Predictors of mortality in bicycle-related trauma: an eight-year experience in a level one trauma center.

DOI: https://doi.org/10.21203/rs.3.rs-1625407/v1

Abstract

Background: Bicycle-related trauma has increased during the last decades, mainly thanks to the antipollution environmental policies and the diffusion of bike-sharing companies. This study investigates the outcome of bicycle-related trauma in our level one trauma center over a period of eight years.

Methods: Data of 446 consecutive bicycle-related trauma admitted to our trauma center from 2011 to 2019 were selected and retrospectively analysed. The sample was divided into 3 age groups: ≤ 17 years, 18–54 years, and ≥ 55 years. Mortality rates were obtained for the overall population and patients with Injury Severity Score (ISS) ≥ 25. Month and seasonal patients’ distribution was described to provide an epidemiological overview of bike-related trauma over the years.

Results: Patients ³ 55 years showed a lower pre-hospital and in-hospital GCS (p≤ 0.001), higher levels of lactates (p< 0.019) and higher ISS (p≤ 0.001), probability of death (p≤ 0.001), and overall mortality (p≤ 0.001). The head and chest Abbreviated Injury Scale (AIS) ≥3 injuries were predictors of mortality, especially in patients over 55 years (p<0.010). However, more than 90% of patients wore a helmet. Patients less than 18 years old showed the higher survival rate (98.7%; Log-rank test ≤ 0.028). Bicycle-related trauma were more frequent during the summer (34%), particularly in July and August.

Conclusions: Age over 55 years old, head and chest injuries, and an ISS > 25 were independent predictors of mortality. Multiple factors probably contributed to mortality rates in these patients. 

Background

In recent years, cycling has become more popular than in the past, both for transport and for sports purposes1. Bicycle-related trauma is widespread in urban areas, representing a consistent percentage of road traffic victims worldwide2. Despite the injury prevention policies promoting helmet use and changes in bicycle infrastructure over the years, bicycle-related trauma seems to be even more frequent3,4. The spread of bicycle use was encouraged by the environmental anti-pollution policies and the diffusion of transport sharing companies3. Unlike the other road traffic trauma, cyclists are more incline to break road rules. This is also related to the geographic area (rural/urban) and to the availability of cycle tracks. Moreover, depending on the country, helmet use is not mandatory, and it is not even provided by most sharing companies5,6.

Most of the studies on bicycle-related trauma investigated the epidemiological and social impact of helmet use on cyclists, often with conflicting results7,8 .

This study, performed in a level-one trauma center, aims to investigate the outcome of bicycle-related trauma according to age, site of injuries, and seasonal distribution to improve the level of care in this type of patient.

Methods

All patients consecutively admitted to Niguarda hospital, a level-one trauma center in Milan, Italy, for bicycle-related trauma from October 2011 to October 2019 were retrospectively retrieved from our Trauma Registry. The study was conducted according to the principles declared by the National Commission for Data Protection and Liberties (CNIL: 2210699) and the ethical principles described in the Declaration of Helsinki. Our Trauma Registry project was approved from the local IRB: 534-102018.

Demographic data, mechanism of trauma, pre-hospital and in-hospital vital signs, trauma aggravating factors, abbreviated injury scale (AIS) of each anatomical region, Injury Severity Score (ISS), and probability of death (PD) estimated by Trauma and Injury Severity Score (TRISS) system were considered. The American Society of Anaesthesiologists (ASA) clinical status summarized patients’ comorbidities. Injuries were grouped by anatomical region according to AIS classification: head, face, chest, abdomen, extremities, and external. To describe their injuries and mortality distribution, patients were divided into three age groups (< 18 years, 18–54 years, and ≥ 55 years), while an ISS equal or higher than 25 was chosen as severity cut-off. Age and ISS cut-off were chosen accordingly to our previous study9 on predictors of mortality among two-wheeled vehicles.

