Pharmacy Blog – Dr Vibore

A number of early studies showed that effects of coumarinoral anticoagulants are unlikely to be changed in those with normal liver function who drink small or moderate amounts of alcoholic beverages such as wine or spirits

The daily consumption of 1 pint (about 560 mL) of Californian white table wine for a 3-week period at meal times by 8 healthy subjects anticoagulated with warfarin, was found to have no significant effects on either serum warfarin levels or anticoagulant response (See reference number 1).

Other studies in both patients(See reference number 2,3)and healthy subjects(See reference number 4,5) taking either warfarin(See reference number 2,3,5) or phenprocoumon(See reference number 4) have very clearly confirmed absence of an interaction with wine(See reference number 3,5) gin,(See reference number 4) or 40 % alcohol (See reference number 2). In one of these studies subjects were given almost 600 mL of a table wine (12% alcohol) or 300 mL of a fortified wine (20% alcohol) without adverse effects on coagulation (See reference number 5)

In contrast to above studies, a 58-year-old man stabilised on warfarin experienced a sharp rise in his INR to 8 when he started to drink half a can of light beer (5.35 g alcohol) every other day. In previous 5 months he had an INR in range of 1.9 to 2.5 with a stable warfarin dose, and no other explanation for change in INR was found. He stopped taking alcohol, and was eventually restabilised on original dose of warfarin (See reference number 6).

In one study,15 alcoholics who had been drinking heavily (250 g ethanol or more daily) for at least 3 months and 11 control subjects (minimal social drinkers or non-drinkers) were given a single 40mg dose of warfarin. The half-life of warfarin was lower in alcoholics (26.5 versus

41.1 hours), but a comparison of prothrombin times with those of healthy subjects found no differences (See reference number 7).

One patient with liver cirrhosis had marked fluctuations in prothrombin times and warfarin levels associated with weekend binge drinking of vodka (See reference number 2). Another patient with abnormal liver function had a fall in plasma warfarin levels and effect when he stopped drinking 50 mL of whiskey daily. When rechallenged with alcohol,warfarin levels and effect rose, and he had a nosebleed (See reference number 8). In contrast,a large retrospective cohort study did not find a significantly increased risk of serious bleeding in 140 patients with a history of alcoholic binge drinking who were taking warfarin. The relative risk was 1.3 (0.8 to 1.9) compared with patients who had no record of alcohol abuse (See reference number 9).

It seems probable that, as in rats,(See reference number 10)continuous heavy drinking stimulates hepatic enzymes concerned with metabolism of warfarin, leading to its more rapid elimination (See reference number 7,11). As a result half-life shortens. The fluctuations in prothrombin times in those with liver impairment(See reference number 2,8) may possibly occur because sudden large amounts of alcohol exacerbate general dysfunction of liver and this affects way it metabolises warfarin. Alcohol may also change ability of liver to synthesise clotting factors (See reference number 12). Constituents of beer other than alcohol may affect warfarin metabolism (See reference number 6).

The absence of an interaction between warfarin or phenprocoumon and alcohol in those free from liver disease is well documented and well established. It appears to be quite safe for patients taking oral anticoagulants to drink small or moderate amounts of wine or spirits. Even much less conservative amounts (up to 8 oz/250 mL of spirits(See reference number 4) or a pint of wine(See reference number 1)) do not create problems with anticoagulant control, so that there appears to be a good margin of safety even for less than abstemious. Only warfarin and phenprocoumon have been investigated but other coumarin anticoagulants would be expected to behave similarly. The single case of increased INR in a patient who started to drink beer is unexplained. Further study specifically with beer is needed to throw light on this possible interaction.

On other hand, those who drink heavily may possibly need above-average doses of anticoagulant, while limited evidence suggests that those with liver damage who binge drink may experience marked fluctuations in their prothrombin times. It might be prudent to avoid anticoagulation in this type of patient unless they can abstain from drinking. Nevertheless,although one cohort study in patients taking warfarin found a slight trend towards serious bleeding events in patients with a history of binge drinking, this was not significant, and other risk factors were more important (highly variable prothrombin time ratio, or prothrombin time ratio greater than 2) (See reference number 9).

1.

O’Reilly RA. Lack of effect of mealtime wine on the hypoprothrombinemia of oral anticoagulants. Am J Med Sci (1979) 277,189–94.

2.

Udall JA. Drug interference with warfarin therapy. Clin Med (1970) 77,20–25.

3.

Karlson B,Leijd B, Hellström A. On the influence of vitamin K-rich vegetables and wine onthe effectiveness of warfarin treatment. Acta Med Scand (1986) 220, 347–50.

4.

Waris E. Effect of ethyl alcohol on some coagulation factors in man during anticoagulanttherapy. Ann Med Exp Biol Fenn (1963) 41,45–53.

5.

O’Reilly RA. Lack of effect of fortified wine ingested during fasting and anticoagulant therapy. Arch Intern Med (1981) 141,458–9.

6.

Havrda DE,Mai T, Chonlahan J. Enhanced antithrombotic effect of warfarin associated with low-dose alcohol consumption. Pharmacotherapy (2005) 25, 303–7.

7.

Kater RMH,Roggin G, Tobon F, Zieve P, Iber FL. Increased rate of clearance of drugs fromthe circulation of alcoholics. Am J Med Sci (1969) 258, 35–9.

8.

Breckenridge A,Orme M. Clinical implications of enzyme induction. Ann N Y Acad Sci (1971) 179, 421–31.

9.

Fihn SD,McDonell M, Martin D, Henikoff J, Vermes D, Kent D, White RH, for the Warfarin Optimized Oupatient Follow-up Study Group. Risk factors for complications of chronic anticoagulation: a multicenter study. Ann Intern Med (1993) 118, 511–20.

Rubin E,Hutterer F, Lieber CS. Ethanol increases hepatic smooth endoplasmic reticulum anddrug-metabolizing enzymes. Science (1968) 159, 1469–70.

Kater RMH,Carruli N, Iber FL. Differences in the rate of ethanol metabolism in recentlydrinking alcoholic and nondrinking subjects. Am J Clin Nutr (1969) 22, 1608–17.

Riedler G. Einfluß des Alkohols auf die Antikoagulantientherapie. Thromb Diath Haemorrh (1966) 16,613–35.

When blood is lost or clotting is initiated in some other way, a complex cascade of biochemical reactions is set in motion, which ends in formation of a network or clot of insoluble protein threads enmeshing blood cells. These threads are produced by polymerisation of molecules of fibrinogen (a soluble protein present in plasma) into threads of insoluble fibrin. The penultimate step in chain of reactions requires presence of an enzyme, thrombin, which is produced from its precursor prothrombin, already present in plasma. This is initiated by factor III (tissue thromboplastin),and subsequently involves various factors including activated factor VII, IX, X, XI and XII, and is inhibited by anti-thrombin III. Platelets are also involved in coagulation process. Fibrinolysis is mechanism of dissolution of fibrin clots, which can be promoted with thrombolytics. For further information on platelet aggregation and clot dissolution,see Antiplatelet drugs and thrombolytics, .

Mode of action of anticoagulants

Anticoagulants may be divided into direct anticoagulants, which have an immediate effect, and indirect anticoagulants, which inhibit formation of coagulation factors, so have a delayed effect as they do not inactivate coagulation factors already formed. See table 1 below,, for a list.

The direct anticoagulants include heparin, which principally enhances effect of antithrombin III, thereby inhibiting effect of thrombin (factor IIa) and activated factor X (factor Xa). Low-molecular-weight heparins are salts of fragments of heparin and act similarly,except that they have a greater effect on factor Xa than factor IIa. They have a longer duration of action than heparin and usually require less monitoring. The heparinoids (such as danaparoid) are similar. A more recent introduction is synthetic polysaccharide fondaparinux, which is an inhibitor of factor Xa. The other group of direct anticoagulants are thrombin inhibitors, which bind to active thrombin site. These include recombinant forms or synthetic analogues of hirudin such as bivalirudin and lepirudin. Megalatran and its oral prodrug ximelagatran act similarly,but have been withdrawn because of liver toxicity.

The indirect anticoagulants inhibit vitamin K-dependent synthesis of factors VII, IX, X and II (prothrombin) in liver, and may also be referred to as vitamin K antagonists. The most commonly used are coumarins, principally warfarin, but also acenocoumarol and phenprocoumon. The indanediones such as phenindione are now less frequently used. The indirect anticoagulants have advantage over currently available direct anticoagulants in that they are orally active. They are often therefore referred to as oral anticoagulants, but this term may become misleading with development of direct-acting oral anticoagulants, such as ximelegatran, which have different monitoring requirements and interactions.

During anticoagulant therapy aim is to give protection against intravascular clotting, without running risk of bleeding. To achieve this, doses of heparin and oral anticoagulants should be individually titrated until desired response is attained. With coumarin and indanedione oral anticoagulants, this procedure normally takes several days because they do not act directly on blood clotting factors already in circulation, but on rate of synthesis of new factors by liver. The successful titration is determined by one of a number of different but closely related laboratory tests,see table 2 below, and below. Note that routine monitoring of anticoagulant effect is not required for low-molecular weight heparins or heparinoids,except in patients at increased risk of bleeding, such as those with renal impairment or who are overweight. Also, note that these tests cannot be used to monitor anticoagulant effect of fondaparinux or direct thrombin inhibitors, and these require no routine monitoring.

