83 / 154
A B S T R A C T S
S53
impact becomes less distinct. Treatment decision-making is
challenging in the older colorectal cancer patients but ostomy
placement should not be withheld based on age alone.
Disclosure of interest:
None declared
Keywords:
Impact, ostomy, survey
P030
HCC ASSOCIATED WITH HIGHEST INPATIENT
POST-OPERATIVE MORTALITY AMONGST GI CANCERS:
ELDERLY AGE HIGHEST PREDICTIVE RISK FACTOR
P. Somasundar
1,
*, G. Cholankeril
2
, A. Ahmad
1
, M. Hu
3
1
Surgery, Roger Williams Medical Center, Providence, USA,
2
Medicine, Roger Williams Medical Center, Providence, Bahamas,
3
Statistics, Brown University, Providence, USA
Introduction:
There is limited data regarding inpatient
mortality and associated risk factors in post-operative
gastrointestinal malignancies. Accordingly, we sought to
evaluate these outcomes in GI malignancies after specific
cancer-directed surgery.
Objectives:
To identify mortality associated with HCC and
age.
Methods:
Using the Healthcare Utilization Project-National
Inpatient Sample (HCUP-NIS) we evaluated age-specific
inpatient mortality and associated independent risk factors
after cancer-directed surgery in GI malignancies from 2004-
2013. GI malignancies were categorized as 1) colon cancer 2)
esophageal 3) gastric 4) hepatocellular carcinoma (HCC) 5)
pancreatic 6) anorectal cancer 7) other GI cancers. All patients
underwent site-specific cancer directed surgery during their
hospitalization. Patients with cancer-directed surgery at
more than one site were excluded. Multivariate analysis
analyzing risk factors associated with mortality included
demographic information (age, gender, ethnicity, insurance
type) and clinical comorbidities which were controlled for
severity of disease using the Elixhauser comorbidity index.
All multivariate logistic and linear regression was performed
using SAS statistical software 9.4.
Results:
Overall from 2004-2013, 276,771 patients under-
went cancer-directed surgery for GI malignancies. Although
HCC (n=1,535, 0.6%) constituted the smallest proportion of
hospitalizations, it was associated with highest inpatient
mortality (6.4%) followed by esophageal (5.4%) and gastric
cancer (4.4%) (Table). Post-operative HCC patients were
significantly younger than all other post-operative GI
malignancy patients (56.5 years versus 67.2 years, p < 0.001).
In our multivariate analysis, HCC was associated with the
highest likelihood for inpatient mortality (reference, colon
cancer; HCC OR, 1.6; 95% CI: 1.3-2.0 p = 0.004). In our sub-
analysis of post-operative HCC patients, we noted advancing
age to be a significant predictive risk factor for mortality
(reference, age < 65; 65-74, OR, 2.4; 75-84, OR, 6.3;
85, OR, 28.3;
p < 0.01).
Conclusion:
Although HCC represents a small proportion
of post-operative GI malignancy, HCC represents the highest
inpatient mortality particularly in the elderly. With the
incidence of HCC expected to rise rapidly in the elderly,
addressing this susceptible population should be prioritized.
Disclosure of interest:
None declared
Keywords:
Hepatocellular cancer
P031
RETROSPECTIVE CORELATION OF BIOPSY WITH PSA
SCREENING VALUES OF GERIATRIC INDIVIDUALS IN A
TERTIARY CARE TEACHING HOSPITAL IN INDIA
R. Unnikrishnan
1,
*, S. Senan
1
1
Geriatrics, Amrita Institute of Medical Sciences, Kochi, Kerala,
India
Introduction:
Prostate Specific Antigen (PSA) screening is a
common investigation employed across various parts of the
world to screen for prostatic malignancies. Various studies
done across a wide spectrum of different populations and
age groups have also revealed an age specific and population
specific variation in cut off of normal values of PSA - which
again adds on to the dilemma whether the current cut-off
values would result in false positive or false negative results.
The purpose behind this study stems from the fact that
multiple papers have suggested that age specific cutoff values
for PSA screening are better than the currently used standard
cutoff values of 4ng/ml and the fact that different races
have their own reference ranges. Similar studies have been
presented for Afro-American and East Asian populations,
but none whatsoever explxoring the same in the Indian
Subcontinent.
Objectives:
Primary Objective :To study P.S.A values of
elderly individuals and postulate the possible normal values
of PSA in different age groups of elderly. Secondary Objective
:To co-relate the PSA values with biopsy reports, if done, thus
hypothesizing the acceptable limits of PSA in the elderly male
in Kerala, beyond which prostatic biopsy maybe indicated.
Methods:
This is a retrospective, cross-sectional study.
Data was collected on the relevant variables from the hospital
records at Amrita Institute of Medical Sciences, Kochi of
patients registered between Januray 2011 and 2013. Total
study population included 1038 patients. Inclusion Criteria:
All male patients aged more than 60 years who are presenting
to the Geriatrics OPD for Comprehensive health check up.
Results:
Of the 1038 patients, only 105 had elevated PSA
and they were classified based on age group in the following
manner. 23 of these patients who had suspicious DRE findings
were subjected to prostatic biopsy, of which only 7 were
reported as malignant.
Table (abstract P031)
Age groups
Sample size
Mean
Standard deviation
60-69 yrs
610
1.55
1.84
70-79 yrs
312
1.40
1.33
80+ years
28
1.20
1.06
Total
950
1.49
1.67