Modeling the impacts of assumptions and nonpulmonary factors on the performance and reliability of indices of oxygenation.

Assessment of oxygenation is fundamental to the care of patients. Numerous indices of oxygenation have been developed that entail variable degrees of invasiveness, complexity and physiologic underpinning. The clinical reliability of these indices has been questioned. This theoretical study uses a steady-state model of blood gas physiology to study the assumptions and nonpulmonary factors that have been hypothesized to impact index performance. A model derived from cardiac and pulmonary Fick expressions was used to calculate the effects of the physiological parameters-shunt, dead space, cardiac output, ventilation, oxygen extraction, carbon dioxide elimination, hematocrit, temperature and base excess-on predicted arterial, mixed-venous and post-capillary oxygen contents and arterial and alveolar oxygen and carbon dioxide partial pressures. Values of these parameters were determined over a range of shunt from 0 to 50% and then used to calculate (1) estimated shunt with the shunt equation and FShunt, and (2) the alveolar-arterial partial pressure of oxygen difference (A-a [Formula: see text] gradient), and the arterial partial pressure of oxygen to fraction of inspired oxygen (Pa/Fi) ratio. Calculations were performed either treating parameters as fixed (assuming several values) or as random variables. Assumptions of constant arterio-venous oxygen content and of alveolar and arterial partial pressures of carbon dioxide being equal were shown to fail in certain settings where shunt and physiologic parameters varied. These effects manifested as calculated indices either over or under-estimating actual shunt by FShunt, or wide unpredictable variability (scatter) when correlating A-a [Formula: see text] gradient and Pa:Fi ratio to actual shunt. Cardiac output and oxygen extraction have noticeable impacts on all calculated indices. The results support the clinical observations that the performance of indices of oxygenation can vary with fraction of inspired oxygen and various nonpulmonary physiological factors that underly heterogeneity present in the clinical population.

Co2 Rebreathing Observed While Using a Bag-Mask Resuscitator With Integrated Manometer: A Case Report.

Bag-mask resuscitators with integrated manometry help reduce the risk of pulmonary injury during manual ventilation. All such devices must function as intended while preventing carbon dioxide rebreathing, as unintended hypercapnia can be harmful in critically ill patients. We describe a case of carbon dioxide rebreathing in a patient suspected of having a brain injury after blunt trauma who was manually ventilated with a widely available bag-mask resuscitator with integrated manometry after emergent intubation. This case highlights the importance of vigilant monitoring of end-tidal carbon dioxide and appropriate troubleshooting and investigation of unexplained findings to mitigate and prevent adverse patient outcomes.

Can variable practice habits and injection port dead-volume put patients at risk?

Injection ports used to administer medications and draw blood samples have inherent dead-volume. This volume can potentially lead to inadvertent drug administration, contribute to erroneous laboratory values by dilution of blood samples, and increase the risk of vascular air embolism. We sought to characterize provider practice in management of intravenous (IV) and arterial lines and measure dead-volumes of various injection ports. A survey was circulated to anesthesiology physicians and nurses to determine practice habits when administering medications and drawing blood samples. Dead-volume of one and four-way injection ports was determined by injecting methylene blue to simulate medication administration or blood sample aspiration and using absorption spectroscopy to measure sample concentration. Among the 65 survey respondents, most (64.52%) increase mainstream flow rate to flush medication given by a 1-way injection port. When using 4-way stopcocks, 56.45% flush through the same injection site. To obtain a sample from an arterial line, 67.74% draw back blood and collect the sample from the same 4-way stopcock; 32.26% use a different stopcock. Mean (SD) dead-volume in microliters ranged from 0.1 (0.0) to 5.6 (1.0) in 1-way injection ports and from 54.1 (2.8) to 126.5 (8.3) in 4-way injection ports. The practices of our providers when giving medications and drawing blood samples are variable. The dead-volume associated with injection ports used at our institution may be clinically significant, increasing errors in medication delivery and laboratory analysis.

Auditory Icon Alarms Are More Accurately and Quickly Identified than Current Standard Melodic Alarms in a Simulated Clinical Setting.

