Nafcillin Therapy for Staphylococcus aureus Endocarditis

Transcription

1 ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Sept. 1978, p /78/ $02.00/0 Copyright ) 1978 American Society for Microbiology Vol. 14, No. 3 Printed in U.S.A. Nafcillin Therapy for Staphylococcus aureus Endocarditis HENRY MASUR, HENRY W. MURRAY, AND RICHARD B. ROBERTS* Division of Infectious Diseases, Department ofmedicine, The New York Hospital-Cornell Medical Center, New York, New York Received for publication 22 February 1978 Semisynthetic penicillinase-resistant penicillins are recommended for therapy of Staphylococcus aureus endocarditis, but evaluation of the efficacy and safety of individual agents has received little attention. At The New York Hospital, 11 heroin addicts and 5 nonaddicts were treated with nafcillin. The 11 addicts did well clinically, but four of the five nonaddicts had severe complications, and three of them died. Important adverse reactions to nafcillin occurred in two patients: one developed leukopenia, and one developed an extensive rash. Methicillin was employed to treat two heroin addicts and four nonaddicts. Five of the six patients were cured bacteriologically, but three patients developed nephritis and one patient developed an extensive rash. Nafcillin appears to be highly efficacious for the treatment of S. aureus endocarditis, yielding results at least equal to those obtained with other drugs. Because adverse reactions appear to occur more frequently with methicillin than with nafcillin, we regard nafcillin as the preferable penicillinase-resistant penicillin for the treatment of S. aureus endocarditis. Semisynthetic penicillinase-resistant penicillins are recommended for therapy of endocarditis due to Staphlococcus aureus (14, 16, 29). Although the clinical syndrome of S. aureus endocarditis in both heroin addicts and nonaddicts has recently been detailed, the relative efficacy and safety of the penicillinase-resistant penicillins have received little attention (2, 20, 27, 28). In this report, our experience with nafcillin in the treatment of S. aureus endocarditis is detailed. MATERIALS AND METHODS Clinical and autopsy records of all patients with two or more blood cultures positive for S. aureus, or a discharge diagnosis of S. aureus endocarditis during the period of 1971 to 1977, were reviewed. A diagnosis of S. aureus endocarditis was established on the basis of one of the following criteria: isolation of S. aureus from at least two separate blood cultures in a patient with a syndrome compatible with endocarditis; isolation of S. aureus from one blood culture during life, in addition to autopsy confirmation of endocarditis; isolation of S. aureus from heart blood or vegetations on the heart, in addition to the presence of gram-positive cocci on the vegetation. S. aureus isolates were identified by the Microbiology Laboratory of The New York Hospital on the basis of morphology, coagulase reaction, and deoxyribonuclease production. Patients were included in this study if therapy with parenteral nafcillin, alone or in combination with other antimicrobial agents, had been instituted after no more than 72 h of some other antimicrobial regimen and administered for at least 48 h. To compare the frequency of adverse reactions associated with nafcillin 457 and methicillin, the clinical records of six additional patients who had received at least 48 h of methicillin for S. aureus endocarditis were reviewed. Antimicrobial susceptibility testing employed the Kirby-Bauer disk diffusion method and a standard twofold broth dilution method (3). Minimum inhibitory concentrations were defined as the concentrations of antibiotic that prevented turbidity in a tube after 24 h of incubation with a 105 inoculum of the microorganism. Minimum bactericidal concentrations were defined as the lowest concentrations of antibiotic that killed 99% of the inoculum when streaked on a blood agar plate at 24 and 48 h. Peak serum inhibitory and bactericidal levels were measured 45 to 90 min after rapid antibiotic infusion. Peak serum inhibitory levels were defined as the highest dilution of serum that prevented turbidity at 24 h. Peak serum bactericidal levels were defined as the highest dilution that killed 99% of the inoculum when streaked at 24 and 48 h. RESULTS From 1971 to 1977, 29 episodes of S. aureus endocarditis in 27 patients were observed at The New York Hospital. Table 1 presents clinical and laboratory data on 16 patients who were treated with nafcillin. Eight heroin addicts and four nonaddicts were treated with nafcillin alone; three heroin addicts and one nonaddict were treated with nafcillin in combination with or in sequence with another antibiotic. Combination therapy reflected the preference of the individual attending physician rather than severity of disease. Thirteen of the 16 S. aureus isolates were resistant to penicillin. The minimum inhibitory

2 458 MASUR, MURRAY, AND ROBERTS TABLE 1. Clinical data on 16 patients treated with nafcillin for S. aureus endocarditis Antimicrobial theory Patient no. Daily Duration Drug dose (g) Days) inhibitory Reciprocal of peak Adverse reactions Comments/outcome level la Nafcillin Well, 4 years later 2a Nafcillin ,024 Rash, week 3 Well at discharge 3a Nafcillin ,024 Well, 3 years later 4a Nafcillin Well, 4 years later 5a Nafcillin Rash, week 3 Well, 1 year later 6a Nafcillin Well at discharge 7a Nafcillin Well at discharge 8a Nafcillin 8 27 NDb Well, 2 years later 9 Nafcillin 9 42 ND Serratia sepsis Well at discharge; sudden death 6 months later 10c Nafcillin Died of a cerebral hemorrhage during therapy 11 Nafcillin Sudden death during therapy 12 Nafcillin Well, 1 year later 13a Nafcillin 8 9 ND Methicillin substituted Methicillin ND when patient transferred to another hospital 14a Nafcillin Well, 3 months later Gentamicin a Nafcillin 8 24d 128 Leukopenia Well, 6 months later Gentaniicin Nafcillin Fever, rash, Relapsed 2 weeks after Vancomycin eosinophilia therapy and died a Heroin addict. b ND, Not done. Chid. d Followed by vancomycin (1.5 g/day) for 18 days. concentrations of nafcillin against these 16 S. aureus isolates were 0.1 to 0.8 ug/ml (mean, 0.4,ug/ml). The minimum bactericidal concentrations of nafcillin against the 16 S. aureus isolates were 0.1 to 1.6 pug/ml (mean, 0.4 pg/ml). The dose of nafcillin employed was 6 to 12 g/day (median, 9 g/day), except in one child who received 4 g/day. Peak serum inhibitory levels ranged from 1:32 to 1:1,024 (median, 1:128); peak serum bactericidal levels ranged from 1:16 to 1:512 (median, 1:128). The extracardiac complications in the 11 heroin addicts included cavitating pneumonia in 4 patients and a cerebral embolus, a pulmonary embolus, a septic joint, and osteomyelitis in 1 patient each. All 11 addicts were discharged from the hospital with no clinical or bacteriological evidence of endocarditis. Six were available for follow-up study-, five had no evidence of relapse 6 months to 4 years after discharge. One patient, however, had another episode of S. aureus endocarditis 1 year after discharge, but had continued to use heroin in the interim period. ANTimICROB. AGENTS CHIZMOTHER. Four of the five nonaddicts had relatively severe complications associated with endocarditis. Patient 10 was admitted to the hospital comatose, developed chronic disseminated intravascular coagulation, and after 22 days of therapy died due to a cerebral hemorrhage despite bacteriological cure. Patient 11 died suddenly of an apparent arrhythmia after 36 days of therapy. Patient 12 had osteomyelitis but is doing well 1 year after discharge. Patient 16 developed an adverse reaction to nafcillin after 24 days of therapy, finished her 6-week course of antibiotics with vancomycin, but relapsed and died shortly thereafter. Patient 16 was the only patient in this series with a prosthetic heart valve. The fifth patient (no. 9) did well for 6 months after termination of therapy, with no evidence for recurrence, but died suddenly. Two of the 16 patients had important adverse reactions to nafcillin One patient (no. 15), who had also received a 2-week course of gentamicin, developed leukopenia (white blood count 1,500/mm3 with 100% lymphocytes) during week

3 VOL. 14, 1978 TABLE 2. NAFCILLIN THERAPY FOR S. AUREUS ENDOCARDITIS of nafcillin therapy, and has been previously reported (30). A bone marrow biopsy showed nornal erythrocyte and platelet precursors but myelocytic maturation arrest at the myelocyte stage. Neutrophils first reappeared in the peripheral blood 4 days after discontinuation of nafcillin. A second patient (no. 16) developed a temperature ranging from 38.0 to 38.7 C, a generalized pruritic maculopapular rash, and eosinophilia (450 to 1,000/ml) during week 4 of nafcillin therapy. Renal function remained stable, and the urine sediment was unremarkable. Antimicrobial therapy was changed to vancomycin with prompt resolution of the fever, rash, and eosinophilia. Two additional patients (nos. 2 and 5) developed a generalized maculopapular rash which did not become confluent. These rashes resolved without specific therapy or cessation of nafcillin. Therapy for S. aureus endocarditis was instituted with methicillin in six other patients, two of whom were heroin addicts (Table 2). Four of the six S. aureus isolates were resistant to penicillin. The minimum inhibitory concentrations of methicillin against these six S. aureus isolates were 1.6 to 6.2 ug/ml (mean, 3.1,ug/ml). The miniimum bactericidal concentrations of methicillin against these six S. aureus isolates were 1.6 to 6.2 ug/ml (mean, 3.1 ulg/ml). The doses of methicillin employed were 6 to 24 g/day (median, 12 g/day). Peak serum inhibitory levels of methicillin ranged from 1:16 to 1:512 (median, 1:64). Peak serum bactericidal levels of methicillin ranged from 1:8 to 1:512 (median, 1:64). Three patients finished 27- to 43-day courses of methicilhin without apparent complications of drug therapy. One patient died in renal failure, which was temporally related to endocarditis rather than drug therapy, and of a gastrointestinal hemorrhage. The other two patients had uncomplicated hospital courses but were lost to follow-up. Three patients who were treated with 12 to 24 g of methicillin per day developed hematuria, pyuria, proteinuria, and serum creatinine elevations (1.5 to 12 mg/ml) 14 to 28 days after initiation of therapy. Two of these patients also had fever and eosinophilia. The Clinical data on six patients treated with methicillin for S. aureus endocarditis Antimicrobial therapy Patient Reciprocal no. Patieno of peak Drug Daily Duration dose (g) (days) serum ihbtr Adverse reactions Comments/outcome level 1 Methicillin Rash, week 4 Well at discharge 2 Methicillin Well at discharge 3 Methicillin Relapse 3 weeks after therapy; died due to renal failure and gastrointestinal hemorrhage 4 Methicillin fol ND0 Interstitial nephri- Renal function returned lowed by tis, rash, eosino- to normal philia Cephalosporin Well at discharge 5b Methicillin fol Interstitial nephri- Renal function returned lowed by tis, fever, eosin- to normal ophilia Cephalosporin Well at discharge a ND, Not done. b Heroin addict. 6b Methicillin fol Interstitial nephri- Renal function returned lowed by tis, fever, eosin- to normal ophilia Cephalosporin Recurrent S. aureus endocarditis 1 year after discharge

4 460 MASUR, MURRAY, AND ROBERTS third patient had a generalized, nonconfluent maculopapular rash. Therapy was changed to a parenteral cephalosporin in each patient. Renal biopsies were not performed. In each instance, the serum creatinine returned to the original value within 21 days, and the urine sediment became normal. DISCUSSION Although penicillinase-resistant semisynthetic penicillins are frequently used for the therapy of S. aureus endocarditis, the efficacy and the safety of any one of these agents has not been reported in this setting (2, 14, 16, 20, 27-29). From 1971 to 1977 at The New York Hospital, 29 episodes of S. aureus endocarditis in 27 patients were documented. Twelve patients were treated with nafcillin alone, two patients were treated with nafcillin in combination with gentamicin, and in two patients nafcillin therapy was replaced by another antimicrobial regimen. Nafcillin was highly efficacious in the eight heroin addicts with right-sided endocarditis who received 27 to 43 days of this antimicrobial agent alone. Complications of endocarditis after therapy was initiated were few; one addict had a cerebral embolus and another a pulmonary embolus during therapy. Each addict was discharged with no clinical or bacteriological evidence of endocarditis. Five of the addicts were available for follow-up study: four had no recurrence of their endocarditis 1 to 4 years after discharge; one was well 3 months after discharge. Two addicts treated with nafcillin and gentamicin, and one patient whose nafdilhin therapy was changed to methicillin upon transfer to another hospital, also