Clozapine-induced gastrointestinal hypomotility: presenting features and outcomes, UK pharmacovigilance reports, 1992-2017.

BACKGROUND: Clozapine-induced gastrointestinal hypomotility (CIGH) affects some 75% of patients treated with clozapine. AIMS: To document the incidence of potentially harmful CIGH in the UK. METHOD: We studied spontaneous UK pharmacovigilance reports recorded as clozapine-related gastrointestinal adverse drug reactions, 1992-2017. RESULTS: There were 527 patients reported with potentially harmful CIGH; 33% (n = 172) died. Deaths averaged 1 per year 1992-1999, 5 per year 2000-2009 and 15 per year 2010-2017. Those who died were older (median 52 years v. 49 years) and had been prescribed clozapine for longer than those who recovered (median 11.3 years v. 4.8 years), but there was no difference in prescribed dose. Within the first 4 years of clozapine treatment, there were 169 reports of CIGH, of which 3% (n = 5) were fatal. At 10-14 years there were 63 reports of CIGH, of which 25% (n = 16) were fatal. Among the deaths, males were younger (median 51, range 22-89 v. median 57, range 24-89 years) with higher clozapine doses (median 450, range 100-900 v. median 300, range 12.5-800 mg/d) than females. In non-fatal CIGH, surgery was the most frequent outcome (n = 92). The procedures included appendectomy, ileostomy, total/partial colectomy, colostomy/stoma and proctosigmoidectomy. Clozapine dosage was reduced in 6 patients, stopped and restarted in 23, 'continued' in 6 and discontinued permanently in at least 76 patients. CONCLUSIONS: The risk of serious morbidity/mortality from CIGH is substantial. The need to actively monitor bowel function and give laxatives to patients treated with clozapine is clear.

Clozapine in the treatment of refractory schizophrenia: a practical guide for healthcare professionals.

BACKGROUND: Clozapine remains the only medication licensed for treating refractory schizophrenia. However, it remains underutilized in part due to concerns regarding adverse events. SOURCES OF DATA: Published literature. AREAS OF AGREEMENT: Common adverse events during clozapine treatment include sedation, hypersalivation, postural hypotension, dysphagia, gastrointestinal hypomotility, weight gain, diabetes mellitus and dyslipidaemia. Rare but serious events include agranulocytosis, cardiomyopathy, myocarditis, pneumonia, paralytic ileus and seizure. AREAS OF CONTROVERSY: It remains unclear how best to minimize clozapine-induced morbidity/mortality (i) during dose titration, (ii) from hypersalivation and (iii) from gastrointestinal hypomotility. It is also unclear how clozapine pharmacokinetics are affected by (i) gastrointestinal hypomotility, (ii) systemic infection and (iii) passive exposure to cigarette smoke. Whether monthly haematological monitoring needs to continue after 12 months of uninterrupted therapy is also a subject of debate. GROWING POINTS: There is a need for better management of serious clozapine-related adverse events in addition to agranulocytosis. There is also a need for better education of patients and carers, general practitioners, A&E and ITU staff and others of the problems posed in using clozapine safely. AREAS TIMELY FOR DEVELOPING RESEARCH: There is a need for more research on assessing clozapine dosage (i) as patients get older, (ii) with respect to exposure to cigarette smoke and (iii) optimizing response if adverse events or other factors limit dosage.

Simple methodology for the therapeutic drug monitoring of the tyrosine kinase inhibitors dasatinib and imatinib.

