Case 10-2005 — A 73-Year-Old Man with Weakness and Pain in the Legs
http://www.100md.com
《新英格兰医药杂志》
Presentation of Case
A 73-year-old man was admitted to the hospital in June because of fever, weakness, and pain in the muscles of his legs.
He had been well until seven days before admission, when a dry cough and rhinorrhea developed and lasted for two days. Six days before admission, he had traveled from Boston to coastal South Carolina to attend a conference. Three days before admission, he had had an intense aching pain in both thighs and a temperature as high as 39.4°C. Shortly after that, generalized weakness developed, and he began to have an intense aching pain in the muscles of his legs. He had increased urinary frequency, with one episode of urinary incontinence. His wife brought him to the emergency department of this hospital. He had no history of shortness of breath, chest pain, production of sputum, nausea, vomiting, or change in bowel habits. Three months before admission, atorvastatin (5 mg daily) had been prescribed for the treatment of elevated serum cholesterol, along with vitamin B complex and folic acid for treatment of an elevated plasma homocysteine level. The patient had stopped taking atorvastatin four days before admission.
The patient's medical history included benign enlargement of the prostate. He had had episodes of chronic low back pain, which radiated from low in his back to his buttocks. He had partial bilateral hearing loss. He had undergone arthroscopic knee surgery and bilateral cataract extractions in the distant past. The patient was a practicing physician. Many years earlier, he had traveled to Japan, China, New Zealand, and South America. Five weeks before admission, he had traveled to Seattle, and three weeks before admission, he had traveled to Cape Cod, Massachusetts. He did not have any history of tick bites.
On examination, the patient appeared acutely ill. The temperature was 39°C, the pulse 84 beats per minute, and the blood pressure 145/80 mm Hg. The conjunctivas were clear, the neck was supple, and there were no enlarged cervical or axillary lymph nodes. There was no cutaneous rash. Inspiratory crackles were heard at the bases of the lungs. A grade 2 systolic ejection murmur was heard at the left sternal border. The abdomen was soft and nontender; the liver and spleen were not enlarged. The prostate was slightly enlarged and nontender. The cranial-nerve functions were preserved. Muscle strength was 5/5 in the arms and 4/5 in the legs. The deep-tendon reflexes were normal, as was sensation. The thighs were slightly tender to touch, and the patient was unsteady on his feet.
Laboratory values are shown in Table 1, Table 2, and Table 3. A chest radiograph showed an elevated left hemidiaphragm, which had been present on previous examinations, with adjacent subsegmental atelectasis. An electrocardiogram revealed a normal rhythm at a rate of 80 beats per minute, with T-wave abnormalities in the V3 and aVF leads. An ultrasonographic study showed normal kidneys, with no evidence of deep venous thrombosis in either leg.
Table 1. Blood Chemical Values.
Table 2. Hematologic Laboratory Values.
Table 3. Urinalysis.
Specimens of blood were obtained for culture and serologic testing, and treatment was begun with trimethoprim–sulfamethoxazole and gentamicin. Fluids were administered intravenously. The patient was admitted to the hospital.
On the second hospital day, the patient's temperature rose to 39°C. Laboratory test results are shown in Table 1 and Table 2. The medications were changed to ampicillin, levofloxacin, and metronidazole. The results of culture and serologic testing are shown in Table 4. A diagnostic test was performed.
Table 4. Results of Microbiologic and Serologic Tests.
Differential Diagnosis
Dr. Howard M. Heller: In light of this patient's urinary symptoms, the history of the benign prostatic hypertrophy, and the fact that common things happen commonly, I think urosepsis and, possibly, prostatitis make sense as initial concerns, even though the prostate gland was not tender. This patient's new cardiac murmur raises the possibility of endocarditis, which also would have been an appropriate initial concern.
Bacterial and Viral Infections
Because the blood cultures were negative and the patient's condition did not respond to broad-spectrum antibiotic therapy, we need to think about other possibilities, especially since the findings of leukopenia and lymphopenia in the blood smear in this case would not have been expected if the patient had had a fulminant bacterial infection. Viral illnesses, such as acute cytomegalovirus infection, can cause myositis and hepatitis with leukopenia but should not make someone as sick as the patient under discussion and usually should not cause disseminated intravascular coagulation. Also, viremia should be detectable with the antigenemia assay. Although we were told that the test for cytomegalovirus antigen was negative, we were not told specifically whether the blood or urine was examined. In persons with acute cytomegalovirus infection, the antigen is usually found in both.
Although a rapid influenza test was not listed in the case summary, I hope that it was part of the initial evaluation. The serologic examination showed only that the patient had had an infection with influenza at some time in the past. The brief illness that he had preceding admission with the rhinorrhea and the cough slightly increased my concern about influenza. However, influenza would probably not cause disseminated intravascular coagulation, and his illness did not occur during peak influenza season.
Parasitic Infections
The only parasitic infection that might cause this constellation of symptoms would be toxoplasmosis. Acute toxoplasmosis can cause myositis and hepatitis, but generally it does not result in such profound illness with high fevers, nor does it lead to leukopenia or disseminated intravascular coagulation. There is nothing in the history to suggest that the patient was immunocompromised, so I am going to dismiss the possibility of a fulminant fungal infection, such as cryptococcosis, which can cause a sepsis-like picture similar to that in this case.
Tick-Borne Diseases
When a patient presents with an acute febrile illness during the spring and summer, we must consider the tick-borne diseases. Diseases carried by the deer tick can occur anytime during warm weather, even during a warm spell in the middle of a New England winter. A few years ago, we saw cases of Lyme disease occurring in January after a one-week period of warm weather. However, infections from deer ticks peak in the spring, because that is primarily when the nymphal forms of the tick are feeding. Since the nymphal forms are tinier than the adult forms, they are able to remain on the human host for longer periods of time without detection, and they have a greater chance of transmitting the infection. Most patients with tick-borne diseases do not recall a tick bite.
