Structure of Report
We have seen what to avoid in reporting. Before I proceed to few more tips in reporting, let us look
in depth about the structure of radiological report.
The radiological report can be divided into 4 parts:
1. Heading/Type of examination
2. Technique/Protocol
3. Description/Main report/Findings
4. Conclusion/Summary/Comment
1. Heading:
The heading should clearly reflect what radiological examination is performed.
Examples:
1. "Right shoulder AP, axial and Y-views" would be better heading than "Shoulder
Radiographs".
2. Rather than writing "CT whole body", we should be more specific, as "CT chest, abdomen
and pelvis".
3. "CTPA" would be better heading than "CT thorax", when CT chest is performed for
pulmonary embolism.
4. "MRI of the brain and IAM" makes more sense than "MRI brain" when the referral has
come from an ENT surgeon to exclude acoustic schwannoma.
Do not use unfamiliar or ambiguous short forms in headings, such as "TVS pelvis"; for few this
may be "transvesical sonography" and for others, "transvaginal sonography".
2. Technique:
Most of us do not bother to mention the technique in the report.
Plain radiograph:
In plain radiograph a good heading, such as, "Right knee lateral and weight bearing AP views"
gives sufficient information about the technique also.
If the chest radiograph is rotated, or underexposed or appears to be expiration, mention it in the
report.
If the film is taken on a trolley in an unconscious patient in expiration, we cannot pick-up
pneumothorax; hence I would put my heading as "Chest AP supine trolley film", which clearly
desribes the technique and its lmitations.
Ultrasound:
It is a good practice to mention the frequency and type of the probe used in the examination, unless
your establishment has got all the types of probes available in the market.
If the procedure involves intimate contact, such as endovaginal scan, mention "verbal/written
consent was taken" in the report.
CT and MRI:
Most of the departments follow fixed protocols for CT scanners. Mentioning the protocol would be
enough most of the time. Example: "CTPA protocol" is sufficient when the CT chest is done for
pulmonary embolism. I personally feel it is worth mentioning the MR sequences in the report.
Interventions:
It is very important to document the technique (i.e. procedure) in the report. This includes drug and
dosage (10ml lignocaine 1% as local anaesthetic), type of the catheter (10G pig-tail catheter),
description of the proceudure, and after-care notes.
3. Description (Main body of the report):
"KISS" approach:
Keep It Short and Simple (KISS). One of my educational tutors used to say, "long report is wrong
report". The description should be clear, concise, accurate and relevant to the clinical question with
a high confidence level.
There are few things which we should try to avoid, about which I have discussed in my previous
blogs.
I will further discuss few more tips about writing radiological description (main body of the report)
in my next post and hopefully conclude this series with tips on writing summary.
What to avoid in radiology reports?
Avoid long descriptions for normal findings.
Example: "The kidneys appear normal" is enough, there is no need to say "Both kidneys appear
normal in size and echotexture. The corticomedullary differentiation is maintained. The margins are
smooth. There is no evidence of calculus or hydronephrosis".
Avoid unnecessary words.
Example: "The Kidneys appear normal" rather than "Both kidneys appear normal", or "The kidneys
appear normal bilaterally"
Avoid repetitions.
Example: "Comparison is made with the CT dated 12/08/07. The low attenuation lesion in segment
7 of the liver is seen again, and has not changed in size since the previous study." Instead, just say,
"Compared with the CT dated 12/08/07, there is no significant interval change."
Avoid abbreviations.
Example: "The features are suggestive of MS". You might be thinking of Multiple Sclerosis when
you dictated the report, but the cardiologist who is reading the report starts worrying about Mitral
Stenosis.
Avoid giving measurements of normal things and incidental findings.
Example: "The right kidney measures 12.5 cm in bipolar length and the left 13.0 cm. The spleen
measures 10.5 cm. The liver span is 13.5 cm. The right ovary measures 38 x 18 mm and the left
ovary measures 26 x 20 mm. There is simple cyst in the upper pole of the right kidney measuring 25
mm in diameter. Another simple cyst is seen in the lower pole of the right kidney measuring 15 x 12
mm. Uterus is anteverted and measures 8.0 x 4.5 x 5.2 cm." Are we tailors to measure each and
everything?
Avoid mentioning in which window you reviewed the images.
Example: "On review of the images in bone window, no metastases are seen". As a radiologist, we
are supposed to view the images in various window settings before we issue the reports. "No bone
metastases are seen" implies that I did look all the images in bone windows.
Avoid "lymphadenopathy".
Lymphadenopathy means pathology in the lymph nodes, and even 4 mm nodes can be metastatic. It
is better to use the term "enlarged lymph nodes based on CT/MR size criteria" than
"lymphadenopathy".
