K V Rolston

University of Texas MD Anderson Cancer Center, Houston, Texas, United States

Are you K V Rolston?

Claim your profile

Publications (367)1694.4 Total impact

  • Kenneth V I Rolston, Lior Nesher, Jeffrey T Tarrand
    [Show abstract] [Hide abstract]
    ABSTRACT: Patients with solid tumors frequently undergo surgical procedures and develop procedure-related infections. We sought to describe the current microbiologic spectrum of infections at various sites following common surgical procedures. This was a retrospective review of microbiologic data between January 2011 and February 2012. The sites studied were those associated with breast cancer surgery, thoracotomy, craniotomy, percutaneous endoscopic gastrostomy (PEG) tube insertion, and abdominal/pelvic surgery. Only patients with solid tumors were included. A total of 368 surgical site infections (SSIs) were identified (68 breast cancer related; 91 thoracotomy related; 45 craniotomy related; 75 PEG-tube insertion related; and 89 abdominal/pelvic surgery related). Of these, 58% were monomicrobial and 42% were polymicrobial. Overall, 85% of the 215 monomicrobial infections were caused by Gram-positive organisms and 13% by Gram-negative bacilli (GNB). Staphylococcus aureus was the predominant pathogen in monomicrobial infections (150 of 215, 70%). Sixty (40%) of these staphylococcal isolates were methicillin resistant (MRSA), and 65% had a vancomycin minimal inhibitory concentration (MIC) ≥1.0 µg/ml. Pseudomonas aeruginosa was the predominant GNB pathogen (19 of 27, 70%). Staphylococci were also the predominant pathogens in polymicrobial infections, while P. aeruginosa and Escherichia coli were the predominant GNB. Overall, 35% of isolates from polymicrobial infections were GNB. Cephalosporins (e.g., cefazolin) or amoxicillin/clavulanate was used most often for surgical prophylaxis, and 47% of organisms from monomicrobial infections (MRSA, P. aeruginosa) were resistant to them. A similar resistance pattern was observed in polymicrobial infections. Staphylococcus species were isolated most often from the sites studied. Polymicrobial infections (42%) and GNB monomicrobial infections (13%) were relatively frequent causes of SSIs. Many of these infections were caused by organisms that are resistant to agents commonly used for surgical prophylaxis. Additionally, 65% of staphylococcal isolates had a vancomycin MIC ≥1.0 µg/ml, suggesting the need for alternative therapeutic agents.
    Infectious diseases and therapy. 11/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background Many transplant centers obtain surveillance blood cultures (SBCs) from asymptomatic allogeneic hematopoietic stem cell transplant (allo-HCT) recipients with central venous catheters for early detection of potential blood stream infections. The aim of this study was to determine the utility of this practice. Methods We conducted a retrospective study of all patients who underwent allo-HCT to determine the frequency, clinical significance, and costs associated with SBCs. Results From 776 patients, 6,801 SBCs were obtained (median, 9 per patient). Most (96.89%) were negative. Of the 211 positive SBCs, 171 (81%) had minimal clinical significance. The remaining 40 positive cultures (19%) were considered potentially significant. The frequency of potentially significant SBCs was 5.1% for the entire cohort and 0.59% of all SBCs drawn. Conclusion All potentially significant cultures and some that were deemed to have minimal significance led to medical intervention, some of which were probably unnecessary. No adverse outcomes occurred in patients with positive SBCs for the first 30 days following the positive result, regardless of the pathogen isolated or the quantitative colony count. The frequency of clinically significant positive SBCs in asymptomatic adult allo-HCT recipients is very low. Routine use of this practice leads to some unnecessary medical interventions and added costs.
    American Journal of Infection Control. 10/2014; 42(10):1084–1088.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background. Concern for serious infection due to β-lactam resistant viridans group streptococci (VGS) is a major factor driving empiric use of an anti-Gram-positive antimicrobial in febrile neutropenic patients. We sought to develop and validate a prediction model for the presence of β-lactam resistance in VGS causing bloodstream infection (BSI) in neutropenic patients. Methods. Data from 569 unique cases of VGS BSI in neutropenic patients from 2000 to 2010 at the MD Anderson Cancer Center were used to develop the clinical prediction model. Validation was done using 163 cases from 2011-2013. In vitro activity of β-lactam agents was determined for 2011-2013 VGS bloodstream isolates. Results. In vitro resistance to β-lactam agents commonly used in the empiric treatment of febrile neutropenia was observed only for VGS isolates with a penicillin minimum inhibitory concentration (MIC) of≥2 µg/mL. 129/732 patients (17%) were infected with VGS strains with a penicillin MIC≥2 µg/mL. For the derivation and validation cohorts, 98% of patients infected by VGS with a penicillin MIC of≥2 µg/mL had at least one of the following risk factors: current use of a β-lactam as antimicrobial prophylaxis, receipt of a β-lactam antimicrobial in the previous 30 days, or nosocomial VGS BSI onset. Limiting empiric anti-Gram-positive therapy to neutropenic patients having at least one of these three risk factors would have reduced such use by 42%. Conclusions. Simple clinical criteria can assist with targeting of anti-Gram-positive therapy to febrile neutropenic patients at risk of serious β-lactam resistant VGS infection.
    Clinical Infectious Diseases 04/2014; · 9.37 Impact Factor
  • George M Viola, Issam I Raad, Kenneth V Rolston
    [Show abstract] [Hide abstract]
    ABSTRACT: Objective. The rate of postmastectomy tissue expander (TE) infection remains excessively high, ranging between 2% and 24%. We hypothesized that current perioperative antimicrobial regimens utilized for breast TE reconstruction may be outdated as a result of recent changes in microflora and susceptibility patterns. Design and methods. We reviewed the records of all patients who had a TE reconstructive procedure and developed a definite breast TE infection between 2003 and 2010 at MD Anderson Cancer Center. Antimicrobials were stratified into 3 groups: systemic perioperative, local irrigation, and oral immediate postoperative antimicrobials. These were considered discordant if they did not target the isolated organisms, while a breakthrough infection was defined as an infection that occurred despite concordant antimicrobial coverage. Results. Overall, 75 patients with a definite TE infection were identified. The most common organisms identified were methicillin-resistant Staphylococcus epidermidis (29%), methicillin-resistant Staphylococcus aureus (15%), and gram-negative rods (26%). The use of systemic perioperative antimicrobials was deemed discordant in 51% of the cases. Although 79% of the patients received broad-spectrum perioperative local antimicrobial irrigation, 63% developed a breakthrough infection. Even though 61% received oral postoperative prophylactic antimicrobials, 63% of the times they were deemed discordant. Conclusions. Contrary to the proven effectiveness of a single dose of perioperative antibiotics, the common use of local antimicrobial irrigation and prolonged postoperative oral antibiotics appears to be an inadequate component of our preventive armamentarium. Also, because methicillin-resistant staphylococcal and pseudomonal infections occurred approximately 60% of the time, at institutions that have observed an increase of these organisms, it may be prudent that perioperative antimicrobials target these microorganisms.
    Infection Control and Hospital Epidemiology 01/2014; 35(1):75-81. · 4.02 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Unlike infections related to chemotherapy-induced neutropenia, postoperative infections occurring in patients with solid malignancy remain largely understudied. Our aim is to evaluate the outcomes and the volume-outcomes relationship associated with postoperative infections following resection of common solid tumors. We used Texas Discharge Data to study patients undergoing resection of cancer of the lung, esophagus, stomach, pancreas, colon, or rectum from 01/2002 to 11/2006. From their billing records, we identified ICD-9 codes indicating a diagnosis of serious postoperative infection (SPI), i.e., bacteremia/sepsis, pneumonia, and wound infection, occurring during surgical admission or leading to readmission within 30 days of surgery. Using regression-based techniques, we estimated the impact of SPI on mortality, resource utilization, and costs, as well as the relationship between hospital volume and SPI, after adjusting for confounders and data clustering. SPI occurred following 9.4 % of the 37,582 eligible tumor resections and was independently associated with nearly 12-fold increased odds of in-hospital mortality [95 % confidence interval (95 % CI), 7.2-19.5, P < 0.001]. Patients with SPI required six additional hospital days (95 % CI, 5.9-6.2) at an incremental cost of $16,991 (95 % CI, $16,495-$17,497). Patients who underwent resection at high-volume hospitals had a 16 % decreased odds of developing SPI than those at low-volume hospitals (P = 0.03). Due to the substantial burden associated with SPI following common solid tumor resections, hospitals must identify more effective prophylactic measures to avert these potentially preventable infections. Additional volume-outcomes research is needed to identify infection prevention processes that can be transferred from high- to lower-volume providers.
