Occult manifestations of bacteraemia in an 82 year old womanBMJ 2014; 349 doi: https://doi.org/10.1136/bmj.g6807 (Published 24 November 2014) Cite this as: BMJ 2014;349:g6807
- Correspondence to: C Gouldthorpe
An 82 year old woman presented with breathlessness, increasing confusion, and a feeling of being generally unwell for the past two weeks. As well as a cough—productive of clear sputum—and dysuria, she also reported pain in her right rib on admission.
Her medical history included hypertension, osteoarthritis, hypothyroidism, and polymyalgia rheumatica, for which she was taking 10 mg prednisolone a day. Two years earlier she had been treated for venous leg ulcers, which were positive for meticillin resistant Staphylococcus aureus (MRSA).
On examination she did not have a fever, appeared pale, and had dry mucous membranes. Her respiratory rate was 18 breaths/min with oxygen saturations of 96% on air. Both lung bases were dull to percussion and reduced breath sounds were noted. No additional heart sounds were noted and nor were there any signs of infective endocarditis. Furthermore, she had no ulcers on presentation.
Initial investigations showed a raised white cell count (17.1×109/L; reference range 4.0-11.0), raised urea mmol/L (9.5; 2.5-7.8), raised creatinine (142 µmol/L; 49-90), and raised C reactive protein (2657 nmol/L; 0-76.2). A midstream urine specimen was negative for micro-organisms. and chest radiography showed no focal consolidation. Subsequent blood cultures were positive for S aureus, sensitive to flucloxacillin and resistant to penicillin. Paraspinal abscesses were detected on magnetic resonance imaging (fig 1⇓).
1. What are the common sources of Staphylococcus aureus bacteraemia?
2. What investigations would you consider in the management of this patient?
3. What are the management options of S aureus bacteraemia?
1. What are the common sources of S aureus bacteraemia?
S aureus bacteraemia commonly arises from skin and soft tissue infections and indwelling venous catheters. It can also arise from pneumonia, endocarditis, or deep seated foci.
S aureus commonly colonises the skin and mucous membranes. Although it is usually harmless, it can result in bacteraemia.1 The organism is capable of upregulating virulence factors to evade the bloodstream, progress with metastatic infection, and seed deep tissues.2 S aureus bacteraemia is a common and important cause of both hospital and community acquired sepsis and is associated with considerable morbidity and mortality. In England, during 2013 to 2014, 1.6 cases of meticillin resistant S aureus (MRSA) bacteraemia per 100 000 population were reported, as well as 17.4 cases of meticillin sensitive S aureus (MSSA) bacteraemia per 100 000 population. Although this was a 6.7% drop in reported cases of MRSA compared with the previous year, a 5.4% increase was seen for MSSA.3
Susceptibility to infection relies on an interaction between bacterial, environmental, and host factors. Increasing age is an independent risk factor for S aureus bacteraemia. Any factors resulting in immunosuppression, including iatrogenic causes such as the use of corticosteroids, further increase the risk, as do local factors such as breaks in the skin or venous catheters.1 It is necessary to exclude leg ulcers in all cases.
In reported cases of MSSA bacteraemia in the previous year, the most common isolated source was infected skin and soft tissue (21.1%), followed by catheters and lines (13.4%). Endocarditis, pneumonia, and osteomyelitis among others were collectively grouped and accounted for 41.5% of cases. In 24.0% of cases the source remained unknown. Similar figures were found for MRSA bacteraemia. Soft tissue infections accounted for 19.4% of cases of MRSA, closely followed by catheters and lines at 13.3%.3 Details of any recent hospital admissions including previous interventions, MRSA infections, or episodes of bacteraemia need to be noted.4 5
2. What investigations would you consider in the management of this patient?
Investigations aim to identify the potential source. Positive blood cultures 48-72 hours after antibiotic treatment point towards an intravascular source, and echocardiography is mandatory in all patients with S aureus bacteraemia. A midstream urine specimen, swabs of any suspicious skin lesions, cultures of removed line tips, chest radiography, abdominal ultrasound, and computed tomography or magnetic resonance imaging may be considered, depending on the clinical presentation.
Important sequelae, which are seen in up to a third of cases of S aureus bacteraemia, are caused by local spread or haematogenous seeding. Sequelae include infective endocarditis, septic arthritis, osteomyelitis, and deep seated abscesses.1 Furthermore, skin and soft tissue infection, infections of prosthetic devices, and involvement of the epidural space are all possible, as is the involvement of intra-abdominal organs such as the kidney and spleen.5 Unfortunately recurrence is also possible.
Several factors influence the severity of the infection and the potential for complications. These include evidence of metastatic infection at presentation, the persistence of bacteraemia 48-96 hours after the initial blood culture, the persistence of a fever more than 72 hours after the start of treatment, community acquisition, age over 60 years, and embolic stroke.5 6
Unfortunately, the site of infection is often occult and asymptomatic. A thorough history and examination should aim to identify any potential sources of infection or foci. This should include any point tenderness, the presence of joint effusions, a new or changing murmur, and any peripheral stigmata of endocarditis.5 Although radiographic imaging can help identify both the source and complications of S aureus bacteraemia, its use should be tailored to the presentation, such as the use of magnetic resonance imaging in back pain to rule out epidural abscesses or osteomyelitis.5
It is crucial to exclude infective endocarditis when assessing a patient with S aureus bacteraemia, in accordance with the British Society for Antimicrobial Chemotherapy (BSAC) and Infectious Disease Society of America (IDSA) guidelines.7 8 Although it can be an important source, it can also be a serious complication. If echocardiography is being considered to assess the presence of vegetations, transoesophageal echocardiography is superior to the transthoracic approach, particularly for small vegetations (<5 cm).6 Transoesophageal echocardiography identified 19% of cases of S aureus bacteraemia with associated infective endocarditis after transthoracic echocardiography had been negative.9 Cardiac complications can be identified in high risk patients, such as those with prosthetic cardiac valves, and echocardiography is useful in low risk settings where a short course of antibiotics is desired.5 This reduces the length of stay, the potential for adverse effects, and the economic burden. Positron emission tomography is an emerging technique in the investigation of complicated cases and is particularly promising in the detection of metastatic infectious foci.10
3. What are the management options in S aureus bacteraemia?
Removal or drainage of potential foci, prolonged antibiotic treatment (at least 14 days), and early detection of complications. Failure to identify an infective source and treat appropriately is strongly associated with an adverse outcome.
Compared with other pathogens implicated in bacteraemia, such as streptococci and enterococci, S aureus is more difficult to treat. The removal of any potential foci, prolonged antibiotic treatment for a minimum of 14 days, and early detection of complications are central to management.11 An undetected infective source is strongly associated with an adverse outcome, highlighting the importance of early detection and appropriate management. A prospective cohort study found that more than 56% of patients relapsed if a foreign body persisted compared with only 16% if the foreign body was removed or no device was identified (P<0.01).12 Failure to remove potential foci increases the likelihood of relapse and treatment failure. A proportion of patients will have an unidentifiable focus even after initial investigations. After the initiation of antibiotics, preferably intravenously, a further set of blood cultures is indicated to enable early recognition of any complications or persisting bacteraemia.11
Management should be carried out under the guidance of a microbiologist or infectious disease physician. Treatment depends on the isolate, antibiotic susceptibility, and the severity of infection. In the past, antibiotics were continued for four to six weeks because of concerns about unidentifiable complications.5 Recent studies suggest that treatment for fewer than 14 days in uncomplicated cases is associated with lower success rates.5 The duration of treatment should be extended to between four and six weeks in more complex cases. Patients who receive the minimum 14 days’ treatment must have any identified foci removed, no localising signs or symptoms, blood cultures repeated within two to four days of the initial culture, and no implanted prostheses.5
A risk assessment of the likelihood of MRSA, including any previous MRSA infections, should be carried out. If the likelihood of MRSA is low, the infection should initially be treated as MSSA with β lactams, such as flucloxacillin, while awaiting blood culture results.13 However, infections with a high likelihood of MRSA should be treated according to the local vancomycin or teicoplanin protocol.13 The use of vancomycin should be limited owing to the prolonged time for clearance of the bacteraemia and the possible development of resistance. Recent studies suggest suboptimal treatment in severe infection with Staphylococcus when using vancomycin.5 Its use is associated with a prolonged clearance of bacteria and a poorer response in MSSA.5
If identified foci are irremovable, the bacteraemia persists, multiple metastatic foci are found, or endocarditis is identified, prolonged treatment (four to six weeks) is indicated.11 The use of rifampicin is controversial and currently under investigation in the ARREST trial.14 Its use as monotherapy is discouraged owing to the possibility of resistance. However, because of its bactericidal activity against S aureus, it may be used successfully as combination therapy.15 Rifampicin has excellent tissue penetration, allowing high intracellular levels, and it can effectively penetrate biofilms.5 15 Unfortunately, there is little evidence that vancomycin improves outcomes in MRSA or native valve endocarditis, but there is evidence of serious adverse effects, related to raised transaminase and drug interactions.5 In addition, the response rate is poorer and bacterial clearance is slower with vancomycin than with β lactams.5
Increasing resistance towards an array of antibiotic regimens has been found among Staphylococcus isolates. Underlying mechanisms include the use of biofilms, whereby bacterial cells and extracellular substances create a matrix that acts as a barrier and encourages a non-growing and more resistant state; the generation of β lactamase, and the incorporation of mec and vanA genes.2 Inappropriate prescribing can therefore be detrimental in cases of S aureus bacteraemia because delayed treatment increases subsequent morbidity and mortality.
No collections were found on abdominal ultrasound and a transthoracic echocardiogram failed to show any vegetations. Because the source remained unknown, magnetic resonance imaging of the spine and computed tomography of the head, chest, abdomen, and pelvis were performed. Discitis at T6/7, as well as paraspinal abscesses and a para-aortic fluid collection, were detected on the short T1 inversion recovery (STIR) sequence of magnetic resonance imaging (fig 2⇓). Although a transthoracic approach was taken, a transoesophageal echocardiogram was considered. However, transoesophageal echocardiography is not performed in patients with evidence of paraspinal abscesses on magnetic resonance imaging and a low likelihood of vegetations. The primary source of this community acquired bacteraemia has not been identified.
After an initial two weeks of flucloxacillin, treatment was extended with a further six weeks of flucloxacillin and rifampicin. Although the abscess and para-aortic fluid were still present they had substantially decreased in size. Because of this improvement the foci were deemed inappropriate for cardiothoracic or neurosurgical input. The patient had made a clinically significant improvement, and her white cell count and C reactive protein had fallen. She stayed in hospital for a prolonged period because no outpatient parenteral antibiotic therapy service was available, and nor were there any rehabilitative or community hospitals in the region to administer antibiotics. She was transferred to a rehabilitative care facility for two weeks before returning home.
Cite this as: BMJ 2014;349:g6807
Competing interests: We have read and understood BMJ policy on declaration of interests and have no interests to declare.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent obtained.