Faculty Bibliography

BACKGROUND:Despite an aging population and increasing number of geriatric trauma patients annually, gaps in our understanding of best practices for geriatric trauma patients persist. We know that trauma center care improves outcomes for injured patients generally, and palliative care processes can improve outcomes for disease-specific conditions, and our goal was to determine effectiveness of these interventions on outcomes for geriatric trauma patients. METHODS:A priori questions were created regarding outcomes for patients 65 years or older with respect to care at trauma centers versus nontrauma centers and use of routine palliative care processes. A query of MEDLINE, PubMed, Cochrane Library, and EMBASE was performed. Letters to the editor, case reports, book chapters, and review articles were excluded. GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology was used to perform a systematic review and create recommendations. RESULTS:We reviewed seven articles relevant to trauma center care and nine articles reporting results on palliative care processes as they related to geriatric trauma patients. Given data quality and limitations, we conditionally recommend trauma center care for the severely injured geriatric trauma patients but are unable to make a recommendation on the question of routine palliative care processes for geriatric trauma patients. CONCLUSIONS:As our older adult population increases, injured geriatric patients will continue to pose challenges for care, such as comorbidities or frailty. We found that trauma center care was associated with improved outcomes for geriatric trauma patients in most studies and that utilization of early palliative care consultations was generally associated with improved secondary outcomes, such as length of stay; however, inconsistency and imprecision prevented us from making a clear recommendation for this question. As caregivers, we should ensure adequate support for trauma systems and palliative care processes in our institutions and communities and continue to support robust research to study these and other aspects of geriatric trauma. LEVEL OF EVIDENCE:Systematic review/guideline, level III.

BACKGROUND:Palliative Care (PC) is indicated in patients with functional dependency and advanced care needs in addition to those with life-threatening conditions. Older trauma patients have PC needs due to increased risk of mortality and poor long-term outcomes. We hypothesized that older trauma patients discharged alive with poor outcomes are not easily identified nor receive PC interventions. METHODS:Prospective observational study of trauma patients 55 years or older. Patients with poor functional outcomes defined by discharge Glasgow Outcome Scale Extended (GOSE) 1-4 or death at 6-month follow-up were analyzed for rate and timing of PC interventions including goals of care conversation (GOCC), do-not-resuscitate (DNR) order, do not intubate (DNI) order, and withdrawal of life supporting measures. Logistic regression was performed for having and timing of GOCC. RESULTS:Three hundred fifteen (54%) of 585 patients had poor outcomes. Of patients who died, 94% had GOCC compared with 31% of patients who were discharged with GOSE 3 or 4. In patients who died, 85% had DNR order, 18% had DNI order, and 56% had withdrawal of ventilator. Only 24% and 9% of patients with GOSE of 3 or 4, respectively, had DNR orders. Fifty percent of the patients who were dead at 6-month follow-up had GOCC during initial hospitalization. The median time to DNR in patients that died was 2 days compared with 5 days and 1 day in GOSE 3 and 4 (p = 0.046). Age, injury severity scale, and preexisting limited physiological reserve were predictive of having a GOCC. CONCLUSION:The PC utilization was very high for older trauma patients who died in hospital. In contrast, the majority of those who were discharged alive, but with poor outcomes, did not have PC. Development of triggers to identify older trauma patients, who would benefit from PC, could close this gap and improve quality of care and outcomes.

INTRODUCTION/BACKGROUND: Older patients are more vulnerable to poor outcomes after trauma than younger patients. Sarcopenia, loss of skeletal mass, is prevalent in trauma patients admitted to the intensive care unit (ICU), and it has been shown to correlate with adverse outcomes, such as mortality and ICU days. Yet, little is known whether it predicts other outcomes. We hypothesized that sarcopenia independently predicts poor functional outcomes in older trauma patients admitted to the ICU. METHODS:We performed a retrospective review of patients aged >55 admitted to a surgical ICU in a Level I trauma center for two years. Sarcopenic status was determined by measuring total skeletal muscle cross-sectional area at the L3 level on admission computed tomography (CT), normalized for height with sex-specific cutoffs. Primary outcome measures were in-hospital mortality, functional outcomes measured by the Glasgow Outcome Scale (GOS) at discharge, and discharge disposition. Multivariable logistic regression was used to determine predictors of primary outcomes. RESULTS:Out of 230 patients, 32% were sarcopenic. The overall mortality was 20%, and 30% were discharged with poor functional outcomes. A higher proportion of sarcopenic patients among survivors had poor functional outcomes at discharge (55% vs. 30%, p=0.002). Sarcopenia was not predictive of in-hospital mortality but was an independent predictor of poor functional outcomes at discharge (OR 2.6; 95% confidence interval [CI] 1.3-5.5), adjusting for age, Glasgow Coma Scale (GCS) on admission, diagnosis of traumatic brain injury (TBI), Injury Severity Score (ISS), and the number of life-limiting illnesses. CONCLUSIONS:Sarcopenia is prevalent in geriatric trauma ICU patients and is an independent predictor of poor functional outcomes. Assessing for sarcopenia has an important potential as a prognostic tool in older trauma patients.

BACKGROUND:Elderly patients presenting with a traumatic brain injury (TBI) often have comorbidities that increase risk of thromboembolic (TE) disease and recurrent TBI. A significant number are on anticoagulant therapy at the time of injury and studies suggest that continuing anticoagulation can prevent TE events. Understanding bleeding, recurrent TBI, and TE risk after TBI can help to guide therapy. Our objectives were to 1) evaluate the incidence of bleeding, recurrent TBI, and TE events after an initial TBI in older adults and 2) identify which factors contribute to this risk. METHODS:Retrospective analysis of Medicare claims between May 30, 2006 and December 31, 2009 for patients hospitalized with TBI was performed. We defined TBI for the index admission, and hemorrhage (gastrointestinal bleeding or hemorrhagic stroke), recurrent TBI, and TE events (stroke, myocardial infarction, deep venous thrombosis, or pulmonary embolism) over the following year using ICD-9 codes. Unadjusted incidence rates and 95% confidence intervals (CIs) were calculated. Risk factors of these events were identified using logistic regression. RESULTS:Among beneficiaries hospitalized with TBI, incidence of TE events (58.6 events/1000 person-years; 95% CI 56.2, 60.8) was significantly higher than bleeding (23.6 events/1000 person-years; 95% CI 22.2, 25.1) and recurrent TBI events (26.0 events/1000 person-years; 95% CI 24.5, 27.6). Several common factors predisposed to bleeding, recurrent TBI, and TE outcomes. CONCLUSIONS:Among Medicare patients hospitalized with TBI, the incidence of TE was significantly higher than that of bleeding or recurrent TBI. Specific risk factors of bleeding and TE events were identified which may guide care of older adults after TBI.

BACKGROUND:Few patients with dysphagia because of stroke receive early palliative care (PC) to align treatment goals with their values, as called for by practice guidelines, particularly before enteral access procedures for artificial nutrition. MEASURES:To increase documented goals of care (GOC) discussions among acute stroke patients before feeding gastrostomy tube placement. INTERVENTION:We undertook a rapid-cycle continuous quality improvement process with interdisciplinary planning, implementation, and performance review to operationalize an upstream trigger for PC referral prompted by the speech and language pathology evaluation. OUTCOMES:During a six-month period, 21 patients underwent gastrostomy tube placement; 52% had preprocedure GOC discussions postintervention, with the rate of compliance increasing steadily from 13% (11/87, preintervention) to 100% (2/2) in the final two months. CONCLUSIONS/LESSONS LEARNED:We effectively increased documented GOC discussions before feeding gastrostomy tube placement among stroke patients. Systems-based tools and education will enhance this upstream trigger model to ensure early PC for stroke patients.

BACKGROUND:Coronary artery bypass graft (CABG) operations are associated with all-cause readmission rates of approximately 15%. In attempts to reduce readmission rates, the Hospital Readmission Reduction Program expanded to include CABG operations in 2015. The aim of this study was therefore to develop a predictive readmission scale that would identify patients at higher risk of readmission after CABG using commonly available administrative data. METHODS:Data of 126,519 patients from California and New York (derivation cohort) and 94,318 patients from Florida and Washington (validation cohort) were abstracted from the State Inpatient Database (2006 to 2011). The readmission after CABG scale was developed to predict 30-day readmission risk and was validated against a separate cohort. RESULTS:Thirty-day CABG readmission rates were 23% in the derivation cohort and 21% in the validation cohort. Predictive factors included older age, female gender (odds ratio [OR], 1.34), African American ethnicity (OR, 1.13), Medicare or Medicaid insurance, and comorbidities, including renal failure (OR, 1.56) and congestive heart failure (OR, 2.82). These were independently predictive of increased readmission rates (p < 0.01). The readmission scale was then created with these preoperative factors. When applied to the validation cohort, it explained 98% of the readmission variability. CONCLUSIONS:The readmission after CABG scale reliably predicts a patient's 30-day CABG readmission risk. By identifying patients at high-risk for readmission before their procedure, risk reduction strategies can be implemented to reduce readmissions and healthcare expenditures.

BACKGROUND:When patients with dementia develop acute surgical abdomen, patients, surrogates, and surgeons need accurate prognostic information to facilitate goal-concordant decision making. Palliative care can assist with communication, symptom management, and family and caregiver support in this population. We aimed to characterize outcomes and patterns of palliative care utilization among patients with dementia, presenting with abdominal surgical emergency. METHOD:We retrospectively queried the National Inpatient Sample for patients aged >50 years with dementia and acute abdominal emergency who were admitted nonelectively 2009-2013, utilizing ICD-9-CM codes for dementia and surgical indication. We characterized outcomes and identified predictors of palliative care utilization. RESULTS:Among 15,209 patients, in-hospital mortality was 10.2%, the nonroutine discharge rate was 67.2%, and 7.5% received palliative care. Patients treated operatively were less likely to receive palliative care than those who did not undergo operation (adjusted OR = 0.50; 95% CI 0.41-0.62). Only 6.4% of patients discharged nonroutinely received palliative care. CONCLUSION:Patients with dementia and acute abdominal emergency have considerable in-hospital mortality, a high frequency of nonroutine discharge, and low palliative care utilization. In this group, we discovered a large gap in palliative care utilization, particularly among those treated operatively and those who are discharged nonroutinely.

BACKGROUND:The purpose of this study is to examine follow-up rates and maturation rates after dialysis access surgery using a fistula-first approach in a county hospital with an indigent population. METHODS:A prospectively maintained dialysis access database was queried for cases performed between August 1, 2009 and September 30, 2011. The follow-up period ended on December 31, 2011. An attempt was made to contact patients who did not have complete follow-up data recorded to the point of fistula maturation for arteriovenous fistulas (AVF). Patients were contacted directly or through their dialysis center. Maturation was defined by successful use of the AVF for hemodialysis for at least 2 weeks for patients who are dialysis dependent or by the clinical assessment of an attending vascular surgeon for those not on dialysis. RESULTS:Two hundred three dialysis access cases were performed. The mean age was 51.4 years. One hundred twenty-six (62%) were male and 175 (86%) were Hispanic. Of these, 194 (95.6%) were AVF. Three AVF were ligated in the postoperative period for steal. Of the remaining 191 AVF, 94 (49%) patients completed their scheduled follow-up appointments. Sixty-six (35%) patients did not come to clinic but were contacted. Twenty-six (14%) patients were completely lost to follow-up and 5 (2.6%) died. Maturation data was obtained on 160 AVF. Of those, 123 (77%) reached maturation with a mean maturation time of 112 ± 99 days (range, 21-483). Twenty patients who completed follow-up underwent at least 1 additional surgical or endovascular procedure in an attempt to achieve maturation. Seventeen (85%) patients who underwent a secondary procedure went on to achieve maturation. There was no significant difference in maturation between the group that completed follow-up and those who were contacted by phone (69 [73%] vs. 54 [81%], P = not significant). CONCLUSIONS:In this indigent population, follow-up does not influence maturation rates of AVF. Despite poor compliance with follow-up in the setting of a public hospital, the maturation rate of an aggressive fistula-first approach is acceptable.