Patients admitted to emergency units in Tuscany, Italy, for tick removal were followed up for 40 days. Epidemiologic and clinical data were collected for each patient by using a standardized questionnaire. History of allergic reactions or hypersensitivity to tick bites was considered and evaluated to avoid mistakes in case definition.

Ticks were classified by using standard identification keys (8) and stored in 70% ethanol until DNA extraction. D. marginatus females were measured, and the degree of engorgement (tick engorgement index [TEI]) was visually estimated. Ticks were ranked by 3 TEI levels: 1 = completely unengorged, 2 = intermediate (idiosoma length ≈2× scutum width), and 3 = engorged (idiosoma length >2× scutum width). Association between TEI levels and occurrence of clinical symptoms was evaluated by using the Fisher exact test. All statistical analyses were conducted by using R statistical software (10). Arcview 3.3 (Environmental Systems Research Institute Inc., Redlands, CA, USA) was used to map the geographic distribution of cases in the study area (Figure 1).

For pathogen detection by PCR, ticks were individually homogenized with a pestle in microcentrifuge tubes and DNA was extracted with the DNeasy Blood and Tissue Kit (QIAGEN, Hilden, Germany). Negative controls (distilled water) were used to check for contamination of samples during this phase. Success of DNA extraction was verified by using PCR for tick mitochondrial 16S rDNA (11). Two PCR assays, targeting citrate synthase A (gtlA) and outer membrane protein A (ompA) genes, were used to identify spotted fever group rickettsiae as described (12).

Cases of abnormal tickbite reaction were observed only in patients bitten by D. marginatus (Table 1). From July 2005 through May 2007, information on 263 patients was recorded in the surveillance system. Removed ticks were classified as Ixodes ricinus (n = 187), Rhipicephalus sanguineus (n = 6), or D. marginatus (n = 17); 53 were unclassified (lost or disrupted).

Five patients (29.4%) (2 adults and 3 children) showed clinical signs typically related to tick-borne lymphadenopathy. We defined tick-borne lymphadenopathy case-patients as those with skin lesion (eschar) at the tick bite site and regional lymphadenopathy (2). The 5 patients were examined by physicians at least twice. Each patient had enlarged lymph nodes (Figure 2 , panel A) at the second examination. Crusted scalp lesions (Figure 2, panel B) ranged in diameter from 8 mm to 25 mm. One patient was surgically treated to remove necrotic tissue from a large tache noire; he showed alopecia >8 months after acute episode (Figure 2, panel C). Three cases were recorded in spring, 1 in autumn, and 1 in winter (Table 2). Association between Dermacentor TEIs and occurrence of symptoms was not statistically significant (p = 0.13). Six (35.3%) of 17 ticks were positive by PCR for either gtlA or ompA. The ompA gene sequences of all positive samples showed similarity of 100% with the R. slovaca (GenBank accession no. U43808).

Until recently, Mediterranean spotted fever (MSF) caused by R. conorii and transmitted by the brown dog tick, R. sanguineus, was considered the only tick-borne rickettsiosis in Italy. Local investigations at ASL 2 showed a decrease in MSF incidence during the past 10 years and fewer Rhipicephalus spp. bites than Ixodes spp. and Dermacentor spp. bites. In the past 2 years, no MSF was officially recorded at ASL 2, and our results suggest an emerging role of R. slovaca as a tick-borne pathogen in the area.

Most bitten patients showed specific clinical manifestations (fever, itching, rash, weariness, and myalgia) and 5 (29.4%) had typical signs of tick-borne lymphadenopathy. Pathogen identification in ticks agreed with the case definition in 50% of cases. Infected ticks were removed from 3 patients not considered to have lymphadenopathy; 1 patient showed no symptoms, probably because the tick was not attached long enough to enable pathogen transmission (TEI = 1).

A study reported that children and women have a higher risk than men for infection caused by R. slovaca (4); all of our patients with tick-borne lymphadenopathy were children or women. This result reflects the higher risk for bite by Dermacentor spp.; 6 (35.3%) of the 17 patients were <10 years of age and 10 (58.8%) were female.

Raoult et al. reported that all cases of tick-borne lymphadenopathy cases recorded in their study were found from October through May (4), confirming the seasonal incidence of tick-borne lymphadenopathy in cold months (2). The activity of D. marginatus adults in cold months has been reported (7,8). Our results confirm this trend, although the absence of cases of tick-borne lymphadenopathy in winter can be explained by the lower human frequency of high-risk areas (Table 2).

Cases reported in this study were concentrated in the northern part of ASL 2, despite the lower population density (Figure 1). Local investigations showed that wild boars, which seem to be important in the epidemiology of R. slovaca (13), are found in this area. In the northern part of this area, woodland habitat and human habits increase the risk for human contact with vectors. TEI did not show any association with cases but study of feeding duration is so far strictly applied to the transmission of Borrelia burgdorferi s.l. by I. ricinus (14).

In recent years, many efforts have been made to characterize distinct tick-borne diseases. A diagnostic approach that includes surveillance of patient symptoms and vectors can be helpful in identifying cases of tick-borne lymphadenopathy (15). Tick identification is also important in diagnosis (4).

Our data indicate a high prevalence of R. slovaca in D. marginatus collected from patients. On the basis of these results, all patients bitten by D. marginatus should be observed to determine whether specific treatments are required. The new surveillance system in Lucca will provide real-time data that will be useful for evaluation of patient health problems.