- Email: [email protected]
An Interdisciplinary Approach to the Prevention and Treatment of Groin Wound Complications After Lower Extremity Revascularization
An Interdisciplinary Approach to the Prevention and Treatment of Groin Wound Complications After Lower Extremity Revascularization Alyssa J. Reiffel,1 Peter W. Henderson,1 John K. Karwowski,2 and Jason A. Spector,1 New York, New York
Background: If not effectively treated, groin wound infections following lower extremity revascularization (LER) may result in graft or limb loss. Methods: A retrospective review was performed of all patients who underwent muscle flap transposition by a single surgeon after LER between 2006 and 2010. Results: Twenty-nine muscle transposition flaps were performed in 24 patients (21 sartorius, 6 rectus femoris, and 2 gracilis). Nineteen were for treatment of groin wound infections, two for treatment of lymphocele, one for coverage of exposed graft in the setting of pyoderma gangrenosum, and seven for infection prophylaxis. Two graft losses followed flap placement. The limb loss rate was 4%. When performed for therapeutic purposes, graft salvage rates were 100% for autogenous and 92% for synthetic grafts. Conclusions: Muscle transposition flaps are an effective means of graft salvage in the setting of groin wound complications following LER and should be considered for infection prophylaxis in high-risk patients.
INTRODUCTION The groin represents the most frequent site of vascular graft infection.1,2 Groin wound infections complicate approximately 0.7e7% of lower extremity revascularization (LER) procedures that utilize open approaches.2-5 Although the exact pathogenesis of groin wound infections is unclear, possible etiologies include the disruption of lymphatic channels and resultant seroma formation, moisture accumulation and skin maceration
1 Division of Plastic Surgery, Weill Cornell Medical College, New York, NY. 2 Division of Vascular Surgery, Weill Cornell Medical College, New York, NY.
Correspondence to: Jason A. Spector, MD, FACS, Division of Plastic Surgery, Weill Cornell Medical College, Payson 709-A, 525 East 68th Street, New York, NY 10065, USA; E-mail: [email protected] Ann Vasc Surg 2012; 26: 365e372 DOI: 10.1016/j.avsg.2011.08.002 Ó Annals of Vascular Surgery Inc. Published online: November 7, 2011
in intertriginous creases, and proximity to the genitalia and perineum, which are known to harbor various bacterial and fungal species.2,5 Historically, the mainstay of treatment for groin infections was total graft excision with or without subsequent extra-anatomic reconstruction.1,2 Such procedures resulted in limb loss and mortality rates of 10e79% and 9e58%, respectively.1,2,6,7 As early as 1963, Carter et al. described in a seven-patient case series a strategy for preservation of exposed or infected Dacron grafts characterized by aggressive wound debridement, antimicrobial therapy, and healing by secondary intention. The authors reported improved rates of graft salvage and decreased amputation and mortality rates compared with traditional treatment methods.8 However, numerous drawbacks to this approach remained: not only was the recovery time long and monetary costs high, but grafts (both autologous and autogenous) left in situ were at risk for thrombosis, superinfection by more virulent hospital-acquired pathogens, and anastomotic rupture.2 365
366 Reiffel et al.
With an improved understanding of the blood supply to the muscles around the groin, muscle transposition flaps rose in popularity as an adjunctive measure in the management of groin infections. Among many advantages, muscle transposition flaps eliminate dead space and cover exposed graft, thereby allowing for more thorough debridement of infected graft beds. Flaps have also been shown to increase local oxygen tension, augment immune cell and antibiotic delivery, and shorten hospital stay.1,2,4 As a result of their more frequent application, both mortality and limb loss rates resulting from groin infections following LER procedures have fallen dramatically (to 0e14% and 0e30%, respectively).1,2 In addition, although less often described, muscle transposition flaps have been successfully used in the treatment of persistent lymphoceles.6 Furthermore, although it has been suggested,9 the prophylactic use of muscle transposition flaps has not yet to our knowledge been investigated. Therefore, to further examine their therapeutic and prophylactic applications, we reviewed our recent experience using muscle transposition flaps in the setting of LER procedures.
METHODS After obtaining institutional review board approval, a retrospective review was performed of all patients who underwent muscle flap transposition by a single surgeon (J.A.S.) following open-approach revascularization procedures using groin incisions between 2006 and 2010. Charts were reviewed for patient demographics and comorbidities, revascularization procedure performed and graft material used, indication for flap placement (i.e., therapeutic flaps for the treatment of infection versus prophylactic flaps for the prevention of infection in the absence of clinical signs and/or symptoms of infection), number of debridements, use of a vacuum-assisted closure (VAC) device, timing and type of flap used, postoperative complications, length of hospital stay, and duration of follow-up. All research was performed in compliance with the Health Insurance Portability and Accountability Act. If muscle transposition was performed for the treatment of infection (typically defined as the presence of erythema, drainage, fever, leukocytosis, wound dehiscence, or positive cultures, or at the discretion of the primary vascular surgeon), charts were reviewed for interval to infection and clinical presentation and microbiology of infection. An ‘‘early’’ infection was defined as an infection
Annals of Vascular Surgery
Table I. Patient demographics Characteristic
N (%)
Male Female Age (years) Mean Range Peripheral vascular disease Coronary artery disease Diabetes mellitus Hypertension Hyperlipidemia Smoking Renal insufficiency/failure Chronic obstructive pulmonary disease
13 (54) 11 (46) 73 46e95 22 (92) 10 (42) 7 (29) 17 (71) 10 (42) 10 (42) 4 (17) 7 (29)
occurring within 30 days of the revascularization procedure. A ‘‘late’’ infection was defined as one occurring after 30 days.
