Rachel K. Gardner, M.D., Daniel A. Solomon, M.D., Allison S. Bloom, M.D., Yee-Ping Sun, M.D., and Bruce D. Levy, M.D.
Traveling Companions / 旅する仲間たち
Clinical Problem-Solving
Caren G. Solomon, M.D., M.P.H., Editor
臨床問題解決
編集: Caren G. Solomon, M.D., M.P.H.
In this Journal feature, information about a real patient is presented in stages (boldface type) to an expert clinician, who responds to the information by sharing relevant background and reasoning with the reader (regular type). The authors’ commentary follows.
A 29-year-old man presented to the emergency department with fever, chills, and elbow pain and swelling. He reported having multiple male sexual partners and using injection drugs. One month earlier, fever, chills, and right ankle pain had developed. At that time, he was evaluated in the emergency department of another hospital; the only available information from the other hospital was a radiograph of his ankle, which showed no fracture. His symptoms were managed with supportive care, and they resolved over the course of a few days. Two days before the current presentation, his fever and chills recurred, followed by pain and swelling in his left elbow. His roommate had received a diagnosis of shigellosis 3 weeks earlier; this patient had not had diarrhea.
The subacute presentation of fever and polyarticular arthritis is suggestive of infection, especially in a patient with underlying risk factors including sexual exposures and injection-drug use. Sexually transmitted infections (STIs) such as human immunodeficiency virus (HIV) infection, syphilis, and gonorrhea may cause arthritis. Reactive arthritis — perhaps caused by chlamydia or shigellosis — should also be considered; chlamydia is more likely to occur without diarrhea than shigellosis. The reduction in pain in his right ankle before the involvement of his left elbow suggests a migratory pattern. Migratory polyarthritis can occur with infections or rheumatic diseases.
The patient also reported a nonpruritic and nonpainful rash on his left hand, both arms, chest, and back that had appeared the day before the current presentation. Testing for severe acute respiratory syndrome coronavirus 2 antigen was negative. The patient had no known drug allergies. His grandmother had rheumatoid arthritis; there was no other family history of autoimmune disease. He spent limited time outdoors or around animals. He reported no recent travel outside New England.
His rash, fever, and arthritis suggest the possibility of an underlying STI. Early secondary syphilis is manifested by a transitory macular rash on the torso and limbs, whereas late secondary syphilis is often characterized by a nonpruritic morbilliform or pustular rash on the palms and soles. Disseminated gonococcal infections are commonly associated with a diffuse, nonblanching petechial or pustular rash on the torso and limbs. Infection with salmonella, campylobacter, or yersinia (among others, including chlamydia and shigella, as previously noted) can precipitate reactive arthritis, which is associated with hyperkeratotic lesions on the palms and soles. Lyme disease should also be considered, although the classic associated skin manifestation is a targetoid lesion. An autoimmune disease, such as systemic lupus erythematosus or vasculitis, is a possibility, owing to the patient’s family history. Cancer and drug reactions should always be considered but are unlikely, given the acute time course and the absence of use of new medications.
The patient reported using intravenous methamphetamines, which he injected twice monthly into his antecubital fossae. His last injection, which had been in his right arm, was 2 weeks earlier. He occasionally reused his own needles, but he never shared injection supplies with others. He reported that he did not drink alcohol or use tobacco or other drugs.
His drug use puts him at risk for injection-site infections. Local infections are usually limited to the skin and soft tissue. Septic arthritis resulting from injection-drug use occurs through direct inoculation of bacteria into the bloodstream or through spread from a soft-tissue infection into a local vascular bed. Reuse of needles increases the risk of soft-tissue infections. Sharing injection supplies increases the risk of bloodborne infections such as HIV infection, hepatitis B virus infection, and hepatitis C virus infection. In addition, adulterants in drugs can cause rashes at injection sites, although they classically cause sores and ulcers.
