Thorac Cardiovasc Surg 2018; 66(S 01): S1-S110
DOI: 10.1055/s-0038-1627976
Oral Presentations
Monday, February 19, 2018
DGTHG: Intensive Care Medicine
Georg Thieme Verlag KG Stuttgart · New York

Changes in Perioperative Creatinine Levels Predict Long-term Survival after Cardiac Surgery with Cardiopulmonary Bypass: A Single-Center Analysis in 18072 Patients

J. Boehm
1   Klinik für Herzchirurgie, Deutsches Herzzentrum München, Technische Universität München, München, Germany
,
M. Krane
1   Klinik für Herzchirurgie, Deutsches Herzzentrum München, Technische Universität München, München, Germany
,
R. Lange
1   Klinik für Herzchirurgie, Deutsches Herzzentrum München, Technische Universität München, München, Germany
› Author Affiliations
Further Information

Publication History

Publication Date:
22 January 2018 (online)

Objectives: Postoperative renal failure (RF) is a severe and frequent complication in patients undergoing cardiac surgery with cardiopulmonary bypass (CPB). The present study analyses the impact of perioperative changes in Creatinine levels as an indicator for long-term survival.

Methods: A total of 18,262 consecutive patients who underwent cardiac surgery with CPB between 2001 and 2015 at our institution were enrolled. After the exclusion of patients suffering from chronic dialysis or NTX, 18,072 patients were analyzed. Data collection was performed by using our standard clinical data acquisition program, where the data had been prospectively collected. Δ Creatinine was defined as the difference between the maximum postoperative creatinine and the preoperative creatinine. Survival for 30 day mortality and long-term mortality was estimated by Cox regression analysis. Included risk factors for Cox regression analysis were Δ Creatinine, LV-function, age, sex and type of cardiac surgery.

Results: Mean Δ Creatinine was 0.22 ± 0.66. Patients underwent CABG (n = 6528, 36.1%), valve surgery (n = 5594, 31.0%), CABG plus valve surgery (n = 2735, 15.1%), aortic surgery (n = 1399, 7.7%) and others (n = 1816, 10.0%). Patients were 66.8 ± 11.9 years old, LV function was 60.1% ± 18.0, an 5665 patients were female (31.3%). Mean follow up was 2.54 ± 3.2 years, ranging up to 16.1years. On overall survival, Cox regression analyses revealed Δ Creatinine as the strongest marker for overall postoperative mortality for (OR = 1.72, p < 0.0001), followed by female gender (OR = 1.60, p < 0.0001), and age (OR = 1.05, p < 0.0001). The other tested factors had an opposite impact on postoperative mortality: LV-function (OR = 0.98, p < 0.0001), CABG (OR = 0.47, p < 0.0001), valve surgery (OR = 0.60, p < 0.0001), and CABG plus valve surgery (OR = 0.65, p < 0.0001). On 30 day mortality, Δ Creatinine was also found to be the strongest marker mortality for (OR = 1.63, p < 0.0001), followed by female gender (OR = 1.15, p < 0.006), and age (OR = 1.06, p < 0.0001), whereas the other tested factors had again an opposite impact on postoperative mortality: LV-function (OR = 0.98, p < 0.0001), CABG (OR = 0.59, p < 0.0001), valve surgery (OR = 0.75, p < 0.0001), and CABG plus valve surgery (OR = 0.79, p < 0.0001).

Conclusion: In the current study, Δ Creatinine appears as a robust marker for the influence of RF on postoperative outcome, independently from age, sex, LV-function and kind of cardiac surgery.