RESULTS Twenty-nine muscle transposition flaps were performed in 24 patients, including 21 sartorius, 6 rectus femoris (RF), and 2 gracilis flaps. Patients ranged in age from 46 to 95 years (mean, 73). Thirteen were male. Three patients required bilateral flap placement. Patient demographics and comorbidities are as listed in Table I. Nineteen flaps were performed for the treatment of groin wound infections, two for the treatment of lymphocele, and one for soft tissue coverage of exposed graft in the setting of pyoderma gangrenosum (PG). Seven flaps were performed for infection prophylaxis in high-risk patients at the time of their revascularization procedure. Twenty-five of 29 flaps (86%) were performed for the coverage of synthetic material including Dacron or polytetrafluoroethylene (PTFE) (Table II). The overall mean time to clinically apparent infection was 164 days (range, 0-1862 days). For patients with ‘‘early’’ wound infections (n ¼ 11), the mean time to clinically apparent infection was 15 days (range, 0e27 days). For patients with ‘‘late’’ wound infections (n ¼ 5), the mean time to clinically apparent infection was 491 days (range, 45-1862 days). Common presenting symptoms included fevers, pain, and drainage from the groin wound (Table III). Coagulase-negative Staphylococcus (53%) and Staphylococcus aureus (32%) were the most common pathogens. Two patients grew Pseudomonas aeruginosa; their wounds ultimately
Vol. 26, No. 3, April 2012
Prevention and treatment of groin wound complications
367
Table II. Flap characteristics and graft material Flap number
Flap type
Indication
Intraoperative microbiology
Synthetic graft?
1
S
Infection
No
2 3 4a
S RF RF
Infection Infection Infection
5a,b 6 7 8 9a 10a,b 11 12 13 14 15 16 17 18a 19a,b 20a,b 21 22 23a
RF S S S S RF S S S S S G G S S RF S S S
Infection Lymphocele PG Prophylaxis Infection Infection Infection Infection Infection Prophylaxis Infection Infection Lymphocele Prophylaxis Infection Infection Prophylaxis Prophylaxis Infection
24a 25 26 27 28 29
RF S S S S S
Infection Prophylaxis Infection Infection Prophylaxis Infection
Klebsiella pneumoniae, Enterobacter cloacae, Pseudomonas aeruginosa, MSSA Coagulase-negative Staphylococcus, Diptheroid species Coagulase-negative Staphylococcus, Escherichia coli Vancomycin-resistant Enterococcus faecalis, MSSA, E. coli Candida albicans Serratia marcescens Coagulase-negative Staphylococcus e Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus MSSA Coagulase-negative Staphylococcus, Enterococcus faecium MSSA e P. aeruginosa Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus e Coagulase-negative Staphylococcus E.coli, MSSA MRSA e E. coli, Propionibacterium acnes, Enterobacter aerogenes, Bacteroides fragilis E. coli, P. acnes, E. aerogenes, B. fragilis e Vancomycin-resistant E. faecium Proteus mirabilis, MRSA, E. faecalis e Coagulase-negative Staphylococcus, E. coli
Yes Yes Yes No Yes No Yes Yes No Yes Yes No Yes Yes Yes Yes Yes No No Yes Yes Yes No Yes No Yes Yes Yes
S, sartorius; RF, rectus femoris; G, gracilis; MSSA, methicillin-sensitive S. aureus; MRSA, methicillin-resistant S. aureus. a Denotes flaps performed in the same patient. b Denotes a flap placed the groin contralateral to the index flap.
healed without complications. Four patients who underwent muscle flap placement for indications other than infection subsequently had positive intraoperative cultures: one patient with persistent lymphocele grew Serratia marcescens while another grew coagulase-negtive Staphylococcus, the patient with PG grew coagulase-negative Staphylococcus, and one patient who underwent prophylactic muscle flap placement grew methicillin-resistant S aureus. Excluding patients who underwent prophylactic flap placement, the median time to flap placement was 38 days after index revascularization procedure (mode, 17; range, 14e1938 days). Median hospital stay was 14 days after flap placement (range, 2e71 days). Prophylactic flaps were placed in patients with multiply reoperated groins, a thin soft tissue envelop due to body habitus or previous scarring,
a large burden of prosthetic material as determined by the vascular surgeon, and/or immune compromise due to multiple comorbidities, including poorly controlled diabetes mellitus. All prophylactic flaps were placed in the presence of synthetic graft. Two patients required return to the operating room for hematoma evacuation. Seven of 27 wounds (25.9%) developed minor complications such as small superficial separation or seroma treated with serial aspiration. Three of 27 wounds (11.1%) demonstrated persistent infection or failure to heal after sartorius muscle flap rotation, two of which (7.4%) were treated with RF placement and one (3.7%) with serial operative debridement. Two patients underwent graft excision and repair of the native vessel with an autologous or synthetic patch before muscle flap coverage. Three patients underwent bypass revision immediately before
368 Reiffel et al.
Annals of Vascular Surgery
Table III. Presenting signs and symptoms of infection among patients who underwent flap placement for the treatment of infection (N ¼ 15) Sign/symptom
N (%)
Fever Pain Drainage Graft exposure Erythema Hematoma/bleeding
8 5 6 4 6 2
(53) (33) (40) (27) (40) (13)
flap placement: one with prosthetic graft and two with native vein. There were two graft losses subsequent to flap placement: one in the patient with PG and another in a patient with a polymicrobial infection of a PTFE axillobifemoral bypass graft. Overall graft and limb loss rates were 10% (2 of 21) and 4% (1 of 24), respectively. When performed for the treatment of groin wound infection, graft salvage rates were 100% (3 of 3) for autogenous vein and 92% (12 of 13) for synthetic graft. Mean follow-up was 221 days (range, 11e615 days). There were two mortalities: one was a patient who underwent axillobifemoral bypass for aortic occlusion, suffered graft and resultant limb loss, and subsequently developed multiorgan system failure due to overwhelming infection; the other was a patient from whom care was withdrawn after he developed acute renal failure due to an unrelated intra-abdominal infection. His graft was functioning well at the time of his death (Table IV). A VAC device was used in nine patients before flap placement. In six of these patients, the VAC dressing was applied over an exposed vessel with an interposed piece of petrolatum-impregnated gauze. Three patients had a VAC dressing placed overlying their flap.