Data were recorded in a computerized spreadsheet (Microsoft Excel 2016; Microsoft Corporation, Redmond; WA) and analysed with statistical software (IBM Corp., released 2012, IBM SPSS Statistics for Windows, Version 21.0; Armonk, NY, IBM Corp.). The sample distribution was evaluated with Kolmogorov-Smirnov and Shapiro-Wilk tests resulting in a non-Gaussian distribution for any examined variable. Continuous variables were compared by independent sample Kruskal-Wallis test, while categorical variables were analysed using Pearson’s chi-squared test. Two logistic regression analyses identified the association between AIS score and mortality and between aggravating factors and mortality by age groups, estimating the adjusted odds ratio (OR) and 95% confidence interval (CI). Variables with a p-value < 0.5 at the univariate analysis were included in the model. Survival curves were obtained with Kaplan-Meier analysis, and the log-rank test was assessed to evaluate differences in cumulative survival among age groups. Mortality rates were obtained for the overall population and patients with ISS ≥ 25, considering the age group stratification. Finally, month and seasonal patients’ distribution was described to provide an epidemiological overview of bike-related trauma over the year.

Results

During the study period, four hundred forty-six bicycle-related trauma were managed at our center. Seventy-five (16.8%) patients were younger than 18 years, 223 (50%) were between 18 and 54 years-old, and 148 (33.2%) were older than 55 years. Patients ≥ 55 years showed lower pre-hospital and in-hospital GCS (p ≤ 0.001), higher levels of lactates (p < 0.019), ISS (p ≤ 0.001), probability of death (p ≤ 0.001) and overall mortality (p ≤ 0.001) as shown in Table 1.

 
 
Table 1

Demographic and trauma related data

Variables

< 18 years

75 (16.8)

18–54 years

223 (50)

≥ 55 years

148 (33.2)

p value

Male n(%)

58 (77.3)

167 (74.6)

120 (81.6)

0.281

GCS [median (IQR)]

15 (15–15)

15 (14–15)

15 (13–15)

0.001*

GCS Hospital [median (IQR)]

15 (15–15)

15 (15–15)

15 (10–15)

0.001*

DBP [median (IQR)]

75 (60–80)

77 (70–80)

80 (70–94)

< 0.001

SBP [median (IQR)]

120 (112–130)

130 (120–140)

146 (125–165)

< 0.001

HR [median (IQR)]

100 (84–115)

80 (70–96)

80 (70–94)

≤ 0.001*

RR [median (IQR)]

18 (14–18)

16 (15–18)

16 (12–18)

0.006*

BE [median (IQR)]

-2 (-3.45-0.25)

-1.05 (-3.52-0.90)

-1.4 (-3.75-0.30)

0.499

Lactate [median (IQR)]

1.82 (1.36–2.35)

1.79 (1.30–2.60)

2.11 (1.49–2.97)

0.019*

OAC [median (IQR)]

1.12 (1.06–1.16)

1.05 (1-1.12)

1.06 (1-1.12)

≤ 0.001*

No Helmet n(%)

6 (8.1)

24 (10.7)

14 (9.5)

0.706

Roll on n(%)

2 (2.6)

7 (3.1)

6 (4.1)

0.825

Roll over n(%)

34 (45.3)

114 (50.9)

69 (46.9)

0.621

Throw n(%)

16 (21.3)

29 (12.9)

13 (8.8)

0.033*

Fall without accident n(%)

25 (33.3)

70 (31.3)

49(33.3)

0.895

Cardioaspirin n(%)

0

0

14 (9.5)

≤ 0.001*

OAC n(%)

0

0

3 (2.0)

0.046*

ASA SCORE n (%)

I

II

III

74 (98.7)

1 (1.3)

0

196 (87.5)

26 (11.6)

2 (0.9)

48(32.7)

80(54.4)

19 (2.9)

≤ 0.001*

Emergency surgery n(%)

14 (18.7)

48 (21.4)

35 (23.8)

0.671

Elective surgery n(%)

6 (8)

39 (17.4)

36 (24.5)

0.010*

Head AIS ≥ 3 n(%)

10 (2.2)

57 (12.8)

70 (15.7)

≤ 0.001*

Face AIS ≥ 3 n(%)

0

3 (0.7)

3 (0.7)

0.460

Chest AIS ≥ 3 n(%)

10 (2.2)

52 (11.7)

47 (10.5)

0.008*

Abdomen AIS ≥ 3 n(%)

8 (1.8)

10 (2.2)

5 (1.1)

0.055

Extremity AIS ≥ 3 n(%)

4 (0.9)

21 (4.7)

24 (5.4)

0.026*

Dead n(%)

1 (1.3)

11 (4.9)

20 (13.6)

≤ 0.001*

Length of stay [median (IQR)]

3 (0–12)

