PE in pregnancy, normal troponins

Summaries from ACP Hospitalist Weekly.

YEARS algorithm adaptation for pregnant women helped avoid CT scans for suspected PE

An algorithm specifically adapted for pregnant patients helped avoid CT pulmonary angiography in many cases of suspected pulmonary embolism (PE), a recent study found.

The prospective study included 510 pregnant women with suspected PE. They were assessed on three criteria from the YEARS algorithm: clinical signs of deep venous thrombosis, hemoptysis, and PE as the most likely diagnosis according to a physician. If none of the criteria were met and the patient's D-dimer level was less than 1000 ng/mL or if the D-dimer level was less than 500 ng/mL, PE was ruled out.

The adapted aspect of the algorithm was to perform compression ultrasonography when symptoms of deep venous thrombosis were present. If compression ultrasonography had positive findings or PE was ruled out, CT pulmonary angiography was not performed; otherwise, it was. Results were published in the March 21 New England Journal of Medicine.

Twelve of the 510 women screened (2.4%) were excluded from the study. PE was diagnosed in 20 of the remaining patients (4.0%) at baseline. During the study's three-month follow-up, no patients had PE and popliteal deep venous thrombosis was diagnosed in one patient (0.21%; 95% CI, 0.04% to 1.2%). CT pulmonary angiography was not indicated in 195 patients overall (39%; 95% CI, 35% to 44%). The algorithm was more efficient earlier in pregnancy; CT was avoided in 65% of patients in their first trimester, 46% in the second trimester, and 32% in the third trimester.

“This decreasing specificity can be explained by the physiological rise in the D-dimer level that commonly occurs during pregnancy,” noted the authors, who also pointed out that radiation is most potentially harmful to the fetus during the first trimester. The study provides evidence that the algorithm can safely be used to rule out suspected PE in pregnant women while reducing the use of CT pulmonary angiography, the authors said.

They noted that the algorithm was applied only to patients with clear suspicion of PE, not used as a primary screening test for patients with nonspecific chest symptoms. Other limitations include the study's nonrandomized design and protocol violations, many of which resulted from physician and patient concerns about radiation exposure, the authors said. “Indeed, the most prevalent risk factor for improper diagnostic management of suspected pulmonary embolism has been reported to be pregnancy,” they wrote.

Many patients in ED and hospital have troponin levels above stated upper limit of normal

Many inpatients and outpatients without symptoms of acute myocardial infarction have high-sensitivity cardiac troponin (hs-cTn) levels significantly above what is considered the upper limit of normal, a recent study found.

The prospective, observational cohort study included 20,000 consecutive inpatients and outpatients undergoing blood tests for any clinical reason at one hospital in the United Kingdom between June 29, 2017, and Aug. 24, 2017. All had their hs-cTnI concentrations measured, and the results were used to determine the 99th percentile for this population, compared to the manufacturer's stated upper limit of normal of 40 ng/L. Results were published by The BMJ on March 13.

Image by Getty Images
Image by Getty Images

In the total study population, the 99th percentile for hs-cTnI was 296 ng/L. After patients diagnosed with acute myocardial infarction (n=122) and patients who had clinical reasons for an hs-cTnI test (n=1,707) were excluded, the 99th percentile was 189 ng/L. By subgroup, the 99th percentiles were 563 ng/L in inpatients (n=4,759), 65 ng/L in outpatients (n=9,280), and 215 ng/L in ED patients (n=3,706). Levels above 40 ng/L were found in 5.4% of the total study population, 14.16% of all medical inpatients, 6.07% of ED patients, and 39.02% of patients on critical care units. Higher hs-cTnI levels were more common in older patients and in men than women.

The results show that the true 99th percentile of hs-cTnI varies by clinical setting and age and sex of the patient and “raise important questions about the applicability of the quoted [upper limit of normal] as an arbiter of type 1 acute myocardial infarction in patients who do not give a typical history consistent with this diagnosis,” the authors said. Clinicians should understand the limitations of troponin testing as a method to diagnose acute myocardial infarction and order the test only when there is a clear indication, the authors advised.

