- Email: [email protected]
Staphylococcus epidermidis arthritis following catheter-induced bacteremia in a neutropenic patient
DIAGNMICROBIOLINFECT DIS 1985;3:119-124
119
Staphylococcus epidermidis Arthritis Following Catheter-induced Bacteremia in a Neutropenic Patient James P. Hutton, Bruce H. Hamory, Joseph T. Parisi, and Larry J. Strausbaugh
Sepsis due to methicillin-resistant Staphylococcus epidermidis occurred in a neutropenic man during management with a Hickman-Broviac catheter. Despite catheter removal and 10 days of i.v. cefazolin therapy, he developed septic arthritis 6 weeks later in a nonprosthetic hip joint. S. epidermidis was isolated from the joint and found to have plasmid and phage susceptibility patterns identical to the previous blood isolate. This case is the first to document a metastatic infection from catheter-associated S. epidermidis bacteremia. It suggests that cephalosporins may not be optimal in such infections despite in vitro sensitivity. Vancomycin appears to be the drug of choice for S. epidermidis bacteremia in the neutropenic population.
INTRODUCTION S t a p h y l o c o c c u s e p i d e r m i d i s is increasingly recognized as a pathogen in serious human disease, and prolonged intravenous catheterization is the most c o m m o n predisposing factor for this n o s o c o m i a l bacteremia (Young, 1981; Christensen et al. 1982; Sanders, 1976). N e u t r o p e n i c patients with hematologic malignancies w h o are managed with surgically i m p l a n t e d catheters appear to have particularly high risk of infections (Young, 1981). We recently investigated a patient w h o d e v e l o p e d S. e p i d e r m i d i s b a c t e r e m i a in this setting; he then d e v e l o p e d septic arthritis due to the same organism. This case e m p h a s i z e s the risk to n e u t r o p e n i c patients, illustrates the potential of bacteremia to i n d u c e s e c o n d a r y foci, suggests potential difficulties of c e p h a l o s p o r i n antibiotic therapy, and demonstrates the usefulness of p l a s m i d and phage analysis for the evaluation of this organism.
CASE REPORT
A 41-year-old m a n was diagnosed as having acute l y m p h o c y t i c l e u k e m i a and was treated with i.v. vincristine, d a u n o r u b i c i n , oral prednisone, and intrathecal methotrexate. He attained a c o m p l e t e hematologic remission.
From the Departments of Medicine (J.P.H., B.H.H., L.J.S.) and Microbiology (J.T.P.) University of Missouri-Columbia, School of Medicine, Columbia, Missouri. Current address (L.J.S.): Veteran's Administration Medical Center, Portland, Oregon. Address reprint requests to: Bruce H. Hamory, M.D., Department of Medicine, University of Missouri-Columbia, School of Medicine, One Hospital Drive, Columbia, MO 65212. Received April 23, 1984; revised and accepted June 18, 1984. © 1985 Elsevier Science Publishing Co., Inc. 52 Vanderbilt Avenue, New York, NY 10017
0732-88931851503.30
120
J.P. Hutton et al.
Six weeks later, he was readmitted for chemotherapy, and a Hickman-Broviac catheter was placed. Fifteen days later, when his granulocyte count was less than 100 cells/mm 3, he developed fever, prompting the initiation of cefazolin, carbenicillin, and tobramycin therapy. Each of four blood cultures drawn over 2 days grew Staphylococcus epidermidis. The catheter was removed, and culture of the tip yielded pure growth of S. epidermidis. One gram of i.v. cefazolin every 8 hr was continued for an additional 10 days (total of 19 days of antibiotics). Four blood cultures obtained after catheter removal and during cefazolin therapy were negative. The patient defervesced and his granulocyte count returned to normal prior to discharge. Six weeks later, the patient was readmitted, complaining of low back and right leg pain of 4 weeks' duration and continued easy fatigability. He denied fever, chills, weight loss, or other bone pain, and his leukemia remained in remission. On examination, he was afebrile and had decreased range of motion of the right hip with pain on rotation. The remainder of his examination was normal. His peripheral leukocyte count was 4,800/mm a, with 63% neutrophils and 4% bands. A plain film of the right hip demonstrated a slightly decreased joint space, but was otherwise unremarkable. On the fourth hospital day, he developed fever to 38.9°C. Cultures of blood, urine, and cerebrospinal fluid (CSF) were obtained and were subsequently negative. A technetium bone scan revealed increased uptake in the right hip, the right distal femur, and the right sacroiliac joint. Aspiration of the right hip yielded pus that showed numerous polymorphonuclear leukocytes, but no organisms on Gram stain. Surgical exploration of the hip revealed copious amounts of pus limited to the joint space. Cultures obtained at surgery and by needle aspiration yielded pure growth of S. epidermidis. Intravenous vancomycin, 500 mg every 6 hr, was begun and was continued for a total of 4 weeks. On follow-up visits over the next 5 months, he remained afebrile but continued to complain of right hip pain. Films of the hip taken 2 months after drainage showed increased narrowing of the joint space and demineralization of the femoral head; there was no evidence of ongoing infection.
