Abstract

The incidence of re-operative median sternotomy is rising. During resternotomy, catastrophic haemorrhage remains a dreaded complication. We describe our approach and experience with the combined use of Mayfield resternotomy retractor and anterior sternal retraction which allows division of adhesions between the sternum and mediastinal structures under direct vision with endoscopic or conventional instruments prior to resternotomy with a standard Hall reciprocating saw. The mean time to divide the retro-sternal adhesions was 26.4±16.7 min. No morbidity related to sternal division was observed. For redo surgery, repeat sternotomy under direct vision may reduce the sternotomy related morbidity (especially the need for cardiopulmonany bypass due to significant haemorrhage) and mortality.

1 Introduction

Median resternotomy has become a commonly performed procedure in cardiac surgery. Society of Cardiothoracic Surgeons of Great Britain and Ireland database (1993–1998), reports the overall mortality for redo cardiac procedures to be 9.1% compared with 3.1% for first time cardiac procedures [1].

The increased risk of redo procedure is partly related to the complications of resternotomy which include severe haemorrhage requiring cardiopulmonary bypass and sternal fractures [2,3]. The exact incidence of these complications is unknown and underestimated in the literature for two reasons: (a) lack of consistent definition and (b) underreporting as it may imply a technical error [2,3].

Factors associated with an increased risk of resternotomy i.e. ‘high-risk resternotomy’ include: more than first time reoperation, enlarged heart chamber or vessel, history of mediastinitis or sternal osteomyelitis, patent grafts and recent previous operation (≪5 years) [4].

The importance of direct visualization of the retro-sternal structures was emphasized initially by the anterior sternal retraction technique [5] and integrated by Mayfield who introduced the combination of the above technique with thoracoscopic equipment [6,7].

2 Material and methods

Between 1998 and 2001, 253 resternotomies (almost 10% of all cardiac cases) were performed at St Mary's Hospital. A retrospective review was undertaken to assess the safety of the approach described in terms of intraoperative injury to mediastinal structures, significant haemorrhage requiring CPB and the need for blood transfusion for substantial bleeding during sternal re-entry and haemorrhage related mortality between the high-risk patients.

2.1 Statistical analysis

Outcome of groups were compared using the Fisher's exact test and a P value of less than 0.05 was considered significant. The statistical software PEPI-INFO version 0.57 was used.

2.2 Technique of video assisted retro-sternal dissection

The usual anaesthetic preparation and haemodynamic monitoring lines were used and two external defibrillator pads were placed on the chest wall.

We perform a slightly longer than standard skin and subcutaneous incision (extending 3 cm distal to the xiphoid process). The sternal wires or other material (ethibond, bands) that had been used for previous sternal closure are left in place. Then the linea alba is divided and the xiphoid process is excised. The patient is placed in Trendelenburg position. The application of the modified Rultract retractor (Rultract, Inc, Cleveland, OH, USA) for ventral and cephalic traction of the lowest part of the sternum allows the insertion of the Mayfield retractor (Genzyme Surgical Products, Cambridge, MA, USA) to dissect the retro-sternal space (Fig. 1 ). For the dissection under direct or video assisted visualization we used conventional or thoracoscopic instruments as indicated. The zero degree 5-mm endoscope (Karl Storz, Charlton, MA, USA) was attached to the Mayfield retractor dissecting close to the sternal plane following the wires. The blade of the Mayfield retractor in the sub-sternal space offers the necessary countertraction to follow the correct plane and to divide thick adhesions with scissors or diathermy (Fig. 2 ). Suction or packing with swabs allows control of minor bleeding points. Dissection continues cranially till the suprasternal notch is reached. The sternal wires or other closure material are then removed and the sternum is divided using a Hall reciprocating saw (Zimmer Ltd).

Fig. 1

Rultract and Mayfield retractors with the attached endoscope.

Fig. 1

Rultract and Mayfield retractors with the attached endoscope.

Fig. 2

Retrosternal adhesions endoscopic view.

Fig. 2

Retrosternal adhesions endoscopic view.

3 Results

Resternotomy was performed using the oscillating saw technique (OSG) in 218 patients and the video assisted retro-sternal (VAR) dissection in 35 patients. For the VAR group, the time needed for retro-sternal dissection ranged between 4 and 84 min (mean 26.4±16.7).

Thirty-one percent (78/218) in OSG and 77% (27/35) in VAR were classified as high-risk redo patients.