The patient was sexually active with multiple male partners. He engaged in both insertive and receptive oral and anal sex. His last sexual encounter was 2 months before the current presentation. He took daily emtricitabine–tenofovir disoproxil fumarate for preexposure prophylaxis against HIV infection. He did not use condoms and did not take doxycycline for postexposure prophylaxis. He had no history of STIs. He reported no diarrhea, dysuria, abdominal pain, nausea, vomiting, or vision changes.
Although his use of emtricitabine–tenofovir disoproxil fumarate reduces his risk of HIV infection, acute or chronic HIV infection remains on the differential diagnosis; a fourth-generation antigen–antibody assay and a polymerase-chain-reaction (PCR) assay for HIV RNA (to assess the viral load) should be performed. His lack of condom use during sexual activity increases his risk of acute or chronic HIV infection and other STIs. He should be screened for syphilis, as well as for gonorrhea and chlamydia at all potential sites of infection on the basis of his sexual activity. Postexposure prophylaxis with doxycycline may be considered for future protection against bacterial STIs.
On physical examination, the patient was afebrile. His heart rate was 80 beats per minute, his blood pressure 109/64 mm Hg, his respiratory rate 18 breaths per minute, and his oxygen saturation 99% while he was breathing ambient air. He had no palpable lymphadenopathy. Cardiac, pulmonary, and abdominal examinations were unremarkable. The posterior aspect of his left elbow was moderately swollen and tender on palpation. There was no erythema. The range of motion of his left elbow was limited by pain. His right ankle was minimally swollen, with no tenderness or erythema. No swelling, tenderness, or erythema was present in any of his other joints. There was no skin breakdown or stigmata of intravenous drug use. A diffuse, faint, erythematous, nonblanching macular rash was observed on his chest, back, and arms but was absent on his palms and soles (Fig. 1). He had no nail changes. No focal neurologic findings were present. A genital examination was deferred owing to the patient’s preference and the absence of symptoms.
Taken together, the generalized, diffuse rash and the arthritis in his left elbow suggest a systemic inflammatory process. Although infective endocarditis should be considered, given the patient’s history of injection-drug use, no physical examination findings were consistent with such a condition (i.e., no new murmurs, splinter hemorrhages, Janeway lesions, or Osler nodes were present).
A photograph shows a diffuse, faint, erythematous, nonblanching macular rash on the patient’s back. The rash was also present on his chest and arms but was not seen in other locations.
Case Presentation – Part 6 / 症例提示 – 第6部
His white-cell count was 8000 per microliter; 44% of the cells were neutrophils and 44% were lymphocytes). His hemoglobin level was 12.4 g per deciliter, and his platelet count was 330,000 per microliter. A comprehensive metabolic panel was normal. The C-reactive protein level was elevated (119 mg per liter; normal value, <10), as was the erythrocyte sedimentation rate (42 mm per hour; normal value, <15). Blood testing for HIV antigen and antibodies was negative; a PCR assay for HIV RNA was also negative. Blood tests for hepatitis A virus IgM, hepatitis B virus core IgM, hepatitis B virus surface antigen, and hepatitis C virus antibodies were all negative. PCR testing for babesia, ehrlichia, and anaplasma was negative. Lyme disease antibodies were not detected. Antinuclear antibody and anticyclic citrullinated peptide antibody tests were negative. A stool culture for campylobacter species, shigella species, and enterovirus was negative, as was testing of a stool specimen for Shiga toxin.
A radiograph showed mild soft-tissue swelling over the posterior aspect of the left elbow (Fig. 2A). A radiograph of the right ankle showed mild soft-tissue swelling and suggested the presence of a small tibiotalar joint effusion (Fig. 2B). A transthoracic echocardiogram showed no vegetations or valvular dysfunction.
A radiograph of the left elbow (Panel A) shows no displaced fractures, as well as normal alignment and normal joint spaces. No effusions are seen. Mild swelling is present over the posterior aspect of the elbow (arrow). A radiograph of the right ankle (Panel B) shows no displaced fractures, along with normal alignment and normal joint spaces. Mild soft-tissue swelling is noted. In addition, a subtle soft-tissue density suggestive of a possible small tibiotalar joint effusion is seen (arrow).
Case Presentation – Part 7 / 症例提示 – 第7部
Tests for chlamydia and gonorrhea that were performed on a urine sample, an oropharyngeal swab, and a rectal swab were negative. A blood test for treponemal antibodies was positive, and a rapid plasma reagin (RPR) test was positive at a titer of 1:8. The patient was treated with a single dose of penicillin (2.4 million units) administered intramuscularly.
The positive treponemal antibody and RPR tests confirm the diagnosis of syphilis. The patient’s fever and diffuse macular rash are consistent with secondary syphilis, and his arthritis most likely reflects hematogenous spread. Elevated levels of inflammatory markers are consistent with secondary syphilis but are nonspecific. A single intramuscular injection of penicillin is highly effective for the treatment of secondary syphilis.
On the second night of admission, his left elbow pain worsened suddenly, and he had new severe pain in his left wrist. His left elbow appeared erythematous, warm, and more swollen. Two sets of blood cultures that had been obtained at the time of admission were negative for bacteremia. Arthrocentesis of the left elbow, left wrist, and right ankle was performed. Analysis of the synovial fluid obtained from his left elbow revealed 84,090 white cells per microliter, 91% of which were neutrophils. Analysis of the fluid from his left wrist revealed 220 white cells per microliter, 34% of which were neutrophils (clotted aspirate), and analysis of the sample from his right ankle showed 460 white cells per microliter, 45% of which were neutrophils. The samples obtained from the elbow and wrist were purulent, but the sample obtained from the ankle showed no purulence. Gram’s staining of the synovial fluid from his left elbow revealed gram-positive cocci. The Gram’s stains of the fluid from his left wrist and right ankle were negative. No crystals were noted. His left elbow and left wrist were irrigated and débrided, and treatment with ceftriaxone and vancomycin was initiated.
The fact that the synovial fluid sample obtained from the patient’s left elbow contained more than 50,000 white cells per microliter is highly suggestive of septic arthritis. However, low white-cell counts in synovial fluid do not rule out septic arthritis and may be seen in patients with disseminated gonococcal infection, a low white-cell count in peripheral blood, or joint replacement. Empirical treatment with antibiotic agents should cover gram-positive cocci — some of the most common types of bacteria — in addition to gram-negative organisms such as Neisseria gonorrhoeae, as well as Pseudomonas aeruginosa (which is relevant given the patient’s history of injection-drug use).
Cultures of the fluid obtained from the left elbow and left wrist grew methicillin-resistant Staphylococcus aureus (MRSA). Culture of the fluid obtained from the right ankle remained negative. The antibiotics he had been receiving were narrowed to vancomycin, which he received for 5 days. Given his injection-drug use, discharge home with a peripherally inserted central catheter for administration of vancomycin was considered to be unsafe. Instead, he received one intravenous dose of dalbavancin (at a dose of 1.5 g) immediately before discharge, with plans for a second dose 2 weeks later.
Although the typical antibiotic treatment for MRSA septic arthritis is a 3-week course of intravenous vancomycin, this approach was avoided in this patient, given his injection-drug use. Dalbavancin, a lipoglycopeptide antibiotic, has excellent MRSA coverage with a very long half-life (approximately 2 weeks), which makes it a great option in this case.
Two weeks after discharge, the patient was doing well, without pain or joint swelling. Given his improvement, the decision was made to defer a second dose of dalbavancin; instead, he received oral doxycycline at a dose of 100 mg twice daily for 1 month.
This patient presented with fever, chills, arthritis, and a diffuse rash; diagnoses of both secondary syphilis and septic arthritis were made. Clinicians who are presented with a case with disparate signs and symptoms often aim to construct a single unifying diagnosis, according to the law of parsimony. This strategy risks premature closure, and in this case, it delayed diagnosis of the patient’s septic arthritis.
Syphilis is caused by infection with the spirochete Treponema pallidum and is spread through sexual or vertical transmission. The incidence of syphilis has increased since 1950, with a record high of 207,255 cases diagnosed in the United States in 2022. Persons infected with T. pallidum follow a predictable disease course that can be divided into four stages: primary, secondary, latent, and tertiary. Primary and secondary syphilis are the sexually transmissible stages. Primary syphilis is characterized by one — or less commonly, more than one — single painless ulcer (chancre) at the site of inoculation, typically the genitals or the mouth. The usual incubation period is approximately 3 weeks to 3 months. The chancre typically resolves in 2 to 6 weeks, even without antibiotic therapy.
Secondary syphilis typically manifests with myalgias and a nonpruritic, disseminated, maculopapular rash, often on the palms and soles. Other manifestations may include alopecia, periostitis, hepatitis, nephritis, and (infrequently) arthralgias. This patient’s previous right ankle pain may have arisen from tenosynovitis due to hematogenous spread of T. pallidum into his tendon, as has been described in case reports. Secondary syphilis lasts an average of 3.6 months and then either resolves or evolves into latent (asymptomatic) infection.
The high percentage of patients in whom latent syphilis is diagnosed as compared with earlier stages indicates that many primary and secondary syphilis cases go untreated or undiagnosed. Latent syphilis can last a lifetime or can progress to tertiary syphilis. Natural history studies have shown that tertiary syphilis occurs in 15 to 40% of untreated patients, and onset ranges from 1 year to decades after an initial infection. Tertiary syphilis is classically subdivided into gummatous syphilis (involving skin, bone, or other organs), which accounts for about half of cases; cardiovascular syphilis (e.g., aortic aneurysms, myocarditis, or coronary arteritis), which accounts for about a third of cases; and late neurosyphilis syndromes (e.g., gait impairments, dementia, general paresis, or tabes dorsalis), otic syphilis, or ocular syphilis, which account for the remaining cases.
Syphilis is diagnosed with the use of treponemal and nontreponemal tests. Both types of tests are needed, since serologic testing (especially nontreponemal tests) can be associated with false positive results. Treponemal tests detect antibodies against T. pallidum; they are highly specific but cannot distinguish active infections from inactive ones. Nontreponemal tests (i.e., the Venereal Disease Research Laboratory test and the RPR test) quantify the amount of antibodies and therefore are useful for detecting active infection and tracking treatment response. Evaluation of nontreponemal titers should be repeated 6 and 12 months after initiation of treatment. With successful treatment, RPR titers should drop by at least a factor of 4 in 6 to 12 months.
Penicillin is the backbone of syphilis treatment. Primary, secondary, and early latent syphilis (infection within the past year) are typically treated with a single intramuscular injection of 2.4 million units of benzathine penicillin G. Late latent and tertiary syphilis are typically treated with three weekly injections of 2.4 million units of benzathine penicillin G. Patients, especially those with early syphilis (i.e., primary, secondary, or early latent syphilis that was acquired within the previous 12 months), should be counseled that fever, chills, myalgias, and headaches may develop within 24 hours after treatment because of cytotoxin release from the breakdown of spirochetes, known as the Jarisch–Herxheimer reaction.
For patients who cannot receive penicillin owing to an allergy or to drug shortages (as have recently occurred), alternative treatments with cure rates similar to those of penicillin can be used; these include oral doxycycline and intramuscular ceftriaxone. Pregnant patients with a penicillin allergy and patients with neurosyphilis should undergo desensitization followed by penicillin treatment.