Several ticks that are found in the United States can cause an illness similar to that of this patient. The deer tick (Ixodes scapularis) transmits three infections: Lyme borreliosis, human granulocytic ehrlichiosis, and babesiosis. The Lone Star tick (Amblyomma americanum), which is found in the southern United States, and not in New England except for a relict population in Narragansett Bay, is the main vector of monocytic ehrlichiosis due to Ehrlichia chaffeensis. The Dermacentor variabilis tick, also known as the dog tick or the wood tick, can transmit Rocky Mountain spotted fever or tularemia. The ornithodoros tick is the only soft-body tick that transmits infections to humans, specifically relapsing fever, but this does not occur in New England.
The nymphal form of the I. scapularis tick is about the size of a poppy seed, and is often not noticed by human hosts. When the nymphs are engorged, they look like blood blisters, but still often go unnoticed. The I. scapularis ticks transmit disease mostly in New England, but also in the upper Midwest. I believe the diagnosis in this case is likely to be an infection transmitted by I. scapularis, and I will focus my discussion on diseases transmitted by this tick. In Massachusetts, the hyperendemic areas are along the coast, especially Cape Cod, and in western Massachusetts in the Berkshires, but infections associated with I. scapularis have been reported in all areas of Massachusetts.
Although Lyme borreliosis is by far the most common tick-borne infection that occurs in North America, it is not likely to be the cause of this patient's illness, because it does not cause such a severe clinical illness, disseminated intravascular coagulation, or leukopenia. Babesia microti infects erythrocytes, and usually the most prominent hematologic finding in patients who are very ill is anemia. Patients who are elderly, immunocompromised, or asplenic are more likely to have severe disease, but this patient is neither immunocompromised nor asplenic. Thus, although babesiosis is possible, the lack of anemia argues against it.
Rocky Mountain spotted fever, caused by Rickettsia rickettsii, can also trigger a profound multiorgan illness, including disseminated intravascular coagulation. However, R. rickettsii causes endothelial infection, thus producing vasculitis at every site in the body where it causes disease. Rash and headache are common. This patient did not have a headache or prominent rash, so I am putting Rocky Mountain spotted fever much lower on my list of possible diagnoses than some other possibilities.
Tularemia is another tick-borne disease that should be considered. Several forms of tularemic illness are differentiated according to the way the infection is acquired. Tick-borne tularemia usually causes ulceroglandular disease, in which an ulcer and then an eschar develop at the site of the tick bite, and regional lymph-node enlargement occurs in the area proximal to the eschar. The absence of an eschar and lymphadenopathy would argue against tick-transmitted tularemia.
The last tick-borne diseases to consider are the ones that have been discovered most recently — namely, human monocytic ehrlichiosis, which is caused by E. chaffeensis, and human granulocytic ehrlichiosis. The agent of human granulocytic ehrlichiosis has been referred to as E. equi and E. phagocytophila, but is now called Anaplasma phagocytophilum. There is controversy regarding nomenclature for this agent and illness, but for the purposes of this discussion, I will refer to the agent as A. phagocytophilum and to the disease as human granulocytic ehrlichiosis.1 E. chaffeensis is transmitted by the Lone Star tick that is commonly found in the southern United States. Although this patient became ill shortly after a trip to coastal South Carolina, his travel to Cape Cod, as well as his residence in the Boston area, put him at continuous risk for being exposed to I. scapularis, which can transmit A. phagocytophilum as well as babesia and Borrelia burgdorferi, the agent of Lyme disease.
Human Granulocytic Ehrlichiosis
The incubation period of human granulocytic ehrlichiosis is 7 to 10 days after the tick bite; multiorgan system involvement is very common. Fever, headaches, myalgias, and malaise are seen in most patients, and the infection resembles that of any acute viral illness. Physicians who practice in areas where deer ticks are found need to think of human granulocytic ehrlichiosis if a patient from the area presents with a virus-like syndrome during tick season and has been at risk for tick bites. Nausea, vomiting, diarrhea, cough, arthralgias, stiff neck, and even mental status changes have been reported, but in fewer than 50 percent of patients.1 Respiratory failure, septic shock, rhabdomyolysis, and rash have also been reported, although more rarely. The mortality rate has varied; some reports have it at about 1 percent; some reports have it as high as 10 percent.
Typical laboratory findings can also mimic viral syndromes, specifically leukopenia, thrombocytopenia, and mild anemia.2 In the first week of the illness, the neutropenia is associated with a left shift and with lymphopenia. Lymphocytosis can occur in the second week, mimicking a viral syndrome. Elevation of the hepatic transaminases can be seen within the first week of the illness. In 20 to 80 percent of the cases, the morula of the organism can be seen in the circulating granulocytes.
Although A. phagocytophilum can cause profound multiorgan system disease, myositis has been reported only rarely. However, since the onset of the myalgias coincided with the onset of the fevers, I think the myositis is probably related to this febrile illness and is not related to the atorvastatin that the patient had been taking for three months.
The lack of response to trimethoprim–sulfamethoxazole, gentamicin, ampicillin, metronidazole, and levofloxacin is consistent with human granulocytic ehrlichiosis. Doxycycline is the treatment of choice, even in children, because human granulocytic ehrlichiosis is a potentially fatal disease. However, we must consider the possibility that this patient had two separate diseases. Since anaplasma, babesia, and B. burgdorferi are all transmitted by the same tick, patients may present with more than one infection. Myositis is seen more commonly with Lyme disease than with human granulocytic ehrlichiosis, so there may have been serologic evidence of concurrent infection with B. burgdorferi, but I believe most of this patient's illness is attributable to human granulocytic ehrlichiosis.
My diagnosis is human granulocytic ehrlichiosis caused by A. phagocytophilum infection. I believe that the diagnostic test was either a serologic test or possibly a careful review of the peripheral smear on which the morulae may have been seen. However, the test was probably serologic.
Dr. Nancy L. Harris (Pathology): Dr. Goodson, since you are the patient's primary care physician, would you comment on your thinking?
Dr. John D. Goodson (Medicine): I thought he had either an acute infection or a drug reaction; his urinalysis was slightly abnormal, he had rather profound weakness, and the level of creatine kinase was elevated. That evening we initiated antibiotic therapy that was directed at urosepsis, and we broadened the coverage the next morning. Because his clinical course was so atypical, I consulted a colleague in infectious disease, Dr. Ryan.
Dr. Edward T. Ryan (Infectious Disease): Taking into account the patient's severe influenza-like illness during tick season, the lack of response to a number of antibiotics, and the impressive leukopenia and thrombocytopenia, I thought the most likely diagnosis was ehrlichiosis. Although morulae were not seen on microscopical analysis of a smear of peripheral leukocytes, we recommended a course of doxycycline. Analysis of a smear of peripheral blood did not disclose evidence of babesiosis.
Dr. Harris: Dr. Telford, from the Harvard School of Public Health and the Tufts Veterinary School, performed the primary diagnostic testing in this case.
Clinical Diagnosis
Human granulocytic ehrlichiosis.
Dr. Howard M. Heller's Diagnosis
A. phagocytophilum infection (human granulocytic ehrlichiosis).
Pathological Discussion
Dr. Sam R. Telford III: As with rickettsial infections, serologic testing does not usually help in the diagnosis of acute ehrlichiosis, because, typically, seroconversion does not occur until at least three weeks into the illness. For this reason, we proceeded to perform a polymerase-chain-reaction (PCR) assay.
We used a primer set targeting a 247-bp portion of the 16S ribosomal DNA (Ehr521/747), which was originally designed to detect and discriminate between A. phagocytophilum and E. chaffeensis. We tested blood samples obtained on the second and third hospital days, and both were positive for A. phagocytophilum. We confirmed this result by amplifying a larger piece of the 16S ribosomal DNA (nearly 1000 base pairs, primers BacA/1448r) and sequencing the amplification product; this was 100 percent homologous to sequences of the human granulocytic ehrlichiosis agent that are deposited at GenBank.
After the PCR analysis, we also used mouse inoculation to recover the agent; recovery of the agent remains the microbiologic gold standard, regardless of the result of PCR, and we always use at least two methods in order to confirm the diagnosis.3 DBA/2 mice were inoculated with blood from the first sample, and on day 7 after inoculation, the mice showed the characteristic inclusions within the white cells. Dr. Branda will discuss the morphologic features.
Dr. John A. Branda: In some cases, a preliminary diagnosis of human granulocytic ehrlichiosis can be made during the acute illness by examining the patient's peripheral-blood or buffy-coat smear.4 A. phagocytophilum is an obligate intracellular, gram-negative, pleomorphic, coccoid bacterium.5 Its favored host cell is the granulocyte, especially the neutrophil.6 The organism enters the host granulocyte by endocytosis and survives within an endocytic compartment by inhibiting lysosomal fusion.5 It reproduces by binary fission, creating a membrane-bound cluster of organisms that is visible by light microscopy as an intracytoplasmic inclusion.4,7 These inclusions, termed morulae, can be visualized with the use of routine Romanovsky's stains, such as Wright's stain or Giemsa stain.7,8,9 The organisms within the vacuole are a deep blue or purple color, which should be different from that of the host cell's nuclear chromatin,4,7 whereas the surrounding vacuolar fluid appears clear. This often results in the classic stippled appearance,7,8 although many other morphologic configurations are possible.8
On the peripheral-blood smear obtained from this patient, rare intracytoplasmic inclusions were identified in granulocytes, which suggested human granulocytic ehrlichiosis morulae, but none were diagnostic. Unfortunately, examination of the peripheral-blood smear is an insensitive test4,10,11,12; sensitivity may depend on the severity of illness10 and the timing of the blood-smear preparation (with perhaps a greater sensitivity in the first week of illness).4 Specificity also varies and largely depends on the examiner's experience.7,9
The peripheral-blood smears from a mouse from the inbred strain DBA that had been inoculated intraperitoneally with 0.2 ml of peripheral blood from the patient showed characteristic human granulocytic ehrlichiosis morulae (Figure 1) in a small proportion of neutrophils (less than 1 percent), which was interpreted as evidence of human granulocytic ehrlichiosis bacteremia (in both the mouse and the patient).
Figure 1. Peripheral-Blood Smears Prepared from a DBA Mouse Previously Inoculated with Blood from the Patient (Giemsa Stain).
Panel A shows an intracytoplasmic human granulocytic ehrlichiosis morula (arrow) with a stippled morphologic pattern. Although the morula and the adjacent neutrophil chromatin overlap, they have different staining characteristics. Panel B shows an intracytoplasmic human granulocytic ehrlichiosis morula (arrow) composed of two deep-blue, crescent-shaped bodies within a clear vacuole.
Dr. Harris: Dr. Goodson, would you comment on treatment and follow-up?
Dr. Goodson: After Dr. Ryan suggested the diagnosis of human granulocytic ehrlichiosis, we switched the patient's medication to doxycycline. By the third hospital day, his fever was almost gone and he agreed to being discharged. When I saw him in my office three days later, he was feeling better and was able to get up from a squatting position. Two months later, I restarted the atorvastatin treatment, and he has done well.
