A hospital-based retrospective cross-sectional study was conducted to compare preventive antibiotic regimens and to determine the prevalence of fracture-related infection (FRI) among patients with open long-bone fractures managed at two referral hospitals in the North West Region of Cameroon. The study involved a review of medical records of eligible patients treated between 1 January 2023 and 31 January 2024.

The study was conducted at Nkwen Baptist Hospital (NBH) and The People’s Clinic (TPC), both located in Bamenda, North West Region of Cameroon. These institutions are among the major referral centers that provide emergency orthopedic and trauma care to patients from the North West Region and neighboring areas.

Nkwen Baptist Hospital is a well-established faith-based tertiary healthcare institution with a specialized orthopedic unit managing both emergency and elective trauma cases. The People’s Clinic is a private referral hospital that also provides comprehensive orthopedic services, including emergency fracture management, surgical stabilization, and postoperative follow-up.

During the study period, the two hospitals adopted different empirical antibiotic prophylaxis protocols for open fractures. Nkwen Baptist Hospital routinely administered cloxacillin monotherapy, whereas The People’s Clinic predominantly used a combination regimen of ceftriaxone, gentamicin, and metronidazole.

The study population comprised all patients diagnosed with open long bone fractures who received definitive management at either study hospital during the study period.

1) Inclusion Criteria

Patients were eligible for inclusion if they had a confirmed diagnosis of an open long bone fracture, were managed primarily at Nkwen Baptist Hospital or The People’s Clinic, had complete medical records documenting fracture characteristics, antibiotic regimen, surgical management, and treatment outcomes, and completed adequate follow-up to determine the presence or absence of fracture-related infection.

2) Exclusion Criteria

Patients were excluded if they had incomplete or missing medical records, received definitive treatment at another health facility before referral, had pathological fractures, had pre-existing chronic osteomyelitis before sustaining the fracture, or were lost to follow-up before their fracture-related infection status could be determined.

3) Follow-UpandAscertainmentofInfection Status

Patients were required to have a minimum documented follow-up period of 12 weeks (3 months) after definitive fracture management to permit assessment of fracture-related infection. The median follow-up duration was 6 months (IQR: 4 - 9 months). Infection status was determined through review of inpatient records, operative notes, wound assessment findings, microbiological results where available, records of additional antibiotic treatment or surgical debridement, and outpatient follow-up documentation. Patients were classified as having an infection when documented clinical and/or confirmatory evidence of infection was identified during follow-up. Patients without documented infection were classified as infection-free only if they had completed at least 12 weeks of follow-up without clinical or confirmatory evidence of infection. Patients lost to follow-up before 12 weeks and for whom infection status could not be reliably established were excluded from the analysis.

A census sampling approach was adopted. All eligible patient records meeting the inclusion criteria during the study period were reviewed. Among 450 long bone fractures managed during the study period, 94 open fractures fulfilled the eligibility criteria and were included in the final analysis.

1) Dependent Variable

The primary outcome variable was the occurrence of fracture-related infection (FRI) following management of an open long bone fracture.

2) Independent Variables

The independent variables included patient demographic characteristics (age and sex), hospital of treatment, mechanism of injury, bone involved, Gustilo-Anderson fracture grade, time from injury to hospital presentation, time to initiation of antibiotic therapy, time to surgical debridement, antibiotic prophylaxis regimen, duration of antibiotic therapy, method of fracture stabilization, presence of associated injuries, and documented comorbidities.

3) Operational Definitions

Fracture-related infection (FRI) was classified in accordance with the international consensus definition proposed by Metsemakers et al. [7]. Confirmatory criteria included the presence of a fistula, sinus, or wound breakdown communicating with the fracture or implant; purulent drainage or intraoperative pus; microbiological confirmation from appropriately collected deep tissue or implant specimens; or histopathological evidence consistent with infection. Clinical, radiological, or laboratory findings that were suggestive but not confirmatory were considered supportive evidence requiring further investigation. Superficial wound infections limited to the skin and subcutaneous tissues, without evidence of communication with or involvement of the fracture, were classified separately and were not considered confirmed FRI.

Data were extracted retrospectively from patients’ medical records using a structured data extraction form informed by established guidance for open-fracture management and FRI diagnosis [7]-[10]. Information collected included demographic characteristics, date and mechanism of injury, anatomical location of the fracture, Gustilo-Anderson fracture grade, initial wound management, timing of antibiotic administration, prescribed antibiotic regimen, timing of surgical debridement, method of fracture stabilization, duration of hospitalization, clinical evidence of fracture-related infection, and patient outcomes at follow-up.

Data collection was conducted by trained investigators to ensure consistency and reliability. All extracted information was cross-checked against operative notes, inpatient records, and outpatient follow-up files to verify accuracy, resolve discrepancies, and maximize dataset completeness.

The primary outcome was the prevalence of fracture-related infection (FRI) among patients with open long bone fractures. Secondary outcomes included a comparison of infection prevalence between the two antibiotic prophylaxis regimens, an assessment of the association between Gustilo-Anderson fracture grade and the occurrence of FRI, and a description of the distribution of open fractures by patient demographic characteristics and anatomical location.

Data were entered into Microsoft Excel, cleaned, coded, and exported to IBM Statistical Package for the Social Sciences (SPSS) version 23.0 for statistical analysis. Continuous variables were summarized using means and standard deviations or medians with interquartile ranges, depending on data distribution. Categorical variables were summarized using frequencies and percentages. Associations between categorical variables were assessed using the Chi-square test or Fisher’s exact test where appropriate. Odds ratios (ORs) with 95% confidence intervals (95% CI) were calculated to estimate the strength of associations. Variables with a p-value < 0.05 were considered statistically significant. Candidate variables for multivariable logistic regression were selected based on clinical relevance and/or evidence of association in bivariable analysis. Because only 19 infection events occurred, the number of predictors that could be reliably evaluated was limited. Accordingly, the multivariable analysis was considered exploratory, and adjusted odds ratios were interpreted cautiously because of the possibility of model overfitting and imprecise estimates.

Ethical approval for this retrospective study was obtained from the appropriate Institutional Review Board/Ethics Committee prior to data collection. Permission to access patients’ medical records was also obtained from the administrative authorities of the participating hospitals. As the study involved a retrospective review of existing medical records with no direct patient contact or intervention, the requirement for individual informed consent was waived by the ethics committee in accordance with applicable national and institutional regulations.

Data were extracted exclusively from hospital records using a standardized data extraction form by trained investigators. To ensure data quality, extracted information was cross-checked against operative notes, inpatient records, and outpatient follow-up files. All patient identifiers were removed during data extraction, and each participant was assigned a unique study identification code to maintain anonymity. The collected data were stored securely and accessed only by authorized members of the research team.

Table 1 shows the socio-demographic characteristics of patients. Young adult males constituted the majority of patients presenting with open long bone fractures. The mean age of the patients was 34.8 years ± 13.7 years (range: 18 - 79 years). Most patients were between 21 and 40 years (58.5%). Males accounted for 69 (73.4%) of the cases, giving a male-to-female ratio of 2.8:1.

Table 1. Socio-demographic characteristics of patients (N = 94).

Table 2 shows that road traffic crashes were the leading mechanism of injury, accounting for nearly two-thirds of open long bone fractures (64.9%). Falls were the second most common cause (18.1%), while assault (8.5%), occupational injuries (6.4%), and sports injuries (2.1%) contributed relatively few cases.

Table 2. Mechanism of injury.

Table 3 shows the anatomical distribution of open fractures. The tibia/fibula was the most frequently affected anatomical site, accounting for half of all open fractures (50.0%). The femur (20.2%) and radius/ulna (16.0%) were the next most commonly involved bones, whereas fractures of the humerus (9.6%) and other bones (4.2%) were relatively uncommon.

Table 3. Anatomical distribution of open fractures.

Table 4 shows the distribution of fractures according to Gustilo-Anderson classification. Grade II fractures were the most common, representing 42.6% of all open fractures, followed by Grade I fractures (36.2%). Grade III fractures accounted for 21.3% of cases, indicating that approximately one-fifth of patients presented with severe open injuries.

Table 4. Distribution of fractures according to Gustilo-Anderson classification.

Preventive antibiotic regimens in the two health facilities are shown in Table 5. Forty-four patients were managed at Nkwen Baptist Hospital using cloxacillin monotherapy, whereas fifty patients received a combination of ceftriaxone, gentamicin, and metronidazole at The People’s Clinic.

Table 5. Preventive antibiotic regimens.

NBH: Nkwen Baptist Hospital; TPC: The People’s Clinic.

Overall, 19 of the 94 patients developed a postoperative infection, corresponding to an overall postoperative infection prevalence of 20.2% (95% CI: 12.8% - 29.6%). Of these 19 postoperative infections, 8 (42.1%) met the consensus criteria for confirmed fracture-related infection (FRI), corresponding to a prevalence of confirmed FRI of 8.5% (8/94). The remaining 11 infections (57.9%) were superficial wound infections confined to the skin and subcutaneous tissues, without evidence of communication with or involvement of the fracture, and were therefore analysed separately and were not classified as confirmed FRI. Among the 8 confirmed FRI cases, 6 represented deep fracture-related infections, and 2 had progressed to chronic osteomyelitis.

Table 6 shows the distribution of consensus-confirmed fracture-related infection according to the antibiotic regimen received. Confirmed FRI occurred in 4 of 44 patients (9.1%) receiving cloxacillin and 4 of 50 patients (8.0%) receiving ceftriaxone, gentamicin, and metronidazole. There was no statistically significant association between antibiotic regimen and confirmed FRI (Fisher’s exact test p = 1.000; crude OR = 1.15, 95% CI: approximately 0.27 - 4.85). Because the antibiotic regimen was completely linked to the treatment hospital, this comparison cannot be regarded as evidence of therapeutic equivalence between the two antibiotic regimens.

Table 6. Comparison of fracture-related infection according to antibiotic regimen.

Fisher’s exact p = 1.000; crude OR = 1.15, 95% CI: 0.27 - 4.85.

Table 7 shows the occurrence of consensus-confirmed fracture-related infection according to Gustilo-Anderson fracture grade. The prevalence of confirmed FRI increased progressively with fracture severity, from 2.9% (1/34) among Grade I fractures, 5.0% (2/40) among Grade II fractures, to 25.0% (5/20) among Grade III fractures. This pattern indicates a strong association between increasing fracture severity and the occurrence of confirmed FRI (Pearson χ² = 8.97, df = 2, p = 0.011).

Table 7. Fracture-related infection according to Gustilo-Anderson grade.

Pearson χ² = 8.97, df = 2, p = 0.011.

Table 8 shows the association between time to surgical debridement and consensus-confirmed fracture-related infection. Confirmed FRI occurred in 1 of 48 patients (2.1%) who underwent debridement within 6 hours compared with 7 of 46 patients (15.2%) who underwent debridement more than 6 hours after injury. Delayed debridement was associated with a higher occurrence of confirmed FRI (Fisher’s exact p = 0.029).

Table 8. Time to debridement and fracture-related infection.

Because only eight patients met the consensus-confirmatory criteria for fracture-related infection, the number of outcome events was considered insufficient to support a reliable conventional multivariable logistic regression model containing the prespecified clinical predictors. Consequently, adjusted estimates were not reported because of the substantial risk of model overfitting and unstable effect estimates. Associations between potential risk factors and confirmed FRI were therefore primarily evaluated using bivariable analyses, with Fisher’s exact test applied where expected cell frequencies were small. Larger prospective studies are required to determine whether fracture severity, delayed debridement, and other clinical factors are independently associated with confirmed FRI.

The overall prevalence of postoperative wound infection in this study was 20.2% (19/94). However, after distinguishing superficial wound infections from infections meeting consensus-confirmatory criteria for fracture-related infection, 8 patients had confirmed FRI, corresponding to a prevalence of 8.5%. The remaining 11 cases were superficial wound infections confined to the skin and subcutaneous tissues without evidence of communication with or involvement of the fracture and were therefore not classified as confirmed FRI. This distinction is important because studies using broader definitions of postoperative infection may report higher infection rates than studies applying contemporary consensus criteria for FRI. Consequently, comparisons between the present findings and previously published infection rates should take differences in case definitions and diagnostic criteria into consideration.

Studies from resource-limited settings nevertheless demonstrate a substantial infection burden. In a Nigerian prospective series of open fractures, 42.5% of wounds became infected [14]. In Cameroon, a study of open tibial fractures reported FRI in 31.4% of patients using the international FRI consensus definition [11]. This should, however, be softened because the cited studies may not have applied exactly the same consensus-confirmatory definition. Nevertheless, these findings support the interpretation that infection after open fracture remains an important problem in sub-Saharan African trauma care.

Several factors may explain these disparities. First, healthcare systems in many LMICs continue to face shortages of orthopedic surgeons, operating theatre capacity, advanced imaging facilities, microbiology laboratories, and essential surgical implants. Second, financial barriers frequently delay patient presentation, diagnosis, and definitive treatment. Third, many patients initially seek care from traditional bone setters before presenting to the hospital, resulting in increased wound contamination and delayed initiation of evidence-based management. Finally, the sociopolitical instability affecting the North West Region of Cameroon during the study period may have further compromised timely access to emergency trauma services. Collectively, these factors likely contributed to the relatively high prevalence of FRI observed in this study.

The predominance of young adult males observed in this study is consistent with trauma patterns reported in African open-fracture series. Approximately three-quarters of the patients were male, and nearly 60% were aged between 21 and 40 years. In the Nigerian series cited above, most patients were aged 20 - 49 years [14], while the Cameroonian open-tibial-fracture study reported a mean age of 37.9 years and a male majority [11]. This pattern is plausibly related to greater exposure to road traffic and occupational hazards among economically active adults.

Road traffic crashes accounted for approximately 65% of all injuries in the present study, making them the leading cause of open fractures. This is consistent with the World Health Organization’s Global Status Report on Road Safety, which identifies road traffic injury as a major cause of death and disability, particularly among younger populations [4].

The tibia and fibula were the most commonly affected bones, accounting for half of all open fractures. This pattern is also reported in open-fracture series from resource-limited settings; in the Nigerian prospective study, the tibia/fibula was the most frequently affected site [14]. The limited soft-tissue envelope over much of the tibia makes it particularly vulnerable to open injury and subsequent contamination.

One of the important findings of this study was the progressive increase in consensus-confirmed FRI with increasing Gustilo-Anderson fracture grade. Confirmed FRI occurred in 2.9% of Grade I fractures, 5.0% of Grade II fractures, and 25.0% of Grade III fractures. The association between fracture severity and confirmed FRI was statistically significant in bivariable analysis. However, because only eight confirmed FRI events occurred, a conventional multivariable logistic regression model was considered insufficiently reliable to establish fracture grade as an independent predictor. The observed association should therefore be interpreted cautiously.

These findings are consistent with previous investigations. Gustilo and Anderson demonstrated the relationship between severe soft-tissue injury and infection risk in open fractures [2], and the later Type III subclassification further characterized severe injuries according to soft-tissue damage, contamination, and vascular injury [6]. Contemporary reviews likewise identify increasing Gustilo-Anderson grade as an important infection risk factor [5].

The biological explanation for this relationship is well established. Severe soft tissue injury compromises local blood supply, reduces oxygen delivery, impairs immune cell migration, and creates an environment conducive to bacterial colonization and biofilm formation. Extensive contamination and larger wound surfaces further increase the inoculum of pathogenic microorganisms. These factors collectively impair fracture healing while simultaneously increasing susceptibility to persistent infection. Consequently, patients with Grade III injuries require particularly aggressive multidisciplinary management, including urgent debridement, broad-spectrum empirical antibiotics when indicated, early soft tissue coverage, and close postoperative surveillance.

Our findings therefore reinforce the continued clinical value of the Gustilo-Anderson classification not only as a descriptive grading system but also as an important prognostic tool for identifying patients at greatest risk of fracture-related infection. Early recognition of severe injuries should facilitate risk stratification, individualized treatment planning, and allocation of specialized reconstructive resources where available.

The principal objective of this study was to compare two empirical antibiotic prophylaxis strategies used at the participating hospitals. Patients managed at Nkwen Baptist Hospital received cloxacillin monotherapy, whereas those treated at The People’s Clinic received ceftriaxone, gentamicin, and metronidazole. Confirmed FRI occurred in 4 of 44 patients (9.1%) managed with cloxacillin and 4 of 50 patients (8.0%) managed with ceftriaxone, gentamicin, and metronidazole. No statistically significant association was observed between hospital/regimen group and confirmed FRI. However, because the antibiotic regimen was completely determined by the treatment hospital, these findings cannot establish equivalence or comparative effectiveness of the antibiotic regimens. Hospital-level differences in patient characteristics, fracture severity, timing of treatment, surgical management, and postoperative care may have confounded the observed comparison. Prevention of infection after open fracture depends on multiple components of care, including early antibiotics, adequate debridement, stabilization, and soft-tissue management [8]-[10][15].

Classic studies established the importance of antibiotic prophylaxis in open fractures [16], and subsequent evidence-based guidance supports starting antibiotics as soon as possible after injury [9][17]. The present study should therefore be interpreted as a comparison of hospital-specific strategies rather than as evidence that cloxacillin and triple therapy are therapeutically equivalent.

A Cochrane review found that antibiotic prophylaxis reduces early infection compared with no antibiotics or placebo, but it did not establish the superiority of one multidrug regimen over another [18]. Thus, the current data do not justify a conclusion that broader combination therapy is inherently more or less effective than a narrower regimen.

