Risk factors for failure after rotational muscle flap coverage for prosthetic knee infections

Abstract

Introduction

Soft tissue defects are a devastating complication of prosthetic joint infections (PJI) after total knee arthroplasty (TKA). Rotational flaps are commonly utilized to address these defects with variable reports of success. This study aimed to identify predictors of poor outcomes in rotational muscle flap placement after prosthetic knee infections. The authors hypothesized that outcomes may vary based on infecting pathogen and treatment characteristics.

Methods

44 cases of rotational muscle flaps for prosthetic knee infection were retrospectively evaluated at a tertiary referral hospital from 2007 to 2020. Muscle flap types included 39 medial and four lateral gastrocnemius, and one anterior tibialis. Minimum follow-up was 1 year (median: 3.4 years). Primary outcome was flap-related complications. Secondary outcomes included recurrent infection requiring additional surgery, final joint outcomes, and mortality.

Results

One-year complication-free flap survivorship was 83.9%, recurrent infection-free survivorship was 65.7%, and amputation-free survivorship was 79%. Multivariable cox regression revealed that rheumatoid arthritis diagnosis (HR: 3.4; p = .028) and methicillin-resistant Staphylococcus aureus-positive culture (HR: 4.0; p = .040) had increased risk, while Coagulase-negative Staphylococcus infections had reduced risk for recurrent or persistent infection (HR: 0.2; p = .023). Final joint outcome was retained TKA implant in 18 (40.9%), amputation in 15 (34.1%) patients, and definitive treatment with articulating spacer in 10 (22.7%). 5-years survivorship from death was 71.4%.

Conclusion

Rotational muscle flaps for soft tissue coverage of the knee are often performed in limb salvage situations with poor survivorship from flap complications, reinfections, and amputation. When considering surgical options for limb salvage, patients should be counseled on these risks.

Keywords

prosthetic knee infection rotational muscle flap soft tissue total knee arthroplasty

Introduction

Total knee arthroplasty (TKA) has become an increasingly common procedure to treat osteoarthritis, providing substantial improvements in pain, joint function, and overall quality of life. Although rare, the incidence of complications after TKA can be devastating. About 11% of patients will develop wound complications, with 0.33% requiring surgical intervention for early wound complications, which is critical to address since it increases the risk of deep periprosthetic joint infection (PJI).^1–3 The incidence of PJI is estimated to range from 1 to 3% and accounts for 25% of revision TKA procedures.^4–8 Treatment algorithms for PJI typically depend on the timing of symptom onset and the mechanism of infection, with acute PJI typically being treated with a combination of debridement, antibiotics, and implant retention (DAIR) and chronic PJI being treated with two-stage revision arthroplasty. Treatment failure occurs in 10-30% of patients undergoing two-stage revision procedures, and soft tissue defect further increases this risk of failure.^9–14

Surgical options to mitigate this increased risk of failure when treating an infected TKA with a soft tissue defect include debridement of the skin edges with primary closure, negative pressure wound therapy with secondary healing, skin grafts at sites of epidermal deficit, and/or rotational or free tissue flaps.¹⁵ Rotational medial gastrocnemius muscle flaps are most commonly used for anterior knee soft tissue defects following TKA due to its reliability and ease of harvest.¹⁵ A secondary perceived benefit is increased local blood supply via the sural vasculature, which offers increased delivery of antibiotics and nutrients for healing.¹⁵ Alternatives to rotational gastrocnemius muscle flaps include vastus lateralis or latissimus dorsi free tissue transfer.¹⁶ Investigation thus far into outcomes following muscle flap coverage following infected TKA has been limited by small cohorts, lack of follow-up, and significant variability in reported outcomes and definitions of success.^17–25 The purpose of this study is to evaluate the outcomes and rates of treatment failure for rotational soft tissue coverage in the course of surgical treatment for PJI following TKA.

Materials and methods

Patient selection

Institutional review board approval was obtained prior to data collection. The institutional electronic health record (EHR) database was queried for patients with a diagnosis of PJI receiving muscle flap surgeries after TKA between 2007 and 2020, identifying 59 cases. Cohort selection was performed as outlined in in the flowchart, Figure 1. Patients receiving flap reconstruction after megaprostheses for oncological resection (n = 5), chronic extensor mechanism deficiency (n = 3), and aseptic revision (n = 1) were excluded. Four patients requiring flap coverage after knee arthrodesis or resection arthroplasty after multiple failed treatment for PJI were also excluded. One latissimus dorsi free flap was removed. Ultimately, 44 patients with rotational muscle flaps after prosthetic knee infection and a minimum of 1 year of follow-up were included. Mean follow-up duration was 4.5 years (range: 1.0–12.5 years).

Figure 1.

Patient cohort selection flowchart.