We also have the good fortune of having the patient here. Dr. Bloom, in retrospect, when do you think you started feeling the manifestations of this condition, and how are you feeling now?
Dr. David Bloom: By the Monday afternoon after returning from a weekend conference at Hilton Head, I had a deep ache in both thighs. I thought I had the flu and got into bed. The aching worsened to such a degree that I could not get out of bed. At that point, my wife had me taken to the emergency room.
When Dr. Ryan gave me the diagnosis, I was surprised, because I was not aware of having been bitten by a tick. By the time I was discharged, my muscles did not hurt, but they were still weak. Within two or three weeks, I went back to playing tennis. However, despite a vigorous exercise program, I have residual weakness and intermittent pain in the thighs.
Dr. Goodson: I wonder if the myositis is a result of the infection or whether it is from something else, such as the atorvastatin.
Dr. Ryan: Some people with ehrlichiosis have concomitant babesiosis, acute Lyme disease, or both. Since serologic assays for Lyme disease are often negative during acute disease, we elected to treat this patient with a 14-day course of doxycycline under the assumption that we would also be treating incipient Lyme disease. There are reports of rhabdomyolysis during ehrlichiosis,13,14,15 and we think that perhaps the combination of the ehrlichiosis, possibly the atorvastatin, and the low-level disseminated intravascular coagulation of severe febrile illness may have contributed to a prominent myositis of the thighs.
Dr. Goodson: A serum sample obtained during the convalescent period was positive for ehrlichiosis (>1:64 IgG with the use of HL60 cultivated human granulocytic ehrlichiosis antigen in an indirect immunofluorescence procedure). The tests for antibodies for babesiosis and Lyme disease during the acute and convalescent phases of the illness were negative.
Anatomical Diagnosis
Human granulocytic ehrlichiosis (A. phagocytophilum infection).
Supported by a grant (AI 39002, to Dr. Telford) from the National Institutes of Health.
Source Information
From the Infectious Disease Unit (H.M.H.) and the Department of Pathology (J.A.B.), Massachusetts General Hospital, Boston; the Division of Infectious Diseases, Tufts University School of Veterinary Medicine, North Grafton, Mass. (S.R.T.); the Departments of Medicine (H.M.H.) and Pathology (J.A.B.), Harvard Medical School, Boston; and the Laboratory of Public Health Entomology, Harvard School of Public Health, Boston (S.R.T.).
References
Dumler JS, Barbet AF, Bekker CP, et al. Reorganization of genera in the families Rickettsiaceae and Anaplasmataceae in the order Rickettsiales: unification of some species of Ehrlichia with Anaplasma, Cowdria with Ehrlichia and Ehrlichia with Neorickettsia, descriptions of six new species combinations and designations of Ehrlichia equi and `HGE agent' as subjective synonyms of Ehrlichia phagocytophila. Int J Syst Evol Microbiol 2001;51:2145-2165.
Bakken JS, Aguero-Rosenfeld ME, Tilden RL, et al. Serial measurements of hematologic counts during the active phase of human granulocytic ehrlichiosis. Clin Infect Dis 2001;32:862-870.
Telford SR III, Dawson JE, Katavolos P, Warner CK, Kolbert CP, Persing DH. Perpetuation of the agent of human granulocytic ehrlichiosis in a deer tick-rodent cycle. Proc Natl Acad Sci U S A 1996;93:6209-6214.
Bakken JS, Dumler JS. Human granulocytic ehrlichiosis. Clin Infect Dis 2000;31:554-560.
Rikihisa Y. Mechanisms to create a safe haven by members of the family Anaplasmataceae. Ann N Y Acad Sci 2003;990:548-555.
Bakken JS, Dumler JS, Chen S-M, Eckman MR, Van Etta LL, Walker DH. Human granulocytic ehrlichiosis in the upper Midwest United States: a new species emerging? JAMA 1994;272:212-218.
Dumler JS, Bakken JS. Human ehrlichioses: newly recognized infections transmitted by ticks. Annu Rev Med 1998;49:201-213.
Dumler JS, Walker DH. Ehrlichial infections. In: Connor DH, Chandler FW, Schwartz DA, Manz HJ, Lack EE, eds. Pathology of infectious diseases. Stamford, Conn.: Appleton & Lange, 1997:441-5.
Walker DH, Dumler JS. Human monocytic and granulocytic ehrlichioses: discovery and diagnosis of emerging tick-borne infections and the critical role of the pathologist. Arch Pathol Lab Med 1997;121:785-791.
Bakken JS, Krueth J, Wilson-Nordskog C, Tilden RL, Asanovich K, Dumler JS. Clinical and laboratory characteristics of human granulocytic ehrlichiosis. JAMA 1996;275:199-205.
Human granulocytic ehrlichiosis -- New York, 1995. MMWR Morb Mortal Wkly Rep 1995;44:593-595.
Aguero-Rosenfeld ME, Horowitz HW, Wormser GP, et al. Human granulocytic ehrlichiosis: a case series from a medical center in New York State. Ann Intern Med 1996;125:904-908.
Javed MZ, Srivastava M, Zhang S, Kandathil M. Concurrent babesiosis and ehrlichiosis in an elderly host. Mayo Clin Proc 2001;76:563-565.
Shea KW, Calio AJ, Klein NC, Cunha BA. Ehrlichia equi infection associated with rhabdomyolysis. Clin Infect Dis 1996;22:605-605.
Shea KW, Calio AJ, Klein NC, Cunha BA. Rhabdomyolysis associated with Ehrlichia chaffeensis infection. Clin Infect Dis 1995;21:1056-1057.(Howard M. Heller, M.D., S)
A 73-year-old man was admitted to the hospital in June because of fever, weakness, and pain in the muscles of his legs.
He had been well until seven days before admission, when a dry cough and rhinorrhea developed and lasted for two days. Six days before admission, he had traveled from Boston to coastal South Carolina to attend a conference. Three days before admission, he had had an intense aching pain in both thighs and a temperature as high as 39.4°C. Shortly after that, generalized weakness developed, and he began to have an intense aching pain in the muscles of his legs. He had increased urinary frequency, with one episode of urinary incontinence. His wife brought him to the emergency department of this hospital. He had no history of shortness of breath, chest pain, production of sputum, nausea, vomiting, or change in bowel habits. Three months before admission, atorvastatin (5 mg daily) had been prescribed for the treatment of elevated serum cholesterol, along with vitamin B complex and folic acid for treatment of an elevated plasma homocysteine level. The patient had stopped taking atorvastatin four days before admission.
The patient's medical history included benign enlargement of the prostate. He had had episodes of chronic low back pain, which radiated from low in his back to his buttocks. He had partial bilateral hearing loss. He had undergone arthroscopic knee surgery and bilateral cataract extractions in the distant past. The patient was a practicing physician. Many years earlier, he had traveled to Japan, China, New Zealand, and South America. Five weeks before admission, he had traveled to Seattle, and three weeks before admission, he had traveled to Cape Cod, Massachusetts. He did not have any history of tick bites.
On examination, the patient appeared acutely ill. The temperature was 39°C, the pulse 84 beats per minute, and the blood pressure 145/80 mm Hg. The conjunctivas were clear, the neck was supple, and there were no enlarged cervical or axillary lymph nodes. There was no cutaneous rash. Inspiratory crackles were heard at the bases of the lungs. A grade 2 systolic ejection murmur was heard at the left sternal border. The abdomen was soft and nontender; the liver and spleen were not enlarged. The prostate was slightly enlarged and nontender. The cranial-nerve functions were preserved. Muscle strength was 5/5 in the arms and 4/5 in the legs. The deep-tendon reflexes were normal, as was sensation. The thighs were slightly tender to touch, and the patient was unsteady on his feet.
Laboratory values are shown in Table 1, Table 2, and Table 3. A chest radiograph showed an elevated left hemidiaphragm, which had been present on previous examinations, with adjacent subsegmental atelectasis. An electrocardiogram revealed a normal rhythm at a rate of 80 beats per minute, with T-wave abnormalities in the V3 and aVF leads. An ultrasonographic study showed normal kidneys, with no evidence of deep venous thrombosis in either leg.
Table 1. Blood Chemical Values.
Table 2. Hematologic Laboratory Values.
Table 3. Urinalysis.
Specimens of blood were obtained for culture and serologic testing, and treatment was begun with trimethoprim–sulfamethoxazole and gentamicin. Fluids were administered intravenously. The patient was admitted to the hospital.
On the second hospital day, the patient's temperature rose to 39°C. Laboratory test results are shown in Table 1 and Table 2. The medications were changed to ampicillin, levofloxacin, and metronidazole. The results of culture and serologic testing are shown in Table 4. A diagnostic test was performed.
Table 4. Results of Microbiologic and Serologic Tests.
Differential Diagnosis
Dr. Howard M. Heller: In light of this patient's urinary symptoms, the history of the benign prostatic hypertrophy, and the fact that common things happen commonly, I think urosepsis and, possibly, prostatitis make sense as initial concerns, even though the prostate gland was not tender. This patient's new cardiac murmur raises the possibility of endocarditis, which also would have been an appropriate initial concern.
Bacterial and Viral Infections
Because the blood cultures were negative and the patient's condition did not respond to broad-spectrum antibiotic therapy, we need to think about other possibilities, especially since the findings of leukopenia and lymphopenia in the blood smear in this case would not have been expected if the patient had had a fulminant bacterial infection. Viral illnesses, such as acute cytomegalovirus infection, can cause myositis and hepatitis with leukopenia but should not make someone as sick as the patient under discussion and usually should not cause disseminated intravascular coagulation. Also, viremia should be detectable with the antigenemia assay. Although we were told that the test for cytomegalovirus antigen was negative, we were not told specifically whether the blood or urine was examined. In persons with acute cytomegalovirus infection, the antigen is usually found in both.
Although a rapid influenza test was not listed in the case summary, I hope that it was part of the initial evaluation. The serologic examination showed only that the patient had had an infection with influenza at some time in the past. The brief illness that he had preceding admission with the rhinorrhea and the cough slightly increased my concern about influenza. However, influenza would probably not cause disseminated intravascular coagulation, and his illness did not occur during peak influenza season.
Parasitic Infections
The only parasitic infection that might cause this constellation of symptoms would be toxoplasmosis. Acute toxoplasmosis can cause myositis and hepatitis, but generally it does not result in such profound illness with high fevers, nor does it lead to leukopenia or disseminated intravascular coagulation. There is nothing in the history to suggest that the patient was immunocompromised, so I am going to dismiss the possibility of a fulminant fungal infection, such as cryptococcosis, which can cause a sepsis-like picture similar to that in this case.
Tick-Borne Diseases
When a patient presents with an acute febrile illness during the spring and summer, we must consider the tick-borne diseases. Diseases carried by the deer tick can occur anytime during warm weather, even during a warm spell in the middle of a New England winter. A few years ago, we saw cases of Lyme disease occurring in January after a one-week period of warm weather. However, infections from deer ticks peak in the spring, because that is primarily when the nymphal forms of the tick are feeding. Since the nymphal forms are tinier than the adult forms, they are able to remain on the human host for longer periods of time without detection, and they have a greater chance of transmitting the infection. Most patients with tick-borne diseases do not recall a tick bite.
Several ticks that are found in the United States can cause an illness similar to that of this patient. The deer tick (Ixodes scapularis) transmits three infections: Lyme borreliosis, human granulocytic ehrlichiosis, and babesiosis. The Lone Star tick (Amblyomma americanum), which is found in the southern United States, and not in New England except for a relict population in Narragansett Bay, is the main vector of monocytic ehrlichiosis due to Ehrlichia chaffeensis. The Dermacentor variabilis tick, also known as the dog tick or the wood tick, can transmit Rocky Mountain spotted fever or tularemia. The ornithodoros tick is the only soft-body tick that transmits infections to humans, specifically relapsing fever, but this does not occur in New England.
The nymphal form of the I. scapularis tick is about the size of a poppy seed, and is often not noticed by human hosts. When the nymphs are engorged, they look like blood blisters, but still often go unnoticed. The I. scapularis ticks transmit disease mostly in New England, but also in the upper Midwest. I believe the diagnosis in this case is likely to be an infection transmitted by I. scapularis, and I will focus my discussion on diseases transmitted by this tick. In Massachusetts, the hyperendemic areas are along the coast, especially Cape Cod, and in western Massachusetts in the Berkshires, but infections associated with I. scapularis have been reported in all areas of Massachusetts.
Although Lyme borreliosis is by far the most common tick-borne infection that occurs in North America, it is not likely to be the cause of this patient's illness, because it does not cause such a severe clinical illness, disseminated intravascular coagulation, or leukopenia. Babesia microti infects erythrocytes, and usually the most prominent hematologic finding in patients who are very ill is anemia. Patients who are elderly, immunocompromised, or asplenic are more likely to have severe disease, but this patient is neither immunocompromised nor asplenic. Thus, although babesiosis is possible, the lack of anemia argues against it.
Rocky Mountain spotted fever, caused by Rickettsia rickettsii, can also trigger a profound multiorgan illness, including disseminated intravascular coagulation. However, R. rickettsii causes endothelial infection, thus producing vasculitis at every site in the body where it causes disease. Rash and headache are common. This patient did not have a headache or prominent rash, so I am putting Rocky Mountain spotted fever much lower on my list of possible diagnoses than some other possibilities.
Tularemia is another tick-borne disease that should be considered. Several forms of tularemic illness are differentiated according to the way the infection is acquired. Tick-borne tularemia usually causes ulceroglandular disease, in which an ulcer and then an eschar develop at the site of the tick bite, and regional lymph-node enlargement occurs in the area proximal to the eschar. The absence of an eschar and lymphadenopathy would argue against tick-transmitted tularemia.
The last tick-borne diseases to consider are the ones that have been discovered most recently — namely, human monocytic ehrlichiosis, which is caused by E. chaffeensis, and human granulocytic ehrlichiosis. The agent of human granulocytic ehrlichiosis has been referred to as E. equi and E. phagocytophila, but is now called Anaplasma phagocytophilum. There is controversy regarding nomenclature for this agent and illness, but for the purposes of this discussion, I will refer to the agent as A. phagocytophilum and to the disease as human granulocytic ehrlichiosis.1 E. chaffeensis is transmitted by the Lone Star tick that is commonly found in the southern United States. Although this patient became ill shortly after a trip to coastal South Carolina, his travel to Cape Cod, as well as his residence in the Boston area, put him at continuous risk for being exposed to I. scapularis, which can transmit A. phagocytophilum as well as babesia and Borrelia burgdorferi, the agent of Lyme disease.
Human Granulocytic Ehrlichiosis
The incubation period of human granulocytic ehrlichiosis is 7 to 10 days after the tick bite; multiorgan system involvement is very common. Fever, headaches, myalgias, and malaise are seen in most patients, and the infection resembles that of any acute viral illness. Physicians who practice in areas where deer ticks are found need to think of human granulocytic ehrlichiosis if a patient from the area presents with a virus-like syndrome during tick season and has been at risk for tick bites. Nausea, vomiting, diarrhea, cough, arthralgias, stiff neck, and even mental status changes have been reported, but in fewer than 50 percent of patients.1 Respiratory failure, septic shock, rhabdomyolysis, and rash have also been reported, although more rarely. The mortality rate has varied; some reports have it at about 1 percent; some reports have it as high as 10 percent.
Typical laboratory findings can also mimic viral syndromes, specifically leukopenia, thrombocytopenia, and mild anemia.2 In the first week of the illness, the neutropenia is associated with a left shift and with lymphopenia. Lymphocytosis can occur in the second week, mimicking a viral syndrome. Elevation of the hepatic transaminases can be seen within the first week of the illness. In 20 to 80 percent of the cases, the morula of the organism can be seen in the circulating granulocytes.
Although A. phagocytophilum can cause profound multiorgan system disease, myositis has been reported only rarely. However, since the onset of the myalgias coincided with the onset of the fevers, I think the myositis is probably related to this febrile illness and is not related to the atorvastatin that the patient had been taking for three months.
The lack of response to trimethoprim–sulfamethoxazole, gentamicin, ampicillin, metronidazole, and levofloxacin is consistent with human granulocytic ehrlichiosis. Doxycycline is the treatment of choice, even in children, because human granulocytic ehrlichiosis is a potentially fatal disease. However, we must consider the possibility that this patient had two separate diseases. Since anaplasma, babesia, and B. burgdorferi are all transmitted by the same tick, patients may present with more than one infection. Myositis is seen more commonly with Lyme disease than with human granulocytic ehrlichiosis, so there may have been serologic evidence of concurrent infection with B. burgdorferi, but I believe most of this patient's illness is attributable to human granulocytic ehrlichiosis.
