CLINICAL PAPER
Endoscope-Assisted Pectoralis Major-Rectus Abdominis Bipedicle Muscle Flap for the Treatment of Poststernotomy Mediastinitis Chieh Chou, MD,*† Ming-jer Tasi, MD,*† Yen-Ting Sheen, MD,*† Shu-Hung Huang, MD,*†‡ Tung-Ying Hsieh, MD,*† Chih-Hau Chang, MD,*† Chung-Sheng Lai, MD, PhD,*† Kao-Ping Chang, MD, PhD,*† Sin-Daw Lin, MD, PhD,*† and Su-Shin Lee, MD*†‡ Introduction: Various management strategies have been reported for sternal wound care; however, they exhibit limited effectiveness or are associated with severe complications. Furthermore, it is difficult for the standard pectoralis major (PM) muscle advance flap to reach the lower third of the sternum. This article examines using the PM-rectus abdominis (RA) bipedicle muscle flap to treat lowerthird deep sternal wound infection. Methods: The outcomes of patients who received a PM-RA bipedicle muscle flap harvest at our institution between 1996 and 2014 were reviewed. The method involves performing a subfascial and subperiosteal dissection of the PM to elevate the muscle flap. Blunt dissection may be performed carefully under an endoscope. Endoscope visualization enables us to identify the critical structures lateral to the PM muscle. In addition, the connective tissue to the RA muscle was preserved. Continuity was carefully preserved from the pectoral-thoracoepigastric fascia to the anterior rectus sheath. The flap could then be transposed to fill the lower-third sternal tissue defect with ease. Results: A total of 12 patients, with a mean age of 71 years (45–89 years), were treated using an endoscope-assisted PM-RA bipedicle muscle flap harvest. Wound microbiology of the 12 patients revealed that 3 patients had methicillinresistant Staphylococcus aureus, 4 had S. aureus, 1 had coagulase-negative Staphylococcus, 1 had Escherichia coli, 1 had Pseudomonas aeruginosa, 1 had Mycobacterium tuberculosis, and 1 had a mixed growth of organisms. One instance of recurrent sternal infection was identified among the patients. Moreover, 1 patient died from heart failure 5 weeks after surgery, but the coverage of the sternal wound was successful. Accidental injury to the surrounding neurovascular structure of the patients was avoided, and only 10 to 15 minutes was required to divide the PM muscle. Conclusions: Performing this harvest method under endoscopic assistance has several advantages, such as preventing excess traction of the skin edge to diminish the skin slough. This method could be an effective alternative for harvesting the PM-RA bipedicle muscle flap to reconstruct the lowerthird sternal wound. Key Words: pectoralis major, rectus abdominis, deep sternal wound infection, bipedicle flap, endoscope (Ann Plast Surg 2016;76: S29–S34)
D
eep sternal wound infection (DSWI) after median sternotomy may severely affect the postoperation course including morbidity, length of hospitalization, cost, and mortality. The DSWI is a
Received October 22, 2015, and accepted for publication, after revision November 6, 2015. From the *Department of Surgery, Faculty of Medicine, College of Medicine, †Division of Plastic Surgery, Department of Surgery, Kaohsiung Medical University Hospital, and ‡Center for Stem Cell Research, Kaohsiung Medical University, Kaohsiung, Taiwan. Conflicts of interest and sources of funding: none declared. Reprints: Su-Shin Lee, MD, Division of Plastic Surgery, Kaohsiung Medical University Hospital, 19 Fl, No.100, Tz-You 1st Road, Kaohsiung 807, Taiwan. E-mail: [email protected], [email protected]. Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0148-7043/16/7603–S029 DOI: 10.1097/SAP.0000000000000693
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
devastating complication after median sternotomy. The worldwide incidence of sternal wound infection reported in the literature ranges between 0.7% and 2.0%,1–5 and the reported mortality rates range from 4.7% to 39%.1–4,6 Several studies have evaluated the risk factors for sternal wound infection. Diabetes mellitus, obesity, end-stage renal disease, chronic obstructive pulmonary disease, expanding use of bilateral mammary arteries, and prolonged mechanical ventilation have been found to be significant independent predictors of mortality despite muscle flap reconstruction.1,7,8 The management of postoperative DSWI varies widely on the basis of the discretion of the cardiovascular surgeon and plastic surgeon.6 Various management strategies have been reported for sternal wound care, including debridement plus rewiring, open-wound care, continuous irrigation, Redon tube closed irrigation, open packing followed by delayed closure, negative pressure wound therapy (NPWT), and different types of flap reconstructions.3,9–13 However, these reconstructive procedures are reported to have limited effectiveness or are associated with complications.8 The gold standards for DSWI management are adequately debriding the infected tissue and covering the defect with well-vascularized tissue. Using the pectoralis major (PM) muscle flap reduces the difficulty of treating wounds in the upper two-thirds of the sternum. However, it is difficult for the standard PM advancement flap to reach the lower third of the sternum. Lee et al9 reported 5 reconstruction methods for lower-third sternal osteomyelitis management1: latissimus dorsi muscle with fasciocutaneous extension flap,2 tripedicle PM musculocutaneous flap,3 PM muscle with rectus abdominis (RA) muscle flap,4 PM muscle with omentum flap, and5 free vastus lateralis muscle flap with skin grafting. However, each flap carries unavoidable limitations and setbacks. Solomon et al14 first described the bipedicle PM-RA muscle (RAM) flap for lower sternal wound treatment. This bipedicle flap can reach the lower third of a sternal wound with limited abdominal wall morbidity. The PM-RAM bipedicle flap is based on the thoracoacromial artery superiorly and deep inferior epigastric vessels inferiorly. Even when the ipsilateral internal mammary artery has been used for coronary artery bypass surgery, this flap can still be harvested. The advantages of this flap are that it is easy to harvest, has sufficient volume, and its pedicle is unrelated to the internal mammary artery. However, the reported drawback of this flap is that it requires a long operative period.11 In this presented series, we used the endoscope to harvest the PM-RAM bipedicle flap. The endoscope provided magnified viewing and a proper light source, which enabled us to harvest the flap through a limited incision and reduced the risk of accidental trauma to the surrounding nerves and vessels at the midaxillary line.
