International evidence-based guidelines on Point of Care Ultrasound (POCUS) for critically ill neonates and children issued by the POCUS Working Group of the European Society of Paediatric and Neonatal Intensive Care (ESPNIC)

Background Point-of-care ultrasound (POCUS) is nowadays an essential tool in critical care. Its role seems more important in neonates and children where other monitoring techniques may be unavailable. POCUS Working Group of the European Society of Paediatric and Neonatal Intensive Care (ESPNIC) aimed to provide evidence-based clinical guidelines for the use of POCUS in critically ill neonates and children. Methods Creation of an international Euro-American panel of paediatric and neonatal intensivists expert in POCUS and systematic review of relevant literature. A literature search was performed, and the level of evidence was assessed according to a GRADE method. Recommendations were developed through discussions managed following a Quaker-based consensus technique and evaluating appropriateness using a modified blind RAND/UCLA voting method. AGREE statement was followed to prepare this document. Results Panellists agreed on 39 out of 41 recommendations for the use of cardiac, lung, vascular, cerebral and abdominal POCUS in critically ill neonates and children. Recommendations were mostly (28 out of 39) based on moderate quality of evidence (B and C). Conclusions Evidence-based guidelines for the use of POCUS in critically ill neonates and children are now available. They will be useful to optimise the use of POCUS, training programs and further research, which are urgently needed given the weak quality of evidence available.


Introduction
The incorporation of Point of Care Ultrasound (POCUS) has represented a transformative change in clinical practice, challenging the traditional diagnostic and procedural "art" of medicine, especially in acute care environments. It has emerged as "the new" clinical tool and it has displaced, in some way, the classical stethoscope. It is performed and interpreted by the bedside clinician with the intent of either answering a focused question or achieving a specific procedural goal.
The use of POCUS by critical care providers has increased significantly in recent years and adult emergency medicine had pioneered this field with the publication of guidelines for implementation of structured POCUS training [1][2][3][4][5][6][7]. These efforts were translated into paediatrics through emergency medicine. Since then, POCUS has gained popularity and expanded across many paediatric disciplines. However, unified guidelines on the use of POCUS and training do not exist yet for the paediatric and neonatal critical care.
Bedside goal-directed echocardiography has been the first POCUS application in paediatric practice with guidelines for implementation. An expert statement was published in 2011, followed by the United Kingdom Expert Consensus Statement on Neonatologist Performed Echocardiography (NPE) and recommendations for NPE in Europe [8][9][10]. The Australian Clinician Performed Ultrasound (CPU) programme is a well-established academic curriculum developed to train neonatologists in the use of POCUS, though applications are limited to the evaluation of the neonatal brain and heart [10,11] and the training programme is restricted to neonatal physiology. Non-cardiac POCUS may carry more opportunities for use as well as a number of benefits for both patients and providers. However, the lack of guidelines is a barrier to widespread adoption.
As more acute care providers invest in POCUS technology, it is critically important to define POCUS main purposes, establish formal training and develop accreditation guidelines, in order to maximise POCUS benefits and reducing risks. Therefore, the European Society for Paediatric and Neonatal Intensive Care (ESPNIC) assembled a group of international paediatric POCUS key opinion leaders to create evidence-based guidelines for the use of current and emerging POCUS applications in the neonatal (NICU) and paediatric intensive care units (PICU) by any clinician working in these units.

Methods
Three lead authors, one neonatologist (YS), one paediatric intensivist (DDL) and one paediatric cardiologist (CT), identified expert colleagues who significantly contributed with publications in the POCUS field and/or have developed POCUS training courses in the last 10 years, similarly to what had been done with previous ESPNIC guidelines [12]. Panellists selection was performed prior to the literature search and for logistic reasons, the number of participants was limited to a maximum of 20. These colleagues should have been fairly representative of all POCUS fields and both Europe and North America and include also non-ESPNIC members. Moreover, at least one co-author should have been an expert in guidelines development and supervised the whole methodology, while literature search was performed by each panellist for their sub-section. All invited experts agreed to participate. Details of the methods used to produce these guidelines are given in the additional file for online supplementary material (see Additional file 1). These guidelines followed relevant ESPNIC internal procedures for manuscript endorsement, and this included an external review by ESPNIC officers not included among the panellists. Guidelines have been prepared according to the international Appraisal of Guidelines, Research and Evaluation (AGREE) [13]. Each recommendation is intended to be applied both for paediatric and neonatal patients, unless otherwise specified.

Results
A total of 41 recommendations on the use of POCUS in neonatal and paediatric critical care were discussed. There was strong agreement on 22 recommendations after first-round voting. Following the discussion in an in-person meeting among the experts (organised within the 2018 ESPNIC congress (Paris, France)), 19 recommendations were re-worded and re-voted in the second round ( Fig. 1). There was agreement on 17 recommendations and disagreement on 2 recommendations as summarised in Table 1 [15][16][17]. Conversely, a collapsed IVC may be suggestive of hypovolemia. Clinicians performing POCUS should be aware of its limitations especially in children on mechanical ventilation and neonates (especially when an umbilical central venous catheter is placed) in whom no validated normative data yet exist [18,19]. Moreover, mechanical ventilation (especially with high mean airway pressure), spontaneous breathing effort and some concomitant conditions may reduce the reliability of IVC evaluation to predict fluid responsiveness [20]. 3. POCUS may be helpful to assess fluid responsiveness in neonates and children-strong agreement (quality of evidence D). The variation of velocity-time integrals (VTIs) measured at the left ventricular outflow tract (LVOT), using pulse wave Doppler (PWD) in apical 5-chamber view, during inspiration and expiration has been reported to predict volume responsiveness. A variation of > 15% has been reported to have a high predictive value with a sensitivity and specificity exceeding 90% [21,22]. This has been validated in several studies, including many studies involving mechanically ventilated children [16,[23][24][25][26][27][28][29][30][31]. Accurate measurement of VTI needs acquisition of apical 5-chamber view and good alignment of ultrasound beam with the across LVOT like any other Doppler assessment, which requires advanced ultrasonography skills. Hence this recommendation is particularly for neonatal/paediatric intensivists with advanced POCUS skills. 4. POCUS may be helpful for qualitative assessment of cardiac function on visual inspection in neonates and children-strong agreement (quality of evidence D). POCUS should be used for qualitative assessment of cardiac contractility ("eyeballing" assessment) in multiple views such as apical 4-chamber (A4C), parasternal long (PLAX) and short-axis (PSAX) and sub-costal short-axis views [8,15]. Subcostal views may be very helpful in patients where parasternal or apical views impossible or sub-optimal, and in younger children or neonates where sub-costal views may provide even better image quality. 5. POCUS is helpful for semi-quantitative assessment of cardiac function in neonates and children [however, a detailed functional assessment should be performed by a person with advanced echocardiography training]-agreement (quality  of evidence C). POCUS can be used to assess cardiac function semi-quantitatively by assessing ejection fraction (EF) and fraction shortening (FS) [14,15]. In experienced hands, POCUS can be used for further assessment of left and right ventricular function as summarised in Table 2 [14, 15, 32-45]. However, a detailed assessment of cardiac function should be performed by a neonatologist/paediatric intensivist with advanced echocardiography training or a paediatric cardiologist. 6. POCUS is helpful for the assessment of pulmonary artery systolic pressure in pulmonary hypertension in neonates and children-strong agreement (quality of evidence B). In the presence of tricuspid regurgitation, pulmonary artery systolic pressure (PASP) can be reliably estimated using bedside POCUS. In the absence of right outflow tract obstruction, the regurgitation velocity represents the difference in pressure between the right atrium and the right ventricle. PASP is derived using a modified Bernoulli equation [14,15,32]. 7. POCUS is helpful for semi-quantitative assessment of pulmonary hypertension in neonates and children-strong agreement (quality of evidence B). In the absence of tricuspid regurgitation, POCUS can be used for semi-quantitative assessment of pulmonary hypertension by evaluating the interventricular septal position and movement at the end of systole and by assessing flow direction and velocities across a patent ductus arteriosus and/or foramen ovale [14,15] [8][9][10]. In experienced hands, it may be used for hemodynamic evaluation of the ductus arteriosus [8,10].  [53][54][55][56][57][58][59][60][61]. The double lung point has been proposed as a pathognomonic finding [62,63] but it is debated, as it does not seem necessary for TTN diagnosis if normal lung areas are evident [64]. High inter-observer agreement between physicians with different lung ultrasound (LUS) expertise has been reported, which makes the differential diagnosis between RDS and TTN reliable, irrespective of the operator [65]. In preterm neonates with RDS, various studies showed that a semiquantitative ultrasound evaluation of  [58,65]. LUS has been reported to have higher diagnostic accuracy compared with chest X-rays for the diagnosis of pneumonia [69]. However, there is no defined threshold for consolidation size or a consensual method of measurement. 3. POCUS is helpful to semi-quantitatively evaluate lung aeration and help the management of respiratory intervention in acute respiratory distress syndrome (ARDS) in neonates and children-agreement (quality of evidence B). LUS in neonatal and paediatric ARDS shows bilateral diffuse areas of reduced lung aeration with areas of the interstitial syndrome and consolidations, pleural line abnormalities and pleural effusion [70][71][72][73]. Although current diagnostic criteria for ARDS do not yet include LUS, it represents a useful tool for its detection [73,74]. Several LUS aeration scores are used to semi-quantitatively measure the effect of fluid restriction, alveolar recruitment and surfactant administration [75][76][77][78][79]. Scores based on the main lung ultrasound semiology (including A lines, alveolar-interstitial pattern and presence of consolidations) should be preferred over the simple B lines count, as they describe better the lung aeration and have been validated with various techniques [80]. 4. POCUS is helpful to recognise meconium aspiration syndrome (MAS)-agreement (quality of evidence C). MAS is now recognised as a cause of neonatal ARDS [73] and shares the same LUS findings. However, this LUS pattern is dynamic and changes with the spread of meconium plugs during mechanical ventilation [81,82]. 5. POCUS is helpful for descriptive purposes in viral bronchiolitis but cannot provide a differential aetiological diagnosis-strong agreement (quality of evidence C). LUS signs in viral bronchiolitis consist of pleural line irregularities, "sub-pleural" consolidations and areas with interstitial pattern [83][84][85]. Good concordance between operators has been shown in wheezing infants [86]. LUS findings in viral bronchiolitis are similar to those seen during influenza outbreaks [87] but it is not currently possible to differentiate between different forms of viral respiratory infections. 6. POCUS is helpful to accurately detect pneumothorax in neonates and children-strong agreement (quality of evidence B). In adults, LUS has a high diagnostic accuracy for the diagnosis of pneumothorax [88,89] and has been reported to be more sensitive than conventional radiology [90]. Neonatal data confirm this high diagnostic performance for tension pneumothorax [91]. 7. POCUS is helpful to insert chest tube or perform needle aspiration in neonatal tension pneumothorax-strong agreement (quality of evidence B). LUS should not only be used to diagnose pneumothorax but also to provide static guidance for pleurocentesis [91]. Thus, POCUS should be used to identify lung margin, hemidiaphragm and sub-diaphragmatic organs throughout the respiratory cycle before needle or tube insertion to safely avoid them. 8. POCUS is helpful to detect pleural effusions in neonates and children-strong agreement (quality of evidence B). Policy statements in adults recommend the use of LUS for the detection of effusions and evaluation of pleural fluid volume to guide management [55,92,93]. In children, LUS shows high accuracy in the diagnosis of pneumonia-related pleural effusion [47]. 9. POCUS is helpful to guide thoracentesis in neonates and children-strong agreement (quality of evidence B). Ultrasound-guided thoracentesis reduces the risk of complications and increases success rates [55,92,93]. It should be used to identify lung margin, hemidiaphragm and sub-diaphragmatic organs throughout the respiratory cycle before needle or tube insertion to safely avoid them. 10. POCUS is helpful to evaluate lung oedema in neonates and children-agreement (quality of evidence C). Although LUS is accurate in detecting extra-vascular lung fluid [94], it cannot distinguish between cardiogenic and noncardiogenic oedema [55,94]. LUS has been used in children to evaluate cardiogenic lung oedema, by assessing extravascular lung fluid counting numbers of B-lines after cardiac surgery [79]. However, extravascular lung water may obviously be affected by the pressure delivered during mechanical ventilation. Thus, in these conditions, LUS globally evaluates the lung aeration rather than extravascular lung fluid.  [125,126]. Alterations in cerebral blood flow allow inferences to be made regarding brain pathology and raised intracranial pressure (ICP) [127,128]. Estimation of flow velocities and calculation of pulsatility (PI) and resistance indexes (RI) are useful tools for non-invasive monitoring of ICP [129]. Age-dependent normal values of blood velocities in different vessels and indexes have been previously published [130]. Paediatric data are not yet sufficient to support its use and caution must be taken when interpreting results [131,132]. 2. POCUS should be used to detect germinal matrix and intraventricular haemorrhage (IVH) in neonates-strong agreement (quality of evidence A). Intraventricular haemorrhage and parenchymal bleeding remain a frequent and serious complication in extremely preterm infants. POCUS is a useful clinical tool to detect IVH and parenchymal changes [133], and assesses the severity according to Papile's classification [134,135]. In environments where imaging resources are limited, brain POCUS should be used for the diagnosis of IVH which may aid in the redirection of care. 3. POCUS is helpful to detect cerebral blood flow patterns suggesting the presence of cerebral circulatory arrest in children with fused skull bones-agreement (quality of evidence C). Transcranial doppler has been used to establish the presence of cerebral circulatory arrest [135,136] by evaluation of the middle cerebral (MCA) and basilar artery. The following patterns are compatible with cerebral circulatory arrest [126,136]: (a) oscillating waveform or sustained reversal of diastolic flow, (b) small systolic spikes and disappearance of all intra-cranial flow, (c) no flow in MCA, (d) reversal of diastolic flow in extracranial Internal Carotid Artery (ICA) and (e) mean velocity in MCA less than 10 cm/s for more than 30 min. However, interpretation of this assessment must be done with caution. 4. POCUS is helpful to detect cerebral blood flow changes secondary to vasospasm in patients with traumatic brain injury and non-traumatic intracranial bleeding-agreement (quality of evidence C). Vasospasm of cerebral arteries results in increased flow velocities. The Lindegaard ratio (LR) is calculated by dividing mean velocity in MCA by mean velocity in ipsilateral extracranial ICA. As hyperaemia may also increase mean velocities, LR is useful to distinguish between hyperaemia and vasospasm (3:6 ratio is considered a sign of mild vasospasm, and > 6 a sign of severe vasospasm) [136]. 5. POCUS is helpful to detect changes in optic nerve sheath diameter (ONSD) indicative of raised ICP in children with fused skull bones-agreement (quality of evidence B). Measurement of the ONSD is suggestive of papilledema and increased ICP [137,138]; however, data conflict on threshold measurements [137][138][139][140] and papilledema may persist despite normalisation of intracranial pressure. Therefore, it must be reminded that this technique may have relevant measurement errors because of the narrow margins that distinguish pathological from normal. 6. POCUS is helpful to detect cerebral midline shift (MLS) in neonates and children-agreement (quality of evidence C). Measurement of the distance from both temporal bones to the centre of the third ventricle through the temporal acoustic window determines the presence of MLS [141,142]. Minimal shifts (less than 5 mm) in adults have proven to identify abnormal conditions [143]. Yet, normative values for MLS identification in children have not been reported, and this may reduce the helpfulness of this application. E. Recommendations for abdominal POCUS 1. POCUS is helpful for the detection of free intraabdominal fluid in neonates and children-strong agreement (quality of evidence C).

