Predicting 30-day mortality following PEG insertion: External validation of the Sheffield Gastrostomy Score and analysis for additional predictors

Background and study aims: The Sheffield Gastrostomy Score (SGS) was devised to stratify patients by calculating their risk of mortality at 30 days following PEG insertion. The aim was to externally validate the SGS and identify any further predictors of 30-day mortality.

Patients and methods: Retrospective review of all PEG insertions performed over a ten year period in our centre. All patients who had a new PEG inserted were identified and the SGS calculated. Additionally, demographic, indication for PEG insertion and other blood results were recorded. Receiver operating characteristic curves were calculated and subsequent univariate and multivariate analysis was performed to identify additional risk factors for 30 day mortality.

Results: The PEG database comprised 1373 patients, of which 808 were suitable for analysis. For each increasing SGS gradation mortality rose, with 4% of those scoring 0 compared to 50% scoring 3. An area under the ROC curve of 0.69 (95% confidence interval 0.64–0.74) indicated good discriminative capacity. Multivariate analysis demonstrated that age ≥60 years (OR = 2.1 p = 0.016), serum albumin concentrations of 25–34 g/l (OR = 2.5 p = 0.001) or <25 g/l (OR = 6.8 p < 0.001), C-Reactive Protein ≥10 mg/l (OR = 2.7 p = 0.009) and lymphocyte count of <1.5 × 10⁹/l (OR = 2.0 p = 0.004) increased the odds of 30-day mortality, whilst referral for PEG placement whilst an inpatient decreased the risk of death (OR = 0.53 p = 0.005).

Conclusions: The SGS displayed reasonable predictive ability but the area under the curve is not high enough for routine clinical use. Modelling of further predictors from a multicentre study could provide scope for updating the SGS potentially improving patient selection.