The study's limitations include the use of a single hs-cTn sample for each patient and the lack of comparison of troponin levels with clinical outcomes. “It is conceivable that the raised hs-cTn concentrations in a patient with stable disease always indicates myocardial injury or unwellness: the so-called ‘never means nothing’ hypothesis,” the authors noted.

Heart failure readmissions were on the decline before HRRP began

Readmissions of patients with heart failure declined between implementation of the Affordable Care Act (ACA) and the Hospital Readmissions Reduction Program (HRRP), regardless of whether heart failure was the primary diagnosis during hospitalization.

The retrospective study included more than 12 million Medicare hospitalizations with a principal or secondary diagnosis of heart failure between January 2008 and June 2015. Hospitalizations were categorized by whether the patient had a principal diagnosis of heart failure (24.8%), a secondary diagnosis of heart failure with a principal diagnosis of acute myocardial infarction (AMI) or pneumonia (9.8%), or a secondary diagnosis of heart failure with another principal diagnosis (65.4%). Results were published by the Journal of the American College of Cardiology on March 4.

All three studied cohorts had stable and fairly similar risk-adjusted 30-day readmission rates before passage of the ACA in March 2010: Mean monthly readmission rates were 26.1% for heart failure as principal diagnosis, 24.9% for AMI or pneumonia as principal, and 24.4% for other principal. Between the ACA's passage and implementation of the HRRP in October 2012, all three declined, by 1.09% (95% CI, 0.51% to 1.68%), 1.24% (95% CI, 0.92% to 1.57%), and 1.05% (95% CI, 0.52% to 1.58%) per year, respectively. After the HRRP launched, readmission rates for all three categories stabilized.

An important finding is how commonly patients with principal diagnoses other than heart failure were admitted and readmitted, the study authors said. “The fact that readmission rates for patients with a secondary [heart failure diagnosis] were nearly the same as the rates for patients with principal diagnosis of [heart failure] in our study suggests that any diagnosis of [heart failure] may be a more important risk factor for readmission than the principal discharge diagnosis,” they wrote. The similar reductions in readmissions across the three categories might have represented a spillover effect of efforts to reduce heart failure readmissions in response to policy changes, the authors suggested.

An accompanying editorial expressed concern about the impacts of policies to reduce readmissions on heart failure patients. “There remains uncertainty regarding whether the reported improvements in risk-adjusted [heart failure] readmission rates during the implementation of HRRP reflect actual improvements in hospital care and transition planning,” the editorialists said, noting that other recent research has shown increasing mortality among heart failure patients. “For HRRP, the evidence for potential harms and gamification of health care metrics should give us pause. Any experimental health policies require close monitoring for adverse consequences,” they wrote.

Biomarkers effective for diagnosing cardiac cause of syncope in ED patients

Measuring B-type natriuretic peptide (BNP), N-terminal proBNP (NTproBNP), and high-sensitivity cardiac troponin (hs-cTn) T and I provides useful diagnostic and prognostic information in ED patients with syncope, a recent study found.

The multicenter prospective study included 1,538 patients over 45 years of age who presented with syncope to an ED. Their BNP, NT-proBNP, hs-cTnT, and hs-cTnI concentrations were measured in a blinded fashion, and two physicians adjudicated whether the patients had cardiac syncope based on all information available, including cardiac workup and one-year follow-up. Results were published by Circulation on Feb. 25 and appeared in the May 21 issue.

Cardiac syncope was adjudicated in 15.2% of the patients, and all four biomarkers were significantly higher in patients with cardiac syncope than those without. The biomarkers' diagnostic accuracy for cardiac syncope, as quantified by the area under the curve (AUC), was 0.77 to 0.78 (95% CI, 0.74 to 0.81), which was superior to the EGSYS (Evaluation of Guidelines in SYncope Study) score. Combining BNP/NT-proBNP with hs-cTnT/hs-cTnI further improved diagnostic accuracy (AUC, 0.81).