A simple HPLC method has been developed to measure imatinib and N-desmethylimatinib (norimatinib) in plasma or serum at concentrations attained during therapy. Adaptation of this method to LC-MS/MS also allows dasatinib assay. A small sample volume (100 µL HPLC-UV, 50 µL LC-MS/MS) is required and analysis time is <5 min in each case. Detection was by UV (270 nm) or selective reaction monitoring (two transitions per analyte) tandem mass spectrometry. Assay calibration was linear (0.05-10 mg/L imatinib, 0.01-2.0 mg/L norimatinib and 1-200 µg/L dasatinib), with acceptable accuracy (86-114%) and precision (<14% RSD) for both methods. A comparison between whole blood and plasma confirmed that plasma is the preferred sample for imatinib and norimatinib assay. For dasatinib, although whole blood concentrations were slightly higher, plasma is still the preferred sample. Despite considerable variation in the (median, range) plasma imatinib and norimatinib concentrations in patient samples [1.66 (0.02-4.96) and 0.32 (0.01-0.99) mg/L, respectively, N = 104], plasma imatinib was >1 mg/L (suggested target for response) in all but one sample from patients achieving complete molecular response. As to dasatinib, the median (range) plasma dasatinib concentration was 13 (2-143) µg/L (N = 33). More observations are needed to properly assess the potential role of therapeutic drug monitoring in guiding treatment with dasatinib.

To restrict or not to restrict? Practical considerations for optimizing dietary protein interactions on levodopa absorption in Parkinson's disease.

Administration of levodopa for Parkinson's disease (PD) has remained the most effective therapy for symptom management despite being in use for over 50 years. Advancing disease and age, changing tolerability and gastrointestinal (GI) dysfunction may result in change in dietary habits and body weight, as well as unpredictable motor fluctuations and dyskinesias. Dietary proteins which convert into amino acids after digestion are implicated as major factors that inhibit levodopa absorption. For people living with PD (PwP) who experience motor fluctuations, low protein diets (LPD) and protein redistribution diets (PRD) may be effective and are often recommended as a non-pharmacologic approach for improving levodopa bioavailability. However, there is a lack of consensus on a standard definition of these diets and appropriate treatment algorithms for usage. This may be due to the paucity of high-level evidence of LPD and PRD in PwP and whether all or specific subgroups of patients would benefit from these strategies. Managing diet and protein intake with proper education and monitoring may reduce complications associated with these diets such as dyskinesias and unintentional weight loss. Additionally, alterations to medications and GI function may alter levodopa pharmacokinetics. In this narrative review we focus on 1) mechanisms of dietary protein and levodopa absorption in the intestine and blood brain barrier, 2) dietetic approaches to manage protein and levodopa interactions and 3) practical issues for treating PwP as well as future directions to be considered.

An automated method for the simultaneous measurement of azole antifungal drugs in human plasma or serum using turbulent flow liquid chromatography-tandem mass spectrometry.

Azole antifungal drugs are important in the prophylaxis and treatment of invasive aspergillosis. Therapeutic drug monitoring may be indicated to (1) monitor adherence, (2) guide dosage and (3) minimise the risk of drug-drug interactions and dose-related toxicity. TurboFlow(TM) technology offers online, automated sample preparation. An Aria Transcend(TM) TLX-II coupled with a TSQ Vantage(TM) MS was used. Centrifuged samples (25 µL) were mixed with internal standard solution (975 µL) and 30 µL injected directly onto a C18-P-XL TurboFlow column. Analytes were focussed onto a Phenomenex Gemini Phenyl analytical column and eluted using a methanol/water gradient (flow-rate, 0.8 mL/min). Analytes were monitored in selected reaction monitoring mode (two transitions per analyte, positive mode APCI). Calibration ranges were as follows: itraconazole, hydroxyitraconazole, and posaconazole 0.05-5.0 mg/L; voriconazole and fluconazole 0.1-10 mg/L. Total analysis time was 12 min. TurboFlow column recovery was >77% for all analytes. Calibration was linear (R (2) > 0.99) for all analytes. Inter- and intra-assay imprecision (% RSD) was <8% and accuracy (nominal internal quality control values) 90-105% for all analytes. The limit of detection was 0.01 mg/L for all analytes. No matrix effects were observed. This method is simple, robust and suitable for measuring these compounds at concentrations attained during therapy.

An automated method for the measurement of a range of tyrosine kinase inhibitors in human plasma or serum using turbulent flow liquid chromatography-tandem mass spectrometry.