BACKGROUND: Current standard audible medical alarms are difficult to learn and distinguish from one another. Auditory icons represent a new type of alarm that has been shown to be easier to learn and identify in laboratory settings by lay subjects. In this study, we test the hypothesis that icon alarms are easier to learn and identify than standard alarms by anesthesia providers in a simulated clinical setting. METHODS: Twenty anesthesia providers were assigned to standard or icon groups. Experiments were conducted in a simulated intensive care unit. After a brief group-specific alarm orientation, subjects identified patient-associated alarm sounds during the simulation and logged responses via a tablet computer. Each subject participated in the simulation twice and was exposed to 32 alarm annunciations. Primary outcome measures were response accuracy and response times. Secondary outcomes included assessments of perceived fatigue and task load. RESULTS: Overall accuracy rate in the standard alarm group was 43% (mean) and in the icon group was 88% (mean). Subjects in the icon group were 26.1 (odds ratio [98.75% CI, 8.4 to 81.5; P < 0.001]) times more likely to correctly identify an alarm. Response times in the icon group were shorter than in the standard alarm group (12 vs. 15 s, difference 3 s [98.75% CI ,1 to 5; P < 0.001]). CONCLUSIONS: Under our simulated conditions, anesthesia providers more correctly and quickly identified icon alarms than standard alarms. Subjects were more likely to perceive higher fatigue and task load when using current standard alarms than icon alarms.

Low End-Tidal Carbon Dioxide at the Onset of Emergent Trauma Surgery Is Associated With Nonsurvival: A Case Series.

BACKGROUND: End-tidal carbon dioxide (EtCO2) is a valuable marker of the return of adequate circulation after cardiac arrest due to medical causes. Previously, the prognostic value of capnography in trauma has been studied among limited populations in prehospital and emergency department settings. We aimed to investigate the relationship between early intraoperative EtCO2 and nonsurvival of patients undergoing emergency surgery at a level 1 academic trauma center as a case series. If there is a threshold below which survival was extremely unlikely, it might be useful in guiding decision-making in the early termination of futile resuscitative efforts. METHODS: After institutional review board approval, a data set was created to investigate the relationship between EtCO2 values at the onset of emergent trauma surgery and nonsurvival. Patients who were admitted and transferred to the operating room (OR) directly from a resuscitation bay were identified using the Ryder Center trauma registry (October 1, 2013, to June 30, 2016). Electronic records from the hospital's anesthesia information management system were queried to identify the matching anesthesia records. The maximum EtCO2 values within 5 and 10 minutes of the onset of mechanical ventilation in the OR were determined for patients undergoing general anesthesia with mechanical ventilation. Patients were divided into 2 groups: those who were discharged from the hospital alive (survivors) and those who died in the hospital before discharge (nonsurvivors). The threshold EtCO2 giving a positive predictive value of 100% for in-hospital mortality was determined from a graphical analysis of the data. Association of determined threshold and mortality was analyzed using the 2-tailed Fisher exact test. RESULTS: There were 1135 patients who met the inclusion criteria. Within the first 5 minutes of the onset of mechanical ventilation in the OR, if the maximum EtCO2 value was ≤20 mm Hg, hospital mortality was 100% (21/21, 95% binomial confidence interval, 83.2%-100%). CONCLUSIONS: A maximum EtCO2 ≤20 mm Hg within 5 minutes of the onset of mechanical ventilation in the OR may be useful in decision-making related to the termination of resuscitative efforts during emergent trauma surgery. However, a large-scale study is needed to establish the statistical reliability of this finding before potential adoption.

Factors Affecting Acoustics and Speech Intelligibility in the Operating Room: Size Matters.

