ANTIBIOTIC GUIDELINES : PHARMACOLOGICALPARAMETERS TO BE CONSIDERED

IDarmansjah, RHH Nehvan

taitfjiUiflikttolCenter for to! SMfBlBrngEtfieac)

Fakultas Kedokteran Universitas Indonesia / Rumah Sakit Dr Cipto Mangunkusumo

Reprinted from : Medical Journal of the University of Indonesia Vol 3, No 1, January-March 1994With permission of the Publisher

Vol 3, No 1, January-March 1994 Antibiotic Guidelines 3

Antibiotic Guidelines : Pharmacological Parameters to be ConsideredI Darmansjah , RHH Nelwan

AbstrakTujuan pedoman penggunaan antibiotik adalah untuk membimbing penggunaannya secara rasional, sehingga antibiotik

digunakan secara efektif dan efisien dengan efek samping sekecil-kecilnya bagi penderita dan masyarakat. Tindakan ini juga berdampakmengurangi biaya tanpa mengorbankan penderita/masyarakat terhadap pemilihan antibiotik yang salah. Pedoman ini harus memenuhipersyaratan tertentu agar dapat diterima oleh pemakainya dan berlaku bagi sebagian besar (kira-kira 90%) kelompok penderita.Antibiotik terpilih yang diusulkan harus efektif tanpa adanya hasil antibiogram. Pedoman ini harus berdasarkan pengetahuanmikrobiologi, farmakologi dan klinik. Parameter farmakologi yang akan dibahas antara lain adalah farmakodinamik, farmakokinetik,efek toksik dan efek samping, hasil uji klinik, percobaan epidemiologi obat serta farmakologi perbandingan antibiotik. Strategi pemilihanterutama berdasarkan penerapan rasio manfaat-risiko serta biaya; manfaat antibiotik yang diperkirakan untuk indikasi tertentu haruslebih besar dari kemungkinan risiko yang dapat ditimbulkannya. Prinsip "Primum non nocere" haruslah merupakan pertimbanganutama. Antibiotik sering menimbulkan efek samping dan karena pemberiannya perlu bertujuan jelas. Menganggap bahwa "bila tidakberefek tentunya tidak berbahaya" tidaklah benar. Pedoman bukanlah suatu aturan yang kaku; pertimbangan ilmiah harus selalumerupakan petunjuk akhir, terutama pada keadaan klinik yang kompleks. Pedoman pemilihan antibiotik perlu diperbaharui ternsmenerus bilamana ada informasi yang baru. Lazimnya tidak perlu ada sanksi terhadap penyimpangan kecil dari pedoman yang ada,namun penyimpangan jauh dari suatu pedoman memerlukan dasar ilmiah yang jelas.

Abstract

The aim of antibiotic guidelines is to provide guidance on the rational use of antibiotics, so that they serve effectively and efficientlywith the least adverse effects on patients and community. The impact of such strategy may also be seen in the containment of cost withoutjeopardizing the patient to unsuccessful attempts of antibiotic treatment. Guidelines have, to satisfy certain requirements in order to beaccepted by its users. It should work for the majority (appro*. 90%) of patient groups and the antibiotic selections suggested should beeffective without or before the results of antibiograms. Antibiotic guidelines must be based on microbiological, pharmacological andclinical knowledge. Among pharmacological parameters of the antibiotic that will be discussed are pharmacodynamics, phar-macokinetics, toxicity and adverse effects, results of clinical trials and drug-epidemiological studies, and the comparative pharmacologyof antibiotics. The main strategy for selection is based on benefit-risk-cost assessment; the presumed benefit that the antibiotic couldoffer for a certain indication should be greater than the possible risks it could give rise to. The principle of treatment: "Primum nonnocere" should always be a prior consideration. In this respect the great potential of antibiotics to cause adversity is very oftenoverlooked, causing the erroneous attitude of "if it does not work, it does not harm either". Guidelines are never meant to be rigid rules;scientific judgement should always be the final guide, especially in complicated clinical conditions. Renewal of choices of antibioticsshould be instituted whenever recent information is available and updating of an antibiotic guideline is mandatory every year or two.Usually there are no sanctions for violating good advice, but major deviations from guidelines must be scientifically justifiable.

Keywords : Antibiotic use, Antibiotic selection, Rational use

INTRODUCTION of microorganisms could be minimized.lf2 The impactof such strategy may also be seen in the containment

The aim of guidelines for antibiotic-use is to provide of cost without jeopardizing the patient to unsuccessfulguidance on the rational use of antibiotics, so that they antibiotic treatment.serve effectively and efficiently with the least adverse Antibiotic guidelines set general guidance on howeffects on patients and community, and that resistance antibiotics should be correctly used. It gives

of Pharmacology, Faculty of Medicine University of Indonesia, Jakarta, IndonesiaDepartment of Internal Medicine, Faculty of Medicine University oflndonesia/Dr. Cipto Mangunkusumo Hospital, Jakarta, Indonesia

Darmansjah and Nelwan MedJ Univlndon

tions on the best antibiotic choice (s) available for thetreatment and prophylaxis of bacterial, fungal andparasitic infections.

Guidelines have to satisfy certain requirements inorder to be accepted by its users. It should work for themajority of patient population and the antibiotic selec-tions suggested should be effective without or beforethe results of sensitivity testings. The majority of in-fections in fact may be treated without the necessity ofdoing any culture of an infected biological specimen.Therefore antibiotic guidelines must be based on theintegration of microbiological, pharmacological andclinical knowledge; the results of clinical trials andepidemiological studies especially should determinethe validity of the selections made. Not to be underes-timated is the value of a correct diagnosis beforeprescribing an antibiotic in order to distinguish the realneeds for eradicating or preventing an infection. Fur-thermore, guidelines should be updated or renewedwhenever new knowledge becomes available.

GENERAL CONSIDERATIONS

Among pharmacological parameters that are importantto be considered are : pharmacodynamics, pharmaco-kinetics, toxicity and adverse effects of the antibioticunder scrutiny, site of infection, special effects on highrisk patient groups in relation to the antibiotic beingconsidered, results of clinical trials, drug-epidemio-logical studies, and above all, the comparative phar-macology of all antibiotics, especially of those belong-ing to the same class.

The main strategy for selection of a treatment isbased on benefit-risk-cost assessment; the presumed orcalculated benefit that an antibiotic could offer for acertain indication should be greater than the possiblerisks it could give rise to. The principle of treatment :"Primum non nocere", which broadly means "do notharm when treating patients", should always be a priorconsideration. Cost should only be decided after allscientific judgements have been made and treatment-cost should be put before drug-unit-cost. When two ormore antibiotic selections can be made with more orless similar predicted outcome, then cost should be thedetermining factor.

As for each guideline, it never should be con-sidered as rigid rules; sound scientific judgement anda scientifically based selection for using antibioticsshould always be the final guide. But, major deviationsfrom guidelines must be scientifically justifiable, be-cause good guidelines should apply for the majority ofsubjects for whom the guidelines are developed.Guidelines also should take into consideration the level

of competence of the health service and prescriber forwhich it is meant; while gentamicin is a good antibioticto be used in a hospital setting where the close obser-vation of the patient is needed and where monitoringof antibiotic blood levels may be determined, in aprimary health setting it may be dangerous.

PHARMACODYNAMIC PARAMETERSAntibiotics are usually chemicals with no or little phar-macodynamic properties; they do not exert significantactions on the organs of the host. This is a goodcharacteristic of an antibiotic, because they preferablyshould only have killing properties on the invadingmicroorganisms without doing harm by stimulating orinhibiting other receptors of the host. Some antibioticshowever, may have actions on intact human organs orgive rise to effects that are adverse in nature, and assuch may limit the use or dose of an antibiotic. Ex-amples of such antibiotics are the aminoglycosides thatinhibit neuromuscular conduction when administeredin high doses. This curari form-like effect is seen withstreptomycin, kanamycin, gentamicin, neomycin, andprobably other aminoglycosides in high doses. Whenused with other muscle relaxants this effect will evenbe enhanced. Circumoral paresthesia is also well-known with drugs like streptomycin; and vestibulardysfunction occurs with relatively high therapeuticdoses.

