Glycopyrroniumbromide for the treatment of sialorrhea in parkinson's disease

Maurits E. L. Arbouw,1,2,3 Kris L. L. Movig,3 Miriam Koopmann,3 Petra J. E. Poels,4 Henk-Jan Guchelaar,5 Toine C. G. Egberts,1,2 Cees Neef6 & Jeroen P. P. van Vugt4
1Department of Clinical Pharmacy, University Medical Center Utrecht, Utrecht, The Netherlands; 2Division of Pharmacoepidemiology and Pharmacotherapy, Utrecht Institute for Pharmaceutical Sciences, Faculty of Science, Utrecht University, Utrecht, The Netherlands; 3Department of Clinical Pharmacy, Medisch Spectrum Twente, Enschede, The Netherlands; 4Department of Neurology, Medisch Spectrum Twente, Enschede; 5Department of Clinical Pharmacy and Toxicology, Leiden University Medical Center, Leiden, The Netherlands; 6Department of Clinical Pharmacy and Toxicology, University Hospital Maastricht, Maastricht, The Netherlands

Objective: To determine the efficacy and safety of glycopyrroniumbromide 1 mg t.i.d. in the treatment of sialorrhea in patients with PD.

Background: Sialorrhea affects approximately 75% of patients with Parkinson's disease (PD). Sialorrhea is often treated with anticholinergics, but central side effects such as drowsiness, confusion or memory impairment limit their usefulness. Glycopyrroniumbromide is an anticholinergic with a quaternary ammonium structure that does not cross the blood-brain barrier in great amounts. Glycopyrroniumbromide exhibits minimal central side-effects, which may be an advantage in patients with PD of whom a significant portion already has cognitive deficits.

Methods: We conducted a 4-week randomized, placebo-controlled, double blind, cross-over study with glycopyrroniumbromide 1 mg t.i.d. in 25 patients with idiopathic PD. The severity of the sialorrhea was scored on a daily basis by the patients or a caregiver with a sialorrhea scoring scale ranging from 1 (no sialorrhea) to 9 (profuse sialorrhea).

Results: Two patients were excluded prior to randomisation because they met the exclusion criteria. Patients included had a mean (SD) age of 70.0 (±7.8) years and a mean (SD) disease duration of 10.2 (±8.6) years. One patient was excluded from the efficacy analysis due to a protocol violation. The mean sialorrhea score (SD) significantly improved from 4.7 (±1.7) with placebo to 3.9 (±1.6) with glycopyrroniumbromide (p = 0.048). Nine patients (40.9%) had a clinically relevant improvement of at least 30% with glycopyrroniumbromide versus one patient (4.5%) with placebo (p = 0.021). Sixteen patients (72.7%) had at least some improvement with glycopyrroniumbromide, compared to five (22.7%) with placebo. There were no significant differences in adverse events between glycopyrroniumbromide and placebo treatment. We did see a trend in dry mouth being reported more frequently with glycopyrroniumbromide. This adverse event is directly linked to the therapeutic effect of glycopyrroniumbromide.

Conclusions: Glycopyrroniumbromide 1 mg t.i.d. is an effective and safe treatment option for sialorrhea in patients with PD.

Phase I and pharmacological study of AZD1152, a selective inhibitor of aurora kinase B

D. S. Boss,1 P. O. Witteveen,2 M. Mergui,1 J. van der Sar,1 P. K. Stockman,3 E. E. Voest,2 J. H. Beijnen4 & J. H. M. Schellens1
1Department of Medical Oncology, The Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands
2Department of Medical Oncology, University Medical Centre, Utrecht, The Netherlands 3AstraZeneca, Alderley Park, Macclesfield, Cheshire, UK 4Department of Pharmacy and Pharmacology, Slotervaart Hospital, Amsterdam, The Netherlands

Background: AZD1152 is a prodrug of AZD1152-hQPA, a potent and selective inhibitor of aurora kinase B (Ki(app)= 0.36 nM). Aurora B is important for mitosis and cytokinesis, and cells bearing P53 mutations are particularly sensitive to aurora B inhibition. AZD1152 induces cellular multinucleation and polyploidy, resulting in apoptosis. AZD1152 potently inhibited the growth of human colon, NSCLC and promyelocytic leukemia tumor xenografts in preclinical studies. This first-in-human study aimed at determining the maximum tolerated dose (MTD), dose limiting toxicities (DLT), and pharmacokinetic profile (PK) of AZD1152 when administered in two different dosing schedules.

inline image

Methods: In part A, AZD1152 was administered as a 2-hour IV infusion once every week. Upon identification of the MTD, part B was initiated at the MTD of part A. In part B, pts received AZD1152 as a 2-hour IV infusion once every 2 weeks. Following identification of the MTD in part B, both schedules were expanded with 10 extra pts in part C. In part C, FDG-PET scans were made at baseline and after 1 cycle of treatment, in order to get more insight into the antitumor activity of AZD1152. All pts provided blood samples for AZD1152 and AZD112 hQPA PK analyses during the first treatment cycle, and toxicity was graded according to CTC-AE version 3.0. Treatment response was evaluated once every 8 weeks using RECIST criteria.

Results: 59 patients were included in this trial. Dose-limiting toxicities (DLTs) in part A included grade (G) 4 neutropenia in 3 out of 6 pts treated at 450 mg, and 2 out of 6 pts treated at 300 mg. The MTD was set at 200 mg. The DLTs in Part B included G4 neutropenia in 3 out of 5 pts treated at 650 mg, and 2 out of 5 pts treated at 550 mg. The MTD was set at 450 mg. PK analyses revealed a rapid conversion to AZD1152-hQPA, and linear pharmacokinetics Best observed response was prolonged stable disease. One patient with SCLC showed an initial response on the FDG-PET scan.

Conclusion: AZD1152 was well tolerated in both schedules. The MTD was set at 200 mg for the weekly schedule and 450 mg when given once every 14 days. The DLT for both schedules was neutropenia. AZD1152 was rapidly converted to the active moiety (AZD1152 hQPA) in plasma, and exposure increased linearly with dose. No objective responses were observed, however long lasting (>1 yr) disease stabilisation was found in a pt with advanced non-small cell lung cancer. Administration of growth factors could reduce the incidence and severity of neutropenia in future trials.

DPYD Single nucleotide polymorphisms (SNPS) in metastatic colorectal cancer (MCRC) patients (PTS) treated with capecitabine, oxaliplatin and targeted agents

Maarten J. Deenen,1 Artur Burylo,1 Jolien Tol,2 Anthonius de Boer,3 Andrew Vincent,1 Henk-Jan Guchelaar,4 Cornelis J. A. Punt,2 Jos H. Beijnen,1,3 Annemieke Cats1 & Jan H. M. Schellens1,3
1Department of Medical Oncology, NKI-AVL, Amsterdam, 2Department of Medical Oncology, UMC Nijmegen, 3Utrecht University, Department of Pharmaceutical Sciences, 4Department of Clinical Pharmacy & Toxicology, Leiden UMC, The Netherlands

Background: Capecitabine is the oral prodrug of 5-FU and is equally effective in the treatment of mCRC. DPYD encodes the DPD protein, which inactivates 5-FU for up to 85%. The aim of this study was to assess the role of DPYD SNPs in the development of severe toxicity in mCRC pts treated with capecitabine-based chemotherapy plus targeted agents.

Methods: Germline DNA was available from 576 previously untreated mCRC pts participating in a phase 3 clinical trial in which pts were randomized between a 3-weekly schedule of capecitabine, oxaliplatin, and bevacizumab or the same regimen with the addition of weekly cetuximab (the CAIRO2 study of the Dutch Colorectal Cancer Group (DCCG)). All 23 exons of DPYD were sequenced in 50 pts who suffered from grade 3/4 capecitabine-related toxicity (i.e. diarrhea, mucositis, leucopenia; NCI-CTC v3) during the first two treatment courses, and in a control cohort of 100 pts randomly chosen from the 576 pts. Exonic SNPs (allele freq. ≥5%) and SNPs that differed significantly in frequency between cases and controls were assessed in the total population of 576 pts and were additionally correlated to overall (OS) and progression-free survival (PFS). In addition, haplotype analysis was performed. Data were statistically analyzed using logistic regression, log-rank test and mixed effect modeling.

Results: 29 SNPs in DPYD were detected in the case and cohort population, of which 11 were assessed in the entire population. The SNPs DPYD*2A, 1236G>A and 2846A>T were significantly associated with capecitabine-related toxicity (OR = 10.4, p = 3x10−7; OR = 3.5, p = 0.003 and OR = 5.1, p = 0.01, resp.). For 496A>G and 3959C>T non-significant associations were observed (OR = 1.7, p = 0.08 and OR = 1.5, p = 0.12). Furthermore, 1236G>A was in high linkage disequilibrium with IVS9–51T>G (D′= 0.91) and with IVS5+18G>A (D′= 0.90). 5 out of 7 patients heterozygous for DPYD*2A or 2846A>T developed grade 3/4 diarrhea (p = 0.01). Furthermore, the average cumulative capecitabine dose per course was lower compared to patients wild type for these SNPs (p < 0.0001 and p = 0.005), with the lowest average dose reduction of 51% and 37%, respectively. For thus far tested SNPs with survival (1601G>A and 1627A>G), no associations with OS or PFS were observed.

Conclusions: These results indicate that DPYD*2A, 1236G>A and 2846A>T are useful predictive markers for toxicity after treatment with capecitabine-based chemotherapy plus targeted agents. A priori dose reductions for pts mutant for DPYD*2A or 2846A>T at start of treatment should be considered.

Influence of antidepressant use on glycemic control in patients with diabetes mellitus: an open-label comparative study

H. J. Derijks,1,2 R. Janknegt,2 E. R. Heerdink,1 G. H. P. De Koning,1,3 M. M. Krekels,4 B. J. Looij4 & A. C. G. Egberts1,5
1Division of Pharmacoepidemiology and Pharmacotherapy, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, The Netherlands; 2Department of Clinical Pharmacy, Orbis Medical Center, Sittard, The Netherlands; 3Association Kring Apotheken The Netherlands, Den Bosch, The Netherlands; 4Department of Internal Medicine, Orbis Medical Center, Sittard, The Netherlands; 5Department of Clinical Pharmacy, University Medical Center Utrecht, Utrecht, The Netherlands

Introduction: Evidence from case reports, animal studies and short term trials suggests that serotonergic antidepressants increase insulin sensitivity, lower glucose levels, decrease HbA1C and decrease insulin requirements. However, evidence on this subject is still scarce. In this open-labeled comparative study we evaluate the change in insulin requirements of four diabetic patients starting with a serotonergic antidepressant compared to eight diabetic patients not using any antidepressant.

Methods: From the outpatients diabetic clinic of a medium-sized teaching hospital in the Netherlands, we identified four diabetic patients who started with a serotonergic antidepressant. We randomly selected two non-users for each starter. All patients were followed for 210 days starting from 30 days prior to the index date. The index date was defined as the start date of the serotonergic antidepressant. Mean relative differences in insulin dose over time (at 0, 30, 60, 120 and 180 days after the index date) were obtained from the patients diabetes diary or the electronic patients record. We also collected the most recent available HbA1C-values before the index date and between 90–180 days after index date.

Results: Mean insulin dose increase in the period from 30 days before the index date to 180 days after index date was 2.4% for the users and 18.3% for the non-users (p = 0.15). Figure 1 shows the standardized mean insulin dose over time for the users and the non-users. At no moment during the follow-up, statistical difference was observed between the standardized mean insulin dose of the users and the non-users. The relative decrease of mean HbA1c-levels during follow up was 7.2% for the users and 0.5% for the non-users (p = 0.37).


Figure 1
Standardized average insulin dose during follow-up. non users (inline image); users (inline image)

Conclusion: This open-label comparative study confirms that serotonergic antidepressants may have insulin-saving effects. Additional research is needed to confirm these results and to establish the clinical relevance of these findings.

Clinical pharmacology education and training at the centre for human drug research

K. L. Franson, J. Burggraaf, A. F. Cohen

Centre for Human Drug Research, Leiden, the Netherlands

Introduction: Centre for Human Drug Research (CHDR) has provided doctoral level clinical research training since it's inception in 1986. In 2000 the CHDR became a Dutch Society of Clinical Pharmacology and Biopharmacy (NVKFB) approved training center for category 3 clinical pharmacologists (non-internists, non-hospital pharmacists). Since that time most CHDR PhD trained scientists have been candidates for the NVKFB approved program.

Aims: Evaluate the perceived relevance of the educational goals for the NVKFB clinical pharmacology program, and identify educational areas CHDR needs to improve.

Table 1. 
Demographics of clinical pharmacology candidates
Educational progressGranted certificate15
Waiting for NVKFB approval1
Still in training18
No longer pursuing certificate1
Previous educationMD23
BioMed/BioPharm sciences7
Current employmentClinical pharmacology unit17
Pharmaceutical company7
Medical or pharmacy practice4
Medical or pharmacy training7

Methods: All CHDR candidates to the NVKFB program had been contacted to determine current job titles. The candidates were emailed a link to an 8 item questionnaire on FreeOnlineSurveys.com®.

Results: Table 1 shows the demographic data for the 35 CHDR candidates to the NVKFB program. 88% of the 24 respondents either agreed or strongly agreed that the clinical pharmacology program was important to their current roles. The specific goals that were perceived as either important or very important are shown in figure 1 as scores ≥3.5.

71% either agreed or strongly agreed that the CHDR was able to meet their educational needs. Topics that 25% perceived needed more attention at the CHDR include pharmacokinetics, interactions, pediatric drug use, and biostatistics.

Conclusions: A majority of the CHDR candidates pursue careers in clinical pharmacology and felt their education was pertinent and met their needs. Some candidates identified areas for improvement. Other programs should use similar methods to assess their educational methods.


