- Email: [email protected]
Clindamycin therapy of staphylococcus aureus endocarditis
CASE REPORTS
Clindamycin Therapy of Staphylococcus
aureus
Endocardi t is Clinical Relapse and Development of Resistance to Ciindamycin, Lincomycin and Erythromycin
CHATRCHAI
WATANAKUNAKORN,
M.D
Cincinnati, Ohio
From the Infectious Disease Division, Department of Internal Medicine, University of Cincinnati College of Medicine, Cincinnati, Ohio 45267. This study was supported by the Morton Hamburger Memorial Fund. Requests for reprints should be addressed to Dr. C. Watanakunakorn. Infectious Disease Division, 7th Floor, U.C. College of Medicine, Cincinnati, Ohio 45267. Manuscript accepted July 11, 1975.
A 42 year old heroin addict with Staphylococcus aureus endocarditis of the mitral valve was treated with clindamycin phosphate, 600 mg intramuscularly, every 6 hours. The intttal ctfnical response was excellent and blood cultures became negative. On the 26th day of clindamycin therapy, fever developed and six blood cuttures taken during a 72 hour period grew Staph. aureus. The patient was subsequently cured with a six week course of nafcillin plus gentamicin fdlowed by cloxaclllin. The Staph. aureus isolated before clindamycin therapy and during relapse phagebtyped 29/52/52A/79/60 and was reststant to penicillin G. The susceptibility of both Staph. aureus isolates to 19 antibiotics was unchanged. However, the Staph. aureus developed marked resistance to clindamycin, lincomycin and erythromycin, to which the original isolate was susceptible. The resistance to clindamycin and lincomycin was heterogeneous whereas the entire cell population became homogeneousfy highly resistant to erythromycin. These antibiotics were not inactivated in vitro by the rapidly growing resistant Staph. aureus. The most likely site of resistance was at the 50 S subunit of the bacterial ribosome.. Staphylococcus aureus endocarditis, a life-threatening but potentially curable infectious disease, usually requires prolonged therapy with massive intravenous doses of antibiotics [ 11. It is also the most common type of endocarditis in narcotic addicts [2]. Almost all “mainliners” abuse every available vein resulting in thrombosis and scars. Prolonged intravenous therapy in these patients poses a significant practical clinical problem. Clindamycin is effective against the majority of Staph. aureus in vitro [3] and high serum antibiotic concentration is achievable after intravenous or intramuscular administration of clindamycin phosphate [4]. Clindamycin phosphate has been used parenterally to treat Staph. aureus endocarditis in a limited number of patients [5-g].
March 1976
The American Journal of Mediclne
Volume 60
419
RELAPSE AND DEVELOPMENT OF STAPH. AUREUS RESISTANT ENDOCARDITIS-WATANAKUNAKORN
NOV. 27:74 1o:c
t TRANSFER
TO CGH
DISCHARGE 1
101 100 99 -
I
98
BLOOD CULTURE ESR mm/hr WBC x 103imm3 HEMOGLOBIN gm CEPHALEXIN
2
3
$+
- _ 70
40
97
4
3.2 9.1
5.7 83
14.1 13.2 IO.9 10.5 IQ2
5
,
6
* 60
56
7.1
7.3
7 8 WEEKS
10.4 IO.8
--
--
9
IO
108 120 92 7.8
II
I2
80
64
vi-
-16
7.3
5.3
8.7
5.5 5.5
6.7
9.5
9.0 9.3 8.5
8.0 8.2
9.9
H
CLINDAMYCIN NAFCI LLI N GENTAMICIN CLOXACILLIN
-
PROBENICID
Figure 1.
The patient’s clinical course.
This report is of a heroin addict who was treated with intramuscular clindamycin phosphate for Staph. aureus endocarditis. The initial clinical response was excellent. However, during the fourth week of clindamycin therapy, he had a clinical and bacteriologic relapse, and was subsequently cured with other antibiotics. Of particular interest was the development of resistance of the Staph. aureus to clindamycin, lincomycin and erythromycin. The degree and nature of this microorganism’s antimicrobial resistance were studied in detail. CASE REPORT A 42 year old man who had been a heroin addict for 25 years was admitted to the hospital on November 27, 1974, because of fever, chills and weakness. For approximately six months before admission, having run out of veins, he had been injecting heroin through a superficial vein underneath an ulcer on the left arm. Three weeks prior to admission, he had fever and chills and was treated with an unknown antibiotic for “urinary tract infection.” He complained of weakness and was feverish most of the time. Two days before admission, he again had chills and felt ill. He denied any urinary symptoms. There was a past history of treated syphilis. Examination revealed an acutely ill man. The temperature was 103.6’F. (39.7’C). blood pressure 120170 mm Hg, pulse rate iOO/min and respiration 24/min. Pertinent findings included a large 10 by 10 cm superficial ulcer on the left arm with some crust formation. There were numerous old needle marks and thrombosed veins. The fundi were unremarkable. The neck was supple with no lymphadenopathy. The chest was clear. There was sinus tachycardia, and no murmur was heard. The liver and spleen were not palpable.
420
March 1978
The American Journal of Medicine
Pertinent laboratory data included a hemoglobin level of 14.1 g/100 ml, a hematocrit value of 42 per cent, a white blood cell count of 9700/mm3 with 64 per cent neutrophils, 3 per cent band forms, 11 per cent lyrhphocytes and 2 per cent monocytes. The erythrocyte sedimentation rate was 40 mm/hour. The latex fixation test for rheumatoid factor was negative. The serologic test for syphilis (VDRL) was weakly reactive and the fluorescent treponemal antibody absorption test was reactive. Urinalysis revealed a pH of 5, no glucose or protein, and 20 to 30 white blood cells/hpf. The blood urea nitrogen level was 16 mg/lOO ml. The electrocardiogram showed sinus tachycardia and no other abnormalities. The chest roentgenogram was within normal limits. The test for hepatitis B associated antigen was negative. Culture from the skin ulcer grew a penicillin-resistant Staph. aureus. The urine culture had no growth. The patient was given cephalexin, 500 mg orally, every 6 hours. His temperature continued to be elevated daily up to 104’F (Figure 1). On the seventh hospital day, a grade 2/6 pansystolic apical murmur was heard. The laboratory also reported that a penicillin-resistant Staph. aureus had grown from four blood cultures taken within 48 hours after admission. The following day, petechial hemorrhages were seen on the palpebral conjuctiva of both eyes. Cephalexin therapy was discontinued and clindamycin, 600 mg intramuscularly every 6 hours, was given. His temperature gradually decreased during the following week. He was transferred to Cincinnati General Hospital on December 13, 1974, for further therapy of Staph. aureus endocarditis. On admission, his temperature was 99.4OF (37.4’C). The skin ulcer was granulating. The apical grade 2/6 pansystolic murmur radiated to the axilla. The spleen was not palpable. The following were pertinent laboratory findings: hemoglobin level 10.5 g/100 ml, hematocrit value 30.8 per
Volume 60
RELAPSE AND DEVELOPMENT OF STAPH. AUREUS RESISTANT ENDOCARDITIS-WATANAKUNAKORN
cent, erythrocyte sedimentation rate 70 mm/hour, blood urea nitrogen 13 mg/lOO ml and serum creatinine 1.1 mg/lOO ml. Urinalysis revealed trace protein and 8 to IO white blood cells/hpf. The electrocardiogram was normal. The latex fixation test for rheumatoid factor was nonreactive. The blood and urine cultures had no growth. Intramuscular clindamycin therapy was continued. The patient tolerated treatment well with no elevation in the liver enzyme levels or symptoms of gastrointestinal adverse reactions. He was also maintained on methadone. On the 26th day of clindamycin therapy, he had a temperature of 102“F (38.8’C) and remained febrile during the next three days (Figure 1). There was no new physical finding except tachycardia and fever. Six blood cultures taken during a three-day period subsequently grew Staph. aureus. On January 1, 1975, the 29th day of therapy, clindamycin therapy was discontinued. Intravenous nafcillin, 1.5 g every 4 hours, and gentamicin, 85 mg every 8 hours, were given via a subclavian catheter. He promptly became afebrile. After nafcillin and gentamicin therapy for 10 days, he again became febrile. There was no new physical finding except that the murmur became grade 416. The subclavian catheter was changed. Culture of the tip of the catheter and blood cultures had no growth. He gradually became afebrile. The administration of gentamitin was discontinued after 18 days and nafcillin after 19 days. Cloxacillin, 1 g orally every 4 hours, was given. Probenecid, 0.5 g orally twice daily, was also given during the first week of cloxacillin therapy. The patient tolerated oral cloxacillin therapy well, and blood cultures remained sterile. The administration of cloxacillin was discontinued when the total period of therapy with nafcillin and cloxacillin reached six weeks. The blood cultures remained sterile after cloxacillin therapy was discontinued. The patient was discharged on February 17, 1975. Three weeks after discharge, he was readmitted for congestive heart failure. He was afebrile. The erythrocyte sedimentation rate was 15 mm/hour. Three blood cultures had no growth. Four months after antibiotic therapy was discontinued, he had no evidence of relapse and was doing quite well on digitalis and diuretic therapy.
