Which endotracheal tube location minimises the device‐related pressure ulcer risk: The centre or a corner of the mouth?

Abstract

The use of an endotracheal tube (ETT), which is required for any mechanical ventilation procedure, involves an inherent risk for facial skin, lip, and mucosal pressure ulcers. The ETT is one of the most common devices associated with medical device‐related pressure ulcers (MDRPUs) among surgical and intensive care unit patients. In the present work, we investigated, for the first time in the literature, the biomechanical effects of the presence and positioning of an ETT in the mouth on lip, mucosal and surrounding facial skin loads. Using two anatomically realistic finite element model variants, two ETT locations were simulated and compared, at the centre versus the corner of the mouth. Our study shows that a central location of the ETT inflicted greater lip and mucosal stress values, but a corner location caused a more widespread and diffused lip, mucosal and facial skin stress exposure. Accordingly, we cannot recommend a “safer” location for ETTs in the mouth; additional preventative measures such as dedicated dressing materials or special cushioning pads applied prophylactically, should be developed to protect from MDRPUs associated with ETT usage. The present modelling framework can be used to study the biomechanical efficacy of such protective technologies, and can therefore aid in the prevention of ETT‐caused MDRPUs.

1. INTRODUCTION

A pressure ulcer (PU) is defined as the localised tissue damage, usually over a bony prominence, caused by the contact of a body site with an interfacing surface or material, resulting in sustained tissue deformations, which trigger the injury spiral.1, 2, 3 Medical device‐related PUs (MDRPUs) are a specific sub‐type of PUs, and are defined as PUs where the injury is known to have been caused by a device applied for a diagnostic or therapeutic purpose. Typically, MDRPUs conform to the pattern or shape of the applied device and the contours of tissue damage match those of the device‐body contact area, or the pattern of the device‐applied forces that caused the sustained tissue deformations.1

The latest published epidemiological data indicate that incidence and prevalence of MDRPUs are 12% and 10%, respectively.4 Devices causing MDRPUs are wide‐ranging, and include endotracheal tubes (ETTs), nasogastric tubes (NGTs), oxygen masks, urinary catheters and other tubing, pulse oximeters, cervical collars, electrodes and wiring, orthopaedic fixations, and even bedpans.5, 6 Facial and mucosal PUs associated with ventilation or feeding equipment and tubing are a considerable portion of the reported MDRPUs. These PUs are, by definition, hospital‐acquired injuries that expose the healthcare professionals involved in the planning and delivery of care as well as the medical institute to litigation acts and will cause increase of insurance premia. From a patient's perspective, facial PUs are also known to substantially compromise the quality of life and cause long‐term psychological effects, such as those related to body image.

An ETT, which is required for any mechanical ventilation procedure, including surgery and in most intensive care unit (ICU) patients, involves an inherent risk for facial skin, lip, and mucosal PUs. Recent studies have shown that an ETT, in particular, is one of the most common devices associated with MDRPUs among surgical and ICU patients. The reported ICU ETT‐caused PU incidence rates range from 7% to as high as 45%, which is clearly unacceptable.5, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 The time points at which PUs become clinically visible post‐intubation also vary, being 2–13 days.9, 14, 16 To prevent accidental extubation or adverse ETT movements, tubes are typically fixed to position using commercial or non‐commercial means of securement, which may use attachment devices or holders and adhesive tapes or strings, respectively.14, 16, 17 Commercial ETT holders restrict potential ETT movements but may exert considerable pressure and shear on facial skin.16, 17 Protocols for prevention of ETT‐caused PUs include risk assessment, routine ETT repositioning in the mouth (e.g. from the centre to a corner of the mouth and vice versa) and cushioning skin areas in contact with the ETT. In practice, that cushioning is often being achieved by padding the tube, facial tissues and lips with dressings or dressing cuts (chosen by nurses according to their clinical judgement and experience), to redistribute the loads and absorb moisture from skin/lip areas in contact with the ETT.1, 6

