The risk of osteomyelitis after mandibular fracture is doubled in men versus women: analysis of 300,000 patients

Postoperative complications following mandibular fracture treatment vary from local wound infections to severe conditions including osteomyelitis and impaired fracture healing. Several risk factors have been associated with the development healing disorders, including fracture localisation, treatment modality and substance abuse. However, limited research on the sex-specific influence of these complications exists. A total of about 300,000 female and male patients with mandibular fractures were examined in two cohorts. After matching for confounders (age, nicotine and alcohol dependence, malnutrition, overweight, anaemia, diabetes, osteoporosis and vitamin D deficiency), two cohorts were compared with propensity-score-matched patients according to outcomes (osteomyelitis, pseudoarthrosis and disruption of the wound) within 1 year after fracture. There were significant differences between female and male patients regarding the occurrence of osteomyelitis (odds ratio [OR] [95% confidence interval]: 0.621 [0.563; 0.686]) and disruption of the wound (OR [95% confidence interval]: 0.703 [0.632; 0.782]). Surprisingly, matching for the expected confounders did not change the results substantially. Sex plays a dominant role in determining the risk stratification for postoperative osteomyelitis and disruption of the wound, after accounting for other potential confounding factors. Additional research is needed to understand the underlying mechanisms and to develop sex-specific strategies to prevent these complications.

depression, underlying metabolic diseases (especially diabetes mellitus), removal of a tooth in the fracture gap and substance abuse, contribute to long-term complications [13][14][15][17][18][19] . Dependng on the severity of impaired bone healing, revision surgery with renewed osteosynthetic treatment, high-dose antibacterial therapy and even bone grafting may be required 20 .Revision surgeries are associated with up to a 32.6% increase in hospital costs, posing an additional burden on both the healthcare system and individual patients 21 .In a large retrospective analysis covering cases of osteomyelitis of various anatomical sites in the United States over 40 years, the annual incidence was significantly higher in male compared with female patients and increase with age (P < 0.001) 22 .However, most of the published data regarding a sex-dependent influence on osteomyelitis are for foot and long bones, and little is known about this effect on the mandible [23][24][25][26][27] .Given this lack of knowledge, the aim of the study was to analyse the influence of sex on the development of osteomyelitis, pseudoarthrosis and disruption of the wound after mandibular fractures.

Assessment, allocation and matching
We considered a total of 302,575 patients who were diagnosed with mandibular fractures (International Classification of Diseases 10th revision [ICD-10] code S02.6).We grouped the patients according to sex (female vs male).The female cohort included 115,051 patients with a mean ± standard deviation (SD) age of 43.9 ± 24.4 years.The male cohort included 187,524 patients with a mean ± SD age of 34.4 ± 19.7 years.There was a significant difference in age between the sexes (P < 0.001).The analyses of the risk factors between the female and male cohort revealed significant differences in nicotine and alcohol dependence, diabetes, overweight, malnutrition, anaemia, vitamin D deficiency and osteoporosis between female and male patients (P < 0.001).Table 1 shows the patient characteristics before and after propensity-score matching.

Risk analysis
We performed statistical analysis to compare three outcomes-osteomyelitis, pseudoarthrosis and disruption of the wound-between female and male patients (Tables 2 and 3).Osteomyelitis within 1 year after mandibular For pseudoarthrosis, there was no significant difference between female and male patients (P = 0.387).We found pseudoarthrosis in 662 female patients and 694 male patients within 1 year of mandibular fracture.The RR was 0.954 (95% CI [0.858; 1.061]) and the OR was 0.954 (95% CI [0.857; 1.062]).
A more in-depth analysis was conducted based on ICD subcodes (Supplementary Table 1).In this context, the examination of individual subcodes (S02.6-S02.69)showed no significant differences in odds ratios compared to the overarching code S02.6.

Discussion
Complications following treatment of mandibular fractures range from mild wound infections and wound healing disorders, wound dehiscence with exposure of fixation material, severe bone infections/osteomyelitis and disturbed/delayed bone healing to failure of fracture healing and the development of pseudarthrosis [28][29][30] .While minor complications can be treated conservatively, revision surgery with refixation might be indicated in cases of severe impaired fracture healing 31 .In a retrospective study by Steffen et al. the main indications for revision surgery with refixation were osteomyelitis (52.9%) and non-union (41.2%) in patients with mandibular fractures 32 .Different factors have been associated with complications after fracture treatment, including smoking and alcohol abuse, increased time from injury to treatment, mandibular fracture severity, treatment modality and tooth extraction 13,17,26,27 .Currently, there is limited data on sex-specific considerations for complications like osteomyelitis following mandibular fracture treatment.Gordon et al. conducted a case-control study on patients who underwent mandibular fracture repair, discovering a higher rate of postoperative inflammatory complications (POIC), including osteomyelitis, among male patients.However, in their bivariate analysis, gender did not exert a significant influence 26 .Lukošiūnas and colleagues analyzed data from patients who developed osteomyelitis after mandibular fracture treatment and compared background factors of complications with a control group.They did not observe a gender-specific effect on the development of osteomyelitis.However, in their logistic regression analysis, the authors identified several significant factors in the development of osteomyelitis in fractured mandibles.These included factors such as immunity dysfunction, oral microflora, presence of caries-affected or intact teeth at the fracture line, mobility of bone fragments, inadequate repositioning, and delayed fixation of bone fragments after trauma 33 .
In a comprehensive retrospective study analysing 760 cases of osteomyelitis across various anatomical sides, the incidence was higher for men than for women and increased with age (P < 0.001).In this retrospective study encompassing Olmsted County, Minnesota residents, only 19% of osteomyelitis cases were linked to traumatic origins.Additionally, craniofacial sites accounted for merely about 5% of all anatomical locations in the study, rendering the interpretation of data on mandibular osteomyelitis less reliable 22 .
To evaluate the sex-specific effect on the development of postoperative wound dehiscence, osteomyelitis and pseudarthrosis of the mandible following fracture treatment, we performed one-to-one matching of male and female patients based on similar covariate distributions, including alcohol and nicotine dependence; diabetes mellitus; malnutrition, overweight, obesity and hyperalimentation; osteoporosis; anaemia; and vitamin D deficiency.Interestingly, there were significant differences in all of these parameters before matching between men and women.Moreover, after matching for confounders, a significant difference between male and female patients regarding the development of postoperative disruption of the wound and the development of osteomyelitis was found.There are several potential reasons for the higher incidence of osteomyelitis following mandibular fracture treatment in men.Males generally exhibit a greater quantity of cortical bone in the mandible, highlighting sexrelated differences 34 .Studies on chronic osteomyelitis reported the invasion and persistence of Staphylococcus aureus in the canaliculi of live cortical bone 35,36 , which may serve as a mechanism for promoting persistent and chronic infection, potentially restricting immune cell access 37 .One factor might be the higher rate of nicotine Table 3. Risk difference, risk ratios and odds ratios for disruption of wound of the female and male cohort after propensity-score matching.The outcome was defined as the occurrence of disruption of the wound within 1 year after mandibular fracture.CI confidence interval.Note that 658 female patients and 857 male patients were excluded from results because they had the outcome prior to the time window.www.nature.com/scientificreports/abuse in men, which has generally been attributed to a higher complication rate 30 .In their review of the potentially modifiable patient factors that could affect mandibular fracture complications, Ahmed et al. identified smoking as the most common potentially modifiable factor (OR 4.04-8.09) 38.To exclude differences in the prevalence of nicotine abuses between both sexes, we conducted confounder matching including nicotine abuse.However, differences in smoking habits have been reported with higher pack-years of smoking and number of cigarettes per day in men 39 .Interestingly, Radabaugh et al. 40 analysed patient compliance following mandibular fracture repair and concluded that current tobacco use is negatively associated with patient compliance.Another aspect is the dominance of interpersonal violence in mandibular fractures in men, leading to a different fracture pattern with a higher susceptibility for osteomyelitis compared with women with mandibular fractures 15,19,41 .

Number of patients Number of patients with disruption of wound
In general, sex-related differences in lifestyle may also affect health status and therefore the prevalence of preexisting conditions 42 .In this regard, social and behavioural characteristics are key factors related to the sex gap in mortality 43 .
Studies in the field of osteoimmunology provide information regarding the modulating effect of the innate and adaptive immune system on bone resorption during inflammations 44 .In men and women, the development and functioning of the immune system are affected in distinct ways by various environmental factors, such as the nutrition status and the composition of the microbiome.These sex-based immunological differences contribute to variations in the incidence and susceptibility to infectious diseases 45 .Interestingly, different human and animal studies focused on fracture healing in long bones have reported impaired bone healing more often in women 46,47 , suggesting possible sex-based differences in bone healing in general 48 .In a large patient database analysis of more than 300.000 fractures in 18 bone, ORs for non-union fractures were significantly increased for different risk factors including male gender (OR 1.21; 95%, CI 1.16-1.25) 49.However, there are several differences between mandibular and long bone fractures.First, mandibular fracture wounds could be contaminated by bacteria of the oral cavity.Second, there is different underlying embryonic bone development/formation: endochondral bone formation in long bones and intramembranous bone formation of the mandible.
Limitations of this study are that the data analysis is conducted on a large patient database, which might lead to unexpected associations as well as to a selection bias 50 .It is important to note that the associations identified in this study do not imply causation.Furthermore, the TriNetX database did not include information on the time elapsed between injury and treatment, potentially introducing an unobserved confounding variable.Several studies have analysed the effects of a treatment delay, reporting conflicting results, which might be attributed to a lack of consensus on the definitions of "early" versus "delayed" intervention 18,27,28,33 .

