PHOENIX – A new model based on clinical, hospital, and operative factors predicted the risk of ureteral injury among patients undergoing colorectal surgery.
Using data from the Nationwide Inpatient Sample, which includes more than 2 million patients in the United States, researchers retrospectively studied patients undergoing surgery in the United States between 2001 and 2010 for colorectal cancer, polyps, diverticular disease, or inflammatory bowel disease.
Less than 1% of patients sustained a ureteral injury, but the incidence rose significantly during the study period and injured patients had sharply higher rates of complications, researchers reported at the annual meeting of the American Society of Colon and Rectal Surgeons.
Eight factors were independently associated with the risk of ureteral injury. A predictive model incorporating these factors had an area under the receiver operating characteristic curve of 0.73. The model-predicted probability of injury ranged from 0.1% to 1.65%, depending on hospital factors, disease type, and procedure type, said Dr. Wissam J. Halabi, a research fellow at the University of California-Irvine, Orange.
Patients had higher adjusted odds of ureteral injury if they had rectal cancer (odds ratio, 1.85), adhesions (OR, 1.83), and metastatic cancer (OR, 1.76); if they had lost weight (OR, 1.08); and if they underwent surgery at a teaching hospital (OR, 1.05). On the other hand, patients had lower odds of injury if they had a laparoscopic procedure (OR, 0.91), a transverse colectomy (OR, 0.90), or a right hemicolectomy (OR, 0.43).
With the new predictive model incorporating these factors, the probability of injury ranged from 0.1% for patients undergoing laparoscopic right hemicolectomy to 1.65% for patients having all five adverse risk factors.
"Diverticulitis did not appear as a predictor in our model," Dr. Halabi said. "As for radiation, one of our predictors was metastatic cancer. So this goes for any cancer at advanced stage that has spread to lymph nodes or distant organ metastasis. In the case of rectal cancer, those are most likely to have received radiation therapy. So part of this effect of radiation therapy was apparent in the metastatic cancer group predictor."
Analyses were based on 2,165,848 colorectal surgery procedures. The overall rate of ureteral injury was 0.28%, he reported.
There was a significant 24% increase in the rate during the study period, from 2.5 per 1,000 cases in 2001-2005 to 3.1 per 1,000 cases in 2006-2010.
On average, the patients sustaining injury were younger and were more likely to be female and to have metastatic cancer, to be immunosuppressed, and to have had weight loss. They were less likely to have certain major comorbidities, such as diabetes and hypertension.
"Interestingly, obesity was similar in the two groups," Dr. Halabi commented.
Patients who sustained ureteral injuries had longer hospital stays and higher rates of a variety of postoperative complications, such as anastomotic leak and acute renal failure, but in-hospital mortality was statistically indistinguishable.
In adjusted analysis, patients with ureteral injury were significantly more likely to die (OR, 1.45) and to experience complications (OR, 1.66), and they had a longer hospital stay (+3.65 days) and total hospital charges (+$31,497).
Ureteral injuries "affect a relatively younger and healthier population. However, they have a significant and dramatic impact on outcomes," he added. "This predictive model can be used for risk stratification and counseling."
Dr. Halabi disclosed no relevant conflicts of interest.