Reducing False Alarm Rates in Neonatal Intensive Care: A New Machine Learning Approach.

In neonatal intensive care units (NICUs), 87.5% of alarms by the monitoring system are false alarms, often caused by the movements of the neonates. Such false alarms are not only stressful for the neonates as well as for their parents and caregivers, but may also lead to longer response times in real critical situations. The aim of this project was to reduce the rates of false alarms by employing machine learning algorithms (MLA), which intelligently analyze data stemming from standard physiological monitoring in combination with cerebral oximetry data (in-house built, OxyPrem). MATERIALS & METHODS: Four popular MLAs were selected to categorize the alarms as false or real: (i) decision tree (DT), (ii) 5-nearest neighbors (5-NN), (iii) naïve Bayes (NB) and (iv) support vector machine (SVM). We acquired and processed monitoring data (median duration (SD): 54.6 (± 6.9) min) of 14 preterm infants (gestational age: 26 6/7 (± 2 5/7) weeks). A hybrid method of filter and wrapper feature selection generated the candidate subset for training these four MLAs. RESULTS: A high specificity of >99% was achieved by all four approaches. DT showed the highest sensitivity (87%). The cerebral oximetry data improved the classification accuracy. DISCUSSION & CONCLUSION: Despite a (as yet) low amount of data for training, the four MLAs achieved an excellent specificity and a promising sensitivity. Presently, the current sensitivity is insufficient since, in the NICU, it is crucial that no real alarms are missed. This will most likely be improved by including more subjects and data in the training of the MLAs, which makes pursuing this approach worthwhile.

Network based statistics reveals trophic and neuroprotective effect of early high dose erythropoetin on brain connectivity in very preterm infants.

Periventricular white matter injury is common in very preterm infants and it is associated with long term neurodevelopmental impairments. While evidence supports the protective effects of erythropoetin (EPO) in preventing injury, we currently lack the complete understanding of how EPO affects the emergence and maturation of anatomical brain connectivity and function. In this case-control study, connectomic analysis based on diffusion MRI tractography was applied to evaluate the effect of early high-dose EPO in preterm infants. A whole brain, network-level analysis revealed a sub-network of anatomical brain connections in which connectivity strengths were significantly stronger in the EPO group. This distributed network comprised connections predominantly in the frontal and temporal lobe bilaterally, and the effect of EPO was focused on peripheral and feeder connections of the core structural connectivity network. EPO resulted in a globally increased clustering coefficient, higher global and average local efficiency, while higher strength and increased clustering was found for regions in the frontal lobe and cingulate gyrus. The connectivity network most affected by the EPO treatment showed a steeper increase graph theoretical measures with age compared to the placebo group. Our results demonstrate a weak but widespread effect of EPO on the structural connectivity network and a possible trophic effect of EPO reflected by increasing network segregation, predominantly in local connections.

Retrospective cohort study of all deaths among infants born between 22 and 27 completed weeks of gestation in Switzerland over a 3-year period.

OBJECTIVES: The aim of this research is to assess causes and circumstances of deaths in extremely low gestational age neonates (ELGANs) born in Switzerland over a 3-year period. DESIGN: Population-based, retrospective cohort study. SETTING: All nine level III perinatal centres (neonatal intensive care units (NICUs) and affiliated obstetrical services) in Switzerland. PATIENTS: ELGANs with a gestational age (GA) <28 weeks who died between 1 July 2012 and 30 June 2015. RESULTS: A total of 594 deaths were recorded with 280 (47%) stillbirths and 314 (53%) deaths after live birth. Of the latter, 185 (59%) occurred in the delivery room and 129 (41%) following admission to an NICU. Most liveborn infants dying in the delivery room had a GA ≤24 weeks and died following primary non-intervention. In contrast, NICU deaths occurred following unrestricted life support regardless of GA. End-of-life decision-making and redirection of care were based on medical futility and anticipated poor quality of life in 69% and 28% of patients, respectively. Most infants were extubated before death (87%). CONCLUSIONS: In Switzerland, most deaths among infants born at less than 24 weeks of gestation occurred in the delivery room. In contrast, most deaths of ELGANs with a GA ≥24 weeks were observed following unrestricted provisional intensive care, end-of-life decision-making and redirection of care in the NICU regardless of the degree of immaturity.

The impact of a register on the management of neonatal cooling in Switzerland.

