European Journal of Cardiovascular Medicine

Closure of the median sternotomy incision is sometimes not possible post cardiac surgery as it may cause compression of the heart and lead to life threathening low cardiac output syndrome. Open chest management and delayed sternal closure is an acceptable technique for management of such patients. Its incidence reported in literature varies from 1.2-4%. Riahi et al. in the year 1975 first coined the term “cardio mediastinal disproportion” and noted that it can be managed by increasing the intra thoracic space and providing place for the heart to heal in a non-constricted environment.^[1,4,8] Furnary et al. in their study postulated the various effects of sternal closure on the myocardial contractility and cardiac output.^[2,3] As this technique came to be more frequently applied, the various advantages of the same became more apparent including^[4,5,6] ease of access for open cardiac inspection and massage, clot evacuation and bleeding control, improve ventricular contraction and output and Relive intra thoracic pressure.

The main disadvantage of the technique still remains the fear of mediastinal infection and the morbidity and mortality associated with mediastinitis. ^[2,4,5] Many studies have reported various methods to isolate the sternum from the external environment. These include the usage of a variety of synthetic material to keep the sternum apart, open sternum with direct skin closure, skin zipper etc. Sterile packing and steri-drape isolation of the mediastinum is the most followed technique proving rapid access and ease of application. ^[4,5,6] The closure is attempted once the patients’ hemodynamics are in the acceptable range. There are many studies in literature praising the advantages of the delayed sternal closure technique in congenital heart diseases in the pediatric population. ^[2,8] However the studies in adult population is limited to isolated case reports and few case series.^[9]

The purpose of this study is to evaluate the incidence, indications, complications morbidity and mortality that is associated with patients with delayed sternal closure. We postulate that despite its disadvantages, open chest management and delayed sternal closure is a viable, useful, therapeutic life saving option to be considered in all patients who do not tolerate primary sternal closure.

This is a retrospective observational study on 100 consecutive patients who fulfil our inclusion criteria. The study has been approved by the institutional ethical committee [tag the number here]

100 patients over a period of 5 yrs from  January 2016 to December 2020 were included in this study The data collected and patient details were recorded with respect to gender, age, pre op diagnosis, cardiopulmonary bypass time, technique of open chest (skin/membrane closure), duration of open chest and post op course (need for re-opening, trachestomy, medistinitis, sepsis, systemic complications-renal failure and need for dialysis, duration of ventilation, need for higher inotropes, total drainage, duration of icu and hospital stay).

The surgeries included in this were performed via median sternotomy. Cannulation stratergies were central for most cases, however in special situations like redo sternotomy, ascending aortic dissections and aneurysms, peripheral cannulation was performed. Operations were done on moderate hypothermic cardio pulmonary bypass and a few cases needed deep hypothermic circulatory arrest. As per institute protocol Delnido Solution was the cardioplegia of choice and repeat doses were given electively at 60 min. All the hemodynamic and metabolic parameters were monitored and maintained within normal limits on the pump.

The procedures performed included

1. On pump Coronary artery bypass grafting- with or without the use of an intra aortic balloon pump.
2. Valve replacements/repair- aortic and/or mitral and/or tricuspid
3. Aortic root replacement with ascending aortic replacement
4. Arch of aorta surgery
5. Descending thoracic aortic bypass
6. Intracardiac repair for complex congenital cases- total correction for tetrology of fallot, total anomalous pulmonary venous connection, ventricular septal defect with severe pulmonary artery hypertension
7. Intra cardiac myxoma and
8. Combined valve and coronary procedures with left ventricular aneurysm or rupture repair.

The cardio pulmonary bypass time was noted for all cases.

Indication For Open Chest.

Open chest and delayed sternal closure was carried out in patients in whom

1. Despite highest levels of pharmacological support persisted to have borderline hemodynamics
2. Reperfusion myocardial odema
3. Uncontrollable bleeding
4. Intractable arrhythmias
5. Very high inotropes
6. Low cardiac output.