INTRODUCTION: Noise in health care settings has increased since 1960 and represents a significant source of dissatisfaction among staff and patients and risk to patient safety. Operating rooms (ORs) in which effective communication is crucial are particularly noisy. Speech intelligibility is impacted by noise, room architecture, and acoustics. For example, sound reverberation time (RT60) increases with room size, which can negatively impact intelligibility, while room objects are hypothesized to have the opposite effect. We explored these relationships by investigating room construction and acoustics of the surgical suites at our institution. METHODS: We studied our ORs during times of nonuse. Room dimensions were measured to calculate room volumes (VR). Room content was assessed by estimating size and assigning items into 5 volume categories to arrive at an adjusted room content volume (VC) metric. Psychoacoustic analyses were performed by playing sweep tones from a speaker and recording the impulse responses (ie, resulting sound fields) from 3 locations in each room. The recordings were used to calculate 6 psychoacoustic indices of intelligibility. Multiple linear regression was performed using VR and VC as predictor variables and each intelligibility index as an outcome variable. RESULTS: A total of 40 ORs were studied. The surgical suites were characterized by a large degree of construction and surface finish heterogeneity and varied in size from 71.2 to 196.4 m (average VR = 131.1 [34.2] m). An insignificant correlation was observed between VR and VC (Pearson correlation = 0.223, P = .166). Multiple linear regression model fits and ß coefficients for VR were highly significant for each of the intelligibility indices and were best for RT60 (R = 0.666, F(2, 37) = 39.9, P < .0001). For Dmax (maximum distance where there is <15% loss of consonant articulation), both VR and VC ß coefficients were significant. For RT60 and Dmax, after controlling for VC, partial correlations were 0.825 (P < .0001) and 0.718 (P < .0001), respectively, while after controlling for VR, partial correlations were -0.322 (P = .169) and 0.381 (P < .05), respectively. CONCLUSIONS: Our results suggest that the size and contents of an OR can predict a range of psychoacoustic indices of speech intelligibility. Specifically, increasing OR size correlated with worse speech intelligibility, while increasing amounts of OR contents correlated with improved speech intelligibility. This study provides valuable descriptive data and a predictive method for identifying existing ORs that may benefit from acoustic modifiers (eg, sound absorption panels). Additionally, it suggests that room dimensions and projected clinical use should be considered during the design phase of OR suites to optimize acoustic performance.

Classifying Alarms: Seeking Durability, Credibility, Consistency, and Simplicity.

Alongside the development and testing of new audible alarms intended to support International Electrotechnical Commission 60601-1-8, a global standard concerned with alarm safety, the categories of risk that the standard denotes require further thought and possible updating. In this article, we revisit the origins of the categories covered by the standard. These categories were based on the ways that tissue damage can be caused. We consider these categories from the varied professional perspectives of the authors: human factors, semiotics, clinical practice, and the patient or family (layperson). We conclude that while the categories possess many clinically applicable and defensible features from our range of perspectives, the advances in alarm design now available may allow a more flexible approach. We present a three-tier system with superordinate, basic, and subordinate levels that fit both within the thinking embodied in the current standard and possible new developments.

Development and Testing of Screen-Based and Psychometric Instruments for Assessing Resident Performance in an Operating Room Simulator.

Introduction. Medical simulators are used for assessing clinical skills and increasingly for testing hypotheses. We developed and tested an approach for assessing performance in anesthesia residents using screen-based simulation that ensures expert raters remain blinded to subject identity and experimental condition. Methods. Twenty anesthesia residents managed emergencies in an operating room simulator by logging actions through a custom graphical user interface. Two expert raters rated performance based on these entries using custom Global Rating Scale (GRS) and Crisis Management Checklist (CMC) instruments. Interrater reliability was measured by calculating intraclass correlation coefficients (ICC), and internal consistency of the instruments was assessed with Cronbach's alpha. Agreement between GRS and CMC was measured using Spearman rank correlation (SRC). Results. Interrater agreement (GRS: ICC = 0.825, CMC: ICC = 0.878) and internal consistency (GRS: alpha = 0.838, CMC: alpha = 0.886) were good for both instruments. Subscale analysis indicated that several instrument items can be discarded. GRS and CMC scores were highly correlated (SRC = 0.948). Conclusions. In this pilot study, we demonstrated that screen-based simulation can allow blinded assessment of performance. GRS and CMC instruments demonstrated good rater agreement and internal consistency. We plan to further test construct validity of our instruments by measuring performance in our simulator as a function of training level.

Intraoperative Autotriggered Pressure Support Ventilation Resistant to Increased Flow Trigger Threshold.

Oscillations from cardiac pulsations are normally transmitted to mediastinal structures without any consequence. Autotriggering (AT) of mechanical ventilation occurs when an inspiratory trigger, typically negative inspiratory flow in anesthesia ventilators, is met in the absence of patient effort. AT can lead to respiratory alkalosis, opioid overdose, prolonged mechanical ventilation, and lung hyperinflation. This entity has been reported in both critical care and operating room environments. Increasing the flow trigger usually resolves AT in all cases. We report a case of AT that failed to respond to increasing the flow trigger threshold to its maximal value on the GE Datex-Ohmeda Avance S5® anesthesia station.

