Abstract
Background:
The operating room start time (ORST) for pediatric femoral fractures is a health-care quality metric used for hospital rankings and accreditation. Factors affecting ORST remain unclear. This study aimed to evaluate the demographic and clinical factors associated with gold-standard (early) ORST (<18 hours) versus delayed ORST (≥18 hours) for pediatric femoral fractures.
Methods:
A retrospective review was conducted of 216 pediatric patients with a femoral shaft fracture admitted to the emergency department (ED) at a pediatric Level-I trauma hospital from 2021 to 2023. Patient demographic and clinical data were analyzed to identify significant factors associated with ORST. Immediate postoperative outcomes were compared across ORST groups.
Results:
In multivariable models, race, ED admission time, comorbidities, and surgery type affected ORST (p < 0.05). Compared with White patients, patients of other racial or ethnic groups, including Hispanic, Black, Asian, and multiracial patients, had 2.4 times higher odds of delayed ORST. Compared with midnight to 6 a.m. ED admissions, the odds of delayed ORST were 6.6 times higher for ED admissions between 6 a.m. and noon and 9.2 times higher for ED admissions between noon and 6 p.m. Patients with comorbidities were 4.7 times more likely to experience delayed surgery compared with healthy patients. Patients who underwent open reduction and internal fixation (ORIF) were 2.5 times as likely as patients who underwent closed reduction (CR) with a spica cast to have delayed ORST. Delayed ORST was associated with longer hospital stay (median, 71 hours) compared with early ORST (median, 41 hours), but not with immediate complications.
Conclusions:
ED admission time, race, method of transfer, comorbidities, and procedure type were associated with ORST for pediatric femoral fractures. Longer ORST led to a disproportionately longer hospital stay.
Level of Evidence:
Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.
In the last few decades, the health-care system has seen a sharp rise in focus on quality-of-care standards, bolstered by quality metrics and rewards for quality care. U.S. News and World Report is one organization whose hospital ranking system is of notable interest to hospitals, physicians, and patients1. Femoral fracture is the most common pediatric orthopaedic injury requiring hospitalization2,3 and has become a focus of U.S. News and World Report, which rewards hospitals that definitively stabilize higher percentages of patients with a femoral fracture within 18 hours. In 2023, U.S. News and World Report increased threshold standards, rewarding 1 point for ≥60% of patients, 2 points for ≥80% of patients, and 3 points for ≥90% of patients1.
Some studies have asserted that the 18-hour cutoff for time to surgery is arbitrary, without a substantial impact on outcomes4,5, whereas others have asserted that a time to surgery of <18 hours is important, as it represents more efficient processes for patient-centered pediatric trauma care, expeditious relief of pain, and decreased length of hospitalization6.
Previous studies have not focused on the factors that contribute to operating room start time (ORST). Therefore, our primary objective was to determine the demographic and clinical predictors of ORST for pediatric femoral shaft fractures. A secondary objective was to investigate the relationship between ORST and immediate clinical outcomes. We hypothesized that patient demographic characteristics, social factors, and injury severity would have an effect on ORST. In addition, we hypothesized that longer ORST would be associated with longer hospital stay.
Materials and Methods
With institutional review board approval, the study was conducted at a tertiary-care, Level-I pediatric trauma hospital. All patients admitted from the emergency department (ED) for a pediatric midshaft femoral fracture between January 1, 2021, and December 31, 2023, were included. Data on standard demographic characteristics (i.e., sex, age, race), language spoken, need for an interpreter, Child Opportunity Index (COI), and insurance type were collected. Clinical data included ED admission time, ambulance transfer from the scene, transfer from an outside hospital, mechanism of injury, Injury Severity Score (ISS), polytrauma, comorbidities, pathologic fracture, ORST, and procedure type. Procedure type was either closed reduction (CR) with a spica cast or open reduction and internal fixation (ORIF), which refers to any surgical instrumentation with or without the need to open the skin at the fracture site. Postoperative outcomes (length of hospitalization, blood transfusion, and immediate complications) were analyzed as a function of ORST, as well as demographic and clinical factors.
Statistical Analysis
Patient demographic characteristics and clinical data were summarized for the overall cohort and were stratified by ORST groups using medians and interquartile ranges (IQRs) for continuous variables and frequencies and percentages for categorical variables. ORST was assessed as a continuous and dichotomous outcome, with <18 hours considered to be early and ≥18 hours considered to be delayed.
Univariate comparisons across the ORST groups (early compared with delayed) were performed using Wilcoxon rank-sum tests for continuous variables and Fisher exact tests or Pearson chi-square tests for categorical variables, as appropriate.
Factors that were significant in the univariate comparisons were included in logistic regression models to assess their association with the ORST group. All significant factors from the univariate logistic regression models were included in a multivariable logistic regression model, using a stepwise selection approach. Odds ratios (ORs) and 95% confidence intervals (CIs) were estimated. A nomogram was constructed on the basis of the final multivariable logistic regression model, providing a visual representation of the probability of delayed ORST based on the combination of significant predictors.
To compare immediate postoperative outcomes (length of hospitalization, complications, and transfusion) across ORST groups, the Wilcoxon rank-sum test or Fisher exact test was used, as appropriate. A multivariable, negative binomial regression model was used to test the association between the ORST groups and length of hospitalization.
The analysis was conducted using R (version 4.3.2; The R Foundation). Significance was set at p < 0.05.
Results
Summary Statistics
The demographic and hospitalization characteristics are presented in Table I. Of 216 patients, 71% underwent surgery <18 hours after admission to the ED, whereas 29% experienced a delay.
TABLE I.
Demographic and Hospitalization Characteristics of Pediatric Patients with Femoral Fracture
| Characteristic | Overall* (N = 216) | Early* (<18 Hours) (N = 154) | Delayed* (≥18 Hours) (N = 62) | P Value† |
|---|---|---|---|---|
| Sex | 0.614 | |||
| Female | 61 (28%) | 45 (29%) | 16 (26%) | |
| Male | 155 (72%) | 109 (71%) | 46 (74%) | |
| Age at surgery (yr) | 6 (2, 11) | 4 (2, 11) | 9 (3, 12) | 0.015 |
| Race (n = 187) | 0.037 | |||
| White | 109 (58%) | 85 (63%) | 24 (45%) | |
| Racial or ethnic minority | 59 (32%) | 35 (26%) | 24 (45%) | |
| Asian or South Asian | 7 (4%) | 3 (2%) | 4 (8%) | |
| Black or African American | 15 (8%) | 8 (6%) | 7 (13%) | |
| Hispanic or Spanish origin | 37 (20%) | 24 (18%) | 13 (24%) | |
| Multiracial | 19 (10%) | 14 (10%) | 5 (9%) | |
| Primary language (n = 214) | 0.658 | |||
| English | 195 (91%) | 141 (92%) | 54 (89%) | |
| Spanish | 10 (5%) | 6 (4%) | 4 (7%) | |
| Other | 9 (4%) | 6 (4%) | 3 (5%) | |
| Need for an interpreter (n = 215) | 12 (6%) | 7 (5%) | 5 (8%) | 0.334 |
| COI (n = 213) | 81 (46, 92) | 81 (47, 93) | 82 (43, 92) | 0.500 |
| COI category (n = 213) | 0.122 | |||
| Very low | 17 (8%) | 8 (5%) | 9 (15%) | |
| Low | 24 (11%) | 19 (12%) | 5 (8%) | |
| Moderate | 21 (10%) | 15 (10%) | 6 (10%) | |
| High | 34 (16%) | 28 (18%) | 6 (10%) | |
| Very high | 117 (55%) | 83 (54%) | 34 (57%) | |
| Insurance | 0.581 | |||
| Public | 87 (40%) | 62 (40%) | 25 (40%) | |
| Private | 108 (50%) | 79 (51%) | 29 (47%) | |
| Other | 21 (10%) | 13 (8%) | 8 (13%) | |
| ED admission time | <0.001 | |||
| 6 a.m. to noon | 15 (7%) | 7 (5%) | 8 (13%) | |
| Noon to 6 p.m. | 70 (32%) | 31 (20%) | 39 (63%) | |
| 6 p.m. to midnight | 98 (45%) | 87 (56%) | 11 (18%) | |
| Midnight to 6 a.m. | 33 (15%) | 29 (19%) | 4 (6%) | |
| Ambulance transfer (n = 215) | 68 (32%) | 36 (24%) | 32 (52%) | <0.001 |
| Outside hospital transfer (n = 215) | 92 (43%) | 77 (50%) | 15 (24%) | <0.001 |
| Mechanism of injury | 0.992 | |||
| Pedestrian or bike struck by a motor vehicle | 10 (5%) | 6 (4%) | 4 (6%) | |
| Motor vehicle accident | 9 (4%) | 6 (4%) | 3 (5%) | |
| Sports-related injury | 23 (11%) | 16 (10%) | 7 (11%) | |
| Fall from ground level | 102 (47%) | 74 (48%) | 28 (45%) | |
| Fall from height | 14 (6%) | 10 (6%) | 4 (6%) | |
| Jumping on surface, trampoline, or bounce house | 9 (4%) | 7 (5%) | 2 (3%) | |
| Other | 38 (18%) | 27 (18%) | 11 (18%) | |
| Unknown or unclear mechanism | 11 (5%) | 8 (5%) | 3 (5%) | |
| ISS | 9 (9, 9) | 9 (9, 9) | 9 (9, 9) | 0.979 |
| ISS category | 0.174 | |||
