Search

Objectives There is controversy whether non-specific symptoms can be related to previous Lyme borreliosis (LB). Positive serology can be considered a proxy for previous or persistent LB. We assessed non-specific symptoms and serology in patients suspected of LB referred to a Lyme centre. Methods Included were adult patients who visited a Lyme centre between 2008 and 2014. Before medical consultation, serum samples were taken and questionnaires on non-specific symptoms completed. The prevalence of non-specific symptoms was calculated for patients with positive and negative IgG serology. Logistic regression was used to obtain odds ratios (ORs) with 95% confidence interval (CI) for an association between positive serology and non-specific symptoms. Results Of 1439 included patients, 31.6% (455/1439) had positive serology. The most common non-specific symptoms were severe fatigue (61.4%, 883/1439), sleep disturbances (54.8%, 789/1439) and stiffness of neck/back (52.6%, 757/1439). The prevalence of severe fatigue was 53.0% (241/455) in patients with positive serology vs. 65.2% (642/984) in patients with negative serology (OR = 0.74; 95% CI, 0.58–0.94). The prevalence of sleep disturbances was respectively 46.2% (210/455) vs. 58.8% (579/984) (OR = 0.73; 95% CI, 0.58–0.93). The prevalence of stiffness of neck/back was respectively 47.7% (217/455) vs. 54.9% (540/984) (OR = 0.85; 95% CI, 0.67–1.06). Conclusions In patients referred to a Lyme centre, non-specific symptoms did not occur more frequently in patients with positive serology compared to patients with negative serology. Hence, a questionnaire on non-specific symptoms cannot be used for identifying patients with possible post-LB symptoms in clinical practice.

A two-step testing strategy is recommended in serological testing for Lyme borreliosis; positive and indeterminate enzyme-linked immunosorbent assay (ELISA) results are confirmed with immunoblots. Several ELISAs quantify the concentration of antibodies tested, however, no recommendation exists for an upper cut-off value at which an IgG ELISA is sufficient and the immunoblot can be omitted. The study objective was to determine at which IgG antibody level an immunoblot does not have any additional predictive value compared to ELISA results. Data of adult patients who visited a tertiary Lyme centre between 2008 and 2014 were analysed. Both an ELISA (Enzygnost Lyme link VlsE IgG) and immunoblot (recomLine blot Borrelia) were performed. Clinical data were extracted from the patient's digital medical record. Positive predictive values (PPVs) for either previous or active infection with Borrelia burgdorferi s.l. were calculated for different cut-off ELISA IgG antibody levels where the immunoblot was regarded as reference test. In total, 1454 patients were included. According to the two-step test strategy, 486 (33%), 69 (5%) and 899 (62%) patients had positive, indeterminate and negative Borrelia IgG serology, respectively. At IgG levels of 500 IU/ml and higher, all immunoblots were positive, resulting in a 100% PPV (95% CI: 97.0-100). At IgG levels of 200 IU/ml and higher, the PPV was 99.3% (95% CI: 97.4-99.8). In conclusion, at IgG levels of 200 IU/ml and higher, an ELISA was sufficient to detect antibodies to Borrelia burgdorferi s.l. At those IgG levels, a confirmatory immunoblot may be omitted in patients referred to a tertiary Lyme centre. Before these results can be implemented in routine diagnosis of Lyme borreliosis, confirmation of the results is necessary in other patient populations and using other quantitative ELISAs and immunoblots.