Enhanced Treatment Effects After Adjunct Antibiotics

In lectures and seminars for graduates and undergraduate students, or many invited talks I have long pointed with regrets to the way how painstakingly collected site-specific periodontal data are frequently treated when aggregating them at the subject level. The main reason for intentionally (or unintentionally?) losing instantly much if not most information by applying this approach is usually mentioned: statistical pundits have pointed to the fact that standard statistical tests require independent observations. On a number of occasions I have used for clarification just an abstract of a published paper by Forsyth Institute’s periodontal group which explains the methods applied,

“Subgingival plaque samples were taken from the mesial aspect of each tooth in 635 subjects with chronic periodontitis and 189 periodontally healthy subjects. The samples were individually analyzed for their content of 40 bacterial species using checkerboard DNA-DNA hybridization (total samples = 21,832). Mean counts, % DNA probe counts and percentage of sites colonized at >105 were determined for each species in each subject and then averaged in each clinical group. Significance of difference between groups was determined using Mann-Whitney test. Association between combinations of species and periodontal status was examined by stepwise logistic regression analysis.” (Emphasis added.)

So, tens of thousands of bacterial samples were analyzed for forty bacterial species using highly sophisticated and expensive molecular biological methods (and most probably clinical signs of periodontal disease were assessed as well) which had been collected in more than 800 subjects; and then all this wealth of information is just dumped when mean counts and percentages of sites more heavily colonized were calculated in each of the subjects. One would have saved enormous an amount of money and time if one pooled sample per subject had been analyzed, and I still would argue that this biological pooling the material would be more reasonable than considering statistically pooled data. And then, data were analyzed by non-parametric Mann-Whitney test which uses ranks of observations rather than observations themselves. Well, as this approach has been adopted by the group in dozens if not hundreds of scientific papers since the late-1990s and hardly ever criticized, a more interesting approach using multilevel modeling had been applied in the summer when very old clinical data of one of Forsyth’s therapy studies had been re-analyzed and -published in a new open-access microbiological journal (implying that the respective microbiological data are upcoming). The paper by Mdala et al. (2012) [pdf], apparently using sites as statistical units in a time series model, would of course be most welcome provided the analysis was flawless and the given information complete. But since conclusions regarding adjunct antibiotic treatment of chronic periodontitis are  grave and may imply even recommendations, a close look at what has been done is mandatory. I had contacted meanwhile the corresponding author of the paper and had been provided with some additional material which is strangely not accessible on the journal’s web page (although mentioned as supplementary material in the text). But my main concerns have not been addressed so far. In the abstract, the authors write (and CAL means here clinical attachment level, SURG periodontal surgery, PD pocket depth; AMOX and MET refer to respective systemic drugs amoxicillin and metronidazole, while TET is a no longer available tetracycline fiber for topical use in periodontal pockets),

“Surgically treated patients experienced most CAL loss. Adjunctive therapy including SURG was most effective in reducing PD. Combining SURG with AMOX, MET, and TET gave significant clinical benefits. Past and current smoking habits were significant predictors of deeper PD. Only current smoking was a significant predictor of CAL loss. Bleeding, accumulation of plaque, gingival redness, and  suppuration were significant predictors of further CAL loss and deeper PD.”

The pretty old study (most data had been collected more than ten years ago, but this information can only be retrieved from another paper by the group describing separate aspects)  has had a complicated design, where eight treatment modalities had randomly be assigned to patients. Scaling and root planing was the standard treatment provided to which the other, increasingly complex, treatments were compared. When the authors claim in the abstract (and I am afraid that many clinicians skip reading the whole, likely incomprehensible, paper) that surgically treated patients experienced most loss of clinical attachment (without giving any numbers), one may already be alerted. Buried on the last page of the manuscript (Table 5), the average difference of change in attachment level two years after periodontal surgery (as compared to what was achieved with scaling and root planing alone) was in fact minus 0.40 mm which in fact compares unfavorably to plus 0.67 mm (as compared to scaling and root planing alone) in the group which received systemic amoxicillin and metronidazole in addition to scaling and root planing. (Note that all subjects received subgingival scaling. Notably, the paper does not favor, as others of the group in fact do, only systemic antibiotics). All other treatment modalities yielded (and it is clear that all estimates were derived from a hierarchical, four-level model) suboptimal (as compared to the group receiving amoxicillin and metronidazole) but superior results than scaling and root planing alone, just between plus 0.67 and minus 0.40 mm. So, all mean results were within 1.07 mm difference to those achieved with scaling and root planing. This might actually demand an exercise on the difference between statistical significance and clinical relevance in another posting. And, of course, on overwhelming evidence of certain periodontal conditions being indications or contraindications for periodontal surgery. Timely randomized controlled clinical studies are most welcome as they provide, if conducted properly, a high level of evidence. But there is progress over time. That periodontal surgery is mainly indicated if periodontitis has resulted in certain anatomical defects, ie. deep infrabony lesions and furcation involvement which are largely inaccessible to scaling and root planing, but not indicated otherwise is meanwhile commonplace. Since these lesions are less frequently found than more accessible periodontal lesions and since periodontal surgery is by definition much more invasive (minimally invasive surgical techniques had apparently not be applied), I would not be too surprised when the net effect in a group of patients treated with (mostly not indicated) periodontal surgery would be slight loss of attachment (as compared to scaling and root planing which was mostly indicated). So far so bad. Now let’s have a look at the model. The authors considered “a four-level model with random intercept and random effect of time on patients at level 3, random intercept for teeth at level 2, and tooth sites at level 1.” They first modeled pocket depth for any given patient at time t corresponding to baseline visit and 3, 6, 12, 18, and 24 months visits on site i, in tooth j nested within patient k. The model (on p. 5) comprises a fixed part with time, treatment and the interaction of the two which was adjusted for subject level co-variates gender, age, ethnicity, nationality and smoking; as well as site level co-variates “baseline pocket depth, plaque, BOP [bleeding on probing], suppuration, [and] gingival redness.” The random part of the model consists of the patient random effects associated with the intercept and time slope; and the random effect of a tooth nested within a patient, a site nested in a tooth which in turn is nested in a patient, and a random error which is assumed to be homogeneous and uncorrelated with mean 0 and variance sigma squared. While baseline pocket depth may in fact determine pocket depth after treatment, it is not so clear why baseline plaque, bleeding on probing, suppuration and gingival redness had to be included in the model. On the other hand, it is well-established that plaque, gingival inflammation, or suppuration after treatment do negatively influence the treatment results (as described as both pocket depth reduction and clinical attachment gain). So, why have respective postoperative co-variates not been included in the model? The model does not contain any tooth-related co-variates, for example multi-rooted as compared to single-rooted teeth, or posterior as compared to anterior teeth, periodontal lesions around the respective former being more demanding (ie. less accessible) than around the latter. The model implies that alterations in pocket depth are linear over time. But that is highly unlikely. In a correct time series analysis, time squared or even a cubic term for time should have been considered as well. Note that a model can yield only reliable estimates in so far as it describes reality. I have exercised this recently when analyzing, in a three-level (occasion, tooth, subject) model, gingival thickness as measured at the gingival margin and at the mucogingival border, and thickness of vestibular lining mucosa after implantation of a bio-resorbable membrane in a surgical root coverage procedure (Müller 2008 [pdf]).  Mucosal thickness was modeled at all three measurement locations in a multivariate time series model with time (linear, quadratic, and cubic components), gingival width, jaw, as well as respective interactions with time as co-variates using specialist software MLwiN. The model (see below, just click on the box twice) was carefully set-up and is admittedly rather complex. But it allows the calculation of all necessary estimates (predictions) and also assessment of the random part with variances and covariances at subject, tooth and occasion levels. Muller 2008 MTH Prediction of mucosal thickness as derived from a multivariate, three-level time series model (Müller 2008) are shown in the figure below. Thickness was measured at the gingival margin (GM), and at the mucogingival border (MB). Thickness of the lining mucosa (LM) was measured as well. Estimates in relation to time (baseline [BL] and at 3-month follow-up examination intervals after surgery, 3–12 months), jaw, and baseline gingival width (GW) are given. Predictions indicate different patterns for teeth in the maxilla and in the mandible. Moreover, gingival thickness was positively correlated at baseline and again after 9 and 12 months, with baseline gingival width. The considerable swelling of mucosa at all maxillary measurement locations and of lining mucosa in the mandible 3 months after surgery was, however, negatively related to baseline gingival width. The random part of the model yielded significant correlations between mucosal thickness measurements at the different locations. Fig. 2 In preparation of the above model which was set-up to yield the respective estimates of fixed effects, or predictions, different models were considered. For instance,

“Variance partition unadjusted time series models revealed that most of the variance of gingival thickness, its width, and the location of the mucogingival border relative to the cemento-enamel junction can be found at the occasion level. Gingival width did not change after surgery, while gingival thickness and the location of the mucogingival border underwent considerable shifts with time including significant linear (velocity), quadratic (deceleration), and cubic terms.”

This recent paper (in a series of articles with the aim of giving a broad overview on the potentials of multilevel modeling of periodontal data) revealed, by using a carefully set-up multilevel model which main purpose was to use all information collected in a longitudinal study on the performance of a bio-absorbable membrane for regenerative surgical root coverage in order to reflect reality as far as possible, an astonishing amount of new information not only as regards reasonable estimates of fixed effects fitting clinical observations but also new insights in the random part (which is of paramount importance when treating future patients and predict variation in outcome). I have occasionally compared multilevel modeling with sort of a microscope which allows for deep views into data structure. In contrast, the above time series model by Mdala et al. (2012) falls far short of providing anything new. Unfortunately, the random parts in the models (whose design has to be considered insufficient) are not explained or even shown (a specific request when having contacted Dr. Mdala by email has not been answered) although one would like to see variance partition at the subject, tooth, and site levels, and whether common rules of thumb in decision making mainly based on site-specific clinical features hold. Frequently reported problems with extrabinomial variation, in particular underdispersion, in multilevel logistic regression models of periodontal data, see for instance Müller and Stadermann 2006 [pdf], are not discussed. The main purpose of applying rather sophisticated models apparently was estimation based on just fixed effects rather than previous simple calculation of averages. The authors apparently had perceived certain problems when setting-up the models as they write,

“The use of multilevel analysis is on the frontier in periodontal research. However, available software for analyzing such models sometimes gives different results, and if the data are huge and depending on the hierarchy of the model to be fitted, they tend to converge slowly and in some cases never converge at all.”

What is of major concern, though, is the authors’ conclusion,

“We have shown that while standard periodontal treatments such as SRP or SURG may improve periodontal conditions, the use of adjunctive systemic or local antibiotic treatments would be a better choice especially if used as combination therapies. If SURG is to be used, it should be combined with AMOX MET TET to maintain the best clinical results.”

I beg to differ. The indications for periodontal surgery are well-established, deep infrabony lesions and advanced furcation involvement. These defects do not disappear after a course of adjunct systemic and/topical antibiotics, be it in conjunction with or without subgingival scaling. If, after periodontal surgery, the anatomical defect however resolves (and that has been shown to be likely to occur in numerous clinical trials, if and only if sufficient plaque control can be maintained), attachment gain and pocket reduction amount to far more than parts of a millimeter, which is clinically irrelevant in particular when considering possible serious side effects of combinations of broad spectrum antibiotics and a possible need for anew prescription after some time. Rather than re-analyzing 10- or 15-yr-old data, which had been collected under completely different premises, new randomized clinical trials should address current open questions regarding true endpoints such as long-term tooth survival (rather than surrogate measures such as mean fractions of a millimeter pocket reduction and/or clinical attachment gain) including the application of sophisticated tools for data analysis which preserve the wealth of collected information. 2 December 2012 @ 13:45. Last modified December 2, 2012.



  1. Pingback: Need For Periodontal Surgery « Periodontology
  2. FMS

    Dear Prof,

    Besides Herrera 2008, there is a recent systematic review by Sgolastra et al. on the use of amox/metronidazole which suggests a positive effect, specially for deep PPD. According to data from Heitz Mayfield, the benefit of surgical periodontal therapy compared to non surgical periodontal therapy, in terms of mm of PPD reduction or CAL gain is comparable to the results obtained with the use of Antibiotics. The report of side effects seems to vary widely between studies but , in any case, are consider minor by the patients and most of the times do not lead to dropout from the studies. As described by Mombelli, from a clinical and economic point of view the use of systemic antibiotics as an adjunct to initial NST can be defended. You can argue that even the possible impact on antibiotic resistance spread is debatable, specially given the inappropriate prescribing patterns shown both in General Medicine and General Dental practice or the well spread inadequate use of antibiotics in agriculture.Obviously, the use of antibiotics is not appropriate for cases of moderate periodontitis, mainly affecting flat surfaces, as they are not likely to provide any benefit. For deep infrabony defects, however, it may not only reduce the need for periodontal surgery (Cionca, Guerrero) but provide a better presurgical enviroment which will enhace the chances of success of the corrective therapy.

    It is a pleasure to read your blog.

    Best Wishes!


    • Muller

      Thanks for giving me the opportunity of being more specific. Apart from awkward statistical analysis in the paper by Mdala et al. (2012) which had inspired me to write this post my main concern is about true endpoints. I would not argue that broad spectrum antibiotic therapy as adjunct to nonsurgical periodontal therapy would result in slightly better results as regards probing depth reduction. But I doubt whether sites with a likely indication for periodontal surgery, i.e. deep infrabony lesions and furcation involvement, would benefit in the long run. These anatomical defects will not just disappear after (insufficient due to access problems) scaling because of systemic antibiotics.

      Mombelli et al., in a paper to be published in J Perio which just went online, see pdf here, report a primary outcome of probing depth of >4 mm and bleeding on probing 3 months after nonsurgical periodontal therapy. This is something else than infrabony lesions or furcations, right? Those who received adjunct systemic amoxicillin and metronidazole had fewer numbers of these sites than those who did not, irrespective of whether they harbored Actinobacillus actinomycetemcomitans (A.a.) or not. Well, when looking at the data and how they were presented (see the box plot in Fig. 2) one has to conclude that few subjects among those not receiving antibiotics had been responsible for most of these sites (>4 mm PD and BOP). The median number in this group was about 3 for those positive for A.a. and 2 for those without A.a. 3 months after subgingival scaling (and admittedly lower in those who received systemic amoxicillin and metronidazole), not too bad. But there were a few subjects with 32, 23, 17, 16 etc. persisting pockets. It would have been helpful if the authors had presented more information about those few patients in order to get an idea who really is in need of antibiotics.

      What is going to happen at sites with rather shallow probing depths after adjunct antibiotics, in which periodontal surgery would have been indicated according to common sense, is not clear. As I mentioned before, I doubt whether the anatomical defects will just disappear. How often do we have to repeat antibiotic therapy then?

      Best, Hans-Peter


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s