Collective Peer review of a Case Report- Peripheral Pulmonary Embolism

What follows below is an original version of the WhatsApp conversational review that began with an invitation statement from the teacher and each participant critically appraised and reviewed the case report. As the discussion progressed, dialogue between the moderator and students centred around study, and a review report was summarized (the moderator is indicated by the title “teacher” and/or initials (R.B.) and student reviewers are indicated by their initials).

MS: By reading the paper the most prominent point I find is for pleuritic chest pain D-dimer test should also be ordered to rule out peripheral PE which is often being missed. The recorded sensitivity, specificity, and accuracy of D-dimer test as regards the final diagnosis by CT pulmonary angiography were 90%, 37.5%, and 76.6% respectively  [1] 

RB: Good point. The first thing that came to mind was to ask the authors to prepare a list (as often adviced in BMJ) of what was known before this case report and what does this case report add to the literature. 

RB: So even here there's a 10% chance to miss a pulmonary thromboembolism (PTE)?

AKG: 90% chance to find an infarct in the body as D dimer is not specific to PTE but to an infarct in body only. So it helped to confirm PTE when there was evidence in CECT. So I guess the chance to miss a P by D Dimer may be very high and the best case will be when other tests also used with D Dimer to confirm PTE. 

RB: Share a reference for your first line (particularly infarct anywhere in the body)

AKG: The title of the paper proves my point. Diagnostic accuracy of D-dimer assay in suspected pulmonary embolism patients. So 90% is when we already suspect a PTE, otherwise, it is not. As it's a test telling if there is any inappropriate blood clot in the body. 

AKG: In the case history section written is, "he was investigated for all possible etiologies of pleuritic chest pain." Is testing for all possible etiologies the way in clinical practice?

AKG: "This case report is to highlight the fact that physicians should consider pulmonary embolism in the differential diagnosis of isolated pleuritic chest pain."

- this statement also needs evidence to support what is the incidence of Peripheral pulmonary embolism. It will help to know why its a missed entity, is it rare or is it commonly missed?

RB: Good point

VP: Should this be a differential diagnosis in any isolated pleuritic chest pain or only after ruling out other potential common causes? Because if we consider it just with isolated pleuritic chest pain, then there is a potential for unnecessary tests even in MSK pain.

RB: I think we need the authors to highlight that it was the persistence of the pleuritic pain leading them to investigate further for pulmonary embolism is a unique selling point of their case report. Let's also search for who else has reported that this persistence a symptom that should increase our suspicion for PTE. Also, can we explain why the pulmonary infarction pleural effusion pain should be persistent and this shouldn't be? What were the initial differentials made by the treating team?

MS: Initially the patient was treated for musculoskeletal pain in a different hospital for which he was given NSAIDS. He did not recover. 

Following that, in the current hospital, the authors mentioned they have ruled out all possible etiologies of pleuritic chest pain but haven't stated what conditions they have ruled out?? 

Then on 2nd day in CECT, they have seen the focal area of collapse-consolidation.  But did not state what made them go for a D-dimer test.  Just stated D-dimer test is elevated so suspected PE.

AKG: When they have seen a focal area in CECT then D dimer test helped them establish it that it may be due to infarct. 

VP: While looking for peripheral versus central pulmonary embolism, I found the following: "Central pulmonary embolism was diagnosed when thrombi were seen in the trunk or in the main pulmonary arteries and peripheral pulmonary embolism when segmental or subsegmental arteries were affected" [2] 

VP: The focal area of collapse-consolidation involving the posterior segment of Rt lower lobe (possible infarct) made them go for the d-dimer. 

AKG: So no need of D dimer in central PE because we can directly see the thrombi?

VP: No, I meant why it is peripheral PE. D-dimer would just mean somewhere in the body there is a thrombus formation. When a patient has respiratory symptoms with elevated d-dimer that increases the suspicion of the source of that thrombus.

MS: So the reason for going for D-dimer is because of the suspension from the CECT.

AKG: The posterior segment of Rt lower lobe was affected while confirming it may be because of thrombus but was the thrombi seen in any major artery to say its central PE?

