Fever in Internal Medicine: Distinguishing Infectious from Autoimmune Etiologies

Dr Neeraj Manikath , claude.ai

Abstract

Fever remains one of the most common presenting complaints in internal medicine, yet distinguishing between infectious and autoimmune causes presents a diagnostic challenge that significantly impacts therapeutic decisions. This review synthesizes current evidence on the pathophysiology, clinical patterns, and diagnostic approaches to differentiate these etiologies, with practical insights for postgraduate trainees navigating complex febrile presentations.

Introduction

Fever, defined as a core body temperature exceeding 38°C (100.4°F), represents a regulated response orchestrated by endogenous pyrogens acting on the hypothalamic thermoregulatory center. While infections account for the majority of acute febrile illnesses, autoimmune and autoinflammatory conditions represent crucial differential diagnoses, particularly in prolonged or recurrent fever syndromes. The distinction carries profound therapeutic implications: inappropriate immunosuppression in infection risks septic complications, while delayed recognition of autoimmune disease perpetuates organ damage.

Pathophysiological Distinctions

The Pyrogenic Cascade

Both infectious and autoimmune fevers share common terminal pathways through interleukin-1 (IL-1), interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-α), which stimulate prostaglandin E2 synthesis in the hypothalamus. However, the upstream triggers differ fundamentally. In infection, pathogen-associated molecular patterns (PAMPs) activate toll-like receptors on immune cells, initiating cytokine release. Conversely, autoimmune conditions generate fever through damage-associated molecular patterns (DAMPs), immune complex deposition, or dysregulated cytokine networks independent of microbial invasion.

Pearl: The fever response itself cannot reliably distinguish etiology because the final common pathway is identical. The clinical context and pattern matter more than absolute temperature.

Inflammatory Biomarker Profiles

Procalcitonin (PCT) has emerged as a valuable discriminator. Bacterial infections typically elevate PCT above 0.5 ng/mL, with levels correlating with severity. Viral infections and most autoimmune conditions maintain PCT below 0.5 ng/mL. However, this distinction has important exceptions: severe systemic autoimmune disease may modestly elevate PCT, and localized bacterial infections may not significantly raise levels.

C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) lack specificity, rising in both infectious and inflammatory conditions. However, certain patterns provide clues: extremely elevated ESR (>100 mm/hr) with disproportionately modest CRP elevation suggests multiple myeloma, temporal arteritis, or polymyalgia rheumatica rather than infection.

Hack: In ambiguous cases, serial PCT measurements over 24-48 hours prove more useful than single values. Bacterial infections show rising or persistently elevated PCT, while inflammatory conditions typically show stable or declining levels without antimicrobial therapy.

Clinical Pattern Recognition

Fever Characteristics

Temperature patterns: Historical emphasis on fever patterns (quotidian, tertian, remittent) has limited modern utility, but certain observations remain valuable. Sustained fever without diurnal variation, particularly exceeding 39.5°C for weeks, favors disseminated infection or lymphoproliferative disease. Quotidian fevers (daily spikes) characterize Still's disease (adult-onset Still's disease, AOSD) and juvenile idiopathic arthritis. Double quotidian patterns (two daily spikes) suggest mixed infections, tuberculosis, or endocarditis rather than autoimmune disease.

Oyster: Artificially produced fever (factitious fever) should enter the differential when temperature-pulse dissociation occurs or when fever patterns seem implausibly regular. Request supervised temperature measurements and examine the thermometer immediately after use.

Associated Symptoms and Physical Examination

Certain clinical constellations strongly suggest specific etiologies:

Infectious clues:

• Rigors and shaking chills strongly favor bacteremia or parasitic infections over autoimmune disease
• Neutrophilic leukocytosis with left shift (bands >10%) indicates bacterial infection
• Purulent drainage, focal consolidation, or abscess formation
• Night sweats soaking bed clothes suggest tuberculosis, lymphoma, or endocarditis

Autoimmune indicators:

• Photosensitive rash, oral ulcers, or malar rash suggest systemic lupus erythematosus (SLE)
• Evanescent salmon-pink rash accompanying fever spikes characterizes AOSD
• Symmetric small-joint polyarthritis points toward rheumatoid arthritis or viral arthropathy
• Purpura, particularly palpable purpura, indicates vasculitis or endocarditis

Pearl: The Koebner phenomenon (lesions appearing at sites of minor trauma) occurs in psoriasis, sarcoidosis, and Behçet's disease but not in infectious exanthems.

Temporal Patterns

Acute fever (<1 week): Infections dominate this timeframe. Consider bacterial pneumonia, urinary tract infection, influenza, or COVID-19. Autoimmune presentations are rare but include acute lupus flares or drug-induced hypersensitivity syndromes.

Subacute fever (1-3 weeks): The differential broadens significantly. Infectious considerations include subacute bacterial endocarditis, tuberculosis, HIV seroconversion, and abscess formation. Autoimmune possibilities include lupus, AOSD, and drug-induced syndromes.

Chronic fever (>3 weeks): This constitutes fever of unknown origin (FUO) when temperature exceeds 38.3°C on multiple occasions without diagnosis despite one week of intelligent investigation. Classic FUO etiologies distribute roughly equally among infections (25-30%), malignancies (20-30%), autoimmune diseases (20-30%), and miscellaneous or undiagnosed causes (15-20%).

Diagnostic Approach

Initial Laboratory Evaluation

The foundational workup should include:

Complete blood count with differential: Leukocytosis with neutrophilia suggests bacterial infection; lymphocytosis indicates viral infection or certain chronic conditions; leukopenia may occur in viral infections, SLE, or overwhelming sepsis. Atypical lymphocytes point toward infectious mononucleosis or CMV.

Hack: Absolute monocyte count >1000/μL (monocytosis) warrants consideration of tuberculosis, subacute bacterial endocarditis, or temporal arteritis.

Acute phase reactants: CRP, ESR, procalcitonin, and ferritin. Extremely elevated ferritin (>2000 ng/mL) with low glycosylated ferritin percentage (<20%) strongly suggests AOSD over infection. Ferritin >10,000 ng/mL narrows the differential to AOSD, hemophagocytic lymphohistiocytosis (HLH), or septic shock.

Liver enzymes: Marked transaminase elevation (>5x upper limit of normal) occurs in viral hepatitis, drug reactions, autoimmune hepatitis, or AOSD. Modest elevation appears in numerous infectious and inflammatory conditions.

Urinalysis: Pyuria without bacteriuria suggests renal tuberculosis, interstitial nephritis, or sterile pyuria of autoimmune disease. Proteinuria and hematuria with dysmorphic RBCs indicate glomerulonephritis.

Microbiological Studies

Blood cultures remain essential, ideally drawn during fever spikes before antimicrobial administration. Minimum two sets from separate venipunctures maximize yield. For suspected endocarditis, obtain three sets over 24 hours. Consider fungal and mycobacterial cultures in appropriate clinical contexts.

Serology deserves cautious interpretation. HIV, EBV, CMV, and hepatitis serologies help establish acute infections. However, positive autoantibodies require clinical correlation: low-titer ANA occurs in 30% of healthy individuals and rises with age. High-titer ANA (>1:640) or specific antibodies (anti-dsDNA, anti-Smith, anti-Scl-70) carry greater diagnostic weight.

Oyster: Infection itself may induce transient positive autoantibodies. Mycoplasma pneumonia produces cold agglutinins and false-positive heterophile antibodies. Bacterial endocarditis generates positive rheumatoid factor and cryoglobulins. Always confirm autoimmune diagnoses with compatible clinical syndromes rather than serology alone.

Imaging Strategies

Initial chest radiography identifies pneumonia, tuberculosis, or sarcoidosis. CT imaging proves invaluable for occult infections: abdominal/pelvic CT detects abscesses, appendicitis, or diverticulitis; chest CT reveals pulmonary emboli or mediastinal lymphadenopathy.

Hack: FDG-PET/CT has revolutionized FUO investigation, demonstrating sensitivity exceeding 85% for inflammatory and infectious foci. However, interpret cautiously: increased uptake occurs in infection, malignancy, and inflammation. PET-positive findings require histopathological confirmation.

Echocardiography (transesophageal superior to transthoracic) remains mandatory when endocarditis enters the differential, particularly with bacteremia, new murmur, or embolic phenomena.

The Role of Biopsy

Tissue diagnosis resolves many diagnostic dilemmas. Temporal artery biopsy definitively diagnoses giant cell arteritis. Lymph node biopsy distinguishes lymphoma from Kikuchi disease or tuberculosis. Bone marrow examination identifies hemophagocytic syndrome, disseminated infections, or hematologic malignancies.

Pearl: When systemic autoimmune disease is suspected but diagnosis remains elusive, consider skin biopsy even of normal-appearing skin. Lupus band test and leukocytoclastic vasculitis may be demonstrated in clinically uninvolved areas.

Specific Clinical Scenarios

Adult-Onset Still's Disease

AOSD exemplifies diagnostic challenges at the infection-autoimmunity interface. Yamaguchi criteria require fever (≥39°C lasting ≥1 week), plus arthralgia, evanescent rash, leukocytosis (≥10,000/μL with ≥80% neutrophils), and exclusion of infection and malignancy. The diagnosis remains clinical because no pathognomonic test exists.

Hack: Calculate the Pouchot score incorporating ferritin (>3000 ng/mL), glycosylated ferritin (<20%), and clinical features. Scores above certain thresholds support AOSD diagnosis, but infection exclusion remains paramount before immunosuppression.

Systemic Lupus Erythematosus

Fever occurs in 50-60% of lupus patients during disease course. However, fever may indicate infection rather than disease activity—a critical distinction given lupus patients' increased infection susceptibility from both disease and immunosuppressive therapy.

Favor lupus flare over infection when fever accompanies:

• New or worsening lupus-specific manifestations (serositis, nephritis, CNS involvement)
• Rising anti-dsDNA titers with falling complement levels
• Absence of localizing infectious symptoms
• PCT <0.25 ng/mL

Conversely, rigors, focal findings, or PCT >0.5 ng/mL mandate infection investigation despite apparent lupus activity. When uncertain, obtain appropriate cultures and consider empiric antimicrobials while pursuing workup.

Hemophagocytic Lymphohistiocytosis

HLH represents a hyperinflammatory syndrome triggered by infections (especially EBV), malignancy, or autoimmune disease. The HScore calculator incorporates clinical and laboratory parameters to estimate HLH probability. Diagnostic features include fever, splenomegaly, cytopenias, hypertriglyceridemia, hypofibrinogenemia, elevated ferritin, and hemophagocytosis on bone marrow.

Oyster: HLH may complicate known autoimmune disease (macrophage activation syndrome in AOSD or systemic juvenile idiopathic arthritis) or unmask underlying rheumatologic conditions. The syndrome itself requires immunosuppressive therapy despite frequent infectious triggers—a therapeutic paradox requiring specialized management.

Drug-Induced Fever and DRESS Syndrome

Drug reaction with eosinophilia and systemic symptoms (DRESS) typically manifests 2-8 weeks after medication initiation with fever, rash, eosinophilia, and internal organ involvement. Common culprits include anticonvulsants, allopurinol, sulfonamides, and minocycline.

Simple drug fever occurs sooner (median 7-10 days) without eosinophilia or organ dysfunction. Both mimic infection and autoimmune disease. Resolution typically occurs within 48-72 hours of drug discontinuation, though DRESS may persist weeks.

Therapeutic Trials as Diagnostic Tools

Empiric Antimicrobials

In critically ill patients, empiric broad-spectrum antimicrobials are mandatory pending diagnostic clarification—the mortality of untreated sepsis vastly exceeds risks of brief unnecessary antibiotics. However, in stable patients with chronic fever, premature antimicrobials complicate diagnosis by suppressing culture yield and obscuring clinical patterns.

Pearl: When empiric therapy seems necessary but diagnosis remains uncertain, obtain all cultures first, then administer antimicrobials. Reassess after 48-72 hours: clear improvement suggests infection, lack of response favors non-infectious causes.

Corticosteroid Trials

Diagnostic corticosteroid trials deserve extreme caution. While dramatic responses to prednisone suggest autoimmune disease, certain infections (tuberculosis, histoplasmosis) may temporarily improve with steroids due to anti-inflammatory effects before subsequently worsening. Furthermore, corticosteroids suppress diagnostic test sensitivity.

Corticosteroid trials should only follow:

• Extensive infection investigation including tuberculosis evaluation
• Clear clinical syndrome suggesting specific autoimmune diagnosis
• Documentation that delayed therapy risks irreversible organ damage
• Close monitoring with predetermined response criteria and reassessment timeline

Conclusion

Distinguishing infectious from autoimmune fever requires systematic clinical reasoning integrating epidemiology, temporal patterns, physical findings, and targeted investigations. While no single test definitively differentiates these etiologies, a methodical approach combining pattern recognition, judicious use of biomarkers like procalcitonin, appropriate imaging, and tissue diagnosis when needed resolves most cases. Premature diagnostic closure and inappropriate empiric therapy represent the greatest pitfalls. In the modern era of advanced diagnostics, fewer fevers should remain genuinely unknown—persistence and thoroughness ultimately yield answers in most cases.

The fundamental principle remains: exclude infection before immunosuppressing, but recognize autoimmune disease before irreversible damage accrues. This balance defines the art of internal medicine.

Key References

1. Mourad O, Palda V, Detsky AS. A comprehensive evidence-based approach to fever of unknown origin. Arch Intern Med. 2003;163(5):545-551.

2. Nakanishi M, Harigai M. New biomarkers for the diagnosis and monitoring of adult-onset Still's disease. Diagnostics (Basel). 2021;11(6):1074.

3. Simon L, Gauvin F, Amre DK, Saint-Louis P, Lacroix J. Serum procalcitonin and C-reactive protein levels as markers of bacterial infection: a systematic review and meta-analysis. Clin Infect Dis. 2004;39(2):206-217.

4. Mulders-Manders C, Simon A, Bleeker-Rovers C. Fever of unknown origin. Clin Med (Lond). 2015;15(3):280-284.

5. Ruscitti P, Cipriani P, Ciccia F, Di Benedetto P, Liakouli V, Berardicurti O, et al. H-ferritin and procalcitonin as diagnostic markers for adult-onset Still's disease. J Rheumatol. 2020;47(3):417-424.

6. Henter JI, Horne A, Aricó M, et al. HLH-2004: Diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer. 2007;48(2):124-131.

7. Wright WF, Auwaerter PG. Fever and fever of unknown origin: review, recent advances, and lingering dogma. Open Forum Infect Dis. 2020;7(5):ofaa132.

8. Govoni M, Bortoluzzi A, Padovan M, et al. The diagnosis and clinical management of the neuropsychiatric manifestations of lupus. J Autoimmun. 2016;74:41-72.