The study also looked at major adverse cardiac events within 720 days (defined as death, CPR, life-threatening arrhythmia, implantation of pacemaker/implantable cardioverter defibrillator, acute myocardial infarction, pulmonary embolism, stroke/transient ischemic attack, intracranial bleeding, or valvular surgery) and found moderate-to-good prognostic accuracy with the biomarkers (AUC, 0.75 to 0.79). The biomarkers were superior to three comparators (ROSE and OESIL scores and San Francisco Syncope Rule) but inferior to the Canadian Syncope Risk Score, leading study authors to conclude that they could be clinically useful.

“The clinical utility of these biomarkers likely is highest in the subgroup of patients in whom the ED diagnosis remains unclear” after standard diagnostic tests, the study authors said. From this population, the biomarkers could identify patients who are safe to discharge from the ED, they explained. “For instance, cut-offs of <22.9 pg/mL for BNP, <97 pg/mL for NT-proBNP, <8 ng/L for hs-cTnT, and <2.9 ng/L for hs-cTnI allowed [the study] to identify 30% of eligible patients with a mortality risk at 30 days of 0%,” the study authors wrote.

Limitations of the study include that the results do not apply to patients who present with syncope in a setting other than the ED or who are age less than 45 years, the authors said. They recommended additional research to determine which components of the patient history, comorbidities, physical examination, and electrocardiogram results could increase the diagnostic and prognostic yield of the studied biomarkers.

Capillary refill time and lactate levels compared as targets for septic shock resuscitation

During early treatment of septic shock, targeting resuscitation based on capillary refill time instead of serum lactate levels resulted in similar 28-day mortality, a recent study found.

The multicenter, randomized trial was conducted in 28 ICUs in five countries. It randomized 424 patients with septic shock (416 of whom completed the trial) to a step-by-step, eight-hour resuscitation protocol that was aimed at either normalizing capillary refill time (peripheral perfusion targeting) or normalizing or decreasing lactate levels at rates greater than 20% per two hours. Patients were enrolled between March 2017 and March 2018. Results were published by JAMA on Feb. 17.

The primary outcome, 28-day mortality, was lower in the capillary refill group, but the difference was not statistically significant (34.9% vs. 43.4%; hazard ratio, 0.75 [95% CI, 0.55 to 1.02]; P=0.06; risk difference, −8.5% [95% CI, −18.2% to 1.2%]). The peripheral perfusion-targeted resuscitation protocol was associated with less organ dysfunction at 72 hours (mean Sequential Organ Failure Assessment [SOFA] score, 5.6 with capillary refill vs. 6.6 with lactate; P=0.045). There were no significant differences on the other secondary outcomes: death within 90 days; days free of mechanical ventilation, renal replacement therapy, or vasopressors within 28 days; and ICU and hospital length of stay.

Overall, the authors concluded that targeting normalization of capillary refill time, rather than lactate levels, did not reduce all-cause 28-day mortality, but they did highlight notable differences between the groups: the lower SOFA scores at 72 hours and a finding of lower 28-day mortality in patients with baseline SOFA scores under 10 (a predefined subgroup) in the capillary refill group. These findings suggesting potential benefit from the strategy should be confirmed by future research, the authors said.

They noted that there can be challenges to the use of capillary refill time in study and practice, including its dependence on age, sex, ambient temperature and light, and pressure applied during the maneuver. Other limitations of the study include that it was unblinded, may have been underpowered, and did not include any patients outside the ICU.

An accompanying editorial noted that the difference between the groups was close to statistical significance. “Does this mean resuscitation based on serial [capillary refill time] examinations could be superior to resuscitation based on lactate level? Perhaps,” said the editorialist. It appears that it is at least not inferior to the lactate approach, the editorial said, noting that capillary refill assessment might be particularly useful in resource-limited settings. Both strategies should continue to be evaluated prospectively, the editorialist concluded.

Postdischarge home visit program appeared most effective for patients with cognitive impairment

In a postdischarge program for high-risk older patients, those with cognitive impairment may have derived the most benefit, based on an analysis of readmission rates.

