Lyme disease and erythema migrans

Meet the patient

A 34-year-old weekend hiker comes back from the woods of southern Connecticut — or, in the European mirror image, a week of camping in Slovenia — with a red patch on her thigh she first blamed on a bruise. Over ten days it has grown into a smooth, ring-shaped plaque now 12 cm across, painless, with a faint flu-ish ache. She does not remember a tick. The registrar reaches for a Lyme serology and a cellulitis swab.[1]

Hold that picture. The question that decides her next two weeks is the one this whole topic turns on: is this erythema migrans — and if it might be, do you treat now or wait for the blood test? The answer is the single most repeated fellowship viva answer in tick-borne disease, and every guideline gives the same reply: erythema migrans is a clinical diagnosis; treat it, do not serologise it.[1]

One tick, one spirochaete, three faces on the skin

Lyme borreliosis is one infection wearing three cutaneous masks at three different times. The agent is a microaerophilic spirochaete of the Borrelia burgdorferi sensu lato complex; the vector is a hard-bodied Ixodes tick; and the skin is involved at every stage — first as the expanding erythema migrans of early-localised disease, then as the bluish-red nodule of borrelial lymphocytoma in early dissemination, and finally as the atrophic, cigarette-paper dermis of acrodermatitis chronica atrophicans in late disease. The rash is so often the first and only clue that the dermatologist and the general physician sit on the front line of diagnosis.[1]

The disease takes its name from Lyme, Connecticut, where an inexplicable cluster of juvenile inflammatory arthritis was investigated in 1975; a few years later the Ixodes tick was implicated, and in 1982 Willy Burgdorfer isolated the spirochaete that now carries his name. Viva gold, one line: originally described as an arthropathy, Lyme is now understood as a staged infection of skin, nervous system, heart, and joints — the rash entered the story late but leads the disease.[1][11]

The three genospecies — and what each one likes to attack

The B. burgdorferi sensu lato complex holds more than twenty genospecies, but only three cause human disease, and they differ in geography and in the organs they favour. This is why European Lyme (rich in afzelii and garinii) looks different from North American Lyme (almost exclusively burgdorferi sensu stricto).[14]

[11]

The one-line discriminator beneath the table: genospecies predicts the organ. Skin disease means afzelii, nerve disease means garinii, joint disease means burgdorferi sensu stricto — a single sentence that earns marks on every European viva.[14]

The three cutaneous manifestations — the dermatology exam answer

The three skin lesions are themselves a high-yield classification, and the fellowship examiner wants the stage and the morphology of each:[1]

[1]

Borrelial lymphocytoma is the rare one to be able to describe: a firm, painless, bluish-red nodule — classically on the earlobe of a European child, or the nipple, areola, or scrotum of an adult — that is a dense polyclonal lymphocytic infiltrate of the dermis with germinal centres. The clinical question it poses is "is this a lymphoma?", and the clue is the site plus endemic exposure.[11]

Epidemiology — where the ticks are

Lyme is reported across the temperate belt of the Northern Hemisphere, and incidence tracks the tick. The United States records roughly 300,000 to 476,000 diagnoses a year (a CDC estimate counting clinically diagnosed and surveillance cases), concentrated in the northeastern and upper-midwestern states and parts of the Pacific coast. Europe reports around 65,000 to 230,000 cases a year, with the highest incidence in central Europe — Slovenia, Austria, southern Germany, Czechia — and Scandinavia and the Baltic states, extending across Russia and temperate Asia.[5]

[5]

The vectors are hard-bodied Ixodes ticks, and the species fixes the map: I. scapularis (the blacklegged or deer tick) in the northeastern and midwestern US, I. pacificus on the North American Pacific coast, I. ricinus (the castor bean or sheep tick) across Europe, and I. persulcatus (the taiga tick) in Russia and Asia. The white-footed mouse, chipmunks, shrews, and ground-feeding birds are the reservoir — the animals on which larval and nymphal ticks acquire the spirochaete. The white-tailed deer is essential to the tick's reproductive cycle but is not a competent reservoir: it clears Borrelia, so it feeds the tick without maintaining the infection. The classic trap here is "deer reservoir" — wrong; the mouse is the reservoir, the deer is the host.[11]

