An Unmeasured Harm of Screening

The US Preventive Services Task Force¹ recommends a 1-time screening for abdominal aortic aneurysm (AAA) by ultrasonography in men aged 65 to 75 years who have ever smoked, on the basis of trials showing a reduction in death from AAA but not improvement in overall mortality.^1,2

Mr R was such a patient. A 65-year-old, obese former smoker with a history of myocardial infarction and persistent risk factors (hypertension and hyperlipidemia), Mr R had undergone a total knee arthroplasty at age 60 years because of osteoarthritis. Thereafter, he was able to walk a mile at a stretch but could go no further because of persistent joint pain. A veteran of Vietnam, Mr R underwent screening ultrasonography for AAA when the clinical reminder system of the US Department of Veterans Affairs prompted his primary care physician. He was found to have an aneurysm of 6.0 × 4.7 cm on the ultrasonogram. Mr R's primary care physician discussed the case with a vascular surgeon, who recommended a follow-up computed tomography (CT) scan with contrast and prompt follow-up in the vascular clinic.

His vascular surgeon considered an open procedure to repair the aneurysm but ultimately decided on an endovascular repair, given Mr R's habitus and history of myocardial infarction. Mr R underwent a detailed informed consent process during which he consented to the procedure, acknowledging the risks of untreated aneurysm as well as those of surgery, including ischemia, bleeding, and infection. He was told in qualitative terms by both his primary care physician and vascular surgeon about a dilation of his aorta and the risk of rupture.

Two weeks later, Mr R underwent placement of an endovascular aneurysm repair (EVAR) stent graft. The vascular closure device maldeployed on insertion but was fixed and correctly positioned, according to the operative note. Six months later, Mr R came back to his primary care physician reporting that his left leg was completely numb. The leg was found to be pulseless and cold. He had experienced a total occlusion of his EVAR graft. He underwent emergency thrombectomy and axillary-femoral bypass to his leg performed by a vascular surgeon. Despite this, Mr R's creatine kinase level began to increase, and his leg was noted to be taut. Three more operations followed, over the course of 2 months, to relieve compartment syndrome and later for wound debridement and dehiscence. Despite these efforts, Mr R was left with a persistent foot drop, requiring use of a walker. He underwent a month of rehabilitation for deconditioning, an experience he calls, “the low point of my life.” Today, Mr R, by his own account, is a shell of his former self. He is gaunt, like “Dick Cheney,” he jokes. He requires a walker. Because of the extent of his thrombosis, his vascular surgeon has recommended lifelong warfarin sodium therapy (Coumadin; Bristol-Myers Squibb). “Not the best idea for someone prone to falling,” Mr R tells me.

Endovascular repair as an alternative to open repair has been criticized because of poor durability and higher rates of long-term complications.^3,4 However, Mr R was not considered a candidate for open surgery. Thus, his only other option was risk-factor modification and observation. Leaving his 6.0-cm abdominal aneurysm untreated is an unnerving proposition. With an aneurysm this size, Mr R faced a 10%⁵ to 13%⁶ per year rate of rupture. But the question before any intervention is not how poor the prognosis is without it, but how much better it is with it. For patients with a large AAA (≥5.5 cm in diameter) but ineligible for an open AAA repair, a multicenter randomized clinical trial (RCT) of 404 patients with median follow-up of 3.1 years found no benefit from endovascular repair over observation on the outcome of death from all causes,⁶ although there was a significant reduction in aneurysm-related mortality. While this trial was published over a year before Mr R's procedure, there appears to be a lag time between the publishing of negative evidence and changes in clinical practice.⁷ Clinical inertia in abandoning a contradicted practice may contribute to harm for some patients, similar to other medical reversals.⁸

For Mr R, the most important thing is that his difficult experience should benefit others. “I hope you doctors are learning from my case,” he says. But, as long as screening trials examine disease-specific, and not all-cause mortality, explaining the risks and benefits of outcomes to patients like Mr R will remain difficult. His life expectancy may have been shortened by complications of treatment. His quality of life is diminished. At his request, I gave Mr R the numbers from the studies of AAA screening.² If 1000 people get screened, and 1000 people do not, an estimated 110 will die after 4 years in each group, but 2 die of AAA in the group that gets screened, and 3 die from AAA in the group that does not.

“Two and three? Wait—110 and 110 . . . ” His voice trails off. A moment later, Mr R shakes his head and laughs. “Had I known those numbers,” he says, “I wouldn't have done it. I thought it was life or death.”

This study highlights a number of complicated issues in health care. First, Mr R was appropriately consented for surgery; however, in hindsight he wishes different information were presented to him. The informed consent process is often long and complicated, but it remains unclear if it best communicates what some patients want to know. Second, translating data from RCTs to individual patients, who often differ from those participants in the RCTs, remains an uncertain affair. Third, data are often extrapolated from older studies. The data from the cited screening example² were collected between 1997 and 1999, prior to the development of endovascular repair. Finally, how quickly should physicians respond to new data, and are data from only 1 or 2 studies sufficient to change practice?

Correspondence: Dr Prasad, Medical Oncology Branch, National Cancer Institute, National Institutes of Health, 10 Center Dr, 10/12N226, Bethesda, MD 20892 (vinayak.prasad@nih.gov).

Published Online: September 17, 2012. doi:10.1001/2013.jamainternmed.682

Acknowledgment: For Mr R, who encouraged me to share his story.

Additional Contributions: I thank the patient for sharing his story and for providing permission to publish it.