The present article consists of a two-year retrospective chart review of a single center’s vascular adverse events (VAE) associated with facial hyaluronic acid (HA) filler injections.1 The authors identified 7 patients (and only 7 patients in all) who fit their definition of a VAE. Upon review of the clinical surgical history of each of these 7 patients, they found that all 7 were former rhinoplasty patients. The authors do not report the total number of patients who were injected during this two-year period, nor do they report how many of their HA filler patients had an uneventful treatment despite their history of rhinoplasty. Unfortunately, this information does not give us the risk associated with fillers in former rhinoplasty patients. There are few estimates of the rate of VAE in filler patients in the literature, estimated as 3 out of 1000 injections.2 The values of the missing numbers in Table 1 make a huge difference to the interpretation of the results. As a valuable heuristic (ie, “rule of thumb”), it is useful to consider correlative data in the form of a fourfold (contingency) table, with the rows and columns divided as shown in Table 1. Table 1. Number of Vascular Adverse Events (VAE) Associated With Former Rhinoplasty Patients Rhinoplasty VAE No Yes Total Yes 0 7 7 No ? ? ? Total ? ? ? Rhinoplasty VAE No Yes Total Yes 0 7 7 No ? ? ? Total ? ? ? View Large Table 1. Number of Vascular Adverse Events (VAE) Associated With Former Rhinoplasty Patients Rhinoplasty VAE No Yes Total Yes 0 7 7 No ? ? ? Total ? ? ? Rhinoplasty VAE No Yes Total Yes 0 7 7 No ? ? ? Total ? ? ? View Large The diagonals of these contingency tables can be used to calculate the Pearson correlation of linear relationships between two variables. Table 1 shows us all the information that the authors have provided us, namely that they had 7 cases of VAE in total, and that all 7 of their VAE patients had a history of rhinoplasty. From this, the authors suggest that history of rhinoplasty may be a risk factor, but clearly this is not the case from these figures alone. The authors I believe did realize this shortcoming (missing data) in their conclusions: “…[The] lack of comparison of vascular event following filler injection with those with no history of rhinoplasty.” However, a correlation cannot be made if 2 of the 4 cells in the contingency table are empty. We do not know the number of former rhinoplasty patients who did NOT have a VAE after having a filler injection, nor do we know how many nonoperated patients also had an uneventful filler treatment (ie, the bottom row is missing). If there were, for example, 1000 HA filler patients, all of whom had a positive history of rhinoplasty, and only 7 of these 1000 patients had a VAE, then we have an incidence of 0.7% (7 out of 1000). If on the other hand, if 993 of the 1000 patients had no history of rhinoplasty, and all 7 of the former rhinoplasty patients had VAE, then rhinoplasty is a serious risk factor, with 100% incidence of VAE in former rhinoplasty patients. The missing row can completely change the information one way or the other. This simple exercise of drawing out a simple contingency table will help you spot single cell, single row, and single column data reporting, very common errors, particularly in newspapers and magazines. I see these kinds of errors all the time in sports, business, and medical data reporting.* The authors do not provide any reasons why they can’t provide the missing information from their computerized database, since it would seem to be an easy matter to simply tabulate all their filler patients (to thus get a rate of vascular complications overall by providing the missing denominator) and to also find out how many of their rhinoplasty patients also had fillers without any problems (to thus provide an estimate of the risk of rhinoplasty with respect to dermal fillers). Using this simple fourfold table analysis that can be done on the back of a napkin, anyone can easily manipulate the figures to see the effects of various scenarios. The most important benefit is that anyone can easily spot missing or incomplete data from textual reports, whether they are in medical journals or in a newspaper. This is a very powerful, but simple technique and I encourage you to get in the habit of scribbling down these little tables in the margins of any textual data reports so that you will be able to spot these errors too. Despite the fact that this information does not support (or deny) the null hypothesis, I feel that the author’s premise is probably correct based on clinical experience. In my opinion, the nose is a high-risk area for injection even without a history of surgery (I think there are high risk and low risk areas, but there are no “safe areas” to inject).3 Partly, this is due to the rich vascularity of the nasal tissues and the fact that the vascular plane is immediately adjacent to the typical plane of augmentation. Because of the typical absence of fatty tissues in the nose, it is a “target rich environment.” If our perverse goal was to cannulate a blood vessel on purpose, the nose would be an excellent target zone. We are also in a sort of double jeopardy with respect to the visual axis, because of the many collateral vessels in the nasal regions that are continuous with the terminal branches of the ophthalmic artery as it enters the face (including the dorsal nasal artery in particular). Surgery (or any type of scarring) does seem to increase the risk of VAE clinically.4 Together, there may be an interaction, where the effects of both together might be worse than either alone but this paper does not help elucidate this either unfortunately. Why might surgery increase risk? As the authors state, the pattern of vascular supply is altered after rhinoplasty by division of several of normal vascular pathways (ie, the vascular anatomy is permanently changed after surgery). Therefore, arteries are not in their usual sizes and arrangements because of the adaptations of the cutaneous circulation as the result of the division of certain vessels (the authors mention the columellar and lateral nasal arteries as examples of vessels commonly divided or injured during rhinoplasty). The second reason why surgery may increase risk is because of tissue scarring. Normal, unscarred tissues allow elastic vessels to more easily roll out of the way of oncoming needles and cannulae. In contrast, a scarred vascular bed tends to hold blood vessels rigidly in place, making it easier to accidentally skewer them. Anyone who has attempted to do an arterial puncture will certainly attest that it is easier to perform when the artery is held firmly in place. And it is hypothesized that the end of the needle need not be inside the lumen, since filler can back-track along the needle path and then enter the arterial lumen.3 VAE occurring in the nose, whether operated or not, seem to be more resistant to treatment (based on clinical experience with these). They often take larger amounts of hyaluronidase (HYAL) to restore blood flow and seem to take longer to reverse. These clinical observations may be unreliable or even incorrect, so please keep that in mind (since by definition, these are not experimental groups and there are no control groups). I’m always cognizant that I may be completely wrong − maybe it doesn’t really take more HYAL, and my clinical impressions are inaccurate because of a skewed case population (I perhaps don’t get involved with any of the easier cases). It is unfortunate that so many injectors are reluctant to use sufficient quantities of hyaluronidase for a sufficient number of treatment cycles, resulting in many patients needlessly suffering with skin loss, scarring, permanent textural changes, hyperpigmentation, and the other sequelae of untreated or undertreated VAE. If it true that the nose is a more difficult area to treat, what are the possible reasons for this? One possibility is because some of the filler HA gel gets into vessels which are effectively shielded from HYAL by bone and cartilage, so that we must contend with the more complicated diffusion dynamics. The HYAL must first diffuse through the vessel wall, then along a static column of blood in the blocked artery, then the HYAL can finally begin to break down the gel “wall” facing the lumen of the artery. At the same time, the HYAL is diffusing out of the artery reaching equilibrium, and if it is in a bony or cartilaginous tunnel, it can then diffuse along the outside of the vessel within the bony canal, and eventually back into the vessel to begin to degrade the HA plug further down. This is a complex environment where many things are happening at once, made even more complicated by the fact that some tissues are dying from hypoxia. Because these vessels are typically rather small, even a modest bolus of filler can track a long way inside the vessels. These are just educated guesses. Another higher risk type of filler injection is the large depot techniques which unfortunately seem to be coming back into fashion. In these techniques, 0.1 to 0.5 mL is deposited in a single area to create a particular volume effect at very specific anatomic locations that are purported to have a low risk. In my opinion, all living tissues will have accessible blood vessels either in them or near them, so there really are no “safe zones” for injection. Aspiration prior to injection is often demonstrated at teaching sessions as good practice, but in my opinion, this is somewhat misleading.5,6 Needle aspiration alone prior to filler injection is not a particularly good indicator that it is safe to inject (because this test is only useful if it is positive, [ie, you get a blood flashback], so that you know you are in a blood vessel, a negative result does not really give you any actionable information). In summary, although this paper supports my own clinical impressions from years in the field as a consultant for two major filler companies in Canada (where in any given week I may be assisting several end users across the country with concurrent major vascular complications), they unfortunately did not provide any good evidence. Although I agree with the conclusions of this paper, I hope the authors will provide a follow-up report which includes the missing data. It seems so simple that they could provide the overall incidence of their VAE cases, as well as to report the number of former rhinoplasty patients who had a successful nasal filler treatment without any adverse events, so as to put a real data-point in the literature. These kinds of figures are typically missing from the literature because so many cases of severe VAE go unreported. At almost every meeting where I give a talk on this complication, I have several attendees who relate yet more cases that were never officially reported. It seems likely that there may be hundreds or maybe thousands of these events across the world that we simply never hear about. Although no one wants more official forms to fill out, it is my hope that regulators will enforce mandatory reporting requirements, allowing for solid numbers to be developed. Only in this manner will we be able to give patients the information they need to give truly informed consent. Disclosures Dr DeLorenzi is Medical Director for Merz Pharma Canada Ltd. (Burlington, ON, Canada) and Allergan Canada, Inc. (Unionville, ON, Canada). In both roles, his main role to assist end users with complications arising from the use of the filler products produced by the respective companies. Funding The author received no financial support for the research, authorship, and publication of this article. REFERENCES 1. Robati RM, Moeineddin F, Almasi-Nasrabadi M. The risk of skin necrosis following hyaluronic acid filler injection in patients with a history of cosmetic rhinoplasty. Aesthet Surg J . 2018. doi: 10.1093/asjour/sjy005. 2. Rzany B, DeLorenzi C. Understanding, avoiding, and managing severe filler complications. Plast Reconstr Surg . 2015; 136( 5 Suppl): 196S- 203S. Google Scholar CrossRef Search ADS PubMed 3. DeLorenzi C. New high dose pulsed hyaluronidase protocol for hyaluronic acid filler vascular adverse events. Aesthet Surg J . 2017; 37( 7): 814- 825. Google Scholar CrossRef Search ADS PubMed 4. DeLorenzi C. Complications of injectable fillers, part 2: vascular complications. Aesthet Surg J . 2014; 34( 4): 584- 600. Google Scholar CrossRef Search ADS PubMed 5. Casabona G. Blood aspiration test for cosmetic fillers to prevent accidental intravascular injection in the face. Dermatol Surg . 2015; 41( 7): 841- 847. Google Scholar CrossRef Search ADS PubMed 6. Carey W, Weinkle S. Retraction of the plunger on a syringe of hyaluronic acid before injection: are we safe? Dermatol Surg . 2015; 41( Suppl 1): S340- S346. Google Scholar CrossRef Search ADS PubMed Footnotes * Some simplified examples of single cell reporting: “2 infants on this medication died.” In our example, perhaps the infant mortality without the medication was 100%, so the medication in fact saved 98 out of 100 infants, but the way the data appears in text, it looks as though the medication took two young lives rather than saving 98. It’s a good habit to look for missing numerators or denominators whenever these statements are made. A single row data example might be “2 out of 100 infants on the medication died.” Unless the author also gives data on how many infants in total were being observed, we don’t know how many infants without the medication died. For example, if there were 1000 infants, and only 100 got the medication, and the other 900 all died, then the medication in fact was very beneficial. This is particularly egregious because lawmakers and regulators might make incorrect inferences from these kinds of sloppy reports and make bad decisions. © 2018 The American Society for Aesthetic Plastic Surgery, Inc. Reprints and permission: firstname.lastname@example.org This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Aesthetic Surgery Journal – Oxford University Press
Published: May 18, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera