Impact of COVID-19 Lockdown on Traumatic Bone Fracture Patterns and Hospital Outcomes in 1646 Elderly Patients: A Retrospective Study in Turkey

Background

Developments in medical care have served to extend human life, which has led to an increase in the elderly population [1]. As age increases, the negative outcomes of injuries become more common, independent of the severity of the injury [1]. Comorbidities and multi-drug use, which are more common in the elderly, also increase the risk of trauma. Additionally, it is known that the rate of trauma cases in the elderly increases every year [2]. It is also worthy of note that the severity of elderly trauma cases is not well known and is triaged as secondary in the emergency department (ED) [3]. After a traumatic event, in the elderly, consequences such as fractures, contusions, bruises, wounds, internal organ lesions, head injuries, and death may occur more frequently [4]. The most dangerous mechanism of injury is craniofacial trauma, which is associated with high mortality, but the most common diagnosis is hip fractures [2]. Of course, trauma mechanisms observed in other age groups are also relevant in the elderly population. However, some mechanisms are known to be more common in the elderly population. The most significant of these mechanisms is falling. Fall-related injuries are more common in elderly patients compared with other populations [5]. Turkey, where the study was conducted, is a country where a wide variety of trauma mechanisms can be seen. Earthquakes occur frequently in Turkey, which is located on active fault lines. Elderly people, who have less effective avoidance reactions in natural disasters, are liable to get injured more frequently [6]. Some differences in trauma courses are predicted in the elderly population due to changes in mechanisms and factors, fewer avoidance actions, and comorbidities [7]. The hospital where the present study was conducted is a very high-capacity hospital in its region where patients are transferred both via their own transportation and by land and air ambulances [8].

Turkey was heavily affected by the Coronavirus Disease-2019 (COVID-19) pandemic [9]. One of the important measures taken to protect the elderly population, one of the risk groups in the course of the pandemic, was to restrict citizens aged 65 and over and people with chronic diseases from leaving their residences and walking around in open areas, in accordance with the circular published by the Ministry of Internal Affairs of the Republic of Turkey. In the later stages of the pandemic, lockdowns were entered into application for all age groups in Turkey [10]. As the number of cases decreased, lockdown periods were relaxed. Gradually, times of normalization began to occur [11]. During lockdown periods, there were changes in trauma mechanisms, as well as some other factors, for the elderly [12, 13]. These lockdowns targeting the elderly population, together with the infectious disease aspect of the pandemic, may have brought about a change in the incidence and mechanisms of trauma, and accordingly, in the experiences of the patients in Turkey. To investigate this possibility, this retrospective study aimed to compare outcomes from emergency admissions to a major trauma center in Turkey of 1646 elderly patients with traumatic bone fractures before, during, and after the COVID-19 pandemic lockdown period.

Material and Methods

ETHICAL APPROVAL AND STUDY DESIGN:

This retrospective study was approved by the local ethics committee with decision number 2021/E1-212192. After this approval, the study was carried out in a single center: the ED of high-capacity Ankara Bilkent City Hospital, which accepts patient transfers both personally and by ground and air ambulance, and offers numerous examination and consultation opportunities to all surgical departments. Within the scope of the study, patients aged 65 and over who arrived in the ED due to trauma between September 15, 2019 and September 15, 2020 and who met the inclusion criteria, were selected. Patients with a traumatic fracture in at least 1 bone were included in the study. Those who arrived in the ED for reasons other than trauma, those who did not arrive in the ED in a stable condition, and those who left the ED with a referral to an external center without completing the diagnosis and treatment processes during their follow-up in the ED or hospital, were excluded during patient selection. Retrospective data screening was a limitation of the research in terms of detecting some chronic diseases. The 1-year patient-selection period was divided into 3 separate periods for analysis. The first of these periods was pre-pandemic lockdown (September 15, 2019 – March 14, 2020; 182 days), and the patients in this period were designated “Group 1”. The second encompassed the lockdown period during the pandemic (March 15, 2020 – May 31, 2020; 78 days), and patients seen during this period were designated “Group 2”. Patients seen during the normalization period after the lockdown was lifted (June 1, 2020 – September 15, 2020; 107 days) were designated “Group 3”. Elderly trauma patients who were admitted in these 3 different periods were evaluated and compared with each other.

DATA COLLECTION:

The patients’ demographic data, trauma mechanism, fracture types, accompanying injuries, COVID-19 polymerase chain reaction (PCR) test results, ED outcome, operation status for hospitalized patients, hospital outcome, length of stay, and cost data were recorded retrospectively from hospital electronic records. The study sample consisted of patients aged 65 and over who were seen in the ED due to trauma between September 15, 2019 and September 15, 2020, and met the research criteria.