had no evidence of disease at the time of hospital discharge. The favorable prognosis of heroin addicts with S. aureus endocarditis in our series is similar to the experience reported from other institutions (2, 14, 20, 27, 28). Although systemic complications of endocarditis were not infrequent in those series, bacteriological failures or deaths were unusual. The efficacy of individual antibiotics was not evaluated, but most patients were treated with either methicillin, a cephalosporin, or a combination of agents. It would appear that these other regimens have no advantage over nafcillin in terms of efficacy. In contrast, the prognosis in the five non-heroin addicts with left-sided endocarditis treated with nafcillin in our series was distinctly unfavorable. The one patient who failed to achieve bacteriological cure had a prosthetic aortic valve. Of the other four patients, two died during therapy; the other two patients had no indication of relapse 6 months after discharge. Experience at other institutions has been similarly ANTIMICROB. AGENTS CHEMOTHER. unfavorable in nonaddicts with S. aureus endocarditis (14, 16, 29, 30). Mortality has been greater than 50% in several of these series. Poor prognosis was associated with age (>50 years), underlying cardiac disease, or underlying systemic diseases such as diabetes mellitus, alcoholism, or cirrhosis (30). Of our five nonaddicts, three had underlying heart disease, and four were >50 years old. In other series, methicillin, nafcillin, a cephalosporin, clindamycin, and vancomycin were used alone or in combination with other drugs (2, 3, 7, 28, 30). No one regimen has been clearly shown to be most efficacious. Adverse reactions to nafcillin have rarely been reported. Nafcillin-induced agranulocytosis similar to that seen in our patient has been reported in two previous cases (19, 24). In each instance, bone marrow examination revealed maturation arrest at the myelocyte stage; neutrophils were detected in the peripheral blood promptly, and white blood counts were nornal within 1 to 3 weeks. Although granulocytopenia was not apparent among our six patients treated with methicillin, methicillin-induced granulocytopenia has been documented in at least eight patients (6, 8, 16, 17, 26). Bone marrow findings and reversibility of the abnormality have been similar to those with nafcillin. The mechanism by which nafcillin or methicillin leads to maturation arrest is unknown, although a direct toxic effect on the bone marrow has been suggested (19). Nafcillin-induced nephritis has been reported previously, although the patient had been treated with methicillin as well (23). Numerous cases of nephritis have been attributed to methicillin (1, 4, 5, 10-13, 15, 22). In our series, three of six patients receiving methicillin developed an active urinary sediment, abnormal creatinine, and fever or rash, all ofwhich promptly remitted after methicillin was discontinued. Similar findings were not observed in patients receiving nafcillin. Although the mechanism by which nafcillin or methicilhin causes nephritis is also unknown, a role for drug-induced antitubular basement membrane antibodies has been proposed (4). Methicillin nephritis does not appear to be dose related, and thus the higher incidence of nephritis with methicillin as opposed to nafcillin is probably unrelated to the higher doses of methicillin that are used (10). The only other adverse reaction to nafcillin noted in our series was a generalized maculopapular rash in two patients. The rash was selflimiting and did not necessitate alteration in antimicrobial therapy. A similar rash was also observed in one of the six patients treated with methicillin. Nafcillin is highly efficacious for the treatment

