AT SSO 2016
BOSTON (FRONTLINE MEDICAL NEWS) – The operation was a success, but the patient died.
It’s an old chestnut for sure, but there is a painful kernel of truth in it, say investigators who found that patients who undergo complex cancer surgery and have serious complications are at significantly increased risk for death for at least 6 months after surgery, compared with patients who undergo the same procedure with few or no complications.
“Our work has important implications for quality assessment. I think in cancer surgery in particular we have to get away from the short-term metrics of survival, and we have to think about the implications of complications for long-term survival, even if at a very high-quality hospital we’re good at salvaging those patients who do experience those complications,” said Dr. Hari Nathan of the University of Michigan, Ann Arbor.
In a retrospective study, results of which were presented at the annual Society of Surgical Oncology Cancer Symposium, Dr. Nathan and colleagues showed that patients who underwent surgery for cancers of the esophagus and lung who had serious complications but survived at least 30 days after surgery had a more than twofold greater risk for death than did patients who had no complications, and patients with serious complications following surgery for cancer of the pancreas had a nearly twofold greater risk.
The effects of serious complications on survival persisted out to at least 180 days after surgery for each of the three procedures.
The findings suggest that just getting the patient through the operation and keeping him or her alive in the ICU is not sufficient cause for celebration by surgeons, Dr. Nathan said.
The investigators conducted the study to examine the incidence of complications following cancer surgery in older patients, the relationship between surgical complications and long-term survival, and whether the effects of complications would diminish or “wash out” over time. They reviewed Surveillance, Epidemiology and End Results–Medicare data on patients aged 65 years and older who underwent surgery with curative intent for esophageal cancer, non–small cell lung cancer, or pancreatic adenocarcinoma from 2005 through 2009.
They defined serious complications as “the appearance of a complication associated with a hospital length of stay greater than the 75th percentile for that procedure.”
The cohort included 965 patients who underwent esophageal surgery, 12,395 who had lung surgery, and 1,966 who underwent pancreatic resection. The proportion of patients over 80 years who underwent the procedures, respectively, were 12%, 18%, and 19%.
Serious complications occurred in 17% of patients with esophageal cancer, 10% of those with lung cancer, and 12% of those with cancer of the pancreas. The respective 30-day mortality rates were 6.%, 3.3%, and 3.9%.
Looking only at those patients with lung cancer who survived at least 30 days after surgery, the investigators found that median survival among those who had no complications was 79 months, compared with 60 months for those who had mild complications, and 33 months for patients who had serious complications (P less than .001)
“And indeed, when we performed adjusted survival analyses looking at all three disease sites, we saw a very consistent story: that those patients who had serious complications had decreased long-term survival for all three malignancies we looked at,” Dr. Nathan said.
Specifically, in survival analyses adjusted for sex, age, and procedure code, hazard ratios for patients with serious complications compared with those who had no complications were 2.55 for esophageal cancer patients, 2.13 for lung cancer patients, and 1.57 for pancreatic cancer patients (all comparisons significant as shown by 95% confidence intervals).
The investigators questioned whether the differences in mortality were due to the late effects of perioperative complications.
“In modern ICUs, we can keep virtually anybody alive for 30 days, and there has been a lot interest in longer-term metrics for perioperative mortality, for example, at 30 or 90 days, so we thought maybe that’s what we were seeing here,” he said. To test this idea, the investigators looked at the effects of complications on patient who survived lung cancer surgery for at least 90 days, and those who lived for at least 180 days after surgery, and they saw that the survival curves were similar to those seen with the 30-day survivors, showing significantly and persistently worse survival for patients with serious complications (P less than .001).
For each of the disease states, patients with serious complications were also significantly less likely than were those with no or mild complications to receive adjuvant chemotherapy, even after adjustment for patient age and cancer stage, two significant determinants of the likelihood of receiving chemotherapy.
And even when the effect of chemotherapy for those who did receive it was added into the survival models, patients with serious complications still had significantly worse overall survival, Dr. Nathan noted.
“Serious complications after these three cancer resections are common and they are associated with dramatically inferior long-term survival. Thirty, 60, 90, and even 180-day measures of mortality do not capture the full impact of complications on long-term survival,” he said.
Asked whether it may be possible to identify those patients at higher risk for serious complications due to comorbidities or other factors, and perhaps suggest withholding surgery from such patients, Dr. Nathan agreed, but added that “the best chance for survival for all of these patients is a high-quality surgical resection, so it’s hard to deny a patient that chance unless you think they have a really high risk of perioperative death.”
The study was internally funded. Dr. Nathan reported no significant disclosures.