| Minor | 0 (0%) | 0 (0%) | 0 (0%) | |
| Moderate | 211 (98%) | 152 (99%) | 59 (95%) | |
| Severe | 2 (1%) | 1 (1%) | 1 (2%) | |
| Very severe | 3 (1%) | 1 (1%) | 2 (3%) | |
| Polytrauma | 25 (12%) | 15 (10%) | 10 (16%) | 0.184 |
| Comorbidities | 55 (25%) | 29 (19%) | 26 (42%) | <0.001 |
| Type of comorbidity (n = 55) | ||||
| Neurodevelopmental disorder | 10 (18%) | 5 (17%) | 5 (19%) | >0.999 |
| Skeletal dysplasia | 3 (5%) | 1 (3%) | 2 (8%) | 0.598 |
| Cardiac disease | 8 (15%) | 2 (7%) | 6 (23%) | 0.131 |
| Neuromuscular disease | 26 (47%) | 15 (52%) | 11 (42%) | 0.485 |
| Other major disease | 31 (56%) | 17 (59%) | 14 (54%) | 0.721 |
| Procedure type | 0.002 | |||
| ORIF | 111 (51%) | 72 (45%) | 42 (68%) | |
| CR with spica cast | 105 (49%) | 85 (55%) | 20 (32%) |
The values are given either as the number of patients, with the percentage in parentheses, or as the median, with the IQR in parentheses.
P values were calculated using the Wilcoxon rank-sum test, Fisher exact test, or Pearson chi-square test.
Most patients were male (72%), with a median age of 6 years (IQR, 2 to 11 years). The cohort was predominantly White (58%) and English-speaking (91%) and approximately half had a very high COI (55%) and private insurance (50%). Most patients were admitted to the ED between 6 p.m. and midnight (45%). Thirty-two percent of patients were transferred by ambulance from the scene, whereas 43% were transferred from an outside hospital. The primary injury mechanism was a fall from ground level (47%), and most injuries had a moderate ISS (98%). Twelve percent of patients sustained polytrauma, and 25% had serious comorbidities. The majority of patients underwent ORIF (53%).
Patients who experienced a surgical delay were older (median age, 9 years) than the early surgery cohort (median age, 4 years) (p = 0.02). The delayed group had a higher proportion of patients transferred from the scene via ambulance (52%) compared with the early group (24%) (p < 0.001). Conversely, the early group had a higher proportion of patients transferred from an outside hospital (50%) compared with the delayed group (24%) (p < 0.001). The delayed group had a higher proportion of patients requiring ORIF (69%) than patients requiring CR with a spica cast (31%) (p = 0.002).
Factors Associated with ORST
Table II summarizes logistic regression analyses of associations between demographic and clinical characteristics and the odds of delayed ORST. In univariate analyses, age, race, ED admission time, ambulance transfer, outside hospital transfer, comorbidities, and procedure type were associated with a delayed time to surgery (p < 0.05). In the final multivariable model, race, ED admission time, comorbidities, and procedure type remained significant predictors of delayed ORST. The ISS was considered an important possible confounding variable and therefore was included in the multivariable analysis, even though it was not a significant predictor of ORST in univariate analyses. When interpreting the final multivariable model, compared with White patients, patients of other racial or ethnic groups, including Hispanic, Black, Asian, and multiracial patients, had 2.4 times (95% CI, 1.0 to 5.9 times) higher odds of delayed ORST, after adjusting for ED admission time, comorbidities, ISS, and procedure type. Patients admitted from 6 a.m. to noon and noon to 6 p.m. had higher odds of delayed ORST compared with patients admitted from midnight to 6 a.m., after controlling for race, comorbidities, ISS, and procedure type. Patients with comorbidities were 4.7 times (95% CI, 2.0 to 11.7 times; p < 0.001) more likely to experience delayed surgery compared with patients without comorbidities, after adjusting for race, ED admission time, ISS, and procedure type. Patients who underwent ORIF were 2.5 times (95% CI, 1.1 to 5.9 times; p < 0.001) more likely than patients who underwent CR with a spica cast to experience delayed ORST. Figure 1 shows the nomogram summarizing the multivariable logistic regression model.