"CT pulmonary angiography which revealed features suggestive of pulmonary embolism in the terminal pulmonary artery branch" [from manuscript]

VP: Yes, central thrombi is found in the trunk or in the main pulmonary artery, unlike peripheral which is found in segmental or subsegmental arteries.

MS: "In the sub-segmental and segmental groups, the patients presented most frequently with incidental CT findings (63.6 and 53.7%, respectively), dyspnea (18.2 and 23.9%, respectively) and leg pain or swelling (13.6 and 20.9%, respectively)" [3]

Here chest pain is seen only in one patient rather dyspnea is seen in many patients for segmental and subsegmental groups. So,  I suppose suspecting a pulmonary embolism for a isolated chest pain is not the right way to go as incidence is so low or must not be included in differentials. 

VP: This could be a differential but I feel they should mention it for a persistent isolated pleuritic chest pain. 

AKG: Why not acute rather than persistent?

VP: If we consider it in the differentials at day 1 with pleuric chest pain then we would have to go with CT and D-dimer for patients who really don't have pulmonary embolism leading to overdiagnosis and potential overtreatment. Because of the tests itself has false positives. 

MS: Exactly so keeping this as a differential is no point at day 1. But again gradually if the pain does not subside they go for a CT which suspects a PE on findings. 

AKG: If other tests come negative on day one, should we treat his MSK or we should do CECT & D Dimer? Otherwise, after a week, he may be admitted to my hospital and the tests need to be repeated.

VP: Yes this time you have a good opportunity to make high suspicion of some other causes. But on day 1, he could be evaluated given that the patient is well-explored for thromboembolic disorders or any other risk factors that could lead to PE. So, keeping it the differential is good if detailed history can be explored to increase the height of suspicion.

MS: No u will be having past history. 

AKG: There was no past history mentioned for this patient in the case report. 

MS: He has a past history of taking medications so they proceeded further for CT on day 2 itself. 

VP: which medications? 

MS: NSAIDS

AKG: The First physician reviewed him for all the possible etiologies and the authors tested him for PTE and suggesting to take PTE in the early differentials. I need to know the incidence of whether it is a rare diagnosis or commonly missed diagnosis? Can both of you summaries your reason why you disagree this must not be included in differentials for day 1? 

VP: They need to answer then how a physician can differentiate PE and MSK pain during isolated pleuritic chest pain?

AKG: What are the sign/ symptom out from above all would not be present in other etiologies?

MS: Did not get you. 

AKG: I mean about the study shared from 2010 above and the current case report where authors are saying that if chest pain is present then test for PTE, if not then there are other clinical features and also author has mentioned which may vary.

I want to say if chest pain (pricking) and leg swelling is one thing which may be absent in other diseases with similar presentation and so this may be an important sign to notice for PTE, otherwise it may not be PTE but if don't want to miss then other signs and tests are also mentioned to go ahead for confirmation.

MS: My point is sample in the above study which I stated, it shows out of all the cases of peripheral PTE, chest pain is seen only in one case. So,  it is very rare to see chest pain in peripheral PTE. As it is rare, keeping it in the differentials and adding a test (which is not again accurate and become the burden to a patient) to isolated chest pain is not necessary.

AKG: But what in case of isolated pleuritic chest pain (emphasizing on isolation). They should also characterize the pain? As they mentioned, "this case report is to highlight the fact that physicians should consider pulmonary embolism in the differential diagnosis of isolated pleuritic chest pain."

VP: They are very non-specific symptoms. But the pain in pulmonary embolism is usually very abrupt and sudden-onset. Musculoskeletal pain can also commonly present with isolated pleuritic chest pain (emphasizing on persistence). It is far more common cause for coming to the emergency department. 

MS: This is good but isn't it better to consider it in the differentials on the 1st day?

VP: Yes, they may consider after ruling out all the possible etiologies. 

AKG: Yes, pricking pain, and abrupt. So needs to notice for the possibility of PTE. This patient had an admission only for chest pain and good that authors thought of PTE diagnosis also. At day 1m when the chest pain started, he was treated for MSK. 

