Article Text

Download PDFPDF

Incidence of hypo- and hyper-capnia in a cross-sectional European cohort of ventilated newborn infants
  1. Anton H van Kaam1,
  2. Anne P De Jaegere1,
  3. Peter C Rimensberger2,
  4. on behalf of the Neovent Study Group
  1. 1Department of Neonatology, Academic Medical Center, Emma Children's Hospital, Amsterdam, The Netherlands
  2. 2Pediatric and Neonatal Intensive Care, Department of Pediatrics, University Hospital of Geneva, Geneva, Switzerland
  1. Correspondence to Dr Anton H van Kaam, Department of Neonatology (Room H3-228), Emma Children's Hospital AMC, PO Box 22700, Amsterdam 1100 DD, The Netherlands; a.h.vankaam{at}


Objective To determine the incidence of hypo- and hyper-capnia in a European cohort of ventilated newborn infants.

Design and setting Two-point cross-sectional prospective study in 173 European neonatal intensive care units.

Patients and methods Patient characteristics, ventilator settings and measurements, and blood gas analyses were collected for endotracheally ventilated newborn infants on two separate dates.

Results A total of 1569 blood gas analyses were performed in 508 included patients with a mean±SD Pco2 of 48±12 mm Hg or 6.4±1.6 kPa (range 17–104 mm Hg or 2.3–13.9 kPa). Hypocapnia (Pco2<30 mm Hg or 4 kPa) and hypercapnia (Pco2>52 mm Hg or 7 kPa) was present in, respectively, 69 (4%) and 492 (31%) of the blood gases. Hypocapnia was most common in the first 3 days of life (7.3%) and hypercapnia after the first week of life (42.6%). Pco2 was significantly higher in preterm infants (49 mm Hg or 6.5 kPa) than term infants (43 mm Hg or 5.7 kPa) and significantly lower during pressure-limited ventilation (47 mm Hg or 6.3±1.6 kPa) compared with volume-targeted ventilation (51 mm Hg or 6.8±1.7 kPa) and high-frequency ventilation (50 mm Hg or 6.7±1.7 kPa).

Conclusions This study shows that hypocapnia is a relatively uncommon finding during neonatal ventilation. The higher incidence of hypercapnia may suggest that permissive hypercapnia has found its way into daily clinical practice.

  • hypocapnia
  • hypercapnia
  • mechanical ventilation
  • survey

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What is already known on this topic

  • Hyperventilation leading to hypocapnia is a well-known complication of mechanical ventilation in newborn infants. Hypocapnia increases the risk of intraventricular haemorrhage, periventricular leucomalacia and bronchopulmonary dysplasia. For this reason, some clinicians advocate the use of higher carbon dioxide levels—that is, permissive hypercapnia.

What this study adds

  • Hypocapnia during neonatal ventilation is a relatively uncommon finding in daily clinical practice. The higher incidence of hypercapnia suggests that permissive hypercapnia has found its way into daily clinical practice. Pco2 is higher in preterm than term infants, and lower during time-cycled pressure-limited ventilation than during volume-targeted or high-frequency ventilation.


Despite the increasing use of non-invasive modes of respiratory support, many newborn infants still require invasive mechanical ventilation during their hospital stay.1 ,2 The basic aim of mechanical ventilation is to restore lung function and gas exchange, while avoiding secondary lung injury.

Overventilation leading to hypocapnia is a well-known complication of mechanical ventilation with serious consequences.3 First, studies in preterm infants have shown that hypocapnia can impair cerebral blood flow leading to an increased risk of intraventricular haemorrhage and periventricular leucomalacia.4–6 Second, hypocapnia is associated with an increased risk of bronchopulmonary dysplasia.7 Alveolar overdistension causing ventilator-induced lung injury and a more direct injurious effect of low carbon dioxide levels on lung tissue are the proposed mechanisms for this association.8

In an attempt to avoid these serious complications, clinicians started to target carbon dioxide levels higher than the normal physiological range—that is, permissive hypercapnia.9 This approach is also supported by animal studies showing that hypercapnia had a direct and independent lung-protective effect during mechanical ventilation.8 Studies in preterm infants showed that permissive hypercapnia was feasible, but failed to show clear benefits.10

It is currently unclear how the aforementioned evidence has affected neonatal ventilation practices, as there are, to our knowledge, no studies reporting the incidence of both hypo- and hyper-capnia in ventilated newborn infants. In this cross-sectional study, we collected data on ventilation practices, including blood gases, in a large European cohort of ventilated newborn infants.


This prospective cross-sectional study was conducted between April 2007 and May 2008 in 21 European countries and included 173 neonatal intensive care units (NICUs) (see the end of this paper for the participating centres). On two predefined dates determined by one of the principle investigators (AHvK), bedside data on ventilation practices were anonymously collected for all patients receiving invasive mechanical ventilation. This included ventilation mode and settings, such as ventilation pressure, expiratory volume and ventilator rate.

Local investigators were also asked to record all blood gas analyses performed in the 24 h before bedside data collection, including the sample site (arterial or capillary).

The following demographic and patient data were collected: level of neonatal care, number of neonatal intensive care beds, yearly average of ventilated patients, gestational age, birth weight, age at the start of ventilation, and age and weight at data collection.

Data are presented as mean±SD or median (IQR) depending on their distribution. Subgroup differences were analysed using the unpaired Student t test or analysis of variance followed by a Bonferroni post hoc test. Possible correlations between ventilation setting and partial carbon dioxide pressure (Pco2) were tested with Pearson's correlation coefficient. p<0.05 was considered significant.

The study protocol was approved by the institutional review board of each hospital.


A total of 535 infants were included in this study, and most participating centres were level 3 (88%) NICUs with more than 10 NICU beds (89%) and ventilating more than 50 patients each year (82%). A detailed report on ventilation practices in this cohort has been published previously.11

In 508 (95%) of the included infants, blood gas analysis was performed in the 24 h before bedside data collection (table 1). The median age at data collection was similar on the two predefined dates.

Table 1

Patient characteristics

A total of 1569 blood gas analyses were performed resulting in a mean Pco2 of 48±12 mm Hg (6.4±1.6 kPa) with a range of 17–104 mm Hg (2.3–13.9 kPa). The corresponding mean pH was 7.32±0.09. Hypocapnia (Pco2<30 mm Hg or 4 kPa) and hypercapnia (Pco2>52 mm Hg or 7 kPa) was present in, respectively, 69 (4%) and 492 (31%) of the blood gases.

Blood gas analysis was performed in 842 samples (54%) obtained via an arterial line and 727 (46%) obtained via the capillary route. The mean arterial Pco2 was significantly lower than for capillary blood gases (46 vs 50 mm Hg or 6.1 vs 6.7 kPa, p<0.001). The mean partial arterial oxygen pressure (Pao2) was 64±35 mm Hg (8.5±4.7 kPa).

Subgroup analysis based on gestational age revealed a significantly higher Pco2 in preterm infants than term infants (49 vs 43 mm Hg or 6.5 vs 5.7 kPa, p<0.001). The opposite was true for the pH (7.32 vs 7.36, p<0.001) and Pao2 (58 vs 88 mm Hg or 7.7 vs 11.7 kPa, p<0.001).

The mean Pco2 in blood samples taken within the first 3 days after birth (n=482) was 44±11 mm Hg (5.9±1.5 kPa) and this was significantly lower than in samples taken on days 3–7 (Pco2=46±11 or 6.1±1.5 kPa, p<0.05, n=292) and >7 days (Pco2=51±13 or 6.8±1.7 kPa, p<0.001, n=795) after birth. In line with this finding, the incidence of hypocapnia was higher in the first 3 days of life (7.3%) compared with 3–7 days (2.7%) and >7 days (3.2%) after birth. The incidence of hypercapnia, however, showed the opposite trend, with an incidence of 17.6%, 23.3% and 42.6% after, respectively, <3, 3–7 and >7 days after birth.