Intraoperative Noise Increases Perceived Task Load and Fatigue in Anesthesiology Residents: A Simulation-Based Study.

BACKGROUND: Operating rooms are identified as being one of the noisiest of clinical environments, and intraoperative noise is associated with adverse effects on staff and patient safety. Simulation-based experiments would offer controllable and safe venues for investigating this noise problem. However, realistic simulation of the clinical auditory environment is rare in current simulators. Therefore, we retrofitted our operating room simulator to be able to produce immersive auditory simulations with the use of typical sound sources encountered during surgeries. Then, we tested the hypothesis that anesthesia residents would perceive greater task load and fatigue while being given simulated lunch breaks in noisy environments rather than in quiet ones. As a secondary objective, we proposed and tested the plausibility of a novel psychometric instrument for the assessment of stress. METHODS: In this simulation-based, randomized, repeated-measures, crossover study, 2 validated psychometric survey instruments, the NASA Task Load Index (NASA-TLX), composed of 6 items, and the Swedish Occupational Fatigue Inventory (SOFI), composed of 5 items, were used to assess perceived task load and fatigue, respectively, in first-year anesthesia residents. Residents completed the psychometric instruments after being given lunch breaks in quiet and noisy intraoperative environments (soundscapes). The effects of soundscape grouping on the psychometric instruments and their comprising items were analyzed with a split-plot analysis. A model for a new psychometric instrument for measuring stress that combines the NASA-TLX and SOFI instruments was proposed, and a factor analysis was performed on the collected data to determine the model's plausibility. RESULTS: Twenty residents participated in this study. Multivariate analysis of variance showed an effect of soundscape grouping on the combined NASA-TLX and SOFI instrument items (P = 0.003) and the comparisons of univariate item reached significance for the NASA Temporal Demand item (P = 0.0004) and the SOFI Lack of Energy item (P = 0.001). Factor analysis extracted 4 factors, which were assigned the following construct names for model development: Psychological Task Load, Psychological Fatigue, Acute Physical Load, and Performance-Chronic Physical Load. Six of the 7 fit tests used in the partial confirmatory factor analysis were positive when we fitted the data to the proposed model, suggesting that further validation is warranted. CONCLUSIONS: This study provides evidence that noise during surgery can increase feelings of stress, as measured by perceived task load and fatigue levels, in anesthesiologists and adds to the growing literature pointing to an overall adverse impact of clinical noise on caregivers and patient safety. The psychometric model proposed in this study for assessing perceived stress is plausible based on factor analysis and will be useful for characterizing the impact of the clinical environment on subject stress levels in future investigations.

Simulating environmental and psychological acoustic factors of the operating room.

In this study, an operating room simulation environment was adapted to include quadraphonic speakers, which were used to recreate a composed clinical soundscape. To assess validity of the composed soundscape, several acoustic parameters of this simulated environment were acquired in the presence of alarms only, background noise only, or both. These parameters were also measured for comparison from size-matched operating rooms at Jackson Memorial Hospital. The parameters examined included sound level, reverberation time, and predictive metrics of speech intelligibility in quiet and noise. It was found that the sound levels and acoustic parameters were comparable between the simulated environment and the actual operating rooms. The impact of the background noise on the perception of medical alarms was then examined, and was found to have little impact on the audibility of the alarms. This study is a first in kind report of a comparison between the environmental and psychological acoustical parameters of a hospital simulation environment and actual operating rooms.

Pitfalls of hemodynamic monitoring in patients with trauma.

This article provides an update for the anesthesiology community on the mechanisms and limitations of common modalities used to assess the early hemodynamic status in patients with trauma. Figures are provided to illustrate important concepts through the use of computer simulation and real-world examples. This article is of value to anesthesiologists whose practice includes management of hemorrhagic shock.

PT-SAFE: a software tool for development and annunciation of medical audible alarms.

BACKGROUND: Recent reports by The Joint Commission as well as the Anesthesia Patient Safety Foundation have indicated that medical audible alarm effectiveness needs to be improved. Several recent studies have explored various approaches to improving the audible alarms, motivating the authors to develop real-time software capable of comparing such alarms. We sought to devise software that would allow for the development of a variety of audible alarm designs that could also integrate into existing operating room equipment configurations. The software is meant to be used as a tool for alarm researchers to quickly evaluate novel alarm designs. METHODS: A software tool was developed for the purpose of creating and annunciating audible alarms. The alarms consisted of annunciators that were mapped to vital sign data received from a patient monitor. An object-oriented approach to software design was used to create a tool that is flexible and modular at run-time, can annunciate wave-files from disk, and can be programmed with MATLAB by the user to create custom alarm algorithms. The software was tested in a simulated operating room to measure technical performance and to validate the time-to-annunciation against existing equipment alarms. RESULTS: The software tool showed efficacy in a simulated operating room environment by providing alarm annunciation in response to physiologic and ventilator signals generated by a human patient simulator, on average 6.2 seconds faster than existing equipment alarms. Performance analysis showed that the software was capable of supporting up to 15 audible alarms on a mid-grade laptop computer before audio dropouts occurred. CONCLUSIONS: These results suggest that this software tool provides a foundation for rapidly staging multiple audible alarm sets from the laboratory to a simulation environment for the purpose of evaluating novel alarm designs, thus producing valuable findings for medical audible alarm standardization.

Simultaneously extracted transient and steady-state evoked responses during general anesthesia: variability of different rates.

Unintended intraoperative awareness occurs in one to two individuals out of every one thousand treated with general anesthesia. Patients that experience intraoperative awareness have significant post-operative psychological sequelae. The ability to detect intraoperative awareness is currently suboptimal because the mechanism employed by anesthetic drugs to impair consciousness remains poorly understood. Studies have suggested that evoked potentials (EP) may be used to monitor the depth of anesthesia. Both transient and steady state responses can be simultaneously extracted using the Continuous Loop Averaging Deconvolution (CLAD) method with specially designed CLAD sequences. 20 Hz and 30 Hz jittered CLAD sequences in addition to 5 Hz isochronic and 40 Hz jittered CLAD sequences were applied in baseline awake and general anesthesia conditions. A qualitative method to assess the extracted EPs was developed in this study, termed Randomized Split Set Average (RSSA). The results showed that EPs extracted during general anesthesia require a greater number of sweeps to obtain a signal-to-noise ratio comparable to that observed in EPs extracted during the awake state. Therefore, the development of a real time or quasi real time EP monitoring system for anesthesia provides an increased challenge. The RSSA employed in this study is a useful method for assessing the signal quality of EP responses.

Influence of auditory stimulation rates on evoked potentials during general anesthesia: relation between the transient auditory middle-latency response and the 40-Hz auditory steady state response.

BACKGROUND: The auditory middle-latency response (transient) and the 40-Hz auditory steady state response (ASSR) are modulated by anesthetics. However, the quantitative relation between these evoked responses is difficult to obtain because of technical limitations of the recording methods used to obtain transients at high stimulation rates. This study uses continuous-loop averaging deconvolution to fill this technical gap and to study the relation between the transient and ASSR waveform during general anesthesia. METHODS: The authors recorded 5- and 40-Hz transients and 40-Hz ASSRs in 13 subjects during general anesthesia. The 5- and 40-Hz transients were used to predict the 40-Hz ASSR by linearly superimposing the transient waveforms. The predicted and recorded ASSRs were analyzed and compared using phasor and Hotelling T(2) analyses. RESULTS: Grand-averaged recordings revealed differences in the early middle-latency peaks between 5- and 40-Hz transients, e.g., the peak P(x) was present only in 5-Hz transient. Only the predicted 40-Hz ASSR derived from the 40-Hz transient matched the actual ASSR. Phasor analysis showed that the early peaks contribute significantly to the steady state waveform, and this explains why 5-Hz transient does not predict the 40-Hz ASSR. Oscillations in both the 5- and 40-Hz transients were observed during anesthesia. DISCUSSION: The 40-Hz ASSR represents a composite waveform and arises when transient waveforms elicited with a 40-Hz stimulation rate are overlapped and superimposed. During general anesthesia, the morphology of the transient is dependent on the rate of stimulus presentation. The composite nature of the ASSR may explain nonmonotonic anesthetic dose-response relations observed by others.

A new paradigm for the design of audible alarms that convey urgency information.

