Miran Brvar

Potential drug-drug interactions and admissions due to drug-drug interactions in patients treated in medical departments

ZusammenfassungZIEL DER STUDIE: Unerwünschte Nebenwirkungen als Folge von Medikamenteninteraktionen spielen eine wichtige Rolle bei der Sicherheit von Medikamenten. Ziel dieser Studie war es, mögliche Medikamenteninteraktionen bei der Spitalsaufnahme und -entlassung zu erfassen, bzw. die Aufnahmen infolge solcher Interaktionen auf internen Abteilungen eines primären Stadtspitals und tertiären Referenzspitals zu evaluieren. METHODEN: Bei 520 zufällig ausgesuchten Patienten des medizinischen Zentrums der Universität Ljubljana wurden Alter, Geschlecht, Nieren- und Leberfunktion, Medikamente, Diagnosen sowie Dringlichkeit und Grund der Zuweisung retrospektiv erhoben. Bei den Patienten, bei denen Information über die bei Aufnahme und Entlassung eingenommenen Medikamente vorlag, wurde auf Medikamenteninteraktionen mittels des Interaktions-Screening Programms Drug-Reax geprüft. Basierend auf den erhobenen Daten wurde der Anteil an Aufnahmen, deren Grund eine Medikamenteninteraktion war, geschätzt. ERGEBNISSE: Bei 14,6% (76/520) Patienten war die Information über die Medikamentennamen bei Aufnahme nicht vollständig. Nach der Behandlung wurden 416 Patienten nach Hause entlassen, bei 52 von diesen (12,5%) war die Information über die Medikamentennamen im Entlassungsbrief ebenfalls unvollständig. Die restlichen 323 Patienten mit kompletter Information wurden in die Studie inkludiert. Mindestens eine Nebenwirkung wegen potentieller Medikamenteninteraktion hatten 51% (166/323) der Patienten bei Aufnahme und 63% (204/323) bei Entlassung (p = 0,001). Schwere potentiell interaktionsbedingte Nebenwirkungen hatten 13% (41/323) der Patienten bei Aufnahme und 18% (59/323) bei Entlassung (p = 0,001). Die gleichzeitige Gabe von ACE-Hemmern mit Spironolacton war die häufigste Interaktion. (20% der Interaktionen bei Aufnahme bzw. 25,6% bei Entlassung). Nur 2,4% (4/166) der Patienten mit potentiell interaktionsbedingter Nebenwirkungen wurden unter dieser Diagnose zugewiesen. 1,2% (2/323) aller Patienten wurden wegen Medikamenteninteraktion aufgenommen. SCHLUSSFOLGERUNGEN: Die Information über die Medikation der Patienten bei Spitals-Aufnahme und -entlassung ist nicht vollständig. Bei Aufnahme hatte die Hälfte und bei Entlassung zwei Drittel der Patienten potentiell interaktionsbedingte Nebenwirkungen. Unerwünschte durch Interaktionen von Medikamenten bedingte Nebenwirkungen machten 1,2% der Zuweisungen zur Aufnahme auf internen Abteilungen eines primären Stadtspitals und tertiären Referenzzentrums aus.SummaryPURPOSE: Adverse drug reactions due to drug-drug interactions (DDIs) are important in drug safety. The aim of this study was to check potential DDIs (pDDIs) on hospital admission and discharge and to evaluate admissions due to DDIs in medical departments of a primary city and tertiary referral hospital. METHODS: Age, sex, presence of renal and liver failure, drug information, diagnosis, and urgency and reason for admission were retrospectively recorded in 520 randomly selected patients in medical departments of the University Medical Center Ljubljana. The screening program Drug-Reax was used to check for pDDIs in patients with drug information on both admission and discharge home, and the proportion of patients admitted as the consequence of a DDI was estimated. RESULTS: Overall, 14.6% (76/520) of patients had incomplete information on drug names in their medical documentation on admission; at the end of treatment 12.5% (52/416) of patients were discharged home with incomplete information on drug names in their discharge letters. A total of 323 patients had complete information on drug names on both admission and discharge and were included in the analysis of pDDIs: 51% (166/323) of patients on admission and 63% (204/323) on discharge had at least one pDDI (P = 0.001). Major pDDIs were found in 13% (41/323) of patients on admission and 18% (59/323) on discharge (P = 0.001). An ACE inhibitor combined with spironolactone was the most common major pDDI, representing 20.0% of all pDDIs on admission and 25.6% on discharge. Among patients with pDDI on admission, 2.4% (4/166) of were admitted because of an ADR caused by a DDI. Overall, 1.2% (4/323) of patients were admitted as the consequence of a DDI. CONCLUSIONS: The information on patient medication on hospital admission and discharge is incomplete. Half of patients on admission and almost two-thirds on discharge had pDDIs. ADRs due to DDIs caused 1.2% of admissions to medical departments in Ljubljanas primary city and tertiary referral hospital.

Haemodialysis clearance of baclofen.

Background Baclofen is a centrally acting gamma-aminobutyric acid agonist used for spasticity of spinal origin and mainly excreted unchanged by the kidneys. We report haemodialysis clearance and the haemodialysis removal rate constant of baclofen in a comatose patient with baclofen overdose due to acute renal failure.Case report A 60-year-old man with spastic tetraplegia on chronic baclofen therapy was admitted due to pneumonia and acute renal failure. The patient became comatose and, as a result of the baclofen dosage being left unchanged despite a deterioration leading to renal failure due to hypotension, the concentration of baclofen was determined to be in the toxic range (0.70xa0mg/L). During a 4-hour-long bicarbonate haemodialysis the patient woke up and became completely orientated and cooperative. Baclofen therapy was subsequently stopped, and the patient remained conscious. The pharmacokinetics calculations revealed a baclofen haemodialysis removal rate constant of 0.152xa0h-1 and a haemodialysis clearance of 2.14xa0mL/s.Conclusions Patients on a stable baclofen regime can develop baclofen toxicity due to acute renal failure. Haemodialysis removes baclofen as effectively as normal kidneys, and it would appear that haemodialysis is a reasonable treatment modality in patients with accidental baclofen overdose due to acute renal failure.

Poisoning with 1-propanol and 2-propanol.

1-Propanol and 2-propanol are isomers of an alcohol with three carbons. They are colorless liquids with a sweet odor. 1-Propanol is metabolized by alcohol dehydrogenase to propionic acid and presents with metabolic acidosis and elevated anion gap, whereas 2-propanol is metabolized by alcohol dehydrogenase to acetone and presents with rapidly developing (within 3–4 h after exposure) ketosis and ketonuria but without metabolic acidosis. We report a patient who simultaneously ingested a lethal dose of 1-propanol and 2-propanol as a hand disinfectant in hospital. The patient lost consciousness and stopped breathing within half an hour after ingestion. He was intubated and artificially ventilated. Initial laboratory results showed mixed acidosis with elevated anion gap, but ketonuria appeared only 12 h after admission and 6 h following the regaining of consciousness. Therefore, laboratory results in simultaneous poisoning with two isomers of alcohol are not just a sum of laboratory results obtained in isolated poisoning with each isomer because they influence each other’s metabolism: 1-propanol retards the metabolism of 2-propanol to acetone. In conclusion, 1-propanol and 2-propanol poisoning presents early with mixed acidosis and elevated anion gap and only later with ketonuria.

Poisoning with insulin glargine.

Insulin glargine (Lantus) is a long-acting recombinant human insulin analog used for basal insulin therapy. It differs from human insulin only in the replacement of amino acid aspargine at position A21 by glycine, and the addition of two arginines to the C-terminus of the B-chain. In subcutaneous tissue it forms microprecipitates from which small amounts of insulin glargine are slowly released. The activity of insulin glargine is relatively constant over 24 hours with no pronounced peak, allowing once daily dosing (1). Searching Medline, we found no publications regarding poisoning with insulin glargine. A 21-year-old woman with insulin-dependent diabetes mellitus treated with Humalog and insulin glargine (13 U/day) presented with hypoglycemia. On the day of admission at 7 p.m. she had normal dinner and took her regular mealtime dose of Humalog (6 U). At 9 p.m. she injected 26 U of the insulin glargine into her leg in a suicide attempt. Two and a half hours later she was found comatose and brought to our Emergency Department. On arrival, her vital signs were tympanic temperature 35.3 C, respiratory rate 17 breaths/min, pulse 68 beats/min and blood pressure (supine) 110/70 mmHg. She had mydriasis and a Glasgow coma scale of 3, but her remaining physical examination was otherwise unremarkable and no focal neurological signs were detected. Her initial laboratory test results were serum sodium 142 mmol/L, potassium 3.3 mmol/L, glucose 1.7 mmol/L, urea 1.9 mmol/L and creatinine 56 mmol/L. An ECG revealed a normal sinus rhythm. She awoke after an intravenous bolus of 50 ml 50% glucose. A continuous infusion of 10% glucose with potassium was started at a rate of 250 ml/h. Serum glucose was measured hourly and intermittent hypoglycemia with neurological signs requiring treatment with 50 ml of 50% glucose or 1–2 deciliters of sugared tea was recorded 15 times during the subsequent hospitalization (Fig. 1). The last episode of hypoglycemia was detected 53 hours after insulin injection and the glucose infusion was discontinued 7 hours later (Fig. 1). Subcutaneous insulin was reinstituted 72 hours after the insulin poisoning. Analysis by gas chromatography coupled to mass spectrometry revealed no drugs in the patient’s blood and urine samples. The insulin levels were not measured. She was discharged after psychiatric evaluation on the fourth day. In this case, prolonged hypoglycemia developed 2 1/2 hours after subcutaneous injection of only twice the standard insulin glargine dose. The duration of hypoglycemia in intermediate-acting neutral protamine human (NPH) insulin poisoning is reported to be dose-dependent, and hypoglycemia lasting more than 54 hours was observed only after massive NPH insulin poisoning (2). The mechanism of prolonged hypoglycemia due to insulin glargine poisoning is not known, since there has been no published report of insulin glargine release from the subcutaneous tissue in overdose. However, after a standard subcutaneous dose of radiolabelled insulin glargine, residual radioactivity over the injection site was observed more than 48 hours and subcutaneous absorption of insulin glargine was slower than subcutaneous absorption of NPH insulin (3,4). Therefore, the prolonged hypoglycemic effect of insulin glargine in overdose could be the result of delayed dose-dependant subcutaneous absorption of insulin glargine. Larger injections of insulin glargine in one place may produce a prolonged action due to the ‘‘depot’’ effect with slow release of insulin from the injection site as described for NPH insulin (3), but this hypothesis still needs confirmation. In conclusion, hypoglycemia appeared in a patient within 2/2 hours after poisoning with twice the standard dose of longacting insulin analog glargine and lasted more than 2 days.

S100B protein in conscious carbon monoxide-poisoned rats treated with normobaric or hyperbaric oxygen.

Objective:To evaluate S100B, an astroglial structural protein, during normobaric and hyperbaric oxygen therapy of conscious carbon monoxide (CO)-poisoned rats. So far, the usefulness of hyperbaric oxygen therapy in conscious CO-poisoned patients has been shown with neuropsychological testing. The S100B protein has been demonstrated as a possible biochemical marker and prognostic parameter in CO-poisoned rats. Design:Randomized, controlled interventional trial. Setting:University laboratory. Subjects:Male Wistar rats weighing 254 ± 14 g. Interventions:The rats were exposed to a mixture of 3,000 ppm CO in air for 60 mins. After CO exposure, the first group of eight conscious rats was exposed to ambient air for 30 mins, the second group of six conscious rats was exposed to 100% normobaric oxygen for 30 mins, and the third group of six conscious rats was exposed to 100% hyperbaric oxygen at 3 bars for 30 mins. Blood samples were taken from the jugular vein just before CO exposure and immediately after oxygen therapy. The level of consciousness was evaluated at the end of exposure, and the survival rate was monitored for 14 days. The S100B concentrations were measured with a commercial immunoluminometric assay. Measurements and Main Results:Analyses of differences in S100B levels between different kinds of therapy before and after treatment showed a global significant difference (p = .002). The post hoc test results showed that S100B levels after therapy of the first group treated with ambient air (0.16 ± 0.07 &mgr;g/L) and the second group treated with normobaric oxygen (0.19 ± 0.05 &mgr;g/L) were similar (p = .741), and both of them were significantly different, with much higher values of S100B levels after therapy, from the third group treated with hyperbaric oxygen (0.06 ± 0.03 &mgr;g/L; p = .018 and p = .002, respectively). All the rats survived. Conclusions:S100B is elevated in conscious CO-poisoned rats left on ambient air or treated with normobaric oxygen, but not in conscious CO-poisoned rats treated with hyperbaric oxygen.

Accidental poisoning with Veratrum album mistaken for wild garlic (Allium ursinum)

Introduction.u2003Veratrum album (white or false hellebore) is a poisonous plant containing steroidal alkaloids that cause nausea, vomiting, headache, visual disturbances, paresthesia, dizziness, bradycardia, atrioventricular block, hypotension, and syncope. It is regularly mistaken for Gentiana lutea (yellow gentian). We report accidental poisoning with V. album mistaken for Allium ursinum (wild garlic), a wild plant used in soups and salads in Central Europe. Case series.u2003Four adults (24–45 years) accidentally ingested V. album mistaken for A. ursinum in self-prepared salads and soups. Within 15–30u2009min of ingestion they developed nausea, vomiting, and abdominal pain. At the same time dizziness, tingling, dimmed and jumping vision, transient blindness, and confusion appeared. On arrival at the ED, all patients had sinus bradycardia and hypotension. Following treatment the patients were discharged well 24–48u2009h after ingestion. Conclusion.u2003In patients presenting with gastrointestinal, neurological, and cardiovascular symptoms a history of wild plant ingestion suggests possible poisoning with V. album mistaken for wild garlic.

Antivenom for European Vipera species envenoming

Abstract Background: European viper bite is relatively uncommon but can cause serious envenoming, particularly swelling and hemorrhage spreading from limb to trunk that can cause long term disability. Systemic features are relatively mild compared to many other venomous species. Moderate-to-severe envenoming requires antivenom, which is given many hundreds of times each year across the continent. Several Vipera spp antivenoms are produced in Europe, but there is little comparative information available for the antivenoms and none is licensed with the European Medicines Agency. We aimed to collect descriptive data on European viper antivenoms and assess their relative effectiveness. Methods: A systematic review of articles relating to antivenom in Europe was performed using the Medline medical database. The following keywords “Europ*” or the individual names of each European country and “antiven*” or “immun*” or “envenom*” and “snake” or “viper*” or “adder” were used. Articles published between 1 January 1996 and 11 March 2016 pertaining to clinical outcome, including case reports, were selected. Referenced articles in the indexed articles were explored for suitability and included if they met any of the criteria: specific antivenom used, route of antivenom administration, adverse reactions to antivenom therapy and length of hospital admission. All accepted abstracts from EAPCCT conferences since 2000 were searched and abstracts relating to Vipera spp envenoming were assessed for suitability. We extracted data on study type, safety and effectiveness. We sought information on antivenoms from manufacturers and individual patient data from authors of publications. Since individual patient data were only rarely available, we compared median length of stay between case series reporting each antivenom. We identified 40 papers and six published abstracts, and one unpublished paper that reported clinical cases and case series of envenomed patients treated with antivenom. No publication reported randomized controlled trials comparing any European Vipera antivenom with either placebo or another antivenom. 25 reports were of retrospective hospital- (nu2009=u200913) or poison center-based (nu2009=u200912) case series including five or more patients; a further 12 reports were either case reports or case series with less than five patients and one paper was a limited literature review. An additional nine papers reported prospective data; seven collected data remotely through poison service telephone communication with the attending physicians. Antivenoms available in Europe: Eight antivenoms are available for European Vipera spp envenoming; a material safety data sheet providing information on manufacture was available for seven. Six are raised against V. berus or V. ammodytes venom; the seventh is raised against a mixture of V. ammodytes, V. aspis and V. berus venom and the eighth is raised against V. ammodytes, Macrovipera lebetina and Montivipera xanthina venom. Six manufacturers recommended intramuscular administration while two recommended intravenous administration. No randomized control trials comparing the effectiveness of antivenoms were identified. Pre-clinical data: We found two papers presenting comparative preclinical data. Clinical data: Clinical studies were predominantly retrospective and contained clinical data on antivenom used in 2602 patients; where the antivenom was identified (nu2009=u20092174), 2061 (94.8%) received Zagreb, ViperFAV or ViperaTAb antivenoms. There were few published data on the other antivenoms. Repeated use of antivenom: Repeat doses were reported in 230/1491 of cases (15.4%) where this information was recorded. Outcome and length of hospital stay: Intravenous administration of antivenom was associated with shorter length of hospital stay (median length of hospital stay in studies of intravenous ViperFAV or ViperaTAb ranged from 1 to 4.8 days versus 2 to 18 days for intramuscular Bulbio or Zagreb antivenoms). Antivenom versus no antivenom: Some small studies demonstrated no difference in the length of hospital stay in patients with equivalent envenomation grading who either did or did not receive antivenom. Adverse events: Adverse reactions were reported in 37 of 2408 cases (1.5%) including seven cases of anaphylaxis. Conclusions: There are very limited pre-clinical comparative data and no randomised controlled trials assessing effectiveness of the antivenoms against different Vipera species. Most descriptive data suggest the efficacy of Zagreb, ViperFAV and ViperaTAb antivenoms by the intravenous route but not intramuscular route, although this is level D evidence. Reported adverse reactions were rare, suggesting that the modern intravenous antivenoms are of good quality. Better and more systematic data, including perhaps randomized controlled trials comparing different antivenoms, are required for the many hundreds of antivenom administrations that occur annually across Europe.

Metabolic acidosis in prometryn (triazine herbicide) self-poisoning.

Introduction. Prometryn is a triazine herbicide, which is one of the most extensively used groups of herbicides. The mechanism of acute triazine herbicide toxicity in humans is not known. We report a first case of acute prometryn poisoning. Case report. A 62-year-old male ingested 50 g of prometryn and ethanol in a suicide attempt. On arrival two hours after ingestion, he was somnolent and vomited. Seven hours after ingestion laboratory tests showed metabolic acidosis with a calculated anion gap of 47.5 mmol/L and lactate of 23.4 mmol/L. Gas chromatography/mass spectrometry revealed serum prometryn concentrations of 48.1 mg/L. Hemodialysis corrected metabolic acidosis, but the serum prometryn concentration increased to 67.7 mg/L. The lactate level after hemodialysis was 11.7 mmol/L and returned within normal limits 47 hours after ingestion. The patient was discharged without any sequelae after psychiatric evaluation. Conclusion. In high anion gap metabolic acidosis we should consider poisoning with prometryn and other triazine herbicides. Hemodialysis corrects metabolic derangements, but it does not lower serum prometryn concentration.

Butanol ingestion in an airport hangar

1–Butanol is a colourless organic solvent with a rancid sweet odour. 1–Butanol ingestion may result in vomiting, abdominal pain, headache, drowsiness and unconsciousness. We present a 47–year–old male with no previous medical history, who was found comatose and soiled after having vomited while unconscious. On arrival, he had a Glasgow coma scale of 3, tachycardia, hypotension, shallow tachypnoic breathing, hypotonic muscles, absent myotatic reflexes and aromatic odour. The patient was intubated and treated with oxygen, dopamine and volume replacement therapy. Gastric lavage was performed and activated charcoal was given. His initial laboratory test revealed hypokaliemia, renal failure, acidosis with elevated lactate and hypercapnic respiratory insufficiency. Twelve hours after admission, the patient started to respond to a painful stimulus and 4 h later he was conscious. He was extubated 23 h after admission. All pathological laboratory results gradually returned within normal limits. The subsequent toxicological examination of gastric content and urine sample by gas chromatography revealed 1–butanol. On awakening, he confirmed ingestion of a solvent stored in an airport hangar. In conclusion, we describe a patient who ingested an unknown dose of 1–butanol. Symptoms were headache, vomiting, abdominal pain, coma, muscular hypotonus, hypotension, respiratory insufficiency and mixed acidosis. The patient totally recovered after supportive therapy over 30 h. In future cases, intravenous administration of ethanol or even hemodialysis can be considered analogous to the treatment of methanol and ethylene glycol poisoning.

Urinary serotonin level is associated with serotonin syndrome after moclobemide, sertraline, and citalopram overdose

Introduction. Altered mental status, autonomic dysfunction, and neuromuscular abnormalities are a characteristic triad of serotonin syndrome. No laboratory tests confirm the diagnosis of serotonin syndrome. Case report. A 35-year-old woman took moclobemide, sertraline, and citalopram in a suicide attempt. She was conscious with mild tachycardia, hypertension, and tachypnea one hour after ingestion. In the second hour after ingestion diaphoresis, mydriasis, horizontal nystagmus, trismus, hyperreflexia, clonus, and tremor appeared. She became agitated and unresponsive. In the third hour after ingestion she became comatose and hyperthermic. She was anesthetized, paralyzed, intubated, and ventilated for 24 hours. Serum moclobemide, sertraline, and citalopram levels were above therapeutic levels. The serum serotonin level was within normal limits and the urinary 5-hydroxyindoleacetic acid:creatinine ratio was below the average daily value. The urinary serotonin:creatinine ratio was increased on arrival (1 mg/g). Discussion and conclusion. The urinary serotonin level is increased in serotonin syndrome due to a monoamine oxidase inhibitor and selective serotonin-reuptake inhibitors overdose. It is possible that urinary serotonin concentration could be used as a biochemical marker of serotonin syndrome.