Deep sternal wound infection (DSWI), also called mediastinitis, is a serious complication after median sternotomy with an incidence of 1% to 5%. While superficial sternal wound infections (SSWI) involve the skin, subcutaneous tissue, and pectoralis fascia only and have much less mortality (0.5% to 9%), DSWI involves retrosternal space, prolongs the hospital stay by an average of 20 days, and is associated with a mortality of 10% to 47% which is double the mortality of those without mediastinitis [1] [2]. The incidence of DSWI is particularly high in the presence of diabetes mellitus (DM), smoking history, chronic obstructive pulmonary disease (COPD), osteoporosis and obesity [1] [2] [3] [4] Prolonged stay in the intensive care unit (ICU), use of assistive devices and reoperation boost the incidence as well. Coronary artery bypass grafting (CABG) is associated with a higher rate of sternal wound infections compared with other surgeries performed through the same surgical approach. Moreover, the technique used in harvesting the internal mammary artery (IMA) for CABG was found to influence the rate of sternal wound infections [5]. When the artery is dissected along the accompanying veins, fascia, adipose tissue and lymphatics (pedicled harvest), the sternal blood flow is decreased by up to 90%, thus increasing the rate of sternal wound infection. In contrast, dissecting the artery free from the surrounding tissues (skeletonized technique) has been shown to preserve the blood supply of the sternum and thereby reduce the rate of sternal wound infections [1]. DSWI is a complication greatly influenced by the surgical technique and can be reduced by a shorter operation and perfusion time and lesser use of electrocautery. On the other hand, shaving with razors, the use of bone wax, reoperation for bleeding, and sternal rewiring are some surgical risk factors that increase the likelihood of this complication [1]. The diagnosis of DSWI could be based on the presence of a group of clinical features such as erythema, fever, drainage and unstable sternum, although a low-grade fever may be the only presentation. According to Pairolero, median sternotomy wound infection could occur within the first week (Type I), or the 2nd to 4th week (Type II) or months to years after surgery (Type III). Most instances of DSWI are of Type II [1]. Unlike SSWI, which is completely resolved with intravenous (IV) antibiotics and local wound care, DSWI is more difficult to cure and requires a much more aggressive treatment regimen. The previous treatment options of DSWI have included closed suction and continuous irrigation while currently, surgical debridement, vacuum-assisted closure (VAC) therapy, flap coverage, and sternal plating are added options [1] [6]. Surgical treatment for angina pectoris was first proposed in 1899. Decades of experimental surgery for coronary artery disease (CAD) finally led to the introduction of CABG in 1964 [7]. Median sternotomy, first proposed by Milton in 1897 [8], was utilized in open heart operation in [3] Iraq for the first time in 1964 by Prof. Yousif D Al-Namaan and his mate Prof. Muayyad M Al-Omeri, cardiothoracic pioneers. Trials of CABG were performed a few years later by Prof. Al-Omeri at the former Republic Hospital (later called Baghdad Medical City), while a modern-time CABG was started by Dr. Najih Al-Asadi (FRCS, a Cardiothoracic & Vascular Surgeon) in 1989 at Al-Rasheed Military Hospital, Baghdad [9]. Despite the seriousness of DSWI after CABG, no study has addressed this problem in our country so far. The current study was conducted in a major tertiary Iraqi cardiac surgical center in order to assess the incidence, clinical and microbial characteristics, perioperative factors, and the outcome of surgical and conservative treatment of this devastating complication in view of the published literature.

From January 2014 to January 2016, 520 patients underwent CABG in the Iraqi Center for Heart Diseases (ICHD). DSWI was diagnosed in 29 patients (20 males, 69% and 9 females, 31%). Pre-operative, intra-operative and post-operative patient’s risk factors and many variables were studied. A questionnaire form was created. History, patients’ charts (ICU, ward, perfusionist, anesthetic, and surgical records) and admission hospital records of the patients were studied retrospectively; physical examination, diagnosis and DSWI management were studied prospectively. The diagnosis of DSWI was based on the history, physical examination, blood investigations, radiological investigations, and CDC definition criteria for DSWI, which includes the involvement of the deep tissues beyond the skin and subcutaneous tissues, including the fascial, muscle layers, sternum and retrosternal space with or without sternal instability. Patients with DSWI following CABG combined with another procedure and those operated upon elsewhere were excluded from this prospective study. The Ethical Committee of our center approved the study protocol and written informed consent of the patients to participate in the study was obtained.

