Undifferentiated Dyspnoea — Written Clinical Reasoning

SAQ 1 — Integrated Diagnostic Approach to the Complex Dyspnoeic Patient (20 marks, 30 minutes)

Prompt: Outline your integrated diagnostic approach to Mr Chen's presentation, addressing: (a) the time-course framework and the prioritised differential; (b) the interpretation of the NT-proBNP, the troponin and the D-dimer, and how each biomarker is interpreted in the clinical context; (c) the relative contributions of COPD, heart failure and infection, and how you discriminate them; (d) the immediate management in the first 60 minutes, including the oxygen strategy and the indications for non-invasive ventilation; (e) the management of the new atrial fibrillation; and (f) the common exam trap in this patient. [1]

Model Answer

(a) Time-course framework and prioritised differential (3 marks): [1]

Mr Chen's dyspnoea is subacute (two days), which generates a differential of pneumonia, COPD exacerbation, decompensated heart failure, pleural effusion, and metabolic acidosis. His background of COPD and ischaemic heart disease, the productive purulent cough and the fever point to an infective COPD exacerbation as the primary driver, but the orthopnoea, the raised JVP, the fine basal crackles and the raised NT-proBNP indicate a coexisting cardiac component. The new atrial fibrillation is almost certainly secondary to the acute illness (the infection, the hypoxia, the fever, the sympathetic surge). The prioritised differential is therefore: first, an infective COPD exacerbation; second, coexisting decompensated heart failure; third, new-onset atrial fibrillation secondary to the acute illness; fourth, a type 2 myocardial injury suggested by the raised troponin. I explicitly do not anchor on a single diagnosis, because the complex multimorbid patient often has overlapping and mutually exacerbating pathologies. [1]

(b) Interpretation of the biomarkers (4 marks): [1]

The NT-proBNP of 1800 pg/mL is above the PRIDE rule-in cutpoint of 900 pg/mL for a patient over 50, which supports a cardiac contribution to his dyspnoea [2]. However, I interpret it in context: he has COPD, atrial fibrillation and an acute infection, all of which can elevate the peptide, and the value does not by itself quantify how much of his dyspnoea is cardiac versus respiratory. The value confirms a cardiac component and shifts my management towards a cautious diuresis alongside the respiratory therapy [1].

The troponin of 45 ng/L (repeat 42 ng/L) is elevated but static over 3 hours. This is most consistent with a type 2 myocardial injury — demand ischaemia from the tachyarrhythmia, the hypoxia and the increased wall stress — rather than a type 1 myocardial infarction. The static trend, the absence of chest pain and the absence of ischaemic ECG changes support the type 2 interpretation. I do not activate the catheter lab; I manage the underlying illness and monitor the trend. [1]

The D-dimer of 1200 micrograms per litre is non-specific and unhelpful in this context — it is elevated by the infection, the COPD and the atrial fibrillation, and it cannot exclude pulmonary embolism. I do not interpret the positive D-dimer as confirming PE; if the clinical picture (pleuritic chest pain, leg swelling, immobility) raises the pre-test probability, I proceed to CTPA, but I do not treat PE on the basis of the D-dimer. [1]

(c) Discriminating COPD, heart failure and infection (4 marks): [1]

I do not try to separate these at the first assessment — I treat them in parallel and reassess. The COPD component is addressed with controlled oxygen (target 88 to 92 per cent given the COPD and the risk of hypercapnia), nebulised salbutamol and ipratropium, and systemic corticosteroids (prednisone 50 mg daily for five days). The infective component is addressed with antibiotics guided by the local guideline (amoxicillin-clavulanate or doxycycline for a COPD exacerbation with increased sputum purulence and volume). The cardiac component is addressed with a cautious intravenous diuresis (furosemide 40 mg), titrated to the clinical response. The reassessment at 30 to 60 minutes tells me which is responding — if the wheeze and the saturations improve with the bronchodilator, the COPD component was dominant; if the JVP and the crackles reduce with the diuresis, the cardiac component was significant. The integrated, parallel approach is safer than forcing a single diagnosis. [1]

(d) Immediate management and oxygen strategy (4 marks): [1]

The immediate management begins with the ABCDE assessment and the controlled oxygen strategy. Because he has COPD and is at risk of hypercapnia, I target a saturation of 88 to 92 per cent (using a Venturi 24 to 28 per cent mask rather than high-flow oxygen) and I check an arterial blood gas within 30 to 60 minutes [6]. If the gas shows a pH below 7.35 with a PaCO2 above 6, he is in acute type 2 respiratory failure and I start non-invasive ventilation (bilevel positive airway pressure) — the indication is the acidotic hypercapnic respiratory failure, not the absolute CO2. In parallel, I give nebulised bronchodilators, systemic corticosteroids, antibiotics, sit him upright, and start the cautious diuresis. I send blood cultures before the antibiotics, I take a sputum culture, and I repeat the troponin and the renal function. The common error is to under-oxygenate the hypoxaemic COPD patient for fear of CO2 retention; I give the controlled oxygen and use the gas to detect and manage the hypercapnia.

