Management of distal humerus fracture-related infections after fixation: A retrospective study

Abstract

BACKGROUND

Distal humerus fracture-related infections (FRIs), particularly those involving the articular surfaces of the elbow, pose a significant challenge as they can lead to severe upper limb functional impairment and systemic complications. Distal humerus fractures represent about 2% of all fractures and approximately 30% of all elbow fractures. Such kind of infections are reported in literature with rates from 4% to 14.5% of distal humerus fractures treated with open reduction and internal fixation. We analyzed current evidence on distal humerus FRIs and report our clinical approach aimed at achieving both fracture healing and infection eradication.

AIM

To present a consecutive series of distal humerus FRIs, and to synthesize treatment strategies and to propose a decision-making algorithm.

METHODS

We performed a retrospective analysis of 43 patients with distal humerus FRIs treated at our institution between 2017 and 2022. Patients were stratified according to fracture healing status and time from index surgery (early/delayed < 6 weeks; late > 6 weeks). Treatment strategies included debridement, antibiotics, and implant retention (DAIR), implant removal, or suppressive antibiotic therapy. Primary outcomes were infection eradication and fracture union. Based on these variables, a treatment algorithm was developed.

RESULTS

Among patients with healed fractures (n = 6), implant removal and debridement resulted in infection eradication in all cases (100%). In early/delayed infections (n = 6), DAIR achieved both fracture union and infection control in 100% of cases. In patients treated with implant removal for unhealed fractures (n = 19), infection eradication and fracture healing were achieved in 14 cases (73.7%), while 5 patients (26.3%) required conversion to total elbow arthroplasty. In cases managed with implant retention and suppressive antibiotic therapy (n = 12), infection control was maintained until fracture union in all patients. Mean follow-up was 45 months. DAIR was effective in resolving acute and delayed infections. For late infections, both eradication and suppression yielded satisfactory outcomes.

CONCLUSION

The literature on the management of distal humerus FRI is limited. Although limited by the small sample size and absence of a control group, the manuscript describes our approach to managing distal humerus FRIs, which could contribute to improved patient outcomes.

Key Words: Fracture-related infection; Distal humerus fractures; Elbow; Debridement, antibiotics, and implant retention; Treatment algorithm

Core Tip: This study evaluates the management of distal humerus fracture-related infections in 43 patients, highlighting the effectiveness of tailored surgical and antibiotic strategies. Our results demonstrate that debridement, antibiotics, and implant retention is highly successful for early and delayed infections. For late infections or unhealed cases, we found that implant removal and suppressive therapy are effective alternatives. Based on these findings, we propose a clinical decision-making algorithm centered on fracture healing status and infection timing to guide surgeons in achieving both infection eradication and bone union.

INTRODUCTION

Fracture-related infection (FRI) is a serious clinical complication associated with substantial morbidity and increased healthcare costs[1]. Distal humerus fractures account for approximately 2% of all fractures[2] and represent about 30% of all elbow fractures[3,4]. Infection rates following open reduction and internal fixation of distal humerus fractures range between 4% and 14.5%, according to the literature[5,6].

Recent consensus definitions and diagnostic criteria have improved the identification of FRIs[7,8]; however, the vast majority of research has focused on infected non-unions of long bones such as the tibia and femur, which have distinct biological and biomechanical characteristics[9-11]. In contrast, there is a notable paucity of high-quality evidence addressing FRIs of the upper limb, and particularly those involving articular surfaces such as the distal humerus. This lack of specific data presents a significant clinical challenge, as the preservation of joint congruity and function is paramount in this anatomically complex region[2,12,13].

Given these gaps, current treatment recommendations for distal humerus FRIs are largely extrapolated from lower limb studies, which may not fully capture the unique considerations required for optimal management. This study aims to fill this critical knowledge gap by: (1) Presenting a consecutive series of complex distal humerus FRI cases treated at our institution; (2) Synthesizing current treatment strategies specifically relevant to distal humerus FRIs; and (3) Proposing a novel decision-making algorithm designed to improve clinical outcomes in this challenging patient population.

MATERIALS AND METHODS

Study design and patient population

We conducted a retrospective case series of patients treated for distal humerus FRI at our institution between January 2017 and December 2022. A total of 43 consecutive patients were included. Fractures were classified according to the Arbeitsgemeinschaft für Osteosynthesefragen/Orthopedic Trauma Association [classification[14] as type C1 (n = 3), C2 (n = 15), and C3 (n = 25)]. The mean patient age was 47.1 years (range, 25-70), and the cohort included 35 men and 8 women.

All patients had previously undergone surgical fracture fixation. Open reduction and internal fixation were performed using plates and screws in 35 cases, screws alone in 6 cases, and K-wires combined with screws or external fixation in 2 cases. Prior to FRI diagnosis, patients had undergone a mean of 1.7 surgical procedures (range 1-3; Table 1).

Table 1 Patients’ data (gender, mean age, fracture classification, method of fixation, number of previous surgeries, microbiology results, antibiotic therapy duration), median (interquartile range).

Overall	43 patients
Gender	35 males, 8 females
Mean age (years)	47.1 (25-70)
Fracture classification (AO/OTA)[14]
C1	3
C2	15
C3	25
Method of fixation
Plates and screws	35
Screws	6
Screws and K-wires	1
K-wires and external fixator	1
Number of previous surgeries	1.7 (1-3)
1	19
2	18
3	6
Microbiology results
Staphylococcus species	24
Streptococcus species	14
Escherichia Coli	5
Antibiotic therapy duration (weeks)	8.7 (6-12)

AO: Arbeitsgemeinschaft für Osteosynthesefragen; OTA: Orthopedic Trauma Association.

Microbiology and antibiotic therapy

Microbiological analysis identified Staphylococcus species in 24 cases, Streptococcus species in 14 cases, and Escherichia coli in 5 cases. The mean duration of antibiotic therapy was 8.7 weeks (range 6-12). An endo-venous broad spectrum antibiotic therapy was started in cases of high suspicion of FRI immediately after surgery. Subsequently, once received an antibiogram, a tailored oral or endo-venous antibiotic therapy was carried out according to the indications of the infectious disease specialist.

Definitions and variables

FRI diagnosis was established according to contemporary consensus criteria. Multiple deep tissue cultures were performed during surgery. Sonication was used on all the removed implants to improve diagnostic sensitivity. Histopathological analysis focusing on the presence of more than five polymorphonuclear neutrophils per high power field was also obtained. Patients were stratified based on: (1) Fracture healing status at presentation (healed vs unhealed); (2) Interval between index surgery and infection onset (early/delayed < 6 weeks; late ≥ 6 weeks); (3) Quality of articular reduction and joint congruency; and (4) Stability of the fixation construct.

Treatment strategy

Based on these variables, patients were managed according to four treatment pathways, including: (1) Implant removal with debridement and antibiotic therapy; (2) Debridement, antibiotics, and implant retention (DAIR); (3) Implant removal with or without revision fixation; and (4) Implant retention with suppressive antibiotic therapy. These pathways were integrated into a decision-making treatment algorithm developed from the analysis of the study cohort.

Outcomes

The primary outcomes were infection eradication and fracture union. Secondary outcomes included the need for additional surgical procedures and conversion to total elbow arthroplasty (TEA). Infection eradication was defined as the absence of clinical, laboratory, and radiological signs of infection at final follow-up. Tertiary outcomes were evaluation of post-operative range of motion (ROM), and functional patient reported outcomes scores, which were assessed at the time of the last follow-up. The Mayo Elbow Performance Score (MEPS)[15], the Oxford Elbow Score[16], the Quick Disabilities of the Arm, Shoulder and Hand (QuickDASH) score[17] and the Visual Analog Scale (VAS) score[18] were investigated.

Statistical analysis

Descriptive statistics were used to summarize patient demographics, clinical characteristics, and treatment outcomes. Continuous variables are presented as mean (interquartile range), while categorical variables are reported as absolute n (%). Comparisons between treatment groups (e.g., DAIR, implant removal, and suppressive therapy) were performed descriptively, given the limited sample size and heterogeneity of the cohort. All analyses were conducted using standard statistical software (XLSTAT 2024.3). A formal inferential statistical analysis was not performed due to the exploratory nature of the study.

Ethical approval

The study was approved by the Institutional Review Board of Garofalo Health Care (Approval No. 0028GHCIRB) and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants.

RESULTS

Healed fractures: Implant removal, debridement, and antibiotic therapy

In our cohort, 6 patients developed a distal humerus FRI after fracture union. They underwent implant removal, thorough debridement, and targeted antibiotic therapy. The mean age was 43.5 years (range 29-58), all patients were male, and all had previously been treated with plate-and-screw fixation. Fracture classification included 3 C2 and 3 C3 fractures. Microbiological cultures identified Staphylococcus species in 3 cases, Streptococcus species in 2 cases, and Escherichia coli in 1 case. Antibiotic therapy lasted 6 weeks in 5 patients and 8 weeks in 1 patient, achieving complete infection eradication in all cases. The mean follow-up for this group was 45 months (range 35-56 months). Average ROM in flexion-extension was 111.6° (range 90°-130°), considered a functional motion for most activities of daily living. Mean MEPS was 86.6 (range 80-90), mean Oxford Elbow Score was 40 (range 35-47), average QuickDASH was 15.5 (range 12.5-19.2) and mean VAS was 1.3 (range 1-2).

Unhealed fractures

DAIR: Six patients with early or delayed FRI underwent DAIR. The mean age was 42 years (range 28-61), with 3 C2 and 3 C3 fractures. All patients had previously received plate-and-screw fixation, and the infection was diagnosed after a single prior surgical procedure. Time from initial fixation to infection diagnosis was 4 weeks in 4 cases and 6 weeks in 2 cases. Pathogens included Staphylococcus species (4 cases), Streptococcus species (1 case), and Escherichia coli (1 case). Postoperative antibiotics were administered for 8 weeks in 5 patients and 12 weeks in 1 patient. Fracture union and infection eradication were achieved in all patients. Mean follow-up was 50 months (range 36-60 months; Figure 1). Average ROM in flexion-extension was 92.5° (range, 70°-105°), almost functional for daily living activities. Mean MEPS was 75.8 (range 70-80), mean Oxford Elbow Score was 40.3 (range 34-47), QuickDASH was 27.3 (range 22.5-30.8), mean VAS 2 (range 1-3).

Figure 1 Case 1. A: Anteroposterior and lateral radiographs, taken at the time of admission, showing an articular distal humeral fracture of a 29-year-old man; B: Postoperative radiograph taken after open reduction and internal fixation with a medial plate. The patient developed symptoms of acute fracture-related infection (local redness, pain and a fistula) 20 days after the open reduction and internal fixation; C: Healed fracture 8 weeks after a debridement, antibiotics and implant retention (debridement, antibiotics, and implant retention procedure).

Implant removal: Nineteen patients underwent implant removal for unhealed fractures. The mean age was 50 years (range 25-70). Fracture patterns included 1 C1, 6 C2, and 12 C3 fractures. Initial fixation methods were plate-and-screw (12 patients), screws only (5 patients), and K-wires with either screws or an external fixator (1 patient each). Thirteen patients had undergone 2 prior procedures, and 6 had undergone 3. 10 of these patients had a first DAIR procedure performed at another institution before the implant removal procedure, which was done at our institution. Time from initial fixation to FRI diagnosis was 4 weeks in 2 cases, 8 weeks in 12 cases, and 12 weeks in 5 cases. Pathogens included Streptococcus species (10 cases), Staphylococcus species (6 cases), and Escherichia coli (3 cases). Antibiotic therapy averaged 7.5 weeks (range 6-12). Fourteen patients achieved fracture healing and infection eradication after a single procedure, while 5 patients (26%) required conversion to TEA due to fragment resorption or failure to achieve a new stable fixation. Mean follow-up was 45 months (range 35-62 months; Figure 2). Average ROM in flexion-extension was 88.4° (range 70°-110°). Mean MEPS was 73.9 (range 55-85), mean Oxford Elbow Score was 39.6 (range 30-47), average QuickDASH was 26.8 (range 13.3-39.2), mean VAS 4.2 (range 2-6).

Figure 2 Case 2. A: Anteroposterior radiographs, taken at the time of admission, showing a displaced open articular distal humerus fracture and temporary stabilization with K-wires and an external fixator. After 10 days, open reduction and internal fixation was performed using two perpendicular distal humeral plates; B: Non-union with hardware failure 3 months after surgery, associated to high erythrocyte sedimentation rate and C-reactive protein blood levels; C: Fracture union after an implant removal and prolonged immobilization in a static posterior splint.

