131. a. Non-surgery vs. Surgery

        b. Statistical methods                                     

HOME           PERIO TOPICS   

This site has not been updated since 2015. To see the complete, updated version of this site, visit PerioAbstracts.com

Rapid Search Topics

  1. Nonsurgical therapy vs Surgical
    1. nonsurgical therapy: scaling and root planing
    2. the effectiveness of scaling and root planing in removing calculus
    3. comparison of hand vs ultrasonic instrumentation
    4. healing in deep pockets after root planing
    5. masking of periodontitis
    6. response following SRP
    7. SRP + antibiotic tx could eliminate the need for sx
    8. reviews
    9. comparison of sx with non sx therapy
  2. Statistics used in periodontal research
    1. sensitivity and specificity
    2. sample size requirements
    3. regression towards the mean
    4. statistical significance vs. clinical significance
    5. the randomized clinical trial
  3. Root Caries
    1. susceptibility to root caries after perio tx
    2. prevention of root caries

What treatments does Non-Surgical Therapy include? What is scaling? What is root planing? What are the endpoints to successful root planing therapy? What studies show the effectiveness of SRP? What are some factors that can limit the effectiveness of SRP? Is SRP equally effective in molars and non-molars? Are more experienced practitioners more effective at SRP?

  1. Cobb C. Non surgical pocket therapy: Mechanical. Ann Periodontol 1996;1;443-490

  2. Buchanan A., Robertson P. Calculus removal by scaling/root planing with and without surgical access. J Periodontol 1987;58:159-163

  3. Loos B et al. Clinical effects of root debridement in molar and non-molar teeth. A 2 year follow up. J Clin Periodontol 1989;16:498-504

  4. Brayer, W et al. Scaling and root planing effectiveness: The effect of root surface access and operator experience. J Periodontol 1989;60:67-72

  5. Badersten A, et al. Effect of nonsurgical periodontal therapy. I. Moderately advanced periodontitis. J. Clin. Periodontol. 8:57-72, 1981.

  6. Badersten A, Nilveus R, Egelberg J: Effect of nonsurgical periodontal therapy. II. Severely advanced periodontitis. J. Clin. Periodontol. 11:63-76, 1984.

  7. Badersten A, Nilveus R, Egelberg J: Effect of non-surgical periodontal therapy. III. Single versus repeated instrumentation. J.Clin.Periodontol.11:114-124, 1984.

  8. Badersten A, et al: Effects of nonsurgical periodontal therapy. IV. Operator variability. J. Clin. Periodontol. 12:190 -200, 1985.

  9. Badersten A, et al: Effect of nonsurgical periodontal therapy. VIII. Probing attachment changesrelated to clinical characteristics. J. Clin. Periodontol. 14:425-432, 1987

  10. Greenwell H, Bissada NF, Dodge JR. Disease masking: A hazard of nonsurgical periodontal therapy. Perio Insights December 1998:14-19.

  11. Matthews D. Conclusive support for mechanical nonsurgical pocket therapy in the treatment of periodontal disease. How effective is mechanical nonsurgical pocket therapy? Evid Based Dent. 6(3):68-9. 2005

  12. Ramfjord S, et al. Results of periodontal therapy related to tooth type. J. Periodontol. 51:270-273, 1980.

  13. Pihlstrom BL, Oliphant TH, McHugh RB: Molar and nonmolar teeth compared over 6.5 years following two methods of periodontal therapy. J. Periodontol. 55:499-504, 1984.

  14. Claffey N, Shanley D: Relationship of gingival thickness and bleeding to loss of probing attachment in shallow sites following nonsurgical periodontal therapy. J Clin Perio 13:654, 1986.

  15. Loesche W, Soehren S, et al. Nonsurgical treatment of patients with periodontal disease. Oral Surg, Oral Med, Oral Pathol 1996; 81: 533-543.

  16. Loesche WJ, Giordano JR, Soehren S, Kaciroti N. The nonsurgical treatment of patients with periodontal disease: results after 6.4 years. Gen Dent. Jul-Aug;53(4):298-306; 2005

  17. Drisko CH. Nonsurgical periodontal therapy. Perio 2000 25:77-88,2001. (Review)

Is Surgical Treatment more effective than non surgical treatment?

  1. Lindhe et al. Long term effect of surgical/non surgical treatment of periodontal disease. J Clin Periodontol 1984;11:448-458

  2. Caffesse, R et al. Scaling and root planing with and without periodontal flap surgery. J Clin Periodontol 1986;13:205-210

  3. Philstrom B, et al: Comparison of surgical and nonsurgical treatment of periodontal disease. A review of current studies and additional results after six years. J Clin Perio 10:524-541, 1983.

  4. Serino, G et al. Initial outcome and long term effect of surgical and non surgical treatment of advanced periodontal disease. J Clin Periodontol 2001;28:910-916

What is the critical probing depth?

  1. Lindhe J, Socransky SS, et al. "Critical probing depths" in periodontal therapy. J Clin Periodontol 9:323-336, 1982.

Statistical Methods

What is sensitivity and specificity? What is positive predictive value and accuracy? What is Prevalence? What is negative predictive value? What is equivalence and superiority? What is regression towards the mean?

  1. D. Brunette: Critical Thinking: Understanding and Evaluating Dental Research 2nd Edition Quintessence Publishing Co, Inc ISBN 978-0-86715-426-9; 2007; pp 163-184, pp192-193

  2. Hujoel PP, Moulton LH, Loesche WJ : Estimation of sensitivity and specificity of site-specific diagnostic tests. J. Periodontal Res. 25:193-196, 1990.

  3. Gunsolley JC, Elswick RK, Devenport JM. Equivalence and superiority testing in regeneration clinical trials. J Periodontol 69:521-527, 1998.

  4. Gunsolley J. Equivalence, superiority, and negative clinical trials. J Periodontol 69:608,1998

  5. Duke SP, Garrett S. Equivalence in periodontal trials: A description for the clinician. J Periodontol 69:650-654, 1998.

  6. Gunsolley, J et al: Is loss of attachment due to root planing and scaling in sites with minimal probing depths a statistical or real occurrence? J Periodontol 2001;72:349-353

What is the difference between statistical significance and clinical significance?

  1. Rethman MP, Nunn ME. Clinical versus statistical significance. J Periodontol 70:700-702,1999.

  2. Greenstein G, Lamster I. Efficacy of periodontal therapy: Statistical versus clinical significance. J Periodontol 71:657-662, 2000.

  3. Hujoel PP, Armitage GC, Garcia RI. A perspective on clinical significance. J Periodontol 71:1515-1518, 2000.

How can we evaluate the quality of a RCT? What is a meta-analysis and how accurate are their results? What are the major pitfalls in clinical trials design?

  1. Jeffcoat MK. Principles and pitfalls of clinical trials design. J Periodontol 1992;63:1045-1051

  2. Cohen ME, Ralls SA : False positive rates in the determination of changes in probing depth related to periodontal measurements. J. Periodontal Res. 23:161-165,1988.

  3. Hujoel PP, Moulton LH : Evaluation of test statistics in split-mouth clinical trials. J. Periodontal Res. 23:378-380, 1988.

  4. Lesaffre E, Garcia Zattera MJ, Redmond C, Huber H, Needleman I; ISCB Subcommittee on Dentistry. Reported methodological quality of split-mouth studies. J Clin Periodontol. Sep;34(9):756-61; 2007. Review.

  5. Needleman I, Worthington H, Moher D, Schulz K, Altman DG. Improving the completeness and transparency of reports of randomized trials in oral health: the CONSORT statement. Am J Dent. Feb;21(1):7-12; 2008

  6. Montenegro R, Needleman I, Moles D, Tonetti M. Quality of RCTs in periodontology--a systematic review. J Dent Res. Dec;81(12):866-70; 2002. Review.

  7. Esposito M, Coulthard P, Worthington HV, Jokstad A. Quality assessment of randomized controlled trials of oral implants. Int J Oral Maxillofac Implants. Nov-Dec;16(6):783-92; 2001. Review.

Root Caries

What factors may influence the incidence of root caries? Does periodontal treatment increase root caries?

  1. Ravald N, Birkhed D, Hamp SE. Root caries susceptibility in periodontally treated patients. Results after 12 years. J Clin Periodontol. Feb;20(2):124-9; 1993

  2. Paraskevas S, Danser MM, Timmerman MF, van der Velden U, van der Weijden GA. Amine fluoride/stannous fluoride and incidence of root caries in periodontal maintenance patients. A 2 year evaluation. J Clin Periodontol. Nov;31(11):965-71; 2004

  3. De Soete M, Dekeyser C, Pauwels M, Teughels W, van Steenberghe D, Quirynen M. Increase in cariogenic bacteria after initial periodontal therapy. J Dent Res. Jan;84(1):48-53; 2005

What treatments does Non-Surgical Therapy include? What is scaling? What is root planing? What are the endpoints to successful root planing therapy? What studies show the effectiveness of SRP? What are some factors that can limit the effectiveness of SRP? Is SRP equally effective in molars and non-molars? Are more experienced practitioners more effective at SRP?

Cobb 1996                    ARTICLE

P: A literature review regarding: 1) progression of untreated periodontal disease which offers a basis for comparing the effects of mechanical non-surgical treatment 2) the effect of scaling and root planing on specific clinical parameters and selected biologic factors (sub-g microbial flora, cementum and root surface roughness), 3) root preparation using power driven instruments.

Disc:

Reduction in % of motile microbes and spirochetes

Increase in cocci and non-motile microbes

Progression of untreated periodontal disease

Distribution of periodontal disease and factors affecting progression

Annual tooth loss rates in patients with and without periodontal therapy

Mechanical non-surgical therapy

The 1989 Proceedings of the World Workshop in Clinical Periodontics defined scaling as “Instrumentation of the crown and root surfaces of the teeth to remove plaque, calculus and stains from these surfaces. Root planing was defined as “A definitive treatment procedure designed to remove cementum or surface dentit that is rough, impregnated with calculus or contaminated with toxins or microorganisms.

Effectiveness of scaling and root planing

Effectiveness of calculus removal with or without flap

Amount of cementum removal during scaling and root planing

Comparison of manual instrumentation to sonic/ultrasonic instrumentation

Buchanan 1987     (efficacy)                ARTICLE

P: To evaluate the presence and extent of calculus on subgingival root surfaces of teeth that received SRP alone, SRP with modified Widman flap, or no treatment.

M+M: 10 pts (28-62 years old) that had at least 3 teeth that needed extractions due to severe periodontitis. Teeth in each pt randomly assigned to one of three groups: SRP alone (29 teeth), SRP w/ MWF (35 teeth), or not tx before extraction (22 teeth). SRP completed with ultrasonic scalers and hand instruments. Teeth were extracted and stained with methylene blue to disclose and looked under 10x stereomicroscope. Each surface was examined to determine the pocket depth, area of root surface exposed to the pocket , and amount of pocket area showing retained calculus. Calculus-positive teeth (CPT) and surfaces (CPS), and percentage of pocket area occupied by calculus (C/A) were derived for each group.

R: The mean treatment time per tooth for SRP alone was 12.9 ± 2.1 minutes, for the SRP w/flap it was 11.5 ± 2.0 minutes per tooth plus an additional 6.6 ±1.9 minutes per tooth for incision, reflection, and removal of granulation tissue, to a total treatment time per tooth of 18.1 ±2.0 minutes.

Pocket depth and area: were similar for all three groups averaged 5.9 mm and 26.0 mm2 respectively.

Calculus:

BL: SRP w/ flap proved to be more effective than SRP alone in reducing the percentage of calculus positive teeth and surfaces. The advantage of SRP w/flap was most dramatic on anterior and premolar teeth, and on facial and lingual surfaces. In pocket depths >6 mm, calculus positive surfaces after SRP alone increased linearly and SS and in pockets deeper than 8 mm, calculus-positive surfaces averaged about 45% after SRP alone whereas after SRP w/ flap they remained constant at about 17%.

Loos 1989                      ARTICLE

P: To report on the longitudinal observations over 2 years of the clinical effects of root debridement in molar furcation sites, molar flat-surface sites and non-molar sites.

M&M: 12 pts with generalized perio included. Sites grouped as either molar furcation, molar flat-surface or non-molar as well as being shallow (<3.5mm), moderate (4.0-6.5) or severe (>7mm). Pts received OHI 3wks prior to the initial therapy, which consisted of only 2 sessions of SRP (1 each jaw). Every 3 months, pts received supra-g prophy and OHI and had their clinical parameters recorded with an electronic, pressure-sensitive probe set at 0.5N for a 2-year period.

R: Initially moderately deep and deep molar furcation sites responded less favorably to therapy compared to non-molar sites and molar flat-surface sites of similar PD. Initial improvements in PDs for moderately deep and deep molar furcation sites were limited and also tended to revert during the observation interval. For moderately deep surfaces at molar flat-surfaces and non-molar sites, the 3-month gain in PAL from 0.2-0.4mm, followed by a return to baseline. There was no change in PAL in furcation sites at 3 months, followed by a gradual mean loss amounting to 0.8mm at 24 months. 25% of molar furcation sites had PALoss, versus 7% of non-molar sites and 10% for molar flat-surface sites. Reduction in BOP was primarily noted in shallow sites. BOP in moderately deep and deep sites remained high.

BL: Furcations require additional treatment after SRP.

Cr: Only 12 pts. 2 years without addressing sub-g sites is not a realistic clinical situation (hopefully). Initial therapy only consisted of SRP (no elimination of overhangs, occlusal adjustment, etc). No sx pocket elimination performed. Therefore, this study cannot be compared to many traditional perio lit maintenance studys.

Brayer 1982                     ARTICLE

Purpose: To investigate 1) if access to the root surface affects an operator’s ability to effectively scale and root plane teeth and 2) if operator skills as measured by experience level affect the ability to effectively perform SRP.

Materials and methods: 29 patients with 114 hopeless teeth were included in the study. Total of four operators performed SRP. Two fully trained Board certified periodontists (EL-1) and two 2nd year perio residents (EL-2). SRP was performed either as a closed procedure or after flap access. Unscaled teeth were retained as controls. PDs and calculus index were recorded. No time limits were placed but time was also recorded. Teeth were then extracted and root surfaces were then microscopically evaluated for residual calculus by a periodontists that did not participate in the operating procedures.

Results: Time to elevate the flap: 6.5 min/tooth (EL-1) vs 7.8 min/tooth (EL-2)

Open SRP time: 5.3 min/tooth (EL-1) vs 6.7 min/tooth (EL-2)

Closed SRP time: 8.1 min/tooth (EL-1) vs 9.5 min/tooth (EL-2)

Calculus free areas in open SRP: 96.4% EL-1 vs 91.2% EL-2

Calculus free areas in closed SRP: 86.2% EL-1 vs 65.7% EL-2

In shallow periodontal pockets no significant difference between groups in % of residual calculus, in pockets 4-6mm open SRP resulted in 92% of calculus free areas comparing to 71% with closed SRP.

The more experienced operators produced significantly greater number of calculus-free root surfaces (89% vs 74%) in 4-6mm pockets and >6mm pockets (91% v 69%). No difference in 1-3mm pockets.

Conclusion: In single rooted teeth with PDs more than 4mm SRP with flap access was more effective and the more experienced operators produced fewer root surfaces containing residual calculus.

Badersten 1981(I)                     ARTICLE

Purpose: To compare the effect of hand versus US instrumentation in patients with moderately advanced periodontitis.

Materials and methods:

Results:

BL: Non-surgical treatment of non-molar teeth with moderate PD is effective, with no difference between HI and US.

Badersten 1984(II)                     ARTICLE

P: To study the healing events in patients with pockets up to 12 mm deep, to compare hand to ultrasonic instruments

M&M: 16 patients, with severe periodontitis, 4-10 teeth in each patient (incisors, canines, premolars), probing depths > 5 mm with calculus & bleeding on probing on > 2 aspects each tooth. OHI over 2-3 visits, extra instructions were given as needed. 3 months after OHI, measurements were taken, then preparation with either ultrasonic or hand instruments in a split mouth design was performed, measuring plaque index, bleeding on probing, attachment level and recession were recorded and repeated every 3 months up to 24 months, instrumentation was performed again at 6 & 9 months. (3 sessions overall)

R: Total of 852 surfaces were treated

- Plaque index: NSD between 2 instruments, higher with reduction with one of the two investigators

- Bleeding On Probing: No change between initial exam & 3 month after OHI. It did, however, decrease after instrumentation (84-90% decreased to 14-18%). Similar decrease in bleeding on probing for both investigators and instruments.

- PD: Initial : 5.5-5.8 mm. 3 months: 5.1-5.3 mm. 12 months: 3.6-3.9 mm. 24 months: residual probing depths decreased irrespective of operator or instrument for all probing depths; 43 surfaces had PD > 7 mm (86% reduction)

- Recession: limited gingival recession during the 3 months but at 12 months 1.6-1.8 mm recession

- PAL:

Residual PD: Loss of attach of ≥1.5 mm in surfaces with residual PD ≤2.5 mm and ≥6 mm. Gain of attachment of ≥1.5 mm for 24-31% of surfaces with residual probing depths of 3-4.5 mm.

Initial PD: Gain/Loss of PAL 1.5 mm or more/less. Found Majority of initial PD < 4-4.5 mm loss of PAL, while PD > 6-6.5 mm gain of PAL.

- NSSD between ultrasonic & hand instruments, but 1 operator used less time with ultrasonic.

BL: Deeper sites had more: recession, residual probing depths, gain in attachment. Sites < 3.5 mm lost attachment, sites > 8 mm gained attachment. Authors mentioned that decision to proceed with Surgery Treatment should be postponed until 6-9 months after initial instrumentation.

Badersten 1984 (III)                     ARTICLE

P: To compare the effect of a single session of subgingival instrumentation using ultrasonic scaler with 3 sessions separated by 3 months.

M&M: 13 patients (30-55yrs) with PD up to 5-11mm were included in the study. Incisors, canines, premolars only, 6-10 teeth in each patient were used in split mouth design. OHI, US debridement with one side received 1 session and the other received 3 sessions at 0, 3, 6 months. PI, BOP, PD, CAL, and recession were recorded at baseline and every 3 months for 24 months (blinded examiner).

R: Avg PD prior to txt 5.8 mm for surfaces to be instrumented once and 5.9 mm for surfaces to be treated 3 times.

PI was sig reduced with both forms of txt.

Initial BOP 78%-80% reduced to 15-20% for both treatments, maintained throughout observation.

PD, similar decrease for both groups: initial PD score 5.5-5.9mm. PD reduced to 4.1-4.2 mm after 3 months, and a further reduction to 3.5-3.7 mm was seen at 9-month exam.

Recession: occurred during 1st 9 months of study and stabilized at an average of 1.8mm.

CAL gain: A mean of 0.4mm if seen after 3 months and remained constant throughout the study.

Sites with <3mm PD showed 0.4-1mm of CAL loss, sites >8mm gained 0.9-2.8mm. Only 25% of PD between 3-4.5 mm gained; >1.5mm of attachment.

Initial healing was the same for both groups. Full effect of healing was not obtained until 6-9 months after instrumentation.

BL: Similar results were recorded with 1 or 3 sessions of SCRP. Repeated instrumentation is of limited value. Clinical parameters remained unchanged during 24 months and did not show recurrence of disease. Full effect of healing not obtained for 6-9 months after instrumentation.

Limitation: molar were excluded from the study

Badersten 1985 (IV)                     ARTICLE

P: To compare the results achieved by other operators in treating severe periodontitis with a single instrumentation.

M&M: 20 patients (28-64 years old) with severe periodontal disease were included. Incisors, cuspids, and premolars in maxilla or mandible were used (each patient had between 6-10 of these teeth). Pockets of at least 5mm with calculus and BOP on at least 2 aspects of each tooth. OHI 2 or 3 times were given during the first month. Additional OHI provided as needed. Periodontal pockets were debrided immediately following the initial examination. Split mouth design was used, one half was assigned to a periodontist, and the other half to a hygienist (5 different hygienist participated); both using US and/or HI for instrumentation. Time of instrumentation was recorded. Pl, BOP, PD, PAL were recorded every 3 months for 24 months.

R: 1056 sites were treated. Significant positive trend was seen in all patients after treatment (improvement at 3 months and essentially the same or slightly better thereafter). Reduction in PI, PD, BOPS with no difference b/w the 2 operator types. The periodontist sites showed slightly greater gains in PAL and less recession than hygienist sites. The average time spent in instrumentation per tooth was 9-12 minutes.

CL: The results of non-sx periodontal therapy b/w different operators were minimal. Overall, it was observed that incisors, cuspids and premolars might be maintained by plaque control and single instrumentation.

Badersten 1987                     ARTICLE

P: To observe the effect of non-sx perio tx on probing attachment changes related to clinical characteristics.

M&M: 1688 proximal sites in non-molar teeth from 49 pts were monitored for 24 mo. following OHI and root debridement. Gain and loss of CAL was recorded, for incisors, canines, and PM, max or mandibular, presence/absence of endo, and for surfaces w/ and w/o the presence of root concavity/furcation involvement. Changes were compared to: initial PD, radiographic bone score, depth of osseous defect (radiographically), and widened PDL.

R: 12% had probing attachment gain, 4% had probing attachment loss (PAL).

CL: outcome of debridement and maintenance in proximal surfaces of non-molar teeth is not compromised by the severity of the initial soft tissue or bony lesion.

More attachment gain is assoc w/ mand cuspids and premolars, deeper initial PD, deeper osseous defects, widened PDL, and w/ no-root concavities/furcas.

Greenwell 1998                    ARTICLE

P: To describe situations in which non-surgical therapy can obscure the disease.

D: Periodontal disease masking results when marginal tissue at 3-4 mm within the gingival margin appears healthy, while apical tissue is unhealthy. This can lead to attachment and bone loss, hidden by the relatively non-inflamed appearance of the gingiva. This is most common after SRP where calculus in the apical part of pocket is missed. Primary goal of post-S/RP is to evaluate for inadequate calculus removal, manifested by residual inflammation, BOP or suppuration. The additional SRP should be directed at calculus removal. Prescribing antibiotics or antimicrobials after SRP would further mask unresolved disease. The most appropriate time for antibiotic therapy is after removal of all calculus & plaque-retentive factors, which is post-surgical, not post-SRP. Reports of abscesses post-prophy demonstrate disease masking. The concept of “soft tissue management” is questionable, since it’s the root, not the soft tissue that needs treatment. Proper non-surgical therapy should include meticulous SRP.

Recognizing disease masking: Absence of BOP & improvement/return of radiographic crestal lamina dura may be the best clinical signs of improved health. Although absence of lamina dura does not mean disease activity present, its reestablishment indicates that disease progression has been halted. Be suspicious of an absent lamina dura. Delayed BOP could be a sign of disease masking and its importance should not be minimized.

Case Report: A 43 year-old healthy patient who underwent surgical therapy for posterior quadrants and was put on 9 SPT/year x 3 years. Despite clinical healthy-looking gingival tissue on anterior areas, BOP was erratic & present about 50 % of time on the area; also, no return in crestal lamina dura & no increase in radiographic density. 2 mm increase in PD was found after 3 years & 1 mm AL around teeth 7 & 8. Sx recommended & burnished calculus found. At 3 years after surgery, increase in bone density was seen, no BOP & Attachment level was stable.

BL: Inadequate SRP or inappropriate antibiotic therapy can potentially mask an area with perio disease. Comparing subsequent exams for BOP & CAL is essential to detect unresolved or recurrent disease.

Matthews 2005                    ARTICLE

P: Systematic review to answer the question, “How effective is mechanical nonsurgical pocket therapy?”

M+M: The authors used the Cochrane Oral health Group List of Systematic Reviews in Dentistry, Database of Abstracts of Reviews of Effectiveness, Medline, Embase, and Scisearch. No date or language restriction was imposed. Reference lists of located reviews were checked for additional references. 12 reviews were selected for inclusion, and a narrative appraisal of the reviews was conducted.

R: Implications for clinical practice were identified.

Initial Therapy:

Maintenance Therapy:

BL: Existing evidence in the form of systematic reviews provides conclusive support for the beneficial effect and efficacy of mechanical nonsurgical pocket therapy in the treatment of periodontal diseases.

Ramfjord 1980                     ARTICLE

P: To determine the influence of tooth type on the results of periodontal treatment over 8 years of a longitudinal study.

M&M: Data from a previous periodontal therapy involving 78 patients over 8 years (Knowles et al., 1979) was analyzed with regard to effect of tooth type on treatment results. Initial probing depths (1-3mm, 4-6mm, and 7-12mm) were used as an expression of the severity of the disease. The dentition was divided into six tooth types: Maxillary molars, mandibular molars, maxillary premolars, mandibular premolars, maxillary anteriors, and mandibular anteriors. Probing depths and attachment levels were measured annually.

R: Tooth type has little influence on the response on the periodontal treatment outcome. Reduction in probing depths and potential for attachment level gain were slightly better in anteriors than molars teeth. Poorest results were seen in maxillary premolars and molars, and one of the reasons could be possibly due to furcation complications. The trend was for probing depths to return more rapidly in deep molar pockets than deep anterior pockets. Anterior teeth sustained gain in attachment better than the rest of the teeth in the arch.

BL: Prognosis for treatment of periodontal pockets is good for all tooth types, and this applies to moderate as well as to deep pockets.

Previous Critique: Measurements taken did not really account for furcation involvement, since the straight buccal probing depths were taken from root prominences rather than furcation areas.

Pihlstrom 1984                     ARTICLE

Purpose: To investigate the periodontal response of molar and nonmolar teeth to either SRP alone or SRP followed by modified Widman flap (MWF).

Materials and methods: 17 subjects 22-59 years old. After initial scoring of the clinical measurements thorough SRP and OHI were performed by a periodontist in training. Overhangs and defective restorations were corrected and occlusal adjustment was performed when needed. Two quads (one maxillary and one mandibular) per patient were selected to receive MWF. Periodontal prophylaxis was then performed 3-4 times/year. Hopeless teeth were not extracted in the initial treatment, but during the maintenance phase if needed. Clinical measurements were obtained prior to any therapy, 6 months after completion of the therapy and then annually for 4 years. PDs and AL were recorded at 6 sites/tooth.

Results: Of the 17 initial subjects 10 remained as participants after 6,5 years.

PDs 4-6mm: There was 0.4mm less pocket depth at baseline for nonmolar teeth treated with SRP and 0.27mm for nonmolar teeth treated with MWF comparing to molars treated with these procedures. This difference increased throughout the study and 6,5 years nonmolar teeth had an average of 1mm less PD irrespective of typed of procedure performed.

Attachment loss was greater at pretreatment baseline was greater for molar teeth (0.74mm more) and tended to remain the same over the 6,5 year period.

PDs of 7mm or more: For teeth treated with SRP there was a difference of 1.86mm and 2.32mm in PD only at 2 and 3 years post-treatment respectively (deeper in molars). No SSD between the teeth group in other time intervals but there was a tendency for more shallow PDs in nonmolar teeth. Teeth treated with MWF nonmolar teeth had 0.41mm less PD at baseline and the magnitude of this difference increased dramatically over the 6,5 years (2.36mm at 6,5 years with 1.22mm standard error). Differences in AL were only SSD at 2 years after flap procedure with nonmolar teeth having 0.93mm less attachment loss than molars.

