SUBMITTED 9 FEB 22 1 REVISIONS REQ. 9 MAY & 18 JUL 22; REVISIONS RECD. 23 JUN & 27 SEP 22 2 ACCEPTED 16 OCT 22 3 ONLINE-FIRST: DECEMBER 2022 4 DOI: https://doi.org/10.18295/squmj.12.2022.065 5 6 The Impact of the COVID-19 Pandemic on the Pattern of Trauma Presenting 7 to a Tertiary Care Trauma Center in Oman 8 Rahma Al Harthi,1 Maram Al Hinai,1 Maather Al Abri,2 Ashjan AlMaamari,3 9 Edwin Stephen,4 *Hani Al Qadhi4 10 11 1Department of General Surgery, Oman Medical Specialty Board, Muscat, Oman; 2Department 12 of General Surgery, 3Sultan Qaboos University, Muscat, Oman; 4Department of General 13 Surgery, Sultan Qaboos University Hospital, Muscat, Oman. 14 *Corresponding Author’s e-mail: halqadhi@squ.edu.om 15 16 Abstract 17 Objective: We noticed a change in the pattern of presentation of trauma cases at SQUH, before 18 the pandemic and during the two waves. Our study aimed at studying this observation. Methods: 19 This retrospective study was from January 2019 to October 2021. Data of all trauma patients was 20 collected from the hospital information system after ethics committee approval. The pattern of 21 trauma was divided into pediatric, adult, and geriatric age groups. Location of trauma was 22 described as outdoor, home and roads along with the details of mechanism of trauma was 23 collected. Patients with incomplete data were excluded. Results: Based on the inclusion criteria 24 589 records were analyzed. The mean age of presentation was 29 years. Majority were male 25 (71%). Adults were (54.2%), pediatrics (34%) and geriatric (11.9%). There was a gradual 26 increase in percentage of pediatric trauma during pre-COVID, COVID wave 1 and COVID wave 27 2 (29%, 32%, 51%), respectively. A significant decline in the number of geriatric trauma by 28 almost 50% between pre-covid and covid phase II. Increase in trauma at home during COVID 29 phase II (65.9%) as was an increase in penetrating trauma during COVID phase II (16.5%). ICU 30 admissions increased during the first wave of the pandemic (10.5%). Conclusion: We noted a 31 true change in the pattern of trauma cases before and during the COVID -19 pandemic. 32 Observations made could lead to better safety guidelines for the pediatric age groups and take 33 steps to reduce penetrating trauma. 34 Keywords: Trauma, coronavirus, COVID-19, Oman, Muscat, Epidemiology. 35 36 Advances in Knowledge 37 ● The rate of pediatric trauma increased during the COVID19 pandemic. 38 ● The rate of trauma has overall decreased during the pandemic. 39 ● The rate of Motor Vehicle Crashes (MVC) was not affected by the restrictions 40 implemented during the pandemic. 41 42 Application to Patient Care 43 ● Public awareness should be raised to prevent trauma during pandemics with special focus 44 on the most vulnerable demographic groups. 45 ● With the overwhelming load of pandemic on the health systems and healthcare workers, 46 understanding trauma patterns during a global pandemic will aid in preparing and 47 planning strategies to deal with such an issue. 48 49 Introduction 50 The COVID-19 pandemic has had a significant impact on the social, economic, educational and 51 health systems all over the world. Many adjustments and restrictions were implemented in an 52 attempt to curb the spread of the virus and minimize the burden of the disease. 53 54 The Sultanate of Oman is located in the southeastern coast of the Arabian Peninsula and has a 55 population size of 4.5 million people.1 The first two cases of COVID-19 infection in Oman were 56 registered in late February 2020.2 The number of cases related to travel gradually increased and 57 community transmission was noted by the end of March the same year. A “Supreme Committee” 58 [SC] was formed to manage the pandemic and released several directives to control the outbreak 59 at regular intervals, based on the case load, morbidity, and mortality within the nation.3 60 61 During the pandemic tertiary hospitals in the Sultanate announced a temporary suspension of all 62 routine non-emergency services such as elective surgeries and procedures, outpatient 63 appointments, etc.4,5 and ours was one such center - The Sultan Qaboos University Hospital 64 [SQUH], located in the capital city of Muscat, with a capacity of 600 beds.6 It is considered one 65 of two major trauma centers in Muscat that accepts cases from all across the Sultanate. 66 67 Several studies conducted around the world showed that the pandemic led to significant 68 reductions in trauma case load and changed the pattern injurie.7,8 However, no national level 69 studies were conducted to evaluate this issue. We noticed a change in the pattern of presentation 70 of trauma cases at SQUH, before the pandemic and during the two waves. The objective of this 71 study is to study the overall trend (increase or decrease) in trauma cases presenting to a major 72 trauma center in the Sultanate. Moreover, it looks at identifying the groups that are at higher risk 73 of trauma and the most common mechanisms of injuries. This will help in raising public 74 awareness to prevent trauma injuries in such overwhelming situations, as well as providing a 75 baseline data for stakeholders to prepare healthcare services to deal with such problems. 76 77 Methods 78 This retrospective, cross-sectional study included all trauma cases that presented to SQUH from 79 1st of January 2019 until 30th October 2021 and was conducted after ethics committee approval. 80 81 To make the comparison between different time periods, we divided the dates of data collection 82 into three phases: pre-COVID (01 Jan 2019 - 29 Feb 2020), COVID phase I (01 March 2020 - 28 83 Feb 2021) and COVID phase II (01 Mar 2021 – 31 Oct 2021). Age groups were divided as 84 pediatrics (less than or equal to 13 years of age), Adults (14 to 64 years of age) and Geriatrics 85 (more than or equal to 65 years of age). 86 87 Electronic medical records of patients’ were reviewed. Data collected included demographics 88 (age and gender), date of presentation to the emergency department (ED), location of trauma, 89 type of trauma (penetrating or blunt), mechanism of trauma, list of injuries, outcomes including 90 disposition from the trauma bay (admission (ICU/ward), discharge home, transfer to another 91 hospital or death and length of hospital stay in days. 92 93 Statistical analysis was conducted using Statistical Package for Social Sciences (SPSS 21.0). 94 Continuous variables were presented as mean, median, and standard deviation. Categorical 95 variables were presented as frequency and percentage. Association/difference between two 96 categorical variables were assessed by using a Chi-square test (Fisher's exact/Likelihood ratio). 97 Appropriate graphs were used to show trends over time. A P-value less than 0.05 was considered 98 statistically significant. All the analyses were carried out in IBM SPSS Statistics version 28.0. 99 100 Results 101 We reviewed 594 files and excluded 5 of them as the data was incomplete, leaving a total of 589 102 cases. 103 104 Demographic data of our study showed that (Table 1) - majority of the patients were males 105 (421;71.5%); and median age was 29 years (age range 7 weeks to 96 years). More than half of 106 the cases were adults (54%), a third of the cases were children (34%) and 12% were in the 107 elderly age group. 108 109 Phase wise analysis revealed a falling trend in the total number of trauma cases presenting to ED 110 (288; 49%, 210; 36%, 91; 15%) (Figure 1) across the adult and geriatric age groups; however, a 111 rising trend was noted for the pediatric age group (29.5%, 32% and 51.6%) which was 112 statistically significant (p <0.05). The gender distribution between the three phases of time was 113 not statistically significant (p 0.061). 114 115 Majority of trauma occurred at home (344; 58%), followed by outdoors (184; 31%) and roads 116 (46; 8%). These differences were not statistically significant (p 0.43). Only 10 cases occurred at 117 work/school, and these were equally divided between the pre-COVID and COVID phase I. 118 Blunt trauma was significantly higher than penetrating trauma (81% vs 11%) with decreasing 119 frequency during the three phases (87%, 79% and 71%). However, the percentage of penetrating 120 trauma increased (7%, 15% and 17%). This change in trend was statistically significant (p 121 <0.05). 122 123 As for the mechanism of trauma (Table 2), falls accounted for 57% of the total number of cases, 124 followed by Motor Vehicle Collision/Accidents [MVC] (11%), penetrating trauma and others 125 (8.8% each). Lastly, sports related injuries, crush injuries, assault, drowning, and suicide 126 attempts accounted for the remaining cases (4.4%, 3.7%, 3.6%, 1.5% and 0.7% respectively). We 127 noted a decreasing trend in falls and sports related injuries through the three phases. 