• 
• 

Three Part Question

In [patients suffering emergency re-sternotomy after cardiac surgery on the ICU] do [Antibiotics or iodine washouts] results in a lower incidence of [sternal wound infections]?

Clinical Scenario

A patient two hours after a double valve and grafts suddenly goes into ventricular fibrillation as you are passing by his bed in the intensive care unit. Three rapid attempts at defibrillation fail and the nurse who was looking after him said that he had been very unstable with a high CVP prior to the arrest. You elect to perform an emergency re-sternotomy, which relieves a tamponade and the heart spontaneously cardioverts into sinus rhythm. A vein proximal anastomosis was bleeding and you repair this and you are eventually happy to re-close the chest. The anaesthetist asks you if you want any more antibiotics and the scrub nurse asks you if you want a betadine washout. You do this as you are not sure how sterile one of your scrubbed colleagues were, but you are not sure if this is necessary.

Search Strategy

Medline 1950–Oct 2007 using the OVID interface.
EMBASE 1980–Oct 2007 using the OVID interface.
Medline:[open chest.mp OR internal cardiac.mp OR resuscitative thoracotomy.mp OR open heart.mp] AND [CPR.mp OR exp Cardiopulmonary resuscitation/or massage.mp]

Embase:[open chest.mp OR internal cardiac.mp OR resuscitative thoracotomy.mp OR open heart.mp] AND [CPR.mp OR exp resuscitation/or massage.mp]

The Cochrane database for systematic reviews and central register of controlled trials was searched using the term ‘open chest’, or ‘internal cardiac’ or CPR. NICE, SIGN, STS, AHA and ESC guidelines were searched.

Search Outcome

Two hundred and sixty-three papers were found in Medline, 256 in EMBASE and eight articles in the Cochrane library. Of these nine were felt to be relevant

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Engleman et al, 2007, USA	Practice Guideline from the Society of Thoracic Surgery (No recommendations directly for patients who suffer a cardiac arrest requiring emergency re-sternotomy)	Guideline (level 1, excellent)	Guideline recommendation	Primary prophylactic antibiotic for elective cardiac surgery is a first generation cephalosporin which is usually cefazolin. (Class IIA) Mupirocin is recommended as a routine prophylactic measure For patients who are considered B-lactam or penicillin allergic, vancomycin is recommended as the primary prophylactic antibiotic with additional gram-negative coverage
			High risk patients	For patients at high risk of Staphylococcal infection, vancomycin may be reasonable. (Class IIB, level of evidence C)
SIGN Publication, 2000, UK	Guideline from the Scottish Intercollegiate Guidelines Network	Guideline (level 1, excellent)	Recommendation	Antibiotics are recommended for all patients undergoing cardiac surgery (Grade B recommendation based on level II evidence)	No specific antibiotics are recommended No specific recommendations for patients requiring emergency re-sternotomy
Kriaras et al, 1996, Greece	All patients (12) who underwent open chest CPR on the day of surgery were included in this study Dates:Dec 1993–Mar 1995 10% iodine spread around the peri-sternotomy skin. Vancomycin 500 mg iv given peri-procedure. Mediastinal iodine and then saline washout at the end of the procedure	Case series (level 5,unsatisfactory)	Cardiac arrest on first postoperative day in ICU	12/2140 patients (0.6%)	The hypothesis that their protocol protects against infection is not sufficiently supported by the evidence presented here in 10 patients Abstract only, full paper not published
			Arrest survival	10/12 patients (83%)
			Survival to discharge	8/12 patients (67%)
			Wound infections	None
McKowen et al, 1985, USA	2 year retrospective audit of 88 resuscitations in 64 patients Dates: Jul, 1982-May 1984 Patients group: Patients undergoing open resuscitation through a midline sternotomy Single centre, Pittsburgh USA Povodine-Iodine preparation of the skin and sterile dressing pack.IV antibiotics are given simultaneously. Bacitracin washout prior to closure also a seperate chest reopening set is used 2 groups, Group 1: primary closure, (n=31). Group 2: secondary closure (n=18)	Retrospective cohort study (level 4, fair)	Wound complications in 49 survivors	Group 1 (primary closure) 2 wound infections Group 2 (secondary closure) , No wound infections. Total 4% wound infection rate.	Only 2 patients actually had a chest wound infection. The antibiotic given was not specified in this paper.
			Wound organism	Patient 1: Staph epidermidis. Patient 2: Escherichia coli
			Cause of chest reopening	Cardiac arrest 15%, pernicious ventricular arrhythmia 14%, shock 30%, tamponade 35%, exsanguinating haemorrhage 7%
			Arrest survival	49/64(77%)
			Survival to discharge	60%
Charalambos et al, 2006, UK	9 year retrospective study of 240 patients Dates:1991-2000 Patient group: Patients who had chest reopening for bleeding or tamponade on the ICU. Patients who arrested were excluded. Majority reopened in the ICU rather than theatres Chest prepared using provodine-iodine antibiotic was usually Flucloxacillin or equivalent antibiotic, duration varied. Single centre: Manchester Royal Infirmary, UK	Retrospective cohort study (level 4, fair)	Complications	7 sternal wound infections (2.9%)	This paper considered all chest re-openings in the ICU rather than in theatres but specifically excluded patients who had a cardiac arrest
			Cause of reopening	20/240 (86%) bleeding. 22/240 (9%) tamponade. 11/240 (5%) both.
			Cause of reopening	125/240 (55%) focal bleeding. 74/240 (33%) diffuse bleeding. 11/240 (5%) both. 25/240 (12%) packed and not closed. 13/240 (10%) further chest reopening
			Reopening survival	224/240 (84%) survived
			Incidence requiring reopening	240 patients (3.4%)
Anthi et al, 1998, Greece	30 month audit of 29 patients who suffered an unexpected cardiac arrest Dates: Dec 1993-Mar 1996 Patient group: Patients in the ICU suffering a cardiac arrest within 24 h of surgery Protocol: CPR and if no restoration of output after 3-5 min then proceed to chest reopening Betadine to skin. No mention of antibiotics Single centre: Onassis Cardiac Surgery Centre, Athens	Prospective cohort study (level 3, good)	Cause of arrest	13/29 (45%) VF/VT 11/29 (38%) Brady arrhythmia 5/29 (17%) EMD	No antibiotics or iodine washout mentioned in the paper
			Interventions during arrest, or cause of arrest	14/29 (48%) MI 5/29 (17%) Tamponade 3/29 (10%) Graft malfunction 7/29 (24%) Unknown
			Arrest survival	Closed chest CPR successful in 13/29 (45%) 14/16 having open chest. CPR recovered an output
			Out of hospital survival	23/29 (79%) survived to discharge
			Resuscitation not in ICU	Excluded in this study
			Incidence requiring CPR	29/3982 (0.7%) required CPR in the first 24 h
			Wound infections	None
Fairman et al, 1981, USA	42 month retrospective audit of 64 patients who had 74 re-thoracotomies Dates: Jan 1977- July 1980 Patient group: Patients who had an emergency re-thoracotomy after cardiac surgery for inadequate circulation University of Pennsylvania Iodine skin preparation, and sterile towel drapes. Re-irrigated prior to closure 3 days of a cephalosporin and aminoglycoside afterwards	Retrospective cohort study (level 4, good)	Cause of arrest	13/64 arrhythmia 15/64 massive bleeding 6/64 suspected tamponade 10/64 unexplained 20/64 progressive deterioration	5% medistinal wound infection rate (2 patients) If the first year surgical resident did the thoracotomy, survival was 29%. Thoracic fellow survival was 41% Of the 20 patients with progressive deterioration, 7 had initially resuscitated and 2 had a tamponade Fairman call for training of nurses in advance of emergency chest reopening
			Arrest survival	37 of 74 (50%) re-thoracotomies
			Out of hospital survival	19/64 discharged (30%) Best survival with tamponade or bleeding. No survivors in the progressive deterioration group
			Incidence requiring CPR	64/2112 (3%) had emergency re-thoractomy in ICU (not necessarily arrested)
			Longest time to reopening resulting in survival to discharge	12 h, 12-24 h - 4, 24-48 h - 7, more than 2 days - 10 patients. Survival by time not documented
El-Banayosy et al, 1998, Germany	2 year retrospective audit of 113 patients Dates: Jan 1993-Dec 1994 Patient group: All patients with circulatory collapse requiring CPR within 7 days of surgery Adults but transplants and paediatric patients excluded Single centre:North Rhine heart Centre, Bad Oeynhausen, Germany Protocol: After 20-30 min of CPR IABP performed. If unsuccessful and operation <48 h - chest reopening. Unsuccessful and operation >48 h - Fem Fem Bypass	Retrospective cohort study (level 4, good)	Complications in the 79 survivors	10 sepsis (8.9%) 11 Renal failure (14%) 10 GI failure (13%) 8 Neurologic (10%) 2 Limb ischaemia 1 Pneumothorax	Duration of CPR 2 to 230 min (mean 30 min) No mention of antibiotic protocols usage mentioned in the paper
			Cause of arrest	58/113 (51%) VF 22/113 (19.5%) EMD 6/113 (5.3%) Asystole
			Interventions during arrest, or cause of arrest	47 MI 9 bleeding 4 heart failure 5 patients had Fem Fem bypass - all died 49/113 had IABP (24-49% survived) 24/113 had resternotomy (13 or 54% survived) 6 patients had a VAD (7 of 47% survived)
Raman et al, 1989, Australia	39 patients who arrested within 72 h of a cardiac surgical operation between 1984 and 1988. 25 CABG, 2 transplants, 12 valves, 1 aneurysmectomy Divided into 2 groups retrospectively Group A (24 patients): Open cardiac massage and resternotomy Group B (15 patients): External cardiac massage only Mean time to reopening 5.6±2 min. After successful chest reopening patient always taken to theatre for closure after povodine-iodine washout. Periresusitative antibiotics were 'recommended' for 48 h	Retrospective cohort study (level 4, good)	Cause of arrest	Group A (after resternotomy) Tamponade 5 (21%) Bleeding 8 (33%) Dissection 1 Graft thrombosis 1 Ruptured ventricle 1 RV failure 1 Arrhythmia 1 Group B (autopsy)	Of note no damage to the heart was noted from any external cardiac compression No sternal wound infection They provided a protocol indicating emergency re-sternotomy after 5 min of unsuccessful resuscitation. Reopening by a cardiac surgeon, return to theatre for closure, IV antibiotics, povodine-iodine washout
			Interventions during arrest, or cause of arrest	Group A Evacuation of clot 4 Regraft or repair 7 (29%) CPB 7 (29%) IABP 4 RVAD 1 Pacing 5
			Arrest survival	Group A 21/24 (87%) Group B 5/15 (33%)
			Infective complications	No wound infections reported

Comment(s)

In 2007, The Society of Thoracic Surgeons published a guideline on antibiotic prophylaxis for elective cardiac surgery [Engleman]. They recommend that a first generation cephalosporin (usually cefazolin) should be the antibiotic of choice for elective cardiac surgery with the addition of vancomycin for patients with increased risk of Staphylococcal infection. Mupirocin ointment is recommended as an additional routine prophylactic measure. The SIGN guidelines recommend antibiotic prophylaxis for patients undergoing cardiac surgery. However, no specific antibiotics were recommended

In these two guidelines some recommendations for high-risk patients are given but neither address emergency re-sternotomy in patients who have recently received these prophylactic antibiotics and may not necessarily have had a sterile reopening.

Kriaras et al. published the only paper on patients after cardiac surgery who had open chest CPR on the day of surgery specifically in order to look at the issue of antimicrobial protection. Twelve patients had 10% iodine spread around the peri-sternotomy skin and vancomycin 500 mg intravenously was given peri-procedure. Mediastinal iodine and then saline washout were given at the end of the procedure. There were no wound infections, and they concluded that this protocol might be a useful intervention in emergency situations.

McKowen et al. reported the outcomes from resuscitation of 64 cardiac surgical patients after emergency re-sternotomy. Their practice was to use povodine-iodine preparation of the skin. Intravenous antibiotics were given simultaneously and bacitracin washout prior to closure. Only 2 out of 49 patients had a wound infection after this (4%).

Charalambos et al. conducted a study on patients who had chest reopening for bleeding or tamponade on the intensive care unit. Patients who arrested were excluded. The sternum was prepared using povodine-iodine solution and prophylactic antibiotic was usually flucloxacillin or an equivalent antibiotic. There was a variation in the duration of antibiotic administered. The incidence of sternal wound infection was 3%.

