Search

We describe 2 cases of endocarditis caused by methicillin-resistant Staphylococcus aureus that failed to respond to intravenous linezolid therapy but were successfully treated with trimethoprim-sulfamethoxazole plus gentamicin and vancomycin plus rifampin.

Serious infections caused by Staphylococcus aureus in HIV-infected patients have been reported. Contributing factors in the development of invasive S aureus infections include a high rate of skin and nasal colonization, frequent dermatologic disease, and the use of intravenous catheters. The authors report three cases of S aureus pericarditis in HIV-infected patients. While cases of viral, mycobacterial, and malignant pericardial effusions in HIV-infected patients have been reported, a review of the literature disclosed only three cases of bacterial pericarditis. Despite appropriate antibiotic therapy and drainage, a patient's condition may abruptly deteriorate and progress to tamponade. Early recognition of bacteremia and pericarditis and monitoring for cardiac tamponade, along with aggressive treatment, can result in a favorable outcome, but mortality remains high, particularly when S aureus is the causative agent.

Disseminated intravascular coagulation (DIC) may complicate severe septicemia caused by Staphylococcus aureus. S. aureus can induce spontaneous platelet aggregation in vitro, the rapidity and degree of which correlates with the severity of DIC in patients with sepsis. Purified peptidoglycan from DIC isolates aggregated human platelets in the presence of staphylococcal protein A with significantly shorter aggregation times than did peptidoglycan from non-DIC isolates. Purified teichoic acid from DIC and non-DIC isolates failed to aggregate platelets in vitro, or in vivo in guinea pigs but inhibited the peptidoglycan-induced aggregation in a dose-response manner. These studies suggest that peptidoglycan may mediate S. aureus-induced spontaneous platelet aggregation in vitro and DIC in vivo. The variability among strains of S. aureus to induce DIC and platelet aggregation may depend on the unique composition of their peptidoglycan and perhaps also the extent of exposure or availability of cell wall teichoic acid.

Sera from 55 parenteral drug abusers with endocarditis due to Staphylococcus aureus were assayed for the presence and titer of rheumatoid factor. Thirteen (24%) of the 55 patients with endocarditis had sera positive for rheumatoid factor at one point or another in their courses; only 2 (7%) of 30 noninfected drug users were found to be positive. It appeared that more severe cases, as evidenced by duration of fever after initiation of antibiotic therapy, were more likely to develop rheumatoid factor.

To study the natural course of catheter-induced endocarditis secondarily infected with Staphylococcus epidermidis, 29 rabbits had catheters introduced surgically through the carotid artery to the aortic valve. Forty-eight hours later the catheters were removed from five rabbits. The rabbits were inoculated intravenously with 10(8) colony-forming units of S epidermidis. Autopsies done at various intervals showed all rabbits with indwelling catheters had noticeable aortic valve vegetations with positive cultures for S epidermidis. In the group with catheters removed after 48 hours, less severe valvular lesions were noted. Metastatic seeding to kidneys, spleen, and liver were noted in both groups. Because of low cure rate in the treatment of S epidermidis endocarditis, this rabbit model could be used to study antibiotic regimens for valvular endocarditis.

10 patients with serious infections caused by Staphylococcus epidermidis (8 cases of endocarditis in non-prosthetic valves, 1 was complicated by osteomyelitis, 1 case of osteomyelitis, and 1 case of septicemia) are described. Clinical and microbiologic features were evaluated including antibiotic sensitivity and synergy studies, phage typing and biotyping. Endocarditis tended to affect the elderly population and the clinical manifestations were quite similar to those caused by Streptococcus viridans. Both patients with osteomyelitis had involvement of the cervical spine with excellent response to antibiotic therapy. The only patient with septicemia acquired via hyperalimentation had delayed clearance of the bacteremia but ultimately responded to intravenous antibiotics. Rifampicin was the most effective of all antibiotics tested. All isolates were sensitive to penicillinase-resistant penicillins and cephalosporins and over half were sensitive to penicillin. Full synergistic activity was demonstrated with cephalothin and nafcillin in combination with rifampicin, and rifampicin-vancomycin was partially synergistic against the majority of the strains. Five of 8 available isolates were non-phage typeable and no definite pattern was established for various types of infections. Four of the 8 isolates were classified as biotype SIIa, 2 biotype SIIc and 2 biotype SVh.

A 25-year-old black female heroin addict presented with Staphylococcus aureus endocarditis. Because of a history of penicillin allergy, therapy was gegun with cephalothin (Keflin); lack of clinical response led to the use of clindamycin. Response was excellent. After six weeks of treatment she was discharged well, only to return six days later with recurrent endocarditis. The organism, confirmed by sensitivity and phage typing, was identifcal to that causing the inital episode. This patient illustrates the apparent failure of clindamycin to eradicate a deep-seated intravascular infection with a sensitive organism.

Counterimmunoelectrophoresis and gel-diffusion techniques were used to study the development of teichoic acid antibodies in eighteen patients with osteomyelitis and seven patients with septic arthritis caused by Staphylococcus aureus. Nine of eleven patients with acute osteomyelitis and three of seven with chronic osteomyelitis had a positive response to tests for teichoic acid antibody. However, only two of seven patients with septic arthritis generated a positive teichoic-acid antibody response. In two patients the test was extremely valuable in the diagnosis and management of osteomyelitis. Antibody detection appears to be a sensitive test for detecting staphylococcal osteomyelitis, especially the acute variety. It does not appear to be a reliable test for septic arthritis. It is also useful for the detection of antibody in patients who had received prior antibiotic therapy, yet have persistent foci of infection. The presence of the teichoic acid antibody, as well as its titer, is of diagnostic value in patients with serious infections caused by Staphylococcus aureus. The detection and quantification of teichoic acid antibodies is of great value for the early diagnosis of patients with acute osteomyelitis caused by Staphylococcus aureus and for assessing the clinical response of such patients.

Staphylococcus aureus is a ubiquitous organism that is normally carried on the skin and body surfaces of man. The nares are sites frequently colonized, and patients and hospital personnel represent the major source of infection. The occurrence of staphylococcal infection depends on the availability of staphylococci and the host resistance to infection. Factors that influence the carrier rate of S. aureus include minimal colonizing dose, effects of antimicrobial therapy, disinfectants in the environment, coincidental respiratory infections, possible effect of immune factors, duration of hospital stay, and regular needle injections. Certain patients such as drug abusers, patients with diabetes, and patients with chronic renal failure are at high risk of S. aureus infections. although underlying immune deficiencies are present, increased carrier rate also might be related to regular needle use, as shown among allergy patients. The significance of carrier state has been defined in outbreaks in hospital nurseries, postoperative patients, and systemic infections such as endocarditis in the drug abuser, the toxic shock syndrome, and dermatologic infections.

Forty patients had staphylococcal endocarditis and a history of parenteral drug abuse. Clinical and microbiologic features of their cases were evaluated. None of our patients were known to have had preexisting valvular disease. The tricuspid valve lesions and their pulmonary complications were the predominant findings. Systemic complications in the form of meningitis, glomerulonephritis, empyema, arthritis, and nosocomial Gram-negative septicemia occurred in 33% of our patients. Of interest was the high incidence of reactions to therapy especially with methicillin sodium, which occurred in 30% of patients. Correlation of phage type and group with the antibiotic sensitivities of individual staphylococci showed that group 3 and phage types 6, 42E, 54, and 75 were much more resistant to penicillin than other groups and types. The clinical outcome did not relate to phage type and group or to antibiotic sensitivity of the organism.

The development of antibodies to teichoic acid was studied in 56 patients with infections due to Staphylococcus aureus. All 28 patients with endocarditis eventually developed teichoic acid antibodies demonstrable both by counterimmunoelectrophoresis and by gel diffusion; however, 7 patients were negative on admission. Eight of 15 patients with S. aureus bacteremia developed antibodies by counterimmunoelectrophoresis and 6 of the 8 were positive by gel diffusion; 4 of those 6 had evidence of seeding of S. aureus. Three of 5 patients with osteomyelitis and 1 of 8 with localized peripheral abscesses had teichoic acid antibodies. Titers of 1:4 or greater by gel diffusion were present in 18 of 28 patients with endocarditis compared with only 1 of 10 patients with nonendocarditic staphylococcal infections. Thus, the demonstration and quantitation of teichoic acid antibodies is of great clinical value in the early diagnosis of infections due to S. aureus and in assessing the likelihood of deep intra- or perivascular seeding.

We studied the carriage rate of Staphylococcus aureus in patients receiving long-term hemodialysis and also noted the incidence of shunt infections, bacteremia, and septicemia in colonized patients. Thirty-one of 50 patients (62%) carried S aureus in the nose, throat, or on the skin, of whom 20 patients developed shunt infections; nine infections resulted in episodes of bacteremia. Patients with chronic renal failure not undergoing hemodialysis had a 21% carriage rate. Thus, there is a high carriage rate of S aureus in asymptomatic patients receiving hemodialysis that is probably related to an increased incidence of shunt infections and bacteremia.

Relative toxicities of methicillin and nafcillin were compared in 70 patients with serious infections caused by Staphylococcus aureus. Of the 29 patients treated with nafcillin, four had fever, rash, and leukopenia and 1 had absolute neutropenia. Of the 41 patients treated with methicillin, 16 experienced 27 reactions. In addition to fever and skin rash, neutropenia and urinary tract abnormalities were common. Methicillin and nafcillin are equally effective in treating S aureus infections, but methicillin was significantly and more frequently associated with adverse drug reaction that was nafcillin.

Sera from patients with endocarditis and bacteremia due to Staphylococcus aureus were compared for peptidoglycan-binding capacity with those from normal blood donors. Those patients treated with beta-lactam antibiotics had higher antigen-binding levels than normal donors and patients treated exclusively with vancomycin (P less than 0.01). The factor responsible for this activity was purified by affinity chromatography from a normal donor and shown to be an immunoglobulin. Specificity studies indicated that the immunodominant determinant was a peptide sequence found in peptidoglycan precursors. Since soluble peptidoglycan molecules having the precursor peptide sequence are known to be secreted by some gram-positive bacteria like Micrococcus luteus when grown in the presence of beta-lactam antibiotics, these soluble molecules may constitute the "natural" immunogen. Such a hypothesis is consistent with the study of the peptidoglycan-binding capacities in the sera of these patients during the course of treatment. For most of the responding patients studied (four of four with bacteremia and seven of nine with endocarditis), a significant increase in peptidoglycan-binding capacity was observed in sera taken 1 to 5 weeks after the initiation of beta-lactam antibiotic therapy (compared with the initial serum studied). No such increase in the peptidoglycan-binding capacity over a similar time span was noted in the sera of people not receiving beta-lactam antibiotics (none of seven).

Twenty-six strains of Staphylococcus aureus obtained from patients with endocarditis were studied for the production of alpha- and/or beta-ribitol teichoic acid (TA), using highly specific anti-TA antibodies prepared in rabbits. A counterimmunoelectrophoretic assay was used. Beta-TA was the predominant residue produced by all strains; alpha-TA was found in all strains, but in smaller amounts and with much strain-to-strain variations. Antibodies in patients' sera were found against beta-TA in higher titers and for longer periods than were anti-alpha-TA antibodies. Antibodies against one or both TA residues were present in all but one of 26 patients.

Because approximately one third of persons injecting illicit drugs carry Staphylococcus aureus in their nose and throat or on the skin, we wondered if the carrier was the person in the drug-injecting population at risk of developing endocarditis. If so, patients with drug-related, staphylococcal endocarditis should have a very high carriage rate of the organism when first admitted to the hospital. Ten such patients were studied within 3 days of admission, and all were carriers of S. aureus. In each case, the phage type of the carried organism matched that of the organism recovered from the blood.

Teicoplanin and rifampicin were evaluated as single and combined agents in the treatment of endocarditis due to Staphylococcus epidermidis in the rabbit model. Rabbits were treated for ten days and the number of bacteria in vegetations determined. At the end of ten days the geometric mean number of bacteria in the vegetations were 5.53 X 10(8), 6.68 X 10(6). 1.10 X 10(4), 2.57 X 10(1) cfu/g of vegetation for control, teicoplanin, rifampicin, and teicoplanin plus rifampicin groups respectively. The MIC and MBC values of the S. epidermidis isolates were 0.78 mg/l for teicoplanin and less than or equal to 0.10 mg/l for rifampicin. In the rifampicin treated group three post-treatment isolates of S. epidermidis tested exhibited marked resistance to rifampicin with MIC and MBC values greater than or equal to 200 mg/l. Teicoplanin and rifampicin were both effective as single agents in the clearance of S. epidermidis from the bloodstream. Rifampicin was more effective than teicoplanin in the clearance of S. epidermidis from vegetations but teicoplanin in combination with rifampicin was more effective than either drug alone.

Two patients with acute endocarditis attributable to Staphylococcus aureus had a clinical syndrome similar to meningococcemia with meningitis. That fulminant S aureus septicemia may on occasion be associated with hemorrhagic skin lesions, thrombocytopenia, coagulation abnormalities, and meningitis is emphasized.

Eight patients with serious staphylococcal infections such as endocarditis, meningitis, epidural abscess, shunt and graft infections were treated with nafcillin, cephalothin or vancomycin in combination with rifampin. In vitro antibiotic sensitivities demonstrated that the Staphylococcus aureus and Staphylococcus epidermidis were highly sensitive to rifampin with minimum inhibitory and bactericidal levels of .0078-.25 micrograms/ml. In all patients reviewed and reported, when serum bactericidal levels were measured before and after the addition of rifampin, there was a positive correlation between microbiological and clinical outcome. Thus, in selected patients with life-threatening infections caused by S. aureus and S. epidermidis rifampin should be considered an adjunctive therapy.

An adult drug abuser with endocarditis due to a group II Staphylococcus aureus developed the staphylococcal scalded skin syndrome (SSSS). Studies of the patient's immune function found cell-mediated immunity (CMI) to be essentially normal; thus, previous suggestions that deficient CMI is required in adults who develop SSSS seem unfounded. Pathogenesis, diagnosis, and therapy of SSSS and its differentiation from drug-induced hypersensitivity reaction are discussed in detail.