HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Evans, R. Articles by Ho-Yen, D. O Search for Related Content PubMed PubMed Citation Articles by Evans, R. Articles by Ho-Yen, D. O Agricola Articles by Evans, R. Articles by Ho-Yen, D. O

Audit of the laboratory diagnosis of Lyme disease in Scotland

Microbiology Department, NHS Highland, Raigmore Hospital, Inverness IV2 3UJ, UK

Correspondence Roger Evans roger.evans{at}haht.scot.nhs.uk

Received 31 December 2004
Accepted 22 August 2005

An audit was performed on the laboratory diagnosis of Lyme disease in Scotland. The problem of a significant number of patients with clinical symptoms of Lyme disease being reported as seronegative or equivocal by the confirmatory Western blot test was identified. Comparisons of current practice were made with American and European standards, and the Western blot scoring system revised. When applied retrospectively (April 2003 to March 2004), 39 (33 %) of 116 serum samples previously negative or equivocal became weak positive or stronger. Thirty-one (80 %) of these 39 samples were from patients with clinical details suggestive of early Lyme disease. The changes were implemented and assessed prospectively for 6 months. There was a significant increase in the proportion of equivocal results, with fewer negatives compared to the same time period 1 year previously. This audit has helped clinicians in the diagnosis of Lyme disease and the management of these patients in Scotland.

	INTRODUCTION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Lyme disease, caused by the spirochaete Borrelia burgdorferi sensu lato is endemic in Scotland (Ho-Yen et al., 1990). The Scottish Lyme disease testing service laboratory, Raigmore Hospital, Inverness, tests approximately 2000 samples each year from throughout Scotland and has seen a significant increase in the number of referrals over the last 10 years. This is mainly due to a greater awareness of Lyme disease among clinicians and the general public (Davidson et al., 2003). It is recognized with early Lyme disease that the diagnosis can be made on clinical grounds but the diagnostic rash, erythema migrans, is sometimes not present or is not recognized and so the diagnosis is dependent upon laboratory tests (Pfister et al., 1994). In late-stage disease the laboratory diagnosis is more important as the clinical spectrum of disease is wide.

The Centers for Disease Control and Prevention guidelines require that the laboratory diagnosis of Lyme disease is a two-step process: a screening enzyme immunoassay then a confirmatory Western blot test (Centers for Disease Control and Prevention, 1995; Wilske, 2003). Although there have been recommendations for the interpretation of band patterns of Western blots in laboratories from Europe and the USA there is no consensus (Robertson et al., 2000). It was recognized that individual laboratories would have to adapt the interpretation of their blotting patterns to the characteristics of Lyme disease in their local area (Robertson et al., 2000). Our laboratory has published its interpretative criteria for the Western blot in Scotland (Davidson et al., 1996). However, it had been noted over the last 2 years that some patients with symptoms and signs characteristic of early Lyme disease produced negative or equivocal serological results. Thus, it was decided to audit the scoring system for the Western blot.

	METHODS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
The in-house Western blot method and the original criteria for routinely scoring Borrelia burgdorferi sensu stricto IgG Western blot results were those published by Davidson et al. (1996). A slight modification has been used since publication; a further category, weak positive, has been included. A weak positive has the same band pattern as a positive score, but the intensity of the bands is weaker (level 1 rather than levels 2–4). For this audit, a comparison of the current practice was made with American and European standards (Dressler et al., 1993; Engstrom et al., 1995; Hauser et al., 1997; Robertson et al., 2000; Wilske, 2003) and a consensus revised scoring system agreed (Table 1). It was implemented at the Scottish Lyme disease testing service laboratory, Raigmore Hospital, Inverness, in April 2004.

Samples that were screening enzyme immunoassay (EIA) positive or equivocal or EIA negative but were from patients with clinical erythema chronicum migrans were tested by the in-house Western blot. Results from these samples tested at the laboratory by the Western blot for B. burgdorferi sensu stricto IgG antibodies from 1 April 2003 to 30 September 2004 were analysed. The revised Western blot scoring system was applied retrospectively to results from April 2003 to March 2004, and the results were compared with those originally reported. The clinical details of patients with samples that would have been assigned to a different result category using the new scoring system were tabulated. The number of samples with blots demonstrating the 58 kDa band was stated for each discrepant result group. A comparison was made prospectively of the number of samples in each result category for the same 6 month period (April–September) in the years preceding and following implementation of the revised scoring system (2003 and 2004, respectively). Statistical analysis was carried out when appropriate.

	RESULTS AND DISCUSSION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Although it is known that the sensitivity of the Western blot is lower in patients with early- compared to late-stage Lyme disease (Hauser et al., 1997), there was concern about the number of seronegative and equivocal results being reported on samples from patients with clinical, early Lyme disease symptoms. The major change to the revised scoring system was the inclusion of the 58 kDa band as one of the specific bands. This band was present in 64 % of the original positive results. All of the samples that were either negative or equivocal by the original scoring system and were now weak positive or positive, respectively, had the 58 kDa present at an intensity of 2–3 (Table 2). When the original scoring system was published, the banding patterns of the Western blot results had similarities to those reported from Europe and the USA (Davidson et al., 1996). However, there were significant differences, suggestive of strain variation in the Scottish Highlands and the rest of Europe. Bands in the range of 55–58 kDa were detected but were not included in the original scoring system (Davidson et al., 1996). Since then, this band has become recognized as one of the highly specific antigens and is often used in the interpretative criteria of other European laboratories (Hauser et al., 1997; Robertson et al., 2000; Wilske, 2003).

When the revised scoring system was applied retrospectively (April 2003 to March 2004), 39 (33 %) of the 116 serum samples previously negative or equivocal became weak positive or stronger (Table 3). Thirty-one (80 %) of these 39 samples were from patients with clinical details suggestive of early Lyme disease (erythema migrans, rash, tick bite) (Table 2). This was significantly more than any other clinical detail (P < 0.001, chi-square test) (Table 2) and confirms the specificity of the revised scoring system. A definitive result is useful for the clinician as antibiotic therapy is indicated and can be started early. Treatment is effective and late-stage sequelae are uncommon (Steere, 2001; Davidson et al., 2003). If a diagnosis is not made, the patient may go on to develop the late-stage sequelae that are more problematic to manage and debilitating for the patient. Additionally the cost of managing late Lyme disease is estimated to be at least three times greater than that of early Lyme disease in Scotland (Joss et al., 2003).

The largest change in the retrospective study was the significant number of negative results that became equivocal (P < 0.01, chi-square test) (Table 3). By the revised scoring system, the proportion of patients with clinical, early Lyme disease in the equivocal group was lower than that in the positive group. However, an equivocal result is likely to prompt the clinician to take further action. This may be either to start treatment if clinically indicated or to send a further serum sample at a later date. Similar results were observed in the prospective study (Table 4). The results of a small number of samples (six) changed from weak positive to equivocal (Table 3). In these samples the change was due to the very low intensity of the specific bands. Again, as an equivocal result is likely to prompt further investigation the small decrease in definitive results was acceptable as true positive Lyme disease cases would be likely to show an increase in band intensity when a later sample is tested. Variation in the intensity of bands has been reported (Cutler et al., 1994; Davidson et al., 1996) and the revised scoring system has emphasized the number of bands rather than intensity as being the decisive factor.

These results emphasize the importance of the regular audit of the interpretation of the Western blot scoring system. The audit of the laboratory diagnosis of Lyme disease in Scotland has been very successful. The importance of interpreting Western blot results according to the characteristics of Lyme disease in the local area has been emphasized. More importantly, Scottish patients will be better managed with improved reporting of Lyme disease blots.

	ACKNOWLEDGEMENTS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
We are indebted to the clinicians, general practitioners and laboratory staff throughout Scotland who were involved in referring samples for testing. We would also like to thank Miss Lesley Dargie for her excellent technical expertise.

	REFERENCES

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES