Vacuum-assisted Wound Closure with Instillation Followed by Nitinol Clips Application to Treat Deep Sternal Wound Infections After Cardiac Surgery: Evolution of a Two-step Approach

DSWI, also known as mediastinitis, is a serious complication of cardiac surgery. It is associated with significant morbidity and mortality, prolonged hospital stay with the need for multiple surgical revisions, and increased cost of treatment. The incidence of DSWI ranges from 0.2% to 3%,¹ with a mortality rate of up to 47%.¹ At Policlinico Umberto I, Rome, Italy, the incidence of DSWI is 1.85% per year. The most commonly used classification system of mediastinitis is that of El Oakley and Wright,² as shown in Table 1, which is based on the presence of risk factors, the time of first presentation, and failed therapeutic trials. Throughout the years, the terms “mediastinitis” and “DSWI” have mostly been used as synonyms to differentiate them from superficial sternal wound infections, which are limited to the skin, subcutaneous tissue, or pectoralis muscle fascia.^3,4 In truth, the term DSWI refers to tissues beyond the fascia, including deep incisional infections and mediastinitis, with osteomyelitis and possible involvement of tissues behind the retrosternal space.^3,4 Moreover, the CDC described specific criteria to define mediastinitis, which is characterized by the presence of microbial organisms identified from tissue or fluid samples, evidence of mediastinitis on gross anatomic or histopathologic exam, fever (>38.0° C), chest pain or sternal instability, and a purulent drainage from mediastinal area or mediastinal widening on imaging.^3,4

Several risk factors for the development of DSWI have been reported,⁵ including obesity, diabetes, advanced age, female sex, use of BIMA grafts, surgical revision and redo surgery, prolonged hospital stay, and emergency surgery.^6-8 Preoperative screening for microorganisms, antibiotic prophylaxis, accurate antiseptic skin preparation, and meticulous sterile and surgical technique, as well as control of modifiable risk factors, can help prevent DSWI.⁹

Diagnosis of DSWI is based on clinical and laboratory parameters. Sternal instability, cutaneous erythema and discharge, retrosternal pain or swelling, fever, and increased values of inflammatory markers are strongly suspicious for sternal wound infection in the postoperative period.^1,10 An example of the appearance of DSWI is shown in Figure 1.

Contrast-enhanced CT is the imaging tool most commonly used to confirm the diagnosis of DSWI and demonstrate the extent of infection. At the time of surgical revision, tissue samples can be obtained and swabs performed to identify the responsible microorganism and establish appropriate antibiotic therapy.^1,9,10 The microorganisms most commonly identified are Staphylococcus aureus and coagulase-negative staphylococci, Pseudomonas aeruginosa, Klebsiella pneumoniae, and Enterobacter species. Fungi such as Candida albicans are less commonly identified.^11-13 Typically, empirical antibiotic therapy is initiated as soon as DSWI is observed and is modified after culture results are available.

Historically, various methods have been used to manage post-sternotomy DSWI. Treatment generally consists of wound debridement, sterilization, and subsequent primary or delayed closure. Open wound dressings, open or closed catheter irrigation, or negative pressure wound therapy can be used prior to sternal closure to ensure sterilization and promote wound healing. Final closure can be achieved with the use of soft tissue flaps, sternal wires (with or without the modified Robicsek technique), sternal plates, and, more recently, nitinol clips.^14,15,16

Since its introduction in 1999 for the treatment of DSWI, successful use of vacuum-assisted closure for sternal dehiscence has provided good outcomes in this patient population. This method involves application of negative pressure, which offers improved perfusion and continuous aspiration of exudate, and it promotes tissue granulation and approximation of sternal halves.^16,17 The system is composed of various foam dressings, sterile adhesive covering, aspiration tubes, and the negative pressure wound therapy unit. The more recently devised V.A.C. TheraFlo (3M+KCI), hereafter referred to as vacuum-assisted wound closure with instillation, adds intermittent automated instillation of irrigation fluids, thus optimizing wound sterilization and further reducing the rate of microorganism growth.^16,17 Unlike closed catheter irrigation, which delivers antimicrobial solution to the mediastinal space, the vacuum-assisted closure device also sterilizes the sternal tissue since it is applied on the anterior facet of the bone.

Nitinol clips are composed of a mixture of metal that gives them thermoreactive properties; these clips are malleable and flexible at low temperatures and return to their original shape at body temperature. These clips are applied in the intercostal spaces, surrounding the anterior facet of the sternum. Thus, they are considerably less invasive than traditional steel wiring and lack the risks associated with sawing the sternum.^18-20

The choice of treatment largely depends on the patient’s overall condition, microbiological culture results, and general appearance of the wound, as well as surgeon preference and experience. Given the possible complications of DSWI, rapid diagnosis and optimization of treatment are critically important.

