# The Limits of Bacterial Infection Tests: Why Lab Results Can't Stand Alone

This review article examines the key limitations of two major diagnostic tests for bacterial infections—serology (antibody blood tests) and PCR (DNA detection tests)—using Lyme disease as a central example. The authors demonstrate that serological tests are prone to cross-reactions because many bacterial species share common genes and proteins, leading to false positives, while PCR testing carries risks of laboratory contamination and cannot distinguish between live and dead bacteria. Drawing on historical examples from syphilis, typhus, and rickettsial diseases, the authors conclude that no microbiological test should ever be interpreted in isolation—the patient's clinical presentation and epidemiological context (where they live, what they were exposed to) are essential for accurate diagnosis.

# The Limits of Bacterial Infection Tests: Why Lab Results Can't Stand Alone

## Table of Contents

- Key Points
- Background: Why This Research Matters
- How This Review Was Conducted
- Understanding Bacterial Serology: How Antibody Tests Work
- The Syphilis Warning: A Cautionary Tale from Medical History
- Historical Cross-Reactions: When Tests Point to the Wrong Bacteria
- PCR Testing: A Powerful Tool with Hidden Pitfalls
- How Reliable Are These Tests Really?
- Clinical Implications: What This Means for Patients
- Limitations of This Review
- Recommendations for Patients
- Frequently Asked Questions
- Source Information

## Key Points

- Serological tests for bacteria, including Lyme, often cross-react, causing false positives.
- PCR detects DNA but cannot tell if bacteria are alive or dead.
- Lab results must be interpreted with clinical symptoms and epidemiological context.
- Early Lyme with a bull's-eye rash should be treated without waiting for lab tests.
- Use only accredited human diagnostic laboratories, not veterinary or unverified services.

## Background: Why This Research Matters

Lyme disease is one of the most controversial infections in modern medicine. Diagnosis is currently based on serology—an indirect diagnostic method that looks for antibodies in the blood rather than detecting the bacteria themselves. This is because the bacteria that cause Lyme disease, from the *Borrelia burgdorferi* sensu lato complex, are extremely difficult to grow in laboratory culture. They also only appear briefly in the bloodstream during the very early stage of infection, a phenomenon called "highly transitory bacteremia" [1].

The only direct diagnostic method that can be useful—with certain specimens such as skin biopsies or aspirated fluids—is PCR (polymerase chain reaction), a technique that amplifies and detects bacterial DNA. This review, published in *Médecine et maladies infectieuses* in 2019, details the main limitations of both serology and PCR testing for diagnosing bacterial infections, using examples drawn from the recent history of microbiology.

The authors emphasize that current controversies about Lyme disease diagnosis highlight a much broader issue: the relevance and reliability of diagnostic tools for all bacterial infections. Understanding these limitations matters because a wrong test result can lead to unnecessary treatments, missed diagnoses, or false assurance.

## How This Review Was Conducted

This is a review article, not a new clinical study. The authors—C. Eldin, P. Parola, and D. Raoult from the IHU-Méditerranée Infection at Aix-Marseille University in France—systematically examined historical and contemporary examples of diagnostic challenges in microbiology. They drew on published case reports, prior research studies, and documented episodes from medical history where serological or PCR testing produced misleading results.

The review covers more than a century of medical evidence, from the 1916 typhus epidemic during World War I to modern studies of Lyme disease seroprevalence in French forestry workers published in 2016. This historical approach allows the authors to demonstrate that the limitations they describe are not theoretical—they have caused real diagnostic errors throughout modern medical history.

## Understanding Bacterial Serology: How Antibody Tests Work

Bacterial serology is fundamentally different from viral serology. Viruses have great genetic variability, which means they produce specific and rare proteins that carry an extremely low risk of cross-reactions. Bacteria, however, are different. They share a pool of common genes—especially genes that encode for the ribosome (the protein-building machinery of cells) and genes encoding for heat-shock proteins (proteins produced in response to stress)—to which human serum is naturally reactive [6].

Additionally, bacterial cell wall components called peptidoglycan and lipopolysaccharide are associated with numerous cross-reactions. This means that antibodies produced against one type of bacteria can mistakenly bind to a different type of bacteria, because they share similar molecular structures.