Current guidance is not completely uniform regarding spectrum. AAOS recommends systemic cefazolin or clindamycin for major extremity trauma, with additional Gram-negative coverage preferred for Type III (and possibly Type II) open fractures [8]. BOAST emphasizes immediate intravenous prophylactic antibiotics, ideally within one hour of injury, and timely orthoplastic management [10]. Antibiotic selection should therefore follow an explicit institutional protocol informed by current guidance and local microbiology rather than routine use of the broadest possible regimen.

The comparable outcomes observed in this study should not be interpreted as evidence that cloxacillin is equivalent to internationally recommended first-generation cephalosporins. Rather, these findings illustrate that within this dataset, broader empirical antimicrobial coverage did not compensate for other important determinants of infection. In real-world clinical practice, antibiotic selection should always follow established evidence-based guidelines while taking into account local antimicrobial resistance profiles and drug availability.

From an antimicrobial stewardship perspective, these observations have important implications for low-resource settings. Broad-spectrum antibiotic combinations are substantially more expensive than narrow-spectrum agents and may increase the risk of nephrotoxicity, gastrointestinal complications, Clostridioidesdifficile infection, and antimicrobial resistance. Hospitals should therefore develop local treatment protocols supported by institutional antibiograms to ensure rational antibiotic use while preserving antimicrobial effectiveness.

Delayed surgical debridement was associated with a higher occurrence of consensus-confirmed fracture-related infection. Confirmed FRI occurred in 1 of 48 patients (2.1%) who underwent debridement within 6 hours compared with 7 of 46 patients (15.2%) who underwent debridement more than 6 hours after injury. Although this association was statistically significant in bivariable analysis, the small number of confirmed FRI events limits the precision of the estimate and precludes a robust assessment of delayed debridement as an independent predictor.

Historically, a “six-hour rule” was used for debridement. More recent evidence has questioned a rigid six-hour threshold, and contemporary guidance emphasizes timely debridement according to injury characteristics rather than an inflexible universal cutoff [10][19]. The present study’s association between debridement after six hours and infection should therefore be interpreted as an observational finding from this cohort, not proof of a universal causal threshold.

Published evidence on the exact timing threshold for debridement is mixed. Some studies have not found a significant increase in infection when surgery occurred after six hours, particularly when antibiotics were administered early [19]. Accordingly, the present finding of higher infection with debridement after six hours requires cautious interpretation and confirmation in larger prospective studies.

In many resource-limited healthcare systems, however, timely debridement remains challenging because of transportation barriers, shortages of orthopedic surgeons, limited operating theatre availability, financial constraints, and delayed referral from peripheral facilities. The sociopolitical instability affecting the North West Region of Cameroon during the study period may also have contributed to delays in presentation and definitive surgical management. These systemic barriers highlight the importance of strengthening trauma systems in addition to optimizing antimicrobial therapy.

The findings suggest that greater fracture severity and delayed surgical debridement are associated with an increased occurrence of consensus-confirmed FRI. These observations underscore the importance of rapid assessment, appropriate antibiotic prophylaxis, timely and adequate surgical debridement, fracture stabilization, and appropriate soft-tissue management in patients with severe open fractures. The absence of a statistically significant difference in confirmed FRI between the two hospital/regimen groups should not be interpreted as evidence that the antibiotic regimens are therapeutically equivalent, particularly because the antibiotic strategy was completely linked to the treatment hospital. The findings also support antimicrobial stewardship and the development of locally appropriate treatment protocols informed by microbiological surveillance and institutional antibiograms.

Several limitations should be acknowledged. The retrospective study design is inherently susceptible to incomplete documentation, missing data, and information bias. Variables such as smoking status, diabetes mellitus, nutritional status, contamination severity, microbiological culture findings, antibiotic timing relative to injury, and patient adherence to postoperative care were not consistently documented and therefore could not be evaluated. In addition, treatment protocols may have varied among surgeons, potentially introducing practice-related heterogeneity. Furthermore, only eight patients met the consensus-confirmatory criteria for FRI. This small number of confirmed outcome events substantially limited statistical power and precluded reliable conventional multivariable logistic regression. Consequently, associations between fracture severity, delayed debridement, antibiotic strategy, and confirmed FRI should be interpreted as exploratory rather than causal or independently predictive. The distinction between superficial wound infection and confirmed FRI also depended on the completeness of retrospective clinical documentation, introducing the possibility of outcome misclassification. Larger prospective studies applying standardized consensus FRI criteria are required to confirm these findings.