Tyrosine kinase inhibitors (TKIs) are used to treat a number of cancers, including chronic myeloid leukaemia and hepatocellular carcinoma. Therapeutic drug monitoring (TDM) may be indicated to (1) monitor adherence, (2) guide dosage, and (3) minimise the risk of drug-drug interactions and dose-related toxicity. On-line, automated sample preparation provided by TurboFlow technology (ThermoFisher Scientific) in conjunction with the sensitivity and selectivity of tandem mass spectrometry (MS/MS) detection may be applied to the analysis of single drugs and metabolites. We report the use of TurboFlow LC-MS/MS for the analysis of nine TKIs and metabolites (imatinib, N-desmethylimatinib, dasatinib, nilotinib, erlotinib, gefitinib, lapatinib, sorafenib, sunitinib) in human plasma or serum for TDM purposes. An Aria Transcend TLX-II system coupled with a TSQ Vantage was used. Samples (50 µL) were vortex mixed with internal standard solution (150 µL imatinib-D(8), gefitinib-D(8), sunitinib-D(10), and nilotinib-(13)C (2) (15) N(2) in acetonitrile) and, after centrifugation 100 µL supernatant were injected directly onto a 50 × 0.5-mm Cyclone TurboFlow column. Analytes were focussed onto a 50 × 2.1-mm (3 µm) Hypersil GOLD analytical column and eluted with an acetonitrile/water gradient. Analytes were monitored in selected reaction monitoring mode (positive APCI). Total analysis time was 7 min without multiplexing. Calibration was linear (R(2) > 0.99) for all analytes. Inter- and intra-assay precision (in percent relative standard deviation, RSD) was <11 % and accuracy 89-117 % for all analytes. No matrix effects were observed. This method is suitable for high-throughput TDM in patients undergoing chronic therapy with TKIs and has been utilised in the analysis of clinical samples.

Methadone-related death in detention.

Oral methadone may be prescribed to detainees with the aim of minimising the risk of fatal opioid poisoning on release. To study the circumstances under which methadone-related deaths can occur in detention, we audited reports of 17 [14 male, 3 female; median (range) age 34 (22-52) years] such deaths, July 2010-December 2011. The median (range) methadone dose was 40 (10-110) mg/d (N = 16). The median (range) post-mortem blood methadone concentration was 0.42 (0.16-1.40) mg/L. Those who died within 7 days of the commencement of methadone treatment were significantly younger (Mann-Whitney U 102.5, p < 0.05), were prescribed a significantly lower dose (U = 80.0, p < 0.05) and had significantly lower blood methadone concentrations at death (U = 106.5, p < 0.02) than in those given methadone long-term. In 8 reports the prisoner had been recorded as either 'sleepy' (N = 7), or 'unwell' in the hours before death. In 13 deaths, the prisoner was either found dead first thing in the morning, or in one instance could not be roused ('snoring heavily'). Pneumonia, tracheobronchitis, end-stage cirrhosis, and ischaemic heart disease/coronary artery atherosclerosis were cited as associated factors in four patients, all of whom were on long term stable methadone treatment. Attention to warning signs of likely methadone toxicity (daytime or excessive drowsiness, snoring, nausea/vomiting) and associated risk factors (use of drugs such as benzodiazepines and gabapentinoids, the presence of respiratory infection, liver or renal disease) could help minimise the risk of unexpected death in patients given methadone.

Basic drug analysis by strong cation-exchange liquid chromatography-tandem mass spectrometry: simultaneous analysis of amisulpride, and of metamfetamine and amfetamine in serum/plasma.

In the HPLC of basic drugs and metabolites, good efficiency and peak shape can often be attained using strong cation-exchange packings with isocratic 100% methanol eluents containing an ionic modifier at an appropriate pH* and ionic strength. Solvent extracts can be analysed directly, and use of ammonium acetate as modifier facilitates the use of atmospheric pressure chemical ionization (APCI)-tandem mass spectrometry, selected reaction monitoring mode. For the analysis of amisulpride and of metamfetamine/amfetamine in plasma (200 µL) after single oral doses in man, a column packed with Waters Spherisorb S5SCX (5 µm average particle size, 100 × 2.1 mm i.d.) was used with methanolic ammonium acetate (40 mmol/L, pH* 6.0, flow rate 0.5 mL/min) as eluent (35°C). Deuterated internal standards were used for each analyte. Detection was by positive-mode APCI. Responses for all analytes were linear over the calibration ranges. Intra-assay precision (RSD) was 2-18%, and inter-assay precision was 2-12%. The limit of detection was 0.5 µg/L for all analytes. No significant matrix effects or isobaric interferences were noted. The total analysis time was 7 min. Similar methodology can be applied to a wide range of basic analytes using MS/MS detection.

