A validated preoperative risk prediction tool for extended inpatient length of stay following anatomic or reverse total shoulder arthroplasty

BACKGROUND

Recent work has shown inpatient length of stay (LOS) following shoulder arthroplasty to hold the second strongest association with overall cost (after implant cost itself). In particular, a preoperative understanding for the patients at risk of extended inpatient stays (≥3 days) can allow for counseling, optimization, and anticipating postoperative adverse events.

METHODS

A multicenter retrospective review was performed of 5410 anatomic (52%) and reverse (48%) total shoulder arthroplasties done at 2 large, tertiary referral health systems. The primary outcome was extended inpatient LOS of at least 3 days, and over 40 preoperative sociodemographic and comorbidity factors were tested for their predictive ability in a multivariable logistic regression model based on the patient cohort from institution 1 (derivation, N = 1773). External validation was performed using the patient cohort from institution 2 (validation, N = 3637), including area under the receiver operator characteristic curve (AUC), sensitivity, specificity, and positive and negative predictive values.

RESULTS

A total of 814 patients, including 318 patients (18%) in the derivation cohort and 496 patients (14%) in the validation cohort, experienced an extended inpatient LOS of at least 3 days. Four hundred forty-five (55%) were discharged to a skilled nursing or rehabilitation facility. Following parameter selection, a multivariable logistic regression model based on the derivation cohort (institution 1) demonstrated excellent preliminary accuracy (AUC: 0.826), with minimal decrease in accuracy under external validation when tested against the patients from institution 2 (AUC: 0.816). The predictive model was composed of only preoperative factors, in descending predictive importance as follows: age, marital status, fracture case, ASA (American Society of Anesthesiologists) score, paralysis, electrolyte disorder, body mass index, gender, neurologic disease, coagulation deficiency, diabetes, chronic pulmonary disease, peripheral vascular disease, alcohol dependence, psychoses, smoking status, and revision case.

CONCLUSION

A freely-available, preoperative online clinical decision tool for extended inpatient LOS (≥ 3 days) after shoulder arthroplasty reaches excellent predictive accuracy under external validation. As a result, this tool merits consideration for clinical implementation, as many risk factors are potentially modifiable as part of a preoperative optimization strategy.

A validated preoperative risk prediction tool for discharge to skilled nursing or rehabilitation facility following anatomic or reverse shoulder arthroplasty

BACKGROUND

As bundled payment models continue to spread, understanding the primary drivers of cost excess helps providers avoid penalties and ensure equal health care access. Recent work has shown discharge to rehabilitation and skilled nursing facilities (SNFs) to be a primary cost driver in total joint arthroplasty, and an accurate preoperative risk calculator for shoulder arthroplasty would not only help counsel patients in clinic during shared decision-making conversations but also identify high-risk individuals who may benefit from preoperative optimization and discharge planning.

METHODS

Anatomic and reverse total shoulder arthroplasty cohorts from 2 geographically diverse, high-volume centers were reviewed, including 1773 cases from institution 1 (56% anatomic) and 3637 from institution 2 (50% anatomic). The predictive ability of a variety of candidate variables for discharge to SNF/rehabilitation was tested, including case type, sociodemographic factors, and the 30 Elixhauser comorbidities. Variables surviving parameter selection were incorporated into a multivariable logistic regression model built from institution 1's cohort, with accuracy then validated using institution 2's cohort.

RESULTS

A total of 485 (9%) shoulder arthroplasties overall were discharged to post-acute care (anatomic: 6%, reverse: 14%, P < .0001), and these patients had significantly higher rates of unplanned 90-day readmission (5% vs. 3%, P = .0492). Cases performed for preoperative fracture were more likely to require post-acute care (13% vs. 3%, P < .0001), whereas revision cases were not (10% vs. 10%, P = .8015). A multivariable logistic regression model derived from the institution 1 cohort demonstrated excellent preliminary accuracy (area under the receiver operating characteristic curve [AUC]: 0.87), requiring only 11 preoperative variables (in order of importance): age, marital status, fracture, neurologic disease, paralysis, American Society of Anesthesiologists physical status, gender, electrolyte disorder, chronic pulmonary disease, diabetes, and coagulation deficiency. This model performed exceptionally well during external validation using the institution 2 cohort (AUC: 0.84), and to facilitate convenient use was incorporated into a freely available, online prediction tool. A model built using the combined cohort demonstrated even higher accuracy (AUC: 0.89).

CONCLUSIONS

This validated preoperative clinical decision tool reaches excellent predictive accuracy for discharge to SNF/rehabilitation following shoulder arthroplasty, providing a vital tool for both patient counseling and preoperative discharge planning. Further, model parameters should form the basis for reimbursement legislation adjusting for patient comorbidities, ensuring no disparities in access arise for at-risk populations.

Appropriate patient selection for outpatient shoulder arthroplasty: a risk prediction tool

BACKGROUND

The transition from inpatient to outpatient shoulder arthroplasty critically depends on appropriate patient selection, both to ensure safety and to counsel patients preoperatively regarding individualized risk. Cost and patient demand for same-day discharge have encouraged this transition, and a validated predictive tool may help decrease surgeon liability for complications and help select patients appropriate for same-day discharge. We hypothesized that an accurate predictive model could be created for short inpatient length of stay (discharge at least by postoperative day 1), potentially serving as a useful proxy for identifying patients appropriate for true outpatient shoulder arthroplasty.

METHODS

A multicenter cohort of 5410 shoulder arthroplasties (2805 anatomic and 2605 reverse shoulder arthroplasties) from 2 geographically diverse, high-volume health systems was reviewed. Short inpatient stay was the primary outcome, defined as discharge on either postoperative day 0 or 1, and 49 patient outcomes and factors including the Elixhauser Comorbidity Index, sociodemographic factors, and intraoperative parameters were examined as candidate predictors for a short stay. Factors surviving parameter selection were incorporated into a multivariable logistic regression model, which underwent internal validation using 10,000 bootstrapped samples.

