BACKGROUND & AIMS: Malnutrition is common in patients eligible for percutaneous endoscopic gastrostomy (PEG). Feeding tube transparietal thickness (TT) may contribute to assess nutritional status. This study aims to: 1) Characterize TT in PEG patients. 2) Determine the association between TT and the currently used tools 3) Define TT best cut-offs to predict undernutrition 4) Assess the correlation between TT and survival.

METHODS

Prospective cohort study including patients who underwent PEG. Nutritional assessment was performed using NRS 2002, anthropometry and serum proteins. Anthropometry included body-mass index (BMI), mid upper arm circumference (MUAC), triceps skinfold (TSF) and mid arm muscle circumference (MAMC). TT was measured immediately after PEG and survival was recorded. TT cut-offs were established by comparison with other anthropometric parameters and using the ROC analysis. The correlation between TT and survival was assessed.

RESULTS

227 patients (161 men and 66 women) aged 23-96 years. Most presented head or neck cancer (51.1%). Undernutrition was identified in 57.7% according with BMI. Median TT was 25 mm (IQR = 10). TT was correlated with BMI (R = 0.5), MUAC (R = 0.5), TSF (R = 0.5) and MAMC (R = 0.4) (p < 0.01), respectively, being accurate in predicting undernutrition (AUROC 0.71 ± 0.033, p < 0.01). TT <20 mm showed positive predictive value of 81.6% and specificity of 84.4% to detect undernutrition. TT was correlated with survival (R = 0.1) (p = 0.05). Head or neck cancer patients' survival was significantly lower if TT ≤ 25 mm (p = 0.03).

CONCLUSIONS

TT is variable among PEG patients but values below 20-25 mm are suggestive of undernutrition. TT defined in the day of the gastrostomy procedure is the easiest anthropometric parameter that can be obtained from a PEG patient. Due to its higher positive predictive value and correlation with survival, TT should be viewed as an additional anthropometric tool specific for PEG patients, with diagnostic and prognostic value.