Essential endodontology: prevention and treatment of apical periodontitis. 3rd ed

Chapter 5. Epidemiology, Treatment Outcome, and Risk Factors for Apical Periodontitis

Lise-Lotte Kirkevang and Michael Vaeth

5.1 Introduction

Apical periodontitis is a common disease; in some populations, about half of the adult population have at least one tooth with apical periodontitis. If untreated, apical periodontitis may cause pain, reduce chewing function, and eventually lead to tooth loss. Prevention and treatment of apical periodontitis have become a central part of a dentist's work and an essential aspect of dental health care planning. Identification of suitable preventive measures and optimal treatment strategies therefore become important.

A patient seeking dental treatment is mainly concerned with elimination of pain and discomfort. The patient expects that treatment decisions are based on best available evidence and wish to be informed about the benefits and risks associated with different treatment alternatives. The dentist must be able to share evidence-based information with the patient so that the patient can select a treatment based on their individual values and preferences. Browsing the internet, patients can easily obtain further information on available treatments, and use it to challenge the clinician's recommendations. The dentist must therefore have up-to-date knowledge about the supporting evidence of the treatment procedures that are presented to the patient.

To select the best available treatment to a patient the prognosis of the different treatment alternatives needs to be known. What is the probability of a successful outcome? The information needed to answer this question can only be provided by clinical and epidemiologic studies [51]. Due to the usually asymptomatic, chronic nature of periapical inflammation, evaluation of endodontic treatment is primarily based on radiographic recording of periapical structures, while clinical signs and symptoms may contribute additional information regarding clinical functioning and patient satisfaction. Information from clinical research represents a best-case scenario of what can be achieved with present knowledge and modern treatment principles, while epidemiologic studies may elucidate to what extent the dental profession is succeeding in controlling and eliminating apical periodontitis.

Moreover, to identify the best treatment, the meaning of “best” must be established. What is the relevant outcome and how do we define a successful outcome? Endodontic outcome studies have usually focused on apical periodontitis, and defined success as the absence or reduction of a radiographic periapical lesion. However, for patients, absence of pain, return of chewing function, or preservation of a tooth may be the most important aspects of a successful treatment. Moreover, additional complications arise when comparing the relative success of endodontic and implant treatment. Comparable outcome measures are needed but rarely used.

Table 5.1 Common epidemiological terms.

Population	A group of people in a defined setting
Sample	A selected part of a population
Selection bias	Systematic overor underestimation of results due to distortion of a sample
Length bias	Distortion of sample due to overrepresentation of long disease duration
Information bias	Systematic overor underestimation of results due to inappropriate data collection or registration
Confounding	Systematic overor underestimation of results caused by a variable that is associated with both outcome and the study factor in question
Confounding by indication	Confounding resulting from the treatment procedure not being chosen at random, but being guided by, either the nature of the disease, or by a standard procedure at an investigating clinic.
		Exposure	A study factor thought to cause disease
Outcome	A measure of present disease
Prevalence	The number/proportion of cases of a disease at a given point in time
Incidence	The number/proportion of new cases developing over a time period
Risk	The probability associated with the outcome in a specified period

Thus, a “successful treatment” has different meanings, and results from different studies may be difficult to compare.

In this chapter, the current knowledge about the prognosis of apical periodontitis is described. This knowledge derives from studies using clinical and epidemiological research approaches. Experimental research focuses primarily on diseased mechanisms or evaluation of technical equipment under strictly controlled laboratory conditions, and these results are seldom directly applicable to a real-life situation. Traditionally, epidemiologic research investigates how diseases occur and persist in different groups of people or in populations, while clinical research concerns the individual patient and the treatment of their health problems. But, epidemiological methods can also be used to analyze disease patterns and identify prognostic factors. Epidemiological information is therefore important for planning and eval- uation of preventive health care and to guide management of patients in whom disease has already developed. Moreover, epidemiologic research methodology is used increasingly in a clinical setting to make predictions about health outcomes based on results from studies of groups of similar patients. Clinical epidemiology plays a central role in evidence-based medicine; a basic understanding of epidemiologic research methodology is therefore required to appreciate the strengths and weaknesses of the different studies. The basic epidemiological terminology is explained in Table 5.1.

5.2 General Aspects of Epidemiology

WHO defines epidemiology as “...the study of the distribution and determinants of health-related states or events (including disease), and the application of this study to the control of diseases and other health problems” [184]. Epidemiologic investigations are traditionally divided in descriptive epidemiology and analytic epidemiology according to the purpose of the research; see Box 5.1 for a fuller explanation.

Traditionally, dentistry focuses on treatment of teeth and dentists may consider results from epidemiological studies less relevant to their daily dental practice. However, proper answers to many important questions depend on epidemiologic data in addition to information from clinical research. Different designs are used in epidemiological studies reflecting the type of question being considered. Box 5.2 gives a brief description of the most common designs used in of epidemiology.

Box 5.1

WHO defines epidemiology as follows: "Epidemiology is the study of the distri- bution and determinants of health-related states or events (including disease), and the application of this study to the control of diseases and other health problems. Various methods can be used to carry out epidemiological investigations: surveillance and descriptive studies can be used to study distribution; analytical studies are used to study determinants" [184].

Traditionally, epidemiology is divided in descriptive epidemiology ad analytic epide- miology: Descriptive epidemiology con- cerns the distribution of health-related states or events, the main aspects being the person (who?), the place (where?), and the time (when?), whereas analytic epidemiology focus on determinants, which are causes and other conditions influencing the occurrence of the health-related states and events (why, how?). In both cases health-related states or events must be determined (what?) and properly defined and these events are used to identify the outcome of the study.

5.2.1 Prevalence

Questions like “What is the frequency of apical periodontitis and how does it depend on age, gender and other person-specific characteristics” and “Which tooth-specific factors and conditions are associated with presence of apical periodontitis?” relate to the prevalence of apical periodontitis in a population and the factors associated with presence of disease. These questions are rel- evant for health care planning and resource allocation and they are usually addressed by cross-sectional studies that include a sample of individuals who participate in a single examination. This gives a snapshot of the distribution of the disease in the population. The outcome is typically absence or presence of apical periodontitis, defined, either by a binary variable or by a scoring system with several ordered categories reflecting an increasing severity of the disease.

Box 5.2 Epidemiological designs

Randomized controlled trial (RCT): A trial, typically conducted to compare the effect of two treatments, in which the patients are allocated at random to one of the treatment groups.

Cohort study: A group of persons/patients are followed forward in time from a specific time point. The starting point may be defined as a given date, or by some event, e.g. a treatment. In population-based cohort study, participants are selected from the general population. In a patient-based cohort study, participants are patients from one, or several, clinics. In a historical cohort study, sometimes denoted a retro- spective cohort study, participants are identified in the past and followed to the present date.

Case-control study: The past exposure history of a group of cases, i.e. diseased persons or patients with an unfavorable outcome, is compared with the exposure history of a control group selected from the source population (the population from which the cases are identified). Many different types of case-control designs have been developed. Case-control studies are sometimes called retrospective studies; they rarely used in endodontic research.

Cross-sectional study: In a cross-sectional study, information about a group of individuals is collected at a given point in time. This design allows estimation of disease prevalence and of the distribution of a condition in the population from which the sample is selected. Causal relations cannot be assessed in a cross-sectional study.

Indices often used to classify sound and diseased periapical bone have their origin in a study from 1956 by Strindberg [169], who conducted a comprehensive longitudinal study on factors related to the results of pulp therapy. The width of the periodontal ligament, the integrity of the lamina dura and the presence of periapical radiolucency were proposed as the main indicators of periapical disease. Brynolf in 1967 investigated to what extent histologic changes were reflected in radiographs by comparing histologic and radiographic appearances of periapical changes in human autopsy material [21]. The study very thoroughly described the characteristics defining the different levels of apical inflammation. The Brynolf study formed the basis for development of a five grade index for registration of periapical inflammation, the Periapical Index (PAI) [128].

To produce valid information the distribution of age, gender, and other relevant factors in the study sample should mirror the distri- bution in the population. A random sample or a stratified random sample is usually the best way to meet this requirement. Other sampling strategies may result in study data that are subject to selection bias and the results of the study would then be of limited use to society or health care planners. A comparison of two studies from Denmark in 1997-98 highlights the problem [87, 88]; see Box 5.3 for details.

The endodontic literature includes a large number of reports from cross-sectional studies, many of which are based on samples of patients from dental schools or specialist clinic. Table 5.2 gives summary information about cross-sectional studies in endodontics. Secular trends in proportion of individuals with apical periodontitis, in proportion of teeth with apical periodontitis, and in proportion of root-filled teeth are shown in Figure 5.1.

Box 5.3

Two cross-sectional studies with participants living in the municipality of Aarhus, Denmark were conducted in 1997-98. One study was based on a random sample of persons born between 1935 and 1975. In this sample the average age was 42.3 years, 42% had at least one tooth with apical periodontitis, and 52% had at least one root filling. The other study was based on a sample of patients enrolled at the Dental School at Aarhus University in 1997-98; the study reported on the endodontic status of the patients prior to the treatment they received at the Dental School. The patient sample was slightly older, the average age being 45.8 years, but here 90% had at least one tooth with apical periodontitis and 99% had at least one root filling. Patients seeking treatment at the dental school obviously differ from the general population with respect to endodontic status and treatment needs [87, 88].

5.2.2 Incidence

The question “Which factors and conditions influence the incidence of apical periodontitis?” concerns the development of disease in healthy teeth. A cross-sectional study can not answer this question because it does not include a time dimension and clinical research focuses on treatment rather than prevention. However, development of appropriate preventive measures hinges on an understanding of the factors and conditions that influence the incidence of apical periodontitis. This knowl edge is only available from population-based cohort studies. Population-based cohort studies are observational studies that allow an investigation of the natural development of the disease and its treatment in the general population. A well-defined group of persons and/or teeth is followed forward in time with periodically examinations of the disease status. Both healthy and diseased persons/teeth are included and followed, and an analysis of the data may identify factors associated with changes in disease status from one examination to the next. The times of examination may not coincide with times of treatment, therefore the exact preoperative status of the tooth will not be known to the researchers, but only be estimated from the information available from a prior examination. Further- more, information on intra-operative factors, e.g. type of irrigant from clinical records, is seldom available. The population-based cohort studies usually rely on radiographic assessment of disease status and the quality of the root filling.

Table 5.2 Summary information about cross-sectional studies with presence of apical periodontitis as outcome.