BACKGROUND: Therapeutic hypothermia following hypoxic ischaemic encephalopathy in term infants was introduced into Switzerland in 2005. Initial documentation of perinatal and resuscitation details was poor and neuromonitoring insufficient. In 2011, a National Asphyxia and Cooling Register was introduced. AIMS: To compare management of cooled infants before and after introduction of the register concerning documentation, neuromonitoring, cooling methods and evaluation of temperature variability between cooling methods. STUDY DESIGN: Data of cooled infants before the register was in place (first time period: 2005-2010) and afterwards (second time period: 2011-2012) was collected with a case report form. RESULTS: 150 infants were cooled during the first time period and 97 during the second time period. Most infants were cooled passively or passively with gel packs during both time periods (82% in 2005-2010 vs 70% in 2011-2012), however more infants were cooled actively during the second time period (18% versus 30%). Overall there was a significant reduction in temperature variability (p < 0.001) comparing the two time periods. A significantly higher proportion of temperature measurements within target temperature range (72% versus 77%, p < 0.001), fewer temperature measurements above (24% versus 7%, p < 0.001) and more temperatures below target range (4% versus 16%, p < 0.001) were recorded during the second time period. Neuromonitoring improved after introduction of the cooling register. CONCLUSION: Management of infants with HIE improved since introducing the register. Temperature variability was reduced, more temperature measurements in the target range and fewer temperature measurements above target range were observed. Neuromonitoring has improved, however imaging should be performed more often.

Self-perceived health status and mental health outcomes in young adults born with less than 1000 g.

AIM: To assess self-perceived health status and mental health outcomes of former extremely low-birth-weight (ELBW) infants at young adulthood compared with community norms and to analyse predictors of poor outcome. METHODS: Fifty-five ELBW adults, 18 men (33%), with median (range) gestational age of 28.7 (25.0-34.0) weeks and birth weight of 930 (680-990) grams, born in Switzerland, were included. They self-rated their health status and mental health at a mean (range) age of 23.3 (21.8-25.9) years. Health status was measured by the Medical Outcomes Study Short Form-36 questionnaire and mental health by the Brief Symptom Inventory. RESULTS: The mean scores for both outcome measures were in the normal range. However, the study group self-rated significantly higher physical health status and lower mental health status compared with the community norms, and scores for self-perceived mental health tended to be worse in the former. ELBW adults reported more problems in socio-emotional role functioning compared with the community norms. Female sex was associated with poorer and bronchopulmonary dysplasia with better mental health status. CONCLUSION: Health status and mental health of former ELBW adults were overall satisfying. However, the comparison with the community norms revealed differences, which may be important for parental and patient counselling and developing support strategies.

Neonatal morbidity in singleton late preterm infants compared with full-term infants.

AIM: The aim of this study was to test the hypothesis that singleton late preterm infants (34 0/7 to 36 6/7 weeks of gestation) compared with full-term infants have a higher incidence of short-term morbidity and stay longer in hospital. METHODS: In this retrospective, multicentre study, electronic data of children born at five hospitals in Switzerland were recorded. Short-term outcome of late preterm infants was compared with a control group of full-term infants (39 0/7 to 40 6/7 weeks of gestation). Multiple gestations, pregnancies complicated by foetal malformations, maternal consumption of illicit drugs and infants with incomplete documentation were excluded. The results were corrected for gender imbalance. RESULTS: Data from 530 late preterm and 1686 full-term infants were analysed. Compared with full-term infants, late preterm infants had a significant higher morbidity: respiratory distress (34.7% vs. 4.6%), hyperbilirubinaemia (47.7% vs. 3.4%), hypoglycaemia (14.3% vs. 0.6%), hypothermia (2.5% vs. 0.6%) and duration of hospitalization (mean, 9.9 days vs. 5.2 days). The risk to develop at least one complication was 7.6 (95% CI: 6.2-9.6) times higher among late preterm infants (70.8%) than among full-term infants (9.3%). CONCLUSION: Singleton late preterm infants show considerably higher rate of medical complications and prolonged hospital stay compared with matched full-term infants and therefore need more medical and financial resources.

How useful is routine amniotic fluid and neonatal surface swab microbiology at Caesarean section?