The 36-to-48-hour rule — why prompt tick removal actually works

The single most important transmission variable is how long the tick has been attached. The spirochaete lives dormant in the tick midgut, tethered there by outer-surface protein OspA. During the blood meal it must downregulate OspA, upregulate OspC, multiply, and migrate to the salivary glands before it can be inoculated — a sequence that takes roughly 36 to 48 hours. Attachment of less than 36 hours rarely transmits, which is the entire biological justification for prompt tick removal and for the narrow window of post-exposure prophylaxis.[1]

[14]

This is also why nymphs cause most human infections. The nymph is pinhead-sized, feeds unnoticed, and so stays attached long enough to deliver the spirochaete; the adult tick is larger, found sooner, and removed before it can transmit. The practical teaching: a recognised, promptly removed tick is a low-risk event; an unrecognised nymph that fed for days is the dangerous one.[1]

Pathophysiology — antigenic switching is how the spirochaete persists

Borrelia burgdorferi is a motile spirochaete with a small genome built around an unusual linear chromosome plus numerous linear and circular plasmids that carry its virulence repertoire — chief among them the outer-surface lipoproteins OspA through OspF and the VlsE antigenic-variation system. It cycles between tick (OspA) and mammal (OspC, VlsE), and that antigenic wardrobe is how it persists in two such different hosts — and partly why the antibody response lags behind the rash.[14]

After inoculation the spirochaete spreads locally outward through the dermis — the slowly expanding ring of erythema migrans — and then haematogenously and lymphatically to distant skin (the multiple lesions of early-disseminated disease), heart, joints, and nervous system. The damage is driven less by a toxin than by the host response: lipoproteins such as OspC engage Toll-like receptor 2, releasing TNF-alpha, IL-1, IL-6, and interferon-gamma. In Lyme arthritis, residual spirochaetal antigens in a genetically susceptible host (HLA-DRB1) drive a T-helper-1 and T-helper-17 autoimmune synovitis that persists after the organism is cleared — the basis of antibiotic-refractory arthritis.[1][14]

The three stages — the spine of the topic

Lyme unfolds in three overlapping stages, mirroring the spirochaete's expansion from skin to bloodstream to deep tissue. Each stage carries a defining syndrome, and examiners test the tempo, morphology, and organ tropism of each.[1]

[1] [1] [1]

Stage 1 — the 5-cm rule, and why size excludes the bite reaction

Erythema migrans is the pathognomonic early lesion and the presenting feature in roughly 70 to 80 per cent of patients. It begins, 3 to 30 days after the bite (median 7 to 14), as an erythematous macule or papule at the inoculation site that expands centrifugally over days to weeks into an annular patch or plaque that must measure at least 5 cm across. The size cut-off is the whole point: a banal tick-bite hypersensitivity reaction stays small, and the 5-cm line is what separates a reaction from an infection.[1][2]

The rash is often joined by a mild flu-like prodrome — low-grade fever, fatigue, headache, myalgia, arthralgia, and regional lymphadenopathy. Frank high fever is uncommon and should make you hunt for a co-infection (anaplasmosis or babesiosis from the same tick) or a different diagnosis. Some patients have no constitutional symptoms at all and notice only the rash.[1]

Face-off — the expanding annular erythema

The differential of an expanding annular erythema is long, and examiners want the one discriminator for each, not a bare list.[1]

[1] [1]

The practical discriminator triad for erythema migrans is centripetal expansion over days to weeks, a maximum diameter of at least 5 cm, and compatible tick exposure in an endemic area. Two absences do not exclude it: most patients do not recall the bite, and most lesions are not targets.[1]

Stage 2 — when the skin seeds, the nerves and heart follow

Untreated, the spirochaete disseminates haematogenously within weeks to a few months, and three organ systems bear the brunt: skin, nerve, and heart.[1]

Multiple erythema migrans is the cutaneous fingerprint of dissemination — secondary annular lesions at sites remote from the original bite, often smaller and more numerous than the primary lesion. This is not multiple tick bites; it is bloodstream seeding of the skin, and it is one of the few rashes that on its own signals bacteraemia.[1]

Neuroborreliosis — think Lyme when the facial palsy is bilateral

Neuroborreliosis occurs in roughly 10 to 15 per cent of untreated patients, and its three classic European-described faces are cranial neuropathy, lymphocytic meningitis, and painful radiculoneuritis. The cranial nerve most often hit is the facial nerve, and the single discriminator from idiopathic Bell palsy is that Lyme facial palsy may be bilateral. Bannwarth syndrome — severe radicular pain, sensory and motor deficits, and CSF lymphocytosis — is the commonest neuroborreliosis pattern in Europe; less common are encephalitis, myelitis, and mononeuritis multiplex.[7]

The bedside teaching to carry: bilateral facial palsy in an endemic area is Lyme until proven otherwise, with sarcoidosis and Guillain-Barre the other two on the differential.[7]

Lyme carditis — the AV block you must not miss

Lyme carditis occurs in roughly 1 to 5 per cent of untreated patients and is dominated by fluctuating atrioventricular block of any degree, from first-degree to complete heart block, which can cause palpitations, dyspnoea, chest pain, presyncope, syncope, and rarely sudden death. Myopericarditis may coexist. The lesion is reversible — but only if you find it before the conduction tissue arrests.[4][8]

[4]

Stage 3 — the knee that comes and goes, and cigarette-paper skin

Late Lyme disease is joints and skin. Two syndromes carry the late stage, and each has a signature the examiner rewards.[1]

Lyme arthritis is the commonest late manifestation — especially with B. burgdorferi sensu stricto in North America — and the picture is unmistakable: an intermittent, mono- or oligo-articular arthritis, most often of a single knee, with a large effusion and comparatively little pain, recurring over months to years. Roughly 60 per cent of untreated North American patients develop it. The trap within the trap: about 10 per cent progress to antibiotic-refractory arthritis — an autoimmune, HLA-DRB1-associated synovitis that persists after the organism is cleared and is managed with NSAIDs and DMARDs, not more antibiotics.[6]

Acrodermatitis chronica atrophicans (ACA) is the late cutaneous manifestation, almost exclusively European and Asian and driven by B. afzelii. It begins insidiously on extensor acral skin — dorsal hands and feet, lower legs, forearms — as an oedematous bluish-red plaque that, over years, evolves into striking atrophy: thin, dry, wrinkled, translucent cigarette-paper or cellophane skin through which veins are visible, with fibrous nodules over joints. It is chronic and slowly progressive but responds to oral antibiotics even years into its course — so a limb scarred by untreated ACA is a preventable harm.[12]

Chronic neuroborreliosis is rare and contested — a slowly progressive encephalomyelitis, peripheral neuropathy (often alongside ACA), or a mild cognitive syndrome. Distinguishing it from post-treatment symptoms needs objective neurological signs and CSF evidence of intrathecal antibody synthesis; a positive serum serology alone is not enough.[1]

Clinical diagnosis first, two-tier serology later — the central principle

[1]

The classic trap: waiting for the serology in early erythema migrans. Antibodies take four to six weeks; the rash appears in one to two. A negative early test is expected, not reassuring, and ordering it only delays treatment. The strategy is therefore stage-dependent:[2]

[4]

Two-tier serology is the standard for any non-EM presentation. The traditional CDC algorithm is a sensitive first-line enzyme immunoassay (ELISA or EIA), often using the C6 peptide of VlsE; if positive or equivocal it is confirmed by a Western immunoblot — IgM (2 of 3 bands, useful in the first month) and IgG (5 of 10 bands, the mainstay for later disease). The two-step design trades a little sensitivity for a lot of specificity, blocking false positives from cross-reacting antibodies (other spirochaetes, autoimmunity, Epstein-Barr).[2][4]

The modified two-tier testing algorithm (MTTT), now endorsed by the CDC, replaces the blot with a second, different EIA — two positive EIAs constitute a positive result. It is simpler, faster, and at least as sensitive in early disease. Whichever algorithm you use, interpret in context: antibodies persist for years after treated or even untreated infection and do not distinguish active from past disease, so a positive serology alone is never a reason to treat without compatible clinical findings.[4]

For suspected neuroborreliosis, perform a lumbar puncture: the CSF shows lymphocytic pleocytosis, elevated protein, and — most specifically — intrathecal synthesis of anti-Borrelia antibodies (a raised antibody index). A normal CSF white count argues against neuroborreliosis even with a positive serum serology. PCR of CSF is insensitive (around 30 per cent); PCR of synovial fluid, by contrast, is highly sensitive and confirms Lyme arthritis when the picture is atypical. Culture in Barbour-Stoenner-Kelly medium is possible from an early EM biopsy but slow, specialised, and not a routine tool — erythema migrans is diagnosed and treated without it.[7]