POCUS is helpful in detecting anaesthesiainduced atelectasis in neonates and children-a-
Abdominal POCUS is widely used as a clinical tool for the management and diagnosis of patients with abdominal pathology [144][145][146][147][148]. Evaluation of free fluid is particularly useful when sudden clinical deterioration and hypotension occur and may provide insight into the ongoing pathophysiologic process [149,150] A renal ultrasound is useful to easily detect the presence of hydronephrosis even by novices [152][153][154]. Urinary retention can be evaluated by POCUS assessing postvoid residual volumes.
In the case of anuria, a simple obstruction requiring a urinary catheter placement can be ruled out. 4. POCUS may assess bowel peristalsis in neonates and children-agreement (quality of evidence D). Bowel POCUS may be used to assess peristalsis [155] but insufficient data yet exist correlating peristalsis with feeding tolerance. However, the presence of peristalsis has strong negative predictive value in adults for ileus and gut ischemia in adults [156]. 5. POCUS may recognise hypertrophic pyloric stenosis [although for a definitive diagnosis a detailed assessment should be performed by a paediatric radiologist]-disagreement (quality of evidence D )[157] 6. POCUS may guide peritoneal drainage or aspiration of peritoneal fluid in neonates and children-strong agreement (quality of evidence D). A paracentesis may be required for either diagnostic or therapeutic purposes. POCUS should be used for pre-procedural planning, identification of epigastric vessels and real-time ultrasonographic guidance. In adult studies, the use of ultrasound has been shown to decrease both bleeding complications and the overall cost of care for patients undergoing in-hospital paracentesis [158]. 7. POCUS is helpful to detect signs of necrotizing enterocolitis (NEC) [although for a definitive diagnosis a detailed assessment should be performed by a paediatric radiologist or a person with specific advanced ultrasound training]-agreement (quality of evidence C).
Ultrasound may be a useful adjunct detecting changes consistent with NEC even when radiographs are inconclusive [159,160]. Furthermore, radiographs have poor sensitivity in the diagnosis of NEC [161,162]. POCUS can provide prognostic value identifying free fluid, bowel wall thickness, pneumatosis intestinalis, portal venous gas and vascular perfusion [159,[163][164][165][166]. The International Neonatal Consortium's NEC subgroup recently revisited the necrotizing enterocolitis (NEC) pathogenesis and new diagnostic criteria were proposed [167]. These were based on the so-called 'two out of three' model which includes pneumatosis intestinalis or portal venous gas by abdominal X-rays and/or POCUS. POCUS was included since various studies demonstrated that it outperformed conventional radiology to this end [159].

Discussion
POCUS is increasingly being utilised in neonatal and paediatric critical care as a valuable adjunct to clinical examination. The use of POCUS by the clinician is different than a complete diagnostic study by the specialist and its role is dynamic-the same provider can perform and interpret the study, rapidly integrate this information within the current clinical setting, and then repeat the study to identify changes associated with the intervention. POCUS involves a focused assessment to answer a specific question, and it provides anatomical and/ or physiological information to be integrated with clinical and laboratory data and make timely and accurate decisions possible. The statements on which panellists disagreed concerned two particular indications (endocarditis and hypertrophic pyloric stenosis), whose diagnosis should be done by a detailed ultrasonography in the hands of an expert paediatric cardiologist or radiologist. The role of POCUS (i.e., point-of-care ultrasound performed directly by neonatal/paediatric intensivists at the bedside) for these indications is not clear, while the role of ultrasound, in general, is indeed important. There remains a significant variation in clinical practice-in indications, training program and clinical governance. The application of POCUS in clinical practice is dependent on many factors including availability of ultrasound machines, providers, hospital setting, local patient population and specialty. Even within a unit, practice varies among clinicians and their expertise. We subdivided POCUS recommendations according to the estimated level of training required for their use (Fig. 2), and this may be helpful for their implementation.
