Unconditional Logistic Stepwise Regression Analysis Of Risk Factors For Sepsis In Severely Burned Adult

Background:Sepsis is defined as a complex clinical inflammatory syndrome that is associated withinfection caused by any of pathogenic microorganisms. Sepsis could lead to septic shock andmultiple organ dysfunction syndrome (MODS). Foreign epidemiological studies show that morethan18,000,000patients suffer with severe sepsis around the world every year, and the numberof patients grows with a raise of1.5%annually. The burden for sepsis reaches16.7billion USDannually. In Europe, the number of patients died of sepsis is estimated over150,000every year.In USA, there are750,000cases of patients with sepsis each year, and215,000people die ofsepsis. It is estimated that there may be3,000,000septic patients every year in our country, andmore than1,000,000people died of sepsis. Sepsis has not only brought great difficulties forclinical treatment, but also a serious threat to human health and economic development. InOctober2002, Europe, United States, and some organizations initiated and signed the "Barcelonadeclaration", aimed to decrease the mortality of sepsis.At present, sepsis is still the greatest threat to severe burn patients. Domestic scientists did aretrospective analysis of657cases with total burn area>30%, reported that the incidence ofsepsis was36.4%, the incidence of multiple organ dysfunction syndrome (MODS) was12.8%,and about60%deaths was due to burn sepsis and MODS. Beijing Ji Shui Tan Hospital burn unitreported that Systemic infection is still the main cause of death. Cincinnati Burn Center of USAreported that the mortality of patients with burn area of more than50%was70%~80%due tosystemic infection. The pathogenesis of sepsis and MODS is very complex, and the course ofSepsis is violent and severe, it is difficult to treat and prognosis for the patients with sepsis whenit evolved to septic shock or MODS. Therefore, if we want to obtain an overall treatmentprogress, we must remove or weaken the triggers of severe sepsis and MODS, rather than using the drug for the established sepsis. Nowadays, the main treatments of burn sepsis is integratedtherapeutic regimen by combining prevention with treatment, which mainly focus on theprevention of severe sepsis and MODS, protection of multiple organ function, treatment of burnsepsis and its complications.ObjectiveIn this study,418adult patients with severe burn from2003to2011are collected. We usestatistic analysis, We will elaborate septic epidemiological characteristics, complications, thetriggers of sepsis, the therapeutic effect and the relationship of incidence of sepsis with differenttreatment, This study aim to provide the basis for reducing the incidence of burn sepsis for largeburn area patients.Material and methods484patients with severe burn adult were collected in burn department of Guangzhou RedCross hospital From2003January to2011December.418cases were enrolled in this study.Average data were collected included gender, age, total body surface area (TBSA), third-degreearea, inhalation injury, the cause of burn, hospitalization time, the first operation time, and theincidence of sepsis, MODS, mortality, the cause of death, healing time. Basing on the long-termclinical observation and literature,8factors were chose, they were age (X1), gender (X2), TBSA(total body surface area (X3)), third-degree area (X4), inhalation injury (X5), the cause of burn(X6), hospitalization time (X7), the first operation time (X8). Sepsis was Y. First, analysis ofsingle factor was carried out in order to eliminate the confounding factors, the statisticallysignificant variables were drawn into Logistic regression model for multivariate analysis, then,screening out the independent risk factors of the burn sepsis.According to the different treatment, adult patients with severe burn were divided into twogroups: group A (form2003to2006year) and group B (form2007to2011year), and comparethe incidence of sepsis, septic patients cure rate and wound healing time between two groups.All data were treated by SPSS17software, t test was used for measurement data, χ2testwas used for enumeration data; The statistically significant variables were drawn into Logistic regression model for multivariate analysis, then, screening out the independent risk of the burnsepsis. the paired t test or paired sample Wilcoxon test was used for paired measurement data,matching χ2test was used for paired enumeration data, P <0.05presents statistics difference.Results1. In this study,418adult patients with severe burn patients are enrolled, male353cases,female65cases; mean age:(36.61±13.33) years; total body surface area: mean (57.97±20.01)%TBSA, Ⅲ°area: mean (27.56±26.63)%TBSA; Admission time: mean (42.09±101.78) h;95cases complicated with inhalation injury (22.73%);148cases were with sepsis, theincidence rate of sepsis was35.41%, the mean onset time was (9.18±4.45) d;31patients werewith MODS;37cases died,21cases (57.76%) died of MODS,10cases (27.02%) died of sepsis;6cases (16.22%) died of inhalation injury.2. By the single factor test, the results showed that5factors are associated with burn sepsis(P<0.05), include the total burn surface area (TBSA), third-degree burn area, hospitalization time,inhalation injury, the time of first operation; and the total burn surface area (TBSA), thirddegree burn area were positively correlated with the incidence of burn sepsis (r=0.506, P <0.05;r=0.508, P <0.05); gender, age, the cause of injuries were eliminate (P>0.05).3. By Logistic regression analysis, the results showed that TBSA (OR>1, P <0.05), third-degree burn area (OR <1, P <0.05) and hospitalization time (OR <1, P <0.05) wereindependent risk factors of burn sepsis.4. Comparison of the A group, the incidence of sepsis in B group was decreased, the curerate was increased, and wound healing time was shortened (P <0.05).Conclusion1. Flame was the main cause of burn in this study, the incidence of burn sepsis is higher; themale of18~50years old had higher incidence rate.2. Total burn surface area (TBSA), three-degree burns, inhalation injury, hospitalization time,the time of first-operation were related to burn sepsis. Among them, total burn surface area,three-degree area and hospitalization time were independent risk factors of burn sepsis. 3. Carry escharotomy, fluid resuscitation, enteral nutrition and the necessary mechanicalventilatory out early for the patient with burn sepsis are important for reducing the incidence rateof burn sepsis, improving the prognosis, and shortening the healing time.