Supportive Care in Cancer (2022) 30:8745–8759

[Link]

REVIEW ARTICLE

To extract or not extract teeth prior to head and neck radiotherapy?

A systematic review and meta‑analysis
Ana Gabriela Costa Normando1 · Maria Eduarda Pérez‑de‑Oliveira1 · Eliete Neves Silva Guerra2 ·
Márcio Ajudarte Lopes1 · André Caroli Rocha3,4 · Thaís Bianca Brandão5 · Ana Carolina Prado‑Ribeiro1,5 ·
Luiz Alcino Monteiro Gueiros6 · Joel B. Epstein7,8 · César Augusto Migliorati9 · Alan Roger Santos‑Silva1 · on behalf
of the Oral Care Study Group, Multinational Association of Supportive Care in Cancer (MASCC)/International
Society of Oral Oncology (ISOO)

Received: 18 January 2022 / Accepted: 9 June 2022 / Published online: 17 June 2022
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022

Abstract
Purpose Teeth with poor prognosis are generally recommended to be extracted prior to head and neck radiotherapy (RT) to
reduce the risk of developing osteoradionecrosis (ORN), although controversies have been reported. The present systematic
review aimed to determine whether tooth extraction prior to head and neck RT may be associated with a reduced risk of
developing ORN compared to dental extraction during or after RT.
Methods The review protocol was registered in PROSPERO (CRD42021241631). The review was reported according to
the PRISMA checklist and involved a comprehensive search of PubMed, Scopus, Embase, Cochrane Library, LILACS, and
Web of Science, in addition to the gray literature. The selection of studies was performed in two phases by two reviewers
independently. The risk of bias of individual studies was analyzed using the Joanna Briggs Institute checklist for cross-
sectional studies, and the certainty of evidence was assessed using the GRADE tool.
Results Twenty-eight observational studies were included in the qualitative synthesis, which showed substantial heteroge-
neity regarding the association between the timing of tooth extraction and ORN development. Twenty-seven of 28 studies
were pooled in a meta-analysis that demonstrated a significant association between an increased risk of ORN and post-RT
tooth extraction (odds ratio: 1.98; 95% CI: 1.17–3.35; p = 0.01).
Conclusion It was confirmed with moderate certainty that dental extractions should be performed prior to the start of head
and neck RT to reduce the risk of ORN.

Keywords Tooth extraction · Radiotherapy · Osteoradionecrosis · Systematic review · Meta-analysis

Ana Gabriela Costa Normando and Maria Eduarda Pérez-de-

Oliveira shared first authorship.

6
* Alan Roger Santos‑Silva Departamento de Clínica E Odontologia Preventiva, Oral
alan@[Link] Medicine Unit, Universidade Federal de Pernambuco, Recife,
PE, Brazil
1
Oral Diagnosis Department, Piracicaba Dental School, 7
Cedars‑Sinai Medical Center, Samuel Oschin Comprehensive
University of Campinas (UNICAMP), Piracicaba, SP, Brazil
Cancer Institute, Los Angeles, CA, USA
2
Laboratory of Oral Histopathology, Health Sciences Faculty, 8
City of Hope Comprehensive Cancer Center, Duarte, CA,
University of Brasilia, Brasilia, DF, Brazil
USA
3
Oral Medicine Department, AC Camargo Cancer Center, 9
College of Dentistry, University of Florida, Gainesville, FL,
São Paulo, SP, Brazil
USA
4
Clinics Hospital of the Medical School of the University
of São Paulo, São Paulo, Brazil
5
Dental Oncology Service, Instituto Do Câncer Do Estado de
São Paulo (ICESP-FMUSP), São Paulo, Brazil

13
Vol.:(0123456789)
8746 Supportive Care in Cancer (2022) 30:8745–8759

Introduction post-RT dental extractions [17–19]. This raises the question

of whether dental extractions before RT are beneficial for
Head and neck cancer (HNC) is a heterogeneous group of preventing the development of ORN.
tumors that may affect the lips, oral cavity, pharynx, lar- Given the inconsistency in the current literature, the aim
ynx, and salivary glands [1]. In 2020, over 900,000 new of this systematic review was to determine whether tooth
HNC cases were estimated worldwide, with approximately extraction prior to head and neck RT is associated with a
460.000 deaths [2]. Squamous cell carcinoma (SCC) of the reduced risk of developing ORN compared to dental extrac-
larynx and hypopharynx is the second most common respir- tion after or during RT.
atory tract cancer, and more than 90% of cancers in the oral
cavity are SCCs [3]. Early-stage SCC may be surgically con-
trolled or treated by radiotherapy (RT) alone, while locally Methods
advanced disease, which encompasses more than 60% of
patients with HNC, is treated mainly based on multimodal Eligibility criteria
approaches that include surgery and RT, with or without
chemotherapy [1]. Radiotherapy plays a key role in the cura- The mnemonic PICOS (Population, Intervention, Com-
tive-intent management of HNC. However, it may lead to a parison, Outcomes, Studies) was used to address a focused
variety of acute and late toxicities, such as oral mucositis, research question and to determine the inclusion criteria in
xerostomia, dysphagia, and osteoradionecrosis (ORN) [4]. this study: (P) HNC patients; (I) tooth extraction prior to
ORN of the jaws is defined as exposed irradiated bone head and neck RT; (C) tooth extraction during/after head and
that fails to heal over a period of 3 months without evi- neck RT; (O) determination of the best time to perform den-
dence of persisting or recurrent tumor [5]. Clinical signs and tal extractions in patients who are receiving head and neck
symptoms of ORN may include ulceration or necrosis of the RT to reduce the risk of developing ORN; and (S) observa-
mucosa with bone exposure, pain, trismus, and suppuration tional studies (cross-sectional, cohort or case–control) and
[6]. ORN progression may lead to pathological fractures of clinical trials.
the maxillofacial bones, intra- or extraoral fistulae, local or We excluded (1) studies that assessed patients with malig-
systemic infection, and difficulties in mouth opening, mas- nant lesions at anatomical sites other than the head and neck;
tication, and speech [7]. Several staging or grading systems (2) studies that did not analyze the impact of dental extrac-
of ORN have been proposed. Despite the differences, all tion or no extraction before and during/after RT on the risk
are based on clinical and radiological findings and clas- of developing ORN; (3) studies that did not clearly report
sify the severity of the injury to provide better management or could not calculate the odds ratio of ORN; (4) reviews,
[5]. Numerous risk factors have been associated with ORN case series/reports, protocols, short communication, per-
development, including radiation dose and fractionation, sonal opinions, letters, conference abstracts, book chapters,
poor oral hygiene, alcohol and tobacco abuse, tumor size and in vitro or in vivo animal studies; (5) studies that did
and location, staging, patient comorbidities (e.g., diabetes), not satisfy a language requirement; (6) studies for which a
and dental extractions [8]. full-text copy was not available; and (7) duplicated samples.
To reduce the risk of ORN and other radiation-related
adverse effects on the oral cavity, it is widely recommended Information sources and search strategy
that all patients being considered for head and neck RT be
evaluated by a dental clinician prior to the beginning of the Individualized search strategies were implemented on March
treatment [9]. It is usually recommended that nonfunctional 6th, 2021 for each of the following databases: Cochrane
teeth or those with a poor restorative or periodontal progno- Library, EMBASE, LILACS, PubMed, Scopus, and Web
sis that will be included in the planned radiation volume be of Science (Appendix 1). Gray literature searches were also
extracted to decrease the risk of ORN. It is recommended carried out on Google Scholar, Open Gray, and ProQuest.
that dental extractions should be performed at least 14 days Additionally, the references of the included studies were
before RT initiation [8–10]. It is assumed that early extrac- manually screened for potential additional studies. The
tions prior to the deleterious effect of RT on the healing retrieved studies were imported into the Endnote Web ref-
capacity of bone will decrease the risk of delayed healing erence manager (Endnote Web, Clarivate Analytics, Phila-
and ORN [11, 12]. Nevertheless, studies have found that pre- delphia, PA), where duplicate references were removed. No
RT extractions were also associated with an increased risk of limits on the publication date were applied in the search
developing ORN [9, 13–16]. In contrast, additional studies strategy.
found no differences in the risk of ORN between pre- and