    Supportive Care in Cancer 10/2013; · 2.09 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate the role of ertapenem versus other standard antibiotic prophylaxis in patients with cancer undergoing intra-abdominal surgery. Our study was a retrospective cohort study consisting 615 patients who underwent intra-abdominal surgery at our institution between January 2007 and December 2010. The groups were divided among patients who received ertapenem as perioperative prophylaxis (ertapenem group) and patients who received other antibiotics (nonertapenem group). Groups were similar with respect to age, gender, and type of surgery. A total of 315 patients underwent colorectal and 300 noncolorectal surgeries. In a multivariate logistic regression model, the main factors associated with risk of surgical site infections (SSI) were as follows: antibiotics within 3 months of surgery (odds ratio [OR] 1.2, 95 % confidence interval [CI] 1.04-1.54; p = 0.05), prior hospitalization within 1 year (OR 1.21, 95 % CI 1.02-1.43; p = 0.05), diabetes mellitus (OR 2.1, 95 % CI 1.7-3.4; p = 0.04), and perioperative prophylaxis other than ertapenem (OR 1.7, 95 % CI 1.2-2.3; p = 0.04). Notably, patients who underwent colorectal surgery and received ertapenem had a lower rate of SSI (4 % ertapenem vs. 13 % nonertapenem, p = 0.01), whereas the frequency of infections was not different in patients who underwent other intra-abdominal surgery whether they received ertapenem or not. The use of ertapenem for perioperative prophylaxis in patients with colorectal surgery was associated with lower rates of SSI, while there was no difference in rates of infection in other intra-abdominal surgery.
    Annals of Surgical Oncology 10/2013; · 4.12 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: Many centers obtain routine surveillance stool cultures (SSC) from allo-HSCT recipients to check for vancomycin resistant enterococcal (VRE) colonization. Our center also performs SSC for Ps. aeru colonization. The aim of this study was to determine the utility of this practice. Methods: We conducted a two-year (2010-2012) retrospective review of all patients who underwent allo-HSCT at MD Anderson Cancer Center to determine frequency of colonization and incidence of infection with Ps. aeru. All patients underwent weekly SSC whenever they were hospitalized. Results: Of the 794 allo-HSCT, 95 (12%) were identified to have either Ps. aeru infection or positive SSC. Only 58 (7.3%) patients had at least 1 positive SSC with 19 (33%) patients developing Ps. aeru infection subsequently. The overall incidence of Ps. aeru infection in the cohort was 7% (56/794), thus, 37 patients (66%) with infection had a negative SSC. Positive predictive value of SSC was 33% and negative predictive value was 95%. Of the 56 patients with infection, 26 (46%) had pneumonia, 20 (36%) blood stream infection, 8 (14%) urinary tract infection, and 2 (4%) at other sites. The incidence of multi drug resistant (MDR) Ps. aeru in the entire cohort (SSC or infection) was 2.2% (18/794), 12 in the SSC group (7 of which developed a MDR Ps. aeru infection later), and 6 in the infection group that were SSC negative for MDR Ps. aeru. Among the19 patients with positive SSC who went on to develop Ps. aeru infection, 8 (42%) had mismatched resistance patterns. This mismatch suggests a potentially different source for some of the infections. No infection related mortality was observed in the first 30 days after infection. Conclusion: The frequency of Ps. aeru colonization on SSC and the development of subsequent infection was low as was the positive predictive value of SSC. Although patients who were not colonized had a low chance of developing Ps. aeru infection, most of the patients who developed Ps. Aeru infection did not have fecal colonization. The value of this SSC for Ps. aeru needs to be revisited at our institution.
    IDWeek 2013 Meeting of the Infectious Diseases Society of America; 10/2013
  • Lior Nesher, Jeffrey J. Tarrand, Kenneth V. I. Rolston
    [Show abstract] [Hide abstract]
    ABSTRACT: Background: Coagulase negative staphylococci (CNS) are common skin colonizers and often cause catheter-related bloodstream infection (CRBSI), but cause other serious infections infrequently. Staphylococcus lugdunensis (S. lug) is a CNS species that is considered more virulent. Most reports are individual cases or small series, and none describe the spectrum of infection or outcomes in cancer patients. Methods: We conducted a retrospective chart review of cancer patients between 1/1/2011 – 3/30/2013 from whom S. lug was isolated from the blood stream or other sterile body sites .Only monomicrobial infections were studied, although 25 polymicrobial infections were also identified. Results: 45 patients had a clinically significant monomicrobial S. lug infection with 26 (58%) being female, 37 (82%) with an underlying solid tumor and 7 (15%), a hematologic malignancy (HM). Only 2(4.4%)were neutropenic. Five (11%) had a blood stream infection (BSI), four of which had HM. Of these 1 had endocarditis / perivalvular abscess and required valve replacement. 36 (80%) had abscesses. All underwent drainage initially 16 (45%) or following inadequate response 20 (55%) to medical therapy. Post-surgical infections occurred in 23 (51%), and implanted devices (expanders, artificial knee, omaya reservoir) were involved in 7 (30%) . Other sites included the urinary tract (2), meningitis and empyema (one each). Echocardiography was done in the 5 patients with BSI with significant findings in only one patient. Oxacillin susceptibility was 89%. All isolates were vancomycin susceptible. Numerous agents were used for therapy. No deaths occurred within 30 days, and 4 (9%) died during follow up (average of 298 days). None died from infection related causes. Conclusion: Unlike other CNS species S. lug cause CRBSI infrequently. Deep-seated abscesses and post-surgical infections involving foreign medical devices are more common. Uniform vancomycin susceptibility was noted. Drainage of abscess in addition to antimicrobial therapy seems to be essential for infection resolution. In cancer patients S. lug should be treated as a potentially serious pathogen and should not be dismissed as a contaminant.
    IDWeek 2013 Meeting of the Infectious Diseases Society of America; 10/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: The goal of this study was to describe the outcomes associated with daptomycin treatment of documented gram-positive infections in patients with neutropenia. All patients with neutropenia (≤500 cells/m(3)) and at least one documented gram-positive culture from 2006-2009 were identified from a retrospective, multicenter, and observational registry (Cubicin® Outcome Registry and Experience (CORE®)). Investigators assessed patient outcome (cured, improved, failed, nonevaluable) at the end of daptomycin therapy. All patients were included in the safety analysis. The efficacy population had 186 patients; 159 (85 %) patients had either cure (n = 108, 58 %) or improved (n = 51, 27 %) as an outcome. Success rates (cure plus improved) by the lowest WBC during daptomycin were 98/116 (84 %) for ≤100 cells/m(3) and 61/70 (87 %) for 101-499 cells/m(3), P = 0.6. Most patients had cancer; 135/186 (73 %) had hematological malignancy; 26/186 (14 %) had solid tumors, and 9 (5 %) had both. One hundred fifty-six (84 %) patients received other antibiotics before daptomycin treatment; 82 % vancomycin, of which 31 % failed vancomycin. The most common infections were bacteremia (78 %), skin and skin structure infections (8 %), and urinary tract infections/pyelonephritis (6 %). The most common pathogens were vancomycin-resistant Enterococcus faecium (47 %), methicillin-resistant Staphylococcus aureus (20 %), and coagulase-negative staphylococci (19 %). The median (min, max) initial daptomycin dose was 6 mg/kg (3.6, 8.3). The median (min, max) daptomycin duration of therapy was 14 days (1, 86). Possibly related adverse events occurred in 12/209 patients (6 %), and 13 patients (6 %) discontinued daptomycin due to adverse event. The results suggest that daptomycin appeared useful and well tolerated in neutropenic patients, and the degree of neutropenia did not affect daptomycin success rates. Comparative clinical trials are needed to confirm these findings.
    Supportive Care in Cancer 08/2013; · 2.09 Impact Factor
  • Lior Nesher, Kenneth V I Rolston
    [Show abstract] [Hide abstract]
    ABSTRACT: Despite advancements in the treatment and supportive care of patients with malignant disorders, neutropenia remains the major side effect of most antineoplastic regimens. Infections occur frequently in neutropenic patients and are associated with considerable morbidity and mortality. The spectrum of infection continues to change, and is influenced by various factors including local epidemiology, the use of chemoprophylaxis, and the use of central venous catheters and other medical devices. Bacterial infections are common in the early stages of neutropenia, with fungal infections emerging if neutropenia persists beyond 7-10 days. Gram-positive organisms cause most bacteremic infections (although this trend appears to be changing), whereas infections at other sites are often caused by Gram-negative bacilli or are polymicrobial, especially if deep tissue infection is present. Candida spp., and Aspergillus spp., remain the most common fungal pathogens, although several opportunistic fungi have emerged. Resistance to antimicrobial and antifungal agents commonly used for the prevention and treatment of infections in neutropenic patients has become a significant problem. The prompt administration of appropriate, empiric, antimicrobial therapy, prior to the availability of microbiological culture results, is the standard of care. Up to date knowledge of the spectrum of infection and local susceptibility/resistance patterns, is critical. In this report, we describe the current spectrum of infection in patients with malignancies and neutropenia, and emphasize the fact that local and geographic differences are not infrequent. We recommend that individual institutions conduct periodic epidemiological surveys in order to have the latest data available for the optimal management of their patients.
    Infection 08/2013; · 2.44 Impact Factor
  • Kenneth V I Rolston
    Clinical Infectious Diseases 05/2013; · 9.37 Impact Factor
  • Infection 03/2013; · 2.44 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Gram-positive organisms are the predominant bacterial pathogens in cancer patients. A survey indicated that coagulase-negative staphylococci (CoNS) (29.5%), Staphylococcus aureus (18.0%), Enterococcus spp. (12.1%) and viridans group streptococci (VGS) (9.1%) are isolated most often. The rate of reduced susceptibility to vancomycin (minimum inhibitory concentration ≥1.0μg/mL) was 100% for meticillin-susceptible S. aureus and 99% for meticillin-resistant S. aureus, and 100% for meticillin-susceptible CoNS and 98% for meticillin-resistant CoNS. More than 98% of these isolates were susceptible to daptomycin and linezolid. Daptomycin and linezolid had comparable in vitro activity to vancomycin against Bacillus spp., Corynebacterium spp., Rhodococcus spp., Micrococcus spp., Stomatococcus mucilaginosus and VGS. Both agents were active against the majority (95%) of vancomycin-resistant organisms, including vancomycin-resistant enterococci, Pediococcus spp. and Leuconostoc spp. These data suggest that daptomycin and linezolid have an adequate antimicrobial spectrum and potent in vitro activity against Gram-positive isolates from cancer patients and may be considered as alternatives to vancomycin for empirical or targeted therapy in this setting.
    International journal of antimicrobial agents 03/2013; · 3.03 Impact Factor
  • Roy A Borchardt, Kenneth V I Rolston
    JAAPA: official journal of the American Academy of Physician Assistants 03/2013; 26(3):18, 25.
  • Roy A Borchardt, Kenneth V I Rolston
    JAAPA: official journal of the American Academy of Physician Assistants 02/2013; 26(2):13, 18.
  • [Show abstract] [Hide abstract]
    ABSTRACT: PURPOSETo provide guidelines on antimicrobial prophylaxis for adult neutropenic oncology outpatients and on selection and treatment as outpatients of those with fever and neutropenia. METHODSA literature search identified relevant studies published in English. Primary outcomes included: development of fever and/or infections in afebrile neutropenic outpatients and recovery without complications and overall mortality in febrile neutropenic outpatients. Secondary outcomes included: in afebrile neutropenic outpatients, infection-related mortality; in outpatients with fever and neutropenia, defervescence without regimen change, time to defervescence, infectious complications, and recurrent fever; and in both groups, hospital admissions, duration, and adverse effects of antimicrobials. An Expert Panel developed guidelines based on extracted data and informal consensus.ResultsForty-seven articles from 43 studies met selection criteria.RecommendationsAntibacterial and antifungal prophylaxis are only recommended for patients expected to have < 100 neutrophils/μ L for > 7 days, unless other factors increase risks for complications or mortality to similar levels. Inpatient treatment is standard to manage febrile neutropenic episodes, although carefully selected patients may be managed as outpatients after systematic assessment beginning with a validated risk index (eg, Multinational Association for Supportive Care in Cancer [MASCC] score or Talcott's rules). Patients with MASCC scores ≥ 21 or in Talcott group 4, and without other risk factors, can be managed safely as outpatients. Febrile neutropenic patients should receive initial doses of empirical antibacterial therapy within an hour of triage and should either be monitored for at least 4 hours to determine suitability for outpatient management or be admitted to the hospital. An oral fluoroquinolone plus amoxicillin/clavulanate (or plus clindamycin if penicillin allergic) is recommended as empiric therapy, unless fluoroquinolone prophylaxis was used before fever developed.
    Journal of Clinical Oncology 01/2013; · 18.04 Impact Factor
  • Roy A Borchardt, Kenneth V I Rolston
    JAAPA: official journal of the American Academy of Physician Assistants 01/2013; 26(1):18, 25.
  • Lior Nesher, Kenneth Rolston
    [Show abstract] [Hide abstract]
    ABSTRACT: Neutropenic enterocolitis (NEC) is a life threatening disease with substantial morbidity and mortality, seen primarily in patients with hematologic malignancies. The frequency of NEC has increased with the wide spread use of chemotherapeutic agents such as the taxanes, which cause severe gastrointestinal mucositis. Neutropenic patients with fever and abdominal symptoms (cramping, pain, distention, diarrhea, GI bleeding), should undergo evaluation of the abdomen for bowel wall thickening of >4 mm, the hallmark of NEC. Clostridium difficile infection should be ruled out, as well as other etiologies such as GVHD. Complications include bacteremia which is often polymicrobial, hemorrhage, and bowel wall perforation / abscess formation. Management includes bowel rest, correction of cytopathies and coagulopathies, and broad spectrum antibiotics and antifungal agents. Surgical intervention may be necessary to manage complications such as hemorrhage and perforation, and should be delayed, if possible, until recovery from neutropenia.
    Clinical Infectious Diseases 11/2012; · 9.37 Impact Factor
  • Samuel A Shelburne, Jeffrey Tarrand, Kenneth V Rolston
    [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVES: To determine the comparative rates, clinical characteristics, and outcomes of invasive infections due to specific streptococcal types in patients with cancer. METHODS: Review of electronic medical records of patients with non-viridans group streptococcal bloodstream infection (BSI) at the MD Anderson Cancer Center from 2000-2011. RESULTS: 550 streptococcal BSI were identified. The largest number of cases were caused by Streptococcus pneumoniae (251), group B Streptococcus (147), and gamma-hemolytic streptococci (55). Risk factors for developing a severe streptococcal infection included older age, being neutropenic at onset of BSI, and having a respiratory source of infection. Between 2000-2001 and 2010-2011, the rates of S. pneumoniae BSI and penicillin non-susceptibility decreased by 55% and 100%. In contrast the rate of group B streptococcal (GBS) BSI increased 34% over the same time period. GBS accounted for >80% of the recurrent infections following streptococcal BSI. Patients with breast cancer and those with soft-tissue/bone BSI sources were at increased risk for recurrent GBS infection but had lower rates of severe GBS disease. CONCLUSIONS: From 2000-2011, our comprehensive cancer center observed a significant decrease in the rates of S. pneumoniae BSI and a significant increase in the rates of GBS BSI.
    The Journal of infection 11/2012; · 4.13 Impact Factor
  • Roy A Borchardt, Kenneth V I Rolston
    JAAPA: official journal of the American Academy of Physician Assistants 10/2012; 25(10):19-20.

Publication Stats

8k Citations
1,694.40 Total Impact Points


  • 1986–2014
    • University of Texas MD Anderson Cancer Center
      • • Department of Medical Specialities
      • • Department of Internal Medicine Specialties
      Houston, Texas, United States
    • University of Texas at Tyler
      Tyler, Texas, United States
  • 2004–2013
    • University of Texas Medical School
      • Division of Infectious Diseases
      Houston, Texas, United States
  • 2011
    • Baylor College of Medicine
      Houston, Texas, United States
  • 2005–2011
    • University of Houston
      Houston, Texas, United States
    • Shizuoka Cancer Center
      Sizuoka, Shizuoka, Japan
  • 2003–2004
    • University of Bristol
      • School of Cellular and Molecular Medicine
      Bristol, ENG, United Kingdom
  • 2001
    • University of Crete
      • School of Medicine
      Réthymnon, Kriti, Greece
  • 1993–1994
    • Southern Illinois University School of Medicine
      • Department of Internal Medicine
      Springfield, IL, United States
  • 1988
    • University of Texas Health Science Center at Houston
      Houston, Texas, United States
  • 1984
    • Wayne State University
      Detroit, Michigan, United States