ABSTRACT
Objective
Temporal bone fractures (TBFs) can result in long‐term adverse outcomes including meningitis, facial nerve disorders (FNDs), and hearing loss (HL) that may require surgical intervention. This epidemiologic study aims to examine the sequelae of adult TBF with HL, including the utilization of cochlear implantation (CI), using data from a large, multinational database.
Methods
Using the TriNetX database, a retrospective cohort study was performed of adults (≥ 18 years) using ICD10 codes for other fractures of the base of the skull and hearing loss, an approximation of TBF with hearing loss. A control group of adults without TBF was used. Measured outcomes included meningitis, CI, FND, cerebrospinal fluid (CSF) leak, and labyrinthitis. Propensity score matching (1:1) was used for cohorts smaller than 33,333,333. Ninety‐five percent of patient data used was from 2006 to 2023.
Results
Adults with TBF and HL (n = 34,878) had a greater risk of meningitis any time after TBF than those without TBF (n = 105,035,185) (RR: 6.65, 95% CI: 5.74–7.70). Labyrinthitis (RR: 3.56, 95% CI: 2.86–4.41), CSF leak (RR: 40.71, 95% CI: 37.91–43.71), and FND (RR: 12.08, 95% CI: 11.62–12.55) were more common after TBF. CI was more common after TBF (RR: 26.22, 95% CI: 22.12–31.07). Meningitis after TBF was associated with an increased risk of CSF leak (RR: 3.0, 95% CI: 1.52–5.93) compared to those without meningitis.
Conclusion
Adults who sustain TBFs resulting in HL have an increased risk of developing meningitis, labyrinthitis, CSF leak, and FND and are more likely to undergo CI for aural rehabilitation compared to a control cohort.
Level of Evidence
Level IV.
Keywords: cerebrospinal fluid leak, cochlear implant, facial nerve disorder, hearing loss, labyrinthitis, meningitis, temporal bone fracture
This study uses a large multinational database to examine long‐term outcomes of adult temporal bone fractures. The risk for meningitis, labyrinthitis, facial nerve disorders, and cerebrospinal fluid leaks was elevated after fracture. Cochlear implantation was also more likely after temporal bone fracture.
1. Introduction
The temporal bone (TB) is a central component of the lateral skull base and borders the middle and posterior cranial fossae. Temporal bone fractures (TBFs) make up 18%–40% of skull base fractures [1] and are frequently the result of significant blunt force trauma, requiring approximately 1300–1875 pounds of force to fracture [2, 3]. Common causes include motor vehicle accidents, assault, and falls [1, 4]. Due to the high‐energy mechanism and the surrounding neurovascular structures, TBFs can result in complications such as hearing loss (HL), meningitis, facial nerve injuries, cerebrospinal fluid (CSF) fistulas, and other cranial nerve injuries. The severity and recovery rates of these complications vary. For patients with conductive hearing loss (CHL) due to a traumatic effusion, for example, complete recovery is expected, while CHL due to ossicular disruption may require surgical reconstruction. Sensorineural hearing loss (SNHL) is less predictable and may require amplification or cochlear implantation (CI), with excellent post‐treatment outcomes [5].
The purpose of this study is to examine and better estimate the rates of complications and outcomes after TBF with hearing loss (TBFHL), including CI for SNHL, utilizing the TriNetX multinational database. To the best of our knowledge, no study has examined this on a large scale across multiple time points, with most publications citing single‐institution data and a population typically less than 1000 patients [2, 4, 6, 7, 8, 9, 10, 11]. We hypothesize that complication rates will be lower than reported in the literature, that CI rates will exceed those of the control, and that meningitis after TBFHL will increase the risk of adverse outcomes.
2. Materials and Methods
This study was reviewed by the University of Utah Institutional Review Board (IRB_00174595) and determined to be exempt as non‐Human Subjects Research. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines were followed [7, 12].
2.1. Study Design and Data Sources
We conducted a population‐based, retrospective cohort study using the TriNetX Research Network, a multinational research platform providing access to data from electronic health records (EHR) from participating healthcare organizations (HCOs). At the time of writing, TriNetX is comprised of 133 million patients from 104 HCOs from 5 countries, with the majority in the USA. The database includes diagnoses, procedures, medications, laboratory results, and genetic information documented in the EHR of each patient's chart at each HCO. Ninety‐five percent of patient data used is from 2006 to 2024. All data are de‐identified in compliance with the HIPAA Privacy Rule.
2.2. Query Design, Cohort Definitions, and Patient Selection
All cohorts were defined using International Classification of Disease (ICD)‐10 (cross‐listed with ICD‐9), Current Procedural Terminology (CPT), or other medical codes. Cross‐listing with ICD‐9 codes allowed patients with an ICD‐9 diagnosis to be included in the ICD‐10 search. Figure 1 and Figure 2 detail the construction of each cohort. Age restrictions were a minimum of 18 years of age. The main cohort of adults with TBFHL was defined using ICD‐10 S02.19 (other fracture of base of skull) and H90 (conductive and sensorineural hearing loss) or H91 (other unspecified hearing loss). This combination was selected to increase the specificity of our approximation for TBF to patients with TBFs, which commonly cause HL (estimated at 57% prevalence of HL in adult TBFs) [1, 8, 13, 14, 15, 16]. Then, the index event (the first day all inclusion criteria are met) was defined as S02.19 and H90 or H91. The group of adults without TBF (“control”) was defined as any adult without S02.19 with any “visit” used as the index event. This “visit” is a random, arbitrary event. Additionally, a cohort of TBFHL with post‐fracture meningitis (TBFHL + Meningitis) was defined as S02.19 and H90 or H91 and G00 (bacterial meningitis, not elsewhere classified) or G03 (meningitis due to other and unspecified causes). A corresponding control of TBFHL without meningitis (TBFHL—Meningitis) was defined as S02.19 and H90 or H91 without G00 or G03. The index events were defined as a diagnosis of S02.19 and G00 or G03 (TBFHL + Meningitis) and S02.19 without G00 or G03 (TBFHL—Meningitis). See Table 1 for a list of codes and corresponding descriptions.
FIGURE 1.
Construction of adult temporal bone fracture and control cohorts. Propensity score matching was unavailable. Outcomes of interest are shown. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
FIGURE 2.
Construction of the adult temporal bone fracture cohort with and without subsequent meningitis. Propensity score matching and outcomes of interest are shown. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
TABLE 1.
List of diagnostic and procedural codes used.
| Descriptor | Code | Label |
|---|---|---|
| Temporal bone fracture | ICD‐10 S02.19 + (H90 or H91) | Other fracture of base of skull AND ([conductive and SNHL] or [other and unspecified hearing loss]) |
| Cochlear implant | CPT 69930 | Cochlear device implantation, with or without mastoidectomy |
|
Meningitis (not using G01 and G02) |
ICD‐10 G00 |
Bacterial meningitis, not elsewhere classified G00.1 = pneumococcal meningitis |
| ICD‐10 G03 | Meningitis due to other and unspecified causes | |
| Labyrinthitis | ICD‐10 H83.0 | Labyrinthitis |
| Facial nerve disorder | ICD‐10 G51 | Facial nerve disorders |
| ICD‐10 R29.810 | Facial weakness | |
| ICD‐10 H02.2 | Lagophthalmos | |
| CPT 67912 | Correction of lagophthalmos, with implantation of upper eyelid load (e.g., gold weight) | |
| Pneumococcal vaccine | CPT 90670 | PCV13, IM |
| CPT 90671 | PCV15, IM | |
| CPT 90677 | PCV20, IM | |
| CPT 90732 | PPSV23, SQ or IM | |
| HCPCS G8864 | Pneumococcal vaccine administered or previously received | |
| HCPCS G0009 | Administration of pneumococcal vaccine | |
| SNOMED 12866006 |
Administration of pneumococcal vaccine |
|
| SNOMED 1119366009 | Administration of PCV13 | |
| CVX 152 | PneumoPCV–Pneumococcal Conjugate, unspecified formulation | |
| CVX 33 | PneumoPPV–pneumococcal polysaccharide PPV23 | |
| CSF Leak | ICD‐10 G96.0 | Cerebrospinal fluid leak |
2.3. Outcome Measures
Outcomes were measured starting one day after the index event. Outcome measures included meningitis, labyrinthitis, CI, facial nerve disorders (FNDs), and CSF leak. Rates of pneumococcal vaccination were analyzed. Outcomes were evaluated and compared at 3 months, 1 year, 5 years, and any time after the index event. Table 1 provides diagnoses and procedure codes used.
2.4. Statistical Analysis
Statistical analysis was performed through the TriNetX platform, which uses programming languages from both R (4.0.2) and Python (3.7). Propensity score matching (PSM) was used for age and sex in a 1:1 ratio, and chi‐square tests were used to assess baseline covariate proportions. PSM is a statistical method that attempts to control for confounding variables and has become a popular alternative to multivariate logistic regression [17, 18, 19]. Additional analyses before and after PSM included absolute risk, risk ratios with 95% confidence intervals (RR, 95% CI), and associated p‐values. A p‐value of < 0.05 was used for significance. Patients who met the index 20 or more years ago were automatically excluded. TriNetX protects patient privacy by reporting any patient counts from 1 to 10 as 10. PSM is only available for cohorts smaller than 33,333,333, with larger cohorts using unmatched data only.
3. Results
3.1. Cohort Demographics
Table 2 describes cohort demographic data before PSM. A total of 34,878 patients with TBFHL and 105,035,185 patients without TBF were included. Table 3 includes rates of TBFHL by decade and CI utilization before and after FDA approval (July 2019) for single‐sided deafness (SSD). All outcomes are detailed in Table 4 (TBFHL vs. control) and Table 5 (TBFHL ± Meningitis). Figure 3 shows risk ratios for all outcomes. PSM was not available for the TBFHL and control cohorts due to the size of the control cohort. For the TBFHL + Meningitis cohort, 160 patients were included compared to 34,694 in the TBFHL—Meningitis cohort. After PSM, each cohort had 158 patients.
TABLE 2.
Demographics and baseline characteristics of cohorts before propensity score matching (PSM).
| Before PSM | ||||||
|---|---|---|---|---|---|---|
| TBFHL Cohort (n = 34,878) | Control Cohort (n = 105,035,185) | p | TBFHL + Meningitis (n = 160) | TBFHL – Meningitis (n = 34,694) | p | |
| Age at index, mean ± SD (years) | 56.1 ± 22.7 | 43.5 ± 20.9 | — b | 49.3 ± 21.7 | 56.2 ± 22.7 | 0.0002* |
| Sex | — b | |||||
| Female | 41.19% | 53.70% | 36.7% | 41.21% | 0.242 | |
| Male | 49.03% | 43.92% | 60.12% | 48.97% | 0.045* | |
| Unknown | 9.78% | 2.38% | 6.32% | 9.82% | — b | |
| Race | — b | — b | ||||
| White | 71.66% | 48.78% | 67.72% | 71.79% | ||
| Unknown | 16.49% | 28.49% | 12.65% | 16.40% | ||
| Black or African American | 6.82% | 10.42% | 10.12% | 6.82% | ||
| Asian | 1.75% | 4.37% | 6.32% | 1.72% | ||
| Native American/Alaskan | 0.33% | 0.25% | 6.32% | 0.33% | ||
| Native Hawaiian | 0.7% | 0.35% | 6.32% | 0.70% | ||
| Ethnicity | — b | — b | ||||
| Not Hispanic | 66.06% | 43.26% | 76.58% | 66.57% | ||
| Hispanic | 7.47% | 7.43% | 10.76% | 7.44% | ||
| Unknown | 26.47% | 49.31% | 12.66% | 25.99% | ||
| Facial Nerve Disorders | 1194 (3%) | 45 (0%) | — b | 20 (13%) | 1164 (3%) | |
| Other/Unspecified Meningitis | 243 (0%) | 22 (0%) | — b | 58 (37%) | 188 (1%) | — b |
| Bacterial Meningitis | 71 (0%) | 10 (0%) a | — b | 32 (20%) | 45 (0%) | — b |
| CSF Leak | 796 (2%) | 10 (0%) a | — b | 43 (27%) | 751 (2%) | — b |
| Labyrinthitis | 194 (0%) | 10 (0%) a | — b | 10 (6%) | (192 (1%) | — b |
Significant values.
Patient numbers between 1‐10 are reported as 10 by TriNetX.
Not calculated due to large sample size.
TABLE 3.
Number of temporal bone fractures with hearing loss by decade and number of CIs before and after FDA approval for single‐sided deafness.
| Number | |
|---|---|
| TBFs with Hearing Loss 2000–2009 | 979 (97.9/year) |
| TBFs with Hearing Loss 2010–2019 | 17,182 (1718.2/year) |
| TBFs with Hearing Loss 2020–2024 | 10,102 (2202.4/year) |
| Cochlear Implants Pre‐FDA Approval | 67 |
| Cochlear Implants Post‐FDA Approval | 70 |
TABLE 4.
Incidence and risk ratios of outcomes at various time frames before propensity score matching (PSM).
| 3‐month | 1‐year | |||||
|---|---|---|---|---|---|---|
| Outcome | TBFHL Cohort (n = 34,878) | Control Cohort (n = 105,035,185) | RR (95% CI) a | TBFHL Cohort (n = 34,878) | Control Cohort (n = 105,035,185) | RR (95% CI) |
| Meningitis | 103 (0.3%) | 27,210 (0.0%) | 11.4 (9.40–13.83)* | 126 (0.4%) | 35,575 (0.0%) | 10.67 (8.96–12.70)* |
| Facial Nerve Disorders | 1613 (4.6%) | 183,464 (0.2%) | 26.48 (25.24–27.78)* | 1823 (5.2%) | 232,528 (0.2%) | 23.61 (22.57–24.69)* |
| Labyrinthitis | 32 (0.1%) | 14,416 (0.0%) | 6.69 (4.73–9.46)* | 44 (0.1%) | 21,417 (0.0%) | 6.19 (4.60–8.32)* |
| Cochlear Implant | 32 (0.1%) | 2497 (0.0%) | 38.59 (27.24–54.69)* | 72 (0.2%) | 6165 (0.0%) | 35.17 (27.89–44.36)* |
| CSF Leak | 628 (1.8%) | 16,197 (0.0%) | 116.76 (107.89–126.37)* | 670 (1.9%) | 22,215 (0.0%) | 90.83 (84.17–98.01)* |
| 5‐year | Anytime | |||||
|---|---|---|---|---|---|---|
| Outcome | TBFHL Cohort (n = 34,878) | Control Cohort (n = 105,035,185) | RR (95% CI) | TBFHL Cohort (n = 34,878) | Control Cohort (n = 105,035,185) | RR (95% CI) |
| Meningitis | 166 (0.5%) | 57,100 (0.1%) | 8.76 (7.52–10.19)* | 177 (0.5%) | 80,219 (0.1%) | 6.65 (5.74–7.70)* |
| Facial Nerve Disorders | 2239 (6.4%) | 389,060 (0.4%) | 17.33 (16.65–18.04)* | 2406 (6.9%) | 599,955 (0.6%) | 12.08 (11.62–12.55)* |
| Labyrinthitis | 70 (0.2%) | 46,407 (0.0%) | 4.54 (3.59–5.74)* | 82 (0.2%) | 69,457 (0.1%) | 3.56 (2.86–4.41)* |
| Cochlear Implant | 112 (0.3%) | 10,135 (0.0%) | 33.28 (27.63–40.08)* | 134 (0.4%) | 15,393 (0.0%) | 26.22 (22.12–31.07)* |
| CSF Leak | 731 (2.1%) | 35,757 (0.0%) | 61.57 (57.26–66.19)* | 752 (2.2%) | 55,634 (0.1%) | 40.71 (37.91–43.71)* |
Significant values.
RR, 95% CI = ratio risk, 95% confidence interval.
TABLE 5.
Incidence and risk ratios of outcomes at various time frames after propensity score matching (PSM).
| 3‐month | 1‐year | |||||
|---|---|---|---|---|---|---|
| Outcome | TBFHL + Meningitis (n = 158) | TBFHL – Meningitis (n = 158) | RR (95% CI) a | TBFHL + Meningitis (n = 158) | TBFH – Meningitis (n = 158) | RR (95% CI) |
| Labyrinthitis | 0 (0.0%) | 0 (0.0%) | — | 0 (0.0%) | 0 (0.0%) | — |
| Cochlear Implant | 0 (0%) | 0 (0.0%) | — | 0 (0%) | 10 (6.3%) b | — |
| CSF Leak | 27 (17.1%) | 10 (6.3%) b | 2.70 (1.35–5.39)* | 29 (18.4%) | 10 (6.3%) b | 2.90 (1.46‐5.75)* |
| 5‐year | Anytime | |||||
|---|---|---|---|---|---|---|
| Outcome | TBFHL + Meningitis (n = 158) | TBFHL – Meningitis (n = 158) | RR (95% CI) a | TBFHL + Meningitis (n = 158) | TBFHL – Meningitis (n = 158) | RR (95% CI) |
| Labyrinthitis | 0 (0.0%) | 0 (0.0%) | — | 0 (0.0%) | 0 (0.0%) | — |
| Cochlear implant | 10 (6.3%) b | 10 (6.3%) b | 1.00 (0.43–2.3) | 10 (6.3%) b | 10 (6.3%) b | 1.00 (0.43–2.3) |
| CSF Leak | 30 (19.0%) | 10 (6.3%) b |
3.00 (1.52–5.93)* |
30 (19.0%) | 10 (6.3%) b |
3.00 (1.52–5.93)* |
Significant values.
RR, 95% CI = ratio risk, 95% confidence interval.
Patient numbers between 1 and 10 are reported as 10 by TriNetX.
FIGURE 3.
Temporal bone fracture with hearing loss risk ratios with 95% confidence interval error bars for each time point analyzed. Risk ratios are unmatched because the cohort size was > 3,333,333. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
3.2. Meningitis
Adults with TBFHL had an increased risk of meningitis within the first 3 months (RR: 11.40, 95% CI: 9.40–13.83), 1 year (RR: 10.67, 95% CI: 8.96–12.70), 5 years (RR: 8.76, 95% CI: 7.52–10.19), and any time after TBFHL (RR: 6.65, 95% CI: 5.74–7.70). In the TBFHL cohort, meningitis occurred in 0.3% (103) within 3 months, 0.4% (126) 1 year, 0.5% (166) 5 years, and 0.5% (177) any time after TBFHL, compared to 0.0% (27,210) within the first 3 months, 0.0% (35,575) 1 year, 0.1% (57,100) 5 years, and 0.1% (80,219) any time after the index in the control.
3.3. Pneumococcal Vaccination
Pneumococcal vaccination status was assessed for each cohort to examine the potential protective effects against meningitis for patients with TBFHL. Data was deemed unreliable. Only 13.1% of adults ages 18–64 had pneumococcal vaccines recorded, significantly less than the current CDC estimate (age 18–64: 22%–36%) [20]. Of the remaining 86.9%, < 0.01% had a vaccine refusal documented.
3.4. Labyrinthitis
Adults with TBFHL were more likely to develop labyrinthitis within 3 months (RR: 6.69, 95% CI: 4.73–9.46), 1 year (RR: 6.19, 95% CI: 4.60–8.32), 5 years (RR: 4.54, 95% CI: 3.59–5.74), and any time after TBFHL (RR: 3.56, 95% CI: 2.86–4.41). In the TBFHL cohort, labyrinthitis occurred in 0.1% (32) within 3 months, 0.1% (44) 1 year, 0.2% (70) 5 years, and 0.2% (82) any time after TBFHL, compared to 0.0% (14,416) within 3 months, 0.0% (21,417) 1 year, 0.0% (46,407) 5 years, and 0.1% (69,457) any time after index in the control. In the TBFHL + Meningitis and TBFHL –Meningitis cohorts, there were no instances of labyrinthitis.
3.5. Cochlear Implantation
Adults with TBFHL were more likely to undergo CI within 3 months (RR: 38.59, 95% CI: 27.24–54.69), 1 year (RR: 35.17, 95% CI: 27.89–44.36), 5 years (RR: 33.28, 95% CI: 27.63–40.08), and any time after TBFHL (RR: 26.22, 95% CI: 22.12–31.07) than the control. In the TBFHL cohort, CI occurred in 0.1% (32) within 3 months, 0.2% (72) 1 year, 0.3% (112) 5 years, and 0.4% (134) any time after TBFHL, compared to 0.0% (2497) within 3 months, 0.0% (6165) 1 year, 0.0% (10,135) 5 years, and 0.0% (15,393) any time after the index in the control cohort. In total, 67 patients received a cochlear implant prior to FDA approval (July 2019) for SSD, compared to 70 afterward.
In the TBFHL + Meningitis cohort, CI occurred in 0.0% (0) within 3 months, 0.0% (0) 1 year, 6.3% (10) 5 years, and 6.3%% (10) any time after TBFHL + Meningitis compared to 0.0% (0) within 3 months, 6.3% (10) 1 year, 6.3% (10) 5 years, and 6.3% (10) any time after TBFHL in the TBFHL—Meningitis cohort. Risk ratios for the “5‐year” and “any time” periods were 1.00 (95% CI: 0.43–2.30).
3.6. Facial Nerve Disorders
Risk of FND was increased in the TBFHL cohort within 3 months (RR: 26.48, 95% CI: 25.24–27.78), 1 year (RR: 23.61, 95% CI: 22.57–24.69), 5 years (RR: 17.33, 95% CI: 16.65–18.04), and any time after TBFHL (RR: 12.08, 95% CI: 11.62–12.55). In the TBFHL cohort, FND occurred in 4.6% (1612) within 3 months, 5.2% (1823) 1 year, 6.4% (2239) 5 years, and 6.9% (2406) any time after TBFHL, compared to 0.2% (183,464) within 3 months, 0.2% (232,528) 1 year, 0.4% (389,060) 5 years, and 0.6% (599,955) any time after the index in the control.
3.7. CSF Leak
Risk for CSF leak was increased in the TBFHL cohort at 3 months (RR: 116.79, 95% CI: 107.89–126.37), 1 year (RR: 90.83, 95% CI: 84.17–98.01), 5 years (RR: 61.57, 95% CI: 57.26–66.19) and any time after TBFHL (RR: 40.71, 95% CI: 37.91–43.71). In the TBFHL cohort, CSF leaks developed in 1.8% (628) within 3 months, 1.9% (670) 1 year, 2.1% (731) 5 years, and 2.2% (752) any time after TBFHL, compared to 0.0% (16,197) within 3 months, 0.0% (22,215) 1 year, 0.0% (35,757) 5 years, and 0.1% (55,634) any time after index in the control.
Adults with meningitis occurring after TBFHL had an increased risk of CSF leak at 3 months (RR: 2.70, 95% CI: 1.35–5.39), 1 year (RR: 2.90, 95% CI: 1.46–5.75), 5 years (RR: 3.00, 95% CI: 1.52–5.93), and any time after TBFHL + Meningitis (RR: 3.00, 95% CI: 1.52–5.93). No cases of CSF leak were reported for patients in the TBFHL—Meningitis cohort during the 3‐month period. In the TBFHL + Meningitis cohort, CSF leaks developed in 17.1% (27) at 3 months, 18.4% (29) 1 year, 19.0% (30) 5 years, and 19.0% (30) any time, compared to 6.3% (10) at 3 months, 1 year, 5 years, and any time in the TBFHL—Meningitis cohort.
4. Discussion
This retrospective cohort study assessed complications and outcomes following TBFHL in adults using a large multinational database. The results highlight the increased risk across multiple time points of meningitis, FND, labyrinthitis, CSF leak, and utilization of CI to treat the severe –profound SNHL that can occur as a sequela of TBF.
4.1. Meningitis
Meningitis as a complication of TBF, neurotrauma, or other head trauma in adults is reported to occur between 1% [21] and 25% [10, 22, 23, 24]. Rates of meningitis vary based on the presence of a CSF leak, ranging from 5%–11% if persisting less than one week, and in up to 88% with an untreated leak [21]. Wilkerson and colleagues reported that meningitis occurs in 1% of cases of TBF without CSF leak [21]. Increased long‐term risk of meningitis may be due to a persistent tegmen or otic capsule defect [25]. The results of this study demonstrate a similar risk of meningitis after TBF, ranging from 0.3% within 3 months up to 0.5% any time after TBFHL.
Among individuals who developed meningitis after TBFHL (TBFHL+ Meningitis cohort), 37% had a diagnosis prior to TBFHL of unspecified meningitis, and 20% had a previous diagnosis of bacterial meningitis. Previous literature has indicated that a history of meningitis is a risk factor for recurrence [26], estimated to be 5% [27].
4.2. Pneumococcal Vaccination
Vaccination status is important to consider, as pneumococcal vaccines may have a protective benefit for meningitis in adults with TBF [24, 27, 28, 29]. Per the CDC, pneumococcal vaccination coverage rates between 2008 and 2022 range from 22% to 36% for adults ages 18–64 [20], compared to approximately 65%–74% for adults 65 and older during the same time frame [20, 30]. TriNetX results were significantly below these rates. Multiple explanations are plausible, such as missing documentation if performed at a facility outside the scope of TriNetX. Another possibility is that the cohort population was too old for the first round of PCV7 (first available in 2000) but also too young for the recommended pneumococcal vaccine at age 65 (first available in 1984) [29].
4.3. Labyrinthitis
Labyrinthitis is characterized by inflammation of the inner ear and is thought to occur most commonly from infection. It can also occur secondary to head trauma and/or TBF [31, 32, 33, 34, 35, 36, 37]. Data is limited, but this condition seems to occur most commonly between ages 30 and 50 [31]. In this study, labyrinthitis occurred in 0.2% of adults following TBFHL, compared to 0.1% in adults without TBF. Meningitis occurring after TBFHL did not increase the risk of labyrinthitis. As both meningitis and head trauma have been reported as potential individual risk factors for labyrinthitis [31, 35, 37], it is surprising that having both did not increase the risk of labyrinthitis in this study. This may be due to the relatively small cohort size of 158.
