Artifact detection in the PO2 and PCO2 time series monitoring data from preterm infants

Background. Artifacts in clinical intensive care monitoring lead to false alarms and complicate later data analysis. Artifacts must be identified and processed to obtain clear information. In this paper, we present a method for detecting artifacts in PCO₂ and PO₂ physiological monitoring data from preterm infants. Patients and data. Monitored PO₂ and PCO₂ data (1 value per minute) from 10 preterm infants requiring intensive care were used for these experiments. A domain expert was used to review and confirm the detected artifact. Methods.Three different classes of artifact detectors (i.e., limit-based detectors, deviation-based detectors, and correlation-based detectors) were designed and used. Each identified artifacts from a different perspective. Integrating the individual detectors, we developed a parametric artifact detector, called ArtiDetect. By an exhaustive search in the space of ArtiDetect instances, we successfully discovered an optimal instance, denoted as ArtiDetector. Results. The sensitivity and specificity of ArtiDetector for PO₂ artifacts is 95.0% (SD = 4.5%) and 94.2% (SD = 4.5%), respectively. The sensitivity and specificity of ArtiDetector for PCO₂ artifacts is 97.2% (SD = 3.6%) and 94.1% (SD = 4.2%), respectively. Moreover, 97.0% and 98.0% of the artifactual episodes in the PO₂ and PCO₂ channels respectively are confirmed by ArtiDetector. Conclusions. Based on the judgement of the expert, our detection method detects most PO₂ and PCO₂ artifacts and artifactual episodes in the 10 randomly selected preterm infants. The method makes little use of domain knowledge, and can be easily extended to detect artifacts in other monitoring channels.     

Bottle- and breast-feeding: effects on transcutaneous oxygen pressure and temperature in preterm infants

Transcutaneous oxygen pressure (tcPO2) and body temperature were monitored for the duration of feeding sessions in five small preterm infants who served as their own controls for bottle-feeding (BoF) and breast-feeding (BrF). Longitudinal data were collected twice weekly for BoF and BrF sessions from infants' first oral feeding until discharge, for a total of 71 feeding sessions: 32 BoF and 39 BrF. Markedly dissimilar for the two feeding methods, tcPO2 patterns suggested less ventilatory interruption during BrF than BoF. Sequential tcPO2 values at baseline, immediately postfeed, and 10 minutes postfeed were significantly different for the feeding methods, with greater declines for BoF. Maximal temperature change, calculated by subtracting the baseline from the most extreme temperature for each feeding session, indicated that infants became significantly warmer during BrF than BoF. Although the small sample size necessitates replication, these results do not support the widely held assumption that BrF is more stressful than BoF for small preterm infants.     

Is transcutaneous PO2 monitoring during exercise a reliable alternative to arterial PO2 measurements?

Summary. Arterial PO₂ measurement during exercise is an important part in the evaluation of pulmonary disease but requires an intra-arterial cannula. However, in clinical work it would be preferable to assess PO₂ non-invasively. To evaluate such a technique, simultaneous measurements of transcutaneous PO₂ (tcPO₂) and arterial PO₂ (PaO₂), sampled from an indwelling arterial radial cannula, were made before, during and after a fatigue or symptom-limited bicycle exercise test in 16 patients referred to hospital because of dyspnoea. In total 181 paired measurements were made. Mean values (range) of PaO₂ and tcPO₂ were 11-2 kPa (5–16) and 9-5 (5–13-3), respectively. The correlation coefficient between PaO₂ and TcPO₂ was only 0–36 (P < 10^-5). By normalizing the values of tcPO₂ and PaO₂ to corresponding values at supine rest before exercise, the correlation coefficient increased to 0–80 (P < 10^-6). Using PaO₂ as golden standard, tcPO₂ described the trend in pO₂ during exercise reasonably well in all cases and this information is often sufficient for assessing the degree of pulmonary insufficiency. Thus, transcutaneous blood gas monitoring during exercise is useful for clinical evaluation of pulmonary disease, but a single arterial blood sample at rest before exercise is recommended for baseline correlation.,     

Evaluation of a single transcutaneous PO2-PCO2 sensor in adult patients

We evaluated a new transcutaneous gas monitor designed to measure simultaneously transcutaneous oxygen (PtcO2) and carbon dioxide (PtcCO2) tensions. A total of 514 simultaneous transcutaneous and arterial gases were obtained in 47 adult ICU patients. Mean PtcCO2 was close (SEE less than 4 torr) to mean PaCO2, but mean PtcO2 was considerably less than mean PaO2. However, PtcO2 changes larger than 15 torr virtually always indicated respective increases or decreases in PaO2. Similarly, PtcCO2 changes larger than 5 torr almost invariably indicated a parallel change in PaCO2. From this study we conclude that monitoring of transcutaneous gases yields reliable trend information on arterial gases and that it is a valuable noninvasive adjunct in the monitoring of gas exchange in adult patients.     

Factors associated with vagal tone responses in preterm infants

The purpose of this study was to examine factors related to vagal tone (VNA) among preterm infants receiving a 10-minute gentle human touch (GHT) intervention three times daily for 10 days. VNA was measured continuously for 10 minutes before, during, and after each 10-minute GHT intervention. Findings indicated that there was a significant relationship between VNA and gestational age, although there were no relationships between VNA and measures of motor activity or behavioral distress. There was no difference in pattern of response to GHT or level of morbidity, average daily weight gain, or behavioral organization among infants with low, moderate, and high baseline VNA levels. There was no difference in VNA comparing infants in the GHT and control groups or during baseline, touch, and posttouch phases for infants in the GHT group. There is a need for further research to examine the usefulness of VNA as a measure of stress vulnerability among preterm infants.     

Factors affecting short-term mortality in very low birth weight infants in Japan

No epidemiological surveys have examined risk factors related to the death of very low birth weight infants (VLBWIs) in Japan. The objectives of this study were to examine the death rate and fatalities related to complications among VLBWIs, and to analyze factors possibly determining the death of VLBWIs. The subjects of this study were 811 VLBWIs admitted to the Neonatal Care Center of Niigata City General Hospital between April 1987 and March 2003. We obtained information on gender, birth weight, gestational age, Apgar scores, single/multiple pregnancy, postnatal transfer, mode of delivery, complications and outcome (alive or deceased) at the time of discharge from medical records. Of the 811 infants, 98 died prior to discharge (12.1%). Logistic regression analysis showed that independent risk factors for death of VLBWIs were male gender (relative risk [RR]: 2.0), low birth weight (RR: 0.56), necrotizing enterocolitis (RR: 58.0), pulmonary hypoplasia (RR: 37.8), chromosomal abnormalities (RR: 36.3), congenital heart diseases (RR: 9.8), persistent fetal circulation (RR: 9.6), neonatal asphyxia (RR: 6.3) and sepsis (RR: 4.4). The risk for death rises 1.8-fold if birth weight decreases by 100 g. A very high risk of perinatal death is associated with necrotizing enterocolitis, pulmonary hypoplasia or chromosomal abnormalities. The risk of death due to congenital heart diseases or neonatal asphyxia is relatively lower, but the incidences of these two disorders are high (8% and 6%, respectively). From the viewpoint of prophylactic treatment aimed at reducing the death rate of VLBWIs, measures to increase birth weight are of primary importance. Furthermore, early treatment and improved perinatal management of congenital heart diseases and neonatal asphyxia are anticipated to reduce the overall death rate of VLBWIs.     

Pedal blood flow and transcutaneous PO2 in normal subjects and in patients suffering from severe arterial occlusive disease

Summary. Pedal transcutaneous partial pressure of oxygen (tcPo₂) has been measured by polarographic method, heating the skin at 44°C. In 50 normal subjects, mean tcPo₂ measured 54 · 5 ± 7 mmHg. Among 43 patients suffering from chronic ischaemia of lower limbs, mean tcPo₃ measured 40·8±8 mmHg in patients with from claudication and 16·1±15 mmHg in patients suffering from rest pain and/or gangrene. The variability of repeated tcPo₂ measurements, expressed as 1 SD of the mean, was 4·5 mmHg in normal subjects and 2·9 mmHg in patients. The relationships between pedal subcutaneous blood flow measured in xenon-133 clearance method and pedal tcPo₂ have been studied in nine normal subjects and in five patients suffering from severe chronic ischaemia of lower limbs (rest pain and/or gangrene). There was a positive correlation between blood flow and tcPo, in normal subjects (r=0·77, P< 0·001). In patients suffering from severe ischaemia, there was no correlation between these two parameters, but measured blood flow was sometimes very high in areas where tcPo₂ was low. It is likely that ¹³³Xe clearance method considerably overestimates local blood flow in these patients, because there is considerably less fat in subcutaneous tissue of chronic severely ischaemic areas. Thus, partition coefficient should be determined in each patient. However, tcPo₂ may constitute an index of nutritional circulation, while ¹³³Xe clearance actually measures total subcutaneous blood flow.     

Comparison of arterial blood gas with continuous intra-arterial and transcutaneous PO2 sensors in adult critically ill patients

We compared the partial pressure of oxygen directly via a continuous intra-arterial probe (PiaO2) and indirectly using a transcutaneous device (PtcO2) with simultaneously obtained arterial blood PaO2. The PiaO2 values were measured using a bipolar oxygen sensor placed through an 18-ga arterial catheter. The PtcO2 values were measured using a transcutaneous O2-CO2 sensor placed on the abdomen. Seven critically ill, hemodynamically stable, ventilator-dependent adult patients were studied. Measurements were obtained at varying concentrations (0.25 to 1.0) of inspired oxygen after a 10-min stabilization. A total of 78 simultaneous values were obtained; by linear regression: PiaO2 = 0.91 PaO2 + 1.39 (r = .98, standard errors of the estimate [SEE] = 18.6); PtcO2 = 0.39 PaO2 + 36.2 (r = .89, SEE = 14.1). To assess these instruments as trend monitors, we compared the changes in simultaneous PaO2, PiaO2, and PtcO2 values; by linear regression: delta PiaO2 = 0.90 delta PaO2 + 3.88 (r = .96, SEE = 27.7); delta PtcO2 = 0.43 delta PaO2 + 5.6 (r = .94, SEE = 15.2). We conclude that, although these instruments correlate highly with the PaO2, the SEE was substantial and therefore may limit their clinical reliability in adults. Any acute or clinically significant change in PiaO2 or PtcO2 should be confirmed with a blood gas PaO2.     

Factors affecting prenatal sonographic estimation of weight in extremely low birthweight infants

Because critical management decisions are based on sonographic estimation of fetal weight in fetuses less than 1000 g, we sought to evaluate the accuracy of birthweight prediction in this range and to identify factors affecting this accuracy. Fetal weight was estimated using several published methods in 53 fetuses with birthweights less than 1000 g. Standard deviations greater than 12.3% indicate more random error in the sonographic weight prediction than has been reported in higher weight groups. No statistically significant differences were found between patient groups with decreased, normal, or increased amniotic fluid volume or portable examination. There was a trend toward lower mean deviation (2.9 vs 6.0%) and standard deviation (8.9 vs. 15.0%) in studies with scan quality judged "good" compared with "poor" based on ability to visualize anatomic landmarks.     

Cardiovascular support in preterm infants

BACKGROUND: Despite increasing investigation in the area of cardiovascular instability in preterm infants, huge gaps in knowledge remain. None of the current treatments for hypotension, including the use of inotropic agents, have been well studied in the preterm population, and data regarding safety and efficacy are lacking. Thus, the labeling information regarding the use of inotropes as therapeutic agents in this population is inadequate. OBJECTIVE: This article reviews the current deficiencies in knowledge with respect to measuring and achieving normal organ perfusion; summarizes the clinical, methodological, and ethical issues to consider when designing trials to evaluate medications for hemodynamic instability in the preterm neonate; and proposes 2 possible trial designs. Unanswered questions and potential obstacles for the systematic study of drugs to treat cardiovascular instability in preterm neonates are discussed. METHODS: The neonatal Cardiology Group was established in 2003 by the US Food and Drug Administration (FDA) and the National Institute of Child Health and Human Development (NICHD) as part of the Newborn Drug Development Initiative. The Cardiology Group conducted a number of teleconferences and one meeting to develop a document addressing gaps in knowledge regarding cardiovascular drugs commonly used in low-birth-weight neonates and possible approaches to investigate these drugs. This work was presented at a workshop cosponsored by the NICHD and the FDA held in March 2004 in Baltimore, Maryland. Information for this article was gathered during this initiative. RESULTS: To develop rational, evidence-based guidelines corroborated by robust scientific data for cardiovascular support in newborns, well-designed and adequately powered pharmacologic studies and clinical trials are needed to evaluate the safety and efficacy of inotropic agents and to determine the short- and long-term effects of these drugs. Trials investigating the currently available and novel therapies for cardiovascular instability in neonates will provide information that can be incorporated into product labeling and a scientific framework for cardiovascular management in critically ill neonates. The Cardiology Group identified and prioritized 2 conditions for investigation of therapeutic options for the management of neonatal cardiovascular instability: (1) cardiovascular instability in preterm neonates; and (2) cardiac dysfunction in neonates after cardiopulmonary bypass surgery. Key research questions in the area of cardiovascular instability in the preterm infant include determining optimal blood pressure (BP) in preterm infants; identifying better measures than BP to determine organ perfusion; optimizing hemodynamic treatments; and clarifying any associations between BP or therapy for low BP and mortality, intraventricular hemorrhage, periventricular leukomalacia, necrotizing enterocolitis, retinopathy of prematurity, and neurodevelopmental outcome. The Cardiology Group concluded that the study of inotropic agents in neonates using outcomes of importance to patients will require a complicated trial design to address the elements discussed. The group proposed 2 clinical trial designs: (1) a placebo-controlled trial with rescue therapy for symptomatic infants; and (2) a targeted BP trial. CONCLUSION: This summary is intended to stimulate and assist future research in the area of cardiovascular support for preterm infants.     

Tactile stimulation and preterm infants

A critical challenge for care providers is improving the outcomes for premature infants. The issues of how to control various kinds of stimulation, provide appropriate sensory stimulation, and maintain the quality of life of premature infants becomes the central focus of care given in neonatal intensive care units. Therefore, intervention research studies that improve the development and quality of life for premature infants are vitally important. This article comparatively analyzes and critiques five intervention studies of premature infants using tactile stimulation and provides future research directions in this area. By examining the effectiveness of the tactile stimulation studies, some evidence and guidance can be provided for researchers generating knowledge in this area as well as nurses involved in clinical care.     

早产儿专用配方奶对早产儿体重影响的观察及护理

目的观察早产儿专用配方奶喂养对早产儿住院期间体重的变化及规律。方法70例早产儿入院后尽早使用早产儿专用配方奶（Ready-to—feed）开奶,从1～4ml／（kg·d）开始喂养,并根据耐受情况逐渐增加奶量直至全胃肠道喂养。每天监测体重,观察有无喂养不耐受情况。结果70例早产儿最低体重出现时间为（4．4±2．5）d,恢复至出生体重时间为（10．3±4．1）d,恢复至出生体重后平均生长速率（22．9±7．8）g／d：达全胃肠道喂养时间（17．1±4．8）d。胎龄、出生体重及恢复至出生体重后的增长速率是影响出院时体重的主要因素（P〈0．05）。结论早产儿专用配方奶（液态奶）喂养早产儿,生理成熟度及恢复至出生体重后增长速度是影响出院体重的主要原因。     

Transcutaneous theophylline collection in preterm infants

Transcutaneous collection of theophylline and its metabolite, caffeine, was undertaken in 33 preterm infants (2 to 89 days old) who were receiving routine theophylline therapy. Collection was done by means of a novel adhesive transcutaneous collection system. The transcutaneous collection system accumulated substances that migrated from the blood to the skin surface by trapping them in an activated charcoal-gel matrix. On one to three occasions, four transdermal collection systems were applied to the back or abdomen of each infant for 4 to 12 hours. During that time, blood samples were obtained for routine monitoring of plasma theophylline levels. Amounts of theophylline (95 +/- 198 ng) and caffeine (83 +/- 77 ng) in the transcutaneous collection system were significantly correlated with the respective average plasma drug concentration and postconceptional age (p less than 0.01). Skin reactions were limited to mild erythema. We concluded that theophylline and caffeine can be collected on the surface of the skin of preterm infants with a novel transcutaneous collection system. Amounts collected by means of the transcutaneous collection system correlated with plasma concentrations consistent with a diffusion process, but they were poor predictors of individual concentrations.