1 (0-9.75)

5 (0–19)

≤ 0.001*

ISS [median (IQR)]

5 [214]

6 [217]

17 [6–29]

≤ 0.001*

Probability of death

[median (IQR)]

0.40 [0.3-1]

0.50 [0.30–1.60]

6.10 [2.5–16.5]

≤ 0.001*

GCS: Glasgow Coma Scale; DBP: Diastolic Bloody Pression; SBP: Systolic Bloody Pression; HR: Heart Rate; RR: Respiratory Rate; BE: Base Excess; INR: International Normalized Ratio; OAC: Oral Anticoagulation; AIS: Abbreviated Injury Scale; ISS: Injury Severity Score; IQR: Interquartile range; * statistical significance.

AIS < 3 head, face, chest, abdomen, and extremity minor injuries were more frequent in patients between 18 and 54 years, while AIS ≥ 3 head, chest, abdomen, and extremity injuries were more frequent in patients over 55 years.

Table 2 shows the mortality distribution for age groups according to AIS grading for each body region. The logistic regression model showed that AIS ≥ 3 head and chest injuries were predictors of mortality in patients over 55 years.

 
 
 
 
Table 2

AIS distribution in age groups

 

Univariate analysis

Multivariate analysis

 

AIS < 3 n(%)

AIS ≥ 3 n(%)

p value

OR (95%IC)

p value

< 18 years old

         

Head

0

1 (1.3)

0.010*

0

0.997

Abdomen

1 (1.3)

0

0.728

   

Face

0

0

0

   

Chest

0

1

0.010*

0

0.997

Extremity

1 (1.3)

0

0.811

   

External

1 (1.3)

-

-

   

18–54 years old

         

Head

0

11 (4.9)

≤ 0.001*

0

0.995

Abdomen

10 (4.5)

1 (0.4)

0.446

   

Face

10 (4.5)

1 (0.4)

0.022*

   

Chest

5 (2.2)

6 (2.7)

0.012*

0.70

(0.18–2.62)

0.597

Extremity

10 (4.5)

1 (0.4)

0.974

   

External

11 (4.9)

-

-

   

≥ 55 years old

         

Head

4 (2.7)

16 (10.9)

0.002*

0.186 (0.05–0.61)

0.005*

Abdomen

20 (13.6)

0

0.367

   

Face

20 (13.6)

0

0.487

   

Chest

7 (4.8)

13 (8.8)

≤ 0.001*

0.198 (0.07–0.56)

0.002*

Extremity

14 (9.5)

6 (4.1)

0.075*

   

External

20 (13.6)

-

-

   

Table 3 shows mortality distribution for age groups following aggravating factors. Roll on and rollover was related to higher mortality in patients over 55 years, while throw was an independent predictor of mortality between 18 and 54 years. Not wearing a helmet was not an independent predictor of mortality in all age groups.

 
 
Table 3

Aggravating factors according to outcome

 

Univariate analysis

Multivariate analysis

Aggravating factors

Survived n (%)

Dead n (%)

p value

OR (95%IC)

p value

< 18 years old

         

No helmet

6 (8.0)

0

0.76

   

Roll on

2 (2.7)

0

0.86

   

Roll over

15 (20)

1 (1.3)

0.05

   

Thrown

33 (44)

1 (1.3)

0.26

   

18–54

         

No helmet

24 (10.7)

1 (0.4)

0.82

   

Roll on

6 (2.7)

1 (0.4)

0.22

   

Roll over

25 (11.2)

4 (1.8)

0.01*

0.30 (0.08–1.15)

0.079

Thrown

104 (46.8)

10 (4.5)

0.006*

0.10 (0.01–0.87)

0.037*

≥ 55 years old

         

No helmet

14 (9.5)

0

0.11

   

Roll on

3 (2)

3 (2)

0.008*

0.11 (0.02–0.76)

0.012*

Roll over

9 (6.1)

4 (2.7)

0.05*

0.25 (0.06–0.94)

0.041*

Thrown

6 (42.2)

7 (4.8)

0.25

   
 

Table 3: OR: odd ratio; CI: confidence interval; *statistical significance

Thirty-two patients (7.2%) died: six patients (1.34%) with an ISS < 24 and 26 patients (26.5%) with an ISS ≥ 25 (log-rank test ≤ 0.001). The survival rates were 98.3% in patients with an ISS < 24 and 73.5% in patients with an ISS ≥ 25, respectively. Figure 1 shows the global mortality distribution among the age groups. Patients over 55 years showed a worse prognosis between the age groups, with a survival rate of 86.4%. The survival rate of patients < 18 years was 98.7%, while between 18 and 54 years, the survival rate was 95.1% (log-rank test ≤ 0.028). Mortality distribution in patients with an ISS ≥ 25 showed higher mortality in patients over 55 years, although not statistically significant (log-rank test ≤ 0.701).