DISCUSSION Since their introduction as an adjunctive measure in the management of groin infections following LER, muscle transposition flaps have become an essential feature of the plastic surgeon’s armamentarium in this setting. Although most authors report the use of a single flap type, we believe it is imperative to tailor flap selection to the specific clinical scenario. The sartorius flap is the most frequently used muscle flap in the setting of groin wound infection at our institution (Fig. 1). It is located directly adjacent to the wound, is relatively easy to harvest (often through the original incision), leaves no functional deficit when relocated, and carries
Table IV. Complications Complication
N (%)
Minor Lymph leak Seroma Superficial dehiscence Majora,b Hematoma Superficial surgical-site infection Deep surgical-site infection Requirement of secondary flap Graft loss Limb loss Death Overallc
6 1 3 2 8 2 3 3 2 2 1 2 13
(21) (3) (10) (7) (28) (7) (10) (10) (7) (7) (3) (7) (45)
a Includes any complication requiring readmission or return to operating room. b Includes 1 patient who developed the following complications: hematoma requiring washout in the operating room, superficial and deep surgical-site infections, graft loss, limb loss (aboveknee amputation), and death. c Includes 1 patient who developed the following complications: hematoma requiring washout in the operating room and seroma treated with serial aspirations (N ¼ 29).
minimal donor-site morbidity. Often-cited drawbacks of the sartorius flap include its segmental blood supply that leaves the rotated muscle at risk for devitalization if two or more segmental vessels are divided, the theoretical risk of flap coinfection due to its proximity to the infected wound, and the high prevalence of superficial femoral artery disease in this population.1,2,9 Most importantly, the sartorius muscle lacks bulk and has a limited arc of rotation, making it suboptimal for large or proximal (i.e., iliac) wounds. In contrast, the RF flap (Fig. 2) is based on the profunda femoris artery, a vessel less likely to have associated atheromatous disease.1,4 Although the RF muscle provides significant bulk and has an arc of rotation that can easily reach the lower abdomen, harvest requires an additional thigh incision.1,4,9 Despite previous suggestions that use of this flap decreased the strength of knee extension, it has recently been demonstrated that careful repair of the distal vastus medialis to the distal vastus lateralis, in combination with aggressive postoperative physical therapy, mitigates most if not all of this loss.10,11 The gracilis, an expendable thigh adductor, has also been described in the treatment of groin infections (Fig. 3). The popularity of this flap has been limited by its relative lack of bulk, associated donor-site morbidity,12 and comparatively difficult access to harvest.
Vol. 26, No. 3, April 2012
Prevention and treatment of groin wound complications
369
Fig. 1. (A) Eighty-two-year-old man with exposed PTFE lower extremity bypass graft; (B) A sartorius flap is harvested via the same incision; (C) By rotating the flap into place, graft coverage is achieved.
The rectus abdominis (RA) is yet another option for groin wound coverage. Although it provides significant bulk, has an extensive arc of rotation, and may be elevated with a large skin paddle, harvesting the RA requires extensive dissection and its use may result in increased risk of ventral hernia.12 Patients within this population may have preexisting disease or even occlusion of the deep inferior epigastric artery, which is the vascular pedicle of the RA flap, prompting some authors to suggest preoperative angiogram.1 This flap is usually reserved as a last resort in patients with large proximal wounds not suitably treated with any of the other less invasive flap options. Despite the numerous muscle flaps available for transposition, graft preservation is possible only when the following criteria are met: the patient must not be systemically septic, the graft must be patent, and the graft-host vessel anastomoses must be intact.1,3,13 The specific patient comorbidities found to predict the development of wound infection vary between studies. Coronary artery disease,9 diabetes mellitus, obesity,5 and hypertension14 have all been suggested as contributing factors. In our study, peripheral vascular disease (92%), hypertension (71%), and coronary artery disease (42%) were most closely associated with the development of groin wound infection. In our series, we had an overall graft salvage rate of 94%. There were two graft losses (one PTFE and
one saphenous vein graft) despite the application of muscle transposition flaps. One of these losses occurred in a patient with multiple comorbidities, including aortic occlusion, who suffered cardiac arrest due to multiorgan system failure shortly after lower extremity (above-knee) amputation. The second case of graft loss occurred in a patient with PG whose femoral-posterior tibial bypass became stenosed, likely due to exposure along its entire course in combination with his autoimmune disease and history of polycythemia vera. We did not observe any significant differences in graft salvage rates following groin wound infection with respect to graft material used (PTFE versus Dacron) or graft wall characteristics (smooth-walled versus graft with interstices). Reported flap survival rates have been excellent across most studies. We report a 100% flap survival rate despite the early application of muscle flaps to infected wounds, in contrast to earlier studies that demonstrated a higher incidence of flap loss following their early application.6 We attribute this success to appropriate flap selection, meticulous surgical technique, proper antibiotic selection, and attentive wound care. In our series, two patients required return to the operating room for the rotation of a secondary muscle flap. In both cases, sartorius flaps were first applied, but were found at reexploration to provide insufficient bulk to appropriately obliterate the dead
370 Reiffel et al.
Fig. 2. (A) Seventy-four-year-old man with exposed PTFE lower extremity bypass graft; (B) A rectus femoris (RF) flap is harvested via a separate thigh incision; (C) The RF flap carries substantial bulk and is easily rotated
Fig. 3. The gracilis flap is also harvested via a separate access incision. It is less frequently used owing to its relative lack of bulk, difficult access, and donor-site morbidity.
space. In one of these cases, the flap had dehisced from its very proximal insertion. In both cases, subsequent reconstruction with an RF flap provided
Annals of Vascular Surgery
into place through a subcutaneous tunnel; (D) Once inset, the graft is covered with healthy, vascularized tissue.
excellent coverage and resulted in salvaged grafts with healed wounds. These cases not only highlight the necessity of proper initial flap choice but also demonstrate the multiple local options available for treatment of the infected groin wound. Our overall complication rate was 45%, which we believe represents an acceptable risk in the setting of successful graft preservation. Two patients required return to the operating room for hematoma evacuation, one of whom was placed on therapeutic heparin immediately postoperatively. Three patients developed superficial surgical site infections: two were treated with debridement and one with parenteral antibiotics. Six patients experienced minor wound complications such as superficial separation or seroma. There were no incidences of anastomotic hemorrhage, recurrent infection, or pseudoaneurysm. Figure 4 illustrates our proposed algorithm for the management of suspected groin wound infection. While multiple imaging techniques have been
Vol. 26, No. 3, April 2012
Prevention and treatment of groin wound complications
371
Table V. Criteria for prophylactic muscle flap placement Criteria for prophylactic muscle flap placement
Irradiated wound bed Thin soft tissue envelope Large prosthetic burden Reoperation Presence of other risk factors for infection/seroma
Fig. 4. Proposed algorithm for the management of suspected groin wound infections. *Gracilis flap is recommended if RF is unavailable.
proposed, operative exploration must be the first step undertaken because it remains the only way to confirm or exclude the presence of infection with certainty. If operative exploration reveals purulent drainage, devitalized tissue, or graft exposure, the wound must be widely and aggressively debrided, serially if necessary. Some authors have described the use of the VAC device in this setting as an alternative to immediate muscle flap closure.3,15 We have found such dressings to be useful adjuncts in graft preservation as a means of temporary graft coverage between serial debridements. However, the VAC dressing alone is rarely sufficient as a primary treatment modality for exposed grafts, as this approach does not result in immediate graft coverage with healthy, vascularized tissue or augment local oxygen tension and immune cell or antibiotic delivery. Furthermore, treatment with VAC dressings alone can result in longer hospital stays and longer time to wound closure. Although interval use of the VAC prevents desiccation of autologous vessels, owing to the at least theoretical risk of anastomotic blowout, we prefer to set the device to 75 mm Hg of suction and place a layer of petrolatum-impregnated gauze overlying the anastomosis. Given the low morbidity inherent to muscle transposition flaps, particularly the sartorius flap, and the minimal additional operative time required
for their application, we have adopted their prophylactic use in patients at increased risk for wound complications such as infection or seroma (Table V). Specifically, patients with an irradiated wound bed, a thin soft tissue envelope, a large burden of prosthetic material, or those undergoing reoperation would benefit from muscle flap placement at the time of their initial LER procedure. Although some of these parameters may be subjective in nature, we believe they are identifiable by an experienced surgeon. Finally, any attempt to salvage an infected vascular graft is a labor-intensive and potentially risky undertaking. For such an effort to be both safe and effective, there must be a strong working relationship between the vascular and plastic surgery teams. Without an interdisciplinary approach, graft salvage is unlikely to be successful.
CONCLUSIONS Our data demonstrate the use of muscle transposition flaps for the treatment of groin wound infections after LER is associated with a 100% salvage rate for autogenous vein and a 92% salvage rate for synthetic graft. Bringing healthy, vascularized muscle to infected wounds allows for the salvage of grafts, and therefore limbs, that would otherwise be lost to infection. Furthermore, due to the relatively low morbidity associated with harvesting local muscle flaps, their prophylactic use should be considered at the time of LER in patients who are at high risk for developing infection. REFERENCES 1. Graham RG, Omotoso PO, Hudson DA. The effectiveness of muscle flaps for the treatment of prosthetic graft sepsis. Plast Reconstr Surg 2002;109:108e13. discussion 114e105. 2. Seify H, Moyer HR, Jones GE, et al. The role of muscle flaps in wound salvage after vascular graft infections: the Emory experience. Plast Reconstr Surg 2006;117:1325e33.
372 Reiffel et al.
3. Dosluoglu HH, Schimpf DK, Schultz R, et al. Preservation of infected and exposed vascular grafts using vacuum assisted closure without muscle flap coverage. J Vasc Surg 2005;42: 989e92. 4. Illig KA, Alkon JE, Smith A, et al. Rotational muscle flap closure for acute groin wound infections following vascular surgery. Ann Vasc Surg 2004;18:661e8. 5. Szilagyi DE, Smith RF, Elliott JP, et al. Infection in arterial reconstruction with synthetic grafts. Ann Surg 1972;176: 321e33. 6. Armstrong PA, Back MR, Bandyk DF, et al. Selective application of sartorius muscle flaps and aggressive staged surgical debridement can influence long-term outcomes of complex prosthetic graft infections. J Vasc Surg 2007;46:71e8. 7. Calligaro KD, Veith FJ, Schwartz ML, et al. Selective preservation of infected prosthetic arterial grafts. Analysis of a 20e year experience with 120 extracavitary-infected grafts. Ann Surg 1994;220:461e9. discussion 469e71. 8. Carter SC, Cohen A, Whelan TJ. Clinical experience with management of the infected dacron graft. Ann Surg 1963;158:249e55. 9. Colwell AS, Donaldson MC, Belkin M, et al. Management of early groin vascular bypass graft infections with sartorius and rectus femoris flaps. Ann Plast Surg 2004;52:49e53.
Annals of Vascular Surgery
10. Sbitany H, Koltz PF, Girotto J, et al. Quadriceps muscle function following harvest of the Rectus Femoris muscle as a pedicle flap for complex groin wound reconstruction. In: 26th Annual Meeting of the Northeastern Society of Plastic Surgeons; Vol. 1; 2009; Charleston, SC. pp. 160e161. 11. Gardetto A, Raschner C, Schoeller T, et al. Rectus femoris muscle flap donor-site morbidity. Br J Plast Surg 2005;58: 175e82. 12. Alkon JD, Smith A, Losee JE, et al. Management of complex groin wounds: preferred use of the rectus femoris muscle flap. Plast Reconstr Surg 2005;115:776e83. discussion 784e775. 13. Calligaro KD, Westcott CJ, Buckley RM, et al. Infrainguinal anastomotic arterial graft infections treated by selective graft preservation. Ann Surg 1992;216:74e9. 14. Herrera FA, Kohanzadeh S, Nasseri Y, et al. Management of vascular graft infections with soft tissue flap coverage: improving limb salvage ratesea veterans affairs experience. Am Surg 2009;75:877e81. 15. Dosluoglu HH, Loghmanee C, Lall P, et al. Management of early (<30 day) vascular groin infections using vacuumassisted closure alone without muscle flap coverage in a consecutive patient series. J Vasc Surg 2010;51:1160e6.
Background: If not effectively treated, groin wound infections following lower extremity revascularization (LER) may result in graft or limb loss. Methods: A retrospective review was performed of all patients who underwent muscle flap transposition by a single surgeon after LER between 2006 and 2010. Results: Twenty-nine muscle transposition flaps were performed in 24 patients (21 sartorius, 6 rectus femoris, and 2 gracilis). Nineteen were for treatment of groin wound infections, two for treatment of lymphocele, one for coverage of exposed graft in the setting of pyoderma gangrenosum, and seven for infection prophylaxis. Two graft losses followed flap placement. The limb loss rate was 4%. When performed for therapeutic purposes, graft salvage rates were 100% for autogenous and 92% for synthetic grafts. Conclusions: Muscle transposition flaps are an effective means of graft salvage in the setting of groin wound complications following LER and should be considered for infection prophylaxis in high-risk patients.
INTRODUCTION The groin represents the most frequent site of vascular graft infection.1,2 Groin wound infections complicate approximately 0.7e7% of lower extremity revascularization (LER) procedures that utilize open approaches.2-5 Although the exact pathogenesis of groin wound infections is unclear, possible etiologies include the disruption of lymphatic channels and resultant seroma formation, moisture accumulation and skin maceration
1 Division of Plastic Surgery, Weill Cornell Medical College, New York, NY. 2 Division of Vascular Surgery, Weill Cornell Medical College, New York, NY.
Correspondence to: Jason A. Spector, MD, FACS, Division of Plastic Surgery, Weill Cornell Medical College, Payson 709-A, 525 East 68th Street, New York, NY 10065, USA; E-mail: [email protected] Ann Vasc Surg 2012; 26: 365e372 DOI: 10.1016/j.avsg.2011.08.002 Ó Annals of Vascular Surgery Inc. Published online: November 7, 2011
in intertriginous creases, and proximity to the genitalia and perineum, which are known to harbor various bacterial and fungal species.2,5 Historically, the mainstay of treatment for groin infections was total graft excision with or without subsequent extra-anatomic reconstruction.1,2 Such procedures resulted in limb loss and mortality rates of 10e79% and 9e58%, respectively.1,2,6,7 As early as 1963, Carter et al. described in a seven-patient case series a strategy for preservation of exposed or infected Dacron grafts characterized by aggressive wound debridement, antimicrobial therapy, and healing by secondary intention. The authors reported improved rates of graft salvage and decreased amputation and mortality rates compared with traditional treatment methods.8 However, numerous drawbacks to this approach remained: not only was the recovery time long and monetary costs high, but grafts (both autologous and autogenous) left in situ were at risk for thrombosis, superinfection by more virulent hospital-acquired pathogens, and anastomotic rupture.2 365
366 Reiffel et al.
With an improved understanding of the blood supply to the muscles around the groin, muscle transposition flaps rose in popularity as an adjunctive measure in the management of groin infections. Among many advantages, muscle transposition flaps eliminate dead space and cover exposed graft, thereby allowing for more thorough debridement of infected graft beds. Flaps have also been shown to increase local oxygen tension, augment immune cell and antibiotic delivery, and shorten hospital stay.1,2,4 As a result of their more frequent application, both mortality and limb loss rates resulting from groin infections following LER procedures have fallen dramatically (to 0e14% and 0e30%, respectively).1,2 In addition, although less often described, muscle transposition flaps have been successfully used in the treatment of persistent lymphoceles.6 Furthermore, although it has been suggested,9 the prophylactic use of muscle transposition flaps has not yet to our knowledge been investigated. Therefore, to further examine their therapeutic and prophylactic applications, we reviewed our recent experience using muscle transposition flaps in the setting of LER procedures.
METHODS After obtaining institutional review board approval, a retrospective review was performed of all patients who underwent muscle flap transposition by a single surgeon (J.A.S.) following open-approach revascularization procedures using groin incisions between 2006 and 2010. Charts were reviewed for patient demographics and comorbidities, revascularization procedure performed and graft material used, indication for flap placement (i.e., therapeutic flaps for the treatment of infection versus prophylactic flaps for the prevention of infection in the absence of clinical signs and/or symptoms of infection), number of debridements, use of a vacuum-assisted closure (VAC) device, timing and type of flap used, postoperative complications, length of hospital stay, and duration of follow-up. All research was performed in compliance with the Health Insurance Portability and Accountability Act. If muscle transposition was performed for the treatment of infection (typically defined as the presence of erythema, drainage, fever, leukocytosis, wound dehiscence, or positive cultures, or at the discretion of the primary vascular surgeon), charts were reviewed for interval to infection and clinical presentation and microbiology of infection. An ‘‘early’’ infection was defined as an infection
Annals of Vascular Surgery
Table I. Patient demographics Characteristic
N (%)
Male Female Age (years) Mean Range Peripheral vascular disease Coronary artery disease Diabetes mellitus Hypertension Hyperlipidemia Smoking Renal insufficiency/failure Chronic obstructive pulmonary disease
13 (54) 11 (46) 73 46e95 22 (92) 10 (42) 7 (29) 17 (71) 10 (42) 10 (42) 4 (17) 7 (29)
occurring within 30 days of the revascularization procedure. A ‘‘late’’ infection was defined as one occurring after 30 days.