(e) Management of the new atrial fibrillation (3 marks): [1]

The AF is almost certainly secondary to the acute illness and will usually settle as the underlying cause is treated. My priority is to treat the cause (the infection, the hypoxia, the electrolytes) rather than the AF itself. I would not give a beta-blocker (relative contraindication in the COPD exacerbation) or a calcium-channel blocker that might worsen the hypotension. The options for rate control that are safer in COPD are digoxin or a rate-limiting calcium-channel blocker (diltiazem) if the rate is poorly controlled. If the AF is itself causing haemodynamic compromise that does not respond to the resuscitation, the options are amiodarone or synchronised DC cardioversion. I anticoagulate him once the acute phase is resolving, weighing the CHA2DS2-VASc (which will be high) against the HAS-BLED. The teaching point is the order: treat the cause of the AF, and the AF usually settles. [1]

(f) The common exam trap (2 marks): [1]

The trap is the anchoring error — stopping at the diagnosis of COPD exacerbation and missing the coexisting heart failure, the new atrial fibrillation, the type 2 troponin rise, and the possible PE that the D-dimer cannot exclude. The registrar who treats only with bronchodilators and corticosteroids has missed the cardiac component and the arrhythmia, and the patient who does not settle at 30 to 60 minutes is telling the registrar that the working diagnosis is incomplete. The second trap is interpreting the elevated troponin as an acute coronary syndrome and activating the catheter lab without weighing the type 2 alternative — the static trend, the absence of chest pain and the AF context make the type 2 injury the more accurate interpretation. [1]

SAQ 2 — The Pulmonary Embolism Diagnostic Algorithm and the Role of Clinical Prediction Rules in Dyspnoea (10 marks)

Prompt: A junior doctor asks you to explain: (a) the two-level Wells score and how it stratifies the pre-test probability of pulmonary embolism; (b) how the Wells score integrates with the D-dimer and CT pulmonary angiography in the validated diagnostic algorithm; and (c) the synthesis of the evidence (with the Christopher study) that you would give a colleague considering the overuse of CTPA in low-risk patients. [1]

Model Answer

(a) The two-level Wells score (3 marks): [1]

The two-level Wells score is a clinical prediction rule that stratifies the pre-test probability of pulmonary embolism into PE unlikely (score 4 or less) and PE likely (score more than 4) [3]. The components and weights are: clinical signs of DVT (leg swelling and pain with palpation of the deep veins) 3; PE is the most likely diagnosis or equally likely 3; heart rate greater than 100 1.5; immobilisation for at least 3 days or surgery in the previous 4 weeks 1.5; previous objectively diagnosed DVT or PE 1.5; haemoptysis 1; and malignancy (on treatment, treated in the last 6 months, or palliative) 1. The single highest-weighted subjective item — PE is the most likely diagnosis — is the one that most often determines which limb the patient enters, and the registrar must weigh it carefully and document the reasoning.

(b) Integration with D-dimer and CTPA (4 marks): [1]

The validated algorithm, established by the Christopher study (van Belle and colleagues, 2006, a prospective cohort of 3306 patients), combines the Wells score with the D-dimer and CTPA [4]. The patient classified as PE unlikely (score 4 or less) undergoes a D-dimer; if the D-dimer is negative, PE is safely excluded without imaging (the 3-month incidence of symptomatic VTE in this group was 0.5 per cent). The patient classified as PE likely (score more than 4), or the patient classified as PE unlikely with a positive D-dimer, proceeds to CTPA. The age-adjusted D-dimer (cutoff equals age multiplied by 10 in micrograms per litre for patients over 50) increases the proportion of patients in whom PE can be excluded without imaging, particularly in the older patient. The key teaching points are: first, the D-dimer is a rule-out test, not a rule-in test — a positive D-dimer does not diagnose PE; second, ordering a D-dimer in a patient with a high pre-test probability is wasteful because the test will be positive and the patient proceeds to CTPA regardless; third, the D-dimer is universally elevated in the postoperative, the malignant, the septic and the pregnant patient, and cannot exclude PE in these contexts.

(c) Synthesis of the evidence for the colleague (3 marks): [1]

The synthesis I would give the colleague is that the validated algorithm — Wells score, then D-dimer if PE unlikely, then CTPA if PE likely or if the D-dimer is positive — safely excludes PE in a defined subset of patients and reduces the overuse of CTPA. The Christopher study demonstrated that the PE-unlikely plus negative-D-dimer strategy carries a 3-month VTE risk of 0.5 per cent, which is below the accepted threshold for clinical safety [4]. The overuse of CTPA — ordering it in every breathless patient without applying the Wells score — exposes patients to unnecessary radiation, contrast nephropathy and the overdiagnosis of clinically insignificant subsegmental PE. The corollary is that the disciplined application of the Wells score and the D-dimer is the single most effective way to reduce inappropriate CTPA, and it is the standard of care. The trap is the reverse — the junior doctor who does not apply the Wells score and who orders a D-dimer in a high-probability patient, then a CTPA on the basis of the inevitable positive D-dimer, has used the tests in the wrong order and has not benefited the patient.