Use of a hybrid-abdominal wound simulated patient in the ACS/ASE medical student simulation skills curriculum.

OBJECTIVE: We incorporated a hybrid-abdominal wound simulation to teach/assess the acquisition of three essential clinical skills in the ACS/ASE Medical Student Simulation-based Surgical Skills Curriculum. METHOD: Third year students (N = 43) attended a workshop based on the ACS/ASE surgical skills curriculum for drain care/removal, staple removal and Steri-Strip application. Following a didactic session and demonstration using a simulated patient, student skill acquisition was assessed using the ACS/ASE module rating tool. Student interest/perceived usefulness of the workshop was evaluated using Keller's Motivational Survey. RESULTS: We used median scores to identify low proficiency (n = 20; scores 17-28) and high proficiency (n = 23; scores 29-35) groups. The high proficiency group was more knowledgeable, performed better drain care, had a higher global score and was more confident than the low proficiency group. The students rated the workshop highly based on the Keller's Motivational Survey. CONCLUSION: All students were proficient in the procedure tasks and communication skills and most felt that the course was beneficial. The ACS/ASE Medical Student Simulation-based Surgical Skills Curriculum was successfully integrated into our third year surgical clerkship.

How Safe Is the Safety Net? Comparison of Ivor-Lewis Esophagectomy at a Safety-Net Hospital Using the NSQIP Database.

BACKGROUND: Recent data suggest that surgical outcomes at hospitals caring for low-income, vulnerable populations are suboptimal compared with outcomes from nonsafety-net hospitals. Therefore, the purpose of our study was to compare outcomes for patients who underwent an Ivor-Lewis esophagectomy at a safety-net hospital with the National Surgical Quality Improvement Program (NSQIP) database. STUDY DESIGN: We retrospectively reviewed the medical records of consecutive patients who underwent an Ivor-Lewis esophagectomy, between September 2013 and January 2017, at a single safety-net hospital. Patient characteristics and outcomes were compared with the 2013 to 2015 NSQIP database. Continuous variables were compared using Student's t-test, and categorical variables were analyzed using chi-square tests. Values of p < 0.05 were considered significant. RESULTS: We identified 78 patients from the safety-net hospital and 1,825 patients in the NSQIP database who underwent an Ivor-Lewis esophagectomy. Baseline characteristics were similar, except the safety-net hospital patients were more likely to have COPD (19.2% vs 8.1%; p = 0.001) and be current smokers (42.3% vs 26.0%; p = 0.001); patients in the NSQIP group had a higher BMI (28 kg/m2 vs 26 kg/m2; p = 0.001). There were no differences between groups for mortality, readmission, discharge destination, or mean operative time. Safety-net hospital patients had significantly fewer complications (16.7% vs 33.3%; p = 0.003), fewer reoperations (6.4% vs 14.5%; p = 0.046), and shorter hospital length of stay (10.3 vs 13.1 days; p = 0.001). CONCLUSIONS: Patients who underwent an Ivor-Lewis esophagectomy at a safety-net hospital had fewer complications and reoperations, and a shorter hospital length of stay compared with a national cohort. These findings illustrate the value of clinical pathways in optimizing the patient outcomes at safety-net hospitals and providing excellent care to their vulnerable patient population.

Acquisition, retention and transfer of simulated laparoscopic tasks using fNIR and a contextual interference paradigm.

BACKGROUND: Using functional near infrared spectroscopy, a noninvasive, optical brain imaging tool that monitors changes in hemodynamics within the prefrontal cortex (PFC), we assessed performance and cognitive effort during the acquisition, retention and transfer of multiple simulated laparoscopic tasks by novice learners within a contextual interference paradigm. METHODS: Third-year medical students (n = 10) were randomized to either a blocked or random practice schedule. Across 3 days, students performed 108 acquisition trials of 3 laparoscopic tasks on the LapSim® simulator followed by delayed retention and transfer tests. Performance metrics (Global score, Total time) and hemodynamic responses (total hemoglobin (µm)) were assessed during skill acquisition, retention and transfer. RESULTS: All acquisition tasks resulted in significant practice schedule X trial block interactions for the left medial anterior PFC. During retention and transfer, random performed the skills in less time and had lower total hemoglobin change in the right dorsolateral PFC than blocked. CONCLUSIONS: Compared with blocked, random practice resulted in enhanced learning through better performance and less cognitive load for retention and transfer of simulated laparoscopic tasks.