BACKGROUND: Our aim was to evaluate the clinical impact of routine amniotic fluid and neonatal surface swab microbiology at Caesarean section. MATERIALS AND METHODS: Microbiology data from 1 537 neonates delivered by Caesarean section were analysed in the light of clinical outcome. RESULTS: 1 340 (87%) neonates had non-pathogenic bacteria or negative culture results from both amniotic fluid and surface swab samples. Of the 197 (13%) neonates with pathogenic bacteria, 22 (1.4%) were diagnosed with infection, but only in 6 (0.4%) were the bacteria presumed to be responsible for the infection. Amniotic fluid and surface swab culture had sensitivities of 54% and 35%, and positive predictive values of 14% and 17%, respectively, for detecting a neonate at risk of infection. CONCLUSION: Amniotic fluid and neonatal surface swab microbiology at Caesarean section contributes little if anything to postnatal management and can be safely dropped from operative routine.

[The early postnatal weight gain as a predictor of retinopathy of prematurity]. / Der frühe postnatale Gewichtsverlauf als Prädiktor einer Frühgeborenenretinopathie.

BACKGROUND: Premature infants are often stressed by the current retinopathy of prematurity (ROP) screening procedure. Additionally, only < 10 % of the screened infants will develop a ROP stadium requiring laser therapy. Therefore the present screening strategy is unsatisfactory. Furthermore, the current guidelines do not take into account postnatal factors. A new method considering postnatal factors is the weight, insulin-like growth factor, neonatal ROP (WINROP) algorithm. This approach is based on the early postnatal weight gain. The aim of this study was to assign the WINROP-algorithm to a preterm population in Switzerland and to analyze its ability for prediction. PATIENTS AND METHODS: In this retrospective study, all preterm infants with a gestational age (GA) < 32 weeks and/or a birth weight (BW) ≤ 1500 g taken care of in the Department of Neonatology at the University Hospital Zurich from January 2003 to December 2008 were included. The weekly postnatal weight gain was analyzed by means of the modified WINROP-algorithm. RESULTS: Altogether 376 preterm infants were analyzed. In 58 infants a "high-risk" alarm was released, thereof eight preterms developed a severe ROP and four of them needed laser therapy. CONCLUSIONS: The high predictive value of the WINROP-algorithm was confirmed in our population of Swiss preterms. This instrument has the potential to simplify the current ROP screening procedure. Accordingly, the amount of ophthalmoscopies could be reduced significantly.

Interobserver variability in assessment of cranial ultrasound in very preterm infants.

BACKGROUND: Cranial ultrasound (cUS) findings help doctors in the clinical management of preterm infants and in their discussion with parents regarding prediction of outcome. cUS is often used as outcome measure in clinical research studies. Accurate cUS performance and interpretation is therefore required. AIMS: The aims of this study were (i) to assess the interobserver variability in cUS interpretation, and (ii) to evaluate whether level of cUS expertise influences the interobserver variability. METHODS: Fifty-eight cUS image series of preterm infants born below 32 weeks of gestation collected within the Swiss Neonatal Network were sent to 27 observers for reviewing. Observers were grouped into radiologists, experienced neonatologists and less experienced neonatologists. Agreement between observers was calculated using Kappa statistics. RESULTS: When cystic periventricular leukomalacia, intraventricular haemorrhage and periventricular haemorrhagic infarction were combined to one outcome, agreement among all observers was moderate. When divided into subgroups, kappa for the combined outcome was 0.7 for experienced neonatologists, 0.67 for radiologists and 0.53 for inexperienced neonatologists. Marked difference in interobserver agreement between experienced neonatologists and radiologists could be found for haemorrhagic periventricular ifraction (HPI). CONCLUSIONS: Our results suggest that interobserver agreement for interpretation of cUS varies from poor to good varying with the type of abnormality and level of expertise, suggesting that widespread structured training should be made available to improve the performance and interpretation of cUS.

Parents' mental health after the birth of an extremely preterm child: a comparison between bereaved and non-bereaved parents.

OBJECTIVE: To assess the impact of extremely preterm birth (24-26 weeks of gestation) on the mental health of parents two to six years after delivery, and to examine potential differences in post-traumatic growth between parents whose newborn infant died and those whose child survived. METHOD: A total of 54 parents who had lost their newborn and 38 parents whose preterm child survived were assessed by questionnaires with regard to depression and anxiety (HADS) and post-traumatic growth (PTGI). RESULTS: Neither group of parents had clinically relevant levels of depression and anxiety. Mothers showed higher levels of anxiety than fathers. Bereaved parents with no other, living child reported higher levels of depression than bereaved parents with one or more children. Mothers reported higher post-traumatic growth compared to fathers. In particular, bereaved mothers experienced the value and quality of their close social relationships more positively compared to the non-bereaved parents. CONCLUSION: In the long term, bereaved and non-bereaved parents cope reasonably well with an extremely preterm birth of a child. Post-traumatic growth appears to be positively related to bereavement, particularly in mothers.

Assessment of jaundice in preterm neonates: comparison between clinical assessment, two transcutaneous bilirubinometers and serum bilirubin values.

AIM: 1) To compare the clinical assessment of craniocaudal progression of jaundice and two transcutaneous bilirubinometers with serum bilirubin values in preterm neonates; 2) to identify factors affecting the difference between non-invasive bilirubin estimation and serum bilirubin. METHODS: Serum bilirubin was clinically estimated in healthy preterm newborn infants (34 to 36.9 gestational weeks) independently by a primary investigator and by nurses, and subsequently compared with separate measures of two transcutaneous bilirubinometers. RESULTS: A total of 107 measurements were performed on 69 infants. Minolta JM-102 showed the best performance, with ROC area under the curve of 0.96, followed by BiliCheck over the sternum (0.89) and over the forehead (0.88), clinical assessment by nurses (0.73) and by a physician (0.70). Serum bilirubin >190 micromol/l can be detected with 95% sensitivity with Minolta JM-102 > or =19 units, with BiliCheck > or =145 micromol/l over the sternum and > or =165 micromol/l over the forehead and with jaundice progression to the trunk or further (Kramer zone > or =2). Gestational age affects all non-invasive methods in the estimation of serum bilirubin, whereas skin colour affects both BiliCheck and clinical assessment. Ambient light affects only clinical assessment. CONCLUSION: Minolta JM-102 showed the best performance, closely followed by BiliCheck, with clinical assessment performing far worse than either transcutaneous method. None of the three methods are recommended as complete substitutes for serum bilirubin values in jaundiced preterm infants.

[Ethical problems regarding extremely preterm infants]. / Ethische Probleme bei extrem unreifen Frühgeborenen.

In the industrialized countries during the last 20 years, ever smaller preterm infants have been re-animated after birth, subjected to mechanical ventilation, fed parenterally and treated with various drugs. The smaller and more premature the infant, the longer the intensive care, the greater pain and suffering for the infants concerned and the worse the long-term results. Although mortality decreased, chronic problems of the lungs, eyes and brain (cerebral palsy) increased in the surviving infants. Therefore specialists as well as the general public began to discuss the limits of using intensive medicine. Investigations among neonatologists and nursing staff all over Europe resulted in a wide spectrum of opinions where the line should be drawn, e.g. concerning gestational age. A majority is however in favour of stopping intensive care measures when these are not in the best interest of the infant, i.e. when treatment becomes painful and inescapable death is only postponed. In order to reach such a difficult medico-ethical decision in a structured, understandable and well-supported way, special procedures were developed, which have stood their test in practice.

Two years outcome of very pre-term and very low birthweight infants in Switzerland.

BACKGROUND: There are only few reports worldwide on the outcome of very pre-term infants and very low birthweight infants for a whole country. In Switzerland official population statistics are based on birthweight only, gestational age not yet being documented. AIM: The aim of the present study was to assess the outcome at two years of age for a geographically defined high-risk neonatal population based on both birthweight and gestational age. METHODS: All infants born in 1996 included in the Swiss Neonatal Network (a national anonymous registry established by the Swiss Society of Neonatology for liveborn infants before 32 completed gestational weeks or weighing less than 1500 g) were divided into three groups according to gestational age and birth weight: Group 1: born <32 completed gestational weeks and weighing =1500 g; group 2: born after 32 completed gestational weeks and weighing <1500 g; group 3: born <32 gestational weeks and weighing <1500 g. Information at 24 months corrected age about growth, neurological outcome, frequency of respiratory infections, prescription of antibiotics and medical consultations during this period was obtained from the paediatricians caring for the infants. Fair outcome was defined as survival without serious neonatal complications or abnormal neurological findings at 24 months corrected for prematurity. RESULTS: 723 infants were born alive in Switzerland between 1.1. and 31.12.1996 before 32 completed weeks or weighing less than 1500 g at birth. Mortality was 4.3% for a total of 163 infants in group 1 (<32 weeks, =1500 g), 4.6% for 108 infants in group 2 (>32 weeks, <1500 g) and 18.6% for 452 infants in group 3 (<32 weeks, <1500 g). 6.5% of group 1 survivors followed up to 24 months corrected age had a poor neurological outcome as compared to 9.3% in group 2 and 10.9% in group 3. Infants in group 1 needed antibiotics less often after hospital discharge (interquartile range IQR: 0-2 courses) than infants in group 2 (0-3 courses) and 3 (0-3 courses). Infants in group 2 suffered from fewer airway infections (interquartile range 2-5 times) than in group 1 (2-6 times) and 3 (1-7 times). Infants in group 3 needed more medical consultations (IQR 12-21) than those in group 1 (10-16) and 2 (11-16). The overall fair outcome at 24 months corrected age was 85.3% in group 1, 80.7% in group 2 and 59.6% in group 3. A close correlation between overall fair outcome and gestational age at birth on the one hand and with birthweight on the other can be observed. CONCLUSIONS: This study gives estimates for mortality, poor and fair outcome at 24 months corrected age for very low birth weight infants (<1500 g) and for very pre-term infants (<32 completed gestational weeks). Gestational age is as important for predicting outcome as birthweight and should therefore be integrated into national statistics.