Defining a scope of practice is fundamental for POCUS, and these ESPNIC guidelines have been developed for the targeted use of ultrasound in NICU and PICU by any neonatologist or paediatric intensivist. For most diagnostic (heart, lungs, brain and abdomen) and procedural (line placement and fluid drainage) POCUS applications, these guidelines can be used at all ages in children or neonates. However, while using cardiac POCUS in neonates, clinicians should be aware that undiagnosed critical congenital heart defects can present just during this period. They should acknowledge the limitations of skills, and it should not be used as a screening tool for diagnosing or excluding congenital heart defects. A patient with a suspected critical congenital heart defect should be quickly referred to a paediatric cardiologist, even if this implies out-ofhospital patient's transportation. However, in cases of suspicion of a non-critical defect or for patients with low pre-test probability a more expectative management can be provided. Moreover, the use of pulse-oximetry screening for congenital heart defects is useful and should be integrated into the clinical decision-making process, as internationally recommended [168,169].
These guidelines may help in standardising clinical practice across acute care settings and physicians. It is not meant to be prescriptive, but rather to outline the important features of structured POCUS applications in clinical practice. The use of these guidelines often will not have any cost implication for the equipment, as ultrasound machine is almost always available in most neonatal and paediatric intensive care units. ESPNIC will promote the diffusion of these guidelines by appropriate advertisement, links and presentation in the society congresses and training events.
These guidelines have strengths and limitations as any other. Strengths are represented by (1) the fairly subdivided multidisciplinary panellists' group (including both neonatologists and paediatric intensivists, expert in different POCUS areas and coming from different geographical regions), and (2) the use of a standardised methodology. In several fields, these recommendations are based upon a few studies with moderate/low evidence or based upon experts' opinions and this is a striking difference compared to similar guidelines in adult critical care. In fact, authors answered to following PICO questions: "Does point-of-care ultrasound for heart/lung/line placement/abdomen/brain provide any clinical advantage in neonatal or paediatric intensive care practice?" (supplementary methods). These questions were purposely quite general, as POCUS is not a therapeutic intervention, but rather a diagnostic and monitoring technique and we expected few randomised controlled trials or high-quality diagnostic accuracy studies about POCUS specifically aiming to improve any particular paediatric/neonatal outcome. Moreover, some of the POCUS applications may not be immediately used in all intensive care units as in ours, because of the specific expertise required in some particular contexts (such as the extremely preterm neonates or non-sedated and non-collaborative patients). However, the existence of formal guidelines may help to improve the local expertise and foster new studies to expand knowledge in this field and refine future recommendations. This is extremely important as further studies are warranted to demonstrate that POCUS actually improves patients' management and outcome, similar to what has been demonstrated in adult critical care [1][2][3][4][5][6][7]55]. Another concern is that the clinicians involved are all enthusiastic users of POCUS and this may have theoretically contributed to have 'too positive' recommendations. The ESP-NIC POCUS Working Group recognises that there are emerging indications for POCUS outside of those listed here and needs for future research steps. Publication of new literature may require future revision and ESPNIC POCUS Working Group is willing to update guidelines accordingly every 3 years.
These guidelines may provide a substrate to develop a POCUS curriculum and structured training programs, which are urgently needed for quality assurance. This guideline paper is not a training statement: requirements and methods of POCUS training should be a sensible follow-up article to this one.

Conclusions
Despite the lack of published evidence-based guidelines specifically for its use in the neonatal and paediatric intensive care units, POCUS is increasingly being used. ESPNIC POCUS guidelines provide the substrate to optimise POCUS use in neonatal and paediatric critical care and guide future research steps according to currently unmet needs. Guidelines may help in making the clinical practice standardised and clinical governance robust.

Supplementary information
Supplementary information accompanies this paper at https://doi.org/10. 1186/s13054-020-2787-9. Authors' contributions YS, CT and DDL conceptualised and designed the development of the whole project. YS and CT wrote the first manuscript draft. DDL took care of the whole methodology and supervised the whole project. All authors performed literature search and analysis, interpreted the literature data with their specific expertise, participate in meetings discussions and voted on recommendations and manuscript preparation. Moreover, all authors critically reviewed the manuscript for important intellectual content, approved it in its final version and agreed to be accountable for all aspects of the work. The participation to the project did not entail any honorarium. All authors read and approved the final manuscript.

Funding
This project has no funding. ESPNIC provided technical and secretarial assistance for free.

Availability of data and materials
Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Competing interests
The authors declare that they have no competing interests.