Abstract
Background
Human granulocytic anaplasmosis (HGA) is an acute tick-borne infectious disease with increasing morbidity and mortality, but is rarely considered in clinical practice. Because human-to-human transfusion or nosocomial transmission can occur, diagnosis is difficult when the history of tick bites is not clear.
Methods
We present clinical features and laboratory data of HGA patients who had no clear tick bite history.
Results
All patients in the study presented with a high fever, petechiae, purpura, nose bleeding and leukopenia, and patients had abnormally high levels of serum ferritin and C-reactive protein. Morulae in leukocytes were observed in three patients. Foamy histiocytes and slight erythrophagocytic activity were only found in severely ill patients.
Conclusion
In patients with fever and thrombocytopenia in whom no other diagnosis is evident on clinical assessment, HGA should be considered in the differential diagnosis, and tested for serologically if possible. For patients in whom the diagnosis of HGA is possible, and to whom tetracyclines can safely be given, it is apparent that these drugs hasten recovery and improve the prognosis.
Background
Human granulocytic anaplasmosis (HGA) is an acute infectious disease caused by Anaplasma phagocytophilum (also called Ehrlichia phagocytophila, Ehrlichia equi and the human granulocytic ehrlichiosis agent), which is an emerging tick-borne pathogen in humans and animals worldwide. 1 Nowadays, HGA is the third most common vector-borne infection in the USA and Europe, and is increasingly recognised as an important vector-borne infection in Asia. In USA, the seroprevalence is in the range of 0.6–14.9%, 2 in 3.5% of blood donors and 4.2% of whom reported having been bitten by a tick. 3 In Europe, the seroprevalence is in the range of 1.4–21%. 1 In Asia, 1.8% of Korean patients with febrile illness had antibodies to A. phagocytophilum. 4 The first HGA case in China was found in Anhui Province in 2006. 5 Since then, at least 557 people, distributed across provinces Hubei, Henan, Anhui, Shandong, Helongjiang, Inner Mongolia, Xingjiang and Tianjin, have been found to be infected and 18 patients died in the past 5 years. 5 Seroprevalence studies showed that 8.8% of farm workers had positive antibodies in northern China, 6 and these were detected in 20% of those at high risk of tick exposure in central and southeastern China. 7 Recently, the morbidity and mortality associated with HGA in China has led to public health concern. 8
The clinical presentation of HGA consists of high fever (>38.5℃), headache, malaise, myalgia and/or arthralgia, and is often accompanied by leukopenia, thrombocytopenia and increased hepatic enzymes. 9 Clinically, the diagnosis of HGA may be difficult as non-specific symptoms, such as nausea, abdominal pain, diarrhoea and cough, proliferate. The combination of high fever and thrombocytopenia is found in several other diagnoses, including dengue, leptospirosis and haemorrhagic fever. It is suggested that HGA may be a major cause of unexplained fever during the tick season. 10 However, it is rarely considered a possible pathogen in clinical practice, and its testing is not listed for routine diagnostic assessment. Thus, the actual number of cases may be considerably underestimated. Misdiagnosis and delayed antibiotic therapy will result in progressive deterioration of the patients. In China, mortality for HGA is much higher than that observed in the USA and Europe.11,12 The estimated HGA case fatality rate is believed to be low in China (range, 0.5–1%); but infection may be severe in the elderly or those immunocompromised. 13
Active antibiotic therapy is advocated for all patients with confirmed HGA. Therefore, aetiology analyses are important, especially as HGA testing is not included in routine laboratory detection. In China, A. phagocytophilum is transmitted by the tick, Ixodes persulcatus. 14 A history of prior tick bite is therefore an important clue, but most do not recall being bitten.15,16 Recently, nosocomial transmission has been documented through blood transfusion respiratory secretions.5,17 In our study, among sera of 133 patients with fever and thrombocytopenia, we found 30 infected by HGA.
Patients
From March 2009 to September 2011, 133 patients with fever (>38.0℃) and thrombocytopenia (platelet counts <100 × 109 /L) were collected. Age, sex, occupation, location of residence, history of tick bite and/or animal contact, prior disease, family disease history and clinical features, including signs and symptoms, laboratory data and outcomes were collected for each patient. This research was approved by the Ethics Committee of Shantou University Medical College and informed consent was obtained from each subject prior to the study.
Methods
To determine HGA, patients were tested for serum IgG of A. phagocytophilum using an indirect immunofluorescence assay (IFA) kit (China Center for Disease Control and Prevention [CCDC]). A confirmed case of HGA was defined in patients with one of the following diagnostic results: a four-fold change in antibody titers to Anaplasma species antigen in the acute and recovery phase serum samples, or a single positive serum antibody IgG titre ≥1: 256 by IFA. To exclude other infections, reverse transcription (RT)–PCR for nucleic acids of Filovirus, Banzi virus, Paramyxo virus, Hanta virus, Alphavirus, Coxsackievirus, Respiratory syncytial virus, Adenovirus, Chlamydia species, Ehrlichia species, Rickettsia species, and Orientia tsutsugamushi were performed. Simultaneous infection with A. phagocytophilum and B. burgdorferi in humans has been reported.18,19 Due to the possibility of co-infection with different tick-borne pathogens, we included additional tests for Babesia IgG and IgM antibodies, and Lyme IgG and IgM antibodies.
Blood and bone marrow (BM) smears were obtained, stained with May-Grünwald-Giemsa and examined for the presence of morulae within the cytoplasm of neutrophils. The presence of morulae in neutrophils was assessed by two haemotopathologists. All patients’ laboratory biochemical tests were determined with a HITACHI 7170A biochemical analyser. Complete blood counts ([CBC], including haemoglobin [Hb], platelet [Plt] and white blood cell [WBC]), were assessed by the Coulter LH 750 haematological analyser.
Treatment
As no initial clinical diagnosis was made, patients were treated symptomatically. Severely ill patients were admitted; 14 were treated empirically with a cephalosporin antibiotic, 13 received systemic corticosteroid and 10 received IVIG treatment. Only two patients received anti-viral treatment, and to a further two with the most severe symptoms, cyclosporine A was added.
Statistical analyses
All studied parameters were characterised by median values and interquartile ranges (IQR) (low quartile–high quartile [LQ–HQ]). The parameters considered at baseline, including clinical and laboratory data, were analysed using SPSS software (version 17.0; SPSS, Chicago, IL, USA).
Results
Demographic features of the patients
A total of 30/133 cases were identified to have HGA infection. All fulfilled the US CDC laboratory criteria for the diagnosis of HGA, 20 and were negative for all viruses mentioned above. In this cohort, no patient had recollection of a recent tick bite, and no patient was found to have a live tick on the body. Among the 30 affected, 20 were admitted and two required intensive care; 10 patients were treated as outpatients. Sixteen (53.3%) patients were male. Four were children, of whom the youngest was 26 days old. Most patients had no detailed records of keeping domestic animals at home.
Medical history of the patients
Clinical manifestations of 20 patients with HGA.
Clinical features
The main clinical features varied, and included sudden onset of fever of unknown cause (mean temperature: 39℃, range, 37.1–39.4℃), which lasted up to 10 days. Among them, three had a fever higher than 40℃. Other symptoms included chills (4/20), cough (8/20), headache (2/20), enlarged lymph nodes (2/20), hepatosplenomegaly (6/20), vomiting (2/20) diarrhoea (2/20) and pharyngitis (5/20) (Table 1). Signs of thrombocytopenia, including petechiae, purpura, bruising or epistaxis were observed in 18. In all, thrombocytopenia was observed more frequently than leukopenia. Only two patients experienced a severe disease course, which encompassed pneumonia, respiratory distress syndrome (ARDS) and disseminated intravascular coagulopathy (DIC), requiring intensive care and mechanical ventilation.
Laboratory data
HGA patients characteristically presented with thrombocytopenia (mean platelet value, 46.3 ± 28.3 × 109/L). Among them, three had platelet counts <10 × 109/L, nine <50 × 109/L and 18 <100 × 109/L, leukopenia (mean WBC value, 3.03 × 109/L, range, 2.4–6.6 × 109/L) and mildly deranged liver function. Elevated C-reactive protein and abnormally high serum levels of ferritin were observed in all (median ferritin concentration, 548 ng/mL vs. normal range [<300 ng/mL]). In the illest patient, levels of ferritin exceeded 1500 ng/mL. In 13 with persistent thrombocytopenia, peripheral and bone marrow (BM) smears were obtained to exclude thrombocytopenia induced by malignant haematological disease. Among these patients, morulae in leukocytes were visualised in three cases. BM smears showed reactive changes in the megakaryocytes of most patients, while increased foamy histiocytes and erythrophagocytic activity were found in the severely ill.
Discussion
An alarming number of infectious disease remains without a laboratory diagnosis. 21 Although A. phagocytophilum has attracted increasing attention as an emerging pathogen in recent years, it is seldom considered in clinical practice, and its testing is not included in routine diagnostic assessment. Viral infection may present in similar fashion.22–24 In children a mistaken diagnosis of ITP is made, but platelet counts do not respond to treatment, although bone marrow megakaryocytes increase reactively, indicating that HGA infection may not interfere with the platelet forming process, unlike the situation in ITP. Intravacuolar bacterial inclusions (morulae), are markers of HGA infection but their identification is 25–75% during the first week, and declines thereafter. 25 Therefore, morulae do not appear to be a sensitive index of infection, as they were detected in only three of 13 (similar to a previous study). 26 However, a significant correlation between HGA severity and serum ferritin was found in our study; this may reflect the downstream effects of macrophage activation that determine clinical severity of disease. 27
Our study could not estimate the seroprevalence of antibodies to A. phagocytophilum. It is known that I. persulcatus is mainly distributed in mountainous and hilly areas covered in rich vegetation, and its epidemic seasons are associated with times of peak larval activity. 28 The area studied (Shantou city of Guangdong Province) is located on the seashore. The majority of our patients live away from forested hilly regions where tick vectors are distributed. Consequently, they are a relatively lower-risk population than those living in the hill regions.
A. phagocytophilum in infected human blood specimens is viable at 4℃ for up to 18 days, 29 and therefore transmission through blood transfusion as well as contact with the blood of infected mammals is possible. 30 Recently, a few reports of presumed transmission of A. phagocytophilum from sources other than ticks have been published.5,17 In our study, 40% of patients had chronic disease prior to disease symptoms, indicating the relevance of being immunocompromised. 31
However, in our study, four (13.3%) were children, two being infants, among whom HGA has rarely been reported. 32 Recently, a case of human HGA developed in a pregnant woman near term and was transmitted perinatally to her infant; this route of transmission may, therefore, be suspected. 17
In HGA, mortality is greater if therapy is delayed, when patients are elderly and complicating opportunistic infections and/or antecedent medical conditions are present. 33 Tetracyclines are the only effective specific compounds approved for use against anaplasmosis. 34 These antibiotics are, however, not the first choice unless the possibility of HGA is taken into account.
There are several limitations of the study: only a few virus and pathogens inducing fever and thrombocytopenia could be tested for. Two patients who died duly had positive HGA tests, but whether this was the primary cause of death could not ascertained because no autopsy results were available. The small simple size and lack of long-term monitoring suggest further studies are advised to confirm the findings, especially regarding seroprevalence and epidemiology of HGA.
Conclusion
General characteristics of patients with HGA.
ALT, glutamic-pyruvic transaminase; AST, glutamic-oxaloacetic transaminase; Ca, calcium; Chol, cholesterol; Cl, chlorine; CRP, C-reactive protein; Fer, ferritin; HDL, high density lipoprotein; HGB, haemoglobin; K, kalium; LDH, L-lactate dehydrogenase; LDL, low density lipoprotein; LPa, lipoprotein a; LY%, percentage of lymphocytes; MPV, mean platelet volume; Na, natrium; NE%, percentage of neutrophils; PLT, platelet; RBC, red blood cell; TRIG, triglycerides. Reference value is from Medical lab of the first affiliated hospital of Shantou University Medical College.
Footnotes
Declaration of Conflicting Interests
All the authors have seen the manuscript and approve it for submission. The authors have no competing interest in the publication of the manuscript to declare.
Funding
This research was supported by grants-in-aid from National Natural Science Foundation of China (30972591), the U.S. Agency for International Development (USAID) Emerging Pandemic Threats Program, PREDICT project, under the terms of Cooperative Agreement Number GHN-A-00-09-00010-00, and by Science and Technology Program of Guangdong Province, China (No. 2010B031600321 and No. 2012B031800217).