Clinical Pharmacology - Wiley Online Library

54

CLINICAL PHARMACOLOGY

172 ANALYSIS OF THE USE OF MEDICINE DURING PREGNANCY IN KYRGYZSTAN Moldoisaeva S., Jzolosheva T., Zurdinov A. Background: One of the least studied issues of clinical pharmacology is the use of medicines during pregnancy. This is due to the impossibility of carrying out large-scale randomized trials in pregnant women because of bioethical reasons. Studies have demonstrated high prevalence of medicine use during pregnancy. A recent systematic review, which evaluated medicine use prescribed during pregnancy in developed countries, showed that 27–93% of the pregnant women received at least one prescription of medicine, excluding vitamins and minerals. Methods: The study has been performed as a multicenter retrospective pharmacoepidemiological research, which has been conducted from 2007 to 2011 in eight maternity hospitals in 5 cities of Kyrgyzstan. 1916 case histories have been analyzed. Inclusion criteria were: gestational age ≤38 weeks, the number of bed-days ≥3 days. Data on applied medicines were recorded in specially designed cards, which then analyzed. Drug safety categories have been determined in accordance with the classification of FDA (USA). Result: Hundred perdcentage pregnant women received prescription of medicine. At least one medicine was prescribed in 3% of cases, 97% – few medicines: 2–5 drug administered in 63.3%, 10–6 in 32.8%, more than 10 medicines in 0.9% of cases. Average number of prescribed drugs was 4.9. 196 different name of medicines have been prescribed. According to the FDA classification, 2.9% of 9350 analyzed all medicine prescriptions – category A, 18.9% belonged to category B; 28.1% – category C, 2.3% – in the category D; 1.9% – the category X, 45.8% of all prescriptions were unspecified safety and have not been included to the classification of the FDA. Conclusions: From the viewpoint of safety the use of medicines during pregnancy is a major concern. Medicines with not enough data on efficacy and safety, with potential risks for pregnant women have been prescribed during pregnancy.

173 ELECTROLYTE EFFECTS DURING INITIATION OF ANTIHYPERTENSIVE THERAPY WITH AMLODIPINE OR HYDROCHLOROTHIAZIDE IN DIABETIC NIGERIANS Iyalomhe G., Omogbai E., Iyalomhe O. There is paucity of information regarding the effects of amlodipine or hydrochlorothiazide therapy on serum and urine electrolyte profiles in hypertensive Nigerians with type 2 diabetes mellitus. Therefore, to evaluate whether amlodipine or hydrochlorothiazide would be preferable to initiate treatment, we randomized 40 newly diagnosed hypertensive subjects with controlled type 2 diabetes mellitus aged 43–68 years to amlodipine and hydrochlorothiazide treatment groups of 20 patients each (10 males, 10 females), and they were treated respectively, with amlodipine 10 mg and hydrochlorothiazide 25 mg, both drugs being given once daily for 48 weeks. Body mass index, blood pressure, 24 h urine volume, serum and urine electrolytes were assessed at baseline and at the end of weeks 1, 3, 6, 12, 24, 36 and 48. The two drugs significantly reduced blood pressure, though the effect of amlodipine was significantly greater compared with that of hydrochlorothiazide (P < 0.01). Diuresis was significant in hydrochlorothiazide group (P < 0.01). Observed male/female serum Na+ loss was 9.18 2.32/ 10.90 2.50 and 13.30 1.34/15.10 1.77 mM for amlodipine and

hydrochlorothiazide subgroups, respectively. There was a parallel significant (P < 0.05) natriuresis. Significant (P < 0.05) hypokalemia occurred in hydrochlorothiazide subgroups and overall male/female serum K+ loss was 0.10/0.08 and 0.16/0.24 mM for amlodipine and hydrochlorothiazide, respectively. However, there was no significant parallel kaliuresis. Significant (P < 0.05) disproportionate hypochloremia occurred in all subgroups, so also was the parallel chloriuria. By providing effective blood pressure control and beneficial biochemical effects, amlodipine therapy appears suitable for treatment of hypertension in these diabetic patients. Similarly, low dose hydrochlorothiazide therapy, which seems to have more marked effects in females, appears to have moderate biochemical complications in these patients and is, therefore, a logical alternative to substitute for or add to amlodipine therapy.

174 A RANDOMIZED TRIAL COMPARING MORPHINE AND IBUPROFEN FOR POST-TONSILLECTOMY PAIN MANAGEMENT IN CHILDREN WITH SLEEP DISORDERED BREATHING Kelly L., Sommer D., Ramakrishna J., Hoffbauer S., Arbabtafti S., Maclean J., Reid D., Koren G. Background: In children, sleep disordered breathing is often caused by hypertrophy of the tonsils and/or adenoids and is commonly managed by tonsillectomy with or without adenoidectomy. There is controversy regarding which postsurgical analgesic agents are safe and efficacious. Methods: This prospective randomized clinical trial recruited children with sleep disordered breathing, who were scheduled for tonsillectomy adenoid removal. Parents were provided with a pulse oximeter to measure oxygen saturation and apnea events the night before, and the night after surgery. Children were randomized to receive 0.2– 0.5 mg/kg oral morphine or 10 mg/kg of oral ibuprofen. The Objective Pain Scale (OPS) and visual analog scale were used to assess analgesic effectiveness on postoperative day 1 and day 5. The primary end point was changes in oxygen saturation during sleep after surgery, as compared to pre-operatively. Results: A total of 91 children aged 1–10 years were randomized. Children receiving ibuprofen showed an improvement in the rate of desaturation events on post-operative day one, whereas children receiving morphine had a significant increase in their rate of desaturation (1.79 7.57 vs. +11.17 15.02, P < 0.01). There were no differences seen in analgesic effectiveness, tonsillar bleeding, or adverse drug reactions. Conclusion: Ibuprofen provides safe and effective analgesia in children undergoing tonsillectomy for sleep disordered breathing. Post-tonsillectomy morphine use should be limited, as it may be unsafe in certain children.

175 A SURVEY ON THE KNOWLEDGE AND ATTITUDE TOWARDS PHARMACOVIGILANCE AMONG MEDICAL STUDENTS IN INDIA Arora E., Katyal J., Gupta Y. Background: Under-reporting is a major limitation of spontaneous reporting systems for suspected adverse drug reactions (ADRs). The

© 2014 The Authors Basic & Clinical Pharmacology & Toxicology © 2014 Nordic Pharmacological Society. Basic & Clinical Pharmacology & Toxicology, 115 (Suppl. 1), 1–374

55 present study was undertaken to evaluate the knowledge and attitude regarding ADR reporting and pharmacovigilance among medical students from across India in order to identify issues and derive at possible measures to improve reporting. Methods: PULSE is the biggest inter-college academic festival of the medical fraternity with participation from all over India. A cross-sectional, questionnaire based study was conducted during PULSE. The survey questionnaire consisted of closed- ended multiple response items. The students were required to mark the most appropriate option. The areas covered included subject knowledge, attitude and awareness towards pharmacovigilance. Results: Out of 350 distributed questionnaires, 253 qualified for analysis – 230 medical, 16 dental and 7 nursing students of 71 medical colleges from 17 states across India. 60% of participants knew the term ‘Pharmacovigilance’, scope; activities involved, who and where to report an ADR but were unaware that ‘Periodic Safety Update Report’ (PSURs) is a mandatory pharmacovigilance activity by the industry. 112 (48.07%) thought thalidomide was a banned drug. While all had taken medicines in the last 2 years, only 19% read the package insert. Majority (93%) felt reporting is a useful practice and causes for underreporting are lack of awareness followed by attitude, misconceptions, fear of litigation and interestingly the least important lack of time. However, most were reluctant to have reporting as mandatory, rather it should be voluntary and preferably with incentive in the form of professional/social recognition and not monetary. For ease of reporting, a toll free number or SMS service was preferred. An ADR monitoring center for every community health center in the country was recommended. Conclusion: Teaching of pharmacovigilance in undergraduate curriculum has led to better theoretical knowledge but awareness needs to be improved further. This can be addressed by educational interventions like widening curriculum and including pharmacovigilance related activities like PSURs. There is a need for conducting regular training programs in the form of workshops and CMEs. Webinars can be a useful concept in this field.

