Validating mortality prediction

The researchers compared eSOFA and SOFA in a sample from 111 U.S. acute care hospitals to see if eSOFA had a comparable predictive validity for mortality. The analysis included 942,360 adults seen between 2013 and 2015. A total of 11.1% (104,903) had a presumed serious infection based on a blood culture order and at least 4 consecutive days of antibiotic use.

The analysis showed that 6.1% of those with infections had a sepsis event based on at least a 2-point increase in SOFA score from baseline (Sepsis-3 criteria), compared with 4.4% identified by at least a 1-point increase in eSOFA score. A total of 34,174 patients (3.6%) overlapped between SOFA and eSOFA, which represented good agreement (Cronbach’s alpha, 0.81). Compared with SOFA/Sepsis-3, eSOFA had a sensitivity of 60%, and a positive predictive value of 82%.

Patients identified by eSOFA were slightly more ill, with more requiring ICU admission (41% vs. 35%), and a greater frequency of in-hospital mortality (17% vs. 14%). Those patients who were identified by SOFA/Sepsis-3, but missed by eSOFA, had an overall lower mortality (6%).

There was a similar risk of mortality across deciles between SOFA- and eSOFA-identified sepsis patients. In an independent analysis of four hospitals from the Emory system, the area under the receiver operating characteristics was 0.77 for eSOFA and 0.76 for SOFA (P less than .001).

The Centers for Disease Control and Prevention and the Agency for Healthcare Research and Quality funded the study. Dr. Rhee and Dr. Khan have no relevant financial conflicts.
 

SOURCE: Rhee C et al. Crit Care Med. 2019;47(3):307-14.

SAN DIEGO – A simplified version of the Sequential Organ Failure Assessment (SOFA) criteria, known as eSOFA, has the potential to make it easier for hospitals to benchmark sepsis outcomes and quality of care, and could propel new sepsis research. The new method replaces some of SOFA’s more subjective criteria with objective measures.

eSOFA relies on electronic health records to reduce reliance on administrative records, which suffer from cross-hospital variability in diagnosis and coding practices, as well as changes in these practices over time. The diagnosis of sepsis itself is also highly subjective. Instead, eSOFA determines dysfunction in six organ systems, indicated by use of vasopressors and mechanical ventilation, and the presence of abnormal laboratory values.

“The SOFA score includes measures like the Glasgow Coma Scale, which undoubtedly at the bedside is a very important clinical sign, but when trying to implement something that is objective for purposes of retrospective case counting and standardization, it can be problematic. The measures we chose [for eSOFA] are concrete, important maneuvers that were initiated by clinicians,” Chanu Rhee, MD, said in an interview.

Dr. Rhee is assistant professor of population medicine at Harvard Medical School and Brigham and Women’s Hospital, Boston. He presented the results of the study at the Critical Care Congress sponsored by the Society of Critical Care Medicine, and the work was simultaneously published online in Critical Care Medicine.

Key elements of SOFA that pose challenges for administrative data include: PaO₂/FiO₂, which are not routinely measured, and can be difficult to assign to arterial or venous samples; inconsistency in blood pressure and transient increases in vasopressor dose; the subjectivity of the Glasgow Coma Scale, which is also difficult to assess in sedated patients; and inconsistent urine output.

eSOFA introduced new measures for various organ functions, including cardiovascular (vasopressor initiation), pulmonary (mechanical ventilation initiation), renal (doubling of creatinine levels or a 50% or greater decrease in estimated glomerular filtration rate, compared with baseline), hepatic (bilirubin levels greater than or equal to 2.0 mg/dL and at least doubled from baseline), coagulation (platelet count less than 100 cells/mcL and at least a 50% decrease from a baseline of at least 100 cells/mcL), and neurological (lactate greater than or equal to 2.0 mmol/L).

“[eSOFA] opens a window into inter-facility comparisons that has not been possible to do. It’s really critical to ask, ‘How am I doing compared to my peer institutions?’ If you’re doing worse, you can look at the whole spectrum of things to try to drive improvements in care,” said Dr. Rhee.

The new tool isn’t just limited to quality improvement research. Shaeesta Khan, MD, assistant professor of critical care medicine at Geisinger Medical Center,Danville, Pa., has found eSOFA to be useful in her research into how genetic polymorphisms play a role in sepsis outcomes. Geisinger has a large population of patients with completed whole genome sequencing, and Dr. Khan began by trying to glean sepsis outcomes from administrative data.

“I explained SOFA scores to our data broker, and he pulled up 3,000 patients and gave everybody a SOFA score based on the algorithm he created, and it was all over the chart. Once I started doing chart review and phenotype verification, it was just a nightmare,” Dr. Khan said in an interview.