A total of 1468 blood samples were obtained during pressure-limited ventilation (72%), volume-targeted ventilation (9%) or high-frequency ventilation (19%). Comparative analysis revealed that Pco2 was significantly lower during pressure-limited ventilation (47±12 mm Hg or 6.3±1.6 kPa) than during either volume-targeted ventilation (51±13 mm Hg or 6.8±1.7 kPa, p<0.05) or high-frequency ventilation (50±13 mm Hg or 6.7±1.7 kPa, p<0.05).

Subgroup analysis of the blood samples taken during conventional ventilation did not reveal a significant correlation between expiratory tidal volume or expiratory minute volume and Paco2.


To our knowledge, this is the first study to report on the incidence of hypocapnia and hypercapnia in a large cohort of ventilated newborn infants. It shows that hypocapnia (Pco2<30 mm Hg or 4 kPa) is relatively uncommon (4% of blood gases), but (permissive) hypercapnia (Pco2>52 mm Hg or 7 kPa) is present in almost a third of the blood gases obtained from ventilated infants.

The strength of this study is that the blood gas data were collected in 173 NICUs across 21 European countries. This makes the results more robust and generalisable than those from single-centre or single-country studies.

The relatively low incidence of hypocapnia is reassuring and is a strong indication that most neonatologists are aware of its adverse effects on the brain and lung, especially in the preterm infant.5–7 This probably, in part, explains why the mean Pco2 was higher in preterm than term infants. It was interesting to observe that Pco2 was significantly lower and the incidence of hypocapnia higher during the first 3 days of life than at later time points. This probably reflects the dynamic changes in lung condition during this period, which often require frequent adjustments of the ventilator setting.

Hypercapnia was present in almost a third of the blood gases analysed. Although this finding seems to suggest that permissive hypercapnia has found its way into NICUs, some caution is warranted as our study cannot establish if these higher Pco2 values were the intentional target of the clinician responsible for mechanical ventilation. It is also important to acknowledge that the benefit of permissive hypercapnia in newborn infants remains to be proven.10 The fact that a small study found an increased risk of adverse long-term outcome in patients subjected to permissive hypercapnia emphasises the importance of further studies.12 It was interesting to observe that hypercapnia was more common in blood gases obtained after the first week of life. This finding may reflect more compromised lung function in infants ventilated after the first week of life. Alternatively, clinicians may be more vigilant for hypercapnia in the fist week of life, as this has been associated with an increased risk of intraventricular haemorrhage.6

Since its introduction in the 1990s, high-frequency ventilation has been associated with an increased risk of hypocapnia and subsequent intraventricular haemorrhage or periventricular leucomalacia.13–15 However, two more recent randomised controlled trials in very preterm infants did not confirm these adverse effects on the brain, suggesting more tightly controlled Pco2 during high-frequency ventilation.16 ,17 The present study seems to confirm this assumption by showing that the mean Pco2 was significantly higher than during conventional mechanical ventilation—that is, time-cycled pressure-limited ventilation.

One of the disadvantages of time-cycled pressure-limited ventilation is the fluctuation of the tidal volume depending on changes in the compliance and resistance of the respiratory system. In volume-targeted ventilation modes, the tidal volume is less variable, resulting in a lower incidence of high tidal volume breaths and blood gases with a low Pco2.18 ,19 Our studies appear to confirm these results in an international and multicentre cohort of ventilated newborn infants, showing that Pco2 during volume-targeted ventilation was significantly higher than in time-cycled pressure-limited ventilation. This finding may also, in part, explain the lower incidence of intraventricular haemorrhage reported during volume-targeted ventilation.19

Interestingly, we did not find a significant correlation between Pco2 and either mechanical expiratory tidal volume or minute volume. There are several possible explanations for this finding. First, the time period of 24 h in which blood gas results were collected may have been too broad to pick up an association with the recorded ventilator parameters. Second, differences in tidal volume during mechanical breaths or the set ventilator rate may have been compensated for by the patient's own spontaneous respiratory drive, which was not recorded in this study. As also reported in a randomised controlled trial on minimal ventilation, this may have attenuated anticipated differences in Pco2 based on ventilator settings.20

Although not the primary goal of this study, we also collected data on Pao2 determined in arterial samples. The mean Pao2 was significantly lower in preterm than term infants, a finding probably best explained by the lower oxygenation targets used in the preterm population.

In conclusion, this study shows for the first time that, in daily clinical practice, hypocapnia is relatively uncommon during neonatal ventilation and that hypercapnia is seen in about a third of blood gases. Furthermore, mean Pco2 tended to be higher in preterm than term infants and during volume-targeted or high-frequency ventilation than in time-cycled pressure-limited ventilation.


We would like to thank J W Theeuwes for database construction and management.



  • For the list of collaborators from the Neovent Study Group please see the end of this paper.

  • Collaborators from the Neovent Study Group The following hospitals and investigator participated in the European cross-sectional study on ventilation practices:

  • University Hospital Gasthuisberg, Leuven, Belgium: A. Debeer; Centre de Pediatrie Gatien de Clocheville, Tours, France: A. Chemin, K. Norbert; Institut de Puériculture, Paris, France: F. Autret; Ipokratio, Kalamaria, Greece: A. Andreou; ErasmusMC-Sophia Children's Hospital, Rotterdam, Netherlands: A. Kroon; Mother and Child Health Institute, Belgrade, Serbia: A. Minić; University Hospital Vienna, Vienna, Austria: J. Schwindt; University Medical Center Utrecht, Wilhelmina Children's Hospital, Utrecht, Netherlands: H. Brouwers; Antwerp University Hospital, Antwerpen, Belgium: P. van Reempts; Children's Hospital University of Ulm, Ulm, Germany: H. Hummler; Emma Children's Hospital, Academic Medical Center, Amsterdam, Netherlands: M. van Veenendaal; Hipokration General Hospital, Thessaloniki, Greece: K. Sarafidis; Leiden University Medical Centre, Leiden, Netherlands: E. Lopriore; Ospedale Maggiore Policlinico, Mangiagalli e Regina Elena, Milano, Italy: F. Mosca; The John Radcliffe, Oxford, United Kingdom: K. McCormick; Universitätsklinikum Mannheim, Mannheim, Germany: T. Schaible; University Medical Center Groningen, Groningen, Netherlands: A. Jaarsma; Bambino Gesù Chidren's Hospital, Roma, Italy: V. Polimeni; General Faculty Hospital Prague, Prague, Czech Republic: R. Plavka, L. Pazderova; Groupe Hospitalier Cochin - Saint Vincent de Paul, Paris, France: J. Patkai, G. Moriette; Hospital de Cruces, Barakaldo, Spain: A. Valls I Soler; Norfolk & Norwich University Hospital, Norwich, United Kingdom: P. Clarke; Ospedale Spedali Civili, Brescia, Italy: C. Migliori; Universitätsklinikum Freiburg, Freiburg, Germany: R. Hentschel; University General hospital, Alexandroupolis, Greece: J. Sigalas; Asklepios Klinik St. Augustin, Sankt Augustin, Germany: M. Ehlen, C. Fremerey; Centre Hospitalier Régional d'Orléans, Orleans, France: M. Roujou-Gris; Clinical Hospital of Obstetrics and Gynecology ‘Cuza-Voda’, Lasi, Romania: M. Stamatin; Great Ormond Street Hospital, London, United Kingdom: Q. Mok; Marienhospital Bottrop, Bottrop, Germany: S. Ata, M. Günther; Städtisches Klinikum Karlsruhe, Karlsruhe, Germany: J. Kühr, U. Seitz; VU University Medical Center, Amsterdam, Netherlands: M. Vermeulen, R. Knol; Centre Hospitalier Universitaire d' Angers, Angers, France:S. Le Bouedec; Charité Universitätsmedizin Berlin, Berlin, Germany: D. Szekessy, R. Wauer; Children's hospital Agia Sofia, Athens, Greece: C. Petropoulou; Hospital Sant Joan de Deu, Barcelona, Spain: J. Moreno Hernando; Karolinska University Hospital, Solna, Stockholm, Sweden: B. Jonsson; Maastricht University Medical Centre, Maastricht, Netherlands: T. Mulder; Royal Maternity Hospital, Belfast, United Kingdom: D. Sweet; Universitätsklinikum Schleswig-Holstein, Lübeck, Germany: E. Herting, W. Goepel; University General hospital of Patras, Rio-Patras, Greece: G. Dimitriou; University Hospital Motol, Prague, Czech Republic: H. Stuchlíková; University Hospital of Ioannina, Ioannina, Greece: M. Baltogianni, S. Andronikou; A. Cardarelli Hospital, Campobasso, Italy: V. Santillo; Azienda Ospedaliera Maggiore Della Carita, Novara, Italy: F. Ferrero; Centre Hospitalier de Troyes, Troyes, France: I. Arnault; Charles University Hospital, Pilsen, Czech Republic: J. Dort; Hopital Charles Nicolle, Rouen, France: T. Blanc; Hospital de São João, Porto, Portugal: G. Rocha, H. Guimarães; Hospital Dona Estefânia, Lisboa, Portugal: D. Virella; Hospital Garcia de Orta, Almada, Portugal: A. Costa; Hospital Geral de Santo António, Porto, Portugal: S. Pedro Frutuoso; Hospital Most, Most, Czech Republic: J. Biolek; I.O.M.C. Polizu Maternity, Bucharest, Romania: S. Stoicescu; Kinder-und Jugendklinik, Universitätsklinikum Erlangen, Erlangen, Germany: M. Schroth; MSCurie Hospital, Bucharest, Romania: C. Cirstoveanu; Radboud University Nijmegen Medical Centre, Nijmegen, Netherlands: W. de Boode; Ulleval University Hospital, Olso, Norway: S. Medbo; Universitätsklinikum Tübingen, Tübingen, Germany: I. Müller-Hansen, C. Poets; University Children's Hospital Bern, Bern, Switzerland: T. Riedel; University Hospital of North Staffordshire, Stoke-on-Trent, United Kingdom: K. Palmer; University of Torino, Torino, Italy: C. Martano; Univesity Medical Center Ljubljana, Ljubljana, Slovenia: I. Stucin Gantar; Azienda Ospedaliera di Verona, Verona, Italy: P. Biban; Bradford Royal Infirmary, West Yorkshire, United Kingdom: S. Chatfield; Centre Hospitalier D'Arras, Arras, France: J. Ghesquiere, B. Theret; Centre Hospitalier Félix Guyon, Saint Denis, France: S. Samperiz; Children's Hospital of Lucerne, Luzern, Switzerland:T. Berger; CHR Citadelle, Liège, Belgium: V. Rigo; Clinica Di Neonatologia Universita Sassari, Sassari, Italy: A. Balato; Complejo Hospitalario Materno-Insular, Las Palmas, Spain: M. Gresa Munoz; Hospital São Francisci Xavier, Lisboa, Portugal: A. Nunes; Isala Clinics, Zwolle, Netherlands: H.Molendijk, S. Beuger; Kaunas Medical University Hospital, Kaunas, Lithuania: A. Puzas; Klinikum der Universität München, München, Germany: S. Hiedl, O. Genzel-Boroviczeny; Klinikum Saarbrücken, Saarbrücken, Germany: D. Anhalt, J. Möller; Länsjukhuset Ryhov, Jönköping, Sweden: F. Ingemansson; Máxima Medical Centre, Veldhoven, Netherlands: F. Halbertsma; Ospedale Civile Spirito Santo, Pescara, Italy: V. de Cesaris; Oulu University Hospital, Oulu, Finland: T. Saarela; Papageorgiou General Hospital, Thessaloniki, Greece: P. Karagianni, C. Tsakalidis; Rikshospitalet-Radiumhospitalet Medical Center, Oslo, Norway: P. Tølløfsrud; Universitiy