A study of PET scan use for follow-up care for U.S. patients with lung and esophageal cancer shows a wide variation among hospitals, but the degree of usage appears to have no impact on survival.
The study, recently published in the Journal of the National Cancer Institute, used Surveillance, Epidemiology and End Results data and Medicare hospital billing data for 97,152 patients with primary lung cancer, and 4,446 patients with primary esophageal cancer patients who were treated in the mid-2000s and received follow-up care through 2011 (2016;108(7):djv429).
There was substantial variation in the use of recurrence detection scans-hospitals at the top of the usage list used follow-up PET scans for patients with lung cancer 14 times more frequently than hospitals that used them the least.
For esophageal cancer patients, there was an eight-fold difference between high users and low users.
"And yet there was no difference in survival for patients who had PET scans at these hospitals," said first author Mark Healy, MD, Surgical Resident and Research Fellow of the Center for Health Outcomes & Policy at the University of Michigan, Ann Arbor. Senior author is Sandra L. Wong, MD, Chair of Surgery at the Geisel School of Medicine at Dartmouth-Hitchcock.
Healy said the study was prompted by concern about the use of PET scans to detect recurrence in patients who are symptomatic, doing well, and have completed their initial anticancer therapy. And the Center for Medicare Services recently changed its policy for Medicare reimbursement to restrict PET use to three scans following a patient's initial anticancer therapy.
PET utilization was calculated in person-years, excluding scans for staging and for follow-up of CT findings.
Hospitals were stratified by quintiles of PET utilization for adjusted two-year survival analysis. Those at the lowest end of the spectrum performed .05 scans per person-year for lung cancer versus 0.70 scans per person-year for those at the highest end.
For esophageal cancer, usage ranged from 0.12 to 0.97 scans per person-year.
But two-year survival rates were virtually the same for all hospitals regardless of PET scan volume.
The lowest and highest utilizing hospitals had adjusted two-year survival rates of 29.0 percent versus 28.8 percent for lung cancer, and 28.4 percent and 30.3 percent respectively for esophageal cancer.
Healy and his colleagues reported the same overuse effect in pancreatic cancer found in prior studies.
The study did not examine variations in practice patterns by region, Healy said.
"Following evidence-based guidelines for clinical follow-up is the way to go. Don't order PET in asymptomatic patients," Healy said.
The study was funded by the National Institutes of Health, Agency for Healthcare Research and Quality, and American Cancer Society.