Anthi et al. reported the outcomes of 16 emergency chest reopenings after a cardiac arrest. They only reported that betadine was applied to the skin and full sterile technique was used. No patients suffered a wound nfection.

Fairman and Edmunds reported 64 patients who had an emergency re-sternotomy after cardiac surgery for inadequate circulation. These patients had iodine skin preparation and sterile towel drapes. Re-irrigation was performed prior to closure. Following that, patients had three days of a cephalosporin and aminoglycoside. Wound infection rate was 5% in survivors.

El Banyosy et al. reported the outcomes of 113 patients who arrested within 7 days of cardiac surgery . They found that 7 of the 79 surviving patients had at least one episode of sepsis after the resuscitation (9%). No mention of antibiotic use was given

Ramen et al. reported the outcomes of 39 patients who arrested after cardiac surgery in 1989. Twenty-one patients had an emergency re-sternotomy with povodine-iodine skin preparation, aseptic reopening on the intensive care unit and perioperative antibiotics for 48 h if successful. In addition, if successful the patient was returned to theatre for povodine-iodine washout and closure.

Of note a few of these papers documented ‘full-aseptic technique’ but no more detail than this was given. Thus, we would propose that a gown and gloves with full patient draping would constitute ‘full-asepsis’ in this setting. We also propose that it is not necessary to wash your hands prior to putting the gown and gloves on due to the difficulty of putting gloves on with wet hands and the necessity for rapid emergency re-sternotomy.

Clinical Bottom Line

For patients who require an emergency re-sternotomy on the intensive care unit, the incidence of sternal wound infection or sepsis after this emergency treatment is around 5%. We found only seven papers that documented the incidence of infection after emergency re-sternotomy. Of these seven papers five documented that they routinely gave additional intravenous antibiotics and an iodine washout. The other two papers did not report whether this was done. We conclude that even though the incidence of subsequent infection is low in the cardiac arrest situation, full aseptic technique including gown and gloves might be regarded as best practice. It is common practice also to give additional antibiotics and a povodine-iodine washout although we could identify no studies other than uncontrolled cohort studies in support of this.

References

1. Morley P, Zaritsky A. The evidence evaluation process for the 2005 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Cardiovascular Care Science With Treatment Recommendations. Resuscitation 2005;67:167–170.
2. Engelman R, Shahian D, Shemin R, et al. The Society of Thoracic Surgeons practice guideline series: antibiotic prophylaxis in cardiac surgery, part II: antibiotic choice. Ann Thorac Surg 2007;83:1569–1576.
3. SIGN. Antibiotic prophylaxis in surgery, Scottish Intercollegiate Guidelines Network, number 45. Thistle Street, Edinburgh, 2000.
4. Kriaras I, Anthi A, Michalopoulos A et al. Antimicrobial protection in cardiac surgery patients undergoing open chest CPR. Resuscitation 1996;31:10–1.
5. McKowen RL, Magovern GJ, Liebler GA et al. Infectious complications and cost-effectiveness of open resuscitation in the surgical intensive care unit after cardiac surgery. Ann Thorac Surg 1985;40:388–392.
6. Charalambous CP, Zipitis CS, Keenan DJ et al. Chest reexploration in the intensive care unit after cardiac surgery: a safe alternative to returning to the operating theater. Ann Thorac Surg 2006;81:191–194.
7. Anthi A, Tzelepis GE, Alivizatos P et al. Unexpected cardiac arrest after cardiac surgery: incidence, predisposing causes, and outcome of open chest cardiopulmonary resuscitation. Chest 1998;113:15–19.
8. Fairman RM, Edmunds LH Jr. Emergency thoracotomy in the surgical intensive care unit after open cardiac operation. Ann Thorac Surg 1981;32:386–391.
9. el-Banayosy A, Brehm C, Kizner L et al. Cardiopulmonary resuscitation after cardiac surgery: a two-year study. J Cardiothorac Vasc Anesth 1998;12:390–392.
10. Raman J, Saldanha RF, Branch JM et al. Open cardiac compression in the postoperative cardiac intensive care unit. Anaesth Intens Care 1989;17:129–135.