Expert-Center Referral After Pancreatectomy Does Not Improve Outcomes

By Will Boggs MD

September 28, 2020

NEW YORK (Reuters Health) - Transfer of patients to expert centers after pancreatectomy does not appear to improve outcomes, according to a database study.

Interhospital transfer (IHT) has been shown to benefit trauma and cardiology patients, but no data are available on the outcomes and characteristics of patients transferred to tertiary centers after such complex procedures as pancreatectomy.

Dr. Mehdi El Amrani and colleagues of the University of Lille and Lille University Hospital, in France, used data from a comprehensive nationwide French database (PMSI) to analyze the outcomes of IHT patients after pancreatectomy.

Of the nearly 20,000 patients who underwent pancreatectomy from 2012 through 2018, 5.8% were transferred. More than half of those (58.2%) had been operated on in hospitals performing fewer than 26 pancreatectomies annually.

The rate of major complications for transfer patients (75.9%) was significantly higher than that for nontransfer patients (52.6%), and more than half of transfer patients had required a reoperation procedure in the transferring facility.

Postoperative in-hospital mortality was 5.2% overall, but was significantly higher in transfer patients (13.3%) than in nontransfer patients (4.7%), the researchers report in Annals of Surgery.

The increase in post-operative mortality was particularly high after IHT of patients who underwent pancreatectomy in low-volume centers (18.9% vs. 5.5% for those not transferred).

In contrast, mortality did not differ significantly by transfer status among patients who underwent pancreatectomy at high-volume hospitals.

The rate of failure to rescue (FTR, defined as mortality among patients with one or more major complications) was significantly higher among transfer patients (17.4%) than among nontransfer patients (8%). As was the case for mortality, the FTR difference was confined to patients operated at low-volume hospitals.

In multivariable analysis, independent risk factors of IHT after pancreatectomy included older age, comorbidities, and major complications after pancreatectomy in low-volume hospitals.

Among patients who underwent pancreatectomy for cancer, age, comorbidity, IHT, and pancreatectomy at low-volume centers were independent predictive factors of FTR.

"Local expertise, resources, and volume of hospitals are mandatory to provide appropriate care after pancreatectomy," the authors conclude.

They add that centralization of pancreatic surgery to high-volume hospitals might improve outcomes, but the complexity and expense of reorganizing the health system make centralization in France unrealistic at this time.

Dr. Marc Besselink of the University of Amsterdam, who studies pancreatectomy, told Reuters Health by email, "Our group in the Netherlands and many other groups have shown that a minimum of 40 pancreatoduodenectomies per center per year is required to obtain acceptable outcomes. With the ongoing differentiation within pancreatic surgery to robotic surgery, to surgery for locally advanced pancreatic cancer, to neoadjuvant treatment, the need for centralization only becomes larger, because for each of these subgroups a sufficient annual volume of patients is required."

He advised, "Only refer patients to high-volume clinics for pancreatic surgery (minimum of 40 pancreatoduodenectomies per year) with the facilities for 24/7 emergency care."

Dr. Besselink, who was not involved in the French research, added, "These surgeons who did transfer their patients are to be commended, because clearly they did this with the patients' best interests in mind. As a surgeon, you have very close contact with your patient, and it is not easy to hand over your care to a colleague in another hospital."

Dr. Kyle H. Sheetz of the University of Michigan, in Ann Arbor, who has researched the association between cancer surgery volume and patient outcomes, told Reuters Health by email, "Beyond the volume of the hospital where transferred patients had their surgery, knowing that those hospitals were also less likely to have advanced care capabilities (e.g., presence of an ICU or interventional radiology) is important."

"The parallels to centralization may not be as clear as the authors suggest," he said. "After all, we don't know that the outcome would have been different had they had their operation at another facility. Transferred patients were sicker at baseline. So one option would be to do a better job matching patients to their hospitals. For example, patients who are more likely to have complications (because they have more medical problems at baseline) should have their surgery at centers with the most experience and most resources available to handle problems when they arise (e.g., needing an ICU)."

Dr. Sheetz added," Discussions around centralizing any healthcare service should be framed around the desired outcome and the tradeoffs we are willing to accept to achieve that outcome. For example, we could centralize all high-risk surgery at one hospital. But how much better would patient safety get? And would we be OK with what that may mean for other things we also assign value to, like patient access or physician training?"

Dr. El Amrani did not respond to a request for comments.

SOURCE: https://bit.ly/3hFomej Annals of Surgery, online August 21, 2020.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....