Epidemiological, clinical, and laboratory characteristics of human granulocytic anaplasmosis in North India

ABSTRACT

Human granulocytic anaplasmosis (HGA) is an emerging, rickettsial tick-borne disease caused by Anaplasma phagocytophilum. Sero-epidemiological data demonstrate that this pathogen has a worldwide distribution. The diagnosis of HGA requires a high index of clinical suspicion, even in endemic areas. In recent years, HGA has increasingly been reported from Asia and described in China, Japan, and Korea. We serologically and molecularly screened 467 patients with clinical suspicion of Anaplasmosis. The present study describes the epidemiology, clinical, and laboratory details of 6 confirmed and 43 probable cases of human granulocytic anaplasmosis. One of the HGA patients developed secondary invasive opportunistic Aspergillus fumigatus and Acinetobacter baumanii infection during the illness, which resulted in a fatal infection. The HGA patients without severe complications had excellent treatment responses to doxycycline. The emergence of this newly recognized tick-borne zoonotic HGA in North India is a significant concern for public health and is likely underdiagnosed, underreported, and untreated. Hence, it is also essential to establish a well-coordinated system for actively conducting tick surveillance, especially in the forested areas of the country.

IMPORTANCE

The results of the present study show the clinical and laboratory evidence of autochthonous cases of Anaplasma phagocytophilum in North India. The results suggest the possibility of underdiagnosis of HGA in this geographical area. One of the HGA patients developed secondary invasive opportunistic Aspergillus fumigatus and Acinetobacter baumanii infection during the illness, which resulted in a fatal infection.

INTRODUCTION

Human granulocytic anaplasmosis (HGA) is an emerging tick-borne zoonosis caused by Anaplasma phagocytophilum (formerly Ehrlichia phagocytophila and E. equi), an obligate intracellular pathogen transmitted by nymphal and adult ticks of the family Ixodidae that infects leukocytes. Since the discovery, the increased incidence has been reported in the USA annually and HGA became the second most common tick-borne zoonosis after Lyme disease (1, 2). In recent years, HGA is increasingly being reported from Asia and described in China, Japan, and Korea (3–7). Sero-epidemiological and molecular evidence of this infection in various geographical areas suggests the possibility of global distribution. However, little data exists on the sero-epidemiological information of HGA in India. An animal study conducted in North East India for A. phagocytophilum in a dog population reported (8) a seroprevalence of 4.7% (n = 191). A recent study demonstrated A. phagocytophilum DNA in stray dogs and Rhipicephalus sanguineus in South India (9). Because of the difficulty in clinical diagnosis and potentially serious complications, it is vital to understand the epidemiology, prevalence, and distribution of HGA in detail in our geographical area to develop effective public health strategies. Earlier, we reported two Lyme neuroborreliosis patients co-infected with A. phagocytophilum (10). Subsequently, we conducted this epidemiological study to describe the clinical and laboratory features and distribution of HGA in patients with acute febrile illness in North India.

MATERIALS AND METHODS

Study design

A cross-sectional and a single institutional observational study was conducted over 2 years (April 2017–April 2019) at All India Institute of Medical Sciences (AIIMS), a referral hospital that represents the highest level of health care in New Delhi, India. The study population includes both inpatients and outpatients and was approved by the Institutional Ethics Committee (IEC-PG-556). The patients with any reported fever and one or more of the following clinical and laboratory features were enrolled in the study; headache, myalgia, anemia, leukopenia, thrombocytopenia, and any hepatic transaminase elevation. During the study period, both serum and EDTA blood samples were collected from 467 acute febrile patients and tested retrospectively for immunoglobulin G (IgG) antibodies to A. phagocytophilum by indirect immunofluorescence assay (IFA), Taq-Man real-time PCR, and Blot-Line A. phagocytophilum IgG Line immunoblot assay (Test Line, the Czech Republic). In addition, the patients diagnosed with other known causes of bacterial and viral acute febrile illness such as typhoid, malaria, scrub typhus, rickettsiosis, and dengue were excluded from the study population.