Bicycle-related trauma was more frequent in summer (34%), followed by spring (28.9%), autumn (22.6%), and winter (14.3%). Seasonal trauma distribution among the age groups is described in Fig. 2, while monthly distribution is in Fig. 3, with July and August more represented.

Discussion

Bicycle-related trauma represents an essential percentage of road traffic victims worldwide2,10. The increasing diffusion of this type of transport responds to anti-pollution policies, representing a valid alternative method of mobility in heavy-traffic urban areas11. Bicycle mobility was also improved by the diffusion of different bike sharing companies worldwide, that have made this popular type of transport easily available in a few minutes11. Given the high diffusion of bicycle use both for sport and transport purposes, analysing the kind of injuries and the mortality distribution of these types of trauma is of paramount importance.

Our study confirmed head and chest injuries as independent predictors of mortality only in patients over 55 years old, as showed in Table 2. Aggravating factors influenced the mortality trends, as roll-on and roll-over were independent predictors of mortality in patients ≥ 55 years old. The throw was an independent predictor of mortality between 18 and 54 years old (Table 3). Only 44 patients (9.8 %) were not wearing helmets in our study. Only one patient died among them, in the group 18–54 years old. In our research, not wearing a helmet was not an independent predictor of mortality, but head and chest AIS ≥ 3 were independent predictors of death. Therefore, despite the protective effect of the helmet, other variables should influence the mortality trends in cycle trauma. These results align with the study of Foley J. et al.3, who showed that different variables (i.e., gender and mechanism of trauma) were independent predictors of mortality in bicycle trauma, and other factors than wearing a helmet could have a role in head injuries. A systematic review conducted by Hoye A.12 showed that mandatory bicycle helmet legislation for all cyclists reduces about 20% of head injuries, significantly affecting severe head injuries. Two meta-analyses13,14 confirmed the positive impact of the helmet only on severe head injuries, also showing a protective result on fatal injury prevention. However, all the studies agreed13,14 on the role of different variables on bicycle mortality.

The survival rate estimated with the Kaplan-Mayer method showed a higher mortality in patients with an ISS ≥ 25, confirming that overall trauma severity influenced mortality.

However, as shown in Fig. 1, patients older than 55 years showed the lowest survival rate, confirming age as an independent predictor of mortality. Interestingly, another study on motorcycle-related trauma9 showed similar results, with a higher mortality trend in older patients (≥ 55 years old).

Finally, Fig. 2 shows the seasonal distribution of bicycle trauma, more frequent in summer (34%), followed by spring (28.9%). These results align with current literature15.

Given its retrospective nature, this study presents some limitations. Although older patients showed a higher mortality rate, it’s possible that patients’ comorbidities and the use of anticoagulant therapy could influence the prognosis. Unfortunately, these variables were not available in our trauma registry and were not considered.

Moreover, although this study was conducted in a level-one trauma center in Italy, no information on prehospital trauma mortality was available. Indeed, our data referred only to our in-hospital experience.

Finally, despite different studies showing a possible correlation between the bicycle infrastructure and mortality16,17, this variable is not reported in our trauma registry, and it has not been considered.

In conclusion, this study showed that different variables influenced bicycle trauma mortality. Older age and aggravating factors are independent predictors of mortality. Despite the protective effect of the helmet, head and chest injuries were confirmed to be independent predictors of mortality in patients ≥ 55 years old. Bicycle-related trauma is more frequent during the warm seasons, especially in July and August. Further multicentric and prospective studies should be advisable to confirm our results, fostering a stronger scientific collaboration with the prehospital care services.