128 Interestingly, penetrating injuries and drowning saw an increase. The rate of MVC remained 129 stable throughout the three-time phases. No particular trend was noted for the remaining 130 mechanism of trauma. 131 132 Injuries to the extremities were the most common (38%, Lower limbs 23% and Upper Limbs 133 15%), (table 3). These included bone fracture/dislocation, soft tissue injuries or neurovascular 134 injuries. Secondly, Head and Neck injuries accounted for 26% of cases and included traumatic 135 brain / ophthalmic / soft tissue injuries to the scalp, face and neck, and maxillofacial fractures. 136 This was followed by polytrauma (10%), Spinal injuries (5%), thoracic injuries (3%), 137 genitourinary injuries (2%), abdominal injuries (1%) and pelvic injuries (0.3%). In 14% of the 138 cases, which included those with falls with late presentation to the ED and foreign body 139 ingestions, we found no injury acquired. 140 141 Ninety five percent of our patients required admission (88% were admitted to the ward, 7% 142 required admission to ICU). Four percent of the cases were discharged home while 0.3% were 143 transferred to a different hospital. No cases were declared dead in the trauma bay. The admission 144 duration varied between one day to 155 days (median duration - 2 days). There was no 145 significant difference in trauma outcome during the three-time phases. 146 147 Discussion 148 The COVID19 pandemic shook the social and medical realms amongst others. This retrospective 149 study looks at the impact of this pandemic on trauma patterns presenting to a major Trauma 150 Center (SQUH) in Oman’s Capital city of Muscat. This was done by comparing all trauma 151 presentations to ED one-year pre-COVID, the first and second phase of COVID. 152 153 589 cases made it to the final analysis. We noted an overall decrease in the total number of 154 trauma cases presenting to ED. When comparing the three-time phases, the total number of cases 155 dropped by a third between the pre-COVID and COVID phase I (288; 49% versus 210; 36%) 156 respectively. This trend continued in the second phase, where we observed a further drop of two-157 thirds (91; 15%). The latter could be an overestimation as data was collected until October 2021, 158 which makes it a shorter period of time compared to the two other time phases. Several 159 international studies have found a similar trend like a multicentre study by Berg et al which 160 found a 32% decrease in the number of trauma cases during the pandemic.7 Other studies showed 161 a decrease ranging from 22% to 57%.8,9 These observations can be attributed to the precautions 162 that were implemented during the pandemic. This included closure of all international border for 163 non-residents, restricting inter-governorate travel to absolute essential, a 70% reduction in the 164 number of employees at workplaces [work from home policy], suspension of classes in schools 165 and universities, banning all public gatherings, closing retail outlets, and recommending social 166 distancing. 167 168 The first lockdown in the Muscat governorate was implemented on the 10th of April 2020, which 169 was subsequently extended during the holy month of Ramadhan and finally lifted on May 170 29th. 10,11 Alongside this was a staged lockdown between governorates in June. Night curfew [NC] 171 was implemented from March 28th till April 8th, 2021, between the hours of 8pm and 5am. The 172 implementation of a lockdown/curfew led to a significant reduction in public and traffic 173 movement. In our study, we found no significant drop in trauma cases during the first lockdown, 174 in fact, the number of cases has remained stable. On the contrary, a drop was noted in the 175 following year during the partial lockdown /NC. 176 177 As for demographics - males predominate with a male to female ratio of 3:1. This male 178 predominance is consistent with trauma epidemiology overall12 and no significant change in the 179 sex distribution was observed during the COVID pandemic. This was true for our study and in 180 other international studies.7,13,14 When it comes to age, adults (13 to 64 years) formed the 181 majority of our study population (54%), and this is expected as this age group