Skip to main content

Setting up a nuclear cardiology service: from the stress lab to the report

A myocardial perfusion service is a stress laboratory and a hot lab before it is a camera. How the workflow fits together, how to size throughput honestly, and when a dedicated cardiac system earns its place.

A cardiologist in Dhaka who wants to know whether a patient's chest pain is ischaemic has two practical routes. Send the patient for a coronary angiogram, which shows the plumbing, or send them for a myocardial perfusion scan, which shows whether the muscle is actually short of blood when it is asked to work. Hospitals add nuclear cardiology when their cardiology service has grown busy enough that pushing every equivocal case into the catheterisation laboratory has become slow, expensive and hard to justify to the patient.

The equipment decision follows the clinical one, not the other way round. A nuclear cardiology service is a stress laboratory, a hot lab, a camera and a reporting workstation, run by people who can supervise a treadmill safely and read a gated study. Buying the camera first and working out the rest afterwards is the common sequence, and it is one reason perfectly good gamma cameras in this country run thin lists on cardiac days.

What the study is asking, and what that forces on your workflow

Myocardial perfusion imaging compares tracer distribution in heart muscle at peak stress against the distribution at rest. A segment that fills normally at rest but poorly under stress is ischaemic. A segment that is poor in both is scar. Everything in the workflow follows from two facts: you need two acquisitions of the same heart in two different physiological states, and the stress tracer has to go in at peak stress, not a few minutes later when the patient has recovered.

Most departments work with Tc-99m labelled agents such as sestamibi or tetrofosmin. The 140 keV photon suits a standard gamma camera with a low energy high resolution collimator, and the six hour half life of Tc-99m makes a same day protocol possible. Thallium-201 is still used in some centres, mainly for viability work: its lower energy photons are more affected by soft tissue attenuation, which shows as breast and diaphragmatic artefact. Build a new service around Tc-99m and treat thallium as an occasional tool.

Gating matters as much as perfusion. An ECG gated acquisition yields ejection fraction, end diastolic and end systolic volumes and regional wall motion from the same data set, at no extra dose and very little extra time. A perfusion service that is not gating is discarding half of what it paid for.

The stress laboratory is the harder half

Departments underestimate the stress laboratory because the camera is the visible purchase. In practice the stress room is where the clinical risk sits, where physician time goes, and where throughput is actually limited. It also has to be a separate room from the camera, because the two activities run in parallel all morning.

Exercise stress

Exercise is preferred wherever the patient can walk, because it adds symptom, ECG and haemodynamic information to the images. The tracer is injected at peak exercise through a cannula sited before the test starts, and the patient continues exercising for a short period afterwards so the tracer distributes at the stressed flow state. That last detail is the one most often lost when a service is staffed by people trained only on static imaging. Stopping the treadmill at the moment of injection wastes the study, and the images will look normal when they should not.

Pharmacological stress

A large share of patients referred for perfusion imaging in Bangladesh cannot exercise adequately: knee and hip disease, peripheral vascular disease, deconditioning, or simply an inability to reach a diagnostic heart rate. Vasodilator stress covers those patients, with dobutamine kept for cases where a vasodilator is contraindicated. Vasodilator protocols require caffeine and methylxanthines to be withheld beforehand, which means written instructions in Bangla at booking and a verbal check on arrival, because nobody thinks of morning tea as a drug.

Choose the protocol before you size the camera

Protocol choice drives scheduling, tracer ordering and camera occupancy, so decide it during planning rather than after installation. It changes how many cardiac slots one camera can carry, and how much waiting space you need for patients sitting out an uptake delay.

ProtocolHow it runsWhere it suits
One day, rest then stressLower activity rest study first, higher activity stress study later the same dayHighest single day throughput and one patient visit. Residual counts from the rest study must be overcome by the stress activity, so total administered activity is higher.
One day, stress then restLower activity stress study first, rest study added later only if the stress study is abnormalLets you cancel the rest half when stress is clearly normal, saving tracer, camera time and patient waiting. Needs a reader available to make that call the same morning.
Two dayStress and rest on separate days at similar activityBest image quality and lowest activity per acquisition. Costs the patient a second trip, which is a genuine problem when they have travelled in from Rangpur or Barishal.

Tracer supply pushes most Bangladeshi departments towards batching. Generator elution and kit reconstitution follow a delivery schedule, and a cardiac agent is not something you reconstitute for one walk-in patient. Fix two or three cardiac days a week, book to them, and hold one overflow slot for the inpatient who deteriorates.

Why dedicated cardiac cameras exist

A general purpose dual head camera images the heart by orbiting large detectors around the whole chest. For most of that orbit the detectors sit a long way from the heart, and geometric resolution on a parallel hole collimator degrades with distance from the collimator face. The collimator also discards the overwhelming majority of emitted photons by design, because that is how it forms an image at all. The result is a study that works perfectly well but takes the time it takes, and gets harder in large patients.

Dedicated cardiac systems exist to attack that geometry. The D-SPECT Cardio from Spectrum Dynamics Medical uses cadmium zinc telluride detectors, which convert gamma photons to charge directly rather than through a scintillator and photomultiplier chain, and places them close to the chest and aimed at the heart, with the patient seated or semi-reclined rather than lying flat with arms above the head. Sampling the heart directly instead of orbiting the whole torso is what buys the shorter acquisition, and that saving can be taken either as faster lists or as lower administered activity. The seated position is also easier for a breathless patient to hold still in, which quietly reduces motion artefact.

The trade-off deserves stating plainly. A dedicated cardiac camera does cardiac work and nothing else. For a hospital running a mixed list where bone, renal and thyroid studies fill most of the week, a cardiac-only system is a poor use of a shielded room and a licence. For a cardiac centre where the cardiac list is the constraint and referrals are already full, it is the right machine. Departments that want both usually specify a general purpose dual head system such as the ECAM SCINTRON family, or a digital CZT SPECT/CT such as the VERITON-CT series, and add a dedicated cardiac system later once volume justifies its own room.

The camera is rarely the bottleneck in a new cardiac service. The stress room, the trained physician and the uptake delays are.: Vvon Technologies commissioning practice

Planning throughput without fooling yourself

Cardiac throughput is over-estimated more often than any other study, because uptake delays hide the real constraint. Work it out on paper before the business case is signed.

  1. Write down the exact protocol you intend to run, including whether the rest study is routine or conditional on the stress result.
  2. Time every step separately: arrival and consent, cannulation, the stress test, the uptake delay, positioning, acquisition, and the technologist's image check before the patient leaves.
  3. Mark which steps occupy the camera. Uptake delays do not, so a second patient can be stressed while the first waits.
  4. Count how many patients can be in the pipeline at once. The limit is usually the stress room and the staff who can supervise, not the detector.
  5. Subtract start of day quality control and the time lost to collimator changes if the same camera also runs general work.
  6. Apply a reality factor for late arrivals, cancellations, patients who ate breakfast after being told not to, and the stress test that goes wrong and takes the room out for an hour.

Staffing a cardiac service

Commissioning: what to insist on before you sign acceptance

  1. Acceptance testing against the manufacturer's stated performance, witnessed by your own medical physicist, with the signed report in your file rather than on the engineer's laptop.
  2. A cardiac phantom acquisition reconstructed on your workstation using the protocols you intend to run, not a factory demonstration data set.
  3. Uniformity, sensitivity and centre of rotation baselines recorded on handover day. Every future fault argument begins with what the system looked like when it was new.
  4. A gated acquisition checked end to end, including how the system handles ectopic beats and arrhythmia rejection, because that is where gated ejection fractions go wrong.
  5. Clinical protocols loaded and locked on both camera and workstation, so they cannot be edited casually by whoever is on duty.
  6. Training signed off against named individuals with dates, and a second visit booked for two to three months later when your staff have real questions.

Back to all Insights