A Perspective Regarding the Role of Local Drug Delivery in the Treatment of Chronic Periodontitis: Things You Should Know

Dentistry Today


Periodontal diseases are infections of the periodontium.1 Once there is an immune response to a bacterial challenge, then the process is referred to as an infection.2 Usually the host response contains the bacterial challenge and there are no signs of inflammation.3 However, if tissue destruction occurs, then the condition is considered a disease.2 The most common form of periodontitis is referred to as chronic periodontitis. It can be localized or generalized (more than 30% of the teeth), and depending on the amount of clinical attachment loss, the severity of the disease process can be labeled as mild (< 3 mm), moderate (3 to 4 mm), or severe (≥ 5 mm).4 It has been determined that approximately 30% of the adult population develops chronic periodontitis, and about 10% manifest signs of severe chronic periodontitis.5

Scaling and root planing or ultrasonic debridement are effective mechanical therapies for periodontitis.6,7 However, in deep or tortuous pockets or sites that do not respond to conventional therapy, it may be beneficial to use adjunctive antimicrobial therapy.8,9 For chemotherapy to be effective, it must meet 3 requirements10: (1) reach the site of disease activity, namely the base of the pocket, (2) be delivered at a bacteriostatic or bactericidal concentration, and (3) remain in place long enough to be effective.

Table 1 illustrates that local drug delivery meets all 3 requirements, whereas the other drug delivery techniques have shortcomings.11 This article addresses the efficacy of local drug delivery and places in perspective its advantages and limitations. Furthermore, it will differentiate between statistically significant and clinically relevant results.

Currently, 3 local drug delivery devices are commercially available in the United States: chlorhexidine chip (PerioChip, Perio Products), doxycycline hyclate gel (Atridox, CollaGenex), and minocycline microspheres (Arestin, OraPharma Inc).12-22 A comparison of characteristics associated with each delivery system is found in Table 2. All 3 are considered controlled delivery devices because they maintain a high intracrevicular drug concentration for more than 24 hours. They are resorbable and easy to place. In the 9-month long studies that evaluated the efficacy of these drug systems, the drug was usually used more than once (Table 2). The unit packaging of both chlorhexidine chips and minocycline microspheres allows drug delivery to one site. If multiple sites or surfaces need to be treated, then additional medicament units need to be employed. In contrast, the doxycycline hyclate gel delivery system provides for treatment of multiple sites.

Numerous clinical trials were conducted to evaluate the effectiveness of local drug interventions. The results from a test (drug) and control group were subjected to hypothesis testing to determine if differences were statistically significant.23 However, the expression statistical significance does not indicate the size of the difference or its importance; it simply denotes that the difference did not occur by chance. Thus, analyses of the data and their relationship to clinical significance are needed to determine if specific procedures are worthwhile. Accordingly, it would be advantageous if studies assessing new treatments addressed several of the determinants for clinical significance listed in Table 3.24

We face a perplexing dilemma when clarifying the term clinical significance, because the definition can vary depending on the situation. For example, at a 6-mm pocket, a 2-mm probing depth reduction may be considered clinically important. However, at a 10-mm pocket, this change would not be significant. Thus, it may be worthwhile to define clinical significance as the minimum change that could cause an alteration in treatment. Consequently, the level of change used to delineate clinical significance can fluctuate, and it will depend on the size of the defect being treated. 
A method that can help clarify the utility of local drug delivery is a calculation referred to as the number needed to treat (NNT).25 It is employed to determine how many more sites would need to be treated with local drug delivery to attain a better result than only scaling and root planing at one additional site. To calculate the NNT, take the difference between the percentages of times an event occurred (eg, 2-mm probing depth reduction) in the test and control groups and divide this number into 100. For example, if a 2-mm probing depth reduction occurred 15% of the time with scaling and root planing (SC/RP) plus an adjunctive device, but only occurred 5% of the time after using SC/RP, then the difference would be 10%. When the difference (10%) is divided into 100, the NNT would be 10. This means that an additional 10 sites on average in that study population would need to be treated with local drug delivery to attain a 2-mm probing depth reduction at one more site than scaling and root planing.


When evaluating the utility of local drug delivery for a particular site in a specific patient, several factors need to be considered:

•What is the magnitude of the effect that can be expected with local drug delivery?
•What is the size of the defect to be treated?
•What is the desired treatment outcome?
•What are the dental and medical histories of the patient?

To help understand numerous issues associated with local drug delivery, a series of questions were developed that are found in Table 4.26 Answers to these questions are presented to help place the utility of local drug delivery into proper perspective.

(1) Best drug to use. With respect to the 3 local drug delivery devices commercially available in the United States, 3 different drugs are being used: doxycycline hyclate, minocycline microspheres, and chlor­hexidine gluconate (see Table 2). All 3 drugs can be effective; there is no one ideal drug for all clinical situations. Advantages of tetracyclines are that they are substantive. This means that they adhere to the tooth surface and tend to be present in the gingival crevicular fluid for a longer period of time. Tetracyclines demonstrate limited tissue penetration, whereas the chlorhexidine molecule is too large to enter the tissue. At the concentrations that these drugs are used, none is clinically toxic to the tissues. Currently, there are no published data regarding the use of multiple drugs or sequencing of different drugs at any one site.
(2) Best local drug delivery device. There is insufficient data to conclude that one device is better than another (Table 5). All 3 devices are resorbable and provide controlled release of the drug reservoir. Each provides a high intracrevicular concentration of drug.
(3) Frequency of drug applications. Table 2 provides the number of drug applications in the phase III clinical trials (phase III trial is a large clinical trial that is provided to the FDA to attain approval for use of a drug). In the phase III trial assessing chlorhexidine, the chips were applied 2 or 3 times in 60% of the patients (baseline, 3 months, and 6 months) during the 9-month study.16 As for the tetracycline class drugs, the minocycline microspheres were applied 3 times (baseline, 3, and 6 months)21 and doxycycline hyclate was applied twice (baseline and at 4 months).12 With respect to the latter 2 devices, improved results were noted in the phase III trials before the drug was applied a second time. Thus, it appears that the drug devices can be placed once and the results should be evaluated before a second application; however, this also indicates that the length of time that the clinical effects last after a single drug application is unknown. Two additional smaller studies evaluated the use of chlorhexidine chips after one applica­tion.18,19
(4) Expected clinical outcomes.

A. Mean summary data.
The overall mean summary data from phase III clinical trials that assessed the 3 commercially available drug delivery systems are provided in Table 5. When chlorhexidine chip plus SC/RP (combined therapy) was compared to SC/RP alone in the phase III trial, a statistically significantly greater pocket reduction was achieved with combined therapy (0.95 mm vs. 0.64 mm).16 When minocycline micro­spheres plus SC/RP was compared to SC/RP alone (phase III trial) in patients with mean probing depths of ≥ 5 mm, the reduction in probing depth was statistically significantly greater for combined therapy (1.32 vs. 1.08 mm).21 Similarly, in patients with mean probing depths of ≥ 6 mm, the result was statistically significantly greater for combined therapy (1.46 vs. 1.05 mm). In patients with a mean probing depth of ≥ 7 mm, the result was 1.99 vs. 0.98 mm. However, this result was not statistically significant, because the study population was too small. Therefore, at deep sites these data need to be interpreted cautiously. When doxycycline hyclate was compared to SC/RP, the results were equivalent.12 Probing depths were reduced 1.1 mm in one investigation and 1.3 mm in another study.12 It should also be noted that if a patient does not respond to SC/RP plus adjunctive local drug delivery, it may be necessary to consider performing drug sensitivity testing to determine the offending bacteria and their susceptibility to specific drugs and/or surgical pocket depth reduction.