Before confirming that one needs to check the incidence and impact of PTE. But I also disagree because it is not the best way for all the possible tests to find any diagnosis. One should follow clinically what is present in a patient and if it doesn't improve then go for higher precision approach and closely follow the history. 

RB: Yes differentials have a prioritization too. One week back MSK topped the list of ddx and one-week later PTE climbed up on the list of priorities due to the evolving information gathered over the week. 

VP: But I really learned that while other doctors just made it MSK pain when they excluded common causes, they further went ahead to evaluate it. But they need to justify in the discussion why despite MSK being a common cause for isolated pleuritic chest pain, peripheral PE should also be considered?

MS: This case has also been found out based on the suspicion of CECT  mostly the peripheral PE presented with dyspnea (often they are asymptomatic). I mean to say on routine clinical practice for isolated pleuritic chest pain, a routine is enough as only very few cases present with chest pain as the presenting symptom, if not treated go with CECT which helps the suspicion of PE (which is stronger). 

VP: But in their discussion of the manuscript, they should have discussed more that: 1. it was the persistence of the pleuritic pain leading them to investigate further for pulmonary embolism. 2.  who else has reported that the persistence of this symptom should increase our suspicion for PE. 3. why the pulmonary infarction pleural effusion pain should be persistent and this shouldn't be? 4. What were the initial differentials made by the treating team? Rather they have discussed cardiac biomarkers, ECG findings, echo findings on PE affecting RV etc. What about ECG of this patient? We see S1Q3T3 in ecg of pulmonary embolism (although authors discussed that it is found only in 20% of cases). Our queries should be asked to the authors asking for justification in the discussion. Currently, their discussion is not centred around this particular case. 

AKG: If this patient had come to them directly at the beginning (before passing 1 week and worsening condition), could they have gone further to evaluate it with such precision and accuracy?

VP: Not sure, but maybe they would?

AKG: How? On what evidence/history? I mean when the incidence of PTE is not known to me yet and whether its a rare entity or not. I am asking because unable to think if I were at the doctor's position 1-week before who even tested for all etiologies, except doing so much to find a PTE. 

RB: We need some details on what the previous doctors did?

VP: Also beside the history of smoking habits, fever, cough, haemoptysis, dyspnoea, syncope, palpitations or weight loss, did the patient have any other risk factors that would cause PE. The risk factors could be around the Virchow's triad of endothelial injury, hypercoagulability, stasis (prolonged immobility for example).They are now considering hemophilic conditions

MS: They would have just gone to CECT which have led them to suspect PE. As they did now.

VP: Exactly, some tertiary care centre has all the facilities and even they would go with that to exclude possible rare etiologies. A possible reason for overdiagnosis!

AKG: So the previous 1 week was also perfect treatment if that was a primary or secondary level clinic?

MS: Maybe, based on the age and other factors they have gone for MSK pain. 

AKG: Because the pain was not severe and the patient was not admitted for during that one week, but rather was went back with NSAIDs only for a week. He got a severe pain and admitted to a tertiary care where the tests were done to the fullest to reach accuracy and precision of what possible in tertiary centres.

RB: Why? Did the pain increase from before or was it just persistence of the pain that led to the CT. We have to ask the authors to tell us. 

MS: If we see the perspective of the healthcare setting then yes but we always define these terms from the patients' outcome so I say no. 

AKG: Even though they have diagnosed the case, the outcomes are indeterminate unless this report details at the end that he lived happily after. 

MS: Yes u are right. After the patient is completely fine then only we say our diagnosis and treatment is precise and accurate.

VP: Here, come the concept of accuracy and precision. They were initially precise, but inaccurate.  In the second time, the doctors were precise in diagnosis and treatment but may or may not be accurate as that would depend on the patient outcome.

RB: Were they imprecise or precisely focused on musculoskeletal chest pain?

AKG: Initially they were accurate and precise based on history and resources. But later it was increased as the history changed, and the patient admitted with progressive PTE and severe pain etc. Moreover, resources were also abundant. Are we saying opposite? 

