In the News

Malpractice claims, fenoldopam after cardiac surgery in acute kidney injury, beta-lactams and community-acquired pneumonia, and more.

Fewer malpractice claims made against hospitalists than other groups

Hospitalists have a lower rate of malpractice claims than nonhospitalist internists, ED physicians, or general surgeons, a recent study found.

Researchers used a liability insurer's database of more than 52,000 malpractice claims to assess the frequency and causes of claims against hospitalists compared to other groups. They found that hospitalists had a rate of 0.52 claim per 100 physician coverage years (PCYs) compared to 1.91 per 100 PCYs for nonhospitalist internists, 3.50 per 100 PCYs for ED physicians, 4.70 per 100 PCYs for general surgeons, and 5.56 per 100 PCYs for obstetrician-gynecologists (P<0.001 for all comparisons).

The most common allegations made against hospitalists were errors in medical treatment (41.5% of claims) and diagnosis (36.0% of claims). Deficiencies in clinical judgment and communication were the most common contributing factors to claims, at 54.4% and 36.4%, respectively. Transition-of-care problems were a factor in 37.9% of the claims, and 50.4% of the claims involved the death of a patient. Results were published in the December 2014 Journal of Hospital Medicine.

The researchers concluded that, contrary to concerns about the risks of the practice model, hospitalists have a significantly lower rate of malpractice claims compared to the studied types of physicians, including other internists. The severity of injury involved in the claims was high, but that may relate to the greater severity of illness in hospitalists' patients. It's uncertain why hospitalists would have lower malpractice rates, but it may be because hospitalists face less diagnostic uncertainty in their patients or because they provide a high quality of care, the authors speculated.

The results also suggest that despite hospitalists' short-term relationships with patients, they are able to build connections that meet patients' expectations, perhaps because of the frequency of interaction, the authors said. Based on the findings, “liability fears should not impede the adoption of the hospitalist model” and the possibility of decreased liability cost should be considered in estimates of hospitalists' cost of care, the study authors concluded.

In addition to continuing to work on improving communication, hospitalists should look for ways to reduce deficiencies in clinical judgment, according to an accompanying editorial. Solutions could include diagnosis-related quality measures or strategies to teach enhanced clinical reasoning skills. However, overall, “hospitalists should remember that the most productive way to approach malpractice risk is reframe the problem as one that attempts to reduce risk for patients, rather than physicians,” the editorialists wrote.

No protective renal effect noted with fenoldopam in AKI patients after cardiac surgery

Fenoldopam does not appear to reduce the need for renal replacement therapy (RRT) in patients with acute kidney injury (AKI) after cardiac surgery, according to a recent study.

Researchers performed a randomized, double-blind, placebo-controlled, parallel-group trial in 19 ICUs in Italy to examine whether the need for RRT after cardiac surgery in critically ill patients with AKI would be decreased by postsurgical use of fenoldopam. Patients admitted to ICUs with early AKI after cardiac surgery were randomly assigned to receive continuous infusion of fenoldopam or placebo for up to 4 days; the starting dose was 0.1 μg/kg/min, with a range of 0.025 to 0.3 μg/kg/min. Early AKI was defined as at least a 50% increase in serum creatinine level from baseline or oliguria for at least 6 hours. The study's primary end point was rate of RRT, while secondary end points were mortality rate in the ICU and at 30 days, as well as hypotension rate during infusion of the study drug. The results were published in the Dec. 3, 2014, Journal of the American Medical Association.

The study took place from March 2008 to April 2013. Three hundred thirty-eight patients were assigned to the fenoldopam group, and 329 were assigned to the placebo group. The mean patient age was 70 years. Four hundred twenty-three patients (64%) were men, and 290 (48%) had New York Heart Association class III or IV heart failure. All patients completed 30 days of follow-up after surgery. After a planned interim analysis, the study's safety committee recommended that the study be stopped because of futility. Overall, 69 of 338 patients in the fenoldopam group (20%) and 60 of 329 patients in the placebo group (18%) experienced AKI progression that required RRT (P=0.47). Mortality rates at 30 days were 23% in the fenoldopam group and 22% in the placebo group (P=0.86). Hypotension rates were 26% in the fenoldopam group and 15% in the placebo group, a statistically significant difference (P=0.001). Systolic blood pressure was also statistically significantly lower in the fenoldopam group compared with the placebo group 1 hour after the study drug was started (120 mm Hg vs. 124 mm Hg; P=0.003).

Limitations of the study included that it was stopped early, that it included fewer patients than had been planned, and that information on intensity and duration of hypotension was not collected, the authors noted. They also pointed out that their findings are different from those of several other previous studies, although those studies were small. However, the authors concluded that in patients with AKI after cardiac surgery, fenoldopam infusion did not reduce the need for RRT or risk of 30-day mortality compared with placebo but was associated with increased hypotension rates. “Given the cost of fenoldopam, the lack of effectiveness, and the increased incidence of hypotension, the use of this agent for renal protection in these patients is not justified,” they wrote.

Trial fails to find noninferiority for beta-lactams vs. beta-lactams plus macrolides in CAP

A noninferiority trial for treatment of community-acquired pneumonia patients was unable to prove that beta-lactam alone was noninferior to beta-lactams plus macrolides.

Researchers conducted an open-label trial from Jan. 13, 2009, to Jan. 31, 2013, of 580 patients from 6 Swiss acute care hospitals who had moderately severe community-acquired pneumonia (CAP). Patients were randomized to receive a beta-lactam and a macrolide, or a beta-lactam alone. The beta-lactam could be cefuroxime (1.5 g, 3 times daily intravenously) or amoxicillin and clavulanic acid (1.2 g, 4 times daily intravenously). The macrolide was clarithromycin, 500 mg twice daily, orally or intravenously. Urine samples were tested regularly for Legionella pneumophila, and a macrolide was added for monotherapy patients who tested positive. The main outcome was the proportion of patients who didn't reach clinical stability at day 7. Stability was described as heart rate <100 beats/min, systolic blood pressure >90 mm Hg, temperature <38.0 °C, respiratory rate <24 breaths/min, and oxygen saturation >90% on room air.

Forty-one percent of patients in the monotherapy arm compared to 33.6% of patients in the combination arm had not reached clinical stability after 7 days of treatment, a difference that was nonsignificant (P=0.07). There were more 30-day readmissions in the monotherapy patients (7.9% vs. 3.1%; P=0.01). Patients infected with atypical pathogens (hazard ratio [HR], 0.33) or with Pneumonia Severity Index (PSI) category IV pneumonia (HR, 0.81) were less likely to reach clinical stability with monotherapy. Patients not infected with atypical pathogens (HR, 0.99) or with PSI category I to III pneumonia (HR, 1.06) had equivalent outcomes in both groups. There were no differences between patient groups in mortality, ICU admission, complications, length of stay, and recurrence of pneumonia within 90 days. Results were published in the December 2014 JAMA Internal Medicine.

Researchers noted the results indicate a trend toward superiority of combination therapy of beta-lactams with macrolides, which may have been statistically nonsignificant due either to chance or to lack of power in the study design. Invited commenters said the trial “pushes the pendulum further in favor of antibiotic therapy covering atypical and typical bacterial pathogens for patients hospitalized with CAP.” While more research on the topic is being done, “dual therapy should remain the recommended treatment for patients hospitalized with CAP,” they wrote.

Clinical guideline offers recommendations on reducing recurrent kidney stones

Drinking more fluid to increase urination may reduce the risk of recurrence of kidney stones, a clinical guideline recommended. Monotherapy with a thiazide diuretic, citrate, or allopurinol can be used if increasing fluid intake is not effective, the guideline also stated.

Researchers reviewed published literature from 1948 to March 2014 to create the evidence-based clinical guideline, which appeared in the Nov. 4, 2014, Annals of Internal Medicine.

Patients who have had kidney stones should increase their fluid intake throughout the day to achieve at least 2 liters of urine per day to prevent a recurrence, according to the guideline (weak recommendation, low-quality evidence). Low-quality evidence from 1 study showed that patients with calcium stones who achieved this fluid intake had less recurrence than the control group (12.1% vs. 27.0%; follow-up, 60 months). Another trial that examined radiographic stone recurrence in patients with calcium stones showed a nonstatistically significant decrease in recurrence in patients with increased fluid intake compared with no treatment (8.0% vs. 55.6%; follow-up, 24 to 36 months).

If increasing fluid intake doesn't work, patients can try monotherapy with a thiazide diuretic, citrate, or allopurinol to prevent recurrent kidney stones (weak recommendation, moderate-quality evidence), the guideline also stated. Moderate-quality evidence from 6 fair-quality, placebo-controlled trials of thiazide diuretic showed that the risk for composite stone recurrence was lower in patients who were treated with thiazide than in those who weren't (24.9% vs. 48.5%). No significant differences were found in the risk for recurrence based on thiazide type or dosage.

Moderate-quality evidence from 6 placebo-controlled trials of citrate monotherapy showed that composite stone recurrence was lower in patients treated with citrate than in controls (11.1% vs. 52.3%), regardless of which type of citrate was used. One fair-quality trial showed no difference between citrate and control groups in risk for radiographic stone recurrence.

Moderate-quality evidence from 4 placebo-controlled trials of allopurinol monotherapy showed a reduced risk for composite stone recurrence with allopurinol. Two trials showed that risk for recurrence was lower in patients treated with allopurinol than in those who received placebo (33.3% vs. 55.4%). One fair-quality study showed a reduction in symptomatic stone recurrence (10.3% vs. 29.0%), although it found no difference in recurrence of radiographic stones between the treatments.

The evidence also showed that patients who decreased intake of soda that was acidified by phosphoric acid had reduced kidney stone recurrence. Low-quality evidence from 1 study showed that patients with baseline soft drink consumption of more than 160 mL per day who were instructed to abstain from drinking soda had a reduced risk for symptomatic stone recurrence compared with no treatment (33.7% vs. 40.6%). The benefit was limited to patients who drank soda that was acidified by phosphoric acid (typically colas) rather than those acidified by citric acid (typically fruit-flavored sodas).

Handoff intervention reduces errors, adverse events at multiple hospitals

A handoff improvement program successfully reduced medical errors and preventable adverse events without slowing workflow, according to a study conducted at 9 pediatric hospitals.

The program included a mnemonic to standardize oral and written handoffs, handoff and communication training, a faculty development and observation program, and a sustainability campaign. To measure its effectiveness, researchers compared adverse events, medical errors (measured by active surveillance), handoff quality (measured by printed documents and audio recordings), and workflow (measured by time-motion observations) in more than 10,000 admissions. Results were published in the Nov. 6, 2014, New England Journal of Medicine.

In the 9 studied hospitals, the medical error rate after the intervention was 23% lower than it had been preintervention (18.8 errors per 100 admissions vs. 24.5; P<0.001). Preventable adverse events also dropped by 30%, from 4.7 per 100 admissions to 3.3 (P<0.001). Nonpreventable adverse events did not change significantly during the study. Handoffs improved significantly in their inclusion of all of the key elements specified by the study (oral elements: illness-severity assessment, patient summary, to-do list, contingency plans, readback; additional written elements: allergies, code status, medication list, dated laboratory results, dated vital signs).

Meanwhile, the duration of oral handoffs remained the same before and after the intervention, as did resident workflow, including patient-family contact and computer time. The results provide evidence that the I-PASS Handoff Bundle, which had been previously tried in a single center, could be successfully implemented across multiple sites without a negative effect on workflow and that “handoff-related errors can be significantly reduced,” the study authors concluded. Although the causal link between the intervention and the improvements isn't definitive, it seems likely, given the parallel improvements across hospitals and the lack of change in nonpreventable adverse events, the authors said.

At 3 of the 9 sites, error rates did not change significantly, and it's unclear why, since these hospitals also had improvements in handoff processes, the study authors said. The bundled nature of the intervention also makes it impossible to tell whether certain elements were more essential than others. Further research will also be necessary to determine how generalizable the results are to settings other than pediatric inpatient units.

In-hospital mortality rates appear worse for inpatient vs. outpatient STEMI

Patients who had an ST-elevation myocardial infarction (STEMI) during a hospitalization for a condition not related to acute coronary syndrome (ACS) received fewer invasive tests and interventions and had higher mortality rates while hospitalized than patients who had STEMI as outpatients, a recent study found.

Researchers performed a retrospective, observational analysis of data from the California State Inpatient Database on STEMIs between 2008 and 2011 to examine treatment and outcomes among patients with STEMI occurring during a hospitalization for a non-ACS condition. Patients who developed a STEMI during a hospitalization for ACS and patients who were identified as having a STEMI when they transferred to the hospital from another facility were excluded. STEMIs were classified as having inpatient or outpatient onset according to present-on-admission codes. Associations among location of STEMI onset, resource use, and hospital characteristics were the main outcome measures. The study results were published online Nov. 16 and appear in the Nov. 19 Journal of the American Medical Association.

Overall, the researchers identified 62,021 STEMIs in 303 hospitals that were present on admission (58,953, or 95.1%) or occurred in the hospital (3,068, or 4.9%). Patients who had STEMI in the hospital were older than those who had STEMI in an outpatient setting (mean age, 71.5 years vs. 64.9 years; P<0.001) and were also more likely to be women (47.4% vs. 32%; P<0.001); in addition, they had almost twice the prevalence of diabetes with complications. In-hospital mortality rates (33.6% vs. 9.2%; adjusted odds ratio, 3.05; 95% CI, 2.76 to 3.38; P<0.001), length of stay (mean, 13.4 days vs. 4.7 days), and inpatient charges (mean, $245,000 vs. $129,000) were higher in the in-hospital STEMI group. These patients were also less likely to be discharged home (33.7% vs. 69.4%; adjusted odds ratio, 0.38; 95% CI, 0.34 to 0.42; P<0.001) and to receive cardiac catheterization (33.8% vs. 77.8%; adjusted odds ratio, 0.19; 95% CI, 0.21 to 0.26; P<0.001) or percutaneous coronary intervention (21.6% vs. 65%; adjusted odds ratio, 0.23; 95% CI, 0.21 to 0.26; P<0.001).

The authors noted that their study was retrospective and observational and did not include information on patients' diagnoses or status at admission or death after discharge. In addition, they said, data on cause-specific mortality were not available. However, they concluded that patients who had STEMI during a hospitalization for a non-ACS condition had higher rates of in-hospital mortality and lower rates of invasive testing and intervention than patients who developed STEMI in an outpatient setting. More research is needed on improving outcomes and determining optimal treatment for hospital-onset STEMI, similar to the quality initiatives aimed at the outpatient STEMI population, they wrote.