Skip to Content

Peri-implantitis: does radiography underestimate bone defects?

The reproducibility of peri-implant radiographic measurements is a well-established fact for...

The challenge of radiographic precision in implantology

The reproducibility of peri-implant radiographic measurements is a well-established datum for marginal bone level (MBL) and crestal bone level (CBL). However, the increasing use of new parameters—such as emergence angle (EA), emergence profile (EP), as well as defect width (DW), defect height (DH), and defect angle (DA)—raises questions about their actual reliability. While these indicators now guide clinical decision-making in the face of peri-implantitis, their inter- and intra-examiner reliability as well as their correlation with actual surgical dimensions remained, until now, largely undefined.

The objective of this study

This exploratory cross-sectional study aimed to bridge this knowledge gap. The researchers evaluated the reliability of measurements obtained on calibrated periapical radiographs (paralleling technique, Rinn system) for all the cited parameters. Furthermore, the central objective was to compare the radiographic precision of DW and DH measurements against the "ground truth": the dimensions of bone defects recorded directly by trans-surgical probing during flap elevation.

Hypotheses tested

The work aimed to determine whether the reproducibility of measurements varies according to the parameter studied and whether periapical radiographs systematically underestimate the surgical dimensions of defects. The study, conducted by eight periodontology students under the supervision of an expert, sought to generate pilot data to standardise measurement and calibration protocols, thereby replacing uncertainty with a quantified database on the clinical reliability of these parameters.

In practical terms, for the practitioner:

Design and experimental protocol

This cross-sectional observational study (IRB HUM00255314) evaluated the reliability and accuracy of calibrated peri-implant radiographic measurements. The dataset is based on 10 periapical radiographs obtained via the paralleling technique (Extension Cone Paralleling system, Dentsply Rinn) and a GXS-700 digital sensor (KaVo Dental/Gendex). Exposure parameters were standardised at 70 kV, 7 mA, and 0.25 s.

The source population comes from a registered clinical study (HUM00169735), including patients with implants (tissue or bone level) evaluated at least 12 months after loading. The protocol was structured into four phases:

  • Training: Eight periodontology fellows were trained on 5 non-study radiographs in the identification of seven parameters (MBL, CBL, EA, EP, DW, DH, DA) using ImageJ software.
  • Calibration: Each image was normalised according to the known implant length.
  • Measurements: Three sessions spaced 2 weeks apart, blinded, with three repeated measurements per parameter for each image.
  • Surgical validation: The radiographic values of DW and DH were compared with direct measurements performed during surgery (UNC probe, Hu-Friedy) after flap reflection.

The statistical analysis (α = 0.05) used intraclass correlation coefficients (ICC) for reproducibility, Fleiss' κ for categorical variables, and Bland-Altman analysis to compare radiographic measurements with actual surgical measurements.

Performance of radiographic measurements

The study reveals a marked disparity in the reproducibility of radiographic parameters. The emergence angle (EA) and the mesial defect width (DW) show excellent inter-examiner reliability. Conversely, the defect height (DH), defect angle (DA), and distal bone levels show poor reliability, highlighting a strong dependence on the examiner's interpretation.

Parameter Inter-examiner reliability (ICC) Intra-examiner reliability
EA (Emergence Angle) Excellent Excellent
Mesial DW Excellent Good
Distal DW Good (0.739) -
Mesial CBL Fair -

Radiography vs surgery comparison

Periapical radiographs systematically underestimate the actual dimensions of defects observed during surgery. The mean measurement differences are as follows:

  • Defect Height (DH): -2.99 mm (mesial) and -3.52 mm (distal).
  • Defect Width (DW): -0.883 mm (mesial) and -0.463 mm (distal).
Announcement

To equip yourself

Delynov products related to this topic:

Delynov Surgery, your supplier for surgical sutures absorbable and non-absorbable, consumables and implant surgery instruments.

The correlation between radiographic and surgical measurements for DH is moderate but statistically significant (mesial: r = 0.479; distal: r = 0.342; p < 0.001). While the mean difference for mesial DH does not reach statistical significance, other measurements confirm a systematic radiographic underestimation. For DW, the mesial correlation is not significant (r = -0.077; p = 0.400), whereas the distal correlation is significant (r = 0.468; p < 0.001).

Comparison of implant types

The agreement for distinguishing bone-level implants from tissue-level implants is judged good (inter-examiner κ = 0.625; intra-examiner κ = 0.719). However, the evaluation of the emergence profile (EP) remains unreliable (κ < 0.40).

Reliability with variable geometry

This exploratory study highlights a frustrating clinical reality: the accuracy of your radiographic measurements depends heavily on the parameter evaluated. While the emergence angle (EA) and defect width (DW) show excellent reproducibility after calibration, other critical indicators such as defect height (DH), defect angle (DA), and distal bone levels are far too sensitive to examiner interpretation. In the practice, this means that these measurements should not be considered as absolute truths.

The blind spot of 2D imaging

The conclusion is clear: periapical radiographs systematically underestimate actual surgical dimensions. With biases reaching −2.99 mm for mesial DH and up to −3.52 mm for distal DH, 2D imaging encounters its structural limit: the planar projection of 3D anatomy. This study confirms that the correlation between radiography and surgical reality is poor (mesial DH r = 0.479; distal r = 0.342) and sometimes non-existent, particularly for mesial DW where no significant correlation was established (r = −0.077, p = 0.400).

Analysis of peri-implant radiographic measurements

This pilot study evaluates the reliability of calibrated radiographic measurements (EA, DW, DH, DA, MBL, CBL) against surgical reality. While the emergence angle (EA) and defect width (DW) show excellent inter-examiner reproducibility, the defect height (DH) and distal bone levels are highly variable and systematically underestimate actual surgical dimensions (mean differences of -2.99 mm and -3.52 mm for mesial and distal DH).

Concretely, for the practitioner:

  • Do not overestimate your images: Periapical radiography significantly underestimates the severity of bone defects. Consider your radiographic measurements as minimum indicators, not as the exact anatomical reality.
  • Make your analyses more reliable: For the most sensitive parameters (DH, DA, distal bone levels), never rely on a single reading. If possible, have measurements correlated by several observers or train yourself on very strict landmark definitions.
  • Prioritise robust data: You can place reasonable confidence in the emergence angle (EA) and defect width (DW) on calibrated images, unlike other measurements which remain too dependent on subjective interpretation.

Technical Glossary

Restoration Emergence Angle (EA): Angle formed between the longitudinal axis of the implant and the contour of the prosthetic restoration, measured on a calibrated periapical radiograph.

Defect Height (DH): Distance measured between the bone crest and the implant platform (for "bone-level" implants) or the transition to the endosseous portion (for "tissue-level" implants).

Defect Width (DW): Distance measured in a plane perpendicular to the long axis of the implant, between the most coronal point of the endosseous part and the bone crest.

Emergence Profile (EP): Morphological parameter related to the transition between the implant and the restoration, evaluated in this study via a categorical approach.

Marginal Bone Level (MBL): Peri-implant bone level measured in relation to anatomical or implant landmarks, used as a radiographic reliability parameter.

Intraclass Correlation Coefficient (ICC): Statistical indicator used here to measure inter- and intra-examiner agreement of continuous radiographic measurements, with a scale ranging from poor (< 0.40) to excellent (0.75–1.00).

Cone Paralleling Technique: Standardized method for performing periapical radiographs used in this study to minimize geometric distortion.


Source

  • Original title: Peri‐Implant Radiographic Measurements Compared With Intra‐Surgical Assessments: Diagnostic Accuracy and Reproducibility
  • Authors: Era Kakar, Parham Hazrati, Andrea Laureti, Zhaozhao Chen, Hom‐Lay Wang
  • Publication: Journal of Periodontal Research - 2026-09-29
  • DOI: https://doi.org/10.1111/jre.70169

Read also on the Delynov blog

06/03/2026
Xiaotong Fan, Yaping Chen, Lin Chen, Jinghan Wang, Liang Shi, Jiaqi Wu, Yilin Rao, ZhongC…, Journal of Visualized Experiments

Rat model of peri-implantitis: when ligature and P. gingivalis validate the clinical study

Peri-implantitis: when the synergy of ligature and P. gingivalis changes the game. Discover a new preclinical rat model to simulate...
19/03/2025
Marco Sorgato, Anna Bottin, Michele Stocchero, Marco Toia, Enrico Savio

Drill wear: an underestimated thermal risk in implantology

The success of osseointegration depends closely on the preservation of bone vitality during drilling. This in vitro study...

Information intended for healthcare professionals. This content may contain errors or truncated summaries. We recommend always checking with the original source article. Delynov disclaims any liability regarding the use of this information. This document is not intended for patients or the general public.

Guided surgery: AI for 3D model registration on imaging in 11ms
Automatic registration between 2D dental imaging and 3D models is a technological bottleneck

Delivery

From France to the whole world.

Europe, overseas territories, worldwide. Everything ships through our own logistics.

With you tomorrow

Ordered before 2 pm, delivered within 24 h. Otherwise, within 72 h.

Delivery included from €500 incl. VAT

Offer limited to mainland France.

A team that answers

Orders, traceability, materiovigilance, your dedicated contact in Épinal. Parcels prepared by the ESAT of Dinozé.

We’ll call you back

References

They trust us.

Public and private healthcare institutions order from Delynov.