Pharmacy Blog – Dr Vibore

Ampicillin and rifampicin markedly reduce colonic release of5-aminosalicylate (the active drug) from sulfasalazine

In a study in 5 healthy subjects conversion and release of active metabolite of sulfasalazine, 5-aminosalicylic acid was reduced by one third when a single 2-g dose of sulfasalazine was given after a 5-day course of ampicillin 250mg four times daily (See reference number 1).

A crossover trial in 11 patients with Crohn’s disease receiving long-term treatment with sulfasalazine found that rifampicin 10 mg/kg daily and ethambutol 15 mg/kg daily reduced plasma levels of both 5-aminosalicylic acid and sulphapyridine by about 60 % (See reference number 3). A similar study in patients taking sulfasalazine 1.5 g to 4 g daily found that plasma sulfapyridine levels were reduced by 57 % when patients were taking rifampicin 10 mg/kg and ethambutol 15 mg/kg daily, when compared with placebo. They also noted an increase in erythrocyte sedimentation rate (ESR) during antibacterial treatment (See reference number 4).

The azo link of sulfasalazine is split by anaerobic bacteria in colon to release sulphapyridine and 5-aminosalicylic acid, latter being active metabolite that acts locally in treatment of inflammatory bowel disease. Antibacterials that decimate gut flora can apparently reduce this conversion and this is reflected in lower plasma levels. Rifampicin also possibly increases metabolism of sulphapyridine.

Information is limited, but interaction appears to be established. However, extent to which these antibacterials actually reduce effectiveness of sulfasalazine in treatment of Crohn’s disease or ulcerative colitis seems not to have been assessed, but be alert for evidence of a reduced effect if ampicillin, rifampicin or any other oral antibacterial is given. Neomycin, which also affects activity of gut microflora, has been seen to interact similarly in animal studies,(See reference number 5) but limited evidence suggests metronidazole does not interact.

Houston JB,Day J, Walker J. Azo reduction of sulphasalazine in healthy volunteers. Br J Clin Pharmacol (1982) 14, 395–8.

Shaffer JL,Kershaw A, Houston JB. Disposition of metronidazole and its effects on sulphasalazine metabolism in patients with inflammatory bowel disease. Br J Clin Pharmacol (1986) 21, 431–5.

Shaffer JL,Houston JB. The effect of rifampicin on sulphapyridine plasma concentrations following sulphasalazine administration. Br J Clin Pharmacol (1985) 19, 526–8.

Shaffer JL,Hughes S, Linaker BD, Baker RD, Turnberg LA. Controlled trial of rifampicin andethambutol in Crohn’s disease. Gut (1984) 25, 203–5.

Peppercorn MA,Goldman P. The role of intestinal bacteria in the metabolism of salicylazosulfapyridine. J Pharmacol Exp Ther (1972) 181, 555–62.

Hypercalcaemia and possibly metabolic alkalosis can develop inpatients who are given high doses of vitamin D and/or largeamounts of calcium if they are also given diuretics such as thiazides, which can reduce urinary excretion of calcium.One case of hypercalcaemia has been reported in a patient usinga high-strength topical tacalcitol with a thiazide diuretic.

An elderly woman taking hydrochlorothiazide 25mg and triamterene 50mg daily became confused,disorientated and dehydrated 6 months after starting to take vitamin D2 50 000 units and calcium 1.5 g daily (as calcium carbonate) for osteoporosis. Her serum calcium level had risen to about 3.5 mmol/L (normal range about 2 to 2.6 mmol/L) (See reference number 1).

A 47-year-old man was admitted to hospital complaining of dizziness and general weakness,which had begun 2 months previously. He was taking chlorothiazide 500mg daily for hypertension,thyroid 120mg daily for hypothyroidism and calcium carbonate 7.5 to 10 g daily for heartburn. On examination he was found to have metabolic alkalosis with respiratory compensation,a total serum calcium concentration of 3.4 mmol/L (range given as 2.15 to 2.6 mmol/L) and an abnormal ECG. He was diagnosed as having milk-alkali syndrome. Recovery was rapid when thiazide and calcium carbonate were withdrawn and a sodium chloride infusion, furosemide and oral phosphates were given (See reference number 3).