METHODS
Blood and hip isolates were tested for coagulase production and identified by the API STAPH-IDENT System (Analytab Products, Plainview, NY). Antibiotic susceptibilities were tested by the method of Bauer-Kirby (Bauer et al., 1966), and a microdilution technique was used in triplicate against cefazolin (MacLowry et al., 1970). Phage typing was performed with 17 phages as described previously (Parisi et al., 1981). Plasmid profiles were determined by a modification of a method described previously (Parisi et al., 1980). Cultures were inoculated in radial streaks onto trypticase soy agar containing 0.3% yeast extract in a petri dish and incubated overnight at 37°C. Approximately 1 inch of growth from the streak was suspended in 1.0 ml of a solution of 2.5 M NaCl and 0.05 M EDTA at pH 7.5. Lysostaphin, lysozyme, and preboiled t ~ a s e 1A (Sigma Chemical Co., St. Louis, MO) were added to final concentrations of 50 p,g/ml (or 30 p~g of t ~ a s e per ml) of cell suspension and mixed. After incubation at 37°C for 30 min, 0.3 ml of a solution containing 1% Brij-58 (Sigma), 1% deoxycholate, and 0.5 M EDTA (pH 8.0) was added and mixed, and the suspension was left at room temperature for 10 min. The contents were then heated at 60°C for 30 rain, after which they were centrifuged in 1.5 ml Eppendorf tubes in a microcentrifuge (Fisher Scientific, St. Louis, MO) for 10 min at 13,000 x g to pellet the proteins. The supernate (cleared lysate) was then poured into a clean tube and two
Staph epidermidis
121
volumes of cold 95% ethanol were added. The mixture was shaken and then stored at -20°C for at least 60 rain to precipitate DNA. Of the contents, 1.4 ml were transferred to a clean tube and centrifuged at 13,000 x g for 15 min. The supernatant was poured off and the procedure repeated again. The precipitated DNA was then suspended in 50 ~1 of 5 mM Tris (pH 7.8) and 20 ~1 were transferred to a clean tube. To this was added 4 I~1 of a solution of 0.025% bromophenol blue and 50% glycerol, 4 p,1 of electrophoresis buffer, and 22 p,1 of deionized water. The contents were mixed, heated at 60°C for 3 min, and 30---40 ~,1 were added to an agarose well. The agarose, electrophoresis buffer, horizontal slab apparatus, and conditions for electrophoresis have already been described (Parisi et al., 1981). RESULTS The laboratory studies indicated that all six isolates of S. epidermidis were identical. Each was coagulase-negative and was classified as S. epidermidis by the API STAPHIDENT system. All isolates were sensitive to cephalothin, chloramphenicol, and cefamadole, but resistant to penicillin, oxacillin, erythromycin, clindamycin, and gentamicin by Bauer-Kirby disc susceptibility (1966). All were lysed by phages Ph 9, Ph 10, and U 14. All six isolates had identical plasmid profiles (Fig. 1). Minimal concentrations of cefazolin needed to inhibit growth and to kill these isolates are shown in Table 1. All isolates appeared to be sensitive to cefazolin when the inoculum tested was 5 loglo colony-forming units. However, when the inoculum was increased to 7 log10 colony-forming units, the MIC and MBC values increased considerably. Both MIC and MBC values appeared to lie outside the therapeutic range when the larger inoculum was used. No differences in susceptibilities between the blood and hip isolates were apparent, suggesting that antimicrobial resistance had not developed from the time of bacteremia to the advent of the arthritis. DISCUSSION Serious S. epidermidis infections have previousy been associated with the implantation of prosthetic materials (Schoenbaum et al., 1975; Rubin et al., 1980; Masur et al., 1980; Wilson, 1977). The use of long-term, indwelling central venous catheters
1 2 3 4
5
6
Figure 1. Plasmid profiles of S. epidermidisisolates. Numbers 1 through 4 represent blood isolates. Number 5 was obtained by needle aspiration and number 6 by open drainage of the right hip.
122
J.P. Hutton et al.
TABLE 1. MIC and MBC Value to Cefazolin at Two I n o c u l u m Sizes of S.
epidermidis Isolate 1 2 3 4 5 6
MIC (p.g/ml)a 5 loglo 0.49 0.49 1.95 1.95 0.98 0.98
MBC (p.g/ml)b 7 loglo 31.3 62.5 62.5 31.5 62.5 31.3
5 loglo 0.98 0.98 1.95 1.95 0.98 1.95
7 loglo ~125 ~125 ~125 ~125 ~125 ~125
~Minimal inhibitory concentration. bMinimal bactericidal concentration. has recently been recognized as a major risk factor for bacteremia and sepsis w i t h this organism (Sanders et al., 1976; Sitges-Serra et al., 1980). One series representing 4,710 p a t i e n t - d a y s of h y p e r a l i m e n t a t i o n via subclavian catheters identified S. epidermidis as the most c o m m o n organism causing catheter sepsis (Sanders et al., 1976). Given these characteristics of S. epidermidis infections, it is not surprising that n e u t r o p e n i c patients m a n a g e d w i t h long-term, right atrial catheters are emerging as a high-risk group. Bacteremia w i t h associated signs of sepsis is most c o m m o n l y found in those patients w i t h severe n e u t r o p e n i a (less than 100 cells/mm 3) a n d serious u n d e r l y i n g diseases, such as acute l e u k e m i a (Wade et al., 1982). Young has reported S. epidermidis as the most c o m m o n isolate from blood in bone m a r r o w t r a n s p l a n t r e c i p i e n t s , - - o u t n u m b e r i n g S. aureus 10 to 1 - - a n d he cited the i n d w e l l i n g H i c k m a n catheter as the most i m p o r t a n t risk factor (Young, 1981). Unlike other cases in the literature in w h i c h no reports of s e c o n d a r y foci of infection due to S. epidermidis b a c t e r e m i a were d o c u m e n t e d , our patient illustrates the s u p p u r a t i v e c o n s e q u e n c e s that m a y follow such infections. No s e c o n d a r y foci of infection were r e p o r t e d in 16 patients w i t h catheter sepsis due to S. epidermidis during parenteral alimentation. (Sitges-Serra et al., 1980). In the n e u t r o p e n i c population, it is a p p a r e n t that generalized sepsis and death can occur, but thus far the occurrence of discrete s e c o n d a r y foci of infection has not been m e n t i o n e d (Young, 1981; Sitges-Serra et al., 1980). With o n l y 13 cases reported, S. epidermidis also has been an u n c o m m o n cause of septic arthritis in n o n p r o s t h e t i c joints (Watkins et al., 1956; S a m i l s o n et al., 1958; W a r d et al., 1960; Baitch, 1962; Nelson, 1972), and a hematogenous source of this infection has not been p r e v i o u s l y demonstrated. We have d o c u m e n t e d that S. epidermidis bacteremia, acquired in a setting of neutropenia, is capable of infecting a p r e v i o u s l y normal, n o n p r o s t h e t i c h i p joint secondarily. The phage t y p i n g and plasm i d profiling of the S. epidermidis establish that the isolates from the bacteremia and the septic arthritis were identical. The fact that this infection occurred in a n e u t r o p e n i c patient is consistent with in vivo data from the rabbit knee, w h i c h suggests that functioning p o l y m o r p h o n u c l e a r leukocytes are a key factor in the clearance of S. epidermidis from the intraarticular space (Johnson et al., 1970). Our case further demonstrates that a methicillin-resistant strain of S. epidermidis was able to establish a metastatic infection d e s p i t e removal of the p r i m a r y b a c t e r e m i a focus and 10 a d d i t i o n a l days of parenteral cefazolin therapy. S i m i l a r failures have been reported in n e u t r o p e n i c patients with methicillin-resistant S. epidermidis infections at various sites (Wade et al., 1982). Despite disc susceptibilities suggesting sensitivity to cefazolin, the success rate w i t h that drug has been d i s a p p o i n t i n g . Recent evidence suggests that cefamandole has superior in vitro activity c o m p a r e d to ceph-
Staph epidermidis
123
alothin, suggesting that it may be more active in vivo (Frongillo et al., 1984). Cefamandole has shown a marginal advantage over cephalothin or cefazolin in other in vitro series, however, (Archer, 1978; Archer et al., 1980), and clinical comparison of these cephalosporins against methicillin-resistant S. epidermidis has not been reported, leaving the question of cefamandole efficacy unanswered. Vancomycin currently appears to provide the best therapeutic results in neutropenic patients with these infections. The occurrence of S. epidermidis bacteremia in this clinical setting may be analogous to that recently reported with methicillin-resistant S. epidermidis causing prosthetic valve endocarditis (Karchmer et al., 1983). Forty-four strains evaluated at inocula of 5 log10 and 7 log10 demonstrated an inoculum effect with the MBC to cefazolin, but not to vancomycin. Therapy with cefazolin was also associated with a higher rate of clinical failure. The in vitro bactericidal properties of vancomycin appear superior to those of cefazolin (Lowy et al., 1979; Archer, 1978). Vancomycin has therefore become the optimal antimicrobial agent for this particular lesion. The absence of polymorphonuclear lankocytes at the site of bacterial proliferation in endocarditis as well as in infections in neutropanic hosts may well underline the need for the best possible bactericidal drug. Our isolate appeared to show a significant inoculum effect with cefazolin when inhibitory and bactericidal activity was evaluated. An explanation for this effect may be that phenotypic expression of high-level methicillin resistance may be seen only in small populations of a given isolate, thereby requiring a large inoculum (7 log10) for its detection (Archer et al., 1980). S. epidermidis is becoming more frequently recognized as a cause of morbidity and mortality in the neutropenic patient. Phage typing and the new technique of plasmid profiling should provide valuable tools for studying the epidemiology of these infections. Their role in establishing the occurrence of metastatic infection is emphasized by our results. The most appropriate course of therapy for bacteremia in the neutropenic patient has not been determined. Removal of all intravenous catheters upon suspicion of S. epidermidis sepsis appears prudent, although one experience suggests that bacteremia may be treated with catheters in place in patients without local cellulitis or tunnel infection (Pizzo et al., 1984). Our experience with this case has led us to the more cautious approach of removing all lines and instituting antimicrobial therapy immediately. The outcome with vancomycin therapy for meth° icillin-resistant strains appears to be superior to that with the ~°lactam antibiotics and we hope that it will decrease the late morbidity of these infections. We prefer to continue therapy for 14 days following removal of the intravascular device, and to advise a longer duration of therapy if neutropenia is severe and prolonged. Secondary foci of infection should be anticipated. As catheter usage increases in this patient population, such foci may be recognized with increasing frequency. REFERENCES
Archer GL (1978) Antimicrobial susceptibility and selection of resistance among Staphylococcus epidermidis isolates recovered from patients with infections of indwelling foreign devices. Antimicrob Agents Chemather 14:353. Archer GL, Tenenbaum MJ (1980) Antibiotic-resistant Staphylococcus epidermidis in patients undergoing cardiac surgery. Antimicrob Agents Chemother 17:269. Baitch A (1962) Recent observations of acute suppurative arthritis. Clin Orthop 22:157. Bauer AW, Kirby WMM, Sherris JC, Turck M (1966) Antibiotic susceptibility testing by a standardized single disk method. Am J C/in Pathol 45:493. Christensen GD, Bisno AL, Parisi JT, McLaughlin B, Hester MG, Luther RW (1982) Nosocomial septicemia due to multiple antibiotic-resistant Staphylococcus epidermidis. Ann Intern Med 96:1.
124
J.P. Hutton et al.
Frongillo RF, Bianchi P, Pasticci MB, Ripa S, Pauluzzi S (1984) Cross-resistance between methicillin and cephalosporins for staphylococci: A general assumption not true for cefamandole. Antimicrob Agents Chemother 25:666. Johnson AH, Campbell WG, Callahan BC (1970) Infection of rabbit knee joints after intra-articular injection of Staphylococcus aureus; comparison with joints infected with Staphylococcus albus. Am J Pathol 60:165. Karchmer AW, Archer GL, Dismukes WE (1983) Staphylococcus epidermidis causing prosthetic valve endocarditis: Microbiologic and clinical observations as guides to therapy. Ann Intern Mud 98:447. Lowy FD, Walsh JA, Mayers MM, Klein RS, Steigbigel NH (1979) Antibiotic activity in vitro against methicillin-resistant Staphylococcus epidermidis and therapy of an experimental infection. Antimicrob Agents Chemother 16:314. MacLowry JD, Jaqua My, Selepak ST (1970) Detailed methodology and implementation of semiautomated serial dilution microtechnique for antimicrobial susceptibility testing. Appl Microbiol 20:46. Masur H, Johnson WD (1980) Prosthetic valve endocarditis. J Thorac Cardiovasc Surg 80:31. Nelson JD (1972) The bacterial etiology and antibiotic management of septic arthritis in infants and children. Pediatrics 50:437. Parisi JT, Robbins J, Lampson BC, Hecht DW (1981) Characterization of a macrolide, lincosamide, and streptogramin resistance plasmid in Staphylococcus epidermidis. J Bacteriol 148:559. Parisi JT, Hecht DW (1980) Plasmid profiles in epidemiologic studies of infections by Staphylococcus epidermidis, l Infect Dis 141:637. Pizzo PA, Commers J, Cotton D, Gress J, Hathorn J, Hiemenz J, Longo D, Marshall D, Robichaud K] (1984) Approaching the controversies in antibacterial management of cancer patients. Am J Med 76:436. Rubin J, Rogers W, Taylor HM, et al. (1980) Peritonitis during continuous ambulatory peritoneal dialysis. Ann Intern Mud 92:7. Samilson RL, Bersani FA, Watkins MB (1958) Acute suppurative arthritis in infants and children; the importance of early diagnosis and surgical drainage. Pediatrics 21:798. Sanders RA, Sheldon GF (1976) Septic complications of total parenteral nutrition; a five-year experience. Am J Surg 132:214. Schoenbaum SC, Gardner P, Shillito J (1975) Infections of cerebrospinal fluid shunts: Epidemiology, clinical manifestations, and therapy. J Infect Dis 131:543. Sitges-Serra A, Puig P, Jaurrieta E, Garau J, Alastrue A, Sitges-Creus A (1980) Catheter sepsis due to Staphylococcus epidermidis during parenteral nutrition. Surg Gynecol Obstet 151:481. Wade JC, Schimpff SC, Newman KA, et al. (1982) Staphylococcus epidermidis: An Increasing Cause of Infection in Patients with Granulocytopenia. Ann Intern Mud 97:503. Ward J, Cohen AS, Bauer W (1960) The diagnosis and therapy of acute suppurative arthritis. Arthritis Rheum 3:522. Watkins MB, Samilson RC, Winters DM (1956) Acute suppurative arthritis, l Bone Joint Surg 38-A:1313. Wilson PD (1977) Joint replacement. So Med J 79:55. Young LS (1981) Nosocomial infections in the immunocompromised adult. Am J Med 70:398.
119
Staphylococcus epidermidis Arthritis Following Catheter-induced Bacteremia in a Neutropenic Patient James P. Hutton, Bruce H. Hamory, Joseph T. Parisi, and Larry J. Strausbaugh
Sepsis due to methicillin-resistant Staphylococcus epidermidis occurred in a neutropenic man during management with a Hickman-Broviac catheter. Despite catheter removal and 10 days of i.v. cefazolin therapy, he developed septic arthritis 6 weeks later in a nonprosthetic hip joint. S. epidermidis was isolated from the joint and found to have plasmid and phage susceptibility patterns identical to the previous blood isolate. This case is the first to document a metastatic infection from catheter-associated S. epidermidis bacteremia. It suggests that cephalosporins may not be optimal in such infections despite in vitro sensitivity. Vancomycin appears to be the drug of choice for S. epidermidis bacteremia in the neutropenic population.
INTRODUCTION S t a p h y l o c o c c u s e p i d e r m i d i s is increasingly recognized as a pathogen in serious human disease, and prolonged intravenous catheterization is the most c o m m o n predisposing factor for this n o s o c o m i a l bacteremia (Young, 1981; Christensen et al. 1982; Sanders, 1976). N e u t r o p e n i c patients with hematologic malignancies w h o are managed with surgically i m p l a n t e d catheters appear to have particularly high risk of infections (Young, 1981). We recently investigated a patient w h o d e v e l o p e d S. e p i d e r m i d i s b a c t e r e m i a in this setting; he then d e v e l o p e d septic arthritis due to the same organism. This case e m p h a s i z e s the risk to n e u t r o p e n i c patients, illustrates the potential of bacteremia to i n d u c e s e c o n d a r y foci, suggests potential difficulties of c e p h a l o s p o r i n antibiotic therapy, and demonstrates the usefulness of p l a s m i d and phage analysis for the evaluation of this organism.
CASE REPORT
A 41-year-old m a n was diagnosed as having acute l y m p h o c y t i c l e u k e m i a and was treated with i.v. vincristine, d a u n o r u b i c i n , oral prednisone, and intrathecal methotrexate. He attained a c o m p l e t e hematologic remission.
From the Departments of Medicine (J.P.H., B.H.H., L.J.S.) and Microbiology (J.T.P.) University of Missouri-Columbia, School of Medicine, Columbia, Missouri. Current address (L.J.S.): Veteran's Administration Medical Center, Portland, Oregon. Address reprint requests to: Bruce H. Hamory, M.D., Department of Medicine, University of Missouri-Columbia, School of Medicine, One Hospital Drive, Columbia, MO 65212. Received April 23, 1984; revised and accepted June 18, 1984. © 1985 Elsevier Science Publishing Co., Inc. 52 Vanderbilt Avenue, New York, NY 10017
0732-88931851503.30
120
J.P. Hutton et al.
Six weeks later, he was readmitted for chemotherapy, and a Hickman-Broviac catheter was placed. Fifteen days later, when his granulocyte count was less than 100 cells/mm 3, he developed fever, prompting the initiation of cefazolin, carbenicillin, and tobramycin therapy. Each of four blood cultures drawn over 2 days grew Staphylococcus epidermidis. The catheter was removed, and culture of the tip yielded pure growth of S. epidermidis. One gram of i.v. cefazolin every 8 hr was continued for an additional 10 days (total of 19 days of antibiotics). Four blood cultures obtained after catheter removal and during cefazolin therapy were negative. The patient defervesced and his granulocyte count returned to normal prior to discharge. Six weeks later, the patient was readmitted, complaining of low back and right leg pain of 4 weeks' duration and continued easy fatigability. He denied fever, chills, weight loss, or other bone pain, and his leukemia remained in remission. On examination, he was afebrile and had decreased range of motion of the right hip with pain on rotation. The remainder of his examination was normal. His peripheral leukocyte count was 4,800/mm a, with 63% neutrophils and 4% bands. A plain film of the right hip demonstrated a slightly decreased joint space, but was otherwise unremarkable. On the fourth hospital day, he developed fever to 38.9°C. Cultures of blood, urine, and cerebrospinal fluid (CSF) were obtained and were subsequently negative. A technetium bone scan revealed increased uptake in the right hip, the right distal femur, and the right sacroiliac joint. Aspiration of the right hip yielded pus that showed numerous polymorphonuclear leukocytes, but no organisms on Gram stain. Surgical exploration of the hip revealed copious amounts of pus limited to the joint space. Cultures obtained at surgery and by needle aspiration yielded pure growth of S. epidermidis. Intravenous vancomycin, 500 mg every 6 hr, was begun and was continued for a total of 4 weeks. On follow-up visits over the next 5 months, he remained afebrile but continued to complain of right hip pain. Films of the hip taken 2 months after drainage showed increased narrowing of the joint space and demineralization of the femoral head; there was no evidence of ongoing infection.
METHODS
Blood and hip isolates were tested for coagulase production and identified by the API STAPH-IDENT System (Analytab Products, Plainview, NY). Antibiotic susceptibilities were tested by the method of Bauer-Kirby (Bauer et al., 1966), and a microdilution technique was used in triplicate against cefazolin (MacLowry et al., 1970). Phage typing was performed with 17 phages as described previously (Parisi et al., 1981). Plasmid profiles were determined by a modification of a method described previously (Parisi et al., 1980). Cultures were inoculated in radial streaks onto trypticase soy agar containing 0.3% yeast extract in a petri dish and incubated overnight at 37°C. Approximately 1 inch of growth from the streak was suspended in 1.0 ml of a solution of 2.5 M NaCl and 0.05 M EDTA at pH 7.5. Lysostaphin, lysozyme, and preboiled t ~ a s e 1A (Sigma Chemical Co., St. Louis, MO) were added to final concentrations of 50 p,g/ml (or 30 p~g of t ~ a s e per ml) of cell suspension and mixed. After incubation at 37°C for 30 min, 0.3 ml of a solution containing 1% Brij-58 (Sigma), 1% deoxycholate, and 0.5 M EDTA (pH 8.0) was added and mixed, and the suspension was left at room temperature for 10 min. The contents were then heated at 60°C for 30 rain, after which they were centrifuged in 1.5 ml Eppendorf tubes in a microcentrifuge (Fisher Scientific, St. Louis, MO) for 10 min at 13,000 x g to pellet the proteins. The supernate (cleared lysate) was then poured into a clean tube and two
Staph epidermidis
121
volumes of cold 95% ethanol were added. The mixture was shaken and then stored at -20°C for at least 60 rain to precipitate DNA. Of the contents, 1.4 ml were transferred to a clean tube and centrifuged at 13,000 x g for 15 min. The supernatant was poured off and the procedure repeated again. The precipitated DNA was then suspended in 50 ~1 of 5 mM Tris (pH 7.8) and 20 ~1 were transferred to a clean tube. To this was added 4 I~1 of a solution of 0.025% bromophenol blue and 50% glycerol, 4 p,1 of electrophoresis buffer, and 22 p,1 of deionized water. The contents were mixed, heated at 60°C for 3 min, and 30---40 ~,1 were added to an agarose well. The agarose, electrophoresis buffer, horizontal slab apparatus, and conditions for electrophoresis have already been described (Parisi et al., 1981). RESULTS The laboratory studies indicated that all six isolates of S. epidermidis were identical. Each was coagulase-negative and was classified as S. epidermidis by the API STAPHIDENT system. All isolates were sensitive to cephalothin, chloramphenicol, and cefamadole, but resistant to penicillin, oxacillin, erythromycin, clindamycin, and gentamicin by Bauer-Kirby disc susceptibility (1966). All were lysed by phages Ph 9, Ph 10, and U 14. All six isolates had identical plasmid profiles (Fig. 1). Minimal concentrations of cefazolin needed to inhibit growth and to kill these isolates are shown in Table 1. All isolates appeared to be sensitive to cefazolin when the inoculum tested was 5 loglo colony-forming units. However, when the inoculum was increased to 7 log10 colony-forming units, the MIC and MBC values increased considerably. Both MIC and MBC values appeared to lie outside the therapeutic range when the larger inoculum was used. No differences in susceptibilities between the blood and hip isolates were apparent, suggesting that antimicrobial resistance had not developed from the time of bacteremia to the advent of the arthritis. DISCUSSION Serious S. epidermidis infections have previousy been associated with the implantation of prosthetic materials (Schoenbaum et al., 1975; Rubin et al., 1980; Masur et al., 1980; Wilson, 1977). The use of long-term, indwelling central venous catheters
1 2 3 4
5
6
Figure 1. Plasmid profiles of S. epidermidisisolates. Numbers 1 through 4 represent blood isolates. Number 5 was obtained by needle aspiration and number 6 by open drainage of the right hip.
122
J.P. Hutton et al.
TABLE 1. MIC and MBC Value to Cefazolin at Two I n o c u l u m Sizes of S.
epidermidis Isolate 1 2 3 4 5 6
MIC (p.g/ml)a 5 loglo 0.49 0.49 1.95 1.95 0.98 0.98
MBC (p.g/ml)b 7 loglo 31.3 62.5 62.5 31.5 62.5 31.3
5 loglo 0.98 0.98 1.95 1.95 0.98 1.95
7 loglo ~125 ~125 ~125 ~125 ~125 ~125
~Minimal inhibitory concentration. bMinimal bactericidal concentration. has recently been recognized as a major risk factor for bacteremia and sepsis w i t h this organism (Sanders et al., 1976; Sitges-Serra et al., 1980). One series representing 4,710 p a t i e n t - d a y s of h y p e r a l i m e n t a t i o n via subclavian catheters identified S. epidermidis as the most c o m m o n organism causing catheter sepsis (Sanders et al., 1976). Given these characteristics of S. epidermidis infections, it is not surprising that n e u t r o p e n i c patients m a n a g e d w i t h long-term, right atrial catheters are emerging as a high-risk group. Bacteremia w i t h associated signs of sepsis is most c o m m o n l y found in those patients w i t h severe n e u t r o p e n i a (less than 100 cells/mm 3) a n d serious u n d e r l y i n g diseases, such as acute l e u k e m i a (Wade et al., 1982). Young has reported S. epidermidis as the most c o m m o n isolate from blood in bone m a r r o w t r a n s p l a n t r e c i p i e n t s , - - o u t n u m b e r i n g S. aureus 10 to 1 - - a n d he cited the i n d w e l l i n g H i c k m a n catheter as the most i m p o r t a n t risk factor (Young, 1981). Unlike other cases in the literature in w h i c h no reports of s e c o n d a r y foci of infection due to S. epidermidis b a c t e r e m i a were d o c u m e n t e d , our patient illustrates the s u p p u r a t i v e c o n s e q u e n c e s that m a y follow such infections. No s e c o n d a r y foci of infection were r e p o r t e d in 16 patients w i t h catheter sepsis due to S. epidermidis during parenteral alimentation. (Sitges-Serra et al., 1980). In the n e u t r o p e n i c population, it is a p p a r e n t that generalized sepsis and death can occur, but thus far the occurrence of discrete s e c o n d a r y foci of infection has not been m e n t i o n e d (Young, 1981; Sitges-Serra et al., 1980). With o n l y 13 cases reported, S. epidermidis also has been an u n c o m m o n cause of septic arthritis in n o n p r o s t h e t i c joints (Watkins et al., 1956; S a m i l s o n et al., 1958; W a r d et al., 1960; Baitch, 1962; Nelson, 1972), and a hematogenous source of this infection has not been p r e v i o u s l y demonstrated. We have d o c u m e n t e d that S. epidermidis bacteremia, acquired in a setting of neutropenia, is capable of infecting a p r e v i o u s l y normal, n o n p r o s t h e t i c h i p joint secondarily. The phage t y p i n g and plasm i d profiling of the S. epidermidis establish that the isolates from the bacteremia and the septic arthritis were identical. The fact that this infection occurred in a n e u t r o p e n i c patient is consistent with in vivo data from the rabbit knee, w h i c h suggests that functioning p o l y m o r p h o n u c l e a r leukocytes are a key factor in the clearance of S. epidermidis from the intraarticular space (Johnson et al., 1970). Our case further demonstrates that a methicillin-resistant strain of S. epidermidis was able to establish a metastatic infection d e s p i t e removal of the p r i m a r y b a c t e r e m i a focus and 10 a d d i t i o n a l days of parenteral cefazolin therapy. S i m i l a r failures have been reported in n e u t r o p e n i c patients with methicillin-resistant S. epidermidis infections at various sites (Wade et al., 1982). Despite disc susceptibilities suggesting sensitivity to cefazolin, the success rate w i t h that drug has been d i s a p p o i n t i n g . Recent evidence suggests that cefamandole has superior in vitro activity c o m p a r e d to ceph-
Staph epidermidis
123
alothin, suggesting that it may be more active in vivo (Frongillo et al., 1984). Cefamandole has shown a marginal advantage over cephalothin or cefazolin in other in vitro series, however, (Archer, 1978; Archer et al., 1980), and clinical comparison of these cephalosporins against methicillin-resistant S. epidermidis has not been reported, leaving the question of cefamandole efficacy unanswered. Vancomycin currently appears to provide the best therapeutic results in neutropenic patients with these infections. The occurrence of S. epidermidis bacteremia in this clinical setting may be analogous to that recently reported with methicillin-resistant S. epidermidis causing prosthetic valve endocarditis (Karchmer et al., 1983). Forty-four strains evaluated at inocula of 5 log10 and 7 log10 demonstrated an inoculum effect with the MBC to cefazolin, but not to vancomycin. Therapy with cefazolin was also associated with a higher rate of clinical failure. The in vitro bactericidal properties of vancomycin appear superior to those of cefazolin (Lowy et al., 1979; Archer, 1978). Vancomycin has therefore become the optimal antimicrobial agent for this particular lesion. The absence of polymorphonuclear lankocytes at the site of bacterial proliferation in endocarditis as well as in infections in neutropanic hosts may well underline the need for the best possible bactericidal drug. Our isolate appeared to show a significant inoculum effect with cefazolin when inhibitory and bactericidal activity was evaluated. An explanation for this effect may be that phenotypic expression of high-level methicillin resistance may be seen only in small populations of a given isolate, thereby requiring a large inoculum (7 log10) for its detection (Archer et al., 1980). S. epidermidis is becoming more frequently recognized as a cause of morbidity and mortality in the neutropenic patient. Phage typing and the new technique of plasmid profiling should provide valuable tools for studying the epidemiology of these infections. Their role in establishing the occurrence of metastatic infection is emphasized by our results. The most appropriate course of therapy for bacteremia in the neutropenic patient has not been determined. Removal of all intravenous catheters upon suspicion of S. epidermidis sepsis appears prudent, although one experience suggests that bacteremia may be treated with catheters in place in patients without local cellulitis or tunnel infection (Pizzo et al., 1984). Our experience with this case has led us to the more cautious approach of removing all lines and instituting antimicrobial therapy immediately. The outcome with vancomycin therapy for meth° icillin-resistant strains appears to be superior to that with the ~°lactam antibiotics and we hope that it will decrease the late morbidity of these infections. We prefer to continue therapy for 14 days following removal of the intravascular device, and to advise a longer duration of therapy if neutropenia is severe and prolonged. Secondary foci of infection should be anticipated. As catheter usage increases in this patient population, such foci may be recognized with increasing frequency. REFERENCES
Archer GL (1978) Antimicrobial susceptibility and selection of resistance among Staphylococcus epidermidis isolates recovered from patients with infections of indwelling foreign devices. Antimicrob Agents Chemather 14:353. Archer GL, Tenenbaum MJ (1980) Antibiotic-resistant Staphylococcus epidermidis in patients undergoing cardiac surgery. Antimicrob Agents Chemother 17:269. Baitch A (1962) Recent observations of acute suppurative arthritis. Clin Orthop 22:157. Bauer AW, Kirby WMM, Sherris JC, Turck M (1966) Antibiotic susceptibility testing by a standardized single disk method. Am J C/in Pathol 45:493. Christensen GD, Bisno AL, Parisi JT, McLaughlin B, Hester MG, Luther RW (1982) Nosocomial septicemia due to multiple antibiotic-resistant Staphylococcus epidermidis. Ann Intern Med 96:1.
124
J.P. Hutton et al.
Frongillo RF, Bianchi P, Pasticci MB, Ripa S, Pauluzzi S (1984) Cross-resistance between methicillin and cephalosporins for staphylococci: A general assumption not true for cefamandole. Antimicrob Agents Chemother 25:666. Johnson AH, Campbell WG, Callahan BC (1970) Infection of rabbit knee joints after intra-articular injection of Staphylococcus aureus; comparison with joints infected with Staphylococcus albus. Am J Pathol 60:165. Karchmer AW, Archer GL, Dismukes WE (1983) Staphylococcus epidermidis causing prosthetic valve endocarditis: Microbiologic and clinical observations as guides to therapy. Ann Intern Mud 98:447. Lowy FD, Walsh JA, Mayers MM, Klein RS, Steigbigel NH (1979) Antibiotic activity in vitro against methicillin-resistant Staphylococcus epidermidis and therapy of an experimental infection. Antimicrob Agents Chemother 16:314. MacLowry JD, Jaqua My, Selepak ST (1970) Detailed methodology and implementation of semiautomated serial dilution microtechnique for antimicrobial susceptibility testing. Appl Microbiol 20:46. Masur H, Johnson WD (1980) Prosthetic valve endocarditis. J Thorac Cardiovasc Surg 80:31. Nelson JD (1972) The bacterial etiology and antibiotic management of septic arthritis in infants and children. Pediatrics 50:437. Parisi JT, Robbins J, Lampson BC, Hecht DW (1981) Characterization of a macrolide, lincosamide, and streptogramin resistance plasmid in Staphylococcus epidermidis. J Bacteriol 148:559. Parisi JT, Hecht DW (1980) Plasmid profiles in epidemiologic studies of infections by Staphylococcus epidermidis, l Infect Dis 141:637. Pizzo PA, Commers J, Cotton D, Gress J, Hathorn J, Hiemenz J, Longo D, Marshall D, Robichaud K] (1984) Approaching the controversies in antibacterial management of cancer patients. Am J Med 76:436. Rubin J, Rogers W, Taylor HM, et al. (1980) Peritonitis during continuous ambulatory peritoneal dialysis. Ann Intern Mud 92:7. Samilson RL, Bersani FA, Watkins MB (1958) Acute suppurative arthritis in infants and children; the importance of early diagnosis and surgical drainage. Pediatrics 21:798. Sanders RA, Sheldon GF (1976) Septic complications of total parenteral nutrition; a five-year experience. Am J Surg 132:214. Schoenbaum SC, Gardner P, Shillito J (1975) Infections of cerebrospinal fluid shunts: Epidemiology, clinical manifestations, and therapy. J Infect Dis 131:543. Sitges-Serra A, Puig P, Jaurrieta E, Garau J, Alastrue A, Sitges-Creus A (1980) Catheter sepsis due to Staphylococcus epidermidis during parenteral nutrition. Surg Gynecol Obstet 151:481. Wade JC, Schimpff SC, Newman KA, et al. (1982) Staphylococcus epidermidis: An Increasing Cause of Infection in Patients with Granulocytopenia. Ann Intern Mud 97:503. Ward J, Cohen AS, Bauer W (1960) The diagnosis and therapy of acute suppurative arthritis. Arthritis Rheum 3:522. Watkins MB, Samilson RC, Winters DM (1956) Acute suppurative arthritis, l Bone Joint Surg 38-A:1313. Wilson PD (1977) Joint replacement. So Med J 79:55. Young LS (1981) Nosocomial infections in the immunocompromised adult. Am J Med 70:398.