Vasoactive and/or inotropic drugs in initial resuscitation of burn injuries: A systematic review

Abstract Background According to current guidelines, initial burn resuscitation should be performed with fluids alone. The aims of the study were to review the frequency of use of vasoactive and/or inotropic drugs in initial burn resuscitation, and assess the benefits and harms of adding such drugs to fluids. Methods A systematic literature search was conducted in PubMed, Embase, Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials, UpToDate, and SveMed+ through 3 December 2021. The search included studies on critically ill burn patients receiving vasoactive and/or inotropic drugs in addition to fluids within 48 h after burn injury. Results The literature search identified 1058 unique publications that were screened for inclusion. After assessing 115 publications in full text, only two retrospective cohort studies were included. One study found that 16 out of 52 (31%) patients received vasopressor(s). Factors associated with vasopressor use were increasing age, burn depth, and % total body surface area (TBSA) burnt. Another study observed that 20 out of 111 (18%) patients received vasopressor(s). Vasopressor use was associated with increasing age, Baux score, and %TBSA burnt in addition to more frequent dialysis treatment and increased mortality. Study quality assessed by the Newcastle‐Ottawa quality assessment scale was considered good in one study, but uncertain due to limited description of methods in the other. Conclusion This systematic review revealed that there is a lack of evidence regarding the benefits and harms of using vasoactive and/or inotropic drugs in addition to fluids during early resuscitation of patients with major burns.


| INTRODUCTION
Burn wounds are dynamic, and both depth and surface area may progress after the initial injury. 1,2 Optimal treatment may reverse vulnerable tissue surrounding the irreversibly damaged center of a burn injury. 1 The direct tissue injury from the burn triggers a release of systemic inflammatory mediators, development of a strong negative interstitial pressure, loss of endothelial glycocalyx structure, increased capillary permeability, and rapid edema formation. 3 This leads to the cardiovascular dysfunction known as the burn shock, and the resulting hypovolemia, hypotension, and/or hypoperfusion may become an urgent threat to life. 2,4,5 Thus, the aim of initial fluid resuscitation of burn patients is to prevent the development of hypovolemic burn shock with as little fluid as possible. 5 Initial fluid resuscitation with crystalloids is generally recommended in major burns covering more than 20% total body surface area (TBSA). [6][7][8] Many different formulae have been suggested to predict fluid requirements in severe burns. The original Parkland, Evans, and Brook formulae are the most well-known and cited. [9][10][11] These formulae and their modified versions still remain in current practice guidelines. 6,8,12 However, resuscitation of severe burns is complicated by the transvascular fluid shifts leading to a rapidly developing burn oedema and the risk of over resuscitation. 13 The resulting fluid overload can become life threatening due to complications such as muscular and/or abdominal compartment syndrome and/or compromised airways. [14][15][16] Strategies to limit the amount of fluid by the use of colloids, subtarget resuscitation, or the use of vasoactive and/or inotropic drugs have been proposed. 17 However, current guidelines do not recommend the use of vasoactive and/or inotropic drugs in the initial resuscitation phase (0-24 h). 6,8,12 The use of vasoactive and/or inotropic drugs may have beneficial effects but might also cause harm due to increased risk of cardiac arrhythmia and reduced perfusion of partially burnt skin. 18 The aims of the study were to review the frequency of use of vasoactive and/or inotropic drugs in initial burn resuscitation, and to assess benefits and harms of adding such drugs to fluids.

| Purpose
The purpose of the present study was to perform a systematic review and meta-analysis of initial resuscitation of burn patients admitted to the intensive care unit (ICU), to identify, evaluate, and summarize the findings of all relevant individual studies on use of vasoactive and/or inotropic drugs.

| Aim
The aims were to review how often vasoactive and/or inotropic drugs (norepinephrine, epinephrine, dopamine, and/or dobutamine) were added to intravenous fluids during initial resuscitation of patients with major burns, and to assess benefits and harms of adding vasoactive and/or inotropic drugs. A PICO (population, intervention, comparison, outcome) diagram is presented in Table 1.

| Study registration and reporting
This systematic review was registered in the Prospero database 20 February 2019 (CRD42019120317). 18 The study protocol with primary outcomes and statistical analyses was published before the study was conducted. Results were reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and a PRISMA checklist is available in Appendix A. 19

| Study selection
Two collaborators (KK and NGA) independently screened studies for eligibility according to predefined study selection criteria (Appendix C). Inclusion criteria were studies of burn patients admitted to the ICU comparing use of fluids alone with addition of vasoactive and/or inotropic drugs during initial resuscitation. Exclusion criteria were use of other study populations, use of vasoactive and/or inotropic drugs more than 48 h after burn injury and lack of a control group receiving intravenous fluid alone. Any disagreement was resolved through discussion with a senior author (SB). Both observational and interventional studies were considered for inclusion in order to summarize all available evidence.

| Data extraction
Two independent collaborators (KK and NGA) extracted available data in duplicate according to a predefined data extraction form (Appendix D). This included data on study design, number of included patients, use of vasoactive and/or inotropic drugs, types of burn injuries, % TBSA burnt and presence of inhalation injury. Additional data were age, sex, risk factors for use of vasoactive and/or inotropic drugs, organ function parameters, and patient outcomes such as length of stay and mortality. Any disagreement was resolved through discussion with a senior author (SB).

| Assessment of study quality
Two authors (KK and NGA) independently assessed the risk of bias of included studies using the Newcastle-Ottawa quality assessment scale, which is a well-established tool for assessing the quality of nonrandomized studies in meta-analyses. 20 Any disagreement was resolved through discussion with a senior author (SB).

| Quantitative data synthesis
According to our protocol, we planned to perform a meta-analysis using random effect models if the number of included studies was three or more, and to perform subgroup analyses covering the following topics: Burn injury population, burn injury mechanism, severity of burn injury and/or severity of organ failures. 18 Reports sought for retrieval (n = 0) Reports not retrieved (n = 0) F I G U R E 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 19 flow chart demonstrating how the publications identified through our systematic literature search were screened for eligibility.

| Study selection
The literature search identified 1058 unique publications that were screened for inclusion. After assessing 115 publications in full text, two retrospective cohort studies were included (PRISMA flowchart presented in Figure 1). 21,22 One of them was published as a conference abstract, the other as a full article. Among records not included, four were not available when sought from two separate medical libraries.

| Data extraction
We extracted data from the two included studies by Pape et al and Adibfar et al. 21,22 The main findings are presented in Table 2. In these studies, vasoactive and/or inotropic drugs were used in 18% and 31% of the patients, respectively. The Adibfar study included 52 patients in one group given vasopressor and fluids (PRESSOR group) (n = 16) and one receiving only fluids (NoPRESSOR group) (n = 36) during the first 48 h following burn injury. 22 Patients in the PRESSOR group had a significantly

| Assessment of study quality
Overall study quality, assessed according to the Newcastle-Ottawa scale, differed in the two studies. 20 In the Pape study, the overall study quality was uncertain due to limited description of study methods. On the contrary, the Adibfar study was given the highest score in every quality assessment domain (Table 3). Both studies consisted of ICU burn patients with adequate assessment of interventions and outcomes. Only the Adibfar study controlled for confounding factors. The length and adequacy of follow-up was uncertain in the Pape study and considered good in the Adibfar study.

| Additional findings regarding use of vasoactive and/or inotropic drugs
Among studies not included there were publications that reported use of vasoactive and/or inotropic drugs in burn patients, but with a research question out of the scope of this systematic review. [23][24][25] Among these, several studies discussed the use of vasoconstrictor drugs in burn patients as a local tumescent injection to limit blood loss during burn surgery, considered irrelevant to resuscitation. 26 Pape et al, who found that high Baux score and high %TBSA burnt were associated with vasopressor use. 21 In the latter study, vasopressors were used in 18% of the patients, and vasopressor use was associated with more frequent dialysis treatment and increased mortality. 21 Clinicians treating patients with major burns resuscitate with crystalloids to avoid hypovolemia and ensure tissue perfusion. In patients who remain hypotensive despite adequate volume therapy, advanced hemodynamic monitoring may help to assess whether the patient has a hypovolaemic shock, or if there are elements of other types of shock such as cardiogenic and/or septic shock. However, in patients who do not respond adequately to volume therapy, there are in fact few alternatives to vasoactive and/or inotropic drugs.
Initial fluid resuscitation of patients with major burns is challenging and there is a tight balance between burn shock and fluid overload. 15,32,33 Reported strategies for fluid resuscitation vary widely, especially concerning the amount of fluid administered. 34 Guidelines typically promote consensus formulae of 2-4 ml/kg/% TBSA burnt of crystalloid fluid administered during the first 24 h after injury. 6,8,12 Vasoactive and/or inotropic drugs might benefit the patient in the acute phase of resuscitation by increasing cardiac output and/or vascular resistance, and thereby improving tissue perfusion and reducing the risk of oedema due to over resuscitation. 17,35 However, the potential constraining effects on the peripheral microcirculation of injured skin may raise concerns on their applicability. 17,36 The systemic effects of vasoactive drugs are quite well recognized, but vasopressors' local impact on injured skin in burn patients is poorly evaluated. Knabl et al revealed that a temporary reduction in skin perfusion due to systemic epinephrine administration seemed to indicate progression of burn necrosis in rabbit models. 35 Currently, use of vasoactive and/or inotropic drugs is not recommended in burn resuscitation guidelines. 6 to what degree vasoactive and/or inotropic drugs should be part of resuscitation of patients with major burns.
In conclusion, this systematic review revealed that there is lack of evidence regarding benefits and harms of using vasoactive and/or inotropic drugs in addition to fluids during early resuscitation of major burn patients. Only two small retrospective studies were identified, and we certainly need more data to decide if vasoactive and/or inotropic drugs should be added to fluids or not. There seems to be a discrepancy between treatment guidelines suggesting use of fluid alone, and clinical practice studied in surveys indicating that vasoactive and/or inotropic drugs are frequently used.

AUTHOR CONTRIBUTIONS
All listed authors have contributed to the manuscript substantially and have agreed to the final submitted version.

ACKNOWLEDGMENT
We want to thank the medical library at the University of Oslo for performing and documenting the literature search.

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

CONFLICT OF INTEREST
None of the authors of this article have any conflict of interest in the manuscript, including financial, consultant, institutional, or other relationships that might lead to bias.