Accuracy of probing attachment levels using a new computerized cemento-enamel junction probe.

BACKGROUND AND OBJECTIVES: The assessment of clinical attachment level (CAL) represents the gold standard for diagnosing and monitoring periodontal disease. The aim of the present study was to evaluate the performance of the newly introduced cemento-enamel junction (CEJ) probe in detecting CAL, using CEJ as a fixed reference point, and to compare the CEJ probe with the Florida stent probe (FSP) as well as with a standard manual probe, University of North Carolina-15 (UNC-15).
MATERIALS AND METHODS: Three examiners recorded the probing attachment level in 384 sites in case group (chronic periodontitis), and in 176 sites, in control group (healthy periodontal status), using the three probes. Subjects included both the sexes and ranged from 35 to 45 years. The experimental design was structured to balance the intra- and inter-examiner consistency at the same site during the two visits.
RESULTS: CEJ probe showed higher intra-and inter-examiner consistency over both FSP and UNC-15 in both the case and control groups. Frequency distribution of differences of various magnitudes of repeated measurements ≤1 mm was in the higher range of 86.8% to 87.5% for CEJ probe. The FSP was more reproducible than UNC-15 in detecting relative attachment level (RAL). INTERPRETATION AND
CONCLUSION: CEJ automated probe was found to have greatest potential for accuracy and consistency in detecting CAL than FSP and UNC-15. The automated probes appeared to be more reproducible than manual probes.

INTRODUCTION

The diagnosis of active periodontal disease has historically relied on several methods to determine changes in attachment apparatus of the teeth. Clinical probing is one procedure that has been used extensively to assess changes in periodontal attachment over time.[1] Increased precision of periodontal probing is of clinical importance because the reported prevalence of disease activity as identified by clinical attachment level (CAL) change clearly depends on the threshold used for identifying whether the loss of attachment has occurred.[2] Moreover, although CAL should be measured from the base of the pocket to the cemento-enamel junction (CEJ), the detection of this anatomic landmark is usually difficult and not reproducible by standard probing methods. Changes in the level of attachment can be caused only by gain or loss of attachment and they provide a better indicator of the degree of periodontal destruction. Pocket depth is less important than the level of attachment because it is not necessarily related to bone loss.[3]

Many investigations have pointed out errors inherent in the use of the conventional probes. The factors which may affect probe measurement include variation in the probing force, graduations of the probe, variations in the size of the probe, probe placement, reading the probe and data recording, and calculation of attachment level.[4] Hence, automated periodontal probes have been developed to reduce the measurement error with probing. These probes are combined with controlled force application, automated measurement, and computerized data collection, and provide a mean of recording the attachment changes over time.[5]

The Florida probe [Figure 1] is one such system which provides measurement of relative attachment level relative to fixed reference point. When using the Florida probe, relative attachment level (RAL) is recorded relative to fixed reference points such as the occlusal surfaces of the teeth (disc probe) or a prefabricated stent (stent probe). Unfortunately, the use of relative reference points provide no information relating to CAL at a single examination, and further, the reference points may change (a tooth may be restored, or a stent may become distorted). For these reasons, it is desirable to measure CAL using the CEJ as the reference point.[1] Recently, a modification of the Florida Probe i.e. CEJ probe was introduced to detect the CEJ and to improve the accuracy and consistency when recording periodontal attachment levels.[16] The CEJ probe was initially tested by Preshaw et al.,[1] His results indicated that the CEJ probe had reproducibility and reliability in detecting the CEJ in human skulls and measuring CAL in humans. Karpinia et al. (2004) evaluated the performance of CEJ probe in detecting CEJ as a landmark and compared with Florida disk probe and standard manual probe. He found CEJ probe to be more efficient than traditional probes.[7] The CEJ probe has a modified sleeve, which includes a 0.125 mm prominent edge to facilitate a “catch” of the CEJ. The width of this edge was considered small enough not to interfere with probing depth measurements offering clinicians, measurement of CAL and probing depth concurrently.[7]

Figure 1

Florida probe system unit

Hence, the aim of the present study was to evaluate the performance of the newly introduced CEJ probe in detecting CAL using the CEJ as a fixed reference point and to compare the CEJ probe [Figure 2] with the Florida stent probe [Figure 3] as well as with a standard manual probe - University of North Carolina-15 (UNC-15) [Figure 4].

Figure 2

CEJ probe handpiece

Figure 3

Florida stent probe handpiece (FSP)

Figure 4

Manual probe (University of North Carolina-15)

MATERIALS AND METHODS

A total of 560 sites were measured for probing attachment level in nine subjects with chronic periodontitis and four healthy subjects of both sexes, 35-45 years old selected from outpatient Department of Periodontics, College of Dental Sciences, Davangere, Karnataka, India.

Patients with chronic periodontitis and healthy periodontium were selected as evidenced clinically and radiologically. Chronic periodontitis patients were selected concurrently with the following criteria, patients with age ≥35 years, 30% of sites involved, periodontal pocket depth ≥4 mm, attachment loss ≥3 mm with severe bone loss (≥50% of root length) radiologically and healthy controls included volunteered paradental staff, with age group ≥35 years, presence of sulcus depth ≤3 mm and radiologically with or without crestal bone loss (CEJ as the landmark).[3] Patients with systemic diseases, aggressive periodontitis, pregnant and lactating women, third molar teeth, and prostheses/crown were excluded.

Supragingival scaling was performed by ultrasonic scaler to remove the local factors which interfered during probing procedure. Alginate impression was taken for both the arches and study models were prepared for the acrylic stents. The two probing appointments (1st and 2nd sitting) were scheduled and the duration between the appointments was one week (a time period in which change in attachment level is unlikely to occur).[8] Three trained examiners participated in this study.

Probing force for the electronic probe was preset at a constant of 15 gm and non standardized insertion probing force was used with manual probe. Each site was probed two times (1st and 2nd measurement) at each sitting. This was to reduce the error in probing angulation. Any sites with a measurement differences ≥1.0 mm was measured again till two of the measurements were within 1.0 mm and the mean was calculated. Four measurements per tooth were recorded (mesiobuccal, midbuccal, distobuccal, and midlingual/mid palatal). Measurements were made with the probe tip parallel to the long axis of the tooth at the midbuccal, midlingual/mid palatal, and as close as possible to the mid interproximal area from the buccal aspect.[9]

The probing order using different type of probe was randomized, but the same site during the same round, per visit was always evaluated by the identical probe. Steam sterilization of three probes was performed every time before recording attachment level in each patient.

Relative attachment level measurement using manual probe

The RAL was measured at selected site by using manual probe (UNC-15). The manual probe (UNC-15) was inserted parallel to the long axis of tooth gently, until resistance was noted and the nearest millimeter was recorded from the lower end of the occlusal stent as a reference point to base of the pocket [Figure 5].

Figure 5

Recording RAL using UNC-15 probe

Clinical attachment level measurement using cej probe

The probe tip reciprocated through a sleeve, the edge of the sleeve is the reference from which measurements were recorded. When the edge of the sleeve is brought into contact with CEJ, the CAL is recorded from flange to probe tip and transferred automatically to the computer when the foot switch is pressed [Figure 6].

Figure 6

Recording CAL using CEJ probe

Relative attachment level measurement using florida stent probe

A modified acrylic stent was used for the Florida stent probe to measure RAL. The probe tip reciprocated through a sleeve, a 2 mm metal ring, attached to the probe's sleeve, rested on the apical border of the stent and it is the reference from which measurements were recorded. The RAL is recorded from ledge to probe tip and transferred automatically to the computer when the foot switch is pressed [Figure 7].

Figure 7

Recording RAL using FSP

Statistical analysis

All the data were collected and subjected to statistical analysis using mean, paired “t” test, standard deviation, Pearson's correlation coefficient, frequency distribution of differences of repeated measurements, and kappa measurement of inter-examiner agreement.

RESULTS

Comparison of intra-examiner differences in CAL/RAL measurements in case group and control group are shown in Tables 1 and 2, respectively.

Table 1

Comparison of intra-examiner differences in CAL/RAL in case group

Table 2

Comparison of intra-examiner differences in CAL/RAL in control group

Comparison of inter-examiner differences in CAL/RAL measurements in case group and control group are shown in Tables 3 and 4, respectively.

Table 3

Comparison of inter-examiner differences in CAL/RAL in case group

Table 4

Comparison of inter-examiner differences in CAL/RAL in control group

Comparison of frequency distribution of differences in repeated measurements in case group is shown in Table 5.

Table 5

Frequency distribution of differences in repeated attachment level measurements in case group

The kappa agreement of inter-examiners using various probes in case group is shown in Table 6.

Table 6

The kappa agreement of inter-examiners using various probes in case group

DISCUSSION

New technology and high-tech computers are becoming the rule rather than the exception during patient treatment. Computer applications and new devices are being developed and marketed to improve diagnosis, enhance therapy, and monitor treatment outcomes. Additionally, computerization offers the entire dental team ideal potential to achieve examiner standardization, such that future comparison of health/disease becomes simpler, more precise and remains cost effective.

Traditional clinical periodontal diagnostic procedures (visual signs of inflammation, bleeding on probing, probing depths, and attachment levels) considered collectively, have provided useful information related to periodontal pathology.[10] Loss of clinical attachment, determined by physical assessment using a periodontal probe measures damage resulting from past episodes of periodontal disease. Currently, the gold standard for recording changes in periodontal status is longitudinal measurement of CAL from the CEJ or relative attachment level from a fixed reference point.[10-12] The CEJ Probe has been developed to improve detection of the CEJ allowing assessment of accurate attachment level changes over time.

In this study, the CEJ probe was found to be the greatest potential for accuracy, intra- and inter-examiner consistency and reliability in detecting clinical attachment level in both the case and control groups compared to both, FSP and UNC-15. These results are comparable with the results of Preshaw et al.,[1] and Karpinia et al.[7]

In case group, the differences of intra examiner consistency and reliability appeared to be better with FSP than UNC-15. However, the accuracy was similar for both the probes. These results are comparable with the Gibbs et al.[5] However, in control group, UNC-15 showed better intra-examiner consistency than the FSP. These results are similar to the study done by Quirynen et al.,[13] and Wang et al.,[14] who reported that manual probe seems to be slightly more reproducible than the automated probe.

In both the case and control groups, the inter-examiner consistency of FSP was found to be better than UNC-15. Similar results were found in the study by Gibbs et al.,[5] and Osborn et al.[15]

In case group, frequency distribution in repeated attachment level measurements and kappa agreement of inter-examiner with various probes indicates agreement that CEJ probe is more reproducible than FSP and UNC-15, whereas both, FSP and UNC-15, were proved to be similar in agreement.

In this study, all the three examiners were trained and experienced postgraduate students in the use of these probes which were being evaluated, due to which the differences between the results obtained by different examiners were minimal in both case and control groups, and thus, reducing the errors of examiner variability. Studies done by Karpinia et al.,[7] and Preshaw et al.,[1] also included calibrated examiners to minimize the variability error.

In this study, subjects with moderate to advanced periodontitis were included, whereas in a study by Marks et al.,[9] subjects with moderate periodontitis and by Osborn et al.,[15] included mild periodontitis. Hassel et al.,[16] stated that degree of inflammation is a factor that has been shown to influence measurement errors. When a comparison of studies yields disparate results, the question arises as whether differences are due to subject type or lack of reproducibility of the probe. However, in this study, probing was done with repeated measurements in subjects of case group with inflamed periodontium and control group with healthy periodontium to avoid the lack of reproducibility of probes. This could be explained by low mean CAL/RAL values presented by studied subjects.

Different types of probes used i.e. electronic, pressure-sensitive probe, and conventional probes differ greatly in design of probing device (differ in tip diameter, probing force, etc) as well as recording system which may determine their performance. By automated measurements, direct computerized data collection, visual observational and transcribing errors are eliminated, which is usually seen with conventional probes.[17] Controlled force (15 gm) application standardizes the probing pressure, thus reducing variations between examiners compared to conventional probe (non standardized force). In our study, the automated probes proved to be better with inter-examiners in both case and control groups compared to manual probe.

Another aspect to be considered is the type of tooth probed which could influence the result. Alves Rde et al.,[18] evaluated only anterior teeth and Osborn et al.,[15] study used only posterior proximal sites with mild periodontitis with probing force 20 gm. In this study, both anterior and posterior teeth were included. However, it must be realized that the studies done are too different to be compared.

Contour of the attachment level can be quite irregular and probing angle has a great effect on the variability as noticed by Watts et al.,[19] and Karim et al.[20] For reduction of errors associated with probing position and angle in this study, acrylic stent was used to fix the probe position and angle.

According to studies by Osborn et al.,[15] Clark et al.,[21] and Fleiss et al.,[22] inter-examiner reliability (ICC) was smaller than intra-examiner reliability for CAL and RAL. In our study, we used multiple examiners where in greater differences of the mean values of the measurements was obtained within the examiners than between examiners, indicating that intra examiner reliability for CAL and RAL was smaller than ICC, according to which multiple examiners are needed to measure the same subjects in clinical trials.

Errors encountered during probing due to the presence of subgingival deposits were reduced by replicate measurements, which is a more reliable data analysis in which detection of attachment level changes can be improved by checking and correcting data.[915] The electronic probe enhances the possibility of immediately identifying sites in which the replicate pair differs by more than 1 mm. If more than 1 mm, third probing was done, if less than 1 mm, the average of the two was calculated.

CONCLUSION

Within the limitations of our study, CEJ probe has proved to have the greatest potential for accuracy and reliability for measurements of CAL than FSP and manual probe (UNC-15), and FSP has proved to be better than manual probe (UNC-15), indicating that automated probes are better as far as accuracy, consistency, and reliability are concerned for measurements of attachment levels.