176 ABSOLUTE VS. RELATIVE DRUG -THERAPY BENEFITS AND SAFETY: ARISTOTLE’S LEGACY Kumana C., Cheung B., Tse H., Lauder I. Background: Aristotle1 was a double blind, double dummy, multicentre, randomised cliinical trial in patients with atrial fibrillation and one other risk factor for stroke, which compared oral anticoagulation with warfarin (targeted international normalized ratio [INR] 2.0–3.0) vs. the new agent apixaban (5 mg 9 2 daily). Whilst apixaban conferred signifantly greater efficacy (stroke/embolic event prevention) and was safer (less frequent bleeding), consideration of outcomes in relative terms can be uninformative. To appreciate the overall cost-effectiveness of such treatment, absolute benefits expressed in a meaning full way should be considered. Methods: As previously described2, we therefore derived unadjusted estimates of relative risk reduction (RRR) and number needed to treat (NNT)/year values together with corresponding 95% confidence intervals (CIs) for prophylaxis with apixaban compared to warfarin for principal efficacy and safety endpoints published in ARISTOTLE. Results: See below Patient Numbers (follow-up) Pre-specified Endpoints % RRR (95% CI) NNT/year (95% CI) Apixaban – 9120 Stroke/systemic embolism 20 (4 to 34) 303 (169 to 1501) Warfarin – 9003 Haemorrhagic stroke 49 (25 to 65) 428 (273 to 987) (Median follow-up 1.8 years) Death from any cause 10 (0 to 20) 238 (119 to infinity) Discussion and Conclusions: Whilst the favourable RRR values pertaining to apixaban for stroke/systemic embolism, haemorrhagic stroke, and death seem striking, in absolute terms (NNT/year) the benefits appear less impressive. Though treatment with the newer oral anticoag-

ulant will inevitably be much more costly than using warfarin, the long-term impact of embolic events and major bleeds are financially and emotionally devastating. Thus, when allocating resources for thromboembolic prophylaxis in patients with atrial fibrillation, the meagre absolute numbers for additional events prevented with apixaban, must be weighed against the consequential long-term accrued benefits as well as those from avoidance of INR monitoring. The NNT/year (absolute benefit) for all deaths appeared more favourable than the values for the other endpoints shown. References: 1. Granger et al., 2011. NEJM 365:981–92; 2. Kumana et al., 1999. JAMA 282:1899–1901.

177 ADMISSIONS DUE TO DRUG RELATED PROBLEMS AT THE EMERGENCY DEPARTMENT OF A TERTIARY HOSPITAL IN ALKHOBAR, SAUDI ARABIA Randhawa M., Al-Ghamdi M., Whaas M., Aljamaan M., Al-Jaizani R. Background: Drug related problems (DRPs) are a frequent cause of admission in hospitals and can occur because of Medication Errors (MEs). MEs are either predictable and, therefore, often possible to avoid (e.g. unnecessary drug therapy, in complete drug therapy, ineffective drug, dosage too low, dosage too high and non-compliance); or unpredictable and thereby difficult to foresee (e.g. hypersensitivity and idiosyncratic reactions). Therefore, not all DRPs are the result of an error, as an unpredictable DRP can occur without an error. An insight into frequency, type and severity of DRPs could help reduce their occurrence. The present study aimed to estimate the prevalence of admissions due to DRPs at Emergency Department (ED) of King Fahd Hospital of the University (KFHU), Alkhobar, Saudi Arabia. Methods: Files of suspected cases of DRPs reporting to ED of KFHU, a tertiary hospital, in the year 2012 were scrutinized. The suspicion was made from the hospital record system based on Diagnosis Code Numbers (ICD-9-CM, Professional 2010). The detection was also achieved from the triggers pointing to DRPs, like signs and symptoms related to DRPs, drugs commonly causing DRPs and relevant laboratory tests. The data from the suspected files was entered in the prescribed questionnaire and analyzed by a committee comprising of a physician, a clinical pharmacologist and a clinical pharmacist. Results: Out of 5574 admissions 253 (4.5%) were DRPs and were categorized as: Overdose toxicity and side effects of drugs 50 (19.8%), drug-interactions 29 (11.5%), accidental and suicidal drug ingestions 26 (10.3%), drug abuse 18 (7.1%), drug allergy 10 (4%), super-infections 8 (3.2%) and non-compliance to treatment 112 (44.3%). Over 60% of DRPs were preventable; 67 (26.5%) required admission in hospital for 7–102 days and 10 (4%) died. Conclusions: Non-compliance to treatment, over dose toxicity, drug interactions and drug abuse are important causes of admissions in hospitals due to DRPs. Awareness of problem and education of prescribers would help to prevent them and improve patient care.

178 ADRS LEADING TO HOSPITALISATION CAUSED BY SELFMEDICATION – RESULTS FROM A GERMAN LONG-TERM, OBSERVATIONAL STUDY Schmiedl S., Rottenkolber M., Hasford J., Rottenkolber D., Farker K., Drewelow B., Hippius M., Salje K., Thuermann P. Background: Self-medication due to over-the-counter (OTC) drugs is common but there are only a few data regarding OTC-related adverse drug reactions (ADRs). For a second type of self-medication, i.e. intake of formerly prescribed drugs without a current recommendation by healthcare professionals, data are scarce. Hence, we aim to analyse ADRs caused by both types of self-medication using data from the German Network of Regional Pharmacovigilance Centres.


56 Methods: In this multi-centre study we included all patients admitted to an Internal Medicine department due to an ADR (hospital catchment area: approximately 500,000 people). Patients were identified by systematic screening of all hospital admissions using ADR-related trigger criteria and spontaneous physicians’ reports. Assessment of prescribing status of medication (OTC, formerly prescribed drugs or questionable) was made by patient’s interviews and chart reviews. All relevant data were documented comprehensively and ADR causality was assessed using Begaud’s algorithm. In this analysis, we included 6,887 patients with an at least ‘possible’ ADR occurring between 01/2000 and 12/ 2008. Results: Self-medication was causative in 266 patients (3.9% of all ADR patients). Regarding the type of self-medication, OTC-compounds could be identified in 132 patients and formerly prescribed drugs taken without current professional advice in 111 patients. In the remaining patients, more than one type of self-medication or questionable medication prescription status was present. The highest number of self medication-related ADRs occurred in elderly patients (women: 70–79 years, men: 60–69 years). Non-steroidal analgesics caused the majority of self-medication related ADRs (mainly acetylsalicylic acid [ASA] taken as OTC). In 68% of patients with a self-medication related ADR caused by analgesics, treatment exceeded a duration of 4 days. A relevant drug-drug interaction with a prescribed drug was present in 102 patients (mainly ASA taken as OTC and prescribed diclofenac). Conclusions: In general, self-medication accounts only for a small proportion of ADRs leading to hospitalisation to Internal Medicine departments under the relatively strict prescribing regulations in Germany. Nevertheless, self-medication contributes particularly to ADRs in elderly patients with polypharmacy, and prevention strategies should focus on this patient group.

179 ADVERSE DRUG EVENTS CAUSED BY POTENTIALLY INAPPROPRIATE MEDICATION IN HOME-BOUND ELDERLY PATIENTS BY NATION-WIDE SURVEY Imai H., Onda M., Nanami Y. Objectives: The objective of this study was to conduct a large-scale survey to shed light on the prevalence of ADEs caused by PIMs in home-bound elderly patients and to identify drugs that cause such ADEs, factors associated with ADEs, measures taken by pharmacists to address ADEs, and the effects of such measures. Methods: A questionnaire was mailed to 3,321 health insurance pharmacies nationwide to ask the pharmacists at those pharmacies who provide home-visit services about the details of their practice with as many as five of the patients receiving home-visit services (survey period: January 23 to February 13, 2013). We conducted a retrospective cross-sectional study in 3321 Japanese community pharmacies by collecting data from the comprehensive assessment forms for 4843 patients aged 65 years and over who were assessed January and February 2013. Potentially inappropriate medications were identified on the basis of the 2003 Beers criteria. Results: One or more medications considered to be PIMs were prescribed to 48.4% of patients (2,053), and pharmacists discovered PIMinduced ADEs in 8.0% of patients (165) during home visits. The top ADE-inducing medications were highly anticholinergic antihistamines, benzodiazepines, sulpiride, and digoxin. ADEs associated with benzodiazepines in particular consisted primarily of frequent lightheadedness, somnolence, and sleepiness, suggesting an increased risk for falls or fractures in elderly patients. The following factors associated with ADEs were identified: patient gender, awareness of prescription issues by pharmacists (contraindications, therapeutic duplication, and interactions, and unnecessarily continued treatment), frequency of visits and time spent at work, and the frequency with which pharmacists checked for adverse reactions in detail. Conclusion: The prevalence of prescriptions for PIMs (48.4%) associated with home healthcare in Japan was nearly twice as high as in the

results of previous studies. The incidence of PIM-induced ADEs was 8.0%, and most were caused by antipsychotics.

