Cutaneous Escherichia coli infection and necrosis after primary total knee arthroplasty: a case report

Abstract

Introduction:

Primary total knee arthroplasty (TKA) is common and increasingly performed. Superficial surgical site infections, though rare, risk progressing to deep periprosthetic joint infection, posing a serious management challenge.

Case presentation:

An 81-year-old woman underwent primary TKA for osteoarthritis. Three months postoperatively, she developed an 8 × 10 cm infected necrotic wound at the incision site after self-applying nonstandard topical agents. A multidisciplinary team performed staged debridement and reconstruction, achieving successful infection control and implant preservation.

Discussion:

This case illustrates a severe superficial infection progressing to extensive soft tissue necrosis due to inappropriate self-treatment. The successful outcome was achieved through timely and coordinated surgical interventions by a multidisciplinary team, highlighting the critical need for structured protocols in managing such complex scenarios. This approach effectively prevented the progression to periprosthetic infection and preserved the functional implant.

Conclusion:

Multidisciplinary collaboration is essential for successfully managing complex post-TKA wound infections, providing a valuable framework for future high-risk orthopedic cases.

Introduction

Primary total knee arthroplasty (TKA) has become a well-established and effective treatment for end-stage osteoarthritis, and its safety and efficacy have been validated through extensive clinical evidence. Globally, more than 1.5 million patients undergo knee arthroplasty annually, a number that is projected to increase with increasing population aging and increasing obesity rates^[1^]. Despite the low incidence of postoperative infections (approximately 0.39%) following primary TKA, superficial surgical site infections (SSIs) are exceptionally rare but clinically consequential, predominantly predominantly within 30 days postoperatively^[2,3^]. In this case, a patient developed cutaneous infection and extensive cutaneous necrosis at the surgical incision site, triggered by inappropriate application of topical herbal patches after primary TKA. Such complications, if mismanaged, may lead to catastrophic outcomes. Studies have demonstrated that patients with impaired wound healing exhibit a 7.5-fold higher risk of deep infection compared to those with normal healing (6% vs. 0.8%)^[4^]. Moreover, periprosthetic joint infection (PJI) is associated with significant morbidity, mortality, and healthcare resource utilization. Specifically, PJI increases mortality rates to 7% in the elderly patients, while treatment costs are 4 times higher than primary arthroplasty^[5,6^].

HIGHTLIGHTS

• This case demonstrates that even extensive wound necrosis after primary total knee arthroplasty can be managed to avert periprosthetic joint infection and salvage the prosthesis through timely and structured intervention.

• It underscores the unique advantage of a staged, multidisciplinary strategy in achieving both infection control and functional preservation in complex soft-tissue defects.

Despite these risks, standardized protocols for managing post-TKA wound infections complicated by skin necrosis remain lacking. The balance between infection control, functional recovery, and cost-effectiveness remains unresolved. Herein, we present a rare case of Escherichia coli-induced cutaneous infection and extensive necrosis following improper topical medication use after primary TKA. The patient underwent staged management, including initial debridement combined with vacuum sealing drainage (VSD) and targeted antibiotic therapy, followed by secondary stamp skin grafting for wound closure, achieving favorable outcomes. This case highlights the importance of physician–patient collaboration in minimizing infection risks and provides a reference for staged strategies in managing complex cutaneous complications post-TKA. This case report has been reported in line with the SCARE checklist^[7^].

Presentation of case

An 81-year-old woman was evaluated at our hospital on 13 December 2024. Her chief complaint was a 6-month history of progressive right knee pain, swelling, and limited range of motion (ROM). Her medical history included hypertension managed with oral antihypertensive agents for over 20 years. Physical examination revealed right knee joint swelling, tenderness, and limited ROM (10°–90° flexion). Preoperative radiographs (Fig. 1A) demonstrated right knee joint space narrowing, osteophyte formation, and subchondral sclerosis, consistent with radiographic features of end-stage osteoarthritis (Kellgren-Lawrence grade IV). After excluding surgical contraindications, primary right TKA was performed under spinal anesthesia via a 12-cm standard anterior midline skin incision (12 cm in length) and medial parapatellar approach. The procedure lasted 120 min with an estimated blood loss of 100 mL. Perioperative antimicrobial prophylaxis with intravenous cefazolin was administered. Multimodal analgesia, low-molecular-weight heparin for thromboprophylaxis, proton pump inhibitors for gastric protection, and daily sterile wound care were implemented postoperatively. The incision achieved Grade A healing (no erythema, discharge, or dehiscence) by postoperative day 7. The right knee ROM had recovered to 0–110°, and postoperative radiographs confirmed proper implant positioning (Fig. 1B). The patient was discharged with instructions for outpatient follow-up.

Figure 1.

Radiographs during the patient’s primary TKA. (A) Preoperative X-ray of the right knee. (B) Right knee radiograph 7 days after surgery.

Three months post-discharge, due to persistent local discomfort and itching at the incision site, the patient self-applied plasters of unspecified composition (reportedly herbal patches). This prompted progressive erythema, pain exacerbation, purulent discharge, and necrotic changes. On 1 March 2025, she was readmitted with right knee swelling and pain. On admission, her temperature was 36.2°C. Physical examination revealed localized warmth, an 8 cm × 10 cm elliptical ulceration over the midline incision (Fig. 2A) with black necrotic eschar and purulent exudate, and restricted active ROM (10–100° flexion). Laboratory tests showed mildly elevated inflammatory markers: white blood cell count 7.69 × 10⁹/L, high-sensitivity C-reactive protein (HsCRP) 4.74 mg/L, and erythrocyte sedimentation rate (ESR) 29 mm/h. Based on the patient’s history, wound characteristics, and laboratory findings, necrotizing wound features, a preliminary diagnosis of SSI with extensive soft tissue necrosis was made.

Figure 2.

Clinical photographs of the patient during readmission. (A) Large skin infection with necrosis found after primary TKA. (B) The trauma condition after the first debridement and VSD. (C) Postoperative trauma condition after the second debridement and VSD. (D) The wound condition after the first stamp skin grafting. (E) The wound condition after the second stamp skin grafting. (F) The wound condition 90 days after discharge.

On the third day following readmission, the patient underwent surgical debridement under combined spinal-epidural anesthesia. Intraoperatively, thorough debridement of the black eschar and necrotic tissue on the surface of the right knee incision was performed. We found the fascial layer was intact, and there was no exposure of the joint capsule, prosthesis, or bone. The wound bed consisted of healthy, bleeding subcutaneous tissue and fascia after debridement. The wound was meticulously irrigated sequentially with sterile saline, chlorhexidine, iodophor, and hydrogen peroxide to ensure complete removal of infected and necrotic tissue. After satisfactory debridement, a VSD-negative pressure device was implanted, and its airtightness was confirmed. Intraoperative wound cultures and sensitivity tests were performed, which revealed Escherichia coli positivity with susceptibility to ticarcillin/clavulanate, piperacillin/tazobactam, and ceftazidime. Based on the sensitivity results, the antimicrobial regimen was adjusted to intravenous tazobactam 4.5 g every 8 h.

After 7 days of continuous VSD therapy, the patient returned to the operating room for dressing change and reassessment. Upon removal of the dressing, the wound exhibited active granulation tissue proliferation and clearer demarcation of necrotic margins (Fig. 2B). A second debridement was performed, followed by reapplication of the VSD system, which remained in situ for an additional 7 days. Upon removal of the VSD, the wound bed was assessed as pink granular granulation tissue without purulent exudate (Fig. 2C), indicating readiness for skin grafting.

Subsequently, the patient underwent repair of the chronic ulcer on the right knee under spinal anesthesia. The wound margins were thoroughly trimmed to normal dermal tissue using a No. 15 scalpel. Hemostasis was achieved with electrocautery, followed by copious irrigation with hydrogen peroxide and normal saline. A split-thickness skin graft (0.04 cm in thickness) was harvested from the lateral thigh using a No. 11 blade knife under local anesthesia, measuring 4 cm × 4 cm. After ensuring no significant bleeding at the donor site, it was covered with sterile gauze and secured under pressure. The harvested graft was then meshed and cut into 1 cm × 1 cm postage-stamp–sized pieces. These stamps were evenly distributed over the prepared wound bed and secured with a nonadherent dressing followed by compressive gauze padding. On postoperative day 11, graft take exceeded 90%, and the wound had contracted to 2 cm × 3 cm (Fig. 2D), prompting a second stamp-grafting procedure. The residual wound measuring 2 cm × 3 cm subsequently underwent a second stamp skin grafting procedure. Under spinal anesthesia, the ulcer bed was prepared with debridement using a No. 15 scalpel, followed by hemostasis and irrigation with hydrogen peroxide and saline. A split-thickness skin graft of corresponding size (2 cm × 3 cm) was harvested from the thigh under local anesthesia using a No. 11 blade. The graft was meshed into 1 cm × 1 cm stamps and applied to the wound bed, secured with a compressive dressing. The transient rise in HsCRP and ESR observed following this surgical intervention (Table 1) was consistent with the expected systemic inflammatory response to operative trauma, and subsequently trended downward with progressive wound healing. One week after this final grafting, the wound achieved complete epithelialization, meeting Grade A healing criteria (Fig. 2E). At the 3-month follow-up, the skin had healed completely, with only minimal, pliable scar tissue that did not restrict joint mobility or function (Fig. 2F). The right knee joint regained an active ROM from 0° to 110°, with no restrictions in passive movement. As shown in Table 1, inflammatory markers had normalized, indicating resolution of infection and successful wound healing.

Table 1.

The blood biochemical indicators of patient in different periods.

Parameter	At admission	Preprimary grafting	Postsecondary grafting	One-week follow-up
WBC (109/L)	6.01	7.65	5.48	7.50
NEUT%	54.80	61.00	46.80	49.40
LYMPH%	30.90	22.20	30.50	35.30
MONO%	13.30	12.40	16.10	11.70
EO%	0.70	4.10	5.50	3.10
BASO%	0.30	0.30	1.10	0.50
NEUT# (109/L)	3.29	4.67	2.57	3.70
LYMPH# (109/L)	1.86	1.70	1.67	2.65
MONO# (109/L)	0.80	0.95	0.88	0.88
EO# (109/L)	0.70	0.31	0.30	0.23
BASO# (109/L)	0.02	0.02	0.06	0.04
HsCRP (mg/L)	1.60	40.10	7.50	6.14
ESR (mm/h)	29	69	62	64

WBC, white blood cell; NEUT%, neutrophil percentage; LYMPH%, lymphocyte percentage; MONO%, monocyte percentage; EO%, eosinophil percentage; BASO%, basophil percentage; NEUT#, absolute neutrophil count; LYMPH#, absolute lymphocyte count; MONO#, absolute monocyte count; EO#, absolute eosinophil count; BASO#, absolute basophil count. HsCRP, high-sensitivity C-reactive protein; ESR, erythrocyte sedimentation rate.

All surgical interventions were jointly performed by a team of experienced attending physicians from both orthopedic and plastic surgery services. Written informed consent was obtained from the patient and legally authorized family members for the publication of this case report, including the use of de-identified clinical photographs and radiographic images for academic and educational purposes.

Discussion

Primary TKA is a highly effective procedure for managing end-stage knee osteoarthritis^[8^]. The introduction of patient-specific instrumentation, aimed at enhancing surgical precision, holds theoretical promise for reducing postoperative complications^[9^]. However, wound complications remain a significant challenge that can jeopardize implant longevity and patient outcomes. These complications, including impaired wound healing, SSIs, thromboembolic events, periprosthetic fractures, and restricted joint mobility, are associated with substantial morbidity and resource utilization. Among these, uncomplicated wound healing is a critical prognostic indicator for the success of primary TKA. Persistent wound drainage, epidermal dissolution, or full-thickness skin necrosis often signals an elevated risk of infection, which may culminate in catastrophic consequences such as implant failure and PJI^[2,10^]. Clinical data indicate that wound complications can increase the incidence of deep infection to 6% within 2 postoperative years and necessitate major surgical interventions, including revision or amputation, in up to 5.3% of cases^[11^]. While traditional inflammatory markers are commonly utilized to monitor for complications, their limitations – primarily due to their normal elevation in response to surgical trauma – underscore the need for a more nuanced approach. As emphasized by Conteduca et al, a combination of clinical suspicion and serial assessment of markers such as CRP and ESR is vital for identifying atypical patterns suggestive of developing complications^[12^]. In practice, this underscores the necessity for healthcare providers to maintain a heightened index of suspicion throughout the postoperative period. This case underscores the critical importance of patient education and early clinical vigilance. As demonstrated here, the patient experienced wound discomfort postoperatively and resorted to self-administered nonstandard topical agents, which precipitated a severe soft-tissue infection. Therefore, it is essential for healthcare providers to clearly instruct patients during hospitalization to seek immediate medical evaluation for any persistent wound drainage, erythema, pain, or signs of skin breakdown. Such proactive monitoring not only aids in the early exclusion of infection but also enables timely intervention – exemplified in this case by prompt debridement and staged reconstruction. Importantly, although the infection and necrosis were extensive, timely management prevented its penetration into the deep fascial layer and joint space, thereby averting a PJI and successfully preserving the prosthesis.

The management of wound complications necessitates a multifaceted approach tailored to factors such as wound size, contamination, infection status, and exposure of muscle, bone, or prosthetic components. In the present case, a patient developed extensive skin necrosis 3 months postoperatively due to superficial wound infection, without evidence of deep tissue exposure or deep infection. Conservative measures, including targeted antibiotic therapy guided by pathogen identification from superficial swabs, were effective in resolving the infection^[10,13^]. Subsequent wound care emphasized meticulous local wound management, frequent dressing changes, and debridement under sterile conditions to remove necrotic and nonviable tissue, thereby promoting secondary healing. Notably, studies have demonstrated that oral antibiotic therapy for superficial wound infections does not increase the long-term risk of PJI, as evidenced by 70-month follow-up data^[14^]. However, secondary healing requires adequate skin coverage over muscle and fascia. In this case, extensive skin necrosis precluded primary closure, necessitating staged wound management. Negative pressure wound therapy via VSD was employed to stabilize the wound and promote granulation tissue formation, creating a favorable environment for subsequent skin grafting.

Traditional split-thickness skin grafting, though reliable for wound closure, imposes functional trade-offs^[15^]. Restricting early knee mobility to prevent graft shear forces risks postoperative stiffness and contractures, particularly detrimental in elderly patients prioritizing functional recovery^[16^]. To address these challenges, stamp skin grafting – a technique involving segmental transplantation of split-thickness skin grafts – was utilized^[17–19^]. This method optimizes donor site utilization, minimizes donor site morbidity, and promotes healing, particularly in irregular or infected wound beds. Importantly, in a case with significant prior infection such as this, staged grafting permits close observation of graft take after each procedure. This is especially critical in elderly patients with potentially compromised healing capacity, as it enables timely adjustment of the treatment plan and reduces the risk of complete graft failure. Given the patient’s age (81 years), the irregular and infected nature of the wound, and the need to preserve knee joint mobility, stamp grafting was deemed the most appropriate approach. The patient achieved over 90% graft survival 11 days postoperatively, with complete wound coverage achieved after a second stamp grafting procedure. At discharge, the wound demonstrated Grade A healing, and knee joint mobility was fully restored.

This case underscores the importance of a tiered therapeutic strategy for managing complex post-TKA wound complications. Timely intervention with debridement, VSD, and staged stamp grafting effectively mitigated infection-related risks and minimized the potential for joint contracture. Notably, early surgical intervention, combined with VSD and stamp grafting, facilitated timely wound closure and functional recovery, even in the presence of extensive necrosis and infection. These findings suggest that this approach may represent a viable alternative to traditional methods, particularly in elderly patients with complex comorbidities.

While this case demonstrates a successful multidisciplinary approach, several limitations must be acknowledged. Primarily, as a single-case report, the generalizability of the findings is inherently limited. The favorable outcome, though encouraging, requires validation through larger, prospective cohort studies or clinical trials. Additionally, the follow-up period, while sufficient to demonstrate complete wound healing and functional recovery, remains relatively short for assessing long-term implant survivorship and scar maturation. Therefore, future research should focus on validating the efficacy and durability of this staged protocol through multicenter studies, particularly in elderly patients with complex comorbidities. Further investigation into the specific role of stamp skin grafting in contaminated or irregular wound beds, as well as its impact on joint mobility and long-term functional outcomes, is also warranted. Such studies would help establish standardized, evidence-based protocols for managing complex soft-tissue complications after TKA and ultimately improve patient care.

Ethical approval

This study was conducted in accordance with the Declaration of Helsinki. Informed consent was obtained from all participants.

Consent

Written informed consent was obtained from the patient for publication of this case report and accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal on request.

Sources of funding

This work was financially supported by the Guangdong Basic and Applied Basic Research Foundation (2023B1515020007), Macau Foundation for Development of Science and Technology (0008/2024/RIA1), Science and Technology Projects in Guangzhou (2024A04J6408), the Joint Funds of the National Natural Science Foundation of China (U24A6013), National Natural Science Foundation of China (82072470), Guangdong Provincial Medical Research Fund (A2024412, B2023328, B2025118), and GDSS2024N017 (Science and Technology Research Project of Guangdong Provincial Sports Bureau).

Conflicts of interest disclosure

The authors declare no competing interests.

Research registration unique identifying number (UIN)

Not applicable.

Author contributions

The authors confirm their contribution to the paper as follows: Q.D. and T.F.Z. were responsible for data collection and reporting. The initial drafting and critical revision of the manuscript were performed by W.W.X. H.T.Z. and S.H.L. were involved in the study conception, planning, and provided final approval of the version to be published. All authors agreed to be accountable for all aspects of the work.

Guarantor

Tengfeng Zhuang, Shenghong Li, Huan-Tian Zhang.