Post-filler photo assessment is the scenario every injector eventually meets and almost nobody has a protocol for: a text message, one image of a lip, and a patient who does not want to drive back. Michelle Langston calls it the worst position she can be put in. The most terrifying situation is when you have a patient send you a photo like this, and we can't get them to come in.
Terrifying is the right word, and the reason is not the image. It is that the image creates an obligation to act on information that is structurally incapable of supporting the decision you now have to make. This piece is about what that information actually is, how to improve it, how to convert the message into an in-person assessment, and what to record when you cannot.
What a physical perfusion assessment actually is
Start by counting the channels, because the count is the argument.
Assessing perfusion in person is a multi-channel examination. You have colour under known lighting. You have capillary refill, which is not a static observation at all but a dynamic test with a defined manoeuvre and a comparison. You have temperature by touch. You have pain on provocation, and pain over time. You have sensation. You have turgor and fluctuance under your fingers. You have the untreated control tissue millimetres away, in the same light, at the same moment. And you have the patient's trajectory in real time, watched rather than reported.
A photograph delivers exactly one of those channels. Degraded. Once.
That asymmetry — eight channels versus one — is the whole clinical content of this article, and everything below is a consequence of it.
What the phone does to the one channel you get
The degradation is not incidental. Modern phone cameras are engineered to produce a flattering, normalised image, and every one of those engineering decisions attacks the specific information you are looking for.
Automatic white balance removes colour casts. The finding you want is a colour cast. A lip that is duskier or paler than surrounding tissue is, to the camera's processing pipeline, an error to be corrected.
Computational tone mapping and HDR lift shadow detail and compress the tonal range. Mottling is a low-contrast pattern of subtle tonal differences; flattening the range reduces it toward uniformity.
Skin smoothing and beauty processing are enabled by default on many front-facing cameras and inside many messaging applications. They are explicitly designed to remove uneven skin tone. That is the finding.
Messaging recompression. An image sent inside a chat application is typically resized and recompressed, and compression artefacts land hardest on exactly the kind of subtle, low-contrast, mottled texture that carries the diagnosis.
Ambient light. Bathroom fluorescent, car interior, overhead office light and evening lamplight all shift skin colour substantially and in different directions. You have no idea which one you are looking at.
Flash washes out pallor, flattens contour and puts specular highlights across wet vermilion — the worst possible illumination for this particular assessment.
No scale, no fixed distance, no fixed angle. Two images taken twenty minutes apart cannot be compared for boundary change if the framing moved, which it always does. The most valuable thing you could extract from a series — is this extending — is precisely the thing casual photography destroys.
Unknown time. The send time is not the capture time. The patient may be sending you something from an hour ago, and will not think to mention it.
Add one further limitation that deserves care rather than a slogan: clinical signs that depend on colour change are harder to assess in more deeply pigmented skin, and the general clinical-assessment literature has flagged both that difficulty and the thinness of the evidence base for doing it well — for example, a recent scoping review of assessing clinical deterioration in children with dark-coloured skin (Kelly et al., J Clin Nurs 2026;35:513-523). That work is not about filler and not about adults, and it should not be over-read. The defensible point is narrow: an assessment channel that is already harder in person does not become easier after white-balance correction and lossy compression.
The evidence on patient-supplied photographs, read honestly
Patient-submitted images are not useless, and it would be dishonest to pretend otherwise. In a randomised study at the Children's Hospital of Philadelphia, concordance between pediatric dermatology diagnoses made from parent-submitted smartphone photographs and those made in person was 83%, rising to 89% among images judged of sufficient quality (O'Connor et al., JAMA Dermatol 2017;153:1243-1248). Notably, giving parents a three-step photography instruction sheet did not produce a statistically significant improvement.
Now read why that result does not transfer to your text message. Those were morphologic diagnoses of dermatologic conditions — pattern recognition on stable lesions. They were not perfusion assessments, which depend on a dynamic manoeuvre the camera cannot perform. They were not time-critical: a diagnosis reached a day later was still a correct diagnosis. And the cost of the 17% that were discordant was, in that setting, a follow-up appointment rather than tissue loss.
So the literature supports a modest claim — photographs carry real diagnostic signal for stable morphology — and offers no support at all for the claim that matters here, which is that a normal-looking photograph excludes an evolving perfusion deficit.
The one rule that survives all of this
Use images asymmetrically.
A photograph can escalate you. A photograph can never stand you down.
If the image looks concerning, that is sufficient to act — you do not need to see it in person to decide that the patient must be seen, and you should not spend twenty minutes trying to improve the picture before saying so. If the image looks fine, you have learned almost nothing, because every degradation listed above works in the direction of making abnormal tissue look normal. A reassuring photograph is the single most dangerous artefact in this entire scenario, because it hands both of you a reason to do nothing.
Ready to put this into practice?
Explore Empire's hands-on, CME-accredited Aesthetic & Injectable Training courses — live patients, expert faculty, and ongoing mentorship.
That rule is the remote-care version of Langston's bedside rule. Any visual component is an immediate stop; if there is any question, you treat. Neither rule contains a clause that lets an image resolve the question in the reassuring direction.
Getting a better image anyway
You still want the best picture you can get, because it may escalate you and because it becomes part of the record. Give short, specific instructions — long ones do not get followed.
- Daylight, near a window, no flash. Face the light, do not stand with your back to it.
- Turn off beauty mode and any filter. Use the rear camera, not the selfie camera.
- Send from the camera roll as a full-size file, not captured inside the chat window, so the application compresses it less.
