Christophe Graf

Efficiency and applicability of comprehensive geriatric assessment in the Emergency Department: a systematic review

Background and aims: Comprehensive geriatric assessment (CGA) may benefit frail or chronically ill patients in the emergency department (ED), but take too much time to be performed routinely in ED. An alternative approach is to use first a screening tool to detect high-risk patients and then perform CGA in these patients only. This systematic review focuses on the use and value of CGA in ED for evaluation of older patients and its influence on adverse outcomes. This approach is compared with an alternative one using existing screening tools, validated in ED, to detect high-risk patients needing subsequent CGA. This review ends by suggesting a short assessment of CGA to be used in ED and ways to improve home discharge management from ED. Methods: A systematic English Medline literature search was conducted in December 2009, with no date limit with the following Medical Subject Heading (MeSH) terms: “Frail Elderly”, “Health Services for Aged”, “Community Health Nursing”, “Emergency Service, Hospital”, “Geriatric Assessment”, “Patient Discharge”, “Risk Assessment” and “Triage”. Results: We selected 8 studies on CGA efficiency and 14 on screening tools. CGA in ED is efficient for decreasing functional decline, ED readmission and possibly nursing home admission in high-risk patients. As CGA takes too much time to be performed routinely in ED, validated screening tools can be applied to detect high-risk patients who will benefit most from CGA. Conclusions: The selected studies demonstrated that screening of high-risk patients is more efficient than age-based screening, and that CGA performed in ED, followed by appropriate interventions, improves outcomes.

Prospective Comparison of 6 Comorbidity Indices as Predictors of 1-Year Post-Hospital Discharge Institutionalization, Readmission, and Mortality in Elderly Individuals

BACKGROUND Older patients often suffer from multiple comorbid conditions. Few comorbidity indices are valid and reliable in the elderly and were rarely compared. OBJECTIVE To compare the performance, relevance, and ability of 6 widely used and validated comorbidity indices--Charlson Comorbidity Index, Cumulative Illness Rating Scale-Geriatrics, Index of Coexistent Diseases, Kaplan, Geriatric Index of Comorbidity (GIC), and Chronic Disease Score--to predict adverse outcomes after discharge (1-year risk of rehospitalization, institutionalization, and death). DESIGN, SETTING, AND PARTICIPANTS Prospective study with 1-year follow-up, between January 2004 and December 2005 in 444 elderly patients (mean age, 85; 74% female) discharged from acute geriatric hospital, Geneva University Hospitals. RESULTS In univariate analyses, Cumulative Illness Rating Scale?Geriatrics and GIC were the predictors with the largest coefficient of determination for mortality with (R(2) of 9.3%, respectively 8.8%). GIC was also the only significant predictor of institutionalization (R(2) = 6.0%). Higher risk of readmission was significantly associated with GIC (R(2) = 14.0%), Cumulative Illness Rating Scale-Geriatrics (R(2) = 5.6%), Charlson Comorbidity Index (R(2) = 3.1%), and Chronic Disease Score (R(2) = 1.7). CONCLUSIONS Understanding how to efficiently predict these adverse outcomes in hospitalized elders is important for a variety of clinical and policy reasons. GIC and Cumulative Illness Rating Scale-Geriatrics may improve hospital discharge planning in a geriatric hospital treating very old patients with acute disease.

Identification of older patients at risk of unplanned readmission after discharge from the emergency department Comparison of two screening tools

STUDY HYPOTHESIS The Identification of Senior At Risk (ISAR) and the Triage Risk Stratification Tool (TRST) are the two most studied screening tools to detect high-risk patients for unplanned readmission after an emergency department (ED)-visit. Since their performance was unclear among ED-patients over 75 years, we evaluated their capacities to predict readmission at 1, 3, 6 and 12 months as well as their usefulness in avoiding unnecessary further comprehensive geriatric assessment (CGA) in negative screened patients. METHODS Historical cohort study with systematic routine data collection of functional status, comorbid conditions and readmission rate of patients released home after an ED-visit between 2007 and 2009 at the Geneva University Hospitals. RESULTS 345 patients were included (mean age 84y; 63% female). Readmission rates were 25%, 38%, 49%, and 60% at 1, 3, 6, and 12 months, respectively. Positive ISAR (≥2/6) and TRST (≥2/5) predicted modestly unplanned readmission at each time point (AUC range: 0.607-0.664). Prediction of readmission with ISAR or TRST was not modified after adjustment for variables significantly associated with readmission (being male, having poor functional or comorbid scores). In case of negative ISAR or TRST, their high negative predictive values (NPV) would safely allow avoiding 64 useless CGA (ISAR <2: 7/64 readmissions at 1 month). CONCLUSIONS Both ISAR and TRST tools predicted modestly unplanned readmission after an ED-visit among patients over 75 years. Nevertheless, due to their low specificity and high NPV these screening tools are useful to select elderly ED-patients who can safely return home without any further CGA.

Similarity and robustness of PET and SPECT binding parameters for benzodiazepine receptors.

The single photon emission computed tomography (SPECT) radiotracer [123I]iomazenil is used to assess benzodiazepine receptor binding parameters. These measurements are relative indices of benzodiazepine receptor concentration (B′max). To evaluate the ability of such indices in accurately accessing the B′max the authors compared them with absolute values of B′max, measured using positron emission tomography (PET). The authors performed SPECT, PET, and magnetic resonance imaging (MRI) studies on a group composed of seven subjects. For SPECT studies, the authors administered a single injection of [123I]iomazenil and estimated the total and specific distribution volumes (DVT SPECT, DVS SPECT) and the binding potential (BP) using unconstrained (BPSPECT) and constrained (BPC SPECT) compartmental models. For PET studies, the authors used a multiinjection approach with [11C]flumazenil and unlabeled flumazenil to estimate absolute values of receptor concentration, B′max, and some other binding parameters. The authors studied the correlation of different binding parameters with B′max. To study the robustness of the binding parameter measurements at the pixel level, the authors applied a wavelet-based filter to improve signal-to-noise ratio of time-concentration curves, and the calculated kinetic parameters were used to build up parametric images. For PET data, the B′max and the DVPET were highly correlated (r = 0.988). This confirms that it is possible to use the DVPET to access benzodiazepine receptor density. For SPECT data, the correlation between DVSPECT estimated using a two-and three-compartment model was also high (r = 0.999). The DVT SPECT and BPC SPECT parameters estimated with a constrained three-compartment model or the DVT″SPECT parameter estimated with a two-compartment model were also highly correlated to the B′max parameter estimated with PET. Finally, the robustness of the binding parameters allowed the authors to build pixel-by-pixel parametric images using SPECT data.

Low fat-free mass as a marker of mortality in community-dwelling healthy elderly subjects

BACKGROUND low fat-free mass has been related to high mortality in patients. This study evaluated the relationship between body composition of healthy elderly subjects and mortality. METHODS in 1999, 203 older subjects underwent measurements of body composition by bioelectrical impedance analysis, Charlson co-morbidity index and estimation of energy expenditure through physical activity by a validated questionnaire. These measurements were repeated in 2002, 2005 and 2008 in all consenting subjects. Mortality data between 1999 and 2010 were retrieved from the local death registers. The relationship between mortality and the last indexes of fat and fat-free masses was analysed by multiple Cox regression models. RESULTS womens and mens data at last follow-up were: age 81.1 ± 5.9 and 80.9 ± 5.8 years, body mass index 25.3 ± 4.6 and 26.1 ± 3.4 kg/m(2), fat-free mass index 16.4 ± 1.8 and 19.3 ± 1.9 kg/m(2) and fat mass index 9.0 ± 3.2 and 6.8 ± 2.0 kg/m(2). Fifty-eight subjects died between 1999 and 2010. The fat-free mass index (hazard ratio 0.77; 95% confidence interval 0.63-0.95) but not the fat mass index, predicted mortality in addition to sex and Charlson index. The multiple Cox regression model explained 31% of the variance of mortality. CONCLUSION a low fat-free mass index is an independent risk factor of mortality in elderly subjects, healthy at the time of body composition measurement.

Body composition and all-cause mortality in subjects older than 65 y

BACKGROUND A low or high body mass index (BMI) has been associated with increased mortality risk in older subjects without taking fat mass index (FMI) and fat-free mass index (FFMI) into account. This information is essential because FMI is modulated through different healthcare strategies than is FFMI. OBJECTIVE We aimed to determine the relation between body composition and mortality in older subjects. DESIGN We included all adults ≥65 y old who were living in Switzerland and had a body-composition measurement by bioelectrical impedance analysis at the Geneva University Hospitals between 1990 and 2011. FMI and FFMI were divided into sex-specific quartiles. Quartile 1 (i.e., the reference category) corresponded to the lowest FMI or FFMI quartile. Mortality data were retrieved from the hospital database, the Geneva death register, and the Swiss National Cohort until December 2012. Comorbidities were assessed by using the Cumulative Illness Rating Scale. RESULTS Of 3181 subjects included, 766 women and 1007 men died at a mean age of 82.8 and 78.5 y, respectively. Sex-specific Cox regression models, which were used to adjust for age, BMI, smoking, ambulatory or hospitalized state, and calendar time, showed that body composition did not predict mortality in women irrespective of whether comorbidities were taken into account. In men, risk of mortality was lower with FFMI in quartiles 3 and 4 [HR: 0.78 (95% CI: 0.62, 0.98) and 0.64 (95% CI: 0.49, 0.85), respectively] but was not affected by FMI. When comorbidities were adjusted for, FFMI in quartile 4 (>19.5 kg/m(2)) still predicted a lower risk of mortality (HR: 0.72; 95% CI: 0.54, 0.96). CONCLUSIONS Low FFMI is a stronger predictor of mortality than is BMI in older men but not older women. FMI had no impact on mortality. These results suggest potential benefits of preventive interventions with the aim of maintaining muscle mass in older men. This trial was registered at clinicaltrials.gov as NCT01472679.

Diabetes, comorbidities and increased long-term mortality in older patients admitted for geriatric inpatient care

AIMS To study the specific impact of diabetes on long-term mortality in very old subjects with multiple comorbidities and functional disabilities. METHODS The prevalence of vascular disorders, global comorbidity load (cumulative illness rating scale [CIRS]) and functional disabilities (activities of daily living [ADL] and Lawtons instrumental ADL [IADL] scores) were determined according to diabetes status in a cohort of 444 patients (mean age 85.3±6.7 years; 74.0% women) admitted to our geriatric service. Also, the specific impact of diabetes on 4-year mortality was analyzed using Cox proportional-hazards models. RESULTS Diabetic patients had higher BMI scores (27.1±4.9 vs. 23.4±4.7 kg/m(2) in controls; P<0.001), and higher prevalences of hypertension (81.9% vs. 65.1%, respectively; P=0.003) and ischaemic heart disease (33.7% vs. 22.2%, respectively; P=0.033), but not of stroke and renal insufficiency. They also had more comorbidities (CIRS score excluding diabetes: 15.1±4.5 vs. 13.8±4.8, respectively; P=0.016) and functional disabilities. Diabetes was associated with mortality (HR: 1.42, 95% CI: 1.02-1.99; P=0.041) after adjusting for age, gender and BMI, and this persisted after adjusting for individual vascular comorbidities, but disappeared after adjusting for CIRS, ADL or IADL scores. CONCLUSION Diabetes was associated with 4-year mortality after adjusting for the inverse relationship between mortality and BMI. This association was better accounted for by the global comorbidity load and functional disabilities than by the individual vascular comorbidities. These findings suggest that the active management of all--rather than selected--comorbidities is the key to improving the prognosis for older diabetic patients.

Telomere length, comorbidity, functional, nutritional and cognitive status as predictors of 5 years post hospital discharge survival in the oldest old

BackgroundTelomere length has been considered in many cross-sectional studies as a biomarker of aging. However the association between shorter telomeres with lower survival at advanced ages remains a controversial issue. This association could reflect the impact of other health conditions than a direct biological effect.ObjectiveTo test whether leukocyte telomere length is associated with 5-year survival beyond the impact of other risk factors of mortality like comorbidity, functional, nutritional and cognitive status.DesignProspective study.Setting and participantsA population representative sample of 444 patients (mean age 85 years; 74% female) discharged from the acute geriatric hospital of Geneva University Hospitals (January–December 2004), since then 263 (59.2%) had died (December 2009).MeasurementsTelomere length in leukocytes by flow cytometry.ResultsIn univariate model, telomere length at baseline and cognitive status were not significantly associated with mortality even when adjusting for age (R2=9.5%) and gender (R2=1.9%). The best prognostic predictor was the geriatric index of comorbidity (GIC) (R2=8.8%; HR=3.85) followed by more dependence in instrumental (R2=5.9%; HR=3.85) and based (R2=2.3%; HR=0.84) activities of daily living and lower albumin levels (R2=1.5%; HR=0.97). Obesity (BMI>30: R2=1.6%; HR=0.55) was significantly associated with a two-fold decrease in the risk of mortality compared to BMI between 20–25. When all independent variables were entered in a full multiple Cox regression model (R2=21.4%), the GIC was the strongest risk predictor followed by the nutritional and functional variables.ConclusionNeither telomeres length nor the presence of dementia are predictors of survival whereas the weight of multiple comorbidity conditions, nutritional and functional impairment are significantly associated with 5-year mortality in the oldest old.

High Levels of Comorbidity and Disability Cancel Out the Dementia Effect in Predictions of Long-Term Mortality after Discharge in the Very Old

Background/Aims: The relative weight of various etiologies of dementia as predictors of long-term mortality after other risk factors have been taken into account remains unclear. We investigated the 5-year mortality risk associated with dementia in elderly people after discharge from acute care, taking into account comorbid conditions and functionality. Methods: A prospective cohort study of 444 patients (mean age: 85 years; 74% female) discharged from the acute geriatric unit of Geneva University Hospitals. On admission, each subject underwent a standardized diagnostic evaluation: demographic variables, cognitive, comorbid medical conditions and functional assessment. Patients were followed yearly by the same team. Predictors of survival at 5 years were evaluated by Cox proportional hazards models. Results: The univariate model showed that being older and male, and having vascular and severe dementia, comorbidity and functional disability, were predictive of shorter survival. However, in the full multivariate model adjusted for age and sex, the effect of dementia type or severity completely disappeared when all the variables were added. In multivariate analysis, the best predictor was higher comorbidity score, followed by functional status (R2 = 23%). Conclusions: The identification of comorbidity and functional impairment effects as predictive factors for long-term mortality independent of cognitive status may increase the accuracy of long-term discharge planning.

Mild cognitive impairment, degenerative and vascular dementia as predictors of intra-hospital, short- and long-term mortality in the oldest old

Background and aims: The relative weight of various etiologies of dementia and mild cognitive impairment (MCI) as predictors of intra-hospital, short- and long-term mortality in very old acutely ill patients suffering from multiple comorbid conditions remains unclear. We investigated intra-hospital, 1- and 5-year mortality risk associated with dementia and its various etiologies in a very old population after discharge from acute care. Methods: Prospective cohort study of 444 patients (mean age 85 years; 74% female) discharged from the acute geriatric unit of Geneva University Hospital. On admission, each subject underwent standardized evaluation of cognitive and comorbid conditions. Patients were followed yearly by the same team. Predictive variables were age, sex, cognitive diagnosis, dementia etiology and severity. Survival during hospitalization, at 1- and 5-year follow-ups was the outcome of interest evaluated with Cox proportional hazard models. Results: Two hundred and six patients were cognitively normal, 48 had MCI, and 190 had dementia: of these, there were 75 cases of Alzheimer’s disease (AD), 20 of vascular dementia (VaD), 82 of mixed dementia (MD) and 13 of other types of dementia. The groups compared were statistically similar in age, sex, education level and comorbidity score. After 5 years of follow-up, 60% of the patients had died. Regarding intra-hospital mortality, none of the predictive variables was associated with mortality. MCI, AD and MD were not predictive of short- or long-term mortality. Features significantly associated with reduced survival at 1 and 5 years were being older, male, and having vascular or severe dementia. When all the variables were added in the multiple model, the dementia effect completely disappeared. Conclusions: Dementia (all etiologies) is not predictive of mortality. The observed VaD effect is probably linked to cardiovascular risk comorbidities: hypertension, stroke and hyperlipidemia.