Software-based planning of ultrasound and CT-guided percutaneous radiofrequency ablation in hepatic tumors.

OBJECTIVES: Radiofrequency ablation (RFA) can be associated with local recurrences in the treatment of liver tumors. Data obtained at our center for an earlier multinational multicenter trial regarding an in-house developed simulation software were re-evaluated in order to analyze whether the software was able to predict local recurrences. METHODS: Twenty-seven RFA ablations for either primary or secondary hepatic tumors were included. Colorectal liver metastases were shown in 14 patients and hepatocellular carcinoma in 13 patients. Overlap of the simulated volume and the tumor volume was automatically generated and defined as positive predictive value (PPV) and additionally visually assessed. Local recurrence during follow-up was defined as gold standard. Sensitivity and specificity were calculated using the visual assessment and gold standard. RESULTS: Mean tumor size was 18 mm (95% CI 15-21 mm). Local recurrence occurred in 5 patients. The PPV of the simulation showed a mean of 0.89 (0.84-0.93 95% CI). After visual assessment, 9 incomplete ablations were observed, of which 4 true positives and 5 false positives for the detection of an incomplete ablation. The sensitivity and specificity were, respectively, 80% and 77% with a correct prediction in 78% of cases. No significant correlation was found between size of the tumor and PPV (Pearson Correlation 0.10; p = 0.62) or between PPV and recurrence rates (Pearson Correlation 0.28; p = 0.16). CONCLUSIONS: The simulation software shows promise in estimating the completeness of liver RFA treatment and predicting local recurrence rates, but could not be performed real-time. Future improvements in the field of registration could improve results and provide a possibility for real-time implementation.

Clozapine: more than 900 mg/day may be needed.

Patients may fail to respond to clozapine treatment despite use of the maximum licensed UK dosage (900 mg/day) because of ultra-rapid metabolism of the drug. We present the findings of a study of a national clozapine/norclozapine assay service for the period 1997-2005 and three individual case studies of patients treated with clozapine in doses greater than 900 mg/day. Clinicians should be alert to the possibility of treatment failure because of rapid clozapine clearance secondary to genetic factors and heavy cigarette consumption. This may necessitate the use of clozapine in doses up to 1400 mg/day, notably in young male smokers. Doses of greater than 900 mg/day are rarely justified in women. Anyone given relatively high-dose clozapine (600 mg/day or more) should be monitored regularly for adverse events and changes in smoking habit.

A petition to Mr Peel: Gideon Mantell and the trial of Hannah Russell.

In the summer of 1826, Hannah Russell was tried for petty treason, viz. the murder of her husband, Benjamin Russell, by poisoning. Their lodger, Daniel Leney, was indicted as her accomplice. The exact circumstances surrounding the death were unclear but Hannah was known to have purchased white arsenic (arsenious oxide). A local surgeon, Thomas Evans, supported at the post-mortem examination by two further surgeons, not only reported severe corrosion of the gastrointestinal tract, but also the recovery of nearly an eighth of an ounce of arsenic from the victim's stomach. Both accused were convicted and sentenced to death. Leney was executed, but Hannah Russell was respited because the trial judge, Sir Robert Graham, had doubts as to a direction he had given to the jury. The surgeon and paleontologist Gideon Mantell took up her case, stressing that death from arsenic could not have taken place as quickly as was alleged and maintaining that the chemical evidence of arsenic poisoning was inconclusive. He gained the support of some eminent chemists and physicians. Subsequently, forensic toxicologists [Sir] Robert Christison and Alfred Swaine Taylor pointed out that Mantell's arguments as to the possible time to death in arsenic poisoning were quite wrong. Moreover, Evans gave details of the analyses he and his colleagues had undertaken to Christison, who pronounced the findings sound, as indeed did Mantell after Evans and his colleagues published details of their investigations in the Sussex Advertiser. Papers in The National Archives show that Hannah was pardoned for the offence for which she was indicted, leaving it open to prefer a lesser charge. That this was never done may have been due to Mantell's campaign, at least in part, but the pardon she did receive was due to the concern of the trial judge as to the implications of the evidence presented at trial.

Plasma and Urine Levamisole in Clinical Samples Containing Benzoylecgonine: Absence of Aminorex.

Aminorex has been reported as a metabolite of levamisole in man, but data on the aminorex concentrations in clinical samples are scant. We thus measured levamisole, aminorex and benzoylecgonine in urine, and levamisole and aminorex in plasma using achiral liquid chromatography-high resolution mass spectrometry. Centrifuged urine (50 µL) was diluted with LC eluent containing internal standard (benzoylecgonine-D3, 25 µg/L) (450 µL). For plasma, sample (200 µL) and Tris solution (2 mol/L, pH 10.6, 100 µL) were added to a 60.5 × 7.5 mm i.d. glass test tube. Internal standard solution (ketamine-D4, 200 µg/L) (10 µL) was added and the tube contents vortex-mixed (5 s). Butyl acetate:butanol (9 + 1, v/v; 200 µL) was added and after vortex-mixing (30 s) and centrifugation (13,680 × g, 4 min), the extract was evaporated to dryness and reconstituted in 10 mmol/L aqueous ammonium formate containing 0.1% (v/v) formic acid (150 µL). Prepared samples and extracts (100 µL) were analyzed using an AccucoreTM Phenyl-Hexyl column (2.6 mm a.p.s., 100 × 2.1 mm i.d.) maintained at 40°C. MS detection was in positive mode using heated electrospray ionization (ThermoFisher Q-ExactiveTM). Intra- and inter-assay accuracy and precision were ±20%, and ≤11%, respectively, for all analytes in both matrices. Lower limits of quantitation were 0.1 and 1 µg/L (all analytes) in plasma and urine, respectively. Of 100 consecutive urine samples submitted for drugs of abuse screening containing benzoylecgonine, levamisole was detected in 72 (median 565, range 4-72,970 µg/L). Levamisole was also measured in eight plasma samples (median 10.6, range 0.9-64.1 µg/L). A number of metabolites of levamisole (4-hydroxylevamisole, levamisole sulfoxide, levamisole glucuronide, and hydroxylevamisole glucuronide) were tentatively identified in urine. Neither aminorex, nor any of its reported metabolites were detected in any sample.

Where next for youth mental health services in Ireland?

To date, Ireland has been a leading light in the provision of youth mental health services. However, cognisant of the efforts of governmental and non-governmental agencies working in youth mental health, there is much to be done. Barriers into care as well as discontinuity of care across the spectrum of services remain key challenges. This editorial provides guidance for the next stage of development in youth mental care and support that will require significant national engagement and resource investment.

Volatile substance abuse and crime: data from U.K. press cuttings 1996-2007.

Volatile substance abuse (VSA, solvent abuse, 'glue sniffing'), carries a risk of sudden death (some 700 deaths in the UK, 1996-2006). However, mortality data take no account of the social cost of the habit. From press cuttings we have identified 508 instances (569 individuals: 507 male, median age 25 yr, range 8-51 yr and 62 female, median age 18 yr, range 11-36 yr) where VSA, either alone or together with alcohol/other drugs, was reported in association with criminal or antisocial behaviour that resulted in a criminal conviction or caution. The frequency of reports decreased from 84 per annum (1997 and 1998) to 20 (2007). The agents reported (17 individuals, two agents) were 'glue' (225), LPG/'butane'/aerosol propellants (176), 'solvents' (158), and petrol (gasoline) (27). The offences cited (most serious crime) were: homicide (35), rape or other sexual assault (34), arson (25), assault or serious threat of assault (192), child neglect/cruelty (6), attempting to pervert the course of justice (2), criminal damage (41), burglarylrobbery/theft/shoplifting (100), nuisance/ breach of the peace/breach of antisocial behaviour order (104), driving whilst impaired and other vehicle-related offence (22), and supply (non-retail) (8). Thirty offenders were given life sentences or detained indefinitely under mental health legislation. Reports came from all parts of the UK, although most were from Northern England, Northern Ireland, and Scotland. There were many reports of recidivists; one 34-year-old male had made 113 court appearances, and had spent approximately nine years in custody. Although there are severe limitations to data derived from press cuttings and notwithstanding that in some cases VSA may have been raised in mitigation, these data provide an additional insight into the problem posed by VSA in the U.K.

Cut Costs at All Costs!

The UK Government decision to close its Forensic Science Service (FSS) in 2010 left the criminal justice system in England and Wales bereft of impartial, high-level scientific support. The private sector was entrusted to fill the gap and to ensure that all results were accurate, timely, fit for purpose, easy to interpret, and above all gave value for money. In the event, however, a major provider has collapsed necessitating a rescue deal to minimise the impact of the fiasco. Moreover, there have been allegations of data manipulation in another private sector laboratory and possible falsification of evidence in a laboratory set up by a police force in an attempt to fill the gap left by the FSS. As to the future, appropriate laboratory regulation and inspection clearly has a part to play, but ironically 'quality management' adds an unnecessary and ever-increasing cost burden that may detract from quality. What is really needed are systems that combine public service and professional integrity with research and development. Involving investigators, coroners/medical examiners/judges, and prosecution and defence lawyers in educational fora would help build cross-professional co-operation and understanding.

Postmortem biochemistry: Current applications.

The results of biochemical analyses in specimens obtained postmortem may aid death investigation when diabetic and alcoholic ketoacidosis is suspected, when death may have been the result of drowning, anaphylaxis, or involved a prolonged stress response such as hypothermia, and in the diagnosis of disease processes such as inflammation, early myocardial infarction, or sepsis. There is often cross-over with different disciplines, in particular with clinical and forensic toxicology, since some endogenous substances such as sodium chloride, potassium chloride, and insulin can be used as poisons. The interpretation of results is often complicated because of the likelihood of postmortem change in analyte concentration or activity, and proper interpretation must take into account all the available evidence. The unpredictability of postmortem changes means that use of biochemical measurements in time of death estimation has little value. The use of vitreous humour is beneficial for many analytes as the eye is in a physically protected environment, this medium may be less affected by autolysis or microbial metabolism than blood, and the assays can be performed with due precaution using standard clinical chemistry analysers. However, interpretation of results may not be straightforward because (i) defined reference ranges in life are often lacking, (ii) there is a dearth of knowledge regarding, for example, the speed of equilibration of many analytes between blood, vitreous humour, and other fluids that may be sampled, and (iii) the effects of post-mortem change are difficult to quantify because of the lack of control data. A major limitation is that postmortem vitreous glucose measurements are of no help in diagnosing antemortem hypoglycaemia.

Antipsychotic-related fatal poisoning, England and Wales, 1993-2013: impact of the withdrawal of thioridazine.

CONTEXT: Use of second generation antipsychotics in England and Wales has increased in recent years whilst prescription of first generation antipsychotics has decreased. METHODS: To evaluate the impact of this change and of the withdrawal of thioridazine in 2000 on antipsychotic-related fatal poisoning, we reviewed all such deaths in England and Wales 1993-2013 recorded on the Office for National Statistics drug poisoning deaths database. We also reviewed antipsychotic prescribing in the community, England and Wales, 2001-2013. Use of routine mortality data: When an antipsychotic was recorded with other drug(s), the death certificate does not normally say if the antipsychotic caused the death rather than the other substance(s). A second consideration concerns intent. A record of "undetermined intent" is likely to have been intentional self-poisoning, the evidence being insufficient to be certain that the individual intended to kill. A record of drug abuse/dependence, on the other hand, is likely to have been associated with an unintentional death. Accuracy of the diagnosis of poisoning: When investigating a death in someone prescribed antipsychotics, toxicological analysis of biological samples collected post-mortem is usually performed. However, prolonged attempts at resuscitation, or diffusion from tissues into blood as autolysis proceeds, may serve to alter the composition of blood sampled after death from that circulating at death. With chlorpromazine and with olanzapine a further factor is that these compounds are notoriously unstable in post-mortem blood. Deaths from antipsychotics: There were 1544 antipsychotic-related poisoning deaths. Deaths in males (N = 948) were almost twice those in females. For most antipsychotics, the proportion of deaths in which a specific antipsychotic featured either alone, or only with alcohol was 30-40%, but for clozapine (193 deaths) such mentions totalled 66%. For clozapine, the proportion of deaths attributed to either intentional self-harm, or undetermined intent was 44%, but for all other drugs except haloperidol (20 deaths) the proportion was 56% or more. The annual number of antipsychotic-related deaths increased from some 55 per year (1.0 per million population) between 1993 and 1998 to 74 (1.5 per million population) in 2000, and then after falling slightly in 2002 increased steadily to reach 109 (1.9 per million population) in 2013. Intent: The annual number of intentional and unascertained intent poisoning deaths remained relatively constant throughout the study period (1993: 35 deaths, 2013: 38 deaths) hence the increase in antipsychotic-related deaths since 2002 was almost entirely in unintentional poisoning involving second generation antipsychotics. Clozapine, olanzapine, and quetiapine were the second generation antipsychotics mentioned most frequently in unintentional poisonings (99, 136, and 99 deaths, respectively). Mentions of diamorphine/morphine and methadone (67 and 99 deaths, respectively) together with an antipsychotic were mainly (84 and 90%, respectively) in either unintentional or drug abuse-related deaths. Deaths and community prescriptions: Deaths involving antipsychotics (10 or more deaths) were in the range 11.3-17.1 deaths per million community prescriptions in England and Wales, 2001-2013. Almost all (96%) such deaths now involve second generation antipsychotics. This is keeping with the increase in annual numbers of prescriptions of these drugs overall (<1 million in 2000, 7 million in 2013), largely driven by increases in prescriptions for olanzapine and quetiapine. In contrast, deaths involving thioridazine declined markedly (from 40 in 2000 to 10 in 2003-2013) in line with the fall in prescriptions for thioridazine from 2001. CONCLUSIONS: The removal of thioridazine has had no apparent effect on the incidence of antipsychotic-related fatal poisoning in England and Wales. That such deaths have increased steadily since 2001 is in large part attributable to an increase in unintentional deaths related to (i) clozapine, and (ii) co-exposure to opioids, principally diamorphine and methadone.

Fatal poisoning with antipsychotic drugs, England and Wales 1993-2002.

Prescription of atypical antipsychotics has increased in recent years. There have also been changes in the guidance on using older drugs, particularly the restriction in the use of thioridazine. We analysed deaths due to poisoning involving antipsychotics in England and Wales, 1993-2002, by age, sex, intent, and agents involved. We also studied antipsychotic prescribing in the community and poisoning deaths in England. Deaths attributed to adverse reactions in the course of normal treatment were not studied because these deaths are not classified as 'poisonings'. The number of deaths involving antipsychotics increased from around 55 per year 1993-1998 to 74 in 2000, and then fell to 53 in 2002. Around 25% of deaths had a verdict of accidental death and in about 60% of deaths a verdict of suicide or an open verdict was recorded. There were no deaths involving thioridazine in 2002, following its removal from use in 2001. However, the number of deaths associated with atypicals, most notably olanzapine and clozapine, has increased. Age-specific death rates were highest in those aged 30-39 and 40-49 years, and were very low in those aged under 20 and 70 or over. Death rates in males were greater than in females. For many drugs the proportion mentions either alone or with ethanol, was 25-45%, but for clozapine and olanzapine such mentions totalled 65-69%. Deaths per million prescriptions (clozapine excluded) were highest for quetiapine (31.3 per million), chlorpromazine (29.4 per million) and thioridazine (15.5 per million).

Fatal poisoning in childhood, England & Wales 1968-2000.

We analysed deaths certified as due to poisoning in England & Wales, 1968-2000, in children aged <10 years by age, sex, circumstances of death, intent, and agents involved. The number of deaths fell from 165 (20.6 per million children) in 1968 to 30 (4.6 per million) in 2000, a decrease of approximately 80%. The age-specific death rates were similar in boys and girls. The rate was initially much higher, and fell more, in those aged <5 years. Most deaths (n=1923) occurred in fires, and had been attributed to inhaling combustion products. A small number (n=104) occurred in fires resulting from motor vehicle and other transport accidents. From 1979 (use of ICD-9) the coding of some of these deaths changed from poisoning with carbon monoxide to poisoning with 'other gases, fumes or vapours'. These 'fire deaths' do not appear as poisonings in mortality statistics based on a single underlying cause of death, and cannot be tabulated as poisoning in many countries. Fire deaths and deaths coded to accidental, deliberate, or undetermined poisoning (n=702) decreased substantially with time, and by 2000 numbered 14 and 10, respectively. Accidental deaths declined from 151 in 1968 to 23 in 2000, but homicides and open verdicts varied from 5 to 20 per year, with no clear trend. Deaths attributed to carbon monoxide and to 'other gases, fumes or vapours' (mostly fire-related) totalled 2431 (84% of all poisoning deaths). Overall, 10% of these deaths were either certified as homicides or open verdicts. However, homicide or open verdict was recorded in half of the 47 fatal opiate poisonings. Opioids have now superseded antidepressants as the commonest agents encountered in fatal poisoning with drugs in children.

Suspected clozapine poisoning in the UK/Eire, 1992-2003.

OBJECTIVE: Toxicological analyses are often performed to investigate suspected poisoning, but the interpretation of results may not be straightforward. We studied suspected poisoning cases 1992-2003 where blood clozapine and N-desmethylclozapine (norclozapine) were measured in order to assess the relationship of these parameters to outcome. METHODS: Samples were referred from clinicians, pathologists/coroners, or via the Clozaril Patient Monitoring Service (CPMS, Novartis). Information was gathered from clinical, post-mortem, or coroners' reports. RESULTS: There were seven fatal [five male, two female; median (range) age 28 (24-41) year] and five non-fatal [four male, one female; median age 35 (26-41) year] clozapine overdoses. The median post-mortem blood clozapine and norclozapine concentrations were 8.2 (3.7-12) and 1.9 (1.4-2.4)mg/L, respectively [median clozapine:norclozapine ratio 4.4 (2.9-5.1)]. The median plasma clozapine and norclozapine concentrations (first or only sample) were 3.9 (1.7-7.0) and 0.40 (0.30-0.70)mg/L, respectively [median clozapine:norclozapine ratio 7.6 (5.3-18)] in the remainder. These overdoses were in patients who were poorly or non-adherent to clozapine, or who had taken tablets prescribed for someone else. In 54 further people who died whilst receiving clozapine [38 male, 16 female; median age 41 (22-70) year], the median post-mortem blood clozapine and norclozapine concentrations were 1.9 (0-7.7, n = 43) and 1.4 (0-6.0, n = 39)mg/L, respectively [median clozapine:norclozapine ratio 1.5 (0.4-7.6, n = 38)]. The median post-mortem increase in blood clozapine and norclozapine as compared to the most recent ante-mortem measurement was 489 (98-5,350)% and 371 (139-831)%, respectively [median sample time before death 14 (0-30, n = 21) days]. CONCLUSION: Clozapine poisoning cannot be diagnosed on the basis of blood clozapine and norclozapine concentrations alone. The analysis of ante-mortem blood specimens collected originally for white cell count monitoring and the blood clozapine:norclozapine ratio may provide additional interpretative information.