Tooth loss: Total tooth loss was 4%. 8/19 teeth before therapy was completed. 11/19 after therapy was completed (2.5% of teeth receiving therapy). 7/11 max molars, 2/11 mand molars, 1/11 deciduous cuspid and 1/11 mand lateral incisor. 5/11 were lost after SRP and 6/11 after SRP + MWF.

Conclusion: 1) Both procedures were effective in treating periodontitis in terms of maintenance of CAL on molar and nonpolar teeth

2) For initial PDs of 4-6mm, there was greater PD and a more apical CAL on molar than nonmolar teeth treated by either method

3) For pockets initially 7mm or more, the flap resulted in less PD on nonmolars than molars but there was no difference in CAL between tooth types for either method of therapy.

Claffey, Shanley, 1986                     ARTICLE

Purpose: To examine the relationship of gingival thickness, bleeding, and the tendency for attachment loss in shallow buccal sites (< 3.5mm PD) following non-surgical periodontal therapy.

Materials & Methods:

Results:

BL: Sites with bleeding prior to instrumentation did not seem to lose attachment. Thin, non-bleeding sites seem to be ones primarily associated with this probing attachment loss.

Loesche 1996      (sx prevention)              ARTICLE

P: To determine whether the short-term use of systemic antimicrobials (metronidazole or doxycycline) and locally delivered antimicrobials (metronidazole, chlorhexidine) in patients with advanced forms of periodontal disease could prevent access surgery.

M+M: Inclusion criteria: presence of an anaerobic infection, spirochetes greater than 20% of the microscopic count and the hydrolysis of benzoyl- DL-arginine napththylamide (BANA-positive reactions) in at least 3 of 4 subgingival plaque samples taken from the site in each quadrant that had the greatest PD. Subjects examined for number of teeth in need of surgery, those that had >4 teeth needing to be extracted were kept in the study. Study was conducted in a double-blind fashion in which patients were randomly assigned to antimicrobial treatment groups. No patient would receive more than two rounds of systemic antimicrobial treatments or more than three rounds of local antimicrobial treatments about individual teeth. A placebo treatment would be used only in the first round of systemic treatment or in the first round of local treatment. Any patient or tooth still in need of treatment after the first round of the systemic or local treatment would be retreated with the opposite medication from what they had received in the first round. All patients progressed through the first round of treatment. After SRP the patients were randomly assigned to receive either placebo, metronidazole, or doxycycline, for 2 weeks unsupervised at home. If anyone had > 6 teeth in need of surgery or extraction after first round of systemic medication, they were retreated with systemic medication. If they had 6 or less teeth, they were treated with ethyl cellulose (EC) films containing either no addition (placebo), 20% metronidazole, or 20% chlorhexidine. If they had no teeth in need of surgery, SPT every 3 months. 90 of 125 patients initially recruited completed all phases of the study and entered into the recall maintenance program.

Loesche 2005                     NO ARTICLE

B: In a previous study involving patients at the dental clinic of a hospital in Detroit, 87% of teeth that initially had been recommended for surgery or extraction were spared either treatment through a combination of debridement and short-term usage of antimicrobial agents.

P: The purpose of the current study was to determine the changes that occurred to these teeth after a median of 6.4 years in the maintenance phase of treatment.

M&M: 90 Patients were scheduled for maintenance therapy at 3-month intervals over a period of 6.4 years. Subjects were diagnosed as having an anaerobic periodontal infection if 3 or more of their plaque samples contained spirochetes exceeding 20% of the microscopic count and were capable of hydrolyzing the synthetic peptide benzoyl-DL-arginine naphthylamide (BANA*-positive).

Treatment phase: Pts with 4 or more teeth requiring sx were entered in the protocol.

After debridement, the patients were randomly assigned in a double-blind design to receive metronidazole (500 mg twice daily), doxycycline (100 mg daily), or placebo tablets/capsules for two weeks. Re-eval after 4-6 weeks. Pts that required surgery on 1-6 teeth were treated with locally delivered antimicrobial agents and > 6 teeth requiring surgery were retreated with systemic agents. No patient received more than two systemic treatments and 3 local treatments. Pts were scheduled for debridement every 3 months.

Maintenance phase: q3 months. OHI given, PD and AL, BOP, root topography and nature of bony defect via x-rays, mobility were assessed. Double-blind design prophylactic antimicrobial with metronidazole 500 BID x1wk or placebo was given.

R: 10-15% of the 90pt who entered the maintenance phase were lost.

The results during the maintenance phase were as follows:

55% - no new surgical needs or reduced surgical needs

24% - new surgery or extraction recommended for either 1 or 2 teeth

15% - new surgery or extraction recommended for 3 or 4 teeth

6% – Extraction recommended of at least 8 teeth

C: These findings indicate that an antimicrobial regimen reduced the initial surgical needs of patients by approximately 85%, and this result can be sustained in the maintenance phase by home care, periodic sessions of SRP and annual prescriptions of 1 week of metronidazole.

BANA test*= examines dental plaque, measuring the presence of an arginine hydrolase possessed by 3 anaerobic species associated with periodontal infections (P. Gingivalis, T. denticola, T. forsuthensis).

Drisko 2001(Review)                     ARTICLE

P: To review literature that addresses the non-surgical approach to treat periodontal disease.

Discussion:

Anti-infective therapy: Successful periodontal therapy depends of the elimination of pathogens. Since perio disease primary etiology is plaque, and most of the pt are not skilled in removing plaque, periodically professional cleaning is indicated. Includes Mechanical and chemotherapeutic approaches. Debridement is performed to produce a root that is biologically acceptable for a healthy attachment.

Risk Factor: Non-compliance or no regular maintenance care. Insufficient debridement. Sys disease. Genetics. Smoking. Furcation involvement. Pockets >5mm may have better result with sx.

Principles to control disease: Plaque control care alone won’t provide a good long term result. Debridement should perform every 3 months. Mechanical therapy: Modified ultrasonic tips reach furcations and pockets >5mm better. Manual and power driven instruments are equally effective. Power driven instruments or it combination with manual may produce the best overall result.

Antimicrobial agents: Due to pt inability of plaque control chemotherapy is often needed. Indicated when mechanical debridement alone may not be effective, especially deep pockets. Poor plaque control, bacteria can’t be eradicated with out antibiotic therapy (A.a). Mouth rinses and dentifrices can reduce plaque accumulation. Sub-gingival re-population occurs within 40-60d with poor supragingival plaque control and 120-240d with a good control.

Local drug delivery: Rationale is to locally kill or reduce the microbes in the pocket. Radvar compared Tetracycline fiber, metronidazole gel and minocycline gel and all improved CAL without difference b/w them. Atridox has showed CAL gain of 0.8mm and PD reduction 1.3mm.

Antibiotics: For cases that not respond as well as expected. Radiographic bone fill and periodontal regeneration has been reported with SRP + Sys Ab in patients with LAP.

Host Modulation: Periostat: Block enzymes associated with alveolar bone and connective tissue loss. Clinical differences are small, but have shown more PD reduction and more CAL.

Total health: Periodontal disease is associated with diabetes, cardiovascular disease, pre-term low birth weight babies.

BL: Nonsurgical therapy remains the cornerstone of periodontal treatment. Attention to detail, patient compliance and proper selection of adjunctive antimicrobial agents for sustained plaque control are important elements in achieving successful long-term results. Frequent re-evaluation and careful monitoring allows the practitioner the opportunity to intervene early in the disease state, to reverse or arrest the progression of periodontal disease with meticulous nonsurgical anti-infective therapy.

Is Surgical Treatment more effective than non surgical treatment?

Lindhe 1984                     ARTICLE

P: To further analyze the role played by the patients’ self-performed plaque control in preventing recurrent periodontal disease and to assess the periodontal conditions of this group of patients 5 years after completion of active treatment with special emphasis on sites with initial pocket depth >3mm.

M&M: 15 subjects w/ mean age of 47.9 selected at random. Following baseline exam, all patients were subjected to treatment utilizing a split mouth design. One side of mouth had SRP w/ a modified Widman flap while the contralateral quadrants were treated by subgingival SRP. Following tx the pts were maintained every 2 weeks with professional cleaning for 6 months. For the following 18 months the recall was extended to 3 months. Pts were evaluated for PI, GI, PD, and CAL at baseline, 6, 12, 24 months. Following 24 months the recall appointments were extended to 4-6 months and maintenance was restricted to OHI and professional supragingival cleaning. Further subgingival instrumentation was avoided. 36, 48, and 60 months after active treatment the quality of the patient’s self-performed plaque control was assessed. At 60 months PI, GI, PD, and CAL were measured.

R: Patients were divided into total sample, subgroup 1 (patients who had excellent standard of OH during the 5 year period), and subgroup 2 (patients who failed to maintain proper standard of oral hygiene). No obvious difference in attachment level alteration between the two groups but loss of attachment was more prevalent at interproximal than buccal surfaces. Subgroup 1 only had loss of attachment in 2-3% of sites where as 95% of the sites remained unchanged or gained attachment. There were no differences between the 2 treatment groups or between interproximal and buccal sites. Subgroup 2 had no sites with a gain in attachment but loss of attachment was found in 20% of the sites. Subgroup 1 had 85% of initial deep pockets ≥4mm reduced (majority more than 2mm) which was less evident in subgroup 2. Gain of attachment occurred more frequently in subgroup 1 than subgroup 2. Subgroup 2 had more sites with clinical attachment loss in the quadrants treated surgically than the ones treated with SRP alone (36% vs 18%).