After struggling with the project, one of her mentors put her in touch with one of Dr. Rhee’s colleagues, and she asked the data broker to modify the eSOFA algorithm to fit her specific criteria. “It was a blessing,” she said.

Now, she has data from 5,000 patients with sepsis and sequenced DNA, and can begin comparing outcomes and genetic variants. About 20 candidate genes for sepsis outcomes have been identified to date, but she has a particular interest in PCSK9, which is an innate immune system regulator. She hopes to present results at CCC49 in 2020.
 

Validating mortality prediction

The researchers compared eSOFA and SOFA in a sample from 111 U.S. acute care hospitals to see if eSOFA had a comparable predictive validity for mortality. The analysis included 942,360 adults seen between 2013 and 2015. A total of 11.1% (104,903) had a presumed serious infection based on a blood culture order and at least 4 consecutive days of antibiotic use.

The analysis showed that 6.1% of those with infections had a sepsis event based on at least a 2-point increase in SOFA score from baseline (Sepsis-3 criteria), compared with 4.4% identified by at least a 1-point increase in eSOFA score. A total of 34,174 patients (3.6%) overlapped between SOFA and eSOFA, which represented good agreement (Cronbach’s alpha, 0.81). Compared with SOFA/Sepsis-3, eSOFA had a sensitivity of 60%, and a positive predictive value of 82%.

Patients identified by eSOFA were slightly more ill, with more requiring ICU admission (41% vs. 35%), and a greater frequency of in-hospital mortality (17% vs. 14%). Those patients who were identified by SOFA/Sepsis-3, but missed by eSOFA, had an overall lower mortality (6%).

There was a similar risk of mortality across deciles between SOFA- and eSOFA-identified sepsis patients. In an independent analysis of four hospitals from the Emory system, the area under the receiver operating characteristics was 0.77 for eSOFA and 0.76 for SOFA (P less than .001).

The Centers for Disease Control and Prevention and the Agency for Healthcare Research and Quality funded the study. Dr. Rhee and Dr. Khan have no relevant financial conflicts.
 

SOURCE: Rhee C et al. Crit Care Med. 2019;47(3):307-14.

SAN DIEGO – When attempting a second extubation, improvements in weaning parameters, compared with the first extubation attempt, do not predict success. Instead, the best predictors were the values of the parameters immediately before the second attempt.

“We hypothesized that the change in parameter values was more important than the actual values right before we tried to re-extubate, and that didn’t turn out to be the case. Because it was a smaller study, we can’t say [change in values] is not useful at all, but we didn’t find a strong association. We showed that the magnitude of the effect with the number measured right before the re-extubation is probably your best bet, but you should obviously evaluate the whole clinical scenario,” commented senior author Michael David Maile, MD, assistant professor of anesthesiology at the University of Michigan, Ann Arbor.

The study was presented at the Critical Care Congress sponsored by the Society of Critical Care Medicine by Suraj Trivedi, MD, who is currently an anesthesiology fellow at Stanford (Calif.) Medicine.

Factors such as rapid shallow breathing index (RSBI), negative inspiratory force (NIF), vital capacity (VC), and partial pressure of arterial carbon dioxide (PaCO₂) have been shown to predict success or failure of an initial extubation attempt.

There is currently little available guidance on how to proceed when a first extubation attempt fails. The researchers had anticipated that RSBI, NIF, VC, and PaCO₂ levels matching the first attempt would be associated with success the second time around.

But their retrospective study of adult patients at the University of Michigan critical care units found that only the change in RSBI values predicted success on a univariate analysis, and that association became statistically insignificant once they corrected for baseline RSBI previous to the re-extubation attempt.

“I think the biggest take-home message is that we have to figure out each attempt to extubate on its own merits. If you’re trying to extubate a patient in the ICU who has potentially been intubated and extubated multiple times, the clinical gut feeling is always that [the patient has] to be better off than the previous attempt. What we are pointing out is that it really doesn’t matter. If the parameters are all within the overall guidelines, it’s still okay to extubate, even if the absolute change in the variables is not better [than the previous attempt],” Dr. Trivedi said in an interview.

“People put a lot of emphasis on the improvement from the first to the second attempt, and this should temper that enthusiasm to put a lot of weight on the change. But I don’t think our data support that the change means nothing,” added Dr. Maile.

The study included 525 patients (42% female). Comorbidities were common: 72% had cardiac arrhythmias, 58% had hypertension, 33% had renal failure, 39% had a pulmonary disorder, and 25% had liver disease.

Univariate analyses showed associations between values of parameters immediately before the second extubation attempt and success in the second extubation attempt, including RSBI (re-extubation success, mean 53.1 vs failure, mean 68.8; P =.0002) and NIF (success, mean –41.2 vs. failure, mean –38.4; P =.036), and VC (success, mean 1009.8 vs. failure, mean 906.8; P =.017).

When the researchers examined changes in parameters between the first and second attempt, only a change in RSBI predicted success (success, value change of 7.1 vs. failure, value change of 0.05; P less than .031). But when they corrected for the RSBI value immediately before the second attempt, the difference was not statistically significant (P = .892).

The study was not funded. Dr. Maile and Dr. Trivedi have no relevant financial disclosures.

SOURCE: Trivedi S et al. CCC48 2019, Abstract 27.

SAN DIEGO – When attempting a second extubation, improvements in weaning parameters, compared with the first extubation attempt, do not predict success. Instead, the best predictors were the values of the parameters immediately before the second attempt.

“We hypothesized that the change in parameter values was more important than the actual values right before we tried to re-extubate, and that didn’t turn out to be the case. Because it was a smaller study, we can’t say [change in values] is not useful at all, but we didn’t find a strong association. We showed that the magnitude of the effect with the number measured right before the re-extubation is probably your best bet, but you should obviously evaluate the whole clinical scenario,” commented senior author Michael David Maile, MD, assistant professor of anesthesiology at the University of Michigan, Ann Arbor.

The study was presented at the Critical Care Congress sponsored by the Society of Critical Care Medicine by Suraj Trivedi, MD, who is currently an anesthesiology fellow at Stanford (Calif.) Medicine.

Factors such as rapid shallow breathing index (RSBI), negative inspiratory force (NIF), vital capacity (VC), and partial pressure of arterial carbon dioxide (PaCO₂) have been shown to predict success or failure of an initial extubation attempt.

There is currently little available guidance on how to proceed when a first extubation attempt fails. The researchers had anticipated that RSBI, NIF, VC, and PaCO₂ levels matching the first attempt would be associated with success the second time around.

But their retrospective study of adult patients at the University of Michigan critical care units found that only the change in RSBI values predicted success on a univariate analysis, and that association became statistically insignificant once they corrected for baseline RSBI previous to the re-extubation attempt.

“I think the biggest take-home message is that we have to figure out each attempt to extubate on its own merits. If you’re trying to extubate a patient in the ICU who has potentially been intubated and extubated multiple times, the clinical gut feeling is always that [the patient has] to be better off than the previous attempt. What we are pointing out is that it really doesn’t matter. If the parameters are all within the overall guidelines, it’s still okay to extubate, even if the absolute change in the variables is not better [than the previous attempt],” Dr. Trivedi said in an interview.

“People put a lot of emphasis on the improvement from the first to the second attempt, and this should temper that enthusiasm to put a lot of weight on the change. But I don’t think our data support that the change means nothing,” added Dr. Maile.

The study included 525 patients (42% female). Comorbidities were common: 72% had cardiac arrhythmias, 58% had hypertension, 33% had renal failure, 39% had a pulmonary disorder, and 25% had liver disease.

Univariate analyses showed associations between values of parameters immediately before the second extubation attempt and success in the second extubation attempt, including RSBI (re-extubation success, mean 53.1 vs failure, mean 68.8; P =.0002) and NIF (success, mean –41.2 vs. failure, mean –38.4; P =.036), and VC (success, mean 1009.8 vs. failure, mean 906.8; P =.017).

When the researchers examined changes in parameters between the first and second attempt, only a change in RSBI predicted success (success, value change of 7.1 vs. failure, value change of 0.05; P less than .031). But when they corrected for the RSBI value immediately before the second attempt, the difference was not statistically significant (P = .892).

The study was not funded. Dr. Maile and Dr. Trivedi have no relevant financial disclosures.

SOURCE: Trivedi S et al. CCC48 2019, Abstract 27.

SAN DIEGO – When attempting a second extubation, improvements in weaning parameters, compared with the first extubation attempt, do not predict success. Instead, the best predictors were the values of the parameters immediately before the second attempt.

“We hypothesized that the change in parameter values was more important than the actual values right before we tried to re-extubate, and that didn’t turn out to be the case. Because it was a smaller study, we can’t say [change in values] is not useful at all, but we didn’t find a strong association. We showed that the magnitude of the effect with the number measured right before the re-extubation is probably your best bet, but you should obviously evaluate the whole clinical scenario,” commented senior author Michael David Maile, MD, assistant professor of anesthesiology at the University of Michigan, Ann Arbor.

The study was presented at the Critical Care Congress sponsored by the Society of Critical Care Medicine by Suraj Trivedi, MD, who is currently an anesthesiology fellow at Stanford (Calif.) Medicine.

Factors such as rapid shallow breathing index (RSBI), negative inspiratory force (NIF), vital capacity (VC), and partial pressure of arterial carbon dioxide (PaCO₂) have been shown to predict success or failure of an initial extubation attempt.

There is currently little available guidance on how to proceed when a first extubation attempt fails. The researchers had anticipated that RSBI, NIF, VC, and PaCO₂ levels matching the first attempt would be associated with success the second time around.

But their retrospective study of adult patients at the University of Michigan critical care units found that only the change in RSBI values predicted success on a univariate analysis, and that association became statistically insignificant once they corrected for baseline RSBI previous to the re-extubation attempt.

“I think the biggest take-home message is that we have to figure out each attempt to extubate on its own merits. If you’re trying to extubate a patient in the ICU who has potentially been intubated and extubated multiple times, the clinical gut feeling is always that [the patient has] to be better off than the previous attempt. What we are pointing out is that it really doesn’t matter. If the parameters are all within the overall guidelines, it’s still okay to extubate, even if the absolute change in the variables is not better [than the previous attempt],” Dr. Trivedi said in an interview.

“People put a lot of emphasis on the improvement from the first to the second attempt, and this should temper that enthusiasm to put a lot of weight on the change. But I don’t think our data support that the change means nothing,” added Dr. Maile.

The study included 525 patients (42% female). Comorbidities were common: 72% had cardiac arrhythmias, 58% had hypertension, 33% had renal failure, 39% had a pulmonary disorder, and 25% had liver disease.

Univariate analyses showed associations between values of parameters immediately before the second extubation attempt and success in the second extubation attempt, including RSBI (re-extubation success, mean 53.1 vs failure, mean 68.8; P =.0002) and NIF (success, mean –41.2 vs. failure, mean –38.4; P =.036), and VC (success, mean 1009.8 vs. failure, mean 906.8; P =.017).

When the researchers examined changes in parameters between the first and second attempt, only a change in RSBI predicted success (success, value change of 7.1 vs. failure, value change of 0.05; P less than .031). But when they corrected for the RSBI value immediately before the second attempt, the difference was not statistically significant (P = .892).

The study was not funded. Dr. Maile and Dr. Trivedi have no relevant financial disclosures.

SOURCE: Trivedi S et al. CCC48 2019, Abstract 27.

Delaying or withholding antibiotics in elderly patients with a urinary tract infection (UTI) increases the risk of sepsis and death, results of a large, population-based study suggest.

andresr/Getty Images

The risk of bloodstream infection was more than seven times greater in patients who did not receive antibiotics immediately after seeing a general practitioner for a UTI versus those who did, according to results of the study based on primary care records and other data for nearly 160,000 U.K. patients aged 65 years or older. Death rates and hospital admissions were significantly higher for these patients, according to the study published in The BMJ by Myriam Gharbi, PharmD, Phd, Imperial College London, and her colleagues.

The publication of these findings coincides with an increase in Escherichia coli bloodstream infections in England.

“Our study suggests the early initiation of antibiotics for UTI in older high risk adult populations (especially men aged [older than] 85 years) should be recommended to prevent serious complications,” Dr. Gharbi and her coauthors said in their report.

The population-based cohort study comprised 157,264 adult primary care patients at least 65 years of age who had one or more suspected or confirmed lower UTIs from November 2007 to May 2015. The researchers found that health care providers had diagnosed a total of 312,896 UTI episodes in these patients during the period they studied. In 7.2% (22,534) of the UTI episodes, the researchers were unable to find records of the patients having been prescribed antibiotics by a general practitioner within 7 days of the UTI diagnosis. These 22,534 episodes included those that occurred in patients who had a complication before an antibiotic was prescribed. An additional 6.2% (19,292) of the episodes occurred in patients who were prescribed antibiotics, but not during their first UTI-related visit to a general practitioner or on the same day of such a visit. The researchers classified this group of patients as having been prescribed antibiotics on a deferred or delayed basis, as they were not prescribed such drugs within 7 days of their visit.

Overall, there were 1,539 cases (0.5% of the total number of UTIs) of bloodstream infection within 60 days of the initial urinary tract infection diagnosis, the researchers reported.

The bloodstream infection rate was 2.9% for patients who were not prescribed antibiotics ever or prior to an infection occurring, 2.2% in those who were prescribed antibiotics on a deferred basis, and 0.2% in those who were prescribed antibiotics immediately, meaning during their first visit to a general practitioner for a UTI or on the same day of such a visit (P less than .001). After adjustment for potential confounding variables such as age, sex, and region, the patients classified as having not been prescribed antibiotics or having been prescribed antibiotics on a deferred basis were significantly more likely to have a bloodstream infection within 60 days of their visit to a health care provider, compared with those who received antibiotics immediately, with odds ratios of 8.08 (95% confidence interval, 7.12-9.16) and 7.12 (95% CI, 6.22-8.14), respectively.

Hospital admissions after a UTI episode were nearly twice as high in the no- or deferred-antibiotics groups (27.0% and 26.8%, respectively), compared with the group that received antibiotics right away (14.8%), the investigators reported. The lengths of hospital stays were 12.1 days for the group classified as having not been prescribed antibiotics, 7.7 days for the group subject to delayed antibiotic prescribing, and 6.3 days for the group who received antibiotics immediately.

Deaths within 60 days of experiencing a urinary tract infection occurred in 5.4% of patients in the no-antibiotics group, 2.8% of the deferred-antibiotics group, and 1.6% of the immediate-antibiotics group. After adjustment for covariates, a regression analysis showed the risks for all-cause mortality were 1.16 and 2.18 times higher in the deferred-antibiotics group and the no-antibiotics group, respectively, according to the paper.

In the immediate-antibiotics group, those patients who received nitrofurantoin had a “small but significant increase” in 60-day survival versus those who received trimethoprim, the investigators noted in the discussion section of their report.

“This increase could reflect either higher levels of resistance to trimethoprim or a healthier population treated with nitrofurantoin, the latest being not recommended for patients with poor kidney function,” the researchers wrote.

This study was supported by the National Institute for Health Research and other U.K. sources. One study coauthor reported working as an epidemiologist with GSK in areas not related to the study.

SOURCE: Gharbi M et al. BMJ. 2019 Feb 27. doi: 10.1136/bmj.l525.

Delaying or withholding antibiotics in elderly patients with a urinary tract infection (UTI) increases the risk of sepsis and death, results of a large, population-based study suggest.

andresr/Getty Images

The risk of bloodstream infection was more than seven times greater in patients who did not receive antibiotics immediately after seeing a general practitioner for a UTI versus those who did, according to results of the study based on primary care records and other data for nearly 160,000 U.K. patients aged 65 years or older. Death rates and hospital admissions were significantly higher for these patients, according to the study published in The BMJ by Myriam Gharbi, PharmD, Phd, Imperial College London, and her colleagues.

The publication of these findings coincides with an increase in Escherichia coli bloodstream infections in England.

“Our study suggests the early initiation of antibiotics for UTI in older high risk adult populations (especially men aged [older than] 85 years) should be recommended to prevent serious complications,” Dr. Gharbi and her coauthors said in their report.

The population-based cohort study comprised 157,264 adult primary care patients at least 65 years of age who had one or more suspected or confirmed lower UTIs from November 2007 to May 2015. The researchers found that health care providers had diagnosed a total of 312,896 UTI episodes in these patients during the period they studied. In 7.2% (22,534) of the UTI episodes, the researchers were unable to find records of the patients having been prescribed antibiotics by a general practitioner within 7 days of the UTI diagnosis. These 22,534 episodes included those that occurred in patients who had a complication before an antibiotic was prescribed. An additional 6.2% (19,292) of the episodes occurred in patients who were prescribed antibiotics, but not during their first UTI-related visit to a general practitioner or on the same day of such a visit. The researchers classified this group of patients as having been prescribed antibiotics on a deferred or delayed basis, as they were not prescribed such drugs within 7 days of their visit.

Overall, there were 1,539 cases (0.5% of the total number of UTIs) of bloodstream infection within 60 days of the initial urinary tract infection diagnosis, the researchers reported.

The bloodstream infection rate was 2.9% for patients who were not prescribed antibiotics ever or prior to an infection occurring, 2.2% in those who were prescribed antibiotics on a deferred basis, and 0.2% in those who were prescribed antibiotics immediately, meaning during their first visit to a general practitioner for a UTI or on the same day of such a visit (P less than .001). After adjustment for potential confounding variables such as age, sex, and region, the patients classified as having not been prescribed antibiotics or having been prescribed antibiotics on a deferred basis were significantly more likely to have a bloodstream infection within 60 days of their visit to a health care provider, compared with those who received antibiotics immediately, with odds ratios of 8.08 (95% confidence interval, 7.12-9.16) and 7.12 (95% CI, 6.22-8.14), respectively.

Hospital admissions after a UTI episode were nearly twice as high in the no- or deferred-antibiotics groups (27.0% and 26.8%, respectively), compared with the group that received antibiotics right away (14.8%), the investigators reported. The lengths of hospital stays were 12.1 days for the group classified as having not been prescribed antibiotics, 7.7 days for the group subject to delayed antibiotic prescribing, and 6.3 days for the group who received antibiotics immediately.

Deaths within 60 days of experiencing a urinary tract infection occurred in 5.4% of patients in the no-antibiotics group, 2.8% of the deferred-antibiotics group, and 1.6% of the immediate-antibiotics group. After adjustment for covariates, a regression analysis showed the risks for all-cause mortality were 1.16 and 2.18 times higher in the deferred-antibiotics group and the no-antibiotics group, respectively, according to the paper.

In the immediate-antibiotics group, those patients who received nitrofurantoin had a “small but significant increase” in 60-day survival versus those who received trimethoprim, the investigators noted in the discussion section of their report.

“This increase could reflect either higher levels of resistance to trimethoprim or a healthier population treated with nitrofurantoin, the latest being not recommended for patients with poor kidney function,” the researchers wrote.

This study was supported by the National Institute for Health Research and other U.K. sources. One study coauthor reported working as an epidemiologist with GSK in areas not related to the study.

SOURCE: Gharbi M et al. BMJ. 2019 Feb 27. doi: 10.1136/bmj.l525.

Delaying or withholding antibiotics in elderly patients with a urinary tract infection (UTI) increases the risk of sepsis and death, results of a large, population-based study suggest.

andresr/Getty Images

The risk of bloodstream infection was more than seven times greater in patients who did not receive antibiotics immediately after seeing a general practitioner for a UTI versus those who did, according to results of the study based on primary care records and other data for nearly 160,000 U.K. patients aged 65 years or older. Death rates and hospital admissions were significantly higher for these patients, according to the study published in The BMJ by Myriam Gharbi, PharmD, Phd, Imperial College London, and her colleagues.

The publication of these findings coincides with an increase in Escherichia coli bloodstream infections in England.

“Our study suggests the early initiation of antibiotics for UTI in older high risk adult populations (especially men aged [older than] 85 years) should be recommended to prevent serious complications,” Dr. Gharbi and her coauthors said in their report.

The population-based cohort study comprised 157,264 adult primary care patients at least 65 years of age who had one or more suspected or confirmed lower UTIs from November 2007 to May 2015. The researchers found that health care providers had diagnosed a total of 312,896 UTI episodes in these patients during the period they studied. In 7.2% (22,534) of the UTI episodes, the researchers were unable to find records of the patients having been prescribed antibiotics by a general practitioner within 7 days of the UTI diagnosis. These 22,534 episodes included those that occurred in patients who had a complication before an antibiotic was prescribed. An additional 6.2% (19,292) of the episodes occurred in patients who were prescribed antibiotics, but not during their first UTI-related visit to a general practitioner or on the same day of such a visit. The researchers classified this group of patients as having been prescribed antibiotics on a deferred or delayed basis, as they were not prescribed such drugs within 7 days of their visit.

Overall, there were 1,539 cases (0.5% of the total number of UTIs) of bloodstream infection within 60 days of the initial urinary tract infection diagnosis, the researchers reported.

The bloodstream infection rate was 2.9% for patients who were not prescribed antibiotics ever or prior to an infection occurring, 2.2% in those who were prescribed antibiotics on a deferred basis, and 0.2% in those who were prescribed antibiotics immediately, meaning during their first visit to a general practitioner for a UTI or on the same day of such a visit (P less than .001). After adjustment for potential confounding variables such as age, sex, and region, the patients classified as having not been prescribed antibiotics or having been prescribed antibiotics on a deferred basis were significantly more likely to have a bloodstream infection within 60 days of their visit to a health care provider, compared with those who received antibiotics immediately, with odds ratios of 8.08 (95% confidence interval, 7.12-9.16) and 7.12 (95% CI, 6.22-8.14), respectively.

Hospital admissions after a UTI episode were nearly twice as high in the no- or deferred-antibiotics groups (27.0% and 26.8%, respectively), compared with the group that received antibiotics right away (14.8%), the investigators reported. The lengths of hospital stays were 12.1 days for the group classified as having not been prescribed antibiotics, 7.7 days for the group subject to delayed antibiotic prescribing, and 6.3 days for the group who received antibiotics immediately.

Deaths within 60 days of experiencing a urinary tract infection occurred in 5.4% of patients in the no-antibiotics group, 2.8% of the deferred-antibiotics group, and 1.6% of the immediate-antibiotics group. After adjustment for covariates, a regression analysis showed the risks for all-cause mortality were 1.16 and 2.18 times higher in the deferred-antibiotics group and the no-antibiotics group, respectively, according to the paper.

In the immediate-antibiotics group, those patients who received nitrofurantoin had a “small but significant increase” in 60-day survival versus those who received trimethoprim, the investigators noted in the discussion section of their report.

“This increase could reflect either higher levels of resistance to trimethoprim or a healthier population treated with nitrofurantoin, the latest being not recommended for patients with poor kidney function,” the researchers wrote.

This study was supported by the National Institute for Health Research and other U.K. sources. One study coauthor reported working as an epidemiologist with GSK in areas not related to the study.

SOURCE: Gharbi M et al. BMJ. 2019 Feb 27. doi: 10.1136/bmj.l525.

SAN DIEGO – A new study of ICU patients in the Netherlands shows a heightened risk of developing new chronic conditions in patients after an intensive care stay. The research showed rising likelihood of conditions such as depression, diabetes, and heart disease.

By merging two existing databases, the researchers were able to capture a more comprehensive picture of post-ICU patients. “We were able to include almost the entire country,” Ilse van Beusekom, a PhD candidate in health sciences at the University of Amsterdam and data manager at the National Intensive Care Evaluation (NICE) foundation, said in an interview.

Ms. van Beusekom presented the study at the Critical Care Congress sponsored by the Society of Critical Care Medicine. The study was simultaneously published in Critical Care Medicine.

The work compared 56,760 ICU survivors from 81 facilities across the Netherlands to 75,232 age-, sex-, and socioeconomic status–matched controls. The mean age was 65 years and 60% of the population was male. “The types of chronic conditions are the same, only the prevalences are different,” said Ms. van Beusekom.

The researchers compared chronic conditions in the year before ICU admission and the year after, based on data pulled from the NICE national quality database, which includes data describing the first 24 hours of ICU admission, and the Vektis insurance claims database, which includes information on medical treatment. Before ICU admission, 45% of the ICU population was free of chronic conditions, as were 62% of controls. One chronic condition was present in 36% of ICU patients, versus 29% of controls, and two or more conditions were present in 19% versus 9% of controls.

The ICU population was more likely to have high cholesterol (16% vs. 14%), heart disease (14% vs. 6%), chronic obstructive pulmonary disease (8% vs. 3%), type II diabetes (8% vs. 6%), type I diabetes (6% vs. 3%), and depression (6% vs. 4%).

The ICU population also was at greater risk of developing one or more new chronic conditions during the year following their stay. The risk was three- to fourfold higher throughout age ranges.

The study suggests the need for greater follow-up after an ICU admission in order to help patients cope with lingering problems. Ms. van Beusekom noted that there are follow-up programs in the Netherlands for several patient groups, but none for ICU survivors. One possibility would be to have the patient return to the ICU 3 months or so after release to discuss their diagnosis, treatment, and any lingering concerns. “A lot of people don’t know that their complaints are linked with the ICU visit,” said Ms. van Beusekom.

Timothy G. Buchman, MD, professor of surgery at Emory University, Atlanta, who moderated the session, wondered why the ICU seems to be an inflection point for developing new chronic conditions. Could it simply be because patients are sicker to begin with and have reached an inflection point of their illness, or could the treatments in ICU be contributing to or exposing those conditions? Ms. van Beusekom believed it was likely a combination of factors, and she referred to data she had not presented showing that even control patients who had been to the hospital (though not the ICU) during the study period were at lower risk of new chronic conditions than ICU patients.

Ms. van Beusekom’s group plans to investigate ICU-related variables that might be associated with risk of chronic conditions.

The study was not funded. Ms. van Beusekom had no relevant disclosures.

SOURCE: van Beusekom I et al. CCC48, Abstract Crit Care Med. 2019;47:324-30.

SAN DIEGO – A new study of ICU patients in the Netherlands shows a heightened risk of developing new chronic conditions in patients after an intensive care stay. The research showed rising likelihood of conditions such as depression, diabetes, and heart disease.

By merging two existing databases, the researchers were able to capture a more comprehensive picture of post-ICU patients. “We were able to include almost the entire country,” Ilse van Beusekom, a PhD candidate in health sciences at the University of Amsterdam and data manager at the National Intensive Care Evaluation (NICE) foundation, said in an interview.

Ms. van Beusekom presented the study at the Critical Care Congress sponsored by the Society of Critical Care Medicine. The study was simultaneously published in Critical Care Medicine.

The work compared 56,760 ICU survivors from 81 facilities across the Netherlands to 75,232 age-, sex-, and socioeconomic status–matched controls. The mean age was 65 years and 60% of the population was male. “The types of chronic conditions are the same, only the prevalences are different,” said Ms. van Beusekom.

The researchers compared chronic conditions in the year before ICU admission and the year after, based on data pulled from the NICE national quality database, which includes data describing the first 24 hours of ICU admission, and the Vektis insurance claims database, which includes information on medical treatment. Before ICU admission, 45% of the ICU population was free of chronic conditions, as were 62% of controls. One chronic condition was present in 36% of ICU patients, versus 29% of controls, and two or more conditions were present in 19% versus 9% of controls.

