Kenneth H Kim

University of North Carolina at Chapel Hill, Chapel Hill, NC, United States

Are you Kenneth H Kim?

Claim your profile

Publications (16)74.17 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Blood products are scarce but essential medical resources. Initially transfusions showed increased perioperative complications, prolonged hospitalizations, and higher mortality. Recently developed restrictive transfusion policies have not shown those adverse affects in critically ill patients. Hospitals adopted these policies to guide blood product administration. The objective of this study is to determine compliance with a restrictive transfusion policy in gynecologic oncology patients. A retrospective chart review of gynecologic oncology patients undergoing transfusion with packed red blood cells (pRBCs) from 12/2008-9/2011 was performed. Cancer type and stage, surgical procedure, hemoglobin values, pRBC transfusions, intraoperative blood loss, and postoperative complications were collected. Each transfusion was classified as compliant or noncompliant. 582 patients requiring 2,276 blood transfusions were identified. The mean age was 55.9years. Ovarian and endometrial cancers were the most common malignancies. Gynecologic oncologists were 81.1% compliant with the restrictive transfusion policy; 59.0% of transfusions were secondary to exceptions. Noncompliant transfusions were commonly given on the day of surgery when intraoperative blood loss was<1500cc and for asymptomatic anemia. Only 64.7% of the transfusions were ordered in single unit increments. There was no significant difference in postoperative infections, thrombotic events, and mortality between compliant and noncompliant transfusions. The majority of gynecologic oncology patients receive transfusions compliant with the restrictive transfusion policy. Morbidity and mortality are not increased with a restrictive transfusion policy. Efforts to improve compliance should focus on limiting transfusions when the hemoglobin is≥7g/dL and transfusing in single pRBCs unit increments.
    Gynecologic Oncology 10/2013; · 3.93 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To determine the radiographic characteristics of ovarian granulosa cell tumors (GCT) and to evaluate the use of CA125 levels >35 in combination with imaging as an algorithm for preoperative diagnosis. A retrospective analysis of women from two academic medical centers who were diagnosed with ovarian GCT between January 1998 and August 2012 was conducted. Clinical data included tumor appearance on preoperative imaging and CA-125 levels. Ovarian cysts were defined as complex if imaging exhibited multicystic areas, hemorrhagic, solid, or cystic and solid components. A CA125 level >35 was abnormal. One hundred and fifteen women were diagnosed with GCTs, of whom 63 underwent pre-operative imaging. Median age at surgery was 46 years (12-87). Forty women had preoperative ultrasounds, 43 had CT scans and 20 underwent both modalities. GCTs were almost exclusively classified as complex cysts in 62 (98%) cases. The most common morphology was solid and cystic (n=44 (70%)). Forty-four (70%) patients had tumors>10cm. Forty-two patients had a pre-operative CA125 performed. Eighteen (43%) patients had complex masses and CA125 >35. Twenty-three (55%) had CA125 <35 with a complex mass, and one (2%) had a unilocular cyst with a CA125 >35. In this study, there was a near equal distribution of patients with complex masses and CA125 levels > or < 35. If established strategies to predict malignancy are applied to GCTs, we will frequently fail to make the diagnosis pre-operatively. Additional research is necessary to generate an appropriate algorithm to guide pre-operative referral to a gynecologic oncologist.