Fig. 1.
Nomogram indicating the probability of delayed ORST using the multivariable logistic regression model. OR = operating room.
TABLE II.
Logistic Regression Models Assessing Factors Associated with Delayed ORST
| Factor | Univariate Analysis | Multivariable Analysis | Final Multivariable Model | |||
|---|---|---|---|---|---|---|
| OR* | P Value | OR* | P Value | OR* | P Value | |
| Age at surgery | 1.07 (1.01 to 1.14) | 0.014 | 1.04 (0.90 to 1.21) | 0.587 | — | — |
| Race | ||||||
| White | Reference | Reference | Reference | |||
| Racial or ethnic minority† | 2.43 (1.22 to 4.87) | 0.012 | 2.45 (1.03 to 5.98) | 0.045 | 2.44 (1.04 to 5.89) | 0.043 |
| Multiracial | 1.26 (0.38 to 3.69) | 0.68 | 2.59 (0.52 to 11.61) | 0.223 | 2.74 (0.57 to 11.76) | 0.185 |
| ED admission time | ||||||
| Midnight to 6 a.m. | Reference | Reference | Reference | |||
| 6 a.m. to noon | 8.29 (2.04 to 39.39) | 0.004 | 6.23 (1.12 to 39.49) | 0.041 | 6.62 (1.26 to 39.92) | 0.030 |
| Noon to 6 p.m. | 9.12 (3.18 to 33.23) | <0.001 | 8.48 (2.35 to 38.16) | 0.002 | 9.19 (2.68 to 39.50) | 0.001 |
| 6 p.m. to midnight | 0.92 (0.29 to 3.51) | 0.889 | 0.79 (0.21 to 3.51) | 0.736 | 0.67 (0.18 to 2.82) | 0.556 |
| Ambulance transfer | ||||||
| No | Reference | Reference | Reference | |||
| Yes | 3.47 (1.87 to 6.51) | <0.001 | 1.68 (0.59 to 4.89) | 0.333 | — | — |
| Outside hospital transfer | ||||||
| Yes | Reference | Reference | Reference | |||
| No | 3.17 (1.67 to 6.32) | 0.001 | 1.32 (0.44 to 3.96) | 0.614 | — | — |
| Comorbidities | ||||||
| No | Reference | Reference | Reference | |||
| Yes | 3.11 (1.63 to 5.97) | 0.001 | 4.27 (1.78 to 10.78) | 0.001 | 4.71 (2.01 to 11.69) | <0.001 |
| ISS | 1.10 (0.98 to 1.25) | 0.121 | 0.99 (0.81 to 1.21) | 0.902 | 0.98 (0.82 to 1.19) | 0.864 |
| Procedure type | ||||||
| ORIF | Reference | Reference | Reference | |||
| CR with spica cast | 2.59 (1.41 to 4.88) | 0.003 | 1.58 (0.34 to 7.04) | 0.554 | 2.49 (1.10 to 5.89) | 0.032 |
The values are given as the OR, with the 95% CI in parentheses.
Other racial or ethnic groups are Asian or South Asian, Black or African American, and Hispanic or Spanish origin.