To mitigate the risk of an ETT‐inflicted injury, manufacturers recommend that an ETT should not be secured to place (i.e. by means of a tube‐fastening device) in patients who have facial oedema, lip oedema, or protruding teeth. Some manufacturers further recommend repositioning the device every 2 hours,18, 19 but as noted from the epidemiological data reviewed above, it is unlikely that ETTs are repositioned that often in the majority of ICUs, and certainly not during surgery, when the stability of ventilation and anaesthesia are critical. In fact, a commonly reported clinical practice is to reposition ETTs as little as once every third day.12 It is almost needless to mention that there is no biomechanical, medical, or clinical research or any other specific evidence to support the aforementioned repositioning frequencies (which, at best, in the case of the 2‐hour manufacturer recommendations are likely an extrapolation of the common clinical practice of 2‐hour cycle of repositioning patients in bed to prevent bodyweight‐related PUs). Moreover, it has been reported that increasing the frequency of ETT repositioning from 12 hours to 4 hours decreased the prevalence of ETT‐related PUs from 16% to 10%. Nonetheless, this reduction was not statistically significant, which likely indicates that ETT repositioning per se is insufficient for effective prevention of ETT‐caused PUs.20

The ETT is a nearly rigid device with respect to soft tissue stiffness, which imposes localised forces and deformations as well as concentrated stresses in the distorted facial, lip, and mucosal tissues of patients who are anaesthetised, sedated, or unconscious. Those MDRPUs that are specifically associated with the use of ETTs are a considerable portion of the overall pool of reported MDRPUs in ICU patients (as reflected by the literature reviewed above). ETTs are further shortlisted among the dangerous devices from a MDRPU‐risk perspective, because mechanical ventilation is required in all patients operated under full anaesthesia, and in the majority of ICU patients. Overall, scientific and medical knowledge regarding the prevention of ETT‐inflicted injuries is extremely poor, despite that ETTs are used for anaesthesia and ventilation since the second half of the 19th century.21 The lack of knowledge and focus on ETT‐caused injuries may be associated with the fact that discovery of the contemporary aetiology of PUs and MDRPUs, and, specifically, the scientific understanding that their direct cause is cell and tissue exposure to sustained deformations (as opposed to the impaired perfusion paradigm) has only been made in the last two decades.2, 3 Clearly, patients requiring mechanical ventilation and who depend on their ETT for survival are often at a critical state, and, hence, prevention of MDRPUs is, naturally, not counted among the highest medical priorities. Despite that, the potential long‐term consequences of scarred lips and face should be considered, and from an organisational perspective, the costs of additional treatment and risk of patient‐initiated litigation should also be taken into account.

Here, we report the first bioengineering study ever conducted concerning the risk for ETT‐caused PUs, which provide novel information on facial, lip, and mucosal tissue exposure to loading in this specific MDRPU scenario. Finite element (FE) modelling, which is a powerful bioengineering method for evaluating tissue loads, including where tissues interact with objects and devices (such as the face‐lip‐ETT interactions), was selected as our research methodology. Our research question has been fundamental, but very practical from a clinical perspective: Which ETT location is biomechanically preferred: At the corner of the mouth or centrally? Answering this question first is also a necessary step towards addressing any ETT‐repositioning‐related questions that may follow. Our findings visually demonstrate the risk to facial, lip, and mucosal tissue viability associated with ETT application, and shed light on tissue exposure to mechanical loads applied by this specific medical device.

2. METHODS

To investigate the biomechanical effects of the presence and positioning of an ETT in the mouth on lip, mucosal, and surrounding facial skin loads, two anatomically realistic finite element (FE) model variants were developed, representing two possible locations of the ETT—at the centre versus the right side of the mouth. A downward/horizontal displacement of the ETT towards the lips was applied based on real‐world ETT curvature data determined from photographs of ventilated patients, as detailed below.

2.1. Geometry

The three‐dimensional model of the mouth region was virtually cut from a model of the entire head, which has been developed and published by our group.22, 23, 24, 25 Briefly, using the visible human (male) project® image database,26 transverse slices of the head were imported to the Scan‐IP module of the Simpleware® software package27 for segmentation of the hard and soft tissues of the head. We separately segmented the skull, mandible (including the teeth), skin, fat, and lips as well as mucosal tissues. The mouth was then virtually opened using the Scan‐IP module of Simpleware®27 to locate the ETT in contact with the lips as in real‐world clinical scenarios. The ETT segment was built using the PreView module of FEBio (version 1.19, University of Utah, Salt Lake City, Utah).28, 29 The volume of the mouth region was 7.7 cm × 10.6 cm × 8.4 cm. The inner and outer diameters of the ETT segment were 8.5 mm and 12 mm, respectively, which are characteristic ETT dimensions. The length of the ETT segment was 20 and 15 mm for the tube at the centre and right side of the mouth, respectively, to consider a long‐enough tube segment for minimising boundary effects according to the Saint‐Venant's principle (Figure 1).

