A patient with chronic haemodialysis with a cardiac pacemaker was admitted for five episodes of bacteraemia with Staphylococcus during an 8-month period. The species identification was complicated since the morphological characters and biochemical reactions were unusual and differing. Molecular biological identification and typing methods revealed that the pathogens for all the episodes were the same strain of Staphylococcus aureus that had small colony variant characteristics. Continuous suppressive antibiotic treatment initiated after the last infection episode has been able to keep the patient free of bacteraemia relapse during the past 24 months without removing the pacemaker.
Statistics from Altmetric.com
Staphylococcus aureus is a common cause of severe foreign body infections that may be difficult to treat effectively without removal of the infected device. Small colony variants (SCVs) are subpopulations of S aureus that are particularly difficult to eradicate and might be implicated in persistent and recurrent infections.1 These phenotypic variants produce small slow-growing non-pigmented, non-haemolytic colonies on routine culture media, which cause them to be frequently undetected or misidentified by standard microbiology procedures, including testing for coagulation in horse citrate plasma. Biochemical characterisation of these variants suggests that they are deficient in electron transport activity.1,2 Poor clinical and microbiological response to prolonged antimicrobial treatment in patients infected with S aureus SCVs has been reported.3,4 The SCV phenotype has also been observed in coagulase negative Staphylococcus (CoNS), Pseudomonas aeruginosa, Burkholderia cepacia and Escherichia coli. We present a case of recurrent bacteraemia with SCVs of S aureus probably due to persistent cardiac pacemaker infection in a patient receiving haemodialysis. After several unsuccessful attempts to eradicate the infection the patient has been kept free of relapse with the use of continuous suppressive antibiotic treatment.
To our knowledge, this is the first case reporting that a chronic foreign body infection caused by SCVs of S aureus can be treated with continuous suppressive antibiotic treatment without removing the foreign body.
A 44-year-old man was admitted five times over an 8-month period beginning in July 2006 because of fever and each time had a positive blood culture with staphylococci.
The patient was diagnosed with insulin-dependent diabetes mellitus (IDDM) in his childhood and had developed diabetic nephropathy, angiopathy, retinopathy and neuropathy. He had been on haemodialysis for 17 years and had been performing home dialysis for the last 5 years through an arteriovenous fistula on his left arm. Due to a total atrio-ventricular blockade in 1999 the patient had received a pacemaker.
Infective endocarditis of the mitral valve was found by transoesophageal echocardiography at the first bacteraemia episode and he was treated for 6 weeks with high doses of cefuroxim, rifampicin and fucidic acid according to our standard regime for infective endocarditis. However, no apparent foci were revealed at the following four episodes. Each time the patient responded quickly to the antibiotics given. Table 1 shows the antibiotic regime given for each episode. The Staphylococcus isolated from blood cultures in all five episodes were finally identified as S aureus. In episode II, S aureus was initially misidentified as CoNS in the laboratory. In episodes III and IV, mixtures of two strains of staphylococci in the blood cultures were suspected: one S aureus, the other CoNS, based on their different phenotypic characteristics. This complicated the diagnostic process because the findings suggested infections with new pathogens rather than relapse due to persistent colonisation. After further investigations using molecular methods, we concluded that the patient was infected with the same clone of SCV of S aureus in all five episodes. This eventually led to the conclusion that the patient had a chronic infection. No certain focus was revealed after thorough clinical investigations supplied with repeated transoesophageal echocardiography, bone scintigraphy and leucocyte scintigraphy. The most probable diagnosis was found to be a chronic pacemaker infection with SCVs of S aureus. The pacemaker was assumed to be infected already at bacteraemia episode I. Due to religious beliefs the patient refused to receive blood transfusion; therefore, it was considered to be associated with too high risk to remove the pacemaker. At the last episode, the patient received intravenous treatment with cefuroxime and vancomycin for 4 weeks. Immediately after the intravenous treatment, the patient was started on continuous prophylactic treatment with cefuroxime axetil tablets 1g four times daily. The patient has not had bacteraemia relapse during the past 24 months since initiation of prophylactic treatment even though the pacemaker was not removed.
There were difficulties in achieving correct identification based on morphological characteristics and phenotypic reactions.
The identification of S aureus is routinely made on Gram positive cocci in clusters that have a positive catalase and a positive coagulase reaction in horse citrate plasma,5 or positive clumping factor/protein A reaction in the rapid slide agglutination method (Staphaurex, Remel, Lenexa, USA). The isolates that were misidentified as CoNS in episodes II, III and IV had four common characteristics: (1) the coagulase reactions using horse citrate plasma had a borderline reaction that was difficult to interpretate, even after 24 hours incubation and repeated testings; (2) all had a negative reaction in the rapid slide agglutination method (Staphaurex); (3) the colonies were small with a slow growth, non-pigmented and non-haemolytic on Danish Blood Agar; (4) the isolates did not ferment mannitol.
Two automated identification systems, API-Staph ID and Vitek G + (bioMerieux, France), could only identify the isolates to the genus level. Both systems suggested S epidermidis, S capitis or S warneri as probable species but with a doubtful profile.
All S aureus isolates were resistant to penicillin and polymyxin but susceptible to dicloxacilin, cefuroxime, rifampicin, fucidic acid and gentamicin. Susceptibility testing was performed according to the agar diffusion method using NeoSensitabs (Rosco, Taastrup, Denmark) on Danish Blood Agar (Statens Serum Institut, Denmark) with semi-confluent growth.6
Molecular diagnostics and typing
Since the S aureus isolate was misidentified as CoNS in episode II, chronic infection/relapse was not suspected. When the patient was re-admitted with bacteraemia episode III (3 months after the second episode), a clinical microbiologist questioned if these episodes were independent of each other or not. To answer this question, the isolates from all three episodes were investigated using molecular methods.
The stored isolates from all three episodes were found to be S aureus protein A (Spa) PCR positive; in addition, partial 16S rRNA gene sequence analysis confirmed that these strains were S aureus. Finally, the correct identification of the isolates from episode II was achieved 3 months after the blood culture turned positive.
When the patient was admitted with fever and had further positive blood cultures in episodes IV and V, we experienced the same difficulties in the laboratory, although based on the experiences from the first three episodes the correct identifications were now achieved on Spa PCR only two days after the blood culture turned positive.
To determine whether the episodes were caused by the same S aureus strain, pulsed-field gel electrophoresis (PFGE) was performed on 13 isolates from the five bacteraemia episodes following standard procedures.7 Figure 1 shows the PFGE result. All S aureus strains from the five episodes (lanes 1, 2, 4–7, 9–13) shared the same PFGE profile regardless of colony morphology. Lanes 3 and 8 were the two probable contaminants in the blood culture from episodes II and III, respectively.
Spa sequencing results confirmed that the S aureus strains from all five episodes were of the same rare spa type t493.8
To find a fast alternative in the microbiological laboratory to make the correct species identification for these SCVs, S aureus/CNS PNA FISH(AdvanDX, Massachusetts, USA) was performed according to the manufacturer’s instruction. This fluorescence in situ hybridisation assay uses fluorescently labelled peptide nucleic acid (PNA) probes that target the species-specific ribosomal RNA (rRNA) in S aureus and CoNS. The PNA FISH results confirmed that the representative isolates from all five episodes were S aureus (fig 2A), while identification of CoNS was made for one of the contaminant isolates (fig 2B). The method can be applied directly on the positive blood culture smear and produce a result in 2.5 hours.
OUTCOME AND FOLLOW-UP
The patient has not had bacteraemia relapse during the past 24 months since initiation of prophylactic treatment even though the pacemaker was not removed.
SCVs are defined as non-pigmented, non-haemolytic colonies approximately 10 times smaller than the parent wild type strain. SCV S aureus has been described as pathogen in serious infections like osteomyelitis,1,9 septic arthritis,10 respiratory tract infections in patients with cystic fibrosis,11 deep-seated abscesses12 and in multi-organ infections.13
The mechanisms behind developing the SCV phenotypes are deficiencies in the electron transport pathway due to auxotrophy for menadione, hemin or thymidine.14 These special phenotypic characteristics pose great challenges for the isolation and correct identification in a routine clinical microbiology laboratory. Application of extended conventional culture media may increase the chance of the recovery and isolation of SCVs.15 The automated identification systems, like API or VITEK, are not sufficient to make the correct species identification for SCVs. Therefore, suspected S aureus SCV isolates should be confirmed by detection of the species-specific genes (Spa, nuc or coa genes) or by hybridisation to the 16s rRNA gene (PNA FISH). Alternatively, 16s rRNA gene sequencing could confirm the diagnosis. In our case, the molecular diagnostic methods were key in revealing that the patient was chronic infected with the same pathogen. This led to change in treatment strategy and the patient could be free of infection relapse in the past 24 months.
The optimal treatment of S aureus SCVs infection has not yet been defined. So far, the experience with treatment of infections due to SCVs S aureus is based on a limited number of case reports and no prospective, randomised studies of treatment protocols are available.
SCVs have reduced susceptibility to aminoglycosides. Deficiency in electron transport results in decreased transmembrane potential resulting in a decreased uptake of aminoglycosides that require a charge difference to be active; therefore, aminoglycosides are not effective for treating SCV infections.
It is documented that SCVs are able to survive in the intracellular milieu under antibiotic pressure.16 The intracellular location of SCVs may provide a survival niche because the microorganisms are protected against several types of antibiotics and host defences.17 The intracellular SCVs, therefore, cannot be completely cleared by cell wall active antibiotics.16 Based on this knowledge, a treatment including antimicrobials with intracellular anti-staphylococcal activity, such as rifampicin in combination with β-lactam antibiotics or vancomycin, seems appropriate.
However, when the infection involves a foreign body, the complete removal of the foreign body seems to be a prerequisite for the infection remission.18–21 Foreign body related infections due to SCVs of S aureus were described in a few case reports (table 3). In most of these cases the correct identification was delayed 1–3 months from the time the Staphylococcus was isolated. The report from Quie in 1969 had no clinical information of treatment and outcome.22 Poor clinical and microbiological responses to prolonged antimicrobial treatment were illustrated in the other cases.18–21 Total eradication of the bacteria could not be achieved with relevant antimicrobial treatment alone and removal of the foreign bodies was necessary for complete cure.
The patient presented in this report rapidly became afebrile each time after initiating relevant antibiotic treatment, indicating that the extracellular bacteria in the bloodstream was easily cleared by a combination treatment regime including rifampicin or fucidic acid or both. The infection, though, relapsed shortly after each course of antimicrobial treatment. This was probably due to biofilm growth on the cardiac pacemaker electrodes that could not be removed. However, continuous suppressive antimicrobial treatment instituted after a 4 weeks antibiotic treatment course for the last bloodstream infection proved to be a sufficient prophylactic measure to avoid new bacteraemia episodes with the pacemaker retained. Although eradication of S aureus biofilm from the pacemaker is unlikely, it is believed that cefuroxim axetil 1g twice daily was able to suppress the growth of S aureus in the biofilm. Therefore, such measures could be considered in patients with foreign body infections caused by S aureus SCVs where removal of foreign bodies is impossible or related to serious clinical risks. The optimal duration of the continuous suppressive antimicrobial treatment has not been defined. Life-long prophylactic treatment is intended in this case.
Small colony variants (SCVs) of Staphylococcus aureus are easily missed or misidentified in a routine clinical microbiology laboratory because of the unique phenotypic characteristics.
In cases of SCVs of S aureus, when it is difficult to determine whether the coagulase reaction is positive or not, the automated identification system with API or VITEK might not be sufficient for correct species identification. Molecular diagnostic methods should be considered to achieve the correct species identification.
In cases with foreign body related infection caused by SCVs of S aureus, where the foreign body cannot be removed, continuous antibiotic suppressive treatment may be considered.
Competing interests: none.
Patient consent: Patient/guardian consent was obtained for publication.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.