Pathology report: what each part means and how to read it after the fact

When a surgeon or radiologist removes tissue, it goes to a pathology laboratory. There it is described, cut, processed into paraffin blocks, sliced thin, stained and put on glass slides. A pathologist reads the slides and writes the report. The report goes to the doctor who ordered the test, and a copy goes into the patient's chart.

There are several kinds. A surgical pathology report covers biopsies and resected tissue. A cytology report covers cells, such as a Pap test or a fine-needle aspirate. A frozen section report is a rapid read during surgery, followed later by the full report on the same tissue. An autopsy report is a separate document with its own structure. This guide covers surgical pathology, which is what most injury, malpractice and disability files contain.

The report is also the start of a paper trail that outlasts it. The blocks and slides are kept after the report is signed, and they can be read again by another pathologist years later. That makes pathology unusual among medical records: the underlying evidence can be re-examined, and a second reading can change the diagnosis.

Terms you will meet

Accession number

The laboratory's unique number for the specimen, such as S26-01234. It ties the report to the blocks and slides.

Gross description

What the specimen looked like to the naked eye: size, weight, color, what was cut and which pieces went into which cassette.

Microscopic description

What the pathologist saw on the slides. Some laboratories leave it out for routine cases.

Final diagnosis

The pathologist's conclusion, usually at the top of the report. It is the line most clinicians read first, and sometimes the only one.

Synoptic report

A structured checklist of required cancer findings, such as tumor type, grade, size, margins and lymph nodes, based on the CAP protocol for that cancer.

Frozen section

A rapid diagnosis on frozen tissue during surgery, used to guide the operation. The final diagnosis comes later on processed tissue.

Addendum

Information added after the report was signed, such as special stain or molecular test results, without changing the original diagnosis.

Amended report

A signed report that has been changed, for example a corrected diagnosis or a fixed patient identification error.

Margins

The cut edges of the removed tissue. A positive margin means disease reaches the edge and may have been left behind.

The mock report below is a breast core biopsy. The layout is invented; every laboratory system differs. The 9 marked parts appear in some form in almost all surgical pathology reports.

What a cancer resection report adds

When a tumor is removed, the report adds a synoptic section built from the CAP cancer protocol for that site. The CAP publishes these cancer protocol templates for each cancer type, and each lists the required data elements. Structure helps completeness: in a comparison of 200 synoptic and 200 narrative reports for colon and lung cancer resections, Schaad and colleagues found 98% of required elements in the synoptic reports and 77% in the narrative ones.

A single specimen can produce several documents over weeks. Most record productions include only the latest one, which can hide the fact that the diagnosis changed.

Amendments are not rare events at the scale of a busy laboratory. The CAP Q-Probes study by Nakhleh and Zarbo covered 1,667,547 surgical pathology cases in 359 laboratories and found 3,147 amended reports, a mean of 1.9 per 1,000 cases. What changed in them:

For a reviewer, 2 things follow. First, request every version of the report and the distribution record for each, because the amendment is only useful if the treating doctor received it. Second, check whether the chart after the amendment reflects the new diagnosis. A clinic note that repeats the original diagnosis weeks after it was corrected is a finding.

Clinical laboratories in the US operate under the Clinical Laboratory Improvement Amendments (CLIA), in 42 CFR part 493. Unlike most parts of the chart, the lab report has a federal rule that lists what it must contain and what the lab must do when it is wrong.

On corrections, the rule is specific:

"Maintain duplicates of the original report, as well as the corrected report."

Retention and the second reading

The retention periods decide what evidence exists. A report and its slides must be kept at least 10 years, so a biopsy from 5 years ago can usually be re-read. Blocks need only be kept 2 years, so a request for new stains or molecular tests on older tissue may fail. Many laboratories keep materials longer under their own policies or state law, so ask. A preservation letter should name the accession numbers and ask for slides and blocks by name. Our guide to getting medical records for a lawsuit covers the request process.

Pathology is interpretation. For most specimens, agreement is high. For some categories of disease, trained pathologists reading the same slide reach different answers often enough that second opinions are routine at referral centers.

At Mayo Clinic, Swapp and colleagues looked at 71,811 cases referred from outside institutions between 2005 and 2010, all re-read before treatment. They found 457 major disagreements, 0.6% of cases. The areas with the most were gastrointestinal and liver pathology (17.5% of the disagreements), lymph nodes (16.0%), bone and soft tissue (10.3%) and genitourinary pathology (9.4%). In a subset of 166 of those cases, follow-up was available for 140: treatment was affected in 126 (90.0%) and prognosis in 129 (92.1%), and the revised diagnosis agreed with the follow-up findings in 84.9% of cases.

