Clinical overview
Ultrasound is the workhorse imaging modality of obstetrics and gynaecology, and in the South African public sector it is frequently the only imaging available at district and regional level. Unlike a radiograph that is captured once and read later, ultrasound is an operator-dependent, real-time examination: the diagnosis is only as good as the image the sonographer makes, and the image is only as good as the operator's command of the machine. "Knobology" — the deliberate, purposeful manipulation of the machine controls (transducer selection, frequency, gain, time-gain compensation, depth, focus, dynamic range, harmonics, and the Doppler controls) — is the practical skill that turns a noisy grey rectangle into a diagnostic study. This is a HOTS (higher-order) objective: the examiner does not want you to recite that "gain controls brightness"; they want you to demonstrate that you can optimise an image and operate Doppler safely in a pregnant patient.
The safety dimension is what distinguishes obstetric scanning from most other ultrasound. Diagnostic ultrasound is generally regarded as one of the safest imaging modalities and has no proven harmful effect at routine diagnostic intensities, but it is not energy-free. Ultrasound deposits energy as heat (thermal effect) and exerts mechanical forces (cavitation, radiation force). The embryo and fetus — particularly developing neural tissue and ossifying bone — are the tissues of greatest theoretical concern. Doppler modes, especially spectral pulsed-wave and colour Doppler, concentrate far more acoustic energy into tissue than B-mode (greyscale) imaging. The governing principle, demanded by obstetric-ultrasound practice and by every safety body, is ALARA — As Low As Reasonably Achievable: obtain the diagnostic information you need with the lowest output and shortest exposure that achieves it.
Core knowledge
Figure F11.1 — The greyscale knob drill: optimise before you diagnose — preset/probe, depth, frequency, focus, gain, TGC, dynamic range and harmonics — a good image first, diagnosis second.
How the image is formed
A transducer converts electrical energy to mechanical (sound) energy and back via the piezoelectric effect. Pulses of high-frequency sound (typically 2–18 MHz in O&G) travel into tissue, reflect at interfaces of differing acoustic impedance, and return as echoes. The machine times each echo's return to place it at the correct depth, and assigns its brightness from echo amplitude — this is B-mode (brightness mode), the standard greyscale image.
Two facts drive most knobology decisions:
- The frequency–penetration trade-off. Higher frequency = better axial resolution but poorer penetration (it attenuates faster); lower frequency penetrates deeper but resolves less detail. Hence a high-frequency transvaginal probe (classically ~5–9 MHz) gives exquisite detail of the pelvis at short range, while a lower-frequency curvilinear abdominal probe (classically ~2–5 MHz) reaches the posterior wall of a gravid uterus or an obese abdomen.
- Attenuation. Sound weakens with depth, so deeper echoes return weaker. The machine and operator must compensate for this to produce an evenly bright image.
The transducers you will use
- Curvilinear / convex abdominal probe (low frequency, wide field) — transabdominal obstetric and gynae scanning.
- Endocavitary / transvaginal probe (high frequency) — early pregnancy, ectopic assessment, the adnexa, cervical length, and detailed pelvic anatomy.
- Linear high-frequency probe — superficial structures (e.g. assessing a perineal/abdominal wall collection, vascular access).
Probe selection is itself a knobology decision: matching the right frequency and footprint to the target is the first optimisation step.
The core greyscale controls
- Gain amplifies all returning echoes uniformly — it brightens (or darkens) the whole image. Crucially, gain amplifies signal and noise; over-gaining fills anechoic structures (bladder, cyst, amniotic fluid) with false low-level echoes and can mimic debris, while under-gaining loses real soft-tissue detail.
- Time-gain compensation (TGC), a bank of sliders, selectively amplifies echoes from specific depths to offset attenuation, producing uniform brightness from near to far field. The classic error is a stepped or banded image from poorly set TGC sliders.
- Depth sets how far the image extends. Set it just beyond the structure of interest so the target fills the screen — excess depth wastes pixels and also slows the frame rate.
- Focus (focal zone) is the depth of narrowest beam width and therefore best lateral resolution. Place the focal marker at (or just deep to) the region of interest; multiple focal zones improve resolution but lower frame rate.
- Frequency / penetration can often be adjusted on a broadband probe (a "Res / Gen / Pen" or kHz toggle) — drop frequency for a difficult, deep, or obese patient; raise it for a thin patient or near-field detail.
- Dynamic range sets the number of grey shades displayed. Wide dynamic range = smoother, softer image (good for soft-tissue gradation); narrow = high-contrast, more black-and-white (useful for cystic/solid distinction).
- Tissue harmonic imaging (THI) uses harmonic frequencies generated within tissue to reduce near-field artefact and clutter — particularly useful in obese patients and to clarify cyst contents.
- Zoom (read/write zoom), cine loop (review the last few seconds frame-by-frame, invaluable for capturing a fleeting fetal view), and measurement calipers complete the everyday toolkit.
Doppler modes and why they matter for safety
Doppler exploits the frequency shift of sound reflected from moving blood cells to interrogate flow.
- Colour Doppler overlays a colour map of mean velocity/direction on the B-mode image within a user-placed box.
- Power (amplitude) Doppler displays the strength of the Doppler signal, is more sensitive to low/slow flow and angle-independent, but gives no directional or velocity information.
- Pulsed-wave (spectral) Doppler samples flow in a small gate and produces the velocity waveform from which indices are derived.
The acoustic output and tissue heating rank, lowest to highest, approximately: B-mode < colour Doppler < pulsed/spectral Doppler. Spectral Doppler concentrates energy by repeatedly insonating a tiny stationary volume, and is the mode of greatest thermal concern — especially over bone (e.g. fetal skull, spine) where heating is greatest, and most especially in the first trimester when the embryo lacks the thermoregulatory buffer of later flowing amniotic fluid and circulation.
The on-screen safety indices: TI and MI
Modern machines display two real-time output indices, mandated by the output display standard:
- Thermal Index (TI) — an estimate of the potential temperature rise; it approximates the maximum temperature increase in degrees Celsius under defined assumptions. It comes in sub-types: TIS (soft tissue — use in early first trimester before bone ossifies), TIB (bone-at-focus — use from late first trimester onward when fetal bone is in the beam), and TIC (cranial bone — transcranial, rarely relevant in obstetrics).
- Mechanical Index (MI) — an estimate of the likelihood of non-thermal (cavitation) bioeffects; relevant mainly where gas bodies exist (not typically the fetus, but relevant principle).
These indices are the operator's dashboard for ALARA. The widely taught precautionary thresholds are that scanning at a TI ≤ 0.7 may proceed without specific time restriction, while higher TIs warrant limiting exposure time — but treat exact numeric thresholds as standard precautionary teaching rather than a fixed legal limit (flagged as such). The unambiguous, non-negotiable rule is to know where TI/TIB is displayed, keep it as low as possible, and minimise dwell time — particularly with Doppler in the first trimester.
Assessment
This objective is demonstrated, so the assessment section is the practical core. Approach every scan as a reproducible optimisation routine.
Before the probe touches the patient
- Consent and explanation, especially in obstetrics — explain purpose and limitations; an ultrasound is not a guarantee of a normal baby. Maintain a chaperone for transvaginal scanning and document it.
- Select the correct preset. Machines have application presets (early pregnancy, fetal, gynae, etc.) that load appropriate output limits and optimised settings — the single most important safety knob is choosing the obstetric/early-pregnancy preset, which constrains acoustic output, rather than a generic or, dangerously, a cardiac preset (cardiac presets permit high output for adult hearts and are wholly inappropriate for a fetus).
- Select the correct transducer for depth and target.
Optimising the greyscale image — a stepwise drill
- Depth — open it up, find the target, then tighten depth so the target fills two-thirds of the screen.
- Frequency / penetration — raise frequency for detail in a thin patient; drop it (or engage harmonics) for penetration in obesity or deep structures.
- Overall gain — set so fluid spaces (bladder, amniotic fluid, cysts) read truly anechoic (black) without low-level fill, while soft tissue retains grey detail.
- TGC — balance near-to-far brightness; correct any banding.
- Focus — place the focal zone at the region of interest.
- Dynamic range / maps — narrow for cyst-vs-solid contrast; widen for soft-tissue gradation.
- Zoom and freeze/cine — magnify and use the cine loop to capture and measure the optimal frame; measure on a frozen, magnified image for caliper accuracy.
A registrar should be able to verbalise why each adjustment helps the specific clinical question — that is the higher-order skill being examined.
Optimising Doppler
When you do use Doppler (e.g. confirming flow in a suspected ectopic "ring of fire", interrogating umbilical artery in growth restriction per placental-insufficiency-response and intrauterine-growth-restriction, or characterising an adnexal mass):
- Keep the colour/sample box small — a smaller region of interest improves frame rate and concentrates the interrogation.
- Set the velocity scale (PRF) to the expected flow: too low causes aliasing in arterial flow; too high misses low venous/placental flow.
- Adjust the angle of insonation — keep it ≤60° for accurate velocity measurement; angle correction is needed for absolute velocities (e.g. MCA peak systolic velocity in fetal anaemia surveillance per rh-isoimmunisation).
- Wall filter removes low-frequency wall-thump artefact but set too high will erase genuine low diastolic flow (which matters when assessing absent/reversed end-diastolic flow).
- Optimise Doppler gain to fill the spectral waveform without noise speckle.
Crucially, watch the TI while doing all of this — colour and especially spectral Doppler will raise the displayed TI substantially.
Management
Demonstrating Doppler safety in practice — the ALARA workflow
Safe Doppler in O&G is a disciplined sequence, not an afterthought. Make this drill automatic:
- Confirm the preset loads obstetric output limits; never carry over a non-obstetric preset.
- Default to B-mode. Make the diagnosis with greyscale (and M-mode for fetal heart rate) wherever possible. M-mode, not spectral/colour Doppler, is the recommended way to document an embryonic/fetal heartbeat in the first trimester, because it delivers far less energy than spectral Doppler to a tiny target.
- Activate Doppler only when clinically necessary and only for as long as needed. Colour before spectral where it answers the question.
- Read the TI/TIB before and during Doppler insonation; keep it low (aim TIS/TIB toward the low end; the commonly taught precautionary ceiling for unrestricted time is TI ≤ 0.7 — standard teaching).
- Minimise dwell time over any single fetal location, and avoid prolonged insonation over fetal bone or the embryo, especially in the first trimester.
- Reduce output power if the diagnostic image allows — raising gain (a receiver function) is preferable to raising output power (a transmit function) because gain does not increase the energy delivered to the patient. This distinction — optimise with receiver gain before increasing transmit output — is a high-yield safety point.
- Avoid first-trimester spectral Doppler for non-medical reasons entirely — there is no justification for souvenir/"keepsake" Doppler or prolonged Doppler simply to let parents hear the heartbeat early; document the heart with M-mode.
South African context
In the South African public sector, ultrasound capacity is unevenly distributed. The NDoH National Integrated Maternal and Perinatal Care Guideline (NDoH, 2024) structures obstetric ultrasound expectations across levels of care — basic dating and viability scanning achievable at district level, with detailed anomaly and Doppler/fetal-medicine studies concentrated at regional and tertiary centres where appropriately trained operators and machines exist. The practical implications:
- Operator training is the rate-limiting safety factor. Where a single midwife-sonographer or medical officer covers a busy district unit, the discipline of correct presets and ALARA must be embedded in training, because there may be no radiologist to audit images.
- Machine maintenance and preset configuration matter: an inherited machine left on an inappropriate preset is a real hazard; part of demonstrating competence is checking the preset every session.
- Targeted, high-yield scanning is the SA norm — confirm viability, number, dating, presentation, placental site, liquor — rather than prolonged exploratory Doppler. This both serves throughput and aligns with ALARA.
- The HIV burden does not change ultrasound safety per se, but it raises the prevalence of growth restriction and placental pathology, increasing legitimate demand for umbilical/MCA Doppler at referral level — making safe, indication-driven Doppler more relevant, not less. Saving Mothers (NCCEMD) themes around missed or delayed diagnosis underline that the bigger SA risk is under-use of competent ultrasound (missed pathology) rather than bioeffects from over-use — but both errors are avoidable with disciplined knobology.
There is no specific bioeffect-related emergency drill for diagnostic ultrasound at correct settings; the "safety emergency" to internalise is the inappropriate-preset / prolonged-first-trimester-Doppler scenario — if you find yourself running spectral Doppler on an embryo, stop, drop to B-mode/M-mode, and reduce output immediately.
Red flags / pitfalls
- Diagnosing on a poorly optimised image. Over-gained anechoic structures generate false "debris" (mistaken haemorrhage or pus); under-gained images hide pathology. Always optimise before you interpret — never report off a suboptimal frame.
- Wrong preset / cardiac preset on a fetus. A high-output preset carried over from another study defeats every safety effort. Check it each time.
- Increasing transmit power when you only needed gain. Reach for receiver gain, harmonics, frequency and TGC before output power.
- First-trimester spectral Doppler. The single most important Doppler-safety pitfall: spectral Doppler to a first-trimester embryo (e.g. to "hear" the heart) is contrary to ALARA — use M-mode.
- Prolonged dwell over fetal bone. TIB rises over ossified bone; do not park a Doppler gate on the fetal spine/skull.
- Ignoring the TI/MI display. If you cannot find or do not watch these indices, you are not practising ALARA.
- Aliasing and angle errors misread as pathology. A too-low PRF aliases normal arterial flow; an uncorrected angle >60° yields spuriously inaccurate velocities (a clinical trap for MCA-PSV in fetal anaemia surveillance).
- Wall filter set too high, erasing genuine low/absent end-diastolic flow and masking the very abnormality (e.g. in severe FGR) you are looking for.
- Transvaginal scanning without consent/chaperone documentation.
- Over-reliance on auto-optimise buttons without understanding them — they help, but the examiner wants to see you reason through manual optimisation.
- Forgetting that ultrasound is operator-dependent and not infallible — a normal scan does not exclude all pathology; communicate limitations.
Evidence anchors
- ISUOG ultrasound safety statements / Bioeffects and Safety Committee guidance — the primary source for ALARA, the thermal (TI: TIS/TIB/TIC) and mechanical (MI) indices, the precautionary approach to first-trimester and Doppler exposure, and the recommendation to use M-mode rather than spectral Doppler to document the early fetal heartbeat. (Listed in
docs/VERIFIED-SOURCES.mdas ISUOG ultrasound safety — ALARA, thermal/mechanical index.) - ISUOG Doppler guidance for umbilical artery, MCA, ductus venosus and cerebroplacental ratio — the clinical indications that justify Doppler in growth restriction and fetal surveillance, framing when Doppler is appropriate.
- NDoH National Integrated Maternal and Perinatal Care Guideline (NDoH, 2024) — South African obstetric source of truth; structures ultrasound provision by level of care and dating/viability standards.
- Saving Mothers / NCCEMD triennial report — context that missed/delayed diagnosis is a leading avoidable factor in SA maternal deaths, reinforcing competent, indication-driven scanning.
- IOTA / O-RADS (per
docs/VERIFIED-SOURCES.md) — for the adnexal-mass application of greyscale and Doppler optimisation, expanded in ultrasound-malignancy-signs.
Note: specific numeric TI thresholds (e.g. TI ≤ 0.7 for unrestricted time, ≤60° Doppler angle, 5–9 MHz / 2–5 MHz probe frequency ranges) are presented here as standard precautionary teaching and conventional probe specifications, not as line-itemed values from a single verified guideline; the durable, examinable principles are ALARA, preset selection, M-mode for the early heartbeat, and minimising Doppler dwell time.