The antibiotic ladder — stage-specific, dose-certain

Treatment is stage-specific, and every regimen is stated as agent, dose, route, and duration. First-line oral therapy cures the great majority of early disease; parenteral ceftriaxone is reserved for neurological, high-grade cardiac, and refractory disease.[2][3]

[1]

Early-localised EM (and early-disseminated disease without neurologic or cardiac involvement)

Doxycycline 100 mg orally twice daily for 14 to 21 days is first-line in adults and children aged 8 years and older — favoured because it is convenient, covers co-infecting Anaplasma, and penetrates the CNS. The principal alternatives, used in pregnancy, young children, and doxycycline intolerance, are amoxicillin 500 mg three times daily and cefuroxime axetil 500 mg twice daily, each for 14 to 21 days. Azithromycin is a third-line macrolide and is less effective than the beta-lactams and tetracyclines.[2][4]

Neuroborreliosis

Ceftriaxone 2 g intravenously once daily for 14 to 28 days is first-line for meningitis, radiculopathy, and parenchymal CNS disease; alternatives are cefotaxime 2 g IV every 8 hours and penicillin G 18 to 24 million units IV daily in divided doses. For mild neuroborreliosis — an isolated facial palsy with a normal CSF — oral doxycycline 100 mg twice daily for 14 to 21 days is an accepted alternative (European guidelines favour it more than US), avoiding an admission and a line; a facial palsy with CSF pleocytosis is treated parenterally.[7][8]

Lyme carditis

Admit for telemetry (see above). First-degree AV block with a normal ventricular response and no symptoms can be managed with oral doxycycline or cefuroxime inpatient initially. Higher-degree or symptomatic block is treated with IV ceftriaxone 2 g daily for 14 to 28 days and a temporary pacemaker if symptomatic or high-grade; conduction usually recovers within a week, at which point oral step-down and discharge are reasonable.[4]

Lyme arthritis

Oral doxycycline 100 mg twice daily or amoxicillin 500 mg three times daily for 28 days is first-line. If arthritis persists, a second 28-day oral course or a switch to IV ceftriaxone 2 g daily for 14 to 28 days follows. About 10 per cent develop antibiotic-refractory arthritis — persistent synovitis months after adequate antibiotics with no evidence of persistent infection — an autoimmune, HLA-associated process managed with NSAIDs, intra-articular corticosteroid, and DMARDs (hydroxychloroquine, methotrexate, TNF inhibitors), and occasionally synovectomy or arthroplasty. The teaching point: more antibiotics do not help this group.[6]

Acrodermatitis chronica atrophicans and borrelial lymphocytoma

ACA responds to oral doxycycline 100 mg twice daily or amoxicillin 500 mg three times daily for 21 to 28 days; extensive disease can take IV ceftriaxone. Improvement is slow (months) and the atrophy is only partly reversible, so early treatment matters. Borrelial lymphocytoma is treated with the same oral regimens for 14 to 21 days.[12]

[1]

Prophylaxis — the single 200-mg doxycycline dose

Post-exposure prophylaxis after a tick bite is narrowly defined and deliberately rare. The classic regimen is a single dose of doxycycline 200 mg orally within 72 hours of removing an engorged, nymphal or adult Ixodes tick acquired in a highly endemic area — the Nadelman regimen, reducing the risk of erythema migrans by roughly 80 per cent.[2]

The 2020 IDSA, AAN, and ACR guidelines moved from routine prophylaxis to shared decision-making, because the baseline risk of infection after a recognised bite is low (around 1 to 3 per cent) and the evidence for universal prophylaxis is modest; the updated meta-analysis underpinning this shows a real but small absolute benefit. The alternative the guidelines endorse is watchful waiting — instruct the patient to seek care if an erythema migrans or a flu-like illness develops within 30 days. The European approach is similarly conservative: no routine prophylaxis, emphasis on prompt removal.[4][13]

The classic trap: prophylaxing every tick bite. Most recognised bites are low-risk; prophylaxis is for the engorged Ixodes nymph from a highly endemic area, removed late, offered by shared decision. For everyone else, teach tick checks and watchful waiting.[13]

Preventable harm — the three things that should never happen

[9]

Post-treatment Lyme disease syndrome — and the harms of treating it with more antibiotics

PTLDS is defined as subjective symptoms — fatigue, musculoskeletal pain, and cognitive complaints — persisting for at least six months after recommended antibiotic therapy, with no objective evidence of persistent infection. It affects roughly 10 to 20 per cent of treated patients, overlaps syndromically with chronic fatigue syndrome and fibromyalgia, and its mechanism is unknown — a post-infectious autoimmune or neuro-inflammatory state is the leading idea. What is not unknown is the evidence on treatment.[9]

Four large randomised, placebo-controlled trials — Klempner (2001), Krupp (2003), Fallon (2008), and Berende (2016) — gave 90 days of doxycycline with or without ceftriaxone and showed no benefit over placebo. Prolonged antibiotics do not help PTLDS, and they cause real harm: line infections, Clostridioides difficile colitis, antimicrobial resistance, allergic reactions, and the displacement of a treatable alternative diagnosis. The evidence-based management is rehabilitative and symptom-directed — explanation and reassurance, graded aerobic exercise, sleep and mood management, cognitive-behavioural therapy, and simple analgesia — with a safety net to return if objective new signs of disseminated disease appear.[4][9]

The contested label chronic Lyme disease attributes a wide range of non-specific symptoms to persistent Borrelia infection in the absence of compatible clinical or serological evidence, and is not accepted by mainstream infectious-disease, rheumatology, or neurology societies. The classic trap: treating PTLDS with months of IV antibiotics — prescribing prolonged intravenous therapy for a serology-positive but symptom-vague patient is the harm, not the cure.[9]

Special populations — pregnancy, children, the elderly

In pregnancy, avoid doxycycline (dental and bone effects on the fetus); use oral amoxicillin 500 mg three times daily or cefuroxime 500 mg twice daily for 14 to 21 days, with IV ceftriaxone for disseminated disease. Transplacental transmission is reported but rare, no consistent congenital syndrome is established, and promptly treated maternal Lyme does not harm the fetus; breastfeeding is not contraindicated once treated.[4]

In children, Lyme is common and the paediatric facial palsy is a classic. Doxycycline was historically avoided under 8 years for dental staining, but current AAP and IDSA guidance permits short-course doxycycline at any age because the staining risk of brief courses is small; when feasible, amoxicillin 50 mg/kg/day in three divided doses (maximum 500 mg per dose) or cefuroxime is the alternative. Neuroborreliosis and carditis take weight-adjusted IV ceftriaxone 50 to 75 mg/kg/day.[4]

In the elderly, presentations are subtler: an unexplained high-grade AV block, or an unexplained mono-articular knee effusion, in an endemic area should prompt the Lyme question, and the atrophic skin of ACA may be mistaken for chronic venous disease or steroid atrophy.[1]

Prevention — a layered strategy

Prevention stacks personal protection, prompt tick removal, targeted prophylaxis, and surveillance.[2]

• Tick avoidance in endemic habitat: long sleeves and trousers tucked into socks, light-coloured clothing, staying on cleared trails, and treating clothing and gear with permethrin.
• Repellents: DEET 20 to 30 per cent on skin (safe in pregnancy and in children over two months at lower concentrations), picaridin, or oil of lemon eucalyptus (not in young children).
• Tick checks: a full-body check — scalp, ears, axillae, groin, popliteal fossae, navel — after outdoor activity, and showering within two hours of coming indoors, which washes off unattached ticks.
• Prompt, correct tick removal: grasp close to the skin with fine-tipped tweezers, pull straight up with steady pressure, clean the site. Do not squeeze, twist, smother with petroleum jelly, or burn — these can make the tick regurgitate. The folk methods (match, Vaseline, alcohol) are ineffective and harmful.
• Targeted prophylaxis: single-dose doxycycline 200 mg within 72 hours for a high-risk engorged Ixodes bite in a highly endemic area, by shared decision.
• Vaccination: VLA15, a multivalent OspA vaccine, completed phase 3 enrolment; if licensed it will be the first human Lyme vaccine in two decades.[1]

[2]