VP: Accuracy and precision not always go hand to hand. Precision is that they have adequately tried to follow the protocols they rule out the possible diagnosis. Inaccurate was that they could not relieve the patient symptoms.

AKG: Also if they do more for the patient then they may find the exact location and can see the clot accurately. Isn't there even more precision possible and they are also imprecise yet that way?

VP: How precision can occur with imprecision? This time this doctors are precise in diagnosis, but not yet accurate as we don't know yet how would be the patient outcome and would thus patient be left without treatment or not.

AKG: Can not. I was thinking that what are the limits of precision diagnosis? Every time when a higher resource modality of management will be available, a higher precision can be possible and previous results will become less precise. My answer to this is - the limits of precision is that when we have the correct diagnosis and can proceed for treatment and the limits of accuracy is when we give a treatment where the patient can live happily after. 

RB: But again do we often have the necessary precise approach to treatment that can produce accurate outcomes?

VP: But the approach to precision may sometime lead to overdiagnosis.

MS: Often in search of precision there might be overdiagnosis. 

AKG: As a doctor, if I ask patients any chief complaints or problems, and the patients tell alright. Should this be enough for me? 

MS: As a doctor, it is enough but it doesn't mean they are happy mentally. 

AKG: If he is unhappy mentally he will complain to the doctor. 

RB: So eventually what was the patient's outcome? Did the authors mention that?

AKG: It is only explaining the patient problems, diagnosis, treatment, but outcomes are not mentioned.

RB: That's a serious issue with the case report then?

VP: The diagnosis and treatment are mentioned. 

AKG: Did the treatment work?

VP: Since the purpose of this case was not to highlight the treatment effect so it's fine to be on follow-up.

RB: Yet for every case report, patient outcomes would be an important purpose that needs to be accounted for. 

VP: But how can they show the outcome of pulmonary embolism in this patient until they take it for a period?

MS: Generally a paper should mainly emphasize one point and here this paper emphasizes the inclusion of PE as a differential diagnosis in pleuritic chest pain. So. We can slide if they have not mentioned follow up. 

RB: A case report is about the patient as a whole and should not have a focused point. 

MS: But this paper is not emphasising more on the patient rather than a focused point then can it be taken any further?

RB: I would put it as a limitation of the authors that they have focused too much on the disease rather than the patient. 

VP: By the patient outcome to what extent they can add? Suppose if the patient is on prophylactic, how can we wait for a second event to occur?

RB: Just this much about how many days now since he is on anticoagulants. What is the average dose he is requiring? What happened to his prior symptoms? More about patient-related outcomes rather than disease-related outcomes. 

VP: Yes, they must add this. They haven't discussed anything much about these.

AKG: In medicine, a case report is a detailed report of the symptoms, signs, diagnosis, treatment, and follow-up of an individual patient (Wikipedia). Is it fine to delay publishing after adding follow up data because its a case report?

RB: I am sure the authors should have the patients follow up by now. I guess the authors have not mentioned the date of their first diagnosis and treatment?

MS: Case report more focuses on the patient-centred whereas research papers more about disease centred? I mean case-control, cohort studies RCTs. 

RB: Unless they are patient-centred research papers. So the question is even in these studies would it be possible to collect patient-centred data. I answered this question in my first JECP paper here [4]. 

VP: "The present EBM literature neglects a lot of events it doesn’t believe to be statistically significant and perhaps here is an area that needs to be improved on" [4]- found it interesting and left me confused in cases where we ignore the insignificant outcomes. 

RB: Yes but there is a difference between statistical and clinical significance. I hope we never left any clinically significant outcome (that were albeit statistically insignificant)?

MS: What about the paper where the chest pain is reported in only one case of peripheral PTE? [2]

RB: One out of how many?

MS: Out of 89. Both central and peripheral PE was a total of 334 patients.  

RB: Was there a difference in the type of the chest pain in central vs peripheral?

MS: No sir except in the table there is nowhere mentioned about chest pain in the paper. 

RB: Have our case report authors detailed the nature of the chest pain in their patient (other than just saying pleuritic)? If they stick to pleuritic we have to think it was peripheral as pleural involvement in pulmonary embolism is generally due to peripheral pulmonary infarcts. 

Let's write up the entire exercise above as a paper on medical education to demonstrate the learning outcomes in 3 UG medical students (preparing for their scheduled semesters as well as USMLE, NEET etc) on how collective peer reviewing around a journal case report can provide them with learning points good enough to cover standard exam questions 

SUMMARY OF CONVERSATIONAL DISCUSSION: 

• The inclusion of multiple tests (CECT and D-Dimer) in the differentials may cause overdiagnosis and overtreatment so to prevent it, the importance of Persistent chest pain needs to be highlighted and to consider it for including in the differential needs to be mentioned more clearly not on the day 1 of the chest pain.
• What was known about pulmonary embolism before this case study and what are the new learning points, this case study adds to the literature?
• The discussion must highlight and concentrate on the patient and the relation of chest pain with the pulmonary embolism.
• The authors have missed to mention what are the potential and possible diagnoses and etiologies they have ruled out. 
• One of the limitations of the study is that the authors that they have focused too much on the disease rather than the patient.
• The authors to highlight that it was the persistence of the pleuritic pain leading them to investigate further for pulmonary embolism is a unique selling point of their case report. Let's also search for who else has reported that this persistence a symptom that should increase our suspicion for PTE. Also can we explain why the pulmonary infarction pleural effusion pain should be persistent and this shouldn't be? What were the initial differentials made by the treating team?
• What is the incidence of PTE and any more data on it being missed? Even chest pain (or isolated chest pain) is very rare in PTE, what about that? [2]
• Need clarification as to whether peripheral PTE is a differential diagnosis in any isolated pleuritic chest pain or would be only after ruling out other potential common causes? Because if we consider it just with isolated pleuritic chest pain, then there is a potential of unnecessary tests even in MSK pain.
• The authors need to detail the nature of the chest pain in their patient (other than just saying pleuritic)? If they stick to pleuritic we have to think it was peripheral as pleural involvement in pulmonary embolism is generally due to peripheral pulmonary infarcts.
• Are there any possibilities for how a physician can differentiate PE and MSK pain during isolated pleuritic chest pain?
• Importantly, authors should detail outcome and follow up of this patient.

FINAL RECOMMENDED REVISIONS FOR THE AUTHORS WHICH TRANSPIRED FROM CONVERSATIONAL REVIEW BY THE TEAM:

Our recommendation for the case report that you sent to us for review is that it is fair and require further revisions as pasted below:

• What was known about pulmonary embolism before this case study and what are the new learning points, this case study adds to the literature?
• One of the potential new learning points that the authors have brought out is perhaps the importance of Persistent pleuritic chest pain that aroused suspicion in their particular case.  
• The author's discussion needs to elaborate further on the relation of chest pain with this patient's pulmonary embolism.
• The authors may also elaborate on what were the other potential and possible diagnoses and etiologies they considered for the patient's pleural effusion before ruling in pulmonary embolism.
• The authors may like to highlight that it was the persistence of the pleuritic pain that leads them to investigate further for pulmonary embolism and this is perhaps a unique point in their case report that could make it a valuable addition to the literature.
• They may also search for who else has reported that this persistence of chest pain as a symptom may increase our suspicion for PTE. 
• Also can they explain why if at all the pulmonary infarction pleural effusion pain should be persistent (as in their case) and pleural pains due to other etiologies shouldn't be? Could it be related to chemical mediators released post infarction? More here in a similar case report http://breathe.ersjournals.com/content/13/2/e46
• One of the important limitations of the authors is that they have focused too much on the disease rather than the patient. It would be nice to see some more data that also address the patient's own concerns about his her disease (more ideas here https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0142070).
• Last but not the least we are interested to know the current follow up of this patient. 

1. https://www.sciencedirect.com/science/article/pii/S0422763813003166

2. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4821092/

3. https://www.karger.com/Article/FullText/277929
4. https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1365-2753.2007.00837.x