Clinical and Health Affairs
Tularemia in Two South Dakota Children
By Nadia A. Sam-Agudu, M.D., DTM&H
■ Tularemia may be relatively rare in the United States, but physicians must be able to recognize it in order to treat it in its earliest stages. They also need to understand that most antibiotics are not effective against the disease. This article presents two cases of tularemia infection among school-aged children in South Dakota who were successfully treated with IV gentamicin and oral antibiotic combinations.
Reported cases of tularemia are relatively rare in the United States. However, the risk of infection is present in almost every state. States in the South Central region have the highest rates; those in the Midwest are not as affected. Occasional outbreaks occur, and bioterrorism is an ongoing concern. Ulceroglandular disease is the most common presentation of tularemia infection. Clinicians should know that most antibiotics used in empiric treatment of lymphadenopathy in children are not effective against tularemia. A high index of suspicion and early diagnosis and treatment are key to combating the infection. This article reports on tularemia infection in two school-aged children from the same area in South Dakota who presented with strikingly similar symptoms within days of each other.
Case 1
A previously healthy 5-year-old Native American girl was referred by a local physician for further management of significant right cervical lymphadenopathy. The patient had not responded to standard empiric antibiotic therapy, and the adenopathy had actually progressed during the antibiotic treatments. The girl lived on a reservation in northeastern South Dakota. About four weeks prior to hospital admission, she experienced intermittent fevers that lasted about one week. A deer tick (Ixodes scapularis) was found on her right parietal scalp one day after the fevers started. A large papule had developed at the site of tick attachment. During the first week of illness, she developed mild cervical lymphadenopathy on the right side. She did not have any myalgias, arthralgias, or rash.
During initial evaluation on Day 3 of her illness, her complete blood count (CBC) was unremarkable. Lyme disease was suspected, and she was treated empirically with 14 days of amoxicillin. However, subsequent Lyme disease titers were negative. At her postamoxicillin evaluation, the fevers had resolved, but the right cervical lymphadenopathy had worsened. The tick attachment site on her scalp was larger and more erythematous. She was given five days of azithromycin to treat cellulitis and/or cat-scratch disease (Bartonella henselae infection). There was no improvement, and the cervical adenopathy progressed even further. There was no spontaneous suppuration.
After four weeks of illness and no response to amoxicillin or azithromycin, the patient was hospitalized for further management. Her Lyme disease titer was rechecked on admission and was again negative. On physical exam, she had impressive cervical adenopathy (Figure 1a) on her right side that was semifluctuant and mildly tender to palpation. The tick attachment site on the patient’s scalp was now ulcerated (Figure 1b). Pediatric infectious disease and otorhinolaryngology consults were obtained. Given the patient’s history and presentation, physical exam, previous laboratory results, and lack of response to antibiotics, tularemia was suspected. Her chest X-ray was normal, as was tuberculosis skin testing. Lymph node aspiration was performed by the otorhinolaryngologist; gram stain and cultures were negative. The microbiology laboratory personnel had been alerted to the possibility of tularemia. A peripherally inserted central catheter (PICC) was placed, and IV gentamicin was initiated, pending results of Francisella tularensis serology. The patient’s F. tularensis titers were markedly elevated at 1:5,120 (>1:160 is positive). Bartonella henselae serology was negative. The patient was discharged from the hospital after two days. She received a 10-day course of gentamicin, with an additional 10 days of oral doxycycline for a satisfactory response and resolution.
Case 2
Exactly two weeks after the first patient was admitted, the same physician referred yet another patient for management of a similar-looking left cervical adenopathy. Given the first patient’s diagnosis, the physician already suspected tularemia prior to the transfer. Patient No. 2 was a previously healthy 6-year-old girl, also Native American, who lived on the same reservation as Patient No. 1. The children were not related, nor were they acquainted with one another. Neither had traveled significantly, been swimming in lakes or rivers, or been in contact with domestic or wild animals prior to the onset of symptoms.
The second patient’s symptoms started within three days of those of Patient No. 1. By the time the former was admitted, she had been ill for almost six weeks. She also had fevers during the first week of illness.
The girl was initially evaluated for large tonsils and odynophagia, and was prescribed amoxicillin for acute tonsillitis, even though rapid group A streptoccocal testing on a pharyngeal swab had been negative.
On postamoxicillin evaluation (two weeks into her illness), the odynophagia and fever had resolved, but she had developed tender left cervical adenopathy. A “bug bite” associated with tenderness and swelling was discovered on her left occipital scalp. No tick was found. She was prescribed oral cefdinir for bacterial cellulitis and cervical adenitis.
The left cervical adenopathy progressed during and after the cefdinir course. She did not report any myalgias, arthralgias, or rash. She was re-evaluated in the clinic on the day prior to admission for progressive left cervical adenopathy. Her CBC was unremarkable. Erythrocyte sedimentation rate was 51mm/hour. Lyme disease and Epstein-Barr virus titers were negative. Given the similarity of her symptoms to those of Patient No. 1, F. tularensis titers were sent, and she was transferred to our institution for PICC line placement and initiation of gentamicin therapy. Her chest X-ray on admission was negative, as was her tuberculosis skin test. The patient’s exam findings were similar to those of Patient No. 1 (Figure 2a); additionally, she had an enlarged left tonsil (Figure 2b). Francisella tularensis titers were elevated at 1: 2,560. Patient No. 2 received 17 days of IV gentamicin and 10 days of oral ciprofloxacin before her symptoms resolved.
At about a year post-treatment, there have been no reported relapses in either child, nor have there been any known pediatric tularemia cases from that particular reservation since these children were evaluated.
Discussion
Tularemia (also known as rabbit fever, deer-fly fever, or meat-cutter’s disease) has become a relatively rare diagnosis in the United States. The annual number of reported cases ranged from 300 to 900 from the 1950s through the mid 1960s; since 1985, the number has averaged between 100 and 200.1 However, the risk for infection continues to be present almost everywhere in the country; tularemia has been reported in every state except Hawaii.1 States in the South Central United States such as Arkansas and Missouri have the highest rates of endemic tularemia. States in the Midwest generally have much lower reported rates. Of note, South Dakota was in the top five states for reported tularemia cases from 1990 to 2000, and from 2000 to 2008.1,2 When compared with other Midwestern states, South Dakota (population 819,000) had 62 cases of tularemia from 2000 to 2008, Minnesota (population 5.3 million) had seven cases, and North Dakota (population 676,000) reported nine cases.2,3 Earlier tularemia outbreaks have been reported in South Dakota, all on Native American reservations; the most recent was an outbreak involving nearly 30 cases in 1984.4
Tularemia typically presents as an acute febrile illness, with or without obvious physical findings. It may present as ulceroglandular or typhoidal disease, with or without pneumonia. Patients with ulceroglandular tularemia may present with a tender maculopapular lesion at the bite/contact site that ultimately ulcerates. They also may have impressive regional adenopathy. As with our two cases, the majority of patients (approximately 75%) present with ulceroglandular disease.5 Typhoidal tularemia occurs in about 25% of patients; it manifests with high fever, splenomegaly, hepatomegaly, and enteritis.
There have been outbreaks of tularemia in various parts of the United States and around the world, especially in Europe. Patients may contract tularemia as a result of bites from or contact with rodents and lagomorphs, cats, contaminated meat/water, deer flies, ticks, and mosquitoes.5 Tularemia is highly infectious; a single organism is enough to establish clinical disease.5 Inhaled F. tularensis organisms may cause pneumonia, which has up to 50% mortality. As such, its potential as an agent of bioterrorism is of great concern.
In these two patients, the presentation was somewhat similar to that of standard bacterial cervical lymphadenitis, albeit more impressive. The children were ill for four to six weeks before tularemia was suspected and worked up. Fortunately, they presented with relatively benign disease, and recovered without sequelae.
Standard recommended treatment for tularemia is parenteral streptomycin or gentamicin for at least 10 days.6 Gentamicin has been successfully used in children.7 Oral agents such as ciprofloxacin and doxycycline are recommended as alternatives6; however, few clinical studies have investigated their efficacy in and ability to be tolerated by children. Snowden and Stovall recently published a retrospective review of 30 pediatric tularemia cases in Arkansas that occurred between 1996 and 2006.8 The authors noted that children who were treated with oral antibiotics alone or with shorter courses (≤ seven days) of initial gentamicin therapy were more likely to relapse or require prolonged therapy.
Most of the antibiotics used in empiric treatment of routine infections in children do not affect tularemia, and the diagnosis is typically confirmed by serological tests. Therefore, only when a clinician specifically considers tularemia can the correct diagnosis be made and proper treatment initiated.
Children younger than 10 years of age account for a significant number of cases in the Centers for Disease Control and Prevention’s surveillance reports;1,2 in the Arkansas study, almost 75% of the children were younger than 6 years of age.8 Clinicians caring for children in every state should be aware of the risk of tularemia and consider this diagnosis, especially when a child presents with fever, adenopathy, or skin lesions that do not respond to standard empiric antibiotics. Physicians should be especially attuned to the possibility of tularemia in subacute or chronic cases of unexplained adenopathy in children. The presence or evidence of an ulcerated arthropod, insect, or animal bite in an area that is drained by, or proximal to, affected lymph nodes is strongly suggestive of tularemia.
The toxicities, age limitations, and lack of Food and Drug Administration approval for routine use of tetracyclines and fluoroquinolones in young children may cause some providers to hesitate prescribing these medications. However, once tularemia is suspected or confirmed, appropriate initial therapy with gentamicin should be initiated. If oral therapy is deemed appropriate for continuation of therapy or if aminoglycosides are contraindicated for a patient, oral ciprofloxacin or doxycycline should immediately be prescribed for the appropriate duration6; the threshold for prolonged treatment should be relatively low in order to ensure successful treatment and complete resolution. The second patient, for example, required a longer course of gentamicin, perhaps because of the duration of illness (six weeks) before receiving appropriate treatment. However, both children tolerated oral antibiotics quite well, and did not experience any relapses.
When compared with Snowden and Stovall’s review (which reported an average of three inpatient cases a year in Arkansas, a highly endemic state), the diagnosis of two pediatric cases of tularemia within two weeks of each other in the Midwest suggests that the incidence of tularemia may be higher than reported. Cases may present in clusters, perhaps based on tick populations and/or greater seasonal exposure to ticks. Therefore, as much as tularemia is relatively rare in the Midwest, an unacceptable number of cases may be missed if clinicians are not aware of the almost universal risk in the United States.
For clinicians who identify a case of tularemia, evidence-based data on treatment of pediatric patients will likely reduce complications and result in the best outcomes for those patients. Some of the greatest gains may be made in providing evidence for clinicians who may otherwise limit doxycycline and ciprofloxacin therapy in children who may need prolonged treatment. Rigorous, multicenter clinical studies on the efficacy and toxicity of tularemia treatment options and combinations would be extremely useful to pediatric providers treating patients with the infection. MM
At the time of writing, Nadia Sam-Agudu was a clinical assistant professor of pediatrics at the University of North Dakota School of Medicine and Health Sciences in Grand Forks, and a pediatric infectious diseases specialist with Sanford Children’s Hospital and Clinic in Fargo. She is currently the technical advisor, pediatrics, for the Institute of Human Virology in Nigeria and an adjunct assistant professor of pediatrics at the University of Minnesota Medical School.
References
1. Centers for Disease Control and Prevention. Tularemia—United States, 1990-2000. MMWR Morb Mortal Wkly Rep. 2002;51(9):181-4.
2. Centers for Disease Control and Prevention. Reported tularemia cases by state-United States, 2000-2008. December 21, 2009.; Available at: www.cdc.gov/tularemia/surveillance/Tul_CasesbyState.html. Accessed April 4, 2011.
3. U.S. Census Bureau. 2010 Resident Population. Available at: www.census.gov/. Accessed April 4, 2011.
4. Centers for Disease Control. Outbreak of tick-borne tularemia—South Dakota. MMWR Morb Mortal Wkly Rep. 1984;33(42):601-2.
5. Nigrovic LE, Wingerter SL. Tularemia. Infect Dis Clin North Am. 2008;22(3):489-504, ix.
6. Dennis DT, Inglesby TV, Henderson DA, et al. Tularemia as a biological weapon: medical and public health management. JAMA. 2001;285(21):2763-73.
7. Cross JT Jr., Schutze GE, Jacobs RF. Treatment of tularemia with gentamicin in pediatric patients. Pediatr Infect Dis J. 1995;14(2):151-2.
8. Snowden J, Stovall S. Tularemia: Retrospective review of 10 years’ experience in Arkansas. Clin Pediatr (Phila). 2011;50(1):64-8. Epub 2010 Sept. 13.