Prognosis — early-treated Lyme is cured Lyme

A 14-to-21-day course of oral doxycycline or amoxicillin cures the overwhelming majority of early erythema migrans, with resolution of the rash and constitutional symptoms within days to weeks and no progression to disseminated disease. Adequately treated arthritis and neuroborreliosis likewise recover fully in most.[3][10]

Acrodermatitis chronica atrophicans progresses slowly over years if untreated, leaving permanent atrophy and fibrosis; treated even years in, it stabilises and partly resolves. Lyme arthritis has a good prognosis with oral therapy, but the antibiotic-refractory subset can develop chronic synovitis and joint damage needing DMARDs and occasionally arthroplasty. Disposition for uncomplicated EM is outpatient oral therapy with safety-net advice to return for facial weakness, palpitations, syncope, or joint swelling; carditis and neuroborreliosis are admitted.[1]

The Jarisch-Herxheimer reaction — transient fever, chills, myalgia, and rash within hours of the first antibiotic dose, from liberated spirochaetal antigens — can occur in early-disseminated disease, is self-limited over 24 hours, and needs only antipyretics; forewarn the patient and do not stop or change the antibiotic.[1]

Guidelines and regional differences

The 2020 IDSA, AAN, and ACR guidelines are the North American standard and have moved toward shared decision-making for prophylaxis, doxycycline first-line for EM (including short courses across age groups), two-tier — now modified two-tier — serology for non-EM presentations, IV ceftriaxone for neuroborreliosis and significant carditis, and an explicit rejection of prolonged antibiotics for PTLDS. The European landscape — EFNS or EAN neuroborreliosis criteria, EUCALB, and UK NICE NG95 — shares most recommendations but leans harder on clinical diagnosis of EM (treat without serology), accepts oral doxycycline for mild neuroborreliosis including isolated facial palsy, gives no routine prophylaxis, and recognises ACA and Bannwarth (B. afzelii and garinii) more readily.[4][8]

[1] [1]

A recombinant OspA vaccine (LYMErix) was licensed in the US in 1998, was about 76 per cent effective, and was withdrawn in 2002 amid low uptake, litigation, and unsubstantiated autoimmune-arthritis concerns — leaving the US without a human Lyme vaccine for two decades. A new generation is now in trials: VLA15 (Pfizer and Valneva), a multivalent OspA vaccine targeting the six commonest Borrelia OspA serotypes, completed phase 3 enrolment, and an mRNA OspA vaccine is in early development. A canine Lyme vaccine is widely used but does not protect humans.[1]

The mantra

EM is clinical; serology is for late disease; a single 200-mg doxycycline dose covers a high-risk bite. Carry that one line, and the antibiotic ladder, the 36-hour rule, and the bilateral-facial-palsy discriminator all fall into place.[1]

Ward-round test — four stems, thirty seconds each

Stem 1 — bilateral facial palsy in a hiker (answer)

A 38-year-old returns from a week camping in Slovenia with sequential bilateral lower-motor-neuron facial palsy over four days, a mild headache, and no rash. What is the most likely diagnosis, the confirmatory test, and the first-line treatment? Model: Lyme neuroborreliosis is the leading diagnosis — bilateral facial palsy in an endemic area is Lyme until proven otherwise (sarcoidosis and Guillain-Barre are the differential). Confirm with two-tier serology and a lumbar puncture looking for CSF lymphocytic pleocytosis and a raised anti-Borrelia antibody index; treat with IV ceftriaxone 2 g once daily for 14 to 28 days. Oral doxycycline is acceptable for an isolated facial palsy with a normal CSF, but bilateral disease with headache warrants parenteral therapy and admission.[7][8]

Stem 2 — syncope in a young camper (answer)

A 26-year-old man from a Lyme-endemic region presents in July with a presyncopal episode, palpitations, and a three-week-old fading rash on his thigh. His ECG shows complete heart block at 38 bpm. What is the diagnosis, and what do you do in the next hour? Model: Lyme carditis with high-grade AV block — the rash was erythema migrans and the spirochaete has seeded the conduction system. Admit for continuous telemetry, place a temporary pacemaker if he is symptomatic or the escape is unstable, start IV ceftriaxone 2 g once daily, and send confirmatory serology. The block is usually reversible within days of antibiotics; a permanent pacemaker is rarely needed. Do not send him home to observe.[4][8]