In both cases thiazide diuretic was thought to be implicated as levels of calcium ingestion were in region of normally recommended doses

In a group of 12 patients treated for hypoparathyroidism with vitamin D (dihydrotachysterol or ergocalciferol),5 patients became hypercalcaemic when they took bendroflumethiazide or methyclothiazide (See reference number 5). A significant rise in plasma calcium levels occurred in 7 patients given vitamin D and methyclothiazide or chlorothiazide,and hypercalcaemia developed in 3 of them (See reference number 6). A study in 12 children taking calcitriol (31 nanograms/kg daily) found that addition of hydrochlorothiazide (1 to 2 micrograms/kg daily) reduced urinary excretion of calcium caused by calcitriol (See reference number 7). Another study in 7 patients with vitamin D-induced calciuria found that addition of hydrochlorothiazide and amiloride reduced urinary excretion of calcium due to calcitriol to a greater extent than hydrochlorothiazide alone. Moreover, addition of amiloride helped to prevent adverse effects associated with use of hydrochlorothiazide, such as hypokalaemia and alkalosis (See reference number 8).

55 mmol/L 28 days after starting tacalcitol ointment and it fell back to within normal range within 7 days of stopping ointment (See reference number 9)

The thiazide diuretics (and triamterene) can cause calcium retention by reducing its urinary excretion. This, added to increased intake of calcium, resulted in excessive calcium levels. Alkalosis (the milk-alkali syndrome, associated with hypercalcaemia, alkalosis, and renal impairment) may also occur in some individuals because thiazide limits excretion of bicarbonate.

The incidence is unknown but reports cited(See reference number 5,6) suggest that it can be considerable if intake of vitamin D and calcium are high. Concurrent use need not be avoided; thiazides have been used clinically to reduce vitamin-D induced hypercalciuria,(See reference number 7,8) but serum calcium levels should be regularly monitored to ensure that they do not become excessive. Patients should be warned about ingestion of very large amounts of calcium carbonate (readily available without prescription) if they are taking thiazide diuretics

The case of hypercalcaemia with use of a topical vitamin D analogue is rare and strength of preparation of tacalcitol used was fivefold higher than current licensed preparation of 4 micrograms/g (Curatoderm). However,bear this case in mind should a patient taking thiazides with a topical vitamin D analogue develop hypercalcaemia.

Drinka PJ,Nolten WE. Hazards of treating osteoporosis and hypertension concurrently withcalcium, vitamin D, and distal diuretics. J Am Geriatr Soc (1984) 32, 405–7.

Parfitt AM. Chlorothiazide-induced hypercalcaemia in juvenile osteoporosis and hyperparathyroidism. N Engl J Med (1969) 281,55–9.

Gora ML,Seth SK, Bay WH, Visconti JA. Milk-alkali syndrome associated with use of chlorothiazide and calcium carbonate. Clin Pharm (1989) 8, 227–9.

Hakim R,Tolis G, Goltzman D, Meltzer S, Friedman R. Severe hypercalcemia associated withhydrochlorothiazide and calcium carbonate therapy. Can Med Assoc J (1979) 121, 591–4.

Parfitt AM. Thiazide-induced hypercalcaemia in vitamin D-treated hypoparathyroidism. Ann Intern Med (1972) 77,557–63.

Parfitt AM. Interactions of thiazide diuretics with parathyroid hormone and vitamin D. Studiesin patients with hypoparathyroidism. J Clin Invest (1972) 51,1879–88.

Santos F,Smith MJ, Chan JCM. Hypercalciuria associated with long-term administration ofcalcitriol (1,25-dihydroxyvitamin D3). Action of hydrochlorothiazide. Am J Dis Child (1986) 140, 139–42.