This case series discusses the gradual change in the authors’ treatment protocol for DSWI, from closed catheter irrigation and sternal synthesis using the modified Robicsek technique, to use of the proprietary vacuum-assisted wound closure with instillation system combined with bilateral and later unilateral pectoralis major flaps. Since 2020, nitinol clips have been used for sternal closure.

The records of 34 patients with post-operative DSWI who were hospitalized at Policlinico Umberto I between January 2012 and December 2020 were retrospectively reviewed. Patients with superficial wound infections, as defined by the CDC, were not included. Patient demographics are summarized in Table 2.

The mean patient age was 66 years. Eleven patients (32%) underwent CABG, 13 (38%) underwent valve replacement surgery, 7 (21%) underwent combined procedures (CABG plus valve replacement), and 3 (9%) underwent surgery for other reasons. Emergency surgery was performed in 6 patients (18%) and redo procedures in 5 (15%). Among the patients who underwent CABG, 5 (15%) underwent BIMA harvesting. A total of 3 patients underwent surgical re-exploration for bleeding (9%). Most patients had comorbidities for the development of DSWI, including obesity (13 patients [38%]), diabetes (12 patients [35%]), congestive heart failure (7 patients [21%]), COPD (6 patients [18%]), and renal impairment (4 patients [12%]), as defined by a serum creatinine level >1.5 mg/dL.

According to protocol, rectal and nasal swabs were performed preoperatively on each patient. Meticulous aseptic technique and sterile procedures were used by the surgical, anesthesia, and perfusion teams. All patients received prophylactic administration of cefazolin 60 minutes before the initial skin incision. Cefazolin was terminated on the second postoperative day.

At each surgical revision wound debridement was performed, with removal of infected material and collection of samples from bone and subcutaneous tissue for biological analysis and microbial culture analysis. The most commonly isolated microorganisms were S aureus (12 swabs), S epidermidis (8 swabs), and carbapenemase-producing K pneumoniae (6 swabs). Other identified microbes were Pseudomonas (2 swabs), C albicans (3 swabs), coagulase-negative staphylococci (3 swabs), Enterobacter species (2 swabs), K oxytoca (1 swab), Streptococcus mitis (1 swab), Bacillus mycoides (1 swab), B pumilus (1 swab), and Serratia marcescens (1 swab). One case each of Corynebacterium tuberculostearicum and Stenotrophomonas maltophilia was observed.

Fifteen patients underwent closed catheter irrigation and subsequent sternal closure with the modified Robicsek technique. Three of these patients were also treated with bilateral pectoralis major muscle flaps. Sixteen patients were treated with vacuum-assisted wound closure with instillation followed by monolateral or bilateral pectoralis major muscle flaps, without direct sternal closure. Three patients received vacuum-assisted wound closure with instillation followed by sternal closure with thermoreactive nitinol clips.

In the patients treated with closed catheter irrigation, following surgical debridement and tissue sampling 2 aspiration tubes were positioned retrosternally and an irrigation tube was connected to a device containing 0.5% povidone-iodine solution. The skin was then sutured with interrupted stitches. This helped create an isolated environment for continuous irrigation of antiseptic solution. Considering the advantages offered by use of the vacuum-assisted wound closure with instillation system, the aforementioned technique was abandoned at Policlinico Umberto I in 2016 in favor of vacuum-assisted closure with instillation therapy.

In the present case series, after necrotic and infected tissue was removed, vacuum-assisted wound closure with instillation was initiated. A piece of white sterile polyurethane foam was sized and inserted between the 2 sternal edges, and a different, grey piece of foam was placed over the sternal stumps to cover the most superficial layers of the wound. The skin then was protected with an adhesive, transparent film to create a closed, sterile environment. A drainage tube was connected from the sternal wound to the vacuum-assisted wound closure device through a hole made in the adhesive film. Negative pressure of −125 mm Hg was applied to the system, as well as intermittent irrigation with a saline solution containing 50% sodium hypochlorite.

According to the protocol of the authors of the present case study, the vacuum-assisted wound closure dressing was changed every 5 days. Irrigation with 100% saline solution was started 12 hours before each surgical revision, and new tissue and culture samples were collected at each dressing change.

The criteria used to interrupt vacuum-assisted wound closure with instillation therapy were absence of fever, reduced C-reactive protein value, reduced absolute neutrophil count, and 2 consecutively sterile results from sternal tissue samples. After wound sterility was achieved and clinical conditions improved, sternal closure was completed using the modified Robicsek technique, pectoralis major muscle flaps, or application of nitinol clips.