Surgical treatment

In this cohort, patients presenting to a tertiary care referral hospital with prosthetic knee infection were treated with debridement, antibiotics, and implant retention (DAIR) for acute PJI or 2-stage exchange for chronic infections. In brief, DAIR involved excisional debridement of necrotic or devitalized tissues and irrigation of the retained implant and joint with antibiotic impregnated solution, 3% acetic acid, diluted betadine, and normal saline. Metal components were scrubbed with a soft brush using chlorhexidine. After modular components were exchanged, the wound was irrigated again with normal saline and dissolvable antibiotic beads (broad spectrum antibiotics added to calcium sulfate) were placed intra-articularly. Jackson-Pratt (JP) surgical drains were placed within the joint (at the discretion of the orthopaedic surgeon) before the arthrotomy was closed as completely as possible, prior to same-surgery handoff to the plastic surgery team performing the flap reconstruction. Decision for flap placement was made by the orthopaedic surgeon preoperatively based on anticipated soft tissue availability, with notification of possible flap requirement to the plastic surgery team. Final decision was made intraoperatively after direct assessment of soft tissue coverage. Patients undergoing 2-stage revision had flap placement during explantation of the implant or prior spacer, or during reimplantation after spacer removal. Of 31 patients treated with an antibiotic spacer, 15 (48.4%) were treated with an articulating spacer (antibiotic cement, primary femoral component, all polyethylene tibial component) and 16 (51.6%) were treated with a static spacer (bulk cement in joint interface with stabilizing device selected by size of bone defect). Intra-operative cultures of the soft tissue, synovial fluid, and bone were collected for culture. For patients receiving flap reconstruction during second stage reimplantation, cultures obtained during prior component removal were reported. All flaps were performed by plastic surgeons. Medial and lateral gastrocnemius flaps were released from the Achilles tendon and the contralateral head, and assessed for vascularity before rotation over the anterior aspect of the knee to cover the defect. In one patient, the gastrocnemius had undergone 100% fat atrophy and a joint decision was made by the orthopedic and plastic surgery teams to perform a rotational anterior tibialis flap. In patients with remaining skin defect over the rotated muscle flap, a split-thickness skin graft (STSG) was harvested from the patient’s ipsilateral. Finally, a wound vacuum-assisted closure (V.A.C.) was placed over the skin graft. Three patients received STSG within 1 week after discharge. All patients received tailored antibiotic treatment under guidance of infectious disease physicians, which typically consisted of 6 weeks of parenteral antibiotic treatment and 6 weeks of oral antibiotic treatment, depending on the isolated microorganism and patients’ response to treatment All patients were allowed to weight bear as tolerated (WBAT) after their procedure except for patients with static spacers who were allowed to WBAT in a knee immobilizing orthosis. After discharge, patients were followed by both a dual orthopaedic-infectious disease musculoskeletal infections clinic and the plastic surgeon.

Study variables

The institutional EHR was manually reviewed and data was collected on patient age, sex, race, comorbidities including smoking status, diabetes, obesity, peripheral vascular disease, rheumatoid arthritis, and overall Elixhauser comorbidity score, wound location and size, prior surgical history, flap surgery, surgical treatment for PJI, and microbiological culture results. Primary outcome was flap-related complications. Flap related complications included any wound complications (eschar, dehiscence), flap failure, and focal necrosis requiring revascularization and debridement. Secondary outcomes were recurrent or persistent infection requiring surgery, amputation, death, flap-related complications, orthopaedic complications, and 90-days ED visits and readmissions.

Statistical analysis

Descriptive statistics comparing 2-stage and DAIR were performed using univariable statistics. Normality of continuous variable distributions were determined by the Shapiro-Wilk normality test and visually via histogram representation. The t test or Mann-Whitney U test were used as appropriate for comparison of continuous variables, and chi-square or Fisher’s exact test was performed for categorical variables. Multivariable cox (proportional hazards) regression analysis, controlling for age, smoking status, Elixhauser score, and PJI treatment type, was performed for calculating hazard ratios for various risk factors for primary (reinfection) and secondary outcomes (amputation and death). p-values reported on survival curve graphs were computed from log rank tests. Post hoc comparison of flap outcomes by flap type was performed using the chi-square test. All statistics were performed using RStudio v 1.1.463 (R Foundation, Vienna, Austria). Findings were considered statistically significant if p-value was <.05.

Results

Patient, PJI, and flap characteristics

Of 44 total rotational flaps for soft tissue defects after prosthetic knee infection, 31 (70.5%) were performed in either stage of a two-stage exchange and 13 (29.5%) at the time of DAIR (Table 1). Of patients undergoing two-stage exchange, 25/31 (80.6%) had flap placement performed at implant explantation and 6/31 (19.4%) at the reimplantation stage. On average, pre-operative BMI was 33.6 kg/m² (sd: 7.5), Elixhauser score was 8.4 (sd: 4.6), 47.7% were current or former smokers, and 43.2% had a diagnosis of type II diabetes.

Table 1.

Patient, PJI, and flap characteristics stratified by PJI surgical procedure.

	Overall	PJI surgical procedure		p-value
	Overall	2-Stage	Early postop/DAIR	p-value
N	44	31	13
Patient demographics
Age, mean (sd)	67.53 (9.97)	67.74 (9.97)	67.01 (10.38)	.828
Male sex, n (%)	22 (50.0)	17 (54.8)	5 (38.5)	.510
Non-caucasian race, n (%)	11 (25.0)	4 (12.9)	7 (53.8)	.008
Patient comorbidities
BMI at flap placement (kg/m^2), n (%)	33.59 (7.50)	33.06 (7.29)	34.86 (8.13)	.473
Current/former smoker	21 (47.7)	13 (41.9)	8 (61.5)	.325
Elixhauser score, mean (sd)	8.36 (4.63)	8.06 (4.75)	9.08 (4.42)	.514
Obesity, n (%)	26 (59.1)	17 (54.8)	9 (69.2)	.507
Diabetes, n (%)	19 (43.2)	13 (41.9)	6 (46.2)	1.000
Peripheral vascular disorders, n (%)	15 (34.1)	8 (25.8)	7 (53.8)	.092
Rheumatoid arthritis, n (%)	8 (18.2)	6 (19.4)	2 (15.4)	1.000
Wound
Wound location, n (%)				.095
Anterior knee	39 (88.6)	29 (93.5)	10 (76.9)
Inferior knee	2 (4.5)	1 (3.2)	1 (7.7)
Lateral knee	2 (4.5)	0 (0.0)	2 (15.4)
Posterior knee	1 (2.3)	1 (3.2)	0 (0.0)
Wound area (cm^2), median [IQR]	44.00 [28.75, 81.25]	50.00 [31.50, 80.00]	42.00 [22.75, 85.75]	.854
Flap
STSG placed at time of flap surgery, n (%)	32 (72.7)	22 (71.0)	10 (76.9)	1.000
STSG size (cm^2), median [IQR]	44.00 [30.50, 71.50]	50.00 [31.50, 74.00]	30.00 [21.00, 39.00]	.360
Flap type, n (%)				.144
Medial gastrocnemius	39 (88.6)	29 (93.5)	10 (76.9)
Lateral gastrocnemius	4 (9.1)	2 (6.5)	2 (15.4)
Anterior tibialis	1 (2.3)	0 (0.0)	1 (7.7)
PJI history
Number of prior aseptic revisions, median [IQR]	0.00 [0.00, 1.00]	0.00 [0.00, 1.00]	0.00 [0.00, 1.00]	.180
Number of prior septic revisions, median [IQR]	2.00 [1.00, 3.00]	2.00 [1.00, 3.00]	1.00 [0.00, 2.00]	.015
Timing of flap in 2-stage: at reimplantation (vs. explantation)	6 (19.4)	6 (19.4)
Follow-up time (months), median [IQR]	40.43 [24.34, 76.47]	41.77 [24.73, 81.35]	36.73 [23.77, 62.53]	.546

The bold indicates significant p-values which were identified as p < .05.

Most wound defects were located on the anterior aspect of the knee (n = 39/44, 88.6%) and median defect size was 44.0 cm² [IQR: 28.8–81.3]. A split-thickness skin graft (STSG) was performed at the time of flap reconstruction in 72.7% of cases. Median STSG size was 44.0 cm² [IQR: 30.50–71.50]. The most common muscle utilized for flap reconstruction was the medial gastrocnemius (n = 39/44, 88.6%), followed by lateral gastrocnemius (n = 4/44, 9.1%), and anterior tibialis (n = 1/44, 2.3%).

The median number of prior septic revisions for patients receiving flaps during 2-stage treatment was 2 [IQR: 1–3] compared to 1 [IQR: 0–2] for patients treated with DAIR (p = .015). For patients who underwent 2-stage exchange arthroplasty, most patients had flap placement during component removal or spacer exchange (n = 25/31, 80.6%) as opposed to during component reimplantation (n = 6/31, 19.4%).

Short- and long-term outcomes after index PJI treatment and flap placement

Overall, there were 13 (29.5%) ED visits and 18 (40.9%) readmissions in the 90-days postoperative window (Table 2, Supp. Table 1). In total, 19/44 (43.2%) patients visited the ED and/or were readmitted in the 90-days postoperative period. Flap-related reasons for readmission were more common in the DAIR group (38.5% vs 9.7%; p = .037) compared to the 2-stage group.

Table 2.

Short- and long-term outcomes after index PJI treatment and flap placement.

	Overall	PJI surgical procedure		p-value
	Overall	2-Stage	Early postop/DAIR	p-value
N	44	31	13
90-day complications
Any ED visit, n (%)	13 (29.5)	9 (29.0)	4 (30.8)	1.000
Orthopaedic-related ED visit, n (%)	6 (13.6)	4 (12.9)	2 (15.4)	1.000
Flap-related ED visit, n (%)	5 (11.4)	3 (9.7)	2 (15.4)	.623
Medical ED visit, n (%)	5 (11.4)	3 (9.7)	2 (15.4)	.623
Any readmission, n (%)	18 (40.9)	11 (35.5)	7 (53.8)	.323
Orthopaedic-related readmission, n (%)	12 (27.3)	9 (29.0)	3 (23.1)	1.000
Flap-related readmission, n (%)	8 (18.2)	3 (9.7)	5 (38.5)	.037
Medical readmission, n (%)	4 (9.1)	2 (6.5)	2 (15.4)	.570
Orthopaedic outcomes
Requiring any orthopaedic surgery, n (%)	24 (54.5)	18 (58.1)	6 (46.2)	.522
Time to ortho surgery (years), median [IQR]	2.77 [0.93, 5.63]	3.38 [1.03, 5.06]	1.58 [0.97, 5.65]	.714
Reinfection requiring surgery at 1 year, n (%)	22 (50.0)	18 (58.1)	4 (30.8)	.185
Reinfection requiring surgery at last follow-up, n (%)	23 (52.3)	18 (58.1)	5 (38.5)	.325
Time to reinfection (years), median [IQR]	3.53 [1.00, 7.38]	3.42 [1.03, 7.42]	4.50 [1.07, 6.83]	.682
Final joint outcome
TKA implant, n (%)	18 (40.9)	9 (29.0)	9 (69.2)	.020
Resection, n (%)	1 (2.3)	1 (3.2)	0 (0.0)	1.000
Arthrodesis, n (%)	0 (0.0)	0 (0.0)	0 (0.0)	-
Destination spacer, n (%)	10 (22.7)	7 (22.6)	3 (23.1)	1.000
Amputation, n (%)	15 (34.1)	14 (45.2)	1 (7.7)	.034
Time to amputation (years), median [IQR]	5.30 [1.95, 10.02]	5.33 [1.81, 11.18]	5.23 [5.23, 5.23]	.817
Mortality, n (%)	12 (27.3)	8 (25.8)	4 (30.8)	.727
Time to death, (years), median [IQR]	1.98 [0.45, 4.10]	3.10 [1.49, 6.47]	0.42 [0.30, 1.00]	.126
Flap-related outcomes
Flap revision/additional flaps, n (%)	4 (9.1)	3 (9.7)	1 (7.7)	1.000
Flap faliure, n (%)	4 (9.1)	4 (12.9)	0 (0.0)	.302
Focal necrosis, n (%)	4 (9.1)	1 (3.2)	3 (23.1)	.071
Repeat STSG, n (%)	4 (9.1)	1 (3.2)	3 (23.1)	.071
Wound dehiscence, n (%)	5 (11.4)	4 (12.9)	1 (7.7)	1.000

The bold indicates significant p-values which were identified as p < .05.

Most patients (n = 24/44, 54.5%) required additional orthopaedic surgery (excluding planned reimplantation in the 2-stage group). 50% required surgery for reinfection at 1 year post-operatively, and 52.3% (n = 23/44) were reinfected at last follow-up. Final joint outcome was successful TKA reimplantation in 18 (40.9%), amputation in 15 (34.1%), definitive treatment with an articulating spacer in 10 (22.7%), and resection arthroplasty in one (2.3%). Overall mortality at final follow-up was 27.3% (n = 12/44).

The most common flap complication was wound dehiscence (n = 5/44, 11.4%). Other flap complications occurred at the same rate (n = 4/44, 9.1%): flap failure, additional flaps, focal necrosis, and repeat STSG.

Overall survivorship from reinfection, amputation, death, and flap complications

1-year infection-free survivorship was 48.7% [95%CI: 35.7–66.3%], amputation-free survivorship was 74.0% [95%CI: 61.9–88.5%], patient survivorship was 90.9% [95%CI: 82.8–99.8%] (Table 3), and complication-free flap survivorship was 77.0% [95%CI: 65.5–90.6%]. Comparatively, 5-years infection-free survivorship was relatively similar at 45.6% [95%CI: 32.7–63.7%], while amputation-free and patient survivorship decreased to 64.8% [95%CI: 51.3–82.0%], 71.4% [95%CI: 57.1–89.4%], respectively. Five-year complication-free flap survivorship was 70.8% [95%CI: 58.0–86.5%].

Table 3.

Overall survivorship from reinfection, amputation, death, and flap complications.

Time (months)	Survivorship probability from outcomes [95% CI]			Flap complications
Time (months)	Reinfection requiring surgery	Amputation	Death	Flap complications
6	65.7 [53.0–81.4]	79.0 [67.6–92.2]	90.9 [82.8–99.8]	83.9 [73.7–95.6]
12	48.7 [35.7–66.3]	74.0 [61.9–88.5]	90.9 [82.8–99.8]	83.9 [73.7–95.6]
24	45.6 [32.7–63.7]	71.6 [59.1–86.6]	86.0 [76.3–97.1]	81.0 [69.9–93.9]
36	45.6 [32.7–63.7]	64.8 [51.3–82.0]	80.2 [68.7–93.7]	77.9 [66.0–92.0]
60	45.6 [32.7–63.7]	64.8 [51.3–82.0]	71.4 [57.1–89.4]	77.9 [66.0–92.0]

Risk factors for reinfection requiring surgery, amputation, death, and flap complications

Multivariable cox regression analysis controlling for patient comorbidities and surgical treatment revealed an increased risk for reinfection requiring surgery for MRSA-positive cultures (HR: 4.02 [95%CI: 1.16–13.98]; p = .028) (Table 4, Figure 2). CoNS-positive PJIs, on the other hand, demonstrated reduced risk for reinfection (HR: 0.19 [95%CI: 0.04–0.93]; p = .040) (Figure 3). Rheumatoid arthritis was also associated with an increased risk for reinfection (HR: 3.44 [95%CI: 1.18–9.97]; p = .023).

Table 4.

Multivariable cox regression of risk factors for unfavorable outcomes.

Variable	Reinfection requiring surgery		Amputation		Flap complications		Death
Variable	HR [95% CI]	p-value	HR [95% CI]	p-value	HR [95% CI]	p-value	HR [95% CI]	p-value
MRSA (n = 6)	4.02 [1.16–13.98]	.028	2.02 [0.41–9.86]	.386	3.92 [0.59–26.20]	.159	1.73 [0.24–12.57]	.589
CoNS (n = 10)	0.19 [0.04–0.93]	.040	—	—	—	—	0.47 [0.09–2.51]	.377
Age	0.98 [0.94–1.03]	.417	0.98 [0.92–1.03]	.391	0.92 [083–1.01]	.093	1.11 [1.03–1.20]	.005
Rheumatoid arthritis (n = 8)	3.44 [1.18–9.97]	.023	3.29 [0.83–13.02]	.089	3.05 [0.48–19.51]	.238	1.04 [0.17–6.35]	.969
Current/former smoker (n = 24)	1.65 [0.63–4.28]	.307	4.53 [1.15–17.81]	.031	1.69 [0.31–9.17]	.543	1.29 [0.29–5.83]	.740
Elixhauser score	1.07 [0.96–1.18]	.238	1.06 [0.93–1.21]	.416	1.10 [0.89–1.35]	.380	1.10 [0.94–1.30]	.242
PJI surgery, reference = 2-stage
Early post-op/DAIR	0.68 [0.24–1.94]	.468	0.11 [0.01–0.94]	.043	2.60 [0.61–11.16]	.197	1.77 [0.50–6.31]	.377

The bold indicates significant p-values which were identified as p < .05.

Figure 2.

Survivorship from reinfection requiring surgery after MRSA PJI.

Figure 3.

Survivorship from reinfection requiring surgery after CoNS PJI.

Patients undergoing DAIR for surgical treatment of their PJI had reduced risk for amputation (HR: 0.11 [0.01–0.99]; p = .043) but increased risk for flap complications (HR: 4.94 [1.36–17.96]; p = .015) compared to 2-stage treatment in the multivariable analysis. Current or former smoker status was associated with increased risk for amputation (HR: 4.53 [1.15–17.81]; p = .031). Older age was associated with increased risk for death after flap placement (HR: 1.11 [95%CI: 1.03–1.20]; p = .005).

Discussion

We found acceptable survivorship of rotational flaps after prosthetic knee infections from flap-related complications but poor 1-year and 5-years survivorship from recurrent or persistent infection of 48.7% and 45.6%, respectively. Large open wounds after prosthetic knee infection are challenging problems for orthopaedic surgeons.^26–28 Full-thickness defects that leave exposed bone and prosthesis significantly increases the risk for deep periprosthetic joint infection and may lead to prosthesis resection, arthrodesis, or limb loss.^15,26 Rotational muscle flaps for defect coverage have been described with large variability in reported success rates for prosthesis and limb salvage.^19,29–31 Although patients presenting with wound defects have significant comorbidites and prior surgical history of the joint, risk factors for successful limb salvage with muscle flaps are not well understood.^16,32–34 The present study identifies several risk factors for poor outcomes following flap placement after prosthetic knee infection.

Our results on survivorship from reinfection lie within the range (42.3–63%) reported by prior studies of rotational flaps for prosthetic knee infection at various endpoints.^19,32,35 The largest study of rotational flaps after TKA by Houdek and colleagues included flaps for aseptic causes of soft tissue defects, such as during primary TKA (n = 18/83), and reported the highest revision survival rate of 68% at 10 years.³³ Although flap reconstruction is utilized in various causes of soft tissue defects, PJI-induced defects have a considerably worse prognosis.³² Notably, our amputation rate at final follow-up was higher at 34.1% (n = 15/44) than previously reported rates of 3.7–23.1% while our rates of other functional limb loss outcomes (i.e., resection or arthrodesis) were lower (n = 1/44).^19,32,35 At our tertiary referral hospital, patients were offered rotational flaps in the setting of multiply-operated limb salvage after recalcitrant PJI, with a median of two prior septic revisions. The majority of amputations in our cohort were performed in patients undergoing either implant explantation or reimplantation during 2-stage revision (45.2% vs 7.7% in DAIR), who also had more prior septic revision surgeries than the DAIR group. This indicates that the 2-stage group had failed more prior PJI treatments. Despite the higher rate of amputation in the 2-stage group, mortality rates were similar for the 2-stage and DAIR groups (25.8 vs 30.8%), which may reflect more aggressive but appropriate treatment decision-making to achieve infection eradication by the surgeon and patient. Our cohort’s overall mortality rate of 27.3% is corroborated by the 26% reported by a prior study, which also supported the notion of advanced age as an independent risk factor.³⁵ Mortality rates are significant after soft tissue reconstruction for PJI, thus. While rotational muscle flaps can fill defects and provide a vascularized wound bed for antibiotic delivery, risk for recurrent infection remains particularly high for PJI patients, and functional limb loss may be unavoidable.

Infection clearance after PJI was associated with a multitude of factors including host status, appropriate treatment method, and pathogen.^36,37 In this study, CoNS was the most common organism identified (n = 10, 21.7%), and MRSA (n = 5) and MSSA (n = 4) were the next most common (Supp. Table 2). Our culture results are consistent with three small studies identifying Staphylococcus spp. most frequently, and one study that found CoNS, specifically, to be the most common pathogen identified (n = 16, 37%).^25,31,32,35 Resident gram-positive skin bacteria comprise a majority of pathogens identified in PJIs with wound complications. Interestingly, our results demonstrated an increased risk of recurrent infection for patients with MRSA-positive PJI and a reduced risk of reinfection for CoNS PJI patients. MRSA has been well-established to have infection recurrence despite appropriate surgical treatment, and although CoNS has recently gained increasing attention for its virulence with rising rates of methicillin-resistance, our cohort highlights MRSA as the pathogen with worse outcomes in the setting of flap reconstruction.^37,38 The relatively lower risk for recurrent infection for CoNS is likely explained by the high occurrence of polymicrobial (n = 8, 18.2%) and enterococcal (n = 6, 13.6%) PJIs in this cohort, which are less common but notorious for high treatment failure rates.^39,40 More importantly, overall outcomes for all patients were dismal with a recurrent or persistent reinfection rate of 52.3%, and the virulence of offending pathogens seen in this cohort emphasize the importance of an intact soft tissue envelope in preventing infections. Regarding host status, patients with a pre-operative diagnosis of rheumatoid arthritis also demonstrated higher risk for infection, commensurate with prior reports of rheumatoid arthritis as an independent risk factor for worse outcomes after PJI.⁴¹

Flap success is also dependent on the timing of flap placement with improved rates of limb salvage and component retention in patients with prophylactic as opposed to reactive flap placement.^23,36,37,42 However, flap placement timing in the setting of a 2-stage revision for PJI has not been sufficiently studied. In a sub-analysis, we found no significant differences in infection clearance rates between patients with flaps placed at prosthesis explantation (n = 10/25, 40%) or reimplantation (n = 3/6, 50%) in the 2-stage revision (p = .676). Similarly, a study by Warren et al. showed that 37.5% (n = 3/8) of patients receiving flap coverage at explantation and 40% (n = 4/10) at reimplantation retained their TKA.³² While no differences were noted for 2-stage timing, our study revealed patients receiving flaps concurrently with DAIR had relatively improved outcomes in regards to amputation compared to the 2-stage group. DAIR patients had higher rates of minor flap-related complications such as repeat STSG (Table 2). All cases of flap failure and most cases of wound dehiscence and additional flap placement occurred in the 2-stage group. Although all flaps were “reactive” in this cohort, the DAIR group had fewer prior septic revisions, representing flap placement at an earlier stage in their overall treatment course for PJI. Our study supports the importance of early recognition of soft tissue insufficiency and appropriate planning for wound management when primary closure is not attainable as compounding wound defects can lead to functional limb loss or amputation.¹⁵

Flap complications were not uncommon in this cohort with a 1- and 5-years complication-free flap survivorship of 83.9% and 77.9%, respectively. Assessing the vascular viability of flaps and not over-tensioning the flap is crucial to avoiding flap failure and necrosis.^15,28 Additionally, infections can precipitate sinus tract formation, wound dehiscence, and flap failure requiring additional flap reconstruction. In this study, there were 5 patients with wound dehiscence and 4 patients requiring additional flaps, three of which received pedicled fasciocutaneous flaps and one required a rotational rectus femoris flap. In a post-hoc analysis, there was no difference in failure rates for medial (n = 3/39) and lateral gastrocnemius (n = 1/4) flaps (p = .333). Patients requiring repeat STSG after failed graft take were common reasons for readmission within the 90-day period and were most common in the DAIR group. Flaps are not without complications and these additional risks should be discussed with the patients when discussing treatment options for limb salvage.

There are several limitations to this study with the major limitation being our study’s small sample size. Exclusion of patients to diminish confounders through a more homogenous cohort reduced the power of our analysis. Next, functional outcome scores for patients were sparse and unattainable for this cohort. Therefore, we are unable to comment on the quality of life for patients with varying final joint outcomes. Additionally, this was a single-institution study and the regional frequencies and virulence of pathogens detected in this study may not reflect those of other institutions. Furthermore, treatment decisions by a variety of infectious disease providers and surgeons over the 13-years study time period may affect outcomes. Nevertheless, the authors believe the present study provides relevant and informative findings on rotational muscle flaps after prosthetic knee infection.

Conclusion

Rotational muscle flaps for soft tissue coverage after prosthetic knee infections have high rates of reinfection, amputation, and mortality. This study is the first to report associations between offending pathogen and risk for recurrent or persistent infection following PJI treatment and flap placement. Recognition of soft tissue insufficiency and flap reconstruction early in a patient’s surgical course for prosthetic knee infection may provide better outcomes for limb salvage. Despite the ability to provide soft tissue coverage with rotational muscle flaps, patients should be advised on the high rates of treatment failure and functional limb loss.

Supplemental Material

Supplemental Material - Risk factors for failure after rotational muscle flap coverage for prosthetic knee infections

Supplemental Material for Risk factors for failure after rotational muscle flap coverage for prosthetic knee infections by Billy I Kim, Colleen M Wixted, Andrew M Schwartz, William A Jiranek, Sean P Ryan and Thorsten M Seyler in Journal of Orthopaedic Surgery.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Billy I Kim

Sean P Ryan

Supplemental Material

Supplemental material for this article is available online.

References

Gaine

Ramamohan

Hussein

, et al. Wound infection in hip and knee arthroplasty. J Bone Joint Surg Br 2000; 82(4): 561–565.

Galat

McGovern

Larson

, et al. Surgical treatment of early wound complications following primary total knee arthroplasty. J Bone Joint Surg Am 2009; 91(1): 48–54.

Saleh

Olson

Resig

, et al. Predictors of wound infection in hip and knee joint replacement: results from a 20 year surveillance program. J Orthop Res 2002; 20(3): 506–515.

Kapadia

Berg

Daley

, et al. Periprosthetic joint infection. Lancet 2016; 387(10016): 386–394.

Fink

Anagnostakos

Winkler

. Periprosthetic joint infection. BioMed Res Int 2019; 2019: 6834680.

Premkumar

Kolin

Farley

, et al. Projected economic burden of periprosthetic joint infection of the hip and knee in the United States. J Arthroplasty 2021; 36(5): 1484–1489.

Kurtz

Lau

Watson

, et al. Economic burden of periprosthetic joint infection in the United States. J Arthroplasty 2012; 27: 61–65.

Schwartz

Farley

Guild

, et al. Projections and epidemiology of revision hip and knee arthroplasty in the United States to 2030. J Arthroplasty 2020; 35(6S): S79–S85.

Bejon

Berendt

Atkins

, et al. Two-stage revision for prosthetic joint infection: predictors of outcome and the role of reimplantation microbiology. J Antimicrob Chemother 2010; 65(3): 569–575.

10.

Triantafyllopoulos

Memtsoudis

Zhang

, et al.

Periprosthetic infection recurrence after 2-stage exchange arthroplasty: failure or fate?

J Arthroplasty 2017; 32(2): 526–531.

11.

Berend

Lombardi

Morris

, et al. Two-stage treatment of hip periprosthetic joint infection is associated with a high rate of infection control but high mortality. Clin Orthop Relat Res 2013; 471(2): 510–518.

12.

Biring

Kostamo

Garbuz

, et al. Two-stage revision arthroplasty of the hip for infection using an interim articulated prostalac hip spacer: a 10- to 15- year follow-up study. J Bone Joint Surg Br 2009; 91(11): 1431–1437.

13.

Sanchez-Sotelo

Berry

Hanssen

, et al. Midterm to long-term followup of staged reimplantation for infected hip arthroplasty. Clin Orthop Relat Res 2009; 467(1): 219–224.

14.

Romano

Albisetti

, et al. Preformed antibiotic-loaded cement spacers for two-stage revision of infected total hip arthroplasty. Long term results. Hip Int 2012; 22(Suppl 8): S46–S53.

15.

Osei

Rebehn

Boyer

. Soft-tissue defects after total knee arthroplasty: management and reconstruction. J Am Acad Orthop Surg 2016; 24(11): 769–779.

16.

Moog

Tinwald

Aitzetmueller

, et al. The usage of pedicled or free muscle flaps represents a beneficial approach for periprosthetic infection after knee arthroplasty. Ann Plast Surg 2020; 85(5): 539–545.

17.

McPherson

Patzakis

Gross

, et al. Infected total knee arthroplasty. Two-stage reimplantation with a gastrocnemius rotational flap. Clin Orthop Relat Res 1997; 341: 73–81.

18.

Gerwin

Rothaus

Windsor

, et al. Gastrocnemius muscle flap coverage of exposed or infected knee prostheses. Clin Orthop Relat Res. 1993; (286): 64–70.

19.

Tetreault

Della Valle

Bohl

, et al.

What factors influence the success of medial gastronemius flaps in the treatment of infected TKAs?

Clin Orthop Relat Res 2016; 474(3): 752–763.

20.

Corten

Struelens

Evans

, et al. Gastrocnemius flap reconstruction of soft-tissue defects following infected total knee replacement. Bone Joint Lett J 2013; 95-B(9): 1217–1221.

21.

Pozzobon

Helito

Guimaraes

, et al. Rotation flaps for coverage after total knee arthroplasty. Acta Ortopédica Bras 2013; 21(4): 219–222.

22.

Suda

Cieslik

Grutzner

, et al. Flaps for closure of soft tissue defects in infected revision knee arthroplasty. Int Orthop 2014; 38(7): 1387–1392.

23.

Kwiecien

Lamaris

Gharb

, et al. Long-term outcomes of total knee arthroplasty following soft-tissue defect reconstruction with muscle and fasciocutaneous flaps. Plast Reconstr Surg 2016; 137(1): 177e–186e.

24.

Markovich

Dorr

Klein

, et al. Muscle flaps in total knee arthroplasty. Clin Orthop Relat Res 1995; 321: 122–130.

25.

Adam

Watson

Jarratt

, et al. Outcome after flap cover for exposed total knee arthroplasties. A report of 25 cases. J Bone Joint Surg Br 1994; 76(5): 750–753.

26.

Amin

Speirs

Simmons

, et al. Total knee arthroplasty wound complication treatment algorithm: current soft tissue coverage options. J Arthroplasty 2019; 34(4): 735–742.

27.

Panni

Vasso

Cerciello

, et al.

Wound complications in total knee arthroplasty. Which flap is to be used? With or without retention of prosthesis?

Knee Surg Sports Traumatol Arthrosc 2011; 19(7): 1060–1068.

28.

Rao

Kempton

Erickson

, et al. Soft tissue reconstruction and flap coverage for revision total knee arthroplasty. J Arthroplasty 2016; 31(7): 1529–1538.

29.

Tetreault

Della Valle

Hellman

, et al. Medial gastrocnemius flap in the course of treatment for an infection at the site of a total knee arthroplasty. JBJS Essent Surg Tech 2017; 7(2): e14.

30.

Nahabedian

Mont

Orlando

, et al. Operative management and outcome of complex wounds following total knee arthroplasty. Plast Reconstr Surg 1999; 104(6): 1688–1697.

31.

Menderes

Demirdover

Yilmaz

, et al. Reconstruction of soft tissue defects following total knee arthroplasty. Knee 2002; 9(3): 215–219.

32.

Warren

Murtaugh

Lakra

, et al. Treatment of periprosthetic knee infection with concurrent rotational muscle flap coverage is associated with high failure rates. J Arthroplasty 2018; 33(10): 3263–3267.

33.

Houdek

Wagner

Wyles

, et al. Long-term outcomes of pedicled gastrocnemius flaps in total knee arthroplasty. J Bone Joint Surg Am 2018; 100(10): 850–856.

34.

Ries

Bozic

. Medial gastrocnemius flap coverage for treatment of skin necrosis after total knee arthroplasty. Clin Orthop Relat Res 2006; 446: 186–192.

35.

Theil

Stock

Gosheger

, et al. Gastrocnemius muscle flaps for soft tissue coverage in periprosthetic knee joint infection. J Arthroplasty 2020; 35(12): 3730–3736.

36.

Uckay

Bernard

. Gram-negative versus gram-positive prosthetic joint infections. Clin Infect Dis 2010; 50(5): 795.

37.

Parvizi

Azzam

Ghanem

, et al. Periprosthetic infection due to resistant staphylococci: serious problems on the horizon. Clin Orthop Relat Res 2009; 467(7): 1732–1739.

38.

Charalambous

Kim

Schwartz

, et al. Prosthetic knee infection with coagulase-negative staphylococcus: a harbinger of poor outcomes. J Arthroplasty 2022; 37(6S): S313–S320.

39.

Kheir

Tan

Higuera

, et al. Periprosthetic joint infections caused by enterococci have poor outcomes. J Arthroplasty 2017; 32(3): 933–947.

40.

Marculescu

Cantey

. Polymicrobial prosthetic joint infections: risk factors and outcome. Clin Orthop Relat Res 2008; 466(6): 1397–1404.

41.

Hsieh

Huang

Shih

. Prosthetic joint infection in patients with rheumatoid arthritis: an outcome analysis compared with controls. PLoS One 2013; 8(8): e71666.

42.

Casey

3rd Rebecca

Krochmal

, et al. Prophylactic flap reconstruction of the knee prior to total knee arthroplasty in high-risk patients. Ann Plast Surg 2011; 66(4): 381–387.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.19 MB