13
Supportive Care in Cancer (2022) 30:8745–8759 8747

Selection process Effect measures

The study selection was accomplished in two phases. The The main objective of this systematic review was to deter-
first phase was performed by two reviewers (AGCN and mine the best time to perform dental extractions in patients
MEPO) who independently read the titles and abstracts of who are being treated with head and neck RT to reduce the
screened studies on Rayyan® [20] and applied the eligibil- risk of developing ORN. The secondary objective was to
ity criteria to select the studies for the second phase. The evaluate the risk of developing ORN following dental extrac-
studies that appeared to meet all inclusion criteria went on tion in patients who are undergoing head and neck RT. The
to the second phase of the selection process, where full texts effect measures that were considered were the odds ratio and
were independently read by the same two authors and the the hazard ratio. For the odds ratio, the collected data were
eligibility criteria were confirmed. The reference lists of all the total number of patients who had tooth extractions before
included studies were hand-screened for potential missing or after RT and the number of events (ORN) in each group.
studies. Divergences in both phases were resolved by discus- In this way, we were able to determine whether the chance
sion and mutual agreement among the authors. of developing ORN is greater in patients who undergo tooth
extraction before versus after RT or whether there is no
Data collection process and data items difference.

The choice of data to be extracted was made in agreement Synthesis methods

with the research team to collect the data that best delin-
eated the sample, interventions, and outcomes. Therefore, A qualitative synthesis was performed by grouping the data
the most important data from the included studies were from all included studies according to feature similarity
collected by one reviewer (AGCN) and crosschecked by a to obtain frequency data for each of the characteristics of
second reviewer (MEPO). Any differences were resolved interest. Statistical data on association, risk, and survival
by discussion and mutual agreement between the authors. were collected to understand the significance of the tooth
The extracted data included (a) publication data (first extraction period for ORN development. The quantitative
author, year, country and journal of publication, and study synthesis was subdivided according to the included sample:
design); (b) sample characteristics (sample size, sex, age, (1) In studies with HNC patients as the main sample, the
habits, comorbidities, HNC diagnosis, tumor site, and clini- total number of patients who underwent pre-RT or post-RT
cal stage); (c) cancer treatment characteristics (RT modal- dental extractions and the proportion of patients who devel-
ity, other treatments, and radiation dose and duration); (d) oped ORN were collected; (2) in studies that only included
dental extraction characteristics (number of extractions that patients who were diagnosed with ORN, the total number
were performed before, during or after RT; sites of dental of patients with ORN and the number of these patients who
extractions; intervals between dental extractions and the underwent dental extractions pre-RT or post-RT were col-
start and end of RT; and procedures that were performed lected. An association meta-analysis was performed fol-
following dental extractions); (e) ORN characteristics (inter- lowing the appropriate Cochrane Guidelines [22]. Review
val between dental extraction/RT and ORN, prevalence and Manager 5.4 software (RevMan 5.4, The Nordic Cochrane
diagnostic criteria of ORN, site and treatment of ORN, and Centre, Copenhagen, Denmark) was used to construct forest
follow-up time); and (f) results (statistical analysis and main plots, with the odds ratio (OR) and 95% confidence intervals
results). (CI) determined at a significance level of 5%. Statistical het-
erogeneity was calculated using an inconsistency index (­ I2),
Risk of bias assessment and a random effect model was applied to address heteroge-
neities. Additionally, a sensitivity analysis was performed
The risk of bias of individual studies was independently by excluding studies with a high risk of bias from the plots.
assessed by two authors (AGCN and MEPO) using the
Joanna Briggs Institute (JBI) Critical Appraisal Tools for Certainty assessment
Analytical Cross-sectional studies [21]. A calibration was
performed between the evaluators prior to individual assess- The certainty of the cumulative evidence was assessed
ments with five studies. Studies were characterized as having using the grading of recommendation, assessment, devel-
a high risk of bias when the “yes” response score was less opment, and evaluation (GRADE) instrument [23]. The
than or equal to 49%, as having a moderate risk of bias when assessment was applied to both outcomes that were eval-
the score was between 50 and 69%, and as having a low risk uated in the present review, and it was based on study
of bias when the score was 70% or higher. Divergences were design, risk of bias, inconsistency, indirectness, impreci-
resolved by mutual agreement. sion, and other considerations, such as publication bias

