Inpatient mortality and length of stay comparison of percutaneous endoscopic gastrostomy and percutaneous endoscopic gastrojejunostomy.

INTRODUCTION

Percutaneous endoscopic gastrostomy (PEG) and percutaneous endoscopic gastrojejunostomy (PEGJ) are endoscopic procedures often performed by surgeons. No recent population-based study has compared inpatient mortality or length of stay between patients who undergo PEG or PEGJ placement during their hospitalization.

METHODS

Patients undergoing inpatient PEG or PEGJ placement and who were at least 18 years old were identified from the 2006 Nationwide Inpatient Sample (NIS) database. Baseline characteristics of each group were compared, and outcomes of risk-adjusted inpatient mortality and length of stay were determined. Means were compared from using a complex sample t-test, and proportions were compared from using a complex sample chi-square test, with an alpha level of 0.05 for significance. Bivariate logistic regression was used to evaluate PEG or PEGJ placement as a risk factor for mortality.

RESULTS

In the 2006 NIS, 187,597 discharges were identified, during which a PEG or PEGJ was placed. Ninety-six percent (179,587) of patients underwent PEG placement, and 4% (8010) had PEGJ tubes placed. Fifty-one percent were men, with the mean age for PEG and PEGJ placement of 71.3 +/- 0.3 (mean +/- standard error) and 64.8 +/- 0.8 years (P < 0.05). In the PEG group, 86% of admissions were nonelective, compared to 79% in the PEGJ group (P < 0.05). The primary discharge diagnoses for both groups of patients included acute cerebrovascular disease, aspiration pneumonitis, septicemia, respiratory failure, and intracranial injury. PEG patients had a higher cumulative incidence of congestive heart failure, chronic lung disease, and diabetes. Crude in-hospital mortality for death was 11% for both PEG and PEGJ patients. No difference in mortality was observed in risk-adjusted analyses accounting for patient severity. Mean length of stay was similar for both groups (PEG 20.9 +/- 0.4 days; PEGJ 22.5 +/- 1.1 days). Neither PEG nor PEGJ was identified as a risk factor for inpatient mortality.

CONCLUSIONS

Comparative analyses of patients undergoing PEG versus PEGJ revealed no detectable difference between inpatient mortality and hospital length of stay in this large observational study. Both procedures can be performed safely in high-risk populations, with no increased mortality or length of stay incurred by jejunal feeding access. However, further analysis is required to compare more specific short-term outcomes between these populations as well as their respective cost-effectiveness.