Peer Reviewed

Photoclinic

Cat-Scratch Disease Presenting as Fever of Unknown Origin

AUTHORS:
Sajiv Alias, MD1 • Nathan F. Bradford, MD2,3 • Zachary J. Grey, MS4

AFFILIATIONS:
1Resident, Department of Family Medicine, AnMed Health Medical Center, Anderson, South Carolina
2Affiliate Professor, Department of Medicine, Medical University of South Carolina, Charleston, South Carolina
3Departments of Internal Medicine and Pediatrics, AnMed Health Medical Center, Anderson, South Carolina
4Medical Student, Medical University of South Carolina, Charleston, South Carolina

CITATION:
Alias S, Bradford NF, Grey ZJ. Cat-scratch disease presenting as a fever of unknown origin. Consultant. 2022;62(4):e28-e30. doi:10.25270/con.2021.08.00009

Received March 17, 2021. Accepted April 9, 2021. Published online August 24, 2021.

DISCLOSURES:
The authors report no relevant financial relationships.

CORRESPONDENCE:
Sajiv Alias, MD, AnMed Health Medical Center, 2000 East Greenville Street, Suite 3700, Anderson, SC 29621 (Sajiv.Alias@anmedhealth.org)


 

A 34-year-old woman presented to the emergency department with a 4-week history of daily fevers, myalgia, headache, cough, vomiting, and diarrhea. She had a medical history significant for migraines, anxiety, depression, and anemia. She was taking antipyretics, ondansetron, and duloxetine.

There was no history of rash, vaginal discharge, blood loss, or dysuria. Her only sexual contact was her boyfriend. Neither he nor her child had been sick. There was no history of sexually transmitted diseases, travel, camping, hiking, or tick bites. However, the patient had recently received 2 tattoos in an accredited facility. In addition, she had been exposed to COVID-19. She had received treatment for community-acquired Clostridioides difficile infection the year prior to presentation. The patient had 2 healthy pet dogs and denied exposure to cats.

Physical examination. At the time of admission, the patient was a well-developed but mildly dehydrated middle-aged woman with an elevated temperature of 39.4 °C and elevated heart rate of 139 beats/min. She had mildly tender bilateral anterior cervical lymphadenopathy and moderate left-sided abdominal tenderness without guarding. No rashes were present. Two tattoos were present without erythema on the left arm.

Diagnostic testing. Results from laboratory studies showed leukocytosis with an elevated white blood cell count of 17,800/μL and 87% neutrophils, a low hemoglobin level of 8.6 g/dL, a low hematocrit level of 27.3%, and an elevated platelet count of 57.1 × 103/μL. Her D-dimer level was also elevated at 659 μg/mL, erythrocyte sedimentation rate was elevated at more than 145 mm/hr, and C-reactive protein level was elevated at 19.3 mg/dL. Her salicylate level was low at 13.8 μg/mL and acetaminophen level was undetectable. Results of a COVID-19 polymerase chain reaction test, as well as HIV and hepatitis panels, were negative.

A chest radiography scan was conducted, results of which were noncontributory. A computed tomography scan of the abdomen showed only bilateral ovarian cysts with a trace of free fluid in the pelvis. Further laboratory studies were conducted for other infectious and rheumatological etiologies.

The patient had a fever of unknown origin as defined by intermittent fevers lasting longer than 3 weeks without an etiology after appropriate workup. Given the presentation, the care team considered a wide differential and were concerned for occult bacteremia, endocarditis, meningitis, prolonged viral syndrome, malignancy, collagen vascular disease such as systemic lupus erythematosus, and Still disease.

The patient was initially administered intravenous fluids, a vancomycin loading dose of 1250 mg with protocol dose adjustments, and meropenem, 1 g, every 8 hours. An infectious disease specialist was consulted, and antibiotics were adjusted to include rifampin, 300 mg, twice daily and doxycycline, 100 mg, twice daily. Additional history was elicited daily to clue into the diagnosis. The patient’s tattoos remained benign in appearance. However, she developed further enlargement of lymph nodes in the left axilla.

Diagnostic imaging showed disseminated lesions in the lymph nodes, liver, and bones (Figures 1 and 2). The patient underwent fine needle aspiration of the left axillary lymph nodes, results of which showed necrotizing granulomas that can be associated with various conditions, including cat-scratch disease, fungal/acid-fast infections, and tularemia (Figure 3).

 

Figure 1. A computed tomography scan showed a lesion in the liver consistent with micro-abscesses, characteristic of disseminated cat-scratch disease.
Figure 1. A computed tomography scan showed a lesion in the liver consistent with micro-abscesses, characteristic of disseminated cat-scratch disease.

Figure 2. A total-body positron emission tomography scan showed focal areas of bone marrow involvement, indicative of cat-scratch disease. Note the areas of increased activity in the anterior inferior left iliac spin, sacroiliac joints bilaterally, left pubic rami, left ischial tuberosity, and the T7 vertebral body.
Figure 2. A total-body positron emission tomography scan showed focal areas of bone marrow involvement, indicative of cat-scratch disease. Note the areas of increased activity in the anterior inferior left iliac spin, sacroiliac joints bilaterally, left pubic rami, left ischial tuberosity, and the T7 vertebral body.

Figure 3. A hematoxylin and eosin stain of the left axillary lymph node aspiration showed granulomata, characteristic of cat-scratch disease.
Figure 3. A hematoxylin and eosin stain of the left axillary lymph node aspiration showed granulomata, characteristic of cat-scratch disease.

 

With these findings, the patient was asked again about any cat exposure, to which she replied, “Well, yes, my daughter and I take care of a stray cat in the neighborhood.” Shortly thereafter, a Bartonella henselae titer returned results showing an elevated level of 1:640, thus confirming the diagnosis of cat-scratch disease. A week later, her titer results doubled to 1:1280. Moreover, results of a Warthin-Starry stain of the previously obtained left axillary fine needle aspiration revealed clusters of rod-shaped structures, further confirming the presence of cat-scratch bacilli.

The patient’s antibiotic regimen was changed to gentamicin, 400 mg daily, rifampin, 300 mg twice daily, and doxycycline, 100 mg twice daily. She continued to have nearly daily fevers and widespread pain. Multiple etiologies for fever were considered, including Jarisch Herxheimer reaction or drug fever, which resulted in discontinuation of rifampin. She continued to have fevers during the remainder of her 3-week hospital course, and it was anticipated that she would continue to have intermittent fevers over the next several months given the extent of her infection. She was discharged from the hospital with a 3-month course of azithromycin, 500 mg daily, doxycycline, 100 mg twice daily, and an extended prednisone taper starting at 60 mg daily.

Discussion. Cat-scratch disease is a relatively rare disease, with fewer than 12,000 cases reported each year.1-3 The infection is caused by Bartonella henselae and is directly linked to cat exposure. Most cases do not require hospital admission. Cat-scratch disease is commonly thought of as a disease of children, although nearly 45% of cases are noted in patients older than age 18 years.3,4 The disease also shows a seasonal distribution, with 50% of the cases occurring between August and January.4 Cat-scratch disease has been well documented in all areas of North America and is most prevalent in the southeastern United States.4

In children, nearly 90% of cases will be self-limited, and approximately one-third of fever of unknown origin in children can be attributed to undiagnosed cat-scratch disease.3 Occasionally, a more severe form of cat-scratch disease may occur with dissemination of the infection to other organs. With lymphoid, spleen, and liver involvement, manifestations can include arthralgia, myalgia, abdominal pain, elevated acute-phase reactants, and elevated liver enzymes, as was the case in our patient.1-3 Cat-scratch disease can also involve the eyes, resulting in optic neuritis, retinitis, or Parinaud oculoglandular syndrome.3 Central nervous system involvement can include encephalitis or meningitis.3 The wide variability in presentation can often lead to significant challenges in diagnosis.

Generally, a presumed diagnosis of cat-scratch disease can be made given a history of cat bite or scratch and appropriate clinical findings including fever and enlarged, painful regional lymphadenopathy near the site of inoculation.1,3 While not routinely indicated, lymph node biopsies showing granulomas that later develop central necrosis can aid in diagnosis, as was the case with our patient. Silver staining, such as the Warthin-Starry stain, have historically been used to visualize the bacterium within granulomas. Culture of biopsy tissue or blood samples for B henselae is generally not indicated, as this bacterium is notoriously difficult to culture. Serology for B henselae can be helpful to confirm the diagnosis, particularly if titers are more than 1:256, which is highly suggestive of active or recent infection, as was the case in our patient.2,5 Despite advances of laboratory techniques, a history of exposure to cats is often the most helpful diagnostic tool and will help inform treatment strategies.

Treatment of the disease is highly dependent on clinical presentation. Currently, limited research regarding optimal regimens is available. Milder cases of cat-scratch disease can be managed with observation alone; lymph node aspiration; oral antibiotics with azithromycin, trimethoprim-sulfamethoxazole, clarithromycin, doxycycline, or rifampin; or a combination of these anitbiotcs.6,7 For disseminated disease, combination therapy with rifampin and either azithromycin or gentamicin is preferred, and combination therapy with doxycycline-rifampin is typically reserved for central nervous system involvement.6,7 Treatment duration should be guided by clinical presentation and evidence of disease resolution. Acute-phase reactants can be helpful in tracking successful response to treatment as well as follow-up imaging of affected organs.6,7 In our case, antibiotic combination therapy with rifampin would have been preferred, but alterations in antibiotic selection were made because of concerns for tolerability. The prolonged course of azithromycin and doxycycline was chosen because of the extent of dissemination and the observed tolerance with clinical improvement on that regimen. 

Patient outcome. The patient had close follow-up with an infectious disease specialist and her primary care provider after discharge. She adhered to her antibiotic regimen and has had no further fevers. Inflammatory markers trended downward. Repeat computed tomography imaging of the abdomen 2 months after hospital discharge revealed resolution of liver lesions. While objective data shows resolution of her illness, she continues to have pain without obvious cause. As a result, she has also experienced worsening of her anxiety and depression since hospital discharge. She no longer interacts with stray cats.

Conclusion. Medicine is both an art and a science. Despite one’s best efforts, the patient’s history is often confounding, as in this case, in which the patient’s lack of recall delayed the diagnosis. Once again, we are reminded of the quote by Sir William Osler, “Medicine is a science of uncertainty and the art of probability.”

References

1. Klotz SA, Ianas V, Elliott SP. Cat-scratch Disease. Am Fam Physician. 2011;83(2):152-155. https://www.aafp.org/afp/2011/0115/p152.html

2. Nelson CA, Moore AR, Perea AE, Mead PS. Cat scratch disease: U.S. clinicians' experience and knowledge. Zoonoses Public Health. 2018;65(1):67-73. https://doi.org/10.1111/zph.12368

3. Baranowski K, Huang B. Cat scratch disease. In: StatPearls. StatPearls Publishing; June 25, 2021. http://www.ncbi.nlm.nih.gov/books/nbk482139/

4. Nelson CA, Saha S, Mead PS. Cat-scratch disease in the United States, 2005-2013. Emerg Infect Dis. 2016;22(10):1741-1746. https://doi.org/10.3201/eid2210.160115

5. Vermeulen MJ, Herremans M, Verbakel H, et al. Serological testing for Bartonella henselae infections in The Netherlands: clinical evaluation of immunofluorescence assay and ELISA. Clin Microbiol Infect. 2007;13(6):627-634. https://doi.org/10.1111/j.1469-0691.2007.01700.x

6. Rolain JM, Brouqui P, Koehler JE, Maguina C, Dolan MJ, Raoult D. Recommendations for treatment of human infections caused by Bartonella species. Antimicrob Agents Chemother. 2004;48(6):1921-1933. https://doi.org/10.1128/aac.48.6.1921-1933.2004

7. Angelakis E, Raoult D. Pathogenicity and treatment of Bartonella infections. Int J Antimicrob Agents. 2014;44(1):16-25. https://doi.org/10.1016/j.ijantimicag.2014.04.00