was the most 182 active. Although the number of trauma cases for both adult and geriatric age groups has 183 significantly dropped during the pandemic, the number of pediatric trauma has risen significantly 184 and exceeded the percentage of adult trauma in COVID phase II (52% vs 41%). This finding was 185 contrary to what was found in other studies which showed a drop in overall pediatrics trauma 186 reaching up to one half.15,16 We also expected the rate to decrease considering the shutdown of 187 schools, parks, public playgrounds etc, but home related injuries can explain this rise. This might 188 be due to the increase of stressors to families resulting from a number of new changes such as; 189 working from home, supervising online teaching, lack of professional childcare services as well 190 as restrictions in seeking extended family support. This in turn led to reduced direct and expert 191 supervision and care of children. 192 193 More than a half of trauma occurred at home followed by a third occurring outdoors. There was 194 no statistically significant difference in the distribution of trauma location between pre-COVID 195 and the two phases of COVID. The proportion of blunt versus penetrating injury, however, 196 changed during the pandemic. A significant decrease in blunt trauma was noted, versus a rise in 197 penetrating trauma. This change was also demonstrated in studies conducted in the United States 198 of America and the United Kingdom.8,17,18 The increase in penetrating injuries has reached up to 199 21% in a multicenter retrospective study from South California, USA.19 This was attributed to 200 the socio-economic stress that resulted from the pandemic such as cutting down working staff 201 and a rise in unemployment rate.7,8,19 Others have attributed this to the rise in homicide, sales of 202 firearms, self-harm, and domestic violence.19,20 However, we cannot relate these findings to our 203 studied population, as only three cases of penetrating injuries were attributed to assault and all of 204 them were stabbing injuries. Another explanation is that more Do It Yourself (DIYs) were 205 conducted at homes, which have resulted in those injuries. This rising trend should alert health 206 care professionals to be prepared to deal with penetrating injuries as they are usually more 207 severe. Moreover, such injuries are usually associated with more blood loss and often require 208 blood transfusion. There was a shortage in blood supply during the pandemic due to reduction in 209 donation21, the health system should therefore be prepared to overcome such problems in the 210 future. 211 212 Different mechanisms of injury were noted in our group, but the majority was due to falls (57%), 213 followed by MVC (11%) and penetrating traumas (8.8%). During the different time periods, we 214 noted a decreasing trend in falls and sports related injuries. Other studies have also found a 215 similar trend, which was statistically significant19 and was attributed to social gatherings being 216 banned, and sports centres being shut down during COVID Phase I and II. The trend of MVC 217 remained unaffected during the pandemic while we expected it to drop given the implemented 218 restrictions that reduced road travel in general. A no change trend in MVC was also noted in 219 other studies8, however, larger studies showed a significant reduction of MVC.7 A possible 220 explanation for our finding is that less severe MVC related injuries were dealt with in regional or 221 non-trauma centers prior to the COVID pandemic. And although the overall trend reduced, the 222 severity of injuries increased during the pandemic and such cases could only be managed in a 223 major trauma center like SQUH. Drowning incidents increased in COVID Phase II. This is 224 probably because private properties with swimming pools were more utilized during the 225 pandemic in an attempt to entertain families and lift the pressure associated with the lockdown. 226 227 The distribution of injuries was classified by anatomical location. Cumulative upper and lower 228 limbs injuries accounted for 38% of all acquired injuries. Head and neck injuries were the second 229 most common and account for a quarter of our cases. These injuries are usually associated with 230 blunt injury which was more common in our cohort. 