Search

The clinical features of COVID-19 pneumonia range from a mild illness to patients with a very severe illness with acute hypoxemic respiratory failure requiring ventilation and Intensive Care Unit admission.To provide a brief overview of the existing evidence for such differences in host response and outcome, and generate hypotheses for divergent patterns and avenues for future research, by highlighting similarities and differences in histopathological appearance between COVID-19 and influenza as well as previous coronavirus outbreaks, and by discussing predisposition through genetics and underlying disease.We assessed the available early literature for histopathological patterns of COVID-19 pneumonia and underlying risk factors.The histopathological spectrum of COVID-19 pneumonia includes variable patterns of epithelial damage, vascular complications, fibrosis and inflammation. Risk factors for a fatal disease include older age, respiratory disease, diabetes mellitus, obesity and hypertension.While some risk factors and their potential role in COVID-19 pneumonia are increasingly recognized, little is known about the mechanisms behind episodes of sudden deterioration or the infrequent idiosyncratic clinical demise in otherwise healthy and young subjects.The answer to many of the remaining questions regarding COVID-19 pneumonia pathogenesis may in time be provided by genotyping as well careful clinical, serological, radiological and histopathological phenotyping.

Graphical abstract, COVID-19 has now spread to all parts of the world and almost all countries are battling against it. This study aimed to assess the efficacy and safety of Integrated Traditional Chinese and Western Medicine (Hereinafter referred to as “Integrated Medicine”) to corona virus disease 2019 (COVID-19). We searched six major Chinese and English databases to identify randomized controlled trials (RCTs) and case-control studies (CCSs) of Integrated Medicine on COVID-19. Two reviewers independently screened, identified studies, and extracted data. Cochrane Risk of Bias tool and the Newcastle-Ottawa Scale were used to assess the quality of included RCTs and CCSs, respectively. Stata (version 13.0; StataCorp) was used to perform meta-analyses using the random-effects model. Risk ratio (RR) was used for dichotomous data while the weighted mean difference (WMD) was adopted for continuous variables as effect size, respectively, both of which were demonstrated in effect size and 95% confidence intervals. A total of 11 studies were included. Four were RCTs and seven were CCSs. The samples of including studies ranged from 18 to 100 (total 982). The traditional Chinese medicine included Chinese medicine compound drugs (QingFei TouXie FuZhengFang) and Chinese patent medicine (such as Shufeng Jiedu Capsule, Lianhua Qingwen granules). Compared with the control group, the overall response rate [RR = 1.230, 95%CI (1.113, 1.359), P = 0.000], cure rate [RR = 1.604, 95%CI (1.181, 2.177), P = 0.002], severity illness rate [RR = 0.350, 95%CI (0.154, 0.792), P = 0.012], and hospital stay [WMD = -1.991, 95%CI (-3.278, -0.703), P = 0.002] of the intervention group were better. In addition, Integrated Medicine can improve the disappearance rate of fever, cough, expectoration, fatigue, chest tightness and anorexia and reduce patients’ fever, and fatigue time (P

A group of patients with pneumonia caused by severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) were reported from China in December 2019. Although several antiviral drugs are widely tested, none of them has been approved as specific antiviral therapy for coronavirus disease 2019 (COVID‐19). Accumulating evidence established a hyperinflammatory states or cytokine storm in COVID‐19. Among these cytokines, IL‐6 plays a key role in cytokine storm and can predict the adverse clinical outcomes and fatality in these patients. Based on the evidence of the significant role of IL‐6 in cytokine storm, diabetes mellitus and cardiovascular diseases as principal comorbidities, it seems that anti‐cytokine therapy may be useful in patients with severe COVID‐19 to reduce mortality. Recent studies demonstrated that herbal‐derived natural products had immunosuppressive and anti‐inflammatory properties and exhibited exceptional act on mediators of inflammation. Parthenolide is the principal sesquiterpene lactones and the main biologically active constituent Tanacetum parthenium (commonly known as feverfew) which has could significantly reduce IL‐1, IL‐2, IL‐6, IL‐8, and TNF‐α production pathways established in several human cell line models in vitro and in vivo studies. Therefore, parthenolide may be one of the herbal candidates for clinical evaluation. This article is protected by copyright. All rights reserved.

Background The pharmacology, toxicology and pharmacokinetics of bioactive preparations derived from natural sources has become a flourishing field of research. However, researching complex extracts and natural products faces numerous challenges. More broadly in recent years the critique of pharmacological research, and specifically its design, the methods used and reporting has intensified. Aims This consensus document provides a perspective on what constitutes best practice in pharmacological research on bioactive preparations derived from natural sources, providing a perspective of what the leading specialist journals in the field consider as the core characteristics of good research. Approach (‘methods’) The editors-in-chief of seven journals developed this best practice statement in an iterative process. A first draft of the guidelines (prepared by MH) was then discussed and amended by the other editors. Outcomes Core to this contribution is a table which provides detailed advice including simple points like a use of appropriate controls and the full taxonomic validity of the material under investigation (see also below), to the relevance of the model for the question being researched (e.g., can specific in silico or in vitro models really address the species anti-inflammatory activity?). Therefore, obviously, researchers must pay detailed attention to reporting and discussing such studies. This information must be discussed critically (as much as it is possible based on the published papers) in terms of their scientific quality and validity. While these points are obvious, as editors, we are aware that they are often not properly implemented. Conclusion We call for an approach which incorporates a careful design, meticulous execution and a detailed reporting of studies focusing on the pharmacology/bioactivity of bioactive preparations. Clearly testable research questions must be developed and investigated experimentally. As the founder of pharmacology Claude Bernard put it already in 1865: ‘…. either the experimenter's hypothesis will be disproved or it will be proved by experiment. When experiment disproves its preconceived ideas, the experimenter must discard or modify it.‘

Tumour lysis syndrome (TLS) is a life-threatening oncological emergency characterized by hyperuricaemia, hyperkalaemia, hyperphosphataemia, and hypocalcaemia [1, 2] due to the rapid lysis of malignant cells, following the initiation of anticancer therapies [3]. Traditionally, therapy for TLS involved intensive hydration, urinary alkalinization and administration of allopurinol [4–6]. Newer guidelines now include rasburicase, with monitoring of electrolytes, white blood cell counts (WCC) and lactate dehydrogenase (LDH) concentrations [1, 7, 8]. Rasburicase, a recombinant urate oxidase enzyme, effectively decreases existing serum uric acid (UA) by oxidizing it to allantoin which is readily soluble and excretable [3]. Although the recommended dose is 0.2 mg kg−1 day−1 for 5–7 days [9], studies have shown the efficacious use of reduced doses for shorter periods of time and subsequent cost savings [5, 6, 10–17]. Expert guidelines by Coieffer et al. [7] in 2008 and Cairo et al. in 2010 [1] on the management of TLS recommend a rasburicase dose of 0.1–0.2 mg kg−1 on the first day, then repeated for up to 7 days [1] or as necessary [7]. We present an analysis of a fixed 3 mg dose of rasburicase administered to adult patients, treated at a tertiary referral centre. The study was approved by the Alfred Health Human Research Ethics Committee and the Monash University Human Research Ethics Committee. Demographic data were collected. Biochemical parameters (serum creatinine, serum UA, phosphate and LDH concentrations), at baseline, 24 h and 72 h after initial administration of rasburicase were recorded and compared. The institution guideline indicates rasburicase to be given before the first dose of chemotherapy in patients considered high risk for TLS. This includes a diagnosis of Burkitt's lymphoma, acute lymhoblastic leukaemia, bulky non-Hodgkin's lymphoma, lymphoblastic lymphoma or acute myeloid leukaemia with one or more of the following: serum UA>0.46 mmol l−1, white cell count (WCC) >50 × 109 l−1 or LDH >two times normal. Patients who were at an ongoing risk of TLS (i.e. elevated UA or LDH or multiple days of aggressive cytoreductive chemotherapy) were allowed a repeat dose of rasburicase 3 mg. Adherence to the guideline was measured. Forty-one patients received 42 courses of rasburicase over a 40 month period (Figure 1A). Diagnosis, demographic and baseline biochemical data are presented in Table 1. Figure 1 Summaryof rasburicase courses and uric acid concentrations. A) Summary of rasburicase courses administered. B) Median uric acid concentrations over time stratified by presentation a baseline. ♦, normal; ▪, hyperuricaemic; ▴, all ... Table 1 Patientcharacteristics Rasburicase was administered as per institution guidelines in 40 (95%) of the patients. Median serum UA concentrations were within normal range at 72 h in all groups; in those who presented with hyperuricaemia, in those who presented with normal baseline serum UA concentrations and overall (Figure 1B). The majority of patients received one dose of rasburicase 3 mg (Figure 1A). In 34 patient episodes requiring one dose only, there was a decline in the median (range) UA concentration from 0.44 mmol l−1 (0.13–1.15) at baseline to 0.22 mmol l−1 (0.02–0.66) at 24 h. This decrease was maintained at 72 h (P < 0.0001) with a median of 0.21 mmol l−1 (0.02–0.52). Serum creatinine concentrations were within normal range (60–105 μmol l−1) at baseline in 74% of patients, with 82% having a normal creatinine at 72 h. Hyperphosphataemia was present in 29% of patients at baseline and increased to 44% at 72 h. Eight patient episodes required more than one dose due to the ongoing risk of TLS. In these patients the median (range) baseline UA was 0.50 mmol l−1 (0.02–2.0), 0.33 mmol l−1 (0.02−1.10) at 24 h and 0.24 mmol l−1 (0.02−1.10) at 72 h (P < 0.0001). Of these patients only 52% had a normal creatinine at baseline, increasing to 83% at 72 h. Mean phosphate concentrations decreased over time but all patients remained hyperphosphataemic at 72 h. No hypersensitivity reactions were noted, no patients required haemodialysis and no deaths were related to the administration of rasburicase. Our results demonstrate that a single fixed dose of rasburicase 3 mg, repeated if required, should be the standard regimen in the management of TLS. Recent studies and published guidelines have shown cumulative support for the safe and efficacious use of off-label dosing regimens of rasburicase [1, 5–8, 10, 11, 16, 17]. A quarter of our patients presented with a baseline WCC>100 × 109 l−1 (Table 1), which is considered a high risk for developing TLS [1, 7]. The Product Information recommends rasburicase 0.1–0.2 mg kg−1 day−1 for 1–7 days [9]. We successfully used a fixed 3 mg dose for these patients. Our data support that presented by Trifilio et al. [11] in a recent study of 287 episodes, the largest published series at this time, of raised UA concentrations successfully treated with a single 3 mg dose of rasburicase, repeated if required. In our cohort, which was smaller in size, a single 3 mg dose was equally effective in both patients who had a normal baseline UA and those with hyperuracaemia. This differed from that published by Trifilio et al., where the single dose was more successful in patients with a lower baseline UA concentration. Our patient cohort also had a higher median LDH. Suboptimal management of hyperphosphatemia was identified in our cohort. More stringent monitoring of patient phosphate concentrations may be warranted in the future to minimize the risk of renal impairment. Serum creatinine, showing a gradual decrease with time, was used as a surrogate maker to indicate an improvement in renal function. Rasburicase was used in conjunction with allopurinol, urinary alkalinazation and intravenous hydration. This strategy is also supported by recent studies and recommendations [1, 11, 16], although the benefit of administering alkalinization with rasburicase needs further investigation [1, 7]. A single fixed 3 mg dose of rasburicase, in the setting of an institution guideline, was efficacious in the management of TLS.

Inhaled corticosteroids and long-acting β2-agonists are medications taken daily on a long-term basis to keep asthma under clinical control. Studies have shown that delivering this therapy in a combination inhaler is as effective as giving each drug separately, but fixed combination inhalers are more convenient for adolescents with asthma, increasing treatment adherence and ensuring that long-acting β2-agonists are always accompanied by inhaled corticosteroids [1,2]. An extrafine fixed combination of beclometasone dipropionate (BDP) and formoterol fumarate (FF) is marketed both as a pressurized metered dose inhaler (pMDI) and as a dry powder inhaler (DPI), as Foster® and Foster® Nexthaler® (Chiesi Farmaceutici S.p.A., Parma, Italy) [3], respectively, at the dose strength of 100/6 μg for patients with asthma ≥18 years of age. The same dose strength is currently being developed in asthmatic adolescents aged 12–17 years. The purpose of this study was to investigate the systemic availability of B17MP (active metabolite of BDP) and formoterol after inhalation of an extrafine fixed combination of BDP/FF 100/6 μg via the NEXThaler® DPI in comparison to the free combination of BDP (Clenil® Pulvinal® 100 μg) and FF (Foradil® Aerolizer® 12 μg) DPIs (Novartis Pharmaceuticals Ltd, Frimley, UK) at a total single dose of 400/24 μg in asthmatic adolescents. Adolescents aged 12–17 years with mild-to-moderate asthma in a stable state participated in this randomized, crossover, single-dose study with two active treatment periods conducted at the Children's Asthma Clinic, Copenhagen, Denmark (ClinicalTrials.gov Identifier: NCT01191424). On each treatment day, blood was sampled for pharmacokinetic and pharmacodynamic assessments before and 5, 15 and 30 min and 1, 2, 4, 6 and 8 h after actuation of the DPIs [4]. Lung function [forced expiratory volume in 1 s (FEV1)] and heart rate were assessed concomitantly and adverse events recorded (see Appendix S1 for details). Pharmacokinetic end-points were as follows: area under the plasma drug concentration–time curve (AUC0–t), maximal plasma concentration (Cmax) and terminal half-life (t1/2) for B17MP and formoterol. The AUC0–t for B17MP was the primary end-point. Pharmacodynamic end-points were as follows: plasma potassium and plasma cortisol minimum plasma concentration (Cmin) and AUC0–t, plasma glucose Cmax and AUC0–2h, time-averaged FEV1 (AUC0–8h/8h) and heart rate (AUC0–8h/8h). All variables were log-transformed and treatment comparisons made using ANOVA models estimating the ratio of the adjusted geometric means test/reference (fixed/free combination) with the corresponding 90% confidence interval (CI) for the pharmacokinetic variables and 95% CI for the pharmacodynamic variables. Bioequivalence for the variables between treatments was demonstrated if the CIs were within the 0.8–1.25 acceptance region [5]. The study design had 90% power to reject the null hypothesis that the upper limit of the 90% CI for the ratio of test/reference means above 1.25 for B17MP AUC0–t. Twenty-seven adolescents [55% males; mean age (SD), 14 years (1.7)] completed the study. The B17MP AUC0–t was slightly higher following BDP/FF NEXThaler® vs. BDP and FF DPIs: test/reference ratio 1.14 (90% CI, 1.03–1.28; Table 1). The plasma cortisol test/reference ratio lower 90% CI bound was just below 0.8 [AUC0–t ratio, 0.92 (0.79–1.06)], but no plasma cortisol levels were below the lower clinical normal limit (5 ng ml−1) in either treatment group. Formoterol AUC0–t was higher following test vs. reference treatment [ratio, 1.46 (1.36–1.56)] and Cmax significantly increased [ratio, 2.83 (2.56–3.13)], but time to Cmax was the same for both treatments (median, 0.08 h). The plasma potassium [AUC0–t test/reference ratio, 0.98 (0.95–1.00)], plasma glucose [AUC0–2h ratio, 1.02 (0.99–1.05)] and heart rate [AUC0–8h/8h ratio, 1.05 (1.01–1.09)] were all within the predefined bioequivalence limit. The treatment effects on FEV1 were similar, and no notable differences in adverse events were observed (see Appendix S1). Table 1 Pharmacokinetic and pharmacodynamic comparison of BDP/FF 400/24 μg inhaled from the fixed BDP/FF 100/6 μg NEXThaler® DPI vs. BDP (Clenil® Pulvinal® 100 μg) and FF ... In adolescents with asthma, the systemic exposure to B17MP following four actuations of BDP/FF 100/6 μg NEXThaler® DPI compared with four actuations of BDP (Clenil® Pulvinal® 100 μg) and two actuations of FF (Foradil® Aerolizer® 12 μg) DPIs was slightly above the 125% bioequivalence limit, with a minor suppression of plasma cortisol levels. The clinical relevance of such suppression should be assessed after repeated administration according to European Medicines Agency guidelines [5]. Formoterol exposure was higher after administration of the fixed vs. free combination, but with comparable systemic effect in terms of heart rate, plasma glucose and plasma potassium. The FEV1 and safety profiles were also comparable. Potassium AUC0–t and Cmin were not affected by the increased formoterol exposure, which may be due to the transient formoterol peak concentration and the time delay between concentration and response or, alternatively, assay insensitivity. We speculate that the increased systemic exposure to formoterol and the slightly increased exposure to B17MP following the fixed BDP/FF NEXThaler® is caused by device-dependent different lung deposition. The BDP/FF NEXThaler® is an extrafine particle DPI with a low minimal inhalation flow threshold [6] in comparison to the Foradil® Aerolizer®, which requires a flow >60 l min−1, below which deagglomeration may be inefficient, resulting in a reduced emitted dose with larger particles and reduced lung deposition. In conclusion, inhalation from BDP/FF 100/6 μg NEXThaler® DPI resulted in a very similar exposure to B17MP but a greater exposure to formoterol compared with the licensed free combination in asthmatic adolescents. However, the greater bioavailability of formoterol following dosing with NEXThaler® had no systemic impact and did not affect the safety profile. For conduction of future bioequivalence trials, we highly recommend usage of devices with comparable lung deposition characteristics, because systemic absorption in adolescents seems to differ according to the device, which may have safety implications.

Inhaled corticosteroids are well established as the first choice of anti-inflammatory therapy for the prophylaxis of asthma in both adults and children. Current treatment guidelines recommend that inhaled corticosteroids should be prescribed for individuals with persistent asthma at all severity levels as long term treatment markedly reduces the frequency and severity of exacerbations. A large number of inhalation devices are used for delivery of asthma medication in children, and it is well known that choice of device is key for the amount of drug that reaches the lungs. For infants and preschool children, in whom active cooperation can be problematic, a pressurized metered dose inhaler (pMDI) used in conjunction with a spacer is usually preferred for maintenance therapy whereas from approximately 6 years onwards a dry powder inhaler (DPI) or breath-activated MDI is recommended [1]. The switch from one delivery device to another generates a possibility of either achieving subtherapeutic drug concentrations or excessive drug exposure with the risk of unwanted side effects. However, there are limited childhood studies available comparing systemic exposure of inhaled corticosteroids using different inhalation devices, and thus no clear rationale for dose selection and dose titration following switch between devices in children with asthma [2]. The objective of this single dose, randomized, double-blind, double-dummy, crossover study was to investigate the effect of delivery device on systemic exposure to fluticasone proprionate (FP) in children with asthma. Due to negligible oral bioavailability of FP and absence of pre-systemic metabolism in the lung, systemic appearance of FP following inhalation is considered equivalent to lung deposition [3]. The study was conducted at the Copenhagen Prospective Study on Asthma in Childhood (COPSAC) clinical research unit in Copenhagen, Denmark. Children with a documented clinical history of mild asthma diagnosed according to the GINA guidelines and able to use the inhalation devices properly were included in the study. Subjects were excluded if they had changed their asthma medication, received systemic corticosteroids, had an airway infection or asthma exacerbation within 4 weeks prior to recruitment or were unable to withdraw from their controller therapy for the duration of the study. The local Ethics Committee provided formal approval of the study (KF 02–107/98, KF 12–136/99). Written informed consent to participate in the study was obtained from subjects and their parents/guardians. Systemic exposure to FP was examined in 13 children aged 12–15 years (10 males), following FP 500 μg inhaled via a DPI (Diskus®) or pMDI with spacer (Volumatic™). The FP dose delivered via pMDI plus spacer was given as two actuations of 250 μg where each puff was inhaled with 30 s of tidal breathing causing an exponential emptying of the aerosol plume. One child had parental consent withdrawn and another child was excluded for technical reasons. Blood samples (5 ml) were collected 10 min before drug administration (baseline) and 20 min, 40 min, 1, 2, 3, 4 and 6 h post-dose for determination of FP plasma concentration [4]. The FP analysis method had a calibration range of 10–1500 pg ml−1 from 0.5 ml plasma. Mean peak inspiratory flow rate obtained during inhalation from the DPI was 107 l min−1 (range 45–126). Delivery of FP 500 μg via a pMDI plus spacer compared with the DPI resulted in a 38% higher maximal plasma concentration (Cmax) (P = 0.02) and 70% higher area under the concentration–time curve (AUC(0,tlast)) (P < 0.01) (Figure ​(Figure1).1). Time to median peak plasma FP concentration (tmax) was furthermore longer with pMDI vs. DPI: 2 h vs. 1 h (P = 0.03) (Table S1 Online). Figure 1 Mean plasma concentrations of fluticasone propionate (FP) following administration of a single 500 μg dose via dry powder Inhaler (DPI) (Diskus®) and pressurized metered dose inhaler (pMDI) plus spacer (Volumatic™) in 12–15-year-old ... The higher systemic exposure after pMDI plus spacer may be due to exponential emptying of the spacer following subsequent inhalations resulting in a greater inhaled volume of the aerosol and increased lung deposition compared to DPI. With DPI, the inhaled volume of drug depends on the turbulence created by each single inhalation and increased impaction in the oropharyngeal region may be a limiting factor for lung deposition [3]. To our knowledge, no previous study has sought to titrate DPI and pMDI to deliver an equivalent lung dose in children, but results in adults have also suggested increased systemic activity with pMDI [5]. Inhaled FP follows linear pharmacokinetics in a wide dose range suggesting that the observed differences between devices will be similar for lower doses [6]. In conclusion, systemic exposure reflecting lung deposition of inhaled FP in 12–15-year-old asthmatic children was significantly higher when the dose was delivered via a pMDI plus spacer (Volumatic™) compared with a DPI (Diskus®). Therefore, we recommend that physicians and patients should be made aware that changing the device may significantly affect the lung dose and hence dose adjustment should be considered.

