Journal Logo



The More You Learn about Cat Scratch Disease, the More Confusing It Gets

Roberts, James R. MD

doi: 10.1097/01.EEM.0000616464.76991.e0
    cat scratch disease
    cat scratch disease:
    Typical scratches from a cat can introduce bacteria (Bartonella henselae) and cause cat scratch disease. Bites from a cat or its fleas can also inoculate the patient. Antibiotics are suggested by most authors, but the benefit has not been substantiated. Antibiotic prophylaxis for CSD from a simple cat scratch is curiously not recommended.

    Cats are common pets, found in about 35 percent of households in the United States. (ASPCA. Cats are generally docile, but they can be unpredictable and occasionally bite or scratch people for no good reason. Numerous diseases can be transmitted this way from cats to humans. Infections following cat puncture wounds are relatively common, and up to 80 percent of cat bites become infected. (Can J Infect Dis. 2000;11[5]:227; Occasionally, septic sequelae, such as meningitis, endocarditis, and osteomyelitis, result.

    Cat bite infections are usually polymicrobial. Pasteurella, Streptococcus, Staphylococcus, and multiple, often unusual, anaerobic organisms are isolated from most infections (an amazing average of five organisms per bite). Common antibiotics used to treat soft tissue infections empirically, such as cefalexin, clindamycin, dicloxacillin, and erythromycin are not recommended for prophylaxis or treating an infected cat bite.

    A potentially confusing complication of being scratched or bitten by a cat is cat scratch disease, also called cat scratch fever. This disease is more common than most clinicians suspect. The diagnosis is much easier if evidence of an encounter with a cat is still visible. But seeing a patient with some of the protean systemic symptoms after the scratches have disappeared can be quite confusing.

    Cat Scratch Disease

    English R

    Pediatr Rev.


    This is a well-done, detailed article describing the etiology, epidemiology, clinical presentation, and laboratory testing for cat scratch disease (CSD). The author stated that this is a relatively common infection in children, but I posit that many emergency clinicians have never knowingly seen a case of systemic CSD. Children who are immunocompetent typically have a self-limited course, but immunocompromised patients can develop serious sequelae following the infection. Antibiotic treatment is usually prescribed for CSD, but its effectiveness is unknown and unproven. The most common feature of CSD is regional lymphadenopathy, which clears up on its own, albeit slowly, but multiple other systemic complications are associated with this infection.

    An enlarged lymph node in a child with cat scratch disease. The nodes are proximal to the bite or scratch, and occur most commonly in the axilla and cervical region. This node is somewhat atypical, being nontender and nonerythematous.

    The author described a 7-year-old boy who presented to his physician with a fever of 103°F for three weeks. He initially complained of a sore throat, diffuse abdominal pain, and intermittent right hip, right leg, and back pain. He received a brief course of amoxicillin clavulanate and azithromycin without relief. He had a cat at home.

    The physical findings were within normal limits, and no tenderness of the extremities or back nor lymphadenopathy were found. Laboratory tests were described as normal except for a rather high RBC sedimentation rate of 99 mm/hr and a C-reactive protein of 2.7 mg/L. A bone scan showed increased uptake in the right ribs, ileum, and thoracic spine. The admitting diagnosis was multifocal osteomyelitis. Intravenous oxacillin was initiated. An MRI revealed inflammatory lesions of the thoracic and lumbar vertebrae and the right acetabulum. The liver and spleen were normal, as was a bone marrow biopsy. A two-week course of azithromycin and rifampin was started. Bartonella henselae titers were markedly elevated, confirming the diagnosis of CSD. He eventually recovered without sequelae.

    CSD is associated with an infection by Bartonella henselae and some related species, including three or more gram-negative bacilli. Many cats harbor the bacteria in the blood and develop antibodies to the organisms. Stray cats seem to have a higher rate of bacteremia and antibody levels. The incubation period from the scratch to the development of lymphadenopathy is one to two weeks. Many patients cannot recall a scratch or bite. Histologic examination of lymph nodes reveals lymphoid hyperplasia and arteriolar proliferation, with occasional granulomas and microabscesses. Diagnosis can also be confirmed by finding the bacilli with the Warthin-Starry stain, but it is not always positive.

    This author said CSD is one of the most common causes of chronic lymphadenitis in children, secondary to a scratch or bite from an infected cat or exposure to cat fleas. At the traumatic site, a brownish red papule can be seen within about a week of the trauma and lymphadenopathy. Adenopathy occurs in a regional site draining the area of the papule, and nodes appear in one to two weeks. Most common are cervical and axillary lymphadenopathies, but numerous nodes can be affected. Some nodes can be small and undetectable, or they may be markedly enlarged to several centimeters. The nodes are often tender, warm, and erythematous, and they can suppurate. Enlarged nodes will persist for weeks to months.

    About half of the patients will have fever, sometimes as high as 104°F. A variety of constitutional symptoms occur, such as malaise, anorexia, and sore throat, but patients usually appear to be quite well. Similar presentations can occur from atypical mycobacteria, tularemia, brucellosis, toxoplasmosis, and tuberculosis. The differential diagnosis of adenopathy also includes lymphomas, leukemia, and sarcoidosis. Often the presentation is a fever of unknown origin without adenopathy in about 30 percent of cases. (Can J Infect Dis. 1997; 8[1]:43; A variety of complications have been described, from headache and cranial or peripheral nerve abnormalities to encephalopathy with mental status changes and seizures. Status epilepticus has also been reported.

    The diagnosis can be confirmed with serologic testing using an immunofluorescent antibody test (the Warthin-Starry stain) and finding elevated Bartonella henselae immunoglobulin titers.

    Warm compresses on the affected nodes and antipyretics are sufficient for immunocompetent patients. The actual value of antibiotic treatment has not been proven and is not associated with marked improvement. Incision and drainage of suppurative lymph nodes is not recommended because that can produce prolonged drainage. Most cases will resolve in two to four months, but lymph nodes may remain enlarged for years.

    Comment: I was confused by this case. The author thought it was a simple and common diagnosis, but the case was rather complicated, and multiple unusual tests were performed. Certainly, the workup would not be appropriate in the ED. The systemic symptoms are protean and nonspecific, and certainly most clinicians would not jump to a diagnosis of CSD when presented with an advanced systemic case. The diagnosis may be suspected in a patient with adenopathy of unknown cause. The lymph nodes can be rather large, and they develop proximal to a bite or scratch. The enlarged lymph nodes are usually tender and often erythematous. Occasionally, nodes will develop pus. There is often a cutaneous lesion at the site of inoculation, but evidence of cat scratches is often gone. I don't think most clinicians in the ED get a history of owning a cat, and patients (or their parents) may not know that their now-cured scratches were important to mention.

    Most cases exhibit only tender regional lymphadenopathy, and it is not known why systemic disease ensues. Disseminated CSD can be serious in an immunocompromised host, such as a patient with HIV. CSD often involves the liver or spleen, and would most likely present as a fever of unknown origin, often associated with abdominal pain and weight loss. An abdominal CT scan may demonstrate necrotizing granulomas, which are characteristic and do not require biopsies. One wonders if radiologists often make this diagnosis.

    Exposure to cat fleas can also cause the disease. The bacteria in the cat's blood and saliva are somehow deposited and thrive on the animal's claws, by which they enter the skin via a scratch. It can also be spread via contact with cat saliva through broken skin or mucosal surfaces. Cats have a bacteremia that does not make them appear ill. Young cats are most commonly affected. Cat fleas spread the infection among cats as well. Inoculation by a cat flea would explain the lack of a history of a bite or scratch by a cat.

    I was somewhat nonplused, if not quite educated, about CSD when researching this topic. Most patients will present with typical clinical features, but many atypical findings can occur with the systemic disease. Even suspecting disseminated CSD in the ED is a challenge, and this disease would not likely even be considered by most emergency clinicians.

    The benefit of antibiotics is unclear, but most authors suggest treatment for all patients with CSD and lymphadenitis, even immunocompetent patients with a mild to moderate form of the disease. Antibiotics seem reasonable because CSD is caused by bacteria. Only a few studies have demonstrated that antibiotics shorten the duration of symptoms. Azithromycin in a standard five-day regimen seems to be the preferred antibiotic regimen. Suggested alternative treatment is trimethoprim-sulfamethoxazole for seven to 10 days. The treatment of disseminated CSD, however, would not be started in the ED because the diagnosis would require a more complicated evaluation.

    The more one reads about cat scratch disease, the more complicated it becomes. It is reasonable to diagnose and treat self-limited adenopathy as an outpatient with follow-up. I don't think I ever saw or even considered systemic CSD, but I have never worked in a pediatric hospital. I had never even heard of the Warthin-Starry silver stain or the Bartonella henselae bacterium.

    Adenopathy plus a puncture wound from a cat scratch or bite are typical for cat scratch disease. Lymph nodes are usually seen in the axilla or cervical region, proximal to the inoculation, and are not generalized. Nodes can be quite large, are usually tender, and can be erythematous. Some nodes can be suppurative, but incision and draining are discouraged. Most cases of cat scratch disease are limited to lymphadenopathy, but systemic infection, usually involving the liver and spleen, and occasionally the CNS, can occur. Lymph nodes can persist for months. A standard five-day course of azithromycin has been reported to decrease lymph node size.

    Antibiotics are recommended, but no data support them for treating CSD. It is unknown whether antibiotic treatment of localized lymphadenopathy reduces the risk of developing systemic disease. Given the length of the disease, it seems unusual that only a five-day course is suggested. But it has been demonstrated that treating typical CSD patients with oral azithromycin for five days did afford a clinically significant benefit as measured by total decrease in lymph node value within the first month. (Pediatr Infect Dis J. 1998;17[6]:447.)

    Finally, there is no indication to treat a simple cat scratch with antibiotics as prophylaxis for CSD, which is odd because this is a bacterial disease.

    Reader Feedback: Readers are invited to ask specific questions and offer personal experiences, comments, or observations on InFocus topics. Literature references are appreciated. Pertinent responses will be published in a future issue. Please send comments to [email protected].

    Read InFocus and Earn CME!

    Earn CME by completing a quiz about this article. You may read the article here, on our website, or in our iPad app, and then complete the quiz, answering at least 70 percent of the questions correctly to earn CME credit. The cost of the CME exam is $10. The payment covers processing and certificate fees.

    Visit for more information about this educational offering and to complete the CME activity. This enduring material is available to physicians in all specialties, nurses, and other allied health professionals. Lippincott Continuing Medical Education Institute, Inc., is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. Lippincott Continuing Medical Education Institute, Inc., designates this enduring material for a maximum of 1 AMA PRA Category 1 Credit™. Physicians should only claim credit commensurate with the extent of their participation in the activity. This activity expires Dec. 31, 2021.

    Learning Objectives for This Month's CME Activity: After participating in this CME activity, readers should be better able to explain the diagnosis and treatment for cat scratch disease.

    Share this article on Twitter and Facebook.

    Access the links in EMN by reading this on our website,

    Comments? Write to us at [email protected].

    Dr. Robertsis a professor of emergency medicine and toxicology at the Drexel University College of Medicine in Philadelphia. Read the Procedural Pause, a blog by Dr. Roberts and his daughter, Martha Roberts, ACNP, PNP, at, and read his past columns at

    Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.