Radiolucent restorative materials

Radiolucent restorative materials are fillings, bases, liners and cements that are less radiopaque than the tooth structure around them, so they read as a radiolucency on the radiograph. Placed at a cavity margin, one can be misread as secondary caries — a mimic that is built into the tooth rather than growing in it.

Synonyms: radiolucent base; non-radiopaque composite; glass ionomer radiolucency; liner mimicking caries

What is a radiolucent restorative material?

It is any restorative that fails to stand out from the tooth on a radiograph. Sometimes that means the material is genuinely dark against everything around it. Often it means something subtler and more dangerous: the material sits between dentine and enamel [1] — more radiopaque than the dentine beside it, less radiopaque than the enamel above it. So it reads as sound tooth in one place and as a lesion in another.

Radiopacity is what lets you read a restoration at all. Adequate radiopacity is what allows a clinician to diagnose secondary caries and marginal defects, which sit most often at the proximal gingival margin [1]. Take it away and the margin stops being a line you can trust.

Which restorative materials are radiolucent?

Radiopacity varies enormously between classes, and the variation is not where you would guess.

Glass ionomers. The radiopacity of glass ionomer-based materials is highly variable [2]. Conventional glass ionomers do not have sufficient radiopacity on their own, which is why radiopaque fillers are added to compensate [2]. In one comparison, resin-based bulk-fill restoratives were significantly more radiopaque than glass ionomer-based ones [3].

Flowable composites. These vary considerably, and that matters more than it sounds, because the flowable is usually the material that goes in first at the gingival floor. In one study of twenty resin composites for posterior use, four of six flowable composites had radiopacity values that fell between dentine and enamel [4]. In another, the lowest-reading flowables sat close to 1 mm of aluminium [5] — which, as the next section explains, is roughly dentine level.

Composites generally. In a study of 56 restorative materials, all were more radiopaque than dentine, but 4 of 33 conventional composites and 3 of 16 flowable composites were significantly less radiopaque than enamel [1]. An earlier study of 21 direct esthetic materials found all but one microfilled resin composite were more radiopaque than dentine and met the ISO 4049 standard [6].

Polymer-based CAD/CAM blocks. Polymer-infiltrated ceramics have been described as fairly radiolucent, with low radiopacity despite radiopacifying elements and high filler content [7]. In a study of six CAD/CAM materials, only three exceeded the ISO minimum, and two polymer-based blocks measured significantly below both dentine and 1 mm of aluminium [8]. Where a block reads lower than dentine, detection of caries around it depends on the radiopacity of the luting cement instead [9].

Cements. The range here is extreme. Zinc phosphate measured 6.4 mm aluminium equivalent while an autopolymerising resin cement measured 1.1 mm [10]. One implant cement could not be detected radiographically at either 1 mm or 2 mm thickness [11] — not faint, absent.

All of these figures come from in vitro work: flat discs of uniform thickness radiographed beside an aluminium step wedge, not restorations sitting in a tooth. Read them as a ranking of materials, not as a prediction of what any individual restoration will look like on a bitewing.

What it looks like

The classic presentation is a radiolucent band at the gingival floor of a proximal box, under an otherwise unremarkable restoration. That is where the first increment of material goes, and it is exactly where secondary caries is looked for — materials with radiopacity lower than enamel might be misinterpreted as secondary enamel caries, especially when applied as the initial increment on the proximal gingival margin [1]. A base is the same problem one layer deeper: if a material less radiopaque than dentine is used as a base, it could be mistaken for carious or decalcified dentine [2].

The corollary is that the first layer placed must be radiopaque enough for the restoration margin to be read at all [2]. When it isn't, you are not looking at a margin. You are looking at the place where a margin should be.

What it's confused with

Being straight about this: the evidence here is about how a material's radiopacity affects whether secondary caries gets detected at all, rather than about a single sign that separates material from lesion on the image. The materials studies establish that the confusion happens and where it happens. So treat any rule you have been taught for telling the two apart as clinical convention unless you can point at the study behind it.

Cervical burnout, residual caries and restoration overhang belong on the same list of things that complicate reading a restored proximal surface. One older review groups restorative materials and sub-bases with cervical burnout and the Mach band effect as factors that influence the radiographic interpretation of caries [12].

Why it happens: what the radiopacity standard actually requires

This is where most people's mental model is wrong.

According to ISO 4049 and ISO 9917, for a manufacturer to claim a product is radiopaque, the material must have radiopacity equal to or greater than that of the same thickness of aluminium [2]. The benchmark is aluminium. Not enamel.

That distinction is the whole entry, because at equal thickness the radiopacity of dentine is almost equal to that of aluminium, and the radiopacity of enamel is almost twice that of aluminium [2]. So a material can pass the standard, be honestly labelled radiopaque, and still be nowhere near the enamel it sits against.

The materials literature reflects this. Most tested materials clear the bar — but "clears the bar" and "reads clearly against enamel" are different statements, and it is the second one that matters at a proximal margin. For the highest accuracy in radiographic diagnosis of secondary caries, the restorative material's radiopacity should be slightly higher than that of enamel [2].

Two things worth knowing:

• The published figures only look inconsistent. Each study reports the threshold at its own specimen thickness: 2.0 mm of aluminium for a 2 mm specimen [4], 1 mm of aluminium for the 1 mm disks used in testing [13]. Both are the same same-thickness rule stated by [2]. There is no dispute about what the standard requires — only a set of studies each quoting it at the thickness they tested.
• Declared values are not always right. Measured radiopacity differed from the manufacturer's declared value by up to 1.53 mm equivalent aluminium for some materials [1].

What it means clinically

It means one class of false positive is manufactured at placement rather than found at review, and it means the restoration record is doing work the image can't.

It also means the ceiling on radiographic detection here is lower than most people assume. In a systematic review and meta-analysis of secondary caries detection, visual, radiographic and laser fluorescence detection had similar sensitivities, between 0.50 and 0.59, and similar specificities, between 0.78 and 0.83 [14]. Those figures come mostly from in vitro studies on permanent posterior teeth, and the review itself notes that the included studies had high risk of bias or applicability concerns [14]. A separate systematic review found that standardised diagnostic criteria, lesion activity assessment and differential diagnosis were described by only a limited number of studies [15].

So: somewhere between four and five in ten secondary lesions missed in those studies, and a specificity that leaves real room for calling something that isn't there. A radiolucent base sits directly in that room.

How to be sure

No radiographic manoeuvre resolves this on its own. What the evidence does support is refusing to decide from the radiograph alone.

Correlate with the clinical examination. A meta-analysis of radiographic optical effects — the Mach band effect and triangular-shaped radiolucency, not restorative materials — found they produced false-positive diagnoses of caries or fracture in roughly 13% of observations, and concluded that clinicians should interpret radiographic findings with caution and always correlate them with a thorough clinical examination [16]. That review graded its own evidence low to very low certainty [16]. The general point holds regardless: a radiolucency with a plausible non-carious explanation is not a diagnosis.

Don't jump to replacement. The revised FDI criteria link each score to a management strategy rather than a binary — reviewing or monitoring, refurbishment or reseal, repair, and replacement [17]. In a two-year clinical evaluation of 88 restorations in 40 patients, the authors concluded that composite restorations with less-than-ideal marginal adaptation and stained margins are better off repaired than replaced [18] — though the significant comparisons in that study were against leaving them untreated, not repair against replacement. That study was about margins and staining rather than radiolucent materials, so read it as a general argument for the conservative option, not as a rule for this situation.

Prevent it next time. This is the only caries mimic on the bitewing that can be designed out. Choosing a first increment with radiopacity slightly above enamel [2] removes the ambiguity from every radiograph that tooth will ever have taken.

Common errors

• Treating "ISO compliant" as "readable against enamel." The standard's reference is aluminium, which sits at about dentine level [2]. Compliance says nothing about contrast with enamel.
• Trusting the manufacturer's number. Declared and measured values have differed by up to 1.53 mm equivalent aluminium [1].
• Reading in vitro millimetres as chairside performance. These are disc specimens on a step wedge.
• Forgetting the opposite failure. Very radiopaque materials cause the other error — masking adjacent lesions by superimposition [2]. A clean-looking margin under amalgam is not the same evidence as a clean-looking margin under enamel-matched composite.
• Calling it from one image. Across visual, radiographic and laser-fluorescence detection, sensitivity for secondary caries clustered between 0.50 and 0.59 [14].