A randomized clinical trial published in BDJ Open on 3 November 2026 found that Class II composites bonded with a radiopaque universal adhesive showed no visible radiolucent line at the tooth-restoration interface on immediate periapical radiographs, while 16.2% of those bonded with a radiolucent adhesive did. The authors stress that this is a difference in how restorations look on an X-ray, not evidence of better bonding or longer survival.
A thin dark line under a proximal composite is one of the more frustrating findings on a recall radiograph. It might be recurrent caries. It might be a marginal gap or a void. Or it might be nothing more than the adhesive layer itself, if the adhesive is radiolucent. As of November 2026, a new trial from Cairo University compared how often that line appears right after placement with a radiopaque versus a radiolucent universal adhesive, and the result matters for anyone deciding whether to replace a restoration based on what the image shows.
What the Trial Tested
The study, by Omar Shaalan and Dina Kamal of the Conservative Dentistry Department at Cairo University's Faculty of Dentistry, appeared in BDJ Open (volume 12). It compared two universal adhesives from the same manufacturer:
- Intervention: Scotchbond Universal Plus (3M ESPE), described as radiopaque.
- Control: Single Bond Universal (3M ESPE), described as radiolucent.
According to the authors, Scotchbond Universal Plus was developed from the chemistry of its predecessor, Single Bond Universal, with a radiopaque brominated dimethacrylate monomer added to the adhesive resin matrix. That is a different route to radiopacity from the traditional one. The paper notes that adhesives have usually been made radiopaque with inorganic fillers such as barium-containing glass, ytterbium trifluoride or zirconium oxide, and that these fillers can affect viscosity, film thickness, filler sedimentation and handling if they are not formulated well.
Who was treated, and how
Seventy-four participants with Class II cavities were randomized into two groups of 37, with one restoration evaluated per participant. Lesions were small to moderate proximal carious lesions scored ICDAS 3 or 4 in vital teeth. Deep lesions within 1 mm of the pulp, teeth with signs of irreversible pulpitis and endodontically treated teeth were excluded. The trial ran in the university's conservative dentistry clinics from October 2025 to March 2026 and was prospectively registered (NCT05907928).
The clinical protocol was tightly standardized:
- Rubber dam isolation and a sectional matrix system for the proximal wall.
- Selective enamel etching with 32% phosphoric acid for 15 seconds.
- Adhesive actively rubbed in with a microbrush for 20 seconds, air-thinned for 5 seconds and light-cured for 10 seconds.
- A nanofilled composite (Filtek Z350XT) placed with a centripetal technique in oblique 2 mm increments, each cured for 20 seconds.
A single operator with more than 15 years of restorative experience placed every restoration. Participants, the two radiographic assessors and the statistician were blinded to group. The operator was not, because the two adhesives handle and look different.
How the outcome was measured
Immediately after placement, in the same appointment, the team took periapical radiographs with a size 2 digital sensor and an extension cone paralleling (XCP) positioning device. Two trained, blinded assessors scored each interface with modified United States Public Health Service (USPHS) criteria: Alpha meant no visible radiolucent line at the tooth-restoration interface, Bravo meant a line was visible. Agreement between the assessors was reported as almost perfect (kappa 0.90).
What the Researchers Found
Every one of the 37 restorations bonded with the radiopaque adhesive scored Alpha. In the radiolucent adhesive group, 31 of 37 (83.8%) scored Alpha and 6 (16.2%) scored Bravo. The difference was statistically significant (P = 0.025), and the authors report an absolute risk reduction of 16.2% for seeing a radiolucent interfacial line (95% CI 3.71% to 28.73%). No adverse events were observed.
To the authors' knowledge, this is the first randomized clinical trial of the immediate radiographic appearance of proximal composites bonded with radiopaque versus radiolucent universal adhesives. Earlier work on the question was mostly laboratory-based.
Why a Dark Line Matters at the Recall Visit
The paper frames the problem around overtreatment. When a radiolucent area shows up beneath a composite, telling apart secondary caries, a marginal gap, a void and a radiolucent adhesive layer can be hard, and the authors note that a definitive diagnosis often comes only after the restoration is removed. If the shadow is only the adhesive, the restoration may be replaced unnecessarily, at the cost of healthy tooth structure.