MATERIALS AND METHODS All the patients who received sternotomy between 1996 and 2014 at our institution were included in this study. Patients with DSWI were identified, and their charts and radiological images were reviewed and analyzed. Exclusion criteria were patients with conditions involving only the skin, subcutaneous tissue, or pectoral fascia. A total of 1880 www.annalsplasticsurgery.com
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
S29
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
Chou et al
procedures were performed during the 9-year period, and 47 patients (2.5%) were documented as having DSWI. All the patients were sorted into Oakley classification type IVa or type IVb.15 The associated morbidity, mortality, length of hospitalization, and outcomes of these patients were reviewed. All the sternal wound debridement and muscle flap reconstructions of these patients were performed by a single plastic surgeon. All necrotic tissue with the sternal wiring was removed, and the wound was then covered using NPWT. The vacuum-assisted closure system (Kinetic Concepts, Inc., San Antonio, TX) was applied to our patients. The negative pressure was maintained up to 125 mm Hg. The negative pressure dressing was changed every 48 hours, when bedside debridement and irrigation were also performed. The second-stage flap reconstruction was not performed until healthy granulation tissue was observed. Reconstruction options depend on the patient's general condition, infection site, and volume of the soft tissue defect. Of the 47 patients, 12 had DSWI involving the lower third of the sternum. The wound was eventually reconstructed using an endoscope-assisted PM-RAM bipedicle flap (Figs. 1 and 2).
FIGURE 2. After the first step, which was ultrasonic debridement, systemic antibiotic negative pressure wound therapy was provided for 2 weeks.
Endoscope-Assisted PM-RA Muscle Bipedicle Flap Harvesting
FIGURE 1. A 67-year-old woman, a victim of mitral valve stenosis and coronary artery disease status post mitral valve replace and CABG, suffered from DSWI over the lower-third sternum. CABG, coronary artery bypass surgery. S30
www.annalsplasticsurgery.com
Each patient was placed in the supine position. Subfascial and subperiosteal dissection of the PM muscle was performed to elevate the muscle flap. The lateral border of the PM muscle was then approached with the endoscope (Fig. 3). An avascular plane is found between the PM and pectoralis minor muscles; dividing its costal insertion allows the PM muscle to be elevated from the chest wall. For our patients, endoscopic visualization enabled the crucial structures lateral to the PM muscle, such as the long thoracic nerve, lateral thoracic artery, and other surrounding vascular pedicles to be identified. In addition, the tissue connected to the RAM was preserved. Continuity was carefully preserved from the pectoral-thoracoepigastric fascia to the anterior sheath of the RAM (Figs. 4 and 5, under endoscopic and gross views). This dissection extended to the lateral margin of the RAM and was undermined from its posterior sheath by blunt dissection through the loose areolar tissue plane. The superior epigastric artery was preserved. The lateral costomarginal and segmental intercostal vessels could be sacrificed. The flap could then be transposed to easily fill the lowerthird sternal tissue defect. Overall, the arterial supply of the whole bipedicle flap was based on the thoracoacromial artery superiorly and deep inferior epigastric vessels inferiorly. No chest tube was necessary. © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
Endoscope-Assisted PM-RA Bipedicle Muscle Flap
FIGURE 3. The position of the operator and assistant during endoscope-assisted PM-RAM bipedicle flap harvesting.
Suction drains were inserted below the muscle flaps. The drains were kept in place for approximately 3 days, and hospitalization extended for approximately 5 postoperative days. Abdominal binders were recommended for use for 1 month.
FIGURE 5. Gross view of the continuity of the PM muscle and RAM.
RESULTS
FIGURE 4. Endoscopic view of the continuity of the PM muscle and RAM. © 2016 Wolters Kluwer Health, Inc. All rights reserved.
A retrospective analysis of the DSWI patients over a 12-year period was performed. A total of 1880 patients received heart surgery during the study period. Overall, 47 patients were referred to the plastic surgery department for sternal wound infection management. At our facility, these patients' injuries involving the sternum were treated using one of the following methods: (1) delayed closure, 8 patients (17%); (2) PM musculocutaneous flap, 21 patients (44.7%); (3) PM muscle with RAM flap, 12 patients (25.5%); (4) PM muscle with omentum flap, 2 patients (4.2%); (5) PM turnover flap, 2 patients (4.2%); and (6) RAM flap, 2 patients (4.2%). The PM-RAM bipedicle flap was harvested under an endoscope for viewing. All PM-RAM bipedicle flap was harvested in this studies were unilateral to cover the defects. The summary of the results is listed in Table 1. Among these 12 patients, the mean age was 71 years, and the male/female ratio was 7:5. The average duration between the first sternotomy procedure and noticeable DSWI was 25.5 days. The interval from the diagnosis of DSWI to the flap reconstruction was 33 days on average. Demographic data and medical history were obtained by retrospective chart review. Data that were collected included: age, sex, comorbid illnesses included renal disease stage, diabetes, chronic obstructive pulmonary disease and body mass index, defect location, defect size, type of flap, flap survival, www.annalsplasticsurgery.com
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
S31
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
Chou et al
TABLE 1. Results of the Different Reconstruction Methods Reconstruction Methods
No. (Percentage)
Morbidity Mortality
Delayed closure 8 patients (17%) PM musculocutaneous flap 21 patients (44.7%) PM muscle with RAM flap 12 patients (25.5%) PM muscle with omentum flap 2 patients (4.2%) PM turnover flap 2 patients (4.2%) RAM flap 2 patients (4.2%) Total 47
3* 1† 0 0 0 0 4
0 0 1‡ 0 0 0 1
*Expired of heart failure during hospital stay. †Died of sepsis, mediastinitis related. ‡Partial wound dehiscence with organism of M. tuberculosis.
interval from cardiac surgery to symptom development of DSWI, interval from the diagnosis of infection to flap operation, and patient survival. Median follow-up interval was 10.5 months (range, from 6 to 18 months). Patient characteristic and demographic data are described in Table 2. The postsurgical courses were uneventful except that 1 patient developed wound tuberculosis infection. Partial wound dehiscence and fascial attenuation were noted; this patient's wound then stabilized and healed after the systemic plus local antituberculous treatment. The wound microbiology among these 12 patients demonstrated that 3 patients had methicillin-resistant Staphylococcus aureus (MRSA), 4 had S. aureus, 1 had coagulase-negative Staphylococcus, 1 had Escherichia coli, 1 had Pseudomonas aeruginosa, 1 had Mycobacterium tuberculosis, and 1 had a mixed growth of organisms. One instance of recurrent sternal infection was reported, and 1 patient expired because of heart failure 5 weeks after surgery. However, the coverage of the sternal wound was still successful in this patient.
FIGURE 6. A long incision wound and excess skin traction are needed to use the conventional PM flap harvesting method.
DISCUSSION Major infection of sternal wounds after open-heart surgery is rare but carries a 50% mortality risk in early series.1–4 Although the incidence of mediastinal wound infection in patients undergoing median sternotomy for cardiopulmonary bypass is less than 1%,15 the associated morbidity, mortality, and cost remain unacceptably high. El Oakley and Wright15 reported that sternal wound debridement and the removal of foreign materials are essential steps as well as asserted the current consensus regarding the ideal treatment options for a complicated median sternotomy wound. The DSWI, or mediastinitis, is classified into 4 subtypes according to the time of first presentation, the presence or
TABLE 2. Patient's Demographic Data
No Sex Age, y
1
M
45
2 3
M F
89 45
4 5 6 7 8 9
M F F F F M
75 56 71 76 85 77
10 M 11 M
83 74
12 M
76
Preoperative Diagnosis
Comorbidities
Diabetics, Chronic CAOD renal failure Diabetics, BMI > 35 CAOD Diabetics, Chronic renal CAOD failure, COPD, BMI > 35 Chronic renal failure CAOD Chronic renal failure, COPD CAOD, valve disease Diabetics CAOD Chronic renal failure Valve disease Diabetics, COPD CAOD Chronic renal failure, CAOD COPD. BMI > 35 Diabetics, COPD CAOD Diabetics, Chronic Valve disease renal failure COPD CAOD
Interval From Cardiac Interval From Surgery to Symptom the Diagnosis Flap Flap Survival Development of of Infection Follow-Up Patient Size, cm (S/F) DSWI (d) to op (d) Interval, m Survival
15 3
S
24
33
6
Y
12 5 16 6
S S
26 17
24 35
6.5 12
Y Y
20 5 S(partial loss) 20 6 S 25 5 S 16 6 S 18 7 S 20 8 S
28 24 34 33 32 27
43 43 33 32 36 24
12 12 12 6 12 18
Y N Y Y Y Y
15 6 16 4
S S
24 22
24 42
12 12
Y Y
16 5
S
15
27
18
Y
M, male; F, female; CAOD, Coronary artery occlusive disease; COPD, chronic obstructive pulmonary disease; BMI, body mass index.
S32
www.annalsplasticsurgery.com
© 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
FIGURE 7. Complication of conventional PM muscle flap harvesting and skin slough.
absence of risk factors, and whether previous attempts to treat the condition have failed. Closed mediastinal irrigation can be successful in patients with type I mediastinitis, whereas surgical debridement plus major reconstructive operation are probably the most effective treatment for patients with mediastinitis types II to V. However, the first step in treating a DSWI is adequate debridement, including bone and cartilage, to initially control the infection. At our institute, ultrasonic debridement was performed for most of the patients. The advantages were efficiency, rapidity, precision, less bleeding, and thorough removal of debris without the sacrifice of viable sternal wound bed tissue.16 After debridement, we subsequently applied NPWT. The reported advantages of NPWT for intractable sternal wound infections include1 protecting the underlying mediasternal structure from infection,2 permitting delayed sternal closure to avoid cardiac compression-induced compromised cardiopulmonary function,3 ensuring the possibility of repeated wound
Endoscope-Assisted PM-RA Bipedicle Muscle Flap
inspection and bedside debridement,4 ensuring the cost-effectiveness of wound care, and5 providing an option to promote sternal wound secondary healing for patients in poor physical condition.12 The PM muscle can serve as a pedicle flap alone or as a myocutaneous flap. It can tolerate and survive an infected or inflamed wound bed. In addition, the muscle provides adequate tissue volume to fill the tissue defects of a DSWI patient. Additionally, the RAM flap can fit ideally for the lower third of the sternal defect. The volume of this muscle flap also provides adequate tissue coverage. Because it is difficult for the PM advance flap alone to reach the lower third of the sternal defect, the combination of the PM and RA bipedicle flap was alternatively chosen for lower-third DSWI. The PM-RA bipedicle muscle flap was first described in 1998 by Solomon et al,14 who also characterized the dissection of this flap as rapid, and the complication rate of dissecting this flap compared favorably with that of other methods. Roh et al17 also explained that the PM–RA bipedicle flap is a practical and efficacious method in the reconstruction of the anterior chest wall defect. In Roh's series, not only the lower third but also the whole sternum was covered by this flap. He concluded that the PM-RAM bipedicle muscle flap could fill the defect with sufficient volume. A skin slough was reported as a common complication of flap harvesting in Roh's series, and a large area of undermined skin and cutaneous tissue as well as concurrent types of disease such as diabetes may easily occur as a result of a skin slough. Otherwise, excess skin traction may also lead to these self-limited problems. Surgeons should be especially aware that a severe (deep) skin slough may occur as a result of using an intraoperative skin traction device (Figs. 6 and 7). These 2 articles discussing a PM-RAM bipedicle flap for DSWI reconstruction are summarized in Table 3. The minimally invasive concept was introduced in the early 20th century, and the endoscope technique has undergone a major breakthrough over the past 2 decades. Endoscopic surgery involves inserting scopes through small incisions or natural body openings to diagnose and treat the disease. Under endoscope viewing, we harvested the PM muscle for sternal wound reconstruction. However, by using the PM muscle flap alone, we had difficulty reaching the lower third sternal wound. Therefore, the PM-RAM bipedicle flap was used for this difficult location defect. In our experience, a PM muscle harvested under endoscope viewing has the following advantages: (1) harvests through a limited incision wound can prevent excess traction of the skin edge and diminish the skin slough; (2) accidental injury to the surrounding neurovascular structure is prevented (in this procedure, endoscopic viewing enables the lateral thoracic artery and long thoracic nerve to be identified before cutting); (3) the PM muscle is cut along the anterior axillary line through clear visualization; and (4) only 10 to 15 minutes are required to divide the PM muscle under endoscopic viewing. Shortening the operation time is critical because DSWI patients typically have other concurrent types of disease such as diabetes, hypertension, heart failure, and pulmonary failure that increase the anesthesia risk. The reported recurrence rates after flap reconstruction varied. One recurrent sternal infection was reported among these patients. A
TABLE 3. Two Relevant Articles of a PM-RA Bipedicle Flap in DSWI Reconstruction Patient Number /Duration
Complication
Mortality
Solomon et al (1998) Plast Reconstr Surg
42 patients/8-y period
There were no deaths in this series
Roh et al (2008) J Thorac Cardiovasc Surg
27 patients/5-y period
Flap failure, 1; persistent infection, 1; pneumothorax, 1; hernia, 1; hematomas, 3; skin slough, 9 Recurrence, 3 Focal dehiscence, 3 Fascial attenuation, 1
© 2016 Wolters Kluwer Health, Inc. All rights reserved.
There were no deaths in this series
www.annalsplasticsurgery.com
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
S33
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
Chou et al
wound culture revealed a MRSA infection. Our recurrence rate was 8%, and we readvanced the flap after proper antibiotic treatment and surgical debridement. The patient improved during this time, and this patient's wound healed 1 month later. The main limitation of this study is its retrospective design and that the data are from a single institution, indicating that the findings may not be generalizable. The assessment results of recurrence or complications may be influenced by the varying expertise and experience of physicians. Despite these limitations, we offered a relatively easy and safe way to reconstruct a lower third sternal wound by using endoscopeassisted harvesting of the PM-RAM bipedicle flap.
CONCLUSIONS The PM-RAM bipedicle flap can be an effective lifesaving boat for lower-third DSWI. In this study, endoscope-assisted harvesting only required limited skin incision wounds, prevented excess traction of the skin edge, and allowed us to identify the surrounding structure before cutting. We could cut the muscle along the anterior axillary line in only 10 to 15 minutes. This could be an effective alternative method for harvesting the PM-RAM bipedicle flap to reconstruct a lower-third sternal wound. REFERENCES 1. Patel NV, Woznick AR, Welsh KS, et al. Predictors of mortality after muscle flap advancement for deep sternal wound infections. Plast Reconstr Surg. 2009;123: 132–138. 2. Sachithanandan A, Nanjaiah P, Nightingale P, et al. Deep sternal wound infection requiring revision surgery: impact on mid-term survival following cardiac surgery. Eur J Cardiothorac Surg. 2008;33:673–678. 3. López-Monjardin H, de-la-Peña-Salcedo A, Mendoza-Muñoz M, et al. Omentum flap versus pectoralis major flap in the treatment of mediastinitis. Plast Reconstr Surg. 1998;101:1481–1485.
S34
www.annalsplasticsurgery.com
4. Ottino G, De Paulis R, Pansini S, et al. Major sternal wound infection after openheart surgery: a multivariate analysis of risk factors in 2,579 consecutive operative procedures. Ann Thorac Surg. 1987;44:173–179. 5. Sarr MG, Gott VL, Townsend TR. Mediastinal infection after cardiac surgery. Ann Thorac Surg. 1984;38:415–423. 6. Cabbabe EB, Cabbabe SW. Immediate versus delayed one-stage sternal debridement and pectoralis muscle flap reconstruction of deep sternal wound infections. Plast Reconstr Surg. 2009;123:1490–1494. 7. Poncelet AJ, Lengele B, Delaere B, et al. Algorithm for primary closure in sternal wound infection: a single institution 10-year experience. Eur J Cardiothorac Surg. 2008;33:232–238. 8. Price J, Rubens F, Bell M. Elastic device facilitating delayed primary closure of sternal wound infection. Ann Thorac Surg. 2007;83:1162–1165. 9. Lee CH, Hsien JH, Tang YB, et al. Reconstruction for sternal osteomyelitis at the lower third of sternum. J Plast Reconstr Aesthet Surg. 2010;63:633–641. 10. Eyileten Z, Akar AR, Eryilmaz S, et al. Vacuum-assisted closure and bilateral pectoralis muscle flaps for different stages of mediastinitis after cardiac surgery. Surg Today. 2009;39:947–954. 11. Greig AV, Geh JL, Khanduja V, et al. Choice of flap for the management of deep sternal wound infection—an anatomical classification. J Plast Reconstr Aesthet Surg. 2007;60:372–378. 12. Lee SS, Lin SD, Chen HM, et al. Management of intractable sternal wound infections with topical negative pressure dressing. J Card Surg. 2005;20:218–222. 13. Sun IF, Lee SS, Chiu CC, et al. Hyperbaric oxygen therapy with topical negative pressure: an alternative treatment for the refractory sternal wound infection. J Card Surg. 2008;23:677–680. 14. Solomon MP, Granick MS. Bipedicle muscle flaps in sternal wound repair. Plast Reconstr Surg. 1998;101:356–360. 15. El Oakley RM, Wright JE. Postoperative mediastinitis: classification and management. Ann Thorac Surg. 1996;61:1030–1036. 16. Cheng LH, Stewart J, Thompson M, et al. Ultrasonic debridement of contaminated facial wounds. Br J Oral Maxillofac Surg. 2002;40:149–150. 17. Roh TS, Lee WJ, Lew DH, et al. Pectoralis major-rectus abdominis bipedicled muscle flap in the treatment of poststernotomy mediastinitis. J Thorac Cardiovasc Surg. 2008;136:618–622.
© 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Endoscope-Assisted Pectoralis Major-Rectus Abdominis Bipedicle Muscle Flap for the Treatment of Poststernotomy Mediastinitis Chieh Chou, MD,*† Ming-jer Tasi, MD,*† Yen-Ting Sheen, MD,*† Shu-Hung Huang, MD,*†‡ Tung-Ying Hsieh, MD,*† Chih-Hau Chang, MD,*† Chung-Sheng Lai, MD, PhD,*† Kao-Ping Chang, MD, PhD,*† Sin-Daw Lin, MD, PhD,*† and Su-Shin Lee, MD*†‡ Introduction: Various management strategies have been reported for sternal wound care; however, they exhibit limited effectiveness or are associated with severe complications. Furthermore, it is difficult for the standard pectoralis major (PM) muscle advance flap to reach the lower third of the sternum. This article examines using the PM-rectus abdominis (RA) bipedicle muscle flap to treat lowerthird deep sternal wound infection. Methods: The outcomes of patients who received a PM-RA bipedicle muscle flap harvest at our institution between 1996 and 2014 were reviewed. The method involves performing a subfascial and subperiosteal dissection of the PM to elevate the muscle flap. Blunt dissection may be performed carefully under an endoscope. Endoscope visualization enables us to identify the critical structures lateral to the PM muscle. In addition, the connective tissue to the RA muscle was preserved. Continuity was carefully preserved from the pectoral-thoracoepigastric fascia to the anterior rectus sheath. The flap could then be transposed to fill the lower-third sternal tissue defect with ease. Results: A total of 12 patients, with a mean age of 71 years (45–89 years), were treated using an endoscope-assisted PM-RA bipedicle muscle flap harvest. Wound microbiology of the 12 patients revealed that 3 patients had methicillinresistant Staphylococcus aureus, 4 had S. aureus, 1 had coagulase-negative Staphylococcus, 1 had Escherichia coli, 1 had Pseudomonas aeruginosa, 1 had Mycobacterium tuberculosis, and 1 had a mixed growth of organisms. One instance of recurrent sternal infection was identified among the patients. Moreover, 1 patient died from heart failure 5 weeks after surgery, but the coverage of the sternal wound was successful. Accidental injury to the surrounding neurovascular structure of the patients was avoided, and only 10 to 15 minutes was required to divide the PM muscle. Conclusions: Performing this harvest method under endoscopic assistance has several advantages, such as preventing excess traction of the skin edge to diminish the skin slough. This method could be an effective alternative for harvesting the PM-RA bipedicle muscle flap to reconstruct the lowerthird sternal wound. Key Words: pectoralis major, rectus abdominis, deep sternal wound infection, bipedicle flap, endoscope (Ann Plast Surg 2016;76: S29–S34)
D
eep sternal wound infection (DSWI) after median sternotomy may severely affect the postoperation course including morbidity, length of hospitalization, cost, and mortality. The DSWI is a
Received October 22, 2015, and accepted for publication, after revision November 6, 2015. From the *Department of Surgery, Faculty of Medicine, College of Medicine, †Division of Plastic Surgery, Department of Surgery, Kaohsiung Medical University Hospital, and ‡Center for Stem Cell Research, Kaohsiung Medical University, Kaohsiung, Taiwan. Conflicts of interest and sources of funding: none declared. Reprints: Su-Shin Lee, MD, Division of Plastic Surgery, Kaohsiung Medical University Hospital, 19 Fl, No.100, Tz-You 1st Road, Kaohsiung 807, Taiwan. E-mail: [email protected], [email protected]. Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0148-7043/16/7603–S029 DOI: 10.1097/SAP.0000000000000693
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
devastating complication after median sternotomy. The worldwide incidence of sternal wound infection reported in the literature ranges between 0.7% and 2.0%,1–5 and the reported mortality rates range from 4.7% to 39%.1–4,6 Several studies have evaluated the risk factors for sternal wound infection. Diabetes mellitus, obesity, end-stage renal disease, chronic obstructive pulmonary disease, expanding use of bilateral mammary arteries, and prolonged mechanical ventilation have been found to be significant independent predictors of mortality despite muscle flap reconstruction.1,7,8 The management of postoperative DSWI varies widely on the basis of the discretion of the cardiovascular surgeon and plastic surgeon.6 Various management strategies have been reported for sternal wound care, including debridement plus rewiring, open-wound care, continuous irrigation, Redon tube closed irrigation, open packing followed by delayed closure, negative pressure wound therapy (NPWT), and different types of flap reconstructions.3,9–13 However, these reconstructive procedures are reported to have limited effectiveness or are associated with complications.8 The gold standards for DSWI management are adequately debriding the infected tissue and covering the defect with well-vascularized tissue. Using the pectoralis major (PM) muscle flap reduces the difficulty of treating wounds in the upper two-thirds of the sternum. However, it is difficult for the standard PM advancement flap to reach the lower third of the sternum. Lee et al9 reported 5 reconstruction methods for lower-third sternal osteomyelitis management1: latissimus dorsi muscle with fasciocutaneous extension flap,2 tripedicle PM musculocutaneous flap,3 PM muscle with rectus abdominis (RA) muscle flap,4 PM muscle with omentum flap, and5 free vastus lateralis muscle flap with skin grafting. However, each flap carries unavoidable limitations and setbacks. Solomon et al14 first described the bipedicle PM-RA muscle (RAM) flap for lower sternal wound treatment. This bipedicle flap can reach the lower third of a sternal wound with limited abdominal wall morbidity. The PM-RAM bipedicle flap is based on the thoracoacromial artery superiorly and deep inferior epigastric vessels inferiorly. Even when the ipsilateral internal mammary artery has been used for coronary artery bypass surgery, this flap can still be harvested. The advantages of this flap are that it is easy to harvest, has sufficient volume, and its pedicle is unrelated to the internal mammary artery. However, the reported drawback of this flap is that it requires a long operative period.11 In this presented series, we used the endoscope to harvest the PM-RAM bipedicle flap. The endoscope provided magnified viewing and a proper light source, which enabled us to harvest the flap through a limited incision and reduced the risk of accidental trauma to the surrounding nerves and vessels at the midaxillary line.
MATERIALS AND METHODS All the patients who received sternotomy between 1996 and 2014 at our institution were included in this study. Patients with DSWI were identified, and their charts and radiological images were reviewed and analyzed. Exclusion criteria were patients with conditions involving only the skin, subcutaneous tissue, or pectoral fascia. A total of 1880 www.annalsplasticsurgery.com
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
S29
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
Chou et al
procedures were performed during the 9-year period, and 47 patients (2.5%) were documented as having DSWI. All the patients were sorted into Oakley classification type IVa or type IVb.15 The associated morbidity, mortality, length of hospitalization, and outcomes of these patients were reviewed. All the sternal wound debridement and muscle flap reconstructions of these patients were performed by a single plastic surgeon. All necrotic tissue with the sternal wiring was removed, and the wound was then covered using NPWT. The vacuum-assisted closure system (Kinetic Concepts, Inc., San Antonio, TX) was applied to our patients. The negative pressure was maintained up to 125 mm Hg. The negative pressure dressing was changed every 48 hours, when bedside debridement and irrigation were also performed. The second-stage flap reconstruction was not performed until healthy granulation tissue was observed. Reconstruction options depend on the patient's general condition, infection site, and volume of the soft tissue defect. Of the 47 patients, 12 had DSWI involving the lower third of the sternum. The wound was eventually reconstructed using an endoscope-assisted PM-RAM bipedicle flap (Figs. 1 and 2).
FIGURE 2. After the first step, which was ultrasonic debridement, systemic antibiotic negative pressure wound therapy was provided for 2 weeks.
Endoscope-Assisted PM-RA Muscle Bipedicle Flap Harvesting
FIGURE 1. A 67-year-old woman, a victim of mitral valve stenosis and coronary artery disease status post mitral valve replace and CABG, suffered from DSWI over the lower-third sternum. CABG, coronary artery bypass surgery. S30
www.annalsplasticsurgery.com
Each patient was placed in the supine position. Subfascial and subperiosteal dissection of the PM muscle was performed to elevate the muscle flap. The lateral border of the PM muscle was then approached with the endoscope (Fig. 3). An avascular plane is found between the PM and pectoralis minor muscles; dividing its costal insertion allows the PM muscle to be elevated from the chest wall. For our patients, endoscopic visualization enabled the crucial structures lateral to the PM muscle, such as the long thoracic nerve, lateral thoracic artery, and other surrounding vascular pedicles to be identified. In addition, the tissue connected to the RAM was preserved. Continuity was carefully preserved from the pectoral-thoracoepigastric fascia to the anterior sheath of the RAM (Figs. 4 and 5, under endoscopic and gross views). This dissection extended to the lateral margin of the RAM and was undermined from its posterior sheath by blunt dissection through the loose areolar tissue plane. The superior epigastric artery was preserved. The lateral costomarginal and segmental intercostal vessels could be sacrificed. The flap could then be transposed to easily fill the lowerthird sternal tissue defect. Overall, the arterial supply of the whole bipedicle flap was based on the thoracoacromial artery superiorly and deep inferior epigastric vessels inferiorly. No chest tube was necessary. © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
Endoscope-Assisted PM-RA Bipedicle Muscle Flap
FIGURE 3. The position of the operator and assistant during endoscope-assisted PM-RAM bipedicle flap harvesting.
Suction drains were inserted below the muscle flaps. The drains were kept in place for approximately 3 days, and hospitalization extended for approximately 5 postoperative days. Abdominal binders were recommended for use for 1 month.
FIGURE 5. Gross view of the continuity of the PM muscle and RAM.
RESULTS
FIGURE 4. Endoscopic view of the continuity of the PM muscle and RAM. © 2016 Wolters Kluwer Health, Inc. All rights reserved.
A retrospective analysis of the DSWI patients over a 12-year period was performed. A total of 1880 patients received heart surgery during the study period. Overall, 47 patients were referred to the plastic surgery department for sternal wound infection management. At our facility, these patients' injuries involving the sternum were treated using one of the following methods: (1) delayed closure, 8 patients (17%); (2) PM musculocutaneous flap, 21 patients (44.7%); (3) PM muscle with RAM flap, 12 patients (25.5%); (4) PM muscle with omentum flap, 2 patients (4.2%); (5) PM turnover flap, 2 patients (4.2%); and (6) RAM flap, 2 patients (4.2%). The PM-RAM bipedicle flap was harvested under an endoscope for viewing. All PM-RAM bipedicle flap was harvested in this studies were unilateral to cover the defects. The summary of the results is listed in Table 1. Among these 12 patients, the mean age was 71 years, and the male/female ratio was 7:5. The average duration between the first sternotomy procedure and noticeable DSWI was 25.5 days. The interval from the diagnosis of DSWI to the flap reconstruction was 33 days on average. Demographic data and medical history were obtained by retrospective chart review. Data that were collected included: age, sex, comorbid illnesses included renal disease stage, diabetes, chronic obstructive pulmonary disease and body mass index, defect location, defect size, type of flap, flap survival, www.annalsplasticsurgery.com
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
S31
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
Chou et al
TABLE 1. Results of the Different Reconstruction Methods Reconstruction Methods
No. (Percentage)
Morbidity Mortality
Delayed closure 8 patients (17%) PM musculocutaneous flap 21 patients (44.7%) PM muscle with RAM flap 12 patients (25.5%) PM muscle with omentum flap 2 patients (4.2%) PM turnover flap 2 patients (4.2%) RAM flap 2 patients (4.2%) Total 47
3* 1† 0 0 0 0 4
0 0 1‡ 0 0 0 1
*Expired of heart failure during hospital stay. †Died of sepsis, mediastinitis related. ‡Partial wound dehiscence with organism of M. tuberculosis.
interval from cardiac surgery to symptom development of DSWI, interval from the diagnosis of infection to flap operation, and patient survival. Median follow-up interval was 10.5 months (range, from 6 to 18 months). Patient characteristic and demographic data are described in Table 2. The postsurgical courses were uneventful except that 1 patient developed wound tuberculosis infection. Partial wound dehiscence and fascial attenuation were noted; this patient's wound then stabilized and healed after the systemic plus local antituberculous treatment. The wound microbiology among these 12 patients demonstrated that 3 patients had methicillin-resistant Staphylococcus aureus (MRSA), 4 had S. aureus, 1 had coagulase-negative Staphylococcus, 1 had Escherichia coli, 1 had Pseudomonas aeruginosa, 1 had Mycobacterium tuberculosis, and 1 had a mixed growth of organisms. One instance of recurrent sternal infection was reported, and 1 patient expired because of heart failure 5 weeks after surgery. However, the coverage of the sternal wound was still successful in this patient.
FIGURE 6. A long incision wound and excess skin traction are needed to use the conventional PM flap harvesting method.
DISCUSSION Major infection of sternal wounds after open-heart surgery is rare but carries a 50% mortality risk in early series.1–4 Although the incidence of mediastinal wound infection in patients undergoing median sternotomy for cardiopulmonary bypass is less than 1%,15 the associated morbidity, mortality, and cost remain unacceptably high. El Oakley and Wright15 reported that sternal wound debridement and the removal of foreign materials are essential steps as well as asserted the current consensus regarding the ideal treatment options for a complicated median sternotomy wound. The DSWI, or mediastinitis, is classified into 4 subtypes according to the time of first presentation, the presence or
TABLE 2. Patient's Demographic Data
No Sex Age, y
1
M
45
2 3
M F
89 45
4 5 6 7 8 9
M F F F F M
75 56 71 76 85 77
10 M 11 M
83 74
12 M
76
Preoperative Diagnosis
Comorbidities
Diabetics, Chronic CAOD renal failure Diabetics, BMI > 35 CAOD Diabetics, Chronic renal CAOD failure, COPD, BMI > 35 Chronic renal failure CAOD Chronic renal failure, COPD CAOD, valve disease Diabetics CAOD Chronic renal failure Valve disease Diabetics, COPD CAOD Chronic renal failure, CAOD COPD. BMI > 35 Diabetics, COPD CAOD Diabetics, Chronic Valve disease renal failure COPD CAOD
Interval From Cardiac Interval From Surgery to Symptom the Diagnosis Flap Flap Survival Development of of Infection Follow-Up Patient Size, cm (S/F) DSWI (d) to op (d) Interval, m Survival
15 3
S
24
33
6
Y
12 5 16 6
S S
26 17
24 35
6.5 12
Y Y
20 5 S(partial loss) 20 6 S 25 5 S 16 6 S 18 7 S 20 8 S
28 24 34 33 32 27
43 43 33 32 36 24
12 12 12 6 12 18
Y N Y Y Y Y
15 6 16 4
S S
24 22
24 42
12 12
Y Y
16 5
S
15
27
18
Y
M, male; F, female; CAOD, Coronary artery occlusive disease; COPD, chronic obstructive pulmonary disease; BMI, body mass index.
S32
www.annalsplasticsurgery.com
© 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
FIGURE 7. Complication of conventional PM muscle flap harvesting and skin slough.
absence of risk factors, and whether previous attempts to treat the condition have failed. Closed mediastinal irrigation can be successful in patients with type I mediastinitis, whereas surgical debridement plus major reconstructive operation are probably the most effective treatment for patients with mediastinitis types II to V. However, the first step in treating a DSWI is adequate debridement, including bone and cartilage, to initially control the infection. At our institute, ultrasonic debridement was performed for most of the patients. The advantages were efficiency, rapidity, precision, less bleeding, and thorough removal of debris without the sacrifice of viable sternal wound bed tissue.16 After debridement, we subsequently applied NPWT. The reported advantages of NPWT for intractable sternal wound infections include1 protecting the underlying mediasternal structure from infection,2 permitting delayed sternal closure to avoid cardiac compression-induced compromised cardiopulmonary function,3 ensuring the possibility of repeated wound
Endoscope-Assisted PM-RA Bipedicle Muscle Flap
inspection and bedside debridement,4 ensuring the cost-effectiveness of wound care, and5 providing an option to promote sternal wound secondary healing for patients in poor physical condition.12 The PM muscle can serve as a pedicle flap alone or as a myocutaneous flap. It can tolerate and survive an infected or inflamed wound bed. In addition, the muscle provides adequate tissue volume to fill the tissue defects of a DSWI patient. Additionally, the RAM flap can fit ideally for the lower third of the sternal defect. The volume of this muscle flap also provides adequate tissue coverage. Because it is difficult for the PM advance flap alone to reach the lower third of the sternal defect, the combination of the PM and RA bipedicle flap was alternatively chosen for lower-third DSWI. The PM-RA bipedicle muscle flap was first described in 1998 by Solomon et al,14 who also characterized the dissection of this flap as rapid, and the complication rate of dissecting this flap compared favorably with that of other methods. Roh et al17 also explained that the PM–RA bipedicle flap is a practical and efficacious method in the reconstruction of the anterior chest wall defect. In Roh's series, not only the lower third but also the whole sternum was covered by this flap. He concluded that the PM-RAM bipedicle muscle flap could fill the defect with sufficient volume. A skin slough was reported as a common complication of flap harvesting in Roh's series, and a large area of undermined skin and cutaneous tissue as well as concurrent types of disease such as diabetes may easily occur as a result of a skin slough. Otherwise, excess skin traction may also lead to these self-limited problems. Surgeons should be especially aware that a severe (deep) skin slough may occur as a result of using an intraoperative skin traction device (Figs. 6 and 7). These 2 articles discussing a PM-RAM bipedicle flap for DSWI reconstruction are summarized in Table 3. The minimally invasive concept was introduced in the early 20th century, and the endoscope technique has undergone a major breakthrough over the past 2 decades. Endoscopic surgery involves inserting scopes through small incisions or natural body openings to diagnose and treat the disease. Under endoscope viewing, we harvested the PM muscle for sternal wound reconstruction. However, by using the PM muscle flap alone, we had difficulty reaching the lower third sternal wound. Therefore, the PM-RAM bipedicle flap was used for this difficult location defect. In our experience, a PM muscle harvested under endoscope viewing has the following advantages: (1) harvests through a limited incision wound can prevent excess traction of the skin edge and diminish the skin slough; (2) accidental injury to the surrounding neurovascular structure is prevented (in this procedure, endoscopic viewing enables the lateral thoracic artery and long thoracic nerve to be identified before cutting); (3) the PM muscle is cut along the anterior axillary line through clear visualization; and (4) only 10 to 15 minutes are required to divide the PM muscle under endoscopic viewing. Shortening the operation time is critical because DSWI patients typically have other concurrent types of disease such as diabetes, hypertension, heart failure, and pulmonary failure that increase the anesthesia risk. The reported recurrence rates after flap reconstruction varied. One recurrent sternal infection was reported among these patients. A
TABLE 3. Two Relevant Articles of a PM-RA Bipedicle Flap in DSWI Reconstruction Patient Number /Duration
Complication
Mortality
Solomon et al (1998) Plast Reconstr Surg
42 patients/8-y period
There were no deaths in this series
Roh et al (2008) J Thorac Cardiovasc Surg
27 patients/5-y period
Flap failure, 1; persistent infection, 1; pneumothorax, 1; hernia, 1; hematomas, 3; skin slough, 9 Recurrence, 3 Focal dehiscence, 3 Fascial attenuation, 1
© 2016 Wolters Kluwer Health, Inc. All rights reserved.
There were no deaths in this series
www.annalsplasticsurgery.com
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
S33
Annals of Plastic Surgery • Volume 76, Supplement 1, March 2016
Chou et al
wound culture revealed a MRSA infection. Our recurrence rate was 8%, and we readvanced the flap after proper antibiotic treatment and surgical debridement. The patient improved during this time, and this patient's wound healed 1 month later. The main limitation of this study is its retrospective design and that the data are from a single institution, indicating that the findings may not be generalizable. The assessment results of recurrence or complications may be influenced by the varying expertise and experience of physicians. Despite these limitations, we offered a relatively easy and safe way to reconstruct a lower third sternal wound by using endoscopeassisted harvesting of the PM-RAM bipedicle flap.
CONCLUSIONS The PM-RAM bipedicle flap can be an effective lifesaving boat for lower-third DSWI. In this study, endoscope-assisted harvesting only required limited skin incision wounds, prevented excess traction of the skin edge, and allowed us to identify the surrounding structure before cutting. We could cut the muscle along the anterior axillary line in only 10 to 15 minutes. This could be an effective alternative method for harvesting the PM-RAM bipedicle flap to reconstruct a lower-third sternal wound. REFERENCES 1. Patel NV, Woznick AR, Welsh KS, et al. Predictors of mortality after muscle flap advancement for deep sternal wound infections. Plast Reconstr Surg. 2009;123: 132–138. 2. Sachithanandan A, Nanjaiah P, Nightingale P, et al. Deep sternal wound infection requiring revision surgery: impact on mid-term survival following cardiac surgery. Eur J Cardiothorac Surg. 2008;33:673–678. 3. López-Monjardin H, de-la-Peña-Salcedo A, Mendoza-Muñoz M, et al. Omentum flap versus pectoralis major flap in the treatment of mediastinitis. Plast Reconstr Surg. 1998;101:1481–1485.
S34
www.annalsplasticsurgery.com
4. Ottino G, De Paulis R, Pansini S, et al. Major sternal wound infection after openheart surgery: a multivariate analysis of risk factors in 2,579 consecutive operative procedures. Ann Thorac Surg. 1987;44:173–179. 5. Sarr MG, Gott VL, Townsend TR. Mediastinal infection after cardiac surgery. Ann Thorac Surg. 1984;38:415–423. 6. Cabbabe EB, Cabbabe SW. Immediate versus delayed one-stage sternal debridement and pectoralis muscle flap reconstruction of deep sternal wound infections. Plast Reconstr Surg. 2009;123:1490–1494. 7. Poncelet AJ, Lengele B, Delaere B, et al. Algorithm for primary closure in sternal wound infection: a single institution 10-year experience. Eur J Cardiothorac Surg. 2008;33:232–238. 8. Price J, Rubens F, Bell M. Elastic device facilitating delayed primary closure of sternal wound infection. Ann Thorac Surg. 2007;83:1162–1165. 9. Lee CH, Hsien JH, Tang YB, et al. Reconstruction for sternal osteomyelitis at the lower third of sternum. J Plast Reconstr Aesthet Surg. 2010;63:633–641. 10. Eyileten Z, Akar AR, Eryilmaz S, et al. Vacuum-assisted closure and bilateral pectoralis muscle flaps for different stages of mediastinitis after cardiac surgery. Surg Today. 2009;39:947–954. 11. Greig AV, Geh JL, Khanduja V, et al. Choice of flap for the management of deep sternal wound infection—an anatomical classification. J Plast Reconstr Aesthet Surg. 2007;60:372–378. 12. Lee SS, Lin SD, Chen HM, et al. Management of intractable sternal wound infections with topical negative pressure dressing. J Card Surg. 2005;20:218–222. 13. Sun IF, Lee SS, Chiu CC, et al. Hyperbaric oxygen therapy with topical negative pressure: an alternative treatment for the refractory sternal wound infection. J Card Surg. 2008;23:677–680. 14. Solomon MP, Granick MS. Bipedicle muscle flaps in sternal wound repair. Plast Reconstr Surg. 1998;101:356–360. 15. El Oakley RM, Wright JE. Postoperative mediastinitis: classification and management. Ann Thorac Surg. 1996;61:1030–1036. 16. Cheng LH, Stewart J, Thompson M, et al. Ultrasonic debridement of contaminated facial wounds. Br J Oral Maxillofac Surg. 2002;40:149–150. 17. Roh TS, Lee WJ, Lew DH, et al. Pectoralis major-rectus abdominis bipedicled muscle flap in the treatment of poststernotomy mediastinitis. J Thorac Cardiovasc Surg. 2008;136:618–622.
© 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
