Clinical Microbiology Lecture 1

27/10/2021

Clinical Microbiology Lecture 1 Transcription: Melis Engin

27/10/2021 Review: Bilinç Kavak
Prof Vittorio Sambri

Introduction
The professor firstly highlighted the course contents:

• Laboratory diagnosis of infectious disease: methods and clinical interpretation of the

microbiological data

• Microbiology of upper and lower respiratory tract infections and infections in the
immunocompromised host.

• Microbiology of the gastro-enteric tract and intra-abdominal infections

• Microbiology of the urinary tract infections

• Pyogenic, fungal, and parasitic infections of skin, bone, and joints. Prosthetic Joint infections

• Microbiology of cardiovascular infections and sepsis

• Microbiology of the central nervous system infections

• Microbiology of the zoonotic infections

Recommended sources by prof:

-Lecture slides that will be available before each class
-Textbook: the 9th edition of Medical Microbiology
by Murray, Rosenthal, and Pfaller.
Prof says that this book could also be useful during our medical career.

-He also suggested several useful websites:

https://www.uptodate.com
https://ecdc.europa.eu/en/home - Website of the European center of disease control (if you’re
curious about infective diseases)
https://www.cdc.gov/DiseasesConditions/ the same concept as ecdc but for the US

There’s theoretically an attendance requirement of 60% but the prof will never take attendance.

The final evaluation is a written, multiple-choice exam. The test, for a total of 33 questions, must be
completed within 45 minutes. (no partial exam)

During our past microbiology course with Prof.ssa Varani, you focused on germs. However, in this
course we will not only focus on germs but also on patients.

I will never tell you about the diagnosis of streptococcus because you should already know it.
I will tell you how to manage the questions that a potential patient with pharyngitis might pose to
you as a doctor. In this course we will try to understand how to get relevant info about theoretical
causes of infective diseases.

1
 Clinical Microbiology Lecture 1
27/10/2021

We will consider the most important epidemiological and clinical points of view in the field of
infectious diseases. I will also give you an idea about the diagnostic tests for those infections.
Our final goal is to make you capable of interpreting the lab reports properly; trying to understand
what is most relevant for our diagnostic process, what is less useful, and what could be potentially
even harmful.

The basic point is that you should never prescribe a lab test unless you have an idea of at least a
hypothetical diagnosis. You cannot just ask a laboratory to solve an issue that you don’t know. The
lab can help you to go in the right direction and conclude with the diagnostic hypothesis, but this
hypothesis depends on you. You cannot just order random tests, it could even be harmful to the
patient. Please bear in mind that the patient is in the centre of our system, and you should receive
the info from the patient and from other diagnostic tools to formulate a hypothesis.
Then we ask the lab to provide confirm whether our hypothesis was right. Otherwise, it does not
make sense.

For exp let’s say that we have a patient with strange dots on his/her face. We need to find out if this
is mumps, rubella, measles, or meningococcal sepsis. It depends on us to find out what it is.

Prof goes on to ask several questions:

Q1) To increase the sensitivity of the Microbiological workflow for CAP (community-acquired
pneumonia) patients which one of the following would you add to the standard culture-based
testing on sputum?: options:

– A)Ab anti S.pneumoniae

– B) PCR on bronchoalveolar lavage

– C) Blood culture

– D) PCR on sputum

The ordinary diagnostic flow is as follows: CAP microscopic examination and culture based on
sputum. However, what could we do to improve the sensitivity? The overall diagnostic sensitivity of
this diagnostic process is probably not more than 75%. That means that 25% of our patients with
CAP will remain without a precise diagnosis.
Let’s go through the possible options:
-S.pneumoniae shares common antigens with other types of streptococci so it doesn’t make sense to
use the technique indicated in A). About 30% of the population under 25 is anyway colonised in the
upper respiratory tract with S.pneumoniae. So probably someone in this class anyway has antibodies
against it, so it isn’t a useful diagnostic method.
PCR is a very sensitive technique so it will pick up any bacteria present on the upper respiratory tract
if we carry it out on the sputum. PCR is a very sensitive tool, so even if have 10 DNA copies of
streptococcus in our sample we will pick it up. Therefore, it isn’t useful for CAP. It could make more
sense for BAL though. BAL is a biological sample that comes from the lung parenchyma so it could
make more sense, but it is still not completely reliable.

The right answer here is blood culture.

The reason is that when CAP starts, more than 70% of the patients have 3-days bacteraemia caused

2
 Clinical Microbiology Lecture 1
27/10/2021

by S.pneumoniae. If the blood culture is collected too early and the result comes to be negative, we
lose very valuable information. However, the sample must be taken before the administration of
antibiotics of any kind.

Q2) Which one of the following parameters is more influencing turnaround time of BCs?:

– A)Interval (hours) between the collection of the sample and the start of incubation

– B)Incubation temperature

– C)Previously given antibiotic therapy

– D)Laboratory working hours

Let’s analyse the options:

A)When nurses take the blood sample, the transport of the blood sample to the lab isn’t always a
necessarily fast procedure. It could be a logistic issue because no one can go to the lab unless it’s
urgent so you need to depend on the regular logistic systems of the hospital. Sometimes these
samples may stay up to a couple of hours on the bench, sometimes even more.

B)If we increase the incubation temperature, it should lead to a shortening of PAT.

(Prof didn’t explain the option C)

D)The incubator gives a signal when it detects CO2 in the incubating bottles (means there’s growth
in the bottle) and this means that we must be there to collect the bottle and start investigating it.
Therefore the right answer is: the laboratory working hours
If a lab Is open 24/7 7 days a week, this means that any time that a positive blood culture bottle s
detected they can pick the bottle and start to proceed. But if a lab doesn’t work at night and/or
during the weekends (like most labs) it means that nobody is taking care of the bottle when the
machine gives a signal of CO2 detection.
Any night that you do not manage the blood culture, leads to a 12-hour long delay in the turnaround
time. Most of the microbiology labs do not have night shifts. Therefore, the information we could
provide during the night shift is not useful from a clinical point of view.

Q3) The microbiologic diagnosis of UTI is based on urine culture, which has a threshold bacterial load
value of 1x105 CFU/ml. In which of the following cases is correct to take also lower values of bacterial
load?

– A)Immune suppressed patients

– B)Elder patients

– C)Patients already treated with antibiotics

– D)Patient with catheter

The lower urinary tract is heavily colonized by various microorganisms. As a consequence, we can
observe all these colonising germs in a regular urine sample. (Prof mentions here discarding the first
(?) part of the urine and using a needle(?) to avoid contamination but the sound quality is so low
here that it’s impossible to decipher exactly what he means)
So, we must keep in mind that any urine sample we analyse in a lab has a high level of bacterial
contamination.
For this reason, if we do not take a high bacterial load as a threshold, we will reach many false

3
 Clinical Microbiology Lecture 1
27/10/2021

positives. This could lead to accidentally unnecessarily treating the colonising bacteria.
Therefore, in the question, above we can see that lower than a certain bacterial load is not
considered to be pathological. Yet there are some cases in which we should also provide information
about the low bacterial load.
The correct answer is C.

The presence of a urinary catheter leads to a higher rate of bacterial colonisation in the lower
urinary tract so that in such case our bacterial load for defining a pathology must be even higher.
Specifically, any catheter that remains positioned for longer than 72 hrs is even exposed to bacterial
biofilm formation.
All the patients who have recently been treated with antibiotics must be considered within the
pathologic range lower than the threshold reported above.

A typical situation is a patient treated with wide spectrum antibiotics such as phosphomycin (in case
of UTI). After 3 days the patient comes back to you and says, “I’m not recovering, I had felt better in
the 48 hrs” This is a clear indication that the empirical therapy that you provided to the patient
wasn’t completely effective. However, the bacterial load might have been reduced anyway. In this
case, the lab must be provided with this information so that they can adjust their threshold values
for the bacterial load.
If you want to have clinically relevant information from the lab, you must get in touch directly with
them.

Q4) In a suspected case of acute viral hepatitis with jaundice, asthenia, and a high level of AST which
one of the following additional tests would you recommend?

– Serum PCR HBV

– US-guided liver biopsy and PCR HBV

– HBeAg

– HBsAg

– HBsAb

We have a strong suspicion that this patient is suffering from acute hepatitis B. If we’d have just one
test to prescribe, which one would be the most reliable one?

The correct answer is HBsAg.

Someone in the class suggests Serum PCR-HBV. However, this is not correct because for this method
to work there must be genomic DNA present in the bloodstream. In general, this is only available
some days after the clinical picture. So, if you have a very acute case, this method is not suggested as
it can give false negatives in the beginning.
Circulating surface antigen is detection methods of today are at least as sensitive as PCR. The first
antigen detected in the serum would be the surface antigen because in this way we can get a very
reliable result even from the first day of the infection.
If you go for a biopsy, it makes no sense since you cannot ask for a biopsy from a suspected HBV
patient.
Serum PCR could be a good alternative but the detection of surface antigen costs way than serum
PCR. Even if the sensitivity of these two tests is quite similar, surface antigen detection is preferred
for cost-effectiveness performance.

4
 Clinical Microbiology Lecture 1
27/10/2021

Detecting HBsAb doesn’t make sense because this would be available only in a patient with
vaccination or previous infection.
This was to give you some idea about how clinical microbiology works.

Microbiological Diagnosis
The relevance of the diagnostic data coming from the lab accounts for up to 70% of the total
diagnostic process, meaning that the laboratory is becoming a very complex and powerful
instrument. Yet, it needs to be used in the right way. The golden rule for good interaction with the
lab is direct interaction with the lab people. What we should not do, is to ask for a specific test.

The interaction that we expect to happen between the clinic and the lab is that the clinical doctor
should provide the lab doctor with a diagnostic hypothesis. It’s impossible for them (assuming he
addresses the doctors in the lab) to know exactly which complete portfolio of tests is available in the
laboratory for each diagnostic hypothesis due to the complicated nature of the issue.

So, when you are interacting with the lab, you must say: “I have a patient with suspected
pneumonia, we are not expecting iatrogenic pneumonia as he came to the hospital himself. The
most possible scenario is a S. pneumonia infection. Please go with that idea.”

Diagnosis of Infection
Clinical microbiology is a clinical method, not a laboratory method. The data you get from
microbiological diagnosis in general prompts you to state the final diagnosis of a patient.

We have two possible strategies:

5
 Clinical Microbiology Lecture 1
27/10/2021

1- Direct Diagnosis: Detection and identification of the pathogen

This is the best option we can have. If you bear in mind the Koch postulate.
Koch postulates: are four obsolete criteria designed to establish a causative relationship between
a microbe and a disease. The postulates were formulated in 1884. The basic statement that correlates
germs and disease. Koch's postulates are the following:

1. The microorganism must be found in abundance in all organisms suffering from the
disease, but should not be found in healthy organisms.
2. The microorganism must be isolated from a diseased organism and grown in
pure culture.
3. The cultured microorganism should cause disease when introduced into a healthy
organism.
4. The microorganism must be reisolated from the inoculated, diseased experimental host
and identified as being identical to the original specific causative agent.
(You don’t need to know the postulates, it’s just to give an idea)

If I can detect the germ I’m suspecting as the causative agent of the infection in my biological
sample, the level of uncertainty is low. Yet, we will see that some level of uncertainty might still be
present.

2- Indirect Diagnosis: Detection and identification of the "traces" left by the pathogen during
infection → serological diagnosis, for exp antibody detection
Most of the results we get by using this method are just for epidemiologic uses. We can use
serologic results in very few cases to state a correlation between the antibody response and the
diagnosis.

One of the most realistic cases in which we can use this method to establish the final diagnosis is
when we detect the IgM. IgM usually appears during the very early stage of infection.
The weak point of this statement is that IgM detection is not that specific. If we think about the
molecular structure of the IgM molecule (pentamer) we can realise that it’s a very complicated and
sticky molecule. It can be trapped into a reaction just because of its complexity. The detection of IgM
is possibly tricky because we must at any time consider the low level of specificity of the result.

So even if we detect IgM, it’s likely that we’re at the early stage of an infection. But there are some
physiological conditions (eg pregnancy) when IgM detection specificity is particularly low.
Pregnant women have something in their bodies that is 50% different from themselves. So there is a
continuous stimulation of the immune system. The direct contact between the foetal circulation and
maternal circulation should be impossible through the placenta. However, what we call impossible in
medicine happens every day in medicine (:P) This means that some low level of antigenic stimulation
from the foetus might get into the maternal circulation, providing a deep level of immune activation
during pregnancy. If you detect IgM during pregnancy, this rather has an epidemiological meaning.

The best solution to get rid of this problem while using the serological approach could be waiting for
the so-called seroconversion. If you detect IgM in the first step, you must wait until the appearance
of IgG which does not appear in less than 10 days. It doesn’t make sense to have a proper report of
the infection 2 weeks after its start though. The true clinical relevance of this subtle data (indicating
serological data) is under debate.

6
 Clinical Microbiology Lecture 1
27/10/2021

Direct Diagnosis
Sampling

The most relevant thing in any diagnostic process is a good sampling. Samples must be properly
collected and stored

The sample must be representative of the disease that we are trying to diagnose.

An example is that we would not initially test the urine of a patient with suspected pneumonia. We
should at least start by sampling from the anatomical site that we suspect is the location of the
infection.

Diagnostic workflow should also respect the same rule. If we are suspecting pneumonia, we should
rather do a test on BAL rather than the sputum. (When we test the BAL, we know that whatever is in
the BAL is certainly coming from the lung parenchyma. But in sputum shows us all the bacterial flora
of the mouth) However, BAL is a very invasive procedure. So, we must see the situation from a
clinical and laboratory point of view. Sometimes obtaining the best sample possible is not easy but it
is of fundamental importance for the correct interpretation of the results.

Sterile vs Nonsterile Sites

For us to interpret the lab reports in the best way, we need to state that the human body has sterile
and non-sterile sites.

Nonsterile: In non-sterile sites, there is microbiota (normal flora) present. All the sites that are open
to the external environment usually contain germs.
Sterile: Areas close to the external environment should not contain any bacteria!
(Deep tissue samples: stuff like muscle. Muscle biopsies should be sterile)

However, these are not facts. Detection of traces of germs in potentially sterile areas is possible.
Hence, we make these statements as “They should be sterile”, not as “They are sterile”.

Saying that the blood is sterile would not be completely true. If you check the presence of 16s rRNA
fragment, a method that uses a specific rRNA segment to identify the bacteria, you can identify
traces of bacteria in at least 40 of 100 healthy blood donors. This means that the bacteria have

7
 Clinical Microbiology Lecture 1
27/10/2021

passed through these samples. They’re no longer there but they have been there, but they were
cleared out by white blood cells.

Thus, the concept of sterile sites is dynamic.

Blood is usually sterile, but you can observe transient bacteraemia and also traces of that.
The frequency of using molecular diagnostic techniques is increasing day by day that allows us to
even detect the presence of small bacterial or viral fragments. The concept of detecting genetic
material of the pathogen instead of the pathogen itself makes the diagnostic process much more
complicated. In the next 5 years (beginning of your medical career), you will be very frequently
exposed to lab reports based on molecular diagnostic techniques.

You have no idea how much the COVID-19 pandemic contributed to do establishment of molecular
diagnostic techniques in labs in the past 19 months. My laboratory is quite huge, and before we used
to perform more than 1 million 200 thousand tests per year, of which 150 thousand were molecular
tests. We had 4 instruments for molecular biology in the workflow. And the number of molecular
biology instruments we have in the lab increased to 40 after the start of the pandemic. Can you
imagine if we’d give up on all this technology that we already paid for after the end of the
pandemic?
Look at the issue from the pov of the companies that manufacture the molecular diagnostic
instruments (reagents for molecular diagnosis). 6 months after the pandemic, the total amount of
reagents in our lab was increased at least by 200 times. Nowadays huge companies like Roche,
Stevens, Abbott invested billions of euros in molecular biology diagnostics, and they made new
manufacturing plans based on that. The diagnostic era post-covid will be a molecular diagnostic era.

Before the covid era, the cost of a syndromic panel was around 30 euros. (syndromic panel: a panel
that combines the same reactions with many different targets for the same syndromic situation.)
Today it costs only 7 euros. It’s less expensive than keeping a diagnostic flow with all these separate
tests. Even people who are in the lab world for a long time can be confused due to the new
developments. Bear in mind that you will receive lots of different information and the clinical
interpretation can be quite complex.

Another example is: if you apply the syndromic panel for suspected gastrointestinal diseases to
healthy people. The symptom that allows you to state that the patient is suffering from a GI
infection is diarrhoea. If you take 100 healthy people (who have no diarrhoea) and apply the
syndromic panel, what we expect to get is 0 (supposedly they are healthy, so they should have no
GI-related bacteria traces) However this expectation wouldn’t be clinically correct. Anyone might
have pieces of DNA of these bacteria, and we don’t know exactly where they came from (exp.
giardia, norovirus, aeromonas, etc).

In other words, this is counterintuitive as we assumed that these people were healthy to start with.
Yet, let’s not forget that we’re detecting traces of DNA, not living bacteria under a microscope.

And if we do this on patients with diarrhoea, we will see the presence of more than 1 pathogen.
Then the question becomes, which one is the real pathogen that causes the problem?
The answer is, we don’t know. The only way to get out of this problem is to keep in touch with the
lab to try to discuss with them to interpret this clinical report in the best way.

8
 Clinical Microbiology Lecture 1
27/10/2021

If we can learn to interpret this powerful method (syndromic panel) in the best way, the patient
prognosis will be vastly improved. It’s faster, sensitive, and more accurate. Yet, we need to bear in
mind that detecting segments of DNA doesn’t necessarily mean detecting living and growing germs.

(The prof now goes back to explaining the methods used for direct diagnosis)

Direct diagnosis: detection and identification of pathogens

a) Direct microscopic examination

It’s old but useful.

An example of the possible practical applications of this method is checking the sputum or BAL. It’s
very informative and very quick (5 mins).
The conclusion of culture-based techniques can take up to 48 hrs so here in 5 mins you can get a lot
of information by direct microscopic examination.
Another field of application is downstream positive blood culture. When the blood culture comes
out positive, we take a smear from the incubation bottle and stain the smear with gram staining.
Then, we look at it under the microscope. This is probably the fastest (just 10 mins) and the most
important report we can combine.
If we are seeing gram positive staphylococci on a smear under the microscope, we have a strong
level of information to set up a specific therapy. If you know exact epidemiology of the staphylococci
in your clinical setting (in your ward), you can start with the specific therapy based on gram positive
staphylococci. That’s totally different from a report with gram positive streptococci. It’s a very
simple but informative information.
If you know that majority of the patients you are treating in your ward are generally affected by S.
aureus, and you know that the prevalence of methicillin resistant staphylococcus aureus in your
ward is over 35% (like any other wards in our country/region) the best solution for the therapy is to
start with vancomycin. You’d use time if you treat your patient with penicillin. This inference is
simply based on the observation of what is inside the bottle and takes only 5 mins.

Bacteriological diagnosis

(Important info: Basically, the infection causing gram-negative cocci are Neisseria. That’s the only
example of gram-negative cocci under the microscope. The shape of pathogenic gram-negative
bacteria is not often cocci. (Here the prof doesn’t talk further about gram negative rods or gram
positive cocci)

9
 Clinical Microbiology Lecture 1
27/10/2021

This is what you can get by direct examination in the case of suspected meningitis.
This is quite easy to interpret as there are plenty of polymorphonuclear leukocytes.
You can see the gram-negative diplococcus Neisseria meningitidis inside the cells.

Gram staining, CSF

If I do a nasal swab and I see polymorphonuclear leukocytes and gram-negative cocci, can I state that
the patient is suffering from Neisseria meningitidis? By nasal swab, no. Nasal secretions are full of
colonising, non-pathogenic Neisseria. This kind of information from a nasal swab means nothing but
instead from CSF, you make the statement of meningitis.

(Prof totally skips the slide with parasitological diagnosis)

b) Direct Antigen Detection

We have various technologies:

• Latex agglutination tests (latex particles covered with specific antibodies)

• Immunochromatographic tests

• Immunoenzymatic tests (a specific antibody is linked to an enzyme)

• Immunofluorescence tests (a specific antibody is linked to a fluorochrome)

In any case, the presence of antigens detected with different technologies can bring you to the
conclusion that the diagnosis is not granted, it’s suspected. (Transcriber note: I cannot make sense of
this sentence so I write it as it is)

Example: Immunochromatographic tests for the detection of Streptococcus pyogenes antigens.

Pharyngeal Swab

10
 Clinical Microbiology Lecture 1
27/10/2021

At the beginning of the lecture, we were discussing the use of urine for the diagnosis of pneumonia.
In the case of pneumonia, we can detect capsular polysaccharide antigens (deriving from the
bacterial metabolism) in urine within only 15 mins.

Let’s say that we have a 65-year-old suspected CAP (community-acquired pneumonia) patient that
has had the clinical symptoms for the past 5 days, and we get the positive urine result from the
immunochromatographic test, you can easily state that the patient that CAP caused by
pneumococci.
Based on this information you must immediately start the therapy.

If you get this positive in a 65-year-old healthy patient, the interpretation will be different. There are
two options:
a) It’s a false positive.
The specificity of this test is unfortunately not that high and therefore something in the urine can
accidentally give us a positive result.

b) It’s a true positive.

This reaction is really specific for antigens deriving from S.pneumonia. Only antigens from
S.pneumonia can be detected. The explanation could be that the patient has been vaccinated
against S.pneumonia less than 1 month ago. So the antigens might show up in the urine.
This is a great example to say that we must always interpret lab results in combination with the
clinical picture of a patient.

11