The patients were thoroughly evaluated. The studied pre-operative variables included age, gender, risk factors such as high body mass index (BMI), DM, COPD and renal impairment, cardiac variables such as LVEF%, number of diseased vessels, type of angina (stable vs. unstable), congestive heart failure, myocardial infarction, previous coronary intervention and pre-operative use of aspirin and low molecular weight heparin (LMWH). Moreover, operative variables such as type of surgery (elective vs. urgent, on-pump vs. off-pump), durations of surgery, cardiopulmonary bypass (CPB) and aortic cross-clamp (ACC), number of grafts and bilateral IMA harvesting were noted. Furthermore, postoperative events such as the length of staying in the ICU and ward, re-exploration for bleeding/tamponade, stroke, coma for >24 hrs and renal impairment were documented. The type of treatment offered to patients with DSWI and death occurring within 30 days of diagnosis of sternal wound infections were reported. All procedures were performed by the same surgical and anesthetic teams. Median sternotomy incision and closure were done according to the standard technique. The number of grafts was dictated by the angiographic and intra-operative findings. Conduits were either the internal mammary artery (IMA) or the great saphenous vein. Most CABG procedures were performed under CPB with topical and central cooling, cross-clamping of the aorta and cardioplegic arrest of the heart, while the off-pump technique was occasionally used. At the end of surgery, the wounds were cleaned with Povidone-Iodine and covered with a 30 cm adhesive gauze plaster, which was kept for 2 days post-operatively.

Parenteral antibiotics (3rd generation cephalosporin and/or penicillin + aminoglycosides) were routinely given for 3 - 5 days post-operatively and then switched to oral antibiotics for 5 days if the patients had an uneventful recovery. Patients with a smooth postoperative course usually stayed for 48 hours in the ICU while those with adverse events stayed longer. The total duration of patients’ stay in the hospital was affected by the presence of the sternal wound infection and other comorbidities such as arrhythmias, myocardial ischemia, renal impairment, cerebrovascular 4 accident, and bleeding. Upon discharge from the hospital, patients prone to sternal wound infection and sternal instability (BMI > 30 kg/m², COPD, DM, age > 75 years) received a thoracic vest for 4 - 6 weeks. DSWI was defined according to the guidelines from the US Centers for Disease Control and Prevention (CDC) for post-sternotomy mediastinitis [10], which includes the involvement of the deep tissues beyond the skin and subcutaneous tissues including the fascial, muscle layers, sternum and retrosternal space with or without sternal instability. DSWIs require the presence of one of the following criteria: 1) An organism isolated from a culture of mediastinal tissue or fluid. 2) Evidence of mediastinitis seen during operation; or histopathological examination. 3) At least one of the following symptoms and/or signs with no other recognized cause: chest pain, sternal instability or fever (>38˚C) in combination with either purulent discharge from the mediastinum or an organism isolated from blood culture or culture of mediastinal drainage, or mediastinal widening on chest radiography [1]. Whenever possible, culture verification of the sternal wound infection was attempted by taking wound swabs or wound drainage for culture and sensitivity tests.

The variables of this study were divided into:

1) Demographic.

2) Pre-operative variables include Age, Gender, BMI, DM, HTN, chronic lung disease, renal disease.

3) Pre-operative cardiac variables include LVEF, NO. of the diseased vessels, angina whether stable or unstable, congestive heart failure, myocardial infarction, previous coronary intervention, pre-operative use of anticoagulants, and continuation of aspirin preoperatively. Discontinuation of the anti-platelets was 5 days prior to surgery except for emergency cases.

4) Operative variables including Status of the procedure whether elective or urgent, off or on pump CABG, CPB time, No. of the grafts, bilateral LIMA harvesting, ACC time, and time of surgery.

5) Post-operative variables include ICU stay, ward stay, re-exploration for bleeding/tamponade, stroke, continuous coma for >24 hrs, and renal impairment.

6) Lines of management and early mortality variables including drainage, debridement and wound closure, drainage, debridement and sternal re-wiring, drainage, debridement, sternal re-wiring and closed irrigation, drainage, debridement, resection of the bone, debridement, VAC, steel wire(s) removal, early mortality within 30 days of the diagnosis of the DSWI.

Regarding the surgical technique, the standard approach was median sternotomy. By using a standard pneumatic sternal hand held saw for all of the patients. Patient preparations: The patient lies in a supine position with the arms secured at the sides. The body hair is shaved the night before the procedure, 1 g of 3^rd generation cephalosporin or 1 g vancomycin antibiotic prophylaxis is given intravenously within 30 minutes of the incision and the patient is draped according to the institutional protocol with disposable sterile towels covering the skin.

Incision: The incision routinely was done by a median vertical line between the sternal notch and the tip of the xiphoid process with 21 or 22 blade knives. The interclavicular ligament has to be carefully divided followed by digital dissection of the rear surface of the sternum from the underlying sternoclavicular ligament. The xiphoid is severed from the underlying tissue of the diaphragm. The midline is identified by palpating the intercostal spaces and the sternochondral junctions at both sides of the sternum. Osteotomy is performed from above downwards. Bleeding is controlled with pinpoint cautery to avoid continuous blood loss during surgery. The use of bone wax to seal the bone marrow. The number of grafts is decided based on the echocardiographic, angiographic findings of each patient and the intra-operative findings. Conduits used were the harvested internal thoracic artery and long saphenous vein; in 19 patients of the 520 patients bilateral LIMA was used. Off-pump CABG was used in 3 patients out of the 29 and the rest were treated under cardiopulmonary bypass with topical and central cooling, cross-clamping of the aorta and cardioplegic arrest of the heart. Sternal closure: Chest tubes after completion of the cardiac procedure, LIMA bed hemostasis is checked. Mediastinal and/or pleural tubes (28,30F) are placed through stab incisions in the epigastrium. A towel is placed between the heart and the sternal edges for protection, placing the drains below the fascia of the rectus muscle. The stab incisions for the mediastinal drains are made in the epigastrium. Five to eight stainless steel wires are used for closure (either singular or figure of eight). Two wire sutures are placed around the manubrium and four are usually placed around the edges of the body of the sternum. The wires are usually either placed parasternal or through the sternal bone. After all, the wires have been set; the towel is removed carefully while lifting the wires upwards. Before closure, check both retro-sternal halves to rule out bleeding. After proper approximation, the wires are loosely twisted and cut. Then, the ends are twisted further until the sternal halves are tightly re-approximated. The twisted ends must not be too long and usually being buried entirely into the presternal tissue especially in very thin patients. The pectoral fascia is closed with one line of running braided polyglactin absorbable suture followed by a second line of the same type of suture for the subcutaneous tissue. The skin is closed according to the surgeon's preference either with an absorbable or non-absorbable subcuticular running suture or with clips. The skin closure technique was variable. Wound care: Cleaning the wound with iodine and covering it with a 30cm adhesive guaze plaster, which will not be removed until 2 days post-operatively. Sternal care and post-operative period: All the patients with risk for sternal wound infection and instability (body mass index > 30, chronic obstructive lung disease, bilateral mammary harvesting, >75 years of age and diabetes) received a thoracic vest for stabilization for 4 - 6 weeks. I.V antibiotics are routinely kept for 3 - 5 days post-operatively, and then oral antibiotics are to be prescribed for 5 days after discharge from the hospital in the uneventful course of the post-operative period. I.V antibiotics that are used are usually 3^rd generation cephalosporins and /or penicillin group antibiotics, with aminoglycosides. The length of ICU stay in average was (48 hrs) for each patient with uneventful post-operative period unless there was an adverse event that required to keep the patient in the ICU. Total hospitalization period post-operatively in average was (seven days) this was affected by complications that necessitated keeping the patient in the hospital, like the following morbidities which includes infection, arrhythmias, myocardial ischemia, renal impairment, cerebro-vascular accident, and bleeding.

Patients of DSWI in this study received one of the following treatments:

1) Drainage, debridement and wound closure;

2) Drainage, debridement and sternal re-wiring;

3) Drainage, debridement, sternal re-wiring and closed irrigation;

4) Drainage, debridement and resection of the bone;

5) Debridement, VAC and steel wire (s) removal.

Each patient with DSWI was followed up for 6 months from the onset of diagnosis of DSWI. Statistical analysis was done using the Excel Sheet of Microsoft office 10. The data were expressed as mean ± SD, ranges, numbers and ratios. P-value was calculated using the chi-square test, Odd’s ratio, and the 2-way contingency table analysis formulas. P value < 0.05 was considered statistically significant.

Throughout the study, 520 patients underwent CABG surgery; 425 males (81.7%) and 95 females (18.6%) for varied indications. Twenty nine (29) out of 520 patients developed DSWI diagnosed based on the CDC definition criteria for DSWI, including nine (31.0%) female patients and 20 (69.0%) male patients. Males were significantly more frequently involved than females (p = 0.0001) with a male to female ratio of 20/9 (2.2:1). Male gender was an independent risk factor for DSWI. The age distribution of the patients is shown in Table 1.

The age of the studied patients ranged between 38 and 75 years, with a mean of 58.1 ± 7.3. Most (n = 25, 86.2%) patients were in the 6th and 7th decades of their lives.

The clinical characteristics and preoperative variables are shown in Tables 2-4. The BMI ranged between 22 and 37.3 with a mean of 27.9 ± 3.4 kg/m² and it