| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Letters to the Editor |
Department of Cardiothoracic Surgery and Anesthesiology, Örebro University Hospital, 701 85 Örebro, Sweden
Received 29 June 2008; accepted 16 July 2008.
* Corresponding author. Tel.: +46 70 6965210; fax: +46 19 611 3943. (Email: orjan.friberg{at}orebroll.se).
Key Words: Cardiac surgery • Postoperative complication • Sternal wound infection • Risk factors • Tracheostomy • Logistic regression analysis
Ngaage et al. report post-sternotomy percutaneous tracheostomy (PDT) to be associated with an approximately three-fold risk for deep sternal wound infections (SWIs) (OR 3.22, 95% CI 1.14–9.31, p < 0.0001) [1]. The following editorial concludes that ‘the authors have shown that early PDT should be discouraged or delayed after cardiac surgery’ [2]. We are less convinced.
There are several potential explanations to the increased incidence of SWI that the authors do not discuss. The majority of SWIs develop after discharge and might even go unrecognised for the cardiac surgical unit unless there is a dedicated post-discharge surveillance program [3]. The controls may thus, due to shorter postoperative stay, have had their SWI-rate underestimated. Nevertheless, we appreciate that the rate of infections in the PDT group was higher than expected.
One principal methodological problem is that PDT is an indicator of a complicated postoperative course. Many patients experiencing severe postoperative complications carry an increased risk for SWI. Several of these complications were not included in the multivariable analysis. For instance reoperation for bleeding/tamponade and septicaemia could have played a directly causative role, whereas other variables could have increased the susceptibility for infection. Furthermore, patients that develop severe SWIs often require readmission to the ICU and eventually PDT due to extended ventilator treatment. Unfortunately, the employed statistical model fails to account for the primary event unless this is specified in the database.
Consequently, in this statistical model PDT could probably have been shown to be a ‘predictor’ of many of the reported complications, such as reopening for bleeding, readmission to the ICU etc.
In another recent study, tracheostomy was not identified as an independent risk factor for deep SWI, it just served as a surrogate for respiratory failure [4]. In our opinion, it is not unlikely that respiratory failure could similarly serve as a surrogate for a complicated postoperative course.
However, even if all registered risk variables had been included, uncertainty would still exist as to whether unregistered data could have introduced bias into the model (e.g. patients in poor clinical condition might have PDT on more liberal indications than patients with good recovery considered likely to be extubated within few days).
Multivariable logistic regression analyses as well as propensity score matching are examples of methods commonly used to compensate for differences in baseline data or risk factors in retrospective studies. It is important to understand the basic presumptions and limitations of these methods, and why they can never replace randomised controlled trials (RCTs) [5].
A properly designed and conducted RCT has the potential to eliminate not only all known, but also possible unknown differences in risk factors between the groups.
There are issues that cannot be studied in an RCT. However, given that endotracheal intubation according to the editorial can be continued safely up to three weeks such a study would be feasible.
In our opinion it is still an open question whether post-sternotomy percutaneous tracheostomy increases the risk for deep SWI or not. Only an adequately powered RCT can sort this out.
References
This article has been cited by other articles:
|
|
 |
|
  D. L. Ngaage Reply to Friberg and Svedjeholm Eur. J. Cardiothorac. Surg., October 1, 2008; 34(4): 931 - 931. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
