Background
The shortened dental arch concept is a potentially cost-effective approach in the management of reduced dentitions. This concept is globally accepted, but not widely practiced [
1]. A body of mainly circumstantial evidence shows that shortened dental arches, comprising all anterior teeth and three to five occluding units, provide a stable and functional dentition with respect to chewing ability, aesthetics and oral comfort [
2‐
5]. The functionality of shortened dental arches has been reflected in outcomes of studies on oral health-related quality of life (OHRQoL). The outcomes are rather controversial: on the one hand, shortened dental arches are found to be related to OHRQoL impairment [
6], especially when first molar contacts were absent [
7] and on the other hand, subjects with a shortened dental arch reported to be satisfied with their oral status [
8].
Oral health care aims at the retention of at least a functional and natural dentition throughout life. Therefore, besides functionality, longevity of shortened dental arches should also be taken into account when considering application of the shortened dental arch concept. The fact that shortened dental arch subjects have lost molars in the past implicitly indicates an increased risk for dental diseases in these subjects. It is reasonable to expect that this predisposition to dental diseases is not eliminated after applying the shortened dental arch concept and hence necessitate more treatment in course of time including additional tooth extractions. Further tooth loss may endanger the longevity of the shortened dental arch status and thus compromise oral function.
The consequences of this predisposition might be even more manifest in subjects with a shortened dental arch restored with removable dental prosthesis (RDP) because RDPs have been associated with increased incidence of caries and periodontal breakdown [
9,
10]. However, a randomised clinical trial revealed no statistically significant difference in frequency of tooth loss after 3 years of follow-up among shortened dental arch subjects with and without RDP [
11]. Furthermore, quality of life levels of subjects with a shortened dental arch with RDP were found to be almost identical to those of subjects with a shortened dental arch without RDP [
12]. In another study, subjects with a shortened dental arch only perceived benefits of RDP from a OHRQoL perspective if anterior teeth replacements are included [
6].
To our knowledge, the longest follow-up reports on shortened dental arches are based on a 9-year observational cohort study (3-, 6- and 9-year observations) [
13‐
15]. It was concluded that shortened dental arches could provide a functional dentition with long-term occlusal stability [
13,
14]. Furthermore, the study reported similar frequencies of signs and symptoms of temporomandibular disorders for shortened dental arches with and without RDP and complete dental arches (CDA) [
15]. The present study is evaluating the initial cohort by analysing 27 to 35 years follow-up data on the basis of information from patient records. The objective of this study was to evaluate the clinical course of shortened dental arches by (1) assessing its longevity; (2) investigating the management of tooth loss and (3) analysing interventions provided during the follow-up period.
It was hypothesised that shortened dental arches have shorter longevity and receive more restorative interventions and tooth extractions than CDA. Moreover, it was hypothesised that these effects are more prone in shortened dental arches plus RDP.
Discussion
This observational cohort study demonstrated that shortened dental arches could be preserved for periods of 27 years and over. On dentition level the number of restorative interventions provided per year was not significantly different among the groups, expect that ‘SDA’ received significantly more indirect restorations in the upper jaw than ‘CDA’. Partially this is due to the high number of abutment crowns needed for the high number of FDPs provided in SDA subjects. This indicates that on dentition level the cost of restorative treatment in a shortened dental arch is at least as high as the costs of restorative treatment in a CDA. However, on tooth level it was found that in SDA group compared to CDA group more direct restorations were provided per tooth per year, except for molars. A plausible explanation for these higher numbers of direct restorations per tooth per year in shortened dental arches is that a predisposition to dental diseases in this group, that caused the loss of molars in the first place, continues to have its effect on the remaining dentition. In course of time this will necessitate additional treatment since new carious lesions develop. However, the reasons for restorative treatment were often not available from the records.
Besides the initial restoration of carious lesions, fillings are also made to replace previous restorations for reasons such as fracture of the filling or secondary caries. A review of 10 surveys including 32.777 direct restorations revealed that more than 50 % of the provided restorations were replacements of previous restorations [
16]. Therefore, it is assumable that considerable numbers of direct restorations provided in this cohort were replacement restorations. Hence, the high numbers of restorations provided per year per tooth in SDA subjects (Table
5) can probably partly be explained by the fact that SDA subjects already had relatively higher restoration levels (i.e. comparable numbers of teeth with restorations but in total significantly fewer teeth) at baseline than CDA subjects (Table
3). The only exception was the number of teeth with direct restorations in the lower jaw (5.83 in CDA group vs. 3.39 in SDA), which can be explained by the large difference in numbers of teeth present among the groups (0.83 teeth absent in CDA group vs. 5.35 in SDA).
Another explanation might be that in shortened dental arches fewer teeth have to bear the loads that occur during chewing. Consequently, increased loading of fewer teeth in shortened dental arches compared to CDAs possibly results in a higher failure rate of fillings in shortened dental arches. However, a study on masticatory performance showed that in shortened dental arches significantly lower occlusal forces could be measured than in complete dental arches [
2]. In line with this, it is feasible that the higher number of direct restorations made in shortened dental arches is not due to overloading and subsequent failure of fillings.
For the present study, information from patient records from the dental school was used. If subjects incidentally visited a general dentist outside the dental school for treatment it is likely that relevant information is missing. In the Netherlands however, where this investigation was conducted, patients are generally loyal to their dentist and only seldom switch temporarily. Therefore we trust that the information in the dental school records is reasonably complete. However, to confirm the accuracy of the records, the data were compared with available X-rays and the investigation forms of the original cohort study.
The small sample size and selected group of dental school patients might limit external validity of this study, e.g. due to stricter maintenance protocols applied to these patients. On the other hand, we consider the quality of the data satisfactory for conclusive outcomes. Sixty-five percent of the dental records of the subjects of the original cohort were not retrievable, mainly of subjects who stopped attending the dental school. According to the Dutch legislation, patient files must be kept at least for 10 years from the moment patients unsubscribe. Fortunately, most files were kept longer, but still a reasonable number of archived files were destroyed. Probably most of these were destroyed at the time that the Nijmegen Dental School changed from paper to electronic patient files in 2003. The destruction of archived records seems to be rather arbitrary and as a result selection bias is considered absent or small.
The number of female subjects was proportionally high for the SDA group and SDA plus RDP group, whereas gender distribution for the CDA group was even (Table
1). This high percentage female subjects is a reflection of the dental school patient population. The CDA subjects, as being a control group, were selected by purpose sampling aiming at equal gender proportions. However analyses did not reveal gender effects and therefore correction for gender was not considered necessary.
Previous longitudinal studies on tooth loss reported mean numbers of 0.03 to 0.24 teeth lost per year [
17,
18]. This is in accordance with incidence of tooth loss found in the present study; incidence varied from 0.03 to 0.12 teeth per year depending on location and group, with the lowest incidence for the lower jaw in CDA group and highest incidence for the upper jaw in SDA plus RDP group. In the majority of SDA subjects in the present study however, tooth loss did not lead to loss of their SDA status: a considerable number of lost teeth were molars without opposing tooth whilst lost teeth causing interruption of the dental arch were replaced by FDPs, by what means the SDA status was maintained. However, especially as incidence of tooth loss has been reported to increase with age, further tooth loss can be expected for this meanwhile aged cohort, which can endanger the functionality of the dentitions [
18,
19].
It is striking that three of the 23 SDA subjects (these were also the subjects who lost their SDA status) were accountable for 52 % of the lost teeth. In contrast to this, 6 SDA subjects did not lose any tooth at all during the follow-up period. Apparently in these subjects the predisposition to dental diseases, as argued above, could be stopped or at least substantially decreased. To what extent and how risk factors for tooth loss exactly continue to have their effect in SDA subjects, is a phenomenon that needs further investigation.
At the start of the study it was hypothesised that adverse effects are more prone in SDA plus RDP subjects than in SDA subjects. However, it appeared that the number of tooth extractions per year was not statistically significant different. This is in accordance with the outcomes of a randomised clinical trial on tooth loss in shortened dental arches with or without RDP; Kaplan-Meier survival rates were not statistically significant different between the two groups in a 3-year follow up period [
11]. Although the incidence of tooth loss is not significantly different between groups, 7 out of 13 SDA plus RDP subjects lost their SDA plus RDP status during the follow-up period whilst only 3 out of 23 SDA subjects lost their SDA status. This is also reflected in the loss of POPS; 62 % of the SDA plus RDP subjects lost one or more POP vs. 22 % of the SDA subjects. However, not all subjects lost their ‘SDA plus RDP’ status due to further tooth loss; three subjects lost this status because they stopped wearing their RDP and two of them actually gained POPs by having their RDPs replaced by free-end FDPs.
The number of restorative interventions for SDA plus RDP group provided per year during the follow period was not significantly different indicating that the costs at dentition level are the same as in SDA. When the costs related to RDP are taken into account, it can be concluded that the total costs in the SDA plus RDP group was higher whereas the longevity appeared to be lower compared to shortened dental arches without RDP. Additionally, apart from costs, every new RDP or RDP adjustment can bring discomfort and will make a considerable appeal on the adaptability of a patient. Moreover, it is questionable whether SDA patients actually benefit from RDP. As stated before, RDPs seem to contribute to OHRQoL only if anterior teeth replacements are included [
6]. Furthermore, Aras et al. showed that RDPs in shortened dental arches did not improve masticatory performance [
2] and McKenna et al. showed [
20] that both prosthetic rehabilitation to a functional dentition as well as full rehabilitation including RDP did not improve the nutritional status as reflected in hematological markers. Several studies, including the present study, showed that RDPs often even have an adverse effect [
10,
21‐
23]. In a randomised clinical trial on caries incidence following restoration of shortened lower dental arches in an elderly sample of patients, it was found that 2 years after restoration, there was a significantly greater incidence of new and recurrent caries lesions in subjects restored with RDPs compared with cantilever resin-bonded bridges [
22]. In the same sample of elderly, it was found that subjects considered restoration with cantilever resin-bonded bridges more comfortable than restoration with RDP [
23]. Also, a higher maintenance frequency for RDPs compared to resin-bonded FDPs in shortened lower dental arches was reported [
10]. In summary, it can be stated that replacement of absent posterior teeth by a free-end RDP in a shortened dental arch is not recommendable; fixed appliances (cantilever (resin-bonded) FDP or implant supported FDP) might be preferable alternatives.
Recently, the body of evidence on the SDA concept was assessed by means of the Grading of Recommendation Assessment and Evaluation [
24].The conclusion of this assessment was that the quality of the evidence for recommendation of the management of shortened dental arches is low because of the lack of evidence provided by randomised clinical trials [
24]. However, conducting well-designed randomised clinical trials may be not feasible because of concerns of ethical and practical nature. Although the present study is not a randomised clinical trial, the strength of the present study is that it is a long-term clinical observational cohort study that provides valuable, long-term clinical data on the clinical course of shortened dental arches.