| 【Introdution】 The risk of late onset bacterial sepsis in preterm infants is higher than in term infants.And the incidence of late onset bacterial sepsis is higher with the smaller gestational age and the lower body weight.There are two possible explanations that preterm infants are highly susceptible to infection.On the one hand,preterm infants have documented immune dysfunction.On the other hand,preterm infants often required prolonged intravenous access,endotracheal intubation,or other invasive procedures.The immune function and thyroid hormones(THs)of preterm infants are proportional to gestational age.Physiological levels of T3 could induce the maturation of dendritic cells and enhance the proliferation of T cell.It has also been reported that T3 can modulate the function of macrophages and T4 has a regulatory role for NK cells.Therefore,THs may reflect the function of immune cells.The most ill preterm infants have more intravenous catheters and other tubes,and these infants often have low serum T4 and T3 concentrations due to non-thyroidal illness.We hypothesize that the preterm infants with perinatal stress may have lower TH levels,which then secondarily affect the function of immune cells,leading to a high incidence of preterm infants with late onset bacterial sepsis.Hence,we use a cohort study to analyze the relationship between TH levels and late onset bacterial sepsis in preterm infants.However,late onset bacterial sepsis in preterm infants has multiple causes.Therefore,a comprehensive,easy-to-use tool that estimates individual risk by incorporating thyroid function and other risk factors could serve as a valuable decision-making tool for clinicians.【Methods】 Data collected from preterm infants who enrolled in the Second Affiliated Hospital of Shantou University Medical College and Child Health Hospital of Guangdong Province from January 2014 to December 2017.We used cohort study to analyze the relationship between THs levels and late onset bacterial sepsis in preterm infants.The exposure in this cohort study was thyroid hypofunction of preterm infants in 3 to 7 days old and the outcome was late onset bacterial sepsis.We built three predicted models of late onset bacterial sepsis in the preterm infants according to the results of multivariable analyses.The data from the Second Affiliated Hospital of Shantou University Medical College was as the primary cohort,and the data from the Maternal and Child Health Hospital of Guangdong Province was defined as the validation cohort in this study.The performance of the nomogram was evaluated by the concordance index(C-index).Comparisons between the difference nomogram models were performed in R software and were tested by the C-index.A larger C-index indicated more accurate prognostic stratification.The total points of each patient in the validation cohort were calculated according to the established nomogram,and then logistic regression in this cohort was performed using the total points as a factor.Finally,the C-index and calibration curve were derived based on the regression analysis.【Results】(1)Overall,211 of 1256 preterm infants(16.8%)had thyroid hypofunction.The principal types of thyroid hypofunction in preterm infants of this study were THOP and low serum T3,which accounted for 8.6% and 7.16% of the study population,respectively.The proportion of thyroid hypofunction became larger with decreasing of gestational age,especially below 31 weeks.According to the results of multivariate logistic regression analysis,the risk of thyroid hypofunction in preterm infants was low birth weight(RR=0.573,95%CI=0.358~0.917),use of dopamine(RR=1.652,95%CI=1.073~2.542),use of albumin(RR=2.156,95%CI=1.441~3.227),and use of antibiotics(RR=1.766,95%CI=1.205~2.59).(2)There were 96 of 1256(7.64%)preterm infants occurred late onset bacterial sepsis.The mean age of onset for the first episode of late onset bacterial sepsis was(21.43±14.62)days.According to the results of multivariate logistic regression analysis,the risk of late onset bacterial sepsis in preterm infants was low birth weight(RR=0.136,95%CI=0.051~0.361),use of endotracheal intubation(EI)(RR=5.195,95%CI=1.797 ~ 15.016),prolong use of umbilical vein catheterization(UVC)(RR=1.346,95%CI=1.194~1.519),prolong use of PICC(RR=0.968,95%CI=0.943~0.994),prolong hospital stay(RR=1.109,95%CI=1.078~1.141),and thyroid hypofunction(RR=4.084,95%CI=2.036~6.262).(3)The incidence of late onset bacterial sepsis in preterm infants with normal thyroid function was 5.45%,while it was up to 18.48% in those with thyroid hypofunction.The results of this cohort study showed that the incidence of late onset bacterial sepsis would increase in preterm infants with thyroid hypofunction.(4)A nomogram was formulated based on multivariable logistic regression analysis of late onset bacterial sepsis in preterm infants.The selection of the final prediction model was performed with a backward step down selection process with the Akaike information criterion.Model 1 only used thyroid hypofunction.Model 2 used the variables of birth weight,use of EI,the time of UVC before infection.Model 3 used the variables of birth weight,use of EI,the time of UVC before infection,and thyroid hypofunction.The C-index of model 2(0.855,95% CI = 0.802 to 0.907)and model 3(0.834,95% CI = 0.775 to 0.894)for predicting late onset bacterial sepsis in the validation cohort was higher than the C-indices of thyroid hypofunction alone(0.606,95% CI = 0.522 to 0.690).With the addition of thyroid hypofunction into model 2,the C-index of model 2 was higher compared the C-index of model 3.A calibration curve showed good agreement between prediction and observation in the probability of late onset bacterial sepsis in model 2 and model 3.A larger C-index indicated more accurate prognostic stratification.Therefore,we established the predicted nomograms for individual risk of late onset bacterial sepsis in preterm infants based on model 2.【Conclussion】 In summary,the results of this study showed that the risk of late onset bacterial sepsis would increase in preterm infants with thyroid hypofunction.A nomogram of late onset bacterial sepsis in preterm infants based on thyroid function,body weight,EI,and the prolong time of UVC,had a good ability to predict late onset bacterial sepsis in preterm infants.The results of this study are conducive to the early detection of preterm infants at risk of developing late onset bacterial sepsis,facilitating the communication of doctors and patients from the perspective of evidence-based medicine,and making effective clinical decisions in the early stages. |
PDF Full Text Request