Author	Country	Year	Population	Age	Method	Number of persons	Persons with AP (%)	Number of teeth	Average number of teeth	Teeth with AP (%)	Teeth with RF (%)	RF teeth with AP (%)
Bergenholtz et al. (1973) [15]	Sweden	1973	DSP	20-70+	F+P	240	57	5472	22.8	6	12.5	31
Kerekes and Bervell (1976) [83]	Norway	1976	DSP	19-81	F	200	34.5	4832	24.2	2.8	5.7	25.4
Allard and Palmqvist (1986) [5]	Sweden	1986	GP	>65	F	183	72	2567	14	9.8	17.6	27
Petersson et al. (1986) [141]	Sweden	1986	GP	-	I	861	-	4985	-	6.6	13.3	33.8
Bergstrom et al. (1987) [16]	Sweden	1987	GPa	21-60	F	250	46.8	6600	26.4	3.5	6.5	28.8
Eckerbom et al. (1987) [36]	Sweden	1987	PP	20-60+	F	200	63	4889	24.4	4.6	13	26.4
Eriksen et al. (1988) [43]	Norway	1988	GP	35	P+I	141	29.8	3917	27.8	1.4	3.4	25.6
Petersson et al. (1989) [140]	Sweden	1989	PP	-	F	567	76.5	11497	20.3	8.7	22.2	26.5
Odesjo et al. (1990) [123]	Sweden	1990	GP	20-80+	F	967	33.2	17430	18.2	2.9	8.6	24.5
Eriksen and Bjertness (1991) [42]	Norway	1991	GP	50	P+I	119	-	2940	24.7	3.5	6	36.6
Imfeld (1991) [75]	Switzerland	1991	GP	66	P+I	143	-	2004	14	8	20.3	31
De Cleen et al. (1993) [32]	Netherlands	1993	PP	20-59+	P	184	44.6	4196	-	6	2.3	39.2
Buckley and Spângberg (1995) [22]	USA	1995	DSP	-	F	208	-	5272	25.3	4.1	5.5	31.3
Eriksen et al. (1995) [41]	Norway	1995	GP	35	P	118	14.4	3282	27.8	0.5	1.3	38.1
Soikkonen(1995) [168]	Finland	1995	GP	76-86	P+I	169	41.4	2355	13.9	6.6	21.5	16.8
Saunders et al. (1997) [154]	England	1997	DSP	20-59+	F	340	67.7	8420	-	4.6	5.6	58.1
Weiger et al. (1997) [182]	Germany	1997	PP	12-89	P+I	323	-	7897	24.4	3.1	2.7	60.9
Marques et al. (1998) [107]	Portugal	1998	GP	30-39	P+I	179	26	4446	-	2	1.6	21.7
Sidaravicius et al. (1999) [161]	Lithuania	1999	GP	25-44	P+I	147	70	3892	26.5	7.2	15.1	35
De Moor et al. (2000) [33]	Belgium	2000	DSP	18-59+	P	206	63.1	4617	-	6.6	6.8	40.4

(Continued)

Table 5.2 (Continued)

Author	Country		Year		Population			Age	Method		Number of persons		Persons with AP (%)		Number of teeth	Average number of teeth		Teeth with AP (%)		Teeth with RF (%)		RF teeth with AP (%)
Kirkevang et al. (2001) [88]	Denmark		2001		GP			20-60+	F		614		42.3		15984	26		3.4		4.8		52.3
Boucher et al. (2002) [20]	France		2002		DSP			18-70+	F		208		-		5373	-		7.4		19.1		29.6
Lupi'Pegurier et al. (2002) [105]	France		2002		DSP			>20	P		344		-		7561	22		7.3		18.9		31.5
Ridell et al. (2006) [148]	Sweden		2006		GP			19	-		-		-		-	-		-		-		-
Dugas et al. (2003) [35]	Canada		2003		DSP			25-40	F+P+I		610		-		16148	26.5		3.1		2.5		45.4
JiméneZ'Pinzon et al. (2004) [78]	Spain		2004		DSP			-	F		180		61.1		4453	-		4.2		2.1		64.5
Georgopoulou et al. (2005) [56]	Greece		2005		PP			16-77	F		320		85.9		7664	24		13.6		8.9		60
Kabak and Abbott (2005) [80]	Belarus		2005		DSP			15-65+	P		1423		80		31212	-		11.7		20.3		45.2
Loftus et al. (2005) [103]	Ireland		2005		DSP			16-75+	P		302		33.1		7424	24.6		2		2		25
Tsuneishi et al. (2005) [177]	Japan		2005		DSP			20-89	F		672		69.8		16232	-		9.4		20.5		40
Peciuliene et al. (2006) [133]	Lithuania		2006		DSP			-	F		83		-		2186	26.3		-		12.9		43.1
Skudutyte-Rysstad et al. (2006) [165]	Norway		2006		GP			35	P		146		16		3971	27.2		1.1		1.5		42.6
Chen et al. (2007) [25]	USA		2007		GP			-	P		244		-		3533	14.5		5.1		4.8		35.5
Sunay et al. (2007) [171]	Turkey		2007		DSP			-	P		375		-		8863	-		4.2		5.3		51.1
Touré et al. (2008) [175]	Senegal		2008		DSP			-	F		208		-		6234	30		4.7		2.6		-
Gulsahi et al. (2008) [60]	Turkey		2008		DSP			-	P		1000		23.8		24344	-		1.4		3.3		18.2
Hollanda et al. (2008) [68]	Brazil		2008		PP			48b	P		1401		-		29467	-		-		21.4		-
Da Silva et al. (2009) [30]	Australia		2009		DSP			-	P		243		-		5647	23.24		-		8.8		21.6
Al'Omari et al. (2011) [3]	Jordan		2011		DSP			16-59	P		294		83.7		7390	-		11.6		5.7		71.9
Matijevic et al. (2011) [109]	Croatia		2011		PP			-	P		1462		76.9		38440	-		8.5		8.5		54
Peters et al. (2011) [134]		Netherlands		2011		PP	18-59+			P	178	36.5		4594			25.8		2.6		4.9		24.1
Gumru et al. (2011) [61]		Turkey		2011		DSP	19			P	1077	-		28974			26.9		2.2		-		42
LopeZ'Lopez et al. (2012) [104]		Spain		2012		DSP	19-70			P	397	34		9390			-		2.8		6.4		23.8
Kalender et al. (2012) [81]		Brazil		2012		DSP	18-50			P+I	1006	68		24730			-		7		8.9		62
Huumonen et al. (2012) [73]		Finland		2012		GP	30-95			P	5244	-		120250			22.9		-		7		-
Jersa and Kundzina (2013) [77]		Latvia		2013		PP	35-44			P	312	72		7065			-		7		18		30.5
Hebling et al. (2014) [67]		Brazil		2014		GP	60-94			F	98	42.9		942			11.6		12.1		13.4		65.1
Berlinck et al. (2015) [18]		Brazil		2015		DSP	0-60+			F	1126	-		25292			-		7.9		6.9		16.7
Oginni et al. (2015) [124]		Nigeria		2015		DSP	-			F	756	67.2		21468			-		-		12.2		41
Huumonen et al. (2017) [72]		Finland		2017		GP	30-95			P	5335	27		120635			-		-		-		39
Kielbassa et al. (2017) [85]		Austria		2017		PP	18-70			P	1000	60.5		22586			-		6.4		11		42.6

Abbreviations: AP: Apical periodontitis; RF: Root filling; DSP: Dental school patients; GP: General population; PP: Patient population; P: Panoramic images; I: Intraoral

radiographs; F: Full mouth survey

Footnotes:

^a Patients from general practice ^b Mean age

Figure 5.1 (a) Proportion of individuals with one or more teeth with apical periodontitis. (b) Proportion of teeth with apical periodontitis. (c) Proportion of root-filled teeth with apical periodontitis. The size of the bubbles reflects the size of the study.

The population-based cohort studies require extended follow-up and are often costly and time-consuming. Moreover, the success of cohort study depends on the cohort members' willingness to continue participation in the examinations. Relatively few population-based cohort studies can be found in the endodontic literature (Table 5.3).

5.2.3 Prognosis

Other important questions relate to the prognosis of the treatment, e.g. “What is the probability that the disease will respond to a specific treatment?” or “Which factors and conditions influence the prognosis of the treatment?” Here, our knowledge derives primarily from clinical studies. These are typically series of cases, all of which have received the same treatment. Some studies compare different treatments, with or without random allocation to the treatment groups.

The predominant clinical study in endodontics describe treatment of patients and/or teeth who are followed from treatment and forward in time. The outcome is observed after a pre-specified period or at a common closing date, and such a clinical study may therefore be viewed as a patient-based cohort study. The patient-based cohort study is often conducted by specialists at a single clinic according to a detailed protocol, there- fore the results may not translate directly to other settings or to general practice. In other words, the external validity may be low.

The clinical setting ensures that patient records are available from which relevant clinical information can be accessed, but patient-based cohort studies are also costly and time-consuming. Sometimes a patient- based cohort study relies on information about previously performed treatments retrieved from patient files. A new examination may then provide outcome information and the study can be conducted quickly, because there is no waiting time for follow-up information. Information from historic records may, however, be hampered by lack of standardization and frequent occurrence of missing data, which may affect the quality of the study. These studies are usually denoted retrospective cohort studies or perhaps more precisely historically prospective cohort studies; such studies are frequent in the endodontic literature. Table 5.4 gives a summary of the findings from patient-based cohort studies.

5.2.4 Prevalence Versus Incidence

The description above shows that the choice of study design and study setting depends on the questions being addressed. Moreover, the distinction between prevalence and incidence must be recognized. Prevalence describes the status of the disease in the members of a population at a given time and gives therefore a static picture of the distri- bution of the disease. Incidence, on the other hand, is related to the dynamics of the disease; in the present context, it describes the occurrence of apical periodontitis in the population over time and is usually quantified as a rate of occurrence or as a proportion of new cases during a specific time interval. Healing and extractions describe other aspects of the disease dynamics. It is also important to distinguish between prevalence and incidence when discussing conditions associated with the apical periodontitis. A high disease prevalence in a certain group could reflect an increased incidence, but delayed healing or reluctance to seek treatment could also be possible explanations.

Table 5.3 Summary information about population-based cohort studies.

Study	Exam year	Country	Number of persons	Follow-up months	Persons with AP	Persons with AP(%)	Persons with RF	Persons with RF(%)	Number of teeth	Teeth per person	Teeth with AP	Teeth with AP(%)	Teeth with RF	Teeth with RF(%)	RF teeth with AP	RF teeth with AP(%)
Petersson et al.	1974	Sweden	351	-	-	25	-	46	2100	6	107	5.1	258	12.3	82	31.8
1991/1993a	1985	Sweden	345	120	-	29	-	52	1962	5.7	121	5.9	323	16.5	94	29.1
[137, 139]
Frisk et al.	1968	Sweden	1220	-	-	41.9	-	18.3	24156	19.8	-	0.7b	-	3.4	-	-
2005 [53]
	1980	Sweden	1023	144	-	41.9	-	22.1	20255	19.8	-	0.6b	-	4	-	-
	1992	Sweden	867	288	-	31.1++	-	21.9	17253	19.9	-	0.5b	-	3.8	-	-
Eckerbom et al.	1975	Sweden	200	-	126	63	-	83.5	4889	24.4	255	5.2	636	13	168	26.4
2007 [38]c
	1980	Sweden	(200) 115	60	44	61.7	98	85.2	2825	24.6	97	3.3	393	13.9	68	17.3
	2002	Sweden	115	240	42	63.2	100	87	2461	21.4	168	6.8	598	24.3	93	21.4
Kirkevang	1997	Denmark	616	-	259	42	319	51.8	16016	26.0	534	3.3	776	4.8	402	51.8
et al. 2012 [92]
	2003	Denmark	473	60	236	49.9	279	58.4	12345	26.1	461	3.7	705	5.7	311	44.4
	2008	Denmark	360	120	189	52.5	214	59.4	9360	26.0	395	4.2	543	5.8	233	42.9

Abbreviations: AP: Apical Periodontitis; RF: Root-filled Footnotes:

^a Molars and premolars ^b Mean

^c Initially patients from a radiographic clinic

Table 5.4 Summary information about patient-based cohort studies with healing of apical periodontitis as outcome.

(Continued)

									Healed(%)
Study	Year	Country	Design	No of patients	No of teeth	No of roots	Follow-up months	AP(%)	No AP	AP	Overall
Strindberg [169]	1956	Sweden	ClinCoh	254	529	-	6-120	42	93	88	90
Seltzer et al. [158]	1963	USA	ClinCoh	2784	2921	-	6	-	94	76	84
Engstrom [40]	1964		ClinCoh	-	306	-	48-60	53	88	76	82
Harty et al. [65]	1970	UK	HistProsp	-	1139	1139	24+/6-24	-	-	-	90
Jokinen et al. [79]	1978	Finland	HistProsp	1199	1782	2459	24-84	33	61	38	53
Kerekes et al. [82]	1979	Norway	HistProsp	-	-	501	36-60	34	92	89	90
Barbakow et al. [10]	1980	S. Africa	HistProsp	-	332	-	12+	-	-	-	87
Oliet [125]	1983	USA	ClinCoh	-	153	-	18	-	-	-	89a
Swartz et al. [172]	1983	USA	HistProsp	-	1007	1770	12+	-	-	-	88a
Bystrom et al. [24]	1987	Sweden	HistProsp	-	-	79	24-60	100	-	85	84
0rstavik et al. [129]	1987	Norway	ClinCoh	-	-	546	12	29	-	-	82
Eriksen et al. [44]	1988	Norway	ClinCoh	-	-	121	36	100	-	82	90
Sjogren et al. [164]	1990	Sweden	HistProsp	356	-	849	96-120	24	96	86	53
Smith et al. [167]	1993	UK	HistProsp	-	821	-	60+	-	-	-	90
0rstavik [126]	1996	Norway	HistProsp	-	-	599	48	-	94	75	87
Sjogren et al. [163]	1997	Sweden	ClinCoh	53	-	-	60+	100	-	-	89a
Trope et al. [176]	1999	USA	RCT	102	-	-	12	-	-	-	88a
Weiger et al. [183]	2000	Germany	RCT	67	-	-	12-60	100	-	-	74
Benenati et al. [13]	2002	USA	HistProsp	-	894	-	6-72	-	-	-	62 (91a)
Hoskinson et al. [69]	2002	UK	HistProsp	167	200	489	48-60	70	88	74	77
Peters et al. [135]	2002	Netherlands	RCT	38	-	-	12-54	100	-	-	76
Huumonen et al. [70]	2003	Finland	RCT	156	-	-	12	100	-	-	76

Table 5.4 (Continued)

Study	Year	Country	Design	No of patients	No of teeth	No of roots	Follow-up months	AP(%)	No AP	Healed(%) AP Overall
Peters et al. [136]	2004	Switzerland	HistProsp	179	233	-	12-36	44	95	76	87
Marending et al. [106]	2005	Switzerland	ClinCoh	66	-	-	30	52	-	-	88
Negishi et al. [115]	2005	Japan	HistProsp	57	114	-	12+	-	-	-	85
Marquis et al. [108]	2006	Canada	HistProsp	325	373	-	48-72	57	93	80	85
Doyle et al. [34]	2007	USA	HistProsp	-	196	-	12+	65	87	75	82
Gilbert et al. [58]	2010	USA	PracBased	-	115	-	14-343	49	81	60	71
Riccucci et al. [147]	2011	Italy	ClinCoh	470	816	1369	60	53	92	83	89
Ng et al.	2011	UK	ClinCoh	534	702	1170	24-48	66	-	-	87
Ng et al. [118]	2011	UK	ClinCoh	-	1617	-	24-48	-	-	-	95b
Setzer et al. [159]	2011	USA	HistProsp	42	50	-	48+	36	-	-	52 (96b)
Bernstein et al. [19]	2012	USA	ClinCoh	1312	-	-	36-60	-	-	-	89b
Arya et al. [7]	2017	-	ClinCoh	46	200	-	12	100	-	-	62
Barborka et al. [11]	2017	USA	HistProsp	-	100	-	60-72	-	-	-	72
Pirani et al. [143]	2017	Canada	HistProsp	94	193(213)	-	60	-	-	-	85 (88b)
He et al. [66]	2017	USA	ClinCoh	54	54	-	12	83	-	-	69

Abbreviations: ClinCoh: Clinical cohort; HistProsp: Historical prospective; RCT: Randomized controlled trial; PracBased: Practice-based; AP: Apical Periodontitis Footnotes: ^a Healing; ^b Survival

Figure 5.2 Length bias in cross-sectional samples. The horizontal lines represent time periods from treatment to healing for a number of teeth. Two equally frequent types of recovery are shown: one group of teeth has a slow recovery, the other heals quickly. The vertical line represents the sampling time of a cross-sectional study. The horizontal lines for teeth included in the cross-sectional sample are red. Here, teeth with slow recovery dominate.

Conditions associated with disease prevalence may therefore not necessarily be potential targets for preventive efforts focused on reducing the disease incidence.

Furthermore, it is well known that cross-sectional sampling results in a length-biased sample, since individuals with protracted disease durations will be over-represented with this sampling strategy. Figure 5.2 illus- trates this phenomenon. The horizontal lines represent the time period from treatment to healing for a number of teeth. Two types of recovery are shown, one group of teeth has a slow recovery; the other heals quickly. The two types have identical incidence, but in a cross-sectional sample, in the figure represented by the vertical line, teeth with slow recovery dominate. Extrusion of root filling material in the periapical tissues may prolong the healing period, so this group of teeth will be over-represented in a cross-sectional sample of root-filled teeth with apical periodontitis, and the effect of a root filling surplus would be overestimated.

Systematic reviews and meta-analyses are often used to gain more information on a specific research question, e.g. the influence of age on root canal treatment outcome [160]. If the studies included in such review and/or meta-analyses are homogeneous, the accumulation of cases may improve the certainty of an estimate. However, often the studies differ so much that valid conclusions are not easily drawn [119]. Figure 5.3 gives an overview of how different study and evidence types compare overall as regards the strength of evidence they provide.

5.3 Elements of an Epidemiologic Study

Endodontic studies of patients/teeth usually focus on treatment or prevention of apical periodontitis, and a typical outcome there- fore describes the absence, reduction or elimination of a periapical lesion as observed on a periapical radiograph. Index systems or scales with ordered categories have devel- oped to describe the extent and severity of the disease [113, 128, 149], but in published research the treatment outcome is often reported as success or failure defined by a suitable cut-point on the outcome scale. A valid comparison of success probabilities from different studies requires that the same outcome scale with the same cut-point is used to define a success and that the outcome is assessed from the same type of radiographs. Moreover, the timing of the treatment outcome is important. The length of the follow- up period should reflect the chosen outcome; for healing from one to several years might be appropriate, for tooth survival a longer period may be relevant, perhaps 5-10 years. Further confusion arises when the success-failure terminology is used when reporting results from a cross-sectional study. To conclude that the treatment is successful an improvement of the disease status must be observed and repeated follow-up measurements of the case, patient or tooth are therefore required.

Figure 5.3 Evidence pyramid showing the relative strength of evidence for different study designs and sources of information.

Even when the outcome is assessed on a well-defined scale, by trained and calibrated observers, misclassifications or measurement errors will occur (Figure 5.4). The reliability and validity of the scoring system is there- fore also important. Reliability concerns the reproducibility of measurement: Will a trained observer always give the same score, if an image is assessed at several occasions? How often will trained observers disagree?

Figure 5.4 Accuracy (validity) and precision (reliability).

Validity measures the trueness of the meas- urement. In a clinical setting, the true value is unknown, but in vitro experiments may allow histological evaluation of the results from the different imaging techniques and thereby provide information on the validity of the method of measurement. Unfortunately, the terminology used to describe properties of a measurement method is not consistent; several names are in use for the same property and a given name may have different definitions. Recently, efforts to establish an international consensus on terminology and definitions of measurement properties have led to the COSMIN checklist [110, 111] Table 5.5 reviews this proposal and other commonly used terms.

Table 5.5 Common terminology for measurement properties.

Measurement error and misclassification
Validity	The COSMIN panel [110] defines validity as “the degree to which an instrument truly measures the construct(s) it purports to measure”. Accuracy is another commonly used name for this property and in statistics the name unbiasedness is used.
Reliability	The COSMIN panel [110] defines reliability as “the degree to which the measurement is free from measurement error” Other commonly used names are repeatability or reproducibility. In statistics reliability is often denoted precision.
Terminology related to measurement of disease status as present or absent
Sensitivity	The probability of correctly classifying a diseased individual
Specificity	The probability of correctly classifying a healthy individual
Positive predictive value	The probability of an individual classified as diseased is classified correctly.
Negative predictive value	The probability of an individual classified as healthy is classified correctly.
ROC curve	The receiver operating characteristic (ROC) curve is a plot of the sensitivity against 1 - specificity as the diagnostic threshold varies.

When the relevant outcome is identified, the next issue becomes how to obtain or measure the outcome and the determinants that you want to include in the study. Again, the answer depends on the aims of the study. If the aim of the study is to give an overall description of the dental health in a large population, panoramic images may be convenient and provide sufficient information to register an outcome such as number of root- filled teeth. However, periapical radiographs are necessary, if the study focuses on the healing of apical periodontitis using the periapical index as the outcome and with quality of the root filling as a possible determinant. Moreover, if the patient’s symptoms or satis- faction is used as treatment outcome, radiographic information may not be needed at all.

5.4 Evaluation of Epidemiologic Data

5.4.1 Data Structure

Statistical evaluation is essential in epidemiol- ogy. Standard statistical methodology assumes independence between different observations, but in dental research so-called multi-level data are frequent and special considerations are therefore often needed. The term multi- level reflects that information describing the patient, e.g. age, gender, and smoking habits, is shared by all the patient’s teeth; whereas information related to the tooth, e.g. tooth type, presence of coronal restoration, may differ among the patient’s teeth.

In endodontic research information is collected from persons, teeth, roots, and occasionally also from root canals, and in the data analysis, the person or the tooth is usually chosen as the unit of analysis. When outcome is measured at the level of individual persons, an assumption of independence is usually appropriate. When the tooth is the unit of analysis, a positive association between outcomes is usually expected for outcomes obtained from teeth from the same person; whereas teeth from different persons are expected to be independent. Part of a dependence may be explained by known factors and conditions and if data are available, the statistical analysis can adjust for their effect. If a positive dependence between outcomes is ignored in the statistical analysis, the uncertainty in the data will be underestimated, confidence intervals become too narrow, and statistical tests will give too many rejections. In analyses where the tooth is the unit of analysis, between-persons comparisons, e.g. smokers versus non-smokers, are more vulnerable to such bias, whereas within- person comparisons, e.g. teeth with coronal restorations versus teeth without coronal restorations are less problematic. Strindberg (1956) [169] in a statistical appendix examines the interdependence between treatment outcomes for two roots of the same tooth and between two teeth in the same mouth. The analysis of his data indicates that both associations are present, but none of them is particularly strong. The impact of the association between treatment outcomes for teeth in the same mouth is illustrated in the example shown in Box 5.4 Data on the determinants, the potential risk factors or prognostic factors, are also multilevelled. When the unit of analysis is the tooth and the outcome therefore is measured on tooth level, information on root canal and root level is usually aggregated to the tooth level; for multi-rooted teeth, the quality of a root filling is not presented as a score for each root, but aggregated to a single score. Similarly, when the unit of analysis is the person, tooth level risk factors are usually aggregated to per- son level variables, e.g. as the number of root- filled teeth or the number of de crown.

Box 5.4

In a Danish population-based cohort study, the incidence of apical periodontitis over a five year period among smokers and non-smokers was compared. A logistic regression with tooth as the unit of analysis gave an odds ratio of 1.77 for smokers relative to non-smokers. Ignoring the association between outcomes in teeth from the same person, the 95% confidence limits for the odds ratio were 1.40 and 2.23, whereas an analysis that accounted for this association gave a slightly wider confidence interval with 1.32 and 2.35 as confidence limits [91].

5.4.2 Adjustment for Confounding in Observational Studies

An investigator planning an experiment may ensure that the experiment is balanced in the sense that the experimental units allocated to the different treatment groups are similar. In epidemiology, most studies are observational; the investigator is an observer and not able to influence the allocation of treatments or interventions.

Also epidemiologic studies in endodontics are usually observational studies; to estimate the influence on the outcome of a specific factor it is therefore necessary to adjust for the effects of confounding factors, i.e. factors related to both the outcome and the factor in question. Separate analyses of each factor with no adjustment or analyses with incomplete adjustment may result in biased estimates of the association between the risk factor and the outcome. To identify the influence of a specific factor it is there fore important to correct for all known con- founding factors in the analysis. Moreover, valid comparisons of estimates from studies with different degree of adjustment may not be possible. An example that illustrates the importance of these points is presented in Box 5.5

5.4.3 Regression Models

Adjustment for confounding factors is usually obtained by some sort of regression anal- ysis. The choice of regression model depends on the outcome selected; see Box 5.6 Modern statistical software packages include modules that will analyze data with any of these regression models (SAS, Stata, R etc.). Note, however, that regression analyses with adjustment for confounding factors and prediction models with many predictors may require large data sets to produce reliable results. A rule of thumb, derived from com- puter simulations, indicates that for a logistic regression the number of records should be at least 15 events per parameter fitted in the model. Similar rules of thumb exist for other are here reduced considerably and the odds ratio for adequate coronal fillings is no longer significantly different from 1. The estimates becomes slightly larger in the third analysis, which also includes adjustment for presence of a root filling. In the final analysis adjustment for presence of crowns and of carious lesions are also included and this has a large impact on the size of the estimate. This change reflects that presence of a crown increases the risk of developing apical periodontitis. In the previous analyses, these teeth were categorized as teeth without a coronal filling leading to an underestimation of the risk associated with coronal fillings. This example illustrates the importance of including corrections for all relevant factors in the analysis and that useful comparisons of estimates from studies with different degree of adjustment may not be possible. Data from [91].