Alon U,Costanzo LS, Chan JCM. Additive hypocalciuric effects of amiloride and hydrochlorothiazide in patients treated with calcitriol. Miner Electrolyte Metab (1984) 10, 379–86.

Kawaguchi M,Mitsuhashi Y, Kondo S. Iatrogenic hypercalcemia due to vitamin D3 ointment (1,24(OH)2D3) combined with thiazide diuretics in a case of psoriasis. J Dermatol (2003) 30, 801–4.

The observation that dependent oedema in a group of epileptics was higher than expected, and that response to diuretic treatment seemed to be reduced, prompted further study. In 30 patients taking phenytoin 200 to 400mg daily with phenobarbital 60 to 180mg daily maximal diuresis in response to furosemide 20 or 40mg occurred after 3 to 4 hrs instead of usual 2 hours, and total diuresis was reduced by 32 % for 20mg dose and 49 % for 40mg dose. When intravenous furosemide 20mg was given, total diuresis was reduced by 50%. Some of patients were also taking carbamazepine, pheneturide, ethosuximide, diazepam or chlordiazepoxide (See reference number 1).

Another study in 5 healthy subjects given phenytoin 100mg three times daily for 10 days found that maximum serum levels of furosemide 20 mg, given orally or intravenously, were reduced by 50 % (See reference number 2).

Not fully understood. One suggestion is that phenytoin causes changes in jejunal sodium pump activity, which reduces absorption of furosemide,(See reference number 2)but this is not whole story because an interaction also occurs when furosemide is given intravenously (See reference number 1). Another suggestion, based on in vitro evidence, is that phenytoin generates a liquid membrane, which blocks transport of furosemide to its active site (See reference number 3).

Information is limited but interaction is established. A reduced diuretic response should be expected in presence of phenytoin

Ahmad S. Renal insensitivity to frusemide caused by chronic anticonvulsant therapy. BMJ (1974) 3,657–9.

Fine A,Henderson JS, Morgan DR, Tilstone WJ. Malabsorption of frusemide caused byphenytoin. BMJ (1977) 2, 1061–2.

Srivastava RC,Bhise SB, Sood R, Rao MNA. On the reduced furosemide response in the presence of diphenylhydantoin. Colloids and Surfaces (1986) 19, 83–8.

Clinical evidence,mechanism, importance and management

A 63-year-old woman who had been taking cyclothiazide/triamterene and acebutolol for 4 years,developed polyuria and polydipsia within 3 weeks of starting to take pravastatin 20mg daily, which gradually worsened. After another 4 months she was hospitalised for hyperglycaemia,which was treated with insulin and later glibenclamide (glyburide). The cyclothiazide/triamterene and pravastatin were stopped and gradually diabetic symptoms began to abate. Five weeks after admission she was discharged without need for any antidiabetic treatment with diabetes fully resolved (See reference number 1). The detailed reasons for this reaction are not understood, but it would seem that pravastatin increased hyperglycaemic potential of thiazide diuretic to point where frank diabetes developed. This is an isolated case and there would seem to be little reason normally to avoid concurrent use of these drugs.

1. Jonville-Bera A-P,Zakian A, Bera FJ, Carré P, Autret E. Possible pravastatin and diuretics-induced diabetes mellitus. Ann Pharmacother (1994) 28, 964–5.

Urapidil does not appear to affect pharmacokinetics of digoxin

Clinical evidence,mechanism, importance and management

In 12 healthy subjects urapidil 60mg twice daily on days 5 to 8 had no significant effects on serum levels of digoxin 250 micrograms twice daily on day one, then 250 micrograms daily on days 2 to 8. Blood pressures and pulse rates were not significantly changed (See reference number 1). No special precautions seem necessary if both drugs are given.

1. Solleder P,Haerlin R, Wurst W, Klingmann I, Mosberg H. Effect of urapidil on steady-stateserum digoxin concentration in healthy subjects. Eur J Clin Pharmacol (1989) 37, 193–4.

Thyrotoxic patients are relatively resistant to effects of digitalis glycosides and may need reduced doses as treatment with antithyroid drugs (carbimazole, thiamazole) progresses, whereaspatients with hypothyroidism may need increased doses of digitalis glycosides as treatment with thyroid hormones progresses. Carbimazole has been shown to reduce serum digoxin in healthysubjects.

The observation of relatively low plasma digoxin levels in a patient taking carbimazole prompted a further study in 10 healthy subjects. In 9 out of 10, steady-state peak serum digoxin levels were reduced by 23 % (from

1.72 to 1.33 nanograms/mL) by a single 60mg dose of carbimazole, but in other subject serum digoxin levels were increased. Other pharmacokinetic parameters were unaffected.

Carbimazole abolished systolic blood pressure decrease seen in first 3 hrs with digoxin, and also reduced duration of digoxininduced diastolic blood pressure fall from 12 to 6 hours. The changes in heart rates,cardiac output and stroke volumes were not statistically significant, but inter-individual differences were large (See reference number 1-3).

A study in 12 patients with hyperthyroidism found that normalisation of serum T3 and T4 by thiamazole treatment did not produce significant changes in pharmacokinetics of digoxin (See reference number 4)

One explanation for changed response to digitalis with carbimazole is that there is a direct and altered response of heart due to raised or lowered thyroid hormone levels. Another is that changes in glomerular filtration rate associated with hypo- or hyperthyroidism result in increased or decreased serum digoxin,respectively (See reference number 5). Why carbimazole reduced serum digoxin in healthy subjects (normal thyroid status) is not known.

As thyroid status is returned to normal by use of drugs (antithyroid drugs or thyroid hormones), dosage of digitalis glycosides may need to be adjusted appropriately. Hyperthyroid patients may need to have their digitalis dosage gradually reduced as treatment proceeds (because initially they are relatively resistant to effects of digitalis and start off needing higher doses). They are also relatively insensitive to chronotropic effects of digitalis (See reference number 6,7). Hypothyroid patients on other hand may need a gradually increasing dosage (because initially they are relatively sensitive to digitalis) (See reference number 5,6). In either of these situations it would be prudent to monitor serum digoxin levels and glomerular filtration rate as treatment continues. The reduction of serum digoxin by carbimazole in healthy subjects does not fit with need to decrease digoxin doses when antithyroid drugs are used in patients. Further study is needed.

Petereit G,Ramesh Rao B, Siepmann M, Kirch W. Influence of carbimazole on the steady stateserum levels and haemodynamic effects of digoxin in healthy subjects. Eur J Clin Pharmacol (1995) 49, A159.

Rao BR,Petereit G, Ebert U, Siepmann M, Kirch W. Influence of carbimazole on the steadystate serum levels and haemodynamic effects of digoxin in healthy subjects. Therapie (1995) 50 (Suppl), 406.

Rao R,Petereit G, Ebert U, Kirch W. Influence of carbimazole on serum levels and haemodynamic effects of digoxin. Clin Drug Invest (1997) 13, 350–4.

Gasińska T,Izbicka M, Dec R. Digoxin pharmacokinetics in hyperthyroid patients treated withmethimazole. J Endocrinol (1997) 152 (Suppl), P285.

Croxson MS,Ibbertson HK. Serum digoxin in patients with thyroid disease. BMJ (1975) 3, 566–8.

Lawrence JR,Sumner DJ, Kalk WJ, Ratcliffe WA, Whiting B, Gray K, Lindsay M. Digoxinkinetics in patients with thyroid dysfunction. Clin Pharmacol Ther (1977) 22, 7–13.

Huffman DH,Klaassen CD, Hartman CR. Digoxin in hyperthyroidism. Clin Pharmacol Ther (1977) 22, 533–8.