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT • Extrapyramidal side-effects induced by antipsychotic agents are treated with dose reduction, switching of antipsychotic medication or the addition of anticholinergic antiparkinson drugs. WHAT THIS STUDY ADDS • The use of anticholinergic antiparkinson drugs appears to be superfluous in a large number of patients. AIMS Extrapyramidal side-effects induced by antipsychotic drugs are treated with dose reduction or substitution with another antipsychotic drug or by the addition of anticholinergic antiparkinson agents. The withdrawal of orphenadrine from the Norwegian market provided a possibility to investigate to what degree these alternative measures were taken in clinical practice. METHODS Data were drawn from the Norwegian Prescription Database on the sales of antipsychotics and one of the two anticholinergic antiparkinson agents marketed in 2004, orphenadrine and biperiden, to a total of 39 758 outpatients. The patients were reinvestigated in 2007. The consequences of the withdrawal of orphenadrine from the Norwegian market in 2005 regarding dosing, switching and cessation of antipsychotics and use of anticholinergics were assessed for orphenadrine users compared with biperiden users. RESULTS Of the patients originally using orphenadrine, 28.4% stopped using the drug without reducing the antipsychotic dose or replacing orphenadrine with another anticholinergic agent. The corresponding number for biperiden users was 19.3%. Only 11.8% of patients switched to another antipsychotic drug, but they used significantly lower antipsychotic doses than those who stayed on the same drug. CONCLUSION The use of anticholinergic antiparkinson agents could be seen as superfluous for at least one-third of patients.

5-Fluorouracil (5-FU) is widely administered as a continuous infusion to treat gastrointestinal tract or head and neck cancers, and as bolus injections in breast cancer chemotherapy regimens. Grade 3–4 toxicity occurs in about 30% of patients receiving 5-FU as a continuous infusion, and proves lethal in 0.5% of these patients [1]. 5-FU-related toxicity has mostly been reported with intravenous administration [1–3]. However, some cases of severe toxicity, including deaths, have been described after oral administration of 5-FU derivatives, such as capecitabine (Xeloda®) [4, 5]. A polymorphism of the dihydropyrimidine dehydrogenase (DPD) gene has been identified as a frequent cause of such toxicity [6, 7]. DPD catalyses the rate-limiting step of fluoropyrimidine catabolism. Partial or total DPD deficiency therefore leads to substantial overexposure in patients treated with the standard dose, exacerbating drug toxicity. This raises questions about possible screening for DPD deficiency before the administration of fluoropyrimidine drugs, including their oral forms. Patients found to have a deficiency on screening before treatment or following signs of toxicity during a previous course are usually given alternative treatments based on nonfluoropyrimidine compounds. However, 5-FU is highly active and its use may be essential in patients who fail to respond to other treatments. We report the case of a patient with DPD deficiency detected due to severe toxicity during capecitabine treatment, who has since received a continuous infusion of 5-FU at almost the standard dose with no significant signs of toxicity.

What is already known about this subject • Homoeopathy and herbalism are increasingly popular among the public and prescribed by general practitioners in the NHS. • Doctors and regulatory authorities have expressed concerns about their efficacy and safety. • Studies from the 1990s suggest that between 5.9 and 7.5% of English NHS general practitioners have prescribed homoeopathy, while less than 1% have prescribed herbal remedies. Current levels of prescribing are unknown but are thought to have increased. What this study adds • Sixty percent of Scottish general practices now prescribe homoeopathic or herbal remedies. • The prevalence of homoeopathic prescribing in those under 16 years has doubled since 2000 and is maximal in children

Aims To characterize the pharmacokinetics of mycophenolic acid (MPA) in Chinese renal transplant patients. Methods Thirty-one renal transplant patients (17 male, 14 female) receiving mycophenolate mofetil (MMF) 1.0 g twice daily were included in this study. A pharmacokinetic study was performed during an interval in dosing after steady state had been reached within 2 months after transplantation. The plasma MPA concentration were measured by high-performance liquid chromatography (HPLC) at 0.5, 1, 1.5, 2, 4, 6, 8, 10 and 12 h after the administration of a single dose. Pharmacokinetic parameters were calculated with 3P97 software. SAS® software was used for statistical analysis. Multiple linear regression analysis was used to determine limited sampling approaches. Results The mean peak plasma concentration (Cmax) and area under the concentration–time curve (AUC0−12) were 19.67 ± 8.21 µg ml−1 and 52.16 ± 12.50 µg h ml−1, but there was large variability in these pharmacokinetic parameters. Regression analysis between each plasma concentration and AUC for the limited sampling strategy of MMF therapeutic drug monitoring demonstrated that each of the concentrations at 0.5, 1, 4 and 10 h was positively correlated with AUC (r = 0.60, P = 0.0004; r = 0.60, P = 0.0003; r = 0.61, P = 0.0003; r = 0.64, P = 0.0001, respectively). The combined use of these four samples explained over 90% of the variance in the total (nine-point) AUC0−12. A formula was obtained for the assessment of MPA AUC based on four samples: MPA AUC = 12.61 + 0.37 × C0.5 + 0.49 × C1 + 3.22 × C4 + 8.17 × C10. Conclusions Chinese renal transplant patients had higher median AUCs than caucasians and African-Americans. As in other studies, there was large interindividual variability. A limited four-point AUC was in good agreement with the 12-h AUC and provided the basis of a predictive formula.

Many frail elderly individuals use several medications, which can lead to unnecessary adverse drug events, resulting in medication-associated hospitalizations of which up to half are potentially preventable [1]. Several methods are available to improve the appropriateness of medications used by the elderly. For example, the Prescribing Optimization Method (POM) [2] evaluates medication adherence and classifies potentially inappropriate prescriptions into different categories. As most clinical specialists do not have a complete overview of all the medicines prescribed or used by patients and often do not have sufficient time or knowledge to optimize polypharmacy, it can be expected that the number of physicians visited contributes to inappropriate prescribing. Previous studies have shown that inappropriate prescribing is associated with multiple prescribers [3–5]. The objective of our cross-sectional study was to investigate the relationship between the number of physicians visited and inappropriate prescribing in geriatric patients in the Netherlands. One of the authors (RV) quantified the original POM questions into the domains to form an index (sumPOM) (see Table 1) to evaluate whether the medications used by elderly patients on admission to an acute geriatric ward of a teaching hospital in Tilburg, the Netherlands had been prescribed inappropriately. Under-prescription (scored 1 point) was assessed in each patient and each drug was evaluated in terms of the remaining seven domains. If a drug was prescribed without its indication being clearly stated, then it was not evaluated for lack of effect. Prophylactic medications were considered effective. Per drug, each domain was scored 0 points if it was appropriate and 1 point if it was inappropriate. The causality of adverse effects was determined with the Naranjo scale. Table 1 Scoring system of the Prescribing Optimization Method Two hundred patients [mean age 82.8 (± 6.4 = SD) years] had a mean of 8.6 (± 4.2) prescriptions and had visited a mean of 3.3 (± 1.7) different physicians in the year before admission. The sumPOM score ranged from 0 to 20. In each patient there were 5 (± 3.8) instances of inappropriate prescribing, involving all classes of drugs. The number of physicians visited was clearly associated with inappropriate prescribing on univariate regression analysis [regression coefficient (β) 0.54, 95% CI 0.24, 0.84] (Table 2), but not after adjustment for the number of prescriptions (β −0.02, 95% CI −0.28, 0.24). The number of physicians visited was also related to the number of prescriptions (β 0.82, 95% CI 0.48, 1.15) (data not shown). The number of prescriptions remained the most important independent predictor of inappropriate prescribing in the full model multivariate regression analysis (β 0.52, 95% CI 0.40, 0.64). In this model, living in warden-assisted or sheltered accommodation was associated with a higher rate of inappropriate prescribing than living independently (β 1.91, 95% CI 0.31, 3.51). In contrast, previous contact with a geriatrician was a significant predictor of fewer inappropriate medications (β −1.32, 95% CI −2.29, 0.35), probably because geriatricians are more aware of potential medication-related problems in elderly patients. Table 2 Results of linear regression analysis using inappropriate prescribing (expressed as sum POM) as an outcome variable and the ‘number of visited physicians in the previous year’ as main predictor of interest In a further analysis (data not shown), individual POM domains were not independently associated with the number of physicians visited. The sumPOM collects information about implicit inappropriate prescribing (whether a drug was indicated), but also about explicit inappropriate prescribing (potential drug–drug and drug–disease interactions). Both may adversely affect the benefit : harm ratio of treatment. We did not investigate the clinical consequences of inappropriate prescribing, but it is known that they are associated with preventable harm [1]. In conclusion, many elderly patients admitted to an acute geriatric ward had been prescribed inappropriate medications. However, a visit to a geriatrician in the previous year was found to be associated with fewer inappropriate prescriptions. Physicians' prescribing skills, especially regarding polypharmacy, can be improved by using a systematic medication review method, such as the POM. Like others [3–5], we found the number of physicians visited to be associated with inappropriate prescribing on univariate analysis, but not after adjustment for the number of prescriptions. This makes it important to adjust for the number of drugs used in future studies of risk factors for inappropriate prescribing.

In their cautious review, Shah and Shah [1] emphasized differences in regulations of personalized medicine (PM) among the three major authorities, the US Food and Drug Administration (FDA), the European Medicines Agency (EMA) and the Pharmaceuticals and Medical Devices Agency (PMDA) in Japan. Specific points regarding the differences, however, were not raised for the drugs they selected for discussion in their review. Given that the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) brings together the regulatory authorities and pharmaceutical industries of the USA, Europe and Japan, scientific and technical aspects of drug registration should be harmonized. To identify differences, if any, in regulations of PM, we investigated approvals of PM drugs in the three regions. As a typical example of PM, we focused on PM [2] drugs whose pharmacogenomic biomaker is required on the label. We also studied ivacaftor and pertuzumab, which were omitted in the list [2] simply because they were approved after publication of the list. The US, European and Japanese approval data on these drugs were obtained from Drugs@FDA (http://www.accessdata.fda.gov/Scripts/cder/DrugsatFDA/), European public assessment reports (http://www.ema.europa.eu/ema/index.jsp?curl=pages/medicines/landing/epar_search.jsp&mid=WC0b01ac058001d125) and the PMDA website (http://www.info.pmda.go.jp/approvalSrch/PharmacySrchInit?), respectively. We defined submission/approval delay as the difference between the date of submission/approval in the USA and that in the EU or in Japan. Of 17 FDA-approved drugs and 18 indications whose biomarker is labelled as required, 13 drugs and 14 indications were approved in the EU, whereas 12 drugs and 12 indications were approved in Japan (Table 1). The median submission delay from the submission in the USA was 0 months in the EU and 21 months in Japan. The median approval delay from the approval in the USA was 6 months in the EU and 28 months in Japan. Table 1 US, EU and Japanese data on the approval of personalized medicine drugs whose pharmacogenomic biomarker is required on the label One would expect that labels would not differ significantly among countries, given that regulatory authorities evaluate the same scientific data. Both biological and nonbiological factors, however, can affect regulatory decisions. For example, a much lower incidence of cystic fibrosis [3] and melanoma [4] in Japan compared with the West could discourage the makers of ivacaftor and vemurafenib to file an application to the PMDA. Denileukin diftitox and tositumomab, which were approved for lymphoma by the FDA in 1999 and 2003, respectively, remain unavailable in both the EU and Japan, probably because better treatment modalities are available now. The approval delay in Japan was observed in other therapeutic areas [5]. The present study shows that three-quarters of the approval delay consisted of delays in submission. The approval delay without submission delay in the EU indicates that the reviews took longer for the EMA than for the FDA. The cross-sectional design of our study makes causal inference of these delays difficult. Our results show some similarities and differences in the approvals of PM drugs among the three regions of the ICH. Further studies are needed to investigate differences in postmarketing regulations of PM drugs, because such regulations are important for risk–benefit assessment of PM and are greatly affected by local factors, such as health polices, culture and financial settings.

De Jong et al. [1] reported that the combined use of nonsteroidal anti-inflammatory drugs (NSAIDs) with selective serotonin reuptake inhibitors (SSRIs) synergistically increased the risk of initiating a treatment course with peptic ulcer drugs as a proxy for gastrointestinal adverse effects. In an attempt to limit confounding by indication, tricyclic antidepressant (TCA) users were used as a control group. As the observed association was not adjusted for potential differences between TCA and SSRI users [1], unmeasured confounding may have affected the estimation of the adverse effects. We therefore addressed the association using another pharmacoepidemiological study design, a so-called prescription sequence symmetry design [2]. The main advantage of this design is that it controls for time-invariant (unmeasured) confounding, because the patient population serves as its own control group. We used data from the same pharmacy prescription database (IADB; http://www.IADB.nl) that de Jong et al. [1] used and applied similar inclusion and exclusion criteria. When assessing the association between initiating treatment A and B, the sequence ratio (SR) is calculated by dividing the number of individuals starting treatment A first and treatment B second by the number of individuals starting treatment B first and treatment A second. For the primary analysis, only patients with a time span between both incident prescriptions between 2 and 28 days were included to limit time-variant confounding. We adjusted for prescription time trends by dividing the crude SR by a null ratio, i.e. the SR obtained assuming no association between both drugs, based on overall prescription rates of both drugs in the total IADB population. We estimated 95% confidence intervals (CIs) of SRs using exact confidence intervals for binomial distributions using STATA 12 software (StataCorp LP, College Station, TX, USA). In total, 50 350 adult patients who initiated SSRI treatment were identified between July 1994 and December 2011. A peptic ulcer drug treatment course was started in 277 of them within a 4 week period of SSRI therapy initiation; 126 (45%) started SSRI therapy prior to peptic ulcer drug treatment, while 151 (55%) patients first started peptic ulcer drug treatment [adjusted (a)SR, 0.83; 95% CI, 0.65–1.06; Table ​Table1].1]. Concurrent use of SSRIs and NSAIDs was associated with a statistically nonsignificant increase in the risk of starting peptic ulcer drug treatment (aSR, 1.48; 95% CI, 0.90–2.49), which did not exceed the risk estimated for NSAID treatment alone (aSR, 2.50; 95% CI, 2.27–2.76). Similar results were obtained for TCA treatment alone and concurrent use of TCAs and NSAIDs (Table ​(Table11). Table 1 Symmetry analysis of selected drug therapy initiation within 4 weeks of peptic ulcer drug therapy initiation using a drug-free run-in period of half a year Given that we used the same database and similar inclusion and exclusion criteria as de Jong et al. [1], the main difference between their study and ours is that we used an alternative design that reduces time-invariant (unmeasured) confounding. In accordance with our results, Dall et al., who applied a case–control design in which they controlled for various potential confounders including alcohol abuse, did not observe an increased risk of uncomplicated peptic ulcers associated with combined use of SSRIs and NSAIDs [3]. As we restricted our analyses to relatively short time spans to reduce the potential influence of time-variant confounding, we could not detect a risk that develops later in the course of treatment. However, de Jong et al. found an increased risk for concomitant use of SSRIs and NSAIDs during an average follow-up of 21 days [1]. In addition, Dall et al. found that the risk of uncomplicated peptic ulcers is greatest during the first 30 days of SSRI treatment [3], indicating that our time span of 4 weeks should be long enough to capture potential drug-related increases in peptic ulcer drug prescribing. We could not capture peptic ulcer drug courses prescribed during hospitalization, measure gastrointestinal adverse effects that required hospitalization or measure over-the-counter NSAID and peptic ulcer drug use. However, de Jong et al. did previously find an increased risk in patients concurrently using SSRIs and NSAIDs, while coping with the same limitations [1]. Therefore, the lack of an increased risk associated with SSRI use, whether used concurrently with NSAIDs or not, is likely not to be related to these potential limitations. In conclusion, our results suggest that the observations reported by de Jong et al. [1] might, at least partly, be attributed to unmeasured confounding. Although comparison with a drug that is used to treat the same indication does limit confounding, it does not eliminate all potential confounders. Therefore, when limited data on potential confounders are available, a self-controlled design can have added value, because it eliminates all time-invariant confounders [4].

Although interventional regimens have rightly become the gold standard treatment for the control of haemoptysis, medication, especially vasoconstrictor medication, is still routinely administered in the clinic. The vasoconstrictor posterior pituitary hormone is the most commonly prescribed drug for the management of haemoptysis in China. However, vasoconstrictor therapy has various adverse effects, some of which may be severe, and is ineffective in a few patients.

In comparison with carbamazepine, oxcarbazepine (OXC) has fewer associated side-effects and a reduced number of reported drug–drug interactions. The only cases of serious toxicity have been documented after suicide attempts using multiple drugs [1], but OXC toxicity alone has never been noted to be the only cause of death [2]. The main drug–drug interactions reported (in particular with other antiepileptics such as carbamazepine, phenobarbital and sodium valproate) actually result in the lowering of plasma OXC concentrations [3, 4]. We report a case of severe OXC toxicity after starting clarithromycin in a patient with refractory epilepsy and discuss the mechanism of this potential interaction. We suggest that clarithromycin may inhibit the efflux proteins of the blood–brain barrier (BBB), which are thought to be overexpressed in drug-resistant patients [5]. This inhibition may result in increased cerebrospinal fluid (CSF) concentrations of OXC and explain the signs of toxicity observed. A 10-year-old boy with known refractory epilepsy presented to our department with symptoms suggestive of acute OXC toxicity. His past medical history included a diagnosis of epilepsy at the age of 18 months when he presented with generalized tonic clonic seizures and associated global developmental delay. There were no other significant illnesses or dysmorphism and he had been born at term without complications. The first child of nonconsanguineous North African parents, there was a family history of epilepsy in three paternal second cousins (one with developmental delay). Diagnostic investigations performed included normal brain magnetic resonance imaging and positron emission tomography scan, karyotyping, genetic screening for fragile X syndrome and a metabolic screen, all of which were unremarkable. His medications at the time of presentation included: OXC 540 mg am/510 mg nocte, lamotrigine 100 mg b.d., sodium valproate 100 mg am/300 mg and topiramate 100 mg b.d. Despite the treatment, between two and three epileptic fits were reported per month. A course of clarithromycin (250 mg b.d.) was started due to a mild respiratory tract infection 3 days after the onset of coryzal symptoms. An hour after he had taken the first dose, the parents noticed their son was unsteady on his feet (Figure 1) and had a brief episode where he appeared unresponsive. Twenty-four hours after starting the antibiotic, he was brought to our paediatric emergency department with an increase in symptoms including vomiting, drowsiness and dizzy spells. Clinical neurological examination revealed hyperkinesia, ataxia and nystagmus. At the time of presentation, the patient was already apyrexial with no respiratory signs other than mild coryzal symptoms. The rest of the clinical examination was unremarkable. On admission, blood tests were essentially normal, including: plasma electrolytes, liver enzymes, renal function and inflammatory markers (sodium 142 mmol l−1, potassium 3.5 mmol l−1, C-reactive protein 57 mg l−1). An initial electroencephalogram showed a decrease in paroxysmal activity. Plasma levels of sodium valproate were dosed as suboptimal (21.7 mg l−1, ref 50–100 mg l−1) and as an incidental finding, the patient was noted to be mildly hypocalcaemic (2.08 mg l−1, ref 2.20–2.80 mg l−1). Twelve hours after admission, the dose of OXC was reduced to 420 mg (80% of the original) and the clarithromycin was stopped. The other medications remained unchanged. After a further 12 h, the dose of OXC was increased back to 540 mg as the clinical symptoms were much improved. Following this increase, once again the patient developed drowsiness and ataxia. As a result, the OXC dose was halved for 24 h. No further symptoms were reported and the dose was then progressively increased back to the initial value over a period of 72 h. No further symptoms or seizures were reported during the rest of his hospital stay and the patient was discharged home after 5 days. Figure 1 Clinical symptoms vs. time since starting clarithromycin OXC is particularly recommended due to the minimal side-effects and drug interactions seen in comparison with carbamazepine. We report the first case of OXC toxicity likely to have been induced by a drug–drug interaction with clarithromycin (250 mg b.d.). After two doses of clarithromycin, the patient needed to be hospitalized as an emergency for suspected toxicity. Many of the symptoms observed in our patient are known to be linked to possible OXC toxicity, including drowsiness, dizziness, nausea, vomiting, hyperkinesia, ataxia and nystagmus. Once the clarithromycin was stopped, a clear improvement was seen (t1/2= 4 h). However, the early re-introduction of OXC (back to the initial dose after 12 h) resulted in a further similar episode. The dose was then halved and progressively increased over a period of 72 h with no further problems. This shows there is a latent period in the mechanism of interaction between clarithromycin and OXC. Furthermore, halving the OXC dose did not lead to any seizures in our patient. In the absence of treatment with clarithromycin, the patient had not previously developed any signs of OXC toxicity (despite being on the medication at the same dose for several months). To date, the only reported OXC interactions are due to modifications in liver metabolism. Ninety-five percent of OXC is metabolized to the active monohydroxy derivative (DMH) and 4% to the inactive dihydroxy derivative (DDH). Pisani et al. have shown that an increase in levels of the active metabolite and a decrease in the inactive form occur when OXC is administered in conjunction with viloxazine (an antidepressant) [6]. This inhibition of the conversion of the active DMH to the inactive DDH form has never been linked to any adverse clinical effects. Studies carried out to research possible interactions between macrolides (erythromycin) and OXC have not shown any interactions [7]. However, such studies were conducted in healthy volunteers with no prior drug resistance, and the results were based on potential changes in plasma concentrations of OXC and DMH. It has been shown that normal OXC or DMH serum levels do not exclude central nervous system (CNS) toxicity or the occurrence of serious neurological adverse effects [8]. The lack of correlation between serum OXC levels and neurological side-effects can be explained by the fact that serum levels do not give a reliable reflection of the actual levels in the CNS. Thus, passage across the BBB may be an important factor in the efficacy of OXC and DMH. Numerous studies have demonstrated the role played by efflux proteins present on the membranes of cells which make up the BBB [9]. Increased expression of efflux proteins such as the P-glycoproteins (multidrug resistance 1 and 2) and the Multidrug Resistance Proteins 1–9 has been observed in some drug-resistant patients, which could be responsible for a decreased concentration of OXC in the CSF, resulting in decreased clinical effects [4, 9–11]. Macrolides are known to be potent inhibitors of P-glycoprotein (50% decrease in activity). Therefore, we propose that the toxic side-effects observed in our patient after starting clarithromycin (whilst already on OXC) may be explained by an increase in brain OXC concentrations due to the inhibition of the active BBB efflux proteins. We recommend physicians consider adjusting the dose of OXC when using a macrolide in drug-resistant patients due to the potential risks of drug toxicity.

Azathioprine (AZA), a prodrug of 6-mercaptopurine, is prescribed in evolving forms of inflammatory bowel disease (IBD), affecting predominantly White adolescents and young adults. Thiopurine S-methyltransferase (TPMT) catabolizes AZA and its genetic polymorphism affects the balance between active 6-thioguanine nucleotides (6-TGN) intracellular concentrations [recommended target 235–400 pmol per 8 × 108 red blood cells (RBC)] and hepatotoxic 6-methylmercaptopurine ribonucleotides (6-MMP) [1]. Three polymorphisms (TPMT*2, *3B, *3C; rs1800462, rs1800460, rs1142345, respectively) account for >95% of variant alleles, but additional rare variants are described (TPMT*1S to TPMT*25) [2]. Approximately 90% of Whites are extensive metabolizers with two wild-type TPMT alleles, 6–11% are intermediate metabolizers and 0.2–0.6% are TPMT-deficient patients, exposed to severe myelotoxicity as they accumulate high 6-TGN concentrations under AZA conventional doses [3]. This case reports a pancytopenia under AZA treatment for IBD. As 6-TGN and 6-MMP concentrations and TPMT genotype based on the three predominant polymorphisms were discordant, we conducted additional genetic analysis and identified a new, never described TPMT mutation affecting TPMT activity.

I doubt that personal spats between foot-soldiers will further solutions to the research governance problems unmasked by Paul Stewart and myself in the BMJ last year [1]. Almost 1 year on, I believe progress is unlikely until the monarchs with the money at grant foundations concede that their largesse is being wasted and take sanctions against the robber barons among regulators. What follows therefore is a mere exhibition joust, aimed mainly at our common and worst foe, RD their incompetence and intimidation were not fully manifest at the time of my article [2], but were exposed by the unanimous obloquy directed at them during the meeting of National Institute of Health Research Senior Investigators in April at the Wellcome Trust (http://www.nihr.ac.uk/files/pdfs/DismantlingBarriersWorkshop.pdf).

While the cure for pulmonary arterial hypertension (PAH) has been elusive, it has been managed by the use of prostanoids, endothelin receptor antagonists (ERA) and the phosphodiesterase-5 (PDE5) inhibitors [1–7]. Recently, the combination of ERA (e.g. bosentan) and PDE5 inhibitors (e.g. sildenafil) has been explored with encouraging outcome [6, 8–10]. The clinical pharmacology of bosentan, the first approved ERA, has been reviewed by Dingemanse and van Giersbergen [11]. Bosentan apparently undergoes complete elimination via hepatic metabolism, followed by biliary excretion of the metabolites with an elimination half-life of 4–5 h. Apart from mild hepatic dysfunction, other conditions of hepatic dysfunction of increasing severity substantially altered the pharmacokinetic disposition of bosentan. The involvement of both cytochrome P450 (CYP) 3A4 and CYP2C9 has been implicated in bosentan's metabolism to form three oxidative metabolites. Interestingly, bosentan decreases the exposure of both CYP3A4 and CYP2C9 substrates due to induction phenomena, whereas a CYP3A4 inhibitor has the propensity to increase dramatically the exposure to bosentan. The pharmacokinetics and clinical pharmacology attributes of sildenafil have been described [12–15]. The metabolic pathways of sildenafil comprise N-demethylation, aliphatic hydroxylation and oxidation, with major involvement of CYP3A4 (70%) and minor involvement of CYP2C9 (20%) isozymes; upon co-administration, both CYP3A4 and CYP2C9 inhibitors can increase exposure of sildenafil, whereas the inducers of these isozymes drastically reduce exposure to sildenafil. The clearance of sildenafil was described to be moderate in humans and the elimination half-life value is about 3 h. The clinical pharmacology and pharmacokinetic characteristics of tadalafil have been reported [15–18]. Tadalafil's metabolism, controlled by CYP3A4, showed formation of an inactive catechol metabolite. A large portion of the catechol metabolite undergoes Phase II metabolism after methylation step to form a glucuronide conjugate. Tadalafil has been documented not to inhibit or induce important CYP isozymes. However, co-administration of tadalafil with CYP3A4 inducer or inhibitor can decrease or increase the exposure of tadalafil. Clearance of tadalafil is slower compared with sildenafil, with an elimination half-life of 17 h. Recently a systematic pharmacokinetic investigation was published that evaluated the interaction potential of both agents when co-administered vs. single agent treatments [19]. There was a 60% reduction in the exposure of sildenafil, which confirmed the earlier findings of decreased exposure of sildenafil when co-administered with bosentan [19, 20]. However, interestingly, an opposite effect was observed in the exposure of bosentan, which showed about 50% increased exposure as a result of co-administration with sildenafil [20]. The authors postulated that sildenafil may play an inhibitory role in the hepatic transporter uptake/biliary clearance of bosentan [20–22]. Since bosentan does not undergo nonhepatic clearance, an inhibitory effect on hepatobiliary clearance mechanisms may lead to the accumulation of the drug. Although sildenafil is an organic anion-transporting polypeptide (OATP) transport inhibitor [22], it was apparent that a greater threshold concentration was necessary to elicit the desired response. It is quite possible that with a dosing of 80 mg sildenafil (t.i.d.), the desired threshold may have been achieved in this study under in vivo conditions [20]. Almost concurrently, another interesting pharmacokinetic interaction report between bosentan and tadalafil in healthy subjects has been published [23]. Whereas, as expected, bosentan reduced the exposure of tadalafil by almost 40%, tadalafil did not appear to increase the exposure of bosentan, in contrast to what was observed previously with that of sildenafil, although numerically the exposure of bosentan (i.e. AUC) was found to increase marginally by 13% [23]. Although it is difficult to rationalize the unique pharmacokinetic interaction observed in the above-cited examples [20, 23], it appears that differential behaviour is exhibited by the two PDE5 inhibitors. It is unknown whether or not tadalafil has a role to play in the OATP transporter uptake inhibition of bosentan. Additionally, a recent study seems to suggest that tadalafil has a tendency to exhibit mechanism-based inhibition of CYP3A4 isozyme with a very low potency [17] and therefore there was a possible opportunity to increase bosentan levels from both speculative counts (CYP3A4 and hepatic uptake inhibition), although it was not supported by study data [23]. These recently reported differential pharmacokinetic interaction data between bosentan vs. PDE5 inhibitors [20, 23] will pave the way for further in vitro and in vivo experiments to understand fully the nature and consequences of such interactions. In this context, another approved ERA agent, ambrisentan, for PAH treatment, may provide an alternative option for co-administration with PDE5 inhibitor [24]. Since ambrisentan is a substrate for both OATP and P-glycoprotein transporter systems as well as CYP3A4 [25], it could be speculated that a similar type of interaction could possibly occur between sildenafil and ambrisentan. However, a single dose of ambrisentan (10 mg recommended dose) did not influence the pharmacokinetic disposition of sildenafil (20 mg t.i.d. dosing) and its active metabolite [25]. Similarly, the pharmacokinetics of ambrisentan was not altered by a single dose of sildenafil [25]. Although no reports have been published, it is unlikely that a drug–drug interaction would have occurred under steady-state conditions between the two agents, because ambrisentan is not documented to be a CYP3A4 inducer or inhibitor [24]. Although drug–drug interaction represents a liability, it could be countered by proper dosage adjustments of either one of the two agents. Although it is customary to dose adjust one agent per the propensity of the documented interaction, it may be difficult to dose adjust both agents such as the one observed between bosentan and sildenafil, especially when the effect of the two agents is observed to act in opposite directions. The other key question is: does increased exposure to bosentan adequately compensate for the reduced exposure to both sildenafil and its metabolite in terms of the pharmacodynamic attributes so that the requirement for dosage adjustment would not arise. Therefore, from all data gathered from the literature there appears to be scientific merit in rationalizing combination use in PAH patients of ERA and PDE5 inhibitors by choosing the appropriate agents that present limited potential for any drug–drug interaction, and/or titrating to the desired pharmacodynamic end-points by taking advantage of the observed pharmacokinetic interactions.

Aims To quantify and compare the incidence of ventricular arrhythmias associated with the use of five nonsedating antihistamines: acrivastine, astemizole, cetirizine, loratadine and terfenadine. The effects of age, sex, dose, duration of treatment, and the interaction with P450 inhibitor drugs were also examined. Methods We carried out a cohort study with a nested case-control analysis using the UK-based General Practice Research Database (GPRD). The study cohort included persons aged less than 80 years old who received their first prescription for any of the five study drugs between January 1, 1992 and September 30, 1996. We estimated relative risks and 95% confidence intervals of idiopathic ventricular arrhythmias with current use of antihistamines as compared with non use. Results The study cohort included 197 425 persons who received 513 012 prescriptions. Over the study period 18 valid cases of idiopathic ventricular arrhythmias were detected. Nine occurred during the current use of any antihistamine, resulting in a crude incidence of 1.9 per 10 000 person-years (95%CI: 1.0–3.6) and a relative risk of 4.2 (95%CI: 1.5–11.8) as compared with non use. Astemizole presented the highest relative risk (RR=19.0; 95%CI: 4.8–76.0) of all study drugs, while terfenadine (RR=2.1; 95%CI:0.5–8.5) was in the range of other nonsedating antihistamines. Older age was associated with a greater risk of ventricular arrhythmias (RR=7.4; 95%CI: 2.6–21.4) and seemed to increase the effect of antihistamines (RR=6.4; 95%CI: 1.7–24.8). The proportions of high dose terfenadine and the concomitant use with P450 inhibitors among current users of terfenadine were 2.7% and 3.4%, respectively over the study period with no single case of ventricular arrhythmias occurring in the presence of these two risk factors. Conclusions The use of nonsedating antihistamines increases the risk of ventricular arrhythmias by a factor of four in the general population. Yet, the absolute effect is quite low requiring 57 000 prescriptions, or 5300 person-years of use for one case to occur. The risk associated with terfenadine was no different from that with other nonsedating antihistamines.

Aims Propofol is a widely used i.v. anaesthetic agent. However, its binding properties to blood components have not been fully studied. Methods We studied the binding of propofol to erythrocytes, to human serum and to isolated serum proteins. Because propofol bound to ultrafiltration and equilibrium dialysis membranes, we used a co-binding technique with dextran coated charcoal and with erythrocytes. Results Propofol free fraction in blood was 1.2–1.7% at total concentrations ranging from 2.80 to 179 μm (0.5 to 32 μg ml−1 ). Fifty percent was bound to erythrocytes and 48% to serum proteins, almost exclusively to human serum albumin. In the clinical range of concentrations (0.5–16 μg ml−1 ) 40% of the molecules bound to erythrocytes are on the red blood cells membranes. No binding to lipoproteins occurred and binding to α1-acid glycoprotein was less than 1.5% Conclusions We conclude that hypoalbuminaemia may increase propofol free fraction particularly during prolonged administration. Since propofol is non-restrictively cleared, no change in clearance is expected to occur, and the increase in free fraction will not be compensated by a parallel increase in clearance. It is also noted that many in vitro studies used concentrations 50 to 500 times the concentration expected to be encountered in the immediate cellular environment.

Many clinical resources for healthcare providers continue to list erythromycin and clarithromycin as inhibitors of cytochrome P4501A2 (CYP1A2) [1]. This classification is based on the pharmacokinetic interactions of these drugs with theophylline, the magnitudes of which are dependent on macrolide dose and the duration of treatment (either no pharmacokinetic changes, or decreases in theophylline clearance by 10–40%; reviewed by Periti et al. and Westphal) [2, 3]. Since theophylline is metabolized primarily by CYP1A2 at clinically relevant concentrations (fmCYP1A2 estimated as approximately 0.8) [4], it is assumed that perturbations in metabolic clearance must result from altered CYP1A2 activity. However, available data suggest that macrolides are poor inhibitors of CYP1A2. In vitro, the 7-hydroxylation of (R)-warfarin by recombinant CYP1A2 was not inhibited by erythromycin or roxithromycin [5], and high concentrations of various macrolides reduced the activity of CYP1A2 in human liver microsomal preparations by ≤10% in competitive inhibition studies [6]. Furthermore, the in vivo indices of CYP1A2 activity, oral caffeine clearance and 6 h paraxanthine to caffeine plasma concentration ratio, were unchanged following clarithromycin dosing in healthy volunteers (500 mg b.i.d. × 7 days) [7]. Drug–drug interactions involving macrolides are typically attributed to mechanism-based inactivation (MBI) of CYP3A enzymes [8]. This type of inhibition is readily detected in vitro by observing time-dependent inhibition (TDI) (CYP activity is reduced by a ‘pre-incubation step’; reviewed by Polasek and Miners [9]). However, the possibility that weak TDI of CYP1A2 by macrolides may contribute to their interactions with theophylline has not been thoroughly explored. To address this, a conventional in vitro approach ([10]) was used in our laboratory to evaluate potential TDI of recombinant (Escherichia coli-expressed) and human liver microsomal CYP1A2 by troleandomycin, erythromycin, clarithromycin, roxithromycin and azithromycin (0.1–1000 µM). Phenacetin O-deethylation was used to measure CYP1A2 activity. The relative contribution of CYP1A2 in human liver microsomes to this reaction was estimated to exceed 90% (see Polasek et al. [10] for details of the experimental method). As evident from Figure 1, all five macrolides were poor inhibitors of phenacetin O-deethylation using either enzyme source (≤25% inhibition at the highest concentration tested), and TDI was not observed. Consistent with previous reports, fluvoxamine (negative control) did not demonstrate TDI, whereas furafylline (5 µM) and clorgyline (5 µM) (positive controls) both reduced phenacetin O-deethylation rate with time, so that approximately 50% of control activity remained following a 10-min pre-incubation. The lack of TDI seen here with all five macrolides supports recent data for erythromycin, which failed to demonstrate TDI of CYP1A2 using a similar experimental approach [11]. Figure 1 Lack of time-dependent inhibition of phenacetin O-deethylation (CYP1A2) by macrolides. Macrolides (1000 µM) were pre-incubated with either (a) human liver microsomes or (b) Escherichia coli-expressed recombinant CYP1A2 and nicotinamide adenine ... If competitive inhibition and MBI of CYP1A2 are excluded, what explains the macrolide–theophylline interactions? It is possible that CYP3A enzymes play a greater role in the metabolism of theophylline in vivo than predicted from in vitro studies. The poor correlation between the 6-h paraxanthine to caffeine plasma ratio and the oral clearance of theophylline [12] suggests that, although the disposition of caffeine and theophylline are mediated primarily by CYP1A2, there are additional CYP (mostly CYP2E1 and CYP3A) involved in the in vivo disposition of theophylline [7, 13]. Indeed, CYP3A may be the principal CYP implicated when CYP1A2 activity is low, a hypothesis substantiated by Tjia et al., who demonstrated partial restoration of theophylline metabolism when human liver microsomal CYP1A2 was fully inhibited by α-naphthoflavone [14]. Furthermore, the 8-hydroxylation pathway can be activated by nifedipine and α-naphthoflavone in microsomes from liver donors treated with dexamethasone, suggestive of allosteric effects on CYP3A [15]. Intriguingly, the rank order of macrolide interaction potential with theophylline corresponds to that of midazolam, a drug exclusively eliminated by CYP3A, i.e. troleandomycin > erythromycin ≈ clarithromycin > roxithromycin > azithromycin for both affected drugs [2, 8]. Similarly, verapamil and diltiazem, also mechanism-based inactivators of CYP3A (but not CYP1A2 [11]) that interact with midazolam, cause between 10 and 25% reduction in theophylline clearance [16]. Despite these observations, the actual extent of CYP3A involvement in the in vivo metabolism of theophylline remains unclear. In fact, other studies with the prototype CYP3A inhibitors ketoconazole [17] and grapefruit juice [18], and the CYP3A substrate nifedipine [19], have demonstrated essentially no effect on theophylline clearance. Another explanation is that macrolides compete with theophylline uptake into hepatocytes via organic anion transporter 2, as shown recently for erythromycin and clarithromycin [20]. It is also known that macrolides inhibit P-glycoprotein [21] and various organic anion transporting polypeptides [22]; however, the degree to which these transporters influence the disposition of theophylline is still unknown (T. Yamamoto, personal communication). Inhibition of CYP2E1 by macrolides is another possibility, although no data are available to support this line of reasoning. It is therefore proposed that the pharmacokinetic interactions between older macrolides and theophylline probably arise via a combination of potent CYP3A inactivation and the inhibition of theophylline uptake into hepatocytes. Substantial evidence is now available to exclude CYP1A2 inhibition as the molecular mechanism responsible for the macrolide–theophylline interactions, and clinical resources for healthcare providers should be updated accordingly.

Many avoidable adverse drug reactions (ADR) in elderly patients are due to an absence of dose adjustment according to renal function [1]. Nephron loss is a normal ageing process, leading to a lower glomerular filtration rate (GFR). Many drugs and/or their metabolites are excreted through the kidney; thus, the risk of ADR is increased if dose is not adjusted as a function of creatinine clearance (CC). However, despite a lower GFR, serum creatinine concentration may be normal in elderly patients because the rate of production of creatinine is dependent on dietary intake and muscle mass [2]. Therefore, CC should be systematically estimated in this population. Physician prescribing behaviour has been assessed for hospitalized patients with concealed renal insufficiency (serum creatinine level within the normal range but low estimated GFR) [3–5], but not in general practice. We aimed to assess general practitioners' (GP) prescribing behaviour for elderly patients with concealed renal insufficiency and knowledge of drugs requiring dose adjustment. The study was carried out in a French departement (a French administrative unit). Fifty GPs were randomly selected from the telephone directory of physicians and were asked by phone to participate in the study. The same interviewer questioned the 50 GPs using a questionnaire based on two questions and two clinical cases. The use of a source of recommendations concerning dose adjustment for renal impairment was allowed. All 50 GPs agreed to participate and completed the questionnaire. This study group included 38 men (76%) and 12 women (24%). Twenty-two GPs worked in a town with a population of >10 000; nine worked in a town with 5000–10 000 people and 19 in a town of

Spontaneous reporting systems remain the cornerstone of the early detection of previously unknown adverse drug reactions (ADRs) [1]. However, a large proportion of ADRs are known and preventable and they are often due to the coadministration of drugs known to interact [2]. Spontaneous reports of known ADRs can provide insight into the inappropriate co-prescribing of medications. The World Health Organization (WHO) ADR Database (Vigibase) contains more than 3.8 million suspected ADR reports from 82 countries [3]. We examined the coreporting in Vigibase of all drugs classified as ‘established’ and ‘clinically important’ in the Swedish, Finnish, INteraction X-referencing drug–drug interaction database (SFINX database) [4] used in a Swedish patient record system. Thirty-five ‘established and clinically important’ drug–drug interactions (DDIs) were identified. Co-prescribing of these drugs was then searched for in VigiBase. Subsequently, data were retrieved on the severity and evidence for interactions involving these pairs, and actions recommended in Stockley's Interaction Alert [5]. Of those 35 ‘established and clinically important’ drug pairs, 31 were reported in Vigibase, involving 9547 reports from 50 countries. The reported DDIs are listed in Table 1. The serious nature of many of the ADRs listed in Table 1 makes this a major patient safety issue. Also, seven pairs had had only theoretical evidence previously available, and another four pairs had had no previous evidence. Table 1 Clinically important drug–drug interactions Amongst a wide range of drugs, the majority of reports concerned anticonvulsants and anticoagulants. Many of the most reported drug pairs included drugs with narrow therapeutic indexes, such as warfarin, carbamezepine, phenytoin and theophylline. This was further emphasized by the nature of the adverse events reported for the drug pairs: ‘therapeutic level increased’, ‘drug level increased’, ‘drug level decreased’ and ‘therapeutic level decreased’. Spontaneous reports sometimes lack detail, which makes the analysis difficult. For example, risperidone/carbamezepine and convulsions could be due to underlying disease or an overdose effect of risperidone on discontinuation of carbamazepine. In many reports one drug only was reported as ‘suspected’. For example, only 149/661 cases of decreased prothrombin and 197/408 cases of gastrointestinal haemorrhage with warfarin and acetylsalicylic acid were reported as co-suspected or interacting. Judging by attribution of suspicion, it seems that potential interactions were not recognized, although we should not assume that all reports are necessarily results of inappropriate use. The continuing reporting of seemingly well-established interactions strongly suggests insufficient impact of drug information on routine prescribing practices. For example, the US Food and Drug Administration (FDA) has warned of concurrent therapy of rosiglitazone and insulin [6], as it was associated with increased risk of cardiac failure [7]. Since that FDA warning in May 2004, 402 reports of the pair, 52 reports of cardiac failure and 46 reports of the most common symptom, peripheral oedema, have been entered into VigiBase. Studies have shown that education can improve both doctors' and medical students' prescribing patterns [8], and electronic advice can significantly improve recognition of dangerous drug combinations [9]. This review has focused on an analysis of spontaneous reports, with no estimate of drug use. The raw number of reports of combinations should not be interpreted as an estimate of incidence. For some combinations, increased recent reporting might be explicable by increased drug use. Many of the drug pairs have been marketed for several years, and for two-thirds of thepairs reporting started >10 years ago. Our results illustrate a longstanding international problem of comedication of contraindicated drugs. This is in agreement with national studies that have shown that physicians fail to recognize [9] and continue to prescribe contraindicated drugs [10]. We contend that VigiBase and other available data sources could, and should, be utilized to identify preventable ADRs through active screening for potential DDIs. Also, much more effort is needed to communicate patient safety findings appropriately to healthcare providers and patients.

Originally, the definition of concordance ‘impl[ied] that the prescriber and patient should come to an agreement about the regimen that the patient will take’, as I wrote in my recent editorial [1]. I also pointed out that compliance can mean ‘Accord, concord, agreement; amicable relations (between parties)’, which seems to me to be in concord with this definition of concordance. This may partly explain the confusion between these terms, reflected in the titles of some publications, in which ‘concordance’ is used as a synonym for ‘compliance’ or ‘adherence’ [2–4]. In fact, the definition of concordance has shifted since it was first invented. The original definition was ‘an agreement reached after negotiation between a patient and a health care professional that respects the beliefs and wishes of the patient in determining whether, when and how medicines are to be taken’ [5]. However, as I argue below, a negotiated agreement is not relevant to the interaction between a clinician and a patient, and the focus of concordance has therefore shifted from the consultation to how to communicate with and support the patient [5], which are relevant. Bell et al. also suggest that a consultation between a clinician and a patient is a negotiation [6]. It is not, as those who regularly treat patients know. A 55-year-old man suddenly develops crushing retrosternal chest pain while mowing his lawn. The pain radiates to his neck and jaw and down his left arm. He has tingling in the fingers of his left hand and a feeling of impending doom. An electrocardiogram shows widespread ST segment elevation in the anterior leads, with Q waves and T wave inversion, and his serum troponin concentration is markedly raised. I tell him that he has had a heart attack and explain why I have reached that conclusion and how it has probably happened. There is no negotiation. The diagnosis is definitive – take it or leave it. If he does not accept the diagnosis he can, if he wants, consult another clinician. But whether he agrees or not, he has had a heart attack. Had the diagnosis been less clear, doubts could have been discussed and further investigations suggested, but there would still be no negotiation. He can refuse further investigations if he prefers, but no amount of agreement or disagreement will alter the circumstances. Now I suggest to him that thrombolytic therapy is likely to be beneficial, since there is good evidence that he will benefit from such treatment and that the potential benefit will outweigh the potential harm. Again there is no negotiation. He can refuse the treatment, but he cannot, for example, negotiate for a delay while he goes home to finish some work, or for treatment with only half the recommended dose, or for a tablet instead of an injection. I have a responsibility to tell him about the benefits and harms and the balance between them, couching my explanations in terms that he will understand, and giving him the opportunity to ask questions. His responsibility is to decide whether the treatment is acceptable to him and to tell me in good time (since the longer he waits the less effective the treatment will be). If he accepts the treatment I have the further responsibility to administer it properly, to monitor its effects and to take further necessary actions, explaining each step. In a discussion with him about resuscitation in the event of a cardiac arrest I would follow his wishes; in that case the position is reversed – I would not want to negotiate otherwise. A search of Pubmed for papers whose titles include the word ‘concordance’ yields about 1300 results, well over 95% of which are nothing to do with agreement between clinicians and patients. They use the term in the sense of correspondence. For example ‘physician and patient gender concordance’ means that the physician and patient are of the same sex [7]. Similarly, ‘patient–physician racial and ethnic concordance’ means that they are of the same race or ethnic origin [8]. ‘Language concordance’ is the use and understanding of the same language [9]. In all these and other cases [10–12] the correspondence is perfect. However, clinicians' views about therapy can never be perfectly concordant with those of patients. As Bell et al. put it, ‘How individual patients value the risks and benefits of a particular medicine may differ from the value assigned by their clinicians’ [6]; indeed, such a difference generally occurs. The best that can be hoped for is that the clinician explains clearly and advises appropriately and the patient decides whether to accept the advice and to adhere to the proffered therapy, with suitable support from the clinician, or to reject the advice, either outright or by subsequent non-adherence (adherence not always being beneficial). Concordance has become a shibboleth that has been accepted uncritically in certain quarters in the absence of evidence of its benefits and harms and of the balance between them. Authors do not write about testing the concept of concordance. They write about achieving concordance [13], promoting it [14], improving it [15], enhancing it [4], even when they admit at the same time that supportive evidence of benefit is rare [13] and although doubts about its benefits have been expressed and possible harms mentioned [16]. What evidence there is, is limited. Take ‘concordance therapy’, for example. A search of Pubmed for the term yields four references, of which three are interventional studies: in the management of akathisia mental health nurses can, it is suggested, ‘safely prescribe psychiatric medication in combination with concordance therapy’ [17] in 19 patients, for whom antidepressants had been prescribed, concordance therapy was said to be ‘acceptable and feasible’ and the authors concluded that it ‘show[ed] sufficient promise of efficacy to justify an adequately powered R[andomized] C[ontrolled] T[rial] [18] in 10 subjects with bipolar I disorder, of whom eight had concordance therapy, there were small improvements in attitudes towards lithium and in self-reported adherence [19]. There is no evidence on the possible harms of such therapies. Some of the types of research on concordance that are desirable have been mentioned in a thoughtful review [5], and the National Institute for Health and Clinical Excellence in the UK is conducting a critical appraisal of the evidence on shared decision making, although the report is not expected until December 2008 [20]. Not all patients want to be more involved in decisions about their management. In a study of 344 patients with rheumatoid arthritis, 78% thought that patients should feel free to make everyday decisions about medical problems, but 75% thought that doctors should make important decisions and >50% thought that patients should go along with doctors' decisions even if they disagreed with them [21]. On the other hand, many patients want more information – in the same study, significantly more patients wanted information than wanted to be involved in making decisions. However, up to 80% of information given to patients during medical consultations is forgotten at once, and almost half of what is remembered is remembered incorrectly [22]; measures to improve this would be welcome. More education of clinicians in the practical aspects of prescribing is also desirable [23]. The word of choice to describe a patient's medicine-taking behaviour is ‘adherence’ [5]. The word ‘concordance’, when used to imply a negotiated agreement between prescriber and patient, is misleading and inaccurate. It was invented with excellent intentions [24] but has failed to live up to expectations. We should ditch it and talk instead about what really matters to patients in their conversations with health care professionals – the quality of communication and support that they receive.

Slow gastric emptying and intolerance to nasogastric feeds occurs in up to 40–50% of critically ill patients [1, 2]. Not only does this result in inadequate nutritional support, but it is also a major risk factor for gastro-oesophageal reflux and aspiration, with potential adverse affects on both morbidity and mortality [2, 3]. The currently available therapeutic options for managing delayed gastric emptying and feed intolerance in critically ill patients are: total parenteral nutrient therapy, prokinetic therapy and postpyloric feeding [4, 5]. Of these, treatment with prokinetic agents is efficacious and frequently used as the first-line therapy to avoid the septic complications associated with total parenteral nutrient therapy and the technical demand of postpyloric feeding [4, 5]. Low-dose erythromycin (1–3 mg kg−1), a motilin agonist, is a widely used prokinetic agent in critically ill patients. However, erythromycin has recently been reported to effect human haemodynamics adversely, with a fall in systolic blood pressure (SBP) of 10 ± 9 mmHg 1.5–2.5 h after an oral dose of 300 mg [6]. The mechanism responsible for the hypotension following erythromycin remains unclear. In animals, infusion of motilin was found to induce endothelial-dependent vasodilation and a transient reduction in blood pressure (BP) [7]. As erythromycin leads to an increase in plasma motilin concentrations in humans, it is conceivable that the hypotension associated the drug is related to the endothelium-mediated vasodilation exerted by motilin [8]. Although erythromycin is frequently used in critically ill patients, its haemodynamic effects in this group are unknown. A reduction of 10 mmHg in SBP after erythromycin may not be important in healthy subjects but may be clinically relevant in patients with potential cardiovascular compromise such as the critically ill [6]. This study therefore aimed to examine the acute effects of a single intravenous prokinetic dose of erythromycin on BP) and heart rate (HR) in critically ill patients, who were intolerant to nasogastric feeds and were not on inotropic supportive therapy. Nineteen mechanically ventilated, critically ill patients (age 52 ± 18 years, M:F 10 : 9, Acute Physiology and Chronic Health Evaluation II score 21 ± 6, length of Intensive Care Unit stay 10 ± 9 days) were randomized to either erythromycin (i.v. 200 mg) or placebo (i.v. normal saline) in a cross-over study, 12 h apart. Intolerance to intragastric feeding was defined as vomiting or regurgitation during feeding, or a 6-hourly gastric aspirate > 250 ml [9]. The cut-off gastric aspirate > 250 ml used in this study is a previously published clinical marker of feed intolerance [9, 10]. The majority of patients were sedated with morphine and midazolam and only two patients received propofol alone. Five patients received nasogastric antihypertensive medications (ACE inhibitors n = 3, β-blockers n = 1 and calcium channel blockers n = 1). Continuous lying BP (systolic, diastolic and mean) and HR were measured via an arterial line at 15-min intervals for 60 min before and 180 min after the administration of prokinetic therapy by using an automatic device (TRAM-RAC 4A; General Electric, Milwaukee, WI, USA). A reduction in SBP of >10% from baseline was defined as a clinically significant change. The study was approved by the Local Research Ethics Committee. Each subject gave written informed consent prior to the study. The effects of erythromycin on BP and HR were analysed using a linear mixed effects model fitted to the data. In this model, group status (erythromycin or placebo) and time were treated as fixed effects, while subject was treated as a random effect. Data are expressed as mean ± SD, except in Figure 1, where the error bars represent 95% confidence intervals. A P-value

Pharmacovigilance is principally concerned with the timely discovery of adverse events that are novel in terms of clinical nature, severity, and/or frequency. With the ever-increasing volume of postmarketing safety surveillance data there is considerable interest in using computer-assisted signal detection algorithms [also known as data mining algorithms (DMAs)] to search extremely large spontaneous reporting system (SRS) databases for statistical dependencies between drugs and events in excess of what would be expected if the drug and event were independently distributed in the database [1–5]. If there is sufficient correlation between the statistical dependencies identified by DMAs and demonstrable causal relationships they could improve pharmacovigilance performance. Data mining algorithms include methods based on simple disproportionality analysis [e.g. proportional reporting ratios (PRRs) [1] and reporting odds ratios (RORs) [2]], and those incorporating Bayesian modelling and other statistical adjustments to reduce the volume of ‘signals’ generated [e.g. Bayesian Confidence Propagation Neural Network (BCPNN) [3] and the multiitem gamma-Poisson shrinker (MGPS) [4]]. A very high volume of signals, including false-positive signals due to confounding, can strain pharmacovigilance resources with too many unfruitful hypotheses. However, components of the statistical modelling and adjustments that reduce the volume of signals may also be associated with deleterious consequences by concurrently degrading the capacity of the DMA to detect true-positive signals of real causal associations. Drug-induced pancreatitis is of considerable concern in pharmacovigilance because of the potential for significant morbidity and mortality and the numerous drugs that may be aetiological agents [6–8]. The ability of DMAs to provide meaningful signals of drug-induced pancreatitis is therefore of interest and can be used to illustrate several important principles related to the practical evaluation and deployment of DMAs. Many drugs have been associated with pancreatitis. Published reports and observational studies have asserted definite causal relationships with various drugs including azathioprine, cimetidine, didanosine, erythromycin, furosemide, hydrochlorothiazide, interferon-alpha, mesalazine, methyldopa, metronidazole, olsalazine, oxyphenbutazone, simvastatin, sulfasalazine, tetracycline, and valproate [6–8]. Although residual uncertainty about causality may still remain even in well-documented case reports involving positive rechallenge [8], it provides an interesting and diverse data set for examining the performance of DMAs since the evidentiary support of the reported associations may be quite strong. Applying two DMAs (PRRs and MGPS) using commonly cited threshold criteria and Medical Dictionary for Regulatory Affairs (MedDRA) Preferred Terms specific for pancreatitis, data from the United States Food and Drug Administration were analysed for possible signals of drug-induced pancreatitis with the 16 aforementioned drugs. For comparison, the time to appearance of replicated findings (i.e. ≥2 case reports) of drug-induced pancreatitis in the published literature (another source of signals) was estimated by manual review of citations generated through a search of MEDLINE using the respective drug name and ‘pancreatitis’ as key words. Proportional reporting ratios appeared to outperform MGPS on several counts in this analysis. PRRs highlighted 15 out of the 16 associations compared with MGPS, which highlighted 11 out of 16. Of the associations highlighted by both methods, PRRs provided a signal of disproportionate reporting from 1 to 14 years prior to MGPS. The number of reports required to generate a signal with PRRs ranged from one to 19 compared with four to 62 for MGPS in the applicable cases. The majority of signals highlighted by PRRs (9/15) were based on three or fewer reports while the majority (6/11) highlighted by MGPS required over 20 reports. Proportional reporting ratios highlighted eight out of the 16 associations (50%) from 1 to 16 years in advance of the published literature compared with three out of 16 (19%) highlighted by MGPS 3–7 years prior to the published literature. The results of this analysis illustrate the potential of both methods to improve the process of signal detection and the potential of simple forms of disproportionality analysis to identify potentially meaningful associations that fail to be identified by certain Bayesian methods such as MGPS or to identify signals highlighted by both at significantly earlier time points. The cost of such enhanced sensitivity is an over abundance of signals that may require additional triage criteria for practical implementation. Performance gradients between simple disproportionality analysis (higher ‘sensitivity’, lower ‘specificity’) and certain methods incorporating Bayesian inference and extensive statistical adjustments (lower ‘sensitivity’, higher ‘specificity’) when used in isolation is important information to bear in mind. However, it does not truly reflect ‘real life’ pharmacovigilance practice, since these methods, if used, should only be used as supplements to, not substitutes for, standard signalling strategies. The observed performance differentials are likely to be significantly mitigated when these methods are used as one element of a comprehensive pharmacovigilance programme. Further complicating any comparisons is the lack of specified standardized data mining procedures (e.g. selection and combination of adverse event terms, lack of reliable and consistent criteria for adjudicating causality and expectedness of adverse events, variations in database and dictionary architectures, multiple biases, confounding factors, and data quality limitations inherent in voluntary reporting systems). The crucial question therefore, is the proper positioning of such techniques within the universe of methods that have been historically used for routine signal detection based on various frequency criteria and clinical pharmacological judgement. Most likely the ideal point on the aforementioned performance gradient for decision-making will be highly situation dependent. Specifically, in assessing the potential added value of DMAs and/or selecting a specific algorithm, potential users of DMAs should assess numerous factors, including the rigour of their standard signalling criteria, the nature and numbers of drugs and adverse events reports to be screened, the respective public health impacts of false-positive and false-negative findings, timing in the product life cycle, and resource constraints. It is quite likely that the incremental utility of DMAs may be higher for health authorities, who have statutory obligations for monitoring the safety of all licensed drugs than for individual pharmaceutical companies, whose surveillance activities are more focused in scope. Another important principle illustrated by this analysis is that the importance of sound clinical, pharmacological and epidemiological judgement in the signal detection process is not diminished, and is in fact increased, with the use of DMAs. This is not only because of the false-positive signals frequently observed with DMAs due to various confounding factors and reporting biases. Pancreatitis is an adverse event that may be included in lists of ‘designated medical events’ based on rarity, medical importance, and/or a high level of drug-attributable risk in general. As few as one to three reports of such events may be considered signals under some circumstances [9]. Given the findings of this analysis, over-reliance on DMAs for detecting such events, especially those that down-weight (‘shrink’) signal scores based on small numbers of reports, may be especially hazardous. Further research is needed to determine if DMAs have utility in the detection of designated medical events. More generally, while PRRs identified a greater number of the associations or identified them earlier than MGPS, both failed to highlight a significant proportion of associations either absolutely or relative to the published literatune. Numerous factors may contribute to the absence of a signal of disproportionate reporting of a drug-induced medical event with DMAs including a high background frequency of the drug and event in the database, the overall reported safety profile of the pharmaceutical, and influences on reporting behaviour. This underscores the increased need for clinical judgement when using DMAs and the caveats that the absence of a signal of disproportionate reporting with any DMA is not adequate to refute a signal generated by traditional criteria and that DMAs should never be used in isolation. The converse is not true, in that a potential signal generated by a DMA can possibly be refuted by expert case-level clinical review. A clear understanding of the strengths and weaknesses of the various DMAs, both with respect to traditional signal detection strategies, and relative to each other, will help identify their appropriate application and help minimize their misapplication and misuse.

We read with interest the paper ‘Thiazolidinediones and associated risk of bladder cancer: a systematic review and meta-analysis’ by Turner et al. [1]. The authors present the results of a meta-analysis including a large number of patients. They concluded that pioglitazone use is associated with a modest but significant increase in the risk of bladder cancer, which further increases with both higher cumulative dose (>28 g) and duration of exposure (>12–24 months). Interestingly, the effect of rosiglitazone on bladder cancer risk was not significant [1]. A few comments might be of interest, based on the results of a recent study (published in August 2013), not included in the meta-analysis [2]. This was an observational follow-up study of patients included in the PROspective pioglitAzone Clinical Trial In macroVascular Events (PROactive) trial, who had received pioglitazone for 2.6 years. PROactive had shown a modest but significant reduction in the composite of all cause mortality, non-fatal myocardial infarction (MI) and stroke in high cardiovascular disease (CVD) risk type 2 diabetes mellitus (T2DM) patients [3]. After a mean follow-up time of 8.7 years (including the PROactive period), the CVD benefit was lost, since only 13.5% continued on pioglitazone [2]. There was no evidence of increased risk of bladder cancer in these patients (89% of them were at the maximum dose of 45 mg day−1) [2]. Interestingly, no significant between group differences were observed in the percentage of patients reporting any other malignancies, except for prostate malignancy, a finding that needs further confirmation [3]. The leading cause of death in patients with T2DM is CVD [3]. As PROactive has shown, 40 patients would need to be treated for 3 years in order to prevent one first CVD event. In other words, the number needed to treat is approximately 120 per year [3], while the number needed to treat for harm (that is to develop one case of bladder cancer) shown by the meta-analysis was 3408 for ever pioglitazone use, 1119 for >28 g cumulative dose and 1404 for >24 months cumulative duration [1]. On the contrary, rosiglitazone use has been associated with increased risk for MI (although not for CVD or all cause mortality) in an older meta-analysis, showing one additional MI per 52 patients treated for 5 years [4]. These results were confirmed by a comparative meta-analysis, showing a significant increase in the odds of MI, congestive heart failure and death (170 excess MIs, 649 excess cases of heart failure and 431 excess deaths for every 100 000 patients receiving rosiglitazone rather than pioglitazone) [5]. It seems therefore prudent to prescribe an agent that has been shown to protect the cardiovascular system, for at least 3 years, rather than avoid it for the fear of causing bladder cancer, the possibility of which for such a period seems very low (taking into account that the cumulative dose of 28 g stands for a 31 month period of 30 mg day−1, which is the usual pioglitazone dose). This becomes more important in an era when clinicians, the Food and Drug Administration and the European Medicines Agency are very cautious about the effect of any anti-diabetic and anti-obesity drug on the cardiovascular system [6]. New studies are already being conducted to test the effect of other drug categories, such as glucagon-like peptide (GLP)-1 based therapies, on the cardiovascular system [6]. In conclusion, we believe that pioglitazone should continue to be considered a reliable therapeutic option in patients with T2DM for a period of up to 2 years (with a potential extension to 3 years), since its benefit and safety on the cardiovascular system overweigh the potential risk for bladder cancer. Patients at high risk for bladder cancer, such as those with a family history, microscopic haematuria and exposure to carcinogenic substances, should be excluded.

Pholcodine is a commonly used antitussive medication. In France, its use is limited because of the risk of an allergic cross-reaction with curare. Here we report the case of a 52-year-old man who ingested 750 mg of pholcodine syrup. An emergency medical team found him at home in nonshockable cardiac arrest. Orotracheal intubation was impossible because of instantaneous rigor associated with trismus, so an emergency cricothyrotomy was performed. Return of spontaneous circulation was obtained 5 min later. Cardiac arrest recurred upon arrival at the intensive care unit, leading to death. Toxicological analysis showed a pholcodine blood level of 2500 ng ml−1 (a lethal dose is >1000 ng ml−1, extrapolated from animal studies). Pholcodine is an opioid derivate with a central antitussive effect that is indicated for unproductive cough in adults. No case of fatal intoxication has ever been reported. We hypothesize that cardiac arrest was the result of hypoxia due to respiratory arrest connected to the opioid nature of pholcodine. Interestingly, instantaneous rigor occurred, which should be not confused with rigor mortis. Instantaneous rigor is a rarely reported condition that persists until rigor mortis and disappears in the stage of secondary flaccidity. Instantaneous rigor can lead to trismus, making orotracheal intubation impossible and necessitating emergency cricothyrotomy. This case shows that pholcodine overdose can be lethal and can lead to instantaneous rigor.