TABLE I
Antimicrobial Susceptibility of Both Isolates of Staphylococcus aureus Minimal Inhibitory Concentration (pg/ml) __~.
-~
Staphylococcus aureus No. 1
Antibiotic
Penicillin Methicillin Nafcillin Oxacillin Cloxacillin Dicloxacillin Cephalothin Cefazolin Cephapirin Cephalexin Cefatrizine Vancomycin Streptomycin Kanamycin Gentamicin Sisomicin Tobramycin Tetracycline Chloramphenicol Erythromycin Lincomycin Clindamycin
>25
.25
1.56 0.2
1.56 0.2
0.39 0.05 0.1 0.05 0.39 0.2 6.25 1.56 0.78 6.25 3.12 0.39 0.2 0.39 0.78 12.5 0.2 0.39 0.1
0.1 0.05 0.05 0.05 0.39 0.2 1.56 1.56 0.78 3.12 1.56 0.39 0.2 0.1 0.39 6.25 >25 >25 3.12
NOTE: No. 1 = isolated before clindamycin lated during relapse.
coagulase-positive,
Staphylococcus aureus No. 2 _
mannitol-positive
therapy.
No. 2 = iso-
and produced
penicillinase.
Antimicrobial Susceptibility. concentration
The minimal inhibitory (MIC) of 22 antibiotics for both isolates
of Staph. aureus was determined simultaneously by the broth dilution method recommended by the International Collaborative Study on antimicrobial sensitivity testing sponsored by the World Health Organization [lo]. Mueller Hinton broth was used. The inoculum was a 10 -3 dilution of an overnight culture. The results are listed in Table I.
LABORATORY The Staph.
STUDIES
aureus
cultured
from the blood before
clindamycin therapy (No. 1) and during relapse (No. 2) phage typed 29/52/52Al79/80. They were both
TABLE II
Lincomycin Erythromycin
NOTE:
the same or within two dilutions. However, the MIC of
lnoculum Effect on the Minimal Inhibitory Concentration (MIC) and Minimal Bactericidal Concentration (MBC) Staphylococcus aureus No. 1
Antibiotic (fig/ml) Clindamycin
The MIC of the beta-lactam antibiotics, vancomytin, the aminoglycosides, tetracycline and chloramphenicol for both isolates of Staph. aureus was either
MIC MBC MIC MBC MIC MBC
No. 1 = isolated before clindamycin
lo-”
lo-’
1o-4
0.1 3.12 1.56 50 0.39 12.5
0.05 1.56 0.39 50 0.1 3.12
0.05 1 .56 0.39 12.5 0.1 1 .56
therapy.
Staphylococcus aureus No. 2 10-z 6.25 50 >200 >200 >200 >200
1o-”
lo-” 1.56 25 50 100 >200 >200
0.2 12.5 3.12 100 >200 >200
No. 2 = isolated during relapse.
March 1976
The American Journal of Medicine
Volume 60
421
RELAPSE AND DEVELOPMENT OF STAPH. AUREUS RESISTANT ENDOCARDITIS-WATANAKUNAKORN
TABLE
Serum Antistaphylococcal
III
Titers
Staphylococcus Staphylococcus aureus No. 2 aureus No. 1
-~
Date
~
Antimicrobial Therapy
Inhibi- Battericidal tory
Inhibitory
Bactericidal
<1:4
Dee 16
Clindamycin
I:128
Jan8
Nafcillin and gentamicin Cloxacillin with probenecid Cloxacillin
1:lS
I:8
I:16
1 :8
I:64
1:32
I:32
I:32
1:32
1:16
I:16
I:16
Jan. 28 Feb. 3
NOTE: No. 1 = isolated before clindamycin lated during relapse.
therapy.
No. 2 = iso-
erythromycin, lincomycin and clindamycin for isolate No. 2 was markedly increased from that for isolate No. 1.
lnoculum Effect.
The inoculum
effect
on the
MIC
and minimal bactericidal concentration (MBC) of clindamycin, lincomycin and erythromycin was investigated next by the same broth dilution method. The MBC was defined as the lowest dilution which gave
CONTROL: S. ayreus No.1
CONTROL: S. oyeus No.2
less than five colonies on subculture of 0.01 ml from the broth at 18 hours of incubation. lnocula of lo-*, culture were used. 10e3 and 10V4 of an overnight The results are presented in Table II. The size of the inoculum had only a moderate effect on the MIC and MBC of these three antibiotics for isolate No. 1. However, it had a marked effect on the MIC and MBC of clindamycin and MIC of lincomytin for isolate No. 2 while its effect on the high MBC of lincomycin and both MIC and MBC of erythromycin for isolate No. 2 was not apparent since 200 pg/ml was the highest concentration included in the tests.
Serum Antistaphylococcal Titers. Serum samples which were taken from the patient during his hospital course and frozen at -20 C were thawed. The serum antistaphylococcal activity against Staph. aureus isolates No. 1 and No. 2 was determined simultaneously, using the same broth dilution method and interpretation. An inoculum of 10m3 dilution of an overnight culture was used, Table Ill shows the serum inhibitory as well as bactericidal titers while the patient was receiving different antibiotic regimens. During clindamycin therapy, the serum inhibitory titer was 1: 128 against the original isolate and only < 1:4 against the isolate recovered during relapse, although the bactericidal titers were <1:4 against both isolates. The serum inhibitory and bactericidal titers against both isolates during therapy with nafcillin plus gentamicin, cloxacillin plus probenecid and cloxacillin alone were comparable, being at least 1:8.
Killing of Staph. aureus by Clindamycin, Lincomytin and Erythromycin. The kinetics of killing of both isolates of Staph. aureus by clindamycin, comycin and erythromycin were investigated.
No.2 RYTHROMYCIN
41 6
No. I
24
Stability of Clindamycin, Lincomycin and Erythromycin. The possibility of inactivation of clindamy-
48
HOURS The growth curves of Staph. aureus No. 7 (isolated before clindamycin therapy) and NO. 2 (isolated during relapse) alone and in the presence of clindamycin, lincomycin and erythromycin 10 tig/ml.
March 1976
overnight culture of each isolate was diluted with Mueller Hinton broth to give between lo6 and lo7 colony-forming units/ml (CFWml) and incubated at 37’C with clindamycin, lincomycin or erythromycin at a final concentration of 10 pg/ml. At 0, 6, 24 and 48 hours of incubation, viable CFU were enumerated by tenfold dilutions and the pour-plate technic. Colonies were counted after incubation at 37’C for 48 hours. The results are depicted in Figure 2. All three antibiotics killed isolate No. 1 slowly, and more than 90 per cent of the original inoculum was killed by 48 hours. However, isolate No. 2 grew in the presence of all three antibiotics at 10 pg/ml.
Figure 2.
422
linAn
The American Journal of Medicine
tin, lincomycin and erythromycin by the actively growing Staph. aureus isolate No. 2 was investigated. An overnight culture of Staph. aureus No. 2 was diluted with Mueller Hinton broth to give approximately lo6 CFU/ml and incubated with clindamycin, lincomycin or erythromycin at a final concentration
Volume 60
RELAPSE AND DEVELOPMENT OF STAPH. AUREUS RESISTANT ENDOCARDITtS-WATANAKUNAKORN
TABLE 1V
Concentrations
in Supernates of Actively
of Antibiotics
Growing Staphylococcus
No, of Colony-Forming Units/ml 0 hr
4 hr
2.1 x 106
0 5.0 x lo6
Antibiotic Clindamycin alone Cllndamycln + Staphylococcus aureus
0
Llncomycin alone Llncomycin + Staphylococcus aureus
0 2.7 x 106
frythromycin atone Erythromycln + Staphylococcus aureus Staphylococcus ‘Isolated
aureus alone
1.8
x lo7
a
0
3.4 x 106
4.9
x
3.4 x 106
1.1
x lo*
106
-
Antibiotic Concentration (pg/ml) 48 hr
0 hr
4 hr
24 hr
48 hr
1.4 x 10’
5.4 x loa
9.4 8.8
8.0 7.2
6.5 6.0
3.8 3.6
0 8.9
0 14x109
9.5 9.5
8.6 8.4
7.8
6.0 5.6
a
0
1.3 x 109
1.8 x 109
9.3 9.2
7.8 7.2
62
4.8 4.5
2.5 x lo9
2.9 x lo9
0
0
0
0
___0
0
aureus No. 2”
24 hr 0
x
10’
7.8
6.7
during relapse
of 10 pg/ml at 37OC in a shaking
incubator. Mueller Hinton broth, with each antibiotic in the same concentration but without organisms, served as controls.
Viable CFU of each culture were determined by the tenfold dilutions and pour-plate technic at 0, 4, 24 and 48 hours. At each time interval, aliquots of the cultures and controls were centrifuged and the supernates filtered through a 0.45 pm filter. The supernates were ffozen at -2O’C immediately. All samples were assayed simultaneously for antibiotic activity by the cup plate method of Bennett et al. [ 111, using Sarcina lutea as the test organism. Standards and unknown samples were run in triplicates. The results are listed in Table IV. All three antibiotics deteriorated slowly at 37’C. The antibiotic activity in the actively growing culture of Staph. aureus No. 2 was comparable with that in
the control broth at all time intervals. Thus, there was no evidence of antibiotic inactivation by the actively growing Staph. aureus No. 2. Population Analyses. Analyses of the cell population of Staph. aureus No. 1 and No. 2 for the degree of resistance to clindamycin, lincomycin and erythromycin were carried out. Mannitol salt agar was incorporated with each of the three antibiotics in the final concentration of 200 pglml and in serial twofold decreasing concentrations. Series of agar plates with a different concentration of antibiotic were inoculated with 0.1 ml of an undiluted and 10-l to lo-’ dilutions of an overnight culture. Series of agar plates without antibiotic were inoculated in the same manner and served as controls. Plates were incubated at 37% and colonies were counted after 48 hours of incubation. Figure 3 shows the number of resistant
9 8 w S. oureus No. I x-x S. oureus No.2
7-
\
65432IO-, ,
LINCOMYCIN IN yg/ml
CLINDAMYCIN IN )tg/rnl Figure 3. Popuktion analyses of Staph. aureus No. 2 [isolated during relapse) given concentrations of clindamycin, lincomycin and erythromycin.
March 1976
showing
the cell population
The Amcrlcan Journal of Medicine
resistant
Volume 60
to
423
RELAPSE AND DEVELOPMENT OF STAPH. AUREUS RESISTANT ENDOCARDITIS-WATANAKUNAKORN
TABLE V
Clindamycin Therapy of Staphylococcus aureus Endocarditis No. of Patients
Reference Romig et al. 151 Menda, Gorbach
[61
Oosaoe
2
450 mg q 8 hr
IM
4-6
800 mg q 8 hr
IM or IV
6wk
et al. [71
1
N S.
8
600-900
Elmore et al. 191
1
Present case
1
900 mg q.6 hr 600 mg q 8 hr 600mgq6hr,
NOTE:
mg q 6 hr
Results
wk
Cured
wk
10 cured, 2 dead, 1 clinical relapse, 1 allergic reaction Cured
IV
4-6
IM or IV
N.S.
IV
32 days
IM
4wk
6 cured, 1 relapse, 1 returned with osteomyelitis Cured, hepatotoxicity Clinical and bacteriologic relapse during therapy
N.S. = not stated.
organisms in isolates No. 1 and No. 2 at .indicated concentrations of clindamycin, lincomycin and erythromycin. Although the majority of the cell population of Staph. aureus No. 1 was highly susceptible to all three antibiotics, there existed a small number of organisms relatively resistant to these antibiotics. There was a marked increase in resistance of the entire cell population of Staph. aureus No. 2 to all three antibiotics. However, the resistance to clindamycin and lincomycin was heterogeneous whereas the entire cell population was homogeneously highly resistant to erythromycin. COMMENT
Since the majority of Staph. aureus is susceptible in vitro to clindamycin, parenteral clindamycin phosphate has been used to treat a limited number of patients with Staph. aureus endocarditis. Of the 27 patients with Staph. aureus endocarditis treated with parenteral clindamycin, 20 (74 per cent) were cured (Table V). Tee administration of clindamycin was discontinued in one patient due to a hypersensitivity reaction. There were two deaths, three relapses and in one patient Staph. aureus vertebral osteomyelitis developed after therapy was completed. All three relapses occurred during therapy but detailed information was available only in the present case. Clindamycin kills Staph. aureus very slowly in vitro resulting in large numbers of surviving organisms at 24 and 48 hours (Figure 2). Since rapidly bactericidal antibiotics are usually required in the therapy of Staph. aureus endocarditis, it is not surprising that the rate of relapse (11 per cent) when clindamycin is used is high. Relapse is very unusual when penicillin G or a semisynthetic penicillin is used. Relapse also occurred in Staph. aureus endocarditis induced experimentally in rabbits and treated with clindamycin whereas it did not occur in rabbits treated with betalactam antibiotics [ 12). In the present case, both clinical and bacteriologic
424
Duration
14
Bente, Jackson [81
Hinthorn
Route
March 1976
The American Journal of Medicine
relapse occurred during the fourth week of clindamytin therapy. The patient’s clinical course during the preceding week was uneventful, and the relapse was unexpected. However, all six blood cultures, taken during a 72 hour period, grew Staph. aureus which had the same bacteriophage pattern as the original infecting organism. The patient was then cured with a course of nafcillin and gentamicin followed by oral cloxacillin. The combination of nafcillin and gentamitin was chosen because of its in vitro enhancement of staphylococcidal activity [3]. The serum inhibitory and bactericidal titers during clindamycin therapy were 1:128 and <1:4, respectively, against the original infecting organism but were <1:4 (both inhibitory and bactericidal) against the Staph. aureus recovered during relapse. There was a large discrepancy between the inhibitory and bactericidal titers against the clindamycin susceptible Staph. aureus, probably due to the slow bactericidal action of clindamycin. Such a large discrepancy was not found during therapy with the penicillins (Table Ill). When using a slow bactericidal antibiotic such as clindamycin, the bactericidal titer is probably more important than the inhibitory titer. When a penicillin is used, the ‘serum inhibitory and bactericidal titers usually do not differ significantly. In such a case the serum inhibitory titer is also useful. The susceptibility of Staph. aureus isolated during relapse was unchanged to 19 unrelated antibiotics. However, the Staph. aureus became resistant to clindamycin, lincomycin and erythromycin (Table I). The increase in resistance to these three antibiotics was confirmed by several laboratory studies. The MIC and MBC of these three antibiotics were markedly increased and the inoculum effect more pronounced (Table II). The Staph. aureus isolated during relapse was more resistant’to killing by these three antibiotics (Figure 2), and this resistance was not due to the inactivation of antibiotics by the actively growing Staph. aureus (Table IV). Population analyses also confirmed the increase in resistance of the
Volume 60
RELAPSE AND DEVELOPMENTOF STAPH. AUREUS RESISTANT ENDOCARDITIS-WATANAKUNAKORN
Staph. aureus and further showed that its resistance to erythromycin was homogeneous whereas there was heterogeneous resistance to clindamycin and lincomycin (Figure 3). In vitro transfer of Staph. aureus in the presence of subinhibitory concentrations of clindamycin has resulted in resistance to clindamycin and cross resistance to lincomycin and erythromycin [ 13,141. The in vivo development of cross resistance of Staph. aureus to erythromycin has been reported in patients treated with lincomycin [15]. Similar in vivo development of cross resistance of Staph. aureus to lincomycin and erythromycin during clindamycin therapy has not been reported previously. Clindamycin, 7 (S)-chloro-7-deoxylincomycin, is a lincomycin derivative [4]. Both lincomycin and erythromycin inhibit protein synthesis at the 50 S subnit of the bacterial ribosome. In a comparative study of lincomycin and erythromycin, it was concluded that lincomycin inhibits the peptidyl transferase of the 50 S subunit whereas erythromycin inhibits the translocation from the A site to P site, both resulting in the inhibition of the synthesis of polyphenylalanine in the
poly U system [ 161. Because of the similarity of the mode of action of clindamycin, lincomycin and erythromycin and the lack of cross resistance of the Staph. aureus to antibiotics with different modes of action, it is postulated that the resistance in the present case was probably due to ribosomal resistance at the 50 S subunit of the bacterial ribosome. The slow bactericidal action of clindamycin, the high relapse rate and the development of bacterial resistance all make clindamycin unsuitable for the treatment of Staph. aureus endocarditis, although there have been reported cures with such therapy. ACKNOWLEDGMENT I thank Albert Sapadin, M.D. for his referral of this patient. Cheryl Glotzbecker and Thomas Bannister provided excellent technical assistance. ADDENDUM After this manuscript had been submitted for publication, a case of clinical and bacteriologic relapse of Staph. aureus endocarditis following clindamycin therapy was reported [ 171.
REFERENCES 1.
2.
3. 4.
5.
6.
7.
a.
Watanakunakorn C, Tan JS, Phair JP: Some salient features of Staphylococcus aureus endocarditis. Am J Med 54: 473, 1973. Watanakunakorn C: The changing epidemiology and newer aspects of infective endocarditis. Adv Intern Med (in press). Watanakunakorn C, Glotzbecker C: Enhancement of the effects of antistaphylococcal antibiotics by aminoglycosides. Antimicrob Agents Chemother 6: 602, 1974. DeHaan RM, Metzler CM, Scheltenberg D, Vandenbosch WD: Pharmacokinetic studies of clindamycin phosphate. J Clin Pharmacol 13: 190, 1973. Romig DA, Cox F, Pohlod D, Quinn EL: Clindamycin, a new alternate antibiotic for bacterial endocarditis (abstract no. 99). Abstracts of the 1 lth Interscience Conference on Antimicrobial Agents and Chemotherapy 1971, p 50. Menda K, Gorbach S: Treatment of staphylococcal endocarditis and septicemia with clindamycin (abstract no. 101). Abstracts of the 12th Interscience Conference on Antimicrobial Agents and Chemotherapy 1972, p 52. Hinthorn DR, Goodpasture HC, Romig DA, Voth DW, Chapin M, Wagoner S, Liu C: Intravenous clindamycin therapy for bacterial endocarditis (abstract no. 219). Abstracts of the 13th Interscience Conference on Antimicrobial Agents and Chemotherapy 1973. Bente N, Jackson GG: Evaluation of clindamycin phosphate treatment of staphylococcal bacteremia and endocardi-
9. 10. 11.
12.
13.
14.
15. 16. 17.
March 1976
tis (abstract no. 77). Abstracts of the 14th Interscience Conference on Antimicrobial Agents and Chemotherapy 1974. Elmore M. Rissing JP, Rink L, Brooks C: Clindamycin-associated heoatotoxicitv. Am J Med 57: 627. 1974. Ericsson H’M, Sherris, JC: Antibiotic sensitivity testing. Acta Pathol Microbial Sect 8217 (suppl): 1, 1971. Bennett JV, Brodie JL, Benner EJ, Kirby WMM: Simplified, accurate method for antibiotic assay of clinical specimens. Appl Microbial 14: 170, 1966. Sande MA, Johnson ML: Antimicrobial therapy of experimental endocarditis caused by Staphylococcus aureus. J Infect Dis 131: 367, 1975. McGehee RF Jr, Barrett FF, Finland M: Resistance of Staphylococcus aureus to lincomycin, clindamycin and erythromyciri. Antimicrob Agents Chemother 1966, p 392. Benner EJ, Adams AP: Unusual resistance of Staphylococcus aureus to lincomycin and 7-chlorolincomycin. Antimicrob Agents Chemother 1969, p 100. Duncan IBR: Development of lincomycin resistance by staphylococci. Antimicrob Agents Chemother 1967, p 723. Hash JH: Antibiotic mechanisms. Ann Rev Pharmacol 12: 35, 1972. Tuazon CU, Sheagren JN: Relapse of staphylococcal endocarditis after clindamycin therapy. Am J Med Sci 269: 145. 1975.
The American Jaurnal of Medlclne
Volume 60
425
Clindamycin Therapy of Staphylococcus
aureus
Endocardi t is Clinical Relapse and Development of Resistance to Ciindamycin, Lincomycin and Erythromycin
CHATRCHAI
WATANAKUNAKORN,
M.D
Cincinnati, Ohio
From the Infectious Disease Division, Department of Internal Medicine, University of Cincinnati College of Medicine, Cincinnati, Ohio 45267. This study was supported by the Morton Hamburger Memorial Fund. Requests for reprints should be addressed to Dr. C. Watanakunakorn. Infectious Disease Division, 7th Floor, U.C. College of Medicine, Cincinnati, Ohio 45267. Manuscript accepted July 11, 1975.
A 42 year old heroin addict with Staphylococcus aureus endocarditis of the mitral valve was treated with clindamycin phosphate, 600 mg intramuscularly, every 6 hours. The intttal ctfnical response was excellent and blood cultures became negative. On the 26th day of clindamycin therapy, fever developed and six blood cuttures taken during a 72 hour period grew Staph. aureus. The patient was subsequently cured with a six week course of nafcillin plus gentamicin fdlowed by cloxaclllin. The Staph. aureus isolated before clindamycin therapy and during relapse phagebtyped 29/52/52A/79/60 and was reststant to penicillin G. The susceptibility of both Staph. aureus isolates to 19 antibiotics was unchanged. However, the Staph. aureus developed marked resistance to clindamycin, lincomycin and erythromycin, to which the original isolate was susceptible. The resistance to clindamycin and lincomycin was heterogeneous whereas the entire cell population became homogeneousfy highly resistant to erythromycin. These antibiotics were not inactivated in vitro by the rapidly growing resistant Staph. aureus. The most likely site of resistance was at the 50 S subunit of the bacterial ribosome.. Staphylococcus aureus endocarditis, a life-threatening but potentially curable infectious disease, usually requires prolonged therapy with massive intravenous doses of antibiotics [ 11. It is also the most common type of endocarditis in narcotic addicts [2]. Almost all “mainliners” abuse every available vein resulting in thrombosis and scars. Prolonged intravenous therapy in these patients poses a significant practical clinical problem. Clindamycin is effective against the majority of Staph. aureus in vitro [3] and high serum antibiotic concentration is achievable after intravenous or intramuscular administration of clindamycin phosphate [4]. Clindamycin phosphate has been used parenterally to treat Staph. aureus endocarditis in a limited number of patients [5-g].
March 1976
The American Journal of Mediclne
Volume 60
419
RELAPSE AND DEVELOPMENT OF STAPH. AUREUS RESISTANT ENDOCARDITIS-WATANAKUNAKORN
NOV. 27:74 1o:c
t TRANSFER
TO CGH
DISCHARGE 1
101 100 99 -
I
98
BLOOD CULTURE ESR mm/hr WBC x 103imm3 HEMOGLOBIN gm CEPHALEXIN
2
3
$+
- _ 70
40
97
4
3.2 9.1
5.7 83
14.1 13.2 IO.9 10.5 IQ2
5
,
6
* 60
56
7.1
7.3
7 8 WEEKS
10.4 IO.8
--
--
9
IO
108 120 92 7.8
II
I2
80
64
vi-
-16
7.3
5.3
8.7
5.5 5.5
6.7
9.5
9.0 9.3 8.5
8.0 8.2
9.9
H
CLINDAMYCIN NAFCI LLI N GENTAMICIN CLOXACILLIN
-
PROBENICID
Figure 1.
The patient’s clinical course.
This report is of a heroin addict who was treated with intramuscular clindamycin phosphate for Staph. aureus endocarditis. The initial clinical response was excellent. However, during the fourth week of clindamycin therapy, he had a clinical and bacteriologic relapse, and was subsequently cured with other antibiotics. Of particular interest was the development of resistance of the Staph. aureus to clindamycin, lincomycin and erythromycin. The degree and nature of this microorganism’s antimicrobial resistance were studied in detail. CASE REPORT A 42 year old man who had been a heroin addict for 25 years was admitted to the hospital on November 27, 1974, because of fever, chills and weakness. For approximately six months before admission, having run out of veins, he had been injecting heroin through a superficial vein underneath an ulcer on the left arm. Three weeks prior to admission, he had fever and chills and was treated with an unknown antibiotic for “urinary tract infection.” He complained of weakness and was feverish most of the time. Two days before admission, he again had chills and felt ill. He denied any urinary symptoms. There was a past history of treated syphilis. Examination revealed an acutely ill man. The temperature was 103.6’F. (39.7’C). blood pressure 120170 mm Hg, pulse rate iOO/min and respiration 24/min. Pertinent findings included a large 10 by 10 cm superficial ulcer on the left arm with some crust formation. There were numerous old needle marks and thrombosed veins. The fundi were unremarkable. The neck was supple with no lymphadenopathy. The chest was clear. There was sinus tachycardia, and no murmur was heard. The liver and spleen were not palpable.
420
March 1978
The American Journal of Medicine
Pertinent laboratory data included a hemoglobin level of 14.1 g/100 ml, a hematocrit value of 42 per cent, a white blood cell count of 9700/mm3 with 64 per cent neutrophils, 3 per cent band forms, 11 per cent lyrhphocytes and 2 per cent monocytes. The erythrocyte sedimentation rate was 40 mm/hour. The latex fixation test for rheumatoid factor was negative. The serologic test for syphilis (VDRL) was weakly reactive and the fluorescent treponemal antibody absorption test was reactive. Urinalysis revealed a pH of 5, no glucose or protein, and 20 to 30 white blood cells/hpf. The blood urea nitrogen level was 16 mg/lOO ml. The electrocardiogram showed sinus tachycardia and no other abnormalities. The chest roentgenogram was within normal limits. The test for hepatitis B associated antigen was negative. Culture from the skin ulcer grew a penicillin-resistant Staph. aureus. The urine culture had no growth. The patient was given cephalexin, 500 mg orally, every 6 hours. His temperature continued to be elevated daily up to 104’F (Figure 1). On the seventh hospital day, a grade 2/6 pansystolic apical murmur was heard. The laboratory also reported that a penicillin-resistant Staph. aureus had grown from four blood cultures taken within 48 hours after admission. The following day, petechial hemorrhages were seen on the palpebral conjuctiva of both eyes. Cephalexin therapy was discontinued and clindamycin, 600 mg intramuscularly every 6 hours, was given. His temperature gradually decreased during the following week. He was transferred to Cincinnati General Hospital on December 13, 1974, for further therapy of Staph. aureus endocarditis. On admission, his temperature was 99.4OF (37.4’C). The skin ulcer was granulating. The apical grade 2/6 pansystolic murmur radiated to the axilla. The spleen was not palpable. The following were pertinent laboratory findings: hemoglobin level 10.5 g/100 ml, hematocrit value 30.8 per
Volume 60
RELAPSE AND DEVELOPMENT OF STAPH. AUREUS RESISTANT ENDOCARDITIS-WATANAKUNAKORN
cent, erythrocyte sedimentation rate 70 mm/hour, blood urea nitrogen 13 mg/lOO ml and serum creatinine 1.1 mg/lOO ml. Urinalysis revealed trace protein and 8 to IO white blood cells/hpf. The electrocardiogram was normal. The latex fixation test for rheumatoid factor was nonreactive. The blood and urine cultures had no growth. Intramuscular clindamycin therapy was continued. The patient tolerated treatment well with no elevation in the liver enzyme levels or symptoms of gastrointestinal adverse reactions. He was also maintained on methadone. On the 26th day of clindamycin therapy, he had a temperature of 102“F (38.8’C) and remained febrile during the next three days (Figure 1). There was no new physical finding except tachycardia and fever. Six blood cultures taken during a three-day period subsequently grew Staph. aureus. On January 1, 1975, the 29th day of therapy, clindamycin therapy was discontinued. Intravenous nafcillin, 1.5 g every 4 hours, and gentamicin, 85 mg every 8 hours, were given via a subclavian catheter. He promptly became afebrile. After nafcillin and gentamicin therapy for 10 days, he again became febrile. There was no new physical finding except that the murmur became grade 416. The subclavian catheter was changed. Culture of the tip of the catheter and blood cultures had no growth. He gradually became afebrile. The administration of gentamitin was discontinued after 18 days and nafcillin after 19 days. Cloxacillin, 1 g orally every 4 hours, was given. Probenecid, 0.5 g orally twice daily, was also given during the first week of cloxacillin therapy. The patient tolerated oral cloxacillin therapy well, and blood cultures remained sterile. The administration of cloxacillin was discontinued when the total period of therapy with nafcillin and cloxacillin reached six weeks. The blood cultures remained sterile after cloxacillin therapy was discontinued. The patient was discharged on February 17, 1975. Three weeks after discharge, he was readmitted for congestive heart failure. He was afebrile. The erythrocyte sedimentation rate was 15 mm/hour. Three blood cultures had no growth. Four months after antibiotic therapy was discontinued, he had no evidence of relapse and was doing quite well on digitalis and diuretic therapy.
TABLE I
Antimicrobial Susceptibility of Both Isolates of Staphylococcus aureus Minimal Inhibitory Concentration (pg/ml) __~.
-~
Staphylococcus aureus No. 1
Antibiotic
Penicillin Methicillin Nafcillin Oxacillin Cloxacillin Dicloxacillin Cephalothin Cefazolin Cephapirin Cephalexin Cefatrizine Vancomycin Streptomycin Kanamycin Gentamicin Sisomicin Tobramycin Tetracycline Chloramphenicol Erythromycin Lincomycin Clindamycin
>25
.25
1.56 0.2
1.56 0.2
0.39 0.05 0.1 0.05 0.39 0.2 6.25 1.56 0.78 6.25 3.12 0.39 0.2 0.39 0.78 12.5 0.2 0.39 0.1
0.1 0.05 0.05 0.05 0.39 0.2 1.56 1.56 0.78 3.12 1.56 0.39 0.2 0.1 0.39 6.25 >25 >25 3.12
NOTE: No. 1 = isolated before clindamycin lated during relapse.
coagulase-positive,
Staphylococcus aureus No. 2 _
mannitol-positive
therapy.
No. 2 = iso-
and produced
penicillinase.
Antimicrobial Susceptibility. concentration
The minimal inhibitory (MIC) of 22 antibiotics for both isolates
of Staph. aureus was determined simultaneously by the broth dilution method recommended by the International Collaborative Study on antimicrobial sensitivity testing sponsored by the World Health Organization [lo]. Mueller Hinton broth was used. The inoculum was a 10 -3 dilution of an overnight culture. The results are listed in Table I.
LABORATORY The Staph.
STUDIES
aureus
cultured
from the blood before
clindamycin therapy (No. 1) and during relapse (No. 2) phage typed 29/52/52Al79/80. They were both
TABLE II
Lincomycin Erythromycin
NOTE:
the same or within two dilutions. However, the MIC of
lnoculum Effect on the Minimal Inhibitory Concentration (MIC) and Minimal Bactericidal Concentration (MBC) Staphylococcus aureus No. 1
Antibiotic (fig/ml) Clindamycin
The MIC of the beta-lactam antibiotics, vancomytin, the aminoglycosides, tetracycline and chloramphenicol for both isolates of Staph. aureus was either
MIC MBC MIC MBC MIC MBC
No. 1 = isolated before clindamycin
lo-”
lo-’
1o-4
0.1 3.12 1.56 50 0.39 12.5
0.05 1.56 0.39 50 0.1 3.12
0.05 1 .56 0.39 12.5 0.1 1 .56
therapy.
Staphylococcus aureus No. 2 10-z 6.25 50 >200 >200 >200 >200
1o-”
lo-” 1.56 25 50 100 >200 >200
0.2 12.5 3.12 100 >200 >200
No. 2 = isolated during relapse.
March 1976
The American Journal of Medicine
Volume 60
421
RELAPSE AND DEVELOPMENT OF STAPH. AUREUS RESISTANT ENDOCARDITIS-WATANAKUNAKORN
TABLE
Serum Antistaphylococcal
III
Titers
Staphylococcus Staphylococcus aureus No. 2 aureus No. 1
-~
Date
~
Antimicrobial Therapy
Inhibi- Battericidal tory
Inhibitory
Bactericidal
<1:4
Dee 16
Clindamycin
I:128
Jan8
Nafcillin and gentamicin Cloxacillin with probenecid Cloxacillin
1:lS
I:8
I:16
1 :8
I:64
1:32
I:32
I:32
1:32
1:16
I:16
I:16
Jan. 28 Feb. 3
NOTE: No. 1 = isolated before clindamycin lated during relapse.
therapy.
No. 2 = iso-
erythromycin, lincomycin and clindamycin for isolate No. 2 was markedly increased from that for isolate No. 1.
lnoculum Effect.
The inoculum
effect
on the
MIC
and minimal bactericidal concentration (MBC) of clindamycin, lincomycin and erythromycin was investigated next by the same broth dilution method. The MBC was defined as the lowest dilution which gave
CONTROL: S. ayreus No.1
CONTROL: S. oyeus No.2
less than five colonies on subculture of 0.01 ml from the broth at 18 hours of incubation. lnocula of lo-*, culture were used. 10e3 and 10V4 of an overnight The results are presented in Table II. The size of the inoculum had only a moderate effect on the MIC and MBC of these three antibiotics for isolate No. 1. However, it had a marked effect on the MIC and MBC of clindamycin and MIC of lincomytin for isolate No. 2 while its effect on the high MBC of lincomycin and both MIC and MBC of erythromycin for isolate No. 2 was not apparent since 200 pg/ml was the highest concentration included in the tests.
Serum Antistaphylococcal Titers. Serum samples which were taken from the patient during his hospital course and frozen at -20 C were thawed. The serum antistaphylococcal activity against Staph. aureus isolates No. 1 and No. 2 was determined simultaneously, using the same broth dilution method and interpretation. An inoculum of 10m3 dilution of an overnight culture was used, Table Ill shows the serum inhibitory as well as bactericidal titers while the patient was receiving different antibiotic regimens. During clindamycin therapy, the serum inhibitory titer was 1: 128 against the original isolate and only < 1:4 against the isolate recovered during relapse, although the bactericidal titers were <1:4 against both isolates. The serum inhibitory and bactericidal titers against both isolates during therapy with nafcillin plus gentamicin, cloxacillin plus probenecid and cloxacillin alone were comparable, being at least 1:8.
Killing of Staph. aureus by Clindamycin, Lincomytin and Erythromycin. The kinetics of killing of both isolates of Staph. aureus by clindamycin, comycin and erythromycin were investigated.
No.2 RYTHROMYCIN
41 6
No. I
24
Stability of Clindamycin, Lincomycin and Erythromycin. The possibility of inactivation of clindamy-
48
HOURS The growth curves of Staph. aureus No. 7 (isolated before clindamycin therapy) and NO. 2 (isolated during relapse) alone and in the presence of clindamycin, lincomycin and erythromycin 10 tig/ml.
March 1976
overnight culture of each isolate was diluted with Mueller Hinton broth to give between lo6 and lo7 colony-forming units/ml (CFWml) and incubated at 37’C with clindamycin, lincomycin or erythromycin at a final concentration of 10 pg/ml. At 0, 6, 24 and 48 hours of incubation, viable CFU were enumerated by tenfold dilutions and the pour-plate technic. Colonies were counted after incubation at 37’C for 48 hours. The results are depicted in Figure 2. All three antibiotics killed isolate No. 1 slowly, and more than 90 per cent of the original inoculum was killed by 48 hours. However, isolate No. 2 grew in the presence of all three antibiotics at 10 pg/ml.
Figure 2.
422
linAn
The American Journal of Medicine
tin, lincomycin and erythromycin by the actively growing Staph. aureus isolate No. 2 was investigated. An overnight culture of Staph. aureus No. 2 was diluted with Mueller Hinton broth to give approximately lo6 CFU/ml and incubated with clindamycin, lincomycin or erythromycin at a final concentration
Volume 60
RELAPSE AND DEVELOPMENT OF STAPH. AUREUS RESISTANT ENDOCARDITtS-WATANAKUNAKORN
TABLE 1V
Concentrations
in Supernates of Actively
of Antibiotics
Growing Staphylococcus
No, of Colony-Forming Units/ml 0 hr
4 hr
2.1 x 106
0 5.0 x lo6
Antibiotic Clindamycin alone Cllndamycln + Staphylococcus aureus
0
Llncomycin alone Llncomycin + Staphylococcus aureus
0 2.7 x 106
frythromycin atone Erythromycln + Staphylococcus aureus Staphylococcus ‘Isolated
aureus alone
1.8
x lo7
a
0
3.4 x 106
4.9
x
3.4 x 106
1.1
x lo*
106
-
Antibiotic Concentration (pg/ml) 48 hr
0 hr
4 hr
24 hr
48 hr
1.4 x 10’
5.4 x loa
9.4 8.8
8.0 7.2
6.5 6.0
3.8 3.6
0 8.9
0 14x109
9.5 9.5
8.6 8.4
7.8
6.0 5.6
a
0
1.3 x 109
1.8 x 109
9.3 9.2
7.8 7.2
62
4.8 4.5
2.5 x lo9
2.9 x lo9
0
0
0
0
___0
0
aureus No. 2”
24 hr 0
x
10’
7.8
6.7
during relapse
of 10 pg/ml at 37OC in a shaking
incubator. Mueller Hinton broth, with each antibiotic in the same concentration but without organisms, served as controls.
Viable CFU of each culture were determined by the tenfold dilutions and pour-plate technic at 0, 4, 24 and 48 hours. At each time interval, aliquots of the cultures and controls were centrifuged and the supernates filtered through a 0.45 pm filter. The supernates were ffozen at -2O’C immediately. All samples were assayed simultaneously for antibiotic activity by the cup plate method of Bennett et al. [ 111, using Sarcina lutea as the test organism. Standards and unknown samples were run in triplicates. The results are listed in Table IV. All three antibiotics deteriorated slowly at 37’C. The antibiotic activity in the actively growing culture of Staph. aureus No. 2 was comparable with that in
the control broth at all time intervals. Thus, there was no evidence of antibiotic inactivation by the actively growing Staph. aureus No. 2. Population Analyses. Analyses of the cell population of Staph. aureus No. 1 and No. 2 for the degree of resistance to clindamycin, lincomycin and erythromycin were carried out. Mannitol salt agar was incorporated with each of the three antibiotics in the final concentration of 200 pglml and in serial twofold decreasing concentrations. Series of agar plates with a different concentration of antibiotic were inoculated with 0.1 ml of an undiluted and 10-l to lo-’ dilutions of an overnight culture. Series of agar plates without antibiotic were inoculated in the same manner and served as controls. Plates were incubated at 37% and colonies were counted after 48 hours of incubation. Figure 3 shows the number of resistant
9 8 w S. oureus No. I x-x S. oureus No.2
7-
\
65432IO-, ,
LINCOMYCIN IN yg/ml
CLINDAMYCIN IN )tg/rnl Figure 3. Popuktion analyses of Staph. aureus No. 2 [isolated during relapse) given concentrations of clindamycin, lincomycin and erythromycin.
March 1976
showing
the cell population
The Amcrlcan Journal of Medicine
resistant
Volume 60
to
423
RELAPSE AND DEVELOPMENT OF STAPH. AUREUS RESISTANT ENDOCARDITIS-WATANAKUNAKORN
TABLE V
Clindamycin Therapy of Staphylococcus aureus Endocarditis No. of Patients
Reference Romig et al. 151 Menda, Gorbach
[61
Oosaoe
2
450 mg q 8 hr
IM
4-6
800 mg q 8 hr
IM or IV
6wk
et al. [71
1
N S.
8
600-900
Elmore et al. 191
1
Present case
1
900 mg q.6 hr 600 mg q 8 hr 600mgq6hr,
NOTE:
mg q 6 hr
Results
wk
Cured
wk
10 cured, 2 dead, 1 clinical relapse, 1 allergic reaction Cured
IV
4-6
IM or IV
N.S.
IV
32 days
IM
4wk
6 cured, 1 relapse, 1 returned with osteomyelitis Cured, hepatotoxicity Clinical and bacteriologic relapse during therapy
N.S. = not stated.
organisms in isolates No. 1 and No. 2 at .indicated concentrations of clindamycin, lincomycin and erythromycin. Although the majority of the cell population of Staph. aureus No. 1 was highly susceptible to all three antibiotics, there existed a small number of organisms relatively resistant to these antibiotics. There was a marked increase in resistance of the entire cell population of Staph. aureus No. 2 to all three antibiotics. However, the resistance to clindamycin and lincomycin was heterogeneous whereas the entire cell population was homogeneously highly resistant to erythromycin. COMMENT
Since the majority of Staph. aureus is susceptible in vitro to clindamycin, parenteral clindamycin phosphate has been used to treat a limited number of patients with Staph. aureus endocarditis. Of the 27 patients with Staph. aureus endocarditis treated with parenteral clindamycin, 20 (74 per cent) were cured (Table V). Tee administration of clindamycin was discontinued in one patient due to a hypersensitivity reaction. There were two deaths, three relapses and in one patient Staph. aureus vertebral osteomyelitis developed after therapy was completed. All three relapses occurred during therapy but detailed information was available only in the present case. Clindamycin kills Staph. aureus very slowly in vitro resulting in large numbers of surviving organisms at 24 and 48 hours (Figure 2). Since rapidly bactericidal antibiotics are usually required in the therapy of Staph. aureus endocarditis, it is not surprising that the rate of relapse (11 per cent) when clindamycin is used is high. Relapse is very unusual when penicillin G or a semisynthetic penicillin is used. Relapse also occurred in Staph. aureus endocarditis induced experimentally in rabbits and treated with clindamycin whereas it did not occur in rabbits treated with betalactam antibiotics [ 12). In the present case, both clinical and bacteriologic
424
Duration
14
Bente, Jackson [81
Hinthorn
Route
March 1976
The American Journal of Medicine
relapse occurred during the fourth week of clindamytin therapy. The patient’s clinical course during the preceding week was uneventful, and the relapse was unexpected. However, all six blood cultures, taken during a 72 hour period, grew Staph. aureus which had the same bacteriophage pattern as the original infecting organism. The patient was then cured with a course of nafcillin and gentamicin followed by oral cloxacillin. The combination of nafcillin and gentamitin was chosen because of its in vitro enhancement of staphylococcidal activity [3]. The serum inhibitory and bactericidal titers during clindamycin therapy were 1:128 and <1:4, respectively, against the original infecting organism but were <1:4 (both inhibitory and bactericidal) against the Staph. aureus recovered during relapse. There was a large discrepancy between the inhibitory and bactericidal titers against the clindamycin susceptible Staph. aureus, probably due to the slow bactericidal action of clindamycin. Such a large discrepancy was not found during therapy with the penicillins (Table Ill). When using a slow bactericidal antibiotic such as clindamycin, the bactericidal titer is probably more important than the inhibitory titer. When a penicillin is used, the ‘serum inhibitory and bactericidal titers usually do not differ significantly. In such a case the serum inhibitory titer is also useful. The susceptibility of Staph. aureus isolated during relapse was unchanged to 19 unrelated antibiotics. However, the Staph. aureus became resistant to clindamycin, lincomycin and erythromycin (Table I). The increase in resistance to these three antibiotics was confirmed by several laboratory studies. The MIC and MBC of these three antibiotics were markedly increased and the inoculum effect more pronounced (Table II). The Staph. aureus isolated during relapse was more resistant’to killing by these three antibiotics (Figure 2), and this resistance was not due to the inactivation of antibiotics by the actively growing Staph. aureus (Table IV). Population analyses also confirmed the increase in resistance of the
Volume 60
RELAPSE AND DEVELOPMENTOF STAPH. AUREUS RESISTANT ENDOCARDITIS-WATANAKUNAKORN
Staph. aureus and further showed that its resistance to erythromycin was homogeneous whereas there was heterogeneous resistance to clindamycin and lincomycin (Figure 3). In vitro transfer of Staph. aureus in the presence of subinhibitory concentrations of clindamycin has resulted in resistance to clindamycin and cross resistance to lincomycin and erythromycin [ 13,141. The in vivo development of cross resistance of Staph. aureus to erythromycin has been reported in patients treated with lincomycin [15]. Similar in vivo development of cross resistance of Staph. aureus to lincomycin and erythromycin during clindamycin therapy has not been reported previously. Clindamycin, 7 (S)-chloro-7-deoxylincomycin, is a lincomycin derivative [4]. Both lincomycin and erythromycin inhibit protein synthesis at the 50 S subnit of the bacterial ribosome. In a comparative study of lincomycin and erythromycin, it was concluded that lincomycin inhibits the peptidyl transferase of the 50 S subunit whereas erythromycin inhibits the translocation from the A site to P site, both resulting in the inhibition of the synthesis of polyphenylalanine in the
poly U system [ 161. Because of the similarity of the mode of action of clindamycin, lincomycin and erythromycin and the lack of cross resistance of the Staph. aureus to antibiotics with different modes of action, it is postulated that the resistance in the present case was probably due to ribosomal resistance at the 50 S subunit of the bacterial ribosome. The slow bactericidal action of clindamycin, the high relapse rate and the development of bacterial resistance all make clindamycin unsuitable for the treatment of Staph. aureus endocarditis, although there have been reported cures with such therapy. ACKNOWLEDGMENT I thank Albert Sapadin, M.D. for his referral of this patient. Cheryl Glotzbecker and Thomas Bannister provided excellent technical assistance. ADDENDUM After this manuscript had been submitted for publication, a case of clinical and bacteriologic relapse of Staph. aureus endocarditis following clindamycin therapy was reported [ 171.
REFERENCES 1.
2.
3. 4.
5.
6.
7.
a.
Watanakunakorn C, Tan JS, Phair JP: Some salient features of Staphylococcus aureus endocarditis. Am J Med 54: 473, 1973. Watanakunakorn C: The changing epidemiology and newer aspects of infective endocarditis. Adv Intern Med (in press). Watanakunakorn C, Glotzbecker C: Enhancement of the effects of antistaphylococcal antibiotics by aminoglycosides. Antimicrob Agents Chemother 6: 602, 1974. DeHaan RM, Metzler CM, Scheltenberg D, Vandenbosch WD: Pharmacokinetic studies of clindamycin phosphate. J Clin Pharmacol 13: 190, 1973. Romig DA, Cox F, Pohlod D, Quinn EL: Clindamycin, a new alternate antibiotic for bacterial endocarditis (abstract no. 99). Abstracts of the 1 lth Interscience Conference on Antimicrobial Agents and Chemotherapy 1971, p 50. Menda K, Gorbach S: Treatment of staphylococcal endocarditis and septicemia with clindamycin (abstract no. 101). Abstracts of the 12th Interscience Conference on Antimicrobial Agents and Chemotherapy 1972, p 52. Hinthorn DR, Goodpasture HC, Romig DA, Voth DW, Chapin M, Wagoner S, Liu C: Intravenous clindamycin therapy for bacterial endocarditis (abstract no. 219). Abstracts of the 13th Interscience Conference on Antimicrobial Agents and Chemotherapy 1973. Bente N, Jackson GG: Evaluation of clindamycin phosphate treatment of staphylococcal bacteremia and endocardi-
9. 10. 11.
12.
13.
14.
15. 16. 17.
March 1976
tis (abstract no. 77). Abstracts of the 14th Interscience Conference on Antimicrobial Agents and Chemotherapy 1974. Elmore M. Rissing JP, Rink L, Brooks C: Clindamycin-associated heoatotoxicitv. Am J Med 57: 627. 1974. Ericsson H’M, Sherris, JC: Antibiotic sensitivity testing. Acta Pathol Microbial Sect 8217 (suppl): 1, 1971. Bennett JV, Brodie JL, Benner EJ, Kirby WMM: Simplified, accurate method for antibiotic assay of clinical specimens. Appl Microbial 14: 170, 1966. Sande MA, Johnson ML: Antimicrobial therapy of experimental endocarditis caused by Staphylococcus aureus. J Infect Dis 131: 367, 1975. McGehee RF Jr, Barrett FF, Finland M: Resistance of Staphylococcus aureus to lincomycin, clindamycin and erythromyciri. Antimicrob Agents Chemother 1966, p 392. Benner EJ, Adams AP: Unusual resistance of Staphylococcus aureus to lincomycin and 7-chlorolincomycin. Antimicrob Agents Chemother 1969, p 100. Duncan IBR: Development of lincomycin resistance by staphylococci. Antimicrob Agents Chemother 1967, p 723. Hash JH: Antibiotic mechanisms. Ann Rev Pharmacol 12: 35, 1972. Tuazon CU, Sheagren JN: Relapse of staphylococcal endocarditis after clindamycin therapy. Am J Med Sci 269: 145. 1975.
The American Jaurnal of Medlclne
Volume 60
425