If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Pediatric Critical Care Medicine and Neonatology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Kerpener Str.62, 50937 Cologne, Germany
Department of Pediatric Critical Care Medicine and Neonatology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Kerpener Str.62, 50937 Cologne, Germany
Department of Pediatric Critical Care Medicine and Neonatology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Kerpener Str.62, 50937 Cologne, Germany
Department of Pediatric Critical Care Medicine and Neonatology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Kerpener Str.62, 50937 Cologne, Germany
Intraosseous (IO)-access plays an alternative route during resuscitation. Our study in preterm and term stillborns was performed to find alternative IO puncture sites beside the recommended proximal tibia.
Methods
The cadavers used were legal donations. 20 stillborns (mean: 29.2weeks, IQR 27.1–39.6) were investigated. Spectral-CT were analysed to calculate the diameter and circumferences of: i) proximal humerus ii) distal femur iii) proximal tibia iv) diaphyseal tibial. Contrast medium was applied under video documentation to investigate the drainage into the vascular system.
Results
In term newborns, diameter of the cortex of the proximal humeral head is 12.1 ± 1.8 mm, distal end of the femur 11.9 ± 3.4 mm and the proximal tibial bone 12.0 ± 2.4 mm with cross-sectional diameter of 113.5 ± 19.7 mm2, 120.6 ± 28.2 mm2 and 111.6 ± 29.5 mm2, respectively. Regarding the preterm groups, there is a strong age-related growth in diameter and cross -sectional size. The diaphyseal area is the smallest in all measured bones with an age-dependent increase and is about half of that of metaphyseal diameters (proximal and distal) and about one third of that of metaphyseal cross sectional areas. The proximal femoral head region has the largest diameter of all measured bones with an egg-shaped formation with an extensive joint capsula. All investigated metaphyseal areas lack a clearly enclosed bone marrow cavity. Infusion of contrast medium into the distal femoral end and the proximal humerus head demonstrate the drainage of contrast medium into the central venous system within seconds.
Conclusion
Proximal humeral head and distal femoral end might be alternative IO areas which may lead to further IO puncture sites in neonates.
is still lacking. Interestingly, alongside this very special and youngest group of patients, establishing IO infusion even seems difficult to proceed with in toddlers, infants and also adolescents.
with different prospects of success, we now seek alternative puncture areas other than the recommended proximal tibial plate. Indeed, in recent studies in human and animal cadavers, it has been demonstrated that alternative bony structures (e.g. mandibular, fibula, calcaneus) could also transfer the injected liquids and drugs into the systemic venous vessels.
Neonatal Life Support 2020 preferred umbilical venous catheterisation as the primary method of vascular access but an IO route seems a “reasonable alternative” during resuscitation.
There are few answers to the question: “What is the potentially best area for neonatal IO access?”. This can be attributed to both limited knowledge of newborn bony and vascular anatomy and limited studies in this area. Weiss et al. demonstrated age-related variability of the pretibial tissue layer, cortical bone thickness and the diameter of the medullary cavity.
in newborn resuscitation, demonstrating that inexperience is a clear disadvantage in performing this procedure. As there are also well known interspecies differences in bone composition and density,
creating suitable newborn osseous dummies remains a challenge in order to provide adequate simulation scenario devices. We therefore investigated other IO areas in preterm and term newborn cadavers in order to give advice on possible alternative puncture sites with respect to the recommended area.
Methods
The cadavers used for the investigations were legal donations to the Centre of Anatomy, University of Cologne, Germany. After ethical committee approval (No: 16-408), twenty (11 female, 9 male) formaldehyde-fixed (12%) stillborns were investigated (Table 1, median gestational age: 29.2weeks (IQR 27.1–39.6). There was no morphological or congenital known malformation in the bodies investigated other than that the cadavers were decapitated for other studies. Three subgroups (a) 25–27 weeks’ gestational age, (b) 28–36 weeks’ gestational age and c) 37−43 weeks’ gestational age were established. The age of gestation of the cadavers was compared to the documented gestational age by measuring the length of the clavicle according to Sherer et al.
Spectral-CT examination (iQon®, Dual-layer Spectral-CT, PhilipsTM, The Netherlands) were performed, raw data were reconstructed using a soft kernel (soft tissue window) and an ultra-hard kernel (bone window), saved in DICOM format and analysed on a clinical PACS workstation (IMPAX-EE® V.R20XVISU2, AGFA, Germany). CT-scans were analysed.
