Starting pathology

Jump to navigation Jump to search

Author: Mikael Häggström

Using this handbook

The goal of this handbook is to make a new pathology trainee able to properly handle at least 80% of cases that are expected at an average general pathology department, including the exclusion of the most pertinent differential diagnoses thereof.

It is recommended to scroll quickly through this handbook to get an idea of its content, and then keeping it at a known location for whenever information is needed for a corresponding patient case. It is not recommended to memorize its content indiscriminately. Other doctors and even laypersons can look up diseases and conditions themselves, including pathology characteristics, without the need for a pathologist consultation, so the expertise of memorizing such readily available information is expendable. In a world where diseases and conditions can readily be looked up, a major skill that distinguishes a pathologist from any person with Internet access is mainly the ability to identify and put words to findings on microscopy or other non-written results, and to conceive most likely diagnoses based on both verbalizable and more abstract appearances. Another major skill is to be able to deal with unusual or equivocal presentations. Unusual or equivocal presentations of very common diseases and conditions are still generally more common than rare diseases, and constitutes a major workload in everyday pathology practice. However, most textbooks still give disproportionately large room for rare diseases compared to such presentations. Nevertheless, strive to master the common conditions (including the most common pitfalls) before diving into the uncommon.

What a pathologist needs to memorize

The best method for memorization is generally through repeated exposure in everyday practice, but the following things are most important for a pathologist to focus on memorizing:

  • Emergent pathology, mostly relating to intraoperative or frozen section consultations. This includes information that usually cannot be timely looked up on the Internet when needed.
  • Main pitfalls: Most common and dangerous situations where a pathologist may not recognize the need to look something up further or ask a senior colleague.
  • Patterns and signs which can be seen grossly or under the microscope. It confers the ability to translate visuals into words that can be looked up if needed.
  • Knowledge of where to find information for various situations. It includes which person or which search engine is most useful for various clinical situations. Google is generally an appropriate search engine, but sometimes more specific or comprehensive databases are necessary, such as for example ClinVar to look up the pathogenicity of specific genetic variants. The chapter will include a directory of major databases and external guidelines.
  • Proficiency in diagnosing equivocal or borderline cases where readily available sources and evidence usually deal with discrete and specific disease entities and subcategories thereof.  Thus, in the question sections, many questions will be in the format of displaying readily available facts about diseases, including their typical immunohistochemistry patterns, but with equivocal or borderline case presentations.
  • Having an idea of one’s unknowns; being aware of unfamiliar fields. For example, a pathologist generally does not need in depth knowledge about cases that are generally sent out to specialized centers (such as pediatric musculoskeletal oncology), as long as that pathologist is aware of lack of knowledge in that field.
What you need to memorize.jpg
  • Dealing with Internet denialists and their exams. With the ease of access to pathology information on the Internet through smartphones and computers, those studying to for everyday practice as a pathologists should not spend time memorizing what can essentially always be conveniently and timely be looked up in times of need. This includes most of the content of books that are sorted by titles of diseases and conditions, because if the name of a disease is already known, it can relatively quickly be looked up when needed. The topics listed above are already immense enough to cover a lifetime of learning. Nevertheless, the path to pathology certification includes one or more exams, whose questions are largely made up of people who act as if they were Internet denialists; as if they do not acknowledge the access to the Internet in everyday pathology practice. Therefore, this resource also includes a chapter on dealing with Internet denialists and their exams.

Using resources

This resource is written with the intention to teach you what to do in various situations you are expected to encounter during your first years of pathology training, at least until the point that you are usually fairly confident about what disease or condition you have at hand, because then you know what words to use to look it up in the vast literature out there. At that point, the fastest way to get more information is generally by Googling the disease or condition name, followed by pathology or pathology outlines. If you just want to see more micrographs of the disease or condition, the fastest way is generally to Google the condition and then clicking the Images tab. Specific searches for specific purposes include:

  • Adding staging in Google searches for definitions of cancer stages, for example Googling prostate cancer staging. The first search will then generally be the one from the American Society of Clinical Oncology.
  • Considering a subscription for ImmunoQuery rather than trying to memorize all immunohistochemistry markers for all conditions.

Ask a colleague at least whenever your own memory or a resource search is not enough, and there is a significant risk that you will do something irreversible that will negatively affect a patient.

Emergent pathology

Surprise frozen sections

While most frozen sections can be predicted from schedules of the operating room and thereby be looked up beforehand, this chapter deals with the most common ones that do not offer such preparation time, thus indicating memorization of how to handle them.

(Gather most common situations.)

