In this study, we describe for the first time the morbidity of analytes of critical laboratory values in burn patients at a provincial burn centre and their impact on outcomes. Furthermore, the main finding of this study was that 5 analytes of critical laboratory values independently associated with in-hospital mortality were screened out: hyperglycaemia, hyperkalaemia, hypofibrinogenemia, hypertroponinemia and hypoxemia.
In the early stage after burn injury, due to the increase in the occurrence of glucose and the impaired uptake of glucose by the tissues, the increase in glucose and lactic acid leads to an increase in blood glucose, that is, stress hyperglycaemia . Although high blood glucose levels allow severe burn patients to cope with organ function after major trauma and provide enough energy to maintain homeostasis , hyperglycaemia has the most profound impact on the outcomes of patients with severe burns . A study  showed that the failure to achieve early blood glucose control was significantly associated with an increased mortality rate in burn patients. The mechanism by which stress-induced hyperglycaemia causes adverse outcomes in severe burn patients is not fully understood. Stress hyperglycaemia is mediated by severe inflammation and neuroendocrine disorders, and blood glucose concentration seems to be positively correlated with the degree of stress . GLUT-1 and GLUT-3 are glucose transporters that allow glucose to enter cells independently of insulin. Acute critical disease states can upregulate these transporters, possibly allowing large amounts of glucose to enter cells without being affected by the normal downregulation response . Therefore, due to acute severe diseases, many tissues may be susceptible to increased glucose toxicity . Compared with stable blood glucose concentrations, acute blood glucose concentration fluctuations induce more endothelial cell apoptosis, greater endothelial dysfunction and a greater oxidative stress response , leading to higher mortality in acute critically ill patients . These studies confirmed our findings that the critical result of hyperglycaemia was an independent risk factor for in-hospital mortality in burn patients.
A study  showed that blood potassium levels above the normal range were always associated with adverse outcomes for all disease states. Due to the increased risk of fatal arrhythmia, hyperkalaemia can lead to acute or even severe clinical manifestations, putting patients at risk . The incidence of cardiac events increases with increasing blood potassium concentration . The above studies support our finding that the critical result of hyperkalaemia is an independent risk factor for in-hospital mortality in burn patients.
In this study, the critical result of hypofibrinogenemia was an independent predictor of in-hospital mortality in burn patients. The following studies may support our findings. A large-scale multicentre severe trauma survey  found that fibrinogen levels ≦1.0 g/L [OR, 1.824; 95% CI (1.029-3.232)] were independently associated with 28-day mortality. Decreased fibrinogen levels on admission of trauma patients were associated with adverse outcomes . The low fibrinogen levels that lead to poor outcomes after severe trauma may be due to impaired haemostasis and subsequent heavy bleeding . It has also been confirmed that diffuse intravascular coagulation scores increase with decreasing plasma fibrinogen levels .
The myocardial cell generator is qualitatively damaged in the shock stage of burn injury, and its main pathogenesis may be related to ischaemia-reperfusion injury, myocardial cell oxygen utilization and energy metabolism disorders, myocardial oedema and uncontrolled inflammation . Since the heart is a circulatory power organ, this myocardial injury leads to cardiac insufficiency and is also a predisposing factor for burn shock and ischaemia-hypoxic injury . Due to the high sensitivity and specificity of troponin I release, it can play a role in the assessment of indirect myocardial injury after traumatic shock . A study  showed that troponin I release was an independent predictor of ICU mortality, suggesting that this biomarker could be used for early stratification of ICU mortality risk. Therefore, the above studies support our finding that the critical result of hypertroponinemia is an independent predictor of in-hospital mortality in burn patients.
The critical result of hypoxemia in this study was an independent risk factor for in-hospital mortality in burn patients, and this finding may be supported by the following studies. Intractable hypoxemia caused by smoke inhalation was an important risk of death in patients with severe burns . Lower PaO2 values was were associated with higher patient mortality . Multivariate logistic regression analysis found that the oxygen delivery index was an independent predictor of poor outcomes in severe burns .
This study may have some value for the improvement of burn treatment. First, having information about the relative importance of critical laboratory results can help laboratories develop strategies to report them first when resources are limited . They can also be built into the logic of an automated critical value result reporting system. Second, each laboratory is responsible for establishing analytes and limits of clinically relevant critical values . However, the analytes included in the critical result report list often significantly differ between laboratories, and laboratories rarely agree on the same analytes . The lack of evidence-based and result-based cut-off thresholds and analyte lists may be an obstacle to achieving practice coordination, which may be related to the lack of data describing the relative importance of each laboratory to the patient’s prognosis . Our data are important for developing a list of analytes to be included in the critical value reporting procedure in burn units. Finally, doctors need to respond to critical value alerts in a timely and appropriate manner . Information about the relative importance of critical laboratory results may lead to better responses to critical value alarms in burn units.
This study has several limitations. First, this study was a retrospective study, and causal inference could not be made. Second, this study was a single-centre study with a limited sample size, and its results may not be extrapolated to other burn centres. In the future, multicentre research with larger samples will be necessary. Third, there are currently no uniform laboratory critical value standards either internationally or nationally. Therefore, the limits and analytes of critical laboratory values in this study may not be extrapolated to other burn units. Finally, this study did not analyse the morbidity of an analyte of critical laboratory value in the same patient, although such an analysis may be meaningful.