The diameters and total circumferences of the intact cortices were determined from the bone window CT- images of each cadaver at four predetermined locations: i) proximal humerus at the position of the tuberculus majus (Fig. 1), ii) distal femur 8−10 mm from the cartilage-bone layer (Fig. 2), iii) proximal tibia at the position of the fibular head and tuberositas tibiae (10 mm from the cartilage-bone layer, Fig. 3) iv) diaphyseal tibial (Fig. 3). The investigations involved the humeral, femoral and tibial bone of both extremities of each cadaver whenever possible.
Fig. 1A + C) Frontal and B) transverse section of the left humerus in a term newborn cadaver (39 + 5GA). The diameters and circumferences of the intact cortex were determined from the bone window CT- images (A + B) at the proximal humerus at the hight of the tuberculus majus outside the cartilage-bone layer (i). A red EZ-IO® IO needle PD 15G (1.8 mm diameter) and a green butterfly® 21G (0.8 mm diameter) needle indicate relative sizes.
Fig. 2A + C) Frontal and B) transverse section of the right distal femur in a term newborn cadaver (39 + 5GA). The diameters and circumferences of the intact cortex were determined from the bone window CT- images (A + B) 8–10 mm from the distal cartilage-bone layer (ii). A red EZ- IO® IO needle PD 15G (1.8 mm diameter) indicates relative size.
Fig. 3A + C) Frontal and B) transverse section of the left tibia in a term newborn cadaver (39 + 5GA). The diameters and circumferences of the intact cortex were determined from the bone window CT-images (A + B) at the hight of the fibular head and tuberositas tibiae (iii) (10 mm from the cartilage-bone layer) and diaphyseal tibial (iv). A red EZ-IO® IO needle PD 15G (1.8 mm diameter) and a green butterfly® 21G (0.8 mm diameter) needle indicate relative sizes.
The cartilage-bone layer was identified in all CT-images and excluded from the measurements. Diameters were calculated as the mean value from the maximum and minimum lengths as the cross-section is not always circular.
Results were estimated using mean and standard deviation [MEAN ± S.D.] to account for possible correlations between the different areas. Results of each gestational age group were analysed using one-way ANOVA or unpaired t-test. P < 0.05 was considered statistically significant.
In order to investigate the drainage of IO injected substances an EZ-IO® IO needles PD 15G (for 3−39 kg body weight with a maximum depth of 15 mm, colour pink, Teleflex Medical, Dublin, Ireland) were applied in the distal femoral or proximal humeral bone in some preparations, followed by injection of 5 ml contrast medium (Imeron®, Fa.Bracco™, Konstanz, Germany). X-ray examinations were performed using “ZIEHM EXPOSCOP 8000 Endo apparatus” (Ziehm® Imaging GmbH, Nuernberg, Germany), digitally photographed (PowerDirector 12, CyperLink Corporation®, New Taipei City, Taiwan) and the distribution of given contrast agent was video documented (supplemental video). Focal length was kept constant in all investigations and positioned in such a way as to reduce parallax to a minimum. All procedures were performed by a clinician neonatologist.
The software GraphPadPrism® (San Diego, California, USA) was used for analysis.
Results
In term newborns (GA 37–43, group c) proximal diameter of the humeral head is 12.1 ± 1.8 mm, distal end of the femur 11.9 ± 3.4 mm and the proximal tibial bone 12.0 ± 2.4 mm (Table 2) with cross sectional diameter of 113.5 ± 19.7 mm2, 120.6 ± 28.2 mm2 and 111.6 ± 29.5 mm2, respectively. There is no statistical significance within these investigated areas regarding diameter and cross-sectional area. Regarding the preterm groups (group a + b) there is a clear age-related growth in diameter and cross-sectional size (see supplemental data). As expected, the diaphyseal area is the smallest but still with an age-dependent increase. The results in this middle bony area are about half of that of metaphyseal diameters (proximal and distal) and about one third of that of metaphyseal cross-sectional areas.
Table 2Diameter [mm] and cross sectional area [mm2] in pre- and term newborns.
GA 25−27 BW: 1.2 kg [0.9−1.4]
GA 28−36 BW: 1.5 kg [1.3−2.2]
GA 37−43 BW: 3.5 kg [3.2−3.9]
Diameter [mm]
Cross sectional area [mm2]
n=
Diameter [mm]
Cross sectional area [mm2]
n=
Diameter [mm]
Cross sectional area [mm2]
n=
Tibia (prox.) Metaphyseal
9.1 ± 1.6
66.5 ± 11.2
20
10.6 ± 2.2
88.2 ± 23.3
24
12.0 ± 2.4
111.6 ± 29.5
16
Humerus (prox.) metaphyseal
9.4 ± 1.2*
67.7 ± 11.7
28
11.2 ± 0.9
85.5 ± 14.5
24
12.1 ± 1.8
113.5 ± 19.7
28
Femur (distal) metaphyseal
8.4 ± 1.7*
57.7 ± 11.9
20
10.2 ± 2.5
85.6 ± 19.3
24
11.9 ± 3.4
120.6 ± 28.2
24
ANOVA
*p < 0.05 n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
Tibia diaphyseal
5.0 ± 0.6
19.8 ± 2.6
20
5.4 ± 0.7
23.9 ± 5.1
24
6.2 ± 0.8
32.2 ± 4.8
16
Tibia (distal) metaphyseal
6.4 ± 0.2
34.2 ± 2.2
20
7.9 ± 0.2
48.9 ± 2.6
24
9.4 ± 0.3
69.3 ± 3.3
16
ANOVA
p < 0.001
p < 0.001
p < 0.001
p < 0.001
p < 0.001
p < 0.001
Predetermined locations as shown in Fig. 1, Fig. 2, Fig. 3 (MEAN ± S.D.). GA = gestational age, ∗ = p < 0.05, ns = not significant, n = number of samples.
When comparing the proximal and distal tibia, proximal were significantly higher in diameter (12.0 ± 2.4 vs. 9.4 ± 0.3 mm) and cross sectional area (111.6 ± 29.5 vs. 69.3 ± 3.3 mm2). Also, as seen in the CT examination (Fig. 3b), the tibial bone is transverse-oval shaped in its proximal head. The volume of the cross-sectional tibial area (Fig. S2) between group b and c did not increase significantly compared to femur (Fig. S4) and humerus (Fig. S6).
The proximal femoral head region has the largest diameter of all measured bones with an egg-shaped formation caused by the trochanter major and its head is built of cartilage with an extensive joint capsule (Fig. S7). In detail, diameter and cross-sectional area differs between proximal and distal femur in term newborn cadavers. The proximal diameter (mean ± SD: 15.0 ± 2.6 vs. 11.9 ± 3.4 mm) and sectional area (mean ± SD: 171.9 ± 26.5 vs. 120.6 ± 28.2 mm2) is significantly larger (p < 0.001) than the distal end of the femur (Figs. S3+S4). In contrast to the tibia, the femur diaphysis is relatively smaller compared to its metaphysis’.
The cross-sectional area of the proximal humerus is nearly circular in contrast to its distal end, which is, due to the developing trochlea, oval-shaped (Fig. S8). The proximal diameter (mean ± SD: 12.1.0 ± 1.8 vs. 8.9 ± 2.8 mm) and sectional area (mean ± SD: 113.5 ± 19.7 vs. 64.3 ± 16.4 mm2) is significantly larger (p < 0.001) than the distal end of the humerus (Figs. S5+S6) in term newborn cadavers.
All investigated metaphyseal areas lack a clearly enclosed marrow cavity (Fig. 1, Fig. 2, Fig. 3). Instead, a trabecular bony structure was found (spongiosa) making it clear that a marrow aspiration test with the inserted IO needle will fail in this age group.
Infusion of contrast medium into the distal femoral end (Video S1 + S2) or the proximal humerus head (Video S3) demonstrated a rapid conversion (3–7 s) into the central venous drainage system
Detailed examination results of each single area are also given as supplementary data.
Discussion
Our cadaver study of human preterm and term newborns present alternative IO access areas if umbilical or intravenous catheterisation has failed in cases of resuscitation in these special groups of patients. Since the proximal humerus head and the distal femoral end offer the same diameters and circumferences as the preferred tibial site, it thus appears logical to investigate these possible insertion areas for IO access.
It has already been shown that, if the IO needle is inserted in the supero-lateral aspect of the humerus in supine position in neonatal cadavers, this area is a safe alternative access point for emergency infusions
without the risk of damage to vulnerable anatomical structures. We verify these results by showing the unhindered drainage of the contrast medium with adequate osseous proportions. The epiphyseal growth plate separates the proximal head of the humerus from the shaft and is located distal to the tuberculus majus of the humerus. It is an advantage that the trapezoidal design of the epiphyseal plate in infants and growing children is more or less ‘wavy’ in newborns, which offers a measure of safety in case of a misplaced needle.
The osseous humeral head consists of spongiosa only, without any detectable marrow cavity. Drainage of the injected contrast medium can easily take place without any considerable delay. Sometimes, as seen in a few video sequences of injected contrast medium into the humeral head, the injected fluid formed a thin layer outside the bone just around the injection site (Video S3) indicated as extravasation without influencing further drainage. Overall, no fractures were seen in all bones investigated. The selected lateral humeral head area is anatomically easily to puncture, and the diameter is wide enough (12.1 ± 1.8 mm) for the 15G-IO needle to be inserted.