Agreement between capillary and arterial lactate in the newborn.

UNLABELLED: Arterial blood lactate is a reliable indicator of tissue oxygen debt and is of value in expressing the degree and prognosis of circulatory failure as a result of various diseases. Therefore, the practical issue of whether capillary lactate measurements might be of equal value was investigated in newborns. In total, 193 simultaneous measurements of capillary and arterial blood lactate concentrations were performed in 25 newborn babies with an indwelling umbilical arterial catheter. A strong linear correlation was found between capillary and arterial lactate concentration (Lcap = 1.02 Lart + 0.04; r = 0.98; p < 0.001). The mean difference was -0.08 mmol/l and the limits of agreement (+/- 2 SD) were +/- 0.69 mmol/l (-0.77 to 0.61 mmol/l). CONCLUSION: Our data show that capillary blood lactate measurements in newborn babies yield lactate concentrations equivalent to arterial measurements over a large concentration range.

Tissue oxygen saturation measured by near infrared spectrophotometry correlates with arterial oxygen saturation during induced oxygenation changes in neonates.

The aim of this study was to compare quantitatively the changes in tissue oxygen saturation (TOS), determined by two algorithms (TOSc and TOSa) based on near-infrared spectrophotometry, to the changes in arterial oxygen saturation (SaO2) measured by pulse oximetry. TOSc is an algorithm derived by the manufacturer (Critikon) based on a modified Beer-Lambert law; TOSa, our own algorithm, uses the diffusion approximation of light transport for the semi-infinite boundary condition. Slow changes of more than 3% in SaO2 were carried out in 20 mechanically ventilated neonates by altering the inspired oxygen fraction. For each change the regression lines of TOSc versus SaO2, TOSa versus SaO2 and TOSc versus TOSa were calculatcd. For each infant the mcan slope, intercept and r2 of these lines were determined. In 18 preterm infants we obtained median 9.5 (range one to 13) measurements corresponding to a total of 166 measurements. The mean SaO2 was 91.6 (SD 2.3)%, TOSc was 64.7 (SD 7.2)% and TOSa was 71.4 (SD 11.0)%. Changes in TOSc and TOSa were strongly correlated to changes in SaO2 (r2 = 0.86 and r2 = 0.87). TOSc considerably but systematically underestimated the size of the change: delta TOSc = 0.49 delta SaO2. TOSa quantified changes reasonably correctly: delta TOSa = 0.90 delta SaO2. Changes in TOSc and TOSa were highly correlated (r2 = 0.98). These results are promising, but the large inter-individual variation requires further work.

Artificial sweetener reduces nociceptive reaction in term newborn infants.

BACKGROUND: Sucrose has been shown to have an analgesic effect in preterm and term neonates. Sucrose, however, has a high osmolarity and may have deleterious effects in infants with fructose intolerance. Furthermore, it may favour caries. We therefore investigated the effects of a commercially available artificial sweetener (10 parts cyclamate and 1 part saccharin), glycine (sweet amino acid) or breast milk in reducing reaction to pain as compared with a placebo. SUBJECTS: Eighty healthy term infants, four days old, with normal birth weight. INTERVENTIONS: The infants were randomly assigned to one of four groups: 2 ml sweetener, glycine, expressed breast milk or water were given 2 min before a heel prick for the Guthrie test. The procedure was filmed with a video camera and analysed by two observers who did not know which medication the infant had received. RESULTS: Using a multivariate regression analysis, the following variables had significant correlation with relative crying time and recovery time: behavioural state before the intervention, the pricking nurse, and the type of medication. Relative crying time and recovery time were significantly less in the sweetener group but not in the glycine and the breast milk group. CONCLUSIONS: The artificial sweetener used in our study reduces pain reaction to a heel prick in term neonates, and thus provides an alternative to sucrose. In contrast, glycine tends to increase pain reaction whereas breast milk has no effect.