BL: Oral hygiene has a decisive influence on long-term result of treatment of periodontal disease. Pts who consistently had a high frequency of plaque-free tooth surfaces showed few signs of recurrent gingivitis, increased PD, or additional CAL. The opposite is true for pts with poor oral hygiene. Initial pocket depth ≥3 mm responded equally well to the non-surgical as the surgical mode of treatment. The critical determinant in periodontal therapy is the removal of inxn and debridement of the root surface is properly performed.

Caffesse 1986                     ARTICLE

P: To evaluate the effectiveness of SRP on calculus removal with or without the use of periodontal flap for access.

M&M: 21 patients, 29-88 years old, with at least 6 teeth, no history of periodontal treatment who were planned for extractions and immediate denture placement were included. Pre-operatively: calculus was scored and PDs were measured. A mark was placed at the level of the free gingival margin to allow differentiation between supra-g and sub-g calculus. Prior to extraction, 2/6 teeth received sc/rp, 2 of the remaining 4 teeth received sr/rp +gingival flap, the rest 2 were left unscaled and served as control. Maxillary and mandibular incisors, cuspids, bicuspids and first and second molars were included. The teeth were extracted and residual calculus was examined by a stereomicroscope. Total number of teeth was 127 (43 scaled, 42 flapped and 42 control teeth).

R: % of calculus free surfaces after treatment

Pocket depth

SC/RP alone

Flap+SC/RP

1-3mm

86%

86%

4-6mm

43%

76%

>6mm

32%

50%

NSSD between anterior and posterior teeth. SSD for % of remaining calculus and PD (probability of leaving calculus increases as PD increases). Areas where residual calculus was found was apical to restorations, furcations, developmental grooves, as well as the CEJ area.

CON: Periodontal flaps for access provide a means for greater reduction of residual calculus in PDs>3mm. The % of residual calculus is related to PD, despite treatment approach. Anterior and posterior teeth respond similarly. Despite open flap approach residual calculus was still found to be significant.

Pihlstrom 1983       ( Comparisons)             ARTICLE

P: A review comparing surgical versus non-surgical treatment, with additional data from a study comparing these 2 methods of therapy over 6.5 years

Review of literature:

Wasserman 1978: 7.1% loss rate for treated teeth in 600 pts over 22 yrs. 40% of patients had surgical intervention (pts who appeared to be more susceptible to disease had more surgery). Tooth retention was more related to susceptibility to recurrent disease or case type than method of therapy.

McFall 1982: 100 surgically treated and maintained pts, 832 teeth, maintenance for 15-29 years. 16% of teeth were lost. 2,627 teeth treated non-surgically, 9.9% were lost (does not mean less teeth are lost w/ non-surgical but that more advanced disease is treated more aggressively). However, 23 of the 100pts accounted for 80% of the tooth loss, so 77% of the pts lost an avg of 0.68teeth.

Ross and Thompson 1971: 180 pts followed for 2-20 years, 15 pts accounted for 60% of tooth loss. These findings suggest that disease activity is more important in retention/prognosis than therapy delivered.

Nyman 1975: Perio surgery w/o a regular maintenance program and OHI will fail. Recall of every 3months is sufficient to maintain effects of therapy (Ramjford 1982).

Listgartenand and Levin 1981: Large subgingival population of spirochetes and motile rods precede detectable CAL loss.

M+M: Longitudonal study analyzing the 6.5 year results. 17 pts (25-59 years old) diagnosed w/ moderate-advanced periodontitis (PDs 1-14mm, 453 teeth) were utilized in a split-mouth design study to compare the effects of SRP alone and combined with MWF surgery. Study started w/ 17 subjects but only 10 pts were available at 6 ½ years. Time for SRP was 2 hours 3-4 appointments Data were collected at baseline, 6 months after treatment, every year up to 4 years, and then at 5.5 and 6.5 years. Recalls for all pts were 3-4 times/year, OHI reinforced.

R:

BL: SRP alone or in combination with MWF surgery resulted in sustained decreases in gingivitis, plaque and calculus but neither procedure appears to be superior with respect to these parameters. Surgery had slightly better 2 year results but showed no advantage by 6.5 years over SRP alone, except for more sustained PD reduction in deep pockets. Decision for or against surgery must be made on individual basis.

Serino 2001                     ARTICLE

P: To determine the initial outcome of non-surgical and surgical access tx in pts with advanced perio and the incidence of recurrent dz during 12 years of maintenance following active therapy.

M&M: 64 pts with minimum of 12 non-molar teeth with deep pockets > 6mm) and with > 6mm of bone loss. Only non-molar teeth were evaluated. Pts were randomly assigned to 2 groups: sx (SU) and non-sx (SRP). After baseline exam (PI, BOP, PD, PAL, RxBL), all pts were given OHI. After Sx (MWF, 4-6 sessions) or SRP (4-6 sessions, 60-90min) and rinsed with 0.2% CHX bid for 2 months. All pts were enrolled in a maintenance program (3-4 visits/year). Sites that at a recall appointment that had BOP and a PD of > 5mm had new sub-g instrumentation. OHI was repeated prn. Comprehensive re-exams were performed after 1, 3, 5, and 13 years of SPT. If a pt between annual exams displayed marked dz progression (ej PALoss > 2mm at > 4 teeth), they were exited from the study and given additional treatment.

R: 4pts (14%) in the SU group and 8 pts (29%) in the SRP group were excluded between years 1 and 3 due to additional dz progression. At different intervals after that time frame, 3 pts in the SU group and 4 in the SRP group were excluded for the same reason as well. NSSD between groups with respect to the various perio parameters examined at baseline. Both groups maintained high standard of OH at 13 years (PI < 15%). In both groups, BOP increased significantly from 18% at year 1, to 20-25% at year 3 and finally to about 30% during the later periods. Mean PD underwent minor changes over time, being 0.6mm in both groups over the 13 years. The SU group had a SS larger proportion of sites with shallow pockets (<4mm) than SRP (70-60% vs 55-40%), and a SS smaller proportion of medium deep (25-35% vs 35-50%) and deep (5-10% vs 10-15%) PD sites. Mean PALoss between SRP and SU was not SS. Shallow sites tended to lose 0.1mm PAL/yr regardless of location (buccal, lingual, interproximal).

-SU was more effective thatn SRP in reducing mean PD and elimination deep pockets

-The number of sites that displayed advanced dz progression after tx was SS smaller in the SU group (14% vs 29%)

-The majority of subjects could be maintained on SPT over a 12-yr period with only minor episodes of recurrent dz

D: The reason for higher incidence of dz progression in SRP group may be due to more residual sites with deep pockets.

BL: Sx therapy is more effective in preventing dz progression than non-sx therapy in pts with advanced perio that are placed on 3-4 month SPT over a 12-yr period. In pts that did not display advanced dz progression and were therefore not excluded from the study, there was NSSD in the periodontal parameters between Sx therapy and SRP.

What is the critical probning depth?

Lindhe 1982                     ARTICLE

Purpose: 1) to calculate the critical PDs (CPD) for one surgical and one non-surgical method of periodontal therapy, 2) To monitor during an 18-month maintenance period sites which following active therapy were associated with PD>4mm with respect to gingival inflammation and attachment level alterations and 3)TO evaluate the effect of the oral hygiene status on PDs and attachment levels during the maintenance care period in patients who following active therapy were recalled for prophy every 3 months.

Materials and methods: 15 subjects 32-57 years old. After baseline examination (PI, GI, PD, AL) and OHI patients were subjected to perio Tx in a split mouth design. In one side debridement was performed in conjunction with modified Widman flap (MWF) and on the other side SRP was performed. In the first 6 months patients were recalled for maintenance visits every 2 weeks and after that every 3 months. Re-eval was performed at 6, 12 and 24 months.

Results/BL: Critical PDs for sites subjected to SRP was 2.9 mm and for the MWF group 4.2 mm. This means that sites with initial PD of less than 2.9mm are likely to show attachment loss if SRP is performed. In sites with initially deeper pockets (7 mm or more) the resulting attachment gain was more pronounced following surgical than non-surgical treatment.

If the plaque score assessed at the 6-month examination is representative of the oral hygiene conditions during the phase of healing, the CPD – value obtained for a given site increase as PI increases.

90% of sites with PD<4mm at 6 months remained in this category during the maintenance phase, 10% became deeper.

60% of sites with PD 4-6mm at 6 months remained in this category, 30% entered in the <4mm category and 1-4% became deeper.

For sites with PD>6mm , 27-60% entered in the 4-6mm or<4mm during maintenance phase and and 40-50% remained unchanged. There was a significant attachment gain during maintenance (1.6mm for MWF and 1.2mm for SRP.

Shallow pockets tend to lose and deeper sites to gain attachment during maintenance phase. Sites with plaque score more than 0 lost attachment during maintenance (0.72mm MWF and 0.55mm SRP).

Statistical Methods

What is sensitivity and specificity? What is positive predictive value and accuracy? What is Prevalence? What is negative predictive value? What is equivalence and superiority? What is regression towards the mean?

D Brunette 2007                     NO ARTICLE

Purpose: Chapter of a book discussing the diagnostic tests and measurements in clinical practice

Discussion

Accuracy

• The overall agreement between the test and the gold standard.

Sensitivity

• The proportion of diseased individuals correctly identified by the test.

• Also known as the true-positive rate

Specificity

• The proportion of non diseased individuals correctly identified by the test.

• Also known as the true-negative rate.

Prevalence

• The overall probability or risk that the disease is present prior to the test

• Also known as the pretest likelihood

• The proportion of individuals in a population who have the disease at a specific point in time; prevalence in a specified population may change over time, and prevalence may change if the definition of the disease changes.

PTL(+)

• Posttest likelihood of a positive test result

• Also known as the positive predictive value

• For an individual with a positive test result, PTL(+) is the probability that the disease is actually present.

PTL(-)

• Posttest likelihood of a negative test result

• For an individual with a negative test result, PTL(-) is the probability that the disease is actually present.

NPV

• Negative predictive value

• For an individual with a negative test result, NPV is the probability that disease is really absent

Regression Toward the mean

Hujoel 1990                     ARTICLE

P: Discussion on the estimation of sensitivity and specificity of site-specific diagnostic tests.

DISC: Sensitivity is the probability that a person with a certain condition will be classified by a test as having that condition. Specificity is the probability that a person without the condition will be classified by the test as being without the condition. Observations on sites within a pt may be incorrectly given SS or NSS. To avoid this situation, site-specific data should be analyzed with statistical methodology that accounts for the dependence of within pt observations. The use of diagnostic tests emphasizes the need for objective procedures to evaluate and compare these tests. Ignoring the correlation between sites can significantly affect the variance estimates and thereby distort differences between diagnostic tests. The correlated binomial model offers advantages such as: 1. It allows for a correct estimation of sensitivity and specificity and their variances. 2. The significance of the within pt correlation coefficient may provide a criterion for assessing the effect of pt-related factors on the performance of the diagnostic tests. 3. The model is flexible and can easily be expanded to provide likelihood ratio tests to compare diff diagnostic test. The clinical value of a diagnostic test may be misjudged and/or comparisons between different diagnostic tests may yield misleading conclusions when large within pt correlation coefficients are present.

Gunsolley1998                     ARTICLE

P: To investigate sample size requirements in both equivalence and superiority studies of products used in regeneration.

B: Superiority testing tests if a new product performs superior to an existing product. Superiority clinical trial is to determine if a new therapy is superior to an estabilished therapy or placebo. Equivalence trials determine if two treatments are equivalent.

M&M: Review of literature conducted. The criteria was: AL gain, sample size, standard deviation, random clinical trials of debridement, DFDBA, HA, or ePTFE .

R: The lower the limit on equivalence the greater the sample size needed (at 10% - need 4X sample size as at 20% regardless of response, defect type, or treatment). Within each limit, similar samples sizes were found for defect type and treatment.

C: Equivalence trials require a much greater sample size than previously thought in trials that investigate perio regeneration.

Gunsolley 1998                     ARTICLE

Editorial: It is a common misconception that failure to detect a SSD b/t tx groups with proving a null hypothesis of no difference between groups are equal. A hypothetical study example is given where two membranes, one experimental and the other “the standard” are compared for GTR and CAL gain in 10 subjects. The results of the study indicated the experimental membrane gave an estimated 2.5mm CAL gain and the “standard membrane” 3.0mm CAL gain. The authors concluded NSD b/t the groups and concluded the experimental membrane could be used as a replacement for the “standard membrane”. The editorial discussed several flaws with this study. One is the subjects were chosen based on pt availability and not by formal statistical size calculations. A second flaw was the study design to determine superiority, not equivalence of the two products. The study’s failure to find a SSD between the two products does not provide evidence they are equivalent. The failure may actually be due to no real difference b/t the products or lack of statistical power. Unfortunately, many “negative clinical trials” are conducted w/o adequate statistical power and no evidence about the superiority or equivalence can be provided.

BL: In planning a clinical investigation, the investigator must determine if the goal is to detect equivalence or superiority of a new treatment and then determine the appropriate sample size.

Duke 1998                     ARTICLE

P: To explain for the clinician a correct method of designing and analyzing studies for purposes of demonstrating equivalence.

D: In clinical trials, a lack of significance when looking superiority b/w 2 treatments is NOT the same as equivalence. The trial has to be designed to look for equivalence as opposed to superiority. One of the first steps in conducting a clinical study is to calculate sample size. For both superiority and equivalence studies, appropriate statistical test and four parameters: Type I (α) and II (β) error, standard deviation (s) and clinical difference sought must be decided. An equivalence study shows the maximum value at which two groups are considered equivalent. With a superiority design (∆), the larger the difference, the smaller the sample size will become, but an equivalence design (δ) must have a clinically justifiable difference. Demonstration of clinical equivalence is both difficult and resource-intensive, meaning a larger sample size is required vs. a superiority design. It has been suggested that for equivalency studies for periodontal productions should have a confidence interval within ±10% of the standard treatment but this would require very large studies. Because the majority of clinical trials focus on superiority rather than equivalency, clinicians have little exposure to draw appropriate conclusions from equivalency trials.

Gunsolley 2001                     ARTICLE

Background: Following sc/rp many studies have implied an association between a loss of clinical attachment at sites with initially shallow PDs (1-3mm) and CAL gain for deeper PDs. However, these effects are also consistent with a statistical phenomenon referred to a regression towards the mean. This principle suggests that extremes values will moderate the next time they are recorded.

P: To estimate the effect that regression to the means has on perceived changes on attachment level after sc/rp.

M&M: 12 patients completed the study. Each participant had a minimum of 20 teeth, at least one quadrant with all teeth present and minimum of 8 sites with a minimum of 5mm PD and 2mm attachment loss. At baseline examination, OHI were provided, 2 investigators performed full-mouth probing. The 2 examiners repeated the probing 2 weeks later before therapy was begun. 2 quadrants were randomly selected to be root planed and scaled. The 2 remaining were not treated. 4-6 weeks after the treatment a 3rd full mouth probing was performed. These probing were performed by the treating periodontist and a periodontist who was blind to the quadrants that had been scaled. Statistical analysis was performed in order to evaluate the effect that regression towards the mean has on perceived changes during scaling and root planing.

R: Both examiners repeated their exam during baseline measurements. Both examiners had similar results and demonstrated that there was a regression towards the mean. In both cases the differences in attachment loss measurements from their first measure were statistically significant. Shallow sites showed average negative differences (-0.27mm, -0.20mm) mimicking CAL loss and deep sites showed average positive values (0.33mm, 0.47mm) mimicking CAL gain. The repeat exams were divided into sites that were scaled and non-scaled so that the effect of therapy could be evaluated. Both shallow non-scaled and scaled sites had similar differences in repeat measures (-028mm, -0.25mm) which were similar to and NSSD from changes after therapy for both non-scaled (-0.21mm) and scaled sites (-0.08mm). For moderate PDs differences in repeat examinations resulted in a negative value for non-scaled sites of

-0.09mm and a small positive value for scaled sites of 0.08mm. For non-scaled sites change over time of 0.12 and SSD from differences in repeat measures. For scaled sites a much larger and SSD CAL gain of 0.38mm was noted. Thus, regression towards the mean explains little of the changes at moderate sites. For deep PDs differences in repeat examinations resulted in positive values of 0.19mm for non-scaled sites and of 0.48mm for scaled sites. For both non-scaled and scaled site changes over time were SSD for scaled sites.

CON: The majority of attachment loss due to scaling of minimal PDs reported in many studies may be due to a statistical phenomenon called regression towards the mean.

What is the difference between statistical significance and clinical significance?

Rethman 1999                     ARTICLE

P: To discuss the relationship between statistical and clinical significance. This study discusses a few concepts that need to be considered by those who seek to discern the clinical value of statistically based conclusions.

Type of clinical studies:

Control vs test: The reason these studies are performed is that the investigators hope that test therapies will work better than and/or perform as well as the control therapy. These comparisons necessitate different statistical analyses. When investigators choose to statistically test numerous outcomes, they increase their chances of making Type I error.  If investigator collects data on 20 different measures of outcome to compare test vs control and sets significance level to 5% (p<0.05) then a SSD will be found in at least 1 out of the 20 outcomes even though no actual difference exists b/w the test and control treatments. That is why, it is prudent to adjust for multiple comparisons when conducting numerous statistical tests.

Longitudinal studies: The investigator may correctly limit the number of outcome measures to 3 or 4 variables that are considered clinically meaningful. An example of this would be for an investigator to collect data probing depths, plaque indices, inflammation indices and bleeding indices. Although these may be related quantities, they reflect different measures of periodontal health and are of interest for the clinician. The problem arises when the investigator collects data for each outcome measures for several time periods and proceeds to analyze each time period in a cross sectional manner. This results into performing 20 t tests (1 t test for each of the 4 outcome measures at each of the 5 time points after baseline measures). So again if you set significance at 5% (p<0.05) then by pure chance alone should have at least one of the comparisons have SS. The solution to this problem is to analyze these data using a longitudinal statistical analysis. The most common method for this is to use repeated measurements ANOVA to take into account all the measurements collected over time. This type of analysis also allows the investigator to not only compare treatments while taking all time periods into account but also to look at the effects of the respective treatments over time.

Equivalence: Another statistical issue that is often misunderstood. Until recent years it was erroneously claimed equivalence between test and control therapy based on an inability to demonstrate a difference between the two.  Example statement :“The therapies were not shown to be different at P<0.05; therefore the compared therapies were equivalent.”

Even though appropriate equivalency tests are finally in common use, abuses of equivalency are still seen.

BL: A claim of the clinical insignificance of a statistically significant result depends on a critical assessment of the context in which a clinical decision must endure. Context considerations need to be more widely and thoughtfully considered in both the design and clinical interpretation of studies. Each clinician must make a determination of clinical significance based on the individual context in which he or she hopes to apply the information to benefit the patient we serve.

Greenstein 2000           editorial           ARTICLE

D: Statistical significance: When 2 treatments are compared with respect to a particular outcome (CAL gain), the finding that one therapy causes a SS better result than the other indicates the likelihood that the detected difference occurred by chance is extremely small. The finding of SS does not quantify the magnitude of difference between tx or the importance of the result. Mean data do not reflect a clear benefit that may occur at any site or tooth within any ‘articular ‘t. Means may be virtually meaningless as a guide to the efficacy of specific types of therapy. Also, the statement that one therapy is SS better than another fails to answer whether the tx makes an important difference with regard to attaining the desired outcome. To compensate for this, investigators could select a difference between therapies (ej. 2mm PD) that is clinically meaningful.

Clinical significance: When 2 tx methods are compared, the smallest difference between therapies with respect to an important outcome (ej. PD) that would result in a decision to modify tx denotes clinical significance. Since clinical trials assess remedies for clinical problems, it would be appropriate to express the conclusion in clinical as opposed to statistical terms.

Absolute criteria: A threshold defining the minimal amount of change that would be considered clinically significant (meaningful) after therapy could be selected.

Cutoff point: A treatment goal (ej, PD < 5mm)

BL: A defined set of parameters describing benefits provided by new therapies would facilitate interpretation of data and selection of the best therapy for different clinical situations.

Hujoel 2000                     ARTICLE
Purpose: To provide a definition of clinical significance.

Discussion: What is clinically important outcome? Clinically important outcomes are measures of how a patient functions, feels or survives. Clinically important outcomes for treatment of periodontitis, include prevention of tooth losss and an increase in quality of life. It is difficult to be associated with a change in surrogate endpoints for all chronic diseases and especially for periodontitis where no large epidemiological studies have been conducted to explore the relationship between commonly used periodontal surrogates and tooth loss. (e.g. no studies have shown that a treatment providing 0.5mm AL gain is also more likely to reduce tooth loss). Therefore providing guidelines as to how to address the issue od clinical significance is difficult. Three indicators of clinical significance have been suggested in 1995: 1) Continuing progressive loss o periodontal support 2)Elimination of clinical signs of inflammation and demonstration of persistent and stable long-term periodontal support 3) The magnitude of improvement in CAL and PD is of less clinical significance than the fact that the parameters represent stability rather than loss. Scientific basis for accepting these indicators as valid surrogates of important long-term outcomes is limited. The larger the change in a surrogate point (0.5mm vs 5mm change in PD) and the longer the duration over which the change in the surrogate endpoint was observed (1-month vs 12-month study) the more likely the observed difference is clinically important. Each study should define whether the surrogates used are based on data or on expert’s opinion. If this is done, others can objectively evaluate the statements.

What magnitude of difference between treatments is considered clinically relevant? To continue a trial after a statistically significant difference is observed, under most circumstances it would be considered unethical. Statistical significance is not the only criterion by which trial results are monitored and decisions regarding the superiority of a treatment are made. Adverse events, results of other ongoing trials and discovery of new information during the trial have a potential impact on the decisions made, but even though these factors play an important role, it is rare that a treatment is decided to be superior to the control in the absence of statistical significance. Treatment cost vs benefit also plays a role in the clinical significance of the results and is evaluated from a patient’s perspective.

What is a phase III or definitive trial? Definitive clinical trials evaluate the effect of an intervention on clinical outcomes of particular relevance to the patient. Problems include failure to guarantee randomization, enrolling too few patients to detect Tx differences, inappropriate analysis of treatment subgroups, post-hoc removal of data from the final analysis and misleading substitutions of surrogate biological markers for clinical endpoints. The conclusion of statistically significant difference is most reliable when these problems are avoided. Post-trial generated findings or findings from exploratory trials are unreliable and cannot be used to indicate clinical significance.

Conclusion: Currently no definitive randomized trials using true endpoints are available indicating that one periodontal treatment is more effective than another in obtaining clinically relevant outcomes. It is safer to claim ignorance regarding the clinical relevance of a periodontal treatment rather than to claim clinical significance based on substantial leap or faith.

How can we evaluate the quality of a RCT (the randomized clinical trial) ? What is a meta-analysis and how accurate are their results? What are the major pitfalls in clinical trials design?

Jeffcoat 1992                     ARTICLE

Purpose: To review selected aspects of clinical trials including the overall clinical experimental design, controls, outcomes, sample size, and pt selection.

Discussion:

Cohen 1988                     ARTICLE

P: To evaluate false positive rates in the determination of changes in probing depth-related perio measurements

M&M/R: This is a simulation study; three phases; Phase I: looks at replicate PD measurements of sites rounded to the nearest 1mm (no expected change). They found that shallower pockets can be measured with accuracy, while deeper pockets are subject to significant measurement error. Phase II: looked at pairs of measurements rounded to nearest 1mm; found acceptable accuracy in detecting true changes with measurements that differed 2mm. Phase III: looked at burst rates; sites measured q 2 months for 1year; true changes determined to be >2mm b/t measurements; pooled measurements showed 1.3% false positive change when there really was no change.

Disc: This paper discussed how a computer model of the continuous disease progression theory is not only different, but less complex, more accurate, less assuming, and just as clinically feasible as the burst theory of progression. Progression of disease in this model went from health to mild to moderate to severe disease in 39 observation periods. The statement was made that treated sites would reenter health and w/o further treatment would become diseased again since they were more susceptible to breakdown.

They concluded 1/3 of detected bursts of perio attachment change may be false (+) attributable to measurement error.

BL: Suggests that 1/3 of tolerance detected “bursts” of periodontal attachment changes may be false positive attributable to measurement error.

Hujoel 1988                     ARTICLE

BG: Split mouth designs can be used in a clinical instigation because it can eliminate the subject factor from experimental error and allows a more economical use of patients. The essential statistical characteristic of this design is that comparisons can be made on a within-patient basis, not a b/w patient basis. If statistics are being used within a single patient, a two-way ANOVA or a paired t test can be used. If statistics are being used on a between-patient basis, will need to utilize other methods.

P: To determine if appropriate statistics test were used for split-mouth clinical trials.

M&M: 22 published trials reviewed; labeled as appropriate, inappropriate, or absent.

R: 5/22 trials reported appropriate stats and 5 did not report or did not use stats; more than half of the reviewed trials reported use of one-way ANOVA or a two-sample t test to investigate data of split-mouth trial; this can lead to an increase in type II errors due to the correlated nature of split mouth data. Type II errors: If there is a conclusion of NSD, when in fact, there are differences.

BL: For studies that are examining data and drawing conclusions in a between patient basis of a split mouth design, it is possible to misrepresent the outcome by using incorrect statistics. A one-way ANOVA and two-sample t test cannot draw conclusions within a patient. The correct methodology for within patient conclusions would be two-factor ANOVA or paired t test

Lesaffre 2007 (Review)                     ARTICLE

BG: Hujoel & Moulton previously questioned the reported quality of split-mouth studies. Since then, there has been little enquiry into the methodology of this study design.

P: To conduct a systematic review of the reported methodology of clinical studies using a split-mouth design published in dental journals over a 1-year period (2004).

M&M: An extension of the CONSORT guidelines for cluster-randomized designs was used to evaluate quality. They evaluated the methods used and quality of reporting split-mouth studies.

R: 34 studies were eligible for this review. The results showed that many papers lack essential qualities of good reporting, e.g. 5 of 34 papers gave the rationale for choosing a split-mouth design, 19 of 34 (56%) used appropriate analytical statistical methods and only 1 of 34 presented an appropriate sample size calculation. Of the 5 studies that used survival analysis, none of them used a paired approach.

BL: Despite some progress in analyzing and reporting split mouth studies, there remains a substantial need for improvement. The errors in these studies can undermine the value and validity of these studies. Involving epidemiologist, statisticians and researches may improve the statistical quality in research.

Needleman 2008                     ARTICLE

P: To discuss the importance of rigorous reporting of trials in oral health and to discuss the impact of CONSORT on trial reporting. Note: CONSORT (The Consolidated Standards of Reporting of Trials) Statement is a guidance to facilitate reporting of randomized controlled trials (RCTs), and to aid clinicians in applying validated findings into practice. 

D: The authors emphasize that study design and methods are of great influence on the size of estimates and study interventions. Bias has also been attributed to causing these effects. Readers/reviewers must have access to this data to effectively appraise trial results. The authors feel that the quality of trials in dentistry is poor and that this hinders the understanding of individual trials. Without complete reporting of the data, a reader is prevented from adequately appraising the quality of the study, increasing the uncertainty of whether the results can be applied to healthcare. Journals that have adopted the CONSORT guidance have shown a marked improvement in the reporting of trials. As of 2008, only five dental journals have adopted the CONSORT criteria; while hundreds of medical journals have implemented this standard. CONSORT consists of two elements: 1) a 22-item checklist to guide the author through thorough trial reporting and 2) a flow chart to account for all study participants throughout the trial. Reporting standards for other research designs are: QUOROM (PRISMA) for systematic reviews, STROBE for observational studies, STARD for studies of diagnostic accuracy.

BL: There is good evidence that bias affects the estimate of effect of an intervention in clinical research. Determining whether a trial is at risk of such bias and therefore whether to introduce new research findings into oral healthcare is dependent on transparent reporting of clinical trial methods and findings.

Montenegro 2002                     ARTICLE

Background: The randomized controlled trial (RCT) provides the highest level of evidence for medical interventions. The elements are used in a carefully designed RCT that will minimize bias, balance confounders, and therefore produce the most reliable estimate of treatment effect.

P: To assess the quality of RCTs in periodontology using evidence-based criteria and its impact on the size of treatment outcome, randomization, allocation concealment (concealing the randomization sequence during pt recruitment), blinding, and handling of withdrawals.

M&M: A search of the Cochrane Oral Health group (COHG) limited to the J of Clin Perio (155 articles-54.8%)., J of Period (126 articles-44.5%), and J of Per Res (2 articles-0.7%) from 1996-1998. 283 articles total. The quality of individual components was assessed (randomization, allocation concealment, blinding). Handling of withdrawals and drop-outs was assessed by analysis of whether all patients that entered the trial were properly accounted for at the end. The pilot study consisted of two aspects: First the lead reviewer was calibrated against another reviewer with experience in conducting systematic reviews. After 4 rounds of calibration, a consistent level of agreement was found. Then all the screening of the articles and data were piloted and minor changes were recommended.

R: The agreement between and among the reviewers was good. Most of the studies described themselves as randomized (91%). However, in most of the studies (83%), there was either no method reported or it was inadequately explained. A clearly inadequate randomization method was utilized in one study only. Allocation concealment was poorly reported, with 93% of the studies giving either no method or an unclear description. Patient and caregiver blinding was absent in 43% and 57% of studies. Only 17% of studies positively demonstrated caregiver blinding, and 43% had patient blinding. Approximately half of the studies (56%) demonstrated proper account for all participants at the end of the trial. The use of analysis to take into account withdrawals was not applicable for 43% of studies where follow up was deemed complete.

C: Quality of RCTs in periodontology frequently does not meet recommended standards. Greater attention to design quality and reporting of RCT’s is needed.

Esposito 2001                     ARTICLE

P: To assess the methodologic quality of published Randomized Controlled Trials (RCTs) of oral implants in an objective and reproducible way.

M+M: A literature search strategy appropriate for a Cochrane systematic review. The Cochrane Oral Health Group specialist register, PubMed, and personal libraries were searched. 74 RCTs were identified. Articles were evaluated only for the information that they included, and no additional reference or information was sought. Trials were not appraised for quality if they included fewer than 10 patients for a parallel study design or fewer than 5 for split-mouth or crossover designs. Each article was assessed by 2 clinical researchers. The statistician evaluated all articles for the quality of statistical analysis and recorded any reason that statistical analyses were performed incorrectly. The final quality score of each article was determined in a consensus meeting by the 3 clinical researchers. After RCTs presenting the same patient population were excluded, 43 articles remained and were assessed in the present investigation. Quality of each study was assessed on 7 questions (Was a sample size calculation undertaken? Randomization & allocation concealment method, Were inclusion /exclusion criteria clearly defined? Was reason for withdrawal specified by study group? Were the control and treatment groups comparable at entry for important prognostic factors? Was there any attempt at blinding? Was the statistical analysis appropriate?)

R:

D: The major limitation of a register of published RCTs is that it could be biased toward positive and “encouraging” results (publication bias). This is because of the fact that “uninteresting” information is less likely to reach the publication stage, which may lead to erroneous conclusions of therapeutic effectiveness. Therefore, it would be of great benefit if unpublished trials could also be identified. Finally, there seems to be considerable potential for improving the design, conduct, statistical analysis, and reporting of RCTs in implant dentistry.

BL: There is considerable potential for improving the design, conduct, statistical analysis, and reporting of RCTs in implant dentistry.

Root Caries

What factors may influence the incidence of root caries? Does periodontal treatment increase root caries?

Ravald 1993                     ARTICLE

P: To study the individual susceptibility to root caries in periodontally treated patients in a long-term follow-up of 12 years.

M+M: 27 pts (14 M, 13 F, mean age 59.2 years old). Tx comprised of OH, SRP, and selective perio surgery. After pts were treated, they were returned to their general practitioners for maintenance care and subsequent annual check-ups. Follow up exams at 1, 2, 4, 8, and 12 years. Measurements recorded at baseline: prevalence of root caries, periodontal condition, plaque score and salivary lactobacillus counts. At the 4-year follow-up, the incidence of root caries (new DFS%), bleeding on probing score, salivary secretion rate and salivary buffer effect were added. At the 8- and 12-year examinations, salivary mutans streptococcus counts and a dietary habit index were included. Oral sugar clearance time was only recorded at the 8-year examination. Root caries was expressed in terms of number (DFS) and percentage (DFS%) of decayed and filled root surfaces. Also retrospectively compared smokers with non-smokers. Statistical analysis was performed. Root caries was defined as “a cavitation or softened area which might or might not involve adjacent enamel or existing restorations (primary and recurrent lesions)” as described by Hix & O’Leary 1976. Active as well as inactive lesions were included.

R: New root caries lesions were recorded in 24/27 patients. In 8 of these, the root caries incidence was between 1 and 5, in 7 patients between 6 and 9, and in 9 patients, 12 or more new DFS. 13 patients with >5 new DFS% during the 3rd 4-year period (years 9-12) differed significantly from 14 patients with <5 new DFS% in salivary mutans streptococcus counts, plaque scores, and new DFS% during the 2nd 4-year period (years 5-8). Risk values among the variables tested at the 8- year examination were about 3 times more prevalent in patients that developed > 5 new DFS% in years 9-12 than in those with < 5 new DFS%. The most prevalent variables with risk values were salivary lactobacillus count (n = 9), salivary mutans streptococcus count (n= 7) and dietary habit index (n = 7). Smokers (14/27 pts) had significantly more root caries than non-smokers after the entire 12-year period (9/14 smokers had > 5 new DFS). No single variable was found to be useful as a predictor for root caries development in the individual patient. However, as the number of variables increased, so did the chances of new DFS to be found.

BL: Root caries in this category of periodontally treated patients is a minor problem although some individuals show a high incidence, and it may be possible to identify patients at risk for development of root caries by using readily available tests in addition to clinical examination and the patient's medical history. Salivary counts of mutans streptococci and lactobacilli, plaque scores and the patient's dietary habits seem to be the most useful variables in the long-term perspective of developing root caries.

Paraskevas 2004       (prevention of rc)              ARTICLE

Purpose: To evaluate root caries in patients using dentifrice and mouthrinse containing amino fluoride (AmF) and stannous fluoride (SnF2) as compared with a dentifrice and mouthrinse containing sodium fluoride (NaF).

Materials and methods: 71 patients 30-65 years old. Pts were healthy, with at least 3 natural teeth in every quad and regular supportive periodontal care (3-4 months) for at least 1 year after treatment. Exclusion criteria: antibiotic therapy within 3 months prior to the study, hypersensitivity to SnF2, NaF, or AmF and systemic disorders or medication that could affect the condition of the periodontal tissues.

Pts were divided randomly in the test group (dentifrice and mouthrinse containing AmF/SnF2) and the control group (dentifrice and mouthrinse containing NaF). Pts continued having periodontal maintenance throughout the study. Normal OH procedures were allowed. Plaque, gingival bleeding and presence of caries were evaluated.

Results: Test group had 33 subjects and control 38. 10 smokers in each group and study lasted 2 years.

SSD less plaque was observed in the test group (25% reduction over 24-month period).

SSD increase n the number of restorations present on root surfaces was observed for both groups at 24 months.

The mean number of active caries lesions at 24 months decrease from 2.1 to 1.8 for the test group and increase from 1.9 to 2.2 in the control group. These changes were NSSD.

91% of the subjects in the test group and 89% in the control group developed at least one new root caries lesion over the study period. NSSD in the number of new caries between the two groups.

Conclusion: No difference was detected in terms of root caries between the two groups.

DeSoete 2005                     ARTICLE

Purpose: To longitudinally examine whether initial periodontal therapy causes an intra-oral shift, both supra and subgingivally, from a periopathogenic to a more cariogenic flora, and if any kind of shift can be prevented.

Materials and methods:

Results:

D/CL:

Privacy Policy  |  Sitemap

Designed By Steven J. Spindler, DDS LLC