Overall incidence of significant organ injury at the time of resternotomy was 9/253 (4%). All these events occurred in high-risk redo patients. Therefore, we compared the outcome of high-risk redo sternotomy between the two groups (Table 1 ). The trends are apparent but all the outcome measures failed to reach statistical significance in this small series. We did not experience any episode of intraoperative arrhythmia requiring defibrillation during the dissection of the anterior surface of the heart.

Table 1

Comparison between Oscillating Saw (OSG) and Video Assisted Retro-sternal Dissection (VAR) in high-risk patients

Table 1

Comparison between Oscillating Saw (OSG) and Video Assisted Retro-sternal Dissection (VAR) in high-risk patients

4 Discussion

The increasing number and surgical difficulty of redo operations remain a challenge for the cardiac surgeon. The use of oscillating saw has been the most common and safe technique of resternotomy [3]. Important drawbacks of this method are the lack of direct vision and the need for assistance.

Few reports emphasized the use of video assisted resternotomy as a safe method for resternotomy in adult and paediatric cardiac surgery [6,7].

The main advantages of this method are: (1) it offers direct visualization of the anterior mediastinal vital structures (2) can be used as an alternative in complex cases with thick adhesions without the need to remove the posterior surface of the sternum and without any other assistance than the scrub nurse. Only limited previous thoracoscopic experience is required and definitely after the first ten cases the operative time can be reduced significantly.

In our high-risk series (patent arterial or vein grafts close to the midline), this approach has enabled us to undertake off-pump redo CABG in a significant proportion of more than 50% (12 off-pump vs. ten-on pump). The combination of preservation of patent grafts and avoidance of CPB in this high-risk group reduces morbidity and mortality [8].

Technical points that can facilitate dissection include the following:

  1. use of zero degree endoscope during the progressive insertion of the blade keeps the same axis for insertion and vision;

  2. following the wires and dissect with diathermy away from the heart in order to avoid damage or arrhythmia is to be recommended;

  3. if the preoperative angiogram shows a patent LIMA close to the midline, we use the slightly ‘off center’ approach (to the right of midline resternotomy);

  4. progressive division of the sternum can be helpful to proceed in points like the manubrium level where usually adhesions are thicker;

  5. lung hyper-inflation can facilitate dissection.

5 Conclusions

In our experience, there was clear trend towards a lower morbidity and mortality in the video assisted group for the high-risk patients and therefore our practice has moved to what we feel is a safer approach.

References

[1]
The Society of Cardiothoracic Surgeons of Great Britain and Ireland, National Adult Cardiac Surgical Database Report, 1998.
[2]
Dobell
A.R.C.
Jain
A.K.
Catastrophic hemorrhage during redo sternotomy
Ann Thorac Surg
 , 
1984
, vol. 
37
 (pg. 
273
-
278
)
[3]
Follis
F.M.
Pett
S.B.
Miller
K.B.
Wong
R.S.
Temes
R.T.
Wernly
J.A.
Catastrophic hemorrhage on sternal reentry: still a dreaded complication
Ann Thorac Surg
 , 
1999
, vol. 
68
 (pg. 
2215
-
2219
)
[4]
Macmanus
Q.
Okies
J.E.
Phillips
S.J.
Starr
A.
Surgical considerations in patients undergoing repeat median sternotomy
J Thorac Cardiovasc Surg
 , 
1975
, vol. 
69
 (pg. 
138
-
143
)
[5]
Eddy
A.C.
Miller
D.
Johnson
D.
Gartman
D.
Gregg
M.
Allen
M.
Verrier
E.D.
Anterior sternal retraction for reoperative median sternotomy
Am J Surg
 , 
1991
, vol. 
161
 (pg. 
556
-
559
)
[6]
Gazzaniga
A.B.
Palafox
B.A.
Substernal thoracoscopic guidance during sternal reentry
Ann Thorac Surg
 , 
2001
, vol. 
72
 (pg. 
289
-
290
)
[7]
Mayfield
W.R.
Endoscopic repeat sternotomy
Heart Surg Forum
 , 
1998
, vol. 
1
 (pg. 
26
-
29
)
[8]
Fanning
W.J.
Kakos
G.S.
Williams
T.E.
Jr
Reoperative coronary artery bypass grafting without cardiopulmonary bypass
Ann Thorac Surg
 , 
1993
, vol. 
55
 (pg. 
486
-
489
)