Clinical

Vaping-associated lung injury cases have now reached 1,479, according to the latest update provided by the Centers for Disease Control and Prevention. Thirty-three deaths have been confirmed.

Carpe89/ThinkStock

E-cigarette–linked lung injuries, now called EVALI, occurred in all U.S. states (except Alaska), the District of Columbia, and the U.S. Virgin Islands. Seventy percent of patients are male, and 79% are under age 35 years.

Information on the substances used over the previous 3 months before symptom onset was available for 849 patients and included the following:

• 78% reported using THC-containing products, with or without nicotine-containing products;
• 31% reported exclusive use of THC-containing products;
• 58% reported using nicotine-containing products, with or without THC-containing products; and
• 10% reported exclusive use of nicotine-containing products.

CDC is now doing additional testing on available samples for chemical in the bronchoalveolar lavage fluid, blood, or urine, as well as lung biopsy or autopsy specimens. CDC is also validating methods for aerosol emission testing of case-associated product samples from vaping products and e-liquids.

In a related development, JUUL, maker of e-cigarette products, has announced that it will suspend the sale of nontobacco, nonmenthol flavors (mango, creme, fruit, and cucumber) in the United States, pending review by the Food and Drug Administration. The JUUL announcement comes in advance of an expected FDA ban on flavored e-cigarettes.

The CDC continues its investigation into EVALI but stated, “Since the specific cause or causes of lung injury are not yet known, the only way to assure that you are not at risk while the investigation continues is to consider refraining from use of all e-cigarette, or vaping, products.”

For more information and resources visit For the Public, For Healthcare Providers, and For State and Local Health Departments pages, as well as the CDC’s Publications and Resources page.

tborden@mdedge.com

Vaping-associated lung injury cases have now reached 1,479, according to the latest update provided by the Centers for Disease Control and Prevention. Thirty-three deaths have been confirmed.

Carpe89/ThinkStock

E-cigarette–linked lung injuries, now called EVALI, occurred in all U.S. states (except Alaska), the District of Columbia, and the U.S. Virgin Islands. Seventy percent of patients are male, and 79% are under age 35 years.

Information on the substances used over the previous 3 months before symptom onset was available for 849 patients and included the following:

• 78% reported using THC-containing products, with or without nicotine-containing products;
• 31% reported exclusive use of THC-containing products;
• 58% reported using nicotine-containing products, with or without THC-containing products; and
• 10% reported exclusive use of nicotine-containing products.

CDC is now doing additional testing on available samples for chemical in the bronchoalveolar lavage fluid, blood, or urine, as well as lung biopsy or autopsy specimens. CDC is also validating methods for aerosol emission testing of case-associated product samples from vaping products and e-liquids.

In a related development, JUUL, maker of e-cigarette products, has announced that it will suspend the sale of nontobacco, nonmenthol flavors (mango, creme, fruit, and cucumber) in the United States, pending review by the Food and Drug Administration. The JUUL announcement comes in advance of an expected FDA ban on flavored e-cigarettes.

The CDC continues its investigation into EVALI but stated, “Since the specific cause or causes of lung injury are not yet known, the only way to assure that you are not at risk while the investigation continues is to consider refraining from use of all e-cigarette, or vaping, products.”

For more information and resources visit For the Public, For Healthcare Providers, and For State and Local Health Departments pages, as well as the CDC’s Publications and Resources page.

tborden@mdedge.com

Vaping-associated lung injury cases have now reached 1,479, according to the latest update provided by the Centers for Disease Control and Prevention. Thirty-three deaths have been confirmed.

Carpe89/ThinkStock

E-cigarette–linked lung injuries, now called EVALI, occurred in all U.S. states (except Alaska), the District of Columbia, and the U.S. Virgin Islands. Seventy percent of patients are male, and 79% are under age 35 years.

Information on the substances used over the previous 3 months before symptom onset was available for 849 patients and included the following:

• 78% reported using THC-containing products, with or without nicotine-containing products;
• 31% reported exclusive use of THC-containing products;
• 58% reported using nicotine-containing products, with or without THC-containing products; and
• 10% reported exclusive use of nicotine-containing products.

CDC is now doing additional testing on available samples for chemical in the bronchoalveolar lavage fluid, blood, or urine, as well as lung biopsy or autopsy specimens. CDC is also validating methods for aerosol emission testing of case-associated product samples from vaping products and e-liquids.

In a related development, JUUL, maker of e-cigarette products, has announced that it will suspend the sale of nontobacco, nonmenthol flavors (mango, creme, fruit, and cucumber) in the United States, pending review by the Food and Drug Administration. The JUUL announcement comes in advance of an expected FDA ban on flavored e-cigarettes.

The CDC continues its investigation into EVALI but stated, “Since the specific cause or causes of lung injury are not yet known, the only way to assure that you are not at risk while the investigation continues is to consider refraining from use of all e-cigarette, or vaping, products.”

For more information and resources visit For the Public, For Healthcare Providers, and For State and Local Health Departments pages, as well as the CDC’s Publications and Resources page.

tborden@mdedge.com

ST. LOUIS – The lower the blood pressure peaks and variability in the first 24 hours after ischemic stroke reperfusion, the better the outcomes, according to a review of 140 patients at the University of New Mexico, Albuquerque. Investigators found that every 10–mm Hg increase in peak systolic pressure boosted the risk of in-hospital death 24% (P = .01) and reduced the chance of being discharged home or to a inpatient rehabilitation facility 13% (P = .03). Results were even stronger for peak mean arterial pressure, at 76% (P = .01) and 29% (P = .04), respectively; trends in the same direction for peak diastolic pressure were not statistically significant.

Also, every 10–mm Hg increase in blood pressure variability again increased the risk of dying in the hospital, whether it was systolic (33%; P = .002), diastolic (33%; P = .03), or mean arterial pressure variability (58%; P = .02). Higher variability also reduced the chance of being discharged home or to a rehab 10%-20%, but the findings, although close, were not statistically significant.

Neurologists generally do what they can to control blood pressure after stroke, and the study confirms the need to do that. What’s new is that the work was limited to reperfusion patients – intravenous thrombolysis with alteplase in 83.5%, mechanical thrombectomy in 60%, with some having both – which has not been the specific focus of much research.

“Be much more aggressive in terms of making sure the variability is limited and limiting the peaks,” especially within 24 hours of reperfusion, said lead investigator and stroke neurologist Dinesh Jillella, MD, of Emory University, Atlanta, at the annual meeting of the American Neurological Association. “We want to be much more aggressive [with these patients]; it might limit our worse outcomes,” Dr. Jillella said. He conducted the review while in training at the University of New Mexico.

What led to the study is that Dr. Jillella and colleagues noticed that similar reperfusion patients can have very different outcomes, and he wanted to find modifiable risk factors that could account for the differences. The study did not address why high peaks and variability lead to worse outcomes, but he said hemorrhagic conversion might play a role.

It is also possible that higher pressures could be a marker of bad outcomes, as opposed to a direct cause, but the findings were adjusted for two significant confounders: age and the National Institutes of Health Stroke Scale score, which were both significantly higher in patients who did not do well. But after adjustment, “we [still] found an independent association with blood pressures and worse outcomes,” he said.

Higher peak systolic pressures and variability were also associated with about a 15% lower odds of leaving the hospital with a modified Rankin Scale score of 3 or less, which means the patient has some moderate disability but is still able to walk without assistance.

Patients were 69 years old on average, and about 60% were men. The majority were white. About a third had a modified Rankin Scale score at or below 3 at discharge, and about two-thirds were discharged home or to a rehabilitation facility; 17% of patients died in the hospital.

Differences in antihypertensive regimens were not associated with outcomes on univariate analysis. Dr. Jillella said that, ideally, he would like to run a multicenter, prospective trial of blood pressure reduction targets after reperfusion.

There was no external funding, and Dr. Jillella didn’t have any relevant disclosures.
 

aotto@mdedge.com

ST. LOUIS – The lower the blood pressure peaks and variability in the first 24 hours after ischemic stroke reperfusion, the better the outcomes, according to a review of 140 patients at the University of New Mexico, Albuquerque. Investigators found that every 10–mm Hg increase in peak systolic pressure boosted the risk of in-hospital death 24% (P = .01) and reduced the chance of being discharged home or to a inpatient rehabilitation facility 13% (P = .03). Results were even stronger for peak mean arterial pressure, at 76% (P = .01) and 29% (P = .04), respectively; trends in the same direction for peak diastolic pressure were not statistically significant.

Also, every 10–mm Hg increase in blood pressure variability again increased the risk of dying in the hospital, whether it was systolic (33%; P = .002), diastolic (33%; P = .03), or mean arterial pressure variability (58%; P = .02). Higher variability also reduced the chance of being discharged home or to a rehab 10%-20%, but the findings, although close, were not statistically significant.

Neurologists generally do what they can to control blood pressure after stroke, and the study confirms the need to do that. What’s new is that the work was limited to reperfusion patients – intravenous thrombolysis with alteplase in 83.5%, mechanical thrombectomy in 60%, with some having both – which has not been the specific focus of much research.

“Be much more aggressive in terms of making sure the variability is limited and limiting the peaks,” especially within 24 hours of reperfusion, said lead investigator and stroke neurologist Dinesh Jillella, MD, of Emory University, Atlanta, at the annual meeting of the American Neurological Association. “We want to be much more aggressive [with these patients]; it might limit our worse outcomes,” Dr. Jillella said. He conducted the review while in training at the University of New Mexico.

What led to the study is that Dr. Jillella and colleagues noticed that similar reperfusion patients can have very different outcomes, and he wanted to find modifiable risk factors that could account for the differences. The study did not address why high peaks and variability lead to worse outcomes, but he said hemorrhagic conversion might play a role.

It is also possible that higher pressures could be a marker of bad outcomes, as opposed to a direct cause, but the findings were adjusted for two significant confounders: age and the National Institutes of Health Stroke Scale score, which were both significantly higher in patients who did not do well. But after adjustment, “we [still] found an independent association with blood pressures and worse outcomes,” he said.

Higher peak systolic pressures and variability were also associated with about a 15% lower odds of leaving the hospital with a modified Rankin Scale score of 3 or less, which means the patient has some moderate disability but is still able to walk without assistance.

Patients were 69 years old on average, and about 60% were men. The majority were white. About a third had a modified Rankin Scale score at or below 3 at discharge, and about two-thirds were discharged home or to a rehabilitation facility; 17% of patients died in the hospital.

Differences in antihypertensive regimens were not associated with outcomes on univariate analysis. Dr. Jillella said that, ideally, he would like to run a multicenter, prospective trial of blood pressure reduction targets after reperfusion.

There was no external funding, and Dr. Jillella didn’t have any relevant disclosures.
 

aotto@mdedge.com

ST. LOUIS – The lower the blood pressure peaks and variability in the first 24 hours after ischemic stroke reperfusion, the better the outcomes, according to a review of 140 patients at the University of New Mexico, Albuquerque. Investigators found that every 10–mm Hg increase in peak systolic pressure boosted the risk of in-hospital death 24% (P = .01) and reduced the chance of being discharged home or to a inpatient rehabilitation facility 13% (P = .03). Results were even stronger for peak mean arterial pressure, at 76% (P = .01) and 29% (P = .04), respectively; trends in the same direction for peak diastolic pressure were not statistically significant.

Also, every 10–mm Hg increase in blood pressure variability again increased the risk of dying in the hospital, whether it was systolic (33%; P = .002), diastolic (33%; P = .03), or mean arterial pressure variability (58%; P = .02). Higher variability also reduced the chance of being discharged home or to a rehab 10%-20%, but the findings, although close, were not statistically significant.

Neurologists generally do what they can to control blood pressure after stroke, and the study confirms the need to do that. What’s new is that the work was limited to reperfusion patients – intravenous thrombolysis with alteplase in 83.5%, mechanical thrombectomy in 60%, with some having both – which has not been the specific focus of much research.

“Be much more aggressive in terms of making sure the variability is limited and limiting the peaks,” especially within 24 hours of reperfusion, said lead investigator and stroke neurologist Dinesh Jillella, MD, of Emory University, Atlanta, at the annual meeting of the American Neurological Association. “We want to be much more aggressive [with these patients]; it might limit our worse outcomes,” Dr. Jillella said. He conducted the review while in training at the University of New Mexico.

What led to the study is that Dr. Jillella and colleagues noticed that similar reperfusion patients can have very different outcomes, and he wanted to find modifiable risk factors that could account for the differences. The study did not address why high peaks and variability lead to worse outcomes, but he said hemorrhagic conversion might play a role.

It is also possible that higher pressures could be a marker of bad outcomes, as opposed to a direct cause, but the findings were adjusted for two significant confounders: age and the National Institutes of Health Stroke Scale score, which were both significantly higher in patients who did not do well. But after adjustment, “we [still] found an independent association with blood pressures and worse outcomes,” he said.

Higher peak systolic pressures and variability were also associated with about a 15% lower odds of leaving the hospital with a modified Rankin Scale score of 3 or less, which means the patient has some moderate disability but is still able to walk without assistance.

Patients were 69 years old on average, and about 60% were men. The majority were white. About a third had a modified Rankin Scale score at or below 3 at discharge, and about two-thirds were discharged home or to a rehabilitation facility; 17% of patients died in the hospital.

Differences in antihypertensive regimens were not associated with outcomes on univariate analysis. Dr. Jillella said that, ideally, he would like to run a multicenter, prospective trial of blood pressure reduction targets after reperfusion.

There was no external funding, and Dr. Jillella didn’t have any relevant disclosures.
 

aotto@mdedge.com

The Food and Drug Administration has approved rivaroxaban (Xarelto) for the prevention of venous thromboembolism (VTE) in hospitalized, acutely ill patients at risk for thromboembolic complications who do not have a high bleeding risk, according to a release from Janssen.

FDA approval for the new indication is based on results from the phase 3 MAGELLAN and MARINER trials, which included more than 20,000 hospitalized, acutely ill patients. In MAGELLAN, rivaroxaban demonstrated noninferiority to enoxaparin, a low-molecular-weight heparin, in short-term usage, and it was superior over the long term, compared with short-term enoxaparin followed by placebo.

While VTE and VTE-related deaths were not reduced in MARINER, compared with placebo, patients who received rivaroxaban did see a significantly reduction in symptomatic VTE with a favorable safety profile.

According to the indication, rivaroxaban can be administered to patients during hospitalization and can be continued after discharge for 31-39 days. The safety profile in MAGELLAN and MARINER was consistent with that already seen, with the most common adverse event being bleeding.

The new indication is the eighth for rivaroxaban, the most of any direct oral anticoagulant; six of these are specifically for the treatment, prevention, and reduction in the risk of VTE recurrence.

“With this new approval, Xarelto as an oral-only option now has the potential to change how acutely ill medical patients are managed for the prevention of blood clots, both in the hospital and for an extended period after discharge,” said Alex C. Spyropoulos, MD, of Northwell Health at Lenox Hill Hospital, New York, and a member of the steering committee of the MAGELLAN trial.

Find the full press release on the Janssen website.

lfranki@mdedge.com

The Food and Drug Administration has approved rivaroxaban (Xarelto) for the prevention of venous thromboembolism (VTE) in hospitalized, acutely ill patients at risk for thromboembolic complications who do not have a high bleeding risk, according to a release from Janssen.

FDA approval for the new indication is based on results from the phase 3 MAGELLAN and MARINER trials, which included more than 20,000 hospitalized, acutely ill patients. In MAGELLAN, rivaroxaban demonstrated noninferiority to enoxaparin, a low-molecular-weight heparin, in short-term usage, and it was superior over the long term, compared with short-term enoxaparin followed by placebo.

While VTE and VTE-related deaths were not reduced in MARINER, compared with placebo, patients who received rivaroxaban did see a significantly reduction in symptomatic VTE with a favorable safety profile.

According to the indication, rivaroxaban can be administered to patients during hospitalization and can be continued after discharge for 31-39 days. The safety profile in MAGELLAN and MARINER was consistent with that already seen, with the most common adverse event being bleeding.

The new indication is the eighth for rivaroxaban, the most of any direct oral anticoagulant; six of these are specifically for the treatment, prevention, and reduction in the risk of VTE recurrence.

“With this new approval, Xarelto as an oral-only option now has the potential to change how acutely ill medical patients are managed for the prevention of blood clots, both in the hospital and for an extended period after discharge,” said Alex C. Spyropoulos, MD, of Northwell Health at Lenox Hill Hospital, New York, and a member of the steering committee of the MAGELLAN trial.

Find the full press release on the Janssen website.

lfranki@mdedge.com

The Food and Drug Administration has approved rivaroxaban (Xarelto) for the prevention of venous thromboembolism (VTE) in hospitalized, acutely ill patients at risk for thromboembolic complications who do not have a high bleeding risk, according to a release from Janssen.

FDA approval for the new indication is based on results from the phase 3 MAGELLAN and MARINER trials, which included more than 20,000 hospitalized, acutely ill patients. In MAGELLAN, rivaroxaban demonstrated noninferiority to enoxaparin, a low-molecular-weight heparin, in short-term usage, and it was superior over the long term, compared with short-term enoxaparin followed by placebo.

While VTE and VTE-related deaths were not reduced in MARINER, compared with placebo, patients who received rivaroxaban did see a significantly reduction in symptomatic VTE with a favorable safety profile.

According to the indication, rivaroxaban can be administered to patients during hospitalization and can be continued after discharge for 31-39 days. The safety profile in MAGELLAN and MARINER was consistent with that already seen, with the most common adverse event being bleeding.

The new indication is the eighth for rivaroxaban, the most of any direct oral anticoagulant; six of these are specifically for the treatment, prevention, and reduction in the risk of VTE recurrence.

“With this new approval, Xarelto as an oral-only option now has the potential to change how acutely ill medical patients are managed for the prevention of blood clots, both in the hospital and for an extended period after discharge,” said Alex C. Spyropoulos, MD, of Northwell Health at Lenox Hill Hospital, New York, and a member of the steering committee of the MAGELLAN trial.

Find the full press release on the Janssen website.

lfranki@mdedge.com

More conservative oxygen therapy during mechanical ventilation in intensive care does not appear to increase the number of ventilator-free days or reduce mortality, according to a study published online in the New England Journal of Medicine.

Diane Mackle of the Medical Research Institute of New Zealand and her co-authors wrote that hyperoxemia in adults undergoing mechanical ventilation has been associated with increased mortality, as well as fewer days free of ventilation, but there was a lack of data to guide oxygen administration.

In a parallel-group trial, 1,000 adults who were expected to require mechanical ventilation – with an intention-to-treat population of 965 – were randomized either to conservative oxygen therapy or usual therapy. For the conservative therapy, the upper limit of the pulse oximetry alarm would sound when levels reached 97% and the F10₂ was decreased to 0.21 if the pulse oximetry was above the acceptable lower limit, while usual therapy involved no specific limiting measures. In both groups, the default lower limit for oxygen saturation was 90%.

At day 28 after ventilation, there was no significant difference between the conservative and usual care groups in the number of ventilator-free days (21.3 days vs. 22.1 days). The patients in the conservative oxygen group spent a median of 29 hours receiving an F10₂ level of 0.21, compared with 1 hour in the usual care group.

The mortality rate at day 180 was 35.7% in the conservative oxygen group, and 34.5% in the usual-oxygen group (HR 1.05, 95% CI 0.85 – 1.30). Researchers also saw no differences between the two groups in paid employment and cognitive function.

In patients with suspected hypoxic-ischemic encephalopathy between-group differences were apparent; At day 28, those in the conservative-oxygen group had a median of 21.1 ventilator-free days, compared with none in the usual-oxygen group. The usual-oxygen group also had a higher 180-day mortality rate than those in the conservative-oxygen group (43% vs. 59%).

“Our data are suggestive of a possible benefit of conservative oxygen therapy in patients with suspected hypoxic-ischemic encephalopathy,” the authors wrote. “It is biologically plausible that conservative oxygen therapy reduces the incidence of secondary brain damage after resuscitation from cardiac arrest, and observational data suggest that exposure to hyperoxemia in such patients may be harmful.”

The authors noted that their trial did not rule out the possibility of benefit or harm had they used a more liberal oxygen regimen in their usual-care group, and that different conservative regimens might also have achieved different outcomes.

The study was funded by the New Zealand Health Research Council. Six authors declared research support for the trial from the study funder, and two declared unrelated research grants from private industry. No other conflicts of interest were declared.

SOURCE: Mackle D et al. NJEM 2019, October 14. DOI:10.1056/NEJMoa1903297.

More conservative oxygen therapy during mechanical ventilation in intensive care does not appear to increase the number of ventilator-free days or reduce mortality, according to a study published online in the New England Journal of Medicine.

Diane Mackle of the Medical Research Institute of New Zealand and her co-authors wrote that hyperoxemia in adults undergoing mechanical ventilation has been associated with increased mortality, as well as fewer days free of ventilation, but there was a lack of data to guide oxygen administration.

In a parallel-group trial, 1,000 adults who were expected to require mechanical ventilation – with an intention-to-treat population of 965 – were randomized either to conservative oxygen therapy or usual therapy. For the conservative therapy, the upper limit of the pulse oximetry alarm would sound when levels reached 97% and the F10₂ was decreased to 0.21 if the pulse oximetry was above the acceptable lower limit, while usual therapy involved no specific limiting measures. In both groups, the default lower limit for oxygen saturation was 90%.

At day 28 after ventilation, there was no significant difference between the conservative and usual care groups in the number of ventilator-free days (21.3 days vs. 22.1 days). The patients in the conservative oxygen group spent a median of 29 hours receiving an F10₂ level of 0.21, compared with 1 hour in the usual care group.

The mortality rate at day 180 was 35.7% in the conservative oxygen group, and 34.5% in the usual-oxygen group (HR 1.05, 95% CI 0.85 – 1.30). Researchers also saw no differences between the two groups in paid employment and cognitive function.

In patients with suspected hypoxic-ischemic encephalopathy between-group differences were apparent; At day 28, those in the conservative-oxygen group had a median of 21.1 ventilator-free days, compared with none in the usual-oxygen group. The usual-oxygen group also had a higher 180-day mortality rate than those in the conservative-oxygen group (43% vs. 59%).

“Our data are suggestive of a possible benefit of conservative oxygen therapy in patients with suspected hypoxic-ischemic encephalopathy,” the authors wrote. “It is biologically plausible that conservative oxygen therapy reduces the incidence of secondary brain damage after resuscitation from cardiac arrest, and observational data suggest that exposure to hyperoxemia in such patients may be harmful.”

The authors noted that their trial did not rule out the possibility of benefit or harm had they used a more liberal oxygen regimen in their usual-care group, and that different conservative regimens might also have achieved different outcomes.

The study was funded by the New Zealand Health Research Council. Six authors declared research support for the trial from the study funder, and two declared unrelated research grants from private industry. No other conflicts of interest were declared.

SOURCE: Mackle D et al. NJEM 2019, October 14. DOI:10.1056/NEJMoa1903297.

More conservative oxygen therapy during mechanical ventilation in intensive care does not appear to increase the number of ventilator-free days or reduce mortality, according to a study published online in the New England Journal of Medicine.

Diane Mackle of the Medical Research Institute of New Zealand and her co-authors wrote that hyperoxemia in adults undergoing mechanical ventilation has been associated with increased mortality, as well as fewer days free of ventilation, but there was a lack of data to guide oxygen administration.

In a parallel-group trial, 1,000 adults who were expected to require mechanical ventilation – with an intention-to-treat population of 965 – were randomized either to conservative oxygen therapy or usual therapy. For the conservative therapy, the upper limit of the pulse oximetry alarm would sound when levels reached 97% and the F10₂ was decreased to 0.21 if the pulse oximetry was above the acceptable lower limit, while usual therapy involved no specific limiting measures. In both groups, the default lower limit for oxygen saturation was 90%.

At day 28 after ventilation, there was no significant difference between the conservative and usual care groups in the number of ventilator-free days (21.3 days vs. 22.1 days). The patients in the conservative oxygen group spent a median of 29 hours receiving an F10₂ level of 0.21, compared with 1 hour in the usual care group.

The mortality rate at day 180 was 35.7% in the conservative oxygen group, and 34.5% in the usual-oxygen group (HR 1.05, 95% CI 0.85 – 1.30). Researchers also saw no differences between the two groups in paid employment and cognitive function.

In patients with suspected hypoxic-ischemic encephalopathy between-group differences were apparent; At day 28, those in the conservative-oxygen group had a median of 21.1 ventilator-free days, compared with none in the usual-oxygen group. The usual-oxygen group also had a higher 180-day mortality rate than those in the conservative-oxygen group (43% vs. 59%).

“Our data are suggestive of a possible benefit of conservative oxygen therapy in patients with suspected hypoxic-ischemic encephalopathy,” the authors wrote. “It is biologically plausible that conservative oxygen therapy reduces the incidence of secondary brain damage after resuscitation from cardiac arrest, and observational data suggest that exposure to hyperoxemia in such patients may be harmful.”

The authors noted that their trial did not rule out the possibility of benefit or harm had they used a more liberal oxygen regimen in their usual-care group, and that different conservative regimens might also have achieved different outcomes.

The study was funded by the New Zealand Health Research Council. Six authors declared research support for the trial from the study funder, and two declared unrelated research grants from private industry. No other conflicts of interest were declared.

SOURCE: Mackle D et al. NJEM 2019, October 14. DOI:10.1056/NEJMoa1903297.

ST. LOUIS – Getting an MRI first for suspected stroke, instead of a CT, saves money by avoiding unnecessary admissions and might lead to better outcomes, according to a review from Johns Hopkins University, Baltimore.

MRI as the first scan leads to “a definitive diagnoses sooner and helps you manage the person more rapidly and appropriately, without negatively affecting outcomes even in stroke patients who receive endovascular therapy,” said neurologist and senior investigator Argye Hillis, MD, director of the Center of Excellence in Stroke Detection and Diagnosis at Hopkins. “Consider skipping the CT and getting an MRI, and get the MRI while they are still in the emergency room.”

Almost all emergency departments in the United States are set up to get a CT first, but MRI is known to be the better study, according to the researchers. MRI is much more sensitive to stroke, especially in the first 24 hours, and pinpoints the location and extent of the damage. It can detect causes of stroke invisible to CT, with no radiation, and rule out stroke entirely, whereas CT can rule out only intracranial bleeding. Increasingly in Europe, MRI is the first study in suspected stroke, and new EDs in the United States are being designed with an in-house MRI, or one nearby.

The ED at Hopkins’ main campus in downtown Baltimore already has an MRI, and uses it first whenever possible. The problem has been that MRI techs are available only during weekdays, so physicians have to default back to CT at night and on weekends. The impetus for the review, presented at the annual meeting of the American Neurological Association, was to see if savings from unnecessary admissions prevented by MRI would be enough to offset the cost of around-the-clock staffing for the MRI scanner.

Dr. Hillis and her team reviewed 320 patients with suspected ischemic stroke who were seen at the main campus in 2018 and had CT in the ED, and then definitive diagnosis by MRI, which is the usual approach in most U.S. hospitals.

A total of 134 patients had a final diagnosis on MRI that did not justify admission; techs were available to give 75 of them MRIs in the ED after the CT, and those patients were sent home. Techs were not available, however, for 59 patients and since the CT was not able to rule out stroke, those patients were admitted. The cost of those 59 admissions was $814,016.

The cost of the noncontrast CTs for the 75 patients who were sent home after definitive MRI imaging was $28,050, plus an additional $46,072 for those who had CT neck/head angiograms. Altogether, skipping the CT and going straight to the MRI would have saved Hopkins $888,138 in 2018, enough to cover round-the-clock MRI staffing in the ED, which is now the plan at the main campus.

Once the facility moves to 24-and-7 MRI coverage, the next step in the project is to compare stroke outcomes with Johns Hopkins Bayview Medical Center, also in Baltimore, which will continue to do CT first. “We know MRI first is cheaper. We want to see if we have better outcomes. If we find they’re much better, I think many hospitals will say it’s worth the 5 minutes longer it takes to get to the MRI scanner,” Dr. Hillis said.

Stroke mimics among the 134 patients included peripheral nerve palsy and migraine, but also people simply faking it for a hot meal and a warm bed. “Its pretty common, unfortunately,” she said.

The average age for stroke admissions at Hopkins is 55 years, with as many men as women.

There was no industry funding, and Dr. Hillis didn’t have any relevant disclosures.

aotto@mdedge.com

ST. LOUIS – Getting an MRI first for suspected stroke, instead of a CT, saves money by avoiding unnecessary admissions and might lead to better outcomes, according to a review from Johns Hopkins University, Baltimore.

MRI as the first scan leads to “a definitive diagnoses sooner and helps you manage the person more rapidly and appropriately, without negatively affecting outcomes even in stroke patients who receive endovascular therapy,” said neurologist and senior investigator Argye Hillis, MD, director of the Center of Excellence in Stroke Detection and Diagnosis at Hopkins. “Consider skipping the CT and getting an MRI, and get the MRI while they are still in the emergency room.”

Almost all emergency departments in the United States are set up to get a CT first, but MRI is known to be the better study, according to the researchers. MRI is much more sensitive to stroke, especially in the first 24 hours, and pinpoints the location and extent of the damage. It can detect causes of stroke invisible to CT, with no radiation, and rule out stroke entirely, whereas CT can rule out only intracranial bleeding. Increasingly in Europe, MRI is the first study in suspected stroke, and new EDs in the United States are being designed with an in-house MRI, or one nearby.

The ED at Hopkins’ main campus in downtown Baltimore already has an MRI, and uses it first whenever possible. The problem has been that MRI techs are available only during weekdays, so physicians have to default back to CT at night and on weekends. The impetus for the review, presented at the annual meeting of the American Neurological Association, was to see if savings from unnecessary admissions prevented by MRI would be enough to offset the cost of around-the-clock staffing for the MRI scanner.

Dr. Hillis and her team reviewed 320 patients with suspected ischemic stroke who were seen at the main campus in 2018 and had CT in the ED, and then definitive diagnosis by MRI, which is the usual approach in most U.S. hospitals.

A total of 134 patients had a final diagnosis on MRI that did not justify admission; techs were available to give 75 of them MRIs in the ED after the CT, and those patients were sent home. Techs were not available, however, for 59 patients and since the CT was not able to rule out stroke, those patients were admitted. The cost of those 59 admissions was $814,016.

The cost of the noncontrast CTs for the 75 patients who were sent home after definitive MRI imaging was $28,050, plus an additional $46,072 for those who had CT neck/head angiograms. Altogether, skipping the CT and going straight to the MRI would have saved Hopkins $888,138 in 2018, enough to cover round-the-clock MRI staffing in the ED, which is now the plan at the main campus.

Once the facility moves to 24-and-7 MRI coverage, the next step in the project is to compare stroke outcomes with Johns Hopkins Bayview Medical Center, also in Baltimore, which will continue to do CT first. “We know MRI first is cheaper. We want to see if we have better outcomes. If we find they’re much better, I think many hospitals will say it’s worth the 5 minutes longer it takes to get to the MRI scanner,” Dr. Hillis said.

Stroke mimics among the 134 patients included peripheral nerve palsy and migraine, but also people simply faking it for a hot meal and a warm bed. “Its pretty common, unfortunately,” she said.

The average age for stroke admissions at Hopkins is 55 years, with as many men as women.

There was no industry funding, and Dr. Hillis didn’t have any relevant disclosures.

aotto@mdedge.com

ST. LOUIS – Getting an MRI first for suspected stroke, instead of a CT, saves money by avoiding unnecessary admissions and might lead to better outcomes, according to a review from Johns Hopkins University, Baltimore.

