Roy G Brower

Johns Hopkins University, Baltimore, Maryland, United States

Are you Roy G Brower?

Claim your profile

Publications (127)804.4 Total impact

  • Neil R Aggarwal, Roy G Brower
    [Show abstract] [Hide abstract]
    ABSTRACT: It was recently suggested that targeting normoxemia (PaO2 85-110 mmHg) in ARDS patients might prevent neurocognitive dysfunction in survivors. However, targeting normoxemia may cause detrimental effects to the lungs from oxygen toxicity. Some have suggested that oxygen is not harmful to the lungs at FiO2s less than 0.6-0.7, but contrasting evidence in normal humans suggests that there can be untoward effects of moderate FiO2s. Furthermore, in experimental models of the acute respiratory distress syndrome, co-existing lung inflammation increases susceptibility to oxygen toxicity. Co-existing lung inflammation may lower the threshold for oxygen toxicity in patients with ARDS or in other acute illnesses in the lung. Moreover, physicians frequently prescribe higher FiO2s than are necessary to achieve their arterial oxygenation goal, further increasing the risk of oxygen toxicity. Targeting normoxemia in ARDS patients may prevent some long-term neurocognitive deficits in survivors, but it may increase lung inflammation and cause worse short-term clinical outcomes. We advocate for a clinical trial in ARDS patients to determine more appropriate goals for arterial oxygenation.
    Annals of the American Thoracic Society. 10/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of the study is to compare neuromuscular electrical stimulation (NMES) vs sham on leg strength at hospital discharge in mechanically ventilated patients.
    Journal of critical care. 09/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Chest wall strapping induces breathing at low lung volumes. Mild to moderate obesity can lead to similar changes in lung volumes, due to chest wall and abdominal restriction. Chest wall strapping is also conceptually similar to a mismatch between significantly oversized donor lungs transplanted into a recipient with a smaller chest cavity. Chest wall strapping increases lung elastic recoil, reduces pulmonary compliance, and substantially increases maximal expiratory flows. The interactions between elastic properties of the lung parenchyma and small airways are critical for pulmonary function. Chest wall strapping lowers residual volume and closing volume, likely from the interdependence between increased elastic recoil and airways leading to greater radial distending forces on small airways and small airway dilation. Chronic obstructive pulmonary disease (COPD) and chronic rejection of the transplanted lung, bronchiolitis obliterans syndrome (BOS), are primarily diseases of the small airways and are characterized by progressive obstruction and subsequent loss of small airways. In COPD, higher body mass index (BMI) (conceptually like being more tightly strapped) is associated with lower lung volumes, increased airway conductance, and with lower risk of progression to emphysema or death. Likewise, in lung transplantation, oversized donor lungs have been linked to higher expiratory airflows, lower risk of BOS, and improved survival. This paper reviews the physiology of chest wall strapping and explores how it could enhance the understanding or even the treatment of small airway diseases such as COPD and BOS.
    Annals of the American Thoracic Society. 08/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: The use of extracorporeal membrane oxygenation (ECMO) for severe acute respiratory failure (ARF) in adults is growing rapidly given recent advances in technology, even though there is controversy regarding the evidence justifying its use. Since ECMO is a complex, high risk, and costly modality, at present it should be conducted in centers with sufficient experience, volume and expertise to ensure it is used safely. This position paper represents the consensus opinion of an international group of physicians and associated healthcare workers, who have expertise in therapeutic modalities used in the treatment of patients with severe ARF, with a focus on ECMO. The aim of this paper is to provide physicians, ECMO center directors and coordinators, hospital directors, health care organizations, regional, national and international policy makers a description of the optimal approach to organizing ECMO programs for ARF in adult patients. Importantly, this will help ensure that ECMO is delivered safely and proficiently, such that future observational and randomized clinical trials assessing this technique may be performed by experienced centers under homogeneous and optimal conditions. Given the need for further evidence, we encourage restraint in the widespread use of ECMO until we have a better appreciation for both the potential clinical applications and the optimal techniques for performing ECMO.
    American Journal of Respiratory and Critical Care Medicine 07/2014; · 11.04 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Between 10-57% of lung transplant (LTx) recipients develop primary graft dysfunction (PGD) within 72 hours of LTx. PGD is clinically and histologically analogous to the acute respiratory distress syndrome (ARDS). In patients at risk for, or with ARDS, lung protective ventilation strategies (low tidal volumes and positive end expiratory pressure) improve outcomes. There is, however, little information on mechanical ventilation strategies after LTx. The aim of this international survey was to describe the current practices of mechanical ventilation immediately after LTx. An electronic survey was sent to the medical and surgical directors of U.S. LTx programs (n=111) and to members of the Pulmonary Council of the International Society for Heart and Lung Transplantation (n=470). One hundred and forty-nine individuals from 18 countries responded to the questionnaire. The most common modes of ventilation were Pressure Assist/Control (37%) and Volume Assist/Control (35%). Tidal volumes were most often determined by recipient characteristics; donor characteristics were rarely considered (35%) and infrequently known by the team managing the ventilator (42%). When presented with a choice of ideal tidal volumes, a majority of respondents selected 6 ml/kg recipient predicted body weight (58%), fewer selected 10 ml/kg (21%), and none selected 15 ml/kg. A majority preferred limiting the fraction of inspired oxygen rather than positive end expiratory pressure (PEEP) (69% versus 31%, p = 0.006). The median minimum PEEP was 5 cm H2O and the median maximum PEEP was 11.5 cm H2O. The presence of PGD increased the perceived importance of monitoring plateau pressure to adjust tidal volumes. The median plateau pressure limit perceived as a threshold triggering reduction in tidal volumes was 30 cm H2O. Most respondents reported using lung-protective approaches to mechanical ventilation after lung transplantation. Low tidal volumes based on recipient characteristics were frequently chosen. Donor characteristics were often not considered and were frequently not known by the team managing mechanical ventilation after LTx.
    Annals of the American Thoracic Society. 03/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background Donor to recipient lung size matching at lung transplantation (LTx) can be estimated by the predicted total lung capacity (pTLC)ratio (donor pTLC/recipient pTLC). We aimed to determine whether the pTLC-ratio is associated with the risk of primary graft dysfunction (PGD) after bilateral LTx (BLT). Methods We calculated the pTLC-ratio for 812 adult BLTs from the Lung Transplant Outcomes Group between 3/2002-12/2010. Patients were stratified by pTLC-ratio>1.0 ("oversized") and pTLC-ratio≤1.0 ("undersized"). PGD was defined as any ISHLT grade 3 PGD within 72 hours of reperfusion (PGD 3). We analyzed the association between risk factors and PGD using multivariable conditional logistic regression. As transplant diagnoses can influence the size matching decisions and also modulate the risk for PGD, we performed pre-specified analyses by assessing the impact of lung size mismatch within diagnostic categories. Results In univariate analyses oversizing was associated with a 39% lower odds of PGD3 (OR 0.61, 95% CI, p=0.003). In a multivariate model accounting for center effects and known PGD risks, oversizing remained independently associated with a decreased odds of PGD3 (OR 0.58, 95% CI 0.38-0.88, p=0.01). The risk adjusted point estimate was similar for the non-COPD diagnoses groups (OR 0.52, 95%CI 0.32-0.86, p=0.01); however there was no detected association within the COPD group (OR 0.72, 95% CI 0.29-1.78, p=0.5). Conclusion Oversized allografts are associated with a decreased risk of PGD3 after BLT; this effect appears most apparent in non-COPD patients.
    The Journal of Heart and Lung Transplantation. 01/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Critically ill patients commonly experience poor sleep quality in the intensive care unit (ICU) because of various modifiable factors. To address this issue, an ICU-wide, multifaceted quality improvement (QI) project was undertaken to promote sleep in the Johns Hopkins Hospital Medical ICU (MICU). To supplement previously published results of this QI intervention, the present article describes the specific QI framework used to develop and implement this intervention, which consists of 4 steps: (a) summarizing the evidence to create a list of sleep-promoting interventions, (b) identifying and addressing local barriers to implementation, (c) selecting performance measures to assess intervention adherence and patient outcomes, and (d) ensuring that all patients receive the interventions through staff engagement and education and regular project evaluation. Measures of performance included daily completion rates of daytime and nighttime sleep improvement checklists and completion rates of individual interventions. Although long-term adherence and sustainability pose ongoing challenges, this model provides a foundation for future ICU sleep promotion initiatives.
    American Journal of Medical Quality 11/2013; · 1.47 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Rationale: The association between a high predicted total lung capacity (pTLC)-ratio (=pTLCdonor / pTLCrecipient), suggestive of an oversized allograft, with improved survival after lung transplantation (LTx) remains controversial. Objective: To characterize the donor-recipient lung size matching based on the pTLC-ratio and to investigate the relationship of the pTLC-ratio with post-LTx survival. Methods: All subjects in the SRTR transplant registry, who underwent first-time LTx in the lung allocation score based system between May 4th 2005 and March 31st 2012 were studied, and the pTLC-ratio was calculated based on gender, height and age. Risk of death after LTx was analyzed using Cox proportional hazards models. Measurements and Main Results: The pTLC-ratio was available for 10,289 of the 10,318 study subjects (99.7%). The mean pTLC-ratio was 1.015 ± 0.175 (interquartile range [IQR] 0.918-1.119). Univariate analysis showed that the pTLC-ratio was strongly associated with death in the first LTx year (p < 0.0001). With the pTLC-ratio entered as a spline there was a non-linear association with declining risk of death with higher pTLC-ratio from 0.5 to about 1.3, where an inflection occurred with rising risk at higher values. Accounting for the pTLC-ratio, recipient and donor gender were not independently associated with death after LTx. The comparison of pTLC-ratio from 1.119 to 0.918 (the IQR) was associated with similar point estimates of reduced risk of death at one year across univariate (HR 0.78) and comprehensive risk adjusted multivariate models (HR 0.86). Conclusion: The pTLC-ratio is an independent predictor of death in the first year after LTx and explains the association of gender with survival after LTx. Incorporating the pTLC-ratio in the lung allocation mechanism could improve outcomes after LTx.
    Annals of the American Thoracic Society. 08/2013;
  • Source
    Roy G Brower
    European Journal of Intensive Care Medicine 08/2013; · 5.17 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: A higher predicted total lung capacity (pTLC)-ratio (=pTLC donor/pTLC recipient), suggestive of oversized allografts, is associated with improved survival after lung transplantation. It is unknown whether the pTLC-ratio has a different association with survival in bilateral (BLT) versus single lung transplantation (SLT). The pTLC-ratio was calculated for all adult patients in the United Network of Organ Sharing lung transplant (LTx) registry who underwent first-time LTx in the post lung allocation score era, between May 2005 and April 2010. The LTx recipients were stratified according to procedure (BLT vs SLT). Risk of death at 1 year after LTx was analyzed using Kaplan-Meier survival and Cox proportional hazards models. In the 4,520 BLT patients, each 0.1 increase in pTLC-ratio conferred a 7% decrease in the hazard for death at 1 year (p < 0.001) in univariate analysis. This association remained significant after controlling for diagnosis, comorbidities, acuity, donor, and transplant factors (hazard ratio [HR] 0.93, 95% confidence interval [CI] 0.88 to 0.98, p = 0.01). Additional adjustment by a propensity score to account for biases to oversizing showed similar results (HR 0.94, 95% CI 0.90 to 0.99, p = 0.018). In the 2,477 SLT patients, each 0.1 increase in pTLC-ratio conveyed a 6% decrease in the hazard for death at 1 year (p = 0.002) in univariate analysis, which did not persist in the multivariate model (HR 1.00, p = 0.8). A higher pTLC-ratio, suggestive of an oversized allograft, is associated with improved survival after lung transplantation. This association is primarily evident in BLT patients.
    The Annals of thoracic surgery 06/2013; · 3.45 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVE:: Deep sedation and delirium are common in the ICU. Mechanically ventilated patients with acute lung injury are at especially high risk for deep sedation, delirium, and associated long-term physical and neuropsychiatric impairments. We undertook an ICU-wide structured quality improvement project to decrease sedation and delirium. DESIGN:: Prospective quality improvement project in comparison with a retrospective acute lung injury control group. SETTING:: Sixteen-bed medical ICU in an academic teaching hospital with pre-existing use of goal-directed sedation with daily interruption of sedative infusions. PATIENTS:: Consecutive acute lung injury patients. INTERVENTION:: A "4Es" framework (engage, educate, execute, evaluate) was used as part of the quality improvement process. A new sedation protocol was created and implemented, which recommends a target Richmond Agitation Sedation Scale score of 0 (alert and calm) and requires failure of intermittent sedative dosing prior to starting continuous infusions. In addition, twice-daily delirium screening using the Confusion Assessment Method for the ICU was introduced into routine practice. MEASUREMENTS AND MAIN RESULTS:: Sedative use and delirium status in acute lung injury patients after implementation of the quality improvement project (n = 82) were compared with a historical control group (n = 120). During the quality improvement vs. control periods, use of narcotic and benzodiazepine infusions were substantially lower (median proportion of medical ICU days per patient: 33% vs. 74%, and 22% vs. 70%, respectively, both p < 0.001). Further, wakefulness increased (median Richmond Agitation Sedation Scale score per patient: -1.5 vs. -4.0, p < 0.001), and days awake and not delirious increased (median proportion of medical ICU days per patient: 19% vs. 0%, p < 0.001). CONCLUSION:: Through a structured quality improvement process, use of sedative infusions can be substantially decreased and days awake without delirium significantly increased, even in severely ill, mechanically ventilated patients with acute lung injury.
    Critical care medicine 03/2013; · 6.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVES:: Lung-protective ventilation with lower tidal volume and lower plateau pressure improves mortality in patients with acute lung injury and acute respiratory distress syndrome. We sought to determine the incidence of elevated plateau pressure in acute lung injury /acute respiratory distress syndrome patients receiving lower tidal volume ventilation and to determine the factors that predict elevated plateau pressure in these patients. PATIENTS:: We used data from 1398 participants in Acute Respiratory Distress Syndrome Network trials, who received lower tidal volume ventilation (≤ 6.5 mL/kg predicted body weight). DESIGN:: We considered patients with a plateau pressure greater than 30 cm H2O and/or a tidal volume less than 5.5 mL/kg predicted body weight on study day 1 to have "elevated plateau pressure." We used logistic regression to identify baseline clinical variables associated with elevated plateau pressure and to develop a model to predict elevated plateau pressure using a subset of 1,188 patients. We validated the model in the 210 patients not used for model development. SETTING:: Medical centers participating in Acute Respiratory Distress Syndrome Network clinical trials. INTERVENTIONS:: None. MEASUREMENTS AND MAIN RESULTS:: Of the 1,398 patients in our study, 288 (20.6%) had elevated plateau pressure on day 1. Severity of illness indices and demographic factors (younger age, greater body mass index, and non-white race) were independently associated with elevated plateau pressure. The multivariable logistic regression model for predicting elevated plateau pressure had an area under the receiving operator characteristic curve of 0.71 for both the developmental and the validation subsets. CONCLUSIONS:: acute lung injury patients receiving lower tidal volume ventilation often have a plateau pressure that exceeds Acute Respiratory Distress Syndrome Network goals. Race, body mass index, and severity of lung injury are each independently associated with elevated plateau pressure. Selecting a smaller initial tidal volume for non-white patients and patients with higher severity of illness may decrease the incidence of elevated plateau pressure. Prospective studies are needed to evaluate this approach.
