Research into the link between sciatica and microorganisms is being undertaken by a team in Birmingham. BackCare is part-funding this research. This is a progress report from the team (31st October 2002).

An association between sciatica and microorganisms

Investigators:

Professor TSJ Elliott¹, Consultant Microbiologist
Dr A Stirling², Consultant Spinal Surgeon
Dr P Lambert³, Reader in Microbiology
Dr T Worthington¹, Clinical Research Scientist
Dr M Jiggins², Research Fellow

1. University Hospital Birmingham NHS Trust, Edgbaston, Birmingham, UK
2. Royal Orthopaedic Hospital, Northfield, Birmingham, UK
3. Aston University, Aston Triangle, Birmingham, UK 28th February 2002

Introduction
The pathogenesis of low back pain and sciatica remains poorly understood despite being one of the commonest causes for consultation in primary care.¹ Since mechanical compression involving the spinal disc space is insufficient alone to account for the pain associated with sciatica, attention has focused on inflammation of the nerve root resulting from exposure to the disc during herniation. The inflammation that is associated with sciatica has raised the question of a microbial aetiology, possibly with low virulent microorganisms, as found for example, with prosthetic hip infections. Over the last two decades the skin commensals, coagulase-negative staphylococci and propionibacteria, considered primarily as contaminants when isolated from patients, are being increasingly recognised as aetiological agents of infection. Tunney and colleagues², for example, have demonstrated the presence of Propionibacterium acnes and coagulase-negative staphylococci in situ on prosthetic joints studied at revision arthoplasty.

The results of our recent research part funded by BackCare, also suggest that these microorganisms, particularly in P.acnes, may be associated with chronic low grade infection in the lower vertebral discs of patients with severe sciatica.³ Indeed, 53% of patients with severe sciatica who were recruited into the study and underwent microdiscectomy, had positive cultures within 7 days of incubation. P.acnes was recovered in pure culture from 84% of the tissue samples. Furthermore, this preliminary study also demonstrated that significant elevations of a specific serum antibody to a novel exocellular bacterial cell wall component termed lipid S, a glycerophospholipid, was present in those patients yielding positive cultures. These results were significantly different from those obtained from control patients. Our preliminary results suggest that there is an association between P.acnes and sciatica, the diagnosis of which could be aided by the use of the novel lipid S antigen.

The aim of our research is to further investigate the role of Gram-positive microrganisms, particularly coryneforms such as P.acnes in the inflammatory reactions around the nerve root which can result in sciatica. If a link between those low virulent skin microorganisms and sciatica can be confirmed, early diagnosis may indeed lead to improved treatment strategies. Early intervention with appropriate antibiotics in selected patients may modify the subsequent clinical progression of this association between microorganisms and sciatica are presented in this progress report. In addition, a detailed plan of our future research extending over a 30-month period and associated costs is also given.

Methods
Microbiological assessment of tissue samples and estimation of serum levels of anti-lipid S IgG

Two-hundred and seven patients diagnosed as having discogenic radiculitis have been recruited into the study to date. The patients underwent microdiscectomy to relieve the unremitting pain associated with their sciatica. Disc and tissue samples were removed during surgery and sent for microbiological investigation. Disc and tissue samples were also taken from 27 control patients including those presenting with scoliosis, trauma and myelomas. All clinical samples were cultured by standard microbiological techniques, and microorganisms were identified by conventional methods. Ten-ml of blood was also taken from all patients for anti-lipid S IgG estimation by ELISA.⁴

Analysis of P.acnes isolates
10 strains of P.acnes isolated from disc material have been investigated for their antigenic composition by western blotting. Strains were grown in brain heart infusion broth (BHI) for 3 days under static, microaerophilic conditions. The culture medium was recovered by centrifugation and precipitated with 2 volumes of ethanol at -20û C. Preticipated material (i.e. potential exocellular antigen) was then subjected to electrophoresis (SDS-PAGE, 12% acrylamide gels) and either stained with coomassie blue to visualise proteins, or western blotted onto mitrocellulose membrane. Antigen profiles on the nitrocellulose were revealed by reacting with patient serum followed by protein A-peroxidase and chromogenic substrate. To investigate antibody response by a number of patients, ethanol-precipotated material from one strain on P.acnes was run in a single perparative lane, the nitrocellulose blot was cut into strips and each reacted with a different patient serum (strip blots).

Results
Patients with Sciatica

Spinal tissue was obtained from 207 patients undergoing microdiscectomy for severe unremitting sciatica. This tissue was examined for the presence of microorganisms. Seventy-six (37%) out of 207 patients yielded positive cultures. The microorganisms recovered from the microdiscectomy samples are shown in Table 1. Of these patients, 26 (34%) had elevated serum levels of anti-lipid S IgG.

One hundred and thirty one out of 207 patients had negative cultures and of these 111 (85%) had corresponding negative lipid S titres.

Table 1
Microorganisms recovered from microdiscectomy samples taken from patients presenting with sciatica
Microorganisms	Number (% of isolates)
Propioninbacterium acnes (pure growth)	49 (64%)
Propionibecterium granulosum	1 (1%)
Propionibacterium acnes/coagulase negative staphylococci	8 (11%)
Coagulase negative staphylococci	11 (14%)
Corynebacterium propinquum	1 (1%)
Micrococcus species	1 (1%)
Heavily mixed cultures	4 (5%)

Control patients
Microdiscectomy samples from 25 out of 27 (93%) of control patients yielded negative cultures. Propionibacterium acnes was recovered from the tissue samples of 2 out of 27 (7%) patients. Out of the control patients, 21 out of 27 (78%) had negative lipid S serology whilst 6 out of 27 (22%) had raised levels.

Statistical analysis of data
Statistical analysis of the data demonstrated the following:

• There is an extremely significant difference in positive serology between patients with positive culture and those with negative culture (p=0.019, Table 2).
• There is a significant difference between the positive cultures of patients with sciatica and controls (p=0.001, Table 3).
• There is a significant difference between the pure cultures of P.acnes from patients with sciatica and controls (p=0.029 Table 4).

Table 2
Positive serology between patients with positive culture and those with negative culture
	Positive serology	Negative serology	Total
Positive culture	26	50	76
Negative culture	20	111	131

Table 3
Patients with sciatica versus controls: total positive cultures
	Patients with sciatica	Controls
Positive cultures	76	2
Negative cultures	131	25

Table 4
Patients with sciatica versus controls: pure growth of P.acnes
	Patients with sciatica	Controls
Pure P.acnes	49	2
Negative culture	118	25
(not inclusive of other microorganisms)	(190-47 25)

Antigenic analysis of P.acnes isolates
Protein profiles (a Coomassie blue stained SDS-PAGE gel) and a representative western blot developed with a patient serum are shown in figure 1 (not reproduced in here) for 10 isolates of P.acnes. Figure 2 (not reproduced here) shows a strip blot for a single strain developed with serum samples from 15 sciatica patients. A prominent cluster of antigens of molecular weight around 29-36kDal is present in all the strains and detected by all patients studied to date. This antigenic material does not appear to contain protein since it does not stain with Coomassie blue. We are currently investigating its composition and properties as a potential market of P.acnes infection and its possible role as a mediator of inflammation and nerve root pain.

Conclusions
These results suggest that low virulent microorganisms, in particular, P.acnes may be associated with infection in the vertebral discs of patients with severe sciatica. This has been confirmed by direct culture of the disc material, the application of a novel serologiacal assay and by comparison to the results obtained for control patients. Results of preliminary western blotting investigations have identified a novel exocellular antigen of P.acnes, which may serve as a further potential serological marker of infection due to the microorganism.

Further work
We propose to follow several lines of investigation.

We will confirm the association between the presence of P.acnes and related microorganisms in disc tissues from sciatica patients using more fully characterised patient groups.

We will investigate the role of the microorganisms in this condition. This will include genotyping and phenotyping of the microorganisms recovered from clinical samples, and studying the potential pro-inflammatory factors and cell-matrix degrading enzymes produced by the bacteria.

The response of the disc to the presence of the microorganisms including host production of inflammatory mediators from disc cells will be investigated. An in vitro model to study the release of inflammatory mediators from disc cells will also be utilised to study local tissue responses to microorganisms.

Patient groups and clinical, microbiological and host-response investigations
The study will be a double blinded, case controlled, randomised investigation.

The groups of patients will be as follows:

1. 100 further patients with sciatica
2. 30 patients undergoing surgery for back pain without radiculitis
3. 50 control patients undergoing surgery for other reasons
4. 50 patients with severe acne without radiculitis

Pre-operative and operative findings (where applicable) will be recorded.

Microbiological investigations including culture of clinical material and estimations of serum levels of anti-lipid S IgG will be undertaken on all patient groups.

Microorganisms recovered from clinical samples will be characterised by phenotypic and genotypic methods.

Tissue samples will be subjected to PCR analysis, in situ hybridisation and immunohistochemistry to reveal microorganisms, which cannot be cultured directly, and to reveal their precise location within the disc.

An investigation to assess the vascularisation of the disc tissue will be undertaken to determine whether or not macrophages infiltrate and dominate the cellular response to microbial mediators.

An investigation to assess the association between microorganisms and levels of metalloproteinases, cytokines and prostaglandins will be undertaken. Furthermore, the ability of microorganisms to stimulate the release of inflammatory mediators will be determined.

Results of the study will be presented at national and international orthopaedic, microbiology and cell biology meetings. Results will be published in medical and scientific journals.

References

1. Office of Population Census and Surveys (OPCS). Morbidity statistics from general practice, 4th national study. Royal College of General Practitioners, OPCS, Department of Health, London. HMSO, 1995.
2. Tunney MM, Patrick S, Gorman SP, Nixon JR, Anderson N, Davis RI, Hnna D, Ramage G. Improved detection of infection in hip replacements. Journal of Bone and Joint Surgery [BR] 1998;80(4): 568-572
3. Stirling A, Worthington T, Rafiq M, Lambert PA, Elliott TSJ. Association between sciatica and Propionibacterium acnes. Lancet 2001;357:2024-2025
4. Worthington T, Lambert PA, Traube A, Elliott TSJ. A rapid ELISA for the diagnosis of intravascular catheter-related sepsis due to coagulase negative staphylococci. Accepted for publication Journal of Clinical Pathology 2001.