    Gynecologic Oncology 06/2013; · 3.93 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVE: The conditionally replicative adenovirus Ad5/3-Δ24 has a type-3 knob incorporated into the type-5 fiber that facilitates enhanced ovarian cancer infectivity. Preclinical studies have shown that Ad5/3-Δ24 achieves significant oncolysis and anti-tumor activity in ovarian cancer models. The purpose of this study was to evaluate in a Phase I trial the feasibility and safety of intraperitoneal (IP) Ad5/3-Δ24 in recurrent ovarian cancer patients. METHODS: Eligible patients were treated with IP Ad5/3-Δ24 for 3 consecutive days in one of three dose cohorts ranging 1 x 10(10)-1 x 10(12) vp. Toxicity was assessed utilizing CTC grading and efficacy with RECIST. Ascites, serum, and other samples were obtained to evaluate gene transfer, generation of wildtype virus, viral shedding, and antibody response. RESULTS: Nine of 10 patients completed treatment per protocol. A total of 15 vector-related adverse events were experienced in 5 patients. These events included fever or chills, nausea, fatigue, and myalgia. All were grade 1-2 in nature, transient, and medically managed. Of the 8 treated patients evaluable for response, six patients had stable disease and 2 patients had progressive disease. Three patients had decreased CA-125 from pretreatment levels one month after treatment. Ancillary biologic studies indicated Ad5/3-Δ24 replication in patients in the higher dose cohorts. All patients experienced an anti-adenoviral neutralizing antibody effect. CONCLUSIONS: This study suggests the feasibility and safety of a serotype chimeric infectivity-enhanced CRAd, Ad5/3-Δ24, as a potential therapeutic option for recurrent ovarian cancer patients.
    Gynecologic Oncology 06/2013; · 3.93 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVE: The study goal was to determine whether prior outpatient exposure to hospice discussion altered the inpatient course and end-of-life (EOL) care among patients ultimately discharged to hospice. METHODS: Medical records from January 2009 - June 2012 were reviewed and data abstracted under an IRB-approved protocol. Hospice discussions were identified in the last outpatient clinical encounter prior to admission. Kaplan-Meier was used to estimate overall survival (OS) and the log-rank test used to test for differences. RESULTS: There were 89 hospitalizations resulting in discharge to hospice care: 41 women with ovarian (46%), 23 with uterine (29%), 19 with cervical (21.3%), and with 6 vulvar/vaginal (6.7%) cancers. 83 patients (93%) had outpatient clinical encounters prior to admission;18% (15/83) were exposed to a hospice discussion (HD) and 82% (68/83) were not (NHD). Median time from last outpatient encounter was 18 days (range 0 - 371). NHD patients had longer inpatient length of stay (median 7 days vs. 4 days, p=0.008) and were less likely to receive palliative care consults than the HD patients (65% vs. 93%, p=0.03). Median OS for HD patients was 33 days (95% CI 22d - 61d) vs. 60 days (95% CI 49d - 84d) for NHD patients (p=0.01). There were no differences detected based on race, ethnicity, or insurance status. CONCLUSIONS: HD patients had significantly shorter OS suggesting that providers were accurate in identifying patients nearing the EOL. Patients exposed to outpatient hospice discussions had a shorter length of stay and increased utilization of palliative care resources.
    Gynecologic Oncology 03/2013; · 3.93 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To develop a cost-minimization analysis of a multivariate index assay (MIA) used for women with complex pelvic masses. A decision analysis model was used to evaluate 81,000 hypothetical patients with a complex pelvic mass requiring surgery. Three strategies were evaluated: (1) referral to a gynecologic oncologist (GO) based on clinical assessment including physical exam, ultrasonography, and CA125 (CLINICAL); (2) utilization of a multivariate index assay (MIA); or (3) referral of all patients to a GO (REFER ALL). Various reoperation rates were evaluated with sensitivity analyses. Actual payer costs were compared between each strategy. The CLINICAL strategy cost $933.9 million (M) and resulted in 72% of patients receiving appropriate initial surgical staging. The REFER ALL strategy cost $939.7 M and all patients were appropriately staged. The MIA strategy cost $976.7 M and resulted in 91% of patients having appropriate initial staging. Using conservative reoperation rates (10-20%), 461 patients required reoperation using CLINICAL strategy compared to 142 patients in MIA strategy. Using aggressive reoperation rates (40-50%), 1715 patients required reoperation using CLINICAL strategy resulting in an incremental cost of $15.2M compared to 529 patients at $4.7 M in MIA strategy. The increased costs associated with an aggressive reoperation rate resulted in the REFER ALL strategy being the least expensive alternative, with the highest rates of appropriate initial surgery. Utilizing an MIA resulted in more ovarian cancer patients receiving appropriate initial surgery, but at increased costs. Referring all patients with complex masses avoids the most reoperations at reduced cost compared to using an MIA.
