SAN ANTONIO – Perioperative risk factors affect postoperative morbidity and mortality differently in emergency and nonemergency surgery, according to an analysis of the ACS National Surgical Quality Improvement Program (ACS NSQIP) database by investigators from Massachusetts General Hospital, Boston.
“Instead of using the same risk-adjustment model for both ... as is currently being done, our findings strongly suggest the need to benchmark emergent and elective surgeries separately,” they concluded.
“Most risk-adjustment models simply have an on/off switch for whether or not the patient underwent emergency surgery and treat comorbidities and other perioperative variables the same.” Those variables, however, “don’t behave the same way in emergency surgery,” said investigator Dr. Jordan Bohnen, a surgical research resident at Mass General.
Because risk adjustment doesn’t take into account variables that have a particularly strong negative impact in emergency settings, acute care surgeons are getting “unnecessarily dinged for having higher complication rates,” he said at the Eastern Association for the Surgery of Trauma scientific assembly.
For example, the team found that preop transfusions and white blood counts (WBC) at or below 4.5 carry a significantly higher risk of 30-day major morbidity or mortality (MMM) in emergency versus nonemergency surgery. Conversely, ascites, preop anemia, and leukocytosis carry a greater MMM risk in nonemergent cases.
The findings come from a comparison of 110,182 nonemergent surgeries to 59,949 emergency cases – generally meaning surgery within 12 hours of emergency department (ED) admission – from the NSQIP database for 2011-2012.
As expected, the overall risk of MMM was significantly higher for emergency cases (16.75% vs. 9.73%; P less than .001), and four procedures – laparoscopic cholecystectomy, exploratory laparotomy, and umbilical and incisional hernia repairs – were relatively riskier when done emergently.
“As surgical quality improvement efforts mature, it’s increasingly important to apply accurate risk-adjustment models to benchmark quality improvement for surgeons, hospitals, and health care systems.” The current “assumption that perioperative variables have an equal impact on outcomes in emergent and nonemergent settings” is incorrect. “Risk factors for bad outcomes change depending on the setting,” Dr. Bohnen said.
Dr. Bohnen has no disclosures.