4.4. Cochlear Implantation
HL is a frequent complication of TBF, with estimates ranging from 26%–66% (CHL) and 5%–23% (SNHL) [1, 8, 13, 14, 15, 16, 38]. Due to the high rates of HL that occur and the lack of a specific ICD‐10 diagnosis code for TBF, we utilized a diagnosis of HL in our inclusion criteria in building the TBFHL cohort. Because of this, we cannot estimate the rate of HL in TBF. To the best of our knowledge, the prevalence of CI after TBF has not been published in the literature. In our study, patients with TBFHL were more likely to undergo CI than those without, occurring in 0.4% at any time after. Despite all patients having HL by inclusion, rates of CI were low and did not significantly increase after FDA approval for SSD. This is not out of line with current estimates of CI utilization at 2.1% when including SSD candidates [39]. Due to the study design, we are unable to determine how many patients qualified for CI, and further investigation is warranted. Hearing aid use rates are estimated to be much higher than CI (24%) [40] in the literature; this outcome was not investigated in the present study. In this study, TBFHL + Meningitis did not appear to be significantly associated with CI, but evaluation is limited by the small sample size (158).
Post‐traumatic CI is complex and involves additional pertinent details not available in the TriNetX database. Not all TBFHL patients are suitable candidates for implantation due to anatomical changes caused by trauma, such as cochlear duct obliteration. While this relationship cannot be elucidated entirely from this database study, it is clear that TBFHL increases the likelihood of needing a CI.
4.5. Facial Nerve Disorders
Injury to the facial nerve is estimated to occur in 7% to 12% of adults after TBF [1, 15, 21], and is more likely after transverse fractures [1]. In our cohort, FND occurred in 4.6% of adults within 3 months and in 6.9% any time after TBFHL.
4.6. CSF Leaks
Estimates for CSF leaks after TBF range from 15% to 45% [1, 21]. Traumatic leaks elevate the risk of meningitis [7, 9, 27, 28]. In the current study, CSF leak occurred at a lower rate of 1.8% within the first 3 months and in 2.2% at any time after TBFHL. The increased lifetime risk may be explained by an unrecognized CSF leak during the first three months that was subsequently recognized. Rates of CSF leak after TBFHL + Meningitis are more in line with currently published literature, occurring in 17.1% of cases within the first three months. This increase may be due to increased scrutiny of the TBFHL due to the infection, with the leak likely occurring before the initial infection. For the TBFHL cohort, the rate is substantially lower than previously reported rates. Our study benefits from a significantly larger sample size but may lack the additional context available in single‐institution studies. Additionally, inaccurate coding may have led to missing data in TriNetX regarding the frequency of CSF leaks.
5. Strengths and Limitations
The primary strength of this study is the large cohort size with access to millions of patient records afforded by TriNetX. The conclusions drawn may be generalizable in a variety of contexts. Due to its retrospective design and observational nature, only associations can be determined.
TriNetX has several limitations inherent in its use. It is considered a “live” database with records being continuously uploaded. This affects reproducibility as cohort sizes change. Data are reported by individual HCOs and may be incorrectly coded or missing if patients transfer to non‐participating HCOs or are lost to follow‐up. We estimate that patient transfer is uncommon, given the median length of follow‐up is 1006 days and greater than 75% having at least 1 year of data. Missing data is more difficult to verify and are likely inevitable, but we expect that major complications are correctly coded at most HCOs. Given the large sample size of the study, mistakes in coding are unlikely to change the results of the analysis. It is important to note that in this study, a random, arbitrary event was used to define the index event for the control. TriNetX reports all patient counts between 1 and 10 as 10, meaning all calculations using these counts may not be accurate, disproportionately affecting small sample sizes. We are limited by reliance on medical codes like ICD‐10 and CPT, necessitating multiple codes to approximate TBF and assuming diagnoses were coded correctly. Lastly, we do not have access to more granular details, such as history, exams, audiograms, geographic distribution of cases, insurance coverage, and other details to provide more detailed insight.
6. Conclusion
This large retrospective cohort study found that TBFHL in adults significantly increased the risk of meningitis, FND, labyrinthitis, CSF leaks, and CI when compared to adults without TBF. The risk of meningitis, FND, labyrinthitis, CSF leaks, and the need for CI remained elevated in most cases for years after TBFHL. Meningitis occurring after TBFHL increased the risk of CSF leaks. Rates of TBFHL appear to be increasing.
Discloser
Meeting Information: Presented at the Triological Annual Meeting at the Combined Otolaryngology Spring Meetings in Chicago, Illinois in May 2024.
Conflicts of Interest
Neil S. Patel, MD has received grant and research support from Cochlear Corp., honoraria from IotaMotion and Zeiss, and is a consultant for Viridian Therapeutics. Richard K. Gurgel, MD has received grant and research support from Advanced Bionics, Cochlear Americas, and Neosensory, and is on the advisory committee and board for Med‐El. No conflicts of interest were declared for the remaining authors.
Steele J. L., Smith H. J., Takkoush S., et al., “Long‐Term Outcomes of Adult Temporal Bone Fractures With Hearing Loss: Results of a Multinational Database Analysis,” The Laryngoscope 135, no. 9 (2025): 3338–3347, 10.1002/lary.32140.
Funding: The authors received no specific funding for this work.