Continuous Time to Operating Room
Table III summarizes ORST in its continuous form across the categorical factors found to be associated with ORST in univariate logistic regression models. The median ORST differed by race or ethnicity. Non-White patients, including Hispanic, Black, Asian, and multiracial patients, experienced a longer delay at 16 hours (IQR, 13 to 20 hours) compared with a delay of 15 hours (IQR, 11 to 18 hours) experienced by White patients. The time to surgery was also influenced by the admission time, ranging from a median time of 21 hours (IQR, 6 to 24 hours) for admissions between 6 a.m. and noon to 9 hours (IQR, 8 to 11 hours) for admissions between midnight and 6 a.m. Patients transferred by ambulance faced a longer median ORST of 18 hours compared with 14 hours for patients transferred by transportation other than an ambulance. Similarly, outside hospital transfers had a shorter median time of 14 hours compared with 17 hours for transfers that were not from an outside hospital. Patients with comorbidities experienced a slightly longer median ORST at 17 hours compared with 15 hours for those without comorbidities. Patients who underwent ORIF had a longer ORST (16 hours) than patients who underwent CR with a spica cast (14 hours). Additionally, patients with polytrauma experienced a slightly longer median time to the operating room at 16 hours compared with 15 hours for those with an isolated femoral fracture; however, this difference was not significant.
TABLE III.
Factors Affecting ORST (N = 216)
| Factor | ORST* (hr) |
|---|---|
| Race (n = 187) | |
| White | 15 (11, 18) |
| Other racial or ethnic groups | 16 (13, 20) |
| Asian or South Asian | 18 (12, 20) |
| Black or African American | 16 (13, 20) |
| Hispanic or Spanish origin | 16 (12, 20) |
| Multiracial | 15 (11, 19) |
| ED admission time | |
| 6 a.m. to noon | 21 (6, 24) |
| Noon to 6 p.m. | 18 (17, 21) |
| 6 p.m. to midnight | 14 (12, 17) |
| Midnight to 6 a.m. | 9 (8, 11) |
| Ambulance transfer (n = 215) | |
| Yes | 18 (14, 21) |
| No | 14 (11, 17) |
| Outside hospital transfer (n = 215) | |
| Yes | 14 (11, 17) |
| No | 17 (13, 20) |
| Polytrauma | |
| Yes | 16 (13, 22) |
| No | 15 (12, 18) |
| Comorbidities | |
| Yes | 17 (12, 21) |
| No | 15 (12, 18) |
| Procedure type | |
| ORIF | 16 (13, 20) |
| CR with spica cast | 14 (11, 17) |
The values are given as the median, with the IQR in parentheses.
Comparing Postoperative Outcomes
Table IV shows the differences in immediate postoperative outcomes between patients who had early surgery and those who had delayed surgery. Patients who experienced delayed surgery had a significantly longer median hospital stay (71 hours) compared with those in the early group (41 hours) (p < 0.001). However, the rates of immediate complications (6% in the delayed group compared with 3% in the early group; p = 0.2) and need for blood transfusion (5% in the delayed surgery group compared with 1% in the early surgery group; p = 0.1) were not significantly different between the 2 groups.
TABLE IV.
Comparison of Postoperative Outcomes by ORST Group
| Outcome | Early* (<18 Hours) (N = 154) | Delayed* (≥18 Hours) (N = 62) | P Value† |
|---|---|---|---|
| Length of hospital stay (hr) | 41 (24, 66) | 71 (44, 95) | <0.001 |
| Immediate complications | 4 (3%) | 4 (6%) | 0.230 |
| Need for blood transfusion (n = 215) | 2 (1%) | 3 (5%) | 0.146 |
The values are given either as the median, with the IQR in parentheses, or as the number of patients, with the percentage in parentheses.
P values were calculated using the Wilcoxon rank-sum test or the Fisher exact test.
Factors Associated with the Length of Hospital Stay
Table V presents the results of the multivariable negative binomial regression models examining the association between the ORST group and the length of hospitalization, while controlling for potential confounding factors. Compared with patients who underwent surgery within 18 hours, those who underwent delayed surgery experienced a 19% longer hospital stay (incidence rate ratio [IRR], 1.19 [95% CI, 1.02 to 1.39]; p = 0.03). Polytrauma was associated with a significant 57% increase in hospital stay (IRR, 1.57 [95% CI, 1.24 to 2.03]; p < 0.001). The presence of comorbidities was associated with a 39% longer hospitalization (IRR, 1.39 [95% CI, 1.05 to 1.09]; p = 0.003). A 1-point increase in ISS was associated with a 5% increase in the length of hospitalization (IRR, 1.05 [95% CI, 1.02 to 1.09]; p = 0.003). Furthermore, patients requiring blood transfusion had a 97% longer hospitalization than those without transfusion (IRR, 1.97 [95% CI, 1.24 to 3.31]; p = 0.003). The occurrence of complications was associated with a 133% increase in the length of hospitalization (IRR, 2.33 [95% CI, 1.68 to 3.35]; p = 0.001). Lastly, patients who underwent ORIF had a 93% (IRR, 1.93 [95% CI, 1.68 to 2.21]; p < 0.001) longer hospitalization compared with patients who underwent CR with a spica cast.
TABLE V.
Multivariable Negative Binomial Regression Model Assessing Factors Associated with Length of Hospital Stay
| Factor | IRR* | P Value |
|---|---|---|
| Time to OR (delayed vs. early) | 1.19 (1.02 to 1.39) | 0.025 |
| Polytrauma (yes vs. no) | 1.57 (1.24 to 2.03) | <0.001 |
| Comorbidities (yes vs. no) | 1.39 (1.19 to 1.63) | <0.001 |
| ISS | 1.05 (1.02 to 1.09) | 0.003 |
| Blood transfusion (yes vs. no) | 1.97 (1.24 to 3.31) | 0.003 |
| Complications (yes vs. no) | 2.33 (1.68 to 3.35) | <0.001 |
| Procedure type (ORIF vs. CR with spica cast) | 1.93 (1.68 to 2.21) | <0.001 |
The values are given as the IRR, with the 95% CI in parentheses.
Discussion
The debate continues as to whether a time of <18 hours to the definitive stabilization of pediatric femoral shaft fractures matters, either in terms of clinical outcomes or as a quality indicator. Several studies have found little meaningful impact, other than shorter intensive care unit and hospital stays, of early surgery on patient outcomes after pediatric femoral fractures4,5,7. The editorial response by Blakemore et al. to Grauberger et al. emphasized that, whether or not clinical outcomes change, expedient care should still be a marker of quality and of ideal health-care processes6. If one considers expedient care to be a priority, it is relevant to determine which patients are particularly at risk for having delayed stabilization of the femoral fracture. In our study, we examined the driving factors of ORST. Over a 3-year period, 71% of our femoral fractures were stabilized within 18 hours of ED admission. In the final multivariable model, race, ED admission time, comorbidities, and procedure type remained significant predictors of delayed time to surgery.
Procedure type, which tracked closely with age, was a significant factor in ORST, which likely reflects a surgeon’s ease of adding on a CR with a spica cast on the day of arrival without substantially prolonging the surgical day, as it requires fewer resources and time. One may also prioritize the fasting status of a young baby or toddler, shifting them to earlier surgical times8. ORIFs require more surgical time and resources and may not be desirable to add on the same day, but rather may be scheduled for the following morning, resulting in a longer wait time. The fact that older children needing ORIF are waiting longer highlights the need to have implant systems readily available and familiar to hospital staff to avoid delays.
The time of ED admission was, not surprisingly, an important factor in ORST. Patients who arrive early in the day but after surgical schedules are made (e.g., 6 a.m. to noon) will likely be admitted for a surgical procedure the following day. Likewise, those admitted from noon to 6 p.m. will not likely be ready for the operating room prior to the end of the normal workday. This may be less applicable to a CR with a spica cast that can be more easily fitted onto the end of a day because of its short duration. Patients admitted overnight are easily fitted into the following day’s schedule. The true challenge comes in accommodating the patient who arrives mid-morning or early afternoon. If the patient is not nil per os (NPO) for a sufficient time, or if the trauma room already has a full roster, accommodating a same-day surgical procedure can be unfeasible. As we become more sensitive to the high rates and harsh realities of physician burnout, pressuring physicians to add on cases in the evening hours in the absence of clear outcome benefits is unlikely to be a winning solution.
It was notable to find that patients transferred from outside hospitals arrived at the operating room faster than those presenting directly by ambulance. This is likely related to the time of admission, because transfers often arrive later in the day or even in the middle of the night. Additionally, outside hospital transfers will often come with appropriate imaging that makes workup at the accepting facility more streamlined9. This underscores the importance of early communication between the ED and the surgical team so that, in the event of a known surgical need, preparation can be accelerated. Every effort should be made to have patients transported expeditiously so that surgical procedures can occur the morning after arrival. Although we document the ORST from admission to the definitive hospital ED, we do not have the infrastructure to document and measure the time from admission to the original ED to ultimate surgical care. Two stops likely add a substantial amount of time to ultimate stabilization for the patient and warrant further study. It is important to expedite care in these patients whose total ORST is not fully captured in the treating hospital’s metrics.
Another important finding is the delayed ORST in patients with comorbidities, who also experienced a significantly longer hospital stay overall. Although not surprising, this emphasizes the importance of expeditious consultations to avoid delay in our most medically complex patients, while not sacrificing a safe and thorough preoperative workup, to minimize discomfort prior to stabilization and long NPO times. Patients with neuromuscular issues were common in our cohort and are particularly vulnerable in the perioperative period. These patients often benefit from pulmonology and cardiology consultations as well as special airway, pharmacological, and positioning considerations to avoid postoperative complications and progression of disease10. Geriatric protocols for preoperative medical optimization have been shown to streamline time to medical readiness, lower health-care costs, and decrease length of stay11,12. Similar pediatric protocols may be beneficial to streamline care for medically complex children.
One of the most important findings of this study was the difference in ORST based on race, which has been described in adults as well. Whether viewed as a categorical or continuous variable, ORST was significantly affected by race, with White patients experiencing shorter ORST than non-White patients. This finding was unexplained by insurance status, COI, or need for an interpreter. Non-White patients had 2.4 to 2.7 times greater odds of delayed surgery. This clear health-care disparity is likely multifactorial, involving institutional, clinical, and patient or cultural factors. There could be unconscious biases and forms of institutional racism that delay care for patients of color. Furthermore, a subtle interplay of injury and demographic characteristics could have evaded our statistical analysis. Although we do not yet understand the precise drivers of this finding, we are acting on it by engaging our Office of Health Equity and Inclusion and by initiating a deeper dive into the bottlenecks to expedient surgical care. A follow-up study has been designed to analyze available time stamps of care milestones following arrival to the ED. Factors such as rooming time, assignment of providers, time to pain medication, time to imaging studies, and time to admission will be analyzed with a critical lens toward health-care disparities.
The final finding that should be addressed is the disproportionate effect of delayed surgery on the duration of hospitalization. The duration of hospitalization has multiple drivers, but being in the delayed surgery group was associated with a 30-hour increase in hospital stay. In line with other studies, patients classified as having multiple traumatic injuries, comorbidities, or higher ISS also had longer hospital stays than their counterparts13,14. Patients who required a postoperative blood transfusion or had surgical complications had significant increases in hospital stay, similar to other studies15,16. Further, patients who underwent ORIF stayed in the hospital longer than patients who underwent CR with a spica cast. These findings are helpful to set expectations with families about the expected duration of stay and present targets for improvement.
This study had limitations, mostly due to its retrospective nature and inability to capture the complicated and nuanced realities of add-on trauma schedules. Furthermore, this study occurred at a tertiary-care, Level-I pediatric trauma hospital and may or may not be reflective of challenges in the ORST at a smaller community hospital or a hospital without a dedicated trauma room.
In conclusion, the results of our study determined that the time of ED admission, race, comorbidities, and procedure type were independently predictive factors of the ORST for pediatric femoral fractures. Although we continue to investigate the clinical implications of delay in care for patients with a femoral fracture, we can all agree that quality care for injured children is a universal priority. This study helps us to identify ways in which we can improve in this endeavor.
Footnotes
Investigation performed at Boston Children’s Hospital, Boston, Massachusetts
Disclosure: The study was funded by the Boston Children’s Hospital Orthopaedic Surgery Foundation. The Disclosure of Potential Conflicts of Interest forms are provided with the online version of the article (http://links.lww.com/JBJS/I529).
Contributor Information
Emi Schwab, Email: emi.schwab@childrens.harvard.edu.
Shanika De Silva, Email: shanika.desilva@childrens.harvard.edu.
References
- 1.Olmsted MG LS, Powell R, Murphy J, Bell D, Blackburn K, Stanley M, Sanchez RT, Allen R. Methodology: U.S. News & World Report Best Children’s Hospitals 2023-24. 2023. Jun 5. Accessed 2025 Feb 4. https://health.usnews.com/media/best-hospitals/BCH_Methodology_2023-2024.pdf
- 2.Loder RT, O’Donnell PW, Feinberg JR. Epidemiology and mechanisms of femur fractures in children. J Pediatr Orthop. 2006. Sep-Oct;26(5):561-6. [DOI] [PubMed] [Google Scholar]
- 3.Nakaniida A, Sakuraba K, Hurwitz EL. Pediatric orthopaedic injuries requiring hospitalization: epidemiology and economics. J Orthop Trauma. 2014. Mar;28(3):167-72. [DOI] [PubMed] [Google Scholar]
- 4.Grauberger J, O’Byrne M, Stans AA, Shaughnessy WJ, Larson AN, Milbrandt TA. Does shorter time to treatment of pediatric femur shaft fractures impact clinical outcomes? J Pediatr Orthop. 2020. Jul;40(6):e435-e439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hedequist D, Starr AJ, Wilson P, Walker J. Early versus delayed stabilization of pediatric femur fractures: analysis of 387 patients. J Orthop Trauma. 1999. Sep-Oct;13(7):490-3. [DOI] [PubMed] [Google Scholar]
- 6.Blakemore L, Garg S, Goodwin R, Gordon JE, Hahn G, Mehlman CT, Mencio G, Rathjen K, Sabatini C, Schmale G, Skaggs DL; USNWR Pediatric Orthopaedic Working Group. Letter to the Editor: Grauberger et al., J Pediatr Orthop, 2020 and Sullivan et al., J Pediatr Orthop, 2020. J Pediatr Orthop. 2023. Nov-Dec 01;43(10):e828. [DOI] [PubMed] [Google Scholar]
- 7.Mendelson SADT, Dominick TS, Tyler-Kabara E, Moreland MS, Adelson PD. Early versus late femoral fracture stabilization in multiply injured pediatric patients with closed head injury. J Pediatr Orthop. 2001. Sep-Oct;21(5):594-9. [PubMed] [Google Scholar]
- 8.Chon T, Ma A, Mun-Price C. Perioperative fasting and the patient experience. Cureus. 2017. May 24;9(5):e1272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Fernandes-Taylor S, Yang Q, Yang DY, Hanlon BM, Schumacher JR, Ingraham AM. Greater patient sharing between hospitals is associated with better outcomes for transferred emergency general surgery patients. J Trauma Acute Care Surg. 2023. Apr 1;94(4):592-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kynes JMBM, Blakely M, Furman K, Burnette WB, Modes KB. Multidisciplinary perioperative care for children with neuromuscular disorders. Children (Basel). 2018. Sep 12;5(9):126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kates SL, O’Malley N, Friedman SM, Mendelson DA. Barriers to implementation of an organized geriatric fracture program. Geriatr Orthop Surg Rehabil. 2012. Mar;3(1):8-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Steffensmeier A, Hoge C, Shah N, Matar R, Rice M, Grawe E, Held J, Budde B, Laughlin R, Sagi HC. Evaluation of a novel multidisciplinary preoperative workup strategy for geriatric hip fractures. J Orthop Trauma. 2022. Aug 1;36(8):413-9. [DOI] [PubMed] [Google Scholar]
- 13.Al-Otaiby M, Almutairi KM, Vinluan JM, Al Seraihi A, Alonazi WB, Qahtani MH, Aljeri T, Alhumud MA, Alobaidi N, Alhurishi SA. Demographic characteristics, comorbidities, and length of stay of COVID-19 patients admitted into intensive care units in Saudi Arabia: a nationwide retrospective study. Front Med (Lausanne). 2022. Jul 13;9:893954. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fenwick A, Pfann M, Mayr J, Antonovska I, Von der Helm F, Nuber S, Förch S, Mayr E. Concomitant fractures in patients with proximal femoral fractures lead to a prolonged hospital stay but not to increased complication rates or in-house mortality if treated surgically: a matched pair analysis. Aging Clin Exp Res. 2023. Mar;35(3):607-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Felix JAA, Soares A, Bastos A, Amaro A, Oliveira C, Rabiais S, Afonso-Silva M, Andreozzi V, Vandewalle B, Ferreira D. Impact of blood transfusions on hospital length of stay and mortality: a single-center experience. Value Health. 2015;18(7):A676-7. [Google Scholar]
- 16.Khan NAQH, Quan H, Bugar JM, Lemaire JB, Brant R, Ghali WA. Association of postoperative complications with hospital costs and length of stay in a tertiary care center. J Gen Intern Med. 2006. Feb;21(2):177-80. [DOI] [PMC free article] [PubMed] [Google Scholar]