The Effect of Transfer on Outcomes in Burns

Abstract

Nearly a half million people in the United States seek treatment for burn injuries each year. An estimated 40,000 of these patients will require hospitalization, with 30,000 of these hospitalizations taking place at tertiary care centers with specialty-focused burn capabilities.¹ Many patients suffering burn injuries are initially triaged at centers without burn capabilities, and decisions for transfer are made based on the extent and type of injury.

Burn injuries are a unique form of trauma, resulting in complex, multisystem pathophysiologic derangements. Burn centers have the comprehensive infrastructure and medical personnel required to care for this patient population. The specialty care and resources these facilities provide allows for improved outcomes in high acuity burn patients.²

The American Burn Association (ABA) maintains a list of criteria to aid emergency medical personnel in determining which patients may benefit from transfer to a burn center.³ While the published criteria can be used as a triage tool to guide provider decision making, they are not strict guidelines for transfer.

Patient transfer between hospitals can often take several hours. These delays in transfer can allow for worsening of patient condition, lapses in care, and delays in treatment.⁴ Conversely, unnecessary transfers with subsequent rapid discharge result in inefficient use of healthcare resources and increase costs and unnecessarily burden patients.⁵ This study aims to assess whether these transfer-associated treatment delays significantly affect patient outcomes in the treatment of burns.

METHODS

Under an institutional review board-approved protocol, patients meeting criteria for transfer to a burn center were retrospectively identified from a burn registry at University of Louisville Hospital. Published ABA guidelines for referral to a burn center were used as the identifying criteria (Table 1).³ Patients over 18 years who were admitted from 2016 to 2019 were included in the analysis.

Cohorts were divided equally between those brought directly to our burn center from the field and those initially taken to an outside facility with subsequent transfer to our burn center. Comprehensive data regarding patient demographics, the mechanism and extent of injury, operations performed, and hospital course were extracted and compared between the two groups. Outcomes between the two cohorts included the rate of infectious complications, the number of ventilator days, the number of emergent (<24 hours after admission) operations (operative debridements) performed, the total number of operations required during the hospital stay, total length of stay (LOS), and mortality.

Statistical calculations were performed using R^® software (The R Foundation). Continuous variables were compared using the t test for independent means. Categorical variables were compared using the chi-square or Fisher’s Exact Test, when appropriate. Multivariable logistic regression analysis of significant outcome variables was performed. P values less than .05 were considered significant.

RESULTS

A total of 122 patients were included (Table 2). Median age was 48 years. Most patients (72%) were male. Of the total cohort, 61 (50%) burn patients were directly admitted to our facility and 61 (50%) patients were transferred from outlying hospitals. There was no difference between the groups with respect to gender (P = .84) or age (P = .45).

Median percent total body surface area burned did not significantly differ between the two groups (10% vs 10%, P = .08) (Table 3). The groups also did not differ in the proportion of patients suffering concomitant injury (0% vs 6.6%, P = .12) or requiring intubation prior to arrival at our facility (8.2% vs 11.5%, P = .76).

There was no difference between the groups with respect to ventilator days (9 vs 3 days, P = .37), total number of operations (0 vs 0, P = .16), total LOS (3 vs 3 days, P = .44), disposition (home [74% vs 85%], subacute rehabilitation facility [10% vs 10%], long-term acute care facility [3% vs 0%], other disposition [3% vs 0%]), or mortality (10% vs 5%, P = .50) (Table 4).

Transfer patients came from a median of 42 miles (range 0.2–266 miles) from our facility (Figure 1) and experienced a longer median time from injury to burn center admission than directly admitted patients (1 vs 8 hours, P < .01) (Table 4). Directly admitted patients were more likely to have inhalational injury (30.0% vs 6.6%, P < .01), require intubation after admission (16.4% vs 3.3%, P = .03), require an emergent procedure (30.0% vs 8.2%, P < .01), and develop infectious complications (23.0% vs 8.2%, P = .04). With the exception of requiring an emergent procedure, these findings remained true on subgroup analysis of patients with less than 15% TBSA burn (Table 5).

Figure 1.

Distance from transferring facilities. Circle denotes county of burn center.

Figure 1.

Distance from transferring facilities. Circle denotes county of burn center.

On subgroup analysis of patients with greater than 15% TBSA burn (Table 6), transfer patients again experienced longer time from injury to burn center admission (1.4 vs 10 hours, P = .03). Directly admitted patients were still more likely to have inhalational injury (38% vs 9%, P = .02) and require an emergent procedure (50% vs 13%, P < .01). However, there was no difference between the two groups with respect to requiring intubation after admission (23% vs 9%, P = .20) or developing infectious complications (27% vs 17%, P = .51).

On multivariable analysis controlling for age, concomitant injuries, and percentage of total body surface area burned, the greater frequency of inhalational injury and emergent procedures for directly admitted patients remained significant (Table 7).

DISCUSSION

Delayed transfer of patients with burn injuries to facilities with burn-specific capabilities runs the risk of adversely affecting patient outcomes. Our study indicates that patients with more severe, higher acuity injuries, such as those with inhalation injury and those who ultimately required intubation, were more likely to be taken directly from the field to our burn facility. For patients with less severe, lower acuity injuries who otherwise met ABA criteria for transfer to a burn center, short delays in transfer did not affect overall patient outcomes.

Adequate burn care is reliant upon the recognition and understanding of the nature and extent of the burn injury. For example, knowledge of the burn mechanism, depth of burn, total body surface area affected, and the presence of concomitant injuries, including inhalation injury, are key metrics in guiding the management of the burned patient. A recent meta-analysis reviewed 37 studies evaluating care prior to arrival at burn centers and found that errors were common, with frequent overestimation of burns ranging from 75% to 3500%, unnecessary intubation in up to one half of patients, over-administration of intravenous fluids, and inadequate application of basic dressings.⁶ Without an ability to recognize and act on these key metrics, there is a distinct potential for patient harm if transfer to a burn center is delayed.

A number of international studies have examined the effects of transfer delays in burn patients, revealing varied results. One study found that while rural populations within British Columbia experienced an increased total risk of mortality, transfer status did not affect inpatient mortality or LOS.⁷ Likewise, time to admission to a burn center did not influence outcomes in Australia and New Zealand, with the exception of a time of >16 hours to definitive care in the case of inhalation injury.⁸ In contrast, a German study found that patients admitted specifically to burn intensive care units as transfers experienced a longer time to first burn debridement, longer ventilator days, and longer total LOS than those admitted directly. Mortality rates, however, were not different between the groups.⁹

These studies examined the effects of burn transfers within the context of each individual country’s emergency response system and inter-hospital transfer apparatus, which are not necessarily comparable to those in place in the United States. While some have performed similar studies within the context of the U.S. healthcare system, they have been performed in the pediatric population and examined the effects of longer delays in transfer to burn centers.^10,11 Our study evaluated the adult burn population within the United States. Within the context of short delays in transfer, there was no discernable difference in outcomes between those who were admitted directly to a burn center and those who were seen at outlying facilities prior to transfer.

The implications of these findings are significant in terms of cost and resource utilization. A study examining costs found that unnecessary transfers accounted for nearly one third of the burn admissions at their facility and comprised nearly one fifth of the cost associated with their transfers.¹² Our findings indicate that patients with significant burn injury and/or compromised airways were more likely to be promptly identified and taken directly to a burn center. However, for patients with lower acuity burns, initial triage at an outlying facility may present an opportunity to avoid unnecessary transfers and ultimately decrease cost and resource utilization without harming patients. This is particularly true with the advent of telemedicine.

A study at our institution found that communication between critical access centers and tertiary referral centers is often insufficient, resulting in frequent unnecessary transfers.¹³ The use of telemedicine and image-sharing has been shown to enhance this communication within the trauma system.¹⁴ Similarly, telemedicine has been shown to improve the assessment of burn patients at non-burn facilities and re-direct their management when needed.¹⁵ At our institution, we are expanding on these findings to facilitate remote burn care within our region. We are currently developing a burn referral line to be available for use by critical access centers at all hours. This will include video evaluation of burn injuries by the on-call provider at our facility. We have also developed telemedicine burn care courses for first responders and referring centers. Similar educational programs have been implemented elsewhere for providers in resource-limited areas in the setting of generalized trauma and have been shown to better optimize transfer timing.¹⁶ Additionally, we have successfully transitioned half of our burn clinic follow-up visits to telemedicine, which is of particular benefit to patients from more remote areas of our region.

This study should be interpreted considering several limitations. It is inherently limited by its retrospective design. In particular, data regarding total intravenous fluid administration was available for only three of our patients. Additionally, because ABA transfer criteria cover a broad range of acuity, this may have hindered our ability to detect a difference in outcomes amongst the most ill patients. However, subgroup analysis of patients with greater than 15% TBSA burns and multivariable analysis controlling for degree of injury were performed to increase the likelihood of detecting such a difference. While our study cohort allowed us to evaluate the effects of short transfer delays (median delay of 8 hours) on outcomes, our data does not speak to the effect of long delays (>24 hours). We also focused on short-term, inpatient outcomes but did not examine the effects of transfer delays on long-term outcomes after discharge.

CONCLUSIONS

This study demonstrates that patients with significant burn injury are more likely to be taken directly to a burn center for definitive burn care. However, for the lower acuity burns, initial triage at a non-burn facility does not negatively affect outcomes. Improving the assessment of burn injuries at outlying facilities may prevent the transfer of burn patients and the associated transfer costs for patients who might otherwise be adequately cared for at a facility without specialized burn care.

Conflict of interest statement. All listed authors of this study have no conflicts of interest or disclosures.

Funding: No funding or financial support was received. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Presented at the 52nd annual American Burn Association meeting, May 2020.

N.E.B. and E.H.B. contributed equally to this manuscript.

REFERENCES

Published by Oxford University Press on behalf of the American Burn Association 2021.

This post was originally published on this site