SEARCH

SEARCH BY CITATION

Keywords:

  • cost effectiveness;
  • cost minimization;
  • implants;
  • maintenance;
  • peri-implantitis;
  • periodontal disease;
  • re-treatment

Abstract

  1. Top of page
  2. Materials and Methods
  3. Results
  4. Discussion
  5. References

Background

Little is known about the cost minimization and cost effectiveness involved in maintaining teeth and implants for patients treated for periodontal disease.

Materials & Methods

A retrospective study was carried out encompassing all patients who had initial periodontal treatment followed by implant placement and maintenance therapy in a specialist practice in Norway. The neighbouring tooth and the contra-lateral tooth were used as controls. The number of disease-free years and the extra cost over and above maintenance treatment for both teeth and implants were recorded.

Results

The sample consisted of 43 patients with an average age of 67.4 years. The patients had 847 teeth at the initial examination and received 119 implants. Two implants were removed 13 and 22 years after insertion. The prevalence of peri-implantitis was 53.5% at the patient level and 31.1% at the implant level. The prevalence of periodontitis was 53.4% at the patient level and 7.6% at the tooth level. The mean number of disease-free years was: implants: 8.66; neighbouring tooth: 9.08; contra-lateral teeth: 9.93. These mean values were not statistically significantly different from each other. The extra cost of maintaining the implants was about five times higher for implants than for teeth.

Conclusion

The number of disease-free years was the same for neighbouring teeth, contra-lateral teeth and implants. However, due to the high prevalence of peri-implantitis, the cost of maintaining implants was much higher than the cost of maintaining teeth.

Periodontal treatment followed by long-term maintenance has been shown to be successful in keeping the majority of patients' teeth (Hirschfeld & Wasserman 1978, Fardal et al. 2004, Fardal & Linden 2008). Maintenance often involves considerable re-treatment and extra cost (Fardal & Linden 2005, Fardal et al. 2012). A number of studies have reported fairly high prevalence levels of peri-implantitis of 7.8%–43.3% (Berglundh et al. 2002, Ferreira et al. 2006, Roos-Jansåker et al. 2006, Zitzmann & Berglundh 2008, Koldsland et al. 2010, Mir-Mari et al. 2012).

A major article estimating cost of periodontal disease control was published by Antczak- Bouckoms & Weinstein (1987). They introduced cost effectiveness, decision-making analysis and utility analysis. In addition, quality-adjusted tooth years (QUATY) was adapted from quality-adjusted life years (QUALY) (Broom 1993). For a number of years there were few studies that addressed the economic aspects of periodontal disease. Recently, a review article by Gjermo & Grytten (2009) concluded that there was a lack of quality studies of the economic aspects of periodontal therapy. Since then, cost effectiveness of adjunctive antimicrobials (Heasman et al. 2010), regenerative devices (Listl et al. 2010), sinus lift operations for implant placement (Listl & Faggion, 2010), supportive periodontal care (Pennington et al. 2011) and oral health educational programmes (Jönsson et al. 2012) have been investigated. In addition, a review article re-examining the cost of periodontal treatment has been published (Listl & Birch 2013).

However, little is known about the efforts and cost of maintaining dental implants in patients treated for periodontitis. A recent European Workshop on implant therapy, concluded that there were few studies on the pathogenesis and treatment of peri-implantitis. Only 6.4% of 982 studies addressed the issue, and only two studies that fulfilled the inclusion criteria were carried out on humans (Vignoletti & Abrahamsson 2012). In a review article it is pointed out that the following needs to be improved: the quality of reporting, case definitions and methods used to study incidence, prevalence and risk factors of peri-implant diseases (Tomasi & Derks 2012).

Few, if any, studies have addressed the cost effectiveness of implant therapy for patients treated for periodontitis. However, a recent study related patients' direct life cost to implant and prosthetic replacements (Fardal et al. 2012). The aim of this article was to compare teeth and implants during maintenance therapy in terms of the number of disease-free years and costs as part of a quality control measure.

Materials and Methods

  1. Top of page
  2. Materials and Methods
  3. Results
  4. Discussion
  5. References

Setting and study population

The patients were from one specialist practice in south-west Norway. The practice receives referrals from general dental practitioners, community dentists and physicians in rural communities with a total population of 25–30,000. The area has approximately 25 dentists divided evenly between private practice and the community dental service. The investigator is a specialist in periodontology, is the only periodontal specialist in the area and works in two practice locations (Egersund and Flekkefjord). Data from this practice have been used in several studies in which the progress of periodontitis has been described, and the effect of different types of periodontal treatment has been investigated (for example see: Fardal et al. 2003, 2004, Fardal & Linden 2008, 2010, Fardal 2006).

Our sample consisted of patients treated for periodontal disease in this practice between 1986 and 2012, and who had received one or several implants either after completion of initial periodontal therapy or during the maintenance phase. All the implants replaced teeth lost due to periodontal disease. The implants were restored with either single crowns, implant-supported bridges or a combination of teeth and implant-supported bridges. To obtain an adequately long observation time, we chose only to include patients with implants that had been in place for 7 years or longer. Our sample consisted of 43 patients. The analyses were also carried out on a sample with patients for whom the observation period was both longer and shorter than 7 years. This did not influence the conclusions. Table 1 presents characteristics of the patients.

Table 1. Description of the patient population
VariableMean/proportionStandard deviationRangeNumber of patients
Baseline values
Number of teeth per patient19.75.5[7–32]43
Number of implants per patient2.82.0[1–8]43
Age5110.8[29–74]43
Proportion of patients who smoked0.63  43
Number of cigarettes smoked per day per patient who smoked15.95.9[5–30]27
Proportion of patients on medication0.32  43
Changes during the observation period
Number of teeth lost per patient4.03.7[0–16]43
Number of implants lost per patient0.0460.21[0–1]43

Treatment prior to implant placement

All the patients completed a similar course of periodontal treatment, which included non-surgical therapy and surgical intervention when appropriate. Initial therapy included oral hygiene instruction, scaling and root planing using standard curettes. In the initial phase, scaling and root planing were completed without the use of local anaesthesia. The whole mouth was treated over a series of visits at 2–4 week intervals. Oral hygiene was reinforced repeatedly based on individual needs. The patients received a detailed explanation of periodontal anatomy and the disease process involved in periodontitis. Special emphasis was placed on the importance of periodontal maintenance therapy, following the initial definitive therapy. Periodontal surgery was prescribed for patients who had sites with bleeding on probing or persistent deep pocketing at reassessment 6 weeks after completion of the initial therapy. The decision to have implant therapy was taken jointly by the patient, the referring dentist and the periodontist (ØF).

Maintenance treatment after the implant placement

Maintenance visits with the specialist practitioner alternated with visits to the general dental practitioner, so that all patients were seen in total between two and four times per year. Written instructions were given both to the referring dentist and the patient outlining the plans for maintenance therapy. During each maintenance visit to the specialist, scaling, root planing and polishing of teeth and implants was routinely performed according to the needs of each patient. The interval between recall visits was shortened or lengthened as appropriate according to the stability of the periodontal/peri-implant condition.

During the maintenance period, sites with increasing probing depth were treated with repeated scaling and root planing. Subsequently, if there were clinical signs of residual subgingival calculus or persistent inflammation, surgical intervention was performed on teeth and/or implants. Standard flap operations designed to access diseased root/implant surfaces for cleaning/irrigations were used. No attempts were made to regenerate lost tissue. In addition, systemic or topical antibiotic therapy was used in cases of acute exacerbation of periodontal/peri-implant disease.

Outcome variables and analyses

Number of disease-free years

On the basis of patient records from 2012, we identified all the patients with implants that had been in place for 7 years or more. We also identified patients who had had peri-implantitis that had required comprehensive treatment. For these patients, we registered the number of years the implant had been in place before treatment for peri-implantitis was required. The diagnosis of peri-implantitis was based on the following criteria: radiographic proximal bone loss of at least three threads when compared with bone levels 1 year after loading in addition to the presence of bleeding and suppuration.

We calculated the sum for the number of disease-free years for all the patients' implants, and divided this sum by the number of patients. Our outcome measure was then the mean number of disease-free years per implant per patient. We then calculated the 95% confidence interval for this mean.

We repeated the same calculation, including the calculation of a 95% confidence interval, for the implant's neighbouring tooth and the contra-lateral tooth. The baseline was the year the implant was inserted. In this way, the number of disease-free years for the implant, the neighbouring tooth and the contra-lateral tooth was measured from the same starting point. It was assumed and verified from the patient records that periodontal condition was stable around the neighbouring tooth and the contra-lateral tooth at the time the implant was inserted. Implants and teeth could then be compared using the time when implants were inserted as the baseline. If the confidence intervals for the implant, the neighbouring tooth and the contra-lateral tooth overlapped, the susceptibility to recurrence of periodontitis and peri-implantitis would be the same for the implant and the control teeth.

In the analyses, we did not include control variables for the patient or the dentist. The comparison between the number of disease-free years was done for the implant and the control teeth for the same patient. In this way, we automatically control for all patient-specific characteristics. Also, it was not necessary to control for specific characteristics of the dentist, as all the periodontal treatment and implant treatment was carried out by the same clinician (ØF).

Cost of treatment

In the case of comprehensive treatment of periodontitis and peri-implantitis, we calculated the cost of treatment of the teeth and the implants separately. The cost was based on the same hourly rate of non-surgical and surgical fees for both implants and teeth (for further details about how the specialist services for periodontists in Norway are organized and financed see Grytten & Skau 2009). The actual cost per implant was then the total cost of treatment of peri-implantitis divided by the number of implants. Correspondingly, the actual cost per tooth was the total cost of re-treatment of periodontitis divided by the number of teeth. To adjust for the variation in time from when the patient completed his or her initial treatment to the time when the patient was re-treated, the cost per implant/tooth were calculated per year. The total costs were calculated based on the fees the patients paid the periodontal specialist for treatment and the cost for antibiotics has been included.

Results

  1. Top of page
  2. Materials and Methods
  3. Results
  4. Discussion
  5. References

Description of the patient population

Table 1 shows descriptive statistics for the patient population. There were 18 male patients and 25 female patients. The mean age at baseline was 51 years. The oldest patient who received an implant was 74 years, the youngest was 29 years. 53.1% of the patients got peri-implantitis associated with one or several implants. 53.4% of the patients got periodontitis associated with one or more teeth.

Altogether, 119 implants were inserted for 43 patients: 2.8 implants per patient. 16 patients received only one implant, while one patient received eight implants. Peri-implantitis occurred with 37 of the 119 implants: 31.1% of all implants. Two implants were removed, one after 13 years and the other after 22 years. There were no differences in the rate of peri-implantits between patients with single or multiple implants.

The numbers of patients (in brackets) according to type of implant were: Nobelbiocare TiU (29), Osseotite 3i (8), Strauman (4), Screw-Vent (2). The majority of implants were placed in the anterior region between the second pre-molars. There were 12 patients with single crowns, and seven patients with implant bridges. 24 patients had a combination of bridges with teeth and implants. Seven out of the twelve single implant crowns (58%) developed peri-implantitis, while five out of the seven implant bridges (71%) and twelve out of 24 combined implant/teeth bridges (50%) developed peri-implantitis.

Altogether, the 43 patients had 847 teeth at the initial examination: 19.7 teeth per patient. None of the patients had less than seven teeth. Periodontal disease developed round 64 of the 847 teeth that were present at the initial examination, and had to be treated, i.e. 7.6% of all teeth. The mean number of teeth that were lost during the observation period was 4.0.

At baseline 27 patients smoked. They smoked on average 15.9 cigarettes per day. There were no differences in the level of peri-implantitis between smokers and non-smokers (8 of 16 non-smokers developed peri-implantitis). 14 patients were taking medication, mainly for cardiovascular diseases.

Main results

For 37 of the patients it was possible to compare neighbouring teeth with the implants. The mean number of disease-free years was 8.66 for the implants and 9.08 for the neighbouring teeth. The confidence intervals for these two means overlapped (Table 2).

Table 2. The number of disease-free years for the implant, the neighbouring tooth and the contra-lateral tooth
Type of comparisonThe mean number of disease-free years per implant/tooth95% Confidence intervalRangeNumber of patients
Implant8.66[7.33, 9.99][2–17] 
Versus   37
Neighbouring tooth9.08[7.96, 10.20][2–27] 
Implant8.66[7.39, 9.93][2–17] 
Versus   29
Contra-lateral tooth9.93[8.49, 11.37][3–22] 

Twenty-nine patients had contra-lateral teeth that could be compared with the implants. The mean number of disease-free years was 8.66 for the implants and 9.93 for the contra-lateral teeth. These confidence intervals also overlapped (Table 2).

In Tables 3, 4 we present additional analyses for smokers versus non-smokers, and for patients with single implant restorations versus patients with multiple implant restorations. Table 3 shows a slightly better disease-free period for smokers than non-smokers. However, this is an unusual finding. For all subsamples, the confidence intervals for the mean number of disease-free years per implant/tooth overlapped. In addition, the sample is too small to investigate this further. Table 4 shows that all the confidence intervals for patients with one implant versus patients with more than one implant overlapped.

Table 3. The number of disease-free years for the implant, the neighbouring tooth and the contra-lateral tooth for smokers and non-smokers
Type of comparisonThe mean number of disease-free years per implant/tooth95% Confidence intervalRangeNumber of patients
Smokers
Implant9.50[8.08, 10.90][4–17] 
Versus   24
Neighbouring tooth9.75[8.53, 11.00][2–22] 
Implant9.28[7.69, 10.80][4–17] 
Versus   18
Contra-lateral tooth10.61[8.56, 12.66][6–22] 
Non-smokers
Implant7.15[5.51, 8.79][2–12] 
Versus   13
Neighbouring tooth7.84[6.30, 9.37][2–12] 
Implant7.64[5.63, 9.65][2–12] 
Versus   11
Contra-lateral tooth8.82[7.11, 10.53][3–13] 
Table 4. The number of disease-free years for the implant, the neighbouring tooth and the contra-lateral tooth for patients with single implant restoration and for patients with multiple implant restoration
Type of comparisonThe mean number of disease-free years per implant/tooth95% Confidence intervalRangeNumber of patients
Single implant restoration   
Implant8.17[6.28, 10.06][4–13] 
Versus   12
Neighbouring tooth8.83[6.28, 11.38][2–17] 
Implant8.17[6.27, 9.99][3–13] 
Versus   12
contra-lateral tooth10.08[8.03, 12.13][7–17] 
Multiple implant restorations   
Implant8.92[7.50, 10.43][2–17] 
versus   25
neighbouring tooth9.20[7.62, 10.78][2–22] 
Implant9.00[7.29, 10.71][2–17] 
versus   17
contra-lateral tooth9.82[7.78, 11.80][3–22] 

The total cost of treatment of peri-implantitis was € 13,100 (non-surgical € 5100 and surgical € 8000). The mean cost per implant for the whole observation period was € 110. After adjusting for the number of years the implants had been in place, the cost was € 10.2 per implant per year (Table 5).

Table 5. Costs for treatment of peri-implantiitis and re-treatment of periodontitis in Euro
 Implants (mean)Teeth (mean)
Total costs13, 10029, 450
Number of implants/teeth119847
Costs per implant/tooth the whole observation period11035
Observation period (in years)1016
Costs per implant/tooth per year10.22.1

The total cost of re-treatment of teeth was € 29,450 (non-surgical € 6450 and surgical treatment € 23,000). The mean cost per tooth for the whole observation period was € 35. After adjusting for the observation time, the cost was € 2.1 per tooth per year (Table 5).

There were no differences in the cost of maintaining the implants compared with neighbouring or contra-lateral teeth. This follows directly from Table 2, which shows that there was no difference in the mean number of disease-free years per implant/tooth for implant versus neighbouring tooth, and for implant versus contra-lateral tooth.

Discussion

  1. Top of page
  2. Materials and Methods
  3. Results
  4. Discussion
  5. References

This study was carried out on a sample of patients who had all been treated for periodontitis. They had received one or more implants, and had been under continual follow-up afterwards. However, the incidence of recurrent periodontitis and peri-implantitis was relatively high, but not markedly different from that reported in other studies (Berglundh et al. 2002, Fardal & Linden 2005, Ferreira et al. 2006, Roos-Jansåker et al. 2006, Zitzmann & Berglundh 2008, Koldsland et al. 2010, Mir-Mari et al. 2012).

An important finding was that the number of disease-free years was the same for implants and teeth. This was also the case when the heterogeneity in the patient population was taken into account. None of these results for smokers versus non-smokers, and for patients with one implant versus patients with more than one implant were different from when the analyses were done on the whole sample. However, the costs for re-treatment was markedly higher if implants were inserted compared to if patients kept their own teeth. This is because there were more implants with peri-implantitis than teeth with recurrent periodontitis: when all teeth were compared with all implants there was a significantly lower prevalence of recurrent periodontitis than that of peri-implantitis. This is in agreement with a systematic review reporting better long-term outcomes for teeth than for implants (Tomasi et al. 2008).

The underlying assumption for our analyses is that the risk of peri-implantits around a fixture which has been placed where a tooth has been lost is similar to the risk of re-emergence of periodontal infection around a tooth that remains in situ. This assumption may be questionable. To our knowledge, there are no studies where this assumption has been examined. One strength of our study is that we use a research design that takes into account the possibility that the risk may vary according to local factors (by comparing implant versus neighbouring tooth), according to site specificity/tooth morphology (by comparing implant versus contra-lateral tooth) and the mouth as a whole (by comparing the total number of implants with the total number of teeth). However, in spite of these comparisons, the clinical conditions may still be different. Even though teeth were lost due to periodontal disease, this does not necessarily imply identical clinical conditions at implant sites and sites where teeth could be maintained. On the other hand, one advantage with our study is that we use three different comparison groups, and the results are similar independent of which group is used.

In this study, the total failure of implants was very low in spite of a high prevalence of peri-implantitis. Only two implants were removed after 13 and 22 years, suggesting that a very active maintenance therapy including surgical treatment of peri-implantitis was successful in maintaining implants previously exposed to disease. The importance of maintenance therapy for implants has also been reported previously (Rocuzzo et al. 2010).

Different types of implant systems were used during the study period from 1986 to 2012. The difference in surface characteristics could have influenced the results in this study as suggested by Renvert et al. (2011). However, as the vast majority of fixtures used were Nobelbiocare TiU this study did not lend itself to a comparison between systems. No apparent differences were observed clinically between the systems.

A relatively high tooth loss rate from periodontal disease was found in this population when compared with previous reports from the same practice setting (Fardal et al. 2004). This underlines the fact that the patients in this study were high-risk patients with marked loss of periodontal support. In spite of maintenance therapy over a number of years, these patients lost several teeth. Some of these teeth were replaced by implants after a healing period.

It is difficult to verify the results from this study due to a lack of comparable studies. Another study from the same practice setting calculated direct life-time patient costs for periodontal treatment (Fardal et al. 2012). However, a number of the outcomes were extrapolated from shorter-term studies. Gaunt et al. (2008) examined the cost of periodontal maintenance treatment, however, this study was not done in real time. The strength of this study is that it is done in real time with no extrapolations, the patients served as their own internal controls and the same experienced clinician carried out all the treatment.

In any health economic analysis, it is important to adhere to underlying welfare economic theories (Vernazza et al. 2012). Usually, it is debatable which tools are best suited for a particular clinical study. It is therefore important to examine the reasons why the specific tools were chosen for this study. First of all, it was decided to define the variables as a measure of technical efficiency, and therefore to utilize cost minimization and cost effectiveness. However, cost minimization analysis ideally should only be used when the interventions or strategies being compared are already known to have equal effectiveness. In this study, the assumption is that the effectiveness of re-treatment of periodontal disease is the same as for treatment of peri-implantitis. This suggestion has not been verified by other studies, but from a clinical standpoint and from the results of the study, it seems reasonable to use this as a hypothesis. A recent Cochrane systematic review (Esposito et al. 2012) reported clinical improvements from the treatment of peri-implantitis. However, they did not find that complex interventions were more effective than control therapies with surgical debridement as chosen for this study. Furthermore, a study from the same clinical setting as in this study reported that a very active maintenance therapy often including surgical retreatment for the recurrence of periodontal disease was effective in the long term, in terms of tooth loss (Fardal & Linden 2005). Second, cost effectiveness is usually measured in natural units, and in this study, disease-free years were used. However, these measurements do not take into account the impact on the patients in a wider sense, and do not answer any questions of allocative efficiency. In a practical sense it is difficult to employ methods like standard gamble, time trade off and quality adjusted tooth years in a private practice setting. A willingness to pay approach has previously been described for periodontal therapy (Matthews et al. 1999). This method was considered for this study, however, it was feared that the patients would misunderstand this approach and use it to bargain with the dental fees quoted.

The results of this study should be interpreted keeping in mind that they were obtained in a single private practice in a particular geographic area with a relatively small, heterogeneous sample with a large proportion of smokers. Thus, generalizing may not be appropriate. Regardless of this, our findings show that the number of disease-free years was the same for neighbouring teeth, contra-lateral teeth and implants. However, due to the high prevalence of peri-implantitis, the cost of maintaining implants was much higher than the cost of maintaining teeth.

References

  1. Top of page
  2. Materials and Methods
  3. Results
  4. Discussion
  5. References
Clinical Relevance

Scientific rationale for the study: Little is known about the efforts and cost of maintaining dental implants in patients treated for periodontitis and how this compares with the active maintenance of teeth.

Principle findings: Neighbouring teeth and contra-lateral teeth and implants have the same susceptibility to recurrence of periodontal disease and peri-implantitis under the same oral conditions.

Practical implications: Active maintenance treatment is important for the long-term success of implants for patients treated for periodontal disease.