RESULTS

In total, 2238 patients (41.4%) were discharged at least by postoperative day 1, with no difference in rates of 90-day readmission (3.5% vs. 3.3%, P = .774) between cohorts with a short length of stay and an extended length of stay (discharge after postoperative day 1). A multivariable logistic regression model demonstrated high accuracy (area under the receiver operator characteristic curve, 0.762) for discharge by postoperative day 1 and was composed of 13 variables: surgery duration, age, sex, electrolyte disorder, marital status, American Society of Anesthesiologists score, paralysis, diabetes, neurologic disease, peripheral vascular disease, pulmonary circulation disease, cardiac arrhythmia, and coagulation deficiency. The percentage cutoff maximizing sensitivity and specificity was calculated to be 47%. Internal validation showed minimal loss of accuracy after bias correction for overfitting, and the predictive model was incorporated into a freely available online tool to facilitate easy clinical use.

CONCLUSIONS

A risk prediction tool for short inpatient length of stay after shoulder arthroplasty reaches very good accuracy despite requiring only 13 variables and was derived from an underlying database with broad geographic diversity in the largest institutional shoulder arthroplasty cohort published to date. Short inpatient length of stay may serve as a proxy for identifying patients appropriate for same-day discharge, although perioperative care decisions should always be made on an individualized and holistic basis.

Cardiovascular findings on chest computed tomography associated with COVID-19 adverse clinical outcomes

STUDY OBJECTIVE

Chest computed tomography (chest CT) is routinely obtained to assess disease severity in COVID-19. While pulmonary findings are well-described in COVID-19, the implications of cardiovascular findings are less well understood. We evaluated the impact of cardiovascular findings on chest CT on the adverse composite outcome (ACO) of hospitalized COVID-19 patients.

SETTING/PARTICIPANTS

245 COVID-19 patients who underwent chest CT at Rush University Health System were included.

DESIGN

Cardiovascular findings, including coronary artery calcification (CAC), aortic calcification, signs of right ventricular strain [right ventricular to left ventricular diameter ratio, pulmonary artery to aorta diameter ratio, interventricular septal position, and inferior vena cava (IVC) reflux], were measured by trained physicians.

INTERVENTIONS/MAIN OUTCOME MEASURES

These findings, along with pulmonary findings, were analyzed using univariable logistic analysis to determine the risk of ACO defined as intensive care admission, need for non-invasive positive pressure ventilation, intubation, in-hospital and 60-day mortality. Secondary endpoints included individual components of the ACO.

RESULTS

Aortic calcification was independently associated with an increased risk of the ACO (odds ratio 1.86, 95% confidence interval (1.11-3.17) p < 0.05). Aortic calcification, CAC, abnormal septal position, or IVC reflux of contrast were all significantly associated with 60-day mortality and major adverse cardiovascular events. IVC reflux was associated with in-hospital mortality (p = 0.005).

CONCLUSION

Incidental cardiovascular findings on chest CT are clinically important imaging markers in COVID-19. It is important to ascertain and routinely report cardiovascular findings on CT imaging of COVID-19 patients as they have potential to identify high risk patients.

In-hospital predictors of 60-day readmission in COVID-19 patients

Introduction

Preventing hospital readmissions can improve a patient's quality of life and decrease healthcare costs. While prior work has focused on pre-existing comorbidities to predict COVID-19 readmissions, the prognostic role of in-hospital data and complications has been less studied.

Methods

Data was collected on adult patients diagnosed with COVID-19 and admitted to a multicenter hospital system in Illinois between March and November 2020. Our cohort consisted of COVID-19 hospitalization survivors excluding those discharged to hospice care. Major adverse events (MAEs) were defined as venous thromboembolism (VTE), myocardial injury (troponin greater than upper limit of normal), stroke, new requirement for renal replacement therapy (RRT), life-threatening arrhythmia, or acute heart failure exacerbation. The primary outcome was readmission within 60 days of initial hospitalization.

Results

From the 1406 survivors of the index hospitalization, 223 (15.9%) patients were readmitted within 60 days. Those readmitted were older and more likely to have underlying comorbidities including atrial fibrillation, coronary artery disease, and hypertension (Table 1). Length of stay between the readmission and non-readmission groups was trending towards statistical significance (10.52 days vs 8.95 days, p=0.053).

Those with one or more MAE during their index hospitalization, when adjusted for age and body mass index, were at an increased risk of readmission (adjusted odds ratio [aOR] 1.90, p&lt;0.01). Readmitted patients were more likely to have VTE during their index hospitalization than those not readmitted (7.2% vs 3.7%, p&lt;0.05). The incidence of new RRT (4.9% vs 2.5%, p=0.083) and myocardial injury (3.6% vs 1.5%, p=0.067) between the groups was also trending towards statistical significance (Table 1). No statistical difference was present between the other individual MAEs; however, this is limited by small sample sizes of certain MAEs. Of the 322 patients with echocardiography during the index admission, 82 (25.5%) were readmitted. In this cohort, left ventricular ejection fraction (LVEF) that was reduced (LVEF &lt;50%) or hyperdynamic (LVEF &gt;65%) was not a statistically significant predictor of readmission (Figure 1).

Lastly, discharge disposition was predictive of readmission as those being sent to acute rehab (OR 2.04, p&lt;0.01), long-term acute care (OR 2.58, p&lt;0.01), or skilled nursing facility (OR 2.67, p&lt;0.001) were at higher risk compared to those who were discharged to home (Figure 1).

Conclusion

In this cohort, the occurrence of any MAE during index COVID-19 hospitalization, particularly VTE, RRT, and myocardial injury, can be used to predict 60-day readmission. Furthermore, discharge disposition, but not LVEF, demonstrated prognostic value in our cohort. Identifying high risk patients prior to discharge helps health care providers focus resources on patients most likely to be readmitted.

Funding Acknowledgement

Type of funding sources: None.

Cardiac arrest in patients hospitalized for COVID-19: a tertiary medical center retrospective cohort study

Background/Introduction

Patients with COVID-19 are at increased risk for mortality during hospitalization. Better definition of the incidence, predictors, and outcomes of cardiac arrest during hospitalization for COVID-19 may support early identification and intervention.

Purpose

To estimate the incidence of in-hospital cardiac arrest in patients with COVID-19, describe the temporal trends in incidence of and survival after cardiac arrest, summarise characteristics of those who experienced a cardiac arrest, and compare the characteristics of survivors versus non-survivors of cardiac arrest.

Methods

We conducted a retrospective cohort study of patients admitted for COVID-19 to a tertiary medical center comprising three hospitals between March and November 2020. Data entry is ongoing for more than 2000 patients admitted through 2021. Clinical variables extracted via review of electronic medical records included age, sex, race/ethnicity, body mass index, history of cardiovascular disease (ie., coronary artery disease, congestive heart failure, atrial fibrillation, or cerebrovascular event), other comorbidities included in the Charlson comorbidity index, date of admission, duration of hospitalization, all cardiac arrest events during hospitalization, presenting rhythm during first cardiac arrest, and death. Data were described using summary statistics. Multivariable logistic regression was used to evaluate associations.

Results

Among 1666 patients, 107 (6.4%) experienced at least one in-hospital cardiac arrest event during hospitalization for COVID-19, of which 25 (23%) survived to hospital discharge. From March to October 2020, there was a decrease in estimated cardiac arrest incidence in-hospital from 8.2% to 3%, whereas estimated survival to hospital discharge after an arrest remained similar at approximately 20% (Figure). Compared to those who did not, patients who experienced in-hospital cardiac arrest were older and more likely to have existing cardiovascular disease, as well as other comorbidities. Similar factors were associated with lower chance of survival after cardiac arrest (Table). Patients with pulseless ventricular tachycardia/fibrillation (VT/VF) as presenting rhythm in cardiac arrest had better survival to hospital discharge compared to those with other rhythms (OR 3.3, p=0.02). Younger age (per 10 years, OR=0.7, p=0.03) and fewer comorbidities (per one fewer comorbidity, OR=1.5, p=0.05) were associated with better survival after cardiac arrest in multivariable logistic regression.

Conclusion

There was a decline in estimated incidence of cardiac arrest during hospitalization for COVID-19 since beginning of pandemic, with survival to hospital discharge after cardiac arrest estimated to be stable at around 20%. Younger age and fewer comorbidities especially cardiovascular disease were associated with better survival after an in-hospital cardiac arrest.

Funding Acknowledgement

Type of funding sources: Public hospital(s). Main funding source(s): Rush University Medical Center Figure 1Table 1

The prognostic value of cardiac troponin for 60 day mortality and major adverse events in COVID-19 patients

BACKGROUND

The variability of coronavirus disease 2019 (COVID-19) illness severity has puzzled clinicians and has sparked efforts to better predict who would benefit from rapid intervention. One promising biomarker for in-hospital morbidity and mortality is cardiac troponin (cTn).

METHODS

A retrospective study of 1331 adult patients with COVID-19 admitted to the Rush University System in Illinois, USA was performed. Patients without cTn measurement during their admission or a history of end stage renal disease or stage 5 chronic kidney disease were excluded. Using logistic regression adjusted for baseline characteristics, pre-existing comorbidities, and other laboratory markers of inflammation, cTn was assessed as a predictor of 60-day mortality and severe COVID-19 infection, consisting of a composite of 60-day mortality, need for intensive care unit, or requiring non-invasive positive pressure ventilation or intubation.

RESULTS

A total of 772 patients met inclusion criteria. Of these, 69 (8.9%) had mild cTn elevation (> 1 to < 2x upper limit of normal (ULN)) and 46 (6.0%) had severe cTn elevation (≥ 2x ULN). Regardless of baseline characteristics, comorbidities, and initial c-reactive protein, lactate dehydrogenase, and ferritin, when compared to the normal cTn group, mild cTn elevation and severe cTn elevation were predictors of severe COVID-19 infection (adjusted OR [aOR] aOR 3.00 [CI: 1.51 - 6.29], P < 0.01; aOR 9.96 [CI: 2.75 - 64.23], P < 0.01, respectively); severe cTn elevation was a predictor of in-hospital mortality (aOR 2.42 [CI: 1.10 - 5.21], P < 0.05) and 60-day mortality (aOR 2.45 [CI: 1.13 - 5.25], P < 0.05).

CONCLUSION

In our cohort, both mild and severe initial cTn elevation were predictors of severe COVID-19 infection, while only severe cTn elevation was predictive of 60-day mortality. First cTn value on hospitalization is a valuable longitudinal prognosticator for COVID-19 disease severity and mortality.

Impact of pre-existing heart failure on 60-day outcomes in patients hospitalized with COVID-19

Background

In the coronavirus disease 2019 (COVID-19) global pandemic, patients with cardiovascular disease represent a vulnerable population with higher risk for contracting COVID-19 and worse prognosis with higher case fatality rates. However, the relationship between COVID-19 and heart failure (HF) is unclear, specifically whether HF is an independent risk factor for severe infection or if other accompanying comorbidities are responsible for the increased risk.

Methods

This is a retrospective analysis of 1331 adult patients diagnosed with COVID-19 infection between March and June 2020 admitted at Rush University System for Health (RUSH) in metropolitan Chicago, Illinois, USA. Patients with history of HF were identified by International Classification of Disease, Tenth Revision (ICD-10) code assignments extracted from the electronic medical record. Propensity score matching was utilized to control for the numerous confounders, and univariable logistic regression was performed to assess the relationship between HF and 60-day morbidity and mortality outcomes.

Results

The propensity score matched cohort consisted of 188 patients in both the HF and no HF groups. HF patients did not have lower 60-day mortality (OR 0.81; p = 0.43) compared to patients without HF. However, those with HF were more likely to require readmission within 60 days (OR 2.88; p < 0.001) and sustain myocardial injury defined as troponin elevation within 60 days (OR 3.14; p < 0.05).

Conclusions

This study highlights the complex network of confounders present between HF and COVID-19. When balanced for these numerous factors, those with HF appear to be at no higher risk of 60-day mortality from COVID-19 but are at increased risk for morbidity.

Fracture Rates in Men With Non-Metastatic Prostate Cancer on Androgen Deprivation Therapy With or Without Anti-Osteoporosis Treatment