Figure 1

Pharmacokinetic-pharmacodynamic modeling of hypertension toxicity in cancer patients treated with the multi-kinase inhibitor E7080

R. J. Keizer,1 J. H. Beijnen,1,4 A. Gupta,2 M. Jansen,2 J. Wanders,2 J. H. M. Schellens,3 M. O. Karlsson5 & A. D. R. Huitema1
1Department of Pharmacy & Pharmacology, the Netherlands Cancer Institute / Slotervaart Hospital, Amsterdam, NL 2Eisai Limited, London, UK 3Division of Clinical Pharmacology, Department of Medical Oncology, The Netherlands Cancer Institute, Amsterdam, NL 4Division of Drug Toxicology, Section of Biomedical Analysis, Department of Pharmaceutical Sciences, Faculty of Science, Utrecht University, Utrecht, NL 5Division of Pharmacometrics, Department of Pharmaceutical Sciences and Pharmacotherapy, Uppsala University, Uppsala, SE

Introduction: The novel anti-angiogenic multiple tyrosine kinases inhibitor E7080 has shown anti-tumor activity in preclinical testing[1,2] and phase I clinical trials[3,4,5]. Although generally well tolerated in clinical studies, one of the observed toxicities was hypertension. We aimed to construct a pharmacokinetic-pharmacodynamic (PK-PD) model for this toxicity, to aid in further clinical development of E7080.

Methods: PK and blood pressure (BP) data were obtained from 67 patients enrolled in a two-center phase I study, investigating qd dosing of E7080 at increasing dose levels. PK and BP data were modeled sequentially, using NONMEM software. The PK model was developed previously, and consisted of a two-compartment model, combined zero- and first-order absorption, and linear elimination. BP data were recorded weekly, over a mean period of 21 weeks (range 1–77 weeks). Prescription of anti-hypertensive (AH) medication was accounted for in the model using the ratio of daily dose/defined daily dose, cumulative over all AH medications, as a negative effect on BP. Several models for BP, baseline BP, and drug effect on BP were evaluated.

Results: The final BP model consisted of two separate indirect-effect models for systolic and diastolic BP. Baseline BPsys and BPdia were estimated at 127 and 77 mmHg (RSE <5%) respectively, with 30% inter-individual variability (IIV) and 66% correlation between systolic and diastolic baseline. The effect of E7080 plasma concentration on the input rates of the indirect effect PD model were 0.77 (RSE 5%) and 1.3 (RSE %) ng/ml.hour−1 for systolic and diastolic BP, leading to a similar absolute effect size (in mmHg). Estimated IIV on drug effect was 72%, (RSE 23%). The overall exponential residual error in BP was estimated to be 9% (RSE %), with 23% correlation between systolic/diastolic errors. Continuous dosing of E7080 at the maximum tolerable dose of 25 mg qd without AH medication, would result in an increase of 13.3 / 12.3 mmHg (systolic/diastolic) for a typical patient, with half of the increase attained after 2.5 weeks. At this dose level, a steady state BP higher than 140/90 would be expected in 41% of all patients.

Conclusion: A PK-PD model was developed that was able to capture the effect of daily treatment with E7080 on BP, which showed a clear exposure-response relation. We were able to account for the use of anti-hypertensive medication in the model. Through simulation studies, the current model may aid in dosing schedule optimization, and design of future clinical trials of E7080.

1 Matsui J et al, Int J Cancer. 122, no. 3 (2008): 664–71.

2 Matsui J et al, Clin Cancer Res 14, no. 17 (2008): 5459–5465.

3 Yamada K et al, J Clin Oncol (Meeting Abstr) 26, no. 15_suppl (2008): 3527.

4 Nemunaitis JJ et al, J Clin Oncol (Meeting Abstr) 26, no. 15_suppl (2008): 14583.

5 Glen H et al, J Clin Oncol (Meeting Abstr) 26, no. 15_suppl (2008): 3526.

Patterns of asthma medication use in the first eight years of life

E. S. Koster,1 A. H. Wijga,2 M. G. P. Zuidgeest,1 J. A. M. Raaijmakers,1 D. S. Postma,3 G. H. Koppelman,3 J. C. de Jongste,4 H. A. Smit2 & A. H. Maitland-van der Zee1
1Division of Pharmacoepidemiology & Pharmacotherapy, UIPS, Utrecht University, the Netherlands 2Center for Prevention and Health Services Research, National Institute for Public Health and the Environment, Bilthoven, The Netherlands 3Department of (Pediatric) Pulmonology, University Medical Center Groningen 4Division of Pediatric Respiratory Medicine, Erasmus University Medical Center Rotterdam, The Netherlands

Introduction: Anti-asthma drugs are the most widely chronically used drugs in childhood. Low persistent use, non-adherence and large differences in use of treatments have been described [1–3]. This study describes changes in patterns of anti-asthma drug use in the first eight years of life.

Methods: Complete medication histories from birth up to age 8 were available for 777 children participating in the PIAMA birth cohort study. Patterns of medication use were assessed in 4 treatment groups: start before third birthday with short-acting B-agonists (SABA) monotherapy ('SABA’ group), inhaled corticosteroid (ICS) monotherapy (‘ICS’ group), SABA and ICS therapy ('SABA+ICS’ group) or none of these treatments (‘no-use’ group). Continuous use was defined as prescriptions continuing up to age 8 years.

Results: Most children (79%) were classified as ‘no-users’, 8% fitted to the ‘SABA+ICS’, 10% to ‘SABA’ and 2% to the ‘ICS’ group. Medication use and prevalence of respiratory symptoms were highest in the ‘SABA+ICS’ group. There were only few continuous users: 6% in the ‘SABA’, 5% in the ‘ICS’ and 9% in the ‘SABA+ICS’ group. Prevalences of different use patterns for the 4 groups are shown in the figure. 42% of the ‘SABA+ICS’ group received trial medication (prescription was filled in 1 or 2 subsequent years), 22% used long-term SABA+ICS (filled prescriptions in 3 subsequent years or more) and 16% switched to monotherapy.


Figure 1
Patterns of medication use in the first 8 years of life for the different treatment groups

Conclusion: Children who started on SABA and ICS therapy seem to have more severe symptoms. But even in this group, not many children use medication continuously during the first 8 years of life. Many received therapy as a trial for relief of respiratory symptoms while others discontinued or switched to monotherapy. This observation is consistent with transient patterns of respiratory symptoms and it should be noted that most of these children are not diagnosed with asthma.

1 Menckeberg TT, Belitser SV, Bouvy ML et al. Distingoushing patterns in the dynamics of long-term medication use by Markov analysis: beyond persistence. BMC Health Serv Res. 7(1): 106. (2007).

2 Breekveldt-Postma NS, Gerrits CM, Lammers JW, Raaijmakers JA, Herings RM. Persistence with inhaled corticosteroid therapy in daily practice. Respir Med. 98(8): 752–9. (2004).

3 Zuidgeest MG, Smit HA, Bracke M et al. Persistence of asthma medication use in preschool children. Respir Med. (2008).

Cocaine and amphetamine use in patients with acute chest pain and suspected acute coronary syndrome in an amsterdam hospital

Kromdijk Wiete,1 Bogaard Kjell,b\2 Somsen G. Aernout2 & Eric J. F. Franssen1
1Department of Clinical Pharmacy, 2Department of Cardiology, Onze Lieve Vrouwe Gasthuis, Amsterdam, The Netherlands

Introduction: Cocaine and amphetamine are frequently used recreational drugs. They can cause cardiovascular complications because of their sympathicomimetic effect; myocardial oxygen demand increases due to an increased heart rate and blood pressure and oxygen supply decreases due to vasoconstriction of smooth muscle cells (McCord, 2008). These cardiovascular complications of cocaine have been extensively studied (Weber, 2003). With this study we aim to: (i) determine the prevalence of cocaine and amphetamine use among patients admitted to the emergency department (ED) with acute chest pain and suspected acute coronary syndrome (ACS) in an Amsterdam hospital, (ii) investigate differences in characteristics of the cocaine and amphetamine positive patients compared to cocaine and amphetamine negative patients and (iii) investigate differences in final outcome.

Methods: We retrospectively evaluated patients admitted to the ED unit of the OLVG from January 2002 to December 2006 subjected to toxicology screening for cocaine and amphetamine. Toxicology screens had been requested by one cardiologist on patients ≤55 years old presenting to the ED with acute chest pain and suspected ACS. Patient demographics and clinical data were reviewed using the electronic patient file (X-care) and, if necessary, medical records. Risk factors for ACS and medical history were based on patient history and medical review. Statistical analysis was performed using SPSS (version 16.0, SPSS Inc.,Chicago).

Results: A total of 216 patients were included in the study. From these patients 14% were tested positive for cocaine and/or amphetamine. The predominant final diagnose was MI in both groups; 45% in the group tested positive for cocaine and amphetamine and 57% in the negative group. Patients tested positive on one of the substances had less often an acute coronary syndrome (55% vs 72%; p < 0.05) and were less likely to have hypertension (7% vs 25%; p < 0.05) or a family history of coronary arterial disease (28% vs 46%; p < 0.05).

Discussion: To our knowledge, this is the first study showing that patients admitted to the ED with acute chest pain and suspected ACS with a positive toxicology screen had less often an ACS than patients with a negative toxicology screen. This can be explained by the fact that cocaine is known for its non-cardiac chest pain. Therefore, it can be anticipated that patients with a positive toxicology screen have a better prognosis. The amount of patients tested positive for cocaine or amphetamine in our study is comparable with other studies (McCord, 2008). One out of 7 patients with suspected ACS have used cocaine and/or amphetamines.

Conclusion: In conclusion, cocaine and amphetamine users with suspected ACS in an Amsterdam study population are at lower risk for death and cardiovascular complications than non-cocaine and amphetamine users.

McCord J et al. Management of cocaine associated chest pain and myocardial infarction: a scientific statement from the American heart association acute cardiac care committee of the council on clinical cardiology. Circulation 2008; 117: 1897–1907.

Weber JE et al. Validation of a brief observation period for patients with cocaine-associated chest pain. N Engl J Med 2003; 348: 510–517.

Trends in antibiotic use in a neonatal intensive care unit over a 16-year period

T. B. Y. Liem,1 A. Van den Hoogen,2 C. M. A. Rademaker,1 A. Fleer,3 L. J. Gerards,2 A. C. G. Egberts1 & T. G. Krediet2
1Dept of Clinical Pharmacy, 2Dept of Neonatology, 3Eijkman-Winkler Center for Microbiology, Infectious Diseases and Inflammation, University Medical Center Utrecht, Wilhelmina Children's Hospital, Utrecht, The Netherlands

Introduction: Antibiotics are among the most frequently used drugs in NICUs.1 To prevent antibiotic resistance, a major threat in NICUs, a strict antibiotic policy is necessary. Multidisciplinary infectious disease teams (IDT) have shown to be effective in the quality of antibiotic use.2 We studied the antibiotic use in our NICU over a period of 16 years where an IDT was active since 1990.

Methods: From all infants admitted to the NICU from 1990–2006 during 2-year time intervals gender, birth weight (BW), gestational age (GA), length of stay (LOS), mortality and clinical data (proven sepsis, presence of central venous catheters [CVCs], total parenteral nutrition [TPN] and mechanical ventilation) were studied in 3 periods (period I:1990,1992,1994;II:1996,1998,2000;III:2002,2004,2006). Antibiotic use was registered from infants admitted during the first 3 months of each two-year interval: duration of treatment with specific antimicrobial agents and of any antimicrobial agent use. Antibiotics are prescribed according to a strict protocol.

Results: Gender (male: period I and II 56%, III 58%), BW < 1500 g (period I 39%, II 36%, III 34%) and GA <32 wks (period I 43%, II 38%, III 40%) were not different between the 3 periods neither were differences noted in use of CVCs, TPN or mechanical ventilation. Mortality decreased from 17% (period I) to 10% (period III). Mean LOS decreased from 20 days in period I to 16 days in period III. The total sepsis incidence during 3 periods was 14%, 17% and 16% respectively, during periods I, II and III. Antibiotic use was consistently high: 85%, 90%, 86% of patients received any antibiotic during period I, II and III respectively. However, a significant decrease in duration of antibiotic use was noted for the three most frequently used antimicrobial agents (amoxicillin- clavulanic acid, p = 0.011; 95% CI −0.197; −0.036; aminoglycosides p = 0.003; 95% CI −0.297; −0.09 and cephalothin/cefazolin p = 0.002; 95% CI −0.134; −0.043). The use of vancomycin was limited (0.1–1.0 days/admission) and did not show significant difference during the years.

Conclusions: The percentage of infants treated with antibiotics remained high, whereas the duration of treatment with amoxicillin-clavulanic acid, aminoglycosides and cephalotin/cefazolin decreased significantly over time, without increase in incidence of sepsis or mortality. Guidelines for antibiotic therapy for neonatal sepsis were unchanged during 16 years. The IDT may have played a key role. Correct identification of infants with sepsis remains a major challenge in attempts to further reduce antibiotic use and postpone the emergence of antibiotic resistant microorganisms.

1 Warrier I et al. J Clin Pharmacol 2006; 46(4): 449–455.

2 Metjian TA et al. Pediatr Infect Dis J 2008; 27(2): 106–111.

Effects of methimazole on the pharmacokinetics of irinotecan chemotherapy

Ron H. Mathijssen,1 Jessica M. van der Bol,1 Theo J. Visser,1 Floris A. de Jong,1 Maarten O. van Aken,1 Andre S. Planting,1 Jan H. Schellens2 & Jaap Verweij1
1Erasmus MC – Rotterdam, the Netherlands, 2NKI – Amsterdam, The Netherlands

Introduction: A 53-year old male patient treated for locally advanced colorectal cancer with single agent irinotecan (350 mg/m2, every 3 weeks) received methimazole co-medication for Graves’ disease. We hypothesized that methimazole could seriously affect the complex pharmacokinetics of irinotecan (Figure 1), i.e. by inhibition of CYP3A and/or induction of UGT1A (Guo et al., 1997).

Methods: Irinotecan pharmacokinetics and side-effects were followed during a total of 4 courses. These included 2 courses with methimazole, followed by 2 without methimazole. In between, radioiodine treatment was given as definite Graves’ treatment.

Results: Area under the curves of the active irinotecan metabolite SN-38 and the inactive metabolite SN-38G were both higher (a mean increase of 14% and 68%, respectively) with methimazole co-medication, compared with irinotecan without methimazole (Table 1). Concentrations of irinotecan, NPC, and APC did not change. The SN-38 glucuronidation rate increased 47% during concurrent treatment. Possible confounding factors (like co-medication use, and smoking behaviour) did not change over time. Specific adverse events due to methimazole co-treatment were not seen.

Conclusion: These results suggest that methimazole treatment affects the (efflux-) transport of SN-38G and/or the metabolism of irinotecan. This leads to higher systemic metabolite drug levels, which could potentially result in increased toxicity. Currently, in vitro experiments are performed to confirm our clinical data. As yet, physicians should be careful in dosing irinotecan in patients co-treated with methimazole.

Table 1. 
Baseline characteristics, and pharmacokinetics during four irinotecan courses
TSH (mU/L)
T3 (nmol/L)
Free T4 (pmol/L)15.510.529.1
AUC (ngxh/mL)   
Glucuronidation rate11.99.67.3

Figure 1
Metabolism of irinotecan

Guo Z et al. Drug Metab Dispos 1997; 25: 390–3.

No evidence for increased activity of the l-arginine/no pathway in migraine patients

B. J. Van der Schueren, R. Verbesselt, A. Van Hecken M. Depré, F. H. Verbrugge & J. N. de Hoon
Center for Clinical Pharmacology, University Hospital Gasthuisberg (K.U. Leuven), Leuven, Belgium

Introduction: Olesen et al. proposed that nitric oxide (NO) plays a crucial role in the pathophysiology of migraine headache. As numerous studies suggest an increased NO production in migraineurs, both ictally and interictally, an increased activity of the L-arginine/NO pathway would be expected.

Aims: To assess whether migraine patients display a chronic nitric oxide synthase (NOS) hyperactivity by comparing the NO production, and biomarkers characteristic for the activity of the L-arginine/NO pathway, before and following a loading dose of L-arginine between migraine patients (interictally) and matched healthy control subjects.

Methods: Twenty healthy volunteers and 20 matched migraine patients participated in a 2-period, randomized, double-blind, placebo-controlled study. Each subject received a 30-min infusion, by peripheral vein, of 30 g L-arginine hydrochloride or placebo (i.e. an equal volume of 0.9% NaCl). In each study period the following biomarkers were assessed just before and for 6 hours following the start of the infusion: exhaled/nasal NO, plasma citrulline and L-arginine and urinary excretion of nitrite/nitrate and cGMP. Data are presented as mean (95% confidence interval).

Results: At baseline, both exhaled (eNO) and nasal NO (nNO) were higher in migraineurs compared to healthy volunteers: 15.9 (8.8, 23.0) ppb compared to 10.8 (7.0, 14.5) ppb for eNO and 76.3 (61.2, 91.4) ppb compared to 61.6 (51.2, 72.0) ppb for nNO, respectively (P = 0.04 and P = 0.03, two sample Student's t-test, respectively). The AUC0–6 for the absolute value of eNO and nNO in ppb did not differ between migraine patients and healthy volunteers following L-arginine or saline infusion. Plasma L-citrulline levels at baseline did not differ averaging 26 (23, 29) and 29 (26, 32) µmol/L for migraine patients and healthy volunteers, respectively. The increase in L-citrulline following L-arginine infusion was slightly lower in migraine patients (15, 13–18 µmol/L) compared to healthy volunteers (19, 16–23 µmol/L; P = 0.046, two sample Student's t-test). Urinary nitrate and cGMP excretion did not differ between migraine patients and healthy volunteers during the placebo period. Interestingly, whereas no significant increase in urinary nitrate and cGMP excretion could be detected in migraine patients between L-arginine and placebo infusion, there was an increase of both biomarkers in healthy volunteers: from 92.07 (66.33, 117.82) to 132.63 (100.24, 165.02) µmol mmol−1 creatinine for nitrate and 39.64 (33.94, 45.34) to 50.53 (42.19, 58.87) nmol mmol−1 creatinine for cGMP, (P = 0.014 and 0.0003, respectively; paired Student's t-test on ln-transformed data).

Conclusion: These data do not support the idea of a generalized increase in NO synthase activity in migraine patients outside of a migraine attack.

Investigation of the effect of macitentan on the pharmacokinetics and pharmacodynamics of warfarin in healthy male subjects

P. N. Sidharta,1 H. Dietrich2 & J. Dingemanse1
1Actelion Pharmaceuticals Ltd, Allschwil, Switzerland; 2ClinPharmCologne – MEDA Manufacturing GmbH, Cologne, Germany

Macitentan is a new, orally active, potent dual endothelin (ET) ETA and ETB receptor antagonist (ERA). In early Phase I trials, macitentan was well tolerated after single and multiple daily oral doses of up to and including 300 mg and 30 mg, respectively, with the most frequent adverse event (AE) being headache.

In this study we investigated the effect of multiple-dose macitentan on the pharmacokinetics (PK) and pharmacodynamics (PD) of a single dose of warfarin. The study was performed in a single-center, open-label, randomized, two-way crossover design. After an eligibility check 14 subjects received one of the two treatment sequences (i.e., A/B or B/A) of the following treatments separated by a washout of at least 2 weeks:

Treatment A: A loading dose of 30 mg macitentan on Day 1. Thereafter 10 mg macitentan q.d. for 8 days. On Day 4, a single dose of 25 mg warfarin was administered concomitantly with macitentan. The observation period was 9 days following the loading dose of macitentan.

Treatment B: A single dose of 25 mg warfarin on Day 1. The observation period was 6 days following drug administration.

During each observation period blood samples for warfarin PK (R- and S-warfarin) and PD (INR and Factor VII) were taken and safety was assessed. In addition, during Treatment A, plasma trough levels of macitentan and its active metabolite, ACT-132577, were determined.

Fourteen healthy male subjects signed the informed consent and were included in this study. During the course of the study, one subject withdrew consent and another subject incorrectly took study drug, resulting in unreliable PK and PD data. Hence, 12 subjects were included in the PK/PD analysis and 14 in the analysis of safety. The plasma concentration-time profiles of R- and S-warfarin (Figure 1) were comparable between administration of macitentan plus warfarin and warfarin alone. The geometric mean ratios and 90% CI for Cmax and AUC0–∞ for R- and S-warfarin were within the bioequivalence range of 0.80 to 1.25.

The PD parameters of INR and Factor VII, i.e., median tmax, mean baseline values for INR and Factor VII activity, and INRmax and VIImax were also comparable between treatments. No impact of warfarin was observed on the trough levels of macitentan and ACT-132577. Both of the study treatments were well tolerated with headache reported as the most frequently occurring AE.


Figure 1
Plasma concentration-time profile of R- and S-warfarin in the absence and presence of macitentan (mean ± SD, n = 12). macitentan + warfarin (inline image); warfarin alone (inline image).

In conclusion, based on these study results, no dose correction of macitentan or warfarin is needed when using these drugs concomitantly.

Impdh type i and ii gene expression correlated to impdh activity and clinical outcomes in MMF treated kidney transplant patients

F. Sombogaard,1 A. M. A. Peeters,2 C. C. Baan,2 R. A. A. Mathot,1 M. E. Quaedackers,2 W. Weimar2 & T. van Gelder1,2
1Dept of Hospital Pharmacy, 2Dept of Nephrology, Erasmus MC University Medical Center, Rotterdam, The Netherlands

Introduction: Mycophenolic acid (MPA) – the active metabolite of MMF– inhibits the enzyme inosine monophosphate dehydrogenase (IMPDH). IMPDH type I is expressed constitutively while type II is expressed after immune-activation. To optimize MMF therapy after kidney transplantation (Tx) monitoring IMPDH activity could be useful. However, IMPDH activity assays are time-consuming and labor-intensive. An alternative method might be measuring IMPDH mRNA. In this study IMPDH activity is correlated to IMPDH mRNA expression in MMF treated kidney Tx patients.

Methods: A cohort of 77 patients was prospectively monitored for pharmacokinetics (MPA levels), pharmacodynamics (IMPDH activity) and IMPDH mRNA during MMF treatment with tacrolimus and steroids. Blood samples were taken pre-Tx, on day 6 post-Tx pre-dose, 0.5, 1, 2, 6 and 12 h after MMF intake and on day 21, 49 and 140 post-Tx pre-dose, 0.5 and 2 h after MMF intake. The enzyme activity of IMPDH in PBMC was measured using a validated HPLC method, IMPDH mRNA was measured using a quantitative real-time polymerase chain reaction.

Results: Pre-dose IMPDH type I and II mRNA expression levels increased over time (p = 0.061; p = 0.004 resp.), while IMPDH activity pre-dose did not alter (p = 0.49). Correlations between IMPDH mRNA and activity were weak (r2 =−0.28 to 0.058) and none of the correlations were significant (p = 0.11 to 0.77). No significant correlations were found between IMPDH mRNA and total or unbound trough MPA levels (r2=−0.30 to 0.10; p = 0.073 to 0.95). On day 6 the ratio between IMPDH mRNA type II and I was significantly lower in patients with a biopsy proven acute rejection (n = 10) within 30 days post-Tx compared to patients without a BPAR (n = 28; 0.55 ± 0.19 vs. 0.84 ± 0.60; p = 0.031 resp.). At visit 4, patients with lower IMPDH type I and II mRNA levels had significant more hematological adverse events (8.1 (2.6–25) vs. 24 (12–46) x103 copies; p = 0.012; 6.9 (1.9–25) vs. 22 (11–45) x103 copies; p = 0.027 resp.). No other significant correlations were found for hematological and gastrointestinal adverse events.

inline image

Conclusion: This study shows that the dynamics of IMPDH mRNA levels do not parallel the IMPDH activity in MMF treated patients. Possibly, gene expression is induced following MPA mediated inhibition of IMPDH activity. Though, the ratio IMPDH mRNA type II/I is significantly lower in patients who have a BPAR within 30 days post-Tx, the clinical utility of monitoring IMPDH activity appears to be superior.

High number of discrepancies between usual and structured history taking of medication use

J. Spee,1 R. J. van Marum,1 A. C. G. Egberts,2 A. C. van Maanen1 & P. A. F. Jansen1
1Department of Geriatrics UMC Utrecht; 2Department of Clinical Pharmacy UMC Utrecht and Department of Pharmacoepidemiology and Pharmacotherapy, Utrecht University, The Netherlands

Introduction: An accurate medication use history at admission to the hospital is of importance to prevent drug-related problems. In about half of the patients an erroneous history of medication use led to at least 1 unintended discrepancy.(1) We have developed a structured history taking of medication use (SHIM) and studied the discrepancies compared to the usual history taking of medication use.

Method: SHIM was compiled consisting of 16 questions related to use of prescription medication, over-the-counter medication (OTC), drug omission, dose, dose frequency, dosage form and medication use experiences (e.g. adverse effects). Between September 2008 and November 2008 all consecutive patients admitted at the department of Geriatrics were invited to participate, with a family member, in the SHIM. They were asked to bring along all the medication they used at home. The SHIM was taken by one of us (JS). The number and type of discrepancies were noted between the SHIM and the medication use at admission as reported in the patient record by the resident and the requested community pharmacy list (usual history-taking). Two geriatricians-clinical pharmacologists (RM and PJ) scored, independently of each other, the discrepancies on clinical relevance according to the method of Cornish et al.(1)

Results: 47 patients, with a mean age of 84 years and mean use of 9.7 drugs, were studied. 37 patients used OTC drugs.

70% of the discrepancies were considered clinically relevant, 24% with the potential to result in severe discomfort or severe clinical deterioration. The number of discrepancies increased significantly with the use of 7 or more medications. The time to take the history of medication use was at mean 12 minutes.

Table 1. 
Number and type of discrepancies at admission of prescribed drugs without and with OTC
SHIM vs medication in patient record
Number of patients with at least one discrepancy4547
Mean number of discrepancies per patient (SD)2.7 (2.2)4.2 (3.0)
Total number of discrepancies130195
Unknown use1215
Incorrect / omitted dose or frequency5466
SHIM vs community pharmacy list  
Number of patients with at least one discrepancy3846
Mean number of discrepancies per patient (SD)2.0 (2.3)3.3 (3.0)
Total number of discrepancies93156
Unknown use1518
Incorrect / omitted dose or frequency4558

Conclusion: This study shows a high prevalence of clinically relevant discrepancies, indicating an additional value of the structured history taking of medication use.

1 Cornish et al. Unintended medication discrepancies at the time of hospital admission. Arch Int Med 2005; 165: 424–9.

Rosiglitazone reduced ischemia-reperfusion–injury measured with annexin A5 scintigraphy.
A randomized, double blind, placebo-controlled cross-over study in subjects with the metabolic syndrome

Dominique J. P. van Uden,1 Alexander J. M. Rennings,1 Patrick Meijer,1,2 Cees J. Tack,3 Paul Smits,1,3 Otto C. Boerman,4 Wim J. G. Oyen4 & Gerard A. Rongen1,3
1Radboud University Nijmegen Medical Centre, departments of Pharmacology-Toxicology, 2Anesthesiology, 3Internal Medicine and 4Nuclear Medicine, Nijmegen, The Netherlands

Introduction: Cardiovascular disease is the leading cause of death in diabetic patients due to both a high event rate and a worse outcome. A pharmacological intervention that reduces ischemia-reperfusion(I/R) injury would improve the outcome of diabetic patients after a cardiovascular event. The thiazolidinedione derivatives are peroxisome proliferator-activated receptor-γ (PPARγ) ligands that are approved for the treatment of hyperglycaemia in type 2 diabetes mellitus. Animal data suggest that PPARγ ligands can protect against I/R injury by improving insulin responsiveness. A meta-analysis of clinical trials with rosiglitazone suggests that this drug increases the incidence of myocardial infarction. However, no human data on a possible benefit on infarct size are available. Recently, our group developed a human in vivo model to quantify ischemia-reperfusion-injury. In this model annexin A5 scintigraphy is used to visualize early and reversible cellular membrane changes that occur in the forearm skeletal muscle vascular bed after ischemic exercise. In the present study, we will use this approach to address the following hypothesis: Rosiglitazone reduces ischemia-reperfusion-injury in humans with insulin resistance, selected by using the criteria for the metabolic syndrome.

Methods: This is a single-centre randomized, double blind, placebo-controlled crossover study with a washout period of 6 weeks comparing eight week treatment with rosiglitazone (RSG) 4 mg bd with placebo (PBO). We included 10 evaluable subjects with the metabolic syndrome, 6 on concomitant statin therapy. At the end of each treatment period and after 24 hours of caffeine abstinence, volunteers were subjected to 10 minutes of forearm ischemia, combined with standardized intermittent handgripping. At reperfusion, 500 MBq 99mTc-Annexin A5 was iv. administered in the control arm. Annexin A5 uptake in the thenar muscle was quantified 1 and 4 hours post-reperfusion using a gamma camera and expressed as % difference between the experimental and control hand (annexin A5 targeting) which quantifies ischemia-reperfusion injury.

Results: Rosiglitazone reduced annexin targeting from 8.4% (median; range 0.6–49%) to 4.7% (0.7–20%) at one hour after reperfusion (P= 0.037) and from 7.4% (0.5–50%) to 4.8% (-0.1–13%) at four hours of reperfusion (P= 0.06). As expected rosiglitazone decreased hematocrite compared to placebo (RSG: −0.02 ± 0.01; PBO: 0.001 ± 0.02; P= 0.03). In addition, rosiglitazone tended to decrease diastolic blood pressure, fasting glucose and waist circumference (P > 0.05). There was no correlation between changes in glycaemic regulation and the changes in annexin targeting.

Conclusion: Rosiglitazone reduces I/R injury in a population with the metabolic syndrome.

This study was supported by GlaxoSmithKline with an unrestricted grant.

Medical students’ performance of therapeutic consultations with real patients during their internal medicine clerkship

R. J. van Unen, J. Tichelaar, A. J. Schneider, P. Nanayakkara, M. C. Richir & T. P. G. M. de Vries
VU University Medical Center, Amsterdam, The Netherlands

Introduction: An important task of clinicians and clinical pharmacologists is to train medical students to prescribe drugs rationally. As shown recently, preclinical pharmacotherapy context learning based on the WHO 6-step approach (Vries de et al., 1994) improves the prescribing behaviour of undergraduate medical students (Richir et al., 2008). However the clinical phase of the current medical curriculum emphasizes the acquisition of diagnostic skills, with relatively little attention being paid to the acquisition of therapeutic skills. For this reason, a pilot pharmacotherapy training programme has been introduced during the internal medicine clerkship at the VU University Medical Center. The aims of this first evaluation were to determine to what extent medical students are capable of carrying out a therapeutic consultation with real patients and to determine whether such a pharmacotherapy training programme is feasible during the internal medicine clerkship.

Methods: From March 2008 to December 2008, 25 medical students were trained to carry out a therapeutic consultation at the outpatient clinic. Individual consultations were observed and assessed by clinicians using the WHO 6-step plan. The student's understanding of the patient's problem, therapeutic knowledge and the student's ability to prepare and carry out the therapeutic consultation were also evaluated. Both students and clinicians also completed a questionnaire about the value and feasibility of the pharmacotherapy programme.

Results: The mean scores (% of maximum score) of medical students for the various WHO 6-step competencies assessed are given in Table 1.

Table 1. 
Mean ± S.D. scores for the WHO 6-step competencies of a therapeutic consultation (n = 25)
Step 1: Defining the patient's problem (‘diagnosis’)84% ± 10%
Step 2: Specifying the therapeutic goals80% ± 16%
Step 3: Use a reliable Standard Treatment (ST)78% ± 16%
Step 4: Verify the suitability of the ST for the patient84% ± 12%
Step 5: Start the treatment and give patient information82% ± 16%
Step 6: Determine monitoring measurements82% ± 12%

Of the students assessed, 79% were considered to have ‘good’ or ‘very good’ insight into patients’ problems and therapeutic knowledge. Moreover, 84% and 79% were considered ‘good’ or very good’ in preparing for and carrying out the therapeutic consultation, respectively. Most clinicians (79%) and students (95%) were ‘neutral’ to ‘very contented’ about the additional work involved. Overall, 73% of the students and 74% of the clinicians ‘agreed’ or ‘strongly agreed’ with the necessity of such a training programme.

Conclusions: The results showed that medical students were capable of carrying out a therapeutic consultation with real patients during their internal medicine clerkship after following a pharmacotherapy training programme. Both clinicians and students were highly appreciative of the programme and considered the extra work and effort involved very worthwhile.

Richir et al. Clin. Pharmacol. Ther. 2008.

Vries de et al. Guide to Good Prescribing. Geneva: WHO, 1994.

Quality of treatment in patients with type 2 diabetes: 2007 monitoring results from giantt

J. Voorham, P. Denig, D. de Zeeuw, F. M. Haaijer-Ruskamp, on behalf of the GIANTT-group.
Department of Clinical Pharmacology, Graduate School for Health Research SHARE, University Medical Center Groningen, The Netherlands

Background: In the Groningen region, the majority of general practitioners (GPs) will join a diabetes management program in 2008/2009, which will include annual benchmarking and quality control. A baseline evaluation was conducted to assess the quality of treatment in patients with type 2 diabetes mellitus (T2DM) in 2007.

Methods: Software has been developed for extracting data from free text and structured tables in electronic patient records (EPR) of general practitioners.1 Treatment and clinical parameters extracted include (management of) HbA1c, blood pressure, lipid spectrum, albuminuria.

Criteria for practice inclusion were: intention to participate in regional program (85% of practices), EPR system with validated extraction software (70% of all practices).

Criteria for patient inclusion were: diagnosed with T2DM at beginning of observation period, complete follow-up, managed by GP (86% of T2DM patients), did not opt-out (99.7% of all patients).

Quality of treatment was assessed in relation to clinical outcomes using a cross-sectional approach, i.e. numbers of patients with elevated risk factor levels being prescribed (specific) treatment, and a prospective approach, i.e. numbers of patients with preceding elevated risk factor levels whose treatment was unchanged, started or intensified within 6 months (when possible).

Results: Data were available for 8536 T2DM patients (120 GPs, 82 practices, covering a total population of over 287,000 people); 52% female, mean age (±sd) was 67 ± 12 years. HbA1c was 6.9 ± 1%; 6% of the patients had poor glycemic control (HbA1c >8.5), and 61% good control (HbA1c <7%). Of the poorly controlled patients, 16% were treated with 1 oral drug, 25% with >1 oral drug, 53% with insulin. In patients with suboptimal control (HbA1c >7%), treatment was not started or intensified in 74% of the cases, which amounts to 25% of all patients. Systolic blood pressure (SBP) was 142 ± 20 mmHg; 13% of the patients had poor control (>160 mmHg), and 44% good control (<140 mmHg). Of the poorly controlled patients, 13% were not treated with medication, whereas 41% received 3 or more different drugs. No treatment start or intensification was seen in 88% of patients with a SBP of 140–160, and 81% of patients with a SBP >160, which amounts to 47% of all patients. Mean LDL-c was 2.3 ± 0.9 mmol/l; 61% of patients had LDL-c <2.5 mmol/l. Of the patients with LDL-c >2.5 mmol/l, 51% received lipid-lowering medication. Almost 94% of patients with an LDL-c >2.5 mmol/l had no treatment start or intensification, which amounts to 32% of all patients. Finally, 34% of patients with (micro) albuminuria were not prescribed a RAS-inhibitor.

Discussion: In 25%–47% of all T2DM patients, no treatment intensification was observed despite suboptimal control of a risk factor, indicating potential undertreatment. This could be partially explained by problems with tolerability or patient compliance.

1 Voorham J, et.al. J Am Med Inform Assoc 2007; 14: 349–54.

Therapeutic drug monitoring for furosemide in neonates on ECMO

M. M. J. van der Vorst,1 A. F. J. van Heijst,2 J. den Hartigh3 & J. Burggraaf4
1Department of Pediatrics, St. Antonius Hospital, Nieuwegein, 2Department of Neonatalogy, University Medical Center Nijmegen, 3Department of Clinical Pharmacy and Toxicology, Leiden University Medical Center, 4Centre for Human Drug Research, Leiden, The Netherlands

Introduction: Neonates treated with extracorporeal membrane oxygenation (ECMO) suffer from volume overload. Furosemide (FURO) is used to limit the detrimental effects of excess volume with a target urine production of 6 ml/kg.hr. FURO can be given as repeated intravenous (IV) boluses or as continuous IV infusion.(Van der Vorst 2006; Van der Vorst 2007) A potential drawback of bolus administrations is that high serum concentrations of FURO are reached that may result in excessive urine output and, when concentrations exceed 50 µg/ml, may become ototoxic.(Rybak, 1985) At present no data are available on serum concentrations of FURO in neonates on ECMO after repeated IV boluses and hearing loss is occasionally observed after ECMO treatment.(Morini, 2008) Therefore, therapeutic drug monitoring (TDM) for serum FURO in neonates on ECMO was introduced.

Methods: The data were collected in 8 neonates admitted to a neonatal intensive care unit for ECMO therapy. At birth the median gestational age was 40 (34.7–41.6) weeks and the median weight was 3192 (2100–4300) gram. The median postnatal age at start of FURO therapy was 6.82 (2.15–17.74) days. All patients were cardiovascular stable and had a normal renal function for age at the start of the FURO therapy.(Rudd, 1983)

The patients received 4–8 FURO bolus doses (1 mg/kg) daily according to the standard treatment protocol. TDM started at the first day of FURO therapy and was continued for 3 days, if applicable. It was at the discretion of the physician at which time the samples were taken, but particular attention was given to ensure sampling shortly (∼10 min) after the drug was given.

Results: On the consecutive study days the median (range) FURO dose was 3.86 (2.00–4.22), 2.80 (1.13–8.00) and 2.00 (1.86–6.33) mg/kg.24 hrs. Maximum serum FURO concentrations ranged between 6.0 and 26.9 µg/ml. The median (range) urine production at day 1, 2 and 3 was 8.94 (2.72–14.56), 9.31 (2.62–11.85) and 8.76 (5.87–13.49) ml/kg.hr respectively.

Discussion and conclusion: The data indicate that after repeated IV boluses of FURO in neonates on ECMO, serum concentrations remain below the presumed ototoxic level. This makes it unlikely that FURO is the cause of hearing loss after ECMO treatment in neonates, although it cannot be excluded that the relatively high FURO concentrations and other potentially ototoxic medications interact. Diuresis tended to be less controlled when compared to data previously presented on continuous IV infusion of FURO in a comparable patient population. (Van der Vorst, 2007) The data suggest that continuous FURO infusion is the preferred treatment modality to reduce volume overload in neonates on ECMO.

Morini F. J Pediatr Surg 2008; 43: 380–4.

Rudd PT. Arch Dis Child 1983; 58: 212–215.

Rybak LPl. Laryngoscope 1985; 95: 1–14.

Van der Vorst MMJ. Crit Care 2006; 10: R168.

Van der Vorst MMJ. Critical Care, 2007; 11: R111.

Dipyridamole induces adenosine receptor-independent protection against ischemia and reperfusion injury

C. W. Wouters,1,2 G. Frederix,1 A. Blenke,3 O. C. Boerman,4 W. J. G. Oyen,4 P. Smits1 & R. A. Rongen1
Depts. of 1Pharmacology-Toxicology, 2Cardiology, 3Clinical Pharmacy, 4Nuclear Medicine Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands

Introduction: An open-label non-randomized study suggested that dipyridamole prevents forearm ischemia-reperfusion injury (Riksen et al, CPT 2005). This study was performed to (1) reassess this action of dipyridamole using a double-blind randomized study design and (2) to test the hypothesis, that this effect of dipyridamole is mediated by endogenous adenosine, using caffeine as an adenosine receptor antagonist.

Methods: Experiment 1: 12 healthy male volunteers were treated with either placebo or dipyridamole (one week, 200 mg slow release twice daily) in a randomized double-blind cross-over design. A washout of at least 3 weeks was respected between the two treatment periods. At the end of each treatment period and after 24 hours of caffeine abstinence, volunteers were subjected to 10 minutes of forearm ischemia, combined with standardized intermittent hand gripping. At reperfusion, 500 MBq 99mTc-Annexin A5 was administered i.v. Annexin A5 uptake in the thenar muscle was measured 1 and 4 hours post-reperfusion using a gamma camera and expressed as % difference between the experimental and control hand (annexin A5 targeting). This quantifies ischemia-reperfusion injury. Experiment 2: 20 healthy male volunteers received a one week oral treatment with dipyridamole 200 mg slow release twice daily. On the last treatment day and after 24 hours of caffeine abstinence, either placebo or caffeine (4 mg/kg) was intravenously administered in a double-blind randomized parallel design. Subsequently, forearm ischemic exercise and annexin A5 scintigraphy was performed as described for the first experiment.

Results: In the first study, we observed a significant carry-over effect: annexin targeting was highest after placebo in the group who was treated with placebo first (27 ± 14 (SD) % and 30 ± 16% at one and four hours respectively) and tended to be significantly decreased after dipyridamole (15 ± 9 and 16 ± 10%; p = 0.06). In the group who was treated with dipyridamole first, these numbers were 17 ± 13 and 19 ± 6% after placebo and 20 ± 14 and 19 ± 6 after dipyridamole (p = 0.4). The interaction between treatment order and treatment effect was statistically significant (p = 0.029). In the second experiment, caffeine did not significantly affect annexin A5 targeting after treatment with dipyridamole: there was no difference in annexin A5 targeting between the placebo group (27.3 ± 7.6% and 28.2 ± 10.7% at respectively 1 and 4 hours after reperfusion) and the caffeine group (27.4 ± 11.4% and the 26.4 ± 11.6% at respectively 1 and 4 hours after reperfusion).

Discussion: This study supports our previous observation that dipyridamole protects against ischemia-reperfusion injury in humans in-vivo. In addition, we observed a significant carry-over effect in our model which is not understood yet.

We found no influence of caffeine administration on annexin targeting after oral dipyridamole treatment. We conclude therefore that this action of dipyridamole does probably not depend on adenosine receptor activation.

C.W.W. is supported by ZonMw (920-03-353) and G.A.R. is an Clinical Established Investigator of the Dutch Heart Foundation (2006T035).