13
8748 Supportive Care in Cancer (2022) 30:8745–8759

and effect magnitude. The certainty of evidence was Study characteristics

scored as high, moderate, low, or very low. A GRADE
evidence profile was developed using the online software All 28 included studies were published in English between
GRADEpro [24]. 1976 and 2021. Most studies were performed in the USA
(n = 8), followed by Japan (n = 4), Belgium (n = 2), Can-
ada (n = 2), and Taiwan (n = 2). All included studies were
Results observational studies, of which 24 were retrospective cross-
sectional studies, 3 were retrospective cohort studies, and 1
Study selection was a single-arm prospective study. The total sample size
was 50,192 patients, and the sample sizes ranged from 33
The searches identified 4528 records in databases and to 25,246 patients among the studies. Most of the included
372 additional studies in the gray literature, for a total patients were male (n = 44,062; 87.8%) with ages that ranged
of 4900 studies. These studies were managed, and dupli- from 7 to 94 years old and with tumors that were mainly
cates were removed. After this process, 3010 references located in the oral cavity (n = 33,568; 66.9%). The most
remained, with 2683 originating from the main data- common diagnosis was squamous cell carcinoma (n = 4585),
bases and 327 from the gray literature. The titles and which represents only 9.1% of the overall sample since the
abstracts of the studies were read for initial screening. studies with larger samples did not include the tumor diag-
After confirming the eligibility criteria and discussing nosis. Regarding confounding factors, deleterious habits
the divergences, 72 studies were selected for full-text were reported by 17 studies that described 1483 patients
reading. Finally, following eligibility criteria confirma- who were current smokers and 1336 who were current alco-
tion, 28 studies were selected for qualitative and quanti- hol drinkers. Comorbidities, which were also considered
tative analysis [8, 9, 11–19, 25–41]. Cohen’s kappa sta- confounding factors, were only reported by ten studies, and
tistic for inter-reviewer agreement in phase 2 was 0.801 the most prevalent was hypertension (n = 17,093), followed
(p = 0.000). The study selection process is summarized by diabetes mellitus (n = 9867). Finally, most patients were
in the flowchart (Fig. 1), and the reasons for exclusion diagnosed with tumors of stages III–IV (n = 3714), which
of each of the articles that were read in full are described may also have impacted the ORN outcome. However, nine
in Appendix 2. studies did not provide this information. Detailed clinical
demographic characteristics of the sample are summarized
in Table 1.

