Evaluation of malnutrition via modified GLIM criteria for in patients undergoing emergent gastrointestinal surgery.

BACKGROUND:Malnutrition remains a critical public health issue in the US, particularly in surgery where perioperative malnutrition is commonly underdiagnosed and undertreated. In 2016, the Global Leadership Initiative on Malnutrition (GLIM) proposed a set of consensus criteria for the diagnosis of malnutrition. Our project aims to assess the post-operative outcomes of patients meeting a modified GLIM-defined (mGLIM) malnutrition criteria undergoing emergent gastrointestinal surgery (EGS) in the NSQIP database. Current GLIM-criteria were modified with addition of admission albumin (a NSQIP-defined malnutrition variable). METHODS:Adapting NSQIP data, mGLIM criteria are (1) BMI of ≤20 for age ≤ 70 and BMI ≤22 for age ≥ 71, (2) weight loss > 10% within the past 6 months, (3) admission albumin ≤ 3.5, and (4) emergent bowel surgery as etiologic criteria of acute disease/injury. All patients undergoing emergent small bowel, colon, and rectal procedures were extracted from the NSQIP database and included in the study. Multivariate linear and logistic regression models controlling for relevant covariates were developed to evaluate mGLIM criteria on length of stay (LOS), mortality, and overall complication rates. RESULTS:We included 31,029 patients who underwent emergent bowel surgeries from years 2011-2016. Demographically, 53.6% (n = 16,622) were female, 13.0% (n = 4023) were African American, and 78.3% (n = 24,292) were Caucasian. Case composition included 71.5% colon operations, 28.0% small bowel, and 0.5% rectal cases. Overall, 1.7% (n = 517) had data necessary to qualify as malnourished as per mGLIM. Controlling for covariates, multivariate linear and logistic regression analyses show that these patients have significantly higher mortality for both colon (p < 0.001, CI 1.55 | 2.61) and small bowel (p = 0.022, CI 1.08 | 2.67) procedures, longer LOS for colon (p < 0.001, CI 1.93 | 4.33) operations, and higher post-operative complications for both colon (p < 0.001, CI 1.61 | 2.62) and small bowel (p < 0.001, CI 1.57 | 3.37) cases. CONCLUSION:This analysis shows that mGLIM criteria malnutrition is associated with poor clinical outcomes following EGS affecting LOS and mortality. Our data indicates the new mGLIM criteria can be a powerful and simple predictive score for malnutrition that can be used to predict malnutrition-related risk of poor outcomes after EGS.