Keywords
dental radiograph - endodontic microsurgery - granuloma - histopathology - periapical
cyst
Introduction
A periapical lesion in a necrotic tooth is induced by microbial infection from the
root canal system, which results in an inflammatory reaction and alveolar bone destruction.
A periapical diagnosis can be symptomatic/asymptomatic apical periodontitis or chronic/acute
apical abscess depending on clinical and radiographic characteristics. The histopathological
diagnosis of periapical lesion includes granuloma (nonepithelialized or epithelialized),
cyst, abscess, or others (e.g., scar or foreign body reaction) depending on the microstructure
in the biopsy specimens.[1]
[2] With regard to the previous histopathologic studies, the prevalence of periapical
granulomas (PGs) and periapical cysts (PCs) were 48 to 85 and 23.5 to 42%, respectively,
while the periapical abscess was reported in a few studies with a wide range of prevalences.[1]
[2]
[3]
[4]
[5]
[6]
[7]
From the histopathological point of view, the granulation tissue infiltrated with
chronic inflammatory cells is initially formed in the periapical area of bone destruction.[8] When the periapical lesion is growing, the growth may stimulate the proliferation
of epithelial rests to encircle the lesion. The center of granuloma begins to lack
blood supply and is gradually degenerated. A cystic cavity is then formed as a periapical
cyst with epithelial lining and fluid content inside.[8] The difference in osmolarity induces tissue fluid diffusion into the cystic cavity,
and the inflammatory mediators involving bone resorption increase an expansion of
the cystic lesion. Hence, the PC tends to be larger than the PG; however, the size
of granuloma may be also large.[9]
[10] For a periapical abscess, the lesion contains the granulation tissue with acute
inflammatory cells such as polymorphonuclear leukocytes (neutrophils) or plasma cells,
which is possibly an early infection stage of the PG.[1]
[11] When the acute stage is subsided, the periapical abscess may turn to the PC if the
proliferation of epithelial lining is formed.[12]
The necrotized tooth with periapical lesion is primarily treated by nonsurgical root
canal treatment. In some circumstances, a failure of root canal treatment can be developed
due to several factors such as persistence of bacteria or inadequate filling of the
canal.[13] An advanced treatment of endodontic microsurgery is then required. The PG is more
likely to be healed after the nonsurgical treatment compared to the PC.[12] Nevertheless, many large cysts can be healed by apoptosis after root canal treatment.[9]
[14] A “pocket” cyst that shows a communication to the apical foramen frequently responds
to the treatment after the source of infection from the root canal is eliminated.[12] In contrast, a “true” cyst with complete epithelial lining and no communication
to the apical foramen is possibly not healed after the conservative treatment, in
which surgical removal by endodontic microsurgery may be later required.[12] However, the PG and the pocket/true cyst can be only diagnosed from the histopathological
examination of the biopsy specimens.
Many studies attempted to find an association between clinical/radiographic data and
histopathological diagnoses of PG and PC.[6]
[10]
[15]
[16]
[17]
[18]
[19]
[20] From the clinical presentation, the PC may be associated with detection of fluid
content draining through the root canal or sinus tract.[21]
[22] On the other hand, characteristics of periapical radiolucency in dental radiographs
and/or cone-beam computed tomography (CBCT) may be used to predict the PCs, for example,
a presence of a radiopaque rim (corticated border), size (such as ≥10 mm or 200 mm2) of lesions, or volume (such as ≥ 250 mm3)/characteristics of lesions in CBCT.[6]
[20]
[23] However, the prediction of histopathology of periapical lesions based on these clinical
and/or radiographic appearances is controversial.[1]
[3]
[10]
[16]
[17]
[18]
[19]
Therefore, this retrospective study aimed to determine the association between histopathological
diagnoses and clinical/radiographic appearances of PGs and PCs obtained from the teeth
treated with endodontic microsurgery.