For the modified Robicsek technique, 2 wires were placed parasternally on each side of the sternal bone and passed in an alternating manner anteriorly and posteriorly to the costal cartilages, from the upper sternal edge down to the xiphoid process (Figure 2).¹⁵ Each wire was then reversed and directed back towards the jugular notch and the cranial sternal edge, passing the wire anteriorly if it had been previously placed posteriorly and vice versa. The 2 ends of each wire were then tied together at the superior border of the first rib, after which a variable number of parasternal wires were passed and tied in the usual manner.¹⁵

Pectoralis major muscle flaps may be either unilateral or bilateral.¹⁴ Regardless which type was used in the present case series, flap harvest was performed by raising the pectoralis muscles from the costal and sternal insertions and leaving the humeral attachment in place, taking care to preserve the internal thoracic arteries. Dissection proceeded inferiorly in the subpectoral plane to allow bilateral advancement of the flaps without tension. Closure was performed in a multilayer fashion.

Application of nitinol clips involved insertion of 3 or 4 clips in select intercostal spaces. Electrocautery was used to create a tunnel through each space in which a clip would be positioned, taking care not to injure the internal thoracic arteries. The distance between the 2 intercostal spaces was then measured to choose the clip size. The clips were cooled with ice to make them deformable and easy to place, set on special forceps, and inserted in the intercostal spaces. Finally, the clips were heated with warm water to return them to their original consistency. A chest radiograph showing nitinol clips in situ is seen in Figure 3.

The 34 patients were divided into 2 groups based on type of treatment for DSWI (Table 3). Group A comprised 15 patients who underwent closed catheter irrigation, and group B comprised 19 patients treated with vacuum-assisted wound closure with instillation, regardless of sternal closure technique used.

In group A, the mean patient age was 69 years, with a male to female ratio of 11:4, and in group B the mean patient age was 63 years, with a male to female ratio of 16:3. Obesity was more prevalent in group B than in group A (27% vs 47%), as was congestive heart failure (20% vs 21%). COPD, diabetes, and renal failure were more prevalent in group A (20% vs 16%, 40% vs 32%, and 13% vs 10%, respectively). BIMA harvesting was performed only in group B patients (0% vs 26%), as was emergency surgery (0% vs 32%) and re-exploration for bleeding (0% vs 16%). Redo surgery was more prevalent in group B (13% vs 16%).

Although risk factors were more prevalent in group B than in group A, there was no mortality in group B, whereas 4 patients in group A died (27%). Mean hospital stay was 7 days shorter for group B compared with group A (75 days vs 68 days). Because of the small sample size, this difference is not statistically significant.

DSWI is a dreaded complication of open-heart surgery. Elimination of the infectious agent and debridement of affected tissue are the cornerstones of treatment. The introduction of vacuum-assisted wound closure with instillation has considerably affected outcomes in a positive manner.^16,17 In the present case series, there was no mortality in the group treated with vacuum-assisted wound closure with instillation, and wound healing was achieved in all patients. The authors of the present study believe that the use of this device could result in improved outcomes, because intermittent instillation of irrigation fluids could help with faster wound decontamination, in addition to all the benefits associated with vacuum-assisted wound closure.

Sternal rewiring can be more complicated following vacuum-assisted wound closure with instillation, because mediastinal structures can become adherent to the posterior surface of the sternum. Bilateral and unilateral pectoralis major muscle flaps were used with success in the patients included in this article; however, such flaps are not ideal for covering the lower one-third of the sternum because of the risk of instability. Additionally, they are associated with the formation of seroma and hematoma, which can potentially require surgical revision. Reconstruction with muscle flaps as well as conventional rewiring can be avoided with the use of nitinol clips in patients with bone tissue of sufficient quality.¹⁸ Use of nitinol clips in the present case series avoided the complications associated with muscle flaps (ie, seroma, hematoma). In addition to being less invasive, with the use of nitinol clips the entirety of the sternum can be approximated and, in the case of recurrence, muscle flaps remain a viable option.

This case series is limited by the small sample size, which is insufficient to perform a statistical analysis. A larger sample size is needed to confirm the results reported herein.

Based on the experience of the authors of this case series, the combination of vacuum-assisted wound closure with instillation and, as possible, sternal closure with nitinol clips appears to be a safe and effective treatment approach in patients with DSWI. Decreased mortality was achieved in the present case series, even in patients with concomitant risk factors such as BIMA grafting or re-exploration for bleeding, and mean hospital stay was reduced. Vacuum-assisted wound closure with instillation represents a step forward in the treatment of DSWI and an evolution from catheter irrigation, sternal wiring, and pectoralis flaps, providing improved patient care following cardiac surgery.