231 232 We have studied the short-term outcome of our cases and found that 95% required admission and 233 only 4% were safe to be sent home. There was no noted effect of the pandemic on the disposition 234 of the patients from the trauma bay. We had a median length of hospital stay of two days which 235 was also not affected by the pandemic. Our median seems to be shorter than that shown in other 236 studies which varied between 4 days17,19 and 5.5 days.18 This might be due to less severe injuries 237 seen in our study population. 238 239 This study aims to look at the trend of trauma patterns before and after the COVID pandemic and 240 does not aim to compare particular blocks of time or absolute date. Being a retrospective study, 241 there are a number of limitations associated with this type of studies such as recall bias, missed 242 data and mis-documentation. Moreover, this is a single center study in the capital of the 243 Sultanate which might not reflect the pattern of trauma in other areas of Oman before or during 244 the pandemic. Also, a number of new implementations have been imposed by the SC to control 245 the pandemic, and the exact direct cause in changing the trauma pattern cannot be specifically 246 identified. 247 248 Conclusion 249 The COVID-19 pandemic has influenced the frequency and pattern of trauma in Oman. There is 250 an overall decrease in the total number of trauma cases presenting to SQUH, however, the 251 proportion of pediatrics and penetrating injuries showed an increase. Despite the limitations of 252 our study, the findings can be taken into consideration when formulating safety guidelines for the 253 pediatric / geriatric age groups with special attention to penetrating trauma, lest there be another 254 phase or pandemic. 255 256 Conflicts of Interest 257 The authors declare no conflict of interests. 258 259 Funding 260 No funding was received for this study. 261 262 Author Contributions 263 Rahma Al Harthti was involved in data collection, data analysis, literature review and manuscript 264 writing. Maram Al Hinaei was involved in data collection and in drafting out the ethical approval 265 application. Maather Al Abri was involved in data collection. Ashjan AlMaamari was involved 266 in data collection. Edwin Stephen was involved in study design and supervising the research. Dr 267 Hani Al Qadhi was involved in supervising the research. All authors approved the final version 268 of the manuscript. 269 270 References 271 1. Population - DATA PORTAL [Internet]. [cited 2021 Dec 3]. Available from: 272 https://data.gov.om/OMPOP2016/population 273 2. MOH Registers First Two Novel Coronavirus (COVID-2019) in Oman - Media Center Display 274 Page - Ministry of Health [Internet]. [cited 2021 Nov 22]. Available from: 275 https://www.moh.gov.om/en/-/--1226 276 3. Statements and Updates - Ministry of Health [Internet]. [cited 2021 Nov 22]. Available from: 277 https://www.moh.gov.om/en/-59 278 4. Observer O. Royal Hospital to stop admitting non-emergency cases [Internet]. Oman Observer. 279 2020 [cited 2021 Nov 22]. Available from: 280 https://www.omanobserver.om/article/14822/Main/royal-hospital-to-stop-admitting-non-281 emergency-cases 282 5. Observer O. Al Nahda Hospital to suspend outpatient appointments [Internet]. Oman Observer. 283 2020 [cited 2021 Nov 22]. Available from: 284 https://www.omanobserver.om/article/14247/CORONAVIRUS/al-nahda-hospital-to-suspend-285 outpatient-appointments 286 6. About Hospital [Internet]. [cited 2021 Nov 22]. Available from: 287 https://www.squ.edu.om/squh/About/About-Hospital 288 7. Berg GM, Wyse RJ, Morse JL, Chipko J, Garland JM, Slivinski A, et al. Decreased adult 289 trauma admission volumes and changing injury patterns during the COVID-19 pandemic at 85 290 trauma centers in a multistate healthcare system. Trauma Surg Acute Care Open. 2021 Feb 291 11;6(1):e000642. 292 8. Chodos M, Sarani B, Sparks A, Bruns B, Gupta S, Michetti CP, et al. Impact of COVID-19 293 pandemic on injury prevalence and pattern in the Washington, DC Metropolitan Region: a 294 multicenter study by the American College of Surgeons Committee on Trauma, Washington, 295 DC. Trauma Surg Acute Care Open. 2021 Jan 19;6(1):e000659. 296 9. Kamine TH, Rembisz A, Barron RJ, Baldwin C, Kromer M. Decrease in Trauma Admissions 297 with COVID-19 Pandemic. West J Emerg Med. 2020 Jul;21(4):819–22. 