Potential risk factors for adverse drug reactions in elderly patients : contribution to safer drug prescribing

Because of demographic aging the proportion of elderly persons in the population is
increasing, especially in industrialized countries. Increasing age is associated with a
higher prevalence of comorbidities possibly necessitating pharmacotherapy. Elderly
persons are not only treated with more drugs than younger ones, but they are also
more vulnerable to adverse drug reactions (ADRs). The aim of the thesis was to
elucidate potential risk factors that increase the risk for ADRs in the elderly with the
purpose to improve safety of medical treatment. First, the literature was reviewed in
order to get an overview on the potential risk factors already known. It has been
shown that not only physiological changes that affect pharmacokinetic and/or
pharmacodynamic effects of drugs, but also specific drugs and drug classes may
increase the risk for ADRs. Two studies were then performed to evaluate specific
aspects of drug prescribing, which may enhance the risk for ADRs.
In the first study age-specific differences in the prevalence of clinically relevant
potential drug-drug interactions (pDDIs) in ambulatory dyslipidemic patients treated
with a statin were evaluated. Practitioners from different parts of Switzerland
collected data for a total of 2’742 patients treated with a statin which attended their
practice. Medical treatment was screened for clinically relevant pDDIs, defined as a
DDI that could have had a potential serious outcome, using an interactive electronic
drug interaction program. The prevalence of clinically relevant pDDIs was
significantly higher in patients aged ³75 years than in patients aged £54 years
(18.4% versus 7.9%; p < 0.001). This was ascribed to a higher number of diseases
(3.5 versus 2.8; p < 0.001) and pharmacologically active substances prescribed (5.8
versus 3.8; p < 0.001). Beside polypharmacy, also heart failure and arrhythmia have
been identified as risk factors for pDDIs in elderly patients. The more frequent
prescription of cardiovascular drugs with a high potential for drug interactions (e.g.
amiodarone and digoxin) was mainly responsible for the observed increase in statin
and non statin pDDIs.
The aim of the second study was to retrospectively evaluate and compare the
prevalence of potentially inappropriate medication (PIM) use and prescription of
drugs with strong anticholinergic properties in 800 elderly patients hospitalized on
general medical or geriatric wards throughout hospital stay. PIMs as defined by the
Beers criteria and anticholinergic drugs have been associated with a higher risk for
ADRs in patients aged ³65 years. At hospital discharge, geriatric patients had a lower
prevalence of use of PIMs that should generally be avoided than at admission
(15.9% versus 22.1%; p < 0.05), whereas no difference was observed in medical
patients. Overall, the three most prevalent inappropriate drugs/drug classes were
amiodarone, long-acting benzodiazepines and anticholinergic antispasmodics. On
the other hand, geriatric patients were discharged with a higher prevalence of use of
PIMs that should be avoided in the presence of specific underlying diseases
compared to medical patients (23.7% versus 11.7%; p < 0.001). The main reason
was the higher prescription rate of benzodiazepines to patients with a history of falls
and syncope. There was neither a difference in the prevalence of patients with
anticholinergic drugs at admission nor at discharge between medical and geriatric
patients. Compared with internists, geriatricians appeared to be more aware of PIMs
that should generally be avoided. However, the results of this study should be
interpreted with caution, because some of the drugs identified as potentially
inappropriate may in fact be beneficial when the patient’s individual clinical condition
is taken into consideration.
Finally, a patient with lithium intoxication as a result of a drug-drug interaction (DDI)
with rofecoxib is presented. This 68-year-old woman had several risk factors that
finally resulted in the clinical manifestation of the DDI, illustrating well the problems of
pharmacotherapy in the elderly. The already impaired renal function (calculated
creatinine clearance 40 mL/min) deteriorated after the addition of rofecoxib, a
selective cyclooxygenase 2 (COX-2) inhibitor. As a consequence, renal clearance of
lithium was impaired, leading to an accumulation of the drug and symptoms of lithium
intoxication such as vomiting, hypokinesia and tremor. Selective COX-2 inhibitors
seem therefore not to be safer than conventional nonsteroidal anti-inflammatory
drugs concerning their effect on renal function, especially in patients with renal
insufficiency.
Depending on the underlying disease, medical treatment with drugs associated with
a high potential for DDIs and/or ADRs may not always be avoided. Knowledge of the
potential risk can help to take appropriate measures to lower the probability for an
adverse outcome, e.g. close monitoring of the patient, dose adjustment or selection
of an alternative drug.