Effectiveness of nonsurgical re‐instrumentation: Tooth‐related factors

Abstract

Background

To investigate tooth‐related factors that influence pocket closure (PC) and the reduction of pocket probing depths (PPD) after nonsurgical re‐instrumentation (NSRI) as part of step 3 therapy.

Methods

A total of 480 patients (10,807 teeth) presenting with residual pockets 6.33 ± 3.79 months after steps 1 and 2 of periodontal therapy were included and retrospectively analyzed before and 5.93 ± 4.31 months after NSRI. Reduction of PPD and PC rates following NSRI were associated with tooth‐related factors, namely tooth type, arch, number of roots, furcation involvement (FI), pulp vitality, mobility, type of restoration, presence of plaque, and bleeding on probing (BOP), using mixed‐effects regression models.

Results

NSRI reduced periodontal pockets persisting after initial cause‐related therapy by (mean ± SD) 1.32 ± 1.79 mm in PPD, and PC rate was 40%. Moderate pockets (4–5 mm) responded better to NSRI than deep pockets (≥ 6 mm) in terms of PC (51% vs. 16%). Both PPD reduction and PC rates of deep residual pockets were significantly influenced by tooth type, arch, number of roots, and presence of BOP.

Conclusion

Tooth type, arch, number of roots, and presence of BOP at re‐evaluation (before NSRI) had a significant and clinically relevant influence on NSRI as part of step 3 therapy. Considering these factors, particularly for deep residual pockets, may allow more tailored re‐intervention.

Plain Language Summary

The present study aimed to investigate the influence of tooth‐related factors on the outcome of repeated nonsurgical therapy of periodontitis. Therefore, 480 patients (10,807 teeth) presenting with clinical symptoms of persistent periodontitis after initial therapy were administered repeated nonsurgical therapy and retrospectively analyzed. Therapy outcomes were associated with tooth‐related factors, namely tooth type, tooth location (maxilla/mandible), number of roots, involvement of the root furcation area in multi‐rooted teeth, pulp vitality, mobility, restoration, presence of plaque and bleeding upon periodontal probing, using mixed‐effects models. The results revealed that repeated nonsurgical therapy was effective in reducing inflammation and clinical signs of disease, with moderate residual periodontal defects responding better than deep defects. Healing of deep defects after repeated nonsurgical therapy; however, was significantly influenced by the factors: tooth type, location, number of roots, and bleeding on probing. Considering these factors, particularly in deep residual defects which are commonly suggested to be treated surgically, may allow less invasiveness and thus a more tailored re‐intervention.

1. INTRODUCTION

In patients affected by periodontitis, nonsurgical periodontal therapy (NSPT) is known to be effective in reducing the majority of periodontal pockets and should be performed regardless of the stage or grade in advance to more invasive therapy approaches. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ However, NSPT is technically sensitive ⁷ , ⁸ , ⁹ and, consequently, inflamed pockets often persist despite thorough subgingival debridement. ¹⁰

Considering residual pockets after steps 1 and 2 of therapy, current guidelines recommend further treatment including nonsurgical re‐instrumentation (NSRI) and/or periodontal surgery (step 3). ¹¹ In their systematic review, Sanz‐Sanchez et al. concluded that in moderate residual pockets (4–5 mm) persisting after steps 1 and 2 NSRI should be performed. In deep residual pockets (≥ 6 mm), on the other hand, periodontal surgery is recommended. ¹²

Regarding the impact of NSRI, however, the literature is inconclusive, especially considering tooth‐related factors that may influence its effectiveness. While Badersten et al. showed no additional benefit of repeated instrumentation, ¹³ the results of more recently published studies showed additional reductions of pocket probing depths (PPD) following re‐instrumentation. ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ Notably, only Tomasi et al. (including 32 patients) and Ferrarotti and co‐workers (25 patients) reported on tooth‐related factors affecting the effectiveness of NSRI. Both studies found therapy more successful in single‐rooted than multi‐rooted teeth. Due to their comparatively small sample sizes, however, the relevance of the data is limited.

The aim of this study was, therefore, to evaluate tooth‐related factors influencing the outcomes of NSRI in a large cohort comprising over 10,000 teeth. Using possible associations may allow you to better tailor the decision for nonsurgical or surgical re‐intervention of specific teeth to achieve the endpoints of therapy.

2. MATERIALS AND METHODS

This is a retrospective analysis of data collected in a prospective clinical trial that was approved by the Ethics Committee of the Medical Faculty of the Ludwig‐Maximilians‐University, Munich (No. 025‐11 [prospective trial], No. 22‐0669 [retrospective analysis]) and conducted by good clinical practice and the principles of the Declaration of Helsinki, as revised in 2013. This study is a continuation of the study published by Werner et al. ¹⁸ included in the same project with the approval of the ethics committee and investigating the same population. The study description followed the guidelines for Strengthening the Reporting of Observational Studies in Epidemiology (STROBE). ¹⁹

2.1. Study population

The study population comprised 480 Patients with a total of 10,807 teeth, who received steps 1 and 2 of periodontal therapy between February 2011 and March 2016 in the undergraduate program at the Department of Conservative Dentistry and Periodontology, University Hospital, LMU Munich. All patients met the following inclusion criteria: (1) age ≥ 18 years; (2) diagnosis of periodontitis according to the 2017 classification ²⁰ ; (3) dental and periodontal charts providing at least PPD, plaque, and bleeding on probing (BOP) at six sites/tooth, tooth mobility, furcation involvement (FI), the type of restoration and pulp vitality at baseline (t0), on average 6 months after conclusion of steps 1 and 2 (re‐evaluation, t1) and on average 6 months after NSRI (t2). The exclusion criteria were: (1) pregnancy at baseline or during the observation period, (2) withdrawal of the consent during the study period, (3) previous periodontal treatment < 2 years before enrollment into the study, and (4) administration to surgical periodontal treatment as part of step 3 therapy.

2.2. Clinical parameters and outcome variables

CE and RH performed all measurements recorded and were calibrated in advance. Calibration phase consisted of repeated periodontal probing using the models introduced by Heym et al. (inter‐rater reliability of periodontal probing, kappa = 0.82). ²¹ , ²² Periodontal re‐evaluation (t1) was performed 6.33 ± 3.79 months after NSPT and 5.93 ± 4.31 months after NSRI (t2). Periodontal and dental examination was conducted at t0, t1 and t2 as described before. ¹⁸ Briefly, PPD were determined to the nearest millimeter at six sites per tooth using a PCP‐12 periodontal probe with a probing force of approximately 0.2–0.3 N. ²² , ²³ BOP was assessed approximately 30 s after probing according to van der Weijden et al. ²⁴ Mobility was assessed as described by Miller. ²⁵ FI was measured with a 2N‐Nabers probe and graded according to Hamp et al. ²⁶ Pulp vitality was tested using a foam pellet sprayed with Endo Ice (Coltène/Whaledent, Altstätten, Switzerland) and applied to the buccal side of the tooth. ²⁷ Plaque was recorded at six sites according to O'Leary et al. ²⁸ Pocket closure (PC) was defined per site, as a PPD of 4 mm in absence of BOP or ≤ 3 mm, according to the current classification. ²⁹ Furthermore, the patient‐related factors age, sex, current smoking status, and diabetes mellitus as well as the intake of systemic antibiotics during NSPT were recorded.

2.3. Nonsurgical periodontal treatment and re‐instrumentation

NSPT was performed by undergraduate students under the supervision of CE and RH as described before. ¹⁸ Before NSPT, patients were given detailed information on the etiology, pathogenesis, risk factors, and treatment of periodontitis and were subjected to oral hygiene instructions and professional mechanical plaque removal as part of step 1 therapy. NSPT was performed under local anesthesia at all sites with PPD > 3 mm using sonic devices in combination with different Gracey curettes (SG5/6, SG7/8, SG 13/14, SG15/16, Hu‐Friedy, Chicago, USA), without limits in time. ³⁰ After re‐evaluation, all residual pockets (4 mm with BOP and ≥ 5 mm) received NSRI in the same manner as NSPT in a single session, additionally enforcing oral hygiene instructions and performing professional mechanical plaque removal at all teeth.

2.4. Sample size

Due to the study's retrospective design, a formal a priori sample size calculation could not be performed. However, post hoc power analysis on the tooth level yielded a chi‐squared value of 159.7, based on the distribution of PC in single‐ and multi‐rooted teeth from the 4103 teeth that underwent NSRI. Using this chi‐squared value with a significance level of p < 0.001, the analysis revealed a power of 1.0. One the patient level the power was calculated at 0.96 assuming an effect size d = 0.76, based on the observed differences in the mean full‐mouth bleeding scores at t1 (Table 1).

TABLE 1.

Patient‐ and tooth‐level characteristics of the study participants.

Variable
Patient‐level factors
No. of participants	480 (100.0)
Age, y	57.4 ± 12.3
Female	224 (46.6)
Smokers	90 (18.8)
Diabetes	36 (7.5)
No. of teeth	22.7 ± 5.9
Systemic antibiotics during NSPT	75 (15.6)
Tooth‐level factors
Teeth	10807 (100.0)
Category at baseline
PPD 1‐3 mm	5194 (48.1)
PPD 4–5 mm	3598 (33.3)
PPD ≥6 mm	2015 (18.6)
Category at re‐evaluation
PPD 1–3 mm	6703 (62.0)
PPD 4‐5 mm	2887 (26.7)
PPD ≥6 mm	1217 (11.3)
Category after NSRI
PPD 1–3 mm	7098 (65.7)
PPD 4–5 mm	2701 (25.0)
PPD ≥6 mm	1008 (9.3)
Tooth‐related factors
Type
Incisors	5112 (47.3)
Premolars	2967 (27.5)
Molars	2728 (25.2)
Single‐rooted	7385 (68.3)
FI ≥2	525 (15.3)
Mobility	1731 (16.0)
BOPREV (positive)	5072 (46.9)
PIREV (positive)	5429 (50.2)
Restoration
None	5269 (48.8)
Filling	2959 (27.4)
Crown	2579 (23.8)
Vitality (positive)	9347 (86.5)

Note: Continuous variables are presented as mean ± standard deviation, and categorical variables are presented as n (%).

Abbreviations: BOPREV, bleeding on probing at re‐evaluation; FI, furcation involvement; NSRI, non‐surgical re‐instrumentation; PIREV, plaque index at re‐evaluation; PPD, probing pocket depth.

2.5. Sources of bias

The retrospective nature of the present study coincides with an increased risk of bias. To address misclassification and performance bias that might have occurred during diagnosis or therapy in the undergraduate program, the supervising operators (CE and RH) underwent a calibration phase in advance (inter‐rater reliability of periodontal probing, kappa = 0.82) ²¹ , ²² to assure consistent levels of clinical measurements and treatment.

2.6. Statistical analysis

The normal distribution of data was tested using the Kolmogorov–Smirnov test. Teeth were categorized into three groups (shallow pockets = 1–3 mm, moderate pockets = 4–5 mm, and deep pockets ≥ 6 mm) at t0, t1, and t2 using their deepest PPD, respectively. Differences between different tooth‐related factors were compared using an analysis of variance (ANOVA) for continuous variables, the Mann–Whitney U test or the Kruskal–Wallis test for ordinal and skewed variables, and the chi‐squared test for categorical variables. For the comparison of PPD reduction between different groups, analysis of covariance (ANCOVA) was included to adjust for significant differences at re‐evaluation. Post hoc pairwise analysis for the Kruskal–Wallis test was done using the Dunn–Bonferroni test. Data were expressed as mean ± standard deviation (SD) and categorical variables were presented as frequencies (with percentage) unless stated otherwise. Mixed‐effects regression models were used to model the association between tooth‐related risk factors and PC, accounting for the clustering of teeth within patients by including the patient ID as a random effect and adjusting for all baseline covariates as fixed effects. Results are shown as adjusted odds ratios (aOR) per 1‐unit change of PC with corresponding 95% confidence intervals (CIs). The two‐sided significance level was set at α = 0.05 for all tests. All analyses were performed using the SPSS (Version 29.0, IBM, Armonk, USA)

3. RESULTS

3.1. Patient characteristics

A total of 480 patients with a total of 10,807 teeth were included in the final analysis having received steps 1 and 2 as well as NSRI as part of step 3 between February 2011 and March 2016. Patients' mean age at t0 was 57.4 ± 12.3, the male‐to‐female ratio was 54.4/46.6%, 18.8% were smokers, and 7.5% had been diagnosed with diabetes mellitus (Table 1). During NSPT 75 (15.6%) of the patients received adjunctive systemic antibiotics in terms of amoxicillin and metronidazole, or metronidazole alone in cases of intolerance to penicillin derivates (Table 1).

Out of the 10,807 teeth, 5112 were incisors, 2967 were premolars and 2728 were molars. A total of 5167 (47.8%) teeth were located in the upper arch and 5640 (52.2%) in the lower. The single‐rooted to multirooted ratio was 68%/32%, and 15% of the multi‐rooted teeth had FI ≥ 2. Also, 16% presented with a degree of mobility ≥ 1, 53% with restorations (i.e., filling or crown), 87% tested positive for pulp vitality and 47% of all teeth showed BOP at more than two sites (Table 1).

3.2. Periodontal characteristics at re‐evaluation

Considering solely periodontal pockets after NSPT (4 mm with BOP and ≥ 5 mm), the mean PPD at re‐evaluation was 5.07 ± 1.28 mm and differed significantly among the groups. The following factors were associated with increased PPD at re‐evaluation: tooth type (molars), arch (upper), number of roots (multi‐rooted), restoration (filling or crown), mobility, and BOP (positive) (see Table 2).

TABLE 2.

Site‐specific probing depth changes for periodontal pockets following NSRI.

Factor	PPD REV (t1)	PPD RECALL (t2)	PPD RED
All	5.07 ± 1.28	3.75 ± 1.63	1.32 ± 1.79
Incisor	4.97 ± 1.23	3.41 ± 1.23	1.62 ± 1.75
Premolar	4.98 ± 1.16	3.57 ± 1.16	1.47 ± 1.67
Molar	5.17 ± 1.35	4.04 ± 1.35	1.06 ± 1.86
p‐value	<0.001 a , c	<0.001 a , b , c	<0.001 a , b , c , d
Single‐rooted	4.96 ± 1.54	3.47 ± 1.64	1.57 ± 1.70
Multi‐rooted	5.15 ± 1.63	3.98 ± 1.67	1.12 ± 1.84
p‐value	<0.001	<0.001	<0.001 d
FI 0/1	5.15 ± 1.33	3.96 ± 1.65	1.19 ± 1.19
FI 2/3	5.26 ± 1.37	4.22 ± 1.85	1.03 ± 1.04
p‐value	0.189	0.012	0.188
No mobility	4.97 ± 1.17	3.69 ± 1.54	1.36 ± 1.72
Mobility	5.32 ± 1.48	3.93 ± 1.83	1.22 ± 1.98
p‐value	<0.001	<0.001	0.002 d
No restoration	4.97 ± 1.14	3.59 ± 1.50	1.44 ± 1.69
Filling	5.15 ± 1.37	3.83 ± 1.67	1.27 ± 1.83
Crown	5.11 ± 1.31	3.85 ± 1.70	1.23 ± 1.87
p‐value	<0.001 a , b	<0.001 a , b	<0.001 a , b , d
BOPREV ‐	4.90 ± 1.18	3.72 ± 1.63	1.29 ± 1.75
BOPREV +	5.24 ± 1.37	3.78 ± 1.62	1.35 ± 1.82
p‐value	<0.001	0.202	0.149 d
PIREV +	5.06 ± 1.26	3.82 ± 1.62	1.24 ± 1.80
PIREV ‐	5.06 ± 1.32	3.64 ± 1.61	1.43 ± 1.79
p‐value	0.273	<0.001	<0.001
Vitality +	5.08 ± 1.27	3.76 ± 1.64	1.32 ± 1.81
Vitality ‐	5.00 ± 1.33	3.72 ± 1.55	1.28 ± 1.72
p‐value	0.075	0.457	0.552

Note: Data are presented as mean ± standard deviation. Bold indicates statistically significant values (p < 0.05). The comparison of the two groups was analyzed using the Mann–Whitney U test. Comparison of more than two groups was analyzed using the Kruskal–Wallis test with the following symbols indicating significant differences between the groups:

^a 1 versus. 3, p < 0.05.

^b 1 versus 2, p < 0.05.

^c 2 versus 3, p < 0.05.

^d Adjusted for significant differences in probing depth at re‐evaluation using ANCOVA.

Abbreviations: BOPREV, bleeding on probing at re‐evaluation; NSRI, non‐surgical re‐instrumentation; PIREV, plaque index at re‐evaluation; PPD, probing pocket depth.

As described before ¹⁸ all teeth were classified into three different categories (shallow pockets = 1–3 mm, moderate pockets = 4–5 mm, and deep pockets ≥ 6 mm) using their deepest pocket. At re‐evaluation, 62% of teeth showed shallow, 27% moderate, and 11% deep pockets. The distribution of pocket categories differed significantly regarding the factors investigated. The number of deep residual pockets was significantly higher in teeth located in the upper arch, in multi‐rooted than single‐rooted teeth, molars in particular, teeth with FI ≥ 2, restorations, mobility, and BOP (see Table S1 in the online Journal of Periodontology).

3.3. Periodontal characteristics following NSRI

3.3.1. Average pocket probing depth reduction

Following NSRI, a mean PPD reduction of 1.32 ± 1.79 mm was observed. Significant differences were present with regards to the tooth type comparing incisors to premolars and molars, upper to lower arch, multi‐rooted to single‐rooted teeth, mobile to nonmobile teeth, and sites exhibiting plaque to those without. BOP, FI, mobility, restoration, and pulp vitality; however, were factors not significantly related to PPD reduction following re‐instrumentation (see Table 2).

Considering teeth with moderate residual pockets, a mean reduction of 0.87 ± 1.42 mm was observed, whereas, in deep residual pockets, a mean reduction of 2.40 ± 2.14 mm was present (see Tables S3 and S3.1 in the online Journal of Periodontology). While in deep residual pockets, the same factors as mentioned for the entire cohort (tooth type, arch, number of roots, and BOP) had a significant impact, in moderate pockets also teeth with restorations and FI ≥ 2 influenced PPD reduction negatively.

3.3.2. Distribution of shallow, moderate, and deep pockets

Across all teeth, there was a significant improvement in pocket categories (shallow = 1–3 mm, moderate = 4–5 mm, and deep ≥ 6 mm) after NSRI (see Figure 1 and Table S2 in the online Journal of Periodontology). 35% of teeth changed from moderate to shallow and 42% from deep to either moderate (31%) or shallow (11%) (p < 0.001). The number of deep pockets persisting after re‐instrumentation was significantly higher in teeth located in the upper arch than those in the lower arch (12%/7%, p < 0.001), multi‐rooted than single‐rooted teeth (19%/5%, p < 0.001), molars than premolars and incisors (21%/7%/5%, p < 0.001), teeth with FI ≥ 2 than those with FI < 2 (34%/17%, p < 0.001), restored than non‐restored teeth (13%/6%, p < 0.001), mobile than nonmobile teeth (17%/8%, p < 0.001), and sites exhibiting BOP than those that did not (13%/6%, p < 0.001).

FIGURE 1.

Distribution of shallow, moderate, and deep pockets following NSRI. Tooth‐related factors are differentiated into the groups anatomy and restoration (A), and tooth‐related features (B) and further stratified to moderate (4–5 mm) and deep (≥ 6 mm) pocket category at re‐evaluation. Shallow pockets = 1–3 mm (green), moderate pockets = 4–5 mm (orange), and deep pockets ≥ 6 mm (red); n = 10,807, * Indicates significant differences (p < 0.05). BOP, bleeding on probing; NSRI, nonsurgical re‐instrumentation.

Considering only deep residual pockets, arch, tooth type, number of roots, restoration, and BOP revealed a significant impact on pocket category changes. In detail, 50% of all incisors and all single‐rooted teeth with deep residual pockets at re‐evaluation were allocated to moderate (32%) or shallow (18%) pockets after re‐instrumentation, while another 50% remained deep. In molars and multi‐rooted teeth, on the other hand, only 38% of deep pockets turned to moderate (33%) or shallow (7%) pockets after re‐instrumentation, while 62% remained deep (see Table S2.1 and S2.2 in the online Journal of Periodontology).

3.3.3. Pocket closure

NSRI resulted in a mean PC rate of 40%. Moderate pockets that persisted after NSPT were significantly more likely to achieve closure than deep residual pockets (51%/16%, p ≤ 0.001) (Table 3, Figure 2). Besides PPD at re‐evaluation, the following factors were significantly associated with less PC than their respective control: tooth type (molars 29%/premolars 46%/incisors 52%, p < 0.001), arch (lower 56%/upper 63%, p < 0.001), number of roots (multi‐rooted 31%/single‐rooted 50%, p < 0.001), FI (FI ≥ 2 28%/FI < 2 51%, p < 0.001), mobility (mobile 34%/ not mobile 42%, p < 0.001) restoration (filling/crown 36%/non‐restored 48%, p < 0.001) and BOP (BOP + 37%/BOP – 45%, p < 0.001). Again, subgroup analysis of deep residual pockets revealed significant differences in terms of PC considering the tooth‐related factors investigated (see Figure 2 and Tables S4 and S4.1 in the Journal of Periodontology). While in incisors, a PC rate of 25% was observed following NSRI, deep residual pockets at molar sites were closed in only 10%. A comparable ratio was found for single‐rooted (24%) and multi‐rooted teeth (11%).

TABLE 3.

PC through different groups following NSRI.

Factor	PC	No PC
All	1657 (40.3)	2447 (59.7)
PPD at REV 4–5 mm	1461 (50.6)	1426 (49.4)
PPD at REV ≥6 mm	196 (16.1)	1021 (83.9)
p‐value		<0.001
Incisor	678 (51.6)	635 (48.4)
Premolar	458 (46.4)	529 (53.6)
Molar	521 (28.9)	1283 (71.1)
p‐value		<0.001
Single‐rooted	1012 (50.2)	1002 (49.8)
Multi‐rooted	645 (30.9)	1445 (69.1)
p‐value		<0.001
FI 0/1	3899 (51.2)	3710 (48.8)
FI 2/3	195 (27.9)	505 (72.1)
p‐value		<0.001
No mobility	1351 (42.1)	1858 (57.9)
Mobility	306 (34.2)	589 (65.8)
p‐value		<0.001
No restoration	746 (48.7)	785 (51.3)
Filling	442 (35.6)	800 (64.4)
Crown	479 (35.7)	862 (64.3)
p‐value		<0.001
PIREV +	1288 (60.3)	847 (39.7)
PIREV ‐	993 (58.3)	710 (41.7)
p‐value		0.206
BOPREV‐	736 (45.2)	893 (54.8)
BOPREV+	921 (37.2)	1554 (62.8)
p‐value		<0.001
Vitality+	1424 (40.8)	2066 (59.2)
Vitality—	233 (38.1)	379 (61.9)
p‐value		0.204

Note: Data were analyzed using the chi‐squared test and are presented as frequencies and percentages in parentheses. Bold indicates statistically significant values (p < 0.05).

Abbreviations: BOPREV (±), bleeding on probing at re‐evaluation (positive/negative); FI (0/1/2/3), furcation with degrees; PIREV (±), plaque index at re‐evaluation (positive/negative); PC, pocket closure; PPD, probing pocket depth; RED, reduction; REV, re‐evaluation.

FIGURE 2.

Periodontal PC of moderate and deep residual pockets following NSRI. PC is defined as probing depths ≤ 3 or 4 mm without BOP. Tooth‐related factors are differentiated into the groups anatomy and restoration (A), and tooth‐related features (B). n = 4,104, * Indicates significant differences (p < 0.05). BOP, bleeding on probing; NSRI, nonsurgical re‐instrumentation; PC, pocket closure.

Mixed‐effects regression model confirmed the aforementioned factors, revealing higher chances for PC in moderate residual compared to deep pockets (aOR 7.3), single‐rooted to multi‐rooted (aOR 1.9), and non‐restored to restored teeth (aOR 1.4), as well as teeth with no degree of mobility compared to mobile ones (aOR 1.5) and those exhibiting BOP (aOR 1.4). Patient‐level factors included in the analysis as well as the tooth‐related factors upper/lower arch, pulp vitality, plaque, and FI were not significantly associated with PC (Table 4). Stratified for deep residual pockets only, PC occurred more often in single‐rooted compared to multi‐rooted teeth (aOR 2.3) and teeth with no degree of mobility compared to mobile ones (aOR 1.7). All other factors included were not significantly associated with the odds of PC in deep residual pockets after NSRI. (see Table S5 in the online Journal of Periodontology).

TABLE 4.

Binary logistic regression model: PC at teeth with periodontal pockets.

Modulator	β‐Coefficient (95% CI)	p‐value
PPD at REV (PPD 4–5 mm)	4.677 (3.880; 5.639)	<0.001
Number of roots (single‐rooted)	1.703 (1.444; 2.008)	<0.001
Restoration type (none)	1.234 (1.012; 1.503)	0.037
FI (0/1)	1.364 (1.034; 1.799)	0.028
Degree of mobility (none)	0.740 (0.615; 0.891)	0.001
PIREV (negative)	0.915 (0.787; 1.064)	0.251
BOPREV (negative)	1.219 (1.050; 1.415)	0.009
Vitality (positive)	1.085 (0.884; 1.333)	0.434

Note: Adjusted for the following covariates: sex, smoking, diabetes mellitus, age, and systemic antibiotic intake during NSPT. Bold indicates statistically significant values (p < 0.05).

Abbreviations: BOPREV, bleeding on probing at re‐evaluation; FI (0/1), furcation involvement with degrees 0 or 1; PC, pocket closure; PIREV, plaque index at re‐evaluation; PPD, probing pocket depth; REV, re‐evaluation.

4. DISCUSSION

4.1. Key findings and objective

The results of the present study indicate a considerable influence of tooth‐related factors on the effectiveness of NSRI as part of step 3 periodontal therapy. Besides PPD at re‐evaluation, arch, tooth type, number of roots, FI, mobility, restoration, and BOP were significantly associated with the responsiveness to NSRI. Clinically most importantly, single‐rooted teeth exhibiting deep residual pockets responded comparatively well to NSRI in terms of PPD reduction, PC, and changes of pocket category compared with multi‐rooted teeth, indicating a possible differential need for periodontal surgery.

4.2. Limitations

Several limitations of this study affecting the generalizability as well as applicability and transferability of the present results need to be critically addressed. As a noncontrolled observational study, not comparing other surgical or nonsurgical treatment protocols, the generalizability of the findings is limited. Although supervising operators underwent a calibration phase in advance, the potential impact of measurement errors must be considered when interpreting the results, regarding both improvement and worsening of sites following NSRI. Furthermore, due to incomplete data collection, the responsiveness to NSRI was assessed clinically and at the tooth level, while radiological findings, such as the presence of intraosseous defects, known to negatively affect the responsiveness to nonsurgical debridement ³¹ were not included in the current analysis. Regarding the criterion number of roots, a simplified categorization (multi‐rooted/single‐rooted) was assumed. The division into one to three roots would feign higher accuracy; however, other anatomical features such as three‐rooted premolars or two‐ and four‐rooted upper molars would be suppressed. The high mean age of the present cohort (57 ± 14 years) is not representative of the general population. The increased prevalence of periodontitis in the elderly population, ³² on the other hand, increases the applicability and transferability of the findings. NSRI was carried out in an undergraduate program under the supervision of experienced periodontists while in the majority of studies reporting on therapy outcomes of NSPT or NSRI, dental hygienists or periodontists performed the therapy. Given the consistency of the clinical outcomes of NSRI with the literature; however, the university setting in which NSRI was carried out increases transferability, especially since it might be expected that therapy outcomes attained by undergraduate students, can also be achieved by general dentists or hygienists despite varying levels of experience and skill. ³³ Further research in terms of well‐designed studies including radiographic data and patient‐related factors is necessary to evaluate the reliability and reproducibility of the presented data.

4.3. Discussion of methods and results

The results of the present study indicate a mean PPD reduction of 1.32 ± 1.79 mm after a second intervention including nonsurgical debridement across all teeth, which can be considered in accordance with previously published data of our group ¹⁸ but also with the literature reviewed by Suvan et al. ² regarding initial NSPT. Similar results can be found in a recent study by Jentsch et al., ¹⁶ addressing the adjunctive use of enamel matrix derivates (EMD) in NSRI. Herein, the PPD reduction of the control group (no adjunctive EMD) accounted for approximately 1.5 mm. Tomasi et al. reported a mean PPD reduction after NSRI of 0.9 mm after 6 months and of 1.1 mm after 9 months. ¹⁴ A subgroup analysis of our data considering only deep pockets (≥ 6 mm) revealed a mean PPD reduction of 1.08 ± 2.37 mm, which can be considered in accordance with the previously mentioned literature. The present analysis further identified significantly more PPD reduction of deep residual pockets in single‐rooted (1.43 ± 2.34 mm) compared to multi‐rooted teeth (0.88 ± 2.36 mm). Consequently, the relative portions of deep pockets persisting after NSRI were significantly elevated in multi‐rooted compared to single‐rooted teeth, which is in accordance with the results of Tomasi et al. ¹⁴

Considering changes in pocket categories, the factors FI, restoration, mobility, and positive BOP were associated with doubled portions of deep residual pockets persisting after NSRI when compared to their controls. These results are supported by previously published data on tooth‐related factors affecting initial NSPT ¹⁸ and confirm the influence of FI on the effectiveness of NSRI reported in the literature. ¹⁴ , ¹⁷

Although the mean PPD reduction in millimeters represents a widely used and unambiguous parameter to describe the effectiveness of periodontal therapy at the tooth‐ and site level, it is not able to display disease remission and therefore less closely related to the goal of therapy. ³⁴ Consequently, the variable PC was included herein to identify tooth‐related factors that influence the outcomes of NSRI with regard to the ultimate endpoint of therapy.

As recently different endpoints of periodontal therapy have been proposed, ¹¹ , ²⁹ , ³⁵ the definition of “pocket closure” used for the analysis of the present data deserves some discussion. While the current classification defines a clinical case of health in a successfully treated periodontitis patient as no presence of pockets > 3 mm with BOP or no pockets ≥ 5 mm, ²⁹ EFP guidelines suggest “no periodontal pockets > 4 mm with BOP or no deep periodontal pockets [6 mm]” as the endpoint of therapy. ¹¹ Feres et al. on the other hand, proposed a less stringent endpoint of therapy considering ≤ 4 sites with PPD ≥ 5 mm to be effective in determining disease remission after active periodontal therapy. ³⁵ This study used the definition of PC as defined by the current classification. ²⁹ The rationale for this more strict endpoint is supported by the data published by Matuliene and co‐workers, revealing comparable tooth loss rates in teeth with residual PPD of 4 mm with BOP and PPD of 5 mm without BOP on the tooth level and more than doubled odds for tooth loss in residual PPD of 5 mm compared to PPD of 4 mm on the site level (OR: 2.6 vs. 5.8). ³⁶

Regarding the entire cohort and using the aforementioned definition, the results revealed a PC rate of 40% following NSRI. These findings are poorer than those reported by Ferrarotti and co‐workers after a single episode of NSRI (67.6%). ¹⁷ However, it must be noted that in the study by Ferrarotti et al., the short healing period of 4–6 weeks after step 2 might have impacted the results. In this context, Suvan et al., concluded that the highest PPD reductions following NSPT might be expected after 3 or even 6 months. Thus, it remains unclear whether the (repeatedly) treated sites would have undergone PC even without NSRI. Furthermore, it is crucial to note that Ferrarotti et al. used the endpoint proposed by the EFP guidelines to investigate the impact of NSRI, hence assuming a PPD of 5 mm without BOP as closed. This difference in reporting most likely explains the considerably higher values of PC compared to the present results. Tomasi et al., on the other hand, defined PC as a PPD ≤ 4 mm and presented rates of 42%–53% following NSRI, which is consistent with our results. However, it is notable that in their definition of PC, BOP was not taken into account, possibly explaining slightly superior outcomes.

Subgroup and multilevel analyses of moderate residual and deep pockets revealed PC rates of 51% and 16%, which is in line with Tomasi et al. (52%–60% for moderate and 12%–17% for deep pockets). Notably, not only residual PPD after NSPT but also tooth‐related factors had a significant and clinically relevant influence on PC following NSRI. While in the entire cohort, including moderate and deep pockets, almost all of the tooth‐related factors investigated had an impact on NSRI, their influence appears to be less pronounced in deep residual pockets. In this sub‐cohort, tooth type and number of roots seem to be the dominant factors affecting treatment outcomes. In detail, incisors, and single‐rooted teeth with persisting PPD of ≥ 6 mm were closed in approximately 25%, while PC rates in molars and multi‐rooted teeth accounted for merely 10%. The decreased accessibility and cleanability of multi‐rooted teeth are related not only to their posterior location and the increased root surface area but also to anatomic variations, such as the furcation entrance area with varying degrees of root separation and root trunk lengths as well as furcation ridges and groves, ³⁷ , ³⁸ further impair the efficacy of NSRI.

Finally, mixed‐effect regression models revealed 7.3 higher chances for PC in moderate compared to deep pockets, clearly supporting the recommendations on NSRI in moderate residual pockets, as part of step 3 therapy. ¹¹ On the other hand, given the poor results for NSRI in deep residual pockets, the rationale for doing so seems questionable. Interestingly, the patient‐related factors investigated seem to play a minor role in the course of NSRI.

5. CONCLUSION

The rationale for repeated nonsurgical debridement as part of step 3 of periodontal therapy has been discussed controversially in the literature. The results of the present study indicate that besides the depths of residual PPD (moderate/deep) following steps 1 and 2 of periodontal therapy, several tooth‐related factors including tooth location in the upper/lower arch, tooth type, number of roots, FI, mobility, restoration, and BOP influence the effectiveness of NSRI. Within the limitations of this study, NSRI may decrease the need for periodontal surgery in deep residual pockets and thus reduce the invasiveness of therapy. At the same time, determining the indication for more invasive approaches as part of step 3 might be supported by considering the identified tooth‐level factors.

AUTHOR CONTRIBUTIONS

Caspar Victor Bumm, Falk Schwendicke, Matthias Folwaczny, and Nils Werner contributed to the study design and conception, data analysis and interpretation, and writing and revision of this manuscript. Christina Ern and Richard Heym contributed to the study design and conception, data acquisition, and writing and revision of this manuscript. Katrin Heck, Elias Walter, and Vinay Pitchika contributed to data analysis, interpretation, and the writing and revision of this manuscript. All authors reviewed and approved the final manuscript and agreed to be accountable for all aspects of the work ensuring integrity and accuracy.

CONFLICT OF INTEREST STATEMENT

All authors report no conflict of interest.

FUNDING INFORMATION

The study was self‐funded by the authors and their institution.

ETHICS STATEMENT

This retrospective study was approved by the Ethics Committee of the Medical Faculty of the Ludwigs‐Maximilians‐University, Munich (22‐0669).

Supporting information

Supporting Information

Bumm CV, Schwendicke F, Pitchika V, et al. Effectiveness of nonsurgical re‐instrumentation: Tooth‐related factors. J Periodontol. 2025;96:748–759. 10.1002/JPER.24-0178

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.