Abstracts

ORIGINAL ARTICLE

^IPharmacist - UFSC

^IIPharmacist - M.Sc., UFSC

^IIIUndergraduate student in Pharmacy, UFSC; Pharmacy Professor of UniSul

^IVChief physician of the Service of Rheumatology, UFSC - Doctor in Rheumatology, USP

^VAssistant physician of the Service of Rheumatology, UFSC - Candidate for Doctor's Degree in Medical Sciences, UFSC

^VICollaborative researcher - Doctor in Production Engineering, UFSC

^VIIPharmacist, Professor of the Post-Graduation Program in Pharmacy, UFSC - Doctor in Natural Sciences by the Pharmazeutisches Institut - Universitat Bonn, Germany

Correspondence to

ABSTRACT

INTRODUCTION: The term polypharmacy, meaning the concomitant use of multiple medications by one individual, has been widely reported in institutionalized or elderly patients. It can, however, occur in patients with chronic diseases, such as rheumatoid arthritis (RA).

OBJECTIVE: To quantify polypharmacy in a group of RA patients and to assess the risk of potential undesirable interactions between medications used for managing RA and those used for non-chronic diseases.

METHODS: A cohort study was carried out with 103 RA patients registered at the Strategy of Access to Medications from the Brazilian Health Ministry, at the School of Pharmacy of the city of Florianópolis, state of Santa Catarina. Patients were monthly followed up by use of form completion. Drug interactions were identified by use of the Drugdex System - Thomson Micromedex^® - Interactions database.

RESULTS: Polypharmacy was found in 95.1% of the patients, and 19 potential undesirable interactions were observed between the drugs used by 74 patients (mean of 3.0 ± 1.2 interactions/patient). All potential interactions were related to methotrexate. Omeprazole was the major representative, accounting for 29.3% of the interactions, followed by diclofenac sodium (17.6%), and metamizole sodium (13.2%).

CONCLUSION: Considering that this study confirms that polypharmacy is a common therapeutic practice in RA patients, it is worth emphasizing the need for greater surveillance regarding the adverse effects or effectiveness reduction of certain drugs due to drug interaction.

Keywords: rheumatoid arthritis, drug interaction, polypharmacy, antirheumatic drugs.

INTRODUCTION

Rheumatoid arthritis (RA) is a chronic inflammatory autoimmune disease, characterized by progressive symmetric polyarticular impairment, and, in some cases, extra-articular manifestations.^1,2

Rheumatoid arthritis is estimated to affect 0.5% to 1% of the world population, women being three to four times more affected than men.^3,4

Ideal therapy varies according to the patient's individual characteristics, such as disease stage, activity, and severity, in addition to the response to previous treatment regimens.^5,6 Currently, the following five classes of drugs that benefit RA patients are available: analgesics; non-steroidal anti-inflammatory drugs (NSAID); corticosteroids; disease modifying anti-rheumatic drugs (DMARD); and biological drugs.

As the prevalence of comorbidities and risk factors, such as dyslipidemias, diabetes mellitus, hypertension, obesity, and osteoporosis, increases with age,⁷ the prescription of concurrent medications is frequent in RA patients. In addition, the use of medications for symptomatic treatments and/or self-medication can further increase the number of drugs used by RA patients.

Polypharmacy deserves special attention, because medications are chemical substances that can interact with each other, with nutrients, or with environmental chemical agents and trigger undesirable or iatrogenic responses.^8,9 Despite the lack of consensus about what number of medications expresses polypharmacy, many authors define it as the association of six or more medications or as the administration of a greater number of medications than that clinically indicated.^10,11

The term polypharmacy has been associated with institutionalized and elderly patients, but it can occur in other groups of patients with chronic diseases, such as RA. Thus, this study aimed at quantifying polypharmacy in RA patients and at assessing the risk of potential undesirable interactions between medications used for managing RA and those used for treating non-chronic affections.

