A machine-learning approach to human ex vivo lung perfusion predicts transplantation outcomes and promotes organ utilization.

Ex vivo lung perfusion (EVLP) is a data-intensive platform used for the assessment of isolated lungs outside the body for transplantation; however, the integration of artificial intelligence to rapidly interpret the large constellation of clinical data generated during ex vivo assessment remains an unmet need. We developed a machine-learning model, termed InsighTx, to predict post-transplant outcomes using n = 725 EVLP cases. InsighTx model AUROC (area under the receiver operating characteristic curve) was 79 ± 3%, 75 ± 4%, and 85 ± 3% in training and independent test datasets, respectively. Excellent performance was observed in predicting unsuitable lungs for transplantation (AUROC: 90 ± 4%) and transplants with good outcomes (AUROC: 80 ± 4%). In a retrospective and blinded implementation study by EVLP specialists at our institution, InsighTx increased the likelihood of transplanting suitable donor lungs [odds ratio=13; 95% CI:4-45] and decreased the likelihood of transplanting unsuitable donor lungs [odds ratio=0.4; 95%CI:0.16-0.98]. Herein, we provide strong rationale for the adoption of machine-learning algorithms to optimize EVLP assessments and show that InsighTx could potentially lead to a safe increase in transplantation rates.

Time to extubation for lung transplant recipients represents a pragmatic end-point to guide the development of prognostic tests.

The field of transplantation would benefit from the integration of advanced precision medicine techniques. Although predictive tests for lung transplantation require a well-defined clinical end-point, there exists no consensus regarding which outcomes are optimal end-points for these purposes. While many possible candidate end-points exist, we propose that time-to-extubation is an optimal end-point for prognostic tests because of its: clinical relevance; objectiveness; stability over time; and association with healthcare expenditure. Herein, we describe the rationale for this selection and present the limitations of alternative outcomes for this purpose. Using a 72-hour cut-off, time to extubation correlated well with Primary Graft Dysfunction Grade 3, intensive care unit and hospital length of stay, and a greater than 2-fold increase in healthcare cost ratios. Given that time-to-extubation is an objective measure that is readily measured by all lung transplant centers, this metric represents a preferred primary end-point for prognostic tests developed for lung transplantation.

A biomarker assay to risk-stratify patients with symptoms of respiratory tract infection.

BACKGROUND: Patients who present to an emergency department (ED) with respiratory symptoms are often conservatively triaged in favour of hospitalisation. We sought to determine if an inflammatory biomarker panel that identifies the host response better predicts hospitalisation in order to improve the precision of clinical decision making in the ED. METHODS: From April 2020 to March 2021, plasma samples of 641 patients with symptoms of respiratory illness were collected from EDs in an international multicentre study: Canada (n=310), Italy (n=131) and Brazil (n=200). Patients were followed prospectively for 28 days. Subgroup analysis was conducted on confirmed coronavirus disease 2019 (COVID-19) patients (n=245). An inflammatory profile was determined using a rapid, 50-min, biomarker panel (RALI-Dx (Rapid Acute Lung Injury Diagnostic)), which measures interleukin (IL)-6, IL-8, IL-10, soluble tumour necrosis factor receptor 1 (sTNFR1) and soluble triggering receptor expressed on myeloid cells 1 (sTREM1). RESULTS: RALI-Dx biomarkers were significantly elevated in patients who required hospitalisation across all three sites. A machine learning algorithm that was applied to predict hospitalisation using RALI-Dx biomarkers had a mean±sd area under the receiver operating characteristic curve of 76±6% (Canada), 84±4% (Italy) and 86±3% (Brazil). Model performance was 82±3% for COVID-19 patients and 87±7% for patients with a confirmed pneumonia diagnosis. CONCLUSIONS: The rapid diagnostic biomarker panel accurately identified the need for inpatient care in patients presenting with respiratory symptoms, including COVID-19. The RALI-Dx test is broadly and easily applicable across many jurisdictions, and represents an important diagnostic adjunct to advance ED decision-making protocols.