The patients were considered to have a diagnosis of probable or confirmed HGA using the following Centers for Disease Control and Prevention (CDC) clinical and laboratory criteria (11). A probable case of HGA was defined as a clinically compatible case (any reported fever and ≥1 of the following: headache, myalgia, anemia, leukopenia, thrombocytopenia, and/or any hepatic transaminase level elevation) with serologic evidence of elevated IgG antibody level (≥1:80) reactive with A. phagocytophilum antigen by IFA. A confirmed case of HGA was defined as a clinically compatible case along with the detection of A. phagocytophilum DNA using real-time PCR. We have used specific IgG line immunoblot assay as additional supportive laboratory evidence to strengthen the detection of anti-A. phagocytophilum antibody in our geographical region. Once the patients were identified, the medical records were reviewed for possible HGA infection by checking for tick exposure history, typical laboratory findings, clinical signs, and symptoms compatible with HGA.

All samples were subjected to a commercial indirect IgG immunofluorescence assay (Fuller Diagnostics, Fullerton, USA) as per the manufacturer's instructions. The IFA assay format can detect both A. phagocytophilum and E. chaffeensis antibodies simultaneously and was developed using semi-purified elementary bodies and morulae from cell culture propagated A. phagocytophilum and E. chaffeensis. The visual intensity, antigen density, and the appearance of the fluorescence pattern of positive samples were compared with negative and positive control. As per the manufacturer's instructions, a serum IgG titer of 1:80 was considered positive for A. phagocytophilum and E. chaffeensis. The reactive clinical samples at the screening dilution of 1:80 were further tested after serial dilution up to 1:1,280 to determine the titer. An IgG endpoint titer of 1:80 to 1:320 is suggestive of (i) exposure to A. phagocytophilum (titers preceding peak levels) and (ii) past exposure. According to the manufacturer's instructions, titers greater than 1:320 when present are a reliable indicator of recent infection.

The clinical samples that reacted positively for A. phagocytophilum IgG antibodies were further tested using the Blot-Line A. phagocytophilum IgG Line immunoblot assay (Test Line, Czech Republic). Briefly, antibodies to specific antigens (p44, Asp62, and OmpA) of A. phagocytophilum were demonstrated using Blot-Line Anaplasma IgG Line immunoblot assay. The testing was performed and interpreted as per the manufacturer's instructions. At least one positive and one borderline band of p44, Asp62, and OmpA can be considered a case of A. phagocytophilum infection. DNA was extracted from EDTA blood samples (200 µL) using a QIAmp DNA mini kit (Qiagen, Hilden, Germany) following the manufacturer's recommendation. The elution step was modified using a final volume of 50 µL of elution buffer. The clinical samples were tested for A. phagocytophilum using a Taq-Man real-time PCR targeting the AnkA (Ankyrin-repeat protein) gene as per the previously published protocol (12).

Statistical analysis

Categorical variables were described as numbers, including percentages, and were compared using Fisher's exact test or Student t-test according to the data distribution. Continuous variables were represented using the median [inter quartile range (IQR)] and compared using the Mann–Whitney U test. The statistical analysis was performed using SPSS software version 25.0 (IBM, NY, USA). All P values were two-sided, and a value of <0.05 was considered significant.