METHODOLOGY

Our data were obtained from a cohort study of RA patients registered at the Strategy of Access to Medications from the Brazilian Health Ministry, at the School of Pharmacy of the UFSC/PMF, in the city of Florianópolis, state of Santa Catarina, from August 2008 to February 2010.

The sample comprised adult RA patients of both sexes, living in the city of Florianópolis and registered at the Strategy of Access to Medications from the Brazilian Health Ministry to receive the following medications: adalimumab, infliximab, etanercept (biologic agents), and leflunomide (DMARD).

Patients who did not want or could not participate in the study were excluded (20), and the following patients were considered as losses: those who did not sign the written informed consent (4); those who abandoned the study (8); and a deceased patient (1). The total sample comprised 103 patients. Total follow-up included one month of initial assessment and monthly follow-up for 12 months.

Data were collected by the major author during monthly interviews with patients, at the occasion of drug dispensation, by using previously validated follow-up forms. The following variables were collected: sex, age, medications used for RA treatment, and other medications used during the study period with or without medical prescription.

Considering the maximum number of drug associations used, the patients were classified according to the presence or absence of polypharmacy. In this study, polypharmacy was defined as the association of six or more medications, regardless of the duration. Topical medications, domestic formulations, and eye solutions were not included in the evaluation.

This study assessed the existence of possible interactions between the medications used for treating non-chronic affections and those used for treating RA. The medications used for treating non-chronic affections were those used by patients at any time of the study and that had no indication to be used in chronic diseases. The drugs used in the symptomatic RA treatment were included in the group of drugs for non-chronic affections.

Those medications used for non-chronic diseases during the follow-up of the RA patients were classified according to the Anatomical Therapeutic Chemical (ATC) Classification¹² into their therapeutic groups.

The ATC Classification is an international classification recommended by the World Health Organization. It consists in classifying medications into different groups and subgroups (levels) according to the organ or system upon which the medications act and according to their chemical, pharmacological, and therapeutic properties. There are five levels, the fifth level being the active chemical substance.¹²

The Drugdex System - Thomson Micromedex^® -Interactions¹³ database was used to assess reports of possible interactions between the drugs previously cited.

An association of the presence of potential interactions was observed with the variables sex, age bracket, presence of polypharmacy, and therapeutic schemes including methotrexate, by using the chi-square (χ2) and Fisher's Exact tests, with 95% confidence interval.

The associations showing potential drug interaction were classified and presented according to their severity, scientific evidence available, and time for the beginning of the adverse effect, according to that same database.

Regarding severity, the drug associations were classified as follows: contra-indicated, when the concomitant administration of drugs is not recommended; major, when the association represents a threat to life, requiring immediate medical intervention; moderate, when the patient's clinical findings get worse, requiring alteration of the medicamentous therapy instituted; minor, when the patient's clinical findings change, but alterations in the medicamentous therapy are not required.¹³

The scientific evidence available about drug interactions was classified as follows: excellent, when controlled clinical trials confirm the existence of drug interaction; good, when documents about the interaction exist, but controlled clinical trials lack; fair, when documents about the interaction are scarce, but pharmacological considerations of the occurrence of the interaction exist; unknown, when no document in the literature confirms the drug interaction.¹³

Regarding the time for the beginning of adverse effects, interactions were classified as follows: rapid, when the adverse effects resulting from the interaction occur within less than 24 hours; slow, when the adverse effects resulting from the interaction do not occur within the first 24 hours; nonspecified, when there is no report in the literature about the time for the beginning of adverse effects after the simultaneous administration of the drugs.¹³

This study was approved by the Committee on Ethics in Research with Human Beings of the Universidade Federal de Santa Catarina (protocol 103/2008).

RESULTS

Regarding the sample, the female sex represented 89.7% of the patients, and the age ranged from 22 to 83 years (median, 57.4 years; mean, 56.9 ± 13.1 years).