Hospital Brussels, Brussels, Belgium: A. Bougatef; Uppsala University Children's Hospital, Uppsala, Sweden: R. Sindelar; Aarhus University Hospital, Skejby, Aarhus, Denmark: K. Wisborg, T. Brink Henriksen; Azienda Ospedaliera G. Salesi, Ancona, Italy: C. Flumini , V. Carnielli; Azienda Ospedaliera Vito Fazzi, Lecce, Italy: S. Giannuzzo; Centre Hospitalier Universitaire Tivoli, La Louvière, Belgium: A. Dussart; Centre Hospitalier Victor Dupouy, Argenteuil, France: D. Brault; Diana Princes of Wales Hospital, Grimsby, United Kingdom: M. Samy; Hôpital Universitaire Des Enfants Reine Fabiola, Bruxelles, Belgium: A. van Wien; Hospital Fernando Fonseca, Amadora, Portugal: M. Cunha, E. Paulino; Klinikum am Steinenberg, Reutlingen, Germany: H. Schneider; Klinikum Braunschweig, Braunschweig, Germany: A. Sandvoss; Klinikum Coburg, Coburg, Germany: P. Dahlem; Klinikum Luedenscheid, Luedenscheid, Germany: B. Koester; Linkoping University Hospital, Linkoping, Sweden: E. Olhanger; Lukaskrankenhaus Neuss, Neuss, Germany: R. Wentzell; Maternidade Bissaya Barreto, Coimbra, Portugal: C. Ramos; Maternidade Julio Dinis, Oporto, Portugal: M. Augusta Areias; North Tees, Stockton on Tees, United Kingdom: I. Verber; Ospedale Cardinale G. Panico, Tricase, Italy: G. Presta; Ospedali Riuniti Azienda Ospedaliero Foggia, Italy: R. Magaldi; S. Giovanni Calibita Fatebenefratelli - Isola Tiberina, Roma, Italy: R. Agostino; Sørlandet Hospital, Kristiansand, Norway: O. Lund, J. Ulriksen; St-Marien Hospital, Bonn, Germany: U. Steder; Universitätsklinikum Giessen und Marburg, Giessen, Germany: D. Faas ; Westküstenklinikum Heide, Heide, Germany: R. Jensen; Alexandra General Hospital, Athens, Greece: G. Baroutis; Centre Hospitalier Universitaire St-Pierre, Bruxelles, Belgium: V. Gouder de Beauregard; County Hospital of Emergency, Cluj-Napoca, Romania: G. Zaharie; Gottfried von Preyersches Kinderspital, Vienna, Austria: A. Eng-Schwartz; Helios Klinikum Wuppertal, Wuppertal , Germany: M. Heldmann; Hospital Pedro Hispano, Matosinhos, Portugal: T. Cezanne, A. Pereira; Inselspital Bern, Bern, Switzerland: M. Nelle; Istituto per L'Infanzia 'Burlo Garofolo', Trieste, Italy: F. Uxa; Karolinska University Hospital Huddinge, Stockholm , Sweden: M. Norman; Kinderkrankenhaus auf der Bult, Hanover, Germany: J. Siegel; Klinikum Deggendorf, Deggendorf, Germany: M. Welsch; Klinikum Nürnberg Süd, Nürnberg , Germany: H. Schiffmann; Klinikum St. Elisabeth, Neuburg/Donau, Germany: L. Haftel, F. Wild; Landeskrankenhaus Feldkirch, Feldkirch, Austria: P. Bühr, B. Simma; Leighton Hospital, Cheshire, United Kingdom: A Thirumurugan; Odense University Hospital, Odense, Denmark: S. Mortensen; Ospedale San Filippo Neri, Roma, Italy: R. Ciccotti; P.O. Camposampiero, Camposampiero, Italy: G. Carli; Royal Victoria Infirmary, Newcastle upon Tyne, United Kingdom: D. Milligan; St-Vincenz Hospital, Coesfeld, Germany: H. Gerleve; The Royal Wolverhampton Hospital, Wolverhampton, United Kingdom: B. Kumararatne; Umeå University Hospital , Umeå, Sweden: S. Hakansson; University College Hospital Galway, Galway, Ireland: D. O'Donovan; University Hospital for