- Three shots, in this order: whole face straight on with a neutral expression; the area of concern close up; and a matched view that includes the untreated tissue — the lower lip, the other side. The third one is the one you actually need and the one nobody sends.
- Repeat the identical set in a stated number of minutes, framed the same way. A pair of comparable images tells you about direction, and direction is the finding.
- A short video of blanch and release. Ask them to press the area firmly with a clean fingertip, release, and keep filming for several seconds, then do the same on the untreated side. This is a crude proxy for capillary refill and it is the only channel besides colour that a phone can carry at all. Tell them explicitly that it is not a substitute for being seen.
That last instruction is not a hedge. The reason to say it out loud is that a patient who successfully completes a technical-sounding task tends to believe the assessment is finished.
Escalating the request
The goal of the conversation is not to diagnose. It is to move the patient from their car to your chair. Reluctance is usually practical — time, childcare, traffic, work, cost, embarrassment — and it is worth attacking the practical reason rather than the reluctance.
Say the concern plainly and without decoration. "I am concerned that the blood supply to part of your lip may be blocked. That is time-sensitive, it is treatable, and it is treatable best right now." Then remove barriers concretely: we will see you immediately, we will stay open, here is the address, bring someone who can drive, tell me what is stopping you and let us solve it.
Avoid two failure modes. Do not minimise — "probably just a bruise, but pop in if you like" gives the patient explicit permission to not come, and you will not get a second call. And do not catastrophise into paralysis, which produces a frightened person googling in a parking lot rather than a person driving toward you.
If they genuinely cannot reach you, they still need to be assessed by someone who can see them in person, and you should say so and help them identify where. What that facility should then do is a clinical conversation between clinicians, not a message to a patient.
When they still will not come
Langston's instruction is one clause long and it is the right one: we would want to document that right away.
Document contemporaneously, not that evening. Record the time of contact and the time of the images as distinct items. Record verbatim what the patient reported — their words, in quotation marks, not your paraphrase. Record what you observed in the images and, importantly, what you could not assess from them. Record what you advised, that you explained the time-critical nature of the concern and the possible consequence of delay, what you offered, and the patient's stated response in their words. Record every subsequent attempt to reach them with times, and retain the original image files rather than screenshots, so that metadata survives.
Two boundaries. First, this is clinical documentation practice, not legal advice; how your practice handles refusal of care, retention and patient communication should be reviewed with your own counsel and your malpractice carrier, and policies vary by state and by carrier. Second, documenting a refusal is not a substitute for continuing to try. Keep calling.
The system fix happens earlier
This scenario is not created by the text message. It is created at the consultation, where a patient learns — or does not — that there is a category of event after lip filler that requires them to return immediately, and that "it looks like a bruise to me" is not their call to make. Langston puts the obligation there: during the assessment and the good faith exam, patients need to know that if we say it needs to be dissolved, then they need to come in and have it dissolved.
A practice whose patients arrive already holding that expectation has a much smaller version of this problem. One whose patients first hear it in a text message at 6pm has the full version. Consent and expectation-setting documentation is a related discipline — what to include in consent forms covers the structural side — and the patient-facing account of ordinary post-treatment lip change, useful to send before treatment rather than after, is common lip filler reactions and how to avoid them.
This reflects Michelle Langston's clinical practice as taught in Empire Medical Training's hands-on curriculum. Technique is learned under supervision; this article is educational and is not a substitute for training, nor is it legal advice.
Empire's Complete Dermal Filler Training and Complete Facial Aesthetic Training build complication recognition and response into the hands-on work; dissolving filler covers the agent you need available before any of this becomes relevant.
Frequently Asked Questions
Can I rule out a vascular occlusion from a photograph?
No. A physical perfusion assessment uses colour, capillary refill, temperature, pain on provocation, sensation, palpation and real-time comparison with untreated tissue. A photograph carries one of those channels, degraded by white balance, tone mapping, smoothing and compression — all of which push abnormal tissue toward looking normal. Images can escalate you; they cannot stand you down.
What should I ask the patient to send?
Three images from the rear camera in daylight without flash or filters: whole face straight on, close-up of the area, and a matched view including the untreated side or lower lip. Then the identical set again after a stated interval. A short video of press-and-release on both the involved and uninvolved tissue adds the only dynamic information a phone can carry.
Is a video better than a still image?
Yes, marginally and importantly. Capillary refill is a dynamic test, and a video of blanch and release is the only remote channel that captures any of it. It remains crude — you cannot control pressure, duration, lighting or comparison — so treat it as a reason to bring the patient in rather than as a completed assessment.
What do I document if the patient refuses to be seen?
Contemporaneously: time of contact, time the images were taken, the patient's report in their own words, what you observed and what you could not assess from the images, what you advised including the time-critical nature and possible consequences, what you offered, their stated response, and every follow-up attempt with times. Retain original files rather than screenshots.
How do I get a reluctant patient to come in?
State the concern plainly, give the time-sensitive reason without minimising or catastrophising, then attack the practical barrier rather than the reluctance — immediate appointment, staying open, directions, someone to drive. "Probably just a bruise, but come in if you want" is the phrasing that guarantees they will not.
Disclaimer
This article reflects the clinical opinions and experience of Michelle Langston, APRN, MSN, FNP-BC, an independent faculty member contributing to Empire Medical Training's curriculum. The views expressed are the author's own and do not necessarily represent those of Empire Medical Training.
It is professional education, not medical advice, and is no substitute for hands-on training or independent clinical judgment. Licensed clinicians remain responsible for their own patient selection, technique and outcomes, for verifying current product labelling, and for practising within their scope and applicable law. Empire Medical Training accepts no liability for reliance on this content.