Breast biopsies show how much depends on the category. In the study by Elmore and colleagues, 115 pathologists in 8 states interpreted 240 breast biopsy slides, 6,900 readings in all, and their diagnoses were compared with a reference diagnosis set by an expert panel:

For a reviewer, these numbers set expectations. A changed diagnosis on second review is not proof of negligence; disagreement exists among careful pathologists, and it is highest in borderline categories. Whether a first reading fell below the standard of care is a question for a pathology expert with the slides in hand. What the record can show is when each reading happened, what it said, and what was done with it.

Where pathology problems show up in a file

What each flag means

1. A recommendation that did not travel. The pathology comment advised excision. The radiology concordance note called the result benign and returned the patient to screening. Whether that was a reasonable call is for experts; the record shows the 2 documents disagree.
2. A changed diagnosis. The amended report replaced atypia with DCIS. The original must still exist, and the laboratory's correction log should show why.
3. A narrow distribution list. The corrected report went to 1 recipient. The question is whether anyone who saw it acted on it.
4. A clinic note that repeats the old diagnosis. 6 weeks after the correction, the primary care note still says benign. Ask whether the amended report was in that clinic's chart when the note was written.

The same file on 1 timeline

The method works for a single biopsy or a cancer case with a dozen specimens. Start by listing every specimen, because each has its own accession number and its own chain of reports.

1. List every specimen. 1 row per accession number: date collected, site, procedure, laboratory.
2. Collect every version. Frozen section, preliminary, final, addenda and amended reports, with sign-out times and distribution lists.
3. Match specimen to procedure. Side, site and date should agree with the procedure or operative note.
4. Read the comment. Pull out every recommendation: more tissue, more stains, correlation with imaging, consultation.
5. Trace each recommendation forward. Find the next note that mentions it. Log the ones that no note mentions.
6. Compare versions. For each amended report, note what changed and the date. Check the chart after that date for the new diagnosis.
7. Compare with imaging. Size, site and suspicion level should be consistent. A benign result on a suspicious lesion needs a documented concordance decision.
8. Preserve the evidence. Send a preservation letter for slides and blocks by accession number before retention periods run.

Is this the diagnosis the treating doctors acted on?

PATHOLOGY REPORT ABSTRACT
Patient: [ID]    Laboratory: [NAME, ADDRESS]
Accession number: [ ]    Version: [frozen / preliminary / final / addendum / amended]
Collected: [DATE]    Received: [DATE]    Signed: [DATE, TIME]
Signed by: [NAME, CREDENTIAL]
Page(s): [ ]

Specimen: [site, side, procedure]
Matches procedure note? [YES / NO, page]
Clinical history given: [ ]

Final diagnosis: [verbatim]
Comment / recommendations: [verbatim]
Synoptic findings (if cancer): type [ ] grade [ ] size [ ]
  margins [ ] nodes [ examined / positive ] LVI [ ] stage [ ]
Addenda / biomarkers: [results, dates]

If amended: what changed [ ], reason given [ ]
Distribution: [recipients]
Next clinical note that uses this result: [DATE, page]
Questions: [ ]

RE: [PATIENT NAME], DOB [ ], accession numbers [LIST]

Please produce, for each accession number listed:
1. Every version of the report: frozen section, preliminary,
   final, addenda, and amended or corrected reports, including
   the original of any corrected report.
2. Sign-out date and time for each version.
3. The distribution record for each version (recipients,
   method, date sent).
4. Records of any critical value or significant finding
   communication (who was called, by whom, when).
5. Outside consultation or second opinion reports.
6. The requisition and clinical history received with the
   specimen.

Please also preserve, and confirm in writing that you are
preserving, all glass slides, paraffin blocks and any remaining
tissue for these accession numbers, and do not discard them
under routine retention schedules.

Before you rely on a pathology report

• Every accession number is listedIncluding frozen sections and outside consults.
• All versions are in handThe original of any amended report as well as the correction.
• Distribution records are requestedThey show who received each version.
• Specimen matches the procedure noteSide, site and date.
• Every recommendation is traced forwardTo the note that acted on it, or to a gap.
• Slides and blocks are preservedBy accession number, in writing.

0 of 6 checked.

Surgical specimens start in the operating room, so read the report next to the operative note; our guide to altered medical records covers what to do when versions conflict, and the EHR audit trail guide covers how to show who opened a result and when. For the wider method, see how to write a medical chronology.

Pathology reports are short, but a cancer file can hold dozens of them across several laboratories and years, mixed in with thousands of pages of clinic notes and imaging. Chapter 7's method is mostly matching: specimen to procedure, version to version, recommendation to the next note. That is where software helps.

AI medical record review can list every pathology report by accession number and date, put each version on the same timeline as the imaging and clinic notes, and cite every line to its page. That turns a scattered file into an AI medical chronology where a corrected diagnosis, or a recommendation with no follow-up, is visible at a glance, and where each finding can be checked against the source in seconds.

The limits are real. A large language model (LLM) summarizing a file can collapse 2 versions of a report into 1 diagnosis, or report the final diagnosis line and drop the comment; hallucination and omission are both risks, which is why every line needs its page. Scanned reports need OCR that keeps synoptic tables intact. Whether a first reading was wrong is a pathologist's judgment made on the slides, which no record review can replace. The human in the loop reads the flagged pages.

Medrecords AI works this way on the files you upload. It builds a cited medical chronology with each pathology version, imaging study and clinic note on 1 timeline, each linked to its page. It tracks how findings like tumor size change across dates with condition progression tracking. It answers questions like "every pathology recommendation and the next note that mentions it" through cited Q&A and record search, and it flags records that should exist but were not produced, such as an original behind an amended report. Flags are signals, not verdicts. It does not read slides or images for diagnosis, and it does not retrieve records from providers. It runs under SOC 2 and HIPAA with a signed business associate agreement (see security and HIPAA). Self-Service bills 10 cents a deduplicated page, down to 5 cents at volume, duplicates free; Enterprise On-Prem is an annual license.

If you are comparing tools, our guide to medical record review covers what to ask any medical chronology software vendor. See how pathology review fits medical malpractice, expert witness, life and disability underwriting and legal nurse consultant work.

What are the parts of a pathology report?

Patient and specimen identifiers, dates, the specimen description, clinical history, the final diagnosis, a comment, the gross description, the microscopic description and the pathologist's signature. Cancer resections add a synoptic section with type, grade, size, margins, lymph nodes and stage.

What is the difference between an addendum and an amended pathology report?

An addendum adds information after sign-out, such as special stain or biomarker results, without changing the diagnosis. An amended or corrected report changes something that was wrong, which can include the diagnosis.

How common are amended pathology reports?

In a CAP study of 1,667,547 cases in 359 laboratories, the mean was 1.9 amended reports per 1,000 cases, and 38.7% of amendments changed the final diagnosis.

How long are pathology slides and reports kept?

Federal rules require laboratories to keep pathology reports and histopathology slides at least 10 years, cytology slides 5 years and paraffin blocks 2 years. Many laboratories keep them longer.

Can a pathology diagnosis be re-examined years later?

Often yes, because the slides are kept. Another pathologist can re-read them. New stains or molecular tests need the blocks, which may have been discarded after 2 years.

How often do pathologists disagree?

It depends on the tissue. At Mayo Clinic, 0.6% of 71,811 referred cases had a major disagreement. On breast biopsies, agreement with an expert reference was 96% for invasive cancer and 48% for atypia.

Why do pathology reports matter in delayed diagnosis cases?

The report often holds the first record of the disease, or a recommendation for more tissue. The case usually turns on when that report was issued, whether it changed, who received it and what was done next.

Can AI read pathology reports accurately?

It can extract diagnoses, synoptic findings and recommendations from the text, and list every version with its page cite. It does not replace a pathologist's reading of the slides, and every extracted line should point back to its source page.

Is it HIPAA compliant to upload pathology reports to an AI tool?

It can be, with a vendor that signs a business associate agreement, holds a SOC 2 report and commits in writing not to train on your data. A consumer chatbot without a BAA is a different matter.

Regulations were read in full text from primary sources in September 2026, and quoted text is verbatim. Study figures come from the published abstracts. The mock report, patient, clinicians, dates and pages in chapters 2, 6 and 7 are hypothetical. Product facts come from this site's product pages. Nothing here is legal or medical advice.

• 42 CFR 493.1291: test report content, critical values, delayed results, corrected reports and patient access.
• 42 CFR 493.1273: histopathology, signatures and terminology.
• 42 CFR 493.1105: retention of reports, slides, blocks and tissue.
• College of American Pathologists, cancer protocol templates.
• Nakhleh RE, Zarbo RJ, "Amended reports in surgical pathology and implications for diagnostic error detection and avoidance: a College of American Pathologists Q-probes study of 1,667,547 accessioned cases in 359 laboratories", Archives of Pathology and Laboratory Medicine 122(4):303-309, 1998.
• Swapp RE, Aubry MC, Salomão DR, et al., "Outside case review of surgical pathology for referred patients: the impact on patient care", Archives of Pathology and Laboratory Medicine 137(2):233-240, 2013.
• Elmore JG, Longton GM, Carney PA, et al., "Diagnostic concordance among pathologists interpreting breast biopsy specimens", JAMA 313(11):1122-1132, 2015.
• Schaad N, Berezowska S, Perren A, Hewer E, "Impact of template-based synoptic reporting on completeness of surgical pathology reports", Virchows Archiv 484(1):31-36, 2024.