Article Abstract

Predictors of blood transfusion and in-hospital outcomes in patients with gastric antral vascular ectasia (GAVE): a nationwide population-based analysis

Authors: Upenkumar Patel, Rupak Desai, Jiten Desai, Nanush Damarlapally, Dipen Zalavadia, Mohamad Yousef, Roxana Coman, Pardeep Bansal, Hemant Goyal

Abstract

Background: Gastric antral vascular ectasia (GAVE) is a rare cause of chronic non-variceal upper gastrointestinal (GI) bleeding and can turn into life-threatening bleed in some patients. Packed red blood cell (PRBC) transfusions are often required in these patients during hospitalization. We aimed to investigate the hospitalization outcomes and predictors of PRBC transfusions in patients with GAVE lesions.
Methods: Using the ICD-9-CM codes (537.82, 537.83), we queried the National Inpatient Sample (NIS) [2010–2014] to recognize hospitalized GAVE patients. A 1:2 random sample was obtained from the non- GAVE cohort and these groups were compared (GAVE vs. non-GAVE). The predictors of PRBC transfusion in GAVE cohort were analyzed with multivariate analysis by using SPSS Statistics 22.0.
Results: We included weighted 89,081 GAVE and 178,550 non-GAVE hospitalized patients. The GAVE patients were tended to be older, female and white. Significantly higher proportions of comorbidities such as congestive heart failure, diabetes, hypertension, hypothyroidism, liver disease, renal failure, Sjogren syndrome, systemic sclerosis and portal hypertension, etc. were present in these patients. The all-cause inpatient mortality was found to be 1.4%. The mean hospital charges and length of stay (LOS) per GAVE hospitalization were $36,059 and 4.63±5.3 days, respectively. A total of 6,276 (weighted 31,102) (34.9%) of these patients received at least one PRBC transfusion during hospitalization. Advanced age, multiple comorbidities, non-elective admissions, male gender, and African American race were the independent factors associated with higher chances of receiving PRBC transfusion.
Conclusions: Our analysis showed that hospitalized patients with GAVE lesions had lower overall mortality rate despite having multiple comorbidities. There was no difference in the LOS and hospital charges between the two cohorts. Nearly 35% of the GAVE patients received at least one PRBC transfusion.