Fig. 1  Flow diagram of the literature search and selection criteria, which were adapted from PRISMA [55]

13
Supportive Care in Cancer (2022) 30:8745–8759 8749

Table 1  Summary of clinicopathological characteristics of the overall Table 1  (continued)

sample (n = 50,192)
Sample characteristics
Sample characteristics
Age, y 7–94
Age, y 7–94 Range
Range
ORN following dental extractions 2618
Sex 4958 Pre-RT 1309
Female 44,062 Post-RT
Male ORN characteristics
Smoking habits 1483 Overall ORN prevalence 2.14–22.1
Current 962 Range, %
Former smoker 464 Site of ORN 804
Non-smoker Mandible 64
Alcohol habits 1336 Maxilla 2
Current 46 Other
Former drinker 14 Interval between RT and ORN 0–30
Non-drinker Range, y
Comorbidities 17,093 Follow-up time 0–34
Cardiovascular disease/hypertension 9867 Range, y
Diabetes mellitus 2883
Liver disease 2610
Chronic renal failure 825
Rheumatologic disease 64 Concerning the cancer treatment modalities, all included
Other patients underwent RT, either alone or in association with
Tumor characteristics other treatments such as surgery and chemotherapy. The
Tumor diagnosis 4585 radiation doses were reported by all studies with the excep-
Squamous cell carcinoma 231 tion of one [40] and ranged from 13.3 Gy [11] to 90 Gy
Nasopharyngeal carcinoma 86
Lymphoma 46 [8]. The radiation treatment duration ranged from 1 [11] to
Undifferentiated carcinoma 37 11 weeks [35]. Most studies did not report the duration of
Adenocarcinoma 29 therapy (n = 20), and most of those that did reported it as a
Adenoid cystic carcinoma 175 minimum and a maximum.
Other
Collectively, 33,662 patients underwent pre-RT tooth
Tumor site 33,568
Oral cavity 5706 extractions, 11,705 underwent extractions after RT, and only
Oropharynx 5240 1265 underwent tooth extractions during RT. The extracted
Hypopharynx 374 teeth were located mainly in the mandible, although 21 stud-
Nasopharynx 216 ies did not report this information. Only thirteen studies pro-
Larynx 127
Salivary glands 45 vided data on the interval between RT and dental extraction.
Pharynx 740 The pre-RT period ranged from 3 to 210 days with a mean
Head and neck other of 24.7 days, while the post-RT extraction interval ranged
Tumor stage 926 from 3 months to 13.4 years.
I-II 3714 The overall prevalence of ORN among the included stud-
III-IV 116
Unknown ies ranged from 2.14% [18] to 22.1% [2], with the interval
RT Modality 10,343 between RT and the development of ORN ranging from 0
IMRT 2357 to 30 years. This information was obtained from 19 studies.
Conventional RT 943 Individual data regarding treatment modalities, radiation
3D Conformal RT 317 doses, the numbers of patients who underwent extractions
Radioactive implants 212
Brachytherapy pre- and post-RT, and the number of patients who developed
Dental extractions characteristics ORN in each group are presented in Table 2.
Dental extractions period 33,682
Before RT 1265 Risk of bias in the studies
During RT 11,705
After RT The majority of the included studies were graded as hav-
Site of dental extractions 289 ing a low (n = 16; 57.1%) or moderate (n = 4; 14.2%) risk
Maxilla 480
Mandible of bias, while eight studies (28.6%) were scored as having
a high risk of bias. Major issues that influenced the risk of
bias were identified in the methods. Close to 47% of the

13
 Table 2  Summary of treatment modalities, dental extractions, and ORN characteristics among included studies (n = 28)
8750

Author (year) Sample size Treatment modalities Radiation doses (Gy) Dental extractions before RT Dental extractions after RT ORN after pre-RT ORN after post-RT
dental extractions dental extractions

13
Acharya et al. (2020) 231 All—CT: 47% ≤ 66 = 61 Only before = 6 Only after = 2 11 7
ORN patients: ≥ 66 = 183 Before + after = 5 Before + after = 5
Surgery + CT + RT = 4
RT = 5
CT + RT = 4
Beech et al. (2017) 190 CT + RT = 167 Mean of 68 Only before = 109 Only after = 10 22 4
RT = 21 (30–77) Before + after = 20 Before + after = 20
Sequential CT and RT = 2
Beumer et al. (1984) 70 RT = 60 50–64.99 = 15 17 22 17 22
RT + RI or with a radiation carrier: 65–69.99 = 27
22 > 70 = 18
RI: 1 (1.2%)
Caparrotti et al. 1,196 RT = 501 70 = 970 72 47 31 18
(2017) RT + CT = 614 60 = 118
RT + EGFRI = 81 64 = 116
50–74 = 19
Chang et al. (2007) 413 Altered fractionation: 327 Median of 75.6 163 51 Only pre-RT: 20; Only post-RT: 6;
BRT + external RT: 20 (50–81.6) Pre-RT + Post-RT: 6 Pre-RT + Post-RT: 6
Induction/concurrent CT: 66
Chopra et al. (2011) 46 Primary RT = 28% Median of 60 27 18 27 18
Adjuvant RT = 72% (50–72.5)
CT = 78% Dose > 60 = 47%
Surgery = 72% Dose < 60 = 53%
Chronopoulos et al. 115 RT = 140 Mean of 63.4 10 40 10 40
(2015) RT and one adjuvant = 10 < 60 = 62
RT and two adjuvant = 3 > 60 = 76
CT = 123
Dumoulin et al. 415 RT = 77 50–70 3 2 3 2
(2021) RT + surgery = 9
CRT = 135
CRT + surgery = 53
CT + surgery + CRT = 5
Surgery + RT = 71
Surgery + CRT = 47
Surgery = 18
Epstein et al. (1987) 146 RT = 146 13.3–62.5 92 42 5 3
Habib et al. (2020) 197 RT = 30 < 55 = 10 48 59 48 59
RT + CT = 71 55–65 = 103
RT + surgery = 67 > 65 = 17
RT + CT + surgery = 29 Unknown = 67
Koga et al. (2008) 405 Surgery + RT = 201 Mean of 63.04 363 57 2 1
RT = 74 (40–75.20)
RT + CT = 72
Surgery + RT + CT = 58
Kojima et al. (2017) 392 RT = 137 Mean of 63.5 132 38 7 11
CRT/BRT = 255
Supportive Care in Cancer (2022) 30:8745–8759
 Table 2  (continued)
Author (year) Sample size Treatment modalities Radiation doses (Gy) Dental extractions before RT Dental extractions after RT ORN after pre-RT ORN after post-RT
dental extractions dental extractions
Kubota et al. (2021) 616 Pre-RT surgery = 238 Median of 68.5 223 49 21 12
Pre-RT mandible surgery = 51 (48.4–77.2)
Induction CT = 53
Concurrent CT = 444

Liao et al. (2021) 16,701 Combined CT = 10,959 ≥ 60 6998 5090 415 692
Pre-RT mandible surgery = 1122
Makkonen et al. 224 RT = 82 24–85 10 25 0 0
(1987) RT + CT = 49
RT + surgery = 93
RT + CT + surgery = 32
Muraki et al. (2019) 67 RT + CT = 52 < 60 = 3 39 4 3 1
RT + surgery = 24 > 60 = 64
Supportive Care in Cancer (2022) 30:8745–8759

Murray et al. (1980) 404 RT = 404 0–20 = 14 132 8 24 7

20–40 = 21
40–60 = 263
60–70 = 250
70–80 = 185
> 80 = 14
Oh et al 81 RT = 29 > 50 = 29 55 5 1 1
(2004) RT + surgery = 41 51 to 60 = 24
RT + CT = 12 61 to 70 = 21
RT + surgery + CT = 11 < 70 = 19
Owosho et al. (2017) 1023 RT = 1023 Tumor area: 66–70 2 6 2 6
Regions of elective nodal
RT: 50–60
Uninvolved low anterior
neck: 45–50
Postoperative RT: 60
Regezi et al. (1976) 130 RT + surgery = 29 Mean of 65 49 10 1 0
RT + CT = 6 Malignant lymphomas: mean
of 40
Reuther et al. (2003) 830 RT = 4 Median of 60 16 18 16 18
RT + CT = 5 (49–90)
RT + CT + surgery = 9
Surgery + RT = 22
Surgery + RT + CT = 28
See Toh et al. (2018) 231 RT = 63 66–70 207 16 4 0
CT + RT = 168
Sulaiman et al. (2003) 187 CT + RT = 49 < 50 = 11 97 127 2 2
RT = 137 50–59 = 29
Surgery = 122 60–69 = 87
> 70 = 46
Not reported = 12
Not started = 2

13
8751
 Table 2  (continued)
8752

Author (year) Sample size Treatment modalities Radiation doses (Gy) Dental extractions before RT Dental extractions after RT ORN after pre-RT ORN after post-RT
dental extractions dental extractions

13
Thorn et al. (2000) 80 RT = 80 45–48 = 2 8 36 8 36
57 = 1
60–62 = 3
64 = 22
66 = 34
68 = 18

Wang et al. (2017) 25,246 RT = 25,246 NR 24,552 5783 1898 320

Surgery = 17,554
Chemotherapy = 19,752
Wanifuchi et al. 33 CT = 25 < 50 = 2 4 6 4 6
(2016) 50–60 = 2
60–70 = 13
> 70 = 16
Widmark et al. (1989) 431 ORN: ORN: mean of 64.3 103 50 32 13
CT Pre-RT = 16 No ORN: mean of 55.0
CT + RT = 7
No ORN:
CT Pre-RT = 20
CT + RT = 12
Willaert et al. (2019) 238 CT + RT = 192 3D-CRT = mean of 72 116 59 4 10
(50–72)
IMRT = mean of 65
(60–72)

Abbreviations: 3D-CRT​, three-dimensional conformal radiotherapy; BRT, brachytherapy; CRT​, chemoradiotherapy; CT, chemotherapy; EGFRI, epidermal growth factor receptor inhibitor;
IMRT, intensity-modulated radiation therapy; NR, not reported; ORN, osteoradionecrosis; RI, radioactive Implant; RT, radiotherapy
Supportive Care in Cancer (2022) 30:8745–8759
Supportive Care in Cancer (2022) 30:8745–8759 8753

Fig. 2  Risk of bias sum-

mary: Reviewers’ judgments
about each checklist item are
presented as percentages across
cross-sectional studies

studies did not clearly define the inclusion and exclusion a significantly higher risk for earlier ORN occurrence than
criteria. Most studies (57.2%) did not use an appropriate sta- those who underwent pre-RT extractions, with hazard ratios
tistical analysis since association analyses, such as odds ratio (HRs) ranging from 1.68 (p = 0.018) to 3.87 (p = 0.025) [9,
analysis, were not performed in the majority of the studies. 32, 40], whereas Epstein et al. [1], Koga et al., [1], and Wani-
In contrast, the detailed description of the subjects and set- fuchi et al. [41] found a higher prevalence of ORN among
ting and the identification of confounding factors were well patients who underwent post-RT extraction, although statis-
carried out by more than 85% of the studies. The assessment tical association tests were not performed, thereby showing
of risk of bias in cross-sectional studies is summarized in how divergent the results of the studies that were evaluated
Fig. 2 and described in detail in Appendix 3. in the present analysis are.
Retrospective studies assessing general potential risk
Results of individual studies factors for the occurrence and severity of ORN were also
included in this systematic review since tooth extractions
There were two types of included studies: those that specifi- were among the risk factors. Caparrotti et al. [14] found
cally assessed tooth extraction as the main risk factor for that pre-RT dental extractions had a statistically significant
ORN and those that screened all potential risk factors for association with ORN (p = 0.045). This result, however,
ORN, including tooth extractions. The studies focused on diverged from the findings of recently published studies
tooth extraction