References
- 1. Johnson F., Semaan M. T., and Megerian C. A., “Temporal Bone Fracture: Evaluation and Management in the Modern era,” Otolaryngologic Clinics of North America 41, no. 3 (2008): 597–618, 10.1016/j.otc.2008.01.006. [DOI] [PubMed] [Google Scholar]
- 2. Diaz R. C., Cervenka B., and Brodie H. A., “Treatment of Temporal Bone Fractures,” J Neurol Surg Part B Skull Base 77, no. 5 (2016): 419–429, 10.1055/s-0036-1584197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Travis L. W., Stalnaker R. L., and Melvin J. W., “Impact Trauma of the Human Temporal Bone,” J Trauma Inj Infect Crit Care 17, no. 10 (1977): 761–766, 10.1097/00005373-197710000-00003. [DOI] [PubMed] [Google Scholar]
- 4. Ahmed Mohammed Islah S. and Rose Jacob N., “A Study of Temporal Bone Fractures With OTOLOGICAL Manifestations,” Journal of Evolution of Medical and Dental Sciences 6, no. 9 (2017): 710–715, 10.14260/Jemds/2017/153. [DOI] [Google Scholar]
- 5. Medina M., Di Lella F., Di Trapani G., et al., “Cochlear Implantation Versus Auditory Brainstem Implantation in Bilateral Total Deafness After Head Trauma: Personal Experience and Review of the Literature,” Otology & Neurotology 35, no. 2 (2014): 260–270, 10.1097/MAO.0000000000000235. [DOI] [PubMed] [Google Scholar]
- 6. Perheentupa U., Kinnunen I., Grénman R., Aitasalo K., and Mäkitie A. A., “Management and Outcome of Pediatric Skull Base Fractures,” International Journal of Pediatric Otorhinolaryngology 74, no. 11 (2010): 1245–1250, 10.1016/j.ijporl.2010.07.023. [DOI] [PubMed] [Google Scholar]
- 7. Exadaktylos A. K., Sclabas G. M., Nuyens M., et al., “The Clinical Correlation of Temporal Bone Fractures and Spiral Computed Tomographic Scan: A Prospective and Consecutive Study at a Level I Trauma Center,” Journal of Trauma: Injury, Infection, and Critical Care 55, no. 4 (2003): 704–706, 10.1097/01.TA.0000038550.11890.A5. [DOI] [PubMed] [Google Scholar]
- 8. Podoshin L. and Fradis M., “Hearing Loss After Head Injury,” Archives of Otolaryngology 101, no. 1 (1975): 15–18, 10.1001/archotol.1975.00780300019004. [DOI] [PubMed] [Google Scholar]
- 9. Poletti‐Muringaseril S. C., Rufibach K., Ruef C., Holzmann D., and Soyka M. B., “Low Meningitis‐Incidence in Primary Spontaneous Compared to Secondary Cerebrospinal Fluid Rhinorrhoea,” Rhinology 50, no. 1 (2012): 73–79, 10.4193/Rhino11.124. [DOI] [PubMed] [Google Scholar]
- 10. Frazee R. C., Mucha P., Farnell M. B., and Ebersold M. J., “Meningitis After Basilar Skull Fracture,” Postgraduate Medicine 83, no. 5 (1988): 267–274, 10.1080/00325481.1988.11700238. [DOI] [PubMed] [Google Scholar]
- 11. Thorén H., Mäyränpää M. K., Mäkitie A., Niemensivu R., Suominen A., and Snäll J., “Otologic Injuries Are Frequent in Pediatric Patients With Temporal Bone Fractures,” Journal of Cranio‐Maxillofacial Surgery 51, no. 1 (2023): 24–30, 10.1016/j.jcms.2023.01.011. [DOI] [PubMed] [Google Scholar]
- 12. von Elm E., Altman D. G., Egger M., et al., “Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for Reporting Observational Studies,” BMJ 335, no. 7624 (2007): 806–808, 10.1136/bmj.39335.541782.AD. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Stanisce L., Chryssofos S., Modi R., et al., “Long‐Term Outcomes of Acute Temporal Bone Fractures,” Laryngoscope 135, no. 5 (2025): 1803–1808, 10.1002/lary.31980. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Honeybrook A., Patki A., Chapurin N., and Woodard C., “Hearing and Mortality Outcomes Following Temporal Bone Fractures,” Craniomaxillofacial Trauma & Reconstruction 10, no. 4 (2017): 281–285, 10.1055/s-0037-1601885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Brodie H. A. and Thompson T. C., “Management of Complications From 820 Temporal Bone Fractures,” American Journal of Otology 18, no. 2 (1997): 188–197. [PubMed] [Google Scholar]
- 16. Kim B. J., Song I., Choi J., and Chan R. Y., “Contralateral Hearing Loss in Temporal Bone Fractures: A Potential Association With Combined Intracranial Injury,” J Int Adv Otol 20, no. 3 (2024): 210–215, 10.5152/iao.2024.22989. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Franchetti Y., “Use of Propensity Scoring and Its Application to Real‐World Data: Advantages, Disadvantages, and Methodological Objectives Explained to Researchers Without Using Mathematical Equations,” Journal of Clinical Pharmacology 62, no. 3 (2022): 304–319, 10.1002/jcph.1989. [DOI] [PubMed] [Google Scholar]
- 18. Benedetto U., Head S. J., Angelini G. D., and Blackstone E. H., “Statistical Primer: Propensity Score Matching and Its Alternatives†,” European Journal of Cardio‐Thoracic Surgery 53, no. 6 (2018): 1112–1117, 10.1093/ejcts/ezy167. [DOI] [PubMed] [Google Scholar]
- 19. Cioci A. C., Cioci A. L., Mantero A. M. A., Parreco J. P., Yeh D. D., and Rattan R., “Advanced Statistics: Multiple Logistic Regression, Cox Proportional Hazards, and Propensity Scores,” Surgical Infections 22, no. 6 (2021): 604–610, 10.1089/sur.2020.425. [DOI] [PubMed] [Google Scholar]
- 20. CDC , “AdultVaxView | General Population Reports | Vaccination Coverage.” (2022).
- 21. Wilkerson B., Brodie H., and Ahmed M., Management of Temporal Bone Trauma (Cummings Otolaryngology: Head and Neck Surgery, 2023). [Google Scholar]
- 22. Rathore M. H., “Do Prophylactic Antibiotics Prevent Meningitis After Basilar Skull Fracture?,” Pediatric Infectious Disease Journal 10, no. 2 (1991): 87–88. [DOI] [PubMed] [Google Scholar]
- 23. Helling T. S., Evans L. L., Fowler D. L., Hays L. V., and Kennedy F. R., “Infectious Complications in Patients With Severe Head Injury,” Journal of Trauma and Acute Care Surgery 28, no. 11 (1988): 1575–1577. [DOI] [PubMed] [Google Scholar]
- 24. Hedberg A. L., Pauksens K., Enblad P., et al., “Pneumococcal Polysaccharide Vaccination Administered Early After Neurotrauma or Neurosurgery,” Vaccine 35, no. 6 (2017): 909–915, 10.1016/j.vaccine.2016.12.065. [DOI] [PubMed] [Google Scholar]
- 25. Sudhoff H. and Linthicum F. H. J., “Temporal Bone Fracture and Latent Meningitis: Temporal Bone Histopathology Study of the Month,” Otology & Neurotology 24, no. 3 (2003): 521. [DOI] [PubMed] [Google Scholar]
- 26. Pimentel J., Escada P., D'Almeida G. N., Oliveira F., Oliveira V., and Marques L., “Subtotal Petrosectomy With Obliteration for the Treatment of a Temporal Bone Fracture Associated With Relapsing Bacterial Meningitis,” J Neurol Surg Part B Skull Base 73, no. S 02 (2012): A510, 10.1055/s-0032-1314408. [DOI] [Google Scholar]
- 27. van de Beek D., Cabellos C., Dzupova O., et al., “ESCMID Guideline: Diagnosis and Treatment of Acute Bacterial Meningitis,” Clinical Microbiology and Infection 22 (2016): S37–S62, 10.1016/j.cmi.2016.01.007. [DOI] [PubMed] [Google Scholar]
- 28. La Russa R., Maiese A., Di Fazio N., et al., “Post‐Traumatic Meningitis Is a Diagnostic Challenging Time: A Systematic Review Focusing on Clinical and Pathological Features,” International Journal of Molecular Sciences 21, no. 11 (2020): 4148, 10.3390/ijms21114148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Chu V., Carpenter D., Winter K., Harriman K., and Glaser C., “Head and Spine Injuries Increase Risk of Streptococcus Pneumoniae Meningitis in Adults,” Open Forum Infectious Diseases 5, no. suppl_1 (2018): S132, 10.1093/ofid/ofy210.341. [DOI] [PubMed] [Google Scholar]
- 30. “Vaccination Coverage Among Adults in the United States, National Health Interview Survey.” (2023).
- 31. Taxak P. and Ram C., “Labyrinthitis and Labyrinthitis Ossificans ‐ A Case Report and Review of the Literature,” J Radiol Case Rep 14, no. 5 (2020): 1–6, 10.3941/jrcr.v14i5.3706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. “Infectious Diseases & Conditions.” (2024).
- 33. Labyrinthitis , “Document ‐ Gale General OneFile” (2024).
- 34. “Labyrinthitis: MedlinePlus Medical Encyclopedia.” (2024).
- 35. Frey R. J. and Watson S., “Labyrinthitis,” Gale Encycl Med 5 (2019): 2939–2941. [Google Scholar]
- 36. Trakimas D. R., Knoll R. M., Castillo‐Bustamante M., Kozin E. D., and Remenschneider A. K., “Otopathologic Analysis of Patterns of Postmeningitis Labyrinthitis Ossificans,” Otolaryngol Neck Surg 164, no. 1 (2021): 175–181, 10.1177/0194599820934748. [DOI] [PubMed] [Google Scholar]
- 37. “Labyrinthitis – ClinicalKey.” (2024).
- 38. Deshmukh K. A., Fatima U., Siddiqui A., and Tegnoor M. S., “Evaluation and Outcomes of Hearing Loss in Temporal Bone Fractures: A Prospective Study,” Cureus 15, no. 10 (2023): e46331, 10.7759/cureus.46331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Nassiri A. M., Sorkin D. L., and Carlson M. L., “Current Estimates of Cochlear Implant Utilization in the United States,” Otology & Neurotology 43, no. 5 (2022): e558–e562, 10.1097/mao.0000000000003513. [DOI] [PubMed] [Google Scholar]
- 40. Haile L. M., Orji A. U., Reavis K. M., et al., “Hearing Loss Prevalence, Years Lived With Disability, and Hearing Aid Use in the United States From 1990 to 2019: Findings From the Global Burden of Disease Study,” Ear and Hearing 45, no. 1 (2024): 257–267, 10.1097/AUD.0000000000001420. [DOI] [PMC free article] [PubMed] [Google Scholar]