The authors cite earlier research in which dentists were roughly twice as likely to recommend replacement when they saw a radiolucency under a composite, even when that radiolucency came only from a thick adhesive layer. The authors suggest a radiopaque adhesive may make that interpretation easier by making a visible adhesive line less likely.
What the Trial Does Not Show
The authors are unusually direct about the limits of their result, and the limits are as useful as the headline number.
- It is a picture, not a performance result. The authors write that Bravo scores should not automatically be read as defective marginal adaptation, secondary caries, restoration failure or weaker bonding. The difference between the groups is a difference in radiographic appearance only.
- No follow-up. Radiographs were taken immediately after placement. The trial says nothing about interface degradation, marginal breakdown, recurrent caries or how long the restorations last.
- Periapicals, not bitewings. The team used standardized periapical radiographs. They acknowledge that bitewings are generally considered more appropriate for assessing proximal margins and may be more sensitive to small interfacial defects.
- No clinical scoring. There was no concurrent clinical USPHS assessment of the restorations.
- One product of each type. Only one radiopaque and one radiolucent universal adhesive were tested, so the authors caution against generalizing to every universal adhesive.
- The mechanism is a hypothesis. The authors suggest viscosity, pooling and film thickness may explain part of the difference, but they did not measure film thickness or interface morphology, and they describe this as a proposed explanation rather than a tested one.
The setup also matters when you read the result against your own operatory. One experienced operator, rubber dam, sectional matrices and a fixed exposure protocol produce cleaner conditions than a busy recall day. The sample size was planned around an earlier study that reported radiolucent interfacial lines under about 16% of proximal composites, which is close to what the control group showed here.
What This Means for Practices in Toronto and Across Canada
For a restorative team in Toronto, Mississauga or anywhere else in Ontario, the practical value of this trial sits in diagnosis more than in purchasing. A few practical suggestions follow (ours, not the authors'):
Pro Tip: When you place a proximal composite, note the adhesive brand in the chart entry. If a dark line appears at a recall visit 12 or 24 months later, knowing whether a radiolucent adhesive was used is one more piece of evidence before you decide to replace.
Pro Tip: A postoperative radiograph taken right after placement, as in this trial, gives you a baseline for later films. A line on that baseline image is still worth checking clinically at the time, since the paper notes it could also be a void or a marginal gap.
The authors' own caution is the right frame: radiographs should always be interpreted together with the clinical examination. A radiopaque adhesive may make that interpretation easier. It does not replace probing, visual inspection or clinical judgment.
The paper does not address regulatory status or availability in Canada. If you are considering any product named in a study, confirm its Health Canada medical device licensing status and its current Canadian availability with your supplier before you change your protocol. Readers who want to compare restorative supply options can browse EBIKO Dental.
EBIKO Dental will continue monitoring restorative research, including any longer-term follow-up that tests whether these radiographic differences change clinical decisions over time.
Sources
- Shaalan O, Kamal D. Radiographic appearance of proximal resin composite restorations bonded with a radiopaque or radiolucent universal adhesive: a randomized clinical trial. BDJ Open. 2026;12:105. doi:10.1038/s41405-026-00494-1 (Europe PMC full text)
- Trial registration: NCT05907928
Frequently Asked Questions
Q: Does a radiopaque universal adhesive make composite restorations last longer?
This trial cannot answer that. It measured only the radiographic appearance of the interface immediately after placement, and the authors state that the result should not be read as evidence of better clinical performance or restoration longevity.
Q: What does a radiolucent line under a new composite mean?
According to the paper, a radiolucent area under a composite may reflect secondary caries, a marginal gap, a void or a radiolucent adhesive layer. In this trial, lines seen right after placement in the radiolucent adhesive group were not treated as proof of a defect.
Q: How big was the difference between the two adhesives?
None of the 37 restorations bonded with the radiopaque adhesive showed a visible line, compared with 6 of 37 (16.2%) in the radiolucent adhesive group. The authors report an absolute risk reduction of 16.2% (95% CI 3.71% to 28.73%), P = 0.025.