Although the proximal femoral head region has the largest diameter of all measured bones in term newborns (15.0 ± 2.6 mm) there are few landmarks for a safe and secure insertion of the IO needle due to extensive joint capsule and accompanying blood vessels and nerves.
By contrast, the distal femoral end seems appropriate for IO access and has been used several times as emergency IO access even in term newborns and infants.
A comparison of proximal tibia, distal femur, and proximal humerus infusion rates using the EZ-IO intraosseous device on the adult swine (Sus scrofa) model.
With a diameter of 11.9 ± 3.4 mm the distal femur end offers enough space for a 15G IO needle. We also demonstrate that the given IO contrast medium reaches the central venous vessels (Vv.femorales) within seconds.
The corticalis diameter of the tibia differs between its larger proximal and narrower distal ends (mean ± SD: 12.0 ± 2.4 vs. 9.4 ± 0.3 mm). Our results reflect the radiological findings of Capobianco et al.
with comparable cortical thickness of the proximal tibia. In contrast to recommendations for using the distal tibial area for IO access (proximal to the medial malleolus in the midline),
the diaphyseal diameter and cross sectional area is significantly smaller than the recommended proximal side and seems unsuitable as i.o access in newborns. So it becomes evident that the recommended proximal IO puncture area is distal to the tibial tuberosity but not too centrally within the diaphysis.
It is of interest that in all metaphyseal areas (humerus, femur, tibia) investigated a real bone marrow cavity was missing. Instead, a trabecular bone structure was found which clearly indicated that a marrow aspiration test of the inserted IO needle would often fail in this age group. This could positively affect the stability of the inserted needle as, the smaller the marrow cavity, the less the dislodgment of the IO needle through tilting movements might be. By contrast, it is also possible that there could be some as yet unexplored explanation for the higher complication rate during IO infusion in neonatal patients
: Since the corticalis and total diameter are restricted, IO needle dislodgment into soft tissue (e.g. muscle and fascia) during pressured infusion (up to 300mmH20) seems not unlikely, as shown in experimental animal studies.
A comparison of proximal tibia, distal femur, and proximal humerus infusion rates using the EZ-IO intraosseous device on the adult swine (Sus scrofa) model.
This would explain the initial successful IO infusion followed by the subsequent dislocation of the needle. With regard to a suitable and secure emergency IO access, further anatomical studies are needed to determine the accuracy and safety of new alternative puncture areas as described in our study.
It seems speculative whether the reported high success rates (more than 80%) for IO devices in young children are in fact true.
By contrast, rates of malposition of IO needles in paediatric cadavers via post-mortem computed tomography showed relatively high malposition rates for ION devices.
As shown in our study, the cortex and total circumference in newborn cadavers only just provide enough space for a commercial IO-needle inserted successfully, verified by X-ray examination.
Our CT-examination results are in agreement with study data from Rodríguez et al., who evaluated the transverse bone growth and cortical bone mass of tibia, femur and humerus to investigate the bone-modelling process during pregnancy.
it appears extremely important to carry out and practise appropriate simulations in true to scale anatomical specimens and to avoid any false sense of security. Currently, simulation training sessions often use unrealistic equipment (e.g. plastic IO doll legs, a variety of chicken bones, pork ribs etc.).
underlines further problems with currently weight-based needle size recommendations. Our findings also demonstrate that newborn bone size and form differ between preterm and term newborns.
Our results may be limited by the small sample size, dictated by the available objects of study. In addition, for better comparability between ours and other studies, we used the mean value of diameters. Our conclusion may also be limited by the postmortem analysis and the preparation of formalin-fixed stillborns, which led to changes in the composition of the organs. The bone marrow aspiration test failed, due to the lack of a marrow cavity. As there is no documented information about any known congenital or genetic conditions, this fact could potentially affect the bony structure or anatomy of the cadavers used.
Conclusion
There might be alternative IO areas which may lead to extra puncture sites in neonates. Further studies are required to assess the optimal position and device.
A comparison of proximal tibia, distal femur, and proximal humerus infusion rates using the EZ-IO intraosseous device on the adult swine (Sus scrofa) model.
00144-1&id=gr1.jpg){kind=link}
00144-1&id=gr2.jpg){kind=link}
00144-1&id=gr3.jpg){kind=link}
00144-1&id=mmc2.jpg){kind=link}
00144-1&id=mmc3.jpg){kind=link}
00144-1&id=mmc4.jpg){kind=link}
00144-1&id=mmc5.jpg){kind=link}
00144-1&id=mmc6.jpg){kind=link}
00144-1&id=mmc7.jpg){kind=link}
00144-1&id=mmc8.jpg){kind=link}
00144-1&id=mmc9.jpg){kind=link}