    Gynecologic Oncology 05/2012; 126(3):364-8. · 3.93 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Ad5.SSTR/TK.RGD is an infectivity-enhanced adenovirus expressing a therapeutic thymidine kinase suicide gene and a somatostatin receptor (SSTR) that allows for noninvasive gene transfer imaging. The purpose of this study was to identify the maximum tolerated dose (MTD), toxicities, clinical efficacy, and biologic effects of Ad5.SSTR/TK.RGD in patients with recurrent gynecologic cancer. Eligible patients were treated intraperitoneally for 3 days with 1 × 10(9) to 1 × 10(12) vp/dose of Ad5.SSTR/TK.RGD followed by intravenous ganciclovir for 14 days. Toxicity and clinical efficacy were assessed using Common Toxicity Criteria (CTC) Adverse Events grading and Response Evaluation Criteria in Solid Tumors (RECIST) criteria. Imaging using In-111 pentetreotide was obtained before and after treatment. Tissue samples were obtained to evaluate for gene transfer, generation of wild-type virus, viral shedding, and antibody response. Twelve patients were treated in three cohorts. The most common vector-related clinical toxicities were grade I/II constitutional or pain symptoms, experienced most often in patients treated at the highest dose. MTD was not identified. Five patients showed stable disease; all others experienced progressive disease. One patient with stable disease experienced complete resolution of disease and normalization of CA125 on further follow-up. Imaging detected increased In-111 pentetreotide retention in patients treated at the highest dose. Ancillary studies showed presence of Ad5.SSTR/TK.RGD virus and HSV1-tk expression in ascites samples collected at various time points in most patients treated within the higher dose cohorts. This study shows the safety, potential efficacy, and possible gene transfer imaging capacity of Ad5.SSTR/TK.RGD in patients with recurrent gynecologic cancer. Further development of this novel gene therapeutic appears to be warranted.
    Clinical Cancer Research 04/2012; 18(12):3440-51. · 7.84 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Debate continues about optimal management of patients with node-positive stage I cervical cancer. Our objective was to determine if patient outcomes are affected by radical hysterectomy in the modern era of adjuvant chemoradiation. Cervical cancer patients diagnosed from 2000 to 2008 were identified. Demographics, therapy, clinicopathologic data, progression free survival (PFS), overall survival (OS), total radiation exposure, and grade 3-4 complications were analyzed by student t, Mann-Whitney, Fisher's exact, Kaplan-Meier, and log rank tests. This single-institution review evaluated forty-one of 334 (13.4%) patients scheduled to undergo radical hysterectomy that had gross nodal disease diagnosed intraoperatively. 15 underwent aborted radical hysterectomy following lymphadenectomy; the remaining 26 underwent radical hysterectomy and lymphadenectomy. Eleven patients undergoing radical hysterectomy underwent whole pelvic radiation therapy (WPRT) while 8 (30.7%) patients underwent WPRT and postoperative vaginal brachytherapy (BT) for local treatment secondary to close margins. All patients undergoing aborted radical hysterectomy underwent WPRT and BT. With mean follow-up of 42.3 months, there were no significant differences in urinary, gastrointestinal, or hematologic complications between groups. When comparing those undergoing radical hysterectomy to aborted radical hysterectomy, there were no significant differences in local recurrence (11.5% vs 26.7%, p=0.39) or distant recurrence (19.2% vs. 33.3%, p=0.45), PFS (74.9 months vs 46.8 months, p=0.106), or OS (91.8 months vs 69.4 months, p=0.886). Treatment of patients with early stage cervical cancer and nodal metastasis may be tailored intraoperatively. Completion of radical hysterectomy and lymphadenectomy decreases radiation exposure without apparently compromising safety or outcome in the era of adjuvant chemoradiation.
    Gynecologic Oncology 04/2012; 126(1):69-72. · 3.93 Impact Factor
  • Kenneth H Kim, Ronald D Alvarez
    [Show abstract] [Hide abstract]
    ABSTRACT: A multivariate index assay recently was developed to assist physicians in the assessment of the risk of ovarian cancer in women with pelvic masses undergoing operative intervention. Its aim is to improve the identification of patients with ovarian malignancy so that these patients can be appropriately referred to a subspecialist with ovarian cancer-management expertise, thereby affording the opportunity for improved outcome. This commentary questions the need to obtain a multivariate index assay test in all women with pelvic masses who are scheduled for operative intervention. Common-sense guidelines for more judicious use of this new triage test in the evaluation of these patients also are provided.