Box 5.5

Comparison of different adjustments for confounding factors.

Box 5.6 Types of outcome and the corresponding regression models
Outcome	Regression method
Two categories (dichotomy) Several categories Several ordered categories Counts Quantitative, normal error	Logistic regression Multinomial logistic regression Ordinal logistic regression Poisson regression, Negative binomial regression Multiple regression

The figure shows the effect of adjustment for confounding factors on the association between the quality of a coronal filling and incidence of apical periodontitis. The association is described by an odds ratio and estimated in a logistic regression using the data on incidence of apical periodontitis in a five year period. The diamonds represents odds ratio estimates for apical periodontitis for teeth with an adequate coronal filling (top) and teeth with an inadequate coronal filling (bottom) relative to teeth with no coronal filling; 95% confidence intervals are shown as bars together with the estimates.

Results are shown from four different analyses with varying degree of adjustment for other risk factors. The first analysis is a simple unad justed analysis, and the second analysis includes adjustment for person-specific factors (gender, age, smoking habit), tooth group (front, premolar, molar), and jaw. The estimates regression models, see Harrell (2015) [64], chapter 4 for a fuller discussion of issues related regression analysis strategies.

Many statistical software packages include options that will allow the user to correct for correlation between outcomes from the same person. Different approaches have been developed, but a detailed description is beyond the scope of the present chapter. Kirkwood and Sterne (2003) [95] chapter 31 gives a brief overview of the available methodology.

In a cohort study of incidence of apical periodontitis with outcome for each tooth registered as absence or present of apical periodontitis a simpler approach may be useful. If interest primarily focuses on the role of tooth-specific risk factors, the impact of person-specific risk factors are disturbing the evaluation and corrections for their influence are therefore needed. An analysis based completely on comparisons of teeth within a person will automatically correct for all person-specific factors, both measured and unmeasured, and may therefore be a conven- ient choice. This approach is implemented in a conditional logistic regression analysis.

Information on treatment outcome in terms of overall success rates is useful as it provides information on what is achieved in different settings and different populations. However, to be able to estimate the prognosis for a specific treatment, the dentist also needs to take factors related to the patient, the tooth, and the treatment, into account. Usually, these factors are described as person or tooth-specific preoperative, intra-operative and postoperative factors.

5.5 Factors and Conditions Associated with Treatment Outcome

When presented with a tooth with apical periodontitis the dentist must decide if the tooth should be treated and if so, which treatment for perform. The prognosis of the different treatment choices depends on conditions of the specific tooth, and to some extent on factors related to the patient. Considering different treatment scenarios, the dentist must therefore assess how the different aspects of the particular treatment session affects the chance of a successful treatment. The current available prognostic information derives mainly from patient- based cohort studies in which the association between tooth- and patient-specific conditions and endodontic disease has been assessed. The different factors are interre- lated, so the apparent association between a condition and treatment outcome may actually rather reflect the confounding effects of other factors. Single-factor analyses may therefore not give valid prognostic information. Factors and conditions associated with the prognosis are often denoted prognostic factors or, if focus is on treatment failure, risk factors. The following is a review of prognostic value of a number of person-specific and tooth-specific factors related to prevention and treatment of apical periodontitis. Focus is on the outcome absence of apical periodontitis, but information related to tooth survival and other patient-related outcomes is also important.

5.5.1 Treatment Outcome

5.5.1.1 Periapical Disease

The focus of most studies in endodontology has been on presence of apical periodontitis assessed by conventional radiographs, some- times supplemented with information about clinical symptoms. Fewer studies have reported on outcomes related to tooth survival and patient satisfaction.

In reviews the calculated success rate of primary root canal treatment varies from around 70% to more than 90% [12, 98, 119, 181]. The main source of information derives from historically prospective cohort studies, as in the review by Ng. et al. (2007) [119]. Reviews are usually considered to generate results at the one of highest levels of evidence, but the validity of a review obviously depend on the quality of the included studies and the validity of results from retrospective studies may be questionable.

Figure 5.5 Endodontic treatments results from patient-based cohort studies. Proportion of successful treatments plotted against the year the study was conducted and stratified according to the type of the study.

In patient-based cohort studies, the success rate of primary root canal treatment has been found to be from 70 to more than 90% (Figure 5.5). The findings in these studies relate typically to treatment of teeth in highly controlled environments, from specialist clinics, dental schools done by dental students, post-graduate students or faculty teachers and represent the results that can be achieved when selected patients are treated under ideal conditions.

Most of the published outcome studies report on primary root canal treatment, or do not distinguish between primary and secondary treatments. In a review of second- ary root canal treatments a success rate of 78-82% was found [118].

In population-based cohort studies the estimated success rates vary extensively depending on the population studied. In the private sector in USA a success rate as high as 89% [185] has been reported. However, the frequency of apical periodontitis before treatment was only 13% in this population. In a Danish study the estimated success rate was 59%, but here the frequency of apical periodontitis at baseline much higher (42%) [92]. In a Swedish study 32% of the teeth had apical periodontitis at baseline. The success rate for the teeth with apical periodontitis was 33%, whereas 83% of teeth that initially had no apical periodontitis were also with out apical periodontitis 10 years afterwards [137]. A recent Swedish study from a public county clinic reported a baseline frequency of apical periodontitis of 61% and the success rate after 3 years of 58% [97]. Few population-based cohort studies allow estimation of retreatment outcomes. In a Danish study, 26% had healed within a 5-year period, but after 10 years, almost 50% had healed [92].

The findings from population-based cohort studies suggest that endodontic treatments performed in the general clinic have lower success rates than those reported in patient- based cohort studies of patients in special- ized clinics and institutions. Cross-sectional studies do not allow estimation of success rates, but the prevalence of apical periodontitis found in these studies reflects the overall treatment outcome and treatment strategy in the population (Table 5.2, Figures 5.1a-c)

5.5.1.2 Patient-Related Outcome

Reviews have reported prevalence of pain one-week postoperatively to be about 14% [132], and this has been confirmed in a recent study where 90% had no or little pain 7 days after endodontic treatment [114], with follow-up periods of at least 6 months the estimate of pain decrease to about 5% [122]. These estimates are supported by recent patient-based cohort studies, which also sug- gest that less than 5% of patients experience persistent pain 3-5 years after endodontic treatment [19, 179].

5.5.1.3 Retention of Root-filled Teeth

Teeth are being root canal treated to eradicate endodontic disease, and for patients to retain their teeth functional and without pain, and even though root-filled teeth are lost more often than non-root-filled teeth (Table 5.6), the retention rate of root-filled teeth in different populations is high.

A review of survival of root-filled teeth found survival rates of 86-93% in patient based cohort studies, with observation periods ranging from 2-10 years [117]. This has been confirmed in recent patient-based cohort studies where 10-year survival rates of 86% [23] and 81.5% [100] was reported, while insurance data show that around 90% of endodontically treated teeth were retained in the oral cavity after 5-10 years [52, 150, 151].

In population-based observational studies of general populations almost 90% of root- filled teeth were retained 10 years after endodontic treatment [93, 139], and in a study covering 20 years approximately 70% root-filled teeth were retained [38]. Tooth loss also depends on the study population, and in another Swedish population receiving high-cost dental care fewer root-filled teeth (65%) were retained after 20 years [138]. Overall, it seems evident that a high proportion of root-filled teeth are retained over periods of 10-20 years.

5.5.2 Person-specific Risk Factors

5.5.2.1 Gender

Most studies have found that the patient's gender has little or no influence the outcome of root canal treatment. Some demonstrate a slightly higher success rates for females - others the opposite effect [69, 91, 116, 118, 120, 147].

Table 5.6 Loss of non-root filled and root filled teeth after 5(A) and 10(B) years. Percentages and relative risk [92].

Panel A
Root filling status		Lost in 2003
in 1997*	Yes	No	Total	Percent	Relative risk
No	65	11754	11819	0.50%
Yes	42	579	621	6.80%	12.3
Total	107	12333	12440	0.90%
* 473 persons
Panel B
Root filling status		Lost in 2008
in 1997**	Yes	No	Total	Percent	Relative risk
No	122	8952	9074	1.30%
Yes	57	398	455	12.50%	9.3
Total	179	9350	9529	1.90%

** 360 persons

It seems prudent to conclude that endodontic outcomes are independent of the patient's gender.

5.5.2.2 Age

Cross-sectional studies show that the prevalence of root canal treated teeth as well as the prevalence of AP among elderly individuals is high. Furthermore elders have a high fre- quency of root canal treated teeth [63].

However, in population-based cohort studies the occurrence of apical periodontitis in root-filled teeth does not seem to be associated either with age or with study population [53, 91, 137]. The proportion of persons who have at least one tooth with a root filling has however been shown to increase with age in some studies [92, 137]. This may affect the overall prevalence of apical periodontitis, since root-filled teeth more often have apical periodontitis than non-root-filled teeth. If teeth are not extracted the prevalence of apical periodontitis will accumulate, mimicking an age dependence. Other studies demon- strate no effect of age or even a slight decrease in persons with apical periodontitis with age [37, 53]. A recent review on patient-based cohorts concludes that the patient's age is not a prognostic factor for non-surgical root canal treatment [160].

In conclusion, older people may be more likely than younger people to have teeth with apical periodontitis, in particular if few extractions are performed, but patient age should not be regarded as a prognostic factor for the outcome of root canal treatment.

5.5.2.3 Smoking

The impact of smoking on apical has previ- ously been investigated in cross-sectional studies and in cohort studies with conflicting findings. Five out of six cross-sectional studies found a significant positive association between presence of apical periodontitis and smoking; one population-based cohort study indicated an increased risk of receiving a root canal treatment, whereas one population-based and one patient-based cohort study found no statistically significant effect of smoking on development of apical periodontitis [180].

Smoking may delay healing, and smoking habits may be related to unfavorable lifestyle choices, but smoking should not be consid- ered a risk factor when deciding whether to perform root canal treatment or not.

5.5.2.4 Socio-economic Status

The influence of socio-economic status has not been investigated in relation to treatment outcome in patient-based cohort studies. In cross-sectional studies have shown that after adjustment for dental status the socio-economic status of the individual did not provide additional information on the periapical status [4, 54, 94].

In conclusion, the socio-economic status of a patient is not associated with presence of apical periodontitis and is not likely to influence the prognosis of a root canal treatment.

5.5.2.5 General Health

Most patient-based cohort studies exclude patients with general diseases. A robust assessment of the effect of a patient's general health on treatment outcome is therefore difficult to obtain. In the study by Strindberg (1956) [169] no effect of the patient's general health was demonstrated. A historic prospective patient-based cohort study from 2005 found that a compromised immune system was associated with poorer prognosis [106].

Studies have investigated the effect of diabetes mellitus on AP, but even though diabetes in some studies has been shown to be associated with a less favorable outcome after endodontic treatment the current evidence is inconclusive and insufficient to claim an association [6, 84].

The association between endodontic lesions and cardiovascular disease has been investigated and inconsistent results have been found. However, only four studies were cohort studies. Three of these studies reported that the presence of AP could affect the cardiovascular system [17].

The findings in cross-sectional studies regarding the association between general health and presence of apical periodontitis are inconsistent [4, 54].

It seems prudent to conclude that relevant information is lacking on the association between general health and endodontic treatment outcome.

5.5.2.6 General Dental Health

Poor dental health and extensive dental treatment experience have been associated with higher risk of a less favorable endodontic treatment outcome in both patient-based and population-based cohort studies [54, 91, 137, 169].

Findings from cross-sectional studies support this finding. Both the extent and quality of previous dental work and signs of present or previous dental disease has been found to be associated with presence of apical periodontitis [59, 131].

In conclusion, the extent and quality of previous dental treatments as well as signs of present or previous dental disease may pre- dict the success of future treatment. This may reflect both to patient habits and quality of dental work.

Overall, person-specific factors seem to be less important in predicting endodontic treatment outcome when adjusted for tooth- specific variables more closely related to treatment [91, 118].

5.5.3 Tooth-Specific Risk Factors

5.5.3.1 Preoperative Risk Factors Preoperative Diagnosis

The preoperative diagnosis (pulpitis, necrosis pulpae, apical periodontitis) is one of the preoperative factors that has been studied most frequently. The presence of a preoper- ative periapical lesion has been shown to be associated with unfavorable treatment outcome in patient-based cohort studies [19, 118, 147, 169], population-based cohort studies [91, 93, 139], and reviews.

In reviews it has been suggested that presence of a periapical lesion may reduce the chance of a successful outcome with 10-20% (Table 5.2) [12] compared to teeth without a preoperative periapical lesion. Moreover, some studies have demonstrated that the size of a periapical lesion size may influence the outcome with larger lesions having a negative effect on outcome [27, 118, 147, 169]. This may reflect that larger lesions may take longer to heal and therefore require longer follow-up time compared to smaller lesions. The importance of the length of follow-up was highlighted in a study by Weiger et al. 1998 [181] that concluded that calculation of success rates without taking the individual observation time into account would result in misleading outcome information. Still, the patient-based cohort study by Ng et al. 2011 [118] reported that larger lesions had a reduced success rate - even after adjusting the results for follow-up time after treatment. Recently, it was demonstrated that different grades of periapical inflammation, as described using the PAI scores, influenced the prognosis of the tooth significantly, the more severe the inflammation was the poorer was the tooth's prognosis [71, 89, 90].

The preoperative diagnosis of a tooth does affect the outcome. Teeth presenting with a preoperative periapical lesion have a poorer prognosis compared to teeth without a pre- operative periapical lesion.

Tooth Type

Most clinical and population-based cohort studies have found that molars have a higher risk of failure [10, 26, 55, 82, 91, 172]. In a review it was demonstrated that mandibular premolars had the lowest risk of failure [120].

Comparisons of results are complicated by differences in reporting; some studies report outcome for teeth other consider separate outcomes for each root, and this may at least in parts explain the differences seen between the findings. If tooth is the unit, multi-rooted teeth have a higher risk of apical periodontitis, let alone because of the number of roots. Assuming independence between treatment outcomes in roots from the same tooth, it follows that if one root canal has a probability of 80% success, then a tooth with three canals will have a probability of success (no apical periodontitis in any canal) of approximately 50%. Similarly, if the root-level probability is 70%, a tooth with three canal will have 30% probability of success. This crude estimation of course does not take root morphology and other relevant factors into account.

Endodontic treatment of multi-rooted teeth may have a lower success probability compared to single-rooted teeth.

Sinus Tract

Some studies have suggested that a sinus tract may facilitate bacteria from the oral cavity to colonize the periapex, and increase the risk for an extraradicular infection, but findings have been inconclusive. Two patient-based cohort studies found that a preoperative presence of a sinus tract did not affect the outcome [27, 147] whereas one study found that presence of sinus tract had a negative effect on the treatment outcome [118].

Little is known about the influence of a sinus tract on the outcome of endodontic treatment, but the studies available suggest that an association, if any, is negative.

Preoperative Pain

Prolonged preoperative pain may be a predictor for pain postoperatively together with negative patient expectations. It has been reported that 5-10% of patients receiving primary root canal treatment may experience postoperative pain even after 6 month [121, 122, 179]. Nevertheless, presence of preoperative pain has not been shown to be associated with an unfavorable treatment outcome of a root canal treatment. [24, 47, 127, 130, 147, 164].

Marginal Bone Level

Several studies have found an association between marginal periodontitis and presence of apical periodontitis, with periodontal involvement having a negative effect on treatment outcome [19, 84, 90, 145, 166]. A compromised periodontal support may be associated with reduced success probability of an endodontic treatment.

Root Filling

Reviews based on patient-based cohort studies have reported success rates for sec- ondary endodontic treatment of more than 70% [116, 178], and when stratified according to length of follow-up period 70% (2-4 years) and 83% (4-6 years) [174].

Very few population-based cohort studies have reported results related to outcome of retreatment. In a Danish study 26% of the re-treated teeth had healed within a 5-year period, and after 10 years almost 50% had healed [92], and Petersson et al. (1991) [139] found that 66% had healed after 10 years.

The presence of a previous root filling may not affect treatment outcome if the dentist is able to gain access to the periapical area and effectively clean the root canal. The findings from population-based studies may suggest that is not always possible in general practice and that a previous root filling therefore may be associated with increased risk of treatment failure.

5.5.3.2 Intra-operative Risk Factors

Treatment procedures and treatment deci- sions are actions, interventions, and choices taken by the dentist as the treatment pro- gresses, often in response to events that occur during cleaning, preparation, and obturation of the root canal. In case series and patient-based cohort studies it may be difficult, or even impossible, to separate the influence of the selected response from the influence of the triggering event on the outcome of the endodontic treatment. In epide- miology, the term confounding by indication is used to describe this situation (Table 5.1). A valid assessment of the treatment proce- dure requires the conduct of a randomized controlled trial, but these trials are unfortu- nately not common in endodontic research. The available evidence from the observational studies may occasionally lead to diverging interpretations.

Aseptic Procedures

Infection with microorganisms is the primary cause of endodontic disease, so minimizing or eliminating contamination of the working area when performing endodontic treatment is essential. Rubber dam has been available to the dental profession for more than a hundred years, and during this time, univer- sities and endodontic specialists have taught and recommended the use of it. Failure to use rubber dam has been shown to be associated with the use of less potent root canal irrigants, and moreover places the patient at risk of swallowing or aspirating materials and instruments [2, 45]. In a historic prospective population-based cohort study the use of rubber dam was shown to increase the survival rate of teeth receiving primary root canal treatment compared to teeth where rubber dam had not been used [102]. Furthermore, studies have shown that patients who had experienced the use of rubber dam during endodontic treatment would prefer rubber dam for eventual future endodontic treatments [2].

Aseptic procedures improve the chance of a successful outcome.

Choice of Instrumentation

The purpose of root canal treatments is to clean and prepare the root canal system as completely and efficiently as possible, with- out ledging, zipping, stripping, or straightening the canal. The mechanical preparation of the root canal is considered one of the most important stages in a root canal treatment, and the last decade has witnessed the intro- duction of many new instruments and techniques. Numerous in vitro studies have demonstrated that NiTi instruments better respect the original root canal anatomy and facilitate root canal preparation compared to the use of traditional stainless steel instruments, especially in curved root canals [74].

So far, relatively few studies have evaluated the effect of different instruments or techniques on treatment outcome in patients. An in vivo study showed that inexperienced operators have fewer procedural errors when using more contemporary techniques [142]. However, the available evidence does not per- mit any firm conclusions regarding differences in outcome of root canal treatment related to use different file materials and systems [8, 74, 86, 155].

A few community trials report that adaptation to newer instrumentation produces result of higher technical quality, assessed in radiographs, than traditional instrumentation procedures and that the improvement was maintained over time [31, 96, 112]. In one community trial, Kock et al. (2013) [96], the effect on outcome was assessed - the proportion of technical adequate root fillings increased, unfortunately without a concomitant effect on the periapical status.

Little is known about the overall effect of different instrumentation techniques on treatment outcome. Root canal preparation with NiTi instruments is easier to perform, and result in better maintenance of the orig- inal canal shape and a more predictable preparation of the root canal, but it has not yet been demonstrated that the use of these instruments has an important impact on the outcome of root canal therapy.

Size of Apical Preparation

Bacteria in the root canal are found in biofilm, on the root canal walls and they may have penetrated into the dentinal tubules where they are difficult to reach with disin- fective agents. To eradicate these bacteria it has been advocated that the infected layers of dentine in the apical part of the root canal should be mechanically removed. Several patient-based cohort studies have investigated the effect of the size of the apical preparation on endodontic treatment outcome, but the results are inconclusive. Most studies find no effect of increasing apical enlargement [69, 82, 118, 130, 135] and even a negative effect of excessive preparation of the apical dentine has been described [169].

The findings were all from patient-based cohort studies and confounding by indication cannot be excluded. Additional studies are needed to establish the optimal apical preparation size to be able to eradicate bacteria at the same time as saving root dentine.

Patency

The concept of retaining patency by letting a thin file bypass the apical foramen repeatedly during root canal preparation has been debated for decades. In some studies apical disturbance has been shown to affect the outcome negatively [1, 12, 14], whereas in other studies have suggested a positive effect of patency [65, 118]. In 1956 Strindberg [169] found more failures if the root canal was blocked and the apical part of the root canal inaccessible for instrumentation.

Also, the effect of patency has only been investigated in patient-based cohort studies without randomization and therefore the conflicting findings may reflect differences in treatment philosophy.

Complications During Treatment Different complications may emerge during treatment related to either obstruction of the root canal or perforation of dentine. Obstruction of the root canal may be a con- sequence of insufficient irrigation from packing of dentine chips in the apical part of the root canal, or caused by a fractured instrument. Both complications may result in impaired cleaning of part of the root canal and this may be problematic, especially in infected root canals. Patient-based cohort studies generally agree that canal obstruction is problematic [29, 115, 164, 169]; in particular, instrument breakage may affect the outcome negatively [76, 169]. Perforation through dentine during endodontic treatment has a negative effect on the treatment outcome [29, 76, 118, 164], in particular if a perforation is situated close to the osseous crest [48, 118, 144]. Moreover, a preoperative lesion at the site of the perforation may affect heling negatively [99, 144].

Treatment complications have a negative effect on treatment outcome.

Irrigation

Thorough irrigation during root canal preparation is considered essential, and many laboratory-based studies have compared different types and concentrations of irrigation solutions [74]. The SBU report (2010) [8] found insufficient scientific evidence to support the use of a specific endodontic irrigant to ensure a favorable treatment outcome. A Cochrane review concluded that, even though there was little evidence, sodium hypochlorite and chlorhexidine appeared to be more effective at reducing bacterial cultures compared to saline [49]. The two irrigants should, however, not be used together since a combination of chlorhexidine and sodium hypoclorite result in an precipitate containing para-chloro-aniline, which, besides being difficult to remove, is suspected of being cytotoxic and carcinogenic.

Based on clinical microbiological and ex vivo microbiological data the ESE guidelines for canal irrigation recommend a solution possessing both disinfectant and tissue-dissolving properties (ESE 2006) [45].

Antiseptic Procedures; Additional Disinfection/Medication

To further improve the cleansing and disinfection of infected root canals different pro- tocols using a variety of disinfective agents has been studied. These include antibiotic or steroid containing substances (e.g. MTAD, Ledermix®), 2% chlorhexidine or iodine potassium iodide, sonic or ultrasonic activation of the irrigant used and photodynamic disinfection using light or laser. It was found that teeth dressed with steroid had lower success rate than those dressed with antibiotics or antiseptics [120]. Some ofthese approaches have showed promising results in in vitro and ex vivo studies [162].

Scientific evidence is lacking to determine if additional antiseptic measures, medication or other, approaches have any effect on the outcome of root canal treatment [8, 162].

Single-visit - Multiple Visits

Randomized controlled trials have estab- lished that treatment outcome does not differ between single-visit and multi-visit treatments. Evidence is, however, lacking regarding dif- ferences in the prevalence of postoperative pain or flare-up following either single or multiple-visit root canal treatment [9, 12, 50, 152, 153, 170].

Root Filling Quality

Root canal treatment should include a com- plete obturation of the root canal space, as voids or gaps invites treatment failure due to infection from remaining or intruding bacteria.

Patient-based cohort studies and reviews have shown that the length of a root filling may affect the treatment outcome. An optimal length (0-2mm) is better than a to short or overextended root filling [12, 98, 118, 120, 147].

A possible interaction between root filling quality and infected/non-infected root canals has also been investigated. The results indicate that the length of the root filling was less important, if no preoperative lesion was detected [46, 57, 167].

If a preoperative lesion was present a suf- ficient length of a root filling resulted in significantly higher success rate compared to teeth where the root filling was either too long or too short. [156, 157, 164, 167].

When a root filling extends beyond the apical foramen, the extruded materials may have an irritating effect on the periapical tissues, this may result in longer healing time, suggesting that longer observation times may be necessary for a trustworthy assessment of treatment outcome [62, 120, 181]. Over-filling is often preceded by an over-instrumentation and pulp remnants and microorganisms may have been pushed beyond the apex, and this may also influence the treatment outcome [12].

The homogeneity of root fillings has been also been investigated, and the results indicate that root fillings without visible voids are associated with a better prognosis [101, 118, 120, 169]. Most of these studies are, however, retrospective patient-based cohort studies so findings should be interpreted with caution. The SBU 2010 [8] found lack of scientific support to determine the effect on root filling quality on outcome.

In cross-sectional studies absence of apical periodontitis is more frequent in teeth with adequate root canal treatment (length/lateral seal) and adequate restorative treatment [59]. In population-based cohort studies, however, the effect of root filling quality on incidence of apical periodontitis, is reduced when adjusted for baseline information of apical periodontitis suggesting the inadequate quality of a root filling could be associated with healing rather than development of apical periodontitis [90, 91].

An inferior quality of root fillings influence the outcome of root canal treatment nega- tively, in particular in teeth with preoperative infection.

5.5.3.3 Postoperative Risk Factors Coronal Restoration

The coronal restoration of a root-filled tooth serves as a barrier against reinfection of the root canal system and restores the functional status of the tooth.

Most studies have found that both the quality of the root filling and of the coronal restoration may affect the periapical status [59, 90, 91, 118, 120, 146, 173]. Chugal et al. 2007 [28], however, found no such effect and argued that lack of stratification on key con- founding factors has resulted in an overestimation of the contribution of an adequate coronal restoration to the success of endodontic treatment.

The quality of the coronal restoration has been found to influence the treatment outcome, with inferior quality being linked to poorer outcome.

5.5.4 Concluding Remarks

A treatment should be based on best available evidence, also when few good-quality studies supporting our diagnostic performance and clinical procedures are available. Additional high-quality studies are urgently needed, and until such studies have been conducted, analyzed, and published we must rely on a cautiously interpretation of the findings in the studies currently available.

Many different factors have been suggested to influence the treatments outcome, but it is difficult to single out which factors are the most important for a successful treatment. We know that bacteria cause apical periodontitis, therefore it is not surprising to find the best success rates of root canal treatments in teeth with no pretreatment apical periodontitis, and when the treatment was performed with the highest clinical standard, lege artis.

Still, it is evident that apical periodontitis is a prevalent disease in most populations. In general populations around 30-40% of

References

1 Adenubi, J.O. and Rule, D.C. (1976) Success rate for root fillings in young patients. A retrospective analysis of treated cases. Br Dent J 141: 237-241.

2 Ahmad, I.A. (2009) Rubber dam usage for endodontic treatment: a review. Int Endod J 42: 963-972.

3 Al-Omari, M.A., Hazaa, A., and Haddad, F. (2011) Frequency and distribution of root filled teeth and apical periodontitis in a Jordanian subpopulation. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 111, e59-65.

4 Aleksejuniene, J. et al. (2000) Apical periodontitis and related factors in an adult Lithuanian population. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 90: 95-101.

5 Allard, U. and Palmqvist, S. (1986) A radiographic survey of periapical conditions in elderly people in a Swedish county population. Endod Dent Traumatol 2: 103-108.

6 Aminoshariae, A. et al. (2017) Association between Systemic Diseases and Endodontic Outcome: A Systematic Review. J Endod 43: 514-519.

7 Arya, S. et al. (2017) Healing of apical periodontitis after nonsurgical treatment in patients with type 2 diabetes. J Endod 43: 1623-1627.

8 Assessment SCoHT (2010) Rotfyllning - en systematisk litteratur oversikt [Methods of Diagnosis and Treatment in Endodontics]. Report 203. Stockholm: Swedish Council on Health Technology Assessment. individuals have at least one tooth with apical periodontitis and 25-50% of root-filled teeth have apical periodontitis. These findings may suggest that endodontic treatments performed in general practice do not always follow the international guidelines for good endodontic treatment [39, 45]. On the other hand, population-based cohort studies have found that few patients experience pain 3-5 years after treatment, and furthermore that retention of teeth 10 years after endodontic treatment is high.

9 Balto, K. (2009) Singleor multiple-visit endodontics: which technique results in fewest postoperative problems? Evid Based Dent 10, 16.

10 Barbakow, F.H., Cleaton-Jones, P., and Friedman, D. (1980) An evaluation of 566 cases of root canal therapy in general dental practice. 2. Postoperative observations. J Endod 6: 485-489.

11 Barborka, B.J. et al. (2017) Long-term clinical outcome of teeth obturated with resilon. J Endod 43: 556-560.

12 Basmadjian-Charles, C.L. et al. (2002) Factors influencing the long-term results of endodontic treatment: a review of the literature. Int Dent J 52: 81-86.

13 Benenati, F.W. and Khajotia, S.S. (2002) A radiographic recall evaluation of 894 endodontic cases treated in a dental school setting. J Endod 28: 391-395.

14 Bergenholtz, G. et al. (1979) Influence of apical overinstrumentation and overfilling on re-treated root canals. J Endod 5: 310-314.

15 Bergenholtz, G.M.E. and Milthon, R. (1973) Endodontisk behandling och periapikalstatus. II. Rontgenologisk bedomning av rotfyllningens kvalitet stalld i relation till forekomst av periapikala destruktioner. Tandlakartidningen 5: 269-279.

16 Bergstrom, J., Eliasson, S., and Ahlberg, K.F. (1987) Periapical status in subjects with regular dental care habits. Community Dent Oral Epidemiol 15: 236-239.

17 Berlin-Broner, Y., Febbraio, M., and Levin, L. (2017) Association between apical periodontitis and cardiovascular diseases: a systematic review of the literature. Int Endod J 50: 847-859.

18 Berlinck, T. et al. (2015) Epidemiological evaluation of apical periodontitis prevalence in an urban Brazilian population. Braz Oral Res 29, 51.

19 Bernstein, S.D. et al. (2012) Outcomes of endodontic therapy in general practice: a study by the practitioners engaged in Applied Research and Learning Network. J Am Dent Assoc 143: 478-487.

20 Boucher, Y. et al. (2002) Radiographic evaluation of the prevalence and technical quality of root canal treatment in a French subpopulation. Int Endod J 35: 229-238.

21 Brynolf, I. (1967) A histological and roentgenological study of the periapical region of human upper incisors Odontologisk Revy 18.

22 Buckley, M. and Spângberg, L.S. (1995) The prevalence and technical quality of endodontic treatment in an American subpopulation. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 79: 92-100.

23 Burry, J.C. et al. (2016) Outcomes of primary endodontic therapy provided by endodontic specialists compared with other providers. J Endod 42: 702-705.

24 Bystrom, A. et al. (1987) Healing of periapical lesions of pulpless teeth after endodontic treatment with controlled asepsis. Endod Dent Traumatol 3: 58-63.

25 Chen, C.Y. et al. (2007) Prevalence and quality of endodontic treatment in the Northern Manhattan elderly. J Endod 33: 230-234.

26 Cheung, G.S. (2002) Survival of first-time nonsurgical root canal treatment performed in a dental teaching hospital. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 93: 596-604.

27 Chugal, N.M., Clive, J.M., and Spângberg, L.S. (2001) A prognostic model for: ssessment of the outcome of endodontic treatment: Effect of biologic and diagnostic variables. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 91: 342-352.

28 Chugal, N.M., Clive, J.M., and Spângberg, L.S. (2007) Endodontic treatment outcome: effect of the permanent restoration. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 104: 576-582.

29 Cvek, M., Granath, L., and Lundberg, M. (1982) Failures and healing in endodontically treated non-vital anterior teeth with posttraumatically reduced pulpal lumen. Acta Odontol Scand 40: 223-228.

30 Da Silva, K. et al. (2009) Cross-sectional study of endodontic treatment in an Australian population. Aust Endod J 35: 140-146.

31 Dahlstrom, L., Molander, A., and Reit, C. (2011) Introducing nickel-titanium rotary instrumentation in a public dental service: the long-term effect on root filling quality. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 112: 814-819.

32 De Cleen, M.J. et al. (1993) Periapical status and prevalence of endodontic treatment in an adult Dutch population. Int Endod J 26: 112-119.

33 De Moor, R.J. et al. (2000) Periapical health related to the quality of root canal treatment in a Belgian population. Int Endod J 33: 113-120.

34 Doyle, S.L. et al. (2007) Retrospective cross sectional comparison of initial nonsurgical endodontic treatment and single-tooth implants. Compend Contin Educ Dent 28: 296-301.

35 Dugas, N.N. et al. (2003) Periapical health and treatment quality assessment of root- filled teeth in two Canadian populations. Int Endod J 36: 181-192.

36 Eckerbom, M., Andersson, J.E., and Magnusson, T. (1987) Frequency and technical standard of endodontic treatment in a Swedish population. Endod Dent Traumatol 3: 245-248.

37 Eckerbom, M., Andersson, J.E., and Magnusson, T. (1989) A longitudinal study of changes in frequency and technical standard of endodontic treatment in a Swedish population. Endod Dent Traumatol 5: 27-31.

38 Eckerbom, M., Flygare, L., and Magnusson, T. (2007) A 20-year follow-up study of endodontic variables and apical status in a Swedish population. Int Endod J 40: 940-948.

39 Endodontists AAE. (2013) Guide to Clinical Endodontics. Sixth Edition. Chicago, IL: AAE

40 Engstrom, B. and Frostell, G. (1964) Experiences of bacteriological root canal control. Acta Odontol Scand 22: 43-69.

41 Eriksen, H.M. et al. (1995) Changes in endodontic status 1973-1993 among 35- year-olds in Oslo, Norway. Int Endod J 28: 129-132.

42 Eriksen, H.M. and Bjertness, E. (1991) Prevalence of apical periodontitis and results of endodontic treatment in middle- aged adults in Norway. Endod Dent Traumatol 7: 1-4.

43 Eriksen, H.M., Bjertness, E., and 0rstavik, D. (1988) Prevalence and quality of endodontic treatment in an urban adult population in Norway. Endod Dent: raumatol 4: 122-126.

44 Eriksen, H.M., 0rstavik, D., and Kerekes, K. (1988) Healing of apical periodontitis after endodontic treatment using three different root canal sealers. Endod Dent Traumatol 4: 114-117.

45 European Society of Endodontology. (2006) Quality guidelines for endodontic treatment: consensus report of the European Society of Endodontology. Int Endod J 39: 921-930.

46 Fabricius, L. et al. (2006) Influence of residual bacteria on periapical tissue healing after chemomechanical treatment and root filling of experimentally infected monkey teeth. Eur J Oral Sci 114: 278-285.

47 Farzaneh, M., Abitbol, S., and Friedman, S. (2004) Treatment outcome in endodontics: the Toronto study. Phases I and II: Orthograde retreatment. J Endod 30: 627-633.

48 Farzaneh, M. et al. (2004) Treatment outcome in endodontics - the Toronto Study. Phase II: initial treatment. J Endod 30: 302-309.

49 Fedorowicz, Z. et al. (2012) Irrigants for non-surgical root canal treatment in mature permanent teeth. Cochrane Database Syst Rev, CD008948.

50 Figini, L. et al. (2008) Single versus multiple visits for endodontic treatment of permanent teeth: a Cochrane systematic review. J Endod 34: 1041-1047.

51 Fletcher, R.H., Fletcher, S.W., and Fletcher, G.S. (2014) Clinical epidemiology. The essentials. Baltimore: Williams & Wilkins.

52 Fransson, H. et al. (2016) Survival of root-filled teeth in the Swedish adult population. J Endod 42: 216-220.

53 Frisk, F. and Hakeberg, M. (2005) A 24-year follow-up of root filled teeth and periapical health amongst middle aged and elderly women in Goteborg, Sweden. Int Endod J 38: 246-254.

54 Frisk, F. and Hakeberg, M. (2006) Socio- economic risk indicators for apical periodontitis. Acta Odontol Scand 64: 123-128.

55 Frisk, F., Hugoson, A., and Hakeberg, M. (2008) Technical quality of root fillings and periapical status in root filled teeth in Jonkoping, Sweden. Int Endod J 41: 958-968.

56 Georgopoulou, M.K. et al. (2005) Frequency and distribution of root filled teeth and apical periodontitis in a Greek population. Int Endod J 38: 105-111.

57 Gesi, A. et al. (2006) Incidence of periapical lesions and clinical symptoms after pulpectomy - a clinical and radiographic evaluation of 1- versus 2-session treatment. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 101: 379-388.

58 Gilbert, G.H. et al. (2010) Outcomes of root canal treatment in Dental Practice- Based Research Network practices. Gen Dent 58: 28-36.

59 Gillen, B.M. et al. (2011) Impact of the quality of coronal restoration versus the quality of root canal fillings on success of root canal treatment: a systematic review and meta-analysis. J Endod 37: 895-902.

60 Gulsahi, K. et al. (2008) Frequency of root-filled teeth and prevalence of apical periodontitis in an adult Turkish population. Int Endod J 41: 78-85.

61 Gumru, B. et al. (2011) Retrospective radiological assessment of root canal treatment in young permanent dentition in a Turkish subpopulation. Int Endod J 44: 850-856.

62 Halse, A. and Molven, O. (1987) Overextended gutta-percha and Kloroperka N-O root canal fillings. Radiographic findings after 10-17 years. Acta Odontol Scand 45: 171-177.

63 Hamedy, R. et al. (2016) Prevalence of root canal treatment and periapical radiolucency in elders: a systematic review. Gerodontology 33: 116-127.

64 Harrell, F. (2015) Regression Modeling Strategies: With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. Switzerland: Springer International Publishing AG

65 Harty, F.J., Parkins, B.J., and Wengraf, A.M. (1970) Success rate in root canal therapy. A retrospective study of conventional cases. Br Dent J 128: 65-70.

66 He, J. et al. (2017) Clinical and patient- centered outcomes of nonsurgical root canal retreatment in first molars using contemporary techniques. J Endod 43: 231-237.

67 Hebling, E. et al. (2014) Periapical status and prevalence of endodontic treatment in institutionalized elderly. Braz Dent J 25: 123-128.

68 Hollanda, A.C. et al. (2008) Prevalence of endodontically treated teeth in a Brazilian adult population. Braz Dent J 19: 313-317.

69 Hoskinson, S.E. et al. (2002) A retrospective comparison of outcome of root canal treatment using two different protocols. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 93: 705-715.

70 Huumonen, S. et al. (2003) Healing of apical periodontitis after endodontic treatment: a comparison between a silicone-based and a zinc oxide-eugenol- based sealer. Int Endod J 36: 296-301.

71 Huumonen, S. and 0rstavik, D. (2013) Radiographic follow-up of periapical status after endodontic treatment of teeth with and without apical periodontitis. Clinical Oral Investigations 17: 2099-2104.

72 Huumonen, S., Suominen, A.L., and Vehkalahti, M.M. (2017) Prevalence of apical periodontitis in root filled teeth: findings from a nationwide survey in Finland. Int Endod J 50: 229-236.

73 Huumonen, S., Vehkalahti, M.M., and Nordblad, A. (2012) Radiographic assessments on prevalence and technical quality of endodontically-treated teeth in the Finnish population, aged 30 years and older. Acta Odontol Scand 70: 234-240.

74 Hülsmann, M.P., Peters, O.A., and Dummer, P.M.H. (2005) Mechanical preparation of root canals: shaping goals, techniques and means. Endodontic Topics 10: 30-76.

75 Imfeld, T.N. (1991) Prevalence and quality of endodontic treatment in an elderly urban population of Switzerland. J Endod 17: 604-607.

76 Imura, N. et al. (2007) The outcome of endodontic treatment: a retrospective study of 2000 cases performed by a specialist. J Endod 33: 1278-1282.

77 Jersa, I. and Kundzina, R. (2013) Periapical status and quality of root fillings in a selected adult Riga population. Stomatologija 15: 73-77.

78 Jiménez-Pinzon, A. et al. (2004) Prevalence of apical periodontitis and frequency of root-filled teeth in an adult Spanish population. Int Endod J 37: 167-173.

79 Jokinen, M.A. et al. (1978) Clinical and radiographic study of pulpectomy and root canal therapy. Scand J Dent Res 86: 366-373.

80 Kabak, Y. and Abbott, P.V. (2005) Prevalence of apical periodontitis and the quality of endodontic treatment in an adult Belarusian population. Int Endod J 38: 238-245.

81 Kalender, A. et al. (2013) Influence of the quality of endodontic treatment and coronal restorations on the prevalence of apical periodontitis in a Turkish Cypriot population. Med Princ Pract 22: 173-177.

82 Kerekes, K. and Tronstad, L. (1979) Long-term results of endodontic treatment performed with a standardized technique. J Endod 5: 83-90.

83 Kerekes, K.B. and Bervell, S.F.A. (1976) En rontgenologisk vurdering av endodontisk behandlingsbehov. Den Norske Tannlægeforenings Tidende 86: 248-254.

84 Khalighinejad, N. et al. (2017) The influence of periodontal status on endodontically treated teeth: 9-year survival analysis. J Endod 43: 1781-1785.

85 Kielbassa, A.M., Frank, W., and Madaus, T. (2017) Radiologic assessment of quality of root canal fillings and periapical status in an Austrian subpopulation - An observational study. PLoS One 12, e0176724.

86 Kirkevang L.L. and Horsted-Bindslev, P. (2002) Technical aspects of treatment in relation to treatment outcome. Endodontic Topics 2: 89-102.

87 Kirkevang, L.L. et al. (2001) A comparison of the quality of root canal treatment in two Danish subpopulations examined 1974-75 and 1997-98. Int Endod J 34: 607-612.

88 Kirkevang, L.L. et al. (2001) Frequency and distribution of endodontically treated teeth and apical periodontitis in an urban Danish population. Int Endod J 34: 198-205.

89 Kirkevang, L.L. et al. (2017) Prediction of periapical status and tooth extraction. Int Endod J 50: 5-14.

90 Kirkevang, L.L. et al. (2015) Prognostic value of the full-scale Periapical Index. Int Endod J 48: 1051-1058.

91 Kirkevang, L.L. et al. (2007) Risk factors for developing apical periodontitis in a general population. Int Endod J 40: 290-299.

92 Kirkevang, L.L., Vaeth, M., and Wenzel, A. (2012) Ten-year follow-up observations of periapical and endodontic status in a Danish population. Int Endod J 45:829-839.

93 Kirkevang, L.L., Vaeth, M., and Wenzel, A. (2014) Ten-year follow-up of root filled teeth: a radiographic study of a Danish population. Int Endod J 47: 980-988.

94 Kirkevang, L.L. and Wenzel, A. (2003) Risk indicators for apical periodontitis. Community Dent Oral Epidemiol 31: 59-67.

95 Kirkwood, B.R.S. and Sterne, J.A.C. (2003) Essential medical statistics. Second Edition ed. Oxford, UK: Blackwell Science Ltd.

96 Koch, M. (2013) On implementation of an endodontic program. Swedish Dental Journal 230: 9-97.

97 Koch, M. et al. (2015) Effect of education intervention on the quality and long-term outcomes of root canal treatment in general practice. Int Endod J 48: 680-689.

98 Kojima, K. et al. (2004) Success rate of endodontic treatment of teeth with vital and nonvital pulps. A meta-analysis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 97: 95-99.

99 Krupp, C. et al. (2013) Treatment outcome after repair of root perforations with mineral trioxide aggregate: a retrospective evaluation of 90 teeth. J Endod 39: 1364-1368.

100 Landys Boren, D., Jonasson, P., and Kvist, T. (2015) Long-term survival of endodontically treated teeth at a public dental specialist clinic. J Endod 41: 176-181.

101 Lee, A.H., Cheung, G.S., and Wong, M.C. (2012) Long-term outcome of primary non-surgical root canal treatment. Clin Oral Investig 16: 1607-1617.

102 Lin, P.Y. et al. (2014) The effect of rubber dam usage on the survival rate of teeth receiving initial root canal treatment: a nationwide population-based study. Journal of Endodontics 40: 1733-1737.

103 Loftus, J.J., Keating, A.P., and McCartan, B.E. (2005) Periapical status and quality of endodontic treatment in an adult Irish population. Int Endod J 38: 81-86.

104 Lopez-Lopez, J. et al. (2012) Frequency and distribution of root-filled teeth and apical periodontitis in an adult population of Barcelona, Spain. Int Dent J 62: 40-46.

105 Lupi-Pegurier, L. et al. (2002) Periapical status, prevalence and quality of endodontic treatment in an adult French population. Int Endod J 35: 690-697.

106 Marending, M., Peters, O.A., and Zehnder, M. (2005) Factors affecting the outcome of orthograde root canal therapy in a general dentistry hospital practice. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 99: 119-124.

107 Marques, M.D., Moreira, B., and Eriksen, H.M. (1998) Prevalence of apical periodontitis and results of endodontic treatment in an adult, Portuguese population. Int Endod J 31: 161-165.

108 Marquis, V.L. et al. (2006) Treatment outcome in endodontics: the Toronto Study. Phase III: initial treatment. J Endod 32: 299-306.

109 Matijevic, J. et al. (2011) Prevalence of apical periodontitis and quality of root canal fillings in population of Zagreb, Croatia: a cross-sectional study. Croat Med J 52: 679-687.

110 Mokkink, L.B. et al. (2010) Inter-rater agreement and reliability of the COSMIN (COnsensus-based Standards for the selection of health status Measurement Instruments) checklist. BMC Med Res Methodol 10, 82.

111 Mokkink, L.B. et al. (2010) The COSMIN checklist for evaluating the methodological quality of studies on measurement properties: a clarification of its content. BMC Med Res Methodol 10, 22.

112 Molander, A. et al. (2007) Improved quality of root fillings provided by general dental practitioners educated in nickel- titanium rotary instrumentation. Int Endod J 40: 254-260.

113 Molven, O., Halse, A., and Grung, B. (1987) Observer strategy and the radiographic classification of healing after endodontic surgery. Int J Oral Maxillofac Surg 16: 432-439.

114 Montero, J. et al. (2015) Patient-centered outcomes of root canal treatment: a cohort follow-up study. J Endod 41: 1456-1461.

115 Negishi, J., Kawanami, M., and Ogami, E. (2005) Risk analysis of failure of root canal treatment for teeth with inaccessible apical constriction. J Dent 33: 399-404.

116 Ng, Y.L., Mann, V., and Gulabivala, K. (2008) Outcome of secondary root canal treatment: a systematic review of the literature. Int Endod J 41: 1026-1046.

117 Ng, Y.L., Mann, V., and Gulabivala, K. (2010) Tooth survival following non-surgical root canal treatment: a systematic review of the literature. Int Endod J 43: 171-189.