Other frozen sections

Although these are generally given on schedules of the operating room, any pathologist may end up suddenly covering for another one, and subsequently be presented with the frozen section case without having had the time to look it up beforehand.

Non-emergent pathology

Also largely a directory of external guidelines and databases.

Non-emergent pathology questions


Which of the following situations does not have comprehensive guidelines available online?


Ask what are the main locations to look for infection in a specific autopsy case.

Provide algorithm of gram stain etc. and table of main bacteria by various locations.

Where to find information


Provide genetic variant in long format.

- Which database is the best to use to look up the pathogenicity of this variant?

 - ClinVar (correct)

 - gnomAD (population frequencies)

 - PolyPhen (likelihood of protein damaging)

 - PharmGKP (associated treatments)

- In ClinVar, what would you enter

 - A1708V (then BRCA1)

Dealing with Internet denialists and their exams

An Internet denialist probably knows about the existence of the Internet, but keeps acting and teaching as if being oblivious of it. An Internet denialist generally takes pride in for example memorizing the chromosome locations of even rare mutations, but when being called in for a frozen section of even relatively common specimens such as brain, lung or skin tissues, they may not know how to differentiate even the most common 90% of diagnoses.

An Internet denialist who has memorized something may assume that pathology trainees should memorize it as well, and entire lectures may largely consist of rants of such items. In reality, when something is encountered and looked up something enough times, it will generally get memorized, and it is generally more efficient to let time tell which situations will be common versus uncommon, rather than trying to memorize knowledge that may never be needed.

Also, while you should initially focus on learning the most common conditions, specialists and subspecialists may already have learned the common conditions, at least in their subspecialty, and they will often distract you from your pursuit by presenting rare conditions to you, because that is interesting to them, but do not spend excessive time or mental effort on such rare conditions during at least your first years.

An Internet denialist exam is basically any exam wherein the examiner does not have access to the Internet, and typically is not allowed to ask colleagues either, even for non-emergent topics that can conveniently and timely be handled by such resources. Since the Internet and teamwork are fundamental parts of everyday practice, such exams are thereby of a different dimension.

Exam studying

Since exams and everyday practice are generally very different (as per previous section), you should generally study specifically for an exam or study specifically for solving each case you encounter everyday, rather than trying to study any material with the intention of covering both purposes. After all, for a multiple choice exam like the American boards, you don't actually need to know the answer, but just have a feeling of what is the most likely answer among the choices. Also, you will become proficient at what you do: If you read textbooks you will be more proficient at reading textbooks, if you study to solve everyday cases on your table then you will become good at that, whereas the best way of becoming more proficient at multiple choice exams like the American boards is to practice qbanks with a similar multiple choice format. There are multiple ones for the American boards (PathPrimer, PathDojo, BoardVitals, ASCP Resident Q bank), and you should exhaust them and repeat at least the ones you failed the first time, before continuing with other study materials. For highest yield, don't read every explanation for every answer, but just the answers that contradicted your belief, just enough to learn why it wasn't what you initially thought. Also, you don't have to memorize every clue to the right answer, but rather get an idea of what makes the right choice most likely, because that's basically all you need to choose that right answer if it would appear in the actual exam. To check if you have remembered a question sufficiently, you may for example review the correct answers for a test until you almost immediately find them to be reasonable rather than unfamiliar, and repeat the question later if you have the time. If the Qbank shows the average percentage of test takers who got a question right, put somewhat less effort on questions with very low percentage, since you generally have less of an expectation of knowing those.

Efficient exam studying will allow you more time and effort to perform for example studying for solving your actual everyday pathology cases (as per the Using your resources section above).




Adipose tissue with crumpling artifact due to insufficient fixation.

Within an hour after removal from the body,[1] tissue samples should generally be placed in vessels with enough fixative to allow them to lie freely in the solution.[2] The standard fixation fluid is generally 10% neutral buffered formalin, which is roughly equivalent to 4% formaldehyde.[3] The ratio of tissue:formalin should be 1:5[4] to 1:10[5].[5]


The duration depends on tissue thickness, where formalin will penetrate and fix the tissue at ~1 mm/hour.[6]

When not to use formalin

The main exception to using formalin are mainly: edit

  • Intraoperative consultation. If a specimen has several parts, and you only want intraoperative consultations on some of them, hold the rest back to avoid potential delays. For all fresh specimens, communicate clearly (such as on a requisition form with the specimen) whether intraoperative consultation is requested or not.
  • A tophus or other specimen suspicious for gout versus pseudogout should be sent in alcohol or dry, since formalin will dissolve the crystals.
  • Lymph nodes (or other lymphoid aggregates) with a suspicion of lymphoma, where samples are generally put in a special solution for flow cytometry.
  • Products of conception with a need to take samples for genetic testing.
  • Cytology specimens, which are preferably sent fresh (such as in red top tubes) to be processed within a few hours. If processing may be after a few hours, put tubes on ice, or add 50% alcohol.[7]

Gross processing

  Following are general notes on selection and trimming in pathology.  

Fresh specimens

The initial processing of fresh specimens is also termed triaging.

  • For intraoperative consults including frozen sectioning, see separate article on Emergent pathology.

When triaging a specimen, the measurements are most important as these may change after fixation. Other descriptions can generally be made after the specimen is fixed.

Before cutting

  • Confirm that the patient identity on the specimen container matches the identity that will be applied to the gross description and cassettes. {{If the referral or requisition form is available, confirm the patient identity on that one as well.}}
For unclear or potentially ambiguous handwriting (here "Right" or "Left" renal stone), look at the referral or requisition form ((and the medical record if available)).
  • (Check for any discrepancy between the specimen description on the container and on the referral or requisition form, such as left versus right.)
  • Don't omit any piece in the container, such as ones hidden in wraps.
  • Generally measure in 3 dimensions, or in volume, but the greatest dimension is generally enough for specimens less than 0.4 cm.
  • Generally weigh entire organs, after having any attached tissue trimmed away if feasible.
  • (Note the color of the sample, even when unremarkable, but do not linger on deciding it.)[note 1]
  • Generally, use inking for resection margins where cancer radicality is important.  
  • (On fatty or greasy surfaces, apply vinegar to emulsify and remove the fat, dry the specimen and then ink. Otherwise, vinegar can be used either before or after inking to "dry" it.)
  • (Preferably photograph or make a drawing where slices have been taken.)[8]
  • Remove any surgical stitches from samples before microtomy.
  • (At least for larger samples, consider looking for medical imaging or biopsy reports in order to better guide the process.)[9]
  • Fix bone in formalin prior to decalcification. Use reminders so not to forget bone that is decalcifying.


  • When cutting with the longer knives, try to cut in one stroke - do not use like a saw (continuous back and forth)
  • Generally, strive to make slices perpendicular to visible interfaces of relevant tissues.
  • Generelly dissect and inspect the entire specimen, while keeping relevant parts intact enough for presentation to seniors and/or maintaining orientation.
  • Trim tissues for microscopy examination to a thickness of maximum 3-4 mm.[note 2]

Perpendicular versus en face sections

Perpendicular and en face sections

Two major types of sections in gross processing are perpendicular and en face sections:

  • Perpendicular sections allow for measurement of the distance between a lesion and the surgical margin.
  • En face means that the section is tangential to the region of interest (such as a lesion) of a specimen. It does not in itself specify whether subsequent microtomy of the slice should be performed on the peripheral or proximal surface of the slice (the peripheral surface of an en face section is closer to being the true margin, whereas the proximal surface generally displays more area and therefore generally has greater sensitivity in showing pathology, also compared to perpendicular sections).
  • A shaved section is a superficial en face slice that contains the entire surface of the segment.

Tissue selection

When sampling sections to submit for microscopic examination, whenever you sample from something that looks abnormal, generally try to also sample from the same type of tissue that looks normal.[note 3]

Biopsy wraps, bags and sponges

Items used for submitting specimens: (Biopsy) wrap, (biopsy) sponge, (tissue processing) cassette and (biopsy) bag.

Put the following types of specimens in bags:

  • Tiny specimens that need to be poured out from their containers.
  • Bloody specimens such as endometrial curettages or products of conception. For products of conception, chorionic villi may otherwise contaminate other specimens. Bloody specimens may stick to wraps, so generally avoid that situation.
  • Friable tissue such as urinary bladder biopsies.

Put the following types of specimens in bags, wraps or sponges:

  • Other tiny specimens
  • ((Any small piece of tissue where there is no leftover specimen to retake sections, since tissues occasionally get lost from cassettes, and the absence of a wrap, sponge or bag in the cassette of such cases points towards a mistake made at gross processing.))

Specimens must be fixed enough to be put on sponges.

H&E staining urgency

(Even in departments where other staff are primarily responsible for determining the urgency of H&E staining of each specimen, still double-check that it is correct if you can, such as by cassette color.) A major indication for rushing cases through H&E staining is a high risk of cancer, especially where immunohistochemistry staining will likely be performed, and the decision and types of staining will be determined by the standard H&E stain. Tissues that are generally rushed are:

  • Brain biopsy.
  • Lung biopsy.
  • Breast needle biopsy.
  • Biopsy from known tumor tissue.
  • Suspected malignant lymph nodes, including lymphoma. However, these are generally not urgent when submitted together with a tumor, except mainly for the following (which are generally urgent):

In both these cases, the cases are rushed so that immunohistochemistry can be performed if a metastasis is not readily detected on standard H&E slides, so that it is available by the time the rest of the slides are out. Immunohistochemistry in these cases can detect micrometastases that are not readily visible on H&E stain, but are evident on cytokeratin AE1/AE3. However, if the lab stains such cases regardless of whether H&E stain shows a metastasis or not, then they do not need to be rushed.

Marking cassettes

Use only hard pencil (or specially purchased histology markers), as marks made with pens, Sharpie markers, or scientific freezer-safe markers can get dissolved in tissue processing.[10]

Evaluation of tumors

  1. REDIRECT Evaluation of suspected malignancies


  Following are general notes on reporting in pathology.

  • Save your digital work frequently, and also when you leave a desk, even if you think you'll be back shortly. It doesn't matter how much time and effort you spend on something if you're just going to let it disappear.
  • Double-check your report, especially if you copy-pasted and adapted a previous report rather than using a template with blank fields or making your report from scratch.


Selection and trimming

From the stage of selection and trimming, a histopathology report should preferably include:

  • Case:
  • Patient identification and/or sample number
  • Type of tissue sample as described on container
  • Dimensions of original tissue[11]
  • Directions or other features of any inked surfaces.
  • Generally the weight of larger samples[11]
  • Dimensions of pathologic components[11]
  • Whether the entire specimen or representative sections were submitted.

Microscopic evaluation

  • Specimen chronology, often A, B, C, etc., at least where there are multiple specimens from the same case. With multiple specimens, preferably write out the chronology for all of them first, so that you don't miss reporting any of them later.
  • Specimen type and/or surgery type, such as "appendix, appendectomy", for clarification. This is not necessary at all departments. For the procedure, use the same term as the operating report whenever possible.
  • Microscopic description. This is not always necessary, but should be included if the diagnosis is uncertain. One systematic approach is to describe findings from largest to smallest ones. For example, a description of a tumor can start with the demarcation of the tumor, followed by texture, cell shapes, nucleus shapes and chromatin appearance.
  • Diagnosis or most probable diagnoses.
  • If the diagnosis does not clearly account for all conditions that were requested, suspected or asked to be ruled out by the referring clinician (such as stated on the requisition form), you need to classify the specimen as "positive for" versus "negative for" for each such condition, or give a reason for why an evaluation thereof could not be made.
  • In case of malignancy or suspected malignancy:
  • Depth or most distant invasion of malignant findings.[11] Depending on location, it may need to exclude important pathways, such as vascular, neural and/or through capsules or other layers.
  • Whether the resection is radical or not.


Factors supporting a relatively more comprehensive report, particularly in the inclusion of negated findings:

  • Lack of explanation from existing evidence. For example, an inflamed appendix that fits the medical history does not need detailed mention of harmless incidental findings.
  • Prospective review: If your report is likely to undergo double-reading by another pathologist before sign-out, it should either be more detailed, because the doctor who will do the double-reading then gets an idea of your thought process, including what you have looked for versus what may still need to be evaluated. If you know who will do the prospective review for a report of yours, you may alternatively convey your thought process by other means such as directly talking to that person.
  • Highly suspected locations, such as given from the referral.
  • Difficulty in obtaining the specimen, such as a CT-guided biopsy versus a skin shave.
  • Defensive precautions, which appears to be more common among doctors in the Unites States compared to for example Europe.[12][13]

Multiple instances of the same type of pathology (such as lung nodules) can often simply be reported as such, at least with a particular mention of the largest or the most severe example thereof.


Words, from
most likely to
least likely
  • (is)
  • positive for
suspicious for
(benign condition)
cannot be excluded
not likely
(malignant condition)
cannot be excluded
  • negative for
  • effectively ruling out

When something looks very much like a specific entity but you are not sure, preferably use "-like" (or when feasible, "-oid" such as squamoid for squamous-like cells).

When the clinical picture strongly favors a certain condition, and the pathology favors it as well, findings are generally described as "consistent with". Sometimes, "bordering on" can be described when the picture almost fits specified criteria of a specific diagnosis.

It is alright to consider the diagnosis of a pathology report to be a combination of the clinical picture and what can be seen on the specimen. For example, if the microscopy picture is uncertain, you may to a certain degree tend towards the diagnosis that best fits the clinical picture. However, mention differential diagnoses if they are still significantly possible, and would confer a different treatment or another substantially different consequence.