Introduction: Androgen deprivation therapy (ADT) decreases bone mineral density and increases osteoporotic fracture (OsteoFx) risk. Hypothesis: To assess OsteoFx incidence most predictive of future OsteoFx among men with prostate cancer on ADT. Methods: 4370 electronic medical records were reviewed of adult men with prostate cancer on cancer therapy +/- anti-osteoporotic therapy (Anti-OsteoRx) from 2011–2019. Cancer therapy included ADT (anti-androgens, GnRH agonists & antagonists, orchiectomy) and supplemental cancer therapy (SupplRx) (prostatectomy, brachytherapy, radiation, immunotherapy, and chemotherapy). Anti-OsteoRx included bisphosphonates, denosumab, and parathyroid hormone analogs. Patients with other cancers within 5 years of initial visit, metastasis or traumatic fractures were excluded. Retrospective analysis was done to determine baseline characteristics, type and duration of ADT, Anti-OsteoRx, SupplRx, and OsteoFx incidence. Results: Fracture rate subgroups: • ADT only - Anti-OsteoRx 37/ 374 fractured (9.89%) • ADT only + Anti-OsteoRx 10/52 fractured (19.23%) • ADT + SupplRx + Anti-OsteoRx 2/19 fractured (10.53%) • ADT + SupplRx + Anti-OsteoRx 13/170 fractured (7.65%) Comparing fracture rates between subgroups: • Comparing ADT only +/- Anti-OsteoRx, statistical significance was observed with higher fracture rate in patients taking Anti-OsteoRx (19.23% vs. 9.89%, p &lt; 0.044) • Comparing ADT + SupplRx +/- Anti-OsteoRx, no significant difference in fracture rates due to small number of fractures Comparing combined subgroups: • ADT +/- SupplRx + Anti-OsteoRx 12/71 (16.9%) fractured • ADT +/- SupplRx - Anti-OsteoRx 50/544 (9.19%) fractured • Statistically significant between groups fracture rates was observed (p= 0.042) in patients treated with Anti-OsteoRX. Discussion: Patients receiving Anti-OsteoRx, regardless of their prostate cancer therapies, had higher rates of fractures (16.9 vs. 9.19%, p= 0.042) due to their being selected for therapy based on greater clinical risks. The Anti-OsteoRx group had a higher percentage of glucocorticoid listed as a historical medication (26.8 vs.15.3% vs, p= 0.023), glucocorticoids administered (50.7 vs. 30.3% p=0.001), and anticonvulsants and proton-pump inhibitor use (45.1 vs. 26.5%, p= 0.002). Conclusion: Higher fracture rates were observed in patients on Anti-OsteoRx that could be related to their being selected for treatment based on risk factors known to be associated with osteoporosis. Limited Anti-OsteoRx use in our study is possibly related to lack of standardized guidelines for prevention of osteoporotic fractures in prostate cancer patients. OsteoFx risk assessment utilizing CRF, DXA, and FRAX may prevent fractures in these high-risk patients. Further long-term prospective studies to address these unresolved queries are warranted.

Clinical Risk Factors for Osteoporotic Fractures in Men With Non-Metastatic Prostate Cancer on Androgen Deprivation Therapy With or Without Anti-Osteoporosis Treatment

Introduction: Androgen deprivation therapy (ADT) decreases bone mineral density and increases osteoporotic fracture (OsteoFx) risk.

Hypothesis: To assess OsteoFx clinical risk factors (CRF) most predictive of future OsteoFx among men with prostate cancer on ADT.

Methods: 4370 electronic medical records were reviewed of adult men with prostate cancer on cancer therapy +/- anti-osteoporosis therapy (Anti-OsteoRx) from 2011–2019. Cancer therapy included ADT (anti-androgens, GnRH agonists & antagonists, orchiectomy) and supplemental cancer therapy (SupplRx) (prostatectomy, brachytherapy, radiation, immunotherapy, and chemotherapy). Anti-OsteoRx included bisphosphonates, denosumab, and parathyroid hormone analogs. Patients with other cancers within 5 years of initial visit, metastasis, and traumatic fractures were excluded. Retrospective analysis was done to determine baseline characteristics, type and duration of ADT, Anti-OsteoRx, SupplRx, and osteoporosis CRF.

Results: 615 men on ADT +/- SupplRx +/- Anti-OsteoRx were included in the study. 10.08% had OsteoFx irrespective of SupplRx or Anti-OsteoRx. Comparing the OsteoFx group to the non-fracture group, the following CRF were found to be statistically significant (p <0.05): age at prostate cancer diagnosis (75.10 +/- 11.80 vs 71.59 +/- 9.80 y), diabetes mellitus (DM) (33.9 vs 19%), pre-existing comorbidities affecting bone (PreCo) (41.9 vs 24.8%), steroid use (11.3 vs 4.0%), and anti-convulsant and proton-pump inhibitor (med) use (45.2 vs 26.8%).

9.89% of 374 men on ADT only without (wo) Anti-OsteoRx fractured. Statistically significant CRF for OsteoFx were age (76.86 +/- 10.55 vs 73.02 +/- 10.06 y), DM (40.5 vs 19.6%), PreCo (45.9 vs. 26.4%), and med use (48.6 vs. 25.5%).

In the following subgroups there were no statistically significant difference in CRF:•7.64% of 170 men on ADT + SupplRx wo Anti-OsteoRx •19.23% of 52 men on ADT only + Anti-OsteoRx •10.52% of 19 men on ADT + SupplRx + Anti-OsteoRx

To increase statistical power, patients on ADT +/- SupplRx were assessed:•Among 71 men on ADT +/- SupplRx + Anti-OsteoRx, there were no statistically significant differences in CRF•Among the 544 men on ADT +/- SupplRx wo Anti-OsteoRx, significant CRF for OsteoFx were age (75.16 + 11.70 vs 71.37 + 9.85 y), DM (38 vs 19.4%), PreCo (38 vs 24.1%), steroid use (12 vs 3.8%), and med use (48 vs 24.3%)

Discussion:

Men with prostate cancer requiring ADT have a higher incidence of osteoporosis defined by DXA prior to initiating ADT compared to age-matched cohorts (Hussain et al). Our study revealed ADT with CRF is associated with OsteoFx irrespective of SupplRx or Anti-OsteoRx. Limitations include inability to evaluate efficacy of Anti-OsteoRx due to insufficient power.

Conclusion: OsteoFx risk assessment utilizing CRF, FRAX, DXA with timely intervention may prevent OsteoFx in these high-risk patients.

Sex Differences in COVID-19 Hospitalization and Mortality

Background: To investigate sex differences in coronavirus disease 2019 (COVID-19) outcomes in a large Illinois-based cohort. Methods: A multicenter retrospective cohort study compared males versus females with COVID-19 infections from March 1, 2020, to June 21, 2020, in the Rush University System. We analyzed sex differences in rates of hospitalization, intensive care unit (ICU) admission, vasopressor use, endotracheal intubation, and death in this cohort. A multivariable model correcting for age and sum of comorbidities was used to explore associations between sex and COVID-19-related outcomes. Results: There were 8108 positive COVID-19 patients-4300 (53.0%) females and 3808 (47.0%) males. Males had higher rates of hospitalization (19% vs. 13%; p < 0.001), ICU transfer (8% vs. 4%; p < 0.001), vasopressor support (4% vs. 2%; p < 0.001), and endotracheal intubation (5% vs. 2%; p < 0.001). Of those who died, 92 were males and 64 were females (2% vs. 1%; p = 0.003). A multivariable model correcting for age and sum of comorbidities showed a significant association between male sex and mortality in the total cohort (odds ratio, 1.96; 95% confidence interval, 1.34-2.90; p = 0.001). Conclusion: Male sex was independently associated with death, hospitalization, ICU admissions, and need for vasopressors or endotracheal intubation, after correction for important covariates.

