COPD meds, endocarditis, and more

Recent research from ACP Hospitalist Weekly.


Long-acting COPD meds often discontinued during hospital stay and not restarted

Long-acting bronchodilators in patients with chronic obstructive pulmonary disease (COPD) are often held at hospital admission and not restarted at discharge, according to recent research.

Researchers performed a retrospective cohort study using health administrative data from 2004 to 2016 in Ontario, Canada, to determine the risk of unintentional discontinuation of long-acting muscarinic antagonists (LAMAs) and long-acting beta-agonist and inhaled corticosteroid (LABA-ICS) combination therapy after hospital discharge in older adults with COPD. Patients with COPD who were ages 66 years and older and who had filled LAMA or LABA-ICS prescriptions continuously for at least one year were included and were categorized as having been hospitalized, having visited the ED, or neither. The study's primary outcome was medication discontinuation, defined as nonreceipt of the medication that the patient had been regularly taking within 90 days after the index date, in those who were hospitalized versus those who were not. Results were published Oct. 1, 2020, by CHEST and appeared in the March 2021 issue.

Image by Getty Images
Image by Getty Images

Overall, 27,613 hospitalizations and 56,057 ED visits in 78,953 patients who had COPD and were highly adherent to medications were included in the study. A total of 18,330 hospitalizations occurred among 69,253 patients continuously using LAMAs, and 9,283 occurred among 36,439 continuous users of LABA-ICS. COPD medications were discontinued 1,466 times among 27,613 discharges. Patients who were hospitalized were at higher risk for medication discontinuation than those who remained in the community (adjusted risk ratios, 1.50 [95% CI, 1.34 to 1.67] for LAMAs and 1.62 [95% CI, 1.39 to 1.90] for LABA-ICS; P<0.001 for both comparisons). Crude rates of discontinuation were 5.2% and 3.3% in the hospitalization group and the community group, respectively, for LAMAs and 5.5% and 3.1%, respectively, for LABA-ICS.

The authors noted that their study was observational and that not all potential causes of medication interruption could be considered, among other limitations. They concluded that based on their results, older adults with COPD who are highly adherent to medications are at risk for unintentional discontinuation of long-acting bronchodilators after hospitalization. “Considering the adverse health outcomes that may be associated with gaps in drug continuity, these findings underscore a need for further prospective clinical study in average, less adherent COPD patients as well as standardized practices to prevent medication errors at hospital discharge in the COPD population,” the authors wrote. “Future studies should also study medication overuse as well.” They noted that quality improvement efforts should focus on safe transitions and on medication reconciliation after discharge.

Addiction treatment rare after hospitalization for endocarditis related to injection drug use

Treatment with medication for opioid use disorder (OUD) was uncommon in a cohort of patients hospitalized with injection drug use-associated endocarditis, a recent study found.

Researchers used the Massachusetts Public Health Data Warehouse to identify residents ages 18 to 64 years who were hospitalized for injection drug use-associated endocarditis and survived for at least two months after discharge between July 1, 2011, and June 30, 2015. They defined injection drug use-associated endocarditis as any hospitalization with a diagnosis of endocarditis and at least one claim in the prior six months for OUD, drug use, or hepatitis C virus. They also assessed the proportion of patients who received medication for OUD, defined as any treatment with methadone, buprenorphine, or naltrexone within three months after discharge (excluding discharge month), before and after hospitalization. The primary outcome was all-cause mortality up to one year after discharge. Results were published on Oct. 14, 2020, by JAMA Network Open.

Of 679 patients (60.8% men; mean age, 39.2 years), 419 (72.3%) had mental illness and 209 (30.8%) experienced homelessness. Overall, 134 (19.7%) received medication for OUD in the three months before hospitalization, and this number increased to 165 (24.3%) in the three months after discharge. Patients were more likely to stop existing OUD treatment than they were to start new medication for OUD after hospitalization. They were also more likely to receive opioid analgesia than medication for OUD. Buprenorphine was the most common treatment for OUD both before (n=86 [64.2%]) and after (n=112 [67.9%]) hospitalization. In the 12 months after discharge, 61 patients (8.9%) died, for a crude mortality rate of 9.2 deaths per 100 person-years. Although receiving medication for OUD within three months after discharge was not associated with reduced all-cause mortality (adjusted hazard ratio, 1.29; 95% CI, 0.61 to 2.72), it was associated with reduced mortality in the month when it was received (adjusted hazard ratio, 0.30; 95% CI, 0.10 to 0.89).

