Night Guards: Hard, Soft or Dual-Laminate, and Upper or Lower?

For most adult bruxers a hard, full-coverage night guard is the best-supported choice; a dual-laminate guard, soft inside and hard outside, is the comfort option for a patient who will not wear a hard one, and a soft guard suits short-term use. Upper or lower matters less than what the guard has to protect.

A patient shows up with fresh wear facets and asks for a night guard, or you have just seated new zirconia on a known grinder. The Rx then asks you to decide: hard, soft or dual-laminate, upper or lower, full coverage or anterior only, how thick, and what guidance.

The evidence on night guards is thinner than their popularity suggests, and some designs carry real risks. Here is what guards can do, how options compare, and what to send. For crowns and implants in grinders, see Restorations for Bruxers.

Quick Answer

For most adult bruxers, a hard, full-coverage guard with even contacts on a flat plane is the best-supported default; the one trial comparing upper and lower splints found no difference in pain. Soft guards raised muscle activity in some patients, and thermoformed materials wore fastest in lab tests. Dual-laminate guards add a soft inner layer for comfort, with little direct evidence either way. Avoid unsupervised anterior-only devices, and ask about snoring before making any guard.

What Can a Night Guard Actually Do?

The 2018 international consensus defines sleep bruxism as masticatory muscle activity during sleep and, in otherwise healthy people, a behavior rather than a disorder. A guard does not reliably change that behavior; it puts a replaceable surface between the arches.

For sleep bruxism: protection, not a cure

A Cochrane review of 5 randomized trials found not enough evidence that occlusal splints treat sleep bruxism, though they might help with tooth wear. A 2020 NIHR review found no trials that measured tooth wear in bruxers, so that benefit remains unproven. Protecting teeth and restorations is a fair goal; stopping grinding is not a promise to make.

For TMD pain: modest and uncertain

The 2020 review pooled 52 trials and found no evidence that splints reduced TMD pain compared with no or minimal treatment. A 2024 Cochrane review of 57 trials rated all its evidence very low certainty. For chronic TMD pain, the 2023 BMJ guideline made a conditional recommendation against reversible occlusal splints and strong recommendations for CBT, supervised jaw exercise and usual care. In a 12-month trial of 200 TMD patients, neither a flat-plane hard acrylic splint nor a soft vinyl splint beat self-care.

Hard, Soft or Dual-Laminate: Which Material?

Hard acrylic

In a 1987 overnight study, a hard splint covering all maxillary teeth with even, simultaneous contacts significantly reduced nighttime muscle activity in 8 of 10 participants. At maximum clench, hard splints lowered masseter and anterior temporalis activity, while soft splints slightly increased it.

Soft

Soft guards are often chosen for comfort, and for TMD symptoms soft vinyl matched hard acrylic at 12 months in the trial above. Not every EMG study agrees: in 16 bruxers clenching after about 6 weeks of wear, activity fell in 6 of 8 soft-splint users and rose in 5 of 8 hard-splint users, yet the authors still judged hard splints better. All these samples are tiny.

Dual-laminate

A dual-laminate guard is soft inside and hard outside. In a Belfast clinic study of 100 migraine-prevention appliances, a laminated design of ethylene vinyl acetate with a light-cured resin needed less adjustment to seat than heat-cured acrylic splints, with similar chair time when changes were needed. We found no trial comparing dual-laminate and hard guards in bruxers.

Full Coverage or Anterior Only?

Anterior-only devices contact a few front teeth and keep the posterior teeth apart. In a review of the NTI-tss, an anterior bite stop, two trials found lower EMG activity than with a stabilization splint, but complications related mainly to single teeth or the occlusion, and the authors advise it only for patients certain to attend follow-up. Full coverage is not risk-free: the review notes rare occlusal changes after full-time wear of a stabilization appliance, and calls them unlikely with night-only wear.

Upper or Lower?

The one randomized comparison we found was in TMD, not bruxism: in 40 patients with myofascial pain, neither an upper Michigan splint nor a mandibular splint reduced pain significantly over 6 months, and muscle activity did not differ. A finite element model built from one patient's CT and MRI data suggested the same arch as the teeth or implants at issue. In practice, choose by what needs protecting, where missing teeth would leave gaps in the contacts, and what the patient will wear.

How Thick, and What Guidance?

In 12 people with a bruxism habit, a splint 3 mm thick at the central incisors lowered nighttime masseter and temporalis activity; a 6 mm splint had no significant effect, and 6 of 12 (masseter) and 7 of 12 (temporalis) did worse with it than with no splint. The finite element model above also favored 2 to 3 mm of anterior thickness over 3 to 4 mm.

On guidance, two small EMG studies in people without jaw dysfunction disagree. On full-coverage splints in 6 subjects, canine guidance reduced elevator muscle activity in excursions more than group function; in 10 subjects, masseter activity did not differ between the two designs. Canine guidance on a flat occlusal plane is a reasonable default, not a proven requirement.

Printed, Milled, Heat-Cured or Thermoformed?

In lab wear tests, the method matters less than whether the material is thermoformed. Clinically, digital guards have shown comfort and fewer visits, not better long-term relief.

From Scan to Guard: The Digital Workflow

A digital guard starts the same way as a crown: with good scans. The steps, and what each one asks of you:

1. Scan both arches. Full arches with every occlusal and incisal surface complete, plus a bite scan on both sides. Holes on a cusp tip become gaps in the guard's fit; our scanning tips cover the common errors.
2. Record the jaw position if it matters. A plain bite scan shows the teeth together; if you want the guard designed at a set opening or position, record it and say so on the Rx (see Digital Bite Registration).
3. Design. The technician sets coverage, thickness at the incisors, a flat plane and the guidance from your Rx on the scanned models.
4. Make. A hard guard can be printed from the design; soft and dual-laminate guards are thermoformed from sheet material over a model, which is printed from the scan.
5. Finish and check. A printed guard is cleaned and cured under light after printing, then the fit and contacts are checked on the models.

The curing step is not a formality. In a 2025 lab study of 180 specimens of a printed splint resin, both print orientation and post-curing time significantly changed its strength and surface hardness, so printed guards are finished to the resin maker's protocol.

How accurate are digital guards? A 2026 lab study made 40 identical splints from one design file: printed splints deviated from the file by 43.9 µm on average and milled ones by 22.0 µm, both in line with earlier studies of digital splints, with the larger printed differences near the back molars. In a 2026 retrospective study of 49 patients who wore CAD-CAM guards for 6 months, guards designed from intraoral scans did as well, with similar patient satisfaction, as guards designed from scanned stone casts, so an impression still works.

Why Ask About Snoring Before Making a Guard?

A maxillary splint can worsen sleep apnea in some patients. In a pilot study of 10 apnea patients, average AHI did not change significantly, but it rose by more than 50% in 5 of them on the splint night. In a randomized crossover trial of 10 patients, mean AHI was 17.4 events per hour with a stabilization splint and 15.9 without, a significant difference the authors judged small.

The pilot study's authors suggest asking about snoring and sleep apnea whenever you recommend a splint. Diagnosis belongs to a physician: the AASM and AADSM guideline says primary snoring should be diagnosed by a sleep physician, not a dentist, because snoring often accompanies OSA; when a sleep physician prescribes an oral appliance for OSA, it suggests a custom, titratable device made by a qualified dentist. How those appliances are chosen is in Oral Appliances for Sleep Apnea.

What Records Should You Send?

• Arches: full-arch scans or impressions of both arches, occlusal and incisal surfaces complete (connection steps by scanner are in How to Send a Scan to a Dental Lab).
• Bite: a bite record; if you want a specific jaw position or opening, record it and say so.
• Rx: material, arch, coverage, thickness, guidance, restorations and implants, previous guards, snoring screened.

For billing, the ADA says a guard with any hard occlusal component, dual-laminate included, is coded as a hard guard; the guard codes are in our CDT codes cheat sheet for lab work.

How Should Patients Clean a Guard, and When Should It Be Replaced?

Cleaners change guard materials. In a lab test, soaking two thermoplastic sheet materials 15 minutes a day for 60 days in liquid soap, 2% chlorhexidine or a mouthwash increased hardness, and liquid soap was best for color. In a 6-month lab simulation, effervescent denture tablets noticeably changed the color of printed and hand-cast splint resins.

We found no clinical study that sets a replacement interval; the wear review calls for studies on which material lasts longest without a remake. Have patients bring the guard to every recall, and check it for wear-through, cracks, loose fit and new facets.

Night Guards at Elegant Dental Laboratory

• We make orthodontic appliances, including hard, soft and dual-laminate night guards, and anti-snoring and sleep apnea appliances on request.
• Our night guards are printed or thermoformed, not milled. When the Rx names no material, we make a hard/soft (dual-laminate) guard; write "hard" if you want a hard one. A night guard takes about 4 days in the lab.
• Every case is made in house at 2308 McDonald Ave, Brooklyn, by more than 35 technicians. Gary Fingerman founded the lab in 2007.
• Digital scans are reviewed before production; we accept STL and PLY files from all major intraoral scanners.
• Our warranty does not cover restorations in patients with bruxism who do not wear a night guard, so a guard protects the crowns as well as the teeth.
• Our own drivers pick up and deliver free every business day in all five New York City boroughs, Long Island and 13 northern New Jersey counties; anywhere else, use a prepaid UPS label or send a case. Call 877-335-5221.

FAQ

Is a hard or soft night guard better for bruxism?

Hard has the better data: in one overnight study it lowered muscle activity in 8 of 10 bruxers; a soft guard raised it in 5.

Will a night guard stop grinding?

There is no good evidence that it will. Present the guard as protection for teeth and restorations.

Should the guard be upper or lower?

In the one randomized comparison we found, in TMD patients, neither arch reduced pain significantly. Choose by what needs protecting.

Are anterior-only guards safe?

Reported complications involve single teeth and the occlusion, including an anterior open bite. Use them only with reliable follow-up.

Should I screen for sleep apnea first?

Yes. A maxillary splint raised AHI by more than 50% in 5 of 10 apnea patients in one pilot study. Refer snorers to a sleep physician for diagnosis.

What is a dual-laminate night guard?

A guard with a soft inner layer against the teeth and a hard outer layer for the bite. It is a comfort option; we found no trial comparing it with a hard guard in bruxers.

Can a night guard be made from a scan?

Yes. Send full-arch scans of both arches and a bite scan; in a 2026 study, guards designed from intraoral scans did as well as guards from scanned casts.

Study values are from published research; diagnosis of sleep apnea belongs to a physician.