Is 30 Ncm Enough Implant Torque?
What You Need to Know About Full-Arch Insertion Torque.
What torque value do we need to load our implants for a full arch prosthesis in the Maxilla? Wait! Before you close this because of how boring this question is, let’s ask — why does this seem so obvious? I suppose because the question has been asked so many times. But here’s another question: what does torque have to do with implants properly integrating? More than this: what is torque, anyway? We’ve gotten so used to referring to “insertion torque” we may not even remember how or why we started thinking in this way. So back to basics we go, then on to some data on how it all relates to full arch treatment.
Torque is just the way of talking about force when we’re rotating something instead of pushing it in a straight line. In the case of bolted joints (where the concept of torque is most useful in assembly and manufacturing) the torque value correlates to a particular clamping force that will remain after the joint is bolted. Good to know when one is putting something together that you wish to remain together under certain stressful conditions. This resultant clamping is due to the bolt itself being elastically stretched as the torque applied turns it and pulls it by means of the screw threads, then the bolt trying to physically snap back to its original size, thereby squeezing on the joint. One might think of a rubber band being pulled over two objects to hold them together — with the bolt acting as a kind of internal rubber band in this analogy.
In the case of the surgical placement of dental implants, insertion torque is a measurement of how hard you have to push to get the implant to keep turning and go further into the bone. The implant itself is not being elastically stretched because it is quite a bit less squishy than the bone (here the relevant physical property is the elastic modulus, which is a measure of how much force you need to bend or flex something, often represented as “λ”). So insertion torque is really just an indirect measure of how dense the bone is at the insertion site.
Why indirect? Because it’s not that simple (it’s NEVER that simple). Torque in a simple system (think of just a bolt clamping a couple of hunks of metal together) is related to the diameter of the bolt, the coefficient of friction, and the desired clamping force. Using our common implant units this would be:
Torque (Ncm) = Implant diameter (in cm) x loading force (Newtons) x μ (coefficient of friction — this has no units)
We’re immediately confounded by the influence of implant diameter. All other things being equal (they NEVER actually are), going from a 4mm to a 5mm diameter implant would increase insertion torque by 25%. Will the same bone heal better around a wider implant? Maybe. It will certainly torque better. Where’s bone density in that equation? It’s in there… hidden as a part of the coefficient of friction. Denser bone — higher coefficient of friction. Further, the clinical bone density is not uniform. We know in general that bone density varies by jaw region (from densest to least dense — anterior MAND, posterior MAND, anterior MAX, posterior MAX) but there is also wide variation within the individual implant site. It has been proposed that as much as 95% of primary implant stability is derived from the cortical layer of bone. Meaning that all the length into the trabecular bone is doing very little in terms of torque achieved. This is of particular concern in our full arch cases, because the appropriate bony reduction for restorative space can obliterate the cortex altogether.
Even still there’s much more to it. Surface roughness of the implant as well as thread pitch, taper, and implant macro design will all influence the coefficient of friction and resulting torque value. Beyond this there are the secondary, behavioral effects on implant companies and clinicians to consider. Once it became clear that research and clinical practice were using torque as a proxy for implant prognosis, all players in the market took notice and worked on design features to maximize torque achieved (within limits). It is perfectly natural and good that implant manufacturers responded to academic work and attempted to optimize their product offerings for excellent clinical outcomes.
But we’re back to begging the question now: i.e., we’re assuming the proxy (torque) is truly representative of the desired outcome (implant prognosis). Does that make sense? Is it true? [N.B. these are two distinct questions: lots of true things do not make sense to us and lots of things that make sense are not true.] It kind of makes sense. It literally feels more stable to place an implant into denser bone with a higher torque. After all, bolting a joint with a higher torque produces a higher clamping force. But we’re not clamping anything here… we’re asking the body to heal around this implant in a certain way (bone not connective tissue, please). How does the torque affect the healing of an implant? We’ll go into the details and theories about implant healing in another post, but for now, for the case of full-arch prostheses in the maxilla, we’ll look at some real data to see how important torque is.
