Laparoscopic Radical Simple Nephrectomy Models

The following patient parameters were recorded during the retrospective chart review (392 patients): age, sex, surgeon, radical versus simple, unilateral versus bilateral,

Our goal was to design algorithms to predict the duration of hospital stay after laparoscopic renal surgery based on preoperative patient parameters.

planned adrenalectomy, planned lymph node dissection, gastro-esophageal reflux disease, hypertension, smoking, diabetes mellitus, hyperlipidemia, chronic obstructive pulmonary disease, coronary artery disease, hematuria (micro or gross), kidney stones, obstructive sleep apnea, congestive heart failure, cerebrovascular accident, polycystic kidney disease depression, fibromyalgia, liver cirrhosis, bleeding disorders, planned transperitoneal versus retroperitoneal approach, side of nephrectomy, tumor size, nodal involvement, renal vein involvement, body mass index, American Society of Anesthesiology grade, planned specimen extraction incision and duration of hospital stay (in days).

These "design group" data were used to create two duration of hospital stay prediction algorithms: to predict if the duration of hospital stay would be greater or lesser than two days, and to predict if the duration of hospital stay would be greater or lesser than one day.

The design group was initially divided into two groups: (i) patients with a duration of hospital stay less than or equal to two days (258 patients) and those with a duration of hospital stay greater than two days (134 patients). Univariate and multivariate logistic regression analysis comparing patient parameters between these groups identified significant predictors of duration of hospital stay. These results were used to generate a linear regression algorithm to predict duration of hospital stay (greater or lesser than two days) (24,25).

The same process was repeated for the greater or lesser than one day duration of hospital stay predictor. The design group in this case was divided into: (i) those patients that had a duration of hospital stay lesser than or equal to one day (123 patients) and those with a duration of hospital stay greater than one day (269 patients).

By combining the above two models, the algorithm would predict if the duration of hospital stay was lesser than or equal to one day, one to two days or greater than two days. These models were then prospectively tested on a separate 29 patient "test group" to assess the duration of hospital stay prediction accuracy. Testing was performed on this separate group to avoid any training bias.

Delicious Diabetic Recipes

Delicious Diabetic Recipes

This brilliant guide will teach you how to cook all those delicious recipes for people who have diabetes.

Get My Free Ebook


Post a comment