For both findings and diagnoses, is preferable to write "negative for..." rather than "no..." to emphasize the possibility of false negative findings.

Synoptic reports

For cancers, generally include a synoptic report, such as per College of American Pathologists (CAP) protocols at However, synoptic reports are generally not needed for tumor metastases.


Whenever possible, give numerical quantities of sizes, rather than descriptions that are subjective (such as "small" or "large") or variable (such as "apple-sized").


The information contained in the reporting sections in this resource assume that the clinician has requested the exam for the topic at hand, but should still be tailored to any particular questions or requests by the clinician. Any relevant findings beyond the issues or questions raised by the clinician should also be mentioned. The reporting templates in this resource do not cover every recurring situation, so it is often more efficient to create your own repository of report templates that you can copy-paste for various cases. When doing so, however, have marks for relevant items that are frequently changed in the template, which should be readily seen as unfinished in the report if you haven't tailored it to the case at hand (such as "...measuring _____."), so as to avoid omissions or even wrongly entered information from templates.

The most important findings can be moved to near the top of the report if feasible, but doctors performing subsequent double-reading may prefer a consistent anatomic order.


If a certain grammatical rule has a risk of making the report less clear to the reader, ignore that rule in that situation.

Restrict acronyms/abbreviations to those who are certainly well known among all doctors, such as "cm".[note 4]

Generally describe what can be seen rather than processes (such as preferring "an abundance of" rather than "proliferation of").

If using a dictation device, avoid "no", and instead use "negative for" (and "positive for" in opposite cases), since there's a risk of "no" not being transcribed and thereby creating the opposite meaning.

Skin excisions

In skin cancers, use "peripheral" or "radial" margins (whereas "lateral" margin should be reserved for the margin opposite to the medial margin).[14]



  1. . Breast pathology grossing guidelines. UCLA Health. Retrieved on 2021-09-09.
  2. Katarzyna Lundmark, Krynitz, Ismini Vassilaki, Lena Mölne, Annika Ternesten Bratel. Handläggning av hudprover – provtagningsanvisningar, utskärningsprinciper och snittning (Handling of skin samples - Instructions for sampling, cutting and incision. KVAST (Swedish Society of Pathology). Retrieved on 2019-09-09.
  3. . Paraformaldehyde, Formadehyde and Formalin. Duke University. Retrieved on 2019-12-17.
  4. . Fixation of Tissues. Approval Date: August 2016, August 2020. Review Date: August 2024|website=Royal College of Pathologists of Australia
  5. 5.0 5.1 Buesa RJ, Peshkov MV (2012). "How much formalin is enough to fix tissues? ". Ann Diagn Pathol 16 (3): 202-9. doi:10.1016/j.anndiagpath.2011.12.003. PMID 22483550. Archived from the original. . 
  6. . How to Submit Tissues for Embedding. University of Pittsburgh, Starzl Transplantation Institute. Revised 04/19/21
  7. . How to send fluid and make good cytology slides. Tufts University.
  8. Monika Roychowdhury. Grossing (histologic sampling) of breast lesions. Topic Completed: 1 August 2012. Revised: 19 September 2019
  9. . Gross Pathology Manual By The University of Chicago Department of Pathology. Updated 2-14-19 NAC.
  10. . Histopathology Services. UNC School of Medicine. Retrieved on 2021-11-15.
  11. 11.0 11.1 11.2 11.3 . An Example of a Melanoma Pathology Report. Melanoma Foundation. Retrieved on 2019-09-24.
  12. Studdert D. M.; Mello M. M.; Sage W. M.; DesRoches C. M.; Peugh J.; Zapert K.; Brennan T. A. (2005). "Defensive medicine among high-risk specialist physicians in a volatile malpractice environment ". JAMA 293 (21): 2609–2617. doi:10.1001/jama.293.21.2609. PMID 15928282. 
  13. Steurer J.; Held U.; Schmidt M.; Gigerenzer G.; Tag B.; Bachmann L. M. (2009). "Legal concerns trigger PSA testing ". Journal of Evaluation in Clinical Practice 15 (2): 390–392. doi:10.1111/j.1365-2753.2008.01024.x. PMID 19335502. 
  14. David Slater, Paul Barrett. Standards and datasets for reporting cancers - Dataset for histopathological reporting of primary cutaneous basal cell carcinoma. The Royal College of Pathologists. February 2019

Cite error: <ref> tags exist for a group named "note", but no corresponding <references group="note"/> tag was found, or a closing </ref> is missing