For infections that can only be diagnosed by serology, these cross-reactions create persistent interpretation difficulties. As the authors note, "The history of microbiology offers several examples of bacterial serologies based on cross-reactions." While serological testing can still contribute to diagnosis when interpreted alongside the clinical context, these inherent limitations mean that bacterial serology "is imperfect and will always be imperfect."

## The Syphilis Warning: A Cautionary Tale from Medical History

Syphilis, caused by the bacterium *Treponema pallidum* (a spirochete similar to *Borrelia*), provides one of the most instructive examples of the dangers of unreliable serological tests. The original serological test for syphilis, known as the "Wasserman test," was based on the detection of a substance called human cardiolipin [2]. This test was positive in many situations other than actual syphilis infection—and could be only temporarily positive—during pregnancy, viral infections, parasitic infections, and autoimmune diseases [3-5].

The consequences were serious. Between the First and Second World Wars, physicians believed that Rh incompatibility between mother and child (a blood type mismatch that can cause complications in pregnancy) was actually due to seropositive or seronegative congenital syphilis. This mistaken belief arose directly from the Wasserman test's frequent false positives. As the authors note, "nontreponemal serological tests used in the diagnosis of syphilis have always lacked specificity."

As early as the 1980s, the authors' team and others demonstrated the existence of cross-reactions among all spirochetes, particularly between the causative agents of syphilis, leptospirosis, and *Borrelia* infections [6]. They found that the cross-reaction between *Borrelia* and *T. pallidum* could be partly prevented by first absorbing the serum with Reiter's treponema (a non-pathogenic strain that can be cultured in the laboratory) [7]. The team routinely used this prior absorption step before performing Lyme serology to avoid cross-reactions with other spirochetes [8].

The development of the confirmatory Western Blot technique (a more detailed test that looks for antibodies against specific bacterial proteins) was associated with better specificity [9,10]. However, even the Western Blot cannot distinguish between *Borrelia* species that cause recurrent fever and those that cause Lyme disease [11]. This distinction matters clinically because the diseases require different treatment approaches.

## Historical Cross-Reactions: When Tests Point to the Wrong Bacteria

The authors present several striking historical examples where cross-reactions led to incorrect diagnostic conclusions and even whole disease theories built on false test results.

### The Weil-Felix Test and Typhus

In 1916, during major typhus epidemics sparked by the First World War, researchers Edmund Weil and Arthur Felix developed a serological test based on the discovery that the serum of patients with typhus cross-reacted with bacteria of the *Proteus* genus [12]. It was only in the 1940s that serological tests based on *Rickettsia* antigens (the true causative agents of typhus) were developed.

Yet for many years—and still today in countries such as India—the Weil-Felix test remained in use, with specific *Proteus* strains used as proxies for different rickettsial diseases [13-15]:

- OX19 (*Proteus vulgaris*): cross-reacts with typhus-group *Rickettsia* species
- OX2 (*Proteus vulgaris*): cross-reacts with spotted-group *Rickettsia* species
- OXK (*Proteus mirabilis*): cross-reacts with *Orientia tsutsugamushi*

The lipopolysaccharides of *Proteus* bacteria are responsible for these cross-reactions, not only with *Rickettsia* species but also with *Legionella* species [14-16].

### The Giroud Microagglutination Test: False Beliefs and Misdiagnoses

Another example comes from the research field of *Rickettsia* species. The microagglutination test developed by Giroud in the 1940s [17] had very poor specificity and produced many false positive results—reportedly 25–30% of results were false positives, as reported by Edlinger and Raoult [18,19].

This technique led several authors to believe that diseases without any proven cause were actually rickettsial infections. These included:

- Multiple sclerosis
- Vasculitis (inflammation of blood vessels), including Buerger's disease (a rare disease of the blood vessels in the hands and feet)
- Schizophrenia
- "Tetany" (a condition of involuntary muscle spasms)

The technique was abandoned in France because of its lack of specificity, but it is still used in some regions of the world, such as South Africa, where some practitioners claim that *Rickettsia* species are responsible for syndromes such as chronic fatigue and fibromyalgia [25]. The authors flag this as an ongoing concern.

### Modern Cross-Reaction Data

These cross-reactions are not just historical curiosities; they have been measured in modern studies. Musso and colleagues demonstrated that up to **35% of patients with *Coxiella burnetii* infection** (Q fever) had a cross-reaction with *Legionella micdadei*, which could lead to incorrect attribution of pulmonary involvement [26]. Another study reported that the serum of patients with a persistent *C. burnetii* infection was positive for *Bartonella henselae* in **50% of cases**, as detected by serology [27]. These cross-reactions are again mainly due to shared proteins and lipopolysaccharide antigens.

## PCR Testing: A Powerful Tool with Hidden Pitfalls

PCR testing—which amplifies and detects bacterial DNA—has contributed to major advances in human microbiology. However, like any diagnostic test, it carries risks of both false positive and false negative results.

### The Contamination Problem

False positive results can occur through nonspecific amplification of environmental contaminants or, in laboratories performing routine PCR tests, of previously amplified products that linger in the workspace. This is why laboratories performing these tests must meet specific certification requirements.

In France, external validation by reference centers for infectious diseases is required for all commercialized tests [28]. These reference centers have extensive sample collections and perfectly documented specimens, including samples from various stages of disease and from "healthy" patients without the disease. Laboratories performing medical diagnoses must also obtain COFRAC certification—this agency has the monopoly on evaluating procedures in French medical laboratories. This certification is now mandatory in both the private and public sectors, and every step of the procedure, every reagent, and every technique must be validated. Positive and negative controls must be performed with each test run.

### The Veterinary Laboratory Problem

The authors raise an important patient safety issue: some patients who "doctor-hop" seeking a diagnosis have their samples tested—sometimes with a prescription from a physician—in **veterinary laboratories**. These laboratories do not have to comply with the obligations and certifications required for human diagnosis. Results from such laboratories are therefore not reliable.

These biological results typically do not inform the patient about:

- The technique used for testing
- Whether positive and negative controls were performed
- The specific method employed

The authors state clearly: "Such practices cannot be accepted in the diagnosis of human infectious diseases."

### The Viability Question: Does Detected DNA Mean Live Bacteria?

Beyond false positives and technical quality, PCR diagnosis raises a fundamental question: the viability of the microorganisms whose DNA is detected in tissues—and whether that DNA has any clinical meaning.

The authors' team reported a case of *Streptococcus pneumoniae* endocarditis (an infection of the heart valves) where PCR remained positive for *S. pneumoniae* on heart valve tissue **seven years after** the endocarditis episode, even though the patient was clinically cured [29]. A similar case was reported with *Streptococcus mutans* endocarditis, with a positive PCR on heart valve tissue **31 months after** the episode [30].

Both cases raise the question of residual bacterial DNA that persists without any particular meaning. Several in vitro studies have also demonstrated that the DNA of bacteria killed by ultraviolet rays, boiling water, acid, or autoclaving (sterilization) could still be detected by PCR. The same likely holds true in the tissues of living organisms [29].

For Lyme disease specifically, the authors emphasize that studies attempting to demonstrate persistent *Borrelia* infection based solely on PCR detection—without any specific clinical signs—cannot prove the existence of an active infection. Dead bacterial DNA can linger for years, giving a positive PCR result that has no clinical relevance.

## How Reliable Are These Tests Really?

How can the "reliability" of a test be assessed? The authors explain that clinicians usually talk about the "performance" of a test, characterized by its sensitivity (how good the test is at detecting the disease when it is truly present) and specificity (how good the test is at avoiding false positives when the disease is absent).

However, the usefulness of a test in a real-world clinical setting depends on another factor: the predictive value. This is determined by the test's intrinsic characteristics (sensitivity and specificity) **and** by the likelihood of having the disease before the test is performed (the prevalence of the disease in the patient's population).

Lyme disease diagnosis does not rely solely on the result of a serological test or a PCR test. Clinical and epidemiological characteristics—like travel history, exposure to ticks, and physical symptoms—must be considered in the test interpretation. The authors note this holds "for Lyme disease, but also for any other disease."

Two examples illustrate this principle:

**Example 1:** A positive serology in a patient who does not present with any objective skin lesions suggestive of Lyme disease (like the characteristic "bull's-eye" rash called erythema migrans), and who does not live in or travel to an endemic area, has a **predictive value close to zero**. This means the test result, while technically positive, is essentially meaningless without supporting clinical and geographical context.

**Example 2:** When performing seroprevalence studies in individuals heavily exposed to the disease—for instance, forest rangers living in Alsace, France—up to **14% of them can have a positive serology** without presenting any sign of the disease [31]. The proportion of people who develop Lyme disease after contact with *Borrelia burgdorferi* is variable, but the presence of the bacterium is not necessarily associated with disease onset in 100% of individuals following a tick bite.

## Clinical Implications: What This Means for Patients

The core message of this review is that biological results cannot be interpreted on their own. Physicians' knowledge is required to interpret biological results in light of the patient's clinical features. This applies to Lyme disease and to all other bacterial infections.

For patients, several important implications follow:

-
**A positive antibody test does not necessarily mean active infection.** It could reflect a past infection, a cross-reaction with another bacterium, or (in highly exposed populations) exposure without disease.
-
**A positive PCR test does not necessarily mean live bacteria are present.** Dead bacteria can leave behind detectable DNA for months or even years.
-
**Testing at unaccredited laboratories can produce unreliable results.** Laboratories performing human diagnostic tests must meet strict certification standards (like COFRAC in France); results from veterinary laboratories or non-certified facilities are not dependable.
-
**For early acute Lyme disease, the clinical picture matters most.** The authors note that early acute infections have well-defined clinical characteristics and "should thus be treated without biological diagnosis." This means that a characteristic bull's-eye rash is enough to diagnose and treat Lyme disease—you don't need to wait for lab results.

## Limitations of This Review

The authors acknowledge that the issue of subjective symptoms—such as fatigue, pain, and cognitive difficulties—in people with tick bites and/or in people who never had confirmed Lyme disease "will not be solved based on current medical and scientific knowledge." This is an honest admission that major knowledge gaps remain.

However, they emphasize that no current literature data suggests that *Borrelia burgdorferi* sensu lato species play a role in causing chronic subjective symptoms. The history of microbiology shows that similar claims were made about other intracellular bacteria (like *Rickettsia* species) based on unreliable serological testing—and those claims were ultimately proven false once better tests were developed.

## Recommendations for Patients

Based on this review, patients should consider the following actionable advice:

1.
**If you have a classic Lyme disease rash (erythema migrans), seek treatment promptly.** According to the authors, early acute Lyme disease has well-defined clinical characteristics and should be treated without waiting for biological diagnosis. Don't delay treatment for lab tests.
1.
**If you receive a positive serology result, ask your doctor how it fits with your symptoms.** A positive antibody test alone, especially without objective signs of disease or exposure in an endemic area, may have a predictive value close to zero.
1.
**Make sure your tests are performed at an accredited laboratory.** In France, this means a laboratory with COFRAC certification. In other countries, ask about laboratory accreditation standards. Avoid veterinary laboratories or direct-to-consumer testing services that don't follow human diagnostic standards.
1.
**Understand the limits of PCR testing.** A positive PCR result does not tell you whether the bacteria are alive. Dead bacterial DNA can persist for years after cure (as shown in the heart valve cases).
1.
**Be skeptical of diagnoses based solely on lab results.** Both serology and PCR must be interpreted in the context of your clinical presentation (symptoms, physical findings) and epidemiological situation (where you live, travel history, tick exposure).
1.
**If you have persistent symptoms after a tick bite without confirmed Lyme disease, know that this is still an unresolved issue.** The authors state that current science cannot resolve these subjective manifestations—but they also emphasize there is no evidence that *Borrelia* species cause them.

## Frequently Asked Questions

### Can a positive Lyme disease antibody test be wrong?

Yes. Antibody tests for Lyme disease can produce false positives because many bacteria share common proteins and structures. Your immune system may make antibodies that react to more than one type of bacteria. A positive result could reflect a past infection, exposure without illness, or cross-reaction with another bacterium. Doctors must interpret it with your symptoms and exposure history.

### Should I get a Lyme disease test if I have a bull's-eye rash?

According to the review, early acute Lyme disease has well-defined clinical features, like the characteristic erythema migrans rash, and should be treated without waiting for laboratory tests. A clinical diagnosis is sufficient for early infection. Delaying treatment for lab results is not recommended. Always consult a healthcare provider promptly for a rash after a tick bite.

### Why do doctors consider where I live and my travel history when interpreting a positive Lyme test?

The predictive value of a test depends on how likely you are to have the disease before testing. If you have no objective signs of Lyme disease and no exposure in an endemic area, a positive serology may have a predictive value close to zero. Even among heavily exposed forest rangers in Alsace, up to 14% had positive serology without any signs of disease.

### Are results from veterinary laboratories reliable for diagnosing human infections?

No. Veterinary laboratories are not required to follow the same certification standards and quality controls as human diagnostic laboratories. Their results may not include information about the technique used, whether positive and negative controls were performed, or the method employed. The authors state such practices cannot be accepted in the diagnosis of human infectious diseases.

### Can a positive PCR test for Lyme mean I have an active infection?

Not necessarily. A positive PCR result only indicates the presence of bacterial DNA, not whether the bacteria are alive. Studies attempting to demonstrate persistent Borrelia infection based solely on PCR detection, without specific clinical signs, cannot prove an active infection. Dead bacterial DNA can remain in tissues for years, so PCR must be interpreted alongside clinical findings.

### What should I do if I have persistent symptoms after a tick bite but no confirmed Lyme disease?

This is still an unresolved issue in medicine. Current scientific knowledge cannot fully explain subjective symptoms like fatigue, pain, and cognitive difficulties in people with tick bites or without confirmed Lyme disease. The review found no evidence that Borrelia species cause these chronic subjective symptoms. Discuss your symptoms with a healthcare provider for proper evaluation.

### When should I seek a second opinion about a Lyme disease test result?

A positive antibody test alone may be nearly meaningless if there are no objective signs like the bull’s-eye rash and no tick exposure in an endemic area. A positive PCR test cannot tell whether bacteria are alive or dead. Testing at unaccredited laboratories, such as veterinary labs, is unreliable. If a diagnosis relies only on lab results, a second opinion can help interpret findings in clinical and epidemiological context. For early Lyme disease with a classic rash, treatment is advised without waiting for lab tests. Diagnostic Detectives Network provides independent expert second opinions.

## Source Information

**Original Article:** "Limitations of diagnostic tests for bacterial infections" (Les limites des tests diagnostiques des infections bactériennes)

**Authors:** C. Eldin, P. Parola, D. Raoult (corresponding author: didier.raoult@gmail.com)

**Affiliations:** IRD, SSA, VITROME, IHU-Méditerranée Infection, Aix-Marseille University, AP–HM, Marseille, France; and IRD, MEPHI, IHU-Méditerranée Infection, Aix-Marseille University, AP–HM, Marseille, France

**Publication:** *Médecine et maladies infectieuses*, Volume 49, 2019, pages 98–101. Published by Elsevier Masson SAS. Received May 24, 2018; accepted December 19, 2018; available online January 24, 2019.

**DOI:** https://doi.org/10.1016/j.medmal.2018.12.004

**Access:** This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

*Note: This patient-friendly article is based on peer-reviewed research published in a scientific journal. It is intended for educational purposes and does not replace professional medical advice. Always consult a qualified healthcare provider for diagnosis and treatment recommendations.*


---
Publisher: Diagnostic Detectives Network (https://diagnosticdetectives.com) — independent multi-expert medical second opinions, worldwide, private-pay.
Author byline: Anton Titov, MD, PhD. Contact: https://diagnosticdetectives.com/pages/contact
Canonical page: https://diagnosticdetectives.com/products/the-limits-of-bacterial-infection-tests-why-lab-results-cant-stand-alone