Try not to begin too many sentences with, "there is a ....".
Example: "There is a large spiculated mass lesion in the right lower lobe of the lung. There is
obstruction of the right lower lobe bronchus. There is complete collapse of the right lower lobe".
All these can be put together, "A large spiculated mass is shown obstructing the right lower lobe
bronchus, leading to right lower lobe collapse."
Avoid words with different meaning to different people.
The words are same, but the meaning is different. Swastik is auspicious for Hindus and Jains, but it
is symbol of Hitler in rest of the world.
Example: "Collection" means abscess to surgeons; hence "use fluid attenuation lesion" when you
are not suspecting abscess.
Avoid double negatives.
Using double negatives is not good English.
Example: "The findings are suggestive of pneumonic consolidation, but a bronchoalevolar
carcinoma cannot be excluded." Say, "Although the features are suggestive of pneumonia,
bronchoalevolar carcinoma should also be considered in the differential."
Avoid highlighting the mistakes of previous report.
Example: Don't mention "The spiculated lesion in the left lower lobe of the lung was clearly seen in
the previous CT as 8 mm density, but was not mentioned". It might be your turn tomorrow.
Downplay the misses, something like "On retrospective look, the left lobe spiculated lesion is seen
as 8 mm density"
Avoid repetitions in conclusion (impression or comment).
Do not repeat what you have already described in 2 pages in your "conclusion" also.
Few more tips in writing the report
• Start with the main abnormality and add supporting radiological abnormalities
centered around it.
Example: Instead of writing,
"Significant hydronephrosis is seen in the left kidney. The right kidney is normal. The left
ureter is also dilated...."
and somewhere in next paragraph,
"There is a large soft tissue mass in the retroperitoneum encasing the aorta and left
ureter",
it is better to say,
"A large retroperitoneal soft tissue attenuation mass is shown encasing the aorta and left
ureter, leading to moderate left hydronephrosis."
• Include important negative findings in the report, depending on the clinical
information and imaging findings.
Example: If there is a speculated mass in the left lung, mention negative findings, such as
“there are no enlarged lymph nodes in the mediastinum. No other lung lesion, no focal
liver or adrenal lesion to suggest metastasis. No lytic bone lesions”.
• Mention, but do not highlight, the incidental findings:
Example: If you think the abnormality is clinically not significant (for example, multiple gall
stones in a patient with known breast cancer showing multiple lung and brain deposits),
mention it as "note is made of multiple gall stones", rather than highlighting it as one of the
abnormalities related to breast cancer. Don’t mention these unrelated clinically not relevant
abnormalities in the “conclusion”.
• If you think the radiological finding is equivocal or indeterminant, do not
hesitate to mention it in the report.
Example: If you are unable to come to conclusion whether the appearance in the caecum
is secondary to a mass or secondary to faecal loading, do not hesitate to mention it. At the
same time, suggest the next step (colonoscopy or CT colonogram) to clarify the doubt.
• Use standard abbreviations (please read this in conjunction with “avoid
abbreviations” in one of my previous posts).
Example: Do not write,
"It was difficult to compare computed tomography images with the magnetic resonance
imaging dated First of February 2008".
It is easily understood if it is written as
"it is difficult to compare CT images with MR images dated 1/2/2008."
• Quantitative assessment is essential.
The report may read "there is large mass." But how large is large? A 2 cm lesion appears
large in orbit, where as a 5 cm lesion appears small in the thigh. It takes a bit of more time,
but it is worth mentioning approximate measurements, at least in the longest axis (please
read this in conjunction with one of my previous posts, where I mentioned there is no need
to measure normal structures).
• Use proper terms when subjectively quantifying.
When quantitative assessment is not possible, for example, as in pleural effusion, use
subjective grading. Use proper terms in quantifying the abnormality using grading.
Example: There is no “mild pleural effusion”, there is “small pleural effusion”; there is no
“severe ascities”, there is “large ascites”.
• Use "clinical correlation might be useful" or "further clinical correlation is
suggested", only if clinical correlation is going to make the difference to
radiological finding.
Use "....is suggested, if clinically indicated" sparingly.
Example: For a left hilar mass seen on CT, you cannot say "bronchoscopy is suggested, if
clinically indicated" as bronchoscopy is necessary in this case, whether clinically indicated
or not.
• Whenever possible compare the images with previous imaging, even if the
modality is different.
Make a note of the date of the previous imaging which you are comparing. It is worth
comparing previous 2 images, especially in the follow-up of cancer patients, even if
different modalities are used.
How to conclude the report?
What do you call it?
