Guide to the radiology report

Review Article Australasian Radiology (2002) 46, 366–369 Guide to the radiology report Lloyd J Ridley Department of Radiology, Concord Hospital, Sydney, New South Wales, Australia SUMMARY The radiology report is an important, but often neglected, aspect of radiology. Suggestions on style and content are discussed. These include: (i) concise reporting; (ii) correct English; (iii) present tense for current reporting; (iv) reading the request with care; (v) recognizing the recipients; (vi) being aware of unintended readers; (vii) stating the most important findings; (viii) where possible, quantifying findings; (ix) reporting some incidental findings; (x) avoiding a tale of woe; (xi) summarizing a longer report; (xii) using inductive logic; (xiii) using non-radiological input (such as probability) if pertinent; (xiv) discussing minor findings in the body of the report; (xv) taking care in suggesting further investigations; (xvi) ensuring the conclusion and recommendations are clear; (xvii) considering alternate diagnoses; and (xviii) considering if a dictated report is sufficient. Key words: radiology; report. GUIDE TO THE RADIOLOGY REPORT The radiology report is the predominant method of communication from the radiologist to the referring doctor. There is only a limited amount written in the literature on the subject, although recent reports do highlight the medicolegal issues that might arise from a poorly constructed report. A US review revealed communication errors are the fourth most frequent primary allegation against radiologists. 1 The need for strategies of risk management in the department and the desire to assist registrars evolve an effective style led to a review of the literature and development of the following suggestions/ guidelines. The subjective nature of what constitutes a good report means that there are alternate views in the literature. 2,3 Individuals will develop their own approach to a report based on personality and experience. The major considerations in writing the report include style, target, introduction, content, conclusion, and follow up. full.5 This survey also noted that while 62% of clinicians said the type of examination should determine the appropriate length, 25% thought that reports longer than one page were too long. A significant consideration is how to report a normal examination. Dr Friedman argues the case for simply writing ‘normal’.6 Variations on this include ‘nonrmal except for…’, ‘normal for age’ (although this begs the question what is normal ageing and what is age-related pathology), or ‘no change from…’. A survey by McLoughlin et al.7 noted that the majority of clinicians preferred more detail, including radiological findings, than proposed by Friedman, particularly for more complex situations (specifically an ultrasound). Correct English It might be helpful to refer to an appropriate text or a dictionary. One such dictionary gives a brief dissertation on ways to improve one’s writing. 8 It discusses the basic units of written communication, being words, sentences and paragraphs. In choosing words, aim to be direct and specific. Consider, for example, the report ‘the lungs are clear’. Compare this with the less specific ‘No evidence of acute cardiopulmonary disease’. In the latter, the perception remains that there might be chronic disease.6 Style Concise reporting4 The information that the radiologist wishes to convey needs to be written in as few words as possible, yet maintain fluency and meaning. The importance of a concise report is emphasized by a survey showing only 38% of referring clinicians read the report in LJ Ridley. Correspondence: Dr Lloyd J Ridley, Department of Radiology, Concord Hospital, Hospital Road, Concord, New South Wales 2139, Australia. Email: xray@email.cs.nsw.gov.au Submitted 18 June 2001; resubmitted 8 March 2002; accepted 28 June 2002. RADIOLOGY REPORT Take advantage of connotations. Consider the different levels of certainty implied by the following different conclusions: ‘There is definite…’, ‘There is possible…’, ‘There is suspected…’, ‘There is equivocal…’. 9 Hall3 gives an extensive list of terminology that is worth reviewing. Remove unnecessary words. In ‘both the lung fields are clear’, the addition of the word ‘fields’ adds little to the meaning, and ‘both’ is clearly implied by using the pleural ‘lungs’. Avoid abbreviations. Non-technical abbreviations in a report give an air of informality and imply you were in a hurry. Technical abbreviations risk the lack of understanding. Consider, for example, the difficulties interpreting radiology requests that use abbreviations. Similarly, jargon, especially radiological jargon, is best avoided. Revak9 uses the example of the Frostberg ‘reverse three’ sign. The referrer might not be aware of this entity, but would understand the significance of a mass at the sphincter of Oddi. Changes in tense or number occur, particularly in more complex sentences. An example is ‘the remainder of the lungs demonstrate no abnormality’. The subject of the sentence is ‘remainder’, which is singular. The singular form of the verb (in this case with an ‘s’) is therefore required. In view of the possibility of transcription errors, some attention should be given to words or phrases that might be mistyped. As an example, ‘asymmetric’ can easily be typed ‘a symmetric’.10 Sentence patterns can also add value. Simple sentences (subject and predicate ‘the lungs are clear’) maintain clarity. Compound sentences (two clauses joined by a conjunction such as ‘and’) allow expression of related ideas of equal importance (‘the lungs are clear and the mediastinurn is normal’). Complex sentences relate ideas of different importance (‘there are three lung metastases, all in the right lower lobe’). Paragraphs allow the organization of ideas. The sentences within each paragraph should be related. In a radiology report, the sentences would typically be grouped anatomically or pathologically, such as ‘A CT of the chest was performed…’, ‘In the lungs there are three lesions’, ‘The largest is located…’, ‘In the mediastinum, there are lymph nodes…’. 367 Recognizing the recipients11 The description and recommendations that a general practitioner requires will be different to that of a person with specialist interest in the field. Albertyn 11 relates an incident where a patient was referred for a CT scan ‘400 miles’ because an ultrasound in his hometown had been reported as showing ‘enhancement’ behind the gallbladder. Being aware of unintended readers Patients, their families and lawyers quite often get access to reports. Consider implications such as whether the patient might think they have a disease (e.g. the elderly with an ‘atrophic brain’). Introduction A report needs to contain basic demographic data. This includes patient identification, referring physician, radiologist, nature of the examination, date (and time) of the examination, date of the report and radiology practice name. 12 For an examination that might be performed in different ways, such as a chest X-ray (e.g. PA, mobile), it is pertinent to state the method used. This is particularly important with cross sectional modalities where issues such as contrast agent administration and different planes and sequences influence the ability to detect different pathologies. 11 This information is relevant both for the referrer and any radiologist who might subsequently encounter the patient.11 Content State the most important findings first11 They are less likely to be overlooked. Relevant negatives and incidental findings follow. Where possible, quantify the findings11 Measurements such as length of lesions can be quoted. Alternatively, use graduated modifiers such as mild, moderate or severe. Using present tense for current reporting9 The past tense for what was done at the time of the examination. As well as being factually correct, this helps provide consistency in the report. Report some incidental findings11 Incidental findings range from those unlikely to be clinically relevant (such as spina bifida occulta) and therefore do not require reporting. This is particularly the case if they have been reported on previous studies. At the other end of the spectrum are normal variants that are the primary cause of disease (such as variants in paranasal sinus anatomy). Excessive discussion of incidental findings runs the risk of misinterpretation. At best, they can cause patient anxiety, at worst, they can result in invasive further investigation. Target Reading the request with care Specifically address any questions raised in the request. This is particularly important when the examination is negative, and where it emphasizes that the radiologist did look for the appropriate findings. It also helps to know the history given to the radiologist. This might explain why a certain conclusion was reached. It might also give information that is relevant to, but not given on, a later request. Avoide a tale of woe11 If the study fails, keep it brief. If the study is inadequate, take action to rectify rather than com-plaining to the referrer. 368 LJ RIDLEY limited to ‘there is interstitial and alveolar oedema consistent with left ventricular failure’. However, the clinical context does not necessarily depend on level of diagnostic confidence. A nodular opacity on a preoperative chest X-ray might be an artefact or carcinoma. It can only be defined at the most basic level. However, it is of high clinical significance. A final point on clinical context should be made. Terms such as ‘if clinically indicated’ should not be overly used, as noted in Berlin’s review of litigation related to ‘vague’ radiology reports. 4 It is difficult to see how clinical correlation can impact on the investigation of an unexpected lung mass, for example. 13 Conclusion The conclusion is the focus for much discussion in the literature, both for an appropriate name and position. A 6 survey reported 32% of clinicians preferred the summary at the beginning, while only 29% expressed the opinion that they preferred it at the end. I prefer the logical progression from imaging findings to an ultimate impression. This is less of an issue provided the report is less than one page long and the conclusion is highlighted. 6 Summarizing a longer report9 The conclusion is what it says, not a repetition of the findings. It has been said that radiologists are paid for using both their eyes and their brains.4 Use of non-radiological input, such as probability, might be pertinent This should be explicit in the analysis so the referrer is able to determine what diagnosis is achievable from the image. 14 Using inductive logic 9 State observations first and develop through to a conclusion. A helpful concept is the hierarchy of terms. 14 This refers to the fact that there are different levels of interpretation and diagnosis. The basic description is that an image is a pattern of light and dark. Anatomical structures comprise the second level of interpretation. It should be possible to locate a finding within an anatomical area. It follows, of course, that accepted anatomical nomenclature is used. Pathological description follows. This might be purely descriptive such as ‘enlarged left atrium’. Alternatively, it might be more pathologically specific such as infection, fracture or neoplasm. ‘Airspace opacification’ (or one of its many synonyms) is at the descriptive level. If there is evidence of a cause, a more specific term might be used, such as ‘pulmonary oedema’. A final diagnosis might ultimately be possible. Developing the above example further, the pulmonary oedema might be ‘cardiogenic’. I believe this can be taken a further step. The diagnosis can be placed in clinical context. Subjective statements such as ‘satisfactory’ or ‘significant’ might be appropriate. In making such statements it is implied, however, that there is an understanding of what makes an appearance ‘satisfactory’. Consider, for example, a post-reduction radiograph of a Colles’ type fracture. A satisfactory position for a patient with a non-functional arm might not be satisfactory for an ipsilateralhanded young manual worker. In this setting, it might be preferable to discuss the anatomical findings, particularly in comparison to the previous examinations. The findings should be interpreted to the highest level possible, using the terminology appropriate to that level. In some cases, it is worthwhile discussing the thought process in reaching this (i.e. by referring to lower levels of the hierarchy). In other cases where the constellation of findings points to a specific pathology or diagnosis, it would be labourious and counterproductive to describe the findings in detail. An example might be pulmonary oedema where a description could be Discussing minor findings in the body of the report This allows the conclusion to concentrate on findings that the radiologist considers clinically important. The length of the body of the report depends on the number of findings, whereas the length of the conclusion varies with the radiologist’s ability to make sense of the findings.4 Taking care in suggesting further investigations There is support for a radiologist ‘recommend[ing] follow-up or additional studies to clarify and confirm the impression … when appropriate’.4 The radiologist needs to balance this against the potential implication that the referrer will need to perform the test simply because of concerns about the legal implications of not performing the test. In suggesting further investigations, consider whether the findings of the current test require this for further clarification, and whether such a test will improve patient management. The radiologist should be circumspect about recommending biopsies. There is concern, for example, about malignancy growing along biopsy tracts.15 The radiologist should be aware of the opinion of the relevant surgeons so as not to preclude the possibility of curative surgery. Ensure the conclusion and recommendations are clear Does the report accurately reflect what is seen, what causes the findings, and what the physician is referred to do next? Does the report convey an appropriate level of certainty? Heilman notes, disturbingly, that even this is not always enough by citing a successful case against a radiologist who didn’t suggest antibiotic treatment for the pneumonia he diagnosed. 16 Consider alternate diagnoses It has been frequently said, ‘if you don’t think of it you won’t diagnose it’.17 Consider ‘what is the best this could be, what is the worst it could be?’. RADIOLOGY REPORT 369 3. 4. 5. 6 7. 8. 9. 10. 11. 12. 13. Hall FM. Language of the radiology report. Primer for residents and wayward radiologists. AJR 2000; 175: 1239–42. Berlin L. Malpractice issues in radiology. Pitfalls of the vague radiology report. AJR 2000; 174: 1511–8. Clinger NJ, Hunter TB, Hillman BJ. Radiology reporting. Attitudes of referring physicians. Radiology 1988; 169: 825–6. Friedman PJ. Radiology reporting and structure. AJR 1983; 140: 171–2. McLoughlin RF, So CB, Gray RR, Brandt R. Radiology reports. How much descriptive detail is enough. AJR 1995; 165: 803–6. Barnhart CL, Barnhart RK. Improving Your Writing. Doubleday, Chicago 1997. Revak CS. Dictation of radiologic reports. AJR 1983; 141: 10. Martin LFW. The language of the radiology report: a response. AJR 2001; 176: 1597. Albertyn LE. Reporting: The neglected aspect of the new imaging modalities. Australas Radiol 1991; 35: 340–2. Berlin L. Malpractice issues in radiology: Radiology reports. AJR 1997; 169: 943–6. Orrison WW, Nord TE, Kinard RE, Hunt JH. The language of certainty: Proper terminology for the ending of the radiological report. AJR 1985; 145: 1093–5. Friedman PJ. Radiology reporting: The hierarchy of terms. AJR 1983; 140: 402–3. Wojtowycz M. Handbook of Interventional Radiology and Angiography, 2nd edn. Mosby, St Louis, 1995. Heilman RS. Practice comer: Advice from the Expert. Radiographics 2000; 20: 1538. Berlin L. Malpractice issues in radiology: Errors in judgement. AJR 1996; 166: 1259–61. Follow up Considering if a dictated report is sufficient Where a condition requires urgent treatment, it is desirable to give a verbal, as well as written, report. Similarly, an unexpected and important finding (classically a cancer on a ‘pre-op’ chest X-ray) might also warrant a verbal discussion (or at least a second method of communication). Any significant discrepancy between the preliminary report and the final written report should be directly communicated with the referrer.1 It is our practice to add a comment regarding pathology results when these are returned after a biopsy. This enables future radiologists to easily ascertain the diagnosis. It also allows a second line of communication with the referrer. ACKNOWLEDGEMENTS The author would like to thank members of the Concord Hospital radiology department for reviewing drafts of the document, particularly Dr Max Schieb, for comments, Dr Wynne Sum, for assistance with literature searching, and Brian Ridley for editing. 14. 15. 16. 17. REFERENCES 1. Cascade PN, Berlin L. Malpractice issues in radiology. American College of Radiology standard for communication. AJR 1999; 173: 1439–42. Naik SS, Hanbidge A, Wilson SR. Radiology reports. Examining radiologist and clinician preferences regarding style and content. AJR 2001; 176: 591–8. 2.