Noise interference with echo delay discrimination in bat biosonar

Echolocating big brown bats (Eptesicus fuscus) were trained in a two-choice task to discriminate differences in the delay of electronic echoes at 1.7 ms delay (30 cm simulated range). Difference thresholds (∼45 μs) were comparable to previously published results. At selected above-threshold differences (116 and 232 μs delay), performance was measured in the presence of wideband random noise at increasing amplitudes in 10-dB steps to determine the noise level that prevented discrimination. Performance eventually failed, but the bats increased the amplitude and duration of their broadcasts to compensate for increasing noise, which allowed performance to persist at noise levels about 25 dB higher than without compensation. In the 232-μs delay discrimination condition, echo signal-to-noise ratio (2E/N₀) was 8-10 dB at the noise level that depressed performance to chance. Predicted echo-delay accuracy using big brown bat signals follows the Cramér-Rao bound for signal-to-noise ratios above 15 dB, but worsens below 15 dB due to side-peak ambiguity. At 2E/N₀ = 7-10 dB, predicted Cramér-Rao delay accuracy would be about 1 μs; considering side-peak ambiguity it would be about 200-300 μs. The bats' 232 μs performance reflects the intrusion of side-peak ambiguity into delay accuracy at low signal-to-noise ratios.

Direct visualization of particle attachment to a pendant drop

An experimental investigation is carried out into the attachment of a single particle to a liquid drop. High-speed videography is used to directly visualize the so-called 'snap-in' effect which occurs rapidly over sub-millisecond timescales. Using high-magnification, the evolution of the contact line around the particle is tracked and dynamic features such as the contact angle, wetted radius and force are extracted from these images to help build a fundamental understanding of the process. By examining the wetted length in terms of an arc angle, ϕ, it is shown that the early wetting stage is an inertial-dominated process and best described by a power law relation, i.e. ϕ ∼ (t/τ)^α, where τ is an inertial timescale. For the subsequent lift-off stage, the initial particle displacement is matched with that predicted using a simple balance between particle weight and capillary force with reasonable agreement. The lift-off force is shown to be on the order of 1-100 μN, whilst the force of impacting droplets is known to be on the order of 10-1000 mN. This explains the ease in which liquid marbles are formed during impact experiments.

Further comments on the induction of lung tumors by plutonium dioxide in beagles

A series of recent papers describes the final results obtained from studies of the effects of inhaled plutonium dioxide on beagle dogs. This note considers the value of a microdosimetric assessment of these data. In particular, it offers support for the existence of a threshold for the induction of lung tumors.

A microdosimetric reassessment of new data on the effects of plutonium dioxide inhalation by beagle dogs

Traditionally, the assessment of the effects of radiation on living tissue has been made in terms of absorbed dose. This concept, however, might not be the most appropriate when considering the effects arising from the inhalation of insoluble radioactive particulates in inhomogeneous tissue such as the lung. We have therefore applied microdosimetric methods to this problem and, in particular, investigated in detail how energy depositions are distributed when alpha particles travel through parenchymal lung tissue. Sections of material derived from rat, beagle and human lung were examined in an image analyzer, and an imaginary plutonium dioxide particulate was placed on the surface of an alveolar sac. The hypothetical alpha particles emitted from it were followed to the ends of their tracks so that the effects of the material's real structure could be followed in detail. It was found that, taking such detail into account, the alpha particles traveled much greater distances than might have been thought on the basis of a uniform, structureless lung. It was also found that the specific energy distributions can cover several orders of magnitude and can differ significantly between tissue as a whole, cells and nuclei at low exposures. Attempts are made to correlate these results with recently published data on beagle dogs that had inhaled graded exposure levels of plutonium dioxide aerosols.

Irradiated lithium formate and amino acids

Radiation-induced free radicals in solids show a microwave saturation effect when studied by electron spin resonance. A comparison is made between such effects in lithium formate and amino acids. The relative effectiveness of neutrons against high-energy photons is also considered.

Oscillatory magnetothermopower and resonant phonon drag in a high-mobility 2D electron gas

Experimental and theoretical evidence is presented for new low-magnetic-field (B<5 kG) 1/B oscillations in the thermoelectric power of a high-mobility GaAs/AlGaAs two-dimensional (2D) electron gas. The oscillations result from inter-Landau-level resonances of acoustic phonons carrying a momentum equal to twice the Fermi wave number at B=0. Numerical calculations show that both 3D and 2D phonons can contribute to this effect.

Echo delay versus spectral cues for temporal hyperacuity in the big brown bat, Eptesicus fuscus

Big brown bats can discriminate between echoes that alternate in delay (jitter) by as little as 10-15 ns and echoes that are stationary in delay. This delay hyperacuity seems so extreme that it has been rejected in favor of an explanation in terms of artifacts in echoes, most likely spectral in nature, that presumably are correlated with delay. Using different combinations of digital, analog, and cable delays, we dissociated the overall delay of jittering echoes from the size of the analog component of delay, which alone is presumed to determine the strength of the apparatus artifact. The bats' performance remains invariant with respect to the overall delay of the jittering echoes, not with respect to the amount of analog delay. This result is not consistent with the possible use of delay-related artifacts produced by the analog delay devices. Moreover, both electronic and acoustic measurements disclose no spectral cues or impedance-mismatch reflections in delayed signals, just time-delays. The absence of artifacts from the apparatus and the failure of overlap and interference from reverberation to account for the 10-ns result means that closing the gap between the level of temporal accuracy plausibly explained from physiology and the level observed in behavior may require a better understanding of the physiology.