Real-time evaluation and visualization of learner performance in a mixed-reality environment for clinical breast examination.

We investigate the efficacy of incorporating real-time feedback of user performance within mixed-reality environments (MREs) for training real-world tasks with tightly coupled cognitive and psychomotor components. This paper presents an approach to providing real-time evaluation and visual feedback of learner performance in an MRE for training clinical breast examination (CBE). In a user study of experienced and novice CBE practitioners (n = 69), novices receiving real-time feedback performed equivalently or better than more experienced practitioners in the completeness and correctness of the exam. A second user study (n = 8) followed novices through repeated practice of CBE in the MRE. Results indicate that skills improvement in the MRE transfers to the real-world task of CBE of human patients. This initial case study demonstrates the efficacy of MREs incorporating real-time feedback for training real-world cognitive-psychomotor tasks.

A portable palpation training platform with virtual human patient.

Palpation (the application of touch to the surface of the body) is an essential clinical skill. Correct palpation is part of a complete physical examination and it assists a clinician in making an accurate diagnosis, while poor palpatory skills can lead to diagnostic errors. As with any clinical skill, palpation is best learned through repetitive practice with constructive feedback. Unfortunately, changes in healthcare provide fewer opportunities for hands-on learning of this essential skill. Unlike other clinical skills, palpation has no immediate feedback to the learner regarding their performance. For example, when students are learning how to insert an intravenous catheter, failure to perform the technique correctly results in no blood return in the catheter. However, students do not know if they are palpating an abnormality if they have never felt it before. This inherent difficulty makes expert feedback even more vital to learning correct palpation. Existing research tools have addressed some of these challenges through simulation techniques that do not require experts, and can provide feedback on palpation pressures and palpation patterns. We describe a novel computer-based palpation training system, leveraging existing approaches, with an emphasis on sensing accuracy, directed-feedback, portability, and user experience.

The predictive value of incidental PET/CT findings suspicious for breast cancer in women with non-breast malignancies.

BACKGROUND: On occasion, patients followed with positron emission tomographic (PET)/computed tomographic (CT) imaging for nonbreast malignancies will have incidental breast findings concerning for second primary breast cancers. The aim of this study was to determine the predictive value of PET/CT imaging to identify breast cancers in these patients. METHODS: Patients with primary nonbreast malignancies and findings concerning for second primary breast cancers were identified from a prospectively acquired nuclear medicine database from January 2005 to July 2008. Chart reviews were then performed. RESULTS: Nine hundred two women underwent PET/CT imaging to evaluate nonbreast malignancies. Nine women (1%) had concerning breast findings, and 5 (56%) had subsequent breast cancer diagnoses. The positive predictive value of PET/CT imaging in these patients was 63%. Evidence of compliance with current screening guidelines was present in only 22% of these patients. CONCLUSIONS: The data suggest that findings concerning for an additional primary breast cancer should be evaluated and that age-appropriate screening tools should not be abandoned.

Incidental PET/CT findings in the cancer patient: how should they be managed?

BACKGROUND: Despite a paucity of evidence-based guidelines, the use of PET/CT (positron emission tomography/computed tomography) in the management of cancer patients is increasing. As widespread clinical application increases, unexpected radiographic findings are occasionally identified. These incidental findings are often suspicious for a second primary malignancy. The purpose of this study was to determine the clinical impact of these incidental PET/CT findings. METHODS: A query of our prospectively acquired Nuclear Medicine database was performed to identify patients with a known malignancy being staged or serially imaged with PET/CT. Patients with incidental findings suggestive of a second primary malignancy were selected. Statistical analysis was performed to determine the ability of PET/CT to identify a second primary malignancy. All PET/CT were interpreted by board certified nuclear radiologists. RESULTS: Of 3,814 PET/CT scans performed on 2,219 cancer patients at our institution from January 1, 2005, to December 29, 2008, 272 patients (12% of all patients) had findings concerning for a second primary malignancy. An invasive work-up was performed on 49% (133/272) of these patients, while 15% (40/272) had no further evaluation due to an advanced primary malignancy. The remaining 36% (99/272) had no further evaluation secondary to a low clinical suspicion determined by the treating team, a clinical plan of observation, or patients lost to follow-up. Of the 133 patients evaluated further, clinicians identified a second primary malignancy in 41 patients (31%), benign disease in 62 patients (47%), and metastatic disease from their known malignancy in 30 patients (23%). The most common sites for a proven second primary malignancy were: lung (N = 10), breast (N = 7), and colon (N = 5). Investigation of these lesions was performed using several techniques, including 24 endoscopies (6 malignant). A surgical procedure was performed in 74 patients (29 malignant), and a percutaneous biopsy was performed on 34 patients (12 malignant). The overall positive predictive value for PET/CT to detect a second primary malignancy was 31% in this subgroup. At a median follow-up of 22 months, 9 of 41 patients with a second primary were dead of a malignancy, 20 were alive with disease, and 12 had no evidence of disease. CONCLUSION: Incidental PET/CT findings consistent with a second primary are occasionally encountered in cancer patients. In our data, approximately half of these findings were benign, a third were consistent with a second primary malignancy or a metastatic focus, and the remainder were never evaluated due to physician and patient decision. Advanced primary tumors are unlikely to be impacted by a second primary tumor suggesting that this subset of patients will not benefit from further investigation. Our data suggests that, despite the high rate of false positivity, incidental PET/CT findings should be investigated when the results will impact treatment algorithms. The timing and route of investigation should be dictated by clinical judgment and the status of the primary tumor. Further investigation will need to be performed to determine the long-term clinical impact of incidentally identified second primary malignancies.

Mixed reality humans: evaluating behavior, usability, and acceptability.

This paper presents Mixed Reality Humans (MRHs), a new type of embodied agent enabling touch-driven communication. Affording touch between human and agent allows MRHs to simulate interpersonal scenarios in which touch is crucial. Two studies provide initial evaluation of user behavior with a MRH patient and the usability and acceptability of a MRH patient for practice and evaluation of medical students' clinical skills. In Study I (n=8) it was observed that students treated MRHs as social actors more than students in prior interactions with virtual human patients (n=27), and used interpersonal touch to comfort and reassure the MRH patient similarly to prior interactions with human patients (n=76). In the within-subjects Study II (n=11), medical students performed a clinical breast exam on each of a MRH and human patient. Participants performed equivalent exams with the MRH and human patients, demonstrating the usability of MRHs to evaluate students' exam skills. The acceptability of the MRH patient for practicing exam skills was high as students rated the experience as believable and educationally beneficial. Acceptability was improved from Study I to Study II due to an increase in the MRH's visual realism, demonstrating that visual realism is critical for simulation of specific interpersonal scenarios.

A pilot study to integrate an immersive virtual patient with a breast complaint and breast examination simulator into a surgery clerkship.

BACKGROUND: We aimed to determine if an immersive virtual patient (VP) with a breast complaint and a breast mannequin could prepare third-year medical students for history-taking (HT) and clinical breast examination (CBE) on a real patient. METHODS: After standardized instruction in breast HT and CBE, students (n = 21) were randomized to either an interaction with a VP (experimental) or to no VP interaction (control) before seeing a real patient with a breast complaint. Participants completed baseline and exit surveys to assess confidence regarding their HT and CBE skills. RESULTS: Students reported greater confidence in their HT (Delta value = 1.05 +/- 1.28, P < .05) and CBE skills (Delta value = 1.14 +/- .91, P < .05) and less anxiety when performing a CBE (Delta value = -.76 +/- 1.10, P < .05). The VP intervention group had a significantly higher mean HT confidence than the control group at the conclusion of the study (4.27 +/- .47 vs 3.50 +/- .71, respectively, P < .05). CONCLUSIONS: A single interaction with a VP with a breast complaint and breast mannequin improves student confidence in breast HT during a surgery clerkship.

Medical student satisfaction using a virtual patient system to learn history-taking communication skills.

Virtual patients (VPs) have the potential to augment existing medical school curricula to teach history-taking and communication skills. A goal of our current efforts to study virtual characters in health professions education is to develop a system that can be independently accessed and thus user satisfaction is an important factor in how readily this technology will be adopted. Twenty-three medical students participated in a study in which they interviewed a virtual patient and were asked to rate the educational value of the experience. Despite some of the limitations in this developing technology, students were generally receptive to its use as an educational tool. Further enhancements to the system, including increased fidelity of the interaction and novel feedback mechanisms, should improve learner satisfaction with and adoption of the virtual patient system.

Survey of student education in surgery.

BACKGROUND: Few data exist comparing medical student surgical education programs across the US and Canada. We conducted a 30-question, Web-based survey of 142 medical schools in the US and Canada about the administration of undergraduate surgical education programs at their institutions. STUDY DESIGN: From October 2005 to January 2006, 142 Association of American Medical Colleges (AAMC) medical schools in the US and Canada were asked to complete a 30-item, Web-based questionnaire. Descriptive statistics were compiled for all responses and data were analyzed using Student's t-test. RESULTS: One hundred two medical schools responded to the survey (90 of 125 US and 12 of 17 Canadian schools, overall response rate,=72%). The majority of surgical clerkships have between 16 and 30 students rotating at a time through 2 to 4 teaching sites. The most frequent overall clerkship duration was 8 weeks, divided most frequently between a 4- to 6-week general surgery rotation and a 4-week subspecialty rotation. There was no consensus about what services constitute general surgery rotations, although the most common subspecialty rotations were urology (68%) and orthopaedics (66%). American medical schools were more likely to use the National Board of Medical Examiners (NBME) subject examination for student assessment (90.9% US versus 50.0% Canadian, p < 0.05); Canadian medical schools were more likely to use an Objective Structured Clinical Examination (69.2% Canadian versus 37.7% US, p < 0.05) and other written examinations (69.2% Canadian versus 37.7% US, p < 0.05). CONCLUSIONS: There is wide variation in the structure and administration of medical student surgical education programs in the US and Canada. These findings underscore the need for a clerkship directors committee and a national dialog about a core student curriculum to ensure consistency in the quality of student education in surgery.

Do medical students respond empathetically to a virtual patient?

BACKGROUND: Significant information exchange occurs between a doctor and patient through nonverbal communication such as gestures, body position, and eye gaze. In addition, empathy is an important trust-building element in a physician: patient relationship. Previous work validates the use of virtual patients (VP) to teach and assess content items related to history-taking and basic communication skills. The purpose of this study was to determine whether more complex communication skills, such as nonverbal behaviors and empathy, were similar when students interacted with a VP or standardized patient (SP). METHODS: Medical students (n = 84) at the University of Florida (UF) and the Medical College of Georgia (MCG) underwent a videotaped interview with either a SP or a highly interactive VP with abdominal pain. In the scenario, a life-sized VP was projected on the wall of an exam room in SP teaching and testing centers at both institutions. VP and SP scripted responses to student questions were identical. To prompt an empathetic response (ie, acknowledging the patients' feelings), during the interview the VP or SP stated "I am scared; can you help me?" Clinicians (n = 4) rated student videotapes with respect to nonverbal communication skills and empathetic behaviors using a Likert-type scale with anchored descriptors. RESULTS: Clinicians rated students interacting with SPs higher with respect to the nonverbal communication skills such as head nod (2.78 +/- .79 vs 1.94 +/- .44, P < .05), and body lean (2.97 +/- .94 vs 1.93 +/- .58, P < .05), level of immersion in the scenario (3.31 +/- .49 vs 2.26 +/- .52, P < .05), anxiety (1.16 +/- .31 vs 1.45 +/- .33, P < .05), attitude toward the patient (3.24 +/- .43 vs 2.89 +/- .36, P < .05), and asking clearer questions (3.06 +/- .32 vs 2.51 +/- .32, P < .05) compared to the VP group. The students in the SP group also had a higher empathy rating (2.75 +/- .86 vs 2.16 +/- .83, P < .05) and better overall rating (4.29 +/- 1.32 vs 3.24 +/- 1.06, P < .05) than the VP group. Empathy was positively correlated with the observed nonverbal communication behaviors. Eye contact was the most strongly correlated with empathy (r = .57, P < .001), followed by head nod (r = .55, P < .001) and body lean (r = .49, P < .001). CONCLUSIONS: Medical students demonstrate nonverbal communication behaviors and respond empathetically to a VP, although the quantity and quality of these behaviors were less than those exhibited in a similar SP scenario. Student empathy in response to the VP was less genuine and not as sincere as compared to the SP scenario. While we will never duplicate a real physician/patient interaction, virtual clinical scenarios could augment existing SP programs by providing a controllable, secure, and safe learning environment with the opportunity for repetitive practice.

Comparing interpersonal interactions with a virtual human to those with a real human.

This paper provides key insights into the construction and evaluation of interpersonal simulators--systems that enable interpersonal interaction with virtual humans. Using an interpersonal simulator, two studies were conducted that compare interactions with a virtual human to interactions with a similar real human. The specific interpersonal scenario employed was that of a medical interview. Medical students interacted with either a virtual human simulating appendicitis or a real human pretending to have the same symptoms. In Study I (n = 24), medical students elicited the same information from the virtual and real human, indicating that the content of the virtual and real interactions were similar. However, participants appeared less engaged and insincere with the virtual human. These behavioral differences likely stemmed from the virtual human's limited expressive behavior. Study II (n = 58) explored participant behavior using new measures. Nonverbal behavior appeared to communicate lower interest and a poorer attitude toward the virtual human. Some subjective measures of participant behavior yielded contradictory results, highlighting the need for objective, physically-based measures in future studies.

The use of virtual patients to teach medical students history taking and communication skills.

BACKGROUND: At most institutions, medical students learn communication skills through the use of standardized patients (SPs), but SPs are time and resource expensive. Virtual patients (VPs) may offer several advantages over SPs, but little data exist regarding the use of VPs in teaching communication skills. Therefore, we report our initial efforts to create an interactive virtual clinical scenario of a patient with acute abdominal pain to teach medical students history-taking and communication skills. METHODS: In the virtual scenario, a life-sized VP is projected on the wall of an examination room. Before the virtual encounter, the student reviews patient information on a handheld tablet personal computer, and they are directed to take a history and develop a differential diagnosis. The virtual system includes 2 networked personal computers (PCs), 1 data projector, 2 USB2 Web cameras to track the user's head and hand movement, a tablet PC, and a microphone. The VP is programmed with specific answers and gestures in response to questions asked by students. The VP responses to student questions were developed by reviewing videotapes of students' performances with real SPs. After obtaining informed consent, 20 students underwent voice recognition training followed by a videotaped VP encounter. Immediately after the virtual scenario, students completed a technology and SP questionnaire (Maastricht Simulated Patient Assessment). RESULTS: All participants had prior experience with real SPs. Initially, the VP correctly recognized approximately 60% of the student's questions, and improving the script depth and variability of the VP responses enhanced most incorrect voice recognition. Student comments were favorable particularly related to feedback provided by the virtual instructor. The overall student rating of the virtual experience was 6.47 +/- 1.63 (1 = lowest, 10 = highest) for version 1.0 and 7.22 +/- 1.76 for version 2.0 (4 months later) reflecting enhanced voice recognition and other technological improvements. These overall ratings compare favorably to a 7.47 +/- 1.16 student rating for real SPs. CONCLUSIONS: Despite current technological limitations, virtual clinical scenarios could provide students a controllable, secure, and safe learning environment with the opportunity for extensive repetitive practice with feedback without consequence to a real or SP.

Virtual patients: assessment of synthesized versus recorded speech.

Virtual patients have great potential for training patient-doctor communication skills. There are two approaches to producing the virtual human speech: synthesized speech or recorded speech. The tradeoffs in flexibility, fidelity, and cost raise an interesting development decision: which speech approach is most appropriate for virtual patients? Two groups of medical students participated in a user study interviewing a virtual patient under each condition. We found no significant differences in the overall impression, speech intelligibility, and task performance. Our conclusion is that if the goal is to train students of which questions to ask, synthesized speech is just as effective as recorded speech. However, if the goal is to teach the student how to ask the correct questions, a high level of expressiveness in the virtual patient is needed. This in turn necessitates the higher cost - even with the lower flexibility - of recorded speech.

Minimizing local recurrence after breast conserving therapy using intraoperative shaved margins to determine pathologic tumor clearance.

BACKGROUND: This study analyzed survival, locoregional recurrence, and reexcision rates after breast conserving therapy, based on the margin analysis technique used at the University of Florida, which incorporates frozen section analysis of shaved breast tissue from the lumpectomy cavity. STUDY DESIGN: A retrospective review was done of 257 patients who underwent 267 consecutive lumpectomy operations and completed radiation therapy at our institution. Margins in 189 patients were assessed using frozen section analysis of shaved margins from the lumpectomy cavity. RESULTS: Breast conserving therapy was performed for 220 (83%) patients with early breast cancer (T1 and T2 tumors) and 47 (17%) with ductal carcinoma in situ. With a median followup of 5.6 years, the crude locoregional recurrence rates for patients who had margins analyzed intraoperatively by frozen section analysis or margins analyzed by permanent analysis were 1.9% (3 of 157) and 3.1% (2 of 63), respectively, for early breast cancer and 15.6% (5 of 32) and 6.6% (1 of 15) for ductal carcinoma in situ (p=NS). Survival rates were 97% and 78%, at 5 and 10 years, respectively, for the early breast cancer patients, and 98% and 98%, respectively, for ductal carcinoma in situ patients. Permanent intraoperative frozen section analysis reexcision rates were 33.3% (26 of 78) and 5.8% (11 of 189). CONCLUSIONS: Regardless of the technique used for margin analysis, breast conserving therapy led to low locoregional recurrence relative to national figures, pointing to the importance of the technique of radiation therapy at our institution. Reexcision rate was reduced with the use of frozen section analysis.

Adjuvant radiation therapy for resectable retroperitoneal soft tissue sarcoma: the University of Florida experience.

BACKGROUND: In the management of retroperitoneal sarcomas it is necessary to achieve local control to ensure survival. The role of adjuvant radiation therapy (RT), either pre- or postoperative, remains controversial. METHODS: Outcomes for 40 patients with retroperitoneal sarcoma treated with surgery and postoperative RT (n = 25) or preoperative RT (n = 15) were analyzed for variables prognostic for local control, survival, and associated complications. RESULTS: Patterns of failure for patients treated by resection and postoperative RT were local (n = 4), local and distant (n = 3), and distant (n = 3). The failure patterns for preoperative RT cases were local (n = 2), local and distant (n = 2); and distant (n = 4). Median time to local recurrence in the postoperative and preoperative RT series were 1 year and 2.5 years respectively. The margin status was predictive for local control (P = 0.0065) and survival (P = 0.0012), regardless of treatment sequence. Absolute 5-year survival was 12% with positive margins versus 69% if negative. Histologic grade was indicative of the risk for distant metastasis (low grade 8% vs high grade 64%; P = 0.1373), and significantly predicted 5-year absolute survival (low grade 77% vs high grade 34%; P = 0.0267). Postoperative RT was associated with significant complications (infection, hemorrhage, and bowel obstruction--2 cases each). CONCLUSION: Compared with the surgery-alone series, adjuvant RT appears to improve the probability of local control. Preoperative RT may be the preferred sequence potentially to improve tumor resectability and local-regional control with less risk of complications than with postoperative RT.

Arginine and cancer.

Arginine is a dibasic, cationic, semiessential amino acid with numerous roles in cellular metabolism. It serves as an intermediate in the urea cycle and as a precursor for protein, polyamine, creatine and nitric oxide (NO) biosynthesis. Arginine is conditionally essential since it becomes necessary under periods of growth and after recovery after injury. Arginine also promotes wound healing and functions as a secretagogue stimulating the release of growth hormone, insulin-like growth factor 1, insulin, and prolactin. Furthermore, arginine has several immunomodulatory effects such as stimulating T- and natural killer cell activity and influencing pro-inflammatory cytokine levels. The discover that l-arginine is the sole precursor for the multifunctional messenger molecule nitric oxide (NO) led to investigation into the role of arginine in numerous physiologic and pathophysiologic phenomena including cancer. Although NO was first identified in endothelial cells, it is now recognized to be generated by a variety of cell types, including several tumor cell lines and solid human tumors. Unfortunately, the precise role of NO in cancer is poorly understood but it may influence tumor initiation, promotion, and progression, tumor-cell adhesion, apoptosis angiogenesis, differentiation, chemosensitivity, radiosensitivity, and tumor-induced immunosuppression. The biological effects of NO are complex and dependent upon numerous regulatory factors. Further research is necessary to enhance our understanding of the complex mechanisms that regulate NO's role in tumor biology. A better understanding of the role of arginine-derived NO in cancer may lead to novel antineoplastic and chemopreventative strategies.

Should decisions on internal mammary lymph node irradiation be based on current lymphoscintigraphy techniques for sentinel lymph node identification?

BACKGROUND: Treatment of internal mammary lymph node (IMN) metastases remains controversial because of the difficulty in predicting involvement, potential treatment-related morbidity, and questionable efficacy. Lymphoscintigraphy with sentinel lymph node biopsy offers a means to identify occult involvement of IMN, allowing appropriate patient selection for IMN treatment. METHODS: The authors retrospectively reviewed 262 lymphoscintigraphies (LS) of 248 patients treated at the University of Florida (Gainesville, FL) between 1998 and 2002. Tumor characteristics were assessed for their value in predicting IMN drainage and their association with IMN radiation. RESULTS: Lymph flow to the IMN was documented with LS in 23 of 262 tumor specimens (9%). Flow to the IMN was not correlated with any of the five factors: tumor location, tumor size, lymphovascular invasion, pathologic lymph node status, and laterality of the involved breast (right vs. left breast). Identification of IMN flow increased from 5.7% to 10.1% with the use of a deep injection technique. IMN radiotherapy was used more frequently in patients with larger tumors (15 of 188 in Tis/T1 vs. 31 of 70 in T2-T4; P<0.0001) and positive lymph nodes (17 of 91 in lymph node-negative patients vs. 28 of 66 in lymph node-positive patients; P<0.0001). In patients with T2N0 tumors (n=32), IMN radiotherapy was used more frequently with medial tumors (5 of 11 [45%]) than with lateral tumors (4 of 21 [19%]). CONCLUSIONS: The incidence of flow to the IMN documented with the current LS technique was low compared with other LS and extended radical mastectomy series. Histopathologic information was obtained for the sentinel IMN when IMN flow was identified on the LS. In the absence of histopathologic information, treatment decisions should continue to be based on clinical factors known to be correlated with occult IMN involvement.