The study authors noted that it is likely they observed no mortality benefit in the intention-to-treat model because more than 40% of patients receiving medication for OUD stopped treatment within six months of discharge. Limitations of the study include its observational design, the possibility of misclassification, and the fact that it excluded individuals who did not survive two months after discharge, they said.

“[W]e found that hospitalization for [injection drug use-associated endocarditis] was an underused opportunity to initiate [medication for] OUD,” they concluded. “The findings suggest that patient navigation through [medication for] OUD initiation, linkage to an outpatient provider to continue pharmacotherapy, and development, evaluation, and implementation of interventions to improve retention are needed.”

Suctioning frequency, high-flow oxygen can determine when to decannulate tracheostomy

Basing the decision to decannulate a tracheostomy tube on suctioning frequency plus continuous high-flow oxygen therapy rather than on the standard 24-hour capping trial reduced the time to decannulation without increasing failure rate, a study found.

Researchers enrolled conscious, critically ill adults who had a tracheostomy tube in five ICUs in Spain. Patients were eligible after weaning from mechanical ventilation. Patients were randomly assigned either to undergo a 24-hour capping trial plus intermittent high-flow oxygen therapy (control group; n=161) or to receive continuous high-flow oxygen therapy with frequency of suctioning being the indicator of readiness for decannulation (intervention group; n=169). Patients underwent decannulation when they had had no more than two aspirations every eight hours during a 24-hour period according to prespecified indications. Results were published Sept. 10, 2020, by the New England Journal of Medicine.

Commercial support came from the maker of the high-flow oxygen therapy device and interface for tracheostomy tubes, Fisher and Paykel Healthcare, which paid for writing assistance with the manuscript but had no role in the study's design or conduct or in the decision to submit the manuscript for publication.

The patients' mean age was 58.3 years; 68.2% were men. The primary outcome of time to decannulation was shorter in the intervention group than in the control group: median, 6 days (interquartile range, 5 to 7 days) versus 13 days (interquartile range, 11 to 14 days) (absolute difference, 7 days [95% CI, 5 to 9 days]).

Secondary outcomes included decannulation failure, weaning failure, respiratory infections, sepsis, multiorgan failure, durations of stay in the ICU and hospital, and deaths in the ICU and hospital. Pneumonia occurred in 16 patients (9.9%) in the control group and seven (4.1%) in the intervention group (difference, 5.8 percentage points; 95% CI, 0.2 to 11.8). Tracheobronchitis occurred in 47 patients (29.2%) in the control group and 32 (18.9%) in the intervention group (difference, 10.3 percentage points; 95% CI, 1.0 to 19.3). Median length of stay was 62 days (interquartile range, 38 to 105 days) in the control group and 48 days (interquartile range, 33 to 71 days) in the intervention group (absolute difference, 14 days; 95% CI, 9 to 33 days). Other secondary outcomes were similar between the two groups.

The authors concluded that this strategy of decannulation based on suctioning frequency plus the use of continuous high-flow oxygen therapy led to faster decannulation and no increase in decannulation failure compared to standard care. “The most plausible explanation for this result is that capping trials are highly demanding, thus delaying the time to decannulation as reflected by the high proportions of patients with capping trials that failed and of patients with weaning failure,” they wrote.

An editorial noted that in the U.S., patients are often transferred to post-acute care facilities for weaning from a mechanical ventilator and for decannulation. This trial took place in ICUs in Spain, where the ratio of nurses and respiratory therapists to patients probably differs. Also, patients in the intervention group spent more time receiving high-flow oxygen therapy than those in the control group, who received high-flow oxygen therapy when the tube was not capped. It is possible that the heated humidification of high-flow oxygen decreased the frequency of suctioning events, the editorial said.

“The authors have taken an important step in building an evidence base to improve care for patients with chronic critical illness,” the editorial stated. “The generation of new knowledge should not end when a patient has survived acute illness.”

PE risk assessment tools modestly effective for estimating mortality

Four commonly used risk assessment methods for acute pulmonary embolism (PE) are modestly effective at estimating mortality, according to a recent study.

Researchers performed a multicenter cohort study between October 2016 and October 2017 at eight hospitals participating in the Pulmonary Embolism Response Team (PERT) Consortium registry. The goal of the study was to examine short-term outcomes in patients with acute PE and assess the ability of different risk scores to determine which patients were at high and low risk for death. The study looked at the Pulmonary Embolism Severity Index (PESI), the simplified PESI (sPESI), and the Bova score, as well as a system developed by the European Society of Cardiology (ESC). Although the ESC usually recommends adding biomarkers and radiologic markers to PESI or sPESI, the researchers instead determined each patient's ESC risk according to presence of shock or hypotension, signs of right ventricular dysfunction on echocardiogram or CT, and elevated cardiac biomarkers; PESI and sPESI scores were not included to avoid collinearity when risk assessments were compared. Patients were included if they presented with acute PE and had enough information available in the medical record for risk scores to be calculated. The study's main outcome measures were all-cause mortality at seven and 30 days and associated discrimination, as assessed by the area under the receiver-operating curve (AUC). Results were published Aug. 26, 2020, by JAMA Network Open.

A total of 416 patients with acute PE were included in the study. The mean age was 61.3 years, and 207 (49.8%) were men. One hundred eighty-eight patients (45.2%) received anticoagulation alone, 32 (77%) received thrombolysis, 11 (2.6%) received thrombectomy, and 106 (25.5%) received other advanced interventions. All-cause death occurred within seven days in 25 patients (6.0%) and within 30 days in 51 patients (12.3%). In groups judged to be low risk according to the risk scores, seven-day mortality rates ranged from 1.3% (n=1) to 3.1% (n=4) and 30-day mortality rates ranged from 2.6% (n=1) to 10.2% (n=13). In groups judged to be at highest risk, seven-day mortality rates ranged from 7.0% (n=18) to 16.3% (n=7) and 30-day mortality rates ranged from 14.4% (n=37) to 26.3% (n=26). Each risk stratification tool was consider to have modest discrimination for seven-day mortality (AUC range, 0.616 to 0.666) and slightly lower discrimination for 30-day mortality (AUC range, 0.550 to 0.694).

The authors noted that their study did not capture all cases of acute PE at the participating centers and that all-cause rather than PE-specific mortality was measured, among other limitations. They concluded that the four commonly used PE risk assessment methods examined in this study showed only modest discrimination and ability to estimate short-term mortality in patients with acute PE. “Furthermore, there appears to be minimal association among the different risk scores for individual patients,” the authors wrote. “Future studies to develop and validate better risk assessment tools would improve both clinical care and PE-related research.”

Risk model assesses function in older patients after hospitalization for acute MI

A new risk model may help predict declines in activities of daily living in older adults who are hospitalized for acute myocardial infarction (MI), a recent study found.

Researchers used data from the prospective SILVER-AMI (Comprehensive Evaluation of Risk Factors in Older Patients With Acute Myocardial Infarction) study to evaluate the incidence of functional decline in a nationwide cohort study of older adults hospitalized with acute MI and develop a risk model to help identify at-risk patients. Data on demographic, cardiac, geriatric, psychosocial, and lifestyle variables were collected during hospitalization and six months afterward. Functional decline was defined as decreased ability to independently perform essential activities of daily living (such as bathing, dressing, transferring, and ambulating) from baseline to six months postdischarge. The primary outcome was a decrease in at least one activity of daily living at six months. The results of the study were published Oct. 1 by the Journal of the American Heart Association.

Image by Getty Images
Image by Getty Images

Overall, 2,555 patients with acute MI were included in the main study analysis. Of these, 1,709 were randomly assigned to the derivation cohort and 846 were randomly assigned to the validation cohort. Mean age was 81.3 years, 56.7% were men, and 90.3% were White. A total of 327 patients (12.8%) reported declines in activities of daily living at six months after discharge, with older age (odds ratio [OR], 1.03; 95% CI, 1.00 to 1.07 per year), longer hospital stay (OR, 1.05; 95% CI, 1.02 to 1.08 per day), mobility impairment during hospitalization (ORs, 3.00 [95% CI, 1.79 to 5.04] and 6.67 [95% CI, 3.89 to 11.40] for moderate and severe impairment, respectively), and depression (OR, 1.92; 95% CI, 1.33 to 2.79) identified as risk factors. In addition, decline in activities of daily living was more common in patients who reported being “about as active” as or “less active” than their peers (ORs, 1.67 [95% CI, 1.16 to 2.41] and 1.71 [95% CI, 1.09 to 2.69], respectively) versus “more active.” Factors associated with decreased risk for functional decline were revascularization during hospitalization (ORs, 0.52 [95% CI, 0.35 to 0.78] for percutaneous coronary intervention and 0.24 [95% CI, 0.12 to 0.48] for coronary artery bypass grafting) and ability to walk a quarter mile before acute MI (OR, 0.70; 95% CI, −0.50 to 0.97). The researchers found that model discrimination (c=0.78) and calibration were both good.

Patients' reported function one month before hospitalization was considered to be their function at baseline, and some patients may have had functional decline at the time of hospitalization that resolved before six months, the authors noted. They also said that their model requires external validation before use in clinical settings. “After external validation, use of this tool may improve treatment planning and shared decision-making for older patients with [acute MI] at risk for this important patient-centered outcome,” they concluded.

Trauma-informed care can help hospitalized patients with PTSD

Physicians seeing hospitalized patients with post-traumatic stress disorder (PTSD) should use the “four R's” approach to provide trauma-informed care, according to a recent narrative review.

The four R's are realization that trauma affects an individual's coping strategies, relationships, and health; recognition of the signs of trauma; an appropriate, planned response to patients identified as having experienced trauma; and resisting retraumatization in the hospital or other care setting, the researchers wrote. Their review including recommendations on how to provide such care in patients with PTSD was published Aug. 19, 2020, by the Journal of Hospital Medicine and appeared in the January 2021 issue.

Physicians should look at the medical history to see if the patient has been diagnosed with PTSD in the past, the authors said. Comorbid conditions often associated with PTSD include mood, anxiety, or substance use disorders. If PTSD is suspected, it is reasonable to consider screening as a next step, according to the review. Partnership with psychiatry colleagues is strongly encouraged in patients with refractory PTSD symptoms, and comorbidities such as alcohol and opioid use disorders must be addressed. The authors also stressed the importance of an individual approach to care that identifies ways to meet each patient's specific needs.

Clinical situations common to hospitalized patients with PTSD include sleep problems, pain, and anxiety and anger, the authors said. Sleep hygiene is often poor in the hospital, and patients with PTSD are in need of a sleep routine. The authors recommended not taking vital signs overnight if possible, asking nursing staff to encourage the patient to turn off the TV at least an hour before a planned bedtime, discouraging napping, avoiding IV fluids in the evening, and planning diuretic doses so that they have worn off by bedtime. Physicians should also ask about the patient's bedtime routines at home and try to incorporate them into the hospital. Patients who use continuous positive airway pressure at home should continue to do so during the inpatient stay, the authors said.

Pain may be associated with the patient's acute problem, but chronic pain and use of substances to self-treat can also be present in PTSD. Physicians should use adjunctive therapies, such as heat, topical agents, and meditation, as first-line pain treatment in patients with PTSD. Nonopioid medications (e.g., acetaminophen, NSAIDs) should always be considered before opioids. If opioids are necessary, physicians should choose oral rather than IV medications and establish a short, fixed time frame for their use, the authors said.

Patients with PTSD who experience anger and anxiety may need extra time and attention from their physicians, and nonpharmacological treatments should be prioritized, according to the authors. They noted that asking patients about their PTSD triggers, such as determining the best way to prevent startle reactions, can help prevent exacerbations. Listening, validating, and negotiating are effective strategies, the authors stressed, noting that benzodiazepines and antipsychotics should be avoided. Other suggestions include using soft music and limiting alarms in the patient's room, asking the patient about lighting preferences at night, and keeping the patient informed about the day's schedule, the authors said.

The authors noted that PTSD is common and that adhering to the principles of trauma-informed care and partnering with patients to develop a specific plan can help make hospitalizations go more smoothly. “Further research should evaluate hospital outcomes that result from a more tailored approach to the care of patients with PTSD,” the authors wrote. “More effective, patient-centered PTSD care could lower rates of leaving against medical advice and improve the inpatient experience for patients and providers alike.”