European Geriatric Medicine - Supplement 1 - Volume 7 - October 2016

38 / 290

drainage (750 cc) which revealed a lung adenocarcinoma. The
gastrointestinal symptomatology improved, but nausea and vomi-
ting worsened refractory to therapy. He developed ipsilateral
Horner’ s syndrome and contralateral facial paresis simultaneously.
Computerized tomography showed a large pulmonary mass in RSL
with metastatic ipsilateral and contralateral lung nodes, supra and
infra-diaphragmatic metastatic lymph nodes, and bilateral adrenal
metastasis. It was already identified 2 secondary bilateral cerebral
lesions at the level of the protuberance and a right cerebellum injury.
The biopsy taken by bronchoscopy confirmed the diagnosis. The
neurological symptoms worsened (left hemiparesis).
The patient died 15 days after diagnosis, about 1.5 months after the
onset of symptoms.
P-012
At what blood urea level should IV fluid replacement start in older
patients?
O. David1. 1RBCH, NHS England, UK
Introduction: The provision of intravenous (IV) fluid replacement is
considered a fundamental aspect of good medical care. The UKs NICE
fluid replacement guideline suggests it should be guided by renal
function, but does not specify when to intervene, and to what specific
values. Since blood Urea (BUN) is often more responsive to change
than creatine it is considered a better acute marker, but what evidence
links it to intervention in the older patient?
Method: A Medline literature search was explored using several
terms for renal function (RF), intravenous fluid replacement (IV) and
comorbidities dementia and heart failure (CM). Without limitations
applied to the search groups (RF and IV), 1998 results were returned
and analysed, but none returned for the older age group. No search
results were returned for RF and IV and CM.
Results: In older patients there is no prospective evidence to recom-
mend when to intervene to any given blood Urea level. Since the
NICE (CG174) guidance on Intravenous fluid, recommends the
reporting of “fluid mismanagement” as a critical incident, this will
be a subjective and controversial area. The expert opinion of
geriatricians for this patient group should remain key.
Conclusion: No prospective research evidence states when to start
intravenous fluid replacement to a set blood Urea level in older
patients. This questions how frequently blood Urea levels should be
requested and how they should be considered. Renal function
interpretation in patients with heart failure and dementia are both
common and problematic to assess, treat and evidently study.
P-013
Prediction of Glomerular Filtration Rate (GFR) in geriatric patients:
which formula to use?
J. Dekoninck1, N. Van Den Noortgate. 1University of Ghent, Belgium
Objectives: To compare four fomulas estimating GFR (eGFR), CKD-EPI,
BIS, Cockcroft&Gault(CG), and MDRD, with a Cr-EDTA clearance as
measured GFR (mGFR).
Methods: Eighty-nine patients (83,5 ± 6,8 y; 25 men) are included
in the study. eGFR’ s are compared to mGFR in the total population as
well in a subpopulation with mGFR <60 mL/min and ≥60 mL/min.
Statistical analysis was perfomed using Spss version22. Bias is
measured as mean difference of eGFR minus mGFR, precision as
standard deviation of bias. The percentage of correct estimates within
30% of the mGFR is used as accuracy.
Results: CG and BIS have the least bias (−3,0 resp −0,26). CG has
the highest precision (16,9) and accuracy (68,5%), followed by BIS
(respectively 20,9 and 66,3%). MDRD is the formula showing the
highest bias (15,4) and the lowest precision/accuracy (23,9/40,4%). In
the older patient with mGFR <60 mL/min, the CG has an even higher
precision/accuracy (9,9/75,4%). The BIS has the second best precision/
accuracy (11,4/64,2%) but overestimates more in mGFR <60 mL/min
(bias 7,3). MDRD remains the least adequate eGFR formula. None of
formulas has an acceptable estimating power for mGFR ≥60 mL/min.
Results are less consistent with the lowest bias, highest precision
and accuracy for respectively MDRD, CG and CKD-EPI. CKD-EPI tends
to perform best taking into account a considerable underestimation
(bias-13,5).
Conclusion: Cockcroft&Gault formula has the lowest bias and highest
precision/accuracy, especially in an older population with mGFR
<60 mL/min. The newer formulas, BIS and CKD-EPI tends to be the
second best choice in respectively the lower and higher mGFR’s.
P-014
A frailty score for the acute hospital setting: identifying vulnerable
patients at point of entry to care
L. Dinesen1,2, A.J. Poots1, F. Ciardi 1, K. Warburton2, J. Soong1,
F.C. Martin3, D. Bell1,2. 1NIHR CLAHRC Northwest London, Imperial
College London, 2Dept. Acute Medicine, Chelsea and Westminster
Hospital, 3King’ s Health Partners, London, UK
Introduction:Our objective is to operationalize a clinically usable tool,
frailty early warning score (FEWS; score range 0–15), which will
identify frailty (>1) and help predict significant outcomes.
Methods: Setting: Acute admission unit, urban university hospital
between 03.06.2015 and 27.08.2015. Sample: 700 admitted patients
>65 y. Data collected from clinical notes taken routinely as part of
the emergency admission process. The national early warning scores
(NEWS, score range 0– 15), which reflects medical acuity, were
simultaneously collected.
Results: Patients had average age of 81 years, and this was stable
regardless of the frailty score. 52.6% were female. Mortality increases
with increasing frailty score, in-hospital mortality is 2.6% for FEWS 0,
whilst 11.9% for FEWS 8. A similar trend is seen with readmissions
rates and length of stay (LOS): 30 day readmission is 21.6% for FEWS 0
and 28.6% for FEWS 8; mean LOS is 7.5 days for FEWS 0 and 12.4 days
for FEWS 8. NEWS and FEWS were cross-tabulated: 246 (35%) patients
or participants scored <3 NEWS (would not trigger escalation), of
which 206 (81%) scored FEWS ≥ 1.
Conclusion: This study describes FEWS as a novel way of predicting a
frail individual’s outcomes. This score can be easily calculated at the
point of care using routinely corrected data. It is fast and simple to use;
it will not require additional clinical assessment. The threshold for
escalation due to frailty risk is as yet undetermined.
P-015
Age over 80 years old in acute pulmonary embolism – does it
influence Manchester triage and time until diagnosis?
K. Domingues, R.S. Almeida, M.J. Vieira, B. Santos, C. Costa, N. Craveiro,
I. Monteiro, M. Leal. Cardiology Department, Internal Medicine
Deparment
Introduction: Early diagnosis of acute pulmonary embolism (PE) is
a challenge in Emergency Departments (ED), because of its diverse
clinical presentation. The Manchester triage system (MTS) depends
on the initial presentation, assigning higher priority to urgent
cases. We aimed to assess whether age over 80 years old (>80 yo)
influences presentation, MTS and time from admission to diagnosis
(TAD) in PE.
Methods: Retrospective study of patients admitted to our ED between
January 2011 and December 2015, with the final diagnosis of PE.
Demographic, clinical and triage data were analyzed.
Results: We included 367 patients, 60.8% women, with a mean age of
71.6 ± 16.4 years. The three most common main complaints were
“indisposition” (38.7%),“dyspnea” (30.8%) and “thoracic pain” (20.4%).
Most patients were triaged as orange (51.5%), followed by yellow
(47.1%), red (0.8%) and green (0.5%). The group of patients >80 yo
(39.8%) had more women (68.5% vs 31.5%; p = 0.014). There was a
significantly higher number of patients with “indisposition” as main
complaint (49.3% vs 31.7%; p = 0,001). Thoracic pain was less frequent
in this group (11.0% vs 26.7%; p < 0.0001). Triage color was not
significantly different between groups. TAD by computed tomography
pulmonary angiogram was marginally correlated with age (r = 0,095)
Poster presentations / European Geriatric Medicine 7S1 (2016) S29–S259 S32