    Obstetrics and Gynecology 02/2012; 119(2 Pt 1):365-7. · 4.80 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Highlights ► The hook effect occurs with extremely high levels of hCG, saturating detection antibodies, leading to falsely low laboratory results. ► In the literature, descriptions of the hook effect are rare in cases of gestational trophoblastic diseases. ► If unrecognized, this can lead to delayed therapy or mismanagement of care.
    Gynecologic Oncology Case Reports. 12/2011; 1(1):6–7.
  • [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this study was to investigate the incidence of mechanical complications associated with low-profile subcutaneous implantable venous access devices in gynecologic oncology patients. Gynecologic oncology patients with low-profile Port-a-Caths implanted between March 2005 and July 2006 were identified into a computerized database. Patient demographics, operative complications, number of chemotherapy cycles, duration of implantation, and mechanical complications were collected. Primary outcomes included port leakage, catheter fracture, and catheter embolization. 112 patients underwent 115 Port-a-Cath placements with low profile single-lumen plastic ports with Groshong-valved catheters. Mean Port-a-Cath indwelling duration was 197 days (range: 4-395) with a mean number of 12 chemotherapy cycles (range 0-64). The cumulative complication rate necessitating removal or replacement was 15%. Of the 14 Port-a-Caths removed, ten (8.7%) were secondary to mechanical malfunction: one for leakage at the port site, two for catheter fracture, and seven for fracture with catheter embolization to the heart or pulmonary vasculature-most commonly the right ventricle. Patients with embolization were asymptomatic and all embolized catheters were successfully retrieved by interventional radiology without complications. The rates of catheter fracture and embolization have previously been reported to be low in patients with subcutaneous Port-a-Caths, and have not been studied in patients receiving low-profile subcutaneous Port-a-Caths. This study suggests that catheter fracture may be more common (8.7%) and must be considered in patients with malfunctioning low-profile Port-a-Caths. Embolized catheters can be removed by interventional radiology without significant adverse affects.
    Gynecologic Oncology 07/2011; 123(1):54-7. · 3.93 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Conditionally replicative adenoviral (CRAd) virotherapy represents a promising therapeutic approach for cancer. We have demonstrated that a serotype chimeric adenoviral 5/3 fiber-knob modification achieves enhanced ovarian cancer infectivity, conditional replication, and oncolytic activity. This study evaluated the safety of intraperitoneal (IP) Ad5/3-Δ24 in advance of a phase I clinical trial in gynecologic cancers. Syrian hamster cohorts were treated with IP Ad5/3-Δ24 or control buffer for 3 consecutive days and euthanized on study days 8, 17, 57, and 89. Blood and tissue samples were harvested from each animal. For biodistribution studies, presence and quantitation of viral levels within samples were determined via quantitative polymerase chain reaction. For safety studies, animals were assessed for adverse vector-related tissue or laboratory effects. In the biodistribution study, low levels of Ad5/3-Δ24 DNA were noted outside of the abdominal cavity. Viral DNA levels in tissues obtained from the peritoneal cavity peaked at day 8 and declined thereafter. In the safety study, no specific histopathologic changes were attributable to virus administration. Hematologic findings noted in the 1 × 10(11) viral particles (vp)/dose group on Days 4 and/or 8 were indicative of an Ad5/3-Δ24-specific generalized inflammatory response; these findings resolved by day 56. The no observable adverse effect level was determined to be 1 × 10(10) vp/dose. This study elucidates the safety profile of IP administration of the serotype chimeric infectivity-enhanced CRAd, Ad5/3-Δ24, and provides guidance for a planned phase I trial for patients with recurrent gynecologic cancers.
    Human gene therapy 07/2011; 22(7):821-8. · 4.20 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Minimally invasive surgery offers advantages for management of obese patients, but technical difficulty often deters its utilization. Compared to laparotomy, robotic surgery should allow comparable staging and improved surgical outcomes. Therefore, we evaluated outcomes in robotic and laparotomy cohorts of obese women with endometrial cancer at our institution. Retrospective robotic and laparotomy cohorts of obese women (BMI ≥ 30 kg/m(2)) undergoing surgical management of primary endometrial cancer from March 2006 to March 2009 were formulated utilizing a computerized database. Patient demographics, operative statistics, peri-operative complications, and pathologic details were collected in an intent to treat analysis. Chi-square or Fisher's exact test and t-test were used for statistical analysis. 73 women underwent robotic surgical management, 11% converted to laparotomy. Mean BMI (39.8 vs. 41.9, p=0.152), number of co-morbidities (2.49 vs. 2.62, p=0.690), number of previous surgeries (0.97 vs. 0.94, p=0.841), and lymphadenectomies performed (65.8% vs. 56.7%, p=0.227) were similar between cohorts. Total lymph nodes obtained were not statistically different between cohorts (8.01 vs. 7.24, p=0.505). Total operative time and room time was significantly longer for robotic surgery; however, estimated blood loss, the percentage of patients receiving transfusion, hospital length of stay, wound complications (4.1% vs. 20.2%, p=0.002) and other complications (9.6% vs. 29.8%, p=0.001) were improved for the robotic cohort. Robotic management of obese women with endometrial cancer yields acceptable staging results and improved surgical outcomes. Although operating time is longer, hospital time is shorter. Robotic surgery may be an ideal approach for these patients.
    Gynecologic Oncology 06/2011; 122(3):604-7. · 3.93 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To determine whether the addition of bevacizumab to paclitaxel and carboplatin for the primary treatment of advanced ovarian cancer can be cost effective. A cost-effectiveness analysis compared the three arms of the Gynecologic Oncology Group (GOG) 218 study (paclitaxel plus carboplatin [PC], PC plus bevacizumab [PCB], and PCB plus bevacizumab maintenance [PCB+B]). Actual and estimated costs of treatment plus the potential costs of complications were established for each strategy. Progression-free survival (PFS) and bowel perforation rates were taken from recently reported results of GOG 218. Sensitivity analysis was performed for pertinent uncertainties in the model. Incremental cost-effectiveness ratios (ICERs) per progression-free life-year saved (PF-LYS) were estimated. For the 600 patients entered onto each arm of GOG 218 at the baseline estimates of PFS and bowel perforation, the cost of PC was $2.5 million, compared with $21.4 million for PCB and $78.3 million for PCB+B. These costs led to an ICER of $479,712 per PF-LYS for PCB and $401,088 per PF-LYS for PCB+B. When the cost of bevacizumab was reduced to 25% of baseline, the ICER of PCB+B fell below $100,000 per PF-LYS. ICERs were not substantially reduced when the perforation rate was equal across all arms. The addition of bevacizumab to standard chemotherapy in patients with advanced ovarian cancer is not cost effective. Treatment with maintenance bevacizumab leads to improved PFS but is associated with both direct and indirect costs. The cost effectiveness of bevacizumab in the adjuvant treatment of ovarian cancer is primarily dependent on drug costs.
    Journal of Clinical Oncology 03/2011; 29(10):1247-51. · 18.04 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To determine whether DNA mismatch repair (MMR) modifies the response to chemotherapy or radiotherapy in patients with endometrial cancer. Immunohistochemistry (IHC) for the DNA MMR proteins MLH1, MSH2, MSH6, and PMS2 was performed on a tissue microarray of specimens of primary endometrial cancer. MMR deficiency was defined as lack of expression of one or more proteins. Expression of all proteins classified a tumor as having an intact MMR system. Recurrence rates were calculated for women treated with platinum-based chemotherapy or pelvic external beam radiation. Comparisons were made using the log-rank test. Multiple comparisons were controlled for by utilizing the Bonferroni correction method. Four hundred seventy-seven cases of endometrial cancer were evaluated on a tissue microarray (TMA). One hundred fifty-eight patients (41%) received chemotherapy. Sixty-six patients (17%) received pelvic teletherapy. Overall and progression-free survival were not different between patients whose tumors had intact MMR and those with defective MMR when stratified by adjuvant treatment with radiation or chemotherapy. Subgroup analyses stratified by histology (non-endometrioid versus endometrioid) and stage did show significant survival differences. There was a significant increase in overall (p=0.003) and progression-free (p=0.004) survival in those with MMR-deficient, non-endometrioid tumors treated with teletherapy compared to those with an intact MMR system. Improved progression-free survival was noted in patients with intact MMR with stage III/IV disease treated with adjuvant chemotherapy (p=0.031). Subgroups of patients with non-endometrioid endometrial cancer and defective MMR may have improved survival after adjuvant radiotherapy. Patients with advanced stage endometrial cancer and defects in mismatch repair may receive less benefit from adjuvant chemotherapy.
