|
|
||||||||
Ann Thorac Surg 2006;81:1549-1550
© 2006 The Society of Thoracic Surgeons
a Imperial College, University of London, Hammersmith Hospital, London, W12 ONN United Kingdom
b St. Marys Hospital, Praed St, London, W2 1NY United Kingdom
(Email: a.weerasinghe{at}ic.ac.uk).
We appreciate the comments mentioned in the letter by Habib and colleagues [1] regarding our article [2].
The definition for major renal adverse outcome (RAO) has evolved with time and has even been subject to revision by The Society of Thoracic Surgeons. We agree with the authors who use the percentage fractional change in serum creatinine (%
Cr) in that it may have been better, but emphasize that if the %
Cr was used, the defining levels for RAO should be derived using the distribution of serum creatinine in a patient subset not having adverse surgical complications impacting renal outcome.
The authors comment that the modeling results for major RAO are the reverse of the unadjusted trend between on-pump coronary artery bypass (ONCAB) and off-pump coronary artery bypass (OPCAB). As some of the patients qualified for more than one RAO group, we did not want to duplicate the events before multivariate analysis, and we classified each subject once in the highest level of the hierarchy. Thus the cumulative trend of major RAO is 10.4% in ONCAB patients (134 of 1,224) compared with 13.9% in OPCAB (114 of 817). An over-modeling artifact is unlikely because the model is simple, with restricted predictors and considerable events, and it accommodates linearity assumptions and multi-colinearity.
Figure 1 is not a direct extrapolation of simple probability but derived from an ordered logistic regression analysis using preoperative creatinine and type of procedure as the two independent predictors of RAO. In addition to the estimates for the covariates regressed, two further coefficients are provided which represent (1) the cutoff value for a minor or major RAO combined, and (2) the cutoff value for a major RAO alone.
The specific objective of our study was to compare the occurrence of functional renal dysfunction with ONCAB versus OPCAB. It was not the aim to investigate the effect of specific factors related to cardiopulmonary bypass (such as hemodilution) as causes for differences in outcome.
We thank the authors for raising these issues, and we thank the editor for the opportunity to respond.
| References |
|---|
|
|
|---|
Related Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |