Back

Pre operative fitness score accurately predicts uneventful post operative course in gastrointestinal and hepatobiliary surgery.

VASAVADA, B.; patel, h.

2020-04-17 surgery
10.1101/2020.04.14.20057612 medRxiv
Show abstract

BackgroundAim of our study was to analyse if we can accurately predict uneventful post operative course pre operatively in gastrointestinal and HPB surgery patients. Material and MethodsWe retrospectively evaluated patients who have undergone gastrointestinal and hepatobiliary surgery at our institute in last 3 years and analysed 90 days mortality and morbidity rates among these patients. We described any 90 day morbidity and mortality as an "event". We performed univariate and multivariate analyses for factors predicting an "event". Then based on pre operative factors that predicted an "event" we formulated a score. Statistical analysis was done using SPSS version 23. ResultsTotal 264 patient operated for gastrointestinal and HPB surgeries between April 2016 to may 2019 were evaluated .Total 45 (17%) events occurred. On univariate analysis CDC grade, ASA score,Operative time,Blood products used, emergency surgeries and open surgeries predicted an event. We developed score based on pre operative factors like ASA score, CDC grade of surgery,open surgery and emergency surgeries included in the score. We proposed score grater than 2 was associated with 90 day event. This score had sensitivity of 77.78%, specificity of 81.65%. low positive predictive value of 46.67% but very high negative predictive value of 94.68%. AUROC showed AUROC of 0.797 (p < 0.0001, 95% confidence interval 0.721-0.874). Pre operative fitness score, Open Surgery and operative time independently predicted an "event" on multivariate analysis. (p =0.003 and 0.026 respectively. ConclusionsPre operative fitness score accurately predicts uneventful post operative course in gastrointestinal and hepatobiliary surgery.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.