Children and Adolescents, Graz, Austria: F. Reiterer; University Hospital of Geneva, Geneva, Switzerland: P. Rimensberger; Aalesund Hospital, Aalesund , Norway: L. Nietsch; Akershus University Hospital, Lorenskog, Norway: B. Nakstad; Azienda Sanitaria Ospedaliera S. Croce e Carle, Cuneo , Italy: P. Gancia; Barnkliniken Malarsjukhuset, Eskilstuma, Sweden: S. Swanstrom; Centre Hospitalier Chrétien - St Vincent, Rocourt, Belgium: P. Maton; CHU Charleroi, Charleroi, Belgium: E. Cavatorta; City Hospital of Panevezys, Panevezys, Lithuania: R. Tvarijonoviciene; Craigavon Area Hospital, Portadown, United Kingdom: M. Hogan; Diakonissen-Stiftungs-Krankenhaus, Speyer, Germany: P. Zinn; GPR-Klinikum, Rüsselsheim, Germany: M. Freff; Haukeland University Hospital, Bergen, Norway: H. Reigstad; Hospital ‘Dr. Dumitru Popescu’, Timisoara, Romania: G. Olariu; Hospital Cuf Descobertas, Lisboa, Portugal: G. Gonçalves; Hospital da Luz, Lisboa, Portugal: M. Escumalha; Hospital Espírito Santo, Evora, Portugal: H. Ornelas, A. Serrano; Hospital Østfold, Fredrikstad , Norway: S. Anderssen; Hospital SAMS, Lisbon, Portugal: P. Garcia; Hospital Santa Maria, Lisboa, Portugal: A. Mendes Da Graça; Kinderklinik VS-Villingen, VS-Villingen, Germany: C. Bender; Klinikum Kreuzschwestern, Wels, Austria: M. Wald; Klinikum Salzgitter, Salzgitter, Germany: M. Bohn; Klinikum St. Marien, Amberg, Germany: A. Schnelke; Klinikum Traunstein, Traunstein, Germany: T. Trips; Länssjukhuset Halmstad, Halmstad, Sweden: J. Ladekjaer; Northampton General Hospital, Northampton, United Kingdom: F. Thompson; Örebro University Hospital, Örebro, Sweden: E. Lindberg; Ospedale Mauriziano Umberto I, Torino, Italy: M. Frigerio; Ospedale S. Chiara , Trento, Italy: F. Pederzini, G. De Nisi; Ostalb-Klinikum Aalen, Aalen, Germany: G. Saur; Ostschweizer Kinderspital, St. Gallen, Switzerland: M. Losa; Panait Sarbu Hospital, Bucharest, Romania: A. Toma, E. Matu; Princess Elisabeth Hospital, St. Martin, United Kingdom: S. Eckhardt; San Bortolo Hospital, Vicenza, Italy: M. Bellettato; St. Elisabethenkrankenhaus, Lörrach, Germany: H. Fahnenstich , P. Hetzel; Stavanger University Hospital, Stavanger, Norway: J. Bland, B. Øglænd; Turku University Hospital, Turku, Finland: L. Lehtonen; Universitätsklinikum Göttingen, Göttingen, Germany: T. Eichler; University Hospital of Lausanne, CHUV, Lausanne, Switzerland: M. Roth; Vestfold Hospital, Tønsberg, Norway: A. Meberg; Vivantes Klinikum Neukölln , Berlin, Germany: T. Kuehn, M. Emeis

  • Contributors AHvK initiated this collaborative project, designed the study, collected and analysed the data and wrote the draft of the manuscript. He is responsible for the overall content of the manuscript. APDJ helped with the design of the study, data analysis and made a significant contribution to finalising the manuscript. PCR helped with the design of the study, data analysis and made a significant contribution to finalising the manuscript.

  • Funding None.

  • Competing interests None.

  • Ethics approval Reviewed by IRB in participating centres.

  • Provenance and peer review Not commissioned; externally peer reviewed.