STATISTICAL ANALYSIS:

Data were recorded and analyzed using IBM SPSS Statistics for Windows, Version 20.0 (Armonk, NY: IBM Corp.). Percentage and median values were calculated for descriptive statistics. Pearson Chi-Square and Fisher’s Exact tests were used for speed comparisons in categorical data, and data were expressed as number of samples, missing data, and percentage. Bonferroni correction was used for P value correction in subgroup analyses. The P value was used for statistical significance, and results with P<0.05 were considered statistically significant.

Results

DEMOGRAPHIC CHARACTERISTICS IN ALL GROUPS:

There were 1646 patients included in the study. The numbers in each group were as follows: 901 (54.7%) of the patients were in Group 1, 226 (13.7%) were in Group 2, and 519 (31.5%) were in Group 3. The average age of the study group was 77.9±8.4 years. No statistically significant difference was found between the groups in terms of age distribution (P=0.772). It was observed that 63.1% (n=1038) of the patients included in the study were female. No statistically significant difference was detected between the groups in terms of gender distribution. (P=0.457). The average age and gender percentages of the 17 patients whose COVID-19 PCR test results were positive were similar to those of the entire patient group (76.5±6.4 years; 58.8% female; n=10).

TRAUMA MECHANISMS:

It was determined that the most common injuries were due to falling from one’s own height on the ground (n=1190; 72. 3%), falling from an elevated height (n=169; 10.3%), and in-vehicle traffic accidents (n=83; 5%). While the rate of falling on the ground was 79.2% in Group 2, it was 69.9% in Group 1 and 73.4% in Group 3. While the rate of injuries due to traffic accidents was 2.2% in Group 2, it was 6.3% in Group 1, and 4% in Group 3 (Table 1).

BONE FRACTURES:

Femur fracture (n=611; 37.1%), rib fracture (n=265; 16.1%), humerus fracture (n=161; 9.8%), lumbar vertebra fracture (n=160, 9.7%), and radius fracture (n=131; 8%) were detected in the fracture evaluation of the patients included in the study. A statistically significant difference was found between the groups with regard to the frequency of ulnar and femur fractures. Fractures of these 2 bones were more common in Group 2 (P=0.026 and P=0.035, respectively). Among the study groups, radius fractures were detected less frequently in Group 1 than in the other groups, and rib fractures were detected at a higher rate in Group 1 than in the other groups, although this difference was not statistically significant (P=0.092; P=0.154 respectively) (Table 2). Femur fractures were detected in 70.6% (n=12), lumbar vertebra fractures in 17.6% (n=3), and humerus fractures in 11.8% (n=2) of the 17 COVID-19 PCR-positive patients.

CLINICAL OUTCOMES AND COSTS FOR PATIENTS IN THE ED:

Regarding the ED outcomes, 53.1% (n=874) of the patients were hospitalized. A statistically significant difference in ED outcome was detected between the study groups (P=0.009): the hospitalization rate was higher in Group 2 than in the other groups. The median value for length of hospital stay of patients admitted from the ED was 6 (4–10) days. No statistically significant difference was found between the groups in terms of length of hospital stay (P=0.765) (Table 3).

When the cost for the patients was calculated, the median cost for all patients included in the study was determined to be 1028.65 (383.33–6087.87) Turkish Liras. The difference is due to the difference between Groups 1–2 and between Groups 1–3. There was no significant difference between Groups 2–3 (Table 3).

Regarding the outcome for patients who had been hospitalized, while 91.7% (n=769) were discharged, 8.3% (n=70) died before discharge. There was no statistically significant difference between the groups in terms of hospital outcomes (P=0.185). However, the high mortality rate in Group 2 is noteworthy (Table 3).

Discussion

In this study, trauma characteristics in elderly patients who were admitted to the ED during pandemic lockdowns were analyzed. Comparisons were made between the pre-lockdown and post-lockdown periods. However, it was during the lockdown when it was observed that ulna and femur fractures were more common, due to more cases of falling from their own height.

Among the patients analyzed, female sex was more common. The mean age of all the patients was 77.9±8.4 years. In general, the most common mechanism of injury was falling on the floor. Falls on the floor were more common in Group 2 (the patients seen in the ED during the lockdown period). Ulna and femur fractures were also more common in Group 2. Hospital costs were lower in Group 1 (seen in the ED before the lockdown). The ED visits mostly led to hospital stays.

In a study of injuries by Gioffre et al, in which a large number of patients who had experienced home accidents and traffic accidents were involved, the higher frequency of injuries in the female sex attracted attention [2]. This situation was interpreted as being due to the higher frequency of osteoporosis in the female sex in this study. In our data, injuries were also more prevalent in women. Injuries may of course be associated with osteoporosis. However, we have no data regarding the presence and frequency of osteoporosis in our patients. Although female patients were more common in the study group, there was no difference in the sex ratios between the groups. Accordingly, we do not think that there are different trauma characteristics between men and women in relation to lockdowns.

In this study, we found that in patients admitted during the lockdown period, the mechanism of falling from their own height on the ground was more common and traffic accidents were less common than in other periods. This situation, of course, seems to be the expected result of being locked in a house. In a review of 35 studies and data from 36 987 patients by Antonini et al, we see that the frequency of major trauma in all age groups was lower during the pandemic period [14]. Our study shows that this reality is also valid for the elderly. Marincovitz et al, similar to the results of our study, revealed that injuries at home were more common during the lockdown in England [12]. Jojczuk et al also emphasized in the data they recorded in Poland that traumas at home were more common during the lockdown periods [13]. It is understood that major traumas such as traffic accidents decreased proportionally because of going out less frequently and taking fewer vehicle trips, and that less severe traumas such as falling from one’s own height on the ground were higher in percentage during lockdown periods. In the future, this result may indicate the need to review some preventive medicine practices. Measures such as more frequent home visits by the state or transfer of these elderly people to places where they can be kept under more frequent supervision, such as a nursing home, may be considered, especially for elderly people living alone.

ED costs were significantly lower for the elderly included in our study who were admitted to the ED (Group 1) when the lockdowns had not yet started. That is to say, when pandemic-related lockdowns and the post-lockdown normalization processes had not yet started, diagnosis and treatment processes were less costly. This situation can of course be interpreted as rooted in the fact that treating trauma patients entails more complications in the context of COVID-19 infection, and this increases the costs. Khan et al emphasized that the costs of hospitalization of COVID-19 patients are a serious burden. However, in such studies, the cost of treatment of COVID-19 infection is emphasized. More comprehensive research with different methodologies is needed to obtain more valid data about whether this situation is directly related to COVID-19 or not.

Elderly trauma patients with at least 1 bone fracture were included in this study. Considering broken bones, it was revealed that ulna and femur fractures were significantly more common in Group 2. This may be related to the fact that there was a higher risk of falling from one’s own height on the ground in Group 2 than in the other groups. Wach et al demonstrated, through a simulation analysis, that when people fall from their own height, they ultimately reach speeds of 30–55 km/h when they hit the ground. The same study revealed that the body falls without making any rotational movement if the body is not exposed to an additional and different rotational force during the fall [15]. Accordingly, it can be thought that the ulna bone of a person who falls by using his or her upper extremity as a shield to protect his or her skull and thoracic region is expected to be the part most frequently exposed to trauma. In our study, the observation of more fractures of the ulna bone during lockdowns, and the fact that falling from one’s own height on the ground was a frequently observed mechanism, can be explained by this prediction. We have previously emphasized the speculation of Gioffre et al that there may be a relationship between the higher incidence of osteoporosis in women and the higher number of traumas experienced by elderly women [2]. This was a predictable relationship in our study as well; since there were more women among the elderly trauma patients, we speculate that osteoporosis was a factor. Likewise, osteoporosis may also be related to our observation of more femur fractures in Group 2, since falls from the patients’ own height to the ground were more common in Group 2 than in the other groups in our study.

Regarding the ED outcome for the patients included in the study, a significantly higher rate of hospitalization in all groups, compared with rates noted in the literature, is noteworthy. In a retrospective study by Kirshenbom et al, it was determined that 73% of elderly patients who were admitted to the ED due to trauma were discharged from the ED, 21.5% were hospitalized, and 3.4% died during the hospital stay [4]. In a retrospective study by Bilgetekin et al, it was reported that the rate of leaving the hospital due to treatment refusal in patients diagnosed with proximal femur fracture was 13.3% in the pre-pandemic period, while it was 35.6% in the pandemic period [16]. When the ED outcomes of the patients included in our study were examined, it was found that they were hospitalized at a higher rate, contrary to the literature. In our opinion, possible reasons for this may be that one of the inclusion criteria for our study was that patients must have at least 1 fracture, and also the fact that the hospital where our study was conducted is a tertiary hospital that receives many referrals from other hospitals.

This study has some limitations. First of all, it was a single-center study, and the data were scanned retrospectively from the hospital’s electronic data. Being a single-center study may have resulted in the patient population being different in terms of some general characteristics. The study had a retrospective design that may have excluded some data, such as chronic diseases, from being reported. Especially regarding some issues, such as osteoporosis, which is more common in the female sex, in the future, more comprehensive prospectively designed studies would be helpful. Likewise, in multi-center studies, different and valuable information can of course be obtained.

Conclusions

The mechanism of falling from one’s own height was more common in elderly people experiencing at least 1 bone fracture who were evaluated in the ED during the pandemic lockdown period. The ulna and femur were the most frequently broken bones in elderly trauma during this lockdown period. During future lockdown periods, precautions should be taken to protect the elderly people against falling from their own height.