Predicting donor lung acceptance for transplant during ex vivo lung perfusion: The EX vivo lung PerfusIon pREdiction (EXPIRE).

Ex vivo lung perfusion (EVLP) has being increasingly used for the pretransplant assessment of extended-criteria donor lungs. Mathematical models to predict lung acceptance during EVLP have not been reported so far. Thus, we hypothesized that predictors of lung acceptance could be identified and used to develop a mathematical model describing the clinical decision-making process used in our institution. Donor lungs characteristics and EVLP physiologic parameters included in our EVLP registry were examined (derivation cohort). Multivariable logistic regression analysis was performed to identify predictors independently associated with lung acceptance. A mathematical model (EX vivo lung PerfusIon pREdiction [EXPIRE] model) for each hour of EVLP was developed and validated using a new cohort (validation cohort). Two hundred eighty donor lungs were assessed with EVLP. Of these, 186 (66%) were accepted for transplantation. ΔPO2 and static compliance/total lung capacity were identified as independent predictors of lung acceptance and their respective cut-off values were determined. The EXPIRE model showed a low discriminative power at the first hour of EVLP assessment (AUC: 0.69 [95% CI: 0.62-0.77]), which progressively improved up to the fourth hour (AUC: 0.87 [95% CI: 0.83-0.92]). In a validation cohort, the EXPIRE model demonstrated good discriminative power, peaking at the fourth hour (AUC: 0.85 [95% CI: 0.76-0.94]). The EXPIRE model may help to standardize lung assessment in centers using the Toronto EVLP technique and improve overall transplant rates.

Electrophysiologic and behavioral outcomes of cochlear implantation in children with auditory nerve hypoplasia.

OBJECTIVES: Hypoplasia of the auditory nerve (AN) refers to significant narrowing of the VIIIth cranial nerve which could compromise stimulation of the nerve by electrical pulses delivered from a cochlear implant (CI), thereby hindering activity in other parts of the auditory pathways. To compensate, high current levels or increased charge may be required to elicit auditory perception causing current to spread to other cranial nerves and potentially resulting in unwanted myogenic responses. Deficits in central auditory activity could reduce perception of speech and language. In the present study, we measured auditory brainstem responses in children with and without hypoplasia of the AN to answer the following questions. In children with hypoplastic ANs, (a) can CI stimulation evoke typical patterns of activity from the AN and brainstem?, (b) do brainstem responses change with CI experience?, (c) are evoked responses dependent on the size of the AN pathway?, and (d) does auditory development measured by behavioral tests of speech perception develop more slowly than in peers with normal AN diameter? DESIGN: Of 807 children using CIs in our program, 20 (2.5%) were identified as having AN hypoplasia using high-resolution computed tomographic scan and/or magnetic resonance imaging. An age-matched control group of children using CIs with normal AN diameter were recruited to compare electrophysiological and behavioral measures. Radiologic imaging was used to measure the diameter of the internal auditory canal (IAC), auditory nerve canal (ANC), and AN. Electrophysiological testing of the evoked compound action potential and auditory brainstem response was performed at CI activation and every 3 mo after initial testing up to 2 yr. Peak latency and waveform morphology were compared between study and control groups. Tests of speech perception and discrimination were attempted every 12 mo after device activation up to 10 yr. RESULTS: : Hypoplastic AN was identified as moderate to critical stenosis of the IAC, ANC, and AN. Initial electrically evoked compound action potential responses were mostly absent in children with AN hypoplasia. In the time window when electrically evoked auditory brainstem responses would be expected, some responses included single amplitude peaks at normal wave eV latencies, but the majority were abnormal, with peaks at atypical latencies or with no observable wave peaks. All evoked responses were inconsistent over time and did not reflect a typical pattern of auditory brainstem development. Speech perception scores were significantly poorer in the study group compared with controls and did not improve with CI experience. The type of abnormal evoked waveform response was independent of IAC, ANC, or AN diameter and also independent of behavioral outcome measures. CONCLUSIONS: : Evoked responses recorded in CI children with AN hypoplasia indicate a high incidence of nonauditory activity with CI use. The range of abnormal responses was not predicted by the severity of the hypoplastic AN or associated structures. This, along with poorer auditory development compared with peers with normal AN diameters, suggests that children with hypoplasia of the AN are poor candidates for cochlear implantation.

