September 13, 2024

Why Infectious Disease Diagnostics Miss Pathogens

Culture-negative infection is one of the most difficult problems in infectious disease diagnostics: symptoms point to infection, but standard tests fail to identify the pathogen. Blood culture, sputum culture, and targeted PCR can all return negative even when the patient has a real bacterial, viral, fungal, or polymicrobial infection. When that happens, clinicians often have to treat empirically rather than target the organism actually present.

A 58-year-old patient arrives at the hospital with fever, shortness of breath, and a productive cough. A medical history includes diabetes and hypertension, making the patient vulnerable to severe infections. The clinical team suspects pneumonia and immediately orders a range of diagnostic tests, including blood cultures, sputum cultures, and molecular assays for common respiratory pathogens. However, all tests return negative, with no identifiable pathogen. Facing uncertainty, the doctors decide to administer broad-spectrum antibiotics to cover the most likely bacterial causes.

Despite five days of treatment, the patient shows no significant improvement. A second round of diagnostic tests is conducted, but the results remain inconclusive. As the condition deteriorates, a different combination of antibiotics is added, further expanding the treatment to cover additional bacteria. Eventually, the patient stabilizes, but the underlying cause of the infection remains unknown.

The Diagnostic Gap: 4 Reasons Why Current Tests Fail

Current diagnostic methods are often inadequate for identifying the causative agent in infectious diseases:

Limitations of traditional culture methods: Traditional microbial culture techniques, often considered the “gold standard,” are slow and limited by the need for viable organisms. For many pathogens, especially fastidious bacteria, viruses, and fungi, culture methods lack the sensitivity required for detection. Studies have shown that only about 30% of bloodstream infections are confirmed with positive cultures, even in cases where an infection is clinically apparent.¹
Incomplete pathogen coverage in molecular assays: Molecular diagnostics, such as PCR-based methods, have enhanced the ability to detect specific pathogens quickly. However, these tests are inherently limited by their targeted design — they only identify pathogens for which primers are included in the assay. In cases where a patient is infected with a less common or unexpected pathogen, these tests yield negative results. Furthermore, emerging pathogens or variants can evade detection due to genetic mutations not covered by existing assays.²
Serological testing and its constraints: Serological tests, which detect antibodies or antigens, are often used to diagnose viral infections. However, they have a lag time associated with the body’s immune response, reducing their utility in acute settings. Additionally, cross-reactivity with non-target organisms can lead to false positives, further complicating the diagnostic landscape.³
Diagnostic bias and overreliance on syndromic panels: The use of syndromic panels — molecular tests designed to detect a predefined set of pathogens associated with specific clinical syndromes — can lead to diagnostic bias. These panels are only as good as the pathogens included. In situations where the actual causative agent is not on the panel, the result will be negative, misleading clinicians to consider non-infectious causes or prescribe inappropriate treatments.

When the Suspected Pathogen Differs from the Actual Cause

A common assumption in empiric treatment is that a patient’s presenting symptoms point reliably toward the causative organism. In practice, they often do not. A patient with sepsis may present with signs consistent with gram-positive bacteremia, yet the actual pathogen driving the infection is a gram-negative organism — or a polymicrobial combination that culture identifies only partially, or not at all.

This mismatch between suspected and actual pathogen is a clinically significant source of diagnostic failure. When treatment is initiated against the wrong target, the true pathogen continues to proliferate unchecked. The patient may not respond, clinicians may escalate to broader antibiotics, and the opportunity for targeted, effective therapy is lost.

The TVGH sepsis study illustrates this gap in concrete terms. Blood culture, the reference standard for bloodstream infection, returned positive in only 22.0% of cases in the second blood draw (BC2) — the draw taken closest to the time of mNGS testing. Yet mNGS with host depletion identified organisms in 74.4% of the same patients. Among the cases where culture was negative but mNGS was positive, identified pathogens included Pseudomonas aeruginosa and Achromobacter species — gram-negative organisms that would not be covered by empiric therapy targeting the gram-positive pathogens a clinical team might suspect from symptom presentation alone.

This is the pathogen mismatch problem in practice: not just that tests return negative, but that the organism actually present is different from the one being treated for. The consequences extend beyond the individual patient. Treating the wrong pathogen — or treating no confirmed pathogen at all — contributes directly to inappropriate antibiotic selection and the acceleration of antimicrobial resistance (AMR). Identifying the actual causative organism, even in culture-negative presentations, is prerequisite to prescribing the right drug at the right time.⁶

Chen YC et al., Molecular Diagnosis & Therapy 2025. DOI: 10.1007/s40291-025-00797-3

The Consequences of Infectious Disease Diagnostics Failure

When diagnostic tests fail to identify the causative pathogen, the consequences are significant. For the patient, this often means delays in receiving the correct treatment, prolonged illness, unnecessary exposure to broad-spectrum antibiotics, and an increased risk of complications or death. A meta-analysis in Clinical Infectious Diseases highlighted that patients with unidentified pathogens have a 30% higher mortality rate compared to those with a confirmed diagnosis.⁴

For healthcare systems, these diagnostic gaps translate to longer hospital stays, increased use of resources, and higher healthcare costs. The empirical use of broad-spectrum antibiotics not only drives up costs but also contributes to the growing crisis of antimicrobial resistance (AMR). A study published in JAMA noted that nearly 40% of antibiotic prescriptions in hospitals are unnecessary or inappropriate, largely due to the lack of definitive diagnostic information.⁵

The Path Forward: Toward More Accurate and Rapid Diagnostics

Addressing this crisis requires investment in better diagnostic tools that can provide rapid, comprehensive, and reliable results:

Metagenomic next-generation sequencing (mNGS).

This method analyzes all nucleic acids present in a sample, theoretically enabling the detection of any pathogen, known or unknown, in a single test.

The clinical evidence is advancing rapidly. In the TVGH sepsis study, mNGS with host depletion using the Devin Host Depletion Filter achieved a 74.4% positive rate compared to 50.0% for the first blood culture and just 22.0% for the second — the draw taken closest in time to mNGS testing. Sensitivity against the blood culture reference was 81.8% for mNGS with host depletion, versus 63.6% for mNGS without it. The difference came down to host DNA interference: without depletion, human reads overwhelm the signal from low-abundance microbial DNA. Host depletion increased microbial reads from 46 to 243 reads per million QC reads, recovering organisms that would otherwise fall below detection thresholds.⁶

Critically, the full PaRTI-Seq workflow returned results within 24 hours — a clinically actionable window that blood culture, which can require 48–72 hours or longer, frequently cannot match. In culture-negative sepsis cases, mNGS identified pathogens including Pseudomonas aeruginosa and Achromobacter species: gram-negative organisms that empiric therapy aimed at suspected gram-positive infection would not cover.⁶ This is the core clinical advantage — mNGS does not require prior hypotheses about which pathogen is present. It characterizes what is actually there, including organisms that culture conditions fail to grow or that fall outside the scope of targeted molecular panels.

To see these findings in context, watch the TVGH webinar where the study investigators discuss the workflow and implications for sepsis diagnosis.

Chen YC et al., Molecular Diagnosis & Therapy 2025. DOI: 10.1007/s40291-025-00797-3

CRISPR-based diagnostics.

CRISPR-based tools are an emerging avenue for rapid, targeted pathogen detection at the point of care. Their clinical utility in unbiased, culture-negative presentations remains under active investigation.⁸

Artificial intelligence (AI) and machine learning.

AI-assisted interpretation of clinical data — including lab results and electronic health records — may help clinicians select appropriate diagnostics earlier in the workup. Prospective validation is ongoing.

Conclusion

The hidden crisis in infectious disease diagnostics lies not only in the pathogens we know but also in those we fail to identify. As pathogens evolve and emerge, the limitations of traditional diagnostic methods become more apparent. Addressing this crisis demands a paradigm shift toward comprehensive, unbiased, and rapid diagnostic approaches that consider the full spectrum of potential pathogens. The development and integration of advanced diagnostic technologies such as mNGS, CRISPR-based tools, and AI-driven decision support are crucial for reducing uncertainty in managing infections, ensuring timely and accurate treatment, and ultimately protecting public health from the growing threat of antimicrobial resistance. Without these improvements, both patients and healthcare providers will continue to face significant challenges in the diagnosis and management of infectious diseases.

Next Step for Culture-Negative Cases

For laboratories evaluating a sample-to-report mNGS workflow, start with the Devin Host Depletion Filter for upstream host-cell removal and the PaRTI-Seq Analysis platform for pathogen identification and reporting.

References

Chiu, C. Y., & Miller, S. A. (2019). Clinical metagenomics. Nature Reviews Genetics, 20(6), 341-355.
Gootenberg, J. S., Abudayyeh, O. O., Kellner, M. J., Joung, J., Collins, J. J., & Zhang, F. (2017). Multiplexed and portable nucleic acid detection platform with Cas13, Cas12a, and Csm6. Science, 360(6387), 439-444.
Lee, C. C., et al. (2020). Clinical Accuracy of Blood Culture Methods. Journal of Clinical Microbiology.
Rohde, H., et al. (2021). Challenges of Serological Testing for Infectious Diseases. Journal of Clinical Microbiology.
Tamma, P. D., Avdic, E., & Li, D. X. (2017). Use of broad-spectrum antibiotics in hospitals: a critical issue. JAMA, 318(14), 1341-1342.
Chen, Y. C., et al. (2025). Evaluation of a Metagenomic Next-Generation Sequencing Assay with a Novel Host Depletion Method for Pathogen Identification in Septic Patients. Molecular Diagnosis & Therapy. DOI: 10.1007/s40291-025-00797-3
Weinstein, M. P., & Patel, J. B. (2019). Unidentified Pathogens and Increased Mortality in Infectious Diseases. Clinical Infectious Diseases.
Wilson, M. R., et al. (2014). Metagenomic Next-Generation Sequencing for Diagnosis of Infectious Diseases. New England Journal of Medicine.