Currently in the United States, about twice as many adults with appendicitis are treated without an operation as 20 years ago, but nonoperative management of an infected appendix is linked to a higher death rate in the hospital, according to the first large U.S. study to look at long-term trends in appendicitis.

Studies in Europe over the past two decades have shown that some patients with uncomplicated appendicitis can be treated successfully with antibiotics alone. These results appear to be driving the change in surgical practice for appendicitis in the United States.

But this new study raises questions about whether a nonoperative approach is a safe option for older or medically complex patients.

After matching the cases in both treatment groups on age, sex and comorbidity index, the investigators reported a nearly sixfold increased death rate in the nonoperative care group. Compared with 0.08% of surgically treated patients with appendicitis, 0.47% of the nonoperatively managed patients died in the hospital.

“I am concerned that surgeons may be overextrapolating the European data and applying nonoperative management of appendicitis to patients for whom it is inappropriate,” said the study’s principal investigator Isaiah R. Turnbull, MD, PhD, assistant professor of acute and critical care surgery, Washington University School of Medicine, in St. Louis. “We as a community of surgeons need to consider whether nonoperative management of uncomplicated appendicitis, such as an antibiotics-first approach, is appropriate for these high-risk patients.”

Dr. Turnbull presented the findings at the 2017 Clinical Congress of the American College of Surgeons. The study was published last month in the Journal of Surgical Research (pii: S0022-4804[17]30661-3).

The researchers analyzed the type of treatment for 477,680 adults with a primary diagnosis of appendicitis between 1998 and 2014 included in the National Inpatient Sample, a database of hospitalized patients representing more than 96% of the U.S. population.

Although the database did not specify whether the appendix had ruptured, investigators excluded patients who had a peritoneal abscess, as well as patients younger than 18 years of age and those who had elective admission.

Over the study period, the proportion of patients treated with an early operation decreased from 94.6% to 92.1%. Use of nonoperative management increased twofold, from 2.3% of patients with appendicitis in 1998 to 4.9% in 2014. Most of the remaining patients had an operation later than the second day in the hospital.

Additionally, in 135,856 patients with appendicitis from 2010 to 2014, the investigators compared patient characteristics and outcomes. A total of 131,162 patients underwent an early operation, primarily an appendectomy on the day of hospital admission or the next day. Another 4,694 patients received nonoperative management, defined as no operation or placement of an abdominal drain. It is not clear from this data set whether these patients received antibiotic therapy.

In this subgroup of patients from the most recent five-year period, those receiving nonoperative management were, on average, eight years older than patients in the early operation group—49 versus 41 years. They also had a significantly higher comorbidity index, with a score of 1.35 compared with 0.78.

The odds of dying of any cause during the hospitalization was 2.4 times higher in patients who had no operation compared with those who underwent an early operation.

“Our findings suggest that U.S. surgeons are selecting elderly, sicker patients for nonoperative management, possibly because they believe these patients are not good candidates for an operation,” Dr. Turnbull said. “However, these patients are at increased risk of a poor outcome if nonoperative management fails because they lack the physiologic reserve or ability to tolerate illness.” Many of the European studies of antibiotic treatment for appendicitis excluded elderly patients and had a patient age that was much younger than in this U.S. database, he said.

This is the first known study to show an association between nonoperative management of appendicitis and death. Mortality is rare in appendicitis, so tracking it requires a large data set.

Dr. Turnbull said he was cautious about making clinical recommendations based on this study alone, saying the primary conclusion from these data is that more information is needed. “That said, for me, I would counsel my colleagues that these data suggest that nonoperative management may not be appropriate for elderly patients or patients with significant medical comorbidities.”

The ongoing CODA trial is expected to provide some answers on how to best treat elderly patients as it is open to adults of all ages. It explicitly excludes several populations of medically complicated patients, such as those who have liver failure, immunologic deficiency, or second infectious conditions such as pneumonia.

Lead investigators in the CODA trial said the study presented at the ACS is interesting, but caution should be applied when interpreting the results. Giana H. Davidson, MD, MPH, assistant professor at the University of Washington, said early results from CODA show that surgeons have significant bias in giving treatment recommendations and patients have preferences that drive treatment decisions, which will not be accounted for using administrative data. “These preferences may also be associated with worse outcomes.”

David R. Flum, MD, MPH, professor of surgery and director of the Surgical Outcomes Research Center at the University of Washington, in Seattle, said the CODA trial was developed to answer some of the questions raised by the new study. “There are real questions about actual rates of adverse outcomes with antibiotic management of appendicitis.” Dr. Flum is a member of the editorial board of General Surgery News.

“While it’s likely that claims-based studies like this may show higher rates of adverse events because of confounding, the finding of increased risk is worrisome. For antibiotics to be offered as an alternative to appendectomy in the United States, we need a rigorous [randomized controlled trial] to establish the comparative rate of adverse events. For this reason, we think antibiotics should remain in the research space until CODA is completed and the results are released in 2020.”

Both Dr. Davidson and Dr. Flum commented based on the study abstract.

Appendicitis is the most common intraabdominal surgical emergency in the United States, with more than 250,000 cases annually.