Investigating the role of the superior colliculus in active vision with the visual search paradigm.

We review here both the evidence that the functional visuomotor organization of the optic tectum is conserved in the primate superior colliculus (SC) and the evidence for the linking proposition that SC discriminating activity instantiates saccade target selection. We also present new data in response to questions that arose from recent SC visual search studies. First, we observed that SC discriminating activity predicts saccade initiation when monkeys perform an unconstrained search for a target defined by either a single visual feature or a conjunction of two features. Quantitative differences between the results in these two search tasks suggest, however, that SC discriminating activity does not only reflect saccade programming. This finding concurs with visual search studies conducted in posterior parietal cortex and the idea that, during natural active vision, visual attention is shifted concomitantly with saccade programming. Second, the analysis of a large neuronal sample recorded during feature search revealed that visual neurons in the superficial layers do possess discriminating activity. In addition, the hypotheses that there are distinct types of SC neurons in the deeper layers and that they are differently involved in saccade target selection were not substantiated. Third, we found that the discriminating quality of single-neuron activity substantially surpasses the ability of the monkeys to discriminate the target from distracters, raising the possibility that saccade target selection is a noisy process. We discuss these new findings in light of the visual search literature and the view that the SC is a visual salience map for orienting eye movements.

Lateralization of interimplant timing and level differences in children who use bilateral cochlear implants.

OBJECTIVES: Interaural level differences (ILD) and interaural timing differences (ITD) are important cues for locating sounds in space. Adult bilateral cochlear implant (CI) users use ILDs more effectively than ITDs. Few studies investigated the ability of children who use bilateral CIs to make use of these binaural cues. Our working hypothesis was that children using bilateral CIs are able to perceive changes in ITDs and ILDs similar to their normal-hearing (NH) peers. DESIGN: Participants were two groups of children; 19 bilateral implant users (CI) and nine NH children. The children in the CI group had received a second CI after 4.9 +/- 2.8 yrs of unilateral use. Children performed a four alternative forced-choice lateralization task in which they were asked to describe stimuli as coming from the left side, right side, middle of the head, or from both sides simultaneously. Stimuli were 500 msec trains of electrical pulses delivered to apical electrode no. 18 (CI group) or clicks (NH group) presented 11 times per second with either ITDs (0, 400, 1000, or 2000 microsec delay between sides) or level differences (0, 10, or 20 Current Units (CI group) or 0, 10, or 20 dB (NH group) difference between sides). ITDs were presented using current levels that were balanced using left and right electrically evoked brain stem responses. Stimulus levels evoking response amplitudes that were most similar were used. RESULTS: Responses from children in the CI group changed significantly with changes in ILD of bilateral stimuli, but not with changes in ITD. Responses from children in the CI group were significantly different from those in the NH group in three ways. Children in the CI group perceived bilaterally presented electrical pulses: (1) to come from the second implanted side more often than the first, (2) to rarely come from the middle, and (3) to come from both sides of the head simultaneously. Perceived changes in lateralization with ILD changes were correlated with differences in amplitudes of electrically evoked brain stem responses by the left versus right CI. CONCLUSIONS: The results of this study illustrate that children who use bilateral CIs can lateralize stimuli on the basis of level cues, but have difficulty interpreting interimplant timing differences. Perceived lateralization of bilaterally presented stimuli to the second implanted side in many of the stimulus conditions may relate to the use of different device generations between sides. Further differences from normal lateralization responses could be due to abnormal binaural processing, possibly resulting from a period of unilateral hearing before the provision of a second implant or due to insufficiently matched interimplant stimuli. It may be possible to use objective measures such as electrically evoked auditory brain stem responses wave eV amplitudes to provide balanced levels of bilateral stimulation in children who have had no binaural hearing experience.

Auditory development in the absence of hearing in infancy.

The auditory brainstem pathways require stimulation to mature, but do they develop in the absence of auditory input? To answer this, peaks of the electrically evoked auditory nerve (wave eN1) and brainstem response (eII, eIII, and eV) were measured in 117 children with early-onset deafness who had received cochlear implants. Data were collected at cochlear implant activation. In the absence of significant input, the interpeak latency eN1-eIII decreased over the first year of life but remained constant thereafter. Chronic cochlear implant stimulation was required to promote significant reduction in eIII-eV interpeak latency. Thus, activity-independent changes occurring in infancy are concentrated in the caudal auditory brainstem and likely involve the auditory nerve. Changes requiring input are most prevalent in rostral brainstem.

Effects of stimulus manipulation on electrophysiological responses in pediatric cochlear implant users. Part I: duration effects.

Discrepancies between electrophysiological and behavioral thresholds in cochlear implant users might be due to differences in stimuli such as the duration and rate of the electrical pulse train. In the present study, we asked: Is there an effect of stimulus duration on electrophysiological responses of the auditory brainstem, thalamo-cortex, and behavioral thresholds? In 5 pediatric cochlear implant users, behavioral thresholds in response to electrical pulse trains at 500 pulses per second (pps) were significantly lower for 40ms than 2ms duration pulse trains. Clear electrically evoked auditory brainstem responses (EABR) and electrically evoked middle latency responses (EMLR) were generated by single electrical pulses and 2, 6, and 10ms pulse trains (500pps) in 5 children. There was a linear decrease in the inter-wave latency between the eV of the EABR and the Na of the EMLR as duration increased. No significant effect of duration was found on eV latency relative to the last pulse in the train or Na latency relative to the onset of the stimuli. Behavioral threshold data is consistent with temporal integration of auditory activity. Electrophysiological data indicates that: (a) recognizable EABR and EMLR waveforms can be recorded in response to electrical pulse trains of up to 10ms; and (b) pulse train stimuli have unique effects on the auditory brainstem compared to thalamo-cortical areas.

Effects of stimulus manipulation on electrophysiological responses of pediatric cochlear implant users. Part II: rate effects.

Electrophysiological thresholds do not accurately predict behavioral thresholds in pediatric cochlear implant users possibly due to differences in rate and duration of pulse presentation. We asked: (1) Is there an effect of rate of stimulus presentation on the electrophysiological responses of the auditory brainstem and thalamo-cortex? and (2) can the relationship between electrophysiological and behavioral thresholds be improved by using the same rate of pulse presentation? Behavioral and electrophysiological (EABR and EMLR) responses were elicited for 14 children to single electrical pulses and pulse trains of 2ms ranging in rate from 500 to 3600 pulses per second (pps). Low rate (500pps) pulse trains resulted in an increase in EABR wave eIII amplitude and a decrease in wave eV amplitude. Further rate increases resulted in smaller EABR wave amplitudes. EMLR amplitudes were unaffected by increases in rate as were EABR and EMLR latencies. Behavioral thresholds decreased with increasing rate, however, there was no associated reduction in electrophysiological thresholds. Correlation between behavioral and electrophysiological thresholds did not improve by using the same rate of electrical pulse stimulation. Results suggest: (1) Higher rates of electrical pulse presentation increase the potential for neural adaptation in the auditory brainstem and (2) using the same rate of electrical pulse presentation does not improve the ability of EABR and EMLR thresholds to predict behavioral thresholds.

Abnormal timing delays in auditory brainstem responses evoked by bilateral cochlear implant use in children.

HYPOTHESIS: A period of unilateral implant use before bilateral implantation affects timing of brainstem processes measured by the electrically evoked auditory brainstem response (EABR). BACKGROUND: EABR latencies decrease with unilateral implant use potentially disrupting binaural timing cues important in auditory brainstem processing of bilateral input. METHODS: EABRs were evoked by electrical pulses from the left, right, and both implants simultaneously in 3 groups of children. All were initially implanted at ages younger than 3 years and had the following: 1) a long delay (>2 yr [n = 16]), 2) a short delay (<1 yr [n = 15]), or 3) no delay (n = 15) between left and right ear implantation. Responses were recorded on the first day of bilateral implant use and 3 and 9 months thereafter. RESULTS: Relative to responses evoked in the experienced ear, the naive ear showed prolonged latency in both the EABR peaks and the binaural difference response. After 3 and 9 months of bilateral implant use, the relative prolongation decreased in the long and short delay groups, but significant differences persisted in the former. No clear differences in latencies evoked by the left versus right implant were found at any time point in children receiving bilateral implants simultaneously. CONCLUSION: Results suggest potential disruptions to binaural brainstem processing based on timing cues in children receiving a second cochlear implant after more than 2 years of unilateral implant use that persist through at least the first 9 months of bilateral implant use.

Effect of increasing duration of stimulation on the electrically evoked auditory brainstem and middle latency responses in pediatric cochlear implant users.

OBJECTIVE: Electrophysiologic responses used to predict behavioural stimulation levels in implant users are typically evoked with a single-pulse stimulus versus higher-rate pulse trains of longer duration. Unfortunately, electrophysiologic measures tend to overestimate behavioural responses. It may be possible to improve the predictive ability of evoked potential thresholds by increasing the duration of the stimulus. We asked if auditory brainstem and middle latency responses can be reliably measured using electrical pulse trains of increasing duration. DESIGN: Evoked potential testing was performed in five pediatric N24RE cochlear implant users aged 7.6 to 14.9 years. SETTING: Cochlear Implant Program, Cochlear Implant Laboratory, The Hospital for Sick Children, Toronto. METHODS: Responses were electrically evoked for durations of 2, 6, and 10 milliseconds by single pulses and pulse trains of 500 pulses per second. MAIN OUTCOME MEASURES: Evoked auditory brainstem response (EABR) and evoked middle latency response (EMLR) waveform latencies and amplitudes were compared between durations. RESULTS: Clear and replicable EABRs and EMLRs were recorded for all durations. There was no significant change in EABR wave eV latency relative to the stimulus offset. There was no significant change in EMLR wave latencies relative to the stimulus onset. This was confirmed by a linear decrease in the interwave latency between the EABR wave eV and the EMLR wave eNa. No significant changes in wave amplitude were found as the pulse train duration increased for the EABR or the EMLR. CONCLUSIONS: EABRs are dominated by the stimulus offset, whereas EMLRs appear to be dominated by stimulus onset. Clear, unchanging EABRs and EMLRs indicated that electrophysiologic measures were a valid tool for increasing durations.

Binaural processing in children using bilateral cochlear implants.

Binaural auditory brainstem processing was examined using evoked potential measures in 40 children who were implanted early and received a second implant simultaneously or after long or short periods of unilateral implant use. Wave latencies were shorter when evoked by the experienced versus naïve implanted ear at initial bilateral activation. Binaural difference waves were detected in most children in response to apical but not basal electrode stimulation and were prolonged in latency in children implanted after long or short delays between implants. Timing differences between the implanted ears in children receiving sequential but not simultaneous bilateral implants reflect a relative immaturity of pathways innervating the second ear and results in abnormal timing of binaural processing at this initial implant stage.