Comparative evaluation of probing depth and clinical attachment level using a manual probe and Florida probe.

BACKGROUND: To compare and evaluate the intra- and inter-examiner efficacy and reproducibility of the first-generation manual (Williams) probe and the third-generation Florida probe in terms of measuring pocket probing depth (PD) and clinical attachment level (CAL).
MATERIALS AND METHODS: Forty subjects/4000 sites were included in this comparative, cross-sectional study. Group- and site-wise categorizations were done. Based on gingival index, PD, and CAL, patients were divided into four groups, i.e., periodontally healthy, gingivitis, mild to moderate periodontitis, and severe periodontitis. Further, based on these parameters, a total of 4000 sites, with 1000 sites in each category randomly selected from these 40 patients, were taken. Full mouth PD and CAL measurements were recorded with two probes, by Examiner 1 and on Ramfjord teeth by Examiner 2.
RESULTS: Full mouth and Ramfjord teeth group- and site-wise PD obtained with the manual probe by both the examiners were statistically significantly deeper than that obtained with the Florida probe. The full mouth and Ramfjord teeth mean CAL measurement by Florida probe was higher as compared to manual probe in mild to moderate periodontitis group and sites, whereas in severe periodontitis group and sites, manual probe recorded higher CAL as compared to Florida probe.
CONCLUSION: Mean PD and CAL measurements were deeper with the manual probe as compared to the Florida probe in all the groups and sites, except for the mild-moderate periodontitis group and sites where the CAL measurements with the manual probe were less than the Florida probe. Manual probe was more reproducible and showed less interexaminer variability as compared to the Florida probe.

INTRODUCTION

Periodontitis is an inflammatory disease of bacterial origin that results in the progressive destruction of the tissues that support the teeth, specifically the gingiva, periodontal ligament, and alveolar bone.[1]

To assess the status of periodontal disease activity, one of the reliable and convenient ways is through the use of periodontal probes.[2] Williams periodontal probe, invented in 1936 by Williams, is the prototype of all the first-generation probes.[3] However, periodontal probing has its limitations. Operator errors, such as incorrect angulation of probe, amount of pressure applied to the probe, misreading the measurement on the probe, recording the data imprecisely, and miscalculating the attachment loss, may occur. Reading errors may result from interference from the calculus, presence of an overhanging restoration, or the crown's contour.[4] Other factors, such as size of the probe-tip, angle of insertion of the probe, precision of probe calibration, and degree of inflammation in the periodontal tissues, affect the sensitivity and reproducibility of measurements.[5]

To overcome the limitations of conventional probing, various modifications in probe designs and probing systems have been developed. Florida Probe® (Florida Probe Corp., Gainesville, FL, USA) devised by Gibbs et al., in 1988,[6] is a third-generation constant pressure electronic probe, consisting of a probe handpiece and sleeve, a displacement transducer, a foot switch, and a computer. The probe tip is hemispheric with a diameter of 0.45 mm and the sleeve has a diameter of 0.97 mm. Constant probing pressure of 15 g is provided by coil springs inside the handpiece. Three types of handpiece are available to be used with the system namely, Florida pocket probe, Florida stent probe, and Florida disk probe.[7] This probe can record probing depth (PD), missing teeth, recession, bleeding, suppuration, furcation involvement, mobility, and plaque assessment.[8] The fourth type is the cementoenamel junction (CEJ) probe that was initially tested by Preshaw et al.,[9] and it consists of a modified sleeve, with a 0.125 mm prominent edge to facilitate a “catch” of the CEJ. The width of this edge was small enough not to interfere with PD measurements offering clinicians, measurement of clinical attachment level (CAL) and PD concurrently.[10]

Hence, the aim of this study was to compare and evaluate the efficacy and reproducibility of the first-generation (Williams) manual probe and the third-generation Florida probe in terms of measuring PD and CAL.

MATERIALS AND METHODS

Patients who reported to the Outpatient Department of Periodontology between June 2012 and June 2013 were screened on the basis of inclusion and exclusion criteria. Systemically healthy patients with or without periodontal disease were included. Patients under medication affecting periodontal health and those having undergone periodontal therapy/surgery in the last 6 months were excluded from the study. A total of 40 patients (28 males and 12 females) who fulfilled the inclusion and exclusion criteria were included in the study.

Selected patients were explained about the study protocol and informed written consent was obtained. In all the selected patients, demographic details and gingival index (GI)[11] were recorded and PD and CAL were assessed and recorded with the help of Williams probe.

Based on the parameters recorded, patients were categorized as either periodontally healthy or diseased. Periodontally diseased patients were divided into three groups, i.e., gingivitis, mild to moderate periodontitis, and severe periodontitis.

Hence, based on PD and CAL, patients were categorized into four groups:

Group A: Periodontally healthy patients [12] (bleeding on probing [BOP] on ≤20% sites, CAL = 0)

Group B: Patients having chronic generalized gingivitis [12] (BOP on >l% sites, CAL = 0)

Group C: Patients having mild to moderate chronic generalized periodontitis with CAL in >l% of sites [13]

Group D: Patients having severe chronic generalized periodontitis with ≥5 mm of CAL in >l% of sites.[13]

For sitewise evaluation of two probes, a total of 4000 sites randomly taken from the included 40 patients, were divided into four categories with 1000 sites in each category based on GI, PD, and CAL measurements:

Periodontally healthy sites – GI score = 0 or 1,[11] CAL = 0[12]

Gingivitis sites – GI score = 2 or 3,[11] CAL = 0[12]

Mild-moderate periodontitis sites – CAL 13]

Severe periodontitis sites – CAL ≥5 mm.[13]

The total number of sites examined in 40 patients was 4196, out of which the periodontally healthy sites were 1050, gingivitis sites were 1063, mild-moderate periodontitis sites were 1083, and severe periodontitis sites were 1000. Group A (number of patients [n] =7) consisted of total of 768 sites, out of which 701 were periodontally healthy sites and 67 were gingivitis sites. Group B (n = 7) consisted of total of 780 sites, out of which 333 were periodontally healthy sites and 447 were gingivitis sites. Out of total of 1108 sites in Group C (n = 10), periodontally healthy sites were 8, gingivitis sites were 387, mild-moderate periodontitis sites were 592, and severe periodontitis sites were 121. Out of total of 1540 sites in Group D (n = 16), periodontally healthy sites were 8, gingivitis sites were 162, mild-moderate periodontitis sites were 491, and severe periodontitis sites were 879. Hence, to equalize, a maximum of 1000 sites in each group were included. The last 50 periodontally healthy sites, 63 gingivitis sites, and 83 mild-moderate periodontitis sites recorded out of 1050, 1063, and 1083 sites, respectively were eliminated. A total of 196 sites were eliminated so as to obtain 1000 sites in each category. To check for interexaminer variability, Ramfjord teeth (16, 21, 24, 36, 41, and 44) were probed twice.

Procedure

All the patients were made to undergo oral prophylaxis to remove deposits which can interfere in probing pockets and detecting CAL. The clinical parameters were recorded in all the patients with the help of two periodontal probes, i.e., Williams probe and Florida probe. To avoid bias and for randomization, on the even number of subjects, the parameters were recorded with the manual probe first and then with the Florida probe after a gap of 15 minutes, and the clinical parameters on the odd number of subjects were recorded with Florida probe first and then with manual probe after a gap of 15 minutes in each group by both the examiners. In all the four groups, PD was measured, and in Groups C and D, CAL was also recorded. The measurements (PD and CAL) were recorded in all the teeth present at four sites namely distobuccal, mid-buccal, mesiobuccal, and lingual/palatal, per tooth with two probes, i.e., with the manual and the Florida probe, by Examiner 1 on the same day, in single examination, and on Ramfjord teeth by Examiner 2 after a gap of 45 minutes.

Measurement of PD by Florida probe was done by advancing the probe tip of the Florida probe with a jiggling motion into the gingival sulcus, and the probe handpiece was pressed down until the leading edge of the sleeve surrounding the probe tip was positioned at the level of the free gingival margin. PD was then determined electronically and recorded on the laptop by depressing a foot switch. Similarly, for the measurement of the CAL by Florida probe, when the edge of the sleeve of the CEJ probe was brought into contact with CEJ, CAL was recorded from flange to probe tip and transferred automatically to the laptop when the foot switch was pressed.

For the measurement of PD by Williams probe, the probe was inserted parallel to the vertical axis of the tooth to reach the deepest point of the pocket, and the distance between the base of the pocket and gingival margin was recorded manually to the nearest millimeter marking. Similarly, CAL was measured as the distance between the fixed reference points, i.e., CEJ to the base of the pocket and recorded manually. With the manual probe when CAL was measured, the CEJ was defined visually wherever possible or with tactile sensation wherever visibility was an issue.

To evaluate interexaminer variability, selected teeth, i.e., 16, 21, 24, 36, 41, 44 (Ramfjord teeth), were examined by the second examiner on the same day after 45 minutes for both parameters by both the probes in the same order as done by the first examiner. The calibration of both the examiners was done before the start of this study. A pilot study on the same pattern with the same instruments and similar category of patients was conducted. Both the examiners recorded 100 sites in each category of patients and calibration was done to reach the accuracy of 95%.

Statistical analysis

The software used for the statistical analysis was Statistical Package for Social Sciences version 11.5 (IBM USA).

The intergroup comparisons of GI between the four groups were done using ANOVA and evaluated through Bonferroni post-hoc test. For the intraexaminer group- and site-wise comparisons between the two probes, the test used for the statistical analysis was paired t-test; ANOVA was used for the intraexaminer group- and site-wise comparison of each probe for PD measurement and independent t-test for CAL measurements. Intraclass correlation coefficient (ICC) was used for assessing interexaminer reliability of manual and Florida probes in measuring PD and CAL both group- and site-wise. The statistical significance levels were present at P ≤ 0.05.

RESULTS

A total of 40 patients (age range 18–64 years, mean age 33.52 years) were enrolled on the basis of the screening criteria in the study. The number of males was 28 (age range 19–64; mean age 34.89 years) and females was 12 (age range 18–41 years, mean age 30.33 years). Based on the parameters recorded, and mean values obtained intra- and inter-examiner, group- and site-wise comparisons were done in all the patients.

Groupwise comparisons

Comparison of full mouth mean GI and standard deviation (SD) for Groups A (1.08 ± 0.28), B (1.57 ± 0.49), C (1.81 ± 0.44), and D (2.17 ± 0.52) by ANOVA test resulted in statistically significant difference between them with P < 0.001. Using Bonferroni post-hoc test, the mean GI in the groups were in the order of A < B < C < D.

Mean PD ± SD and mean CAL ± SD for Groups A, B, C, and D are shown in Table 1.

Table 1

Full mouth and Ramfjord teeth mean probing depth±standard deviation and clinical attachment level±standard deviation measurements of Groups A, B, C, D

Intraexaminer comparisons were made between different groups, i.e., between manual probe readings obtained in all four groups and Florida probe reading of all four groups. Then, the comparisons between manual and Florida probe were made as measured by Examiner 1, both full mouth (M vs. F) and Ramfjord teeth (M1 vs. F1) and also Examiner 2 recordings on Ramfjord teeth, i.e., M2 versus F2. Full mouth intraexaminer comparison (Examiner 1) of mean PD ± SD by both manual and Florida probes respectively for both within the groups and between different groups resulted in statistically significant difference between them with P < 0.001 [Tables 1 and 2]. Similar results were obtained on comparing the mean PD ± SD between manual and Florida probes for Groups A, B, C, and D, in Ramfjord teeth by Examiner 1 (M1, F1) and Examiner 2 (M2, F2) in all groups except in Group A wherein the difference between M1, F1 was 0.054 [Table 2].

Table 2

Groupwise intraexaminer comparison of mean probing depth and clinical attachment level difference between manual and Florida probes in all four groups

For assessing interexaminer reproducibility of manual (M1, M2) and Florida (F1, F2) probes in measuring PD, groupwise ICC assessment was used which showed consistently higher values with manual probe as compared to Florida probe. Groupwise interexaminer comparison in Ramfjord teeth by manual and Florida probes (M1 with F2 and M2 with F1) is shown in Table 3.

Table 3

Groupwise interexaminer comparisons (Ramfjord teeth) of mean probing depth and clinical attachment level difference in all four groups

Similarly, intraexaminer comparison of full mouth mean CAL ± SD by Examiner 1 for both manual and Florida probes for both within and between Groups C and D was done; and statistically significant difference was found between them with P < 0.001 [Tables 1 and 2].

The comparison of mean CAL and SD by manual and Florida probes for Groups C and D, in Ramfjord teeth by Examiner 1 (M1, F1) and Examiner 2 (M2, F2) is shown in Table 2. Groupwise interexaminer comparisons in Ramfjord teeth for CAL measurements (M1 vs. M2, F1 vs. F2, M1 vs. F2, and F1 vs. M2) are shown in Table 3.

Sitewise comparisons

Mean values of PD ± SD and CAL ± SD on measuring 4000 sites with 1000 sites in each category are shown in Table 4.

Table 4

Full mouth and Ramfjord teeth mean probing depth±standard deviation and clinical attachment level±standard deviation measurements of healthy, gingivitis, mild-moderate periodontitis, and severe periodontitis sites

Full mouth intraexaminer (Examiner 1) comparison of mean PD and SD within and between different sites when measured with manual and Florida probes resulted in significant differences between them with P < 0.001 [Tables 4 and 5]. Similarly, intraexaminer comparison of Ramfjord teeth mean PD ± SD by both examiners (Examiners 1 and 2) between manual and Florida probe (i.e., M1, F1 and M2, F2) resulted in significant differences with P < 0.001 with Florida probe measurements being lower than manual at all sites [Table 5].

Table 5

Intraexaminer comparison of mean probing depth and clinical attachment level difference between manual and Florida probes in healthy, gingivitis, mild-moderate periodontitis, and severe periodontitis sites

Sitewise interexaminer comparison in Ramfjord teeth (M1 vs. M2, F1 vs. F2, M1 vs. F2, and F1 vs. M2) is shown in [Table 6].

Table 6

Sitewise interexaminer comparison (Ramfjord teeth) of mean probing depth and clinical attachment level difference in healthy, gingivitis, mild-moderate periodontitis, and severe periodontitis sites

Full mouth intraexaminer (Examiner 1) comparison of mean CAL ± SD between mild-moderate and severe periodontitis sites for manual and Florida probes was −3.69 and −2.67, respectively, resulting in a statistically significant difference between them and also within the sites for both the probes with P < 0.001 [Table 4]. Similarly, intraexaminer comparison in Ramfjord teeth by both the Examiners 1 and 2 between both the probes resulted in statistically significant difference with only exception being Examiner 2 where difference between 2 probes in mild-moderate periodontitis group was 0.332 [Table 5].

Sitewise interexaminer reliability of CAL measurement by both the probes showed higher ICC value with manual probe in both mild-moderate as well as severe periodontitis groups, i.e., 0.81 as compared to Florida probe with ICC value being 0.50 and 0.58, respectively [Table 6].

DISCUSSION

To estimate the severity of soft tissue destruction and disease progression as well as the response to periodontal treatment, measurements of PD, CAL, and BOP are currently the most frequently used and the most informative parameters.[14] Variations in probing force appear to be evident not only between different examiners but also with a single examiner.[15] The use of constant pressure electronic probes overcomes certain limitations of the manual probe by providing a constant probing force and also electronic data collection eliminates transcription errors with higher resolution than the manual probes. The reproducibility of the probe is how a probe could give the same measurement every time.[16]

Group- and site-wise categorizations in the study were done as in groupwise categorization, any patient in a particular group may have a combination of various sites present, i.e., periodontally healthy, gingivitis, mild to moderate periodontitis, and severe periodontitis sites, or the disease is not uniform at all sites. Further, in our daily clinical practice, an individual patient is considered as an entity. Periodontitis is a site-specific disease. In sitewise categorization, all the sites in that particular category had the same severity of disease or the disease was uniform, thereby controlling for the variations in periodontal inflammation.

In the present study, comparison and reproducibility of mean PD ± CAL measurements with manual and Florida probe were evaluated.

The mean GI increased significantly from Group A to Group D [Table 1] which could be attributed to the increase in the gingival inflammation which further indicates the increase in the severity of the disease. This increase in the gingival inflammation may affect PD and CAL measurements. The inflamed tissues offer less resistance to probing as compared to the healthy tissues, so the probe tip penetration in healthy tissues is less than the inflamed tissues.[17]

Between two registrations, a time interval of at least 15 minutes was introduced to eliminate a possible bias due to examiners memory so that the second measurement could not be influenced by the first, and it also provided adequate rest to the patient in between different probing registrations.

To minimize the effect of bias and for the authenticity and validity of the data, two examiners were chosen.

When considering the full mouth mean PD ± SD as recorded by manual probe and Florida probe for all the four groups as well as sites by Examiner 1, it was observed that the mean PD recorded by manual and Florida probes increased from Group A to Group D and also from periodontally healthy to severe periodontitis sites, with a significant difference between them, suggesting the increase in the severity of the disease with the degree of inflammation present. Group- and site-wise intraexaminer comparisons between the manual probe and Florida probe by Examiner 1 revealed that full mouth PD obtained with the manual probe was statistically significantly deeper than that obtained with the Florida probe.

Similar results were obtained by both examiners in Ramfjord teeth, with manual probe measurements being statistically significantly deeper as compared to Florida probe (M1 > F1 and M2 > F2) in all groups and sites.

When interexaminer group- and site-wise comparison of PD measurements by manual and Florida probes (M1, F2 and M2, F1) in Ramfjord teeth was done, the measurements recorded by manual probe of Examiner 1 and Florida probe of Examiner 2 increased from Group A to Group D and from periodontally healthy to severe periodontitis sites with statistically significant differences between them. PD measurements recorded by manual probe by Examiners 1 and 2 were deeper as compared to Florida probe in all the groups and sites (M1 > F2 and M2 > F1). The results were in accordance with the studies of Gibbs et al.,[6] Osborn et al.,[8] Gupta et al.,[16] Magnusson et al.,[18] Perry et al.,[19] Osborn et al.,[20] Rams and Slots,[21] Quirynen et al.,[22] Tupta-Veselicky et al.,[23] Hull et al.,[24] Breen et al.,[25] and Barendregt et al.[26]

In the interexaminer group- and site-wise comparisons, ICC for manual probe was higher as compared to the Florida probe. The manual probe proved to be more reliable than electronic probe in PD measurements, in accordance with reports by Buduneli et al.,[14] Quirynen et al.,[22] Barendregt et al.,[26] Wang et al.,[27] Wang et al.,[28] Mayfield et al.,[29] and Khocht and Chang.,[30] who concluded that such automated devices offered no significant advantage over manual probes which was in contrast to the studies of Samuel et al.,[7] Magnusson et al.,[18] Rams and Slots,[21] and Araujo et al.[31]

In our study, the differences in the PD with manual recordings being deeper than the Florida probe measurements may be attributed to uncontrolled probing force while using manual probe and also lack of tactile sensitivity and placement problems of Florida probe, especially in proximal areas because of its size and sleeve. Rounding off of the measurements of the manual probe, difficulty in visual read out may have led to deeper manual probe recordings. The lesser reproducibility of Florida probe as compared to manual probe may be due to the lack of familiarity with the automated probe.

Full mouth mean CAL ± SD by manual probe and Florida probe for Groups C and D and also their respective sites by Examiner 1 increased from Group C to Group D and from mild-moderate to severe periodontitis sites with the differences being significant. The reason for this increase may be attributed to the increase in the severity of the disease with subsequent increase in the gingival inflammation.

Group- and site-wise intraexaminer comparison between the manual probe and Florida probe by Examiner 1 and by both examiners in Ramfjord teeth revealed that CAL obtained with the manual probe in Group C and mild-moderate periodontitis sites was statistically significantly lower than CAL obtained with the Florida probe, whereas CAL obtained with the manual probe in Group D or severe periodontitis sites was statistically significantly higher than CAL obtained with the Florida probe.

Similar results were also obtained when the interexaminer group- and site-wise comparison of CAL measurements by manual and Florida probes (M1, F2 and M2, F1) were done.

The Group D and severe periodontitis sites are following the same trend as that of PD with the measurements by manual probe being greater than the Florida probe by both the examiners, as the factors affecting the PD measurement by manual and Florida probe are applicable to CAL measurement as well. The contrasting results of the CAL measurement in Group C and mild-moderate periodontitis sites could be attributed to the fact that in our study, in mild to moderate periodontitis group, out of total of 713 CAL sites, 38.2% of sites had gingival recession whereas 61.8% sites had no recession, so identification of CEJ by Florida probe in this group was difficult because of hidden CEJ. Identification of the CEJ is often complicated by its subgingival location and lack of clear demarcation felt by probing. In addition, lack of familiarity with the new probe (Florida probe) might be the contributing factor in difficulty in CEJ detection. Hence, the Florida probe demonstrated higher mean CAL in mild to moderate periodontitis group as compared to severe periodontitis group, whereas in severe periodontitis group, out of total of 1390 CAL sites, 66.6% of sites had gingival recession and 33.4% sites had no recession. Hence, detection of CEJ in this group was easier as the CEJ was exposed and the CEJ probe could easily measure the CAL. This was also supported by Preshaw et al.[9] and Karpinia et al.[32]

In the interexaminer group- and site-wise comparison, ICC for manual probe was higher as compared to the Florida probe which was in accordance with reports by Wang et al.,[28] Jeffcoat and Reddy,[33] Reddy et al.,[34] and Oringer et al.[35] but in contrast to the studies of Gibbs et al.,[6] Deepa and Prakash,[10] and Jeffcoat et al.[36]

In our study, the differences in CAL measurement between manual and Florida probe in mild-moderate and severe periodontitis groups and sites may be attributed to the fact that Florida probe is able to identify and locate CEJ easily by the sleeve of the probe in cases with gingival recession. The lesser reproducibility of Florida probe as compared to manual probe may be due to lack of familiarity with the automated probe and also to the increased resolution of Florida probe which might have led to greater variation in repeated measurements.

CONCLUSION

Mean PD and CAL measurements were found to be deeper with the manual probe as compared to the Florida probe in all the groups and sites, except for the mild-moderate periodontitis group and sites where the measurements with manual probe were found to be less than the Florida probe. Manual probes are more reliable and show less interexaminer variability as compared to the electronic probes.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.