B. Other criteria that can be used to evaluate clinical significance.
•Changes of ≥ 2 mm in Probing Depth—Table 6 lists how often 2-mm changes were achieved when a local drug delivery device was employed. NNT values were calculated to indicate how many additional sites on average would need to be treated to achieve this result at one additional site: Chlorhexidine chip16 9; minocycline microspheres21 12.5; doxycycline hyclate12 not significant. Since the NNT is an average, the result could happen sooner or later than what is predicted by the NNT. In general, NNT calculations indicate that numerous sites need to be treated with adjunctive local delivery to attain an improvement at one additional site.12,14,16,21
•Changes at Deep Probing Sites—Table 7 reveals that only a few studies addressed the issue as to whether there was a greater improvement with respect to probing depth reduction at deep sites after local drug delivery when compared to SC/RP alone.12,17,21 A better result was not attained when doxycycline hyclate was compared to SC/RP.12 As indicated with minocycline microspheres, the result did not reach statistical significance.21 However, in another study that assessed the efficacy of chlorhexidine chips, Soskolne et al17 observed a better result with combined therapy than SC/RP alone (Table 7). Overall, there is a dearth of data demonstrating that local delivery enhanced results in deep pockets.
•Reduction of Probing Depths at Treated Sites To < 5 mm—Only 2 studies addressed the ability of local delivery to reduce probing depths to < 5 mm.14,21 In patients with a mean probing depth of ≥ 5, when SC/RP plus adjunctive minocycline spheres were used, there was a 1.59 times greater chance than SC/RP alone to achieve a mean probing depth of < 5 mm. Similarly, with combined therapy, patients with an initial mean probing depth of ≥ 6 mm had a 2.86 times greater chance than SC/RP alone to attain a mean probing depth of < 5 mm.21 However, the percentage of patients with this improvement was not reported. It should also be noted that the odds ratio cannot be used to determine the frequency that this will occur. For example, when an odds ratio of 3 is reported, it may only reflect 3 vs. 1 patients or it may represent 30 vs. 10 patients, etc. In another study, when doxycycline hyclate was used in
conjunction with ultrasonic debridement and compared to the efficacy of SC/RP, it was reported that 58% vs. 50% of the sites were reduced to probing depth of < 5 mm.14 Note that if the NNT is calculated for these data, it indicates that 12.5 additional sites need to be treated to result in one additional site becoming < 5 mm.
•Inhibition of Disease Progression—Insufficient data exist to suggest that local drug delivery results in less disease progression. (5) Duration of the results. There are no data to indicate that local delivery facilitates a longer interval between maintenance visits.

(6) Which diseases can be treated. With regard to the commercially available local drug delivery devices, all studies were conducted in patients with chronic periodontitis. There are no data with respect to commercially available local delivery systems concerning treatment of other types of periodontitis (eg, aggressive periodontitis). One article in the literature addresses the use of a local delivery system not available in the Unites States (metronidazole gel) around implants.27
(7) Types of defects that can be treated and bone gain. With respect to the commercially available local delivery systems, there is only data for minocycline microspheres concerning the treatment of furcations. When it was used as an adjunct to SC/RP, it achieved a statistically significantly better result for probing depth reduction than SC/RP alone in 2 studies (1.1 vs. 0.7 mm21 and 1.26 vs. 0.99 mm).22 However, it should be noted that these results were attained after 3 applications of the drug during a 9-month period.

Only a few studies ad­dressed the relationship be­tween use of local drug delivery and bone deposition.20,22 It appears that there is no significant gain of bone when local delivery is employed. One study noted that use of a chlorhexidine chip plus SC/RP 8 weeks prior to bone grafting and a GTR procedure enhanced bone deposition.28 However, additional data are needed to corroborate this finding.

(8) Stages of therapy: when to employ local drug delivery. Most patients respond to SC/RP or ultrasonic debridement, thus the routine administration of adjunctive drugs during active therapy may be unnecessary. Consequently, the decision to employ local drug delivery as an aid to SC/RP during active treatment to try to enhance results should be based on an understanding of the data in the literature, lesion severity, and the patient’s medical and dental history. Overall, drug therapy may be most beneficially used at sites nonresponsive to conventional therapy.7,8 This clinical situation may occur during active treatment or during maintenance.
(9) Local drug delivery as an alternate therapy to SC/RP. When the effectiveness of SC/RP was compared to doxycycline hyclate as a monotherapy, there were no statistically significant differences with regard to probing depth reduction or gain of clinical attachment (Table 5).12,13 This could be interpreted to mean that it is not necessary to perform SC/RP or that is unnecessary to administer drug therapy because the same results could be attained without its application. Since mechanical instrumentation removes calculus, eliminates endotoxin on the root surfaces, and disperses biofilms, it is prudent to perform root instrumentation, and local delivery when used should be administered as an adjunct to mechanical instrumentation.
(10) Cause of bacterial resistance. Five studies concluded that there was a transient selection of resistant microbes after therapy, and that after several months there was no increase in the number of resistant organisms.29-33 However, these were all short-term (6-month) studies. In contrast, the literature indicates with regard to use of systemic antibiotics that there has been an increase in the level of bacterial resistance to drugs.34-36 Therefore, it is suggested that judicious pharmacological principles should also be followed when using local drug delivery because long-term studies after multiple uses of locally delivered drugs have not been performed.
(11) Local vs. systemic antibiotic therapy. Local and systemic drug delivery appear to attain the same results.37-39 However, only a few studies addressed this question. The following are benefits attributed to local drug delivery: high concentrations, minimum side effects, and no reliance on patient compliance. In contrast, systemic drug delivery facilitates treating multiple sites simultaneously, and this may be more efficient and cost-effective for some patients. In addition, systemic drug delivery facilitates treating reservoirs of bacterial re-infection such as the tonsils, saliva, and tissue-invasive organisms. In general, when multiple sites need to be treated, it may be prudent to consider the use of systemic antibiotics.
(12) Impact of local drug delivery on tissue-invasive bacteria.Certain bacteria, such as Actinobacillus actinomycetemcomitans (Aa) may be detected within the periodontal tissues.40 In this regard, several studies noted that local drug delivery has not been effective against tissue-invasive organisms.41-43 Therefore, in patients that may have forms of periodontitis that are associated with Aa (eg, localized aggressive periodontitis formerly called juvenile periodontitis), it may not be appropriate to employ local drug delivery.
(13) Reducing the need for surgery. A study by Loesche et al42 indicated that local drug delivery diminished the need for periodontal surgical procedures, but they used subjective criteria to assess the need for surgery, and surgical decisions were done prematurely at the first evaluation.42 In general, after SC/RP with or without antibiotics there often is a decrease in the perceived need for surgery. This is consistent with the idea that surgical needs should be determined at periodontal re-evaluation, after enough time has elapsed for healing to occur. The concept that systemic and/or local delivery of antimicrobials in combination with instrumentation might reduce the need for surgery requires further evaluation in well-designed, longitudinal, long-term studies.

This article discussed the utility of local drug delivery systems in the treatment of chronic periodontitis with regard to their statistical and clinical relevance. The data indicated that there are statistically significant and at times clinically significant benefits associated with using local drug delivery as an adjunct to mechanical instrumentation. Nevertheless, periodontal health can often be attained without the adjunctive use of drugs. Therefore, clinicians need to choose treatments for their patients based on both statistical evaluations and careful interpretation of the data as they relate to clinical findings.