Techinique

In all cases both pleural cavities were opened to provide additional space for the heart to expand and allow for closure and to prevent tamponade in case of uncontrolled bleeding. Where feasible the skin was approximated leaving the sternum unapproximated. When skin was direct closed it was done so in continuous fashion with non-absorbable monofilament suture. Else the closure was accomplished by suturing skin edges to a collapsible infusion bag(membrane) that been additionally sterilised by soaking in 5 % Glutaraldehyde solution for a duration of 30 min. A non-absorbable monofilament suture was used to suture the skin edges to the bag over lapping the skin by 3-5mm. The membrane was fit without any tension or dog ears to prevent exit of tissue fluid or influx of micro-organisms. Once this was complete, the skin membrane interface was sealed additionally with soaked povidone iodine guaze pieces. And an air tight seal was reinforced with 3M ioban.

The advantage of the transparent bag was that the mediastinal contents could be visualised and appropriate action was taken as and when needed-immediate resuscitation in case of arrest, clot evacuation.

Skin or membrane closure was decided based on the hemodynamic instability. In cases of incontrollable bleeding, the raw surfaces were tightly packed with gauze that in some instances were additionally soaked with hemostatic agents. The approximated skin or membrane were further reinforced with an opsite ioban to maintain sterility and act as a dressing.

ICU Protocol

All the patients were mechanically ventilated at the rate of 7-10ml per kg body weight with acceptable air way pressures. Patients were sedated and paralysed with infusions of fentanyl and vecuronium till sternal closure. In cases where deep hypothermic circulation arrest was instituted, sedation was started after confirming neurological status.

Upto 48 hrs post initial procedure all patients were maintained on fluid replacement. 48 hrs later enteral feeds were started. Blood and blood products were used liberally to maintain volume and hematocrit. Diuresis was instituted from post op day 0 itself to maintain the patient in an overall negative balance. Some patients also had dialysis (peritoneal or hemodialysis) as and when needed.

Emergency re-explorations were carried out in the intensive care unit itself. And if found stable closure was attempted in the ICU only.

Antibiotic protocol

Broad spectrum antiobiotic coverage as per established protocol was used. Combined antibiotics of Cefaperazone+Sulbactum alomg with Amikacin was used in all patients. In patients who under went  multiple re-exploration before closure, persistent increase in white cell count, fever, were prophylactically started on Meropenem and Colistin.  Mediastinal swabs were sent for cultures at the time of closure. Endo tracheal secretions in case of prolonged ventilation, and blood cultures in case of increase in total white cell count, fever were sent for culture sensitivity and antibiotic changes were done as per sensitivity report.

CLOSURE of STERNUM

Most closures were performed in the operation theatre itself. The timing of closure was decided by improvement in hemodynamics, accepted pharmacological support, correction of coagulation abnormalities, and response to temporary re approximation of sternum.

After removing the membrane and opening the skin, the swabs inside were sent for culture. The heart was noted for control of bleeding and improvement in contractility. Grafts were checked for patency. Mediastinum was irrigated with normal saline and any coagulum was removed. Betadine and saline wash was given before closure. Sternal closure was achieved by using no 5 steel wires for females and no 6 steel wires for males. Subcutaneous tissue was approximated with vicryl in continuous fashion and skin with monofilament suture with interrupted mattress technique.

The duration of open chest and timing of successful closure was noted. And in case of re opening of the delayed closed sternum and the indication for the same has also been recorded.

All major complications in the peri operative period like bleeding, prolonged ventilation, tracheostomy, need for re-opening or re exploration, low cardiac output etc. were recorded.

Out of the 2961 open hearts performed in our institute from 2016 to 2020, prolonged open chest and delayed sternal closure was instituted in 772 patients. In this, the patients who were included on our study who fulfilled our inclusion and exclusion criteria were 100(3.38%). The patient details were collected from hospital records and entered in excel sheet.

Table-1: The results are expressed as mean and standard deviation.

The male to female percentage was 68% vs 32%.

Figure-1: Procedures followed in present study

Figure-2: Indications for Open sternum in present study

Our patients had undergone chest approximation either via direct closure of the skin keeping the sternum open or approximating the skin using a bag sterilised in 0.5% glutaraldehyde for 30 min.

Figure-3: Type of Closure in present study

The patients underwent sternal closure during different time periods as per the surgeons’ discretion. 4 patients in this study did not meet the criteria for sternal closure and te=hey succumbed before an attempt at closure could be done.

Table-2: Duration of open chest procedure

Not all secondary closures were successful, 13 patients who underwent delayed sterna closure were re opened again for reasons such as bleeding, tamponade, arrthymias, and compromised haemodynamics.

Before undergoing sternal closure, a total of 23 patients out of the 100 underwent re explorations. Out of the 23, 11 (47.82%) underwent re-exploration once, 8 (34.78%) underwent re-exploration 2 times, 3 (13.04%) underwent 3 times and 1 (4.35%) underwent re exploration 4 times.

All patients who had delayed sternal closure had a pump time ranging from 95 min to 402 min. Averaging at 199 min.

Table-3: Cardio-pulmonary bypass time

The patients who had open sternum were invariably patients who were more compromised. Due to the pre operative status, or intra operative findings and experiences, they needed higher usage of dual or triple inotropes-dopamine, adrenaline and noradrenaline. In some cases due to the unstable hemodynamics, some patients were even started on vasopressin. Based on the patients need, the vaso active inotrope score was used to calculate the inotrope  usage.

Table-4: Vaso Active Inotrope Score in present study

The addition of vasopressin to the score drastically increases the score to more than 30, sometimes reaching to scores of 100 in one patient. These are those patients who were hemodynamically compromised, and the addition of vasopressin as a life saving drug to improve hemodynamics was considered.

Delayed sternal closure also affected the duration the patients needed mechanical ventilator support. 12 patients needed tracheostomy due to prolonged ventilation and difficulty in weaning from ventilator. Out of the 87 patients who survived delayed sternal closure in this study, 9 (10.34%) patients underwent tracheostomy and prolonged mechanical ventilator support.

Figure-4: Ventilator Support Duration

Bleeding was one of the main indications for open chest management. These patients had more than normally accepted drain output.

Figure-5: Drain Output in present study

Hospital survival was 87%.

Amongst the 87 patients who survived, morbidity was 33 (37.93%)

16 patients developed increased in serum creatinine. In 8 patients it was managed conservatively, 6 patients improved with peritoneal dialysis. Remaining 2 patients needed Sustained low efficiency dialysis.

19 patients needed higher antibiotics than those as per protocol. The causes were fever, increased leukocyte count, prophylactically in patients who underwent more than 2 re explorations, patients with lung collapse or as per reports of culture sensitivity reports.

9 patients needed tracheostomy as there was difficulty in weaning from the ventilator. They were discharged with the tracheostomy tube in situ.

Due to increased bleeding and need for higher quantities of transfusion of blood and blood products, 6 patients had transient increased in liver function tests that was represented in increased values of direct bilirubin and SGOT and SGPT.

2 patients had ischemic hemiplegia and were managed conservatively.

Table-5: Causes of morbidity and mortality in present study

The overall in-hospital mortality has been 13 (13%). Icu and total post operative stay for patients who underwent delayed sternal closure. The average stay has been 5.76 days ± 3.26 days and the average duration of hospital stay has been 9.59 days ±2.83 days.

Table-6: ICU and hospital stay in present study

There were no pre operative predictors of mortality and morbidity. Potential intra operative determinants like cardiopulmonary bypass time, type of surgery performed, indication for open chest did not influence post operative mortality and morbidity. However, post operative predictors of outcome have been identified as Renal function and need for dialysis. Persistence of high inotropes at the time of closure. On going coagulopathy and need for massive transfusions.

In the initial days of cardiac surgery, the sternum post operatively had to be closed to avoid the morbidity and mortality associated with mediastinitis.^[3] In 1975 when riahi and coworkers first introduced the concept of cardiomediastinal disproportion and stated that cardiac output can be improved by increased the intra thoracic space.^[1] They had a closed chest but increased the intra thoracic space by having upward traction sutures on the approximated sternum. This was progressively released as and when there was improvement in the patient haemodynamics. Since then delayed sternal closure had been used in the post operative management of patients who did not tolerate sternal approximation. There have been many studies of this technique being used in complex congenital heart disease, paediatric population and case reports in adult heart disease. However, it was not until 1981 when the first series of adult patients who underwent delayed sternal closure was published by Gielchinky et al.^[10]  It was in the article of 1992 that Antony Furnary^[3] and co-workers explained the beneficial effect of open chest on hemodynamics. Despite preserved velocity of myocardial fibre shortening, sternal closure causes significant reduction in cardiac output by reducing the diastolic filling. This is due to non-compliance of the closed chest wall, myocardial ischemia, reperfusion, oedema and also tamponade. Open chest prevents this restriction of diastolic filling by increasing the cardiac index by 59% and systolic blood pressure by 18% without significant increase in filling pressures.

The incidence of open chest in our adult patients post cardiac surgery is 3.38%. This is comparable to the study of U. Boeken and coworkers ^[7] who reported an incidence of 3.5% in their patients. Wong and collegues^[9] had an incidence of 5% in their study spanning 6 yrs. All studies quote a male preponderance in open chest management and delayed sternal closure. Ours is 68%. Udo Boeken and colleagues ^[7] had 70 % male preponderance, Yasa and colleges 67 %.^[6] The most common procedure that needed open chest management was coronary artery bypass grafting-33% followed by 22% for combined procedures and valve surgeries for 17%. Yasa and collegues^[6] found that 32.6% who underwent coronary bypass grafting needed delayed sternal closure. Combined procedures accounted for 8.7%, valves for 26%. Anderson and colleagues ^[4] reported coronary surgeries to be 18.4%, valves 17.24, & combined 54.02%. Udo Boeken^[7] had 26.9% coronary surgeries, 9.9% valve procedures and 46.2% combined surgeries. Long cardiopulmonary bypass run at the end of surgery contributes significantly to the development of myocardial oedema. These patients fare well with delayed sternal closure. The indication for open chest in our study has been multifactorial with some cases having more than one specific reason for open chest. The most common indication for open chest in our study has been Bleeding-58%. Myocardial oedema accounted for 45%. Yasa and colleagues ^[6] bleeding was 45.7%. Wong et. al^[9] reported caogulopathy to be the main cause in 70.7%. And hemodynamic instability to be 21 %. Sternal closure was performed at a mean of 21 hrs in our series. 4 patients in our study did not reach the hemodynamic criteria for closure and succumbed before attempt at chest closure was done. Anderson et al^[4] performed theirs at 3.7 days with 8 of the 87 patients dying before closure. Yasa and colleagues ^[6] closed chest at 3.48 days.  There had been no prior decision to keep the sternum open electively. All the decisions were made on table keeping in mind the haemodynamic status, inotrope support, myocardial oedema, bleeding condition. Re-explorations of the mediastinum were performed only if indicated. The number of reexplorations in our study were 23 %. This is in comparison to Wong and collegues^[9] who has a re-exploration rate before closure of 29.9%.

High doses of inotropes were essential in our patients, we used the vasoactive inotrope score to calculate the concentrations and predict the mortality and morbidity. Though the patients needed high doses in the initial post operative period, as and when the hemodynamics stabilised, we were able to wean the supports rapidly and, in most cases, bring them to acceptable range at the time of closure. There has been no correlation between the strength of the inotropes and mortality and morbidity when it came to our patients with delayed strenal closure.  Timo Koponen et al in their study noted that the vaso active inotrope score is a predictive indicator of mortality in adult patients undergoing cardiac surgery.^[11] Anderson et al^[4] in their study noted that at the time of open chest, 64 of their patients were on high inotropes, but at the time of closure, only 24 of them remained. The bypass time in our patients averaged at 198 min. Anderson et al^[4] had 88 min. Yasa and collegues^[6] 147 min. Wong et al^[9] had 175 min.

Most of our patients were extubated between 24-48 hrs post closure 42%, with 9% patients who underwent tracheostomy. Yasa and co workers^[6] had an average of 7 days with range of 5-102 days. 15.4 % pt needed tracheotomy. Anderson et al^[4] had 24 % patients who needed tracheostomy.

The most common cause of morbidity in our patients is an elevated serum creatinine in 16 patiesnts-48.48 %.and 57.57% needed higher antibiotics. 12.12% pulmonary problems. Yasa and collegues^[6] 11 % renal failure. 15.4% pulmonary problems.

Emergencies in our study contributed to 28%. Anderson et al^[4] had 29% patients who underwent emergency surgery, U Boeken and co workers 22.9%^[7] ,Yasa and collegues^[6] 22 % Redo sternotomies were 15 % of our study population, U Boeken et al^[7] had 31.7% patients who underwent resurgeries. Yasa and co workers^[6] 31.9%. Wong et al 40 %^[9].

