Is immediate wound coverage possible in bite-related digital amputations?: a retrospective cohort study

Article information

Arch Hand Microsurg. 2026;31(1):1-9

Publication date (electronic) : 2025 March 27

doi : https://doi.org/10.12790/ahm.25.0034

Bong Gyu Choi, Soo Jin Woo, Kwang Hyun Park, Hyun Jae Nam, Sang Hyun Woo

W Institute for Hand and Reconstructive Microsurgery, W General Hospital, Daegu, Korea

Corresponding author: Sang Hyun Woo W Institute for Hand and Reconstructive Microsurgery, W General Hospital, 1632 Dalgubeol-daero, Dalseo-gu, Daegu 42642, Korea Tel: +82-53-550-5000 Fax: +82-53-552-4000 E-mail: handwoo303@gmail.com

Received 2025 September 22; Revised 2025 November 12; Accepted 2025 November 20.

Abstract

Introduction

Bite wounds caused by animals may result from dogs, cats, rats, wild animals, or humans; among these, dog bites are the most common, accounting for approximately 90%, followed by cat bites and human bites [1,2]. In the United States, more than 28,000 patients with dog bite injuries undergo surgical operations annually [3]. In Korea, the proportion of households raising companion animals has been steadily increasing due to declining marriage and birth rates [4], and consequently, the prevalence of bite injuries has also been rising [5].

In the treatment of bite wounds, infection is the most frequent and clinically significant complication. In a study of 5,248 patients with bite wounds, 500 patients underwent immediate wound closure by suture or surgery, and 75 of them developed infections, corresponding to an incidence of 15% [6]. For hand wounds specifically, an infection rate of 4.17% after primary closure with passive drainage has also been reported [7]. The general principles of bite wound management include thorough irrigation and empirical antibiotic administration (typically ampicillin/sulbactam plus cefoxitin) [1]. Tetanus prophylaxis is also recommended. Wounds are usually left open, with delayed wound closure performed only after swelling and erythema subside and the wound bed and tissue margins appear clean [1,8]. However, there is no universal consensus regarding all aspects of bite wound management. While most surgeons agree on the necessity of early and copious irrigation until the wound is adequately cleansed, considerable debate remains regarding immediate versus delayed wound closure and the optimal use of empirical antibiotics. Numerous studies have therefore addressed these issues [9].

There are relatively few studies concerning bite injuries that result in finger amputation, aside from a limited number of case reports [10-13]. This scarcity may be because amputations directly caused by bites are less common than those resulting from other mechanisms. When faced with a patient with finger amputation due to a bite injury, the surgeon must weigh multiple factors in deciding between replantation and alternative treatment options. A major concern in this setting is the risk of infection. Surgeons may assume that primary wound closure in bite-related amputations is inherently associated with a higher infection rate, especially given the complex anatomical structure of the hand and fingers, which involve enclosed spaces and intricate compartments that predispose to infection and complicate treatment.

Therefore, it is important to clarify the actual complications, particularly infection rates, associated with the treatment of bite-related amputations. At our institution, various treatment strategies have been applied to patients with amputations caused by bite injuries, and we sought to analyze these outcomes.

Methods

A retrospective chart review was conducted of patients who underwent surgical treatment for finger amputations caused by human or animal bites at our institute between January 2009 and December 2024. Data collected from medical records included demographic information, injury mechanism, level of amputation, types of surgery, and postoperative complications. Patients were divided into two groups based on the timing of wound coverage: those who underwent coverage at the initial surgery (immediate wound coverage group) and those whose wounds were left open after irrigation and subsequently covered at a secondary procedure (delayed wound coverage group). Major infection was defined as cases that required a return to the operating room for irrigation and debridement due to definitive signs, including purulent drainage, foul odor, or progressive wound deterioration. Replantation success was determined by preservation of finger length, regardless of the need for additional procedures (e.g., skin grafting or flap surgery) for partial necrosis.

All surgical procedures included thorough wound irrigation, aggressive debridement of necrotic tissue, prophylactic antibiotic administration, and tetanus immunization when indicated. The antibiotic protocol consisted of intravenous ampicillin–sulbactam (1.5 g in adults; 60 mg/kg in children) administered in three divided doses during hospitalization, followed by a 1-week course of oral amoxicillin–clavulanate (1,500/375 mg per day divided into three doses in adults; 40/10 mg/kg per day divided into three doses in children) after discharge. All patients were routinely hospitalized for 7 days and were discharged when they demonstrated clinical improvement, including the absence of fever and resolution or improvement of local symptoms such as pain, swelling, and erythema. At the time of discharge, intravenous antibiotics were transitioned to oral antibiotics. A few patients were discharged one to 2 days earlier or later based on personal circumstances at the patient’s request. For cases requiring replantation or flap surgery, hyperbaric oxygen therapy was provided during hospitalization, and dressings were changed every 2 to 3 days.

The major infection rate in the immediate wound coverage group was compared with the infection rate reported by Mandic and Yeboah [7] for primary closure of hand animal bite injuries. Effect size was quantified using Cohen’s h (0.2, small; 0.5, medium; and 0.8, large), and 95% CIs were calculated.

Functional outcomes were compared between the immediate coverage and delayed coverage groups. Grip and pinch strength were measured with the patient seated in a standardized posture, maintaining the shoulder in neutral alignment and the elbow flexed at 90°. The mean value from three consecutive trials was used for analysis. Grip and pinch strengths were quantified with a hand dynamometer and pinch gauge (E-Link system, Biometrics Ltd., Newport, UK). Grip and Pinch strength of the affected hand were compared with those of the contralateral hand. Sensory recovery was assessed by static two-point discrimination (2-PD) testing. Range of motion (ROM) was measured in all patients who underwent replantation.

All statistical analyses were performed using IBM SPSS Statistics ver. 26.0 (IBM Corp., Armonk, NY, USA). Continuous variables were compared using the Mann-Whitney U-test. Categorical variables were compared using the Fisher exact test. A p-value <0.05 was considered statistically significant.

Results

Among the 58 patients included in this study, 42 (72.4%) were male and 16 (27.6%) were female, with a mean age of 52.4 years (range, 7–91 years). The mean follow-up duration was 7.8±16.2 months (median, 4.0 months [interquartile range (IQR), 2.0–5.0]; range, 1–96 months). By anatomical distribution, the index and ring fingers were the most commonly affected, each accounting for 14 cases (24.1%), followed by the long finger in 11 cases (19.0%), the thumb in 10 cases (17.2%), and the small finger in nine cases (15.5%). Of the total cohort, 43 patients sustained complete amputations, whereas 15 patients had incomplete amputations. Analysis of injury mechanisms by digit revealed that dog bites were the most common (29 cases), followed by human bites (27 cases); cow and pig bites were observed in one case each (Table 1). All baseline characteristics showed no significant differences between groups. The mean duration of intravenous antibiotics was 6.89±0.64 days (median, 7 days [IQR, 7–7]; range, 5–8 days) in the immediate coverage group and 7.10±0.54 days (median, 7 days [IQR, 7–7]; range, 6–8 days) in the delayed coverage group, with no significant difference between groups (p=0.152). All patients underwent surgical treatment within 24 hours of injury. The mean time from injury to surgery was 13.77±4.08 hours (median, 13.82 hours [IQR, 11.52–15.88]; range, 3.80–20.57 hours) in the immediate coverage group and 14.51±4.09 hours (median, 15.52 hours [IQR, 12.93–17.23]; range, 3.45–21.82 hours) in the delayed coverage group, with no significant difference between groups (p=0.339).

Table 1.

Patients’ characteristics

Functional outcomes were evaluated in the immediate and delayed coverage groups. In the immediate coverage group, to reduce confounding attributable to digital nerve reconstruction, outcomes for patients who did not undergo replantation were compared with those of patients in the delayed coverage group. Outcomes for patients in the immediate coverage group who underwent replantation were analyzed separately.

Functional outcomes were thus assessed in 13 patients in the immediate coverage group excluding those who underwent replantation, and in 24 patients in the delayed coverage group. Static 2-PD <15 mm was achieved in 9 of 13 patients in the immediate coverage group (69.2%) and in 16 of 24 patients in the delayed coverage group (66.7%). Mean static 2-PD was 13.3 mm in the immediate coverage group and 14.1 mm in the delayed coverage group. In the immediate coverage group, the mean grip strength of the injured hand was 61.7% of the contralateral (uninjured) side; in the delayed coverage group, it was 66.5%. Mean pinch strength of the injured hand was 68.0% of the contralateral side in the immediate coverage group and 63.9% in the delayed coverage group. No statistically significant between-group differences were observed in static 2-PD, grip strength, and pinch strength (Table 2).

Table 2.

Functional outcomes in the immediate coverage (without replantation) and delayed coverage groups

In the immediate coverage group, functional outcomes were also assessed in six patients who underwent successful replantation. Static 2-PD <15 mm was achieved in all six patients (100%), with a mean static 2-PD of 10.5 mm. Mean grip and pinch strength of the injured hand were 73.2% and 70.8% of the contralateral side, respectively. According to Chen criteria for ROM assessment, five patients had excellent outcomes and one patient had a good outcome. Four patients reported cold intolerance (Table 3).

Table 3.

Functional outcomes in the replantation group

Of the 58 patients, 31 (53.4%) underwent massive irrigation and debridement with the wound left open at the primary surgery, followed by delayed wound coverage at secondary surgery. The remaining 27 patients (46.6%) underwent immediate wound coverage by means of replantation, flap surgery, or revision amputation. The surgical distribution of the immediate coverage group was as follows: replantation in 10 cases (37.0%), revision amputation in seven cases (26.0%), local flap in eight cases (29.6%), regional flap in one case (3.7%), and distant flap in one case (3.7%). In the delayed coverage group, revision amputation was performed in 14 cases (45.2%), local flap in nine cases (29.0%), distant flap in seven cases (22.6%), and skin graft in one case (3.2%). Replantation was attempted in 10 cases, with an 80% survival rate; all failures were attributable to vascular complications, with no infection-related loss. The first case of replantation failure involved an 80-year-old male who sustained a dog bite resulting in an incomplete amputation of the left small finger at Tamai zone IV. Except for a small portion of dorsal skin, the bone, tendon, nerve, and digital artery were all severely damaged. The patient had a medical history of hypertension and diabetes mellitus. Following the initial replantation, arterial insufficiency developed; however, as the patient requested revision amputation, further revascularization was not pursued and amputation was performed instead. The second failure case was a 45-year-old female with a complete fingertip amputation of Ishikawa zone I caused by a dog bite. Following replantation, arterial insufficiency developed, and subsequent debridement was performed. Coverage was achieved using a V-Y advancement flap (Table 4).

Table 4.

Surgical procedures and complications

Infection was not observed in either group. The 95% CIs for infection were 0% to 12.8% in the immediate coverage group and 0% to 11.2% in the delayed coverage group, with a pooled upper bound of 6.2% when all 58 patients were combined. Relative to the 4.17% infection rate following primary closure in hand bite injuries reported by Mandic and Yeboah, our observed 0% infection rate in the immediate wound coverage group corresponded to a Cohen’s h of –0.41.

Symptoms suggestive of minor infection, such as erythema, swelling, and pain, were observed in some patients. However, no definitive signs of infection (e.g., pus, foul odor) were observed, and all cases resolved with conservative management without additional surgery.

Excluding 13 patients in whom wound cultures were not performed, 45 patients remained for analysis. Among these, no organism was identified in 28 cases, while positive cultures revealed Serratia marcescens (n=2), methicillin-resistant Staphylococcus aureus (n=3), methicillin-resistant Staphylococcus epidermidis (n=5), Klebsiella aerogenes (n=2), Burkholderia cenocepacia (n=1), Acinetobacter baumannii (n=1), Chryseobacterium gleum (n=1), and Aeromonas sobria (n=2) (Table 5). Culture results typically required approximately 1 week to finalize. By that time, all patients had demonstrated satisfactory wound healing without definitive signs of infection under the empirical antibiotic regimen. Consequently, antibiotic therapy was not modified based on culture results in any case.

Table 5.

Aerobic and facultative anaerobic bacteria isolated from 45 patients with animal or human bite wounds

Discussion

The generally accepted principles of bite wound management include prompt and thorough irrigation, initiation of empirical antibiotic therapy, and favoring delayed over primary wound closure [9,14]. The rationale for favoring delayed closure is the risk that bacteria may become trapped within a closed space, exacerbating infection. Hand injuries, in particular, have higher infection rates than other sites due to the unique anatomical features of the hand, which allow even small wounds to spread infection aggressively, especially when tendon sheaths are involved [14].

Empirical antibiotics most commonly used are ampicillin/sulbactam in combination with cefoxitin. In cases of deep wounds or systemic signs of infection, piperacillin–tazobactam may be substituted [15]. However, as noted above, there is still no consensus regarding the use of empirical antibiotics and the timing of wound closure. Several studies have suggested that the most important factor influencing infection rates is the time interval between injury and hospital presentation, while antibiotic administration itself does not significantly affect outcomes. These studies emphasized that the timing of surgical interventions—including irrigation, removal of foreign bodies, debridement of infected material, and excision of nonviable tissue—plays a critical role [9,15]. The question of whether bite wounds should undergo immediate or delayed primary closure also remains controversial. Although delayed primary closure is generally preferred, some authors have advocated immediate loose closure in cases of large gaping wounds [9,16].

The risk of infection also differs by the animal responsible for the bite. Cat bites often take the form of puncture wounds, with bacteria penetrating deeply through sharp teeth [17,18]. These wounds are difficult to irrigate effectively and are often underestimated and treated with simple dressings, leading to high infection rates. “Fight bites” may lead to deeper infections because tendon gliding can transport contaminants into deep spaces [19]. By comparison, amputations caused by human or dog bites present with widely open wounds that are easier to irrigate, potentially lowering infection risk. Moreover, in amputations, the proximal stump of the tendon retracts proximally, while the distal part loses active movement, reducing the risk of deep-space infection compared to fight bites, in which tendon position varies with open or clenched hand posture. To validate this hypothesis, an important area for future investigation is the comparison of infection rates between bite-related amputations and non-amputated bite wounds. While our series demonstrated no major infections in bite-related amputations, direct comparative data with non-amputated bite injuries are lacking in the literature. Future comparative studies examining infection rates between these injury patterns, stratified by timing of closure, would provide valuable evidence to clarify whether the unique characteristics of amputation wounds indeed confer a protective effect against infection.

At our institution, a variety of treatment methods have been applied to finger amputations caused by bite injuries. From 2009 to 2010, all patients were treated with debridement and open dressing, followed by delayed wound coverage using revision amputation, flap surgery, or skin grafting (Fig. 1). In 2011, immediate wound coverage was performed for the first time, and its use has gradually increased (Figs. 2, 3). Against this background, we analyzed our institutional series to evaluate whether immediate coverage indeed increased infection risk. As shown in our results, no major infections requiring irrigation and debridement were observed in patients who underwent immediate wound coverage.

Fig. 1.

(A–C) A 68-year-old man sustained a dog bite–induced amputation of the right long finger at Tamai zone I; the amputated part could not be found. Methicillin-resistant Staphylococcus aureus was cultured from the wound. (D) Open dressing was performed for 7 days. (E, F) Delayed wound coverage was achieved using a thenar flap elevated from the ipsilateral thenar region. (G, H) Clinical photographs at the 7-year postoperative follow-up.

Fig. 2.

(A–C) A 25-year-old woman sustained an amputation of the left small finger at Tamai zone I during an altercation with another person. She presented without the amputated part, and the distal phalangeal bone was slightly damaged. Wound culture yielded no identifiable organisms. (D, E) Immediate wound coverage was achieved using a V-Y advancement flap elevated from the volar aspect of the left small finger. (F) Clinical photograph at the 8-year postoperative follow-up.

Fig. 3.

(A–C) A 48-year-old woman sustained an incomplete amputation of the left ring finger at Tamai zone II after being bitten by a dog. Chryseobacterium gleum was cultured from the wound. (D–F) After massive irrigation, immediate replantation was performed. Following replantation, the wound healed without the need for any additional procedure. (G, H) Clinical photograph at the 1-year postoperative follow-up.

Using 95% CIs and effect size analysis with Cohen’s h, we evaluated the clinical significance of the absence of infection in our series. Although no infections were observed, the 95% CIs remained wide (0–12.8% in the immediate coverage group and 0–11.2% in the delayed coverage group; pooled upper bound 6.2%), indicating that a small risk of infection cannot be completely excluded. Compared with the 4.17% infection rate reported for primary closure with drain in hand bite injuries, our observed rate of 0% corresponded to a Cohen’s h of –0.41 (|h|=0.41, small-to-moderate). These findings support the safety of immediate wound coverage in bite-related finger amputations.

We found no statistically significant differences in functional outcomes—grip strength, pinch strength, or static 2-PD—between the immediate coverage and delayed coverage groups. A possible explanation is the uniformly low burden of infectious complications in both groups. Infection is a major driver of poor hand function through prolonged immobilization, additional debridements, soft tissue scarring, tendon adhesions, and joint stiffness. However, the analysis may be underpowered to detect small between-group differences, and unmeasured confounders (e.g., injury severity, contamination burden, amputation level, or flap selection) could still influence outcomes. Future studies with larger samples and stratification by replantation status, amputation level (e.g., Tamai zone), and contamination severity are warranted to validate these observations.

These results provide clinically relevant insights that may guide surgical decision-making in the management of bite-related digital amputations. In practice, surgeons are often reluctant to attempt replantation or immediate coverage in bite-related amputations due to presumed high infection risk. However, amputations present the unique feature of a completely open wound, which facilitates irrigation and may potentially reduce infection risk. Our study provides data suggesting that such procedures are feasible without increasing major infection rates, provided that essential surgical principles are followed. This consideration should be taken into account when determining the treatment strategy.

This study has some limitations. First, infection rates may have been underestimated because only cases requiring surgical intervention were classified as major infections. Unlike typical bite wounds, amputation wounds usually involve bone surgery and multiple procedures, which can cause swelling, pain, and erythema even in the absence of infection. Nevertheless, no case demonstrated unequivocal signs of infection such as pus or foul odor. Second, the choice between immediate and delayed wound coverage was based on the operating surgeon’s judgment rather than strict criteria, raising the possibility that immediate closure was more often selected in cases perceived to have a lower infection risk. Third, the immediate closure group included patients treated with a heterogeneous set of procedures—replantation, revision amputation, and various flaps. Subgroup analysis was not feasible due to small sample sizes, but this limitation may be addressed in future studies with larger cohorts. Despite these limitations, our findings provide clinically relevant insights into the management of bite-related amputations. Fourth, multivariable analysis is valuable in identifying independent risk factors by adjusting for potential confounding variables. However, in the present study, no postoperative infections occurred in either the immediate or delayed coverage groups; therefore, multivariable modeling was not feasible.

Conclusion

In this cohort study of 58 patients with bite-related finger amputations, no major infections were observed among the 27 patients who underwent immediate wound coverage. While a few patients experienced minor infections manageable with antibiotics, no severe infections requiring surgical intervention occurred. These results clearly demonstrate that immediate wound coverage—including replantation when indicated—can be performed safely in bite-related finger amputations, challenging the conventional hesitation toward such procedures and reinforcing their feasibility as a reliable treatment option. This evidence provides strong support for hand surgeons to consider immediate closure or replantation without undue concern for infection risk in appropriately managed cases.

Article information Continued