My diagnosis is human granulocytic ehrlichiosis caused by A. phagocytophilum infection. I believe that the diagnostic test was either a serologic test or possibly a careful review of the peripheral smear on which the morulae may have been seen. However, the test was probably serologic.
Dr. Nancy L. Harris (Pathology): Dr. Goodson, since you are the patient's primary care physician, would you comment on your thinking?
Dr. John D. Goodson (Medicine): I thought he had either an acute infection or a drug reaction; his urinalysis was slightly abnormal, he had rather profound weakness, and the level of creatine kinase was elevated. That evening we initiated antibiotic therapy that was directed at urosepsis, and we broadened the coverage the next morning. Because his clinical course was so atypical, I consulted a colleague in infectious disease, Dr. Ryan.
Dr. Edward T. Ryan (Infectious Disease): Taking into account the patient's severe influenza-like illness during tick season, the lack of response to a number of antibiotics, and the impressive leukopenia and thrombocytopenia, I thought the most likely diagnosis was ehrlichiosis. Although morulae were not seen on microscopical analysis of a smear of peripheral leukocytes, we recommended a course of doxycycline. Analysis of a smear of peripheral blood did not disclose evidence of babesiosis.
Dr. Harris: Dr. Telford, from the Harvard School of Public Health and the Tufts Veterinary School, performed the primary diagnostic testing in this case.
Clinical Diagnosis
Human granulocytic ehrlichiosis.
Dr. Howard M. Heller's Diagnosis
A. phagocytophilum infection (human granulocytic ehrlichiosis).
Pathological Discussion
Dr. Sam R. Telford III: As with rickettsial infections, serologic testing does not usually help in the diagnosis of acute ehrlichiosis, because, typically, seroconversion does not occur until at least three weeks into the illness. For this reason, we proceeded to perform a polymerase-chain-reaction (PCR) assay.
We used a primer set targeting a 247-bp portion of the 16S ribosomal DNA (Ehr521/747), which was originally designed to detect and discriminate between A. phagocytophilum and E. chaffeensis. We tested blood samples obtained on the second and third hospital days, and both were positive for A. phagocytophilum. We confirmed this result by amplifying a larger piece of the 16S ribosomal DNA (nearly 1000 base pairs, primers BacA/1448r) and sequencing the amplification product; this was 100 percent homologous to sequences of the human granulocytic ehrlichiosis agent that are deposited at GenBank.
After the PCR analysis, we also used mouse inoculation to recover the agent; recovery of the agent remains the microbiologic gold standard, regardless of the result of PCR, and we always use at least two methods in order to confirm the diagnosis.3 DBA/2 mice were inoculated with blood from the first sample, and on day 7 after inoculation, the mice showed the characteristic inclusions within the white cells. Dr. Branda will discuss the morphologic features.
Dr. John A. Branda: In some cases, a preliminary diagnosis of human granulocytic ehrlichiosis can be made during the acute illness by examining the patient's peripheral-blood or buffy-coat smear.4 A. phagocytophilum is an obligate intracellular, gram-negative, pleomorphic, coccoid bacterium.5 Its favored host cell is the granulocyte, especially the neutrophil.6 The organism enters the host granulocyte by endocytosis and survives within an endocytic compartment by inhibiting lysosomal fusion.5 It reproduces by binary fission, creating a membrane-bound cluster of organisms that is visible by light microscopy as an intracytoplasmic inclusion.4,7 These inclusions, termed morulae, can be visualized with the use of routine Romanovsky's stains, such as Wright's stain or Giemsa stain.7,8,9 The organisms within the vacuole are a deep blue or purple color, which should be different from that of the host cell's nuclear chromatin,4,7 whereas the surrounding vacuolar fluid appears clear. This often results in the classic stippled appearance,7,8 although many other morphologic configurations are possible.8
On the peripheral-blood smear obtained from this patient, rare intracytoplasmic inclusions were identified in granulocytes, which suggested human granulocytic ehrlichiosis morulae, but none were diagnostic. Unfortunately, examination of the peripheral-blood smear is an insensitive test4,10,11,12; sensitivity may depend on the severity of illness10 and the timing of the blood-smear preparation (with perhaps a greater sensitivity in the first week of illness).4 Specificity also varies and largely depends on the examiner's experience.7,9
The peripheral-blood smears from a mouse from the inbred strain DBA that had been inoculated intraperitoneally with 0.2 ml of peripheral blood from the patient showed characteristic human granulocytic ehrlichiosis morulae (Figure 1) in a small proportion of neutrophils (less than 1 percent), which was interpreted as evidence of human granulocytic ehrlichiosis bacteremia (in both the mouse and the patient).
Figure 1. Peripheral-Blood Smears Prepared from a DBA Mouse Previously Inoculated with Blood from the Patient (Giemsa Stain).
Panel A shows an intracytoplasmic human granulocytic ehrlichiosis morula (arrow) with a stippled morphologic pattern. Although the morula and the adjacent neutrophil chromatin overlap, they have different staining characteristics. Panel B shows an intracytoplasmic human granulocytic ehrlichiosis morula (arrow) composed of two deep-blue, crescent-shaped bodies within a clear vacuole.
Dr. Harris: Dr. Goodson, would you comment on treatment and follow-up?
Dr. Goodson: After Dr. Ryan suggested the diagnosis of human granulocytic ehrlichiosis, we switched the patient's medication to doxycycline. By the third hospital day, his fever was almost gone and he agreed to being discharged. When I saw him in my office three days later, he was feeling better and was able to get up from a squatting position. Two months later, I restarted the atorvastatin treatment, and he has done well.
We also have the good fortune of having the patient here. Dr. Bloom, in retrospect, when do you think you started feeling the manifestations of this condition, and how are you feeling now?
Dr. David Bloom: By the Monday afternoon after returning from a weekend conference at Hilton Head, I had a deep ache in both thighs. I thought I had the flu and got into bed. The aching worsened to such a degree that I could not get out of bed. At that point, my wife had me taken to the emergency room.
When Dr. Ryan gave me the diagnosis, I was surprised, because I was not aware of having been bitten by a tick. By the time I was discharged, my muscles did not hurt, but they were still weak. Within two or three weeks, I went back to playing tennis. However, despite a vigorous exercise program, I have residual weakness and intermittent pain in the thighs.
Dr. Goodson: I wonder if the myositis is a result of the infection or whether it is from something else, such as the atorvastatin.
Dr. Ryan: Some people with ehrlichiosis have concomitant babesiosis, acute Lyme disease, or both. Since serologic assays for Lyme disease are often negative during acute disease, we elected to treat this patient with a 14-day course of doxycycline under the assumption that we would also be treating incipient Lyme disease. There are reports of rhabdomyolysis during ehrlichiosis,13,14,15 and we think that perhaps the combination of the ehrlichiosis, possibly the atorvastatin, and the low-level disseminated intravascular coagulation of severe febrile illness may have contributed to a prominent myositis of the thighs.
Dr. Goodson: A serum sample obtained during the convalescent period was positive for ehrlichiosis (>1:64 IgG with the use of HL60 cultivated human granulocytic ehrlichiosis antigen in an indirect immunofluorescence procedure). The tests for antibodies for babesiosis and Lyme disease during the acute and convalescent phases of the illness were negative.
Anatomical Diagnosis
Human granulocytic ehrlichiosis (A. phagocytophilum infection).
Supported by a grant (AI 39002, to Dr. Telford) from the National Institutes of Health.
Source Information
From the Infectious Disease Unit (H.M.H.) and the Department of Pathology (J.A.B.), Massachusetts General Hospital, Boston; the Division of Infectious Diseases, Tufts University School of Veterinary Medicine, North Grafton, Mass. (S.R.T.); the Departments of Medicine (H.M.H.) and Pathology (J.A.B.), Harvard Medical School, Boston; and the Laboratory of Public Health Entomology, Harvard School of Public Health, Boston (S.R.T.).
References
Dumler JS, Barbet AF, Bekker CP, et al. Reorganization of genera in the families Rickettsiaceae and Anaplasmataceae in the order Rickettsiales: unification of some species of Ehrlichia with Anaplasma, Cowdria with Ehrlichia and Ehrlichia with Neorickettsia, descriptions of six new species combinations and designations of Ehrlichia equi and `HGE agent' as subjective synonyms of Ehrlichia phagocytophila. Int J Syst Evol Microbiol 2001;51:2145-2165.
Bakken JS, Aguero-Rosenfeld ME, Tilden RL, et al. Serial measurements of hematologic counts during the active phase of human granulocytic ehrlichiosis. Clin Infect Dis 2001;32:862-870.
Telford SR III, Dawson JE, Katavolos P, Warner CK, Kolbert CP, Persing DH. Perpetuation of the agent of human granulocytic ehrlichiosis in a deer tick-rodent cycle. Proc Natl Acad Sci U S A 1996;93:6209-6214.
Bakken JS, Dumler JS. Human granulocytic ehrlichiosis. Clin Infect Dis 2000;31:554-560.
Rikihisa Y. Mechanisms to create a safe haven by members of the family Anaplasmataceae. Ann N Y Acad Sci 2003;990:548-555.
Bakken JS, Dumler JS, Chen S-M, Eckman MR, Van Etta LL, Walker DH. Human granulocytic ehrlichiosis in the upper Midwest United States: a new species emerging? JAMA 1994;272:212-218.
Dumler JS, Bakken JS. Human ehrlichioses: newly recognized infections transmitted by ticks. Annu Rev Med 1998;49:201-213.
Dumler JS, Walker DH. Ehrlichial infections. In: Connor DH, Chandler FW, Schwartz DA, Manz HJ, Lack EE, eds. Pathology of infectious diseases. Stamford, Conn.: Appleton & Lange, 1997:441-5.
Walker DH, Dumler JS. Human monocytic and granulocytic ehrlichioses: discovery and diagnosis of emerging tick-borne infections and the critical role of the pathologist. Arch Pathol Lab Med 1997;121:785-791.
Bakken JS, Krueth J, Wilson-Nordskog C, Tilden RL, Asanovich K, Dumler JS. Clinical and laboratory characteristics of human granulocytic ehrlichiosis. JAMA 1996;275:199-205.
Human granulocytic ehrlichiosis -- New York, 1995. MMWR Morb Mortal Wkly Rep 1995;44:593-595.
Aguero-Rosenfeld ME, Horowitz HW, Wormser GP, et al. Human granulocytic ehrlichiosis: a case series from a medical center in New York State. Ann Intern Med 1996;125:904-908.
Javed MZ, Srivastava M, Zhang S, Kandathil M. Concurrent babesiosis and ehrlichiosis in an elderly host. Mayo Clin Proc 2001;76:563-565.
Shea KW, Calio AJ, Klein NC, Cunha BA. Ehrlichia equi infection associated with rhabdomyolysis. Clin Infect Dis 1996;22:605-605.
Shea KW, Calio AJ, Klein NC, Cunha BA. Rhabdomyolysis associated with Ehrlichia chaffeensis infection. Clin Infect Dis 1995;21:1056-1057.(Howard M. Heller, M.D., S)