RESULTS Twenty-nine muscle transposition flaps were performed in 24 patients, including 21 sartorius, 6 rectus femoris (RF), and 2 gracilis flaps. Patients ranged in age from 46 to 95 years (mean, 73). Thirteen were male. Three patients required bilateral flap placement. Patient demographics and comorbidities are as listed in Table I. Nineteen flaps were performed for the treatment of groin wound infections, two for the treatment of lymphocele, and one for soft tissue coverage of exposed graft in the setting of pyoderma gangrenosum (PG). Seven flaps were performed for infection prophylaxis in high-risk patients at the time of their revascularization procedure. Twenty-five of 29 flaps (86%) were performed for the coverage of synthetic material including Dacron or polytetrafluoroethylene (PTFE) (Table II). The overall mean time to clinically apparent infection was 164 days (range, 0-1862 days). For patients with ‘‘early’’ wound infections (n ¼ 11), the mean time to clinically apparent infection was 15 days (range, 0e27 days). For patients with ‘‘late’’ wound infections (n ¼ 5), the mean time to clinically apparent infection was 491 days (range, 45-1862 days). Common presenting symptoms included fevers, pain, and drainage from the groin wound (Table III). Coagulase-negative Staphylococcus (53%) and Staphylococcus aureus (32%) were the most common pathogens. Two patients grew Pseudomonas aeruginosa; their wounds ultimately
Vol. 26, No. 3, April 2012
Prevention and treatment of groin wound complications
367
Table II. Flap characteristics and graft material Flap number
Flap type
Indication
Intraoperative microbiology
Synthetic graft?
1
S
Infection
No
2 3 4a
S RF RF
Infection Infection Infection
5a,b 6 7 8 9a 10a,b 11 12 13 14 15 16 17 18a 19a,b 20a,b 21 22 23a
RF S S S S RF S S S S S G G S S RF S S S
Infection Lymphocele PG Prophylaxis Infection Infection Infection Infection Infection Prophylaxis Infection Infection Lymphocele Prophylaxis Infection Infection Prophylaxis Prophylaxis Infection
24a 25 26 27 28 29
RF S S S S S
Infection Prophylaxis Infection Infection Prophylaxis Infection
Klebsiella pneumoniae, Enterobacter cloacae, Pseudomonas aeruginosa, MSSA Coagulase-negative Staphylococcus, Diptheroid species Coagulase-negative Staphylococcus, Escherichia coli Vancomycin-resistant Enterococcus faecalis, MSSA, E. coli Candida albicans Serratia marcescens Coagulase-negative Staphylococcus e Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus MSSA Coagulase-negative Staphylococcus, Enterococcus faecium MSSA e P. aeruginosa Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus e Coagulase-negative Staphylococcus E.coli, MSSA MRSA e E. coli, Propionibacterium acnes, Enterobacter aerogenes, Bacteroides fragilis E. coli, P. acnes, E. aerogenes, B. fragilis e Vancomycin-resistant E. faecium Proteus mirabilis, MRSA, E. faecalis e Coagulase-negative Staphylococcus, E. coli
Yes Yes Yes No Yes No Yes Yes No Yes Yes No Yes Yes Yes Yes Yes No No Yes Yes Yes No Yes No Yes Yes Yes
S, sartorius; RF, rectus femoris; G, gracilis; MSSA, methicillin-sensitive S. aureus; MRSA, methicillin-resistant S. aureus. a Denotes flaps performed in the same patient. b Denotes a flap placed the groin contralateral to the index flap.
healed without complications. Four patients who underwent muscle flap placement for indications other than infection subsequently had positive intraoperative cultures: one patient with persistent lymphocele grew Serratia marcescens while another grew coagulase-negtive Staphylococcus, the patient with PG grew coagulase-negative Staphylococcus, and one patient who underwent prophylactic muscle flap placement grew methicillin-resistant S aureus. Excluding patients who underwent prophylactic flap placement, the median time to flap placement was 38 days after index revascularization procedure (mode, 17; range, 14e1938 days). Median hospital stay was 14 days after flap placement (range, 2e71 days). Prophylactic flaps were placed in patients with multiply reoperated groins, a thin soft tissue envelop due to body habitus or previous scarring,
a large burden of prosthetic material as determined by the vascular surgeon, and/or immune compromise due to multiple comorbidities, including poorly controlled diabetes mellitus. All prophylactic flaps were placed in the presence of synthetic graft. Two patients required return to the operating room for hematoma evacuation. Seven of 27 wounds (25.9%) developed minor complications such as small superficial separation or seroma treated with serial aspiration. Three of 27 wounds (11.1%) demonstrated persistent infection or failure to heal after sartorius muscle flap rotation, two of which (7.4%) were treated with RF placement and one (3.7%) with serial operative debridement. Two patients underwent graft excision and repair of the native vessel with an autologous or synthetic patch before muscle flap coverage. Three patients underwent bypass revision immediately before
368 Reiffel et al.
Annals of Vascular Surgery
Table III. Presenting signs and symptoms of infection among patients who underwent flap placement for the treatment of infection (N ¼ 15) Sign/symptom
N (%)
Fever Pain Drainage Graft exposure Erythema Hematoma/bleeding
8 5 6 4 6 2
(53) (33) (40) (27) (40) (13)
flap placement: one with prosthetic graft and two with native vein. There were two graft losses subsequent to flap placement: one in the patient with PG and another in a patient with a polymicrobial infection of a PTFE axillobifemoral bypass graft. Overall graft and limb loss rates were 10% (2 of 21) and 4% (1 of 24), respectively. When performed for the treatment of groin wound infection, graft salvage rates were 100% (3 of 3) for autogenous vein and 92% (12 of 13) for synthetic graft. Mean follow-up was 221 days (range, 11e615 days). There were two mortalities: one was a patient who underwent axillobifemoral bypass for aortic occlusion, suffered graft and resultant limb loss, and subsequently developed multiorgan system failure due to overwhelming infection; the other was a patient from whom care was withdrawn after he developed acute renal failure due to an unrelated intra-abdominal infection. His graft was functioning well at the time of his death (Table IV). A VAC device was used in nine patients before flap placement. In six of these patients, the VAC dressing was applied over an exposed vessel with an interposed piece of petrolatum-impregnated gauze. Three patients had a VAC dressing placed overlying their flap.
DISCUSSION Since their introduction as an adjunctive measure in the management of groin infections following LER, muscle transposition flaps have become an essential feature of the plastic surgeon’s armamentarium in this setting. Although most authors report the use of a single flap type, we believe it is imperative to tailor flap selection to the specific clinical scenario. The sartorius flap is the most frequently used muscle flap in the setting of groin wound infection at our institution (Fig. 1). It is located directly adjacent to the wound, is relatively easy to harvest (often through the original incision), leaves no functional deficit when relocated, and carries
Table IV. Complications Complication
N (%)
Minor Lymph leak Seroma Superficial dehiscence Majora,b Hematoma Superficial surgical-site infection Deep surgical-site infection Requirement of secondary flap Graft loss Limb loss Death Overallc
6 1 3 2 8 2 3 3 2 2 1 2 13
(21) (3) (10) (7) (28) (7) (10) (10) (7) (7) (3) (7) (45)
a Includes any complication requiring readmission or return to operating room. b Includes 1 patient who developed the following complications: hematoma requiring washout in the operating room, superficial and deep surgical-site infections, graft loss, limb loss (aboveknee amputation), and death. c Includes 1 patient who developed the following complications: hematoma requiring washout in the operating room and seroma treated with serial aspirations (N ¼ 29).
minimal donor-site morbidity. Often-cited drawbacks of the sartorius flap include its segmental blood supply that leaves the rotated muscle at risk for devitalization if two or more segmental vessels are divided, the theoretical risk of flap coinfection due to its proximity to the infected wound, and the high prevalence of superficial femoral artery disease in this population.1,2,9 Most importantly, the sartorius muscle lacks bulk and has a limited arc of rotation, making it suboptimal for large or proximal (i.e., iliac) wounds. In contrast, the RF flap (Fig. 2) is based on the profunda femoris artery, a vessel less likely to have associated atheromatous disease.1,4 Although the RF muscle provides significant bulk and has an arc of rotation that can easily reach the lower abdomen, harvest requires an additional thigh incision.1,4,9 Despite previous suggestions that use of this flap decreased the strength of knee extension, it has recently been demonstrated that careful repair of the distal vastus medialis to the distal vastus lateralis, in combination with aggressive postoperative physical therapy, mitigates most if not all of this loss.10,11 The gracilis, an expendable thigh adductor, has also been described in the treatment of groin infections (Fig. 3). The popularity of this flap has been limited by its relative lack of bulk, associated donor-site morbidity,12 and comparatively difficult access to harvest.
Vol. 26, No. 3, April 2012
Prevention and treatment of groin wound complications
369
Fig. 1. (A) Eighty-two-year-old man with exposed PTFE lower extremity bypass graft; (B) A sartorius flap is harvested via the same incision; (C) By rotating the flap into place, graft coverage is achieved.
The rectus abdominis (RA) is yet another option for groin wound coverage. Although it provides significant bulk, has an extensive arc of rotation, and may be elevated with a large skin paddle, harvesting the RA requires extensive dissection and its use may result in increased risk of ventral hernia.12 Patients within this population may have preexisting disease or even occlusion of the deep inferior epigastric artery, which is the vascular pedicle of the RA flap, prompting some authors to suggest preoperative angiogram.1 This flap is usually reserved as a last resort in patients with large proximal wounds not suitably treated with any of the other less invasive flap options. Despite the numerous muscle flaps available for transposition, graft preservation is possible only when the following criteria are met: the patient must not be systemically septic, the graft must be patent, and the graft-host vessel anastomoses must be intact.1,3,13 The specific patient comorbidities found to predict the development of wound infection vary between studies. Coronary artery disease,9 diabetes mellitus, obesity,5 and hypertension14 have all been suggested as contributing factors. In our study, peripheral vascular disease (92%), hypertension (71%), and coronary artery disease (42%) were most closely associated with the development of groin wound infection. In our series, we had an overall graft salvage rate of 94%. There were two graft losses (one PTFE and
one saphenous vein graft) despite the application of muscle transposition flaps. One of these losses occurred in a patient with multiple comorbidities, including aortic occlusion, who suffered cardiac arrest due to multiorgan system failure shortly after lower extremity (above-knee) amputation. The second case of graft loss occurred in a patient with PG whose femoral-posterior tibial bypass became stenosed, likely due to exposure along its entire course in combination with his autoimmune disease and history of polycythemia vera. We did not observe any significant differences in graft salvage rates following groin wound infection with respect to graft material used (PTFE versus Dacron) or graft wall characteristics (smooth-walled versus graft with interstices). Reported flap survival rates have been excellent across most studies. We report a 100% flap survival rate despite the early application of muscle flaps to infected wounds, in contrast to earlier studies that demonstrated a higher incidence of flap loss following their early application.6 We attribute this success to appropriate flap selection, meticulous surgical technique, proper antibiotic selection, and attentive wound care. In our series, two patients required return to the operating room for the rotation of a secondary muscle flap. In both cases, sartorius flaps were first applied, but were found at reexploration to provide insufficient bulk to appropriately obliterate the dead
370 Reiffel et al.
Fig. 2. (A) Seventy-four-year-old man with exposed PTFE lower extremity bypass graft; (B) A rectus femoris (RF) flap is harvested via a separate thigh incision; (C) The RF flap carries substantial bulk and is easily rotated
Fig. 3. The gracilis flap is also harvested via a separate access incision. It is less frequently used owing to its relative lack of bulk, difficult access, and donor-site morbidity.
space. In one of these cases, the flap had dehisced from its very proximal insertion. In both cases, subsequent reconstruction with an RF flap provided
Annals of Vascular Surgery
into place through a subcutaneous tunnel; (D) Once inset, the graft is covered with healthy, vascularized tissue.
excellent coverage and resulted in salvaged grafts with healed wounds. These cases not only highlight the necessity of proper initial flap choice but also demonstrate the multiple local options available for treatment of the infected groin wound. Our overall complication rate was 45%, which we believe represents an acceptable risk in the setting of successful graft preservation. Two patients required return to the operating room for hematoma evacuation, one of whom was placed on therapeutic heparin immediately postoperatively. Three patients developed superficial surgical site infections: two were treated with debridement and one with parenteral antibiotics. Six patients experienced minor wound complications such as superficial separation or seroma. There were no incidences of anastomotic hemorrhage, recurrent infection, or pseudoaneurysm. Figure 4 illustrates our proposed algorithm for the management of suspected groin wound infection. While multiple imaging techniques have been
Vol. 26, No. 3, April 2012
Prevention and treatment of groin wound complications
371
Table V. Criteria for prophylactic muscle flap placement Criteria for prophylactic muscle flap placement
Irradiated wound bed Thin soft tissue envelope Large prosthetic burden Reoperation Presence of other risk factors for infection/seroma
Fig. 4. Proposed algorithm for the management of suspected groin wound infections. *Gracilis flap is recommended if RF is unavailable.
proposed, operative exploration must be the first step undertaken because it remains the only way to confirm or exclude the presence of infection with certainty. If operative exploration reveals purulent drainage, devitalized tissue, or graft exposure, the wound must be widely and aggressively debrided, serially if necessary. Some authors have described the use of the VAC device in this setting as an alternative to immediate muscle flap closure.3,15 We have found such dressings to be useful adjuncts in graft preservation as a means of temporary graft coverage between serial debridements. However, the VAC dressing alone is rarely sufficient as a primary treatment modality for exposed grafts, as this approach does not result in immediate graft coverage with healthy, vascularized tissue or augment local oxygen tension and immune cell or antibiotic delivery. Furthermore, treatment with VAC dressings alone can result in longer hospital stays and longer time to wound closure. Although interval use of the VAC prevents desiccation of autologous vessels, owing to the at least theoretical risk of anastomotic blowout, we prefer to set the device to 75 mm Hg of suction and place a layer of petrolatum-impregnated gauze overlying the anastomosis. Given the low morbidity inherent to muscle transposition flaps, particularly the sartorius flap, and the minimal additional operative time required
for their application, we have adopted their prophylactic use in patients at increased risk for wound complications such as infection or seroma (Table V). Specifically, patients with an irradiated wound bed, a thin soft tissue envelope, a large burden of prosthetic material, or those undergoing reoperation would benefit from muscle flap placement at the time of their initial LER procedure. Although some of these parameters may be subjective in nature, we believe they are identifiable by an experienced surgeon. Finally, any attempt to salvage an infected vascular graft is a labor-intensive and potentially risky undertaking. For such an effort to be both safe and effective, there must be a strong working relationship between the vascular and plastic surgery teams. Without an interdisciplinary approach, graft salvage is unlikely to be successful.
CONCLUSIONS Our data demonstrate the use of muscle transposition flaps for the treatment of groin wound infections after LER is associated with a 100% salvage rate for autogenous vein and a 92% salvage rate for synthetic graft. Bringing healthy, vascularized muscle to infected wounds allows for the salvage of grafts, and therefore limbs, that would otherwise be lost to infection. Furthermore, due to the relatively low morbidity associated with harvesting local muscle flaps, their prophylactic use should be considered at the time of LER in patients who are at high risk for developing infection. REFERENCES 1. Graham RG, Omotoso PO, Hudson DA. The effectiveness of muscle flaps for the treatment of prosthetic graft sepsis. Plast Reconstr Surg 2002;109:108e13. discussion 114e105. 2. Seify H, Moyer HR, Jones GE, et al. The role of muscle flaps in wound salvage after vascular graft infections: the Emory experience. Plast Reconstr Surg 2006;117:1325e33.
372 Reiffel et al.
3. Dosluoglu HH, Schimpf DK, Schultz R, et al. Preservation of infected and exposed vascular grafts using vacuum assisted closure without muscle flap coverage. J Vasc Surg 2005;42: 989e92. 4. Illig KA, Alkon JE, Smith A, et al. Rotational muscle flap closure for acute groin wound infections following vascular surgery. Ann Vasc Surg 2004;18:661e8. 5. Szilagyi DE, Smith RF, Elliott JP, et al. Infection in arterial reconstruction with synthetic grafts. Ann Surg 1972;176: 321e33. 6. Armstrong PA, Back MR, Bandyk DF, et al. Selective application of sartorius muscle flaps and aggressive staged surgical debridement can influence long-term outcomes of complex prosthetic graft infections. J Vasc Surg 2007;46:71e8. 7. Calligaro KD, Veith FJ, Schwartz ML, et al. Selective preservation of infected prosthetic arterial grafts. Analysis of a 20e year experience with 120 extracavitary-infected grafts. Ann Surg 1994;220:461e9. discussion 469e71. 8. Carter SC, Cohen A, Whelan TJ. Clinical experience with management of the infected dacron graft. Ann Surg 1963;158:249e55. 9. Colwell AS, Donaldson MC, Belkin M, et al. Management of early groin vascular bypass graft infections with sartorius and rectus femoris flaps. Ann Plast Surg 2004;52:49e53.
Annals of Vascular Surgery
10. Sbitany H, Koltz PF, Girotto J, et al. Quadriceps muscle function following harvest of the Rectus Femoris muscle as a pedicle flap for complex groin wound reconstruction. In: 26th Annual Meeting of the Northeastern Society of Plastic Surgeons; Vol. 1; 2009; Charleston, SC. pp. 160e161. 11. Gardetto A, Raschner C, Schoeller T, et al. Rectus femoris muscle flap donor-site morbidity. Br J Plast Surg 2005;58: 175e82. 12. Alkon JD, Smith A, Losee JE, et al. Management of complex groin wounds: preferred use of the rectus femoris muscle flap. Plast Reconstr Surg 2005;115:776e83. discussion 784e775. 13. Calligaro KD, Westcott CJ, Buckley RM, et al. Infrainguinal anastomotic arterial graft infections treated by selective graft preservation. Ann Surg 1992;216:74e9. 14. Herrera FA, Kohanzadeh S, Nasseri Y, et al. Management of vascular graft infections with soft tissue flap coverage: improving limb salvage ratesea veterans affairs experience. Am Surg 2009;75:877e81. 15. Dosluoglu HH, Loghmanee C, Lall P, et al. Management of early (<30 day) vascular groin infections using vacuumassisted closure alone without muscle flap coverage in a consecutive patient series. J Vasc Surg 2010;51:1160e6.