SAN FRANCISCO – Postoperative complications are prognostic for poor outcomes in patients undergoing resection of gastric cancer, researchers reported at the annual Gastrointestinal Cancers Symposium sponsored by the American Society of Clinical Oncology.
The team retrospectively analyzed data collected in the U.S. Gastric Cancer Collaborative, a joint effort of seven moderate- to high-volume academic institutions across the United States. Analyses were based on 850 patients who underwent curative resection of gastric or gastroesophageal junction adenocarcinoma by abdominal approach from 2000 to 2012.
Complications occurred within a month of surgery in 40% of the patients, reported coauthor Dr. Ryan C. Fields of Washington University in St. Louis. These patients were 20% less likely to receive adjuvant therapy than were their peers with an uncomplicated postoperative course. And during a median follow-up of almost 3 years, those with postop complications were 60% more likely to die after potential confounding factors were taken into account.
"The particular strength of this data set is the cross-section of real-world treatments and outcomes that it represents," Dr. Fields noted.
Patients who had complications were older, had higher levels of comorbidity and complexity, underwent more extensive surgery, and had more advanced disease. "We could potentially identify preoperatively [these patients] in an attempt to reduce the chance of complications," he said.
Not included in this analysis was "the finding of muscular sarcopenia as measured from the psoas muscle on a preoperative CT scan. There are ongoing studies using this measure to identify patients who may benefit from a short course of intensive physical therapy and nutritional supplementation prior to surgery in an attempt to reduce complications," he added.
The study was limited by its retrospective and nonrandomized design, and factors such as selection bias and missing data may have affected outcomes, Dr. Fields acknowledged. The researchers are therefore conducting further analyses using propensity matching and imputation.
The data set "may not be representative of (patients in) other countries or importantly of the nearly 40% of patients with gastric cancer that are treated at non-university, community-based hospitals in the United States," he added. "I would argue, however, that if we accept the relationship between volume and outcomes that has been shown to be significant in gastrectomy, the effect of complications on outcomes would be expected to be even more significant in these settings."
Groups stratified by Clavien-Dindo score were too small to draw definitive conclusions, but there were trends whereby worse complications had a greater negative impact on outcome, Dr. Fields said.
"One of the problems that we have – as an oncologic community –is there is no real consistent definition of complications in a lot of this literature," he added, which may be generating some of the disparate findings. For example, the severity of a wound infection can range widely, with more severe infections affecting decisions about whether to delay or even withhold adjuvant therapy.
In the study, there were a total of 699 complications in 342 patients, "highlighting the ... observation that often complications can beget further complications," according to Dr. Fields.
More than one-third of the complications (37%) were of Clavien-Dindo grade III or IV, requiring care more invasive than simple treatment at the bedside or with intravenous medications.
The most common were anemia/bleeding (12%), wound infections (8%), pneumonia (6%), and respiratory failure requiring reintubation (6%).
Relative to counterparts without complications, patients who had complications were on average significantly older and had significantly higher American Society of Anesthesiologists scores (reflecting greater preoperative comorbidity and complexity), more extensive and complex surgery, and more advanced cancer.
"Importantly, we did not find neoadjuvant therapy to be associated with an increase in postoperative complications, and this is consistent with literature in other solid tumor types, such as pancreatic cancer and rectal cancer," noted Dr. Fields, who disclosed no conflicts of interest related to the study. Also, the extent of lymphadenectomy was not associated with this outcome.
A minority of patients had a laparoscopic surgery, and the number was too small to say whether this factor influenced complication rate, he said.
"What we can surmise, one, from the literature, and two, is that the specific complications that are oftentimes reduced with laparoscopy and minimally invasive surgery certainly relate to wound infection and wound problems for sure," Dr. Fields commented. "So I think if we can do anything to achieve the same oncologic goals and reduce the morbidity of an operation, and improve the ability of patients to get on adjuvant therapy, that’s going to be critical. And laparoscopy may be a way to do that in gastric cancer. It may be more challenging in some of the bulkier, more advanced tumors to apply it globally, but I think it certainly holds promise for that exact point."