The average stay has been 5.76 days ± 3.26 days and the average duration of hospital stay has been 9.59 days ±2.83 days Yasa and co workers^[6] had icu stay of 7 days and 16 days. Hospital survival in our series is 87% which is a better outcome compared to Anderson and collegues^[4] whose study reported survival of 76 %. And Udo Boeken^[7] 72%. Wong and colleagues^[9] 84 %. Further, Anderson and co workers^[4] in their paper on analysis of multiple studies with open chest management have quoted a survival advantage due to open chest between 48-80%.   

We would like to postulate that open chest management and delayed sternal closure is a safe and effective method to employ in post cardio-pulmonary bypass patients who have complications in the form of Bleeding, Need for high inotropes, Persistent arrhythmias and Myocardial oedema. Delayed closure can be achieved in most patients with low incidence of morbidity and mortality. Due to advances in techniques and icu care and strict antibiotic protocol, there is low incidence of sternal instability. We propose that cardiac surgeons keep this tool in mind when dealing with post cardiotomy unstable patients. And be familiar with the care needed and post operative management of such patients. Delayed sternal closure is one of the best opportunities that can be given to provide survival advantage in post cardiotomy instability.

The main aim of this study is to demonstrate the survivial advantage that delayed sternal closure provides in unstable post cardiotomy patients. Patients who were stable and had an uncomplicated intra operative and post operative course were not a part of this study. So all cases considered in this study were high risk cases and had variable hemodynamics with high need for inotropes.

There were many study limitations in this retrospective observational study. The most common being to determine the primary reason for delayed closure as it was multifactorial and involved a combination of causes. Since all cases included were primarily left open in the theatre, there has been no documentation if an attempt was made at closure before making the decision for leaving them open.

The duration the sternum has been kept open has been arbitrary and was affected by delays like availability of operation theatre, availability of surgeon, delay due to late shifting of case after primary surgery, emergency procedures that were managed over night and closure hence deflected.

There has been no proper documentation in cases where re explorations were done for causes other than bleeding, no documentation of, if present, superficial sternal wound infection or sterna instability.

The knowledge of these limiting factors can help make significant contributions in improving the care for patients with delayed sterna closure.

1. Riahi M, Tomatis LA, Schlosser RJ, et. al. Cardiac compression due to closure of median sternotomy in open heart surgery. Chest. 1975 January 1; 67(1):113-4.
2. Rachel Eikelboom and Michael H. Yamashita, Commentry:Delayed sternal closure-an open and not so such case. JCTVS Techniques June 2020, 80-81
3. Furnary AP, Magovern JA, Simpson KA et. al. Prolonged open sternotomy and delayed sternal closure after cardiac operations. The Annals of Thoracic Surgery. 1992 August 1; 54(2):233-239.
4. Anderson CA, Filsoufi F, Aklog L, et al. Liberal use of delayed sternal closure for post cardiotomy hemodynamic instability. The Annals of Thoracic Surgery. 2002, 73: 484-488.
5. Khosrow Hashemzadeh, Shahryar Hashemzadeh. In hospital outcomes of delayed sternal closure after open cardiac surgery. Journal of Cardiac Surgery 2009. January-February 24(1):30-33.
6. Yasa H, Lafci BB, Yilik L et. al. Delayed sternal closure:an effective procedure for life saving in open heart surgery. Anatolian Journal of Cardiology. 2010. April 1;10(2).
7. Udo Boeken, Alexander Assmann, Arash Mehdiani et. al. Open chest management after cardiac operations: outcome and timing of delayed sternal closure. European Journal of Cardio Thoracic Surgery 2011. November 40(5):1146-1150.
8. Siavash Saadat, Molly Schulthesis, Joeseph Romero et. al. Delayed sternal closure in cardiac surgery. Journal of Clinical and Experimental Cardiology. 2015. 6:6.
9. Joshua K. Wong, Devang K. Joshi, Amber L. Melvin et.al. Early and later outcomes with prolonged open chest management after cardiac surgery. The Journal of Thoracic and Cardiovascular Surgery 2017.September 154(3):915-924.
10. Gielchinky I, Parsonnet V, Krishnan B, Silidker M, Abel RM. Delayed sternal closure following open-heart operation. Ann Thorac Surg 1981; 32: 273-7
11. Timo Koponen, Johanna Karttunen, Tadeusz Musialowicz et.al. Vasoactive-inotropic score and the prediction of morbidity and mortality after cardiac surgery. Br.J. Anaesth. 2019 Apr;122(4):428-436.