替考拉宁血小板问题

International Journal of Antimicrobial Agents10(1998)143–152Comparative safety of teicoplanin and vancomycinA.P.R.Wilson*Department of Clinical Microbiology,Uni6ersity College Hospital Outpatient Building,Grafton Way,London,WCIE6DB,UKReceived18February1998;accepted18March1998AbstractTeicoplanin have different safety profiles which can affect choice.Nephrotoxicity is significantly less likely to occur during treatment with teicoplanin than vancomycin when an aminoglycoside is being given concurrently.‘Red man’syndrome is a troublesome effect of vancomycin infusion which is extremely uncommon with teicoplanin use.Rash and fever can be dose-related phenomena but patients reacting to one glycopeptide may not react to both.Although thrombocytopenia is more frequent with teicoplanin,it is reversible and seldom seen at standard doses.©1998Elsevier Science B.V./International Society of Chemother-apy.All rights reserved.Keywords:Nephrotoxicity;‘Red man’syndrome;Safety;Teicoplanin;Vancomycin1.IntroductionOver the past10years,the rise in device-related infections associated with coagulase-negative staphylo-cocci has resulted in a large increase in the use of glycopeptide antibiotics.In addition,Gram-positive bacteria resistant to i-lactam antibiotics,such as me-thicillin-resistant staphylococci,penicillin-resistant pneumococci and the enterococci,are becoming com-mon causes of nosocomial infection.Early preparations of vancomycin were associated with a high incidence of adverse effects,probably caused by the many impurities present.Highly purified forms are now available(Vancocin®,Eli Lilly)but some adverse events remain.Nephrotoxicity is seen when vancomycin is used together with aminoglycosides. Serum vancomycin assays are still requested by most clinicians to determine the appropriate dose,and avoid toxic levels.The‘red man’syndrome,a combination of erythema,pruritus and sometimes hypotension,can be a troublesome immediate adverse event of vancomycin infusion.Rigors have been reported,even with modern preparations(Vancoled®,Lederle)[1].Teicoplanin has a similar structure to vancomycin but seldom causes‘red man’syndrome,and is less nephrotoxic when combined with aminoglycosides. Routine serum assays are not necessary.In a series of 3377patients given teicoplanin,the most common re-ported adverse events were hypersensitivity(89patients, 2.6%),abnormal liver function(57patients, 1.7%), fever(27patients,0.8%),local intolerance(55patients, 1.6%),abnormal renal function(22patients,0.7%)and ototoxicity(11patients,0.3%)[2].The manufacturer holds data on6696patients treated with teicoplanin,of whom315(4.7%)had skin reactions,205(3.1%)gas-trointestinal events and229(3.4%)had other gener-alised symptoms(Hoechst Marion Roussel,data on file).Comparative trials of teicoplanin and vancomycin in the USA showed a similar incidence of adverse events, with the exception of thrombocytopenia(Table1)[3]. In a single,unpublished trial of treatment of endocardi-tis,11(79%)of14patients developed thrombocytope-nia while receiving teicoplanin,30mg/kg per day.This is well above the normal recommended dose.A meta-*Corresponding author.Tel.:+441713809518;fax:+441713888514;email:p.wilson@0924-8579/98/$19.00©1998Elsevier Science B.V./International Society of Chemotherapy.All rights reserved. PII S0924-8579(98)00025-9A.P.R.Wilson/International Journal of Antimicrobial Agents10(1998)143–152 144analysis of11comparative trials found significantly higher number of adverse events in patients given van-comycin(21.9%)than teicoplanin(13.9%)[4].Pooled information from six European trials studying serious Gram-positive infections at various sites showed a sig-nificantly higher incidence of rash and pruritus with vancomycin than with teicoplanin(27/193(14.0%)vs six of196(3.1%),P B0.001,x2-test)(Hoechst Marion Roussel,data onfile).There was also a greater inci-dence of abnormal renal function with vancomycin (24/193(12.4%)vs11/196(5.6%),P B0.05).Other re-views presented a similar conclusion[2,5].Of ten pub-lished and six unpublished comparative trials reviewed by Wilson et al.[5],five reported significantly more adverse events with vancomycin(usually rash and ab-normal renal function).The remainder showed no sig-nificant difference,but none suggested that vancomycin was the better tolerated antibiotic.2.NephrotoxicityAlthough the incidence of nephrotoxicity was much higher with the early,less purified preparations of vancomycin,today only5%of patients treated with vancomycin develop abnormal renal function,usually when trough levels exceed30mg/l[6,7].Acute intersti-tial nephritis has been demonstrated on renal biopsy from a patient treated with vancomycin for3weeks for Staphylococcus aureus endocarditis[8].In rats,van-comycin nephrotoxicity is dose dependent and,using urinary malate dehydrogenase activity as a marker,was demonstrable at doses\25mg/kg[9].One review found very few cases of nephrotoxicity in the literature not affected by confounding factors and doubted van-comycin was responsible for the nephrotoxicity re-ported in5%of patients in rigorous studies[10].Of greater importance is the synergistic nephrotoxic-ity between vancomycin and aminoglycosides.Weight loss and uraemia in rats during a10-day course of vancomycin,200mg/kg per day,plus tobramycin,80 mg/kg per day,were significantly greater than with either agent alone[11].The release of brush border enzymes in rats given vancomycin,200mg/kg i.p.,and gentamicin,100mg/kg i.m.,was significantly greater than in rats given gentamicin alone[12].A meta-analy-sis of eight studies suggested that the incidence of nephrotoxicity in patients given vancomycin plus an aminoglycoside was13%greater than in those given vancomycin alone(P B0.01)and4.3%greater than in those given an aminoglycoside alone(P B0.05)[13].In another study,the incidence of nephrotoxicity was14/ 63patients(22%)when vancomycin was used in combi-nation with aminoglycosides,compared with11% (11/103patients)when gentamicin was used alone[7]. However,one review found studies equally split be-tween those that did or did not support the observation of synergistic nephrotoxicity[10].In animals,the nephrotoxicity of teicoplanin is simi-lar to that of vancomycin on a weight-for-weight basis. Focal tubular necrosis develops in rats and dogs at a dose of40mg/kg per day given for1month.However, in clinical practice the usual maximum dose of tei-coplanin is15mg/kg per day,which is around half that of vancomycin.Interstitial nephritis has been reported in an elderly patient given teicoplanin,1200mg o.d., for40days[14].The incidence of nephrotoxicity with teicoplanin,6mg/kg per day,plus an aminoglycoside is lower than that with vancomycin plus an aminogly-coside(Table2)[15–24].A meta-analysis of nine com-parative trials showed a statistically significant higher incidence of nephrotoxicity in patients given van-comycin(58/54410.7%)compared with teicoplanin(28/ 585 4.8%),despite similar proportions given other nephrotoxic drugs including gentamicin[4].In a dou-ble-blind,randomised trial in febrile neutropenic bone marrow transplant recipients,six of25patients(24%) given vancomycin plus gentamicin developed cy-closporin-induced renal toxicity compared with none of 25patients given teicoplanin plus gentamicin(P=0.02) [19].Increases in serum creatinine were more common in patients given vancomycin than in those given tei-coplanin in a randomised trial of patients undergoing intensive care[22].In pooled trial results(Hoechst Marion Roussel,data onfile),two of143patients(1%) given teicoplanin versus20of227patients(9%)given vancomycin had serum creatinine levels increase toTable1Adverse events in collected US comparative trials of teicoplanin and vancomycinNumber of patients(%)Adverse eventTeicoplanin Vancomycin(n=406)(n=417)48(11.5)49(12.1)Hypersensitivity17(4.2)Fever12(2.9)Pruritus9(2.2)11(2.6)36(8.8)Rash33(7.9)5(1.2)Rigors3(0.7) Diarrhoea23(5.7)29(7.0)Nausea/vomiting15(3.6)11(2.7)5(1.2)Hearing and balance6(1.4)Liver function abnormali-5(1.2)6(1.4)tiesNephrotoxicity25(6.2)27(6.5)16(3.9)Haematological abnormali-5(1.2)ties14(3.4)Thrombocytopenia2*(0.5)These adverse events are irrespective of any likely causal relation-ship.The excess comprised patients given teicoplanin\12mg/kg per day.*P=0.003,Fisher’s exact test.A .P .R .Wilson /International Journal of Antimicrobial Agents 10(1998)143–152145T a b l e 2N e p h r o t o x i c i t y i n p u b l i s h e d a n d u n p u b l i s h e d c o m p a r a t i v e t r i a l s o f t e i c o p l a n i n a n d v a n c o m y c i nO t h e r a n t i b i o t i c sR e f e r e n c e D o s e a n d d u r a t i o n (l o a d i n g )I n d i c a t i o nN e p h r o t o x i c i t y V a n c o m y c i n O d d s r a t i o T e i c o p l a n i n V a n c o m y c i n D e fin i t i o n T e i c o p l a n i n 95%C .I .1g /12h ,9–35R i s e i n c r e a t i n i n e 0.5m g /S e v e r e i n f e c t i o n i n i m -0/37200–400m g /24h ,(4003/37V a n d e r A u w -—N o n em g ×3/12h ),4–17d a y s d a y s e r a e t a l .[15]m u n o c o m p r o m i s e d h o s t d l 1g /12h ,1–59U n e x p l a i n e d r i s e i n c r e a -4/2752/2521.8(0.3−10)C e f t a z i d i m e M e n i c h e t t i e t F e b r i l e n e u t r o p e n i a400m g /24h ,(600m g ×1d a y s h ),3–40d a y s a l .[16]t i n i n e a m i k a c i n (a l l )1g /12h —0/240/35—C o n y -M a k h o u l C e f t a z i d i m eF e b r i l e n e u t r o p e n i a400m g /24h ,(400m g ×3/12h ),d u r a t i o n n o t e t a l .[17]s t a t e d 400m g /24h ,(400m g ×—0/220/20—C e f t a z i d i m e C h o i e t a l .[18]500m g /8h ,7–F e b r i l e n e u t r o p e n i aa z t r e o n a m (a l l )3/12h ),7–12d a y s 12d a y s K u r e i s h i e t a l .1g /12h ,m e a n F eb r i l e n e u t r o p e n i aR i s e i n c r e a t i n i n e f r o m 400m g /24h ,(400m g ×P i p e r a c i l l i n t o -0/256/25—16d a y s3/12h ),m e a n 22d a y sb r a m yc i n (a l l )n o r m a l t o \1.1m g /d l [19](m e n )o r \1.0m g /d l (w o m e n )200–400m g /24h ,(400U n e x p l a i n e d r i s e i n c r e a -H i c k m a n c a t h e t e rS m i t h e t a l .[20]1/351g /12h ,M e d i a n 5/370.2(0.02–1.7)P i p e r a c i l l i n g e n t a m -m g ×1–2/12h ),M e d i a n i c i n (85%)t i n i n e \44.2v M7d a y s 8d a y s T r a n s i e n t s l i g h t i m p a i r -M R S A2/12V a n L a e t h e m e t 2/91g /12h ,10–480.7(0.08–6.2)—400m g /24h ,10–80d a y s d a y s a l .[21]m e n tV a r i o u s500m g /8h ,3–C h a r b o n n e a u e t \50%R i s e i n c r e a t i n i n e6/2417/320.3(0.09–0.93)N e t i l m i c i n (a l l )400m g /24h ,(400m g ×42d a y s 3/12h ),5–31d a y s a l .[22]V a r i o u s\50%R i s e i n c r e a t i n i n eN e v i l l e e t a l .1/28200–400m g /24h ,4005/280.2(0.02–1.6)s i x v e r s u s 14G i v e n 0.75–1g /12h ,a m i n o g l y c o s i d e s m g ×1h ),4–30d a y s [23]1–19d a y s —0/220/18E n d o c a r d i t i s ,v a s c u l a r—G i l b e r t e t a l .V a r i o u s (7%)400m g /24h ,(400m g ×1g /12h 3/12h )[24]200–400m g /24h ,(4000.5–1g /12h ,\50%R i s e i n c r e a t i n i n e14/11429/1090.4(0.2–0.8)V a r i o u s (70%)V a r i o u sC 02–5,8m e a n 14d a y s m g ×3/12h ),m e a n 12d a y s \50%R i s e i n c r e a t i n i n e2/1173/1210.7(0.1–4.2)V a r i o u s (51%)400m g /24h ,(400m g ×102–0091g /12h ,14–28V a s c u l a r d a y s 3–9/12h ),14–28d a y s 1g /12h ,14–726–10m g /k g p e r 12–24B a c t e r e m i a /e n d o c a r d i t i s—1/502/560.6(0.05–6.3)V a r i o u s (38%)102–014d a y s h ,(400m g ×1–9/12h ),14–72d a y s 102–0191g /12h ,9–486–30m g /k g p e r 24h ,—9/1806/1721.5(0.5–4.2)V a r i o u s (45%)B a c t e r e m i a /e n d o c a r d i t i s(6–30m g /k g ×3/12h ),d a y s 7–46d a y sN B ,t h e l a s t f o u r t r i a l s a r e u n p u b l i s h e d .A.P.R.Wilson/International Journal of Antimicrobial Agents10(1998)143–152 146\220v M(P=0.004,Fisher’s test).However,a large trial comparing teicoplanin with vancomycin in neu-tropenic patients also receiving amikacin showed no significant difference in nephrotoxicity[16].The incidence of increased creatinine levels has varied from four of1430patients(0.3%)tofive of614patients (0.8%)in large,non-comparative trials of teicoplanin [25,26].In an unpublished study(109–019,Hoechst Marion Roussel,data onfile),patients with a serum trough teicoplanin level\60mg/l had a higher inci-dence of elevated serum creatinine levels than those with teicoplanin levels of20–40mg/l(four of36(11%) vs14/43(33%),P=0.04,Fisher’s test).However, aminoglycoside usage may have been a confounding factor.3.OtotoxicityNeither teicoplanin nor vancomycin has been shown to cause ototoxicity in animals,even when administered with ethacrynic acid[27,28].Ototoxicity with van-comycin was only seen when aminoglycosides were also administered[29].There were no changes in auditory brain stem thresholds after administration of doses of vancomycin as large as200mg/kg to guinea pigs for 11–17days[30].In man,ototoxicity has been said to occur with vancomycin at trough concentrations of 13–32mg/l and peak levels of25–50mg/l,but it is uncommon[31].Brummett and Fox reviewed the many case reports of hearing loss associated with the use of vancomycin and found them to be confusing,particu-larly in relation to concurrent or preceding aminogly-coside use[27].In another study,no ototoxicity was reported in100courses of vancomycin therapy,though onlyfive patients had serial audiograms[6].One patient who had vertigo may have been given aminoglycosides. Using prospective audiometry,acute hearing loss of30 dB at5000Hz was found in one of23courses of vancomycin,but the patient was not aware of the change[32].Another patient given vancomycin and gentamicin was found to have a20dB hearing loss,but again this was asymptomatic.Of11other patients monitored by audiometry,one was found to have a 15–20dB loss,but gentamicin had been given concur-rently[33].In further studies,17episodes in12patients treated with vancomycin plus one dose of gentamicin, 1–2mg/kg i.v.,showed no evidence of auditory toxicity [34].Ototoxicity associated with teicoplanin may be no more than could be expected in post-surgical or other-wise debilitated patients[2].Of298patients given tei-coplanin and43given vancomycin who had repeated audiograms,only two had significant bilateral hearing loss;one patient given vancomycin plus aminoglycoside and another given teicoplanin,9mg/kg per day,for250days[3].Others have reported individual cases of oto-toxicity.Maher et al.[35]described a Down’s syndrome patient who developed high-frequency hearing loss dur-ing a prolonged course of treatment with teicoplanin. In a European open trial of teicoplanin including au-diometry,this was the only instance of ototoxicity in 101patients[36].Of64patients studied with serial audiograms in uncontrolled trials of various serious infections,three had deterioration of a previously nor-mal audiogram in the absence of other potentially ototoxic drugs[37–39].Only one of these had a‘serious reaction’as defined by the Council for International Organizations of Medical Sciences.The risk of ototoxicity and nephrotoxicity associated with vancomycin alone is low and not clearly related to serum levels.Routine assays may be unnecessary if vancomycin is given as a single agent,and some advise performing only trough assays[40].Serum level moni-toring is recommended,however,for patients with renal failure,pseudomembranous colitis or those given con-comitant aminoglycosides.Accumulation of van-comycin is a potential risk in elderly patients because of reduced clearance,and serum assays are then needed to judge the appropriate dose[31].Vancomycin serum concentrations can be predicted from the estimated creatinine clearance and the number of previous doses if a single peak and trough level have been taken within 48h of the start of treatment[41].Alternatively,serum concentrations of concomitant aminoglycosides can be used as a surrogate marker.Serum level monitoring of teicoplanin is only needed occasionally to ensure thera-peutic dosage,not to avoid toxicity.It is recommended when treating staphylococcal endocarditis,or patients who are drug abusers or who have renal failure,but not as a routine measure[5].84.‘Red man’syndrome‘Red man’syndrome(or‘red neck’)comprises ery-thema,pruritus andflushing of the upper torso and is often associated with rapid(B1h)infusion of thefirst dose of vancomycin(to be differentiated from‘red man’syndrome associated with rifampicin toxicity, sometimes called‘red man’s’syndrome).Less fre-quently,hypotension and angioedema can occur,and other reports include sudden pain and spasm in the chest or parasternal muscles from a normal electrocar-diogram.In many patients,the syndrome is a mild, evanescent pruritus at the end of the infusion which goes unreported.Most patients do not experience the same severity of reaction on subsequent doses,but it can occur for thefirst time after several doses or with a slow infusion.The syndrome has been reported with highly purified vancomycin and even oral preparations, suggesting that the vancomycin molecule itself is re-A .P .R .Wilson /International Journal of Antimicrobial Agents 10(1998)143–152147Table 3Occurrence of red man syndrome with vancomycin Dose (g)Infusion time (min)Reference Proportion of reported cases (%)Patient /volunteer 0.5Polk et al.[44]60Volunteer 0/11(0)1.0609/11(82)1.060Sahai et al.[50]11/12(92)Volunteer 1.060Volunteer 9/12(75)Sahai et al.[51] 1.0Healy et al.[52]60Volunteer 8/10(80)1.01203/10(30)1.060Rybak et al.[49]9/10(90)Volunteer 1.060Patient 0/15(0)1.020Schifter et al.[53]8/101(8)Patient 1.010Patient 11/76(14)Newfield and Roizen [54]15mg /kg 60Odio et al.[55]7/20(35)Patient 1.060Patient 8/17(47)Wallace et al.[56]O’Sullivan et al.[57]Patient1.0601/29(3)sponsible,rather than any impurities [42–45].Intraperi-toneal administration has also been implicated [46].Rash and hypotension can occur in new-born infants [47],and the incidence in a 3year study in children was 1.5%[48].The incidence of ‘red man’syndrome is uncertain,being reported in up to 47%of patients,but in up to 90%of volunteers (Table 3)[49,56].Differences in dose,infusion rate,concomitant drugs and the closeness of monitoring are thought to be responsible for the higher incidence in volunteers [44].Schifter et al.[53]observed the reaction in eight of 101patients (8%)given two 1g doses of vancomycin over 20min with a 4h interval.When infused over 10min,a 1g dose produced hy-potension in 14%of patients (11/76),but none was affected if vancomycin was infused over 30min [54].The effects resolve quickly when the infusion is discontinued.Some studies implicate histamine release as the cause of the characteristic symptoms of ’red man’syndrome,and suggest the lower incidence in patients compared to volunteers is the result of depletion of histamine stores by infection or trauma.In tissue culture,vancomycin causes degranulation of rat mast cells [58].Polk et al.[44]observed the reaction during a 1h infusion of vancomycin,1g,in 9of 11volunteers (82%),which was associated with a rise in plasma histamine levels.No reaction occurred with a 500mg dose.Healy et al.[52]noted symptoms in 8of 10volunteers (80%)given vancomycin,1g,over 1h but in only 3of 10(30%)given the same dose over 2h.Total histamine release was greater with the faster infusion.The effects of ‘red man’syndrome can be relieved by antihistamines.In a double-blind crossover volunteer study,pre-treatment with hydroxyzine,but not rani-tidine,significantly reduced the occurrence of erythema and pruritus after a l h infusion of vancomycin,1g (nine of 12(75%)vs one of 12(8%))[50].In another double-blind study,administration of diphenhydramine to patients before starting vancomycin infusion (1g over 1h)prevented all cases of ’red man’syndrome with the first dose of vancomycin (none of 16patients given diphenhydramine vs eight of 17patients (47%)not given this drug)[56].In this study only severe cases were associated with high plasma histamine levels.Oth-ers have failed to confirm the role of histamine,how-ever [57],and prolonged hypotension has been reported without changes in histamine levels [51].In another crossover study of vancomycin versus placebo in healthy volunteers,there was no significant relationship between serum histamine concentrations and the sever-ity of reaction [49].Hypotension can be troublesome if it occurs during anaesthesia following the use of vancomycin for surgi-cal prophylaxis [59,60].Vancomycin is also used as prophylaxis against bacterial endocarditis in penicillin-allergic patients undergoing dental or upper respiratory tract operations.The risk of allergic reaction during bolus injection and the closeness of blood pressure monitoring ensures that most cases of hypotension in surgical patients are detected [61].One trial was discon-tinued prematurely because of rash in seven of 20patients (35%)given vancomycin prior to CSF shunt insertion [55].When vancomycin,1g,given over 30min was used in addition to cefazolin during cardiac surgery,there was a significantly greater requirement for noradrenaline infusion to support blood pressure [62].Vancomycin has been administered as a rapid bolus into the oxygenator reservoir during cardiopul-monary bypass,however,with only a minor fall in arterial pressure and increase in venous capacitance [63].Given over 30–60min,administration of van-comycin before anaesthesia does not significantly alter blood pressure compared with administration after anaesthesia has begun [64].Teicoplanin does not cause ‘red man’syndrome in healthy volunteers [49,65],and is a very uncommon cause in patients with various Gram-positive infectionsA.P.R.Wilson/International Journal of Antimicrobial Agents10(1998)143–152 148[66–68].Just three cases(0.04%)were reported in6696 patients in clinical trials[3].In a double-blind, crossover study of12volunteers,11(91.7%)experi-enced‘red man’syndrome and a rise in plasma his-tamine when given vancomycin,15mg/kg,over1h,but with teicoplanin,15mg/kg,given over30min,no cases occurred and there was no change in histamine concen-trations[65].5.Rash and feverTreatment with vancomycin or teicoplanin results in a similar incidence of rash and fever but not necessarily in the same patients.Access to both agents in the hospital pharmacy can be useful to treat patients with adverse reactions to one or other glycopeptide. Maculopapular or erythematous rashes have been reported in2–8%of patients given vancomycin[31,33], and were seen in three of100vancomycin treatment courses studied retrospectively[6].Even oral adminis-tration of vancomycin(for colitis)has been implicated [69].Phlebitis was noted in all of20patients given vancomycin through a peripheral cannula in one series [33],but in only13of100courses in another[6]. Exfoliative dermatitis has been described in six patients having continuous ambulatory peritoneal dialysis after vancomycin administered by the i.p.route[6,70].Im-munological studies have defined a vancomycin-associ-ated linear IgA dermatosis.This has been reported in seven patients who had an acute blistering dermatosis, and linear deposits of IgA demonstrable at the der-moepidermal basement membrane zone[71–74]. Hypersensitivity reactions are the most common rea-son for discontinuation of teicoplanin therapy,and were reported in225of1545patients(14.6%)in US trials.Of these,7%had rash and6%fever(Hoechst Marion Roussel,data onfile).Urticaria,fever or rash usually developed in thefirst2weeks of treatment. However,the incidence of hypersensitivity appeared similar to that following vancomycin therapy in com-parative trials(49/406(12.1%)for teicoplanin therapy vs48/417(11.5%)for vancomycin therapy).In febrile neutropenic patients,a rash was significantly more common in those given vancomycin than those receiv-ing teicoplanin(15/252(6%)for vancomycin therapy vs four of215(2%)for teicoplanin therapy,P B0.05)[16]. In open,clinical trials of teicoplanin,2–3%of patients developed a rash at usual clinical doses[67,75]. Patients treated with high doses of teicoplanin for prolonged periods are most likely to experience adverse events.Fever and rash developed infive of18patients (28%)given teicoplanin,]12mg/kg,for osteomyelitis [38].Of98patients treated for osteomyelitis or septic arthritis with teicoplanin,6–12mg/kg per day,13 patients(13%)developed a rash,13(13%)fever,12(12%)nausea and seven(7%)diarrhoea,but most events were mild and did not require treatment to be stopped[76].In an unpublished trial(102-012,Hoechst Marion Roussel,data onfile),38of342patients (11.1%)developed fever and42(12.3%)a rash after receiving teicoplanin,\10mg/kg.Fever may be dose-related,being seen in six of272patients(2%)given teicoplanin,3mg/kg per day,33of779(4.2%)given teicoplanin,6mg/kg,and29of274(10.6%)given teicoplanin,12mg/kg,in comparative US trials (Hoechst Marion Roussel,data onfile).Allergic cross-reactions can occur between van-comycin and teicoplanin.Grek et al.[77]and McElrath et al.[78]reported single patients who developed rash when teicoplanin was administered,having previously developed rash and fever during treatment with van-comycin.Davenport described a patient who developed a rash during teicoplanin treatment of CAPD peritoni-tis,having suffered generalised urticaria with previous vancomycin therapy[79].Knusden et al.[80]reported an immunological study of a patient with erythema who had become allergic to vancomycin,and who experienced a cross-reaction to teicoplanin that was IgE mediated[11].Nevertheless,Smith et ed tei-coplanin successfully in9haematological patients who had developed’red man’syndrome or rash during vancomycin treatment[81].Schlemmer et al.[82]used teicoplanin without adverse events in a patient who previously had developed neutropenia and rash with vancomycin.A patient who developed a rash with teicoplanin therapy was effectively treated with van-comycin in a study by de Vries et al.[83].6.Haematological effectsA number of anecdotal reports of reversible van-comycin-induced neutropenia and thrombocytopenia have been published[84–86].In100courses of treat-ment in98patients,two patients developed neutropenia [6].Morris et al.[87]noted a high incidence of leukope-nia associated with vancomycin therapy in a cardiotho-racic unit which affected18%of patients(nine of49 patients).The use of vancomycin together with AZT for the treatment of staphylococcal infection in AIDS patients has also been associated with neutropenia[88]. The white blood cell count recovers quickly once van-comycin is stopped.Eosinophilia in the cerebrospinal fluid has been associated with intraventricular adminis-tration of vancomycin for shunt infections,and might be mistaken for treatment failure[89].Vancomycin-de-pendent antiplatelet antibodies have been demonstrated in two patients with leukaemia.These patients devel-oped a thrombocytopenia which resolved when van-comycin was stopped,but recurred in one patient when therapy was restarted[90].A.P.R.Wilson/International Journal of Antimicrobial Agents10(1998)143–152149Teicoplanin causes spontaneous platelet aggregationin vitro and inhibits aggregation caused by other fac-tors,but only at concentrations that far exceed thoseachievable clinically(5000–10000mg/l)[91].In com-parative US trials,however,there was a significantlyhigher incidence of thrombocytopenia with teicoplanin, \12mg/kg per day,than with vancomycin,15mg/kg 12hourly.Most cases were seen in an unpublished trial(102-019,Hoechst Marion Roussel,data onfile[3])inwhich teicoplanin.30mg/kg per day,was used to treatendocarditis.When analysed with two other unpub-lished US trials in which doses of6mg/kg per day wereused,the difference in the incidence of thrombocytope-nia between teicoplanin and vancomycin,15mg/kg12hourly,was no longer significant(21/217(9.7%)vs14/221(6%),Hoechst Marion Roussel,data onfile).Aclinically significant fall in platelet count occurred ineight of58patients(14%)with trough teicoplaninconcentrations\60mg/1,compared with12of251patients(5%)with lower trough levels(P B0.05, 2test)The association was lost when only patients with abaseline platelet level of\150000/mm3were included[5].However,Novakova et al.[92]showed that theneed for platelet transfusions when using a ceftazidime-teicoplanin combination for the treatment of febrileneutropenia was similar to that needed for patientsrandomised to receive ceftazidime alone.Neutropeniaand eosinophilia have been reported infrequently[93,94].Del Favero et al.[95]reported neutropeniaoccurring after20days therapy for endocarditis,whichresolved when the drug was stopped but recurred onrechallenge.These authors reviewed the records of an-other1500patients treated in European trials andfoundfive cases of neutropenia,four of which wereonly’possibly’attributable to teicoplanin.Although theeffect of dosage has not been formally studied,Livor-nese et al.[96]reported neutropenia in one10patients(10%)receiving teicoplanin,30mg/kg per day,as op-posed to none of seven receiving teicoplanin,12mg/kgper day.7.Effects on the liver and the gastrointestinal tract Neither teicoplanin nor vancomycin cause severe liver toxicity.Menichetti et al.[16]monitored liver function in a comparative trial of both agents in527 patients and reported no abnormalities,though in North American trials there was an incidence of1.3% in patients receiving teicoplanin(three of238)and vancomycin(three of239)[5].However,liver dysfunc-tion is not normally regarded as an adverse event of vancomycin therapy and the abnormality may be re-lated more to the serious infections being treated.The glycopeptides are seldom associated with gastrointesti-nal adverse events.8.ConclusionsWhen treating infections in which teicoplanin and vancomycin have similar efficacy,teicoplanin has the advantage because‘red man’syndrome can be avoided, and serum monitoring of teicoplanin is not necessary when used with an aminoglycoside.Teicoplanin can cause reversible thrombocytopenia,but only at doses above those in common use,and it has not been associated with clinical diathesis.Clinicians have greater clinical experience with van-comycin,and nephrotoxicity is uncommon when this agent is used alone.Routine monitoring could be avoided,except when aminoglycosides are given con-currently,and even then,doses could be adjusted on the basis of aminoglycoside concentrations or of age, weight and renal function[10].Some unitsfind a low incidence of‘red man’syndrome,but this is often the result of a failure to report mild reactions to the clinician.Teicoplanin is associated with a lower incidence of ‘red man’syndrome and hypotension during surgery than vancomycin.It can be used in patients having rash or fever with vancomycin without there being necessar-ily any cross-reaction.Nephrotoxicity appears less com-mon with teicoplanin than vancomycin although arguments remain as to the proportion due to the glycopeptide or a concomitant aminoglycoside.Tei-coplanin has an advantage over vancomycin in its ease of administration and choice of routes.Garrelts et al.[97]summarised the cost of vancomycin therapy in100 patients and noted that,out of a total of US$30000, US$1700was spent on assays and US$4700on the treatment of adverse events.However,vancomycin is now available in generic form at a lower cost than teicoplanin.Although both should be available,the choice of which glycopeptide to use asfirst line has to be made locally on whether savings in administration time and the treatment of adverse events are likely to outweigh the difference in acquisition cost. References[1]Conley NS,Weiner RS,Hiemenz JW.Rigors with vancomycin.Ann Intern Med1991;115:330.[2]Davey PG,Williams AH.A review of the safety profile ofteicoplanin.J Antimicrob Chemother1991;27(Suppl B):69–73.[3]Wilson APR,Gru¨neberg RN.Teicoplanin:The First Decade.Abingdon:The Medicine Group(Education),1997;pp.126–128,139,143.[4]Wood MJ.The comparative efficacy and safety of teicoplaninand vancomycin.J Antimicrob Chemother1996;37:209–22. [5]Wilson APR,Gru¨neberg RN,Neu H.A critical review ofteicoplanin in Europe and the USA.Int J Antimicrob Agents 1994;4(Suppl1):S1–S30.[6]Farber BF,Moellering RC.Retrospective study of the toxicity ofpreparations of vancomycin from1974to1981.Antimicrob Agents Chemother1983;23:138–41.。