RESULTS

A total of 467 serum samples collected from acute febrile illness patients were screened for HGA using IgG IFA and real-time PCR. Of the 467 patients, 318 (68.1%) were adults (≥18 years), and 265 (56.7%) were males. Exposure history to tick-infested areas was reported in 365 (78.1%) patients. Of the 467 sera samples, 45 (9.6%) reacted with A. phagocytophilum by IgG IFA and among these two (0.4%) sera reacted simultaneously with A. phagocytophilum and E. chaffeensis, respectively. On further dilution, these two samples showed higher antibody titers to A. phagocytophilum (1:1,280, 1:640). The six HGA cases were diagnosed using a real-time PCR targeting the AnkA gene of A. phagocytophilum (Fig. 1) and among these, only two (33.3%) patients had IgG antibodies to A. phagocytophilum by IFA. The antibody titers of the reacted samples of HGA are shown in Table 1. Additionally, all the IgG IFA-positive sera (n = 45) were subjected to A. phagocytophilum-specific IgG immunoblot which yielded 20 (44.4 %, n = 45) positive and seven borderline results, and were reacted with p44 (MSP-2), Asp62, or OmpA specific antigens (Fig. 2). Furthermore, the antibody reaction patterns of various Anaplasma-specific antigens with immunoblot-positive sera could be divided into two groups; 14 sera reacted with the 44 kDa, OmpA, and Asp62 antigen combination, and six sera reacted with the combination of two antigens (Asp62 and OmpA).

Fig 1.

Real-time PCR linear amplification plot of A. phagocytophilum six positive samples

TABLE 1.

Characteristics of patients with HGA according to case category^a

Characteristic	Case category
	HGA confirmed (n = 6)	HGA probable (n = 43)
Days after the onset of symptoms before the collection of samples
<7 days	5 (83)	2 (5)
7–14 days	1 (17)	18 (42)
>14 days	0	23 (53)
Duration of fever (days)	6 (5–8)	7 (5–10)
Laboratory tests used to diagnose HGA
Real-time PCR	6/6	0/43
Serology
IgG Immunofluorescence titer
Titer at 1:80	1	4
Titer at 1:160	1	3
Titer at 1:320	0	18
Titer at 1:640	0	14
Titer at 1:1,280	0	4
IgG immunoblot positive	1/2	19/43

^a Values shown are numbers (%) or median (IQR).

Fig 2.

A. phagocytophilum-specific IgG immunoblot results of three representative samples. (a) Positive for p44, Asp62, and borderline for OmpA antigens; (b) positive for Asp62 and OmpA antigens; (c) negative sample.

Demographic characteristics

Of the 49 patients diagnosed with HGA, 30 were male (61.2%), and 45 were adults (≥18). The median age of the patient was 33 years (4.5–76 years). Almost all age groups were affected, with higher frequency among cases above 21. Exposure history, including gardening, camping, or outdoor activities in forest areas, was reported in 41 (83.7%, n = 49) patients. The positive cases were mainly reported from Bihar, Haryana, Uttar Pradesh, Madhya Pradesh, Rajasthan, and Uttarakhand (Fig. 3). Most patients presented with clinical symptoms between April and October. Demographic and other clinical features are shown in Table 2.

Fig 3.

The geographical location of HGA-positive cases. (The figure was created using MapChart.)

TABLE 2.

Characteristics of patients with HGA and non-HGA according to case category^a

Presentation	Case category		P value	Odds ratio
	HGA positive (n = 49)	HGA negative (n = 418)
Age (median [IQR])	33 (25–50)	24 (15–42)	0.000f
Gender, Male	30 (61.2)	235 (56.2)	0.504
Clinical signs or symptoms
Fever duration	6 (5–8)	7 (5–10)	0.455
Chills/rigours	9 (18)	73 (17)	0.875	1.063
Rash	7 (14)	110 (26)	0.066	0.467
Headache	30 (61)	149 (36)	0.000	2.851
Myalgia	33 (67)	168 (40)	0.000	3.069
Arthralgia	8 (16)	57 (14)	0.607	1.236
Abdominal pain	5 (10)	92 (22)	0.054	0.403
Vomiting	4 (8)	87 (21)	0.034	0.338
Diarrhea	4 (8)	27 (6)	0.553	1.287
Dyspnea	14 (29)	117 (28)	0.932	1.029
Cough	8 (16)	88 (21)	0.439	0.732
Hepatomegaly	5 (10)	137 (33)	0.001	0.233
Splenomegaly	3 (6)	68 (16)	0.061	0.336
Acute kidney injury	21 (43)	131 (31)	0.104	1.643
Laboratory findings
Anemiab	30 (61)	306 (73)	0.077	0.578
Hemoglobin (g/dL)	10.9 (9.8–13.1)	10.2 (8.5–12)	0.135f
Leukopeniac	19 (39)	91 (22)	0.008	2.276
WBC count (×103 µL)g	6 (3.7–8.5)	8 (4.6–12.9)	0.000f
Thrombocytopeniad	38 (78)	238 (57)	0.005	2.613
Platelet count (×103 µL)	120 (94–141)	135 (100–197)	0.013f
Elevated liver enzymese	40 (82)	262 (63)	0.009	2.646

^a Values shown are number (%) or median (IQR).

^b Reference value 12.9–17.3 g/dL (male); 11.7–15.5 g/dL (female).

^c Reference value 4.1–10.9 × 10³ μL.

^d Reference value 175–450 × 10³ μL Thrombocytopenia < 150 × 10³ μL.

^e Reference value Alanine transaminase 10–49 U/L (elevated level >40 U/L); Aspartate transaminase 14–36 U/L (elevated level >40 U/L); Acute kidney injury (AKI) was defined as an increase in serum creatinine (≥0.3 mg/dL within 48 hours, or ≥1.5 times increase from the baseline). Categorical variables were analyzed using Fisher’s exact or χ² test, and continuous variables were compared using the Wilcoxon rank-sum test (Mann–Whitney U test). .

^f Mann–Whitney U test Hepatomegaly and splenomegaly >15 cm (ultrasonography).

^g WBC, white blood cell.

Clinical and laboratory results

All patients had a fever from 38.5°C to 40.2°C. The patients with confirmed or probable infection were older (33 years) than HGA negative population (24 years; P ≤ 0.01). Overall, patients with HGA in our cohort presented with fever (100%), headache (61%), myalgia (67%), acute kidney injury (43%), dyspnea (29%), chills (18%), arthralgia (16%), hepatomegaly (10%), abdominal pain (10%), vomiting (8%), diarrhea (8%), and splenomegaly (6%). Frequent laboratory abnormalities include elevated liver enzymes [n = 40 (82%), P = 0.009], thrombocytopenia [n = 38 (78%), P = 0.005], and leukopenia [n = 19 (39%), P = 0.008]. The rash was present only in 7 (14%). The median platelet count of the HGA cases was 120,000 /µL (range, 94,000–141,000). At admission, HGA cases had lower leukocyte count (median, 5,500 cells/µL; range, 3,700–8,500 cells/µL) when compared to non-HGA cases (median, 7,740 cells/µL; range, 4,600–12,900 cells/µL). The lowest leukocyte count reported in the HGA cohort was 3,100 cells/µL. A minority had arthralgia, dyspnea, hepatomegaly, splenomegaly, and involvement of the gastrointestinal tract. Acute kidney injury was observed among 21 HGA patients but was also more frequent in the non-HGA group and was statistically insignificant. We have compared both confirmed and probable infections of HGA to see the clinical differences. We found few differences in the clinical manifestations of HGA confirmed and probable infection (Table 3).

TABLE 3.

Clinical presentation of patients with confirmed and probable HGA^a

Presentation	Case category		P valueb	Odds ratio
	Confirmed (n = 6)	Probable (n = 43)
Age [median (IQR)]	39 (30–45)	32 (24–51)	0.855
Gender, male	5 (83)	25 (58)	0.384
Clinical signs or symptoms
Fever duration	9 (6–10)	6 (5–8)	0.106
Chills/rigours	1 (17)	8 (19)	1	1.143
Rash	1 (17)	6 (14)	0.631	1.212
Headache	6 (100)	24 (56)	0.069	1.25
Myalgia	6 (100)	27 (63)	0.159	1.22
Arthralgia	1 (17)	7 (16)	1	0.972
Abdominal pain	2 (33)	3 (7)	0.107	0.150
Vomiting	1 (17)	3 (7)	0.418	0.375
Diarrhea	0 (0)	4 (9)	1	0.867
Dyspnea	2 (33)	12 (28)	1	0.774
Cough	1 (17)	7 (17)	1	0.972
Hepatomegaly	1 (17)	4 (9)	0.495	0.513
Splenomegaly	0 (0)	3 (7)	1	0.870
Acute kidney injury	4 (67)	17 (40)	0.381	0.327
Laboratory findings
Anemia	2 (33)	28 (65)	0.19	3.73
Hemoglobin (g/dL)	13.2 (11.5–14)	10 (8.1–13)	0.053
Leukopenia	4 (67)	15 (35)	0.190	0.268
WBC count (×103 µL)c	4 (3.6–10)	5.5 (3.7–8.5)	0.272
Thrombocytopenia	6 (100)	32 (74)	0.315	1.18
Platelet count (×103 µL)	117 (49–122)	123 (98–152)	0.272
Elevated liver enzymes	6 (100)	34 (79)	0.574	1.17

^a Values shown are number (percentage) or median (IQR).

^b Categorical variables were analyzed using Fisher’s exact test, and continuous variables were compared using Wilcoxon rank-sum test (Mann–Whitney U test).

^c WBC, white blood cell.

Treatment, complications, and outcome

Twenty-seven (55%) patients in our cohort required hospitalization, while 22 (45%) were treated as outpatients. Among 49 patients, 35 (71.4%) received antibiotic treatment: 26 (74%) received doxycycline, 3 (9%) received azithromycin, 5 (14%) received ceftriaxone, and 1 (3%) received levofloxacin. Forty-five (92%) patients had no complications and were cured, out of which 25 received doxycycline. Fourteen patients (29%) were cured without any antibiotic treatment. However, we noted that four patients (8%) developed severe complications and died, of which one received doxycycline. The average time between hospital admission and death was 8 days. Other findings observed were acute pancreatitis (1/4), renal failure (4/4), hepatic injury (3/4), altered sensorium (2/4), type II diabetes mellitus (2/4), chronic alcoholic liver disease (2/4), hypertension (1/4), and pericardial effusion (1/4). One patient had developed invasive pulmonary aspergillosis (A. fumigatus isolated from Broncho-alveolar lavage) and multi-drug resistant A. baumanii bloodstream infection during the course of HGA. Another patient developed sepsis and septic shock due to Escherichia coli.

DISCUSSION

Rickettsial pathogens are the leading tropical vector-borne zoonotic agents, contributing to the burden of undifferentiated acute febrile illness in India. Limited evidence and information are available for the presence of other Rickettsia-like pathogens, including A. phagocytophilum, in this geographical area. The present study demonstrated a seroprevalence of A. phagocytophilum of about 9.6% (n = 45). The sero-epidemiological data from Asia, Europe, and the USA demonstrated high seroprevalence of HGA (1.4%–28%) in endemic areas, and these data suggest that many of the HGA cases go unnoticed (13–17). We diagnosed 49 patients with HGA in North India. Of the 467 patients tested, 6 (1.2%) and 43 patients (9.2%) met the confirmed and probable criteria of the CDC for HGA, respectively. In most of the patients, the blood samples were taken after 1 week of illness and after antibiotic treatment. The infected neutrophils can no longer be detected by PCR after doxycycline treatment or later than the second week of HGA infection if untreated (18). This may be a reason for getting low PCR positivity in our cohort of patients. It is well-recognized that serum samples from patients with HGA cross-react with E. chaffeensis (19). In the present study, only two (0.4%) samples were positive for both A. phagocytophilum and E. chaffeensis. Further testing of these samples showed higher titer antibodies to A. phagocytophilum and was considered as probable HGA.

Most patients presented with clinical symptoms between April and October, in North India while most cases in the USA have been reported between May and August (14, 20). All of the HGA patients reported in this study resided in areas where Lyme disease is emerging (10). The A. phagocytophilum has been found mainly in regions where Lyme disease is common (21). The distribution of cases was primarily in the Northwestern states and agricultural areas in North India, where ticks were commonly found. Limited data are available on the seasonal activity of ticks in North India. The highest infestation with ticks was observed during the rainy season (June–September) with a peak in August–September due to high humidity followed by summer. Rainfall appeared to be an important macroclimate factor influencing the seasonal activity of ticks in these areas (22). A. phagocytophilum infection has been found in several species of ticks in Asia. In Asia, Haemophysalis spp., R. sanguineus, and Ixodes ovatus seem to act as vectors (23–25). A. phagocytophilum DNA was demonstrated in stray dogs and R. sanguineus in South India (9). However, no active systematic tick surveillance has been conducted in India to detect A. phagocytophilum until now. Haemophysalis bispinosa, H. spinigera, H. intermedia, Rhipicephalus microplus, R. sanguineus, R. haemaphysaloides, R. turanicus, Hyalomma anatolicum, H. marginatum isaaci, H. detritum, H. hussaini, H. kumari, Amblyomma testudinarium, Dermacentor auratus, Ixodes acutitarsus, and Ixodes ovatus are the common tick species infesting wild and domestic animals found in previously conducted zoological surveys in North India (26). Additionally, Ixodes granulatus, Ixodes himaleyensis, Ixodes kashmericus, and Ixodes ricinus are also reported from some parts of North India (27–30). Based on the previous surveys, it is highly probable that one or a few of the above-mentioned tick species which are commonly found in areas where HGA has been detected in our study are more likely to be involved in local transmission.

HGA has a wide range of clinical manifestations and sometimes may be asymptomatic. The patients develop high-grade fever, severe headache, myalgia, and malaise. Additionally, arthralgia, anorexia, non-productive cough, and nausea are also reported frequently in the literature (14, 31, 32). The most frequently described clinical signs or symptoms in this study include fever, headache, and myalgia. In addition, most of our HGA patients developed leukopenia, thrombocytopenia, and elevated transaminase levels. Less frequent signs or symptoms were arthralgia, hepatomegaly, splenomegaly, dyspnea, and diarrhea, and were not statistically significant. Even though acute kidney injury occurred in 21 patients, this was not statistically significant when compared with non-HGA cases. (Table 2). Only 7 of the 49 (14%) patients had developed the rash (six macular and one maculopapular). The rash was more commonly reported in HME, which is less common (1%–11%) in HGA (31, 32). In our cohort, 19 patients had a leukocyte count below the lower limit (3,700 cells/µL). Leukopenia was only observed in 76 (52.8%) of 144 laboratory-confirmed HGA previously and was observed especially during the first week and returned to normal during the second week (33). The clinical and laboratory findings reported in the present study were similar to those reported in the USA and Europe. However, European HGA is generally milder (32, 34, 35). Dyspnea and cough were present in 14 (29%) and 8 (16 %) patients. However, infiltrates were only found in chest radiographs in two.

The presence of specific IgG antibodies in probable HGA indicates contact with A. phagocytophilum. However, this data does not always indicate the presence of a recent infection. This might be due to present, previous symptomatic, or previous asymptomatic infection. A comparison of confirmed and probable HGA in our study revealed similar clinical and laboratory findings. Stanka et al. reported a significant difference between confirmed and probable HGA in their study (36). Additionally, A. phagocytophilum IgG immunoblot was performed among 45 IFA-positive sera, which showed 20 positive and 7 borderline results. Among these 20 immunoblot-positive cases, most of them (n = 14) had IFA titers above 1:320 and 6 patients (n = 6) had IFA titers ranging from 1:80 and 1:160. This discrepancy between IFA and immunoblot has been documented previously. This happens probably due to the sensitivity and specificity of the assays (37, 38). These 27 immunoblot-positive patients were also included in the probable HGA group and served as additional supportive laboratory evidence to strengthen the detection of anti-A. phagocytophilum antibody in our geographical region.

In this study, the HGA without severe complications had excellent clinical responses to treatment with doxycycline. Twenty-seven (55%) cases in our cohort of HGA required hospitalization while 22 were treated as outpatients. Bakken et al. reported similar findings in their study (32). Of 49 HGA cases, 35 (71.4%) received antibiotic treatment and 26 received doxycycline. Forty-five (91.8%) were completely cured without any complications. Out of 26 (74.2%) patients who received doxycycline therapy, 25 patients were completely cured. Doxycycline is the recommended drug of choice, with good results for HGA in adults and children older than 8 years. Rapid response to treatment with a marked clinical improvement within 24–48 hours is usually seen (39).

Life-threatening complications can occur in 7%–9% of HGA patients including sepsis, disseminated intravascular coagulation (DIC), and systemic inflammatory syndromes (34, 40). The severity of complications is higher in older patients (over 40 years of age) with HGA, including acute renal failure, acute respiratory distress syndrome (ARDS), or rhabdomyolysis (14). Similarly, our data also support this finding in which some of our older patients presented with severe HGA. Additionally, no obvious long-term sequelae were observed in those patients who recovered. Both leukopenia and thrombocytopenia at initial presentation were normalized with or without doxycycline treatment after 7 days. However, four HGA patients developed complications and died, of which one received doxycycline treatment which underscores the importance of early diagnosis and specific therapy. In this cohort, all deaths were among patients with older age, concomitant immune suppressing co-morbidities including diabetes mellitus (n = 2) and chronic alcoholic liver disease (n = 2). The median age of these HGA patients (median, 50; range 42–69) was relatively higher. Furthermore, a patient who had type II diabetes mellitus as a risk factor developed invasive pulmonary aspergillosis and bloodstream infection caused by multi-drug resistant A. baumanii. Additionally, another patient with a risk factor of alcoholic liver disease developed sepsis and septic shock due to E. coli infection. All of the patients were older and the diagnosis was delayed. Four patients developed severe complications, including multi-organ failure, renal failure, and acute pancreatitis. Fatal infections of HGA are often associated with disseminated opportunistic fungal or viral infections, suggesting an immunosuppressive effect and also due to chronic co-morbid conditions such as diabetes mellitus. In addition, there were reports of opportunistic infections such as candida esophagitis, herpes simplex virus esophagitis with cryptococcal pneumonia, and invasive pulmonary aspergillosis during HGA, which resulted in fatal complications (20, 35, 41, 42). The mortality observed in HGA is higher if treatment is delayed, when patients are elderly, or due to opportunistic secondary infections and concomitant chronic illnesses (43). In view of the above complications, HGA could potentially contribute to previously unrecognized morbidity and mortality.

Limitations of the study

Our study has a few limitations. The findings of this study pertain to a single institution and may not accurately represent the overall prevalence of HGA, as it may differ across different geographical areas. It would be helpful to assess the seroconversion or fourfold increase in titer between paired serum specimens, which was not possible for single serum samples tested in the present study. However, the antibody titer of all the serology-positive cases of HGA has been determined using IFA. We have additionally supplemented the serology results with A. phagocytophilum-specific immunoblot.

Conclusions

The results of the present study show the clinical and laboratory evidence of autochthonous cases of A. phagocytophilum in North India. The results suggest the possibility of underdiagnosis of HGA in this geographical area. Lack of awareness among healthcare providers may increase the likelihood of undiagnosed and untreated infections in the community. Most of the patients with an HGA diagnosis recovered without significant complications. HGA must be suspected in previously healthy individuals who develop a fever with cytopenia or elevated transaminases following outdoor activities in tick-infested areas.