Other reactions may be seen with antibiotics, suchas bone-marrow depression with chloramphenicol orcephalosporins, hypoprothrombinaemia with moxa-lactam, etc. These may not be regarded as pharmaco-dynamic because they occur seldom and are only seenwith very high doses or are a result of idiosyncraticreactions and are therefore unpredictable for a givenpatient. Phenotyping and genotyping of the way in-dividuals do metabolize certain drugs may in the futuredetermine which patient would be prone to the abovementioned adverse effects.4'5

The pharmacodynamic effects on the micro-or-ganisms are reflected in the eradicating properties ofthe antibiotic itself; they may be bactericidal or bac-teriostatic. The degree of eradicating abilities may beexpressed by the minimum inhibitory concentrations(MICs) of the antibiotic. While this simple in vitro testis an important parameter, it may not be used as thesingle determinant to select an antibiotic fortherapeutic or prophylactic purposes, because in vivoconditions do not always produce the same results. Theclinical evaluation of the patient must always be takeninto account.

An important point for consideration is thedestructive specificity for each antibiotic class; they

Vol 3, No 1, January-March 1994 Antibiotic Guidelines

may be divided into groups of microorganisms. Thusthere are antibiotics that mainly attack Gram positivebacteria such as benzylpenicillin, phenoxymethyl-penicillin, cloxacillin or flucloxacillin, erythromycin,spiramycin, and other macrolides. The fluoro-quinolones are more specific against Gram negativebacteria and although it may be active against Grampositive cocci, it is not to be used when the narrowspectrum antibiotics are sensitive to it and thereforefluoroquinolones are not recommended as the firstchoice for definite Gram positive infections such asstreptococcal or staphylococcal. In a group howeverthe antibiotics may have individual specific actions;gentamicin for instance would be more effectiveagainst pseudomonas than kanafnycin, and thecephalosporins may have different emphasis on an-tibacterial activity according to the time of develop-ment of this class of drugs during the last 20 years.

PHARMACOKINETIC PARAMETERS

First it must be ascertained that the antibiotic chosenis absorbed by the proposed route and is available atthe site of infection in the required concentration in itsactive form. Many antibiotics, although having an ade-quate MIC may not qualify for some of the otherparameters mentioned and therefore may not produceclinically good results.

Orally available antibiotics such as macrolides(erythromycin, spiramycin, etc), lincomycin, certainbeta-lactams like ampicillin and phenoxymethyl-penicillin are incompletely absorbed. Despite theirclinical usefulness it may not produce sufficientenough antibiotic concentrations in certain individualsto rely on in severe infections when given by mouth,especially when taken with food. Injectable dosageforms are most appropriate when high antibiotic levelsare required immediately, such as in contaminatedsurgical prophylaxis.

Tissue levels of an antibiotic is an importantparameter for clinical efficacy; some may show rela-tively high concentrations at the site of infectiondespite a low bioavailability (spiramycin, azithro-mycin). Chemical configuration of the antibiotic maydetermine this property; fat-soluble antibiotics, havinghigher afinity to tissue, will pass cell-membranes andthe blood-brain barrier at greater ease (clindamycinversus lincomycin).

Other important pharmacokinetic parameters are: speed of absorption, peak and trough levels, con-centration half-lifes, formation of active metabolites,and pathways of excretion.

SPECIAL PRECAUTIONSThere are special precautions that need to be takenwhen choosing antibiotics in patients with the follow-ing characteristics :* premature babies and geriatric patients* child-bearing women and nursing mothers* organ deficient patients, especially organs of excre-

tion (liver, gastrointestinal tract, and kidney)* specific diseases* patients who are on concomitant treatment with

drugs that are incompatible with the antibioticgiven.

ADVERSE EFFECTS OF ANTIBIOTICS

Although antibiotics may be life-saving when needed,it may produce adverse effects that are not to be dis-regarded. These side effects rank among the highest infrequency of all drug classes. It may therefore mini-mize the usefulness of the antibiotics prescribed. Ad-verse reactions varying from skin rashes, nausea andabdominal discomfort (the most frequent) to blooddyscrasia, hepatitis and fatal anaphylactic reactions arewell known. The propensity of one antibiotic givingmore frequent adverse reactions than another shouldbe taken into account in the selection process.