5 VOL. 14, 1978 of S. aureus endocarditis in heroin addicts, and yields results comparable to those obtained from other drugs for the treatment of S. aureus endocarditis in nonaddicts. Since serious adverse reactions to nafcillin have been rare in our experience, but relatively common with methicillin, we regard nafcillin as the preferable penicillinase-resistant penicillin for the treatment of S. aureus endocarditis. Oxacillin, however, was not evaluated in this study. Few adverse reactions have been associated with oxacillin, and limited data suggest that oxacillin may be as efficacious and safe as nafcillin (2, 9). Finally, the poor prognosis of non-heroin addicts with S. aureus endocarditis in this as well as other studies warrants careful consideration of other possible antimicrobial regimens such as the combination of a penicillinase-resistant penicillin with an aminoglycoside (21, 25). ACKNOWLEDGMENTS This study was supported in part by Public Health Service training grant AI from the National Institute of Allergy and Infectious Diseases and by a grant from Wyeth Laboratories. LITERATURE CITED 1. Baldwin, D. S., B. B. Levine, and R. T. McCluskey Renal failure and interstitial nephritis due to penicillin and methicillin. N. Engl. J. Med. 279: Banks, T., R. Fletcher, and N. Ali Infective endocarditis in drug addicts. Am. J. Med. 55: Barry, A. L The antimicrobial susceptibility test: principles and practices. Lea and Febiger, Philadelphia. 4. Bowder, W. A., D. H. Lehman, and J. D. Egan Antitubular basement membrane antibodies in methicillin associated interstitial nephritis. N. Engl. J. Med. 291: Brauninger, C. E., and J. S. Remington Nephropathy associated with methicillin therapy. J. Am. Med. Assoc. 203: Bullock, W. E., Jr Evaluation of 2,6-dimethoxyphenyl penicillin therapy in staphylococcal sepsis, p Antimicrob. Agents Chemother Burch, K H., E. L Quinn, F. Cox, T. Madhavan, E. Fisher, and D. Romig Intramuscular clindamycin for therapy of infective endocarditis. Am. J. Cardiol. 38: Caldwell, J. R., and L. E. Cluff Adverse reactions to antimicrobial agents. J. Am. Med. Assoc. 230: Dismukes, W. E Oxaciilin-induced hepatic dysfunction. J. Am. Med. Assoc. 226: Ditlove, J., P. Weidmann, M. Bernstein, and S. G. Massry Methicillin nephritis. Medicine 56: NAFCILLIN THERAPY FOR S. AUREUS ENDOCARDITIS Doyle, W. F., F. R. Davey, and R. E. Chojnacki Interstitial nephritis associated with methicillin therapy. Milit. Med. 139: Fergin, R. D., and A. Fiascone Hematuria and proteinuria associated with methicillin administration. N. Engl. J. Med. 272: Gilbert, D. N., R. Gourley, and A. d'agostino Interstitial nephritis due to methicillin, penicillin and ampicillin. Ann. Allerg. 28: Hook, E. W., and R. L. Guerrant Therapy of infective endocarditis, p In D. Kaye (ed.), Infective endocarditis. University Park Press, Baltimore. 15. Jensen, H. A., A. B. Halvig, and K. L Savnamaki Permanent impairment of renal function after methicillin nephropathy. Br. Med. J. 4: Kaye, D Changes in the spectrum, diagnosis and management of bacterial and fungal endocarditis. Med. Clin. North Am. 57: Levitt, B. H., A. J. Gottlieb, and I. R. Rosenberg Bone marrow depression due to methicillin, a semisynthetic penicillin. Clin. Pharmacol. Ther. 5: McElfresh, A. E., and N. N. Huang Bone marrow depression resulting from the administration of methicillin. N. Engl. J. Med. 266: Markewitz, S. M., M. Rothkopf, F. D. Holden, D. M. Stitch, and R. J. Duma Nafcillin induced agranulocytosis. J. Am. Med. Assoc. 232: Menda, K. B., and S. L Gorbach Favorable experience with bacterial endocarditis in heroin addicts. Ann. Intern. Med. 78: Murray, H. W., F. M. Wigley, J. J. Mann, and R. P. Arthur Combination antibiotic therapy in staphylococcal endocarditis. Arch. Intern. Med. 136: Nolan, C. M., and R. S. Abernathy Nephropathy associated with methicillin therapy. Arch. Intern. Med. 137: Parry, M. F., W. D. Ball, J. E. Conte, and S. N. Cohen Nafcilihn nephritis. J. Am. Med. Assoc. 225: Sandberg, M., C. U. Tuazon, and J. N. Sheagren Neutropenia probably resulting from nafcillin. J. Am. Med. Assoc. 232: Sande, M. A., and K. B. Courtney Nafcillingentamicin synergism in experimental staphylococcal endocarditis. J. Lab. Clin. Med. 88: Simon, R. J., and L. A. Rantz Newer penicillins: 11 clinical experiences with methicillin and oxacillin. Ann. Intern. Med. 57: Tuazon, C. U., T. A. Cardella, and J. N. Sheagren Staphylococcal endocarditis in drug users-clinical and microbiologic aspects. Arch. Intern. Med. 135: Watanakunakorn, C., andl. M. Baird Prognostic factors in Staphylococcus aureus endocarditis and results of therapy with a penicillin and gentamicin. Am. J. Med. Sci. 273: Watanakunakorn, C., J. S. Tan, and J. P. Phair Some salient features of Staphylococcus aureus endocarditis. Am. J. Med. 54: Wolf, D. J., and G. D. Resnick Nafcillin-induced neutropenia. N.Y. State J. Med. 78: