Overdose outcomes, chest pain triage, and more

Summaries from ACP Hospitalist Weekly.

Mortality in year after overdose lower with methadone or buprenorphine treatment

In the 12 months after nonfatal opioid overdose, patients who received methadone or buprenorphine had significantly lower mortality risk than those who received no medication treatment for opioid use disorder, a recent study found.

Researchers used datasets from Massachusetts government agencies to assess mortality in 17,568 adults without cancer who survived an opioid overdose between 2012 and 2014. They identified exposure to methadone maintenance treatment, buprenorphine, and naltrexone at monthly intervals and examined the effect of the medications on all-cause and opioid-related mortality, adjusting for age, sex, baseline anxiety diagnosis, depression diagnosis, and other variables.

Results were published online on June 19 by Annals of Internal Medicine and appeared in the Aug. 7 issue.

About 30% of survivors received medications for opioid use disorder in the 12 months after overdose: 2,040 (11%) enrolled in methadone maintenance treatment, for a median of five months (interquartile range, two to nine months); 3,022 (17%) received buprenorphine, for a median of four months (interquartile range, two to eight months); and 1,099 (6%) received naltrexone, for a median of one month (interquartile range, one to two months). About 5% of participants received more than one medication for opioid use disorder.

At 12 months, overall all-cause mortality was 4.7 per 100 person-years (95% CI, 4.4 to 5.0 per 100 person-years), and opioid-related mortality was 2.1 per 100 person-years (95% CI, 1.9 to 2.4 per 100 person-years). After adjustment, compared with no medications for opioid use disorder, methadone was associated with decreased all-cause mortality (hazard ratio [HR], 0.47; 95% CI, 0.32 to 0.71) and opioid-related mortality (HR, 0.41; 95% CI, 0.24 to 0.70). Buprenorphine was also associated with decreased all-cause mortality (HR, 0.63; 95% CI, 0.46 to 0.87) and opioid-related mortality (HR, 0.62; 95% CI, 0.41 to 0.92).

There was no association between naltrexone and all-cause mortality (HR, 1.44; 95% CI, 0.84 to 2.46) or opioid-related mortality (HR, 1.42; 95% CI, 0.73 to 2.79). However, the study authors noted that few events among those who received naltrexone precluded them from making firm conclusions.

They added that limitations of the study include its observational design, the likelihood of residual confounding, and the potential for data misclassification. They also noted the possibility of limited generalizability outside Massachusetts, a state with higher-than-average opioid-related mortality and insurance coverage.

The study highlights the underuse of effective medications for opioid use disorder and the need to increase treatment retention, according to an accompanying editorial.

“A great part of the tragedy of this opioid crisis is that, unlike in previous such crises America has seen, we now possess effective treatment strategies that could address it and save many lives, yet tens of thousands of people die each year because they have not received these treatments,” the editorialists wrote.

Stress echo and CCTA show similar safety, effectiveness for ED chest pain triage

For patients seen in the ED with low- to intermediate-risk chest pain, stress echocardiography (SE) worked as well as coronary computed tomography angiography (CCTA), a recent study found.

The trial included 400 consecutive patients without known coronary artery disease and a negative initial serum troponin level. They were randomized to either immediate CCTA (n=201) or SE (n=199). The mean patient age was 55 years, and 43% of patients were women. Most patients were ethnic minorities (46% Hispanic, 32% African-American). Results were published by JACC: Cardiovascular Imaging on June 13.

Hospitalization was significantly more common among patients who received CCTA (19%, 39 patients) than those received SE (11%, 22 patients; difference, 8% [95% CI, 1% to 15%]; P=0.026). Among patients who weren't hospitalized, ED length of stay was also longer with CCTA: 5.4 hours (interquartile range [IQR], 4.2 to 6.4 hours) versus 4.7 hours (IQR, 3.5 to 6.0 hours; P<0.001). The same was true for median hospital length of stay: 58 hours (IQR, 50 to 102 hours) versus 34 hours (IQR, 31 to 54 hours; P=0.002). During a median of two years of follow-up, major adverse cardiovascular events did not differ significantly between groups (11 with CCTA and seven with SE; P=0.47). Radiation exposure was significantly lower with SE than with CCTA.

The two-year results (which also included similar rates of repeat ED visits and hospitalizations in the two groups) support the safety of SE as a triage method for chest pain patients, the study authors said. They noted that the finding of longer length of stay with CCTA differed from previous research results and is of uncertain etiology. CCTA did provide diagnoses of more noncardiac causes of pain, although incidental findings could also have led to overdiagnosis, the authors speculated. The CCTA patients also received more pharmacotherapy for coronary artery disease, they noted.

Limitations of the study include that it was conducted at a single center during daytime weekday hours when results of either imaging method could be interpreted by experts. The study hospital did not have access to high-sensitivity troponin testing, which would have reduced the number of patients requiring any imaging.

An accompanying editorial praised the study for bringing attention to SE, “an often under-utilized testing option which is attractive given the lack of radiation and relatively low cost.” However, the editorialists felt that the preferred cardiac test still depends on a number of factors, including goals of testing, patient factors and preferences, availability of testing and expertise, and safety.

“A safe and efficient triage strategy for patients with possible acute coronary syndrome (ACS) is highly desirable, but taken together, the evidence suggests that no one test is best for evaluating all, or even the majority, of the 8 to 10 million patients presenting to the ED each year with acute chest pain,” the editorial said.

Hospital at home plus transition care reduced readmissions, improved patient satisfaction

A trial that bundled hospital-at-home (HaH) care with 30 days of postacute transitional care found shorter lengths of stay, fewer readmissions, and higher patient satisfaction compared to usual hospital care.

The case-control study included 295 patients treated under this HaH-Plus model from Nov. 18, 2014, to Aug. 31, 2017, in New York City. All had fee-for-service Medicare and acute medical illness requiring inpatient-level care. The control group consisted of 212 patients who met HaH eligibility but declined participation or were seen in the ED when an HaH admission could not be initiated. The mean age of the study patients was 74.6 years, and 68.6% were women. Results were published by JAMA Internal Medicine on June 25 and appeared in the August issue.

Image by Thinkstock
Image by Thinkstock

Compared to the controls, HaH patients had shorter acute length of stay (3.2 days vs. 5.5 days; difference, −2.3 days [95% CI, −1.8 to −2.7]; P<0.001) and lower rates of readmission (8.6% vs. 15.6%; difference, −7.0% [95% CI, −12.9% to −1.1%]; P<0.001), ED revisits (5.8% vs. 11.7%; difference, −5.9% [95% CI, −11.0% to −0.7%]; P<0.001), and skilled nursing facility admission (1.7% vs. 10.4%; difference, −8.7% [95% CI, −13.0% to −4.3%]; P<0.001). They were also more likely to rate their hospital care highly (68.8% vs. 45.3%; difference, 23.5% [95% CI, 12.9% to 34.1%]; P<0.001).

This HaH intervention differed from previous ones by including more diagnoses (19, representing 65 diagnosis-related groups) and bundling HaH with 30 days of postacute care to improve care coordination, facilitate access, and enhance postacute illness self-management. “Although a randomized trial design might have provided a more precise figure for the efficacy of postacute care coupled with HaH care, our findings suggest that this extension of HaH care enhances its effect,” the authors said.

The authors also highlighted some noteworthy differences between the groups—the HaH patients had lower urinary catheter insertion rates but also lower ratings of their pain management. The study was limited by its observational design and was subject to selection bias, although HaH patients were older and had more functional impairment than controls, which would be expected to bias outcomes in favor of controls. The results of this study led the Physician-Focused Payment Model Technical Advisory Committee (PTAC) to recommend full implementation of this model as a Medicare alternative payment model, the authors reported.

An accompanying commentary was more cautious about the model's potential, saying that “a rigorous test of the HaH-Plus payment model in select conditions seems more appropriate than the wide-scale implementation recommended by the PTAC.” Challenges include maintaining quality and safety, avoiding unintended consequences, and designing a bundled payment, the editorialists said.

Adding telemedicine to ICUs in VA hospitals reduced transfers of sickest patients

ICUs that implemented telemedicine had reductions in interhospital transfers of critically ill patients, according to a recent study of Veterans Affairs (VA) hospitals.

Researchers compared 52 ICUs in VA hospitals that implemented telemedicine support between 2011 and 2015 to 254 VA ICUs that did not have telemedicine. In total, 553,523 ICU admissions (97,256 in telemedicine hospitals and 456,267 in nontelemedicine hospitals) were analyzed. Results were published by CHEST on June 15 and appeared in the July issue.

When admissions before and after implementation of telemedicine were compared, both kinds of hospitals had decreases in the rate of interhospital transfers: from 3.46% to 1.99% in the telemedicine hospitals and from 2.03% to 1.68% in the nontelemedicine facilities (P<0.001). After adjustment for demographics, illness severity, admission diagnosis, and facility, ICU telemedicine implementation was associated with an overall reduction in transfers (relative risk [RR], 0.79; 95% CI, 0.71 to 0.87; P<0.001).

The reduction occurred in patients in all categories of illness severity except low (RR for transfer by severity: moderate, 0.77 [95% CI, 0.61 to 0.98; P=0.034]; moderate to high, 0.79 [95% CI, 0.63 to 0.98; P=0.035]; and high, 0.73 [95% CI, 0.60 to 0.90; P=0.003]). Transfers also decreased significantly among nonsurgical patients (RR, 0.82; 95% CI, 0.73 to 0.92; P=0.001) and in patients admitted with gastrointestinal (RR, 0.55; 95% CI, 0.41 to 0.74, P<0.001) and respiratory diagnoses (RR, 0.52; 95% CI, 0.38 to 0.71; P<0.001). There was no change in 30-day mortality associated with implementation of ICU telemedicine; mortality did decrease over the study period in the nontelemedicine ICUs, but their baseline mortality rates were higher than in the ICUs that got telemedicine.

The cause of the reduction in transfers of patients with gastrointestinal diagnoses is unclear, but the effect on patients with respiratory diagnoses can be explained by the importance of critical care expertise for treating patients with respiratory disease, especially those on mechanical ventilation, the authors said. Similarly, telemedicine consults could have made ICU staff more comfortable treating patients with higher severity of illness, the authors suggested. Telemedicine “may have helped to triage and not transfer patients that could be appropriately treated in local ICUs by providing remote access to critical care expertise,” they concluded.