| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Case Report |
1 Institute of Medical Microbiology, Virology and Hygiene, Rostock University Hospital, Germany
2 Department of Neurosurgery, Rostock University Hospital, Germany
Correspondence
Franka Lestin
franka.lestin{at}helios-kliniken.de
Received 8 January 2008
Accepted 7 April 2008
Abbreviations: GGS, group G streptococci; MRI, magnetic resonance imaging.
Present address: Institute for Laboratory Medicine, HELIOS-Kliniken Schwerin, Teaching Hospital of the Rostock University, Wismarsche Str. 393-397, 19049 Schwerin, Germany. 
A supplementary figure showing the confirmation of the identity of the three group G streptococcal isolates by PFGE is available with the online version of this paper.
 |
Case report |
|---|
 TOP Case report DiscussionREFERENCES |
|---|
Careful history taking revealed a history of coronary heart disease, a myelodysplastic syndrome with refractive anaemia, chronic recurrent pancreatitis and a seminoma (relapse-free for 14 years). The patient had also suffered from leg ulcers for about 9 months due to circulatory disturbances of both legs and a post-thrombotic syndrome after a pelvic vein thrombosis years ago.
On admission, we saw a critically ill, moderately obese patient (89 kg/173 cm) who was alert and fully oriented. Body temperature was slightly elevated (37.4 °C). The blood pressure was normal at 105/60 mmHg. The heart sounds, skin aspect and nails beds as signs for endocarditis as well as results from transthoracic echocardiography were unremarkable. Arterial pulses were palpable except for the Aa. dorsalis pedis on both sides. There were leg ulcers on both legs with clinical signs of bacterial infection such as inflamed rims, pus, smear and odour (Fig. 1
). The microbiological cultures from superficial swab material taken from both legs grew β-haemolytic group G streptococci (GGS; Streptococcus dysgalactiae subsp. equisimilis) and Staphylococcus aureus. Other parameters of the physical examination were unremarkable.
|
On admission, the white blood cell count was slightly elevated with 12.4x109 cells l–1 (normal range 4–9x109 cells l–1). Other infection parameters such as C-reactive protein (241 mg l–1; normal <5 mg l–1), procalcitonin (2.42 ng ml–1; normal <0.5 ng ml–1) and erythrocyte sedimentation rate (105 mm h–1) were remarkably elevated.
The red blood cell count showed a normocytic anaemia with decreased values for haemoglobin (4.5 mmol l–1; normal range 8.6–12 mmol l–1) and haematocrit (23 %; normal range 40–51 %); thrombocytes were normal (334x109 cells l–1). Liver enzymes were within the normal range except for an elevated alkaline phosphatase (182 U l–1; normal range 38–126 U l–1). A diminished creatinine clearance (0.43 ml s–1; normal range 1.63–2.66 ml s–1) was associated with an elevated serum creatinine (245 µmol l–1; normal range 62–106 µmol l–1) and potassium (5.8 mmol l–1; normal range 3.6–5.1) level. Glucose levels were normal upon repeated measurements with no evidence of diabetes mellitus. All other laboratory parameters obtained were normal. A peripheral blood culture performed on admission grew large colony forming GGS.
The X-ray of the spine showed destruction at the level of L4/5 suggestive of severe spondylodiscitis. Magnetic resonance imaging (MRI) of the spine confirmed a spondylodiscitis and additionally showed a paraspinal subfascial abscess on the left side (Fig. 2).
|
The Gram-stained smear from the surgical site showed Gram-positive cocci arranged in clusters and chains. Histological examination of tissue specimens confirmed an acute purulent, destructive and necrotizing spondylodiscitis without evidence of malignancy. Antibiotic treatment with clindamycin (600 mg three times daily) and penicillin (10 Mio IU three times daily) was immediately started and continued for 18 days.
Two days after surgery, large colony forming β-haemolytic Lancefield GGS were grown in pure culture from all specimens. All GGS isolates were subjected to biochemical identification by the commercial test system Rapid ID 32 Strep (bioMérieux) and appeared as Streptococcus dysgalactiae subsp. equisimilis (code no. 45022041100).
Disc diffusion testing for determination of antibiotic resistance was performed and the isolates were found to be susceptible to penicillin, clindamycin, ciprofloxacin, erythromycin, vancomycin, tetracycline and linezolid according to the criteria of the CLSI/NCCLS (2006). Following genomic DNA preparation and digests with SmaI restriction endonuclease, an analysis of three GGS isolates from the leg ulcers, the blood culture and the intraspinal specimen by PFGE showed identical band patterns, thus strongly indicating the potential route of infection (Supplementary Fig. S1 available with the online version of this paper).
During the first days of antibiotic treatment, the patient recovered well and the infection parameters declined, i.e. C-reactive protein level dropped to 95 mg l–1 on day 9 post-surgery. On day 13 after surgery, the wound healing seemed to be disturbed accompanied by a slight increase of the infection parameters. Repeat MRI of the lower spine confirmed an infection of the left facet joint L4/5 and retained material in the former paraspinal abscess in the same region. Subsequently the wound was revised via a dorsal approach and the abscess was drained. After this second procedure, the patient improved rapidly and infection parameters fell almost to baseline. The patient was mobilized and penicillin administration was reduced to 1 Mio IU three times daily on day 18 after the first surgery. Also the leg ulcers appeared to be healed at that time.
On day 27 after the first surgery, due to a decision by the surgical outpatient clinic the antibiotic therapy was changed to oral ciprofloxacin (500 mg two times daily) plus oral clindamycin (600 mg three times daily) for 4 additional weeks. The patient was discharged to 3 weeks neurological rehabilitation in a specialized hospital.
At the follow-up examination 1 year after surgery, the patient presented with a normal neurological status. He denied any further pain attacks of the lower back and did not experience further episodes of fever or unintended weight loss.
|
Discussion |
|---|
|
TOPCase reportDiscussionREFERENCES |
|---|
Spondylodiscitis is predominantly caused by bacteria transported by the bloodstream as opposed to bacteria penetrating from neighbouring anatomical structures (e.g. from contamination after medical interventions). Due to the special blood supply of the spine, the two vertebrae and the linking intervertebral disc are normally affected. Skin and soft tissue infections, urinary tract infections, endocarditis, respiratory tract infections and intravasal device-associated infections are common primary sources for the haematogenous route of infection (Huttner & Opravil, 2006).
Staphylococcus aureus is the major infective agent responsible for 50–75 % of all cases, followed by a wide range of rarer micro-organisms such as coagulase-negative staphylococci, Gram-negative rods and streptococci (Butler et al., 2006). Depending on the geographical area, Mycobacterium tuberculosis and Brucella spp. could also have some aetiological importance. β-Haemolytic streptococci, predominantly group B streptococci, are occasionally described pathogens in spondylodiscitis patients with underlying diseases such as neoplasia, diabetes mellitus and heart disease, those that have undergone a splenectomy and in those receiving corticosteroid treatment (Falagas et al., 2006; Narváez et al., 2004).
A thorough search of the literature (keywords: ‘spondylodiscitis, spondylitis, discitis, spinal osteomyelitis or vertebral osteomyelitis’ in combination with ‘G streptococcus, Streptococcus dysgalactiae, Streptococcus equisimilis or streptococcal’ in the NCBI and DIMDI databases) identified only five published cases of human spondylodiscitis involving GGS as causative agents (Tobias et al., 1992; Castellarin et al., 1993; Hall & Williams, 1993; Hayashi et al., 2007). Three patients were male and two female. Except for one female, the patients were older than 50 years. Three of them had underlying malignancies as risk factors. Diagnoses were confirmed by fine needle biopsy of the spine or positive blood cultures. The route of infection remained unclear in all cases. Additionally, Kumar et al. (2005) reported three cases of spondylodiscitis caused by Streptococcus dysgalactiae subsp. equisimilis Lancefield group C. The three male patients had no known underlying diseases. In one, infective endocarditis was identified as the potential source for spondylodiscitis.
Using biochemical identification, large colony forming GGS can be classified as Streptococcus dysgalactiae subsp. equisimilis and Streptococcus canis. In addition, Streptococcus dysgalactiae subsp. equisimilis includes large colony forming Lancefield group C streptococci. Whereas Streptococcus canis is associated with ulcer infections in dog owners, Streptococcus dysgalactiae subsp. equisimilis apparently does not have an animal reservoir (Woo et al., 2001). GGS share the same ecological niche with group A and C streptococci, exhibit considerable overlap in their disease profiles and share some pathogenicity factors such as M protein and extracellular enzymes, e.g. haemolysin.
GGS may colonize the human pharynx, skin, and intestinal and genitourinary tracts. They are well known as pathogens causing pharyngitis, possibly complicated by glomerulonephritis, lower respiratory tract infections, sepsis and, much less frequently, endocarditis and meningitis (Kaufhold & Ferrieri, 1993). Since most GGS patients with invasive infections have some underlying condition such as diabetes mellitus, malignancy, cardiovascular disease, bone and joint diseases or liver cirrhosis, GGS are assumed to have a generally low pathogenic potential (Liu et al., 1995). Since GGS still await full genome sequencing, the genetic differences from Streptococcus pyogenes which could explain the lower virulence remain obscure.
Although deep-seated infections at other anatomical sites as potential steps of the infectious process cannot formally be excluded, the present case demonstrated how a local skin infection with GGS could lead to a complicated spondylodiscitis. The genetic identity of the isolates from the leg ulcer, the blood culture and the spine confirmed the route of infection. Remarkably, the patient did not present any diabetes mellitus or consuming malignancies during the entire infection period and up to 1 year after the hospital admission. His myelodysplastic syndrome with refractive anaemia only involved the red blood cell line and did not progress. Generally, myelodysplastic syndrome patients suffer predominantly from viral infections. That corresponds to the findings of Kumar et al. (2005), who reported three spondylodiscitis cases without underlying diseases caused by Streptococcus dysgalactiae subsp. equisimilis (Lancefield group C).
On the whole, GGS are susceptible to penicillins, cephalosporins and vancomycin. Erythromycin resistance is only occasionally observed. Thus, in systemic GGS infections such as pyogenic spondylodiscitis, a conservative intravenous antibiotic therapy over 4–12 weeks leads to good results (Mueller et al., 2004; Grados et al., 2007). The drug of choice is penicillin. Combinations with gentamicin or clindamycin are recommended for life-threatening or specially located infections such as endocarditis and osteomyelitis. Identification of the responsible bacterium via needle biopsy sampling or blood cultures and subsequent susceptibility testing is essential for this strategy (Phadke et al., 2001). Additional surgery is recommended in patients with neurological deficits, epidural abscesses, spinal deformity or a protracted course of disease (e.g. continuous rise of infection parameters under standard antibiotic therapy judged as effective by in vitro testing).
Because of the neurological deficit, our patient underwent immediate surgery and resurgery after persistently raised infection parameters and non-improved radiological imaging. This strategy led to an excellent clinical result after 1 year.
The current case demonstrates the risks especially for older (>60 years), predominantly male, patients of developing invasive disease upon exposure to Streptococcus dysgalactiae subsp. equisimilis, even if classical underlying diseases such as diabetes mellitus or actual known malignancies are absent. The pathogenic potential of GGS should not be underestimated, despite a commonly successful antibiotic therapy.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPCase reportDiscussionREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