The prothrombin time test (PT, Pro-Time, tissue factor induced coagulation time) is most common method employed in clinical situations. It measures time taken for a fibrin clot to form in a citrated plasma sample containing calcium ions and tissue thromboplastin. The PT is usually reported as International Normalised Ratio (INR).

International normalised ratio (INR). The INR was adopted by WHO in 1982 to standardise (using International Sensitivity Index) oral anticoagulant therapy to take into account sensitivities of different thromboplastins used in laboratories across world. The formula for calculating INR is as follows: INR = (patient’s prothrombin time in seconds/mean normal prothrombin time in seconds)(See reference number ISI)The PT values obtained from patient’s sample are compared to a control, and this gives INR. The higher INR, higher PT value so if patient’s ratio is 2, this means PT (and therefore clotting) is twice as long as normal plasma. The British Corrected Ratio is essentially same, but was calculated to a standard British thromboplastin.

Quick Value. The Quick Value is expressed as a percentage; lower value, longer blood takes to coagulate. Therefore as Quick Value increases, corresponding INR value gets smaller and vice versa.

The activated partial thromboplastin time (aPTT) is second most common method for monitoring anticoagulant therapy, measuring all clotting factors in intrinsic pathway as opposed to PT test, which measures extrinsic pathway.

Other tests used, which in some instances offer more sensitivity to specific aspects of therapy, include prothrombin-proconvertin ratio (PP), thrombotest, thrombin clotting time test (TCT, activated clotting time, activated coagulation time), platelet count and bleeding time test. The use of most appropriate test will depend on situation and desired result.

Stable oral anticoagulant therapy is difficult to achieve even during close monitoring. For example, in one controlled study in patients with atrial fibrillation, only 61 % of INR values were within target range of 2 to 3, despite monitoring INR monthly and adjusting warfarin dose appropriately.(See reference number 1) A large number of factors can influence levels of coagulation, including diet, disease (fever, diarrhoea, heart failure, thyroid dysfunction), and use of other drugs. It must therefore be remembered that it is particularly difficult to ascribe a change in INR specifically to a drug interaction in a single case report,and single case reports or a few isolated reports for widely used drugs do not prove that an interaction occurs. Nevertheless, either addition or withdrawal of drugs may upset balance in a patient already well stabilised on an anticoagulant. Some drugs are well known to increase activity of anticoagulants and can cause bleeding if dosage of anticoagulant is not reduced appropriately. Others reduce activity and return prothrombin time to normal. Both situations are serious and may be fatal, although excessive hypoprothrombinaemia manifests itself more obviously and immediately as bleeding and is usually regarded as more serious. The interaction mechanism may be pharmacodynamic or pharmacokinetic: pharmacokinetic mechanisms are particularly well established and important for coumarin anticoagulants.

(a) Metabolism of coumarins

The coumarins,warfarin, phenprocoumon and acenocoumarol, are racemic mixtures of S– and R-enantiomers. The S-enantiomers of these

coumarins have several times more anticoagulant activity than R-enantiomers. Reports suggest for example,that S-warfarin is three to five times more potent a vitamin K antagonist than R-warfarin. The S-enantiomer of warfarin is metabolised primarily by cytochrome P450 isoenzyme CYP2C9, and to a much lesser extent, by CYP3A4. The metabolism of R-warfarin is more complex,but this enantiomer is primarily metabolised by CYP1A2, CYP3A4, and CYP2C19. S-warfarin is eliminated in bile and R-warfarin is excreted in urine as inactive metabolites. There is much more known about metabolism of warfarin compared with other anticoagulants, but it is established that S-phenprocoumon and S-acenocoumarol are also substrates for CYP2C9 and that they differ from warfarin in their hepatic metabolism, and stereospecific potency (See reference number 2).

It makes sense to assume therefore,that an inhibitor of CYP2C9 (e.g. fluconazole, ) is likely to increase concentration of coumarin and enhance anticoagulant effect. Drugs that induce CYP2C9

(e.g. rifampicin, ) reduce plasma levels of coumarins by increasing clearance.

Genetic differences, , in genes for these cytochrome P450 isoenzymes may have an important influence on drug metabolism of coumarins. For example, different versions of gene encoding CYP2C9 exist and enzymatic activity of most clinically important CYP2C9 variants, CYP2C9*2 and CYP2C9*3, is significantly reduced. Studies have suggested an association between patients possessing one or more of these variants and a low-dose requirement of warfarin. Similar observations have been seen with CYP2C9*3 variant and acenocoumarol.

While metabolism of coumarins, especially warfarin, are well known, numerous interaction pathways and variability in patient responses, makes clinical consequences difficult to predict.

Some drugs, such as colestyramine, , may also prevent absorption of coumarins and reduce their bioavailability. See also Drug absorption interactions,. Additive anticoagulant effects can occur if anticoagulants are given with other drugs that also impair coagulation by other mechanisms such as antiplatelets,. Coumarins and indanediones act as vitamin K antagonists,and so dietary intake of vitamin K,

can also reduce or abolish, their effects. Protein-binding displacement, is another possible drug interaction mechanism but this usually plays a minor role compared with other mechanisms (See reference number 3).

When prothrombin times become excessive,bleeding can occur. In order of decreasing frequency bleeding shows itself as ecchymoses, blood in urine, uterine bleeding, black faeces, bruising, nose-bleeding, haematoma, gum bleeding, coughing and vomiting blood.

Vitamin K is an antagonist of coumarin and indanedione oral anticoagulants. The British Society for Haematology has given advice on appropriate course of action if bleeding occurs in patients taking warfarin, and this is readily available in summarised form in British National Formulary.

If effects of heparin are excessive it is usually sufficient just to stop heparin, but protamine sulfate is a specific antidote if a rapid effect is required. Protamine sulfate only partially reverses effect of low molecular weight heparins.

There is currently no known specific antidote for fondaparinux, or for direct thrombin inhibitors.

Stroke Prevention in Atrial Fibrillation Investigators. Adjusted-dose warfarin versus low-intensity,fixed dose warfarin plus aspirin for high-risk patients with atrial fibrillation: StrokePrevention in Atrial Fibrillation III randomised clinical trial. Lancet (1996) 348, 633–8.

Ufer M. Comparative pharmacokinetics of vitamin-K antagonists. Warfarin,phenprocoumonand acenocoumarol. Clin Pharmacokinet (2005) 44, 1227–46.

Sands CD,Chan ES, Welty TE. Revisiting the significance of warfarin protein-binding displacement interactions. Ann Pharmacother (2002) 36, 1642–4.

Table 1 Anticoagulants
Direct anticoagulants	Indirect anticoagulants
Fondaparinux sodium	Coumarins
Heparin	Acenocoumarol
Heparin calcium	Dicoumarol
Heparin sodium	Ethyl biscoumacetate
Heparinoids	Phenprocoumon
Danaparoid	Warfarin
Dermatan sulfate	Indanediones
Pentosan polysulfate	Fluindione
Suleparoid	Phenindione
Sulodexide
Low-molecular-weight heparins
Bemiparin
Certoparin
Dalteparin
Enoxaparin
Nadroparin
Parnoparin
Reviparin
Tinzaparin
Thrombin inhibitors
Argotroban
Bivalirudin
Desirudin
Hirudin
Lepirudin
Melagatran
Ximelagatran

Table 2 Coagulation tests
Test	Normal range	Therapeutic/diagnostic range
Activated partial thromboplastin time	20 to 39 seconds after reagents added	1.5 to 2.5 x control
Bleeding time	1 to 9 minutes depending on method used	Critical value greater than 15 minutes
International normalised ratio	0.9 to 1.2	2 to 4 depending on indication for anticoagulation
Plasma thrombin time test	10 to 15 seconds	Greater than 15 seconds
Prothrombin-proconvertin ratio	70 to 130 %	10 to 30 %
Prothrombin time	10 to 15 seconds	1 to 2 x control
Quick value	70 to 130 %	10 to 20 %
Thrombin clotting time	70 to 120 seconds	150 to 600 seconds depending on indication for anticoagulation
Thrombotest	100%	10 to 20 %

Ginkgo biloba does not appear to affect pharmacokinetics orpharmacodynamics of donepezil

Clinical evidence,mechanism, importance and management

In a pharmacokinetic study 14 elderly patients with Alzheimer’s disease were given donepezil 5mg daily for at least 20 weeks,after which Ginkgo biloba extract 90mg daily was also given for a further 30 days. Concurrent use did not affect pharmacokinetics or cholinesterase activity of donepezil, and cognitive function appeared to be unchanged (See reference number 1). Therefore, over course of 30 days, concurrent use appears neither beneficial nor detrimental.

1. Yasui-Furukori N,Furukori H, Kaneda A, Kaneko S, Tateishi T. The effects of Ginkgo biloba extracts on the pharmacokinetics and pharmacodynamics of donepezil. J Clin Pharmacol (2004) 44, 538–42.

A number of drugs can affect myasthenia gravis,often by increasing muscular weakness. This is, strictly speaking, a drug-disease interaction, but such effects may be expected to oppose actions of drugs used to treat myasthenia gravis. A number of drugs (e.g. chlorpromazine,methocarbamol, and propafenone) are clearly contraindicated in patients with myasthenia, and, as this is not strictly a drug interaction, they

See Beta blockers + Anticholinesterases,p. 834.

