Arthropod-borne diseases are the most common zoonoses involving wildlife in the northern hemisphere, especially the Old World [1]. Borrelia burgdorferi, the causal agent of Lyme borreliosis (LB), is found worldwide and comprises at least 18 genospecies [2,3]. Discovered in 1982, B. burgdorferi is now considered an emerging (and possibly under-reported) pathogen that could become an important public health problem [4]. Ixodes ricinus is the tick that usually transmits this pathogen to humans in Europe [5].

In recent years there has been a substantial increase in interest in B. burgdorferi infection, and different epidemiological studies have been made in different parts of Spain [6,7,8,9]. Clinical cases have also been reported [10,11,12,13]. Thus far, the prevalence of LB has been studied in high-risk groups, in the general population, and in patients showing possible signs of infection [12,14,15,16,17]. Ongoing studies at the local level are required to construct more efficient risk prediction models. In response to this need the present work reports a 17-year epidemiological surveillance study of B. burgdorferi infection in two areas of northern Spain where this pathogen is known to exist.

Specific B. burgdorferi antibodies were detected in 83 (13.3%) patients (Figure 3); 43 males (12.46%) and 40 females (14.38%).

Seven positive sera came from people who lived in the Burgos region (31.8%) and 76 from people in the Province of Palencia (12.6%). All seropositive patients had contact with animals, and all age groups were represented (range 3–87 years). Table 1 shows there were no significant differences among age groups in terms of seropositivity.

Table 2 shows there were no differences in seroprevalence in terms of patient occupation. No relationship was seen between seropositivity and province of residence. Most patients had been bitten in spring (315 people [50.56%]) or autumn (119 people [21.18%]).

After examining all patient sera, i.e., including sera taken from patients up to three times, 49 (7.8%) were found to be IgM positive and 60 (9.6%) IgG positive; 27 patients (4.33%) had both IgM and IgG antibodies (Table 3 shows the distribution by years).

In those patients for whom more than one serum sample was available (n = 498), 8 (1.61%) showed rising IgG titres, five showed rising IgM titres (1.00%), and 4 (0.80%) showed increases in both IgG and IgM. Retro-seroconversion with respect to IgG occurred in six people (1.20%), and with respect to IgM in one person (0.20%). In both cases, seroconversion and retroconversion, occurred during the first years of the study, in particular between 1997 and 2004. Some patients remained completely asymptomatic while others showed influenza-like symptoms (fever, fatigue, headaches and joint pain).

Five species of ticks (17 ticks) were identified to have bitten the seropositive patients: Dermacentor marginatus (41.5%; 14.2% males and 85.71% females), Dermacentor reticulatus (11.76%; 50% males and 50% females), Rhiphicephalus sanguineus (17.64%; 66.6% males and 33.3% females), Rhiphicephalus turanicus (5.88%; 100% males) and I. ricinus (23.52%; 25% males and 75% females).

IgM antibodies expressed against B. burgdorferi were found in four patients who had been bitten by D. marginatus, in two patients bitten by D. reticulatus, in two patients bitten by R. sanguineus, and in one bitten by I. ricinus. A male of D. reticulatus was infected with B. burgdorferi. The tick was removed from the scalp of a 13-year-old woman in May 1997. The patient presented a week after the bite the presence of regional lymph nodes close to the area of the bite without fever. The patient had high titer of IgM and IgG antibodies against B. burgdorferi, but not seroconversion.

Though Lyme disease is thought to have been present in Spain since 1977, its existence was not serologically confirmed until 1987 [10,21,22]. Since then, B. burgdorferi infections have been frequently recorded, with differences in prevalence from area to area, and its geographical distribution may be increasing as a result of climate change, changes in land uses, and evolving socioeconomic factors [10,23]. The present work is the longest-term (17 years) epidemiological surveillance study of B. burgdorferi ever performed in Spain. The seroprevalence of 13.3% detected in the present work falls within the range previously reported for Spain (0.3–14.8%) and other European countries (1.8%–19.68%) [24,25,26,27,28,29].

Ticks are abundant in rural areas, and it is reported that those persons whose work implied contact with animals in the rural setting are at higher risk of infection [30,31,32,33]. However, the prevalence of B. burgdorferi infection in high risk populations is not well known and seems to vary widely from area to area. In the present study, no significant differences in seroprevalence were seen between different occupation groups and by age, but others authors report studies that shown significant differences in seroprevalence data respect to occupation and/or age [32,34]. In the present work, similar seropositivity results were seen in different seasons, suggesting that bites can occur throughout the year.

B. burgdorferi antibodies are developed relatively late in some patients, and indeed never in others. However, since several serum samples were available from some patients, a few cases of acute (subclinical or clinical) infection (presence of IgM and rising or falling IgG antibody titres) were clearly identified. The small number detected may be explained in that all patients presenting with fever symptoms were treated with antibiotics in accordance with the policy for the prevention and control of tick-borne diseases designed by the Servicio Territorial de Sanidad y Bienestar Social de la Junta de Castilla y León (the regional health service). Thus the true number of infections might have been masked by this early treatment [35]. Indeed, some authors have now clearly indicated that the early administration of antibiotics can postpone or inhibit the development of a response to the pathogen [36,37]. The present results may support these findings. Though some authors indicate that prophylactic antibiotic treatment should be administered after a tick bite others suggest it is not indicated given the low incidence of eventual disease [13,38,39]. Nevertheless, since the transmission of B. burgdorferi never occurs within the first 24 h of tick attachment (indeed, 48 h may be required), removing ticks as soon as possible is a good prevention method. Sometimes bites from several ticks are needed before antibodies to B. burgdorferi become detectable.

In Europe, B. burgdorferi is usually transmitted to humans via the tick I. ricinus, and mainly by its nymphs [2,3,40,41,42,43]. These are more active from spring to autumn and need 85% humidity in order to thrive. Tick species other than I. ricinus were, however, identified on the patients in the present work, including D. marginatus, D. reticulatus, R. sanguineus and R. turanicus.

PCR detected B. burgdorferi DNA in D. reticulatus males. Dermacentor spp. play an important role in the transmission of tick-borne pathogens that is not well recognized in Europe [44,45]. The latter authors reported the first epidemiological characterizations of pathogens spread via Dermacentor spp. in France. In the present work, B. burgdorferi DNA was detected in a D. reticulatus male. Other authors suggest D. reticulatus to have expanded its area of distribution [46]. Certainly, its growing presence in Serbia was reported [47]. Further experiments are needed to determine to the vector competency of this pathogen.

In summary, it is important to increase awareness of B. burgdorferi infection, and that better surveillance should be implemented. Further studies are planned, including studies using more antigens (like other serotypes and species of the B. burgorferi sensu stricto, B. afzelii, B. garinii, to avoid the loss of positive samples), more epidemiological studies and analyses of suspected clinical cases, to try to determine the true importance of B. burgdorferi infection.