MRI as the first scan leads to “a definitive diagnoses sooner and helps you manage the person more rapidly and appropriately, without negatively affecting outcomes even in stroke patients who receive endovascular therapy,” said neurologist and senior investigator Argye Hillis, MD, director of the Center of Excellence in Stroke Detection and Diagnosis at Hopkins. “Consider skipping the CT and getting an MRI, and get the MRI while they are still in the emergency room.”

Almost all emergency departments in the United States are set up to get a CT first, but MRI is known to be the better study, according to the researchers. MRI is much more sensitive to stroke, especially in the first 24 hours, and pinpoints the location and extent of the damage. It can detect causes of stroke invisible to CT, with no radiation, and rule out stroke entirely, whereas CT can rule out only intracranial bleeding. Increasingly in Europe, MRI is the first study in suspected stroke, and new EDs in the United States are being designed with an in-house MRI, or one nearby.

The ED at Hopkins’ main campus in downtown Baltimore already has an MRI, and uses it first whenever possible. The problem has been that MRI techs are available only during weekdays, so physicians have to default back to CT at night and on weekends. The impetus for the review, presented at the annual meeting of the American Neurological Association, was to see if savings from unnecessary admissions prevented by MRI would be enough to offset the cost of around-the-clock staffing for the MRI scanner.

Dr. Hillis and her team reviewed 320 patients with suspected ischemic stroke who were seen at the main campus in 2018 and had CT in the ED, and then definitive diagnosis by MRI, which is the usual approach in most U.S. hospitals.

A total of 134 patients had a final diagnosis on MRI that did not justify admission; techs were available to give 75 of them MRIs in the ED after the CT, and those patients were sent home. Techs were not available, however, for 59 patients and since the CT was not able to rule out stroke, those patients were admitted. The cost of those 59 admissions was $814,016.

The cost of the noncontrast CTs for the 75 patients who were sent home after definitive MRI imaging was $28,050, plus an additional $46,072 for those who had CT neck/head angiograms. Altogether, skipping the CT and going straight to the MRI would have saved Hopkins $888,138 in 2018, enough to cover round-the-clock MRI staffing in the ED, which is now the plan at the main campus.

Once the facility moves to 24-and-7 MRI coverage, the next step in the project is to compare stroke outcomes with Johns Hopkins Bayview Medical Center, also in Baltimore, which will continue to do CT first. “We know MRI first is cheaper. We want to see if we have better outcomes. If we find they’re much better, I think many hospitals will say it’s worth the 5 minutes longer it takes to get to the MRI scanner,” Dr. Hillis said.

Stroke mimics among the 134 patients included peripheral nerve palsy and migraine, but also people simply faking it for a hot meal and a warm bed. “Its pretty common, unfortunately,” she said.

The average age for stroke admissions at Hopkins is 55 years, with as many men as women.

There was no industry funding, and Dr. Hillis didn’t have any relevant disclosures.

aotto@mdedge.com

MADRID – In patients managed on mechanical ventilation in an intensive care unit following cardiac surgery, a fully automated system provides more reliable ventilatory support than highly experienced ICU nurses, suggest results of a randomized trial.

The study’s control group received usual care, which means that nurses adjusted mechanical ventilation manually in response to respiratory rate, tidal volume, positive end-respiratory pressure (PEEP), and other factors to maintain ventilation within parameters associated with safe respiration. The experimental group was managed with a fully automated closed-loop system to make these adjustments without any nurse intervention.

For those in the experimental group “the proportion of time in the optimal zone was increased and the proportion of time in the unsafe zone was decreased” relative to those randomized to conventional nursing care, Marcus J. Schultz, MD, reported at the annual congress of the European Respiratory Society.

Conducted at a hospital with an experienced ICU staff, the study had a control arm that was managed by “dedicated nurses who, I can tell you, are very eager to provide the best level of care possible,” said Dr. Schultz, professor of experimental intensive care, University of Amsterdam, the Netherlands..

The investigator-initiated POSITiVE trial randomized 220 cardiac surgery patients scheduled to receive postoperative mechanical ventilation in the ICU. Exclusions included those with class III or higher chronic obstructive pulmonary disease (COPD), a requirement for extracorporeal membrane oxygenation (ECMO), or a history of lung surgery.

The primary endpoint was the proportion of time spent in an optimal zone, an acceptable zone, or a dangerous zone of ventilation based on predefined values for tidal volume, maximum airway pressure, end-tidal CO₂, and oxygen saturation (SpO₂).

The greatest between-group difference was seen in the proportion of time spent in the optimal zone. This climbed from approximately 35% in the control arm to slightly more than 70% in the experimental arm, a significant difference. The proportion of time in the dangerous zone was reduced from approximately 6% in the control arm to 3% in the automated arm. On average nurse-managed patients spent nearly 60% of the time in the acceptable zone versus less than 30% of those in the automated experimental arm.

A heat map using green, yellow, and red to represent optimal, acceptable, and dangerous zones, respectively, for individual participants in the trial provided a more stark global impression. For the control group, the heat map was primarily yellow with scattered dashes of green and red. For the experimental group, the map was primarily green with dashes of yellow and a much smaller number of red dashes relative to the control group.

In addition, the time to spontaneous breathing was 38% shorter for those randomized to automated ventilation than to conventional care, a significant difference.

There are now many devices marketed for automated ventilation, according to Dr. Schultz. The device used in this study was the proprietary INTELLiVENT-ASV system, marketed by Hamilton Medical, which was selected based on prior satisfactory experience. Although not unique, this system has sophisticated software to adjust ventilation to reach targets set by the clinician on the basis of information it is receiving from physiologic sensors for such variables as respiratory rate, tidal volume, and inspiratory pressure.

“It is frequently adjusting the PEEP levels to reach the lowest driving pressure,” said Dr. Schultz. Among its many other features, it also “gives spontaneous breathing trials automatically.”

Uncomplicated patients were selected purposefully to test this system, but Dr. Schultz said that a second trial, called POSITiVE 2, is now being planned that will enroll more complex patients. Keeping complex patients within the optimal zone as defined by tidal volume and other critical variables has the potential to reduce the lung damage that is known to occur when these are not optimized.

“Applying safe ventilatory support in clinical practice remains a serious challenge and is extremely time consuming,” Dr. Schultz said. He reported that fully automated ventilation appears to be reliable, and “it takes out the human factor” in regard to diligence in monitoring and potential for error.

Overall, these results support the potential for a fully automated system to improve optimal ventilatory support, reduce risk of lung injury, and reduce staffing required for monitoring of mechanical ventilation, according to Dr. Schultz.

Relative costs were not evaluated in this analysis, but might be another factor relevant to the value of fully automated ventilation in ICU patients.
 

MADRID – In patients managed on mechanical ventilation in an intensive care unit following cardiac surgery, a fully automated system provides more reliable ventilatory support than highly experienced ICU nurses, suggest results of a randomized trial.

The study’s control group received usual care, which means that nurses adjusted mechanical ventilation manually in response to respiratory rate, tidal volume, positive end-respiratory pressure (PEEP), and other factors to maintain ventilation within parameters associated with safe respiration. The experimental group was managed with a fully automated closed-loop system to make these adjustments without any nurse intervention.

For those in the experimental group “the proportion of time in the optimal zone was increased and the proportion of time in the unsafe zone was decreased” relative to those randomized to conventional nursing care, Marcus J. Schultz, MD, reported at the annual congress of the European Respiratory Society.

Conducted at a hospital with an experienced ICU staff, the study had a control arm that was managed by “dedicated nurses who, I can tell you, are very eager to provide the best level of care possible,” said Dr. Schultz, professor of experimental intensive care, University of Amsterdam, the Netherlands..

The investigator-initiated POSITiVE trial randomized 220 cardiac surgery patients scheduled to receive postoperative mechanical ventilation in the ICU. Exclusions included those with class III or higher chronic obstructive pulmonary disease (COPD), a requirement for extracorporeal membrane oxygenation (ECMO), or a history of lung surgery.

The primary endpoint was the proportion of time spent in an optimal zone, an acceptable zone, or a dangerous zone of ventilation based on predefined values for tidal volume, maximum airway pressure, end-tidal CO₂, and oxygen saturation (SpO₂).

The greatest between-group difference was seen in the proportion of time spent in the optimal zone. This climbed from approximately 35% in the control arm to slightly more than 70% in the experimental arm, a significant difference. The proportion of time in the dangerous zone was reduced from approximately 6% in the control arm to 3% in the automated arm. On average nurse-managed patients spent nearly 60% of the time in the acceptable zone versus less than 30% of those in the automated experimental arm.

A heat map using green, yellow, and red to represent optimal, acceptable, and dangerous zones, respectively, for individual participants in the trial provided a more stark global impression. For the control group, the heat map was primarily yellow with scattered dashes of green and red. For the experimental group, the map was primarily green with dashes of yellow and a much smaller number of red dashes relative to the control group.

In addition, the time to spontaneous breathing was 38% shorter for those randomized to automated ventilation than to conventional care, a significant difference.

There are now many devices marketed for automated ventilation, according to Dr. Schultz. The device used in this study was the proprietary INTELLiVENT-ASV system, marketed by Hamilton Medical, which was selected based on prior satisfactory experience. Although not unique, this system has sophisticated software to adjust ventilation to reach targets set by the clinician on the basis of information it is receiving from physiologic sensors for such variables as respiratory rate, tidal volume, and inspiratory pressure.

“It is frequently adjusting the PEEP levels to reach the lowest driving pressure,” said Dr. Schultz. Among its many other features, it also “gives spontaneous breathing trials automatically.”

Uncomplicated patients were selected purposefully to test this system, but Dr. Schultz said that a second trial, called POSITiVE 2, is now being planned that will enroll more complex patients. Keeping complex patients within the optimal zone as defined by tidal volume and other critical variables has the potential to reduce the lung damage that is known to occur when these are not optimized.

“Applying safe ventilatory support in clinical practice remains a serious challenge and is extremely time consuming,” Dr. Schultz said. He reported that fully automated ventilation appears to be reliable, and “it takes out the human factor” in regard to diligence in monitoring and potential for error.

Overall, these results support the potential for a fully automated system to improve optimal ventilatory support, reduce risk of lung injury, and reduce staffing required for monitoring of mechanical ventilation, according to Dr. Schultz.

Relative costs were not evaluated in this analysis, but might be another factor relevant to the value of fully automated ventilation in ICU patients.
 

MADRID – In patients managed on mechanical ventilation in an intensive care unit following cardiac surgery, a fully automated system provides more reliable ventilatory support than highly experienced ICU nurses, suggest results of a randomized trial.

The study’s control group received usual care, which means that nurses adjusted mechanical ventilation manually in response to respiratory rate, tidal volume, positive end-respiratory pressure (PEEP), and other factors to maintain ventilation within parameters associated with safe respiration. The experimental group was managed with a fully automated closed-loop system to make these adjustments without any nurse intervention.

For those in the experimental group “the proportion of time in the optimal zone was increased and the proportion of time in the unsafe zone was decreased” relative to those randomized to conventional nursing care, Marcus J. Schultz, MD, reported at the annual congress of the European Respiratory Society.

Conducted at a hospital with an experienced ICU staff, the study had a control arm that was managed by “dedicated nurses who, I can tell you, are very eager to provide the best level of care possible,” said Dr. Schultz, professor of experimental intensive care, University of Amsterdam, the Netherlands..

The investigator-initiated POSITiVE trial randomized 220 cardiac surgery patients scheduled to receive postoperative mechanical ventilation in the ICU. Exclusions included those with class III or higher chronic obstructive pulmonary disease (COPD), a requirement for extracorporeal membrane oxygenation (ECMO), or a history of lung surgery.

The primary endpoint was the proportion of time spent in an optimal zone, an acceptable zone, or a dangerous zone of ventilation based on predefined values for tidal volume, maximum airway pressure, end-tidal CO₂, and oxygen saturation (SpO₂).

The greatest between-group difference was seen in the proportion of time spent in the optimal zone. This climbed from approximately 35% in the control arm to slightly more than 70% in the experimental arm, a significant difference. The proportion of time in the dangerous zone was reduced from approximately 6% in the control arm to 3% in the automated arm. On average nurse-managed patients spent nearly 60% of the time in the acceptable zone versus less than 30% of those in the automated experimental arm.

A heat map using green, yellow, and red to represent optimal, acceptable, and dangerous zones, respectively, for individual participants in the trial provided a more stark global impression. For the control group, the heat map was primarily yellow with scattered dashes of green and red. For the experimental group, the map was primarily green with dashes of yellow and a much smaller number of red dashes relative to the control group.

In addition, the time to spontaneous breathing was 38% shorter for those randomized to automated ventilation than to conventional care, a significant difference.

There are now many devices marketed for automated ventilation, according to Dr. Schultz. The device used in this study was the proprietary INTELLiVENT-ASV system, marketed by Hamilton Medical, which was selected based on prior satisfactory experience. Although not unique, this system has sophisticated software to adjust ventilation to reach targets set by the clinician on the basis of information it is receiving from physiologic sensors for such variables as respiratory rate, tidal volume, and inspiratory pressure.

“It is frequently adjusting the PEEP levels to reach the lowest driving pressure,” said Dr. Schultz. Among its many other features, it also “gives spontaneous breathing trials automatically.”

Uncomplicated patients were selected purposefully to test this system, but Dr. Schultz said that a second trial, called POSITiVE 2, is now being planned that will enroll more complex patients. Keeping complex patients within the optimal zone as defined by tidal volume and other critical variables has the potential to reduce the lung damage that is known to occur when these are not optimized.

“Applying safe ventilatory support in clinical practice remains a serious challenge and is extremely time consuming,” Dr. Schultz said. He reported that fully automated ventilation appears to be reliable, and “it takes out the human factor” in regard to diligence in monitoring and potential for error.

Overall, these results support the potential for a fully automated system to improve optimal ventilatory support, reduce risk of lung injury, and reduce staffing required for monitoring of mechanical ventilation, according to Dr. Schultz.

Relative costs were not evaluated in this analysis, but might be another factor relevant to the value of fully automated ventilation in ICU patients.
 

The Centers for Disease Control and Prevention has released an updated interim clinical guidance for health providers for evaluating and treating patients with lung injury associated with e-cigarette use or vaping.

In a telebriefing, Anne Schuchat, MD, CDC principal deputy director, and her colleagues answered questions about the current investigation into the source of this lung injury outbreak and the updated clinical guidance. Dr. Schuchat said, “I can’t stress enough the seriousness of these injuries.” She added, “We are not seeing a drop in cases” but a continuation of the trend of hospitalization and deaths that started in August 2019.

Investigation update

The investigation to date has yielded some information about current cases of lung injury related to vaping:

• The acronym EVALI has been developed to refer to e-cigarette, or vaping products use associated lung injury;

• 1,299 EVALI cases have been reported as of Oct. 8;

• No single compound or ingredient has emerged as the cause of these injuries, and more than one substance may be involved;

• Among the 573 patients for whom data are available on vaping products used in the previous 90 days, 76% reported using THC-containing products; 58% reported using nicotine-containing products; 32% reported exclusive use of THC-containing products, and 13% reported exclusive use of nicotine-containing products;

• Of the 700+ samples sent to the CDC for analysis, most had little or no liquid remaining in the device, limiting content analysis. In 28 THC-containing samples, THC concentrations were found to be 13% - 77% (mean 41%).

• A “handful” of cases of readmission have been reported and the CDC is currently investigating whether these cases included patients who took up vaping again or had some other possible contributing factor.

• The CDC is currently developing an ICD-10 code relevant to EVALI.

Clinical guidance update

The CDC provided detailed guidance on evaluating and caring for patients with EVALI. The recommendations focus on patient history, lab testing, criteria for hospitalization, and follow-up of these patients.

Detailed history of patients presenting with suspected EVALI is especially important for this patient population, given the many unknowns surrounding this condition. The updated guidance states, “All health care providers evaluating patients for EVALI should ask about the use of e-cigarette, or vaping, products and ideally should ask about types of substances used (e.g.,THC, cannabis [oil, dabs], nicotine, modified products or the addition of substances not intended by the manufacturer); product source, specific product brand and name; duration and frequency of use, time of last use; product delivery system, and method of use (aerosolization, dabbing, or dripping).” The approach recommended for soliciting accurate information is “empathetic, nonjudgmental” and, the guidelines say, patients should be questioned in private regarding sensitive information to assure confidentiality.

A respiratory virus panel is recommended for all suspected EVALI patients, although at this time, these tests cannot be used to distinguish EVALI from infectious etiologies. All patients should be considered for urine toxicology testing, including testing for THC.

Imaging guidance for suspected EVALI patients includes chest x-ray, with additional CT scan when the x-ray result does not correlate with clinical findings or to evaluate severe or worsening disease.

Recommended criteria for hospitalization of patients with suspected EVALI are those patients with decreased O₂ saturation (less than 95%) on room air, are in respiratory distress, or have comorbidities that compromise pulmonary reserve. As of Oct. 8, 96% of patients with suspected EVALI reported to CDC have been hospitalized.

As for medical treatment of these patients, corticosteroids have been found helpful. The statement noted, “Among 140 cases reported nationally to CDC that received corticosteroids, 82% of patients improved

The natural progression of this injury is not known, however, and it is possible that patients might recover without corticosteroids. Given the unknown etiology of the disease and “because the diagnosis remains one of exclusion, aggressive empiric therapy with corticosteroids, antimicrobial, and antiviral therapy might be warranted for patients with severe illness. A range of corticosteroid doses, durations, and taper plans might be considered on a case-by-case basis.”

The report concludes with a strong recommendation that patients hospitalized with EVALI are followed closely with a visit 1-2 weeks after discharge and again with additional testing 1-2 months later. Health care providers are also advised to consult medical specialists, in particular pulmonologists, who can offer further evaluation, recommend empiric treatment, and review indications for bronchoscopy.

Mitch Zeller, JD, director, Center for Tobacco Products with the Food and Drug Administration emphasized the extraordinary complexity of the EVALI problem but noted that the FDA and CDC “will leave no stone unturned until we get to the bottom of it.”

tborden@mdedge.com

The Centers for Disease Control and Prevention has released an updated interim clinical guidance for health providers for evaluating and treating patients with lung injury associated with e-cigarette use or vaping.

In a telebriefing, Anne Schuchat, MD, CDC principal deputy director, and her colleagues answered questions about the current investigation into the source of this lung injury outbreak and the updated clinical guidance. Dr. Schuchat said, “I can’t stress enough the seriousness of these injuries.” She added, “We are not seeing a drop in cases” but a continuation of the trend of hospitalization and deaths that started in August 2019.

Investigation update

The investigation to date has yielded some information about current cases of lung injury related to vaping:

• The acronym EVALI has been developed to refer to e-cigarette, or vaping products use associated lung injury;

• 1,299 EVALI cases have been reported as of Oct. 8;

• No single compound or ingredient has emerged as the cause of these injuries, and more than one substance may be involved;

• Among the 573 patients for whom data are available on vaping products used in the previous 90 days, 76% reported using THC-containing products; 58% reported using nicotine-containing products; 32% reported exclusive use of THC-containing products, and 13% reported exclusive use of nicotine-containing products;

• Of the 700+ samples sent to the CDC for analysis, most had little or no liquid remaining in the device, limiting content analysis. In 28 THC-containing samples, THC concentrations were found to be 13% - 77% (mean 41%).

• A “handful” of cases of readmission have been reported and the CDC is currently investigating whether these cases included patients who took up vaping again or had some other possible contributing factor.

• The CDC is currently developing an ICD-10 code relevant to EVALI.

Clinical guidance update

The CDC provided detailed guidance on evaluating and caring for patients with EVALI. The recommendations focus on patient history, lab testing, criteria for hospitalization, and follow-up of these patients.

Detailed history of patients presenting with suspected EVALI is especially important for this patient population, given the many unknowns surrounding this condition. The updated guidance states, “All health care providers evaluating patients for EVALI should ask about the use of e-cigarette, or vaping, products and ideally should ask about types of substances used (e.g.,THC, cannabis [oil, dabs], nicotine, modified products or the addition of substances not intended by the manufacturer); product source, specific product brand and name; duration and frequency of use, time of last use; product delivery system, and method of use (aerosolization, dabbing, or dripping).” The approach recommended for soliciting accurate information is “empathetic, nonjudgmental” and, the guidelines say, patients should be questioned in private regarding sensitive information to assure confidentiality.

A respiratory virus panel is recommended for all suspected EVALI patients, although at this time, these tests cannot be used to distinguish EVALI from infectious etiologies. All patients should be considered for urine toxicology testing, including testing for THC.

Imaging guidance for suspected EVALI patients includes chest x-ray, with additional CT scan when the x-ray result does not correlate with clinical findings or to evaluate severe or worsening disease.

Recommended criteria for hospitalization of patients with suspected EVALI are those patients with decreased O₂ saturation (less than 95%) on room air, are in respiratory distress, or have comorbidities that compromise pulmonary reserve. As of Oct. 8, 96% of patients with suspected EVALI reported to CDC have been hospitalized.

As for medical treatment of these patients, corticosteroids have been found helpful. The statement noted, “Among 140 cases reported nationally to CDC that received corticosteroids, 82% of patients improved

The natural progression of this injury is not known, however, and it is possible that patients might recover without corticosteroids. Given the unknown etiology of the disease and “because the diagnosis remains one of exclusion, aggressive empiric therapy with corticosteroids, antimicrobial, and antiviral therapy might be warranted for patients with severe illness. A range of corticosteroid doses, durations, and taper plans might be considered on a case-by-case basis.”

The report concludes with a strong recommendation that patients hospitalized with EVALI are followed closely with a visit 1-2 weeks after discharge and again with additional testing 1-2 months later. Health care providers are also advised to consult medical specialists, in particular pulmonologists, who can offer further evaluation, recommend empiric treatment, and review indications for bronchoscopy.

Mitch Zeller, JD, director, Center for Tobacco Products with the Food and Drug Administration emphasized the extraordinary complexity of the EVALI problem but noted that the FDA and CDC “will leave no stone unturned until we get to the bottom of it.”

tborden@mdedge.com

The Centers for Disease Control and Prevention has released an updated interim clinical guidance for health providers for evaluating and treating patients with lung injury associated with e-cigarette use or vaping.

In a telebriefing, Anne Schuchat, MD, CDC principal deputy director, and her colleagues answered questions about the current investigation into the source of this lung injury outbreak and the updated clinical guidance. Dr. Schuchat said, “I can’t stress enough the seriousness of these injuries.” She added, “We are not seeing a drop in cases” but a continuation of the trend of hospitalization and deaths that started in August 2019.

Investigation update

The investigation to date has yielded some information about current cases of lung injury related to vaping:

• The acronym EVALI has been developed to refer to e-cigarette, or vaping products use associated lung injury;

• 1,299 EVALI cases have been reported as of Oct. 8;

• No single compound or ingredient has emerged as the cause of these injuries, and more than one substance may be involved;

• Among the 573 patients for whom data are available on vaping products used in the previous 90 days, 76% reported using THC-containing products; 58% reported using nicotine-containing products; 32% reported exclusive use of THC-containing products, and 13% reported exclusive use of nicotine-containing products;

• Of the 700+ samples sent to the CDC for analysis, most had little or no liquid remaining in the device, limiting content analysis. In 28 THC-containing samples, THC concentrations were found to be 13% - 77% (mean 41%).

• A “handful” of cases of readmission have been reported and the CDC is currently investigating whether these cases included patients who took up vaping again or had some other possible contributing factor.

• The CDC is currently developing an ICD-10 code relevant to EVALI.

Clinical guidance update

The CDC provided detailed guidance on evaluating and caring for patients with EVALI. The recommendations focus on patient history, lab testing, criteria for hospitalization, and follow-up of these patients.

Detailed history of patients presenting with suspected EVALI is especially important for this patient population, given the many unknowns surrounding this condition. The updated guidance states, “All health care providers evaluating patients for EVALI should ask about the use of e-cigarette, or vaping, products and ideally should ask about types of substances used (e.g.,THC, cannabis [oil, dabs], nicotine, modified products or the addition of substances not intended by the manufacturer); product source, specific product brand and name; duration and frequency of use, time of last use; product delivery system, and method of use (aerosolization, dabbing, or dripping).” The approach recommended for soliciting accurate information is “empathetic, nonjudgmental” and, the guidelines say, patients should be questioned in private regarding sensitive information to assure confidentiality.

A respiratory virus panel is recommended for all suspected EVALI patients, although at this time, these tests cannot be used to distinguish EVALI from infectious etiologies. All patients should be considered for urine toxicology testing, including testing for THC.

Imaging guidance for suspected EVALI patients includes chest x-ray, with additional CT scan when the x-ray result does not correlate with clinical findings or to evaluate severe or worsening disease.

Recommended criteria for hospitalization of patients with suspected EVALI are those patients with decreased O₂ saturation (less than 95%) on room air, are in respiratory distress, or have comorbidities that compromise pulmonary reserve. As of Oct. 8, 96% of patients with suspected EVALI reported to CDC have been hospitalized.

As for medical treatment of these patients, corticosteroids have been found helpful. The statement noted, “Among 140 cases reported nationally to CDC that received corticosteroids, 82% of patients improved

The natural progression of this injury is not known, however, and it is possible that patients might recover without corticosteroids. Given the unknown etiology of the disease and “because the diagnosis remains one of exclusion, aggressive empiric therapy with corticosteroids, antimicrobial, and antiviral therapy might be warranted for patients with severe illness. A range of corticosteroid doses, durations, and taper plans might be considered on a case-by-case basis.”

The report concludes with a strong recommendation that patients hospitalized with EVALI are followed closely with a visit 1-2 weeks after discharge and again with additional testing 1-2 months later. Health care providers are also advised to consult medical specialists, in particular pulmonologists, who can offer further evaluation, recommend empiric treatment, and review indications for bronchoscopy.

Mitch Zeller, JD, director, Center for Tobacco Products with the Food and Drug Administration emphasized the extraordinary complexity of the EVALI problem but noted that the FDA and CDC “will leave no stone unturned until we get to the bottom of it.”

tborden@mdedge.com

Statin use for a minimum of 3 months was significantly associated with an increased risk of skin and soft tissue infections (SSTIs), according to a sequence symmetry analysis of prescription claims over a 10-year period reported in the British Journal of Clinical Pharmacology.

In the study, statin use for as little as 91 days was linked with elevated risks of SSTIs and diabetes. However, the increased risk of infection was seen in individuals who did and did not develop diabetes, wrote Humphrey Ko, of the school of pharmacy and biomedical sciences, Curtin University, Perth, Australia, and colleagues.

The current literature on the impact of statins on SSTIs is conflicted, they noted. Previous research shows that statins “may reduce the risk of community-acquired [Staphylococcus aureus] bacteremia and exert antibacterial effects against S. aureus,” and therefore may have potential for reducing SSTI risk “or evolve into promising novel treatments for SSTIs,” the researchers said; they noted, however, that other data show that statins may induce new-onset diabetes.

They examined prescription claims (for statins, antidiabetic medications, and antistaphylococcal antibiotics) from 2001 to 2011 from the Australian Department of Veterans’ Affairs that included more than 228,000 veterans, war widows, and widowers. Prescriptions for antistaphylococcal antibiotics were used as a marker of SSTIs.

Overall, statins were significantly associated with an increased risk of SSTIs at 91 days (adjusted sequence ratio, 1.40). The risk of SSTIs from statin use was similar at 182 (ASR, 1.41) and 365 days (ASR, 1.40). In this case, the ASRs represent the incidence rate ratios of prescribing antibiotics in statin-exposed versus statin-nonexposed person-time.

Statins were associated with a significantly increased risk of new onset diabetes, but the SSTI risk was not significantly different between statin users with and without diabetes. Statin users who did not have diabetes had significant SSTI risks at 91, 182, and 365 days (ASR, 1.39, 1.41, and 1.37, respectively) and statin users with diabetes had similarly significant risks of SSTIs (ASR,1.43, 1.42, and 1.49, respectively).

In addition, socioeconomic status appeared to have no significant effect on the relationship between statin use, SSTIs, and diabetes, the researchers noted.

The findings were limited by several factors including the inability to account for patient compliance in taking the medications, a lack of dosage data to determine the impact of dosage on outcomes, and potential confounding by the presence of diabetes, they said. However, the results suggest that “it would seem prudent for clinicians to monitor blood glucose levels of statin users who are predisposed to diabetes, and be mindful of possible increased SSTI risks in such patients,” they concluded. Statins, they added, “may increase SSTI risk via direct or indirect mechanisms.”

More clinical trials are needed to confirm the mechanisms, and “to ascertain the effect of statins on gut dysbiosis, impaired bile acid metabolism, vitamin D levels, and cholesterol inhibition on skin function,” they wrote.

The study was supported in part by the Australian Government Research Training Program Scholarship, the Curtin Health Innovation Research Institute Biosciences Research Precinct Core Facility, and the School of Pharmacy and Biomedical Sciences (Curtin University). The researchers had no financial conflicts to disclose.

SOURCE: Ko H et al. Br J Clin Pharmacol. 2019 Oct 9. doi: 10.1111/bcp.14077.

Statin use for a minimum of 3 months was significantly associated with an increased risk of skin and soft tissue infections (SSTIs), according to a sequence symmetry analysis of prescription claims over a 10-year period reported in the British Journal of Clinical Pharmacology.

In the study, statin use for as little as 91 days was linked with elevated risks of SSTIs and diabetes. However, the increased risk of infection was seen in individuals who did and did not develop diabetes, wrote Humphrey Ko, of the school of pharmacy and biomedical sciences, Curtin University, Perth, Australia, and colleagues.

The current literature on the impact of statins on SSTIs is conflicted, they noted. Previous research shows that statins “may reduce the risk of community-acquired [Staphylococcus aureus] bacteremia and exert antibacterial effects against S. aureus,” and therefore may have potential for reducing SSTI risk “or evolve into promising novel treatments for SSTIs,” the researchers said; they noted, however, that other data show that statins may induce new-onset diabetes.

They examined prescription claims (for statins, antidiabetic medications, and antistaphylococcal antibiotics) from 2001 to 2011 from the Australian Department of Veterans’ Affairs that included more than 228,000 veterans, war widows, and widowers. Prescriptions for antistaphylococcal antibiotics were used as a marker of SSTIs.

Overall, statins were significantly associated with an increased risk of SSTIs at 91 days (adjusted sequence ratio, 1.40). The risk of SSTIs from statin use was similar at 182 (ASR, 1.41) and 365 days (ASR, 1.40). In this case, the ASRs represent the incidence rate ratios of prescribing antibiotics in statin-exposed versus statin-nonexposed person-time.

Statins were associated with a significantly increased risk of new onset diabetes, but the SSTI risk was not significantly different between statin users with and without diabetes. Statin users who did not have diabetes had significant SSTI risks at 91, 182, and 365 days (ASR, 1.39, 1.41, and 1.37, respectively) and statin users with diabetes had similarly significant risks of SSTIs (ASR,1.43, 1.42, and 1.49, respectively).

In addition, socioeconomic status appeared to have no significant effect on the relationship between statin use, SSTIs, and diabetes, the researchers noted.

The findings were limited by several factors including the inability to account for patient compliance in taking the medications, a lack of dosage data to determine the impact of dosage on outcomes, and potential confounding by the presence of diabetes, they said. However, the results suggest that “it would seem prudent for clinicians to monitor blood glucose levels of statin users who are predisposed to diabetes, and be mindful of possible increased SSTI risks in such patients,” they concluded. Statins, they added, “may increase SSTI risk via direct or indirect mechanisms.”

More clinical trials are needed to confirm the mechanisms, and “to ascertain the effect of statins on gut dysbiosis, impaired bile acid metabolism, vitamin D levels, and cholesterol inhibition on skin function,” they wrote.

The study was supported in part by the Australian Government Research Training Program Scholarship, the Curtin Health Innovation Research Institute Biosciences Research Precinct Core Facility, and the School of Pharmacy and Biomedical Sciences (Curtin University). The researchers had no financial conflicts to disclose.

SOURCE: Ko H et al. Br J Clin Pharmacol. 2019 Oct 9. doi: 10.1111/bcp.14077.

Statin use for a minimum of 3 months was significantly associated with an increased risk of skin and soft tissue infections (SSTIs), according to a sequence symmetry analysis of prescription claims over a 10-year period reported in the British Journal of Clinical Pharmacology.

In the study, statin use for as little as 91 days was linked with elevated risks of SSTIs and diabetes. However, the increased risk of infection was seen in individuals who did and did not develop diabetes, wrote Humphrey Ko, of the school of pharmacy and biomedical sciences, Curtin University, Perth, Australia, and colleagues.

The current literature on the impact of statins on SSTIs is conflicted, they noted. Previous research shows that statins “may reduce the risk of community-acquired [Staphylococcus aureus] bacteremia and exert antibacterial effects against S. aureus,” and therefore may have potential for reducing SSTI risk “or evolve into promising novel treatments for SSTIs,” the researchers said; they noted, however, that other data show that statins may induce new-onset diabetes.

They examined prescription claims (for statins, antidiabetic medications, and antistaphylococcal antibiotics) from 2001 to 2011 from the Australian Department of Veterans’ Affairs that included more than 228,000 veterans, war widows, and widowers. Prescriptions for antistaphylococcal antibiotics were used as a marker of SSTIs.

Overall, statins were significantly associated with an increased risk of SSTIs at 91 days (adjusted sequence ratio, 1.40). The risk of SSTIs from statin use was similar at 182 (ASR, 1.41) and 365 days (ASR, 1.40). In this case, the ASRs represent the incidence rate ratios of prescribing antibiotics in statin-exposed versus statin-nonexposed person-time.

Statins were associated with a significantly increased risk of new onset diabetes, but the SSTI risk was not significantly different between statin users with and without diabetes. Statin users who did not have diabetes had significant SSTI risks at 91, 182, and 365 days (ASR, 1.39, 1.41, and 1.37, respectively) and statin users with diabetes had similarly significant risks of SSTIs (ASR,1.43, 1.42, and 1.49, respectively).

In addition, socioeconomic status appeared to have no significant effect on the relationship between statin use, SSTIs, and diabetes, the researchers noted.

The findings were limited by several factors including the inability to account for patient compliance in taking the medications, a lack of dosage data to determine the impact of dosage on outcomes, and potential confounding by the presence of diabetes, they said. However, the results suggest that “it would seem prudent for clinicians to monitor blood glucose levels of statin users who are predisposed to diabetes, and be mindful of possible increased SSTI risks in such patients,” they concluded. Statins, they added, “may increase SSTI risk via direct or indirect mechanisms.”

More clinical trials are needed to confirm the mechanisms, and “to ascertain the effect of statins on gut dysbiosis, impaired bile acid metabolism, vitamin D levels, and cholesterol inhibition on skin function,” they wrote.

The study was supported in part by the Australian Government Research Training Program Scholarship, the Curtin Health Innovation Research Institute Biosciences Research Precinct Core Facility, and the School of Pharmacy and Biomedical Sciences (Curtin University). The researchers had no financial conflicts to disclose.

SOURCE: Ko H et al. Br J Clin Pharmacol. 2019 Oct 9. doi: 10.1111/bcp.14077.

Influenza vaccination modestly reduces the risk of hospitalizations associated with laboratory-confirmed influenza in people with chronic obstructive pulmonary disease (COPD), according to data published in the Journal of Infectious Diseases.

Cynthia Goldsmith/CDC photo #10073

“To the best of our knowledge, this is the first large, real-world population study to examine vaccine effectiveness in people with COPD using the test-negative design and influenza-specific study outcomes,” wrote Andrea S. Gershon, MD, of Sunnybrook Health Sciences Center in Toronto and colleagues. “These findings emphasize the need for more effective influenza vaccines for older COPD patients and other preventive strategies.”
 

A test-negative study design

Data suggest that 70% of COPD exacerbations are caused by infection, and influenza often is identified as the cause. Although all major COPD practice guidelines recommend seasonal influenza vaccination, the evidence indicating that vaccination reduces hospitalizations and death is limited. The inherent or corticosteroid-induced decrease in immune response to vaccination and respiratory infection among patients with COPD may reduce the effectiveness of influenza vaccination, wrote Dr. Gershon and colleagues.

The investigators used a test-negative design to evaluate how effectively influenza vaccination prevents laboratory-confirmed influenza–associated hospitalizations in community-dwelling older patients with COPD. They chose this design because it attenuates biases resulting from misclassification of infection and from differences in health care–seeking behavior between vaccinated and unvaccinated patients.

Dr. Gershon and colleagues examined health care administrative data and respiratory specimens collected from patients who had been tested for influenza during the 2010-2011 to 2015-2016 influenza seasons. Eligible patients were aged 66 years or older, had physician-diagnosed COPD, and had been tested for influenza within 3 days before and during an acute care hospitalization. The researchers determined influenza vaccination status using physician and pharmacist billing claims. They obtained demographic information through linkage with the provincial health insurance database. Multivariable logistic regression allowed Dr. Gershon and colleagues to estimate the adjusted odds ratio of influenza vaccination in people with laboratory-confirmed influenza, compared with those without.
 

Effectiveness did not vary by demographic factors

The investigators included 21,748 patients in their analysis. Of this population, 3,636 (16.7%) patients tested positive for influenza. Vaccinated patients were less likely than unvaccinated patients to test positive for influenza (15.3% vs. 18.6%). Vaccinated patients also were more likely to be older; live in an urban area; live in a higher income neighborhood; have had more outpatient visits with a physician in the previous year; have received a prescription for a COPD medication in the previous 6 months; have diabetes, asthma, or immunocompromising conditions; have a longer duration of COPD; and have had an outpatient COPD exacerbation in the previous year.

The overall unadjusted estimate of vaccine effectiveness against laboratory-confirmed influenza–associated hospitalizations was 21%. Multivariable adjustment yielded an effectiveness of 22%. When Dr. Gershon and colleagues corrected for misclassification of vaccination status among people with COPD, the effectiveness was estimated to be 43%. Vaccine effectiveness did not vary significantly according to influenza season, nor did it vary significantly by patient-specific factors such as age, sex, influenza subtype, codiagnosis of asthma, duration of COPD, previous outpatient COPD exacerbations, previous COPD hospitalization, previous receipt of inhaled corticosteroids, and previous pneumonia.

One limitation of the study was the possibility that COPD was misclassified because not all participants underwent pulmonary function testing. In addition, the estimates of vaccine effectiveness in the present study are specific to the outcome of influenza hospitalization and may not be generalizable to vaccine effectiveness estimates of outpatient outcomes, said the investigators. Finally, Dr. Gershon and colleagues could not identify the type of vaccine received.

“Given that a large pragmatic randomized controlled trial evaluating influenza vaccination would be unethical, this is likely the most robust estimate of vaccine effectiveness for hospitalizations in the COPD population to guide influenza vaccine recommendations for patients with COPD,” wrote Dr. Gershon and colleagues.

An Ontario Ministry of Health and Long-Term Care Health Systems Research Fund Capacity Grant and a Canadian Institutes of Health Research operating grant funded this research. One investigator received grants from the Canadian Institutes of Health Research during the study, and others received grants from pharmaceutical companies that were unrelated to this study.

egreb@mdedge.com

SOURCE: Gershon AS et al. J Infect Dis. 2019 Sep 24. doi: 10.1093/infdis/jiz419.

Influenza vaccination modestly reduces the risk of hospitalizations associated with laboratory-confirmed influenza in people with chronic obstructive pulmonary disease (COPD), according to data published in the Journal of Infectious Diseases.

Cynthia Goldsmith/CDC photo #10073

“To the best of our knowledge, this is the first large, real-world population study to examine vaccine effectiveness in people with COPD using the test-negative design and influenza-specific study outcomes,” wrote Andrea S. Gershon, MD, of Sunnybrook Health Sciences Center in Toronto and colleagues. “These findings emphasize the need for more effective influenza vaccines for older COPD patients and other preventive strategies.”
 

A test-negative study design

Data suggest that 70% of COPD exacerbations are caused by infection, and influenza often is identified as the cause. Although all major COPD practice guidelines recommend seasonal influenza vaccination, the evidence indicating that vaccination reduces hospitalizations and death is limited. The inherent or corticosteroid-induced decrease in immune response to vaccination and respiratory infection among patients with COPD may reduce the effectiveness of influenza vaccination, wrote Dr. Gershon and colleagues.

The investigators used a test-negative design to evaluate how effectively influenza vaccination prevents laboratory-confirmed influenza–associated hospitalizations in community-dwelling older patients with COPD. They chose this design because it attenuates biases resulting from misclassification of infection and from differences in health care–seeking behavior between vaccinated and unvaccinated patients.

Dr. Gershon and colleagues examined health care administrative data and respiratory specimens collected from patients who had been tested for influenza during the 2010-2011 to 2015-2016 influenza seasons. Eligible patients were aged 66 years or older, had physician-diagnosed COPD, and had been tested for influenza within 3 days before and during an acute care hospitalization. The researchers determined influenza vaccination status using physician and pharmacist billing claims. They obtained demographic information through linkage with the provincial health insurance database. Multivariable logistic regression allowed Dr. Gershon and colleagues to estimate the adjusted odds ratio of influenza vaccination in people with laboratory-confirmed influenza, compared with those without.
 

Effectiveness did not vary by demographic factors

The investigators included 21,748 patients in their analysis. Of this population, 3,636 (16.7%) patients tested positive for influenza. Vaccinated patients were less likely than unvaccinated patients to test positive for influenza (15.3% vs. 18.6%). Vaccinated patients also were more likely to be older; live in an urban area; live in a higher income neighborhood; have had more outpatient visits with a physician in the previous year; have received a prescription for a COPD medication in the previous 6 months; have diabetes, asthma, or immunocompromising conditions; have a longer duration of COPD; and have had an outpatient COPD exacerbation in the previous year.

The overall unadjusted estimate of vaccine effectiveness against laboratory-confirmed influenza–associated hospitalizations was 21%. Multivariable adjustment yielded an effectiveness of 22%. When Dr. Gershon and colleagues corrected for misclassification of vaccination status among people with COPD, the effectiveness was estimated to be 43%. Vaccine effectiveness did not vary significantly according to influenza season, nor did it vary significantly by patient-specific factors such as age, sex, influenza subtype, codiagnosis of asthma, duration of COPD, previous outpatient COPD exacerbations, previous COPD hospitalization, previous receipt of inhaled corticosteroids, and previous pneumonia.

One limitation of the study was the possibility that COPD was misclassified because not all participants underwent pulmonary function testing. In addition, the estimates of vaccine effectiveness in the present study are specific to the outcome of influenza hospitalization and may not be generalizable to vaccine effectiveness estimates of outpatient outcomes, said the investigators. Finally, Dr. Gershon and colleagues could not identify the type of vaccine received.

“Given that a large pragmatic randomized controlled trial evaluating influenza vaccination would be unethical, this is likely the most robust estimate of vaccine effectiveness for hospitalizations in the COPD population to guide influenza vaccine recommendations for patients with COPD,” wrote Dr. Gershon and colleagues.

An Ontario Ministry of Health and Long-Term Care Health Systems Research Fund Capacity Grant and a Canadian Institutes of Health Research operating grant funded this research. One investigator received grants from the Canadian Institutes of Health Research during the study, and others received grants from pharmaceutical companies that were unrelated to this study.

egreb@mdedge.com

SOURCE: Gershon AS et al. J Infect Dis. 2019 Sep 24. doi: 10.1093/infdis/jiz419.

Influenza vaccination modestly reduces the risk of hospitalizations associated with laboratory-confirmed influenza in people with chronic obstructive pulmonary disease (COPD), according to data published in the Journal of Infectious Diseases.

Cynthia Goldsmith/CDC photo #10073

“To the best of our knowledge, this is the first large, real-world population study to examine vaccine effectiveness in people with COPD using the test-negative design and influenza-specific study outcomes,” wrote Andrea S. Gershon, MD, of Sunnybrook Health Sciences Center in Toronto and colleagues. “These findings emphasize the need for more effective influenza vaccines for older COPD patients and other preventive strategies.”
 

A test-negative study design

Data suggest that 70% of COPD exacerbations are caused by infection, and influenza often is identified as the cause. Although all major COPD practice guidelines recommend seasonal influenza vaccination, the evidence indicating that vaccination reduces hospitalizations and death is limited. The inherent or corticosteroid-induced decrease in immune response to vaccination and respiratory infection among patients with COPD may reduce the effectiveness of influenza vaccination, wrote Dr. Gershon and colleagues.

The investigators used a test-negative design to evaluate how effectively influenza vaccination prevents laboratory-confirmed influenza–associated hospitalizations in community-dwelling older patients with COPD. They chose this design because it attenuates biases resulting from misclassification of infection and from differences in health care–seeking behavior between vaccinated and unvaccinated patients.

Dr. Gershon and colleagues examined health care administrative data and respiratory specimens collected from patients who had been tested for influenza during the 2010-2011 to 2015-2016 influenza seasons. Eligible patients were aged 66 years or older, had physician-diagnosed COPD, and had been tested for influenza within 3 days before and during an acute care hospitalization. The researchers determined influenza vaccination status using physician and pharmacist billing claims. They obtained demographic information through linkage with the provincial health insurance database. Multivariable logistic regression allowed Dr. Gershon and colleagues to estimate the adjusted odds ratio of influenza vaccination in people with laboratory-confirmed influenza, compared with those without.
 

Effectiveness did not vary by demographic factors

The investigators included 21,748 patients in their analysis. Of this population, 3,636 (16.7%) patients tested positive for influenza. Vaccinated patients were less likely than unvaccinated patients to test positive for influenza (15.3% vs. 18.6%). Vaccinated patients also were more likely to be older; live in an urban area; live in a higher income neighborhood; have had more outpatient visits with a physician in the previous year; have received a prescription for a COPD medication in the previous 6 months; have diabetes, asthma, or immunocompromising conditions; have a longer duration of COPD; and have had an outpatient COPD exacerbation in the previous year.

The overall unadjusted estimate of vaccine effectiveness against laboratory-confirmed influenza–associated hospitalizations was 21%. Multivariable adjustment yielded an effectiveness of 22%. When Dr. Gershon and colleagues corrected for misclassification of vaccination status among people with COPD, the effectiveness was estimated to be 43%. Vaccine effectiveness did not vary significantly according to influenza season, nor did it vary significantly by patient-specific factors such as age, sex, influenza subtype, codiagnosis of asthma, duration of COPD, previous outpatient COPD exacerbations, previous COPD hospitalization, previous receipt of inhaled corticosteroids, and previous pneumonia.

One limitation of the study was the possibility that COPD was misclassified because not all participants underwent pulmonary function testing. In addition, the estimates of vaccine effectiveness in the present study are specific to the outcome of influenza hospitalization and may not be generalizable to vaccine effectiveness estimates of outpatient outcomes, said the investigators. Finally, Dr. Gershon and colleagues could not identify the type of vaccine received.

“Given that a large pragmatic randomized controlled trial evaluating influenza vaccination would be unethical, this is likely the most robust estimate of vaccine effectiveness for hospitalizations in the COPD population to guide influenza vaccine recommendations for patients with COPD,” wrote Dr. Gershon and colleagues.

An Ontario Ministry of Health and Long-Term Care Health Systems Research Fund Capacity Grant and a Canadian Institutes of Health Research operating grant funded this research. One investigator received grants from the Canadian Institutes of Health Research during the study, and others received grants from pharmaceutical companies that were unrelated to this study.

egreb@mdedge.com

SOURCE: Gershon AS et al. J Infect Dis. 2019 Sep 24. doi: 10.1093/infdis/jiz419.

Artificial intelligence (AI) and machine learning (ML) are promoted as the solution to many health care problems, but the area risks becoming technology led – with only secondary consideration to the safe clinical application of the technology, says Robert Challen, PhD.

Dr. Challen, of the University of Exeter (England), is the lead author of a recent paper that examines the short-, medium-, and long-term issues with medical applications of AI. “In the short term, AI systems will effectively function like laboratory screening tests, identifying patients who are at higher risk than others of disease, or who could benefit more from a particular treatment,” Dr. Challen said. “We usually accept that laboratory tests are useful to help make a diagnosis; however, clinicians are aware that they might not always be accurate and interpret their output in the clinical context. AI systems are no different in that they will be a useful tool so long as they are designed with safety in mind and used with a pragmatic attitude to their interpretation.”

The paper also suggests a set of short-and medium-term clinical safety issues that need addressing when bringing these systems from laboratory to bedside.

In the longer term, as more continuously learning and autonomous systems are developed, the safety risks will need to be continuously reevaluated, he added. “Any new technology comes with limitations and understanding those limitations is key to safe use of that technology. In the same way a new screening test has limitations on its sensitivity and specificity that define how it can be used, AL and ML systems have limitations on accuracy and which patients they can be used on,” Dr. Challen said. If hospitalists understand these limitations, they can participate better in their development.

Dr. Challen recommends that hospitalists help the development of AI tools by participating in studies that assess AI applications in the clinical environment. “Try to make sure that where AI research is taking place, there is strong clinical involvement.”

Reference

1. Challen R et al. Artificial intelligence, bias and clinical safety. BMJ Qual Saf. 2019 Jan 12. doi: 10.1136/bmjqs-2018-008370.

Artificial intelligence (AI) and machine learning (ML) are promoted as the solution to many health care problems, but the area risks becoming technology led – with only secondary consideration to the safe clinical application of the technology, says Robert Challen, PhD.

Dr. Challen, of the University of Exeter (England), is the lead author of a recent paper that examines the short-, medium-, and long-term issues with medical applications of AI. “In the short term, AI systems will effectively function like laboratory screening tests, identifying patients who are at higher risk than others of disease, or who could benefit more from a particular treatment,” Dr. Challen said. “We usually accept that laboratory tests are useful to help make a diagnosis; however, clinicians are aware that they might not always be accurate and interpret their output in the clinical context. AI systems are no different in that they will be a useful tool so long as they are designed with safety in mind and used with a pragmatic attitude to their interpretation.”

The paper also suggests a set of short-and medium-term clinical safety issues that need addressing when bringing these systems from laboratory to bedside.

In the longer term, as more continuously learning and autonomous systems are developed, the safety risks will need to be continuously reevaluated, he added. “Any new technology comes with limitations and understanding those limitations is key to safe use of that technology. In the same way a new screening test has limitations on its sensitivity and specificity that define how it can be used, AL and ML systems have limitations on accuracy and which patients they can be used on,” Dr. Challen said. If hospitalists understand these limitations, they can participate better in their development.

Dr. Challen recommends that hospitalists help the development of AI tools by participating in studies that assess AI applications in the clinical environment. “Try to make sure that where AI research is taking place, there is strong clinical involvement.”

Reference

1. Challen R et al. Artificial intelligence, bias and clinical safety. BMJ Qual Saf. 2019 Jan 12. doi: 10.1136/bmjqs-2018-008370.

Artificial intelligence (AI) and machine learning (ML) are promoted as the solution to many health care problems, but the area risks becoming technology led – with only secondary consideration to the safe clinical application of the technology, says Robert Challen, PhD.

Dr. Challen, of the University of Exeter (England), is the lead author of a recent paper that examines the short-, medium-, and long-term issues with medical applications of AI. “In the short term, AI systems will effectively function like laboratory screening tests, identifying patients who are at higher risk than others of disease, or who could benefit more from a particular treatment,” Dr. Challen said. “We usually accept that laboratory tests are useful to help make a diagnosis; however, clinicians are aware that they might not always be accurate and interpret their output in the clinical context. AI systems are no different in that they will be a useful tool so long as they are designed with safety in mind and used with a pragmatic attitude to their interpretation.”

The paper also suggests a set of short-and medium-term clinical safety issues that need addressing when bringing these systems from laboratory to bedside.

In the longer term, as more continuously learning and autonomous systems are developed, the safety risks will need to be continuously reevaluated, he added. “Any new technology comes with limitations and understanding those limitations is key to safe use of that technology. In the same way a new screening test has limitations on its sensitivity and specificity that define how it can be used, AL and ML systems have limitations on accuracy and which patients they can be used on,” Dr. Challen said. If hospitalists understand these limitations, they can participate better in their development.

Dr. Challen recommends that hospitalists help the development of AI tools by participating in studies that assess AI applications in the clinical environment. “Try to make sure that where AI research is taking place, there is strong clinical involvement.”

Reference

1. Challen R et al. Artificial intelligence, bias and clinical safety. BMJ Qual Saf. 2019 Jan 12. doi: 10.1136/bmjqs-2018-008370.