Persisting myasthenic symptoms,including muscular weakness, attributed to prior chloroquine use. Development of myasthenic symptoms in 3 patients,one who took chloroquine in overdose.

Aggravation of myasthenic symptoms in a patient taking pyridostigmine,and unmasking of myasthenia in one patient.

8,9

20,21

Carmignani M,Scoppetta C, Ranelletti OF, Tonali P. Adverse interaction between acetazolamide and anticholinesterase drugs at normal and myasthenic neuromuscular junction level. Int J Clin Pharmacol Ther Toxicol (1984) 22,140–4.

Argov Z,Brenner T, Abramsky O. Ampicillin may aggravate clinical and experimental myasthenia gravis. Arch Neurol (1986) 43,255–6.

McDowell IFW,McConnell JB. Cholinergic crisis in myasthenia gravis precipitated by ketoprofen. BMJ (1985) 291,1094.

De Bleecker J,De Reuck J, Quatacker J, Meire F. Persisting chloroquine-induced myasthenia? Acta Clin Belg (1991) 46,401–6.

Robberecht W,Bednarik J, Bourgeois P, van Hees J, Carton H. Myasthenic syndrome caused by direct effect of chloroquine on neuromuscular junction. Arch Neurol (1989) 46,464–8.

Sghirlanzoni A,Mantegazza R, Mora M, Pareyson D, Cornelio F. Chloroquine myopathy and myasthenia-like syndrome. Muscle Nerve (1988) 11,114–19.

Pichon P,Soichot P, Loche D, Chapelon M. Syndrome myasthenique induit par une intoxication a la choroquine: une forme clinique inhabituelle confirmee par une atteinte oculaire. Bull Soc Ophtalmol Fr (1984) 84,219–22.

Moore B,Safani M, Keesey J. Possible exacerbation of myasthenia gravis by ciprofloxacin. Lancet (1988) 1,882.

Mumford CJ,Ginsberg L. Ciprofloxacin and myasthenia gravis. BMJ (1990) 301,818.

Haddad M,Zelikovski A, Reiss R. Dipyridamole counteracting distigmine in a myasthenic patient. IRCS Med Sci (1986) 14,297.

Absher JR,Bale JF. Aggravation of myasthenia gravis by erythromycin. J Pediatr (1991) 119,155–6.

O’Riordan J,Javed M, Doherty C, Hutchinson M. Worsening of myasthenia gravis on treatment with imipenem/cilastatin. J Neurol Neurosurg Psychiatry (1994) 57,383.

Neil JF,Himmelhoch JM, Licata SM. Emergence of myasthenia gravis during treatment with lithium carbonate. Arch Gen Psychiatry (1976) 33,1090–2.

Rauser EH,Ariano RE, Anderson BA. Exacerbation of myasthenia gravis by norfloxacin. DICP Ann Pharmacother (1990) 24,207–8.

Vincent A,Newsom-Davis J, Martin V. Anti-acetylcholine receptor antibodies in D-penicillamine-associated myasthenia gravis. Lancet (1978) i,1254.

Masters CL,Dawkins RL, Zilko PJ, Simpson JA, Leedman RJ, Lindstrom J. Penicillamine-associated myasthenia gravis,antiacetylcholine receptor and antistriational antibodies. Am J Med (1977) 63,689–94.

Ferro J,Susano R, Gómez C, de Quirós FB. Miastenia inducida por penicilamina: ¿existe interacción con los antidepresivos tricíclicos? Rev Clin Esp (1993) 192,358–9.

Russell AS,Linstrom JM. Penicillamine-induced myasthenia gravis associated with antibodies to acetylcholine receptor. Neurology (1978) 28,847–9.

Brumlik J,Jacobs RS. Myasthenia gravis associated with diphenylhydantoin therapy for epilepsy. Can J Neurol Sci (1974) 1,127–9.

Drachman DA,Skom JH. Procainamide – a hazard in myasthenia gravis. Arch Neurol (1965) 13,316–20.

Kornfeld P,Horowitz SH, Genkins G, Papatestas AE. Myasthenia gravis unmasked by antiarrhythmic agents. Mt Sinai J Med (1976) 43,10–14.

Stoffer SS,Chandler JH. Quinidine-induced exacerbation of myasthenia gravis in patient with Graves’ disease. Arch Intern Med (1980) 140,283–4.

Weisman SJ. Masked myasthenia gravis. JAMA (1949) 141,917–18.

are not dealt with here. A number of case reports (see table 1 below, (above)) describe worsening or unmasking of myasthenia gravis with a range of different drugs. The evidence for many of these interactions is very sparse indeed,and in some instances they are simply rare and isolated cases. It would therefore be wrong to exaggerate their importance, but it would nevertheless be prudent to be alert for any evidence of worsening myasthenia if any of drugs listed are added to established treatment.

Table 1 Case reports of drugs aggravating or unmasking myasthenia gravis
Drug	Effect seen	Refs
Acetazolamide 500mg intravenously	Aggravation of muscular weakness in patients with myasthenia gravis taking unnamed anticholinesterases.	1
Ampicillin up to 1.5 g daily	Aggravation of myasthenic symptoms in 2 patients taking pyridostigmine.	2
Aspirin	Mild aggravation of myasthenic symptoms in a patient taking neostigmine.	3
Beta blockers	See Beta blockers + Anticholinesterases, p. 834.
Chloroquine	Persisting myasthenic symptoms, including muscular weakness, attributed to prior chloroquine use. Development of myasthenic symptoms in 3 patients, one who took chloroquine in overdose.	4-7
Ciprofloxacin	Aggravation of myasthenic symptoms in a patient taking pyridostigmine, and unmasking of myasthenia in one patient.	8, 9
Dipyridamole* 75mg three times daily	Aggravation of myasthenic symptoms in a patient taking distigmine.	10
Erythromycin 500mg intravenously	Precipitation of a myasthenic crisis in an undiagnosed 15-year-old girl.	11
Imipenem/cilastatin 500mg four times daily	Aggravation of myasthenic symptoms in a patient taking pyridostigmine.	12
Ketoprofen 50mg daily	Aggravation of myasthenic symptoms in a patient taking neostigmine.	3
Lithium carbonate 600mg daily	Unmasking of myasthenia in one patient.	13
Norfloxacin*	Aggravation of myasthenic symptoms in a patient taking pyridostigmine.	14
Penicillamine	Aggravation of myasthenic symptoms in numerous patients taking anticholinesterases. Amitriptyline and imipramine also implicated in 2 cases.	15-18
Phenytoin 100mg three times daily	Aggravation of myasthenic symptoms in an untreated patient.	19
Procainamide* 250mg	Serious aggravation of myasthenic symptoms in a patient taking pyridostigmine. Two other less severe cases also reported.	20, 21
Quinidine* up to 970mg daily	Mild aggravation of myasthenic symptoms in one patient taking pyridostigmine and another taking neostigmine. Development of myasthenic symptoms in 2 undiagnosed patients.	21-23

Clinical evidence,mechanism, importance and management

Vancomycin is both potentially nephrotoxic and ototoxic,and its manufacturers therefore suggest that it should be used with particular care, or avoided in patients with renal impairment or deafness (See reference number 1). They also advise avoidance of other drugs that have nephrotoxic potential, because effects could be additive. They list amphotericin B,aminoglycosides, bacitracin, colistin, polymyxin B, viomycin and cisplatin. They also list etacrynic acid and furosemide as potentially aggravating ototoxicity.

The monograph Aminoglycosides + Vancomycin interaction,outlines some of evidence that additive nephrotoxicity can occur with aminoglycosides, but there seems to be no direct evidence about other drugs. Even so, general warning issued by manufacturers to monitor carefully is a reasonable precaution.

1. Vancomycin hydrochloride. Mayne Pharma plc. UK Summary of product characteristics,December 2003.

The anticholinesterase drugs (or cholinesterase inhibitors) can be classified as centrally-acting, reversible inhibitors such as donepezil (used in treatment of Alzheimer’s disease), reversible inhibitors with poor CNS penetration, such as neostigmine (used in treatment of myasthenia gravis), or irreversible inhibitors, such as ecothiopate and metrifonate. The centrally-acting anticholinesterases and reversible anticholinesterases form basis of this section, and these are listed in table 1 below,(see below). Interactions where anticholinesterases are affecting other drugs are covered elsewhere in publication.

Due to their differing pharmacokinetic characteristics, centrally-acting anticholinesterases have slightly different interaction profiles, although they share a number of common pharmacodynamic interactions. Tacrine(See reference number 1) is metabolised by cytochrome P450 isoenzyme CYP1A2, and so interacts with fluvoxamine, , a potent inhibitor of this isoenzyme, whereas there is no evidence to suggest other centrally acting anticholinesterases do. On other hand, donepezil(See reference number 1) and galantamine(See reference number 1) are metabolised by cytochrome P450 isoenzymes CYP3A4 and CYP2D6, and so they may interact with ketoconazole, and quinidine, , respectively, whereas tacrine would not be expected to do so. Rivastigmine,(See reference number 1) which is metabolised by conjugation,seems relatively free of pharmacokinetic interactions. Consideration of concurrent drug use would therefore seem to be an important factor in choice of centrally-acting anticholinesterase.

Note that,organophosphorus compounds such as insecticides are also anticholinesterases.

1. Jann MW,Shirley KL, Small GW. Clinical pharmacokinetics and pharmacodynamics ofcholinesterase inhibitors. Clin Pharmacokinet (2002) 719–39.

6. McSwain ML,Forman LM. Severe parkinsonian symptom development on combination treat-Anticholinesterases; Centrally acting + ment with tacrine and haloperidol. J Clin Psychopharmacol (1995) 15, 284.

7. Maany I. Adverse interaction of tacrine and haloperidol. Am J Psychiatry (1996) 153,1504.

Table 1 Anticholinesterase drugs; reversible
Centrally-acting inhibitors used principally for Alzheimer’s disease	Inhibitors with poor CNS penetration used principally for myasthenia gravis
Donepezil	Ambenonium
Galantamine	Distigmine
Rivastigmine	Edrophonium (mainly used diagnostically)
Tacrine	Neostigmine
	Physostigmine
	Pyridostigmine (also used for glaucoma)

Ketoconazole modestly increases levels of donepezil. Eventhough this was not considered to be clinically significant,themanufacturers suggest that potent inhibitors of CYP3A4 willraise donepezil levels and inducers of CYP3A4 will lower donepezil levels. Galantamine levels are also increased by ketoconazole.

Clinical evidence,mechanism, importance and management

Donepezil 5mg daily was given to 18 healthy subjects with ketoconazole 200 mg daily, which is a specific and potent inhibitor of cytochrome P450 isoenzyme CYP3A4. After one week of concurrent use, maximum serum levels and AUC of donepezil were increased by less than 30%. Donepezil had no effect on pharmacokinetics of ketoconazole (See reference number 1). None of increases in donepezil levels were considered to be clinically relevant, and authors suggest that no dose modifications will be required with ketoconazole or other CYP3A4 inhibitors (See reference number 1). Despite this, UK manufacturer recommends that donepezil should be used with CYP3A4 inhibitors with care, and they specifically name itraconazole and erythromycin. Furthermore, both US and UK manufacturers suggest that CYP3A4 inducers (they name carbamazepine, dexamethasone, phenobarbital, phenytoin and rifampicin) may lower donepezil levels (See reference number 2,3). Be aware that a reduction in donepezil levels is possible with these drugs,but that a clinically significant interaction seems unlikely.

The manufacturers note that ketoconazole increased bioavailability of galantamine by 30%, probably as a result of CYP3A4 inhibition. They therefore predict that ketoconazole (and other potent CYP3A4 inhibitors such as ritonavir) may increase incidence of nausea and vomiting with galantamine, and suggest that, based on tolerability, a decrease in maintenance dose be considered (See reference number 4,5). Whether this is in fact necessary in practice remains to be established. Erythromycin,a moderate CYP3A4 inhibitor, only increased galantamine bioavailability by about 10%,(See reference number 4,5) and so a clinically significant interaction would not be expected.

Tiseo PJ,Perdomo CA, Friedhoff LT. Concurrent administration of donepezil HCl and ketoconazole: assessment of pharmacokinetic changes following single and multiple doses. Br J Clin Pharmacol (1998) 46 (Suppl 1), 30–34.

Aricept (Donepezil hydrochloride). Eisai Ltd. UK Summary of product characteristics,January2007.

Aricept (Donepezil hydrochloride). Eisai Inc. US Prescribing information,October 2006.

Reminyl (Galantamine hydrobromide). Shire Pharmaceuticals Ltd. UK Summary of productcharacteristics,July 2005.

Razadyne (Galantamine hydrobromide). Ortho-McNeil Neurologics,Inc. US Prescribing information, August 2006.

Memantine does not appear to attenuate anticholinesterase effects of donepezil, galantamine, or tacrine, nor affect pharmacokinetics of galantamine or donepezil.

Clinical evidence,mechanism, importance and management

An in vitro study in rats suggested that memantine does not attenuate anticholinesterase effects of donepezil at therapeutic concentrations (See reference number 1). In a later study 19 healthy subjects were given memantine 10mg before and on last day of taking donepezil (5 mg daily for 7 days then 10mg daily for 22 days). The pharmacokinetics of both drugs were not significantly affected by concurrent use, and effects of donepezil on anticholinesterase were also unaffected (See reference number 2). Furthermore, an efficacy and safety study of one year’s duration has reported that combination is well tolerated and beneficial (See reference number 3).

An in vitro study in rats suggested that memantine does not attenuate anticholinesterase effects of galantamine at therapeutic concentrations (See reference number 1). A study in 15 healthy subjects found that concurrent use of extended-release galantamine 16mg daily with memantine 10mg twice daily for 12 days did not affect pharmacokinetics of galantamine and generally did not increase incidence of adverse effects, although dizziness may have been more common (See reference number 4).

Furthermore, a review of efficacy studies suggested that effects of galantamine on anticholinesterase are unaffected, that combination is safe and generally well tolerated (See reference number 5).

An in vitro study in rats suggested that memantine does not attenuate anticholinesterase effects of tacrine at therapeutic concentrations (See reference number 1)

Wenk GL,Quack G, Moebius H-J, Danysz W. No interaction of memantine with anticholinesterase inhibitors approved for clinical use. Life Sci (2000) 66, 1079–83.

Periclou AP,Ventura D, Sherman T, Rao N, Abramowitz WT. Lack of pharmacokinetic or pharmacodynamic interaction between memantine and donepezil. Ann Pharmacother (2004) 38, 1389–94.

Tariot PN,Farlow MR, Grossberg GT, Graham SM, McDonald S, Gergel I, for the Memantine Study Group. Memantine treatment in patients with moderate to severe Alzheimer Disease already receiving donepezil: a randomized controlled trial. JAMA (2004) 291, 317–24.

Yao C,Raoufinia A, Gold M, Nye JS, Ramael S, Padmanabhan M, Walschap Y, Verhaeghe T, Zhao Q. Steady-state pharmacokinetics of galantamine are not affected by addition of memantine in healthy subjects. J Clin Pharmacol (2005) 45, 519–28.

Grossberg GT,Edwards KR, Zhao Q. Rationale for combination therapy with galantamine and memantine in Alzheimer’s disease. J Clin Pharmacol (2006) 46, 17S–26S.

Clinical evidence,mechanism, importance and management

Clofazimine 100mg daily, given to 15 patients with leprosy taking rifampicin 600mg daily and dapsone 100mg daily, had no effect on pharmacokinetics of rifampicin (See reference number 1). A single-dose study similarly found that bioavailability of clofazimine remained unaltered when rifampicin was given, although a reduction in rate of absorption was seen (See reference number 2). No special precautions would seem to be necessary on concurrent use.

1. Venkatesan K,Mathur A, Girdhar BK, Bharadwaj VP. The effect of clofazimine on the pharmacokinetics of rifampicin and dapsone in leprosy. J Antimicrob Chemother (1986) 18, 715–

18.

2. Mehta J,Gandhi IS, Sane SB, Wamburkar MN. Effect of clofazimine and dapsone on rifampicin (Lositril) pharmacokinetics in multibacillary and paucibacillary leprosy cases. Indian J Lepr (1985) 57, 297–310.

The absorption of rifampicin can be reduced up to about one-third by antacids, but clinical importance of this is uncertain.

When 5 healthy subjects took a single 600mg dose of rifampicin with various antacids absorption of rifampicin was reduced. The antacids caused a fall in urinary excretion of rifampicin as follows: 15 or 30 mL of aluminium hydroxide gel 29 to 31%; 2 or 4 g of magnesium trisilicate 31 to 36%; and 2 g of sodium bicarbonate 21 % (See reference number 1)

Three groups of 15 patients with tuberculosis were given a single oral dose of rifampicin 10 to 12 mg/kg,isoniazid 300mg and ethambutol 20 mg/kg either alone or with about 20 mL of antacid. A significant number of patients had peak rifampicin concentrations below

6.5 micrograms/mL (serum level quoted as necessary to achieve adequate lung concentrations) in group receiving Aludrox (aluminium hydroxide), but no significant effect was noted in group receiving Gelusil (aluminium hydroxide plus magnesium trisilicate) (See reference number 2). However,in a further study in 14 healthy subjects, 30 mL of Mylanta (aluminium/magnesium hydroxide) given 9 hrs before, with and after rifampicin had no effect on rifampicin pharmacokinetics (See reference number 3).

It has been suggested that rise in stomach pH caused by these antacids reduces dissolution of rifampicin and thereby inhibits its absorption. In addition,aluminium ions may form less soluble chelates with rifampicin, and magnesium trisilicate can adsorb rifampicin, both of which would also be expected to reduce bioavailability (See reference number 1).

Direct information seems to be limited to these reports. The effects of 20 to 35 % reductions in rifampicin absorption do not appear to have been assessed,but if antacids are given it would be prudent to be alert for any evidence that treatment is less effective than expected. The US manufacturers of rifampicin advise giving rifampicin 1 hour before antacids (See reference number 4).

Khalil SAH,El-Khordagui LK, El-Gholmy ZA. Effect of antacids on oral absorption of rifampicin. Int J Pharmaceutics (1984) 20, 99–106.

Gupta PR,Mehta YR, Gupta ML, Sharma TN, Jain D, Gupta RB. Rifampicin-aluminium antacid interaction. J Assoc Physicians India (1988) 36, 363–4.

Peloquin CA,Namdar R, Singleton MD, Nix DE. Pharmacokinetics of rifampin under fastingconditions, with food, and with antacids. Chest (1999) 115, 12–18.

Rifadin (Rifampicin). Sanofi-Aventis US LLC. US Prescribing information,March 2007.