    Gynecologic Oncology 02/2010; 117(2):234-8. · 3.93 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To compare outcomes between robotic versus laparoscopic hysterectomy and lymphadenectomy in patients with endometrial cancer. A cohort study was performed by prospectively identifying all patients with clinical stage I or occult stage II endometrial cancer who underwent robotic hysterectomy and lymphadenectomy from 2006-2008 and retrospectively comparing data using the same surgeons' laparoscopic hysterectomy and lymphadenectomy cases from 1998-2005, prior to our robotic experience. Patient demographics, operative times, complications, conversion rates, pathologic results, and length of stay were analyzed. 181 patients (105 robotic and 76 laparoscopic) met inclusion criteria. There was no significant difference between the two groups in median age, uterine weight, bilateral pelvic or aortic lymph node counts, or complication rates in patients whose surgeries were completed minimally invasively. Despite a higher BMI (34 vs. 29, P<0.001), the estimated blood loss (100 vs. 250 mL, P<0.001), transfusion rate (3% vs. 18%, RR 0.18, 95%CI 0.05-0.64, P=0.002), laparotomy conversion rate (12% vs. 26%, RR 0.47, 95%CI 0.25-0.89, P=0.017), and length of stay (median: 1 vs. 2 nights, P<0.001) were lower in the robotic patients compared to the laparoscopic cohort. The odds ratio of conversion to laparotomy based on BMI for robotics compared to laparoscopy is 0.20 (95% CI 0.08-0.56, P=0.002). The mean skin to skin time (242 vs. 287 min, P<0.001) and total room time (305 vs. 336 min, P<0.001) was shorter for the robotic cohort. Robotic hysterectomy and lymphadenectomy for endometrial carcinoma can be accomplished in heavier patients and results in shorter operating times and hospital length of stay, a lower transfusion rate, and less frequent conversion to laparotomy when compared to laparoscopic hysterectomy and lymphadenectomy.
    Gynecologic Oncology 01/2009; 113(1):36-41. · 3.93 Impact Factor
  • Kenneth H. Kim, Warner K. Huh
    [Show abstract] [Hide abstract]
    ABSTRACT: Endometrial cancer is the most common gynecologic malignancy in the United States. The vast majority of these cases are discovered when the disease is still limited to the uterus. There is a consensus agreement that the initial treatment and management should be surgical staging; however, there remains much controversy as to the use and role of adjuvant radiation in the setting of early-stage endometrial cancer. In the literature to date, there has not been resounding evidence to support the use of adjuvant radiation therapy in this early disease state. This is somewhat troublesome because this is the most common disease state of the most common gynecologic malignancy; finding the optimal treatment would be of massive benefit to patients and physicians as well as to society as a whole. Herein, the relevant literature is reviewed to demonstrate that the use of adjuvant radiation therapy is not clearly supported in the setting of early-stage endometrial cancer.
    Clinical Ovarian Cancer 01/2009; 2(2):90-93.

Publication Stats

201 Citations
74.17 Total Impact Points


  • 2013
    • University of North Carolina at Chapel Hill
      • Department of Obstetrics and Gynecology
      Chapel Hill, NC, United States
  • 2011–2012
    • University of Alabama at Birmingham
      • • Department of Obstetrics and Gynecology
      • • Division of Gynecologic Oncology
      Birmingham, AL, United States