Case report

A 35-year-old man presented with a 6-week history of rigors, night sweats, anorexia and weight loss following a dental extraction for which he had received appropriate antibiotic prophylaxis. His aortic valve had been replaced by a Carpentier-Edwards bioprosthesis 14 years previously for congenital aortic stenosis. On examination he was pyrexial (temperature 39°C), he had no peripheral signs of infective endocarditis or heart failure, but there was evidence of mild aortic regurgitation. His transthoracic echocardiogram showed a calcified aortic prosthesis with mild aortic regurgitation and good left ventricular function. No vegetations were seen. His blood cultures grewStreptococcus mitis sensitive to penicillin, fusidic acid and vancomycin.

He was treated with intravenous penicillin and gentamycin for 16 and 6 days, respectively, followed by oral amoxycillin for a further 4 weeks. His temperature settled within 2 days of starting treatment and he was discharged after his course of intravenous therapy. Although the initial response to antibiotic therapy was dramatic, he represented with shortness of breath, tiredness, pyrexia and malaise. He was referred for further management. His repeat echocardiogram on this occasion showed an echogenic mass consistent with a large vegetation on the aortic valve and significant aortic regurgitation.

He was re-admitted and started on intravenous benzyl penicillin and vancomycin. The latter was given as 1 g over 2 hours, every 12 hours. He was operated on a week later and his aortic valve was replaced by a Sorin bi-leaflet prosthesis. At exploration, there was considerable evidence of destruction of the tissue prosthetic leaflets. Post-operatively, he continued on benzyl penicillin and vancomycin and during the first few days, he had a swinging pyrexia, peaking at 39–40°C which continued despite discontinuation of the benzyl penicillin. At this point he developed an extensive purpuric rash and erythema, which progressed to exfoliative dermatitis. The C-reactive protein was over 200 mg/l. This was identified as Red man syndrome secondary to vancomycin, which was subsequently replaced by teicoplanin on microbiological advice. His rash began to fade away and he made a remarkable improvement with the C-reactive protein falling to 35 mg/l and his temperature settled to normal for 4 days. After the fifth day on teicoplanin, however, his temperature started to climb and there was a recurrence of the swinging pyrexia, peaking at 39–40°C during the day but settling overnight. His C-reactive protein increased to 88 mg/l. There was no recurrence of rash. Blood cultures failed to grow any organisms including Candida and other fungal organisms, and serology was negative for Coxsackie, cytomegalovirus and Epstein Barr virus. A repeat echocardiogram showed mild paravalvular aortic regurgitation but no para-aortic abscesses. After a battery of further negative tests it was felt that this was a drug-induced fever and the teicoplanin was discontinued. Two days after stopping the drug he became apyrexial, and the C-reactive protein fell to normal 6 days later. His temperature remained normal and he was subsequently discharged on warfarin.

Discussion

Vancomycin is a glycoprotein antibiotic that has been associated with an anaphylactoid reaction termed the Red-man syndrome. It usually consists of erythema, flushing and pruritis of the face and upper torso and occasionally progresses to include dyspnoea, chest pain and hypotension.1-4 Originally it was thought that Red-man syndrome occurred secondary to impurities in the formulation (at one time vancomycin was labelled “Mississippi Mud”).5 The reaction, however, continued to be reported despite purification of the product. The incidence of Red-man syndrome in healthy volunteers has consistently been reported as 70–90% in studies that administer vancomycin in a 1-g dose intravenously over 60 minutes.6 7 Interestingly, the incidence in infected patients has been reported to be much lower, varying between 3.4% and 47%.5 8 This difference has been explained by the different definitions of Red-man syndrome and by the fact that some patients may have been on antihistamine agents. This syndrome has been related to the release of endogenous histamine although some patients have been documented to have Red-man syndrome independent of any change in histamine levels.5

Red-man syndrome has been considered to be a mild reaction, occurring primarily on the first dose, with subsequent reactions, if present, of lesser severity.6 9 This was not the case in our patient. Not only was the initial reaction very severe, resulting in exfoliative dermatitis, but it worsened with subsequent doses. Vancomycin was stopped 4 days after the patient first complained of a skin problem and failure to identify the reaction and persistence of vancomycin may have resulted in the severe reaction. Antihistamine prophylaxis,5 lower and more frequent vancomycin dosing,10 and 2-hour infusions11 have been shown to reduce the frequency and severity of Red-man syndrome.

Teicoplanin, a glycopeptide antibiotic (figure 2), similar in antibacterial spectrum to vancomycin, has not been shown to cause Red-man syndrome in cross-over studies, and has been suggested as an alternative agent to vancomycin.12 13 Case reports have, however, documented cross-reactivity. In the case reported by McElrath and colleagues,14 vancomycin was associated with a diffuse erythematous rash and fever on both occasions that it was administered, over a 5-month period, to a heroin addict with methicillin-resistantStaphylococcus aureus infection. Teicoplanin had been started a week after vancomycin had been discontinued on the second occasion and after initial improvement the patient developed a pyrexia up to 40°C and a rash, described now as maculopapular.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Structure of vancomycin

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Structure of teicoplanin

The dramatic teicoplanin-induced pyrexia seen in our patient cannot be labelled Red-man syndrome but would further suggest that there is an element of cross-reactivity in patients who show the classical Red-man syndrome. The reason for this cross-reactivity, if any, is unclear, and it may well be due to the fact that teicoplanin is structurally related to the glycopeptide group of antibiotics which includes vancomycin. Alternative agents would be chosen according to the sensitivities of the microbe being treated and in our case would have included fusidic acid and ciprofloxacin. Teicoplanin may not be the most appropriate substitute for all cases of vancomycin-induced Red-man syndrome.