The patient’s second diagnosis was septic arthritis. Septic arthritis most frequently arises from occult bacteremia. Small breaks in the skin, such as from intravenous drug use, can allow skin flora (most commonly staphylococcus and streptococcus species) to enter the bloodstream; similarly, gastrointestinal and genitourinary infections can allow predominantly gram-negative species to enter the bloodstream. This patient’s history of intravenous drug use increased his risk of transient bacteremia from direct inoculation of bacteria into his bloodstream or from bacterial entry through minute breaks in his skin. Other recognized risk factors for septic arthritis include preexisting joint disease and immunocompromise. This patient had no known preexisting joint disease, although tenosynovitis due to secondary syphilis may have been susceptible to seeding from transient bacteremia.
Septic arthritis is classically manifested by fever and a warm, swollen, exquisitely painful joint; however, in a series involving patients in the United Kingdom who had septic arthritis, only 58% had fever and only approximately half had leukocytosis. Although most of the cases were monoarticular, involvement of more than one joint occurred in up to 20% of the patients and was most common among patients with inflammatory arthritides, immunocompromise, or sustained bacteremia. An elevated erythrocyte sedimentation rate and an elevated C-reactive protein level are very common among patients with septic arthritis but are nonspecific. Blood cultures should be obtained before starting antibiotics, although they are positive in only 50 to 70% of patients with nongonococcal septic arthritis. The most common cause of septic arthritis is S. aureus, followed by streptococcus species.
The evaluation for septic arthritis includes analysis and Gram’s staining of the synovial fluid, aerobic and anaerobic cultures of the synovial fluid, and measurement of the white-cell count and differential count in the synovial fluid. A white-cell count that exceeds 50,000 per microliter is highly suggestive, but not diagnostic, of septic arthritis. Gram’s stains are positive in 60 to 80% of cases, and cultures are positive in more than 90% of cases. No pathognomonic imaging findings are associated with this infection, but baseline plain radiography is recommended to rule out bone or joint disease.
Treatment of septic arthritis involves the use of antibiotics, along with joint drainage by means of closed-needle aspiration, arthroscopy, or arthrotomy (an open procedure). Empirical treatment with antibiotics targeting S. aureus (including MRSA) and streptococcus species should be started immediately after obtaining synovial fluid and then tailored on the basis of results of microbiologic testing. Empirical regimens typically include vancomycin or daptomycin (for MRSA coverage) plus a third- or fourth-generation cephalosporin (for gram-negative–organism coverage). For patients with risk factors for pseudomonas infection (e.g., immunosuppression or injection-drug use), a cephalosporin with antipseudomonal activity should be used. Source control by means of surgical débridement or arthrocentesis is vital for joint decompression, removal of bacteria, and improved blood flow. A switch to oral antibiotics with high bioavailability (i.e., fluoroquinolones or doxycycline) after source control is associated with good clinical outcomes and obviates the need for durable intravenous access. In a retrospective observational study, a high cure rate for serious gram-positive infections was observed with dalbavancin, but prospective data comparing dalbavancin with standard care are limited. The typical duration of antibiotic therapy for nongonococcal septic arthritis is 2 to 4 weeks.
This case highlights the importance of considering additional diagnoses when the clinical picture is unclear. Furthermore, this case reminds us to consider the varied clinical manifestations of syphilis, for which it has earned the name “the great imitator,” as well as the many complications associated with injection-drug use.
Ratnaraj F, Brooks D, Walton M, Nagabandi A, Abu Hazeem M. Forgotten but not gone! Syphilis induced tenosynovitis. Case Rep Infect Dis 2016;2016:7420938.
Garnett GP, Aral SO, Hoyle DV, Cates W Jr, Anderson RM. The natural history of syphilis: implications for the transmission dynamics and control of infection. Sex Transm Dis 1997;24:185-200.
Satyaputra F, Hendry S, Braddick M, Sivabalan P, Norton R. The laboratory diagnosis of syphilis. J Clin Microbiol 2021;59(10):e0010021.
Clement ME, Okeke NL, Hicks CB. Treatment of syphilis: a systematic review. JAMA 2014;312:1905-17.
Ross JJ. Septic arthritis of native joints. Infect Dis Clin North Am 2017;31:203-18.
Earwood JS, Walker TR, Sue GJC. Septic arthritis: diagnosis and treatment. Am Fam Physician 2021;104:589-97.
Li H-K, Rombach I, Zambellas R, et al. Oral versus intravenous antibiotics for bone and joint infection. N Engl J Med 2019;380:425-36.
英語原文:“A 29-year-old man presented to the emergency department with fever, chills, and elbow pain and swelling. He reported having multiple male sexual partners and using injection drugs.”
英語原文:“Although the typical antibiotic treatment for MRSA septic arthritis is a 3-week course of intravenous vancomycin, this approach was avoided in this patient, given his injection-drug use. Dalbavancin, a lipoglycopeptide antibiotic, has excellent MRSA coverage with a very long half-life (approximately 2 weeks), which makes it a great option in this case.”
flowchart TD
A[患者来院:\n29歳男性\n発熱・悪寒・左肘の痛みと腫れ] --> B{リスク因子の評価}
B --> |複数の男性パートナー\n注射薬物使用| C[性感染症のスクリーニング]
B --> |発疹の存在| D[皮膚所見の評価]
B --> |関節症状| E[関節評価]
C --> C1[トレポネーマ抗体検査\nRPR検査]
C --> C2[HIV検査\n肝炎ウイルス検査]
C --> C3[淋菌・クラミジア検査]
D --> D1[広範囲の斑状発疹\n(胸部・背中・腕)]
E --> E1[X線検査]
E --> E2[炎症マーカー\nCRP・赤沈]
C1 --> |陽性| F[梅毒の診断]
C2 --> |陰性| G[HIV・肝炎は否定的]
C3 --> |陰性| H[淋菌・クラミジアは否定的]
D1 --> I[二期梅毒を示唆]
E1 --> |軟部組織腫脹| J[関節穿刺の検討]
E2 --> |上昇| K[全身性炎症の存在]
F --> L[二期梅毒の確定診断]
L --> M[ペニシリンG筋注\n単回投与]
J --> |入院後の症状悪化| N[関節穿刺実施]
N --> O[左肘関節液の分析:\n白血球84,090/μL\n好中球91%\nグラム陽性球菌]
O --> P[化膿性関節炎の診断]
P --> Q[関節洗浄・デブリドマン\n経験的抗生物質投与]
Q --> R[培養結果: MRSA]
R --> S[MRSA化膿性関節炎の確定診断]
S --> T[抗MRSA薬投与:\nバンコマイシン→ダルババンシン]
I --> L
K --> N
style A fill:#2196B9,stroke:#2196B9,color:white
style L fill:#EF7C1B,stroke:#EF7C1B,color:white
style S fill:#E84518,stroke:#E84518,color:white
subgraph 重要な臨床的決断ポイント
U["複数の診断を考慮する\n(単一診断の罠を避ける)"]
V["関節穿刺は診断の決め手\n(白血球数・培養)"]
W["患者背景を考慮した\n治療薬の選択"]
end
subgraph 診断の鍵となった所見
X["トレポネーマ抗体陽性\nRPR陽性 (1:8)"]
Y["関節液白血球数 > 50,000/μL\n好中球優位"]
Z["MRSAの培養同定"]
end
Claude 3.7 Sonnet凄いですね。人間は完全に追い抜かされそうです。論文を読む、ということの持つ意味が変わりつつあるように思います。従来は”自分で必要なら文献を探して読んで比較しながら理解していく、内容について自分の従来の知識と新しい知識の関連付けをしていく(ネットワークを作る)、というのが論文抄読だと考えていたのですがAIを利用すると”AIにまず論文を読んでもらいAIに教えてもらう”というのがこれからの論文抄読になりそうです。
コメント