Stem 3 — the 12-cm thigh patch and a negative serology (answer)

The hiker from the top of the topic has a 12-cm expanding annular patch on her thigh and a Lyme serology that has come back negative. Treat, or repeat the test? Model: Treat now — do not repeat the serology. Erythema migrans is a clinical diagnosis; antibody responses take four to six weeks to develop and serology is only about 30 to 40 per cent sensitive at this stage, so a negative early result is expected. A compatible expanding rash of at least 5 cm in an endemic area is sufficient to start oral doxycycline 100 mg twice daily for 14 to 21 days (amoxicillin or cefuroxime if pregnant). Waiting for seroconversion risks dissemination.[1][2]

Stem 4 — the swollen knee that keeps coming back (answer)

A 45-year-old from the northeastern US has had three episodes of a large, relatively painless right-knee effusion over a year, each resolving spontaneously. Serology is IgG-positive. Do you give more antibiotics? Model: First treat as Lyme arthritis with oral doxycycline 100 mg twice daily or amoxicillin 500 mg three times daily for 28 days; if it persists, a second oral course or IV ceftriaxone 2 g daily for 14 to 28 days. If synovitis persists months after adequate antibiotics with no evidence of persistent infection, that is antibiotic-refractory arthritis — an autoimmune, HLA-DRB1 synovitis managed with NSAIDs, intra-articular corticosteroid, and DMARDs. More antibiotics do not help this group — and that is the trap.[6]

References

1. [1]Steere AC, Strle F, Wormser GP, et al. Lyme borreliosis Nat Rev Dis Primers, 2016.PMID 27976670
2. [2]Sanchez E, Vannier E, Wormser GP, et al. Diagnosis, Treatment, and Prevention of Lyme Disease, Human Granulocytic Anaplasmosis, and Babesiosis: A Review JAMA, 2016.PMID 27115378
3. [3]Kullberg BJ, Vrijmoeth HD, van de Schoor F, et al. Lyme borreliosis: diagnosis and management BMJ, 2020.PMID 32457042
4. [4]Lantos PM, Rumbaugh J, Bockenstedt LK, et al. Clinical Practice Guidelines by the Infectious Diseases Society of America (IDSA), American Academy of Neurology (AAN), and American College of Rheumatology (ACR): 2020 Guidelines for the Prevention, Diagnosis and Treatment of Lyme Disease Clin Infect Dis, 2021.PMID 33417672
5. [5]Mead P. Epidemiology of Lyme Disease Infect Dis Clin North Am, 2022.PMID 36116831
6. [6]Arvikar SL, Steere AC. Lyme Arthritis Infect Dis Clin North Am, 2022.PMID 36116835
7. [7]Rauer S, Kastenbauer S, Fingerle V, et al. Lyme Neuroborreliosis Dtsch Arztebl Int, 2018.PMID 30573008
8. [8]Lantos PM, Rumbaugh J, Bockenstedt LK, et al. Clinical Practice Guidelines by the Infectious Diseases Society of America, American Academy of Neurology, and American College of Rheumatology: 2020 Guidelines for the Prevention, Diagnosis, and Treatment of Lyme Disease Neurology, 2021.PMID 33257476
9. [9]Wong KH, Shapiro ED, Soffer GK. A Review of Post-treatment Lyme Disease Syndrome and Chronic Lyme Disease for the Practicing Immunologist Clin Rev Allergy Immunol, 2022.PMID 34687445
10. [10]Smith RP. Lyme Disease Ann Intern Med, 2025.PMID 40354663
11. [11]Stanek G, Wormser GP, Gray J, et al. Lyme borreliosis Lancet, 2012.PMID 21903253
12. [12]Gade A, Matin T, Rubenstein R, et al. Acrodermatitis Chronica Atrophicans 2026.PMID 33085436
13. [13]Zhou G, Xu X, Zhang Y, et al. Antibiotic prophylaxis for prevention against Lyme disease following tick bite: an updated systematic review and meta-analysis BMC Infect Dis, 2021.PMID 34749665
14. [14]Strnad M, Rudenko N, Rego ROM. Pathogenicity and virulence of Borrelia burgdorferi Virulence, 2023.PMID 37814488