  1. Haffajee AD, Socransky SS. Microbial etiological agents of destructive periodontal diseases. Periodontol 2000. 1994;5:78-111.
  2. Evans AS. Epidemiological concepts. In: Evans AS, Brachmen PS, eds. Bacterial Infections of Humans: Epidemiology and Control. New York, NY: Plenum Pub Co; 1991: 3-58.
    3. Offenbacher S. Periodontal diseases: pathogenesis. Ann Periodontol. 1996;1:821-878.
    4. Armitage GC. Development of a classification system for periodontal diseases and conditions. Ann Periodontol. 1999;4:1-6.
    5. Oliver RC, Brown LJ, Löe H. Periodontal diseases in the United States population. J Periodontol. 1998;69:269-278.
    6. Cobb CM. Non-surgical pocket therapy: mechanical. Ann Periodontol. 1996;1:443-490.
    7. Drisko CL, Cochran DL, Blieden T, et al for the Research, Science and Therapy Committee of the American Academy of Periodontology. Position paper: sonic and ultrasonic scalers in periodontics. J Periodontol. 2000;71:1792-1801.
    8. Tonnetti MS. The topical use of antibiotics in periodontal pockets. In: Lang NP, Karring T, Lindhe J, eds. Proceedings of the 2nd European Workshop on Periodontology: Chemicals in Periodontics. Chicago, Ill: Quintessence Pub Co; 1997: 78-109.
    9. Hanes P, Purvis JP. Local anti-infective therapy: pharmacological agents. A systematic review. Ann Periodontol. 2003;8:79-98.
    10. Goodson JM. Controlled drug delivery: a new means of treatment of dental diseases. Compend Contin Educ Dent. 1985;6:27-36.
    11. Killoy WJ. Controlled local drug delivery of antimicrobials in the treatment of periodontitis. Paper presented at: Periodontal Disease Management Conference, American Academy of Periodontology, 1994.
    12. Garrett S, Johnson L, Drisko CH, et al. Two multi-center studies evaluating locally delivered doxycycline hyclate, placebo control, oral hygiene, and scaling and root planing in the treatment of periodontitis. J Periodontol. 1999;70:490-503.
    13. Garrett S, Adams DF, Bogle G, et al. The effect of locally delivered controlled-release doxycycline or scaling and root planing on periodontal maintenance patients over 9 months. J Periodontol. 2000;71:22-30.
    14. Wennstrom JL, Newman HN, MacNeill SR, et al. Utilisation of locally delivered doxycycline in non-surgical treatment of chronic periodontitis. A comparative multi-centre trial of 2 treatment approaches. J Clin Periodontol. 2001;28:753-761.
    15. Grisi DC, Salvador SL, Figueiredo LC, et al. Effect of a controlled-release chlorhexidine chip on clinical and microbiological parameters of periodontal syndrome. J Clin Periodontol. 2002;29:875-881.
    16. Jeffcoat MK, Bray KS, Ciancio SG, et al. Adjunctive use of a subgingival controlled-release chlorhexidine chip reduces probing depth and improves attachment level compared with scaling and root planing alone. J Periodontol. 1998;69:989-997.
    17. Soskolne WA, Heasman PA, Stabholz A, et al. Sustained local delivery of chlorhexidine in the treatment of periodontitis: a multi-center study. J Periodontol. 1997;68:32-38.
    18. Heasman PA, Heasman L, Stacey F, et al. Local delivery of chlorhexidine gluconate (PerioChip) in periodontal maintenance patients. J Clin Periodontol. 2001;28:90-95.
    19. Azmak N, Atilla G, Luoto H, et al. The effect of subgingival controlled-release delivery of chlorhexidine chip on clinical parameters and matrix metalloproteinase-8 levels in gingival crevicular fluid. J Periodontol. 2002;73:608-615.
    20. Jeffcoat MK, Palcanis KG, Weatherford TW, et al. Use of a biodegradable chlorhexidine chip in the treatment of adult periodontitis: clinical and radiographic findings. J Periodontol. 2000:71:256-262.
    21. Williams RC, Paquette DW, Offenbacher S, et al. Treatment of periodontitis by local administration of minocycline microspheres: a controlled trial. J Periodontol. 2001;72:1535-1544.
    22. Meinberg TA, Barnes CM, Dunning DG, et al. Comparison of conventional periodontal maintenance versus scaling and root planing with subgingival minocycline. J Periodontol. 2002:73:167-172.
    23. Greenstein G. Clinical versus statistical significance as they relate to the efficacy of periodontal therapy. J Am Dent Assoc. 2003;134:583-591.
    24. Greenstein G, Lamster I. Efficacy of periodontal therapy: statistical versus clinical significance [guest editorial]. J Periodontol. 2000;71:657-662.
    25. McQuay HJ, Moore RA. Using numerical results for systematic reviews in clinical practice. Ann Intern Med. 1997;126:712-720.
    26. Greenstein G, Polson A. The role of local drug delivery in the management of periodontal diseases: a comprehensive review. J Periodontol. 1998;69:507-520.
    27. Mombelli A, Feloutzis A, Bragger U, et al. Treatment of peri-implantitis by local delivery of tetracycline. Clinical, microbiological and radiographical results. Clin Oral Implants Res. 2001;12:287-294.
    28. Reddy MS, Jeffcoat MK, Geurs NC, et al. Efficacy of controlled-release subgingival chlorhexidine to enhance periodontal regeneration. J Periodontol. 2003;74:411-419.
    29. Goodson JM, Tanner A. Antibiotic resistance of the subgingival microbiota following local tetracycline therapy. Oral Microbiol Immunol. 1992;7:113-117.
    30. Larsen T. Occurrence of doxycycline resistant bacteria in the oral cavity after local administration of doxycycline in patients with periodontal disease. Scand J Infect Dis. 1991;23:89-95.
    31. Preus HR, Lassen J, Aass AM, et al. Bacterial resistance following subgingival and systemic administration of minocycline. J Clin Periodontol. 1995;22:380-384.
    32. Larsen T, Fiehn NE. Development of resistance to metronidazole and minocycline in vitro. J Clin Periodontol. 1997;24:254-259.
    33. Walker CB, Godowski KC, Borden L, et al. The effects of sustained release doxycycline on the anaerobic flora and antibiotic-resistant patterns in subgingival plaque and saliva. J Periodontol. 2000;71:768-774.
    34. Walker C, Karpinia K. Rationale for use of antibiotics in periodontics. J Periodontol. 2002;73:1188-1196.
    35. Neu HC. The crisis in antibiotic resistance. Science. 1992;257:1064-1073.
    36. Murray BE. Problems and dilemmas of antimicrobial resistance. Pharmaco­therapy. 1992;12:86S-93S.
    37. Palmer RM, Matthews JP, Wilson RF. Non-surgical periodontal treatment with and without adjunctive metronidazole in smokers and non-smokers. J Clin Periodontol. 1999;26:158-163.
    38. Purucker P, Mertes H, Goodson JN, et al. Local versus systemic adjunctive antibiotic therapy in 28 patients with generalized aggressive periodontitis. J Periodontol. 2001;72:1241-1245.
    39. Noyan U, Yilmaz S, Kuru B, et al. A clinical and microbiological evaluation of systemic and local metronidazole delivery in adult periodontitis patients [published correction appears in J Clin Periodontol. 1997;24(6):447-448]. J Clin Periodontol. 1997;24(3):158-165.
    40. Lamont RJ, Yilmaz O. In or out: the invasiveness of oral bacteria. Periodontol 2000. 2002;30:61-69.
    41. Mandell RL, Tripodi LS, Savitt E, et al. The effect of treatment on Actinobacillus actinomycetemcomitans in localized juvenile periodontitis. J Periodontol. 1986;57:94-99.
    42. Riep B, Purucker P, Bernimoulin JP. Repeated local metronidazole-therapy as adjunct to scaling and root planing in maintenance patients. J Clin Periodontol. 1999;26:710-715.
    43. Mombelli A, Schmid B, Ratar A, et al. Local antibiotic therapy guided by microbiologic diagnosis. J Clin Periodontol. 2002;29:743-749.
    44. Loesche WJ, Giordano J, Soehren S, et al. Nonsurgical treatment of patients with periodontal disease. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 1996;81:533-543.

Dr. Greenstein is a clinical professor of the Department of Periodontology at the University of Medicine and Dentistry of New Jersey in Newark, NJ. He also has a private practice in Freehold, NJ. He can be reached at (732) 780-1450.