Kamal, Arif H.

9592 Background: Palliative care (PC) is an essential part of the continuum of care for cancer (CA) patients (pts). Little is known about the aggregate characteristics and survival of pts receiving inpatient palliative care consultation (PCC). Methods: We reviewed data prospectively collected on patients seen by the Palliative Care Inpatient Consult Service at Mayo Clinic - Rochester from 2003–2008. Demographics, consult characteristics, and survival were analyzed. Kaplan-Meier survival curves and a Cox model of survival were produced. Results: 1794 total patients were seen over the five year period. Cancer is the most common primary diagnosis (47%). Growth in annual PCC has risen dramatically (113 in 2003 vs. 414 in 2007) despite stable total hospital admissions. Patient are predominantly men (52% vs. 48%, p=0.02); median age is 76. General medicine, medical cardiology, and medical intensive care unit services refer most often. Most frequent issues addressed are goals of care, dismissal planning, and pain control (29%, 19%, 17%). PCC in actively dying pts have increased with 27% of all non-operating room, non-trauma in-hospital deaths being seen. Although CA pts have the highest median survival after PCC vs. other diagnoses (17 days, p = 0.018), we observed a five-year trend of decreasing survival from admission to death and PCC to death. Median time from admission to death in CA pts is 36 days in 2003 and 19 days in 2008 (p<0.01). Median time from PCC to death is 33 versus 11.5 days (p<0.01). Despite this, median hospital length of stay and time from PCC to discharge have remained fixed at 8 and 2.5 days, respectively. A Cox model of survival to discharge and <6 months survival (hospice eligibility) shows hospital length of stay, time from consult to discharge, and dismissal location from hospital are all prognostic factors. Conclusions: Survival window for PC intervention for CA pts is lessening. With the trend of shorter survival after PCC, PC professionals have little over two days to implement a comprehensive, ongoing care plan. This highlights the importance of earlier outpatient palliative care involvement with advanced cancer patients and families. No significant financial relationships to disclose.

Abstract All-trans retinoic acid (ATRA) syndrome is a complex constellation of symptoms that arise in about 15% of patients who receive retinoic acid as therapy for acute promyelocytic leukemia (PML). This syndrome typically consists of fever, dyspnea, weight gain, pulmonary infiltrates, pleural or pericardial effusions, episodic hypotension, renal dysfunction, and leukocytosis. Rarely does this entity present as typical chest pain suggestive of an acute myocardial infarction. A 39-year-old man diagnosed with acute PML treated with idarubicin and ATRA was admitted for febrile neutropenia, nausea, vomiting, cough, and epigastric pain despite being on ciprofloxacin and metronidazole at home. He continued to have low-grade fevers despite being on antibiotics and antifungals. On hospital day three, he was transferred to the CCU for worsening non-radiating left substernal chest pain associated with nausea and vomiting, new inferior lead 2 mm ST segment elevation, lateral lead T wave inversion, troponin and CKMB elevation, and hypotension not responding to frequent fluid boluses. A Swan-Ganz catheter was placed and pressors were started. The patient was found to have a dramatic neutrophil-predominant leukocytosis only hours after a CBC showed neutropenia. He also developed anuric acute renal failure. A cardiac catheterization was considered but could not be performed because of thrombocytopenia and the patient’s renal status. Subsequent EKG’s revealed more impressive changes. New 3 mm ST segment elevations and T wave inversions were noted in the lateral leads, the inferior lead ST segment elevations persisted and new inferior lead T wave inversions were present. An emergent echocardiogram was obtained which showed a significantly decreased LVEF at 36% and global LV hypokinesis without pericardial effusion. The constellation of symptoms was more suggestive of ATRA syndrome versus acute coronary syndrome and intravenous dexamethasone was started. The patient’s chest pain resolved spontaneously over the next three hours, the dramatic EKG findings and elevated cardiac enzymes resolved over the next few days, and the patient was transferred back to the admitting hematology service. A follow-up echocardiogram before discharge revealed a normal LVEF at 74% and no residual wall motion abnormalities. He eventually required hemodialysis for continued renal failure, which was continued as an outpatient after discharge. Of note, the patient’s fever resolved after beginning dexamethasone and multiple blood, urine, stool, and sputum cultures obtained remained negative throughout the hospital stay. This clinical situation presents a very distinct dilemma in how to proceed with clinical care. The presenting symptoms, signs, diagnostics, and imaging alone were all suggestive of a worsening ischemic cardiac process. But in considering the clinical presentation as a whole, especially the concomitant renal failure, fever in the setting of broad-spectrum coverage, and dramatic leukocytosis, we were more convinced of a systemic reaction to recent ATRA use. We maintained a low threshold for beginning an acute coronary syndrome protocol involving antiplatelet agents but were supported in our diagnosis by the patient’s remarkable recovery shortly after beginning intravenous steroids.

Abstract Systemic capillary leak syndrome (SCLS) is a vascular disorder characterized by intermittent episodes of hypotension and edema from increased vascular permeability. Patients present with resultant dramatic shifts in plasma volume from intravascular to extravascular spaces. Diagnosis requires observation of severe hemoconcentration and hypoalbuminemia during acute attacks. Typically, laboratory evaluation of hematocrit and albumin is normal when patients are not symptomatic. Concomitantly, and of unknown significance, a monoclonal protein is also frequently discovered. We present the case of a patient whose increased hematocrit suggested polycythemia vera, but whose total clinical picture of hypotension, hemoconcentration and hypoalbuminemia confirmed the diagnosis of SCLS. A 54-year old gentleman was referred for evaluation of polycythemia vera that was diagnosed after a series of dramatic hospitalizations for near-syncope. The patient had no chronic illnesses, negative family history for hematologic abnormalities, and was otherwise in his usual state of good health when a complaint of sudden lightheadedness and pre-syncope led to an emergency room evaluation. There he was found to be severely hypotensive and was resuscitated with intravenous fluids. Initial laboratory evaluation showed a hematocrit of 60%. Further workup revealed an increased red cell mass. For the presumptive diagnosis of polycythemia vera, the patient was phlebotomized once and, interestingly, did not require any further phlebotomy for almost three years. During this interim, he remained relatively asymptomatic with only occasional bouts of lightheadedness that would abort with increased fluid hydration. Phlebotomy was again used when the patient presented with near-syncope, hypotension, and vomiting accompanied by significant extremity swelling. The hematocrit was found to be greater than 80%. The patient recovered with aggressive fluid resuscitation and was discharged on a low-dose aspirin with the continued empiric diagnosis of polycythemia vera. During that year, the patient recovered from four other similar episodes that resolved after fluid resuscitation and where phlebotomy was used. The most severe episode involved a protracted ICU stay for severe hypotension complicated by severe extremity edema and resultant compartment syndrome and ischemic muscle necrosis. With the concern of adrenocortical dysfunction leading to these hypotensive episodes, the patient was started on low dose daily steroids. While on this, the patient reported less frequency and severity of his almost-daily symptoms of lightheadedness and edema. In fact, attempts at tapering the steroid dose down resulted in a return of the minor symptoms and with continued steroids, no further hospitalizations were needed. During our evaluation, we confirmed a normal hematocrit and albumin while asymptomatic and the presence of a monoclonal protein. This case represents a common clinical history of patients who are ultimately diagnosed with SCLS. Often, the significance of the observed hypotension in concert with laboratory evidence of hemocentration and hypoalbuminemia is not initially appreciated. Although current attack prophylaxis regimens for SCLS use terbutaline and theophylline, steroids had historically shown occasional benefit in some patients. SCLS still carries a significant mortality risk and prompt recognition of this syndrome and initiation of an appropriate prophylaxis regimen is crucial.

576 Background: In node-negative, ER + breast cancer, gene expression profiling can identify level of risk and, in the case of ODX, may also identify pts with a higher chance of benefiting from adjuvant chemotherapy. Because the gene profile in ODX includes an assessment of ER, HER2, and proliferation, we hypothesized that clinicians using standardized criteria could discriminate risk (high versus low/intermediate) as specified by ODX. Methods: We identified Mayo Clinic patients with node-negative, ER + breast cancer, for whom ODX scores were available. Tumor slides were reviewed by an expert breast pathologist to confirm tumor size, histology, and tumor grade. Both ER and PR were quantitated; HER-2 was determined by IHC (FISH, if 2+). These clinical cases were presented to six academic oncologists, blinded to the ODX score, to predict ODX risk (low, intermediate, or high) and give their recommendation for chemotherapy (CTX) (yes/no). Afterwards, they were presented with the same cases with the actual ODX score, to give recommendations regarding CTX. Results: ODX scores in tumors from 31 patients were low in 18 pts, intermediate in 10 pts, and high in 3 pts. Concordance between predicted and actual ODX scores being low/intermediate vs high exceeded 87% for each oncologist. The most frequent discrepancies were actual low scores predicted as intermediate (31/80 discordant) and actual intermediate scores predicted as low (29/80 discordant). Overall agreement of predicted scores (high vs low/intermediate) among the oncologists was substantial (kappa=0.75, p<0.0001). CTX recommendations following provision of the ODX scores changed on average 18.2% (range 12.9%-25.8%) of the time, with slightly fewer changing for, versus against, a CTX recommendation. Conclusions: Our findings suggest that the proper evaluation and interpretation of traditional prognostic criteria will identify most node negative, ER + patients at high risk of recurrence (as predicted by ODX) but poorly discriminate low versus intermediate risk. The provision of ODX data changed the recommendation for CTX in approximately 20% of cases. No significant financial relationships to disclose.

91 Background: Formal mechanisms to share data on quality remain immature in specialty palliative care. As the field grows, infrastructure that promotes collaboration among academic and community-based practice will be required to foster comparisons and benchmarking of data to inform areas for quality improvement. Further, such relationships will create a palliative care “quality improvement laboratory”, where proposed guidelines and best practices can be developed, implemented, and tested. Methods: We set out to bring together specialty palliative care practices with a shared vision for collaborative quality improvement. We modeled our approach after the Institute for Healthcare Improvement Breakthrough Series alongside our Rapid Learning Quality Improvement paradigm. We use a set of common data collection procedures, across an electronic point-of-care platform called Quality Data Collection Tool (QDACT), alongside a centralized data registry. Further, we meet and discuss challenges and issues, compare best practices, and brainstorm new projects through biweekly conference calls. Results: We have created a multi-institutional collaboration for quality assessment and improvement in specialty palliative care. Termed the Global Palliative Care Quality Alliance, we have brought together 11 academic and community organizations, both general and oncology-specific, across six states to study various areas of quality practice. Short-term, we will conduct rapid-cycling quality improvement projects addressing National Quality Forum domains for quality palliative care, including documentation of spiritual assessment and timely advance care planning. Long-term, we aim to study the link between quality measure adherence and outcomes and further align our initiatives with those of other large consortia, like the Palliative Care Research Cooperative and Palliative Care Quality Network. Conclusions: Collaborative quality improvement is needed in specialty palliative care across a national platform. Developing the infrastructure to perform standardized quality improvement is achievable across multiple palliative care settings.

90 Background: Increasingly, oncology teams and specialty palliative care (PC) clinicians are collaborating in the care of patients with serious or advanced cancer. Because of such partnerships, PC services have grown tremendously, with teams now available in most US cancer centers. Despite increased penetrance of services, little is known about the workforce that delivers this care, including their demographics, training, and future plans. Methods: Members of the American Academy of Hospice and Palliative Medicine (e.g. advanced practice provider (APP), registered nurses, chaplains) completed a workforce characteristics survey through electronic invitation in the second half of 2013; respondents were recruited via e-mail, blog and Facebook posts, and Twitter tweets. Descriptive statistics were calculated. Results: 1,241 clinicians responded, representing approximately a third of those who received the survey; 68% were physicians and 14% were APPs. Most (56%) were older than 50 years of age; 65% were female. Of physicians, 58% were board-certified through the experiential track, while 22% had completed fellowship training. Of clinical responsibilities, about half (48%) split time between consultative PC and hospice; 30% performed PC consultations only. Regarding organization size, 35% work with 1-3 other colleagues; 8% work alone. Regarding tenure, 34% have worked in PC for <5 years; 22% for >10 years. 82% reported they would become a PC clinician again, if asked to revisit their career choice. Of motivations for joining clinical PC, leading reasons were “opportunity for personal growth” and “influential experience during practicing another specialty”. About half (47%) intend to leave the field within 10 years. Of reasons to leave, 49% will retire; 24% may leave due to burnout; and 20% reported dissatisfaction with practice. Conclusions: The specialty PC workforce is diverse and faces sustainability challenges due to advancing physician age, impending retirements, evolving burnout, and dissatisfaction with the profession. If patients are going to have adequate access to services, research and approaches are needed to develop solutions that sustain and grow specialty PC.

87 Background: The increasing demand for specialist palliative care (PC) in cancer care requires an available and responsive clinical workforce. But as the volume of work expands, PC providers face an increased risk of burnout, characterized by depersonalization and work-related emotional exhaustion. The prevalence and predictors of burnout among PC providers is poorly understood. Methods: Members of the American Academy of Hospice and Palliative Medicine (e.g. advanced practice provider, registered nurses, chaplains) completed the Maslach Burnout Inventory (MBI) via electronic survey in the second half of 2013; respondents were recruited via e-mail, blog and Facebook posts, and Twitter tweets. The MBI determine burnout severity across two domains, Emotional Exhaustion (EE) and Depersonalization (DP). Severity is reported as “low”, “moderate”, or “high”; “high” is consistent with burnout. Demographic and work-related variables associated with burnout in other studies were determined via stepwise logistic regression. Results: We received surveys from 1,241 clinicians. We estimated a response rate of 30%. 68% of respondents were physicians. Most respondents were over age 50 (57%), female (65%), were married or partnered (82%) and had worked in the field for less than 10 years (67%). 42% took overnight call regularly, 30% reported working 50 hours per week or more, and 57% had at least 4 colleagues. Regarding burnout, 24% reported high DP, 59% reported high EE, and 62% of reported high burnout symptoms on either EE or DP scales. In logistic regression, younger physicians, those working more than 50 hours per work, and those with fewer colleagues within their practice were at greatest risk of burnout (p<0.02). Conclusions: PC clinicians report a burnout rate of 62%. Even with the limitations of our methods and potentially biased sampling, this prevalence is remarkable and higher than data reported in medical oncology (45%). Burnout severity is associated with working in isolation and working longer hours. Further studies on how burnout affects sustainability of the PC workforce are needed, especially since this workforce is so critical to the provision of high quality cancer care.

Context: Differences among patient populations that present to consultative palliative care are not known. Such an appreciation would inform health-care delivery tailored to unique populations. Objectives: We aimed to compare characteristics and palliative care needs of African Americans (AAs) and whites during initial palliative care consultation. Methods: We analyzed patient-reported, clinician-entered clinical encounter data from a large, multisite community-based, nonhospice palliative care collaborative. We included first specialty palliative care consultations from January 1, 2014, to July 2, 2015, across 15 sites within the Global Palliative Care Quality Alliance registry. Demographics, disease, performance status, advance care planning, and symptom prevalence/severity were compared. Results: Of 775 patients, 12.9% (N = 100) were AA. African Americans were younger (63 vs 75.4 years, P < .0001). A larger proportion of AAs had a diagnosis of cancer (45.0% vs 36.3%, P = .09) and in the hospital (71% vs 61.8%, P = .07). African Americans were more likely to have a Palliative Performance Score of 0 to 30 (35.6% vs 23.7%, P = .049). Around 50% in both racial groups were full code; slightly more than 40% had an advance directive. Nearly two-thirds in both racial groups reported 3 or more symptoms of any severity; one-third reported 3 or more moderate or severe symptoms. A larger proportion of Africans than whites reported pain of any severity (66.0% vs 56.1%, P = .06). Conclusion: All patients present to palliative care consultations with significant symptom and advance care planning needs. Further research is needed to identify how to deliver palliative care: earlier, in noncancer conditions, and improve pain management in AA populations.