Figure 1.

Computational finite element modelling of the mouth region with an endotracheal tube (ETT), the applied loading conditions on the ETT and the volume of interest (VOI) for calculating volumetric exposures of soft tissues to stresses induced by the ETT for the following conditions: A, An ETT located at the centre of the mouth, B, An ETT positioned at the right side of the mouth

2.2. Mechanical properties

Constitutive laws and mechanical properties of all the tissues included in the head model, that is, the facial skin and fat, lips and mucosa, skull, and mandible, and also, those of the ETT were adopted from the literature. The mechanical properties used in the present modelling work with respective references to the literature are listed in Table 1. The mandible bone and ETT were assumed to be isotropic elastic materials. The lips and mucosa were assumed to be a non‐linear isotropic material, with their large deformation behaviour described using an uncoupled Neo‐Hookean constitutive model with the following strain energy density function W:

$$W=\frac{G_{\mathit{ins}}}{2}\left({\lambda }_1^2+{\lambda }_2^2+{\lambda }_3^2-3\right)+\frac{1}{2}K{\left(\ln J\right)}^2$$ (1)

where G _ins is the instantaneous shear modulus, λ _i (i = 1, 2, 3) are the principal stretch ratios, K is the bulk modulus, and J = det(F), where F is the deformation gradient tensor. Facial skin and fat tissues were represented using the Mooney‐Rivlin constitutive model:

$$W=C_1\left(I_1-3\right)+C_2\left(I_2-3\right)-2\left(C_1+2C_2\right)\mathit{lnJ}+\frac{\lambda }{2}{\left(\mathit{lnJ}\right)}^2$$ (2)

where I ₁ and I ₂ are the first and second invariants of the right Cauchy‐Green deformation tensor, respectively, and J is the determinant of the deformation gradient tensor. The lip and mucosal, skin and fat tissues were assumed to be nearly incompressible (ie, Poisson's ratio of 0.495).

Table 1.

Mechanical properties of the model components, including hard and soft head tissues as well as the endotracheal tube (ETT), and characteristics of the finite element mesh

Model component	Elastic modulus [MPa]	Shear modulus [MPa]	Bulk modulus [MPa]	Poisson's ratio	Number of elements
Skina	‐	0.0319	3.17937	0.495	30 969 to 31 115
Fatb	‐	0.000286	0.0285	0.495	89 671 to 91 273
Lips and mucosac , d	0.0337	‐	‐	0.495	11 396 to 14 705
Skulle	6483.59	‐	‐	0.2	33 598 to 33 787
Mandiblef	5000	‐	‐	0.23	16 654 to 18 057
ETTg	1.12	‐	‐	0.4	22 654 to 27 717

^aLinder‐Ganz et al.30

^bSopher et al.31

^cLuboz et al.32

^dGefen and Haberman.33

^eMoore et al.34

^fHorgan et al.35

^gJones et al.36

2.3. Boundary and material transition conditions

To stimulate the effects of a contacting ETT on tissue deformations and loads, downward and horizontal displacements of 3.4 and 3.54 mm were applied to the ETT at the centre and at the right side of the mouth, respectively. These displacements were determined based on a mid‐range value of lower‐lip indentations by ETTs, which we have evaluated to be in the range of 2 to 6 mm. The aforementioned range of lip indentations has been measured from multiple digital photographs of clinical cases of adult intubated patients. The comparison between the simulation cases was conducted under the same reaction force acting from the lips, 0.6 N (differences up to 3% from the target reaction force were allowed). The back surface of the mouth region was fixed for all translations and rotations. Frictional sliding was defined between the lips and the ETT; the coefficient of friction for these interfaces was set as 1 in all the simulations (indicating a relatively high friction at the interfaces, eg, because of wetness37, 38, 39).

2.4. Numerical method

Four‐node linear tetrahedral elements were used in all model components. The mouth region was meshed using the Scan‐IP module of Simpleware®,27 with finer meshes used in specific tissue regions of interest associated with the ETT contact sites and ETT‐induced MDRPUs, that is, lip and mucosal tissues near the ETT segment. Meshing of the ETT segment was performed using the Preview module of FEBio.28, 29 Numbers of elements in each of the model components are specified in Table 1.

All simulations were conducted using the PreView module of FEBio (version 1.19), analysed using the Pardiso linear solver of FEBio (http://mrl.sci.utah.edu/software/febio) (version 2.5.0), and post‐processed using PostView of FEBio (version 2.3.2).28, 29 The runtimes of the simulations were up to 2 hours, using a 64‐bit Windows 7‐based workstation with an Intel Core i5‐3470 3.20 GHz CPU and 8 GB of RAM.

2.5. Biomechanical outcome measures

In each model variant, the effective and maximal shear stresses in the lips, mucosal tissues, and facial skin were calculated and compared, separately for each tissue type. Additionally, volumetric exposures of these parameters were plotted and compared across the two model variants, as per our published methodology to quantify device‐tissue interactions in the context of PUs and MDRPUs in particular.22, 25, 40, 41, 42, 43, 44 In order to allow systematic comparison between the model variants, all data analyses were performed for the soft tissues elements near the ETT segment with effective stress values greater than 3 kPa, which has been set as the threshold for defining the size of the volume of interest (VOI) (Figure 1).

3. RESULTS

Effective and maximal shear stress distributions in the soft tissues of the mouth region, when the ETT has been located at either the centre or right side of the mouth, are shown in Figure 2. Application of the ETT at the centre of the mouth is associated with substantially lower effective and maximal shear stresses in the lip, mucosal and skin tissues with respect to its application at the right mouth side. Specifically, for the central location of the ETT, both the average effective stress and average maximal shear stress decreased by 7% in lip and mucosal tissues, and by 18.5% in skin, relatively to the case where the ETT was positioned at the right mouth side. Comparisons of the cumulative volumetric exposures of lip, mucosal and skin tissues to effective and maximal shear stresses for the central (VOI1) and right side (VOI2) locations of the ETT are shown in Figure 3. Again, this volumetric exposure data demonstrate that a central ETT location is overall advantageous compared with a side location, although when focusing on just the high stress domains, stress exposure differences between the locations become less distinct.

Figure 2.

Effective (left column) and maximal shear (right column) stress distributions in the soft tissues of the mouth region for A, An endotracheal tube (ETT) located at the centre of the mouth, B, An ETT positioned at the right side of the mouth

Figure 3.

Cumulative volumetric exposures of the, A, lips and mucosal, and, B, skin tissues to effective (left column) and maximal shear (right column) stresses, in the corresponding volume of interest (VOI) when the endotracheal tube is located at the centre of the mouth (VOI1 as depicted in Figure 1) and right side of the mouth (VOI2)

4. DISCUSSION

In the present work, we investigated, for the first time in the literature, the biomechanical effects of the presence and positioning of an ETT in the mouth on lip, mucosal, and surrounding facial skin loads. Using two anatomically realistic FE model variants, we compared the biomechanical loading states of the above soft tissues for two possible ETT locations, the centre versus the right corner of the mouth. Tissue stress concentrations at the contact regions between the ETT and lower lip, which spread to deep lip, mucosal, and facial skin tissues, were observed for both the central and side ETT locations, but were more diffused for the corner position of the tube (Figures 2 and 3). Anatomically, the corner of the mouth has a more curved geometry compared with the central lower lip segment, which theoretically suggests that a side location of the ETT would be more prone to greater stress magnitudes and stress concentrations, particularly because it is known, from classic engineering mechanics, that small radii of curvature in a structure are associated with material stress concentrations. Hence, from a purely theoretical perspective, and without conducting the present analyses, it could have been reasonable to assume that the side location of the ETT would be more susceptible to an ETT‐caused MDRPU.45 Interestingly, however, the findings of our present study are counter‐intuitive. Our data showed that, in fact, for the high stress domain, the effective and maximal shear stresses in the lower lip and mucosal tissues were mildly greater for the centre position of the ETT, not the corner location (Figure 3A). A possible explanation for this phenomenon is the proximity of the lip and mucosal tissues at the centre of the mouth to the frontal teeth (which are nearly rigid elements compared with lip and mucosal tissue stiffness). That steep tissue stiffness gradient between the teeth and lip/mucosal tissues appears to intensify the lip and mucosal stresses when the ETT is deforming the central lower lip, and, essentially, squeezing it towards the frontal teeth. Another relevant biomechanical phenomenon is that the corner of the mouth can be seen as a concave surface which is interfacing with the convex ETT surface, and this should theoretically allow for a greater lip‐tube contact area, and therefore, reduced stress concentrations for the side ETT position. All of the above‐mentioned biomechanical factors and interactions play concurrent roles in influencing the stress states of tissues (for either the side or the centre tube positions), and the simulation data reported here show their effective contributions (at each ETT position), altogether.

Peko Cohen et al44 previously investigated the interactions between non‐invasive ventilation masks and facial soft tissues. Their study, which included both experimental measurements and computational modelling, also resulted stress values that were under 7 kPa, similarly to our presently reported maximal shear stress values applied by ETTs.44 Furthermore, our resultant difference in tissue stress states is also in agreement with the above published work concerning facial tissue stresses under oxygen masks,44 in terms of the stress intensity ranges. Nevertheless, it should be considered that in real‐world conditions, PU risk profiles of patients requiring intubation vary remarkably. Accordingly, the above differences may or may not be clinically significant, depending on the fragility of the facial tissues of the individual.46 This tissue fragility depends, in turn, on age, health status, and any chronic or acute conditions (that are either local or systemic). It specifically depends on the biomechanical tissue tolerances to loads, which are difficult to predict given the aforementioned large biological and pathophysiological variability across patients. The present work is the first study to ever describe the influence of ETT position on facial tissue stress states. We found that application of the ETT at the centre of the mouth decreased both the average effective stress and the average maximal shear stress, by 7% in lip and mucosal tissues, and by 18.5% in skin, in comparison to a side location. These are notable reductions of the MDRPU risk, which may provide vital protection for some patients by keeping them just below their individual tissue tolerance levels, but will not be sufficient for others, who are inherently or temporarily more fragile.46

In clinical practice, facial MDRPUs caused by ETTs have been reported either at the corner of the mouth or at central locations, for patients undergoing surgical procedures exceeding 4 hours or in mechanically ventilated patients.14, 16, 17, 47 It is difficult to determine, from the above‐cited clinical reports, if a central ETT location is epidemiologically more susceptible to MDRPUs than a side location, as some of these clinical datasets are contradicting in this regard. Published work does suggest that the method of device securement, time of exposure to the ETT,^* administration of vasopressors, and, clearly, background diseases and conditions all interact with the biomechanical effects described here, and will altogether influence and determine the clinical outcome in individual cases.20 Given these complex interactions, which are specific to the individual, we cannot recommend a “safer” ETT location in the mouth, based on our present modelling work. Taken together with the epidemiological data reviewed in the Introduction section, which suggests that cyclic centre‐corner‐centre ETT repositioning is an insufficient measure, our findings strongly indicate that new preventative bioengineering technologies, such as dressing materials or special cushioning pads applied prophylactically, should be developed to protect from ETT‐inflicted MDRPUs. Our results specifically suggest that from a bioengineering perspective, a central location is not necessarily advantageous over a side location, despite the “naïve” engineering experience associating stress concentrations with curved (ie, corner‐of‐the‐mouth) sites. The mechanical conditions in tissues differ in nature between the two studied ETT locations (Figures 2 and 3), but no location can be considered “safer” than the other. This is because generally, a central location inflicted greater lip and mucosal tissue stress values, but a corner location caused a more widespread and diffused (but lower‐magnitude) lip, mucosal, and facial skin stress exposure (Figures 2 and 3).

Limitations of our work are that (a) it used a specific head and face geometry, which does not necessarily represent all facial (mouth) structural features of ventilated patients; (b) the present results cannot be extrapolated to paediatrics where facial anatomy and relative dimensions of anatomical features are inherently different from those of adults; and (c) we did not consider the method of securement of the tube, nor did we account for potential padding or other means for tissue protection. Despite these limitations, this work is pioneering in PU research as it provides, for the first time, quantitative biophysical mapping of lip, mucosal and facial tissue load magnitudes and distributions caused by ETTs. This advancement, by itself, has unique educational value in the field, as the mouth‐ETT biomechanical interactions have never been studied or visualised before.

In closure, the present paper shows facial tissue stress states associated with ETT application, and should promote additional research, now focusing on how to improve these tissue‐loading conditions. In future work, our unique and original modelling framework can be used to study the biomechanical efficacy of protective technologies, improved ETT designs, securement methods, and different practices that account for the MDRPU risk and attempt to mitigate it.

Amrani G, Gefen A. Which endotracheal tube location minimises the device‐related pressure ulcer risk: The centre or a corner of the mouth? Int Wound J. 2020;17:268–276. 10.1111/iwj.13267