Can near infrared spectroscopy of the liver monitor tissue oxygenation?

Measurement of the hepatic oxygenation index by near infrared spectroscopy is a suitable method to estimate the oxygenation and can be a non-invasive means to continuously monitor tissue perfusion and to detect early haemodynamic disturbances in critically ill children.

Continuous monitoring of liver oxygenation with near infrared spectroscopy during naso-gastric tube feeding in neonates.

Near infrared spectroscopy (NIRS) is a non-invasive method of estimating the haemoglobin concentration changes in certain tissues. It is frequently used to monitor oxygenation of the brain in neonates. At present it is not clear whether near infrared spectroscopy of other organs (e.g. the liver as a corresponding site in the splanchnic region, which reacts very sensitively to haemodynamic instability) provides reliable values on their tissue oxygenation. The aim of the study was to test near infrared spectroscopy by measuring known physiologic changes in tissue oxygenation of the liver in newborn infants during and after feeding via a naso-gastric tube. The test-retest variability of such measurements was also determined. On 28 occasions in 25 infants we measured the tissue oxygenation index (TOI) of the liver and the brain continuously before, during and 30 minutes after feeding via a gastric tube. Simultaneously we measured arterial oxygen saturation (SaO2), heart rate (HR) and mean arterial blood pressure (MAP). In 10 other newborn infants we performed a test-retest analysis of the liver tissue oxygenation index to estimate the variability in repeated intra-individual measurements. The tissue oxygenation index of the liver increased significantly from 56.7 +/- 7.5% before to 60.3 +/- 5.6% after feeding (p < 0.005), and remained unchanged for the next 30 minutes. The tissue oxygenation index of the brain (62.1 +/- 9.7%), SaO2 (94.4 +/- 7.1%), heart rate (145 +/- 17.3 min-1) and mean arterial blood pressure (52.8 +/- 10.2 mm Hg) did not change significantly. The test-retest variability for intra-individual measurements was 2.7 +/- 2.1%. After bolus feeding the tissue oxygenation index of the liver increased as expected. This indicates that near infrared spectroscopy is suitable for monitoring changes in tissue oxygenation of the liver in newborn infants.

[Ultrasonically established cystic periventricular leukomalacia (PVL): incidence and associated factors in Switzerland 1995-1997]. / Sonographisch nachgewiesene zystische periventrikuläre Leukomalazie (PVL): Inzidenz und assoziierte Faktoren in der Schweiz 1995-1997.

BACKGROUND: Cystic periventricular leukomalacia (PVL) is an ischemic brain lesion that mainly affects preterm infants and causes severe neurological damage. Diagnosis is made by cranial ultrasonography. Objectives of this study were to determine the incidence, to identify associated factors and to evaluate the frequency of neurological abnormality at discharge. PATIENTS AND METHODS: Infants with PVL in Switzerland were systematically registered (Swiss Pediatric Surveillance Unit, Swiss Neonatal Network) over three years (1995 to 1997). They were compared to a control group matched for gestational age. RESULTS: Over the three year period 40 infants with PVL defined as at least 2 cysts with diameter = 2 mm in the periventricular region were registered (35 of them were preterm babies). In comparison with the matched controls the infants with PVL had received significantly less frequently antenatal corticosteroids (44 vs 78%, Event Rate Ratio 0.57, 95% Confidence Interval 0.38-0.68), they had lower umbilical cord arterial pH and lower Apgar scores; there was a trend to arterial hypotonia and hypocapnia associated with PVL. The infants of the study group needed more often mechanical ventilation or nasal CPAP (92% versus 67%; ERR 1.38, CI 1.07-1.77) and had more often intracranial hemorrhage (39 versus 14%; ERR 2.8, CI 1.13-6.96). 56% of the infants with PVL were considered abnormal at the neurological examination at discharge compared to 28% in the control group (p < 0.02). CONCLUSION: The incidence of PVL in Switzerland is 1.2% for preterm infants with a birth weight less than 1500 g. Cranial ultrasonography on infants at risk for PVL is important because 44% of the infants with PVL didn't show neurologic abnormalities at discharge.