Materials and Methods
The protocol of the study was approved by the institutional ethic committee (MU-DT/PY-IRB
2021/075.2308). The clinical, radiographic, and histological information were collected
from the teeth treated with endodontic microsurgery at the Faculty of Dentistry, Mahidol
University, Bangkok, Thailand from 2016 to 2021. All teeth were treated by endodontists
using microsurgical technique. The soft tissue of periapical lesions were removed
during the surgery and delivered to pathologists for interpretation. The inclusion
criterion was the teeth with histopathological reports of PG or PC. The exclusion
criteria were (a) inadequate clinical record, (b) absence of radiographic data, (c)
unclear histopathological diagnosis, and (d) other histopathologic diagnoses, that
is, periapical abscess, scar, or foreign body reaction.
Data Acquisition
The clinical information were collected as follows: sex (male/female); age (years
old); site of lesion (anterior maxilla/posterior maxilla/anterior mandible/posterior
mandible); type of teeth (incisor /canine/ premolar/molar); history or presence of
these parameters (yes/no)—sinus opening, swelling, pus, or exudate; tooth mobility
(yes/no); pain on percussion (yes/no); and clinical diagnosis (symptomatic/asymptomatic
apical periodontitis, acute/chronic apical abscess).
The digital periapical (parallel) radiographs (X-Mind, Acteon, Olgiate Olona (VA),
Italy) were taken before the surgery. CBCT (3D Accuitomo 170, J. Morita, Kyoto, Japan)
was occasionally taken in a few cases depending on the operators' decision. The radiographic
data of periapical lesions were collected including lesion size (<1 cm, ≥1 cm); radiopaque
rim (fully, partially, without); margin (well-defined, ill-defined); root resorption
(yes, no); and multiple-teeth involvement (yes, no). In addition, the data of periapical
lesions were collected from CBCT (if presented) including the perforation of cortical
bone (with, without), and volume of the lesion (mm3). The radiographic evaluation of periapical lesions was performed by a qualified
oral and maxillofacial radiologist (RA). The evaluation of intraobserver reliability
revealed intraclass correlation coefficient ranging from 82 to 98% that indicated
good reliability.
The histopathological diagnoses were obtained from the specimens prepared by the serial-section
technique. The specimens were microscopically re-examined and classified as PG or
PC by a certified oral pathologist (PL) according to the criteria modified from Nobuhara
and Del Rio[24] and Ricucci et al[25]; the presence of a cystic cavity or granulation tissues, detection of epithelium
lining, and any content in the cavity. The PG contained granulation tissue with no
epithelial lining (or with minor epithelial strand) and no cavity or fluid content.
The PC had a cystic cavity with major epithelium lining and occasionally contained
cellular debris within the cavity.
Statistical Analysis
The statistical analysis was performed using STATA version 17 (StataCorp LLC, College
Station, Texas, United States). The descriptive statistics were used to describe the
distribution of periapical granulomas and cysts according to the considerable factors.
For the radiographic evaluation, the intraobserver reliability was assessed by intraclass
correlation coefficient. The association between clinical /radiographic appearances
and histopathological diagnoses was determined using the bivariate analysis (chi-squared
and Fisher's exact test) and followed by the multinomial logistic regression analysis.
Results
A total of 111 teeth were initially included according to the inclusion criteria,
and then 28 teeth were excluded according to the exclusion criteria. A total of 83
teeth remained for data analysis, of which 68 periapical lesions (18.1%) were granulomas
and 15 periapical lesions (81.9%) were PCs. The 83 teeth were obtained from 83 patients:
25 males and 58 females aged between 14 and 75 years old with a mean age of 48.7 ± 13.8
years. The majority of the teeth were the incisors in the anterior maxillary region.
All general information, clinical, radiographic, and cone-beam CT data are shown according
to the PG and cyst groups in [Tables 1]
[2]
[3]. Representatives of the radiographs and histopathology of the PG and PC are presented
in [Fig.1].
Fig. 1 The upper row—the representative radiograph (A) and the histopathology (B, C at 40x and 200x magnifications) of periapical granuloma. The lower row—the representative radiograph (D) and the histopathology (E, F at 40x and 200x magnifications) of periapical cyst.