Clinical evidence,mechanism, importance and management

Retrospective analysis of data from a large randomised,placebo-controlled study of HRT (conjugated estrogens/medroxyprogesterone 0.625/2.5 mg daily) in women with coronary heart disease was conducted to see if there were any subgroups of patients who responded differently. Use of digitalis was associated with a fivefold excess rate of cardiovascular events in first year in women receiving HRT, when compared with control group. A lower 1.5-fold excess rate was seen over whole duration of study (average 4.1 years). Possible mechanisms could be a drug-drug interaction or a drug-disease (HRT with congestive heart failure) interaction (See reference number 1).

However, it is impossible to say whether this represents a true effect, because number of positive sub-group analyses in this study was same as number predicted by chance alone. Confirmatory evidence is required (See reference number 1).

1. Furberg CD,Vittinghoff E, Davidson M, Herrington DM, Simon JA, Wenger NK, Hulley S.Subgroup interactions in the Heart and Estrogen/Progestin replacement study. Lessons learned.Circulation (2002) 105, 917–22.

A study in 8 healthy subjects found that ginkgo biloba leaf extract80 mg three times daily had no significant effects on pharmacokinetics of a single 500-microgram dose of digoxin

1. Mauro VF,Mauro LS, Kleshinski JF, Khuder SA, Wang Y, Erhardt PW. Impact of Ginkgobiloba on the pharmacokinetics of digoxin. Am J Ther (2003) 10, 247–51.

Guanadrel did not affect pharmacokinetics of a single dose ofdigoxin

Clinical evidence,mechanism, importance and management

In 13 healthy subjects guanadrel 10mg orally every 12 hrs for 8 days did not affect pharmacokinetics of a single intravenous dose of digoxin given on day 5. One subject experienced a 10-minute episode of asymptomatic second-degree heart block (Wenckebach) 3 hrs after dose of digoxin, but reason for this effect was not clear (See reference number 1). There seem to be no reports of adverse interactions between digitalis glycosides and any of guanethidine-like antihypertensive drugs.

1. Wright CE,Andreadis NA. Digoxin pharmacokinetics when administered concurrently withguanadrel sulfate. Drug Intell Clin Pharm (1986) 20, 465.

Plasma digoxin levels are unaltered or only modestly increased bythe use of flecainide,but this is not likely to be important in mostpatients.

In contrast, a study in 15 healthy subjects found that flecainide 200mg twice daily increased trough and peak plasma levels of digoxin 250 micrograms by 24 % and 13%, respectively (See reference number 2). The changes observed in vital signs were not clinically significant. Based on results of a single-dose study steady-state digoxin levels were predicted to rise by about 15 % during use of flecainide 200mg twice daily (See reference number 3).

It is suggested that any changes may be due to alterations in volume of distribution (See reference number 3)

Documentation is limited but what is known suggests that either no interaction occurs,or any changes are small and unlikely to be clinically relevant in most patients. However, UK manufacturers of flecainide recommend that digoxin plasma levels should be measured not less than 6 hrs after any digoxin dose, before or after administration of flecainide (See reference number 4). The US manufacturers do not advise any additional monitoring (See reference number 5). The authors of one of reports(See reference number 2)suggest that patients with high drug levels, atrioventricular nodal dysfunction, or both, should be monitored during concurrent treatment.

McQuinn RL,Kvam DC, Parrish SL, Fox TL, Miller AM, Franciosa JA. Digoxin levels in patients with congestive heart failure are not altered by flecainide. Clin Pharmacol Ther (1988)43, 150.

Weeks CE,Conard GJ, Kvam DC, Fox JM, Chang SF, Paone RP, Lewis GP. The effect of flecainide acetate, a new antiarrhythmic, on plasma digoxin levels. J Clin Pharmacol (1986) 26, 27–31.

Tjandramaga TB,Verbesselt R, Van Hecken A, Mullie A, De Schepper PJ. Oral digoxin pharmacokinetics during multiple-dose flecainide treatment. Arch Int Pharmacodyn Ther (1982)260, 302–3.

Tambocor (Flecainide acetate). 3M Health Care Ltd. UK Summary of product characteristics,May 2006.

Tambocor (Flecainide acetate). 3M Pharmaceuticals. US Prescribing information,June 1998.