Call it “Summary”, “Impression”, “Conclusion”, “Opinion”, or “Comment” (I call it “Summary”
in this discussion), but avoid "Diagnosis", because diagnosis is more specific, and is a
combination of clinical, radiological, biochemical, microbiological, and histopathological
factors.
Summary is not mandatory.
Not all reports should have a summary. If the main report itself is short and precise, there
is no need to repeat the same thing again in summary.
Summary needs to be short and precise.
It is a medicolegal requirement on the part of the clinician to read the whole report.
Despite, most of the clinicians rely on summary only. Hence, summary should be short and
precise, should answer the clinical questions, and/or shows an unexpected clinically
relevant finding. There is no need to repeat the whole episode of the main report in the
summary too (just like some of the TV serials' "recaps", which are as long as 10 minutes!).
Some of the orthopaedicians and neurosurgeons do not bother to read the report which is
all together a different issue.
Summary addresses the clinical question, contains clinical relevant conclusion, and
ignores irrelevant positive incidental findings.
Example: if the request asks for pulmonary mass, the summary should specifically state
whether mass present or not.
Example: CT chest, abdomen and pelvis with a history of colonic cancer has following
findings (from top to bottom): 2 thyroid nodules in a retrosternal thyroid, abberent right
subclavian artery, pulmonary emboli, multiple lung metastasis, multiple liver metastasis,
multiple gall stones, ascites, degenerative lumbosacral spine, and haemangioma in L3
vertebral body. I would put my “summary” as follows:
“In a known case of colonic carcinoma, there are multiple liver and lung metastases with
generalised ascites, which is also likely to be malignant. Bilateral pulmonary emboli in the
lobar branches need urgent clinical attention.”
I would not put other findings in my “conclusion”; they are there in the “main report”.
More importantly, I would pick up the phone and ring the relevant team to tell about the
pulmonary embolism which needs urgent anticoagulation, and I would also document the
communication in my report.
Commit, even if your opinion may turn wrong.
Everyone make mistakes. Radiological report is a small part of clinical investigations. A
wise clinician knows how much weight to attach to a radiology report, when he/she is
making the complete diagnosis.
Summary is for clinicians.
Summarise in words which clinicians understand. The main report is always better
understood by fellow radiologists, and is very useful for the radiologist who is comparing
the previous images.
Example: If a CT shows focal fluid attenuation area with thick wall enhancement, then
summarise it as “abscess” rather than “fluid attenuation cystic lesion.
Do not make too many recommendations, especially the non-radiological ones.
If another radiological investigation is capable of answering the radiological dilemma of the
current investigation, and can give more definitive answer, then do not hesitate to
recommend it. It not only helps the clinician to take next relevant step in establishing the
diagnosis, but also safeguards you medicolegally.
Example: Illdefined low attenuation area in the left centrum semiovale may be a white
matter infarct, but in a 30 year old lady, do not hesitate to recommend MRI to look for
demyelination.
If another radiological investigation might help to clarify the doubt of current investigation,
but clinically not going to change the management, then one can suggest it in summary.
Example: A few rounded low attenuation lesions in the liver are most likely to represent
“simple cysts”, but suggesting ultrasound would be useful, if the patient is known to have
a malignancy.
When recommending a non-radiolgical investigation, be extra cautious or your ignorance
may exposed.
Use “clinical correlation is needed” sparingly.
The terms like "further clinical correlation is suggested" and "CT is suggested, if clinically
indicated" are too defensive way of practicing radiology, and should avaoided as much as
possible. These kind of sentences should be reserved if you think clinical correlation
makes a difference in reducing differential diagnosis. Whenever you feel like writing
“further clinical correlation is suggested”, be specific about “what further clinical correlation
is needed”
Example: In reporting a HRCT of an interstitial lung disease, “further correlation with
occupational history is suggested” sounds better than “further clinical correlation is
suggested”.
Add a personal touch.
Adding personal touch is a great way of building a good rapport with the referring clinician.
It also makes you sound you are an experienced radiologist.
Examples: "I think this is unlikely to be any clinical significance". “In my opinion, this is
tuberculosis, unless proven otherwise”. “I would suggest MRI of the brain”. “I explained the
risks and benefits to the gentleman and took a written consent for ultrasound guided
biopsy”.
Keep your report open-ended.
When a radiological investigation is performed and reported, do not presume the patient
no longer belongs to you, and leave everything back into the hands of the referring
clinician. Hence, do not shut the door of the report, keep it open. Open-ended summaries
not only increase the rapport with the referring clinician, but also can, to some extent, can
safeguard against medico-legal issues.
Example: “I am more than happy to review the findings if you can provide the previous
CT imaging".