A lethal combination: fulminant liver failure due to Babesia microti and Borrelia burgdorferi co-infection

Abstract

Babesiosis and Lyme disease are tick-borne infections endemic to the northeastern and upper Midwestern United States. Although both diseases can involve multiple organ systems, acute liver failure (ALF) is exceedingly uncommon. Co-infection with Babesia microti and Borrelia burgdorferi may exacerbate disease severity, create diagnostic uncertainty, and delay appropriate therapy. We describe a 77-year-old man with type II diabetes mellitus, stage IV chronic kidney disease, and heart failure with preserved ejection fraction who presented with dyspnea, cough, and peripheral edema. Initial evaluation suggested decompensated heart failure and acute kidney injury consistent with cardiorenal syndrome. Within 24 h, however, he developed ALF (total bilirubin 6.8 mg/dL; aspartate aminotransferase 2656 U/L; alanine aminotransferase 1421 U/L; INR 3.5), thrombocytopenia (95 × 10⁹/L), acute anemia (hemoglobin 7.7 g/dL), and encephalopathy. Initial infectious, autoimmune, and ischemic evaluations were unrevealing. Despite empiric antimicrobial therapy with intravenous ceftriaxone, vancomycin, and doxycycline, his condition rapidly deteriorated, and he died. Postmortem testing identified B. microti parasitemia of 2.4% and B. burgdorferi co-infection. This case highlights the importance of considering tick-borne coinfections in patients presenting with ALF and shock, even in the absence of classic features such as fever, rash, or overt hemolysis. Increased clinician awareness of atypical presentations of endemic tick-borne diseases may help reduce diagnostic delays and improve outcomes in severe or unusual clinical scenarios.

Plain language summary

A rare and severe case of liver failure caused by babesiosis and Lyme disease without typical symptoms

Babesiosis and Lyme disease are tick-bourne infections transmitted by Ixodes ticks and are increasingly prevalent in the northeastern and upper midwestern U.S. While each pathogen independently can cause systemic illness, co-infection is not uncommon and may present with overlapping or atypical features, complicating timely diagnosis and treatment. We present a case of a 70 year old man who presented to the hospital with dyspnea, dry cough and leg edema. Initial workup supported acute decompensated heart failure and cardiorenal syndrome. Despite appropriate diuresis and supportive care, the patient rapidly developed oliguric renal failure requiring dialysis, hepatic encephalopathy and fulminant acute liver failure (ALF) with severe coagulopathy and shock. Notably, he lacked classic features of tick-bourne illness during presentation, such as fever, rash, hemolysis. Broad spectrum empiric antibiotics were initiated, including doxycycline, but the patient deteriorated and died on hospital day five. Postmortem testing revealed Babesia microti (2, 4% parasitemia) and Borrelia Burgdorferi by PCP testing, confirming tick-bourne co-infection. This case highlights the diagnostic challenge of atypical presentations of endemic zoonoses. Although hepatic involvement in babesiosis is recognized, fulminant ALF remains rare. Lyme disease is typically associated with mild, transient LFT elevations and has not been linked to ALF. The patient’s clinical decline may reflect the additive effect of Babesia induced multiorgan failure. Clinicians should maintain a high index of suspicion for tick-bourne infections, even in the absence of hallmark features in endemic regions, particularly when faced with unexplained multiorgan failure. Early recognition and empiric therapy is imperative to improve outcome in similar cases.

Background

Babesiosis and Lyme disease are emerging tick-borne infections transmitted by Ixodes ticks. They are endemic to the northeastern and upper Midwestern regions of the United States. Babesiosis, caused by the protozoan parasite Babesia microti, primarily affects erythrocytes and leads to hemolytic anemia, thrombocytopenia, splenomegaly, acute kidney injury (AKI), hepatitis, pneumonia, and, in rare cases, splenic infarct and rupture.^1–3 Babesiosis is particularly severe in elderly and immunocompromised patients. ¹ Lyme disease, caused by B. burgdorferi, typically presents in its early stages with a characteristic erythema migrans rash and, if left untreated, may progress to disseminated disease leading to neurological, musculoskeletal, or cardiac complications.^4–6 Thrombocytopenia and hemolytic anemia are not laboratory features of Lyme disease.

Due to climate change and global warming, which affect tick population proliferation, the incidence of both diseases is expected to rise in the coming decades.^4–6 Co-infection with both B. microti and B. burgdorferi is well documented in humans due to their shared tick vector. Overlapping clinical features, including fever, fatigue, and multiorgan involvement, might lead to diagnostic uncertainty and delay in establishing an accurate diagnosis. ⁷ Based on the most recent data from the IDSA 2020 Guideline on Diagnosis and Management of Babesiosis, co-infection is not uncommon: in endemic areas, 2%–40% of patients with early Lyme disease have babesiosis, and up to half of babesiosis cases have concurrent Lyme disease. ⁸ Co-infected patients frequently exhibit more severe constitutional symptoms, such as persistent fever and fatigue, and more pronounced laboratory abnormalities, including thrombocytopenia, leukopenia, and anemia, compared with Lyme disease alone. ⁹ Evidence suggests that babesiosis may increase the severity of Lyme disease, though the effect of Lyme disease on babesiosis severity appears limited. ¹⁰

Hepatic involvement, and in particular fulminant ALF, is an exceptionally rare complication in babesiosis and thus far has not been reported as a complication of Lyme disease. Whether co-infection is associated with a more severe clinical presentation and increased mortality has been a topic of debate, and research on this topic remains inconclusive.^7,10,11 This case report illustrates a co-infection of B. microti and B. burgdorferi in a patient who presented without classical clinical findings of either disease, which led to a delay in treatment and a lethal outcome. In this patient, the earliest manifestation of illness was an acute exacerbation of congestive heart failure, which preceded the development of other clinical and laboratory abnormalities.

Case presentation

Our patient was a man in his 70s with a medical history of insulin-dependent type II diabetes, stage IV chronic kidney disease, and heart failure with preserved ejection fraction (HFpEF). He presented to the emergency department with a two-week history of worsening dyspnea, dry cough, and bilateral lower extremity edema. His medications included insulin, aspirin, metoprolol, and torsemide. The cough was nonproductive and worsened in the supine position. The edema was symmetric, nontender, and without erythema. The dyspnea was exertional and orthopneic, without chest pain. The patient denied fever, chills, abdominal pain, diarrhea, or vomiting.

On admission, his vital signs revealed a blood pressure of 110/60 mmHg, a heart rate of 98 beats per minute, a respiratory rate of 22 breaths per minute, and an oxygen saturation of 92% on room air. Physical examination was notable for signs of volume overload, including elevated jugular venous distention, pulmonary crackles, and 3+ bilateral pitting edema. The skin examination did not show any rashes or evidence of cellulitis. Chest X-ray demonstrated bilateral pulmonary vascular congestion and pleural effusions, without lobar infiltrates. Laboratory testing showed an elevated pro-BNP (blood urea nitrogen) and a serum creatinine of 4.0 mg/dL, up from a baseline of 2.5 mg/dL. Electrolytes and liver enzymes were within normal limits on admission. Complete blood cell count showed stable chronic normocytic anemia consistent with anemia of chronic kidney disease, as illustrated in Table 1. Importantly, there was no evidence of hemolysis or thrombocytopenia at the time of admission.

Table 1.

Laboratory values on admission (day 1), evening of day 3 when hemodialysis was initiated and last day (day 5) when the patient deceased.

Laboratory value	Normal value	Baseline value	Day 1	Day 3	Last day/day 5
Hemoglobin	13.2–16.6 g/dL	9.3	8.6	7.0	7.7
Leucocyte count	3.4–9.6 × 109/L	6.1	4.4	6.1	4.3
Platelet count	135–317 × 109/L	263	160	158	95
Eosinophil count	0.03–0.48 × 109/L	0.11	<0.03	—	—
Neutrophil count	1.56–6.45 × 109/L	3.56	3.03	—	—
Total bilirubin	0.0–1.2 mg/dL	0.8	1.1	3.4	6.8
Direct bilirubin	0.0–1.2 mg/dL	0.2	—	2.6	4.2
ALT	7–55 U/L	20	19	532	1421
AST	8–48 U/L	24	31	1000	2656
Lactate	0.5–2.2 mm/L	1.2	—	7.8	3.3
Potassium	3.6–5.2 mm/L	4.0	4.6	6.3	5.1
BUN	8–24 mg/dL	55	75	112	31
Creatinine	0.74- 1.35 mg/dL	2.5	4.89	7.21	1.89
Anion gap		15	14	23	13
Total protein	6.3–7.9 g/dL	7.9	6.9	6.6	6.3
Albumin	3.5–5/0 g/dL	4.4	3.7	3.5	3.1
INR	0.9–1.1	1.2	—	2.1	1.9
Fibrinogen	200–393 mg/dL	—	—		253
LDH	122–222 U/L	—	—	—	>2250
Haptoglobin	30–200 mg/dL	—	—	—	<17

ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; INR, international normalized ratio; LDH, lactate dehydrogenase.

The patient was diagnosed with acute decompensated HFpEF and cardiorenal syndrome and was started on intravenous furosemide. Despite the use of intravenous diuretics, he became oliguric over the subsequent 24 h and required initiation of hemodialysis for fluid overload.

On the following day (48 h after admission), the patient developed lethargy, thrombocytopenia, and a hepatocellular pattern of acute liver injury (Table 1). He was unable to provide further history due to encephalopathy that developed despite adequate hemodialysis. Workup for acute viral hepatitis was negative. Ischemic hepatitis was ruled out due to the absence of hypotension and shock. Drug-induced liver injury was deemed unlikely, as no new medication was started, and the insulin, torsemide, and metoprolol the patient was receiving had been well tolerated in the past and have a low risk for hepatotoxicity. Abdominal ultrasound and Doppler studies excluded cholecystitis and biliary obstruction and confirmed patent portal and hepatic veins. Urinalysis could not be obtained as the patient became anuric. Transthoracic echocardiogram showed preserved ejection fraction, normal valvular function, and absence of pericardial effusion or tamponade.

As the patient’s condition deteriorated, signs of sepsis and hypotension emerged on day 3 of hospitalization (48–72 h after admission). Empiric antibiotic therapy with intravenous ceftriaxone 2 g daily, vancomycin (renally dosed 15 mg/kg/day), and doxycycline 100 mg IV every 12 h was initiated for community-acquired sepsis without a clear source. His hypotension worsened, and he developed shock, requiring inotropic support with norepinephrine, vasopressin, phenylephrine, and dobutamine. Despite aggressive resuscitation efforts, the patient succumbed to his illness on hospital day five.

Peripheral blood smear was reported postmortem and demonstrated intraerythrocytic ring forms consistent with B. microti, with a quantified parasitemia of 2.4%. Smear diagnosis was supported by characteristic microscopic features, including small pleomorphic ring forms within erythrocytes (Figure 1). A blood Plymerase Chain Reaction (PCR) assay detecting B. microti DNA confirmed acute infection. B. microti blood PCR was performed at the Mayo Clinic Clinical Microbiology Laboratory. Nucleic acid was extracted using the automated MagNA Pure (Roche) bead-based system, followed by target amplification on the LightCycler 480 platform. The PCR assay targets the nuclear small-subunit ribosomal RNA gene (18S rDNA) using species-specific primers and fluorescence resonance energy transfer (FRET) probes, with amplicon detection and species confirmation by melting-curve analysis.

Figure 1.

Peripheral blood smear at 1000× magnification, showing intraerythrocytic parasitic inclusions (thick blue arrow) consistent with babesiosis.

A tick-borne panel sent on hospital day 2 (when thrombocytopenia and encephalopathy developed) returned positive by blood PCR assay detecting B. burgdorferi s.s. as well as positive for IgM and IgG antibodies. These tests were also performed at the Mayo Clinic Clinical Microbiology Laboratory. A PCR assay detecting Borrelia DNA was performed using a FRET-based assay, with melting-curve analysis confirming B. burgdorferi s.s and excluding other genospecies within the B. burgdorferi sensu lato complex (B. mayonii, B. afzelii, and B. garinii). Together, these results confirmed co-infection with acute B. microti and B. burgdorferi sensu stricto.

Trends in laboratory values over the patient’s hospital course are shown in Table 1.

Discussion

Acute liver failure is a life-threatening condition defined by the development of acute liver injury with coagulopathy (INR > 1.5) and any grade of hepatic encephalopathy in individuals without preexisting liver disease. It is classified by the O’Grady criteria into hyperacute (<7 days), acute (7–28 days), and subacute (28 days–12 weeks) presentations. The etiology of ALF varies geographically: viral hepatitis (particularly hepatitis A and E) is the leading cause in under-resourced regions, while drug-induced liver injury (DILI), most commonly from acetaminophen, predominates in developed countries. Complementary and alternative medicines and anti-tuberculosis drugs are major causes in parts of Asia. Other less frequent causes include ischemia, Budd–Chiari syndrome, herpes viruses, Wilson’s disease, malignancy, and pregnancy-related conditions. Etiology significantly influences prognosis; ALF due to acetaminophen, viral hepatitis A, and pregnancy-related causes tends to have better outcomes, while idiosyncratic DILI, autoimmune hepatitis, Wilson’s disease, and hepatitis B are associated with higher mortality and a higher likelihood of requiring liver transplantation.^12,13

Acute liver failure in severe babesiosis results from a combination of hemolysis-induced hepatocellular injury, microvascular obstruction, and inflammatory cytokine-mediated damage^14,15 While ALF caused by B. microti has been rarely documented,^16,17 ALF in a patient co-infected with B. microti and B. burgdorferi has not been reported thus far. A large retrospective study by Ssentongo et al. ¹⁸ reported no ALF cases linked to B. microti and B. burgdorferi co-infection.

Hepatic involvement in Lyme disease is generally mild and subclinical, with elevated liver enzymes seen more frequently in early disseminated disease (66%) than in localized disease (34%). ¹⁹ In early disseminated disease, mildly elevated liver enzymes, particularly gamma-glutamyl transferase and ALT, are relatively common, occurring in up to 66% of patients. ¹⁹ These abnormalities are generally transient and resolve with appropriate antibiotic therapy. Benedix et al. ¹⁹ described a patient with multiple erythema migrans due to B. afzelii, who had mild hepatitis and hepatomegaly, both of which improved following treatment with ceftriaxone and doxycycline. Similarly, a prospective study conducted by Nadelman et al. ²⁰ found that 37% of patients with culture-confirmed B. burgdorferi had hepatitis, but none developed clinically significant liver disease, particularly ALF. In addition, isolated elevations in AST may reflect Lyme-associated myositis rather than true hepatic injury, according to some studies. ²¹ Although rare cases of clinically significant hepatitis have been reported, such as the case described by Duffau et al. ²² involving B. garinii in an immunocompromised patient, these are exceptionally rare and do not cause acute liver failure. Given the typically benign hepatic involvement in Lyme disease and the fact that the patient received ceftriaxone and doxycycline, both of which would treat Lyme disease, it is unlikely that Lyme infection alone accounted for the ALF in our case. More likely, ALF was part of severe babesiosis, which is known to be life-threatening in older patients with comorbidities such as the patient we report.

This case highlights an unusual, severe, life-threatening, and atypical presentation of an endemic infection and the diagnostic complexity in cases of co-infection with B. microti and B. burgdorferi. The co-infection rate of B. microti and B. burgdorferi across the United States is 42%, ¹⁸ while it is 37% in Wisconsin. ¹⁰ These findings differ from an earlier study based in the Northeast region of the United States, where the co-infection rate was as low as 10%. ⁷ The mortality rate from babesiosis ranges from 1.6% to 13%. ¹⁸ It is even higher, up to 20%, for hospitalized patients who are immunocompromised or who contracted babesiosis through blood transfusion. ¹ The most recent retrospective study reported an overall 90-day mortality rate of 1.4% across the United States and noted it to be lower in the B. microti and B. burgdorferi co-infection group than in isolated babesiosis. ¹⁸

In animal models, co-infection has been shown to exacerbate the severity of Lyme disease, with the hypothesis that babesiosis could cause suppression of adaptive immunity, thereby exacerbating Lyme disease symptoms by altering the host’s immune response.^10,11 However, other studies in humans demonstrated no such risk. ¹⁸

The severity of babesiosis depends on host immunity and the degree of parasitemia. It ranges from asymptomatic mild infection to moderate or severe infection. Severe disease is seen in immunocompromised patients, particularly those with asplenia, B-cell–associated lymphoid malignancy, and patients on B-cell–depleting therapies such as rituximab, or those with high parasitemia >10%.^1,23 A high degree of parasitemia can cause hemolytic anemia and its complications. In addition, a cytokine storm due to an inappropriately intense immune response to high parasitemia can lead to the development of acute respiratory distress syndrome and shock. ¹ Interestingly, in cases of splenic complications such as splenic rupture and infarction, disease severity does not depend on host immune status or parasitemia level; in fact, most cases of splenic rupture are seen in young, healthy males with a low parasitemia index. ³

This case emphasizes specific diagnostic challenges inherent to B. microti. Because the patient initially lacked fever, hemolysis, cytopenias, or a reported tick exposure, babesiosis was not considered as a trigger for CHF exacerbation. Diagnosis of babesiosis relies heavily on peripheral blood smear microscopy, which has limited sensitivity at low parasitemia. Parasitemia levels below 1%–2% are frequently missed, as B. microti organisms are extremely small, requiring examination of at least 300 microscopic fields under oil immersion to achieve adequate detection. Even at 2%–3% (as reported in this case), organisms may be overlooked due to their small size, morphologic similarity to artifacts, and the technical demands of microscopic examination. ⁸ PCR (including real-time and 18S rRNA reverse transcriptase PCR) offers significantly higher sensitivity and specificity but is not always readily available, may require send-out processing, and often has a turnaround time that delays confirmation in real-world clinical settings. ⁸

Cardiac complications are a significant and underrecognized feature of severe babesiosis, occurring in nearly 20% of hospitalized patients, as shown in a retrospective study of 163 adults at Yale–New Haven Hospital between 2011 and 2021. ²⁴ The most common cardiac manifestations included atrial fibrillation, congestive heart failure, QTc interval prolongation, and type 2 myocardial infarction, and these occur independently of traditional cardiovascular risk factors or preexisting cardiac disease. ²⁴ We believe that in our case, babesiosis was a trigger for acute decompensated HFpEF and that it was the first sign of infection, despite the initial absence of fever, cytopenia, and hemolysis.

Babesiosis-induced AKI is typically seen in the context of hemolysis and cast nephropathy. However, in this case, it was unlikely, as it presented early in the absence of significant hemolysis. A case reported by Luciano et al. described a patient with severe babesiosis and AKI in the absence of typical heme pigment–induced nephropathy. ² Urine microscopy revealed the novel finding of large urinary macrophages phagocytosing infected erythrocytes, raising suspicion for an inflammatory renal process. This case illustrates that B. microti may precipitate AKI via infection-triggered immune mechanisms, including interstitial inflammation, even in the absence of severe hemolysis. ² Unfortunately, due to oliguria, urinalysis could not be obtained in our patient. We speculate that acute interstitial nephritis was the primary cause of AKI in this patient, as he did not have hemolytic anemia nor a high degree of parasitemia, making hemolysis a less likely cause. Lyme disease is not recognized as a cause of AKI.

Finally, it is important to address Lyme disease in this case. The diagnosis in our patient was confirmed by a blood PCR assay detecting B. burgdorferi s.s. DNA, which was ordered as part of a tick-borne disease evaluation in light of diagnostic uncertainty at the time of patient decompensation. A positive blood PCR for B. burgdorferi DNA is diagnostic of active infection. Our patient also tested positive for both IgM and IgG antibodies. IgM antibodies are markers of early infection, whereas IgG antibodies may be positive due to prior exposure, particularly since the patient lived in an endemic area with a high incidence of Lyme disease and may have had a previous infection. The seroprevalence of IgG in individuals residing in highly endemic areas—such as the county where our patient lived—can be as high as 15%.^6,25

Conclusion

This case highlights co-infection with B. microti and B. burgdorferi presenting atypically with multiorgan failure, including fatal fulminant ALF. Clinicians, particularly those practicing in tick-borne endemic areas, should remain vigilant for atypical manifestations of these emerging zoonoses. Co-infection with B. microti and B. burgdorferi is common, and B. microti should be included in the differential diagnosis of patients presenting with ALF. In addition, B. microti infection can exacerbate congestive heart failure, even at low parasitemia levels.