Electron Microscopy

[Enkidu]

---

Members don't see this ad.

It seems like in medical school we learn about electron microscopy in relation to kidney disease, mostly. Do other branches of pathology routinely diagnose based on EM findings, and how difficult is it to do? I would imagine that it takes some time to examine an electron micrograph and it takes some specialization in it as well.

I was reading Differential Diagnosis in Surgical Pathology, and it listed EM findings for a lot of diseases, but it wasn't clear to me that this was actually part of a normal workup when they appear on the differential.

 

[pathstudent]

It seems like in medical school we learn about electron microscopy in relation to kidney disease, mostly. Do other branches of pathology routinely diagnose based on EM findings, and how difficult is it to do? I would imagine that it takes some time to examine an electron micrograph and it takes some specialization in it as well.

I was reading Differential Diagnosis in Surgical Pathology, and it listed EM findings for a lot of diseases, but it wasn't clear to me that this was actually part of a normal workup when they appear on the differential.

Before IHC, EM was routinely used in tumor pathology. IHC put EM out of business.

 

[Enkidu]

Before IHC, EM was routinely used in tumor pathology. IHC put EM out of business.

So are you saying that EM is literally only used in medical kidney?

 

[Ruination]

So are you saying that EM is literally only used in medical kidney?

Kidney and maybe the occasional muscle biopsy. Most kidney biopsies aren't "native" and likely won't require EM for diagnosis. Very rarely you will see someone resort to EM out of desperation to classify a tumor.

 

[Enkidu]

Kidney and maybe the occasional muscle biopsy. Most kidney biopsies aren't "native" and likely won't require EM for diagnosis. Very rarely you will see someone resort to EM out of desperation to classify a tumor.

Well, this completely answers my question. I hate to prolong the thread any further, but what do you mean by "native" biopsies, and what does that have to do with requiring EM?

 

[caffeinegirl]

Most kidney biopsies are done of transplanted kidneys to assess for rejection.
However, there still are a number of kidney biopsies that are done to assess for medical renal diseases (ie. "native" biopsies). In this case, the diagnosis is made based on a combination of H&E, immunofluorescence, and electron microscopy findings. The EM findings are for looking at the glomerular basement membrane, presence and location of immunotactant deposits, presence of tubuloreticular inclusions, etc.

Electron microscopy is also used for muscle biopsies (looking for such things as inclusion bodies), and sometimes to classify difficult tumors. In addition, electron microscopy analysis (such as microprobe analysis) is used to determine elemental properties of renal/ureteral calculi, and also can be used to determine presence of asbestos bodies or other particles in lung tissue. EM can also be used to look for viral particles (although not used so much anymore given immunoperoxidase antibodies), and also to determine the structure of cilia (looking for ciliary dysmotility).

In academics with an electron microscopy unit, you can have an EM rotation where you learn about the processing and visualizing of EM. Also, in academics EM is more often used on classifying of tumors (at least it was where I trained), whereas in private practice I've only seen it used in medical kidney and muscle biopsies.

As for the boards, there are a few questions on EM (maybe 3?), and they're usually esoteric and not from routine practice.

 

[pathstudent]

In academics with an electron microscopy unit, you can have an EM rotation where you learn about the processing and visualizing of EM. Also, in academics EM is more often used on classifying of tumors (at least it was where I trained), whereas in private practice I've only seen it used in medical kidney and muscle biopsies.

Your program seems like it is way behind the times because using EM to classify tumors is what private practice used to do before IHC. The only role for EM in tumor classification in this day and age is the 1 in 1000 cases where immunos don't pan out.

 

[caffeinegirl]

I guess I should clarify:

in my program, EM was used to classify those rare tumors where immuno's don't pan out (which happened more often than not being that MGH is a large tertiary care hospital that gets lots of weird cases). Also this is in keeping with the history of the program, which has a strength in diagnostic electron microscopy (see Dickersin's text), and which encourages taking samples for EM of rare tumors.

in my current private practice setting, EM isn't used for those weird cases...we send them out for consultation to the academic centers which have many more of the fancy IHC antibodies available.

Even though EM is "arcane" in the current world of immunos, I always did find it fascinating to look at the cellular aspects of the cell at super high magnification...from the tonofibrils to the basement membrane to the pinocytotic vesicles. It is really cool in an academic way, and also helps you remember just what antigens your keratin and myosin antibodies are targeting with IHC 🙂

 

[lurksalot]

Ditto caffeinegirl, and really, only 1/1000 immunos don't pan out?!? I need to work where you're training, pathstudent.

 

[2121115]

EM is the future and you would do well to learn all you can about it. In the future, more and more tumors will be diagnosed with EM and EM findings will have prognostic and therapeutic influence. You will be left behind without it. Do not underestimate it's importance.



Sincerely,

The 1980's

 

[pathstudent]

Ditto caffeinegirl, and really, only 1/1000 immunos don't pan out?!? I need to work where you're training, pathstudent.

Yeah. In my experience, greater than 99% of tumors can be diagnosed via H&E and immunos. At least to the point where it clinically relevant. When I was a resident we routinely saved tissue for possible E.M. but only sent it when all else failed and when all else failed usually E.M. didn't help. Now the residents only save tissue for E.M on clinically malignant soft tissue tumors and some pediatric tumors and then it is only sent less than 5% of the time. They never even save it on prostate cancers, endometrial cancers, ovarian cancers, lymphomas, testicular tumors, esophageal cancers, skin cancers, stomach cancers, colon cancers, pancreatic tumors, lung cancers, brain tumors, thyroid tumors, head and neck tumors. H&E and immunos are enough to relevantly classify those almost every single time.

If MGH is running E.M. on tumors where it doesn't clinically add to the diagnosis then they are just burning other people's money because it is fun to look at and that shouldn't be the definition of "academic pathology".

 

[caffeinegirl]

Woah pathstudent...
I never said at MGH they ran EM for "fun." There's a difference when collecting tissue for EM and running the EM...we ran the EM in tumors where the immunos didn't pan out and where it makes a clinical difference...not just to see the pretty pictures. The protocol for residents getting tissue for EM is similar to your institution.

 

[KCShaw]

I.e., electron microscopy is currently a niche market. It has value in that relatively low volume market, not a lot outside of that, and only marginal overlap with the more common methodologies. It's not overly difficult technically speaking, as long as you have the hardware and submit the sample adequately, but can be modestly difficult to interpret unless you have some regular experience with it, simply to know what you're looking at.

I gather it was considered the end-all-be-all of morphologic diagnosis at one time until people remembered the first impression at 2-4x is often the most accurate, then gave way to immunohistochemistry, which is trying to be superceded by various molecular-genetic diagnostics. For the most part they're all still subservient supplements to clinical information + gross + H&E. Not to say these extras don't have value, there's just been nothing to "replace" the basics.