Abbreviations

GCS

Glasgow Coma Scale

AIS

Abbreviated injury scale

ED

Emergency department

ISS

Injury Severity Score

PD

Probability of death

TRISS

Trauma and Injury Severity Score

ASA

American Society of Anaesthesiologists

OR

odds ratio (OR)

CI

confidence interval

HR

heart rate

SBP

systolic blood pressure

DBP

diastolic blood pressure

BE

Basic Excess

INR

International Normalized Ratio

EPP

Extraperitoneal pelvic packing

DCL

Damage control laparotomy

DCT

Damage control Thoracotomy

TI

Tracheal intubation

RTS

Revised Trauma Score

Declarations

Consent to publication: Not applicable

Availability of data and materials: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request

Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author’s contribution: ER, SPBC, FV, MA and AS contributed to the manuscript to concept the study design, data analysis, data interpretation, and article drafting. SC and OC contributed to data interpretation and manuscript critical revisions. All authors read and approved the final version of the manuscript.

Acknowledgements: Not applicable 

Competing interest: The authors declare that they have no competing interests

Ethics approval and consent to participate: Not applicable

Informed consent: All patients signed an informed consent. 

References

  1. de Guerre LEVM, Sadiqi S, Leenen LPH, Oner CF, van Gaalen SM. Injuries related to bicycle accidents: an epidemiological study in The Netherlands. Eur J Trauma Emerg Surg. 2020;46:413–8.
  2. Difino M, Bini R, Reitano E, Faccincani R, Sammartano F, Briani L, et al. Epidemiology of trauma admissions in a level 1 trauma center in Northern Italy: a nine-year study. Updates Surg. 2021;73:1963–73.
  3. Foley J, Cronin M, Brent L, Lawrence T, Simms C, Gildea K, et al. Cycling related major trauma in Ireland. Injury. 2020;51:1158–63
  4. Fraser SDS, Lock K. Cycling for transport and public health: a systematic review of the effect of the environment on cycling. Eur J Public Heal. 2011;21:738–43.
  5. Robinson DL. Bicycle helmet legislation: Can we reach a consensus? Accid Anal Prev. 2007;39:86–93.
  6. Webman R, Dultz LA, Simon RJ, Todd SR, Slaughter D, Jacko S, et al. Helmet use is associated with safer bicycling behaviors and reduced hospital resource use following injury. J Trauma Acute Care Surg. 2013;75:877–81.
  7. Teschke K, Koehoorn M, Shen H, Dennis J. Bicycling injury hospitalisation rates in Canadian jurisdictions: analyses examining associations with helmet legislation and mode share. BMJ Open 2015;5:e008052.
  8. Walker I. Drivers overtaking bicyclists: Objective data on the effects of riding position, helmet use, vehicle type and apparent gender. Accid Anal Prev. 2007;39:417–25.
  9. Granieri SS, Reitano EE, Bindi FF, Renzi FF, Sammartano FF, Cimbanassi SS, et al. Motorcycle-related trauma:effects of age and site of injuries on mortality. A single-center, retrospective study. World J Emerg Surg. 2020;15:18
  10. Scott LR, Bazargan-Hejazi S, Shirazi A, Pan D, Lee S, Teruya SA, et al. Helmet use and bicycle-related trauma injury outcomes. Brain Inj. 2019;33:1597–601.
  11. Thompson DC, Thompson RS, Rivara FP. Incidence of bicycle-related injuries in a defined population. Am J Public Health. 1990;80:1388.
  12. Hoye A. Recommend or mandate? A systematic review and meta-analysis of the effects of mandatory bicycle helmet legislation. Accid Anal Prev. 2018;120:239–49.
  13. Olivier J, Creighton P. Bicycle injuries and helmet use: A systematic review and meta-analysis. Int J Epidemiol. 2017;46:278–92.
  14. Høye A. Bicycle helmets – To wear or not to wear? A meta-analyses of the effects of bicycle helmets on injuries. Accid Anal Prev. 2018;117:85–97.
  15. Nordengen S, Andersen LB, Riiser A, Solbraa AK. National Trends in Cycling in Light of the Norwegian Bike Traffic Index. Int J Environ Res Public Health. 2021;18(12).
  16. Pucher J, Buehler R. Safer cycling through improved infrastructure. Am J Public Health. 2016;106:2089–91.
  17. Goerke D, Zolfaghari E, Marek AP, Endorf FW, Nygaard RM. Incidence and Profile of Severe Cycling Injuries After Bikeway Infrastructure Changes. J Community Health. 2020;45:542.