118 Ng, Y.L., Mann, V., and Gulabivala, K. (2011) A prospective study of the factors affecting outcomes of non-surgical root canal treatment: part 2: tooth survival. Int Endod J 44: 610-625.

119 Ng, Y.L. et al. (2007) Outcome of primary root canal treatment: systematic review of the literature - part 1. Effects of study characteristics on probability of success. Int Endod J 40: 921-939.

120 Ng, Y.L. et al. (2008) Outcome of primary root canal treatment: systematic review of the literature - part 2. Influence of clinical factors. Int Endod J 41: 6-31.

121 Nixdorf, D.R. et al. (2016) Frequency, impact, and predictors of persistent pain after root canal treatment: a national dental PBRN study. Pain 157: 159-165.

122 Nixdorf, D.R. et al. (2010) Frequency of persistent tooth pain after root canal therapy: a systematic review and meta- analysis. J Endod 36: 224-230.

123 Odesjo, B. et al. (1990) Prevalence of previous endodontic treatment, technical standard and occurrence of periapical lesions in a randomly selected adult, general population. Endod Dent Traumatol 6: 265-272.

124 Oginni, A.O., Adeleke, A.A., and Chandler, N.P. (2015) Root canal treatment and prevalence of apical periodontitis in a Nigerian adult subpopulation: a radiographic study. Oral Health Prev Dent 13: 85-90.

125 Oliet, S. (1983) Single-visit endodontics: a clinical study. J Endod 9: 147-152.

126 0rstavik, D. (1996) Time-course and risk analyses of the development and healing of chronic apical periodontitis in man. Int Endod J 29: 150-155.

127 0rstavik, D. and Horsted-Bindslev, P. (1993) A comparison of endodontic treatment results at two dental schools. Int Endod J 26: 348-354.

128 0rstavik, D., Kerekes, K., and Eriksen, H.M. (1986) The periapical index: a scoring system for radiographic assessment of apical periodontitis. Endod Dent Traumatol 2: 20-34.

129 0rstavik, D., Kerekes, K., and Eriksen, H.M. (1987) Clinical performance of three endodontic sealers. Endod Dent Traumatol 3: 178-186.

130 0rstavik, D., Qvist, V., and Stoltze, K. (2004) A multivariate analysis of the outcome of endodontic treatment. Eur J Oral Sci 112: 224-230.

131 Pak, J.G., Fayazi, S., and White, S.N. (2012) Prevalence of periapical radiolucency and root canal treatment: a systematic review of cross-sectional studies. J Endod 38: 170-6.

132 Pak, J.G. and White, S.N. (2011) Pain prevalence and severity before, during, and after root canal treatment: a systematic review. Journal of Endodontics 37: 429-438.

133 Peciuliene, V. et al. (2006) Apical periodontitis in root filled teeth associated with the quality of root fillings. Stomatologija 8: 122-126.

134 Peters, L.B. et al. (2011) Prevalence of apical periodontitis relative to endodontic treatment in an adult Dutch population: a repeated cross- sectional study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 111: 523-528.

135 Peters, L.B. and Wesselink, P.R. (2002) Periapical healing of endodontically treated teeth in one and two visits obturated in the presence or absence of detectable microorganisms. Int Endod J 35: 660-667.

136 Peters, O.A., Barbakow, F., and Peters, C.I. (2004) An analysis of endodontic treatment with three nickel-titanium rotary root canal preparation techniques. Int Endod J 37: 849-859.

137 Petersson, K. (1993) Endodontic status of mandibular premolars and molars in an adult Swedish population. A longitudinal study 1974-1985. Endod Dent Traumatol 9: 13-18.

138 Petersson, K. et al. (2016) Twenty-year follow-up of root filled teeth in a Swedish population receiving high-cost dental care. Int Endod J 49: 636-6345.

139 Petersson, K. et al. (1991) Follow-up study of endodontic status in an adult Swedish population. Endod Dent Traumatol 7: 221-225.

140 Petersson, K. et al. (1989) Endodontic status and suggested treatment in a population requiring substantial dental care. Endod Dent Traumatol 5: 153-158.

141 Petersson, K. et al. (1986) Technical quality of root fillings in an adult Swedish population. Endod Dent Traumatol 2: 99-102.

142 Pettiette, M.T., Delano, E.O., and Trope, M. (2001) Evaluation of success rate of endodontic treatment performed by students with stainless-steel K-files and nickel-titanium hand files. J Endod 27: 124-127.

143 Pirani, C. et al. (2017) Survival and periapical health after root canal treatment with carrier-based root fillings: five-year retrospective assessment. Int Endod J.

144 Pontius, V. et al. (2013) Retrospective evaluation of perforation repairs in 6 private practices. J Endod 39: 1346-1358.

145 Pretzl, B. et al. (2016) Endodontic status and retention of molars in periodontally treated patients: results after 10 or more years of supportive periodontal therapy. J Clin Periodontol 43: 1116-1123.

146 Ray, H.A. and Trope, M. (1995) Periapical status of endodontically treated teeth in relation to the technical quality of the root filling and the coronal restoration. Int Endod J 28: 12-18.

147 Ricucci, D. et al. (2011) A prospective cohort study of endodontic treatments of 1,369 root canals: results after 5 years. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 112: 825-842.

148 Ridell, K. et al. (2006) Periapical status and technical quality of root-filled teeth in Swedish adolescents and young adults. A retrospective study. Acta Odontol Scand 64: 104-110.

149 Rud, J., Andreasen, J.O., and Jensen, J.E. (1972) Radiographic criteria for the assessment of healing after endodontic surgery. Int J Oral Surg 1: 195-214.

150 Salehrabi, R. and Rotstein, I. (2004) Endodontic treatment outcomes in a large patient population in the USA: an epidemiological study. J Endod 30: 846-850.

151 Salehrabi, R. and Rotstein, I. (2010) Epidemiologic evaluation of the outcomes of orthograde endodontic retreatment. Journal of Endodontics 36: 790-792.

152 Sathorn, C., Parashos, P., and Messer, H. (2008) The prevalence of postoperative pain and flare-up in single- and multiple- visit endodontic treatment: a systematic review. International Endodontic Journal 41: 91-99.

153 Sathorn, C., Parashos, P., and Messer, H. H. (2005) Effectiveness of single- versus multiple-visit endodontic treatment of teeth with apical periodontitis: a systematic review and meta-analysis. International Endodontic Journal 38: 347-355.

154 Saunders, W.P. et al. (1997) Technical standard of root canal treatment in an adult Scottish sub-population. British Dental Journal 182: 382-386.

155 Schafer, E. and Burklein, S. (2012) Impact of nickel-titanium instrumentation of the root canal on clinical outcomes: a focused review. Odontology 100: 130-136.

156 Seltzer, S. et al. (1967) Endodontic failures - an analysis based on clinical, roentgenographic, and histologic findings. I. Oral Surg Oral Med Oral Pathol 23: 500-516.

157 Seltzer, S. et al. (1967) Endodontic failures - an analysis based on clinical, roentgenographic, and histologic findings. II. Oral Surg Oral Med Oral Pathol 23: 517-530.

158 Seltzer, S., Bender, I.B., and Turkenkopf, S. (1963) Factors Affecting Successful Repair after Root Canal Therapy. J Am Dent Assoc 67: 651—662.

159 Setzer, F.C. et al. (2011) Long-term prognosis of endodontically treated teeth: a retrospective analysis of preoperative factors in molars. JEndod 37: 21-25.

160 Shakiba, B. et al. (2017) Influence of increased patient age on longitudinal outcomes of root canal treatment: a systematic review. Gerodontology 34: 101-109.

161 Sidaravicius, B., Aleksejuniene, J., and Eriksen, H.M. (1999) Endodontic treatment and prevalence of apical periodontitis in an adult population of Vilnius, Lithuania. Endod Dent Traumatol 15: 210-215.

162 Siqueira, J.F., Jr. and Rôças, I.N. (2011) Optimising single-visit disinfection with supplementary approaches: a quest for predictability. Aust Endod J 37: 92-98.

163 Sjogren, U. et al. (1997) Influence of infection at the time of root filling on the outcome of endodontic treatment of teeth with apical periodontitis. International Endodontic Journal 30: 297-306.

164 Sjogren, U. et al. (1990) Factors affecting the long-term results of endodontic treatment. J Endod 16: 498-504.

165 Skudutyte-Rysstad, R. and Eriksen, H.M.(2006) Endodontic status amongst 35- year-old Oslo citizens and changes over a 30-year period. International Endodontic Journal 39: 637-642.

166 Skupien, J.A. et al. (2016) Survival of restored endodontically treated teeth in relation to periodontal status. Braz Dent J 27: 37-40.

167 Smith, C.S., Setchell, D.J., and Harty, F.J. (1993) Factors influencing the success of conventional root canal therapy - a five- year retrospective study. Int Endod J 26: 321-333.

168 Soikkonen, K.T. (1995) Endodontically treated teeth and periapical findings in the elderly. Int Endod J 28: 200-203.

169 Strindberg, L.Z. (1956) The dependence of the results of pulp therapy on certain factors. An analytic study based on radiographic and clinical follow-up examinations. Acta Odontologica Scandinavia 14.

170 Su, Y.Y., Wang, C.L., and Ye, L. (2011) Healing Rate and Post-obturation Pain of Single- versus Multiple-visit Endodontic Treatment for Infected Root Canals: A Systematic Review. Journal of Endodontics 37: 125-132.

171 Sunay, H. et al. (2007) Cross-sectional evaluation of the periapical status and quality of root canal treatment in a selected population of urban Turkish adults. International Endodontic Journal 40: 139-145.

172 Swartz, D.B., Skidmore, A.E., and Griffin,J.A., Jr. (1983) Twenty years of endodontic success and failure. J Endod 9: 198-202.

173 Tolias, D. et al. (2012) Apical periodontitis in association with the quality of root fillings and coronal restorations: a 14-year investigation in young Greek adults. Oral Health Prev Dent 10: 297-303.

174 Torabinejad, M. et al. (2009) Outcomes of nonsurgical retreatment and endodontic surgery: a systematic review. J Endod 35: 930-937.

175 Touré, B. et al. (2008) Prevalence and technical quality of root fillings in Dakar, Senegal. Int Endod J 41: 41-49.

176 Trope, M., Delano, E.O., and 0rstavik, D. (1999) Endodontic treatment of teeth with apical periodontitis: single vs. multivisit treatment. J Endod 25: 345-350.

177 Tsuneishi, M. et al. (2005) Radiographic evaluation of periapical status and prevalence of endodontic treatment in an adult Japanese population. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 100: 631-635.

178 Van Nieuwenhuysen, J.P., Aouar, M., and D'Hoore, W. (1994) Retreatment or radiographic monitoring in endodontics. Int Endod J 27: 75-81.

179 Vena, D.A. et al. (2014) Prevalence of persistent pain 3 to 5 years post primary root canal therapy and its impact on oral health-related quality of life: PEARL Network findings. J Endod 40:1917-1921.

180 Walter, C. et al. (2012) Association of tobacco use and periapical pathosis - a systematic review. International Endodontic Journal 45: 1065-1073.

181 Weiger, R., Axmann-Krcmar, D., and Lost, C. (1998) Prognosis of conventional root canal treatment reconsidered. Endod Dent Traumatol 14: 1-9.

182 Weiger, R. et al. (1997) Periapical status, quality of root canal fillings and estimated endodontic treatment needs in an urban German population. Endod Dent Traumatol 13: 69-74.

183 Weiger, R., Rosendahl, R., and Lost, C. (2000) Influence of calcium hydroxide intracanal dressings on the prognosis of teeth with endodontically induced periapical lesions. Int Endod J 33:219-226.

184 WHO (2017) Epidemiology. https://www. who.int/topics/epidemiology/en/(accessed 21 June 2019).

185 Zhong, Y. et al. (2008) Extension and density of root fillings and postoperative apical radiolucencies in the Veterans Affairs Dental Longitudinal Study. J Endod 34: 798-803.