Conclusion and future perspectives
Sex plays a dominant role in determining the risk stratification for postoperative osteomyelitis and disruption of the wound, after accounting for other potential confounding factors.Sex-specific treatment recommendations should be considered to account for the sex-specific risk for the development of osteomyelitis.A possible recommendation is prolonged peri-and postoperative antibacterial therapy in men with the corresponding risk profile and risk factors as well as an extended follow-up observation.

Data acquisition and inclusion and exclusion criteria
We used TriNetX, a global federated health research network providing access to statistics on electronic medical records (diagnoses, procedures, medications, laboratory values, genomic information) from patients in large Healthcare Organizations predominately.As a federated network, TriNetX received a waiver from Western IRB since only aggregated counts, statistical summaries of de-identified information, but no protected health information is received, and no study-specific activities are performed in retrospective analyse.This retrospective study is exempt from informed consent.The data reviewed is a secondary analysis of existing data, does not involve intervention or interaction with human subjects, and is de-identified per the de-identification standard defined in Section §164.514(a) of the HIPAA Privacy Rule.The process by which the data is de-identified is attested to through a formal determination by a qualified expert as defined in Section §164.514(b)(1) of the HIPAA Privacy Rule.This formal determination by a qualified expert refreshed on December 2020.The TriNetX network was accessed on June 23rd, 2023.The query was run on the platform with a group of 81 health care organisations (HCOs).The database was searched for electronic medical records up to 20 years before the access date for patients with mandibular fractures according to the ICD-10 (International Statistical Classification of Diseases and Related Health Problems) code S02.6.Based on the subdivision of the S02.6 code (S02.60-S02.69), a subgroup analysis of the individual codes is displayed in Supplements (Supplementary Table 1).
Figure 1 displays a modified Consolidated Standards of Reporting Trials (CONSORT) flow chart.We grouped patients according to sex (female vs male).Before matching, there were 115,051 patients from 71 HCOs in the female cohort (cohort I) and 187,524 patients from 73 HCOs in the male cohort (cohort II).We applied propensity-score matching to reduce confounding variables and to ensure the groups were based on similar covariate distributions.We performed one-to-one matching for alcohol and nicotine dependence; diabetes mellitus; malnutrition; overweight, obesity and hyperalimentation; osteoporosis, anaemia; and vitamin D deficiency.After matching, each cohort had 96,245 patients.

Data analysis
We defined the index event as the day of the mandibular fracture; the observation period was 1 year after the mandibular fracture.We defined the outcomes as osteomyelitis (ICD-10 code M86), pseudoarthrosis after fusion (ICD-10 code M96.0) and disruption of the wound (ICD-10 code T81.3).We excluded patients with the and warrants that it has all necessary rights, consents, approvals, and authority to provide the data to TriNetX under a Business Associate Agreement (BAA), so long as their name remains anonymous as a data source and their data are utilized for research purposes.The data shared through the TriNetX Platform are attenuated to ensure that they do not include sufficient information to facilitate the determination of which HCO contributed which specific information about a patient (https:// trine tx.com/ trine tx-publi cation-guide lines/).Access to the database is closed. https://doi.org/10.1038/s41598-023-48235-w

Table 1 .
Patient characteristics before and after propensity-score matching.Std.standardised; Vit.Vitamin.Percentage refers to the respective cohorts.P-value refers to the comparison between both cohorts (log-rank test).

Table 2 .
Risk difference, risk ratios and odds ratios for osteomyelitis of the female and male cohort after propensity-score matching.The outcome was defined as the occurrence of osteomyelitis within 1 year after mandibular fracture.CI confidence interval.Note that 924 female patients and 1426 male patients were excluded from the results because they had the outcome prior to the time window.