This review attempted to assess the use of pacing to prevent irregular heart rhythms after coronary bypass surgery. The authors concluded that biatrial pacing was effective, but that there is insufficient evidence pertaining to single atrial pacing. Methodological weaknesses and poor reporting of the review process mean that the conclusions should be treated with caution.

To critically review the trial evidence regarding the efficacy of epicardial atrial pacing (AP) to prevent post coronary artery bypass graft (CABG) atrial fibrillation (AF).

MEDLINE was searched for papers published in the English language before March 2003; the search terms were given. Abstract books from major U.S. and European scientific meetings were checked for relevant abstracts from 1996 onwards.

Study designs of evaluations included in the review

Randomised controlled trials (RCTs) only were included.

Specific interventions included in the review

The inclusion criterion was epicardial AP compared with control (not defined). In the included studies, right, left, or biatrial AP was used. The pacing protocols differed between the included studies and were given in the paper. Comparisons with control (unspecified) and between different pacing sites were made. The authors stated that different concomitant drugs were used in the trials, but no details were given.

Participants included in the review

Studies of people undergoing CABG were sought. The majority of the people in the included studies underwent isolated CABG. However, some also had valve surgery either with or without CABG.

Outcomes assessed in the review

The primary outcome was the incidence of postoperative AF. The secondary outcome of interest was the duration of postoperative hospitalisation. The outcomes in the included studies were either AF or a combination of AF and flutter. The duration of arrhythmia was greater than 1 minute to greater than 1 hour. The duration of follow-up ranged from 24 hours on the second postoperative day, to total in-hospital stay. Details of complications and the subsequent discontinuation of pacing were also discussed.

How were decisions on the relevance of primary studies made?

The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.

The authors did not state that they assessed validity.

The authors did not state how the data were extracted for the review, or how many reviewers performed the data extraction. The data extracted included the incidence of post operative AF (as a percentage) and the number of days of hospital stay for the treatment and control groups. P-values were presented for the differences within individual studies.

How were the studies combined?

The studies were combined in a narrative discussion and outcome data were tabulated.

How were differences between studies investigated?

The authors discussed the differences between studies within the narrative.

Thirteen RCTs (1,387 participants) were included. Seven of these studies (874 participants) reported results for the duration of hospital stay.

In studies that compared biatrial AP with control, the incidence of AF was generally less in the pacing groups, although this was statistically significant in only five of the nine studies (see CRD Commentary for comment on the interpretation of P-values in the results tables). In one study the incidence was 50% in the pacing group and 22.2% in the control group; because of this the study was stopped early after the recruitment of 21 people (out of a planned 200).

In studies that compared right and left AP with control, the incidence of AF was generally similar, although one study found a markedly lower incidence in the right AP group compared with control left AP and biatrial AP.

In two of the seven studies that reported on the length of hospital stay, this was significantly shorter in the biatrial AP group compared with the control. In a third study the length of stay was similar, but the time spent in the intensive care unit was significantly less in the biatrial AP group. The remaining studies showed no difference in the incidence of AF between the pacing and control groups.

Adverse effects: in some studies pacing was discontinued early because of lead failure. No incidences of cardiac tamponade were reported.

None. However, the authors noted that the duration of hospital stay (the secondary outcome of the review) is related to the overall cost of treatment.

Postoperative biatrial AP appears to be effective for preventing atrial fibrillation after CABG. The effectiveness of single-site right or left AP was less clear.

The aims of this review were clearly stated. The search was limited to English language publications and MEDLINE was the only database searched. In addition, the search terms listed in the paper were phrases which appear inadequate for a systematic search. It is possible that studies have been missed and this could affect the outcomes of the review. The authors gave no details about the methods of the review, how the studies were selected, or how study quality was assessed. Two of the thirteen trials included were reported in abstract form only. It is possible that decisions made during the review process can introduce bias that could affect the results of a review.

No information about the included participants, severity of illness, or concomitant treatment regimens was given. It is therefore difficult to know what type of patients the results of this review are generalisable to. The results were tabulated and discussed in a narrative, although the studies were not adequately grouped in the latter. In addition, it was sometimes difficult to tell whether the comparisons were made with the control or between different pacing methods; this confusion also relates to the tabulated P-values. If the results had been presented as a formal meta-analysis, instead of a narrative discussion, the interpretation of the outcomes would have been clearer. The authors' recommendations include those relating to cost-effectiveness, however, cost-effectiveness does not appear to have been considered in the aims of the review. In view of these comments, the authors' conclusions should be treated with caution.

Practice: The authors did not state any implications for practice.

Research: The authors stated that further research is needed, both to determine the effectiveness of single-site pacing and the cost-effectiveness of pacing strategies post CABG.

Reviewer's comment: The authors did not say that they searched for cost-effectiveness studies, and they did not discuss this within the review.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.