Randomised control trial

Vander Salm TJ, Okike ON, Pasque MK, Pezzella AT, Lew R, Traina V et al.
Reduction of sternal infection by application of topical vancomycin.
Journal of Thoracic & Cardiovascular Surgery
1989;98(4):618-22.

Submitted by:Tom Barker - SHO, Cardiothoracics
Institution:MRI
Date submitted:11th November 2002

Before CA, i rated this paper: 4/10

1 Objectives and hypotheses

1.1 Are the objectives of the study clearly stated?

Objective is to test the hypothesis that topical vancomycin applied to the cut sternal edges will reduce the incidence of sternal infections

2 Design

2.1 Is the study design suitable for the objectives

The study design is suitable for the objectives although double rather than single blinding would have been more desirable.

2.2 Who / what was studied?

All patients of 3 surgeons during May 1986-July 1987 who underwent cardiac operations through a median stenotomy. The incidence of sternal infection (which included mediastinitis but excluded superficial sternal wound infections) were compared between the experimental group (topical vancomycin/gelatin paste) and control groups (gelatin paste alone).

2.3 Was this the right sample to answer the objectives?

The right sample was used. Consecutive patients having median sternotomys for cardiac surgery.

2.4 Is the study large enough to achieve its objectives? Have sample size estimates been performed?

Samples size estimates have not been included. The sample size is large enough to provide a significant result, although larger numbers would add strength to the study's conclusions.

2.5 Were all subjects accounted for?

416 patients in total. 223 in the vancomycin group and 193 in the control group. Therefore the patient numbers add up in the abstract. However, 417 patients are stated in the method. This is probably a typing error.

2.6 Were all appropriate outcomes considered?

All outcomes were considered for the original hypothesis. Was there a sternal infection or not? They have defined a sternal infection.

2.7 Has ethical approval been obtained if appropriate?

Approval by the Human Studies Committee was obtained.

2.8 Were the patients randomised between treatments?

Patients were randomised between the two groups.

2.9 How was randomisation carried out?

Randomisation using hospital record number was performed. Although consecutive patients were used this is not an ideal method of randomisation.

2.10 Are the outcomes clinically relevant?

The outcomes are clinically relevant. It showed a signifcant reduction in infection rates. However, larger numbers would add support to the conclusions.

3 Measurement and observation

3.1 Is it clear what was measured, how it was measured and what the outcomes were?

It is clear what was measured: sternal infection rates. It is clear how it was measured: sternal infections (included sternal and mediastintis which involved a major reoperaton) as opposed to superficial infections (those which only necesitated laying open of the skin) which were excluded.

3.2 Are the measurements valid?

The validity of the measurements in this study are questionable as infections were graded on the clinicians opinion. It does not state the seniority of clinicians that made the decisions. If infections were confirmed by microbiological means (sternal bone cultures or swabs taken from the mediastinum) this would have been more valid.

3.3 Are the measurements reliable?

The reliability of the measurements is questionable although in practice the decision about whether a major debridement or a superficial laying open is necessary, is relatively clear. However, clinicial opinion is subjective.

3.4 Are the measurements reproducible?

The measurements would be reasonably reproducible although there will always be variability between clinican's opinions.

3.5 Were the patients and the investigators blinded?

The patients were blinded but the investigators were unblinded.

4 Presentation of results

4.1 Are the basic data adequately described?

There are clear tables descibing characteristics,operative and postoperative details of the 2 groups. In addition there is statistical analysis of other important variables and how these correlate with sternal infection. More detail is included in the infected patients table.

4.2 Were groups comparable at baseline?

The 2 groups were compared statistically for baseline characteristics.

4.3 Are the results presented clearly, objectively and in sufficient detail to enable readers to make their own judgement?

The tables are clear and provide adequate detail to allow conlusions to be drawn.

4.4 Are the results internally consistent, i.e. do the numbers add up properly?

There are several instance where numbers do not add up.
1) Table I: 283 males; Table IIA: 284 males
2) Table I: 84 diabetics; Table II: 83 diabetics
3) Table II: 6 Infected IMA+ patients; Table V: 7 IMA+ patients

4.5 Were side effects reported?

It was stated that no side effect occur with topical vancomycin.

5 Analysis

5.1 Are the data suitable for analysis?

The data is suitable and appropriate for analysis.

5.2 Are the methods appropriate to the data?

1) Fisher's test is the appropriate test as the expected count in one cell is less than five.
2) t-test is appropriate for the continuous variables however comparison of age is often a non-parametric variable. They have not proven to us that it is parametric.

5.3 Are any statistics correctly performed and interpreted?

The statistics are correctly performed and interpreted.

6 Discussion

6.1 Are the results discussed in relation to existing knowledge on the subject and study objectives?

The results are discussed in relation to knowledge on topical antibiotics in other fields of surgery and also in animal studies. However, although they have discussed evidence for systemic antibiotic use in cardiac surgery, the lack of evidence for topical antibiotics in cardiac surgery has prevented further discussion.

6.2 Is the discussion biased?

There discussion is biased in that, despite recommending the use of topical vancomycin, they have not discussed the implications regarding the possiblity of development of resistant strains in the future. This should be considered which making such recommendations.

7 Interpretation

7.1 Are the authors' conclusions justified by the data?

The conclusion that topical vancomycin reduced sternal infection rates is justified. However, further support to these conclusions would be obtained if microbiological confirmation of these infections had been made. In addition, larger numbers would also provide more support.

7.2 What level of evidence has this paper presented? (using CEBM levels)

7.3 Does this paper help me answer my problem?

This paper does give support to using topical vancomycin. However, this study only deals with topical vancomycin/gelatin gel applied to the cut edges of the bone. Some surgeons practice is to apply vancomycin powder to the cut sternal edges and surrounding wound. This study does not necessarily answer whether this is effective. Also, this study does not assess the possibility of development of resistant strains which is a possibility. This would have implications in recommending its universal use.

After CA, i rated this paper: 5/10

8 Implementation

8.1 Can any necessary change be implemented in practice?

This study might not persuade surgeons who use only systemic antibiotic prophylaxsis to change their practice because of the possibility of development of resistant bacteria. Other local surgeons use topical vancomycin powder. This study considered vancomycin/gelatin gel only. Therefore, these results might help implement change to the gel or alternatively instigate a randomised controled trial using powdered topical vancomycin.

8.2 What aids to implementation exist?

Further evidence from other randomised studies. This is the only study available that addresses this question.

8.3 What barriers to implementation exist?

Possibility of resistance development if this is universally used.