Computed tomography - Radiation Dose Safety and Artifact Management

Image Quality, Radiation Dose, and Artifacts in CT Imaging Introduction Computed tomography (CT) scanning offers exceptional diagnostic power, but this capability comes with an important tradeoff: exposure to ionizing radiation. Understanding how radiation dose affects image quality, learning to recognize and manage artifacts, and appreciating the associated safety risks form the foundation of responsible CT practice. This chapter explores these interconnected concepts to help you produce diagnostically excellent images while minimizing harm to patients. Radiation Dose and Image Quality The fundamental relationship: Higher radiation dose improves image quality, while lower dose degrades it. This creates a constant tension in medical imaging: we want to deliver just enough radiation to make an accurate diagnosis, but no more. When you increase the radiation dose delivered to a patient, two things happen in the resulting image: Reduced noise: The detector captures more X-ray photons, leading to better signal-to-noise ratio. Images appear "cleaner" with less grainy appearance. Improved spatial resolution: More photons allow for sharper definition of anatomical edges and smaller structures. Conversely, when dose is reduced to minimize patient exposure, the images become noisier and less sharp. The challenge is determining how much quality degradation is acceptable for the clinical question being asked. Modern Dose-Reduction Strategies Rather than simply increasing radiation dose whenever image quality seems marginal, modern CT relies on sophisticated techniques to maintain diagnostic quality at lower doses. Iterative reconstruction algorithms are a major breakthrough. Traditional CT reconstruction (called filtered back-projection) is fast but requires relatively high doses to avoid excessive noise. Iterative algorithms, such as iterative sparse asymptotic minimum variance (ISAM), work differently: they start with an initial image estimate and repeatedly refine it by comparing simulated projections against the measured data. This process can produce high-quality images with as much as 30-50% dose reduction compared to traditional methods. The tradeoff is increased computational time, though modern scanners complete this within acceptable processing timeframes. Dose tailoring to patient characteristics represents another key strategy: Body size: A dose appropriate for a 70 kg adult would be excessive for a 20 kg child. Modern protocols adjust mA (tube current) automatically based on patient size. Organ of interest: A lung CT focused on detecting pulmonary nodules may require higher doses than an abdominal CT performed primarily to rule out free fluid. Clinical indication: A CT used to rule out serious pathology may justify higher doses than a follow-up scan to assess known, stable disease. Understanding Radiation Dose Metrics Before discussing artifacts and safety, you need to understand how radiation dose is measured in CT. Gray (Gy) and milligray (mGy): The absorbed dose—the actual energy deposited in tissue—is measured in gray. One gray equals one joule of energy per kilogram of tissue. Because CT doses are typically small, we use milligray (mGy), where 1 Gy = 1000 mGy. For context: A plain chest X-ray delivers about 0.01–0.05 mGy to the chest A typical abdominal CT delivers 10–20 mGy to organs in the scanned region Specialized CT exams (such as some cardiac protocols) may reach 80 mGy Computed Tomography Dose Index (CTDI): Manufacturers and hospitals use CTDI as a standardized measure of scanner output. CTDI estimates the absorbed dose in a specific region if you were to scan repeatedly at the same position (even though clinical scanning moves the table). CTDI values are reported in mGy and printed on the scanner console after each scan. Typical CTDI values range from 10–40 mGy depending on protocol. Dose-Length Product (DLP): Since actual CT scans cover a range of anatomical length (not just one position), we combine CTDI with scan length to estimate total dose: $$\text{DLP (mGy·cm)} = \text{CTDI (mGy)} \times \text{Scan Length (cm)}$$ A 30 cm abdominal scan with CTDI of 15 mGy would have a DLP of 450 mGy·cm. Effective Dose: This is where things become more complex. Different organs have different radiosensitivity—the thyroid is more sensitive to radiation damage than muscle tissue. Effective dose attempts to weight organ doses by their sensitivity and combine them into a single number representing whole-body cancer risk. Effective dose is measured in sievert (Sv) or millisievert (mSv). The relationship between effective dose and cancer risk is estimated from data on atomic bomb survivors and other exposed populations. A rough guideline: a single abdominal CT delivering 8 mSv adds an estimated 0.05% (1 in 2,000) lifetime risk of developing cancer. This sounds small, but when multiplied across millions of CT scans performed annually, it translates to significant numbers. Biological Effects of Radiation Understanding radiation damage at the cellular level clarifies why we're concerned about CT doses. Deterministic effects occur when radiation dose is high enough to kill or damage significant numbers of cells: Threshold dose: Usually requires several gray (thousands of mGy) Examples: Skin burns, sterility, cataracts Important note: Typical CT doses (10–80 mGy) are far below these thresholds for deterministic effects Stochastic effects are different. These are probabilistic injuries arising from mutations in individual cells: Cancer induction: Radiation may damage DNA in a way that leads to malignant transformation Heritable effects: If germ cells are affected, mutations may pass to offspring Critical feature: There is no "safe threshold." Any dose, no matter how small, carries a small risk of inducing cancer Dose relationship: Risk increases with dose, but the relationship is not perfectly linear (it becomes less steep at very low doses) This distinction is crucial: CT doses are far too low to cause deterministic injury, but they do carry a quantifiable stochastic risk that must be weighed against diagnostic benefit. Common CT Artifacts Artifacts are false or distorted information in the image not representing actual patient anatomy. Four types are particularly important in CT practice. Beam-Hardening and Streak Artifacts What causes them: X-ray beams are polyenergetic—they contain photons of many energies. Low-energy photons are preferentially absorbed as the beam passes through tissue. This means the beam becomes progressively enriched in high-energy (harder) photons—hence "beam hardening." When a high-density material (especially metal) lies in the beam's path, it absorbs nearly all photons, creating a region of photon starvation. The detectors receive minimal signal, making it impossible to reconstruct the true attenuation at those locations. The reconstruction algorithm can't distinguish between "very dense material" and "no signal," producing dark streaks and bands radiating from the metal object. Clinical scenarios: Metal dental work, orthopedic implants, pacemakers, and surgical clips are common sources. The artifact severity depends on the metal's density and size. Mitigation strategies: Metal-artifact reduction software: Modern algorithms modify the reconstruction to minimize streaking around metal Virtual monochromatic imaging: Using the data from dual-energy scans, you can synthesize images as if they were acquired with a single energy, reducing beam-hardening effects Patient positioning: Sometimes repositioning can angle the metal away from critical anatomy Iterative reconstruction: Particularly helpful because these algorithms can model metal's effects more accurately than filtered back-projection Motion Artifacts What causes them: CT acquisition takes time—even fast modern scanners require several seconds for a complete scan. If the patient moves (breathing, swallowing, cardiac motion) during this acquisition, the projection data becomes inconsistent. The reconstruction interprets this inconsistency as anatomy in multiple positions, creating blur and ghosting (duplicate images). Motion artifacts are particularly problematic in slower spiral scans and in organs affected by respiratory motion (lungs, liver, kidneys). Mitigation strategies: Faster scanner rotation speeds: Modern scanners rotate in 0.3–0.5 seconds per revolution, minimizing the time window for motion Breath-hold instructions: Clear directions to "hold your breath in" and coaching during the scan markedly reduce respiratory artifact Cardiac gating: For heart imaging, synchronizing acquisition with the cardiac cycle (using ECG triggering) eliminates motion artifact Marker-based tracking: Research systems track external markers to prospectively adjust scanning, though this remains uncommon clinically Partial-Volume Artifacts What causes them: CT images are composed of voxels (volumetric pixels)—three-dimensional blocks of tissue that are displayed as one number. When a thick imaging slice contains different tissue types (say, bone and air), the reconstruction averages their attenuation values, producing an intermediate number that doesn't reflect reality. This averaging reduces the apparent density of high-density objects and raises the apparent density of low-density objects—boundaries become blurry, and small structures disappear if they're smaller than the slice thickness. Clinical impact: A bone cortex passing through a thick slice appears less dense than it truly is; a small nodule might be completely obscured by averaging with surrounding lung. Mitigation strategies: Thinner slices: Modern thin-slice imaging (0.5–1 mm) dramatically reduces partial-volume effects Reconstruction in multiple planes: Reviewing images in coronal and sagittal planes (not just axial) provides additional information to interpret anatomy Partial-volume correction algorithms: Software estimates the true tissue fractions within voxels to improve quantitative accuracy Appropriate window/level settings: Adjusting display settings can emphasize remaining edge information Ring Artifacts What causes them: Modern CT detectors must be precisely calibrated to report identical values when they receive identical radiation. Imperfect detector calibration leads to systematic errors—one detector might report values consistently 5% too high. During reconstruction, this manifests as concentric circles or rings in the image centered on the scanner's axis. Ring artifacts are less common with modern detectors but can still appear with aging equipment or after detector damage. Mitigation strategies: Adaptive center determination: Software automatically corrects for small misalignments of the image reconstruction center Eigen-flat-field normalization: Advanced calibration methods compensate for individual detector variations Pediatric Radiation Concerns Children deserve special mention because they are significantly more radiosensitive than adults. This increased risk stems from two factors: Longer life expectancy: Children have more years ahead in which radiation-induced cancer might develop Developing cells: Growing tissues are inherently more vulnerable to radiation damage than mature tissues The "Image Gently" campaign, endorsed by radiological societies, emphasizes age-appropriate dose reduction: Low-kV protocols: Using lower tube voltage (80–100 kV instead of 120 kV) in thin children reduces dose while maintaining image quality Iterative reconstruction: Particularly valuable in children because it enables high-quality imaging at lower doses Careful justification: Ensuring each pediatric CT is truly indicated and not performed out of habit A pediatric abdominal CT that might deliver 3–5 mSv in a child contributes a higher proportional cancer risk than the same dose in an adult, reinforcing the importance of strict dose justification. Cumulative and Long-Term Dose Considerations A single CT scan may carry acceptable risk, but patients often undergo multiple scans over their lifetime. Cumulative dose problem: Consider a patient who has: Baseline CT for an incidental finding: 8 mSv Follow-up CT three months later: 8 mSv Another follow-up one year later: 8 mSv Total: 24 mSv over a short interval Repeated exposures within 1–5 years further elevate cancer risk in ways that may not be simply additive. This is why tracking dose history and applying strict justification criteria—"Is this scan necessary now?"—is essential. Best practices: Review previous imaging before ordering new studies; a prior CT may answer your question Use imaging follow-up intervals supported by evidence, not automatically repeating studies Consider alternative imaging (ultrasound, MRI) when appropriate Maintain dose tracking systems to identify patients receiving cumulative high doses Contrast-Induced Nephropathy Iodinated contrast agents used in CT carry their own risks, distinct from radiation exposure. What is contrast-induced nephropathy (CIN)? It is acute kidney injury occurring within 48–72 hours of contrast administration, defined as a ≥25% increase in serum creatinine from baseline or ≥0.5 mg/dL absolute increase. Incidence and risk factors: Affects 2–7% of patients receiving iodinated contrast overall Dramatically higher in patients with pre-existing renal impairment (especially eGFR <30 mL/min) Additional risk factors: diabetes, dehydration, older age, concurrent nephrotoxic medications Note: Risk is now considered considerably lower than previously thought, as awareness of risk factors has improved prophylaxis Prevention strategies: Pre-screening: Check serum creatinine before contrast administration to identify at-risk patients Hydration: IV hydration before and after contrast administration is the most evidence-based intervention, reducing CIN risk in patients with mild-to-moderate renal impairment Low-osmolar agents: Using lower-osmolarity contrast media (rather than high-osmolarity agents) reduces but doesn't eliminate risk Metformin considerations: In diabetic patients with renal impairment, metformin should be held after contrast exposure (patient-specific timing depends on renal function) because the combination of contrast and metformin increases lactic acidosis risk Severe renal impairment (eGFR <30): Iodinated contrast should be avoided or alternative imaging techniques (ultrasound, non-contrast MRI, non-contrast CT) should be used. If contrast is absolutely necessary, extra precautions (aggressive hydration, brief interval before dialysis for dialysis patients) are required. Allergic-Type Reactions to Contrast Iodinated contrast agents can provoke immediate hypersensitivity reactions: Mild: Urticaria, pruritus Moderate: Mild bronchospasm, hypotension Severe: Anaphylaxis with severe bronchospasm, cardiovascular collapse Risk factors: History of atopy, asthma, and prior contrast reaction substantially increase risk. Prevention in high-risk patients: Premedication with corticosteroids (and antihistamines) several hours before contrast administration significantly reduces reaction risk Using non-ionic, iso-osmolar contrast agents when possible (though modern non-ionic agents carry low reaction rates regardless) <extrainfo> Historical context: Prior concerns about iodine allergy have been largely debunked—true allergy to iodine itself is rare, and the risk is primarily from the contrast molecule structure rather than iodine content. This distinction matters because iodine avoidance in patients with contrast reactions is usually unnecessary. </extrainfo> Patient Communication and Professional Standards Current guidelines emphasize informed decision-making about radiation and contrast risks. What patients should know: The approximate radiation dose (can be communicated in terms of multiples of annual background radiation if helpful) The stochastic cancer risk in terms they can understand (e.g., "This CT adds about a 1 in 2,000 lifetime risk of cancer") Contrast risks if contrast is planned, particularly if they have renal disease or prior reactions The clinical indication for the CT and why it's justified despite these risks Professional practice: Order CT only when there is clear clinical justification Follow evidence-based screening intervals (not arbitrary 3- or 6-month follow-ups) Maintain communication with referring providers about cumulative dose if patients are undergoing multiple studies Document the clinical indication in the medical record Summary Table of Key Dose Values | Examination | Typical Absorbed Dose (mGy) | Typical Effective Dose (mSv) | |---|---|---| | Plain chest X-ray | 0.01–0.05 | 0.02 | | Abdominal CT | 10–20 | 8–15 | | Chest CT | 8–15 | 7–12 | | Cardiac CT (specialized) | Up to 80 | 15–20 | | Head CT | 40–60 | 2–3 | These values vary by scanner, protocol, and patient size, but give you a sense of the relative doses delivered by different exams. Putting It Together: A Clinical Example A 55-year-old diabetic patient with eGFR of 45 presents with symptoms suggesting possible small bowel obstruction. You're considering an abdominal/pelvic CT with IV contrast. Your thought process should include: Justification: Is CT indicated? Yes—clinical evaluation alone can't definitively diagnose obstruction, and CT guides management decisions. Dose consideration: This CT will deliver 8–15 mSv of effective dose. Is the diagnostic benefit worth this cancer risk? Yes, because missing an obstruction has serious consequences. Contrast safety: eGFR 45 indicates mild-to-moderate renal impairment. Check baseline creatinine. Provide IV hydration before and after contrast. Consider holding metformin. Watch for CIN symptoms (rising creatinine) in the 48–72 hours post-exam. Artifact prevention: Scout the patient for metal implants. Ensure proper breath-hold instruction. Use modern iterative reconstruction if available to optimize image quality at the dose used. Cumulative exposure: Verify you're not repeating a recent CT unnecessarily. Documentation: Record the clinical indication and relevant risk factors. This approach balances diagnostic excellence with radiation safety—the essence of responsible CT practice.