Table 1
The general information of the periapical granuloma group and the periapical cyst
group (total, n= 83)
|
General information
|
Periapical granuloma (n = 68)
|
Periapical cyst (n = 15)
|
|
Age (mean ± SD, years)
|
49.5 ± 13.7
|
45.0 ± 14.0
|
|
Gender
|
|
|
|
Male
|
20 (29.4%)
|
5 (33.3%)
|
|
Female
|
48 (70.6%)
|
10 (66.7%)
|
|
Tooth type
|
|
|
|
Incisor (n = 65)
|
52 (76.5%)
|
13 (86.7%)
|
|
Canine (n = 3)
|
3 (4.4%)
|
0
|
|
First premolar (n = 9)
|
9 (13.2%)
|
0
|
|
Molar(n = 5)
|
3 (4.4%)
|
2 (13.3%)
|
|
Combination (n = 1)
|
1 (1.5%)
|
0
|
|
Site of lesion
|
|
|
|
Anterior maxilla (n = 63)
|
51 (75%)
|
12 (80%)
|
|
Posterior maxilla (n = 13)
|
11 (16.2%)
|
2 (13.3%)
|
|
Anterior mandible (n = 6)
|
5 (7.3%)
|
1 (6.7%)
|
|
Posterior mandible (n = 1)
|
1 (1.5%)
|
0
|
Abbreviation: SD, standard deviation.
Table 2
The clinical and radiographic information of the periapical granuloma group and the
periapical cyst group (total, n = 83) with a bivariate analysis (chi-squared and Fisher's exact test)
|
Factors
|
Periapical granuloma (n = 68)
|
Periapical cyst (n = 15)
|
Bivariate analysis
(p-Value)
|
|
Clinical appearances
|
|
|
|
|
History of fistula
|
|
|
|
|
Yes (n = 36)
|
32 (47.1%)
|
4 (26.7%)
|
0.149[a]
|
|
No (n = 47)
|
36 (52.9%)
|
11 (73.3%)
|
|
History of swelling
|
|
|
0.108[a]
|
|
Yes (n = 23)
|
16 (23.5%)
|
7 (46.7%)
|
|
No (n = 60)
|
52 (76.5%)
|
8 (53.3%)
|
|
History of pus
|
|
|
1
|
|
Yes (n = 17)
|
14 (20.6%)
|
3 (20%)
|
|
No (n = 66)
|
54 (79.4%)
|
12 (80%)
|
|
History of exudate
|
|
|
0.003[a]
|
|
Yes (n = 5)
|
1 (1.5%)
|
4 (26.7%)
|
|
No (n = 78)
|
67 (98.5%)
|
11 (73.3%)
|
|
Tooth mobility (n = 77)[b]
|
|
|
0.617
|
|
Yes (n = 7)
|
5 (8.1%)
|
2 (13.3%)
|
|
No (n = 70)
|
57 (91.9%)
|
13 (86.7%)
|
|
Pain on percussion (n = 81)[b]
|
|
|
0.358
|
|
Yes (n = 24)
|
18 (27.3%)
|
6 (40%)
|
|
No (n = 57)
|
48 (72.7%)
|
9 (60%)
|
|
Clinical diagnosis
|
|
|
|
|
AAP (n = 22)
|
17 (25%)
|
5 (33.3%)
|
0.633
|
|
SAP (n = 9)
|
8 (11.8%)
|
1 (6.7%)
|
|
AAA (n = 15)
|
11 (16.2%)
|
4 (26.7%)
|
|
CAA (n= 37)
|
32 (47%)
|
5 (33.3%)
|
|
Radiographic appearances
|
|
|
|
|
Tooth involvement
|
|
|
0.002[a]
|
|
Single tooth (n = 73)
|
64 (94.1%)
|
9 (60%)
|
|
Multiple teeth (n = 10)
|
4 (5.9%)
|
6 (40%)
|
|
Size of lesion
|
|
|
0.003[a]
|
|
< 1 cm (n =50)
|
46 (67.6%)
|
4 (26.7%)
|
|
≥ 1 cm (n = 33)
|
22 (32.4%)
|
11 (73.3%)
|
|
Margin
|
|
|
0.212[a]
|
|
Well-defined (n = 58)
|
45 (66.2%)
|
13 (86.7%)
|
|
Ill-defined (n = 25)
|
23 (33.8%)
|
2 (13.3%)
|
|
Radiopaque rim
|
|
|
0.149[a]
|
|
None (n = 57)
|
49 (72.1%)
|
8 (53.3%)
|
|
Partial (n = 8)
|
7 (10.3%)
|
1 (6.7%)
|
|
Full (n = 18)
|
12 (17.6%)
|
6 (40%)
|
|
Root resorption (n = 79)[b]
|
|
|
0.591
|
|
Yes (n = 40)
|
32 (49.2%)
|
8 (57.1%)
|
|
|
No (n = 39)
|
33 (50.8%)
|
6 (42.9%)
|
|
Abbreviations: AAA, acute apical abscess; AAP, asymptomatic apical periodontitis;
CAA, chronic apical abscess; SAP, symptomatic apical periodontitis.
a The factors with a p-value≤0.25 were further included in the multinomial logistic regression model.
b Data was not available in all cases.
Table 3
The cone-beam computed tomography information of the periapical granuloma group and
the periapical cyst group (total, n= 23)
|
Cone-beam computed tomography factors
|
Periapical granuloma
(n = 21)
|
Periapical cyst
(n = 2)
|
|
Volume of lesions
|
|
|
|
Mean
|
222.1 ± 270.7 mm3
|
2238.5 ± 789.1 mm3
|
|
Range
|
6.678-882.518 mm3
|
1680.5- 2796.404 mm3
|
|
Perforation of cortical bone (n)
|
15 (71.4%)
|
2 (100%)
|
For the clinical appearances ([Table 2]), the PC group showed a higher prevalence of history of swelling or exudate compared
to the PG group—46.7 versus 23.5% and 26.7 versus 1.5%, respectively. In contrast,
the PC group showed a lower prevalence of history of fistula (26.7%) compared to the
PG group (47.1%). However, the prevalence of history of pus was the same between the
two groups (20% approximately). The common clinical diagnoses in both groups were
chronic apical abscess and asymptomatic apical periodontitis.
For the radiographic appearances ([Table 2]), the PC and PG groups typically showed different characteristics. The PC group
had a higher prevalence than the PG group in multiple-teeth involvement (40 vs. 5.9%),
a well-defined margin (86.7 vs. 66.2%), and partial/full radiopaque rim (46.7 vs.
27.9%). Moreover, the lesion size of the PC group was commonly larger than 1 cm (73.3%),
while the lesion size of the PG group was frequently smaller than 1 cm (67.6%). Approximately
half of either PC or PG group had a sign of root resorption.
The CBCT data was available only in 23 out of 83 teeth ([Table 3]), including 2 teeth in the PC group and 21 teeth in the PG group. The mean volume
of PC was 2238.5 ± 789.1 mm3 while that of PG was 222.1 ± 270.7 mm3. Moreover, 100% of PC presented the perforation of cortical bone, whereas 71.4% of
PG did.
The results of the bivariate analysis are presented in [Table 2]. In comparison to the PG group, the factors probably associated with the diagnosis
of PC are the three clinical appearances including history of fistula, swelling, and
exudate, as well as the four radiographic appearances including multiple-teeth involvement,
lesion size more than 1 cm, well-defined margin, and radiopaque rim (p < 0.25). However, the history of exudate was excluded from the multinomial logistic
regression model since only one case was observed in the PG group that was not appropriate
for further statistical analysis. Among the remaining factors, the multiple-teeth
involvement was only a significant factor in the multinomial logistic regression analysis
(p < 0.05; [Table 4]). The periapical lesion with the radiographic appearance of multiple-teeth involvement
showed approximately six times higher chance to be PC than PG.
Table 4
The multinomial logistic regression analysis of the factors selected from the bivariate
analysis as a predictor of periapical cyst compared to periapical granuloma
|
Factors
|
Relative risk ratio
|
p-Value
|
95% confidence interval
|
|
History of swelling
|
1.560
|
0.661
|
0.212 − 11.445
|
|
History of fistula
|
1.334
|
0.728
|
0.262 − 6.789
|
|
Margin (well-defined)
|
1.063
|
0.951
|
0.150 − 7.510
|
|
Radiopaque rim (yes)
|
1.432
|
0.414
|
0.604 − 3.393
|
|
Size of lesion (≥1 cm)
|
3.015
|
0.175
|
0.612 − 14.843
|
|
Tooth involvement (multiple)
|
6.307
|
0.035[a]
|
1.139 − 34.903
|
a Multiple-teeth involvement was only the significant factor with a p-value < 0.05 and a relative risk ratio of 6.307.