OBJECTIVE: The current international standard (IEC 60601-1-8) stipulates that medical device audible alarms should be priority-encoded and validated for efficacy. Evidence suggests that the melodic alarms described in the standard are not functioning as originally intended. We present a multi-disciplinary, human factors paradigm for audible alarm development whereby urgency information is encoded via modulation of the physical characteristics of sounds. We also test the feasibility of this approach using information measures. METHODS: We designed series of experimental sounds that varied along controlled physical and acoustical dimensions. Subjects rated these sound series for perceived urgency. Based on these ratings, selected sounds from each series were assigned a priority category from 'low' to 'high' - we call these resulting sets of sounds 'urgency-codecs'. The method of categorical judgments (based on information theory) was used to compare each urgency-codec for ability to convey urgency information. RESULTS: Subjects were consistent in their ratings of the three series of experimental sounds for perceived urgency. The urgency data pertaining to one of the series (harmonic interval) was successfully fit to a psychophysical empirical law. The urgency-codec derived from another sound series (melodic interval) was found to have the highest signal (correct interpretation of urgency level by subjects) transmission rate. CONCLUSIONS: The proposed paradigm is feasible, and it offers an evidence-based strategy for alarm sound design and testing. This approach would be performed before implementation of new alarm sounds in clinical settings, and should result in development of alarm sounds that satisfy the requirements of priority-encoding and validation.

Early management of the traumatized airway.

Penetrating face and neck trauma is usually obvious, but blunt trauma mandates high index of suspicion to recognize its existence. Comprehensive understanding of the injury is mandatory to plan the best timing and method to secure the airway.

Expression and localization of Muc4/sialomucin complex (SMC) in the adult and developing rat intestine: implications for Muc4/SMC function.

Muc4/sialomucin complex (SMC) is a high molecular mass heterodimeric membrane mucin, encoded by a single gene, and originally discovered in a highly metastatic ascites rat mammary adenocarcinoma. Subsequent studies have shown that it is a prominent component of many accessible and vulnerable epithelia, including the gastrointestinal tract. Immunoblot and immunofluorescence analyses demonstrated that Muc4/SMC expression in the rat small intestine increases from proximal to distal regions and is located predominantly in cells at the base of the crypts. These cells were postulated to be Paneth cells, based on their location, morphology, and secretory granule content. Immunohistochemistry indicated the presence of Muc4/SMC in these granules. Muc4/SMC expression was higher in the rat colon than small intestine and was abundantly present in colonic goblet cells, but not in goblet cells in the small intestine. Immunohistochemistry also suggested the presence of MUC4 in human colonic goblet cells. Biochemical analyses indicated that rat colonic Muc4/SMC is primarily the soluble form of the membrane mucin. Analyses of Muc4/SMC during development of the rat gastrointestinal tract showed its appearance at embryonic day 14 of the esophagus and at day 15 at the surface of the undifferentiated stratified epithelium at the gastroduodenal junction, then later at cell surfaces in the more distal regions of the differentiated epithelium of the small intestine, culminating in expression as an intracellular form in the crypts of the small intestine at about day 21. Limited expression in the colon was observed during development before birth at cell surfaces, with expression as an intracellular form in the goblet cells arising during the second week after birth. These results suggest that membrane mucin Muc4/SMC serves different functions during development of the intestine in the rat, but is primarily a secreted product in the adult animal.

Sialomucin complex in tumors and tissues.

Sialomucin complex (SMC) is a high Mr glycoprotein heterodimer, originally discovered on the cell surfaces of ascites sublines of the highly metastatic 13762 rat mammary adenocarcinoma, and composed of mucin (ASGP-1) and transmembrane (ASGP-2) subunits. SMC is encoded by a single gene and synthesized as a large precursor protein which is cleaved into its subunits early in its transit to the cell surface. SMC exhibits behavior typical of both membrane and secreted mucins. In the ascites cells, it is found only in the membrane form, creating a protective barrier at the cell surface to reduce cell adhesiveness and protect the tumor cell from immune killing. Normal tissues express both the membrane formand a non-membrane form, which may be secreted by either constitutive or regulated, secretory granule mechanisms. This soluble form is proposed to contribute to multilayer mucus gels which protect epithelia, though it may also play other roles. ASGP-2 contains two EGF-like domains, one of which binds the receptor tyrosine kinase ErbB-2. Thus, SMC may be a bifunctional protein, the mucin serving a protective function and the transmembrane domain possibly playing a role in the proliferation of metastatic tumor cells or repair processes necessary for the maintenance of damaged epithelia.