I believe in the power of microbiology results but when nothing grows from blood cultures it can seem as if we are advising without any evidence! A recent patient comes to mind; a middle aged man who presented with a fever and whose routine echocardiography investigation revealed a large vegetation on his aortic valve. The team discussed the patient with the Microbiologist and three sets of blood cultures were taken, no antibiotics were started whilst the results were pending as the patient was not septic. Five days later the blood cultures were negative and the clinical team were getting nervous about their patient not being on treatment.
Infective endocarditis is an infection of the endocardium of the heart. Diagnosis is based upon Duke’s criteria. Infective endocarditis usually implies infection of the heart valves; however infection can also be related to transmural thrombosis and congenital heart defects such as atrioseptal defects and ventriculoseptal defects. Untreated endocarditis has 100% mortality.
Treatment is guided by the specific causative microorganism. The mainstay of investigation is the humble blood culture, which is positive in 95% of cases of infective endocarditis. However, what do you do if your patient is one of the 5% of culture-negative endocarditis patients? It is best to take a systematic approach to identifying the causes of infective endocarditis, especially when nothing grows from blood cultures.
Systematic investigation of culture-negative endocarditis
In patients with culture-negative infective endocarditis it is important to be selective and systematic about investigating the underlying cause. Some causes are more common than others and it is sensible to look for these first. If a blunderbuss approach is used there is a higher likelihood of missing a diagnosis or getting a false positive test for a rare cause which may lead to mistakes in changing antibiotic treatments.
So what are the possible reasons for a culture-negative result?
Previous administration of antibiotics
If antibiotics are given before the blood cultures are taken, the rate of positive blood cultures is reduced by 35-40%. The rate depends on whether the infecting organism is sensitive to the antibiotic given and how long the patient has been on the antibiotics. If the patient has only received a few days of antibiotics then stopping them and waiting for a few days more and then repeating the blood cultures usually allows the bacterium to grow. As long as the patient is not septic they are very unlikely to come to any harm in this short time period and it is in their benefit to know exactly what is being treated as they can then be given the optimal and safest treatments.
Non-culturable bacteria
There are a number of bacterial causes of infective endocarditis which are virtually impossible to grow in routine microbiology laboratories including the more common Coxiella burnetii (Q fever) and Bartonella spp., as well as the much rarer Tropheryma whipplei, Mycoplasma spp., Chlamydia spp. and Legionella spp. These bacteria are not detectable using blood cultures but instead are detected either by serology or molecular methods which cost more (approx. £150 per test) and take longer (up to 2 weeks).
Slow growing or fastidious bacteria
There are a number of bacterial causes of infective endocarditis which grow very slowly or need specific growth requirements. Routine processing of a blood culture will not always discover these bacteria.
Brucella spp. and the HACEK bacteria (see previous blog) often will not grow within the standard five days, therefore if these are suspected (e.g. risk of exposure or a history of dental problems) indicate these risk factors on the request form and the laboratory will then incubate the blood cultures for up to 14 days.
Other bacteria require extra pyridoxine (vitamin B6) and cysteine (an amino acid) in order to grow, e.g. Abiotropohia defectiva (previously known as nutritionally variant Streptococcus). The traditional method for identifying these bacteria was to streak a line of Staphylococcus aureus on a blood agar plate which would break down the blood releasing these extra nutrients which in turn allowed the A. defectiva to grow as satellite colonies close to the S. aureus. Nowadays, an easier way of growing A. defectiva is to use a Fastidious Anaerobic Agar (FAA) plate and incubate in a CO2 incubator as this agar has specifically had extra nutrients added to help fastidious, fussy, bacteria grow. It’s a bit like adding tomato ketchup to kids’ meals!
Fungi
Most Candida spp. will grow very well in blood cultures but some can be problematic. One such species is Malassezia furfur, a rare fungus associated with infective endocarditis secondary to giving total parenteral nutrition (TPN) through a central venous catheter. This fungus requires lipid to grow and therefore supplemented agar should be used.
Moulds such as Aspergillus spp. are a very rare cause of infective endocarditis. They do not grow in blood cultures. Diagnosis would only be made if the fungi were seen by histological examination of a surgically removed heart valve.
Non-infectious
There are a number of rare non-infectious causes of endocarditis. In these situations a vegetation is formed but it is sterile, there is no microorganism present. Diagnosis is difficult, and can be delayed whilst waiting for the investigation of bacterial causes.
Non-infectious causes of endocarditis include:
Don’t forget to take a detailed history when making your list of differential diagnoses. It is also critical to take a thorough clinical history about potential exposure to fastidious or non-culturable bacteria e.g. dietary, occupational or travel risk factors, however treatment for these bacteria is not routinely started as they are very rare, and treating them may compromise the treatment of more common causes.
The following steps take into account a decreasing likelihood of causes of endocarditis in the UK:
Treatment is guided by the specific causative microorganism. The mainstay of investigation is the humble blood culture, which is positive in 95% of cases of infective endocarditis. However, what do you do if your patient is one of the 5% of culture-negative endocarditis patients? It is best to take a systematic approach to identifying the causes of infective endocarditis, especially when nothing grows from blood cultures.
Systematic investigation of culture-negative endocarditis
In patients with culture-negative infective endocarditis it is important to be selective and systematic about investigating the underlying cause. Some causes are more common than others and it is sensible to look for these first. If a blunderbuss approach is used there is a higher likelihood of missing a diagnosis or getting a false positive test for a rare cause which may lead to mistakes in changing antibiotic treatments.
So what are the possible reasons for a culture-negative result?
Previous administration of antibiotics
If antibiotics are given before the blood cultures are taken, the rate of positive blood cultures is reduced by 35-40%. The rate depends on whether the infecting organism is sensitive to the antibiotic given and how long the patient has been on the antibiotics. If the patient has only received a few days of antibiotics then stopping them and waiting for a few days more and then repeating the blood cultures usually allows the bacterium to grow. As long as the patient is not septic they are very unlikely to come to any harm in this short time period and it is in their benefit to know exactly what is being treated as they can then be given the optimal and safest treatments.
Non-culturable bacteria
There are a number of bacterial causes of infective endocarditis which are virtually impossible to grow in routine microbiology laboratories including the more common Coxiella burnetii (Q fever) and Bartonella spp., as well as the much rarer Tropheryma whipplei, Mycoplasma spp., Chlamydia spp. and Legionella spp. These bacteria are not detectable using blood cultures but instead are detected either by serology or molecular methods which cost more (approx. £150 per test) and take longer (up to 2 weeks).
Slow growing or fastidious bacteria
There are a number of bacterial causes of infective endocarditis which grow very slowly or need specific growth requirements. Routine processing of a blood culture will not always discover these bacteria.
Brucella spp. and the HACEK bacteria (see previous blog) often will not grow within the standard five days, therefore if these are suspected (e.g. risk of exposure or a history of dental problems) indicate these risk factors on the request form and the laboratory will then incubate the blood cultures for up to 14 days.
Other bacteria require extra pyridoxine (vitamin B6) and cysteine (an amino acid) in order to grow, e.g. Abiotropohia defectiva (previously known as nutritionally variant Streptococcus). The traditional method for identifying these bacteria was to streak a line of Staphylococcus aureus on a blood agar plate which would break down the blood releasing these extra nutrients which in turn allowed the A. defectiva to grow as satellite colonies close to the S. aureus. Nowadays, an easier way of growing A. defectiva is to use a Fastidious Anaerobic Agar (FAA) plate and incubate in a CO2 incubator as this agar has specifically had extra nutrients added to help fastidious, fussy, bacteria grow. It’s a bit like adding tomato ketchup to kids’ meals!
Fungi
Most Candida spp. will grow very well in blood cultures but some can be problematic. One such species is Malassezia furfur, a rare fungus associated with infective endocarditis secondary to giving total parenteral nutrition (TPN) through a central venous catheter. This fungus requires lipid to grow and therefore supplemented agar should be used.
Moulds such as Aspergillus spp. are a very rare cause of infective endocarditis. They do not grow in blood cultures. Diagnosis would only be made if the fungi were seen by histological examination of a surgically removed heart valve.
Non-infectious
There are a number of rare non-infectious causes of endocarditis. In these situations a vegetation is formed but it is sterile, there is no microorganism present. Diagnosis is difficult, and can be delayed whilst waiting for the investigation of bacterial causes.
Non-infectious causes of endocarditis include:
- Antiphospholipid syndrome associated with the presence of antiphopholipid antibodies which allow vegetations to form on the heart valves even though these are sterile. This syndrome has been associated with rheumatological disorders e.g. systemic lupus erythematosis (SLE) as well as malignancies
- Malignancy e.g. atrial myxoma, carcinoid and direct invasion by cancer cells
- Autoimmune e.g. rheumatic heart disease, SLE, polyarteritis nodosa, Behcet’s disease
- Post-valvular surgery e.g. stitches promoting sterile clot formation
- Miscellaneous e.g. eosinophillic heart disease, ruptured chordae and myxomatous degeneration all promoting clot formation
Don’t forget to take a detailed history when making your list of differential diagnoses. It is also critical to take a thorough clinical history about potential exposure to fastidious or non-culturable bacteria e.g. dietary, occupational or travel risk factors, however treatment for these bacteria is not routinely started as they are very rare, and treating them may compromise the treatment of more common causes.
The following steps take into account a decreasing likelihood of causes of endocarditis in the UK:
Click for larger image
Download a PDF of this flow chart, it prints onto a standard postcard (4x6 inches) and can be used to bookmark the endocarditis pages in your copy of Microbiology Nuts & Bolts.
Empirical treatments
It is important to base empirical treatment on the type of heart valve present (be it native or prosthetic) as well as speed of onset and how unwell the patient is.
Empirical treatments
It is important to base empirical treatment on the type of heart valve present (be it native or prosthetic) as well as speed of onset and how unwell the patient is.
Click for larger image
This approach was employed for the middle aged man who presented with a fever, discussed above. The blood culture incubation was extended to 14 days and serology was arranged for Coxiella burnetii and Bartonella spp. as well as antiphospholipid antibodies and rheumatoid factor. After 3 days the blood culture was positive with a Gram-positive coccus in chains which was plated to routine plates plus FAA. The fourth day an Abiotrophia defectiva was confirmed. The patient was started on IV Benzylpenicillin 2.4g 4 hourly and IV Gentamicin 1mg/kg BD. Serology came back negative on day 10.
After 4 weeks he had made a full recovery with no complications and was discharged home.
After 4 weeks he had made a full recovery with no complications and was discharged home.
RSS Feed