Discussion
This retrospective study is one of a few clinical studies comprehensively evaluating
an association between clinical/radiographic data and histopathological diagnoses
of periapical lesions. Most of the other studies only investigated the association
from either clinical or radiographic data.[1]
[16]
[17]
[18]
[26]
[27]
The multiple-teeth involvement in periapical radiographs was significantly associated
with the diagnosis of PC. This can be explained by the pathogenesis of PG and PC,
in which PC tends to be larger than PG.[21]
[22] The prolonged inflammation of PG may induce the proliferation and differentiation
of the epithelial rests of Malassez to form the epithelial lining that encloses the
highly vascularized granulation tissue. After that, the center of granulation tissue
lacks blood supply that consequently induces liquefaction necrosis and becomes a cystic
cavity with enclosing epithelial lining. The osmolarity of cystic fluid contributes
to the increase of fluid content via transportation of serum fluid from the neighboring
tissue accumulating into the cystic cavity, and, as a result, the enlargement of PC
especially in the horizontal dimension.[14] In our study, the prediction of PC was significantly related to the multiple-teeth
involvement with sixfolds higher chance than that of PG.
The radiographic size of periapical lesion at or greater than 1 cm was not a significant
predictor for the cyst in our study. Other studies[10]
[17]
[19] also failed to show an association between the histopathological diagnosis of the
PC and the lesion size. In contrast, Mortensen et al[15] and Natkin et al[6] proposed that PC could be differentiated from PG when the size of the lesion was
equal to or larger than 1 cm. This discrepancy maybe due to a difference of the sample
size and different study cohorts. However, the prediction of PC from the radiographic
size of the lesion remains controversial and may be unreliable.[10]
[19]
[28] However, the horizontal enlargement of periapical lesion involving the adjacent
tooth was a significant predictor in our study. This result may be explained by the
tendency of cystic expansion in the horizontal dimension rather than in the vertical
dimension, as previously mentioned.
It was believed that a radiopaque rim around the periapical lesion was the bone reaction,
as a defense mechanism, to enclose the slow-progressing cystic lesion. However, the
present study did not find a significant association between the radiopaque rim and
the histological diagnosis of PC, corresponding to the results of Ricucci et al.[16]
Most of the clinical studies have reported no relationship between clinical characteristics
and histopathological diagnoses of PC or granuloma.[3]
[7]
[26]
[28] In our study, the history of inflammatory exudate tended to be a predictor of the
PC; however, the number of cases with the exudate in our study was limited (only 5
teeth; 1/68 teeth in the granuloma group, and 4/15 teeth in the cyst group) and not
appropriate for statistical analysis. The exudate is commonly the content in the cystic
cavity and may be used as a predictor for the PC.[22] Our retrospective study collected the data from dental records, in which the presence
of minor exudate may be neglected and not documented.
Our study found a trend in the three-dimensional CBCT images that the PC (2238 ± 789 mm3) had a higher volume than the PG (222 ± 270.7 mm3), which was close to what was reported in a previous study.[20] We believed that there is a tendency to use the CBCT volume of a periapical lesion
for predicting the PC. However, the number of cases with CBCT data in our study was
limited due to preoperative CBCT that has not been yet a standard of examination before
endodontic microsurgery in our daily practice. A relationship between the CBCT volume
and the histopathology of the periapical lesion should be further investigated.
Based on our findings, the multiple-teeth involvement may be used as a predictor to
distinguish between the PG and PC, that probably imply the prognosis of treatment.
The PG is commonly healed after nonsurgical root canal treatment, and the PC is also
likely to be healed after the nonsurgical treatment by the apoptosis of epithelial
lining.[14] However, it has been proposed that the pocket cyst (with a communication to the
root canal) can be healed by the root canal treatment, while a true cyst (completely
isolated from the root canal) is self-sustaining and less likely to respond to the
nonsurgical treatment.[12] A possible association (if any) between the clinical/radiographic findings and the
histological diagnosis may be useful to predict the prognosis of root canal treatment.
Conclusion
In our studied population, the incidences of PG and cyst were approximately 82 and
18%, respectively. The significant factor for prediction of PC was the multiple-teeth
involvement of the periapical lesion in a dental radiograph.