Figure 1 shows the therapeutic regimens used by patients on study occasions T0, T4, T8, and T12, corresponding to 0, 4, 8, and 12 months, respectively. The combinations of DMARDs (49%, 44%, 39%, and 36% at T0, T4, T8, and T12, respectively) and of DMARDs and biological agents (35%, 37%, 35%, and 35% at T0, T4, T8, and T12, respectively) predominated, and methotrexate was present in approximately 60% of the therapeutic regimens. Throughout the study, a reduction in the association of DMARDs and an increase in the use of a biological agent in isolation (from 3% at T0 to 8% at T12) and of a DMARD in isolation (from 15% at T0 to 24% at T12) were observed.

The association of leflunomide and methotrexate was the predominant therapeutic regimen throughout the study.

The presence of polypharmacy was observed in 95.1% of the patients (Table 1). The means of the minimum and maximum numbers of medications associated per patient were 7.5 ± 3.2 and 12.2 ± 4.1, respectively.

Of the medications used by the patients studied, only those for treating non-chronic diseases were selected for assessing the risk of drug interaction, and they corresponded to 105 different active substances of 19 therapeutic groups.

Figure 2 shows the percentage frequency of the therapeutic groups used in the RA patients for treating their non-chronic diseases during the period studied. The three most frequently used therapeutic groups were the anti-inflammatory products (17.6%), analgesics (17.6%), and corticosteroids (15.8%). In such groups, the most used drugs were diclofenac sodium (36.4%), paracetamol (43.2%), and prednisone (72.3%), respectively.

The literature consulted evidenced 19 potential interactions, all related to methotrexate (Chart 1). Drug associations with potential interactions were used by 74 patients, in a total of 205 potential interactions and mean of 3.0 ± 1.2 potential interactions per patient. Nine patients showed only one potential interaction (Chart 2).

The literature also shows three potential drug interactions referring to chloroquine diphosphate and four others related to methotrexate, but no patient studied reported the use of such associations.

Of the variables assessed, only the presence of polypharmacy showed a positive correlation with the existence of drug interactions (Table 2).

Omeprazole was the major representative of the risk for potential drug interaction identified in the study, corresponding to 29.3% of them, followed by diclofenac sodium (17.6%), and metamizole sodium (13.2%).

Regarding the potential interactions identified in the study, 78.9% were classified as major, 21.0% had fair scientific evidence, and, in 21.0%, the beginning of adverse effects was rapid (Chart 2).

DISCUSSION

This study aimed at providing information about the risk of potential drug interactions in RA patients, considering that, in Brazil, that type of information is rare.

Studies have shown that combinations of DMARDs are more effective than monotherapy.^14,15,16 In addition, biological agents are more effective when combined with DMARDs, in particular methotrexate.^17,18,19 This justifies the strategy of combining DMARDs or biological agents found in most patients in this study.

The means of the minimum and maximum numbers of medications used per patient in this study were 7.5 ± 3.2 and 12.2 ± 4.1, respectively. In a survey about medications used by retired elderly, the mean number of medications used was 4.1 ± 3.0.²⁰ In 2006, Loyola Filho et al.,²¹ studying the prevalence of medication consumption in 1,598 elderly, reported that the mean number of medications used was 2.18.

The mean number of medications used and the frequency of polypharmacy in our study were higher than those found in other national studies.^22,23 However, studies about medication use and the term polypharmacy are usually associated with institutionalized and/or elderly patients, hindering the comparison of the results obtained with those of other studies.

According to Prybys et al.²⁴, in elderly patients, the risk of adverse effects, including drug interactions, increases 13% with the use of two medications, 58% with the use of five medications, reaching 82% in the presence of seven or more medications.

Polypharmacy is associated with an increase in the risk and severity of adverse reactions, in the risk of triggering drug interactions, of causing cumulative toxicity and mistakes during medication administration, and of reducing adherence to treatment. Therefore, it is directly related to health care costs, which include medications and the repercussions of their use, such as costs of visits to specialists, emergency health care, and hospital admission.^22,25

In this study, the most frequently used therapeutic groups were the anti-inflammatory products (17.6%), analgesics (17.6%), and corticosteroids (15.8%). These groups and their representatives are used for treating RA, and, in more severe cases, their full doses can be administered,^6,26 which explains their greater frequency of use.

It is worth noting that, in this study, the detection of potential drug interactions was based in an information technology tool included in the Drugdex System - Thomson Micromedex^® - Interactions¹³ database, which cannot consider aspects related to patients, dosages, and sequence and time of medication administration. This, thus, may overestimate the incidence and risk of potential drug interactions.

All potential drug interactions found in this study were related to methotrexate. In the literature consulted, no potential interaction of the medications selected in this study with those used for RA management (abatacept, adalimumab, etanercept, infliximab, leflunomide, and hydroxychloroquine sulfate) has been reported.^13,27,28

Methotrexate is the most used initial therapy in RA patients, except for those with liver disease.²⁶ Despite its excellent efficacy and tolerability profiles, in addition to its low cost,²⁹ methotrexate can interact with other medications, resulting in toxicity.

In this study, 19 types of potential interactions with methotrexate were observed, 15 of which were classified as major and four as moderate. This emphasizes the need for being cautious when prescribing and dispensing other medications to patients already using methotrexate in their treatment regimens, as well as the need for instructing patients regarding self-medication.

The interactions related to methotrexate and classified as major can result in an increase in the serum concentrations of its active metabolite, potentiating its adverse effects, such as leukopenia, thrombocytopenia, anemia, liver toxicity, nephrotoxicity, and mucosal ulcerations.^1,13,27,28

An increase in the serum concentrations of methotrexate, with signs and symptoms of toxicity, was observed when the drug was coadministered with a variety of penicillins, such as amoxicillin, benzylpenicillin, mezlocillin, piperacillin.^30,31 Because of the structural similarity between penicillins and methotrexate, a competitive inhibition of methotrexate tubular secretion can occur, increasing its half-life.^28,32,33

According to the Drugdex System - Thomson Micromedex^®-Interactions¹³ database, tetracyclines and other oral antibiotics can reduce the effectiveness of methotrexate by reducing its intestinal absorption or interfering with its enteropathic circulation, inhibiting the intestinal flora and suppressing the drug metabolism by bacteria. However, other authors believe that tetracyclines and their derivatives increase methotrexate serum concentrations.^27,34,35

The association of high doses of methotrexate with doxycycline can result in gastrointestinal and hematologic toxicity.³⁴ The interaction mechanism remains unknown, but it is believed to be related to the removal of methotrexate from its binding sites in serum proteins, the competition for renal tubular secretion, or the inhibition of the renal synthesis of prostaglandins.^13,27,34,35

No potential interaction between minocycline and methotrexate has been reported, but as minocycline is a tetracycline derivative, its association with methotrexate should be monitored.

Pancytopenia and myelotoxicity have been reported after the concomitant administration of methotrexate and sulfamethoxazole + trimethoprim.^36,37,38 The literature suggests the following two mechanisms of interaction: 1) sulfonamides can remove methotrexate from its binding site to serum proteins or reduce its renal excretion; and 2) additive inhibition of dihydrofolate reductase by methotrexate and trimethoprim.^36,39

In addition to pancytopenia and myelotoxicity, that association can increase the risk of megaloblastic anemia because methotrexate and sulfonamides can produce a folate deficiency by suppressing dihydrofolate reductase.^1,13,27,28

The concomitant use of methotrexate and ciprofloxacin can result in an increase in the serum concentrations of the former due to the inhibition of its renal tubular transportation caused by the latter.^1,13,27,28 Two cases of severe toxicity have been reported, and, thus, the association of high doses of methotrexate and ciprofloxacin should be avoided.⁴⁰

The association of methotrexate and NSAIDs can cause several complications, such as severe hematologic and gastrointestinal toxicity.^13,35 In addition, preexisting renal dysfunctions (or NSAID-induced renal dysfunctions) potentiate the risk for adverse reactions.^13,27,28 That interaction can occur through four different mechanisms: 1) competition between methotrexate and NSAIDs for renal tubular secretion; 2) removal of methotrexate, or its active metabolite, from the binding site to serum proteins; 3) a reduction in the liver metabolism of methotrexate caused by NSAIDs; and 4) NSAIDs inhibit the synthesis of prostaglandins (vasoconstrictors of renal capillaries), which results in a decrease in renal blood flow, and, thus, in the glomerular filtration of methotrexate.^41,42

Some studies have reported that the coadministration of methotrexate, piroxicam, naproxen, ketoprofen, and ibuprofen in RA patients did not affect the pharmacokinetic profile of methotrexate.^43,44

Although the concomitant administration of NSAIDs and methotrexate can potentially cause severe toxicity, low doses of that association are considered well tolerated, but the appearance of severe adverse effects should always be cautiously monitored.^13,35

Although metamizole sodium and acetylsalicylic acid have been classified as analgesics according to the ATC Classification, in the literature consulted, their mechanisms of interaction with methotrexate have been compared with the general mechanism of NSAIDs.^{13,27,28,45,46}

The major representative of the potential drug interactions identified was omeprazole, accounting for 29.3% of them. Its interaction was classified as major and the beginning of adverse effects was rapid. The concomitant use of methotrexate and omeprazole can increase the risk of toxicity of the former,^13,27,28,47 because, according to Suzuki et al.,⁴⁸ the coadministration of proton pump inhibitors can delay the excretion of methotrexate and potentiate its adverse effects.Omeprazole serum concentrations decrease rapidly after interruption of its use.^13,27,28,47 Thus, patients receiving that association should be strictly monitored to avoid possible damage resulting from the high concentration of methotrexate in their bodies.^13,27,28

The interaction between pantoprazole and methotrexate is less severe due to the degree of activation of pantoprazole according to the pH of the medium. With a pH of approximately 5, as that found in the renal tubules, pantoprazole is less active than omeprazole, having, thus, a lower inhibitory effect on methotrexate tubular secretion.⁴⁹

A double-blind, placebo-controlled study with 15 adult asthmatic patients was carried out to assess the effects of the concomitant use of theophylline and methotrexate.⁵⁰ Eight patients received methotrexate and seven received placebo for six weeks. After six weeks, theophylline excretion reduced from 48 to 38.9 mL/h/kg in patients treated with methotrexate. The serum concentrations of theophylline should be monitored when methotrexate is introduced, discontinued, or has its dosage altered.^13,50

Patients with RA undergo treatment protocols with multiple medications and for prolonged time. Those are important factors, which, when combined, can contribute to the occurrence of adverse reactions and undesirable drug interactions, which can worsen the pathophysiological scenario already installed and/or require the suspension or change of the therapeutic regimen.

It is worth emphasizing the need for assessing the risk-benefit relation of each drug association, as well as the adoption of measures that can reduce negative effects, such as changing the time of drug administration and monitoring adverse effects related to drug association.

The practice of self-medication, mainly for pain control, is common among RA patients, which puts them at greater risk for drug interactions, adverse effects, and toxicity. The therapeutic group with the highest potential of interaction with the medications used for RA treatment was that of the anti-inflammatory drugs.

To minimize or prevent the potential risk of adverse reactions and drug interactions, contributing to the rational use of medications and improvement of the health conditions of RA patients, the following is recommended: multidisciplinary teams; pharmacotherapeutic follow-up; and education of patients regarding the practice of self-medication.

REFERENCES

Publication Dates

History