Image by Getty Images
Image by Getty Images

The retrospective study included 315 community-dwelling, hospitalized, older adults (≥60 years of age) who were at high risk for readmission (based on having an Elder Risk Assessment score ≥16). All were discharged home between January 1, 2011, and June 30, 2013, and enrolled in a care transitions program before discharge. Under the program, patients were seen at home within one to five days of discharge and again the following week by a nurse practitioner (NP) who provided medication reconciliation; chronic illness management; education on self-care and advance care planning; and acute illness, mobility, safety, and cognition assessments. Home visits by an NP or a doctor continued as needed for at least one month, along with weekly calls from a nurse case manager.

The study's primary outcome was all-cause readmissions within 30 days of the first home program evaluation, and this occurred in 17.1% of the patients. The intensity of the program varied, with 17.1% of participating patients receiving only one visit and 50.8% getting three or more within 30 days. The patients who were readmitted were less likely to have cognitive impairment than those who weren't readmitted (29.6% vs. 46.0%; P=0.03). They also were more likely to be on at least 14 medications (59.3% vs. 44.4%; P=0.047) and had more visits to the ED (0.8 vs 0.4; P=0.03) and primary care (4.0 vs. 3.0; P=0.018) during follow-up. Results were published by the Journal of Hospital Medicine on Feb. 20 and appeared in the June issue.

The study tried to determine whether any of the program components (e.g., medication reconciliation, advance directives discussion) were particularly associated with reduced readmissions, but no single component was found to reliably predict readmission risk. “This result is consistent with prior works that demonstrated the need for multifaceted and intensive interventions to reduce readmission risk among highly complex and multimorbid patients,” the authors said. They noted that other research has found advance directive discussions to reduce readmissions and that toward the end of the study, they extended their program for patients with limited life expectancy and goals of care.

Multivariable analysis of the study's results found only cognitive impairment and previous ED visits to be significant predictors of readmission. The study was limited by its retrospective and single-center design, among other limitations, the authors said. Still, they concluded that the results “suggest that compared with cognitively functional patients, cognitively impaired patients with conservative goals of care may be more likely to avoid burdensome hospitalizations when provided with early intervention in their home.”

New tetracycline noninferior to standard-of-care antibiotics for two common infections

Two recent phase 3 double-blind randomized clinical trials showed that omadacycline, a new-generation tetracycline, was noninferior to standard-of-care antibiotics for acute bacterial skin and skin-structure infections (ABSSI) and community-acquired bacterial pneumonia (CABP). Results of the manufacturer-funded trials, which led to FDA approval in fall 2018, were published online on Feb. 7 by the New England Journal of Medicine.

In the ABSSI trial, researchers enrolled adults with infections involving a contiguous surface area of at least 75 cm2 and exhibiting clear evidence of erythema, edema, or induration. They randomly assigned patients to receive intravenous omadacycline (100 mg given every 12 hours for two doses, then given every 24 hours; n=329) or intravenous linezolid (600 mg given every 12 hours; n=326). After three days, a transition to oral omadacycline (300 mg every 24 hours) or oral linezolid (600 mg every 12 hours) was permitted, for a total treatment duration of seven to 14 days. The primary endpoint was early clinical response at 48 to 72 hours, defined as survival with a 20% or more reduction in lesion size without rescue antibacterial therapy.

In a modified intention-to-treat population, omadacycline (n=316) was noninferior to linezolid (n=311) regarding early clinical response (rate of response, 84.8% vs. 85.5%, respectively; difference, −0.7 percentage points [95% CI, −6.3 to 4.9 percentage points]). In both groups, the drug's efficacy was similar for methicillin-susceptible and methicillin-resistant Staphylococcus aureus infections. Adverse events were reported in 48.3% of patients in the omadacycline group and 45.7% of those in the linezolid group, and the most frequent events in both groups were gastrointestinal (18.0% and 15.8% of patients, respectively). Serious adverse events were reported in 3.7% of patients in the omadacycline group and 2.5% of those in the linezolid group, and the mortality rate in each group was 0.3% and 0.6%, respectively.

In the CABP trial, researchers randomly assigned hospitalized adults with CABP who were not in the ICU to receive intravenous omadacycline (100 mg every 12 hours for two doses, then 100 mg every 24 hours; n=386) or intravenous moxifloxacin (400 mg every 24 hours; n=388). After three days, a transition to oral omadacycline (300 mg every 24 hours) or oral moxifloxacin (400 mg every 24 hours) was permitted, for a total treatment duration of seven to 14 days. CABP-causing pathogens were identified at baseline in 49.9% of patients in the intention-to-treat population. The primary endpoint, early clinical response, was defined as survival with improvement in at least two of four symptoms (cough, sputum production, pleuritic chest pain, and dyspnea) and no worsening of symptoms at 72 to 120 hours, without receipt of rescue antibacterial therapy.

For early clinical response, omadacycline was noninferior to linezolid (rate of response, 81.1% vs. 82.7%, respectively; difference, −1.6 percentage points [95% CI, −7.1 to 3.8 percentage points]). Adverse events after treatment initiation, most commonly gastrointestinal, were reported in 41.1% of patients in the omadacycline group and 48.5% of those in the moxifloxacin group. Serious adverse events were reported in 6% to 7% of patients in each group, and the mortality rate was 2.1% in the omadacycline group and 1.0% in the moxifloxacin group.

Both of the trials were of high quality and enabled FDA approval of omadacycline to treat patients with ABSSI and CABP, according to an accompanying editorial. “Yet within the context of the growing threat of drug-resistant bacterial pathogens (carbapenem-resistant gram-negative pathogens most especially) and the urgent need for new agents active against them, one must ask: So what?” the editorialist wrote.

Omadacycline has few advantages over the various agents that are already available to treat these infections, and its role for treating infections caused by multidrug-resistant pathogens, while promising, remains unknown, the editorial said. “Well designed clinical trials of omadacycline for the treatment of infections caused by multiple-drug–resistant gram-negative pathogens are needed to determine its real value as an antibacterial agent,” the editorialist wrote.

Oseltamivir effectiveness may vary by flu strain in severely ill patients

Prompt treatment with oseltamivir is indicated in severely ill patients with suspected influenza, but its effectiveness appears to vary by strain, according to a recent study.

The study included adults with laboratory-confirmed influenza who were hospitalized in ICUs in Greece during eight influenza seasons (from 2010 to 2011 through 2017 to 2018), all of whom received oseltamivir. Researchers looked at the association between early oseltamivir treatment and death and cause-specific and subdistribution hazards for death and discharge, stratifying effect estimates by type of influenza. Early treatment was defined as oseltamivir within 48 hours of symptom onset. The primary outcome was death in the ICU versus live discharge from the ICU, with length of ICU stay as a secondary outcome. The study results were published by Clinical Infectious Diseases on Feb. 7.

Overall, 1,330 patients were included in the study. Most patients were men, were older than 50 years of age, and had at least one comorbid condition. Influenza A/H1N1 was the most common strain, affecting slightly more than half of patients; the remainder had influenza A/H3N2 and influenza B. Overall, 622 patients (46.8%) died in the ICU. Early treatment with oseltamivir was associated with significantly lower mortality rates in patients with influenza A/H3N2 (relative risk, 0.69; 95% credible interval [CrI], 0.49 to 0.94), due entirely to a higher cause-specific hazard for discharge. Patients with influenza A/H3N2 who survived had a model-predicted median length of ICU stay that was 1.8 days shorter with early treatment than with late treatment. Early treatment with oseltamivir did not seem to have an effect on mortality in patients with influenza A/H1N1 or those with influenza B.

The authors noted that no data were available on dosage and duration of oseltamivir treatment, that safety outcomes of the drug were not evaluated, and that their study was observational. However, they concluded that “in the absence of randomized trials, our study provides strong and important new evidence about oseltamivir use in critically ill influenza patients” and “changes current thinking, by indicating that oseltamivir effectiveness may depend on influenza type.” They called for additional randomized studies to replicate their findings in patients with differing illness severity.