The Research
The study we’re looking at comes from the Maló clinic, where Dr. Paolo Maló and his colleagues have been performing exceptional treatment and independent research for many years. One of the joys of scientific writing is precision in Title/headline writing: so this one is called “Immediate function dental implants inserted with less than 30N·cm of torque in full-arch maxillary rehabilitations using the All-on-4 concept: retrospective study.” Not exactly clickbait, but accurate. Dr. Maló’s clinic does a high volume of full-arch cases (he did INVENT the All-on-4 so, you know..), so gathering retrospective data (all warnings apply) from his clinic can yield some valuable insight.
The study included 83 patients having received 332 implants as part of full-arch treatment in the maxilla, with the specific inclusion criteria that at least one of the implants was placed with a final insertion torque < 30Ncm. Each patient was followed up at 1 year to measure implant success and marginal bone loss. The implants placed with < 30Ncm showed a 98.3% success rate, compared to the implants placed with ≥ 30Ncm at a 97.5% success rate. This difference was not statistically significant. The bone loss for the ❤0Ncm group was 1.14 ± 0.38 mm, compared to 1.39 ± 0.49 mm for the ≥ 30Ncm group, and this difference was statistically significant (though not clinically, one would really have to squint to see that quarter millimeter under the final full-arch prosthesis). One might object that any of the implants in the ❤0Ncm group were stabilized through the prosthesis by several implants in the ≥ 30Ncm group, so the authors kindly included the overall distributions by patient:
One implant < 30Ncm; three implants ≥ 30Ncm — 58 patients (69.9%)
Two implants < 30Ncm; two implants ≥ 30Ncm — 18 patients (21.7%)
Three implants < 30Ncm; one implant ≥ 30Ncm — 2 patients (2.4%)
Four implants < 30Ncm — 5 patients (6.0%)
Another distinction of interest is the location of the implants within the arch, as force analysis shows different stresses placed on posterior vs. anterior implants in a full-arch restoration. Theoretically we might have some more concern with the posterior because the biomechanical forces are greater in the posterior, and the mechanical stabilization from the prosthesis and other implants is only coming from one side. The data here show in the < 30Ncm group, one implant failed in the right posterior and one in the left anterior. In the ≥ 30Ncm group, two implants failed in the right anterior and one in the left anterior. The gross number of failures is too low to draw statistically significant conclusions, but there is at least no immediately obvious problem with either anterior or posterior implants in the < 30Ncm group.
So in Dr. Maló’s clinic, it seems there is no discernable difference or problem in utilizing implants placed with under 30Ncm of torque for maxillary full-arch restorations. A study like this with only one site has questions about generalizability. After all, maybe they’re just very good at this (spoiler: they are), maybe their conversion and provisional protocol is exceptionally gentle and accurate, maybe they do something we don’t even know about or understand. No question, multi-center data would be more generalizable, still there’s no need to discount these findings. There were clear inclusion/exclusion criteria, less than 20% of patients lost to follow up, and well defined measurements, so the internal validity of the data is high.
So What?
What are the implications of this for our practices? It seems we can feel confident with lower insertion torques for full arch cases. How low? The data reported here don’t give us any indication, but there are some practical clinical limitations. Many multi-unit abutments have final torques at 20Ncm, so it would be nice to achieve that so the abutments can be torqued at the time of surgery. If we get too far below 20Ncm, attaching anything (multi-unit abutment, impression coping, scan body, temp cylinder) may move the implants and make provisionalization difficult. One very irritating scenario (one this author has experienced, sadly) that can occur is the attempted removal of something like a scan body results in removal of the whole scan body+multi-unit abutment+implant. Perhaps the most important implication is the most obvious; we should boldly load full-arch cases even when torque values are low. It may seem counter-intuitive, but until evidence is presented to the contrary, it’s what’s best for our patients.