Implant retention with infection suppression: Twelve patients were treated with implant retention combined with prolonged antibiotic therapy to suppress infection until fracture union. The mean age was 47 years (range 32-68). Fracture classification included 2 C1, 3 C2, and 7 C3 fractures. Eleven patients had plate-and-screw fixation, and 1 had screws only. Ten patients had a single prior procedure, and 2 had undergone 2 procedures before FRI diagnosis. Infection was diagnosed at 4 weeks in 10 patients and 8 weeks in 2 patients. Pathogens included Staphylococcus species in 11 cases and Streptococcus species in 1 case. Antibiotic therapy averaged 11 weeks (range 8-12 weeks), successfully suppressing infection in all cases until fracture healing. Mean follow-up was 42 months (range 31-60). At the final follow-up, mean ROM in flexion-extension was 96.6° (range 90°-110°). The mean MEPS was 77.5 (range 60-85), the mean Oxford Elbow Score was 40 (range 35-46), average QuickDASH was 22.9 (range 13.3-35.8) and mean VAS was 3.6 (range 2-5).

DISCUSSION

Until recently a clear definition for the infections arising after fracture management with internal metal implants was not available[19]. As a result, the diagnostic criteria for FRI were often taken from periprosthetic joint infections[20]. Yet, appropriate terminology is crucial both for accurate diagnosis and treatment optimization[21]. In 2018, a consensus conference determined the confirmatory and suggestive criteria for FRIs[8], which were validated in 2022[22]. The confirmatory criteria include a fistula, sinus or wound breakdown; presence of pus or purulent drainage; phenotypically indistinguishable pathogens identified by culture from at least two separate deep tissue/implant specimens; and presence of more than five polymorphonuclear neutrophils per high power field on histological examination. The suggestive criteria comprise clinical signs, such as pain increasing over time, local redness and swelling, increased local temperature or fever; radiological and nuclear imaging signs; pathogenic organisms identified by culture from a single deep tissue/implant specimen; elevated serum infection markers (white blood cell count, erythrocyte sedimentation rate, C-reactive protein); persistent or increasing wound drainage; and new onset of joint effusion. The diagnosis of FRI thus involves a combination of clinical, imaging, microbiological, histological, and serological findings[23]. Elevated serum infection markers are suspicious but are not diagnostic for FRI, nor can low inflammatory marker levels exclude a low-virulence chronic FRI[24,25]. Likewise, radiological signs of FRI are uncommon, even in presence of radiolucent lines around the implant and of a periosteal reaction[26]. In addition, the clinical confirmatory criteria may not be present in up to 23% of patients with FRI[27]. Altogether, conventional diagnostic techniques are valuable, especially if used in combination as suggested by the new consensus definition of FRI[28].

FRI treatment strategies vary. The goal is to achieve both stable fracture union and to eradicate the infection[29]. FRI therapy hinges on a combination of antimicrobials and surgical treatment[30]. The surgical treatment rests on two mainstays: DAIR and debridement, antibiotics and implant removal/exchange[31]. In distal humerus fractures, conversion to TEA is performed in patients with extensive bone loss due to resorption or in those where a new stable fixation cannot be achieved[32]. In contrast to patients with periprosthetic joint infections, in these cases the plates and screws are merely temporary implants that ensure sufficient stability until fracture healing. In case of infection, after bone union is achieved, plates and screws can be removed. Analysis of the data reported above enabled us to devise our FRI treatment strategy. Below, we outline some typical FRI scenarios to illustrate its application (Figure 3). We were inspired by algorithms already present in literature[33-35] and applied existing concepts to the distal humerus FRIs.

Figure 3 Our treatment strategy for distal humerus fracture-related infections. TEA: Total elbow arthroplasty.

A healed fracture is the least complex situation: The hardware can be removed and antibiotic therapy begun[29]. The implants are sent to the laboratory for cultures and a targeted antibiotic therapy can then be started. Thorough debridement is critical to prevent infection recurrence[36]. This is considered as the gold standard approach in patients with a healed fracture[37]. An unhealed fracture is more difficult to treat. In these cases, we first evaluate the interval from surgery to FRI diagnosis and divide patients into those with early/delayed FRI (with infection arising within 6 weeks of surgery) and late FRI (where infection has developed more than 6 weeks from surgery). In patients with early/delayed FRI we consider the quality of the joint reduction and fixation. If both are acceptable, the patient can undergo DAIR and a targeted antibiotic therapy for 12 weeks, or until the fracture has healed[38]. In patients with an unacceptable reduction or fixation, we remove the hardware, stabilize the fracture with bone sutures, and place the elbow in a posterior static splint for the duration of the antibiotic therapy. Once the infection has resolved, the patient undergoes a new procedure with a bone graft, or TEA in case of a comminuted or necrotic fracture. In late FRIs, treatment is guided by three criteria: Whether the reduction of the articular surface is anatomical, whether the quality of fixation is stable or can be improved, and whether DAIR has already been performed. If the answers to these three questions are negative, we opt for eradication therapy, which involves hardware removal, debridement and temporary stabilization with bone sutures. Eventually, a further osteosynthesis procedure can be performed using a bone graft or TEA. If the answer to the three questions is affirmative, we can act to suppress the infection, opting for thorough debridement with improvement of the fixation, if necessary, and a prolonged antibiotic therapy, to help retain the hardware until fracture union. After fracture healing, the hardware can be removed and debridement performed again.

FRIs are a global burden[39,40]. The key to treatment is clearly a correct definition of the condition, and the criteria adopted by the 2018 international consensus conference on FRIs have provided a clear benefit[8,22,41]. A multidisciplinary approach with co-operation between the orthopedic surgeon, the infectious disease specialist and microbiologist is considered important and may be a positive factor influencing better outcomes[42,43]. Our study focuses on the treatment of FRIs of the distal humerus. Even though the elbow joint is at high risk of postoperative infection, due to its relatively thin soft tissue envelope and the shear forces acting on it[44], studies of FRIs of the distal humerus are limited. The incidence of this complication ranges from 1.4%[5] to 2.2%[45] and up to 9%[46,47]. In general, the treatment of FRI in patients with healed fractures is well established and consists of hardware removal with meticulous debridement of the infected area followed by a targeted antibiotic therapy[29,34,36]. If the fracture has not healed, the treatment is more complex. The first criterion is the interval from surgery to FRI development: If the infection is acute or delayed, DAIR[48] can resolve the infection and allows to retain the hardware until fracture healing. The rate of infection control with DAIR can be up to 94%, with different results depending on the type of fixation and the need for additional surgical procedures[49]. The presence of an intramedullary nail is a risk factor for failure of DAIR procedure of long bones[50] and implant fixation exchange with an external fixator can be an option for diaphyseal long bone FRIs[11]. Since the distal humerus is an articular surface, it is vital to restore joint congruency in the best way possible[51], and we feel that debridement with implant retention is a good strategy in such patients. The timing of DAIR is debated. According to a systematic review by Morgenstern et al[48], good outcomes can be obtained under 6 weeks from the trauma[52-55], whereas similar yet lower success rates can be achieved from 6 to 10 weeks after surgery[53,55,56]. Findings are similar for periprosthetic hip and knee joint infections, where there is a grey zone between 1 month and 3 months from surgery when DAIR may or may not be useful[57]. In the chronic setting, DAIR can be performed but outcomes are less reproducible and yield a lower success rate[33,55], hence it was not performed in our cohort of patients. In patients with a chronic infection, treatment can be by eradication or by suppression. In presence of inadequate anatomical reduction and fixation and of previous DAIR procedures, the possible solution may be to remove the hardware, debride, apply bone sutures to stabilize the fragments, and place the elbow in a splint to help fracture union. If consolidation is not achieved, a possible solution is a new fixation procedure or TEA, a useful salvage procedure after fixation failure[32,58]. If the FRI is chronic, but an anatomical reduction has been achieved, a good treatment option is debridement with fixation improvement (where necessary) and suppression antibiotic therapy until fracture healing. Functional outcomes in our series are consistent with those of previous series treating distal humerus FRIs[59,60]. Patients undergo limitations such as elbow stiffness in various cases, yet functional scores (MEPS, QuickDASH and VAS) in our cohort are acceptable in most of the cases, as in other literature reports[61,62].

This study has some strengths: First, it is a homogenous cohort of patients with FRIs of the distal humerus and all the patients were treated by the same senior surgeon applying the same algorithm. Moreover, it talks about a frequent complication of distal humerus fractures trying to advocate a useful algorithm for enhanced patient care.

However, this manuscript has some limitations: The retrospective design, the absence of a control group and the single center setting inherently limit our ability to establish causal relationships. Furthermore, the small sample size (n = 43) did not allow the authors to perform inferential statistical analysis, which could restrict the generalizability of the findings. Future prospective randomized controlled studies with larger cohorts are warranted to validate these preliminary results.

CONCLUSION

The literature on the management of elbow FRI is limited, particularly where the distal humerus is concerned. Based on our clinical experience, we describe our approach to managing distal humerus FRIs, which may contribute to improved patient outcomes.