NEW YORK (Reuters Health) – The Albumin-Bilirubin (ALBI) score more accurately predicted outcomes in patients undergoing pulmonary, elective colorectal and adrenal operations than the Model for End Stage Liver Disease-sodium (MELD-Na) in a medical records analysis.
“ALBI is an excellent predictor of 30-day mortality in major non-liver surgery and is a good predictor of morbidity as well,” Dr. Kelvin Kwan Lau of the Lewis Katz School of Medicine Temple University Hospital in Philadelphia told Reuters Health by email. “Moreover, ALBI is either equivalent to or better than MELD-Na at predicting mortality and morbidity in non-liver surgery, especially lung resections, elective colectomy, and adrenalectomy.”
“MELD-Na was not statistically better than ALBI in any of the procedures we examined,” he added, “although it was close to being a more accurate predictor of mortality in emergent open abdominal aortic aneurysm (AAA) repair.”
As reported in the Journal of the American College of Surgeons, the team analyzed more than 258,000 patients in the 2015-2018 ACS-NSQIP database (median age, 60; 50% women; 70%, white) who underwent cardiac, pulmonary, esophageal, gastric, gallbladder, pancreatic, splenic, appendix, colorectal, adrenal, renal, hernia, and aortic surgery.
Fifty-one percent were ALBI grade 1; 42%, grade 2; and 6.8%, grade 3. The median MELD-Na was 7.50, with 78% of patients having a score of less than 10.
Overall, the 30-day mortality rate was 2.7% and the morbidity rate was 28.6%.
Increasing ALBI grade was significantly associated with mortality and morbidity for ALBI grades 2 (OR, 5.24 and OR, 2.15, respectively) and 3 (OR, 25.6 and OR, 6.12).
ALBI outperformed MELD-Na, with increased accuracy in pulmonary, elective colorectal, and adrenal operations on receiver operating characteristic analysis.
Dr. Lau said, “We only investigated broad surgical outcomes such as postoperative pneumonia, urinary tract infections, surgical site infections, etc. The next investigations we will pursue include examining ALBI’s predictive ability in procedure-specific outcomes – e.g., anastomotic leak in colectomy.”
Nonetheless, ALBI “is not perfect,” he noted. “Surgical societies such as the American College of Surgeons or the Society of Thoracic Surgeons have risk calculators which take many patient factors into account, not just their preoperative serum albumin and bilirubin. Perhaps the inclusion of ALBI into these models may enhance their predictive abilities. Further investigation is warranted.”
Dr. Myron Schwartz, The Henry Kaufmann Professor of Surgery at Icahn School of Medicine at Mount Sinai in New York City, noted in an email to Reuters Health, “The study population in this paper is far more similar to the cohort used to develop the ALBI score than that from which the MELD-Na score was derived.”
“Patients with hepatorenal syndrome and hyponatremia, who drove the inclusion of creatinine and sodium into the MELD-Na model, are not candidates for the various surgeries undergone by the patients included in the current study,” he said. “It was thus predictable that the ALBI score would perform better in this setting.”
“Each of these scoring systems has its shortcomings,” he continued. “The variables included, while reflective of liver dysfunction, may also be abnormal for other reasons. Additionally, portal hypertension is a well-validated factor associated with outcome in patients with cirrhosis that is not included in either model.”
“Ultimately,” Dr. Schwartz concluded, “decisions in individual patients must consider the individual circumstances; while scoring systems can provide a frame of reference, they should not be considered the ultimate basis for decision-making in the clinic.”
SOURCE: https://bit.ly/3n05JEB Journal of the American College of Surgeons, online December 17, 2020.