An important fact to be considered is the frequen-cy of use of a certain antibiotic; the more frequent it isused the more adverse reactions it will elicit. Assumingthe incidence rates of adverse reactions to a certainantibiotic to be constant, the absolute number of caseswill therefore increase and become noticeable as theantibiotic is used more frequently. Two examples mayhighlight this statement. When in earlier years penicil-lin G procaine injections were so common that almostevery other patient got a penicilline shot, anaphylacticreactions were rampant. Although most of these injec-tions were grossly inappropriate and therefore did onlycontribute falsely to cure, this practice went on formore than 2 decades. Some cases with fatal outcomeat last brought a drastic change because of cases oflitigation. Although it strangely ended with the verdict"not guilty", the practice of penicillin injections disap-peared and now 70% of primary health centres do noteven use penicillin any more, a phenomenon to bedeplored. This suggests that if antibiotics were usedmore appropriately, fewer patients would have beeneligible for penicillin injections with a resultantdecrease in the number of adverse effects. The oppositehappened with flucloxacillin; a total of 179 cases ofhepatitis associated with the use of flucloxacillin hasbeen reported to the Australian Drug Adverse ReactionCommittee until June 1992.7'8 This drug has been

6 Darmansjah and Nelwan MedJ Univ Indon

widely promoted in the Australian Antibiotic Guide-lines'* as the drug of choice against staphylococcalinfections. The resultant steep increase of its use hascaused a drastic rise of flucloxacillin associatedhepatitis which in turn might not justify the benefit-riskequation for using flucloxacillin routinely any more.The indication for flucloxacillin use has recently beenlimited to include only severe staphylococcal infec-tions, and caution was mandated when presribing thedrug for older people or when the drug is used for morethan 14 days.

RESULTS OF CLINICAL TRIALS

The results of clinical trials are the ultimate "proof ofthe pudding" that the antibiotic is efficacious for aspecific indication. However, the accessability of theresults on several clinical trials to make an informedjudgement on the profile of an antibiotic is usuallyscarce. The information is scattered in many journalsand contradictory results are difficult to unify, not theleast because of the industry's promotional activitiesin the form of seminars and literature. Evaluatedtextbooks should give an objective view on the prob-lem. The American Medical Association's DrugEvaluations9 is an annual which is an objective guidefor selecting drugs including antibiotics; it describesdrug classes and individual drugs. Another book en-titled Therapeutic Drugs10 is a monograph that hasbeen published recently; each monograph deals,among others, with "Major outcome trials" whichdescribes in a nutshell the results of important clinicalstudies. The generalizability or external validity ofsuch trials, however, are sometimes difficult to assess.Therefore antibiotic guidelines would be very usefulin the everyday application of these potent drugs.Studies, involving larger populations, such as in drugepidemiological research are also valid to supplementthe smaller scale clinical trials.

LEGAL ASPECTS OF AN ANTIBIOTIC GUIDE-LINE

The status of a scientific guideline should necessarilydeal with scientific issues. Some may have the idea thata guideline of this sort should also be the standard ofmeasure when one is confronted with legal implica-tions when deviations from it become a matter ofdispute in a litigation case.

We must emphasize that the legality for suchpurpose cannot be fully attributed to an antibioticguideline or any other treatment guidelines. Treatmentmodalities are a complex subject and adhering simply

to a guideline may sometimes be erroneous when oneis confronted by a complex situation or ramificationwhich necessitates one to deviate from a guideline. Itmay however not be assumed that a guideline can beviolated any time for invalid reasons. One mayreasonably speculate that a guideline works for themajority (90% or more) of cases, but if one departssubstantially from it there must be valid reasons tojustify one's choice. In a court case where drugs mighthave caused injury or death to a patient, there willusually be one or more experts who will testify whetherthe treatment had caused the injury. But even then it isnot easy to establish causality as it was exemplified bythe "Bendectin case" recently in an American Court-room. Bendectin is an antihistaminic used againtsvomiting in early pregnancy that was on the Americanmarket from 1956 until 1983, when it was bannedbecause of accusations that its may have causedphocomelia in 2 cases. The Court accepted only peerreviewed journals as "good science" and in 1991 aCalifornia appeals court rejected the parent's suit. Thiswas challenged by other scientists and the SupremeCourt will reopen the case at the time of this writing.The problem becomes more entangled because of theinvolvement of the notorious American lawsuit onhealth related matters and not the least, the statementmade that " 'experts' will testify to anything, for aprice".11