180 ADVERSE DRUG REACTIONS IN PATIENTS WITH HIV OR TUBERCULOSIS REPORTED TO THE MEDICINES INFORMATION CENTRE IN CAPE TOWN, SOUTH AFRICA Njuguna C., Swart A., Chisholm B., Jones J., Mouton H., Stewart A., Blockman M., Maartens G., Cohen K. Background: The Medicines Information Centre (MIC) at the University of Cape Town provides advice to health care workers (HCWs) regarding management of suspected adverse drug reactions (ADRs). We aimed to describe ADRs reported to the MIC by HCWs, concordance with advice and patient outcomes. Methods: We analysed the first 200 queries from HCWs about suspected ADRs in patients with HIV and/or tuberculosis (TB) received from May 2013 onwards. We performed causality assessment using World Health Organization (WHO) criteria. A suspected ADR rated as certain, probable or possible by those criteria was classified as an ADR. Patient outcomes were collected telephonically after the next patient visit. Results: Median age of patients was 33 (IQR 28–42) years, 114/200 (57%) were females, 49/200 (25%) were on tuberculosis treatment and 196/200 (98%) were HIV-infected, 192/196 (98%) of whom were on antiretrovirals. The 200 queries yielded 202 suspected ADRs. We classified 174/202 (86%) suspected ADRs, from 172 patients, as ADRs, with 7/174 (4%) certain, 55/174 (32%) probable and 112/174 (64%) possible. Most frequent ADRs were rash in 44/174 (25%), hepatitis in 21/174 (12%), and renal impairment in 21/174 (12%). The ADR was graded as severe in 20% of rashes, 86% of hepatitis and 27% of renal impairment cases. The most frequently implicated drugs were efavirenz, TB treatment and tenofovir, for rashes, hepatitis and renal impairment respectively. Patients were outpatients in 126/172 (73%) of ADR queries, inpatients in 28/172 (16%) of queries and hospitalisation status unknown in 18/172 (11%). In 103 cases with follow-up data, HCWs followed advice given by the MIC in 90/103 (87%) of cases, did not follow advice in 7/103 (7%) and response to advice was unknown in 6/103 (6%). In patients with follow-up data, 72/103 (70%) improved, 12/103 (12%) were unchanged, 2/103 (2%) deteriorated and 17/103 (17%) had unknown outcomes. Conclusion: HCWs frequently sought advice from the MIC on management of rashes, hepatitis and renal impairment. Management of these ADRs should be prioritised for inclusion in HCW training. In cases with follow-up information, concordance with advice was good, and HCWs reported clinical improvement in the majority.

181 ADVERSE REACTIONS IN CRIMEA, UKRAINE. 2013 ANNUAL REPORT Matvieiev O., Ezernitsky O., Koniaieva O. Background: Analysis of adverse reactions (ADR) of medicinal products (MP) is important activity of regulatory agency necessary for permanent improve of patients` safety. The responsible authority for pharmacovigilance in Ukraine is State Expert Center of Ministry of Health which has regional offices in every administrative region which ensure and support registration of ADRS locally. Methods: The only method actively used in Ukraine for the ADR registering is spontaneous report analysis. In 2013 the Regional Office in Crimea region has received 1129 spontaneous reports from 89 hospitals. The Crimea population is 1.9 million. Data from all reports were recorded in local database ARCADe handled by FileMaker software. This data is corresponding to results of previous years (1166 in 2011; 1135 in 2012).


57 Results: Most ADRs were registered in female patients (60.8%) but subgroups analysis revealed predominance of boys in childhood (53.6%). A curve of distribution of reports accordingly to patient`s age had two peaks: in first year children and in adults 46–60 years old (8% and 22.2%). The most ADRs were not serious, but 2 (0.2%) led to death of patient (Metamizole, Ceftriaxone), 112 (9.9%) were lifethreating, 60 and 54 (5.3% and 4.8%) resulted in hospitalization or its prolongation, 54 and 6 (4.8% and 0.5%) caused loss of ability to work (temporary and permanent). Causality assessment revealed that in 482 cases it was ‘probable’, in 455 – ‘certain’, in 167 – ‘possible’, in 16 – ‘doubtful’. Complicated allergy anamnesis was found in 118 (10.5%). In comparison with previous years the frequency of polypharmacia cases significantly decreased (P < 0.05) mainly due to decrease of cases with 4 and more accompanying MPs. In 22% of patients product was prescribed to treat respiratory nosology, in 17% – cardiovascular pathology and in 9.7% – infection. Most frequent clinical presentation of ADR was urticaria. 68% of ADRs required medicinal correction. As earlier, the ‘leading’ group of MP was antibiotics (39.2%), inside which cefalosporins (42%) caused the most ADRs. Conclusions: The knowledge of ADR structure allows pharmacovigilance specialists to turn efforts to concrete groups of MPs, nosologies and patients` groups.

182 AN ECONOMIC ANALYSIS OF THE INADEQUATE PRESCRIPTION OF ANTIULCER MEDICATIONS FOR INHOSPITAL PATIENTS IN A THIRD LEVEL INSTITUTION IN COLOMBIA Machado J. Introduction: The prescription and costs of antiulcer medications for in-hospital level have increased during the last years with reported inadequate use. Aim: To determine the patterns of prescription-indication and also the economic analysis of over cost caused by the non-justified use of antiulcer medications in a third level hospital in Colombia. Methods: Cross-sectional study, of prescription-indication of antiulcer medications for patients hospitalized in ‘Hospital Universitario San Jorge’ of Pereira during July of 2012. Adequate or inadequate prescription of the first antiulcer medication prescribed was determined as well as for the one prescribed during the hospital stay, supported by the clinical practice guidelines from the Zaragoza I sector workgroup, the clinical guidelines from the Australian Health Department, and finally the American College of Gastroenterology Criteria for stress ulcers prophylaxis. Daily Defined Dose per bed/day was used, the cost for 100 beds/day, and the cost of each bed/drug. A multivariate analysis was carried out using SPSS 21.0 Results: Seven hundred and seventy-eight patients were analyzed, 435 men (55.9%) and 343 women, mean age 56.6 20.1. The number of patients without justification for the prescription of the first antiulcer medication was 377 (48.5%), and during the whole in-hospital time it was 336 (43.2%). Ranitidine was the most used medication in 438 patients (56.3%). The cost/month for not indicated antiulcer medications was €3.335.6. The annual estimated cost for the inadequate prescription of antiulcer medications was €16.770.0 for 100 beds. Conclusion: A lower inadequate prescription of antiulcer medications was identified compared with other studies, however it still keeps being high and worrying because of the important costs that these inadequate prescriptions generates to the institution.

183 AN IMPROVED METHODOLOGY, BASED ON PHARMACOKINETIC, PHARMACODYNAMIC AND PHARMACOGENETIC EVALUATIONS, FOR BIOEQUIVALENCE ASSESSMENT OF SELECTIVE SEROTONIN REUPTAKE INHIBITORS (SSRIS): STUDY DESIGN ON HEALTHY VOLUNTEERS Gori G., Ferrannini E., Blandizzi C., Del Tacca M. Background: SSRIs represent one of the most prescribed pharmacological classes in USA and EU and, due to patent expiration, several pharmaceutical companies are marketing their generic formulations. Physicians must pay attention when prescribing generic SSRIs owing to their potential lower efficacy and/or higher toxicity, such as haemorrhagic and pro-arrhythmic risks, as compared to originators. Although the disclosure of pre-authorization data is important for prescribers, several countries don’t supply pre-marketing documentation of generic drugs. Moreover, post-marketing studies are raising doubts on actual therapeutic interchangeability between branded and generic drugs available in the pharmaceutical market. In addition, chronic treatments with different generic formulations might be harmful (biocreep phenomenon), particularly for antidepressants. Accordingly, we propose a more accurate evaluation of therapeutic equivalence between branded and generic escitalopram formulations through a more comprehensive methodology, which, besides pharmacokinetic testing, compares some markers of SSRI pharmacodynamics (e.g. platelet aggregation and heart repolarization), and comprises a pharmacogenetic assessment of polymorphic genes involved in escitalopram metabolism (CYP2C19 *2/*3) or activity (5-HTTLPR S/L). Methods: A no-profit, monocentric, randomized, 20-mg single-dose, open label, three-period, cross-over study has been designed to test three escitalopram formulations on 30 healthy volunteers, according to three sequences (R-T1-T2, T1-T2-R, T2-R-T1), where R is the reference formulation, while T1 and T2 are two generic products randomly selected from the EU market. The primary objective is to compare the clinical pharmacokinetics of escitalopram formulations in terms of Cmax and AUC0-∞. These data will be correlated with basal and Tmax values of QTc and platelet aggregation as well as with CYP2C19 and 5-HTTLPR genotypes. The study has been approved by the local Ethics Committee and is going to start. Cmax and AUC0-∞ will be compared using geometric mean ratios. The paired-sample T test will be applied to evaluate the significance of differences between basal and post-treatment values of QTc and platelet aggregation. Results and Conclusions: The proposed methodology is intended to implement more careful investigations on the equivalence of generic SSRI formulations, by a combination of pharmacodynamic and pharmacogenetic endpoints with pharmacokinetics, in order to improve the quality of bioequivalence studies thus achieving better inference of therapeutic equivalence.

184 AN INTER-PROFESSIONAL LEARNING MODULE IN CLINICAL PHARMACOLOGY Athuraliya N., Newby D., Schneider J. Introduction: The study developed and assessed an inter-professional learning (IPL) module centred on a patient in transition from hospital to the home focusing on a collaborative home-based medication review (HMR) by a pair (medical and pharmacy) of students. Outcomes assessed : (i) Changes in student confidence in (a) conducting a medication review (b) communication with the General Practitioner (c) working with another professional student to achieve same goal (d) identify barriers to medication management during transition. (ii) Acceptability of the module among students. Methods: Participants: Students volunteers from the final year medical and pharmacy professional courses; patient volunteers requiring a HMR at hospital discharge. The IPL Module consisted of : an online medication review, a face to face resource session on communication and team work, allocation of


58 each student pair a patient requiring a HMR and the discharge summary, a HMR session, a resource session with a supervisor, writing a report, assessment of student learning &module evaluation with prepost test and focus group discussions. Results: Sixteen students volunteered completed the online and face -face module. Seven did not complete. Due to difficulties in coordination with another professional course student, patient and/or ongoing academic programme. Results of pre-post tests are attached. Focus group discussions: Positive aspects of the module: Experience in authentic learning: ‘Its more hands on..handle situation as it comes rather than a simulator..take seriously’ ‘good opportunity to apply what you have learnt than theorising. . .actually responding to a real person’. Learning through feedback: ‘practising using the language knowing what is relevant..to be communicated to the doctor and having a feedback was really good.’ ‘Good to have feedback. . .being encouraged to push the point. . ..’ Opportunity of inter-professional learning: ‘getting a taste of what its like working as a multidisciplinary team’. ‘They have strength where I have weaknesses, the reverse is true. . .can combine our strength effectively’. Areas for improvement: Difficulties in coordination with another course student and ongoing academic programme. Conclusions: The module had a positive impact on student confidence in their professional role as medication reviewers. Successful implementation requires a fixed time allocation at the Faculty level. Fig. 1. Pre-post questionnaire test : overall data. Pre-Post Test :raƟng from 1-4

4.00 3.50 3.00 2.50

Pre-test

2.00

post-test 1.50 1.00 0.50 0.00 q1

q2

q3

q4

q5

q1. Confidence in carrying out a medication review q2 Confidence in identifying major and minor ADE q3. confidence in communicating with the GP on medication matters q4. confidence in working with student of another professional course q5. confidence in identifying barriers to medication management patient faces during transition period

Fig. 2. Pre-post questionnaire data (rating 1–4) Pharmacy and Medical students. 4

4

Pharmacy

3.5

Medical

3.5

3

3

2.5

2.5 pre-test

2

pre-test

2

poat test

1.5

post-test

1.5

1

1

0.5

0.5

0 Q1

Q2

Q3

Q4

0

Q5

Q1

Q2

Q3

Q4

Q5

q1. Confidence in carrying out a medicaƟon review q2 Confidence in idenƟfying major and minor ADE q3. confidence in communicaƟng with the GP on medicaƟon maƩers q4. confidence in working with student of another professional q5. confidence in idenƟfying barriers to medicaƟon management paƟent faces during transiƟon period

Fig. 3. Feedback on the IPL module Overall student responses (rating 1–4). Total student parƟcipants' raƟng 4 3.5 3 2.5 2

rating

1.5 1 0.5 0 Q1

Q2

Q3

Q4

Q5

Q6

Q7

Q8

Q9

Q10 Q11

Q1. Resource sessions in preparing for HMR Q2. Clinical Pharmacology is about rational management of medications Q3. Medication review: made me confident in carrying out MR Q4. Resources were useful Q5. awareness of importance of other professional in MR Q6. Resource session with supervisor Q7. Team effort in resolving medication issues Q8. Awareness of barriers to medication management by patient during transition Q9. Previous knowledge supports final year programme demand Q10. IPL module prepared me to carry out a comprehensive MR Q11. Awareness of importance of multidisciplinary approach

Fig. 4. Feedback on the IPL module: Total student responses (rating 1–4). 4 3.5 3 2.5 Pharmacy rating

2

Medical rating

1.5 1 0.5 0 Q1

Q2

Q3

Q4

Q5

Q6

Q7

Q8

Q9 Q10 Q11

Q1. Resource sessions in preparing for HMR Q2. Clinical Pharmacology is about rational management of medications Q3. Confidence in carrying out MR Q4. Resources were useful Q5. Awareness of importance of other professional in MR Q6. Resource session with supervisor Q7. Team effort in resolving medication issues Q8. Awareness of barriers to medication management by patient during transition Q9. Previous knowledge supports final year programme demand Q10. Online and face to face modules prepared me to carry out a comprehensive MR Q11. Awareness of importance of multidisciplinary approach

185 AN LC-MS/MS METHOD FOR THE DETERMINATION OF LOPINAVIR (LPV), RITONAVIR (RTV), EFAVIRENZ (EFV) AND NEVIRAPINE (NVP) FROM 10 lL HUMAN PLASMA BY LC-MS/MS USING PROTEIN PRECIPITATION Van Wyk A., Castel S., Smit P., Norman J., Wiesner L. Background: Efavirenz (EFV) is a non-nucleoside reverse transcriptase inhibitor (NNRTI) and is used as part of highly active antiretroviral therapy (HAART) for the treatment of a human immunodeficiency virus (HIV) type 1. Nevirapine (NVP) is another NNRTI used to treat HIV-1 infection and AIDS. Lopinavir (LPV), an antiretroviral of the protease inhibitor class, is used in a fixed-dose combination with ritonavir. Ritonavir (RTV) is an antiretroviral drug also from the protease inhibitor class. HIV rapidly develops resistance if any of these drugs are used alone. It is therefore recommended that therapy consists of combinations of three or more antiretrovirals. This has prompted the development of this multiplex assay for the determination of these four antiretrovirals in human plasma. We describe here a sensitive and robust LC-MS/MS method coupled with simple protein precipitation extraction for the determination of LPV, RTV, EFV and NVP from small volumes of plasma. Methods: A method was developed for the quantification of these four drugs in 10 ll human plasma, and consisted of protein precipitation with acetonitrile, followed by high performance liquid chromatography with MS/MS detection using an AB Sciex API 4000 mass spectrometer. The calibration curves fits quadratic (weighted by 1/concentration2) regressions for each drug over the ranges 0.0195–20 lg/ml for LPV, EFV and NVP, and 0.00488–5.00 lg/ml for RTV. The following transitions were monitored; m/Z 629.5?120.2 for LPV, m/z 721.5?296.1 for RTV, m/Z 316.1?244.0 for EFV and m/Z 266.9?198.2 for NVP. Results: Accuracy and precision were assessed over three consecutive, independent runs. The combined accuracy statistics during the validation of all quality controls were between 94.0% and 98.5% for LPV, 93.6% and 98.2% for RTV, 95.0% and 100.2% for EFV, and 90.8% and 100.8% for NVP. The overall precision was 0.05) in PTH, ALP, SCr, CrClCG and TmPO4/GFR among the three groups. Mean PTH and ALP (bone resorption markers) were highest in the TDF exposed group, 53 pg/mL and 108.1 U/L, respectively but was not statistically significant. Conclusion: In this pilot study no statistical significant differences in bone density and renal function markers were observed. However the increase in parathyroid hormone (although not statistically significant) in the tenofovir exposed women needs further investigation within a larger sample size.

194 BUILDING A DECISION-TREE-ANALYSIS MODEL AS A TOOL FOR COST COMPARISON OF METHICILLINRESISTANT STAPHYLOCOCCUS AUREUS (MRSA) MANAGEMENT IN THE CHARLOTTE MAXEKE JOHANNESBURG ACADEMIC HOSPITAL (CMJAH) Ahmed-Hassen S., Moch S., Miot J. Background: Infections due to MRSA increase costs of patient care within hospitals. Patients can be screened for MRSA using the conventional culture method or the new, more rapid Polymerase Chain Reaction (PCR) testing. International studies have compared the costs and treatment pathways when using these two methods of MRSA screening. However in the South African context where socio-economic status and access to healthcare may contribute different influences compared with other countries, no such models exist. Therefore, the aim of this study was to develop a decision-tree-analysis model as a baseline for comparison of costs associated with using the conventional culture method vs. PCR testing for MRSA in a Johannesburg tertiary services hospital. Method: Using TreeAge Softwareâ a baseline decision-tree-analysis model was developed to evaluate the management of suspected MRSA positive and negative patients using the current conventional culture screening method at CMJAH. The model was populated with input parameters using data from retrospective patient records, observations in hospital wards and diagnostic laboratories as well as published literature. Results: Initial observation in the Septic and Tumour Orthopeadic ward 369, revealed that nurses measured the patients’ temperature twice daily and only if febrile (temperature > 37.5°C), was a septic workup initiated, this included but was not limited to: urine MCS, blood/tissue/pus cultures and administering a broad-spectrum antibiotic while waiting for results. It was also observed that post-operation, intravenous gentamicin was administered to patients until three negative culture results were obtained. If all three results were negative, the intravenous gentamicin was stopped and the patient could be discharged, however if a positive result was obtained, the antibiotic was changed accordingly. Conclusion: This model demonstrates the use of empiric antibiotics while waiting 2 or more days for culture confirmation of suspected

infection. This model will then be compared to a theoretical model of using a more rapid PCR screening to evaluate the differences in clinical management and costs incurred in using these two different methods of MRSA screening.