    Critical care medicine 01/2013; · 6.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVES:: To determine if a quality improvement intervention improves sleep and delirium/cognition. DESIGN:: Observational, pre-post design. SETTING:: A tertiary academic hospital in the United States. PATIENTS:: 300 medical ICU patients. INTERVENTIONS:: This medical ICU-wide project involved a "usual care" baseline stage, followed by a quality improvement stage incorporating multifaceted sleep-promoting interventions implemented with the aid of daily reminder checklists for ICU staff. MEASUREMENTS AND MAIN RESULTS:: Primary ICU outcomes were perceived sleep quality and noise ratings (measured on a 0-100 scale using the valid and reliable Richards-Campbell Sleep Questionnaire) and delirium/coma-free days. Secondary outcomes included ICU and hospital length of stay and mortality. Post-ICU measures of cognition and perceived sleep quality were evaluated in an ICU patient subset. During the baseline and sleep quality improvement stages, there were 122 and 178 patients, respectively, with more than one night in the ICU, accounting for 634 and 826 patient-days. Within the groups, 78 (63.9%) and 83 (46.6%) patients received mechanical ventilation. Over the 826 patient-day quality improvement period, checklist item completion rates ranged from 86% to 94%. In multivariable regression analysis of the quality improvement vs. baseline stages, improvements in overall Richards-Campbell Sleep Questionnaire sleep quality ratings did not reach statistical significance, but there were significant improvements in daily noise ratings (mean ± SD: 65.9 ± 26.6 vs. 60.5 ± 26.3, p = 0.001), incidence of delirium/coma (odds ratio: 0.46; 95% confidence interval, 0.23-0.89; p = 0.02), and daily delirium/coma-free status (odds ratio: 1.64; 95% confidence interval, 1.04-2.58; p = 0.03). Improvements in secondary ICU outcomes and post-ICU outcomes did not reach statistical significance. CONCLUSIONS:: An ICU-wide quality improvement intervention to improve sleep and delirium is feasible and associated with significant improvements in perceived nighttime noise, incidence of delirium/coma, and daily delirium/coma-free status. Improvement in perceived sleep quality did not reach statistical significance. (Crit Care Med 2013; 41:0-0).
    Critical care medicine 01/2013; · 6.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVES Mechanical ventilation tidal volumes are usually set according to an estimate of patient size in millilitres (ml) per kilogram (kg) body weight. We describe the relationship between donor-recipient lung-size mismatch and postoperative mechanical ventilation tidal volumes according to recipient- and donor-predicted body weights in a cohort of bilateral lung transplant patients.METHODSA most-undersized (10 patients with lowest predicted total lung capacity [pTLC] ratio = pTLC-donor/pTLC-recipient), a most-oversized (10 patients with highest pTLC ratio) and best-matched subset (10 patients with predicted total lung capacity ratio closest to 1.0) were selected within a cohort of 70 patients. All tidal volumes during mechanical ventilation in the first 96 h after bilateral lung transplantation were recorded. Tidal volumes were expressed in ml and ml/kg-recipient-predicted body weights and ml/kg-donor-predicted body weights.RESULTSPostoperative absolute tidal volumes (in ml) were comparable between subsets of patients with undersized, matched and oversized allografts (552 ± 103 vs 581 ± 107 vs 582 ± 104 ml), and tidal volumes in ml/kg-recipient-predicted body weights were also similar (8.8 ± 1.4 vs 9.3 ± 1.1 vs 9.8 ± 2.1). However, tidal volumes in ml/kg-donor-predicted body weights revealed significant differences between undersized, matched, and oversized subsets (11.4 ± 3.1 vs 9.4 ± 1.2 vs 8.1 ± 2.1, respectively; P < 0.05). Two patients developed primary graft dysfunction grade 3, both in the undersized subset. Four patients in the undersized group underwent tracheotomy (vs none in matched and one in oversized subset).CONCLUSIONS During mechanical ventilation after bilateral lung transplantation, undersized allografts received relatively higher tidal volumes compared with oversized allografts when the tidal volumes were related to donor-predicted body weights.
    Interactive Cardiovascular and Thoracic Surgery 12/2012; · 1.11 Impact Factor
  • Canadian Medical Association Journal 11/2012; · 6.47 Impact Factor
  • European Journal of Intensive Care Medicine 09/2012; 38(10):1731-2. · 5.17 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Our objective was to revise the definition of acute respiratory distress syndrome (ARDS) using a conceptual model incorporating reliability and validity, and a novel iterative approach with formal evaluation of the definition. The European Society of Intensive Care Medicine identified three chairs with broad expertise in ARDS who selected the participants and created the agenda. After 2 days of consensus discussions a draft definition was developed, which then underwent empiric evaluation followed by consensus revision. The Berlin Definition of ARDS maintains a link to prior definitions with diagnostic criteria of timing, chest imaging, origin of edema, and hypoxemia. Patients may have ARDS if the onset is within 1 week of a known clinical insult or new/worsening respiratory symptoms. For the bilateral opacities on chest radiograph criterion, a reference set of chest radiographs has been developed to enhance inter-observer reliability. The pulmonary artery wedge pressure criterion for hydrostatic edema was removed, and illustrative vignettes were created to guide judgments about the primary cause of respiratory failure. If no risk factor for ARDS is apparent, however, objective evaluation (e.g., echocardiography) is required to help rule out hydrostatic edema. A minimum level of positive end-expiratory pressure and mutually exclusive PaO(2)/FiO(2) thresholds were chosen for the different levels of ARDS severity (mild, moderate, severe) to better categorize patients with different outcomes and potential responses to therapy. This panel addressed some of the limitations of the prior ARDS definition by incorporating current data, physiologic concepts, and clinical trials results to develop the Berlin definition, which should facilitate case recognition and better match treatment options to severity in both research trials and clinical practice.
    European Journal of Intensive Care Medicine 08/2012; 38(10):1573-82. · 5.17 Impact Factor
  • Article: Response.
    Chest 08/2012; 142(2):543. · 7.13 Impact Factor
  • Chest 06/2012; 141(6):1379-82; discussion 1386-7. · 7.13 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Unnecessary variation in clinical care and clinical research reduces our ability to determine what healthcare interventions are effective. Reducing this unnecessary variation could lead to further healthcare quality improvement and more effective clinical research. We have developed and used electronic decision support tools (eProtocols) to reduce unnecessary variation. Our eProtocols have progressed from a locally developed mainframe computer application in one clinical site (LDS Hospital) to web-based applications available in multiple languages and used internationally. We use eProtocol-insulin as an example to illustrate this evolution. We initially developed eProtocol-insulin as a local quality improvement effort to manage stress hyperglycemia in the adult intensive care unit (ICU). We extended eProtocol-insulin use to translate our quality improvement results into usual clinical care at Intermountain Healthcare ICUs. We exported eProtocol-insulin to support research in other US and international institutions, and extended our work to the pediatric ICU. We iteratively refined eProtocol-insulin throughout these transitions, and incorporated new knowledge about managing stress hyperglycemia in the ICU. Based on our experience in the development and clinical use of eProtocols, we outline remaining challenges to eProtocol development, widespread distribution and use, and suggest a process for eProtocol development. Technical and regulatory issues, as well as standardization of protocol development, validation and maintenance, need to be addressed. Resolution of these issues should facilitate general use of eProtocols to improve patient care.
    International Journal of Clinical Monitoring and Computing 04/2012; 26(4):305-17.

Publication Stats

5k Citations
804.40 Total Impact Points

Institutions

  • 1988–2014
    • Johns Hopkins University
      • • Division of Pulmonary and Critical Care Medicine
      • • Department of Physical Medicine and Rehabilitation
      • • Department of Medicine
      Baltimore, Maryland, United States
  • 2012
    • Intermountain Medical Center
      Salt Lake City, Utah, United States
  • 2011
    • University of Maryland, Baltimore
      • Division of Pulmonary and Critical Care Medicine
      Baltimore, MD, United States
    • University of Texas Health Science Center at Houston
      Houston, Texas, United States
    • University of California, San Francisco
      San Francisco, California, United States
    • National Institutes of Health
      Maryland, United States
  • 1990–2011
    • Johns Hopkins Medicine
      • • Division of Pulmonary and Critical Care Medicine
      • • Department of Medicine
      Baltimore, Maryland, United States
  • 2008
    • Massachusetts General Hospital
      • Division of Pulmonary & Critical Care Medicine
      Boston, MA, United States
  • 2007
    • Johns Hopkins Bloomberg School of Public Health
      • Department of Environmental Health Sciences
      Baltimore, MD, United States
  • 2006
    • The University of Chicago Medical Center
      • Department of Medicine
      Chicago, IL, United States
  • 2004
    • University of Pennsylvania
      Philadelphia, Pennsylvania, United States