QUICK REFERENCE FOR ANTIBIOTIC IN-DICATIONS

It may be useful to design a quick reference in atable-form in which diseases are matched with theappropriate antibiotics. Table 1 depicts such aguidance; a first and subsequent choice is provided toaccomodate for situations in which the first one cannotbe used. This gives a first impression on what to useroutinely before going into the details of the clinicalcondition. It will serve the reader with a useful an-tibiotic to start with, which is correct according to theauthors' view at the time of this publication. It must beemphasized however that in making the final judge-ment, many of the above mentioned parameters shouldbe taken into consideration. This guideline should beused in conjunction with other (authoritative) literatureso that nuances caused by individual conditions of thepatient can be accomodated.

There are certain general principles that are incor-porated into this model guideline, some of them are :a. When antibiotic treatment has been initiated before

an antibiogram is made and clinical improvementhas been demonstrated clearly, it would be avisable

Vol 3, No 1, January-March 1994 Antibiotic Guidelines

Table 1. Antibiotic Choice In Selected Infections

Infection Microorganism (s) Involved Firsl choice Subsequent choice

Endocarditis,bacterial; acute

subacuteCystitisPyelonephritisGonorrhoeaUrethritis, nonspec.Lymphogranuloma ven.Infiltrate/furunculosisErysipelasImpetigoInfection.mouth/teethCandidiasis, oralTonsillitis, acutePharyngitis, acuteCommon cold, fluMeaslesChicken poxParotitis, epidemic

Pneumonia : adultchild

Otitis mediaSeplicemia.

intra-abdotninalTetanusGas gangreneDiphteriaDiarrhea, nonspecificTyphoid fever

Staph., strep, enterococciStriviridans, enterococciE.coli, proteus, staphylococcicolifonn bact.N.gonorrhoeaeChlamydiaChlamydiaStr.pyogenes, staphylococciSlr.pyogcnesSlr.progenes, Staph. aureusAerob + anaerobic microorg.CandidaStreptococcusVirus (> 90%), Str.pyogenesVirusVirusVirusVirus

Str.pneumoniaeH.influen/aeH.influenzae. streptococciAnything possibleColifonn, BacteroidcsCl.tetaniCl.perfringensC.diphteriaeVirus, fcxxl, or bacteria (rare)S.typhi, S.paratyphi

pen.G + clox,+ gcntamicinpen.G + gcntamisincontrimoxazole, ampicillincontrimoxazole, ampicillinpen.G, ampicillin, ceftriaxonetctracyclinc/doxycyclinetetracycline/doxycyclinepcn.G,crythromycin, (flu) cloxacillinpen.G.pen.G, pen.V.pen V, pen G, erythro, macrolidem y cost at i n lozenges/suspensionpen.G, pen.V, erythromycinwithout antibioticwithout antibioticwithout antibioticwithout antibiotic-bath regularlyno antibiotic-chew on chewing gumis most effectivepen.G, erythromycinamoxycillinampicillincephalosporin G3 (+ aminoglyc.)aminoglyc. + penicillin + metronid.pen.Gpen.Gpen.Gwithout antibiotic - bistnust saltchloramphenicol, ampicillin

vancomycm + gcntamicinidem

nitrofurantion, fluoroquinolonefluoroquinoloneerithromycin, doxycycline, fluoroquinoloneerithromycinsulfonamidc, erithromycinspiramycin, clindamycinerithromycin, clindamycinerithromycin, spiramycin, flucloxacillin+ mctronidazol (chronic infection)amphotericin lozengesspiramycin, clindamycinpen.V, pen.G, erithromycinamoxycillin, erithromycin, doxycyclineamoxycillin, erithromycin, doxycyclineamoxycillin, erithromycin, doxycycline

chloramphenicol, clindamycin + aminoglycchloramphenicol, cotrimoxazolecrithromycin, contrimoxazoleclindamycin + aminoglycclindamycin + cephalosporin G3clindamycin, chloramphenicol + metronidclindamycin, chloramphenicolerithromycintetracyclinc (when bacterial [V.cholerae])cotrimoxa/ole, fluoroquinolone

Adapted from Katzung B, ed. Basic and Clinical Pharmacology, 4th ed., 1989:619-21, Oilman AG, Goodman LS, Rail TW, Murad F, cd.The Pharmacological Basis of Therapeutics, 8lh ed., 1990, 1024-33, and Antibiotic Guidelines, 6th ed 1990-91, Health Department of Victoria,Australia.

** Only when secondary Infection is evident

not to change the antibiotic initially used foranother that was more sensitive according to theantibiogram results. These results may be used laterwhen the first treatment fails.

b. The local experience based on scientific judgementand the local sensitivity pattern of pathogens testedmust be taken into account when making antibioticselections. When recent sensitivity patterns arelacking, Table 1 may be used as a guide.

c. Ampicillin may be interchanged with amoxycillin,while cloxacillin with flucloxacillin, without caus-ing significant clinical outcome.

d. Although penicillin G and penicillin V have almostthe same antibiotic spectra, penicillin V shouldnever be used for serious infections. Its action is tooweak and aborption of the drug is limited, so that ahigh concentration in the blood cannot be attained.

e. Different macrolides may be used interchangeblyfor the indication cited in the table. Althougherythromycin is the prototype and is the moststudied, it may cause unbearable gastric irritationfor some patents. Other macrolides may be moretolerated by the stomach. It should always be given1/2 an hour before meals to guarantee adequateabsorption.

f. Cotrimoxazole may be substituted by trimetoprimalone, especially in urinary tract infections. Thesulfa component in cotrimoxazole gives rise tofrequent and sometimes serious adverse reactions,while the efficacy of trimetoprim alone is not dif-ferent.

g. When mycostatin lozenges are not available, it maybe substituted by drops or syrup or even tabletsdissolved in a little water. This may be used as a

8 Dannansjah and Nelwan MedJ Univ Indon

mouth-wash; the solution should be kept in themouth for a few minutes.

h. The viral infections listed in the table should nor-mally not be treated with an antibiotic, but situa-tions may arise where a patient may be having acomplication with a bacterial superinfection eachtime he/she suffers from flu and in this case anantibiotic is well justified. Lately, epidemicparotitis have been treated with all kinds of an-tibiotics and isoprinosine. None of these are effec-tive and the only sensible "treatment" is to givechewing gum to chew on, so that the movement ofthe jaws will open the canal of the parotis gland,thereby draining the contents. Patients with chick-enpox should also take a bath regularly with soapand water so as tq keep the skin clean, preventingsecondary infections. The use of potassium per-manganate in the bath water is redundant.

i. Fluoroquinolones may not be used in trivial condi-tions, and where other antibiotics are very effective,fluoroquinolones should be reserved when the firstchosen antibiotic has failed to produce satisfactoryresults.

j. Finally, guidelines have to be regularly updated.

REFERENCES

1. Parker MT. Antibiotic resistance in pathogenic bacteria.WHO Chronicle 1982;36(5): 191-6.

2. Control of antibiotic resistant bacteria : Memorandum froma WHO meeting. Am J Hosp Pharmacy 1984;41:1329-37.

3. Antibiotic Guidelines 6th ed. 1990-1991, Victorian MedicalPostgraduate Foundation Inc. Australia.

4. Vessel ES. Pharmacogenetic perspectives : Genes, drugs,and disease. Hepatology 1984;4:959-65.

5. Weber WW, Hein DW, Litwin A, Lower GM. Relationshipof acetylator status to isoniazid toxicity, lupus erythema*tosus, and bladder cancer. Federation Proceedings 1983;42:3086-97.

6. Pudjiadi LL, Dannansjah I. Survei syok anaphylactik akibatpenisilin G di Puskesmas, Jakarta. Maj Farmakol TerapiIndonesia 1988;5:9-12.

7. Flucloxacillin hepatitis. Australian Adverse Drug ReactionsBulletin 1992 February.

8. Australian Adverse Drug Reactions Advisory Committee.Personal communication.

9. AMA Drug Evaluations, Annual 1993. American MedicalAssociation, Chicago, Illinois.

10. Dollery, ed. Therapeutic Drugs Livingstone, 1992. LondonChurchill.

11. The meaning of junk. Newsweek 1993 March 29:42-4.