298 10. Muscat lockdown from Friday, schools suspended until further notice [Internet]. [cited 2021 299 Nov 22]. Available from: 300 https://web.archive.org/web/20200726112845/https://www.omanobserver.om/muscat-301 supreme-commitee/ 302 11. Observer O. Muscat lockdown extended, ban on all Ramadhan gatherings [Internet]. Oman 303 Observer. 2020 [cited 2021 Nov 22]. Available from: 304 https://www.omanobserver.om/article/13832/CORONAVIRUS/muscat-lockdown-extended-305 ban-on-all-ramadhan-gatherings 306 12. Sauaia A, Moore FA, Moore EE, Moser KS, Brennan R, Read RA, et al. Epidemiology of 307 trauma deaths: a reassessment. J Trauma. 1995 Feb;38(2):185–93. 308 13. Okoye OG, Olaomi OO, Gwaram UA, Apollo KD. The impact of COVID-19 lockdown on 309 acute trauma patients seen at the National Hospital Trauma Centre Abuja, Nigeria. Pan Afr 310 Med J. 2021 Apr 30;38:414. 311 14. Zsilavecz A, Wain H, Bruce JL, Smith MTD, Bekker W, Laing GL, et al. Trauma patterns 312 during the COVID-19 lockdown in South Africa expose vulnerability of women. S Afr Med J. 313 2020 Oct 28;110(11):1110–2. 314 15. Sanford EL, Zagory J, Blackwell J-M, Szmuk P, Ryan M, Ambardekar A. Changes in 315 pediatric trauma during COVID-19 stay-at-home epoch at a tertiary pediatric hospital. J 316 Pediatr Surg. 2021 May;56(5):918–22. 317 16. Shi Y, Kvasnovsky C, Khan S, Jain S, Sargeant D, Lamoshi A, et al. Impact of the COVID-19 318 pandemic on trauma activations at a pediatric level 1 trauma center in New York. Pediatr Surg 319 Int. 2021 Jul 15;1–6. 320 17. Sheets NW, Fawibe OS, Mahmoud A, Chawla-Kondal B, Ayutyanont N, Plurad DS. Impact 321 of the COVID-19 Pandemic on Trauma Encounters. Am Surg. 2021 Jul 322 4;00031348211029858. 323 18. Ghafil C, Matsushima K, Ding L, Henry R, Inaba K. Trends in Trauma Admissions During 324 the COVID-19 Pandemic in Los Angeles County, California. JAMA Netw Open. 2021 Feb 325 22;4(2):e211320. 326 19. Yeates EO, Grigorian A, Barrios C, Schellenberg M, Owattanapanich N, Barmparas G, et al. 327 Changes in traumatic mechanisms of injury in Southern California related to COVID-19: 328 Penetrating trauma as a second pandemic. J Trauma Acute Care Surg. 2021 Apr;90(4):714–329 21. 330 20. Olding J, Zisman S, Olding C, Fan K. Penetrating trauma during a global pandemic: Changing 331 patterns in interpersonal violence, self-harm and domestic violence in the Covid-19 outbreak. 332 The Surgeon. 2021 Feb;19(1):e9–13. 333 21. Stephen E, Al-Adawi SSH, Abdelhady I, Mawali HA, Al-Wahaibi K. Managing Vascular 334 Surgery Emergencies and Referrals During the COVID-19 Pandemic at a Tertiary Centre in 335 Oman. Sultan Qaboos Univ Med J SQUMJ. 2021 Mar 15;21(1):e116-119. 336 337 338 Figure 1: The number of trauma cases presenting to ED from January 2019 till September 2021 339 340 Table 1: The demographic data of our studied population. 341 Pre-COVID (288) COVID Phase I (210) COVID Phase II (91) Total number (589) Sex Male Female 193 (67%) 95 (33%) 160 (76%) 50 (24%) 68 (75%) 23 (25%) 421 (71.5%) 168 (28.5%) Age groups Pediatric Adult Geriatric 85 (29.5%) 161 (55.9%) 42 (14.6%) 68 (32%) 121 (58%) 21 (10%) 47 (51.6%) 37 (40.7%) 7 (7.7%) 200 (34%) 319 (54%) 70 (12%) 342 Table 2: Mechanism of injury 343 Pre-COVID (n 288) COVID Phase I (n 210) COVID Phase II (n 91) Total (n 589) Fall 175 (61%) 115 (55%) 48 (53%) 338 (57%) MVC 32 (11%) 23 (11%) 10 (11%) 65 (11%) Penetrating 14 (5%) 27 (13%) 11 (12%) 52 (8.8%) Sports 19 (6.6%) 5 (2%) 2 (2%) 26 (4%) Crush 11 (4%) 7 (3%) 4 (4%) 22 (3.7%) Assault 10 (3.5%) 9 (4%) 2 (2%) 21 (3.6%) Drowning 4 (1%) 1 (0.5%) 4 (4.4%) 9 (1.5%) Suicide attempt 2 (0.7%) 1 (0.5%) 1 (1%) 4 (0.7%) Others 21 (7%) 22 (10.5%) 9 (10%) 52 (9%) 344 Table 3: Anatomical location of the injuries 345 Pre-COVID COVID Phase I COVID Phase II Total Head and Neck 62 (21.5%) 58 (28%) 34 (37%) 154 (26%) Lower Limbs 80 (28%) 44 (21%) 13 (14%) 137 (23%) Upper Limbs 49 (17%) 32 (15%) 9 (10%) 90 (15%) Polytrauma 26 (9%) 26 (12%) 6 (7%) 58 (10%) Spine 13 (4%) 10 (5%) 4 (4%) 27 (5%) Thorax 9 (3%) 5 (2%) 2 (2%) 16 (3%) Genitourinary 4 (1%) 4 (2%) 2 (2%) 10 (1.7%) Abdomen 3 (1%) 3 (1%) 0 (0%) 6 (1%) Pelvis 2 (0.7%) 0 (0%) 0 (0%) 2 (0.3%) Others 4 (1%) 0 (0%) 0 (0%) 4 (0.7%) Note: There are excluded data for patients who had no or minimal injury (36, 28 and 21 patients 346 from the pre-COVID, COVID phase I and COVID phase II respectively. 347