| BACKGROUNDAcute hypoxemic respiratory failure (AHRF) causes significant morbidity and mortality in pediatric intensive care unit (PICU), often develops to acute respiratory distress syndrome (ARDS). Incidences of AHRF and ARDS in developed countries are0.7-4.2%of total PICU admissions and its death rate varies at20-30%, but in developing countries its death rate is50-75%. Corresponding incidence and death rate of AHRF in the Chinese PICU are4%and41.8%, respectively, whileas ARDS was at1.44-2.7%and46.4-60.3%, respectively. AHRF is commonly caused by pneumonia from upper and lower respiratory tract bacteria or viruses and inadequate treatment. With advanced technology of cardiopulmonary resuscitation, respiratory support and life sign monitoring in modern emergency and PICU facilities, most AHRF may be prevented or cured without compromise with long term adverse outcome. This is especially becoming true in the domestic hospitals of municipal, provincial and sub-provincial cities in the past two decades, and even extending to county level pediatric services in coastal provinces. There is an increasingly new challenge for the attending staffs to use ventilator support technique at daily intensive care service to combat AHRF in ranges from common lung infection to non-typical pneumonia caused by new type of viral respiratory infection, such as hand-foot-mouth disease syndrome, avian and swine influenza which often prevail paroxysmally and cause complex multiple organ failure.Lung protective ventilation strategy includes restricted tidal volume settings, permissive hypercapnia, ventilation with prone position, high frequency oscillatory ventilation and open lung manouvre, etc., aiming at preventing non-physiological, periodical stretching induced lung injury and facilitating ventilation-perfusion matching at the lower risk of adverse cardiovascular dysfunction. A large multicenter clinical trial in adult ARDS (American ARDS-Net trial) reveals that ventilation with tidal volume at6ml/kg in volume-controlled mode and a plateau pressure<30cm H2O were associated with nearly9%decrease in mortality (number-needed-to-treat,11) compared with higher tidal volume ventilation controls (12ml/kg). When it comes to the clinically conventional ventilation therapy, the result is not notable. There is no multicenter randomly controlled trial on lung protective ventilation strategy in PICU patients with ARDS yet. The fact that physiological parameters of mechanical ventilation between adult and children are different casts an uncertain question as to whether ventilation with lower tidal volume be regarded as a solution in children for lung protective ventilation. This collaborative study group completed a clinical epidemiologic study of incidence, management and outcome of ARDS in25PICUs in2004-2005, and found that ARDS occurred in children at a median age of2years old, and63.8%were below6years old; pressure-control mode was dominant (86.6%), and measured expiratory tidal volume (Vte) at3days after enrollment was10.4±5.4ml/kg. In2006-2007, this collaborative study group perfomed another study in AHRF and ARDS in26PICUs and found that AHRF occurred in children at a median age of11months old, and83%were below6years old, pressure-controlled mode was still used in93%, Vte at3days after enrollment was8.8(8-12) ml/kg. Another finding in this study was that a higher positive fluid balance value at7days after enrollment was associated with increased mortality. In these regards, we designed current multicenter, prospective clinic study targeting to AHRF with ages between29d to less than6years old (5y), with a special protocol for restricted tidal volume ventilation and fluid intake (balance) as non-controlled intervention.OBJECTIVES1To evaluate effect and safety of respiratory support protocol with restricted tidal volume and fluid intake for AHRF.2To observe trend and spectra of the incidence, causes, mortality and its risk factors, and costs in AHRF and ARDS in comparison with two previous multicenter clinic studies in infants and children in PICUs of major cities in China.3To estimate difference in the participating centers with regard to AHRF patient enrollment and therapeutic strategies.4To improve compliance among the collaborative PICUs in conducting multicenter non-randomised study that may lead to randomised, controlled trial.SUBJECTS AND METHODSThe study was designed as a descriptive clinical epidemiologic investigation, conducted prospectively from Jan.2009through Dec.2009for12consecutive months. Based on the previous pediatric AHRF collaborative study group, this study included24PICUs in which eleven from economically developed (newly industrilased) and thirteen from developing (underdeveloped) regions. All PICU admissions were screened according to a domestic pediatric critically ill score and American guidelines for admission and discharge policies for PICU that enrolled in all critically ill patients aged from29days to5years old (before6th birth day). AHRF was defined as:acute onset (<2weeks) of respiratory problems; PaO2≤50mmHg or PaO2/FiO2≤250mmHg for≥12h; requiring endotracheal intubation and mechanical ventilation (FiO2>0.3and positive end-expiratory pressure [PEEP]>2cmH2O to maintain PaO2>60mmHg or SpO2>90%); PaO2/FiO2remained≤250mmHg after12h of ventilation. ALI/ARDS was diagnosed according to the1994America-European Consensus Conference definitions (AECC).Data were recorded daily, including demographic, physiologic, intervention, medication, outcome and cost. Lab of Pediatric Respiratory and Intensive Care at Children’s Hospital of Fudan University served as coordinating center and in charge of collection of data and communication among the participating PICUs. All data were analyzed with statistic software SPSS13.0. Continuous data were presented as means and standard deviation (SD) or median and interquartile range (IQR) where appropriate. Differences of stratified/group data were compared by Mann-Whitney test, and categorical differences were compared with chi-square test. A p value<0.05was considered statistically significant. Assessment of individual clinical risk factors associated with mortality was performed according to previously reported methods for uni-and binary logistic regression. Results were reported as odds ratio (OR) or relative risk and95%confidence interval (CI).RESULTSDuring the12-month study period,15,462patients were admitted to these24PICUs, and12,176identified as critically ill. A total of401met the AHRF criteria,396of them fulfilled ALI criteria and348developed ARDS. Thus they accounted for3.29%,3.25%and2.86%, respectively, of critically ill patients in the total PICU admission.Median age of AHRF patients was0.8year (change to months)(range0.2-2.3y),53.9%(216/401) were less than1year old. Out of401patients,396(98%) were Han ethnic and272(67.8%) were male. More AHRF were diagnosed during the winter-spring season than in the summer-autumn.Most AHRF had pneumonia (95%) or sepsis (35.6%, sepsis+pneumonia) as underlying diseases, among them there were infectious diseases of type Ⅰ H1N1swine influenza (9,2died), measles (5,2died), and hand-foot-mouth disease (5,3died).In hospital of AHRF was30.2%(121/401) and the mortality of90days after enrollment was30.9%(124/401). The median of survival days of AHRF was6.6days (3.4-12.2). The median of ventilation free days (VFD) in28days after enrollment was18.7days (0-23.0), and for the survivors,21.8(17.8-23.7) days.The median of survival days among the non-survivors of AHRF with sepsis was8.5(4.8-13.2) days, in contrast to5days (2.5-11.4) in those without. The median of ventilation days in the non-survivors of AHRF with sepsis was8.8(5.1-14.1) days versus5(2.5-11.4) days in those without.The predominant cause of death was underlying diseases (56.1%,69/123), followed by multiple organ system dysfunction/failure (18.7%,23/123) and respiratory failure (15.4%,19/123). Median cost of AHRF was22,452yuan, twice that of the average for a critically ill patient, more than the average urban resident’s annaul disposable income and4times that of the rural resident’s annual disposable income in China in the same period. Median in-patient stay time of AHRF was11days (7-17.5), with the survivors (13.6days,8.9-19.9) longer than the non-survivors (8.1days,4.6-14.8).All the patients received mechanical ventilation, and93%(346/373) were ventilated on pressure-controlled mode. Median ventilation period was6(0.6-47.6) days. Upon enrollment, median values for Vte were7.5(4.3-12.3) ml/kg, PIP21(12-35) cmH20, MAP12(7-20) cmH20and PEEP (2.2-11.8)5cmH2O. There is no difference in Vte and PIP between the survivors and non-survivors. Median Vte was higher in subgroup with age<1y than those≥1y old. The incidences of VDF<14d in those with age<1y decreases with increasing Vte.The median fluid intake during the first3days was121ml.kg-1.d-1(90-162), and in the first7days,128ml.kg-1.d-1(96-172), along with a fluid balance (fluid in minus out) of31(9-75) ml.kg-1.d-1and33(12-74) ml.kg-1.d-1, respectively. The fluid balance of the survivors was higher than those of the non-survivors on day7and during the first7days. There were no statistical difference in fluid intake between the survivors and non-survivors.By binary logistic analyses, age<1y with cardiovascular system dysfunction as well as levels of PEEP and PaO2was independently associated with mortality in AHRF patients. PRIMS Ⅲ as well as PIP, PEEP and PaO2was independently associated with the incidence of VDF<14d.The incidence, underlying diseases, ventilation settings and fluid management of ARDS were similar to those of AHRF. Cardiovascular system dysfunction, PEEP, PaO2and MV by binary logistic analyses were independently associated with mortality in ARDS patients. And age<1y with cardiovascular system dysfunction as well as PEEP, PIP, Crs and PaO2was independently associated with the incidence of VDF<14d.The median percentage of critically ill patients in all PICU admissions was86.3%(63.3%-91.2%) among the24PICUs, with3less than50%. The number of enrolled patients, incidence, mortality and ventilation rate in the critically ill patients varied greatly among PICUs. The burden of illness in the hospitals from developed regions seemed lower than that in the hospitals from underdeveloped ones. The incidence of withdraw and PICU stay days in university affiliated hospitals were lower than in the non-university ones. Median Vte in the university hospitals was higher than in the non-university ones druing the first4days, and fluid intake and balance were lower than in the non-university ones too. The incidences of VDF<14d in different level by Vte of24PICUs decreased gradually with increasing Vte (P=0.045).CONCLUSIONS1The incidence of AHRF was close to that reported from2006-2007, but the mortality of AHRF and ARDS were lower than that reported in both2004-2005and2006-2007. The patients from29days and below6years old were more suscepteble to the lower tide volume treatment. PEEP and PaO2were the very important facts associated with prognosis in AHRF and ARDS patients.2The incidences of VDF<14d in those with ages<1y subgroup decreased with increasing Vte levels.Ventilation with restricted tide volume and/or fluid intake did not have substantial impact on survival in AHRF patient, but may shorten the course of clinical pathology.3The median of survival days and ventilation days in the non-survivors of AHRF with sepsis was longer than those without. Ventilation with lower Vte had impact on prolonged survival days but did not have impact on the mortality of AHRF with sepis.4Despite an average level of Vte among24PICUs was kept at lower range, there were variations among the participating PICUs with different median levels and ranges of Vte. Only13(54%) PICUs performed mechanical ventilation with Vte in the recommended ranges.5The burden of illness in hospitals from newly industrialized regions tended to be lower in contrast to the hospitals from developing regions. The incidence of withdrawal and PICU stay days in the university hospitals were lower than in the non-university ones. For the ventilation and fluid management the university hospitals performed better according to the study protocol than the non-university ones. |
PDF Full Text Request