Ashraff, Suhel and Raza, Zahid and Santos, Derek and Siddiqui, Muhammad A and Carline, Tom (2012) Development of a prognostic model for stump healing in patients with advanced atherosclerosis. In: Leading Innovative Vascular Education Conference, May 24-27, 2012., Alexandroupolis, Greece.
Introduction This study aimed to explore the role of haematological markers and predictive factors on the stump healing in patients who underwent lower extremity amputation surgery at the Royal Infirmary of Edinburgh. Methods Retrospective analysis of 300 patients was retrieved who had undergone lower extremity amputation surgery between 2006 and 2010. A predictive model was developed using backward stepwise logistic regression. The model discrimination was assessed by the receiver operating characteristics curve and its calibration by the Hosmer and Lemeshow goodness of fit test. Result Three variables were identified which influenced stump healing. Serum Sodium – The lower limb stump healing was 75% likely in patients with normal serum sodium compared to that of patients with abnormal serum sodium (odds ratio [OR] 1.756; 95% confidence interval [CI] 1.048-2.942) Serum Creatinine– Patients with normal serum creatinine were 66% more likely to have their stump healed (OR 1.664; 95% CI 0.94 to 2.946) Serum High Density Lipoproteins – A normal serum HDL resulted in a 75% more likely chance of healing compared to those with abnormal serum HDL (OR 1.753; 95% CI 1.061 to 2.895). There was a good calibration as indicated by Hosmer and Lemeshow goodness of fit test (P=0.879) and the c-index was 0.612. Conclusion Serum Sodium, Serum High Density Lipids and Serum creatinine are useful predictors of lower limb stump. The clinical utility of these risk categories in the healing of a lower limb stump requires further clinical evaluation in a prospective study.
|Item Type:||Conference or Workshop Item (Lecture)|
|Divisions:||School of Health Sciences > Podiatry|
|Date Deposited:||04 Jun 2012 13:21|
|Last Modified:||24 Nov 2015 12:58|
Actions (login required)