The ICU population was more likely to have high cholesterol (16% vs. 14%), heart disease (14% vs. 6%), chronic obstructive pulmonary disease (8% vs. 3%), type II diabetes (8% vs. 6%), type I diabetes (6% vs. 3%), and depression (6% vs. 4%).

The ICU population also was at greater risk of developing one or more new chronic conditions during the year following their stay. The risk was three- to fourfold higher throughout age ranges.

The study suggests the need for greater follow-up after an ICU admission in order to help patients cope with lingering problems. Ms. van Beusekom noted that there are follow-up programs in the Netherlands for several patient groups, but none for ICU survivors. One possibility would be to have the patient return to the ICU 3 months or so after release to discuss their diagnosis, treatment, and any lingering concerns. “A lot of people don’t know that their complaints are linked with the ICU visit,” said Ms. van Beusekom.

Timothy G. Buchman, MD, professor of surgery at Emory University, Atlanta, who moderated the session, wondered why the ICU seems to be an inflection point for developing new chronic conditions. Could it simply be because patients are sicker to begin with and have reached an inflection point of their illness, or could the treatments in ICU be contributing to or exposing those conditions? Ms. van Beusekom believed it was likely a combination of factors, and she referred to data she had not presented showing that even control patients who had been to the hospital (though not the ICU) during the study period were at lower risk of new chronic conditions than ICU patients.

Ms. van Beusekom’s group plans to investigate ICU-related variables that might be associated with risk of chronic conditions.

The study was not funded. Ms. van Beusekom had no relevant disclosures.

SOURCE: van Beusekom I et al. CCC48, Abstract Crit Care Med. 2019;47:324-30.

SAN DIEGO – A new study of ICU patients in the Netherlands shows a heightened risk of developing new chronic conditions in patients after an intensive care stay. The research showed rising likelihood of conditions such as depression, diabetes, and heart disease.

By merging two existing databases, the researchers were able to capture a more comprehensive picture of post-ICU patients. “We were able to include almost the entire country,” Ilse van Beusekom, a PhD candidate in health sciences at the University of Amsterdam and data manager at the National Intensive Care Evaluation (NICE) foundation, said in an interview.

Ms. van Beusekom presented the study at the Critical Care Congress sponsored by the Society of Critical Care Medicine. The study was simultaneously published in Critical Care Medicine.

The work compared 56,760 ICU survivors from 81 facilities across the Netherlands to 75,232 age-, sex-, and socioeconomic status–matched controls. The mean age was 65 years and 60% of the population was male. “The types of chronic conditions are the same, only the prevalences are different,” said Ms. van Beusekom.

The researchers compared chronic conditions in the year before ICU admission and the year after, based on data pulled from the NICE national quality database, which includes data describing the first 24 hours of ICU admission, and the Vektis insurance claims database, which includes information on medical treatment. Before ICU admission, 45% of the ICU population was free of chronic conditions, as were 62% of controls. One chronic condition was present in 36% of ICU patients, versus 29% of controls, and two or more conditions were present in 19% versus 9% of controls.

The ICU population was more likely to have high cholesterol (16% vs. 14%), heart disease (14% vs. 6%), chronic obstructive pulmonary disease (8% vs. 3%), type II diabetes (8% vs. 6%), type I diabetes (6% vs. 3%), and depression (6% vs. 4%).

The ICU population also was at greater risk of developing one or more new chronic conditions during the year following their stay. The risk was three- to fourfold higher throughout age ranges.

The study suggests the need for greater follow-up after an ICU admission in order to help patients cope with lingering problems. Ms. van Beusekom noted that there are follow-up programs in the Netherlands for several patient groups, but none for ICU survivors. One possibility would be to have the patient return to the ICU 3 months or so after release to discuss their diagnosis, treatment, and any lingering concerns. “A lot of people don’t know that their complaints are linked with the ICU visit,” said Ms. van Beusekom.

Timothy G. Buchman, MD, professor of surgery at Emory University, Atlanta, who moderated the session, wondered why the ICU seems to be an inflection point for developing new chronic conditions. Could it simply be because patients are sicker to begin with and have reached an inflection point of their illness, or could the treatments in ICU be contributing to or exposing those conditions? Ms. van Beusekom believed it was likely a combination of factors, and she referred to data she had not presented showing that even control patients who had been to the hospital (though not the ICU) during the study period were at lower risk of new chronic conditions than ICU patients.

Ms. van Beusekom’s group plans to investigate ICU-related variables that might be associated with risk of chronic conditions.

The study was not funded. Ms. van Beusekom had no relevant disclosures.

SOURCE: van Beusekom I et al. CCC48, Abstract Crit Care Med. 2019;47:324-30.