195 CAN SEVERE ADVERSE REACTIONS OF NSAIDS BE PREVENTED? Matvieiev O., Ezernitsky O., Matvieieva N. Background: Non-steroidal anti-inflammatory drugs (NSAIDs) is a large group of drugs which is widely used in therapy of different conditions. It is possible due to such effects of NSAIDs as anti-inflammatory, antipyretic and analgesic. Wide use of NSAIDs leads to high frequency of their adverse reactions (ADR) even those which are very rare. The situation worsens by belonging of many NSAIDs to overthe-counter group of drugs. Methods: We have analyzed a data of 3430 spontaneous reports sent to Regional office of pharmacovigilance in 2011–2013 in Crimea, Ukraine and recorded in local database ARCADe handled by FileMaker software. 306 reports (8.92%) informed about ADRs of NSAIDs. Severity was estimated accordingly to Karch-Lasagna method, LDS points and Hartwig-Siegel scale and simplified to variables ‘nonsevere’, ‘moderate’ and ‘severe’ (included lethal ones). Preventability of ADRs was assessed by Hallas and Schumock-Thornton criteria and Granada test and simplified to variables ‘preventable’ and ‘non-preventable’. StatsDirect software was used for statistical calculations. Results: Most ADRs were caused by Diclofenac (68 cases; 22.2%), Ibuprofen (57; 18.6%), Paracetamol (36; 11.7%), Metamizole (34; 11.1%), Ketorolac (24; 7.8%), Acetylsalicylic acid (24; 7.8%) and Nimesulide (21; 6.9%). Agreement analysis revealed moderate strength between chosen methods of preventability (Cohen-Fleiss`s j = 0.58) and fair strength for methods of severity assessment (j = 0.34). We did not find any significant relationship between preventability and severity for all mentioned products except Diclofenac. For latter we found moderate (Cramer’s V = 0.41) significant (Fisher P = 0.005) relationship between the preventability measured accordingly to Schumock -Thornton criteria and severity measured in LDS points. The same methods revealed low (V = 0.15) significant (P = 0.023) association for all 306 ADRs of NSAIDs. This finding was also confirmed by significant trend in mean scores for pairs ‘Granada test and LDS points’ (P = 0.027) and ‘Schumock -Thornton and Karch-Lasagna criteria’ (P = 0.023). Conclusion: Our results do not confirm or refute statement that preventable ADRs of NSAIDs are more severe that non-preventable however we found one for whole group and Diclofenac. Obviously, the results of similar studies appreciably depend on chosen methods. 196 CARE WHEN INTRODUCING PRESCRIBING RESTRICTIONS; IMPLICATIONS FOR HEALTH AUTHORITIES Godman B., Bennie M., Sakshaug S., Simoens S., Vlahovic-Palcevski V., Malmstrom R. Background: Unsustainable growth in pharmaceutical expenditure, rising by more than 50% in the past decade among OECD countries, has resulted in multiple initiatives across Europe to lower generic prices and enhance their utilisation. These include prescribing restrictions where all drugs in the class are seen as essentially similar, e.g. PPIs, renin-angiotensin inhibitor drugs and statins. However, there have been concerns with their impact on subsequent quality of care and their influence in reality given differences in enforcement and timing. As a result, need to (i) Review their influence on subsequent utilisation alongside differences in their nature, drug class and timing; (ii) Ascertain whether prescribing restrictions can be added to existing


62 demand-side measures to further enhance prescribing efficiency; (iii) Whether such measures compromise the subsequent quality of care. Method: Principally a narrative review of published studies. Results: Prescribing restrictions have a variable impact on the subsequent utilisation of patented products, with their influence depending on their nature, follow-up and timing rather than class. This is seen by (i) more aggressive follow-up of prescribing restrictions for patented statins led to a greater reduction in their subsequent use in Austria (66%) vs. Norway (44%), (ii) limited impact of prescribing restrictions for patented statins in Sweden when introduced some 6 years after multiple measures among the regions to encourage the prescribing of generic statins, (iii) lower use of patented ARBs vs. generic ACEIs in Croatia than Austria with more aggressive physician follow-up, (iv) limited impact on esomeprazole utilisation in Norway with first prescription or recommendation in hospital where restrictions do not apply, (v) significant increase in losartan utilisation in Austria and Belgium once restrictions removed (still apply for patented ARBs), (vi) limited impact of prescribing restrictions for duloxetine in Sweden as need to tailor drugs – but increased venlafaxine utilisation, (vii) prescribing restrictions can be successfully added to existing measures, e.g. statins in Austria, (viii) restrictions do not appear to compromise subsequent quality of care. Conclusions: Prescribing restrictions can be a successful strategy. However, health authorities need to take care with their planning including follow-up else they could be disappointed with their outcome.

197 CHANGES IN DRUG PRESCRIPTIONS AFTER THE SMART TRIAL – SECONDARY DATA ANALYSIS OF BAVARIAN ASTHMA PATIENTS Rottenkolber M., Fischer R., Iba~nez L., Fortuny J., Reynolds R., Gerlach R., Tauscher M., Thuermann P., Hasford J., Schmiedl S. Background: Following the publication of the SMART trial (2006) which evaluated the safety of salmeterol (long-acting beta-2-agonist (LABA)) in asthma patients, drug authorities explicitly recommended to use LABA only in patients receiving inhaled corticosteroids (ICS). In addition, benefits of LABA/ICS fixed combination drugs were widely discussed. We aim to describe LABA- and ICS-related prescription patterns after the SMART trial in Germany. Methods: Patients documented in the Bavarian Association of Statutory Health Insurance Physicians database (covering 85% of the population, i.e. approximately 10.5 million people) were included if they had a diagnosis of asthma and at least one prescription of LABA and/ or ICS between 2004 and 2008. Annual period prevalence rates (PPRs) per 10,000 insured persons were calculated for all LABA and ICS compounds separately. For five mutually exclusive patient categories ‘concomitant LABA and ICS users’, ‘switchers’, ‘non-concomitant LABA and ICS users’, ‘LABA users without ICS’, and ‘only ICS user’, stratified analyses by calendar year, age and sex were made. For ‘concomitant LABA and ICS users’, fixed vs. non-fixed LABA/ICS prescriptions were compared. Cochrane Armitage test was used to analyse trends over time. Results: Highest annual PPRs were found for budesonide (between 76 and 91 per 10,000 persons) and the fixed combination of salmeterol/ fluticasone (between 62 and 73 per 10,000 persons). The fraction of ‘concomitant LABA and ICS users’ increased significantly from 52% to 58% within the study period, whereas for ‘LABA users without ICS’ a significant decrease from 6.5% to 5.4% was found. The fraction of patients with at least one LABA prescription without concomitant ICS was highest in elderly, male patients (approximately 20%). In ‘concomitant LABA and ICS users’, the fraction of patients with two separate inhalers decreased significantly from 12.1% to 9.5% within the study period. Conclusions: In a representative German population, we found in general an increase in guideline-adherent LABA prescriptions in terms of concomitant ICS prescribing. Nevertheless, elderly men received a sig-

nificant number of LABA presciptions without concomitant ICS which might increase the risk for LABA-related adverse events.

198 CHANGES IN PRESCRIBED DAILY DOSAGES OF ANTIDEPRESSANTS IN COMPLIANT VS. NON-COMPLIANT PATIENTS Slabbert N., Lubbe M., Harvey B., Brink T. Purpose: To compare changes in the prescribed daily dosages (PDD) of antidepressants (AD) in compliant vs. non-compliant patients. Methods: We conducted a prospective, cohort study, analysing AD medicine claims (N = 35,175) for 14,135 patients, obtained from a South African Pharmaceutical Benefit Management (PBM) company, over a 6 year study period, viz. 1 January 2006 to 31 December 2011. The medicine possession rate (MPR) was used as proxy to determine patient compliance and its relation to the generic name and class of AD medication. The following inclusion criteria were applied: only patients older than 18 years and treatment initiated by a psychiatrist. A patient was considered compliant to AD treatment if the MPR was ≥80% and ≤110% and with a >4 month treatment period. Changes in prescribed daily dosages (PDD) (in mg) were determined for each antidepressant dispensed to compliant vs. non-compliant patients. PDD changes were adjusted for differences in initial PPD, based on correlation results (P > 0.0001, r > 0.5) between initial PDD and PDD changes. Results: The mean MPR of 35 175 AD items was 98.2% (95%CI, 96.6–99.9). Only 50.8% of dispensed items were associated with an acceptable MPR (≥80% and ≤110%). After the application of the >4 month treatment period, only 34% of patients were compliant. The ten most dispensed antidepressant medications represented 86.6% of all AD drugs. These include venlafaxine, escitalopram, duloxetine, mirtazapine, citalopram, fluoxetine, bupropion, sertraline, amitriptyline and trazodone. A statistically significant association was found between the type of active ingredient consumed and compliance (P < 0.0001). Only 26.2% of patients who received amitriptyline-containing products were complaint compared to 38.8% and 38.7% in the cases of venlafaxine and duloxetine, respectively. Although, statistically significant differences (P < 0.05) in the adjusted PDD changes were found between complaint and non-compliant patients on venlafaxine, escitalopram, mirtazapine and amitriptyline therapy no practical significance was found (d > 0.8). Conclusion: Only one-third of MDD patients found to be compliant with AD medication, which correlates with the current global noncompliance data. The treatment period has an outspoken effect on the MPR.

199 CHEBYSHEV’S INEQUALITY METHOD FOR PHARMACOKINETIC STUDIES Kadusevicius E., Noreikaite A., Stankevicius E., Saint-Marcoux F. Abstract: Objective: To implement Chebyshev’s inequality method for cyclosporine therapeutic range calculation. Methods: The cyclosporine exposure level was based on calculation of the mean area under the concentration-time curve AUC(0–12) using a Bayesian estimator and a 3-point limited sampling strategy, at 0, 1, and 3 h post-dose. Blood samples of 175 patients with target AUC(0– 12) exposure of 3.8 mg h/L (kidney post-transplantation time >1 year) were evaluated. The therapeutic window of cyclosporine was calculated by the Chebyshev’s inequality method with a 99% guarantee (a = 0.01) using the IBM SPSS Statistics 20 software. Results: It was found, that the therapeutic window of cyclosporine differs in different post-transplantation time groups: the estimated AUC(0–12) exposure range is 3.05–3.75 mg h/L (an average AUC(0–


63 3.60 mg h/L) in the group of patients who have a graft for 1– 5 years, while in the group of patients who have a graft longer than 5 years is 2.70–2.98 mg h/L (an average AUC(0–12) exposure is 2.84 mg h/L). In the 1–5-year post-transplantation time group, there were 17 patients (37.0%) within the estimated therapeutic range and 28 patients (21.7%) in the group of who have a graft longer than 5 years vs. 29 patients (16.6%) in all study groups of 175 patients. Conclusions: All study patients had the same theoretical AUC(0–12) exposure target of 3.8 mg h/L, but cyclosporine therapeutic window, calculated by Chebyshev’s inequality method according to the posttransplantation time, better defines the dosage accuracy. Chebyshev’s inequality could be an appropriate and more precise method to determine the therapeutic window for CsA monitoring than the target AUC(0–12), but further studies should be conducted to evaluate patients with postoperative time C), rs7793861 (CYP51A1_c.*251G>C), rs16947 (CYP2D6_2850C>T) showing values of >0.48. Among the rest of the 1494 gene variants, we were also unable to detect minor alleles for 44% of variants, whereas 68 variants showed MAF values of >0.45. These included three NAT2, 5 SLC, three UGTB and four ABC variants showing MAF values on >0.49. Comparing with other ethnic populations, these results show a distribution of some variations which appears to be unique for our study population. This points to the need to verify the applicability of a SNP profile in a given population prior to pharmacogennetic screening for personalized medicine.

210 DETERMING FACTORS ASSOCIATED WITH ADHERENCE RATES TO HIGHLY ACTIVE ANTIRETROVIRAL THERAPY AMONG PATIENTS ATTENDING A PUBLIC HEALTH CARE FACILITY IN MTHATHA, EASTERN CAPE Katende-kyenda N., Lubbe M., Apalata T. Background: Adherence to HAART is a major determinant to HIV-infection control. Good ART-adherence achieves best virologicalresponse, reduces risk of drug-resistance, morbidity/mortality. Therefore the aim of this study was to determine factors associated with poor adherence to HAART in HIV-infected patients attending a primary healthcare centre in Mthatha. Methods: A descriptive-cross-sectional study was conducted in June 2013 among a convenience sample of ARV clinic attendees. Face-toface exit interviews using a standardized questionnaire was combined with patient record reviews to collect data of interest. Data were analysed using SPSS version 21. Pill counts were performed and the computed adherence-rate of ≥95% was considered acceptable. Univariate factors associated with poor adherence to HAART were assessed with


66 ANOVA. A logistic regression model was constructed to exclude confounders and to determine significant predictors of poor adherence. A P value of ≤0.05 was considered statistically significant. Results: A total of 102 HIV-infected patients on HAART were recruited, of which 85 (83.3%) were females and 17 (16.7%) males. The mean age (SD) of participants was 35.09 (9.3) years. Patients were on HAART for 24.68 (20.5) months. Univariate factors associated with poor adherence rates to HAART were: CD4 + T count >350 cells/mm3 (v2 = 46; P = 0.05), age >35 years (v2 = 28.75; P = 0.011), primary school level (Grade 7 and below) education (v2 = 9.18; P = 0.027), HAART regimen 1A-TDF (v2 = 14.37; P = 0.003), and more than 4 tablets (including ARVs) (v2 = 11.87; P = 0.001). There was a linear correlation between age >35 years and primary school level (Grade 7 and below) education (r = 0.538; P < 0.0001). After adjusting for univariate confounders, primary school level (Grade 7 and below) education (P = 0.020) and more than 4 tablets (including ARVs) (P = 0.026) were identified as significant predictors of poor adherence to HAART. Conclusions: Although there are recently an increase number of HIVinfected patients receiving HAART, findings from this study indicated that many of these patients will not adhere to their treatment once they start improving clinically (CD4 + T count > 350 cells/mm3). The underlying assumption remains lack of education and complexity of combined ARV tablets with other drugs received by HIV/AIDS patients.

211 DIABETIC MACULAR EDEMA TREATED WITH INTRAVITREAL RANIBIZUMAB AND GLICOMETABOLIC STATUS Eandi C., Serpe L., Eandi M., Alovisi C. Backround: Diabetic macular edema is a complication of diabetic retinopathy causing severe visual loss in working population. Several studies showed the efficacy and safety of intravitreal (IVT) ranibizumab for the treatment of this condition. However, the correlation between the effectiveness of this treatment and the glycometabolic status needs to be clarified. The aim of this pilot study is to analyse the response to IVT ranibizumab in patients with diabetic macula edema (DME) in relationship with metabolic parameters. Methods: Patients with DME at baseline visit underwent to visual acuity (BCVA) with ETDRS charts, complete ophthalmic evaluation, fluorescein angiography (FA, Heidelberg Spectralis), and optical coherence tomography (OCT, Heidelberg Spectralis). Thereafter, BCVA and OCT were performed monthly. Systolic blood pressure, glycaemia, glycate haemoglobin, cholesterol, and triglyceride were also measured at baseline and every 4 months. Follow up was 1 year. BCVA, central retinal thickness (CRT) and number of injections were recorded and analysed. Patients were treated according to the label. Results: Ten eyes of ten patients were included in this pilot study. Mean age was 66 years, 7 were female. Systemic parameters were within normal range in 3 of 10 patients at baseline and remained stable during follow up. Mean BCVA at baseline and at last follow up was 20/63 and 20/50, respectively. Mean CRT at baseline and at month 12 was 415 and 335 microns, respectively. Mean number of injections was 4.5. No correlation was found between metabolic parameters and CRT or number of injections. Conclusions: This is a pilot study with several limitations. First of all the sample size; then the relatively short follow up. Further investigations are necessary to understand the role of glycaemic metabolism in the treatment of DME.

212 DIFFERENT INITIATIVES ACROSS COUNTRIES TO ENHANCE THE PRESCRIBING OF GENERIC PROTON PUMP INHIBITORS; FINDINGS AND IMPLICATIONS Godman B., Bennie M., Campbell S., Wettermark B., Truter I. Background: Pharmaceutical expenditure has grown considerably in recent years driven by factors including ageing populations and premium priced drugs. This can compromise governments’ abilities to provide comprehensive and accessible healthcare. Increasing the use of low cost generics in a class where all products are seen as similar, e.g. PPIs, is one way to alleviate this. However, there is considerable variation in the utilisation of generic PPIs and their costs across countries. Methods: Retrospective observational studies in over 20 European countries and South Africa. Demand-side measures compared with PPI utilisation. Results: Compulsory generic substitution in South Africa in 2003 mirroring Sweden (2002). In addition, many medical aid schemes in South Africa only reimburse generic products and patients pay the price difference for a more expensive product, mirroring some European countries. Wide variety of demand-side measures in Europe varying from limited measures in Ireland to extensive initiatives in Netherlands, Sweden and UK. Generic prices also low in these three countries at 2– 10% of pre-patent loss prices. PPI expenditure in Scotland in 2010 was 56% below 2001 levels despite threefold increase in utilisation. In the Netherlands, expenditure fell by 58% in 2010 vs. 2000 despite threefold increase in utilisation. Expenditure adjusted for population size in Ireland was over ten times greater in 2007 vs. Sweden with increased prescribing of patented PPIs. In South Africa in 2010, omeprazole was the most prescribed PPI (51% of all prescriptions) followed by esomeprazole (20%) and lansoprazole (18%) enhanced by lower costs. Average cost/prescription for lansoprazole and omeprazole was 154.42 and 157.76 Rand respectively vs. 308.97 for esomeprazole, with high use of generic lansoprazole and omeprazole (98–99% of prescriptions) enhanced by lower costs (average of 52% to 67% of originator prices). High use of generic PPIs in Europe. Conclusions: Multiple demand-side measures enhanced prescribing of low cost generics to realise considerable savings without compromising outcomes, with no concerns with generic PPIs. Further measures needed in South Africa to lower generic prices similar to the Netherlands. Further measures needed to moderate high PPI utilisation due to increasing concerns with long term use.

213 DISCRETE EVENT SIMULATION OF HAEMOPHILIA A REAL-LIFE MANAGEMENT STRATEGIES WITH FVIII: INTEGRATING PK/PD, EPIDEMIOLOGY, CLINICAL EVIDENCE, AND COST-EFFECTIVENESS ANALYSIS Eandi M., Pradelli L., Della Pepa C., Povero M. Background: The improvement in haemophilia A (HA) management has extended survival and raised average quality of life of the affected population; yet, unfortunately, this is not true for all patients. The evolution of the strategies has brought up a series of new challenges, such as the choice of the most appropriate prophylaxis (PRO) regimens and the selection of patients for the induction of tolerance to inhibitors (ITI). Aim of the present study is to compare different HA treatment strategies, with plasma-derived (pdFVIII) and recombinant factor VIII (rFVIII), by discrete event simulation (DES) of both clinical outcomes and cost-effectiveness from the perspective of the Italian NHS. Methods: A DES model was developed to comprehensively microsimulate lifetime management strategies of individual HA patients: ondemand (OD) treatment of bleeding, alternative PRO programs with FVIII, and ITI. Patient characteristics and disease evolution were modeled on data available in clinical literature to represent the current profile of HA management. Individual bleeding risks are modeled sampling distributions fitted on epidemiological data; bleeding control is modeled using a continuously updated FVIII PK/PD function.


67 Alongside current HA management, alternative strategies are simulated to explore their cost impact for the Italian NHS. Only direct sanitary costs are considered in the simulation, with a 3.5% annual discount rate. Results: Total costs are strongly and positively correlated with HA severity: mean lifetime cost per patient increases from €86,269 (mild HA) to €1,509,231 (severe HA) for patients treated with pdFVIII (average €1,332,373), and from €147,900 to €2,621,540 in patients treated with rFVIII (average €2,013,222). The incremental cost-effectiveness ratio (ICER) of primary prophylaxis on severe HA patients, in comparison with OD, is estimated at €642 and €3,728 per avoided bleeding, for pdFVIII and rFVIII, respectively. Conclusions: Management of HA patients is complex and difficult to optimize; despite involving a limited number of patients, costs for the Italian NHS are extremely high. Our DES model, integrating PK/PD functions with epidemiology and clinical evidence, permits to realistically micro-simulate any HA management strategy, both in terms of regimen and of selected product. Long-term primary prophylaxis is estimated more cost-effective than OD treatment.

214 DOPAMINE PROTECTS AGAINST DNA DAMAGE DURING CELLULAR HYPOTHERMIA AND REWARMING IN CULTURED SMOOTH MUSCLE CELLS Tolouee M., Trentadue G., Henning R. Introduction: Ischemia is a condition suffered by cells and tissues when deprived of blood flow due to inadequate nutrient and oxygen supplementation. This is mainly due to the rapid generation of reactive oxygen species (ROS) at the start of reperfusion and characterized by apoptotic cell death. Similarly, many mammalian cell types are vulnerable to prolonged and profound hypothermic storage related to a burst of ROS upon rewarming, which may be attenuated by treatment with dopamine. Cooling/rewarming is thus thought to represent ischemia/ reperfusion condition in cell lines. As oxidative stress is causing DNA damage, here we investigated whether cooling/rewarming injury leads to DNA strand breaks by Comet assay. Methods: Smooth muscle aortic cells (SMAC) were cultured in 6 well plates until confluency, cooled at 4°C for 24 h with or without subsequent rewarming to 37°C. Untreated cells (37°C) served as a control. Dopamine (50 lM) was added throughout the whole procedure, or during parts of the cooling-rewarming protocol. Single and double strand DNA breaks, DNA cross-links and oxidative damage in single cells was assessed by Comet assay. Per condition >60 Comets were analyzed and damage was expressed as the percentage of DNA in the Comet’s tail (% TailDNA) using ImageJ software. Results: Compared to control cells, the %TailDNA was substantially increased both by cooling and cooling/rewarming, as reflected by an increase in the median %TailDNA from 8.4% to 87.3% and 76.4%, respectively. Dopamine treatment throughout the protocol strongly reduced DNA damage, as evidenced by a substantial decrease in the median of %TailDNA to 49.9%. Notably, when treatment with dopamine was restricted to 30 min. Prior to cooling, reduction in DNA damage was absent. Conclusion: Our results demonstrate that dopamine protects SMAC cells from DNA damage induced by hypothermia/rewarming by protection of cells in the hypothermic and rewarming phases. Thus, dopamine represents a pharmacologically promising compound for several clinical applications that require hypothermia and rewarming, such as the cold preservation of transplant organs.

215 DRUG ADVERSE REACTIONS ASSOCIATED WITH THE USE OF DISEASE MODIFYING ANTIRHEUMATIC DRUGS IN COLOMBIAN PATIENTS Machado J., Machado M. Objective: Describe the adverse drug reactions (ADR) that occurred on patients diagnosed with Rheumatoid Arthritis, who were treated in a specialized Centre within the Health System of Colombia. Methods: Cross-sectional study, using information from all patients of any age and gender, treated in a specialized healthcare Centre in the following Colombian cities: Bogota, Cali, Manizales, Medellın and Pereira, from December 1st 2009 and August 31st 2013. The data was obtained from clinical records, and ADRs were classified according to WHO Adverse Reaction Terminology (WHO-ART). Analysis was conducted using SPSS 22.0 Results: Four hundred and seventy-three ADR reports were obtained in 371 patients, from a cohort of 1364 individuals (27.2%). There was a female preponderance (n = 324, 87.3%) and a mean of 52.7 13.1 years. The frequency of patients that presented any ADR were: Infliximab (36.2%), adalimumab (22.1%), etanercept (15.6%), methotrexate (15.0%), cloroquine (11.4%), and leflunomide (10.4%). The most frequent ADRs were ‘hepatic and biliary duct disorder’ and ‘gastrointestinal system disorders’. 88.7% of the ADRs were Type A; among them 37.0% were classified as mild, 39.0% as moderate and 24.0% as severe. Medication was suspended on 76.4% of patients experiencing an ADR. Conclusions: The occurrence of ADR on treated Rheumatoid Arthritis patients are frequent, especially the ones associated to the use of antirheumatic biotechnological drugs; report and following should be encouraged to reduce the risk on these patients.

216 DRUG UTILIZATION REVIEW OF ERGOTAMINE IN COLOMBIA Machado J., Morales C. Introduction: Ergot derivatives are drugs designed to abort migraine attacks, with vasoconstrictor effects. To determine how ergotamine derivatives are being prescribed by physicians in Colombia and variables associated with inappropriate prescribing and their potential interactions in patients in 2012. Methods: We reviewed 86,411 prescriptions made during April 2012. We identified the drug, dose, interval, time of use and indication. We interviewed 288 randomly selected patients in whom also sought to concomitant use of: (i) antihypertensive (ii) ischemic heart disease medications (iii) antiretrovirals (iv) other antimigraine (v) macrolides, because of their interactions. Results: We obtained 801 prescriptions in 27 cities with mean age: 35.1 14.1 years, 82.5% of prescriptions in women, 96.5% held by the primary care physician; (n = 524, 65.4%) cases for migraine, were found 26 different prescription forms and 797 incorrect prescriptions regarding improper usage recommendations (99.5%). Inappropriate prescribing was significantly associated with health service (P = 0.005). In patients who were contacted by telephone (n = 266, 92.4%) took it according to the incorrect indication. We found 54 (6.7%) patients taking antihypertensive drugs, 24 (2.9%) patients with macrolides, five more other antimigraine drugs. Discussion: Most patients are receiving improperly ergotamine, plus possible interactions that increase the risk of health problems such as ergotism and coronary events. Assessment measures should be implemented, updating and training for physicians.


68 217 DRUG-DRUG INTERACTIONS – A PROBLEM OF MULTIPLE DRUGS OR MULTIPLE PRESCRIBERS? Böttiger Y., Andersson M., Eiermann B. Background: Drug-drug interactions, leading to adverse effects as well as lack of effect, represent an increasing problem in health care. In Sweden, 2% of all patients above the age of 65 are prescribed drug combinations that should better be avoided, despite the fact that most prescribers have access to computerized decision support (CDS) integrated in their electronic health record system. Are these unsuitable drug combinations a result of different prescribers, not being fully aware of all prescriptions to the same patient, or do they arise from single prescribers? Aim: To investigate to which extent possible drug-drug interactions arise from prescriptions issued by different prescribers as compared with single prescribers. Methods: All prescriptions to patients who visited any of 20 primary health care centers (PHC) in Stockholm during two 4 month periods were retrieved from the Swedish Prescribed Drug Register. All drugs prescribed to one patient within the same 4 month period were considered to be used concomitantly. Possible drug-drug interactions were identified and classified according to the SFINX database. Prescriptions issued from the patients ‘home’ PHC were separated from those issued by other prescribers. Results: In total, 489,000 prescriptions to 110,000 patients were included in the study. 68% of these were issued from the patients home PHC and 32% from other prescribers. 2836 D-interactions (serious) and 25,507 C-interactions (may demand dose adjustment) were identified. 62% of D-interactions and 48% of C-interactions were the result of prescriptions issued by different care givers. Conclusion: According to this study, drug-drug interactions in primary health care settings arise more often as a result of multiple prescribers, than from a single care giver prescribing multiple drugs. Thus, one common patients0 drug list and communication between different care givers must be considered important tools, beside the knowledge about drug-drug interactions, in preventing adverse outcomes from drug treatment.

218 EDUCATION TO PATIENTS ON ORAL ANTICOAGULANT TREATMENT (OAT): ASSESSMENT AND OPTIMIZATION OF KNOWLEDGE Izazola-Conde C., Majluf-Cruz A., Mandoki J., Molina-Guarneros J. Background: Insufficient education, polypharmacy and INR’s values above therapeutic range have been reported as predictive factors for bleeding complications in patients on OAT. It has been reported that over 60% of aged patients have not received any education about their OAT. Some patients must remain on OAT during their entire life for different indications (thrombophilia, prosthetic cardiac valves, atrial fibrillation, etc.). Patient education about OAT has limitations, knowledge assessment lacks frequently and while some authors have reported that patient education has a positive effect on quality of treatment control, others have failed to show or dismissed an improvement. Objectives: To assess (i) Knowledge of patients about their OAT, (ii) to improve patients’ Knowledge about OAT, (iii) to assess retention of information about OAT after an educational intervention, (iv) to improve quality of treatment with oral anticoagulant antagonists of vitamin K (AVK) drugs. Methods: Thirty three patients attending an anticoagulation clinic and receiving OAT with AVK answered a validated 87 items questionnaire before an educational intervention and were followed at weeks 1–8 and at weeks 24–36 after educational intervention for knowledge assessment about their OAT. Patients INR’s were registered at least 6 months before and 6 months after the educational intervention. Results: Knowledge about OAT after educational intervention was significantly improved. Correct answers to the questionnaire in the first

postintervention assessment showed a 30% increase, knowledge was retained after 6 months. All topics explored by the questionnaire improved, however some of the topics remained below 80% of correct answers: mechanism of action, adverse effects, interactions with other drugs and interactions with alcohol. Percentage of INR in range per individual patient did not show a significant improvement. Conclusions: Knowledge about OAT may be improved in patients on OAT. Patient education about OAT should be provided and evaluated, information given to patients may not be understood or retained without a specific educational intervention. We are currently exploring more efficient ways to deliver education to the patients on OAT. One possible explanation to lack of effect on INR control may be variability of duration of treatment in patients involved in this study.

219 EFFECT OF EDUCATIONAL INTERVENTIONS ON REPORTING OF ADVERSE DRUG REACTIONS BY PRESCRIBERS Desai C., Patel S., Patel P., Dikshit R. Background: Under reporting is a major drawback of Spontaneous adverse drug reaction (ADR) reporting. Various interventions are reported to improve spontaneous ADR reporting. We assessed the impact of certain educational interventions on ADR reporting at a tertiary care teaching hospital in India. Methods: Following approval from Institutional Ethics Committee, 169 randomly selected prescribers from different specialties of the tertiary care teaching hospital were assigned to Groups A and B. Study tools included a prevalidated knowledge, attitude and practice (KAP) questionnaire, ADR Reporting Forms and Educational Interventions (EI). Data was collected in pre- intervention (Pre-IP), intervention I and II (IP-I, IP-II) and postintervention (Post-IP) phases, each of 3 months. Both groups were educated through a Pharmacovigilance Awareness Programme and posters. Additionally, Group A (n = 84) also received SMS alerts (IP-I) and personal briefings (IP-II) while Group B (n = 85) received e- mails (IP-I) and information leaflets (IPII). The KAP scores, rate and quality of ADR reporting (measured by completeness score (CS) of ADR reporting form) were compared in both groups, in all phases using Chi Square and Student’s ‘t’ test and INSTAT 3. Results: Both groups were comparable at baseline. The mean Pre-IP KAP scores were similar in both groups (P = 0.35). Post-IP, the scores were significantly higher in group A than in group B (P < 0.0001). A total of 190 ADRs were reported (Group A-119 and B-71). A higher number were reported in IP-I (n = 48) and IP-II (n = 57) as compared to Pre-IP (n = 8; P < 0.05). A higher number of serious ADRs were reported in IP-II than in Pre-IP (P < 0.05). Mean CS was significantly higher in IP-II as compared to Pre-IP (P < 0.001). Rate of reporting was better in group A than group B in IP- I (P = 0.02) and II (P = 0.01). However, Post-IP, the number of ADRs including serious ADRs was lower as compared to IP-I and IP-II (P < 0.05). Conclusions: Educational interventions improve rate and quality of ADR reporting, albeit temporarily. SMS alerts and personal briefings have a better impact than email alerts and information leaflets. Sustained use of a combination of educational interventions is recommended to improve ADR reporting.

220 EFFECTIVENESS OF ANTIRETROVIRAL TREATMENT IN COLOMBIA Machado J. Objective: To evaluate the effectiveness of antiretroviral therapies and factors associated with HIV/AIDS control in a population of patients treated by the Colombian Social Security Health System (SGSSS).


69 Methods: This was a descriptive study of 510 HIV/AIDS patients treated with antirretroviral therapies in 19 cities in Colombia from June 1992-April 2011. Factors assessed from each patient’s clinical history were: first, annual and last viral load (controlled viral load 250 cells/mm3), antiretroviral treatment regimens and switches of these, prescribed daily doses of medications, length of disease evolution, time duration of therapy, number of antiretroviral pills/day, report of treatment failure, report of any antiretroviral drug intolerance, report of non adherence (from 1 day of fault), time between diagnosis and initiation of therapy, comorbidities, diagnosis of anxiety or depression and drug costs. Results: Patients were predominantly male (75.1% males vs. 24.9% women), with a mean age of 41.0 11.4 years and an average length of disease progression of 72 months. All recommended treatment regimens were prescribed at recommended daily doses. Treatment was effective in 65.3% of patients (viral load T, 354G>T, or 16T>C) was significantly (P < 0.03) increased, when compared to that (9.88 6.24 nM/mg) in patients with wild types of these genes. Adjusted MTX-PGs in patients with SLC19A1 and/or other GGH gene variants were not changed from that in patients with wild types of these genes. Conclusion: GGH may play an important role in regulation of the intracelluler MTX-PGs. Since intracellular MTX-PGs level is known as a surrogate marker of the efficacies and/or adverse responses of MTX, measurement of the GGH polymorphisms may predict clinical responses to MTX therapy.

226 GENERIC USE IN ANTIDEPRESSANT SALES IN GREECE Papaioannidou P., Ntaralas A. Background: The use of generics in the Greek market of medicines is generally lower than in other European countries. An effort has been made recently by Greek authorities to encourage generic prescribing among physicians of the National Health System (NHS). In this study the use of generics in antidepressant sales was calculated in a sample of sales from the market of Thessaloniki, the second largest city in Greece. Methods: A sample of antidepressants registered sales was collected from the new registration system, which has been applied during the last 2 years in Greece. The sample corresponds only to a small amount of sales from the market of Thessaloniki, during the years 2012 and 2013. The percentage of generics in the sale of each medicine was calculated out of a variety of brand names in each class of antidepressants. The amount of medicines was estimated in Defined Daily Doses (DDDs) of the reference drug and its generics. Results: Selective Serotonin Reuptake Inhibitors (SSRIs) and Serotonin-Norepinephrine Reuptake Inhibitors (SNRIs) have displaced older antidepressants, corresponding to 88% of total antidepressants sales. Generic use corresponded to 26% of total sales and varied greatly among various SSRIs, being only 15% of fluoxetine sale, 36% of sertraline sale and reaching up to 53% of citalopram sale. Conclusions: Generic use in antidepressants is quite low in Greece, and should be increased, in order to lower the burden of medicines cost in NHS.

Methotrexate (MTX) is an anchor drug for treatment of rheumatoid arthritis (RA). MTX is intracellularly converted to polyglutamate com© 2014 The Authors Basic & Clinical Pharmacology & Toxicology © 2014 Nordic Pharmacological Society. Basic & Clinical Pharmacology & Toxicology, 115 (Suppl. 1), 1–374

71 227 GENOTYPING OF PGENI-RELATED GENE VARIANTS BY GENOME-WIDE SINGLE NUCLEOTIDE POLYMORPHISM ANALYSIS Dzimiri N., Wakil S.M., Nguyen C., Muiya P., Andres E., Baz B., Morahan G., Meyer B.F. Recently a number of single nucleotide polymorphisms (SNPs)/genes have been identified through various studies and assembled in a PGENI platform on the DMET chip for clinical use in personalized medicine. However, the clinical relevance of these variants is likely to vary dependent on ethnicity. Hence, the identification of the relevant variants of interest is a prerequisite for a given population as basis for employing the data clinically. We determined the distribution of some of the PGENI variants in a population of 5600 Saudi individuals randomly selected from our cardiovascular disease database by sequencing on the Affymetrix platform using the SNP 6 chip. In all, 27 cytochrome P450 (CYP450) variants were genotyped, including three CYP3A5 variants on chromosome (chr) 7, three CYP2C19, two CYP2C9 and three CYP2C8 on chr 10, four CYP2A6 and three CYP2B6 on chr 19, as well as eight CYP2D6 variants on chr 22. Apart from the CYPs, 38 variants were also genotyped in genes encoding other important enzymes, receptors and signaling mediators, among others. These include the ADRB2, MTHFR, NAT2, SCNSA, TPMT and VKORC1 gene variants. Several of the variants exhibited very low minor allele frequencies (MAF) of