Resistivity of dilute 2D electrons in an undoped GaAs heterostructure

We report resistivity measurements from 0.03 to 10 K in a dilute high mobility 2D electron system. Using an undoped GaAs/AlGaAs heterojunction in a gated field-effect transistor geometry, a wide range of densities, 0.16 x 10(10) to 7.5 x 10(10) cm(-2), are explored. For high densities, the results are quantitatively shown to be due to scattering by acoustic phonons and impurities. In an intermediate range of densities, a peak in the resistivity is observed for temperatures below 1 K. This nonmonotonic resistivity can be understood by considering the known scattering mechanisms of phonons, bulk, and interface ionized impurities. Still lower densities appear insulating to the lowest temperature measured.

Zener tunneling between landau orbits in a high-mobility two-dimensional electron gas

Magnetotransport in a laterally confined two-dimensional electron gas (2DEG) can exhibit modified scattering channels owing to a tilted Hall potential. Transitions of electrons between Landau levels with shifted guiding centers can be accomplished through a Zener tunneling mechanism, and make a significant contribution to the magnetoresistance. A remarkable oscillation effect in weak field magnetoresistance has been observed in high-mobility 2DEGs in GaAs -Al Ga 0.3As (0.7) heterostructures, and can be well explained by the Zener mechanism.

New class of magnetoresistance oscillations: interaction of a two-dimensional electron gas with leaky interface phonons

We report on a new class of magnetoresistance oscillations observed in a high-mobility two-dimensional electron gas (2DEG) in GaAs-Al(x)Ga(1--x)As heterostructures. Appearing in a weak magnetic field ( B < 0.3 T) and only in a narrow temperature range ( 2 K < T < 9 K), these oscillations are periodic in 1/B with a frequency proportional to the electron Fermi wave vector, k(F). We interpret the effect as a magnetophonon resonance of the 2DEG with leaky interface-acoustic phonon modes carrying a wave vector q = 2k(F). Calculations show a few branches of such modes existing on the GaAs-Al(x)Ga(1--)xAs interface, and their velocities are in quantitative agreement with the observation.

Syndrome of short stature, widow's peak, ptosis, posteriorly angulated ears, and joint problems: exclusion of the Aarskog (FGD1) gene as a candidate gene

A syndrome encompassing postnatal onset of short stature, widow's peak, ptosis, posteriorly angulated ears, and limitation of forearm supination is reported in a boy and his mother. The boy has not yet experienced dislocation of patella or other joint anomaly except for limitation of supination of the forearms. On the other hand, the mother has a milder limitation of supination only on the left arm and is devoid of ptosis. Their condition is reminiscent of that described in the family reported by Kapur et al. [1989: Am. J. Med. Genet. 33: 357-363.], which showed an X-linked dominant mode of inheritance. DNA study on our family using an intragenic polymorphism of the Aarskog syndrome (FGD1) gene and four other adjacent markers convincingly excludes the possibility that their condition could be caused by a mutation of the FGD1 gene. Our family and the family reported by Kapur et al. may suggest segregation of a novel X-linked dominant condition.

Neural responses to overlapping FM sounds in the inferior colliculus of echolocating bats

The big brown bat, Eptesicus fuscus, navigates and hunts prey with echolocation, a modality that uses the temporal and spectral differences between vocalizations and echoes from objects to build spatial images. Closely spaced surfaces ("glints") return overlapping echoes if two echoes return within the integration time of the cochlea ( approximately 300-400 micros). The overlap results in spectral interference that provides information about target structure or texture. Previous studies have shown that two acoustic events separated in time by less than approximately 500 micros evoke only a single response from neural elements in the auditory brain stem. How does the auditory system encode multiple echoes in time when only a single response is available? We presented paired FM stimuli with delay separations from 0 to 24 micros to big brown bats and recorded local field potentials (LFPs) and single-unit responses from the inferior colliculus (IC). These stimuli have one or two interference notches positioned in their spectrum as a function of two-glint separation. For the majority of single units, response counts decreased for two-glint separations when the resulting FM signal had a spectral notch positioned at the cell's best frequency (BF). The smallest two-glint separation that reliably evoked a decrease in spike count was 6 micros. In addition, first-spike latency increased for two-glint stimuli with notches positioned nearby BF. The N(4) potential of averaged LFPs showed a decrease in amplitude for two-glint separations that had a spectral notch near the BF of the recording site. Derived LFPs were computed by subtracting a common-mode signal from each LFP evoked by the two-glint FM stimuli. The derived LFP records show clear changes in both the amplitude and latency as a function of two-glint separation. These observations in relation with the single-unit data suggest that both response amplitude and latency can carry information about two-glint separation in the auditory system of E. fuscus.

Spectral cues and perception of the vertical position of targets by the big brown bat, Eptesicus fuscus

Big brown bats (Eptesicus fuscus) were trained to discriminate between vertical angles subtended by paired beads suspended from fishing line. Bats were rewarded for choosing the smaller of the two angles presented. The difference between the angles was changed systematically using a transformed up-down procedure and the bats' ability to detect the difference was measured at different vertical locations. When the beads were centered at +20 degrees (above the horizon), at 0 degree (the horizon), and at -20 degrees (below the horizon), vertical angle acuity (VAA) was maintained between 2.9 degrees and 4.1 degrees. At more extreme vertical positions both bats showed loss of acuity; when the beads were centered around -40 degrees, VAA was 6.7 degrees or 8.3 degrees and at +40, VAA was worse than 21 degrees (the largest difference tested). When the tragi of both ears were bent down and glued to the side of the face, bats showed severe loss of acuity for beads centered at -20 degrees (VAA 18.3 degrees and 20.1 degrees), but maintained their angle acuity for beads centered at +20 degrees (VAA 3.8 degrees and 4.9 degrees). The results are consistent with the spectral cues created by the filtering of the external ear.

The track structures of ionizing particles and their application to radiation biophysics. II. Results for various organisms irradiated with protons, deuterons and alpha-particles

Using knowledge of the track structure generated by ionizing particles together with details of the organisms being irradiated, the application of a new analytical method to two biophysical models to explain the inactivation of cells by radiation has been developed. It is shown that both models are equally successful in predicting experimental results and that good agreement is found with the data for single-strand phage, Bacillus subtilis spores, various strains of Escherichia coli, haploid and diploid yeast, and human diploid fibroblasts. The only significant discrepancy arose with T1-phage, for which a tentative explanation is offered. The differences in inherent radiosensitivity between organisms, after allowance is made for differences in target size, are attributed to differences in enzymatic repair systems and in the packing of the DNA.

The track structures of ionizing particles and their application to radiation biophysics. I. A new analytical method for investigating two biophysical models

A new approach to the interpretation of the effects of radiation on cells is described, in which sample particle tracks are constructed using a Monte Carlo computer program and the exposure of cellular targets to these tracks is simulated using a second program known as BIOPHYS. Data on the shapes and DNA contents of the cell nuclei are obtained from the literature. It is assumed that the sensitive material is DNA, and that the target is divided into cubes of approximately 2 nm (the diameter of the DNA helix) per side; the numbers of these cubes containing different numbers of ionizations are derived. Two different methods of analysing the output of BIOPHYS are described. In the first, it is assumed that lethality is caused by the occurrence of a number of ionizations equal to or greater than a certain threshold in one cube; in the second method, it is assumed that only two ionizations are required, in different parts of the cube, but that only some fraction of the cube is sensitive. These models have been applied to the interpretation of the variation of radiosensitivity with a linear energy transfer (LET) of spores of Bacillus subtilis exposed wet and dry, and good fits to the published experimental data were obtained using both models. Fits to experimental data for a range of other cell lines will be presented in a second paper.

Echo-delay resolution in sonar images of the big brown bat, Eptesicus fuscus

Echolocating big brown bats (Eptesicus fuscus) broadcast ultrasonic frequency-modulated (FM) biosonar sounds (20-100 kHz frequencies; 10-50 microseconds periods) and perceive target range from echo delay. Knowing the acuity for delay resolution is essential to understand how bats process echoes because they perceive target shape and texture from the delay separation of multiple reflections. Bats can separately perceive the delays of two concurrent electronically generated echoes arriving as little as 2 microseconds apart, thus resolving reflecting points as close together as 0.3 mm in range (two-point threshold). This two-point resolution is roughly five times smaller than the shortest periods in the bat's sounds. Because the bat's broadcasts are 2,000-4,500 microseconds long, the echoes themselves overlap and interfere with each other, to merge together into a single sound whose spectrum is shaped by their mutual interference depending on the size of the time separation. To separately perceive the delays of overlapping echoes, the bat has to recover information about their very small delay separation that was transferred into the spectrum when the two echoes interfered with each other, thus explicitly reconstructing the range profile of targets from the echo spectrum. However, the bat's 2-microseconds resolution limit is so short that the available spectral cues are extremely limited. Resolution of delay seems overly sharp just for interception of flying insects, which suggests that the bat's biosonar images are of higher quality to suit a wider variety of orientation tasks, and that biosonar echo processing is correspondingly more sophisticated than has been suspected.

Frequency tuning, latencies, and responses to frequency-modulated sweeps in the inferior colliculus of the echolocating bat, Eptesicus fuscus

Neurons in the inferior colliculus (IC) of the awake big brown bat, Eptesicus fuscus, were examined for joint frequency and latency response properties which could register the timing of the bat's frequency-modulated (FM) biosonar echoes. Best frequencies (BFs) range from 10 kHz to 100 kHz with 50% tuning widths mostly from 1 kHz to 8 kHz. Neurons respond with one discharge per 2-ms tone burst or FM stimulus at a characteristic latency in the range of 3-45 ms, with latency variability (SD) of 50 microseconds to 4-6 ms or more. BF distribution is related to biosonar signal structure. As observed previously, on a linear frequency scale BFs appear biased to lower frequencies, with 20-40 kHz overrepresented. However, on a hyperbolic frequency (linear period) scale BFs appear more uniformly distributed, with little overrepresentation. The cumulative proportion of BFs in FM1 and FM2 bands reconstructs a scaled version of the spectrogram of FM broadcasts. Correcting FM latencies for absolute BF latencies and BF time-in-sweep reveals a subset of IC cells which respond dynamically to the timing of their BFs in FM sweeps. Behaviorally, Eptesicus perceives echo delay and phase with microsecond or even submicrosecond accuracy and resolution, but even with use of phase-locked FM and tone-burst stimuli the cell-by-cell precision of IC time-frequency registration seems inadequate by itself to account for the temporal acuity exhibited by the bat.

Mapping of 228 ESTs and 26 genes into an integrated physical and genetic map of human chromosome 17

We have integrated genetic and physical mapping data for chromosome 17 subdivided into 26 bins, by using a panel of chromosome 17 deletion somatic cell hybrids. One hundred four short tandem repeat and STS markers have been localized into these bins and have enabled the ordering of 288 ESTs and 26 genes, including 142 ESTs that had not been previously sublocalized on chromosome 17. The mapping information of several genetic maps, as well as information obtained by radiation hybrid and STS content mapping of YACs, has been integrated using this hybrid panel. Although existing mapping information for chromosome 17 was generally consistent for many ESTs previously mapped, the map presented here further refines the location of ESTs, as well as demonstrating a number of discrepancies found in the 17q24-q25 region. We attribute these discrepancies to the fact that the current radiation hybrid panels were selected for retention of the thymidine kinase gene at 17q25, as well as to a low concentration of YAC contigs in this region. These data illustrate the benefit of combining multiple mapping techniques to obtain the greatest accuracy. The integration of maps developed by different methods will generate the most accurate genome maps, which may then be used for the generation of large insert clone contigs for chromosome sequencing. Additionally, accurate transcript maps generated by ESTs will greatly speed the isolation of genes linked to disease loci.

Localization of a growth suppressor activity in MCF7 breast cancer cells to chromosome 17q24-q25

Chromosome 17 is one of the most frequently altered chromosomes in malignant breast cancer. At least four genes implicated in breast cancer reside on chromosome 17 (p53, 17p13; Her-2/neu/ERBB2, 17q12; BRCA1, 17q21; and nm23, 17q22). In addition, allelic imbalance has been described for at least five regions of chromosome 17. We have previously shown that the introduction of a normal human chromosome 17 into the breast cancer cell line MCF7 by microcell mediated chromosome transfer (MMCT) results in the in vitro growth arrest of these cells within 8 weeks, suggesting the presence of a growth suppressor on chromosome 17. Additionally, we have shown that the tumor suppressor gene p53 is not responsible for this phenotype, as it is wild type in MCF7 cells, and overexpression has no effect on either the in vitro or in vivo growth of these cells. We have further localized this growth suppressor gene to 17q24-q25 by transfer of chromosome 17 hybrids containing defined deletions. Whereas transfer of hybrids that contained an intact 17q (delta43/A9 and delta26/A9) resulted in growth arrest, two hybrids with overlapping deletions at 17q24-q25, had no effect on growth of MCF7 cells. Molecular analyses revealed that 50/70 (71%) of the resulting delta2/MCF7 or delta624/MCF7 MMCT clones retained an intact introduced chromosome 17. In contrast, only 8/34 (24%) of delta43/MCF7 revertants (deleted for 17p13.1-pter) which escaped growth arrest showed no breakage of the introduced chromosome 17. We did not observe a preferential loss of an intragenic BRCA1 marker in the MMCT hybrids, excluding BRCA1 as the gene responsible for this growth arrest phenotype. These data therefore implicate a new growth suppressor gene involved in breast cancer that is localized to chromosome 17q24-q25.

Sound source elevation and external ear cues influence the discrimination of spectral notches by the big brown bat, Eptesicus fuscus

Measurements of external ear transfer functions in the echolocating bat Eptesicus fuscus have revealed a prominent spectral notch that decreases in center frequency (50 to 30-35 kHz) as elevation decreases [Wotton et al., J. Acoust. Soc. Am. 98, 1423-1445 (1995)]. To examine the influence of this notch, four Eptesicus were trained to discriminate between two sets of electronically generated artificial echoes. The negative (unrewarded) stimulus contained a test spectral notch at a specific frequency that varied from 30 to 50 kHz, while the positive (rewarded) stimulus contained no test notch. The vertical position of the loudspeakers delivering these simulated echoes was changed daily. When echoes were returned from an elevation at which the external ear introduced a spectral notch at the same frequency as the test notch, then the discrimination should have been difficult. The bats' performance conformed to this prediction: All bats discriminated the presence of a 35-kHz notch at all elevations except -10 degrees. As the frequency of the synthesized notch increased, the elevation at which bats could not perform the discrimination also increased. The movement of the bat's "blind spot" for the test notch of different frequencies followed the movement of the external ear notch at different elevations.

Representation of perceptual dimensions of insect prey during terminal pursuit by echolocating bats

The echolocating big brown bat, Eptesicus fuscus, broadcasts brief frequency-modulated (FM) ultrasonic sounds and perceives objects from echoes of these sounds returning to its ears. Eptesicus is an insectivorous species that uses sonar to locate and track flying prey. Although the bat normally hunts in open areas, it nevertheless is capable of chasing insects into cluttered environments such as vegetation, where it completes interceptions in much the same manner as in the open except that it has to avoid the obstacles as well as catch the insect. During pursuit, the bat shortens its sonar signals and increases their rate of emission as it closes in to seize the target, and it keeps its head pointed at the insect throughout the maneuver. In the terminal stage of interception, the bat makes rapid adjustments in its flight-path and body posture to capture the insect, and these reactions occur whether the bat is pursuing its prey in the open or close to obstacles such as vegetation. Insects can be distinguished from other objects by the spectrum and phase of their echoes, and Eptesicus is very good at discriminating these acoustic features. To identify the insect in the open, but especially to distinguish which object is the insect in clutter, the bat must have some means for representing these features throughout the interception maneuver. Moreover, continuity for perception of these features is necessary to keep track of the prey in complex surroundings, so the nature of the auditory representations for the spectrum and phase of echoes has to be conserved across the approach, tracking, and terminal stages. The first problem is that representation of changes in the phase of echoes requires neural responses in the bat's auditory system to have temporal precision in the microsecond range, which seems implausible from conventional single-unit studies in the bat's inferior colliculus, where the temporal jitter of responses typically is hundreds of microseconds. Another problem is that echoes do not explicitly evoke neural responses in the inferior colliculus distinct from responses evoked by the broadcast during the terminal stage because the delay of echoes is too short for responsiveness to recover from the emissions. In contrast, each emission and each echo evokes its own responses during the approach and tracking stages of pursuit. How does the bat consistently represent the phase of echoes in spite of these evident limitations in neural responses? Local multiunit responses recorded from the inferior colliculus of Eptesicus reveal a novel format for encoding the phase of echoes at all stages of interception. Changes in echo phase (0 degree or 180 degrees) produce shifts in the latency of responses to the emission by hundreds of microseconds, an unexpected finding that demonstrates the existence of expanded time scales in neural responses representing the target at all stages of pursuit.

Survival and yields of chromosome aberrations in hamster and human lung cells irradiated by alpha particles

The effects of alpha-particle irradiation on hamster and human lung cells have been studied. In both cases two end points were taken, cell death and the induction of chromosome aberrations. The hamster cells were common stock V79 cells; the human ones were freshly derived from fetal material. For both types of cells, the survival curves could be described by straight lines in the conventional exponential plot, with values of D(o) of 0.78 and 0.37 Gy for the hamster and human cells, respectively. The rate of induction of chromosome aberrations could also be described by straight lines with slopes of 0.30 and 0.62 aberration per cell per gray. Thus, for this second end point also, it appears that human cells are twice as sensitive to the effects of alpha-particle irradiation as hamster cells.

Spatially dependent acoustic cues generated by the external ear of the big brown bat, Eptesicus fuscus

To measure the directionality of the external ear of the echolocating bat, Eptesicus fuscus, the left or right eardrum of a dead bat was replaced by a microphone which recorded signals received from a sound source that was moved around the stationary head. The test signal was a 0.5-ms FM sweep from 100 kHz to 10 kHz (covering all frequencies in the bat's biosonar sounds). Notches and peaks in transfer functions for 7 tested ears varied systematically with changes in elevation. For the most prominent notch, center frequency decreased from about 50 kHz for elevations at or near the horizontal to 30-40 kHz for elevations 30 degrees-40 degrees below the horizontal. A second notch shifted from about 85 kHz to 70 kHz over these same elevations. Above the horizontal, a peak that flanks these notches changed in amplitude by 15 dB with changes in elevation. Removal of the tragus from the external ear disrupted the systematic movement of notch frequencies with elevation but did not disrupt changes in the peak's amplitude. Smaller changes in notch frequency also occurred with changes in azimuth, so monaural notch information alone cannot determine the position of sound sources away from the median plane. However, because bats routinely keep the head pointed at the target's azimuth, median-plane localization occurs with monaural cues delivered to the two ears. Corresponding changes with elevation occurred in the impulse-response, which consists of a series of 3-6 peaks spaced 10-20 microseconds apart. The time separation of two prominent impulse peaks systematically increased from 22-26 microseconds above the horizontal to about 36-40 microseconds below the horizontal, and removal of the tragus disrupted this time shift below the horizontal.