番茄社区

Urine tests in the emergency department to rule out urinary tract infection (UTI) are often unnecessary, keep patients there longer than needed, and prompt too many antibiotic prescriptions, new research suggests.

Among 16,191 patients presenting with chest pain at two San Diego hospitals, 19% had urine tests even though fewer than 0.01% would have been expected to UTIs, reported Richard Childers, MD, of the University of California San Diego (UCSD), at the .

And for 3,146 of those patients who completed a urine test, 10.5% received prescriptions for antibiotics used to treat UTIs compared to 1.4% of patients not getting urine tests. More than 75% of those with abdominal pain took urine tests, and 17% of those who did received UTI antibiotic prescriptions (all P<0.001).

The retrospective findings aren't definitive, Childers told 番茄社区. But they raise more questions about the widespread use of routine urine tests when patients aren't showing symptoms of bacteriuria.

"If they don't have any urinary tract symptoms, you shouldn't test them," said Childers.

Physicians have been sounding the alarm about the overuse of urine tests since at least 2013, when a team of Michigan physicians that "efforts targeting use of urine cultures in hospitalized patients will play an important role in improving antimicrobial stewardship."

Childers and colleagues retrospectively tracked adult discharged patients from 2015-2019 at two UCSD hospitals. They looked at patients with chief complaint groups not typically linked to UTIs -- chest pain, vaginal bleeding in pregnancy, and abdominal pain.

Women aged 65 and older with weakness, altered mental state, and/or confusion underwent urine tests 78% of the time, and 27% of those were treated with UTI antibiotics.

Overall, the findings don't prove that many patients had asymptomatic bacteriuria that didn't need treatment, but Childers suspects this is the case. He noted that many older women have positive urine findings regardless of symptoms: "We treat it and we feel better, but we're not really treating the patient. Then they get exposed to all these antibiotics."

In a , the Infectious Diseases Society of America recommends against routine screening and treatment of this condition in most populations, including functionally impaired older people.

Separately, the UCSD researchers retrospectively tracked all discharged ED patients – 228,494 visits -- from 2015-2019. The average length of stay was 78 minutes longer in those who took urine tests. Among all those who had lab tests (111,082 visits), those who took urine tests had length of stays that were 21 minutes longer (P<0.001).

Lindsay Petty, MD, an infectious disease specialist at the University of Michigan who led a in the hospital, told 番茄社区 that the new study findings are "important" because they provide details about practice in the ED.

"However, the studies do not drive down to specifics of patient symptoms, urine test results and concomitant infections, so we cannot really conclude that 'inappropriate' urine testing led to more antibiotics or increased ED length of stay," she cautioned.

Petty noted that her study found that 83% of asymptomatic bacteriuria was treated in the hospital setting. "The strongest factor associated with treatment was an abnormal urinalysis, which likely represents a misinterpretation of the UTI diagnostic criteria," she said. "Clinicians should not presume that pyuria or nitrites in the absence of urinary symptoms is diagnostic of a UTI, and this is an important area to focus on in education and stewardship interventions."

She added: "If a urine test is being considered only for the indication for a possible infection, clinicians in the ED should ask the patient if they have specific urinary signs or symptoms: e.g., dysuria, urgency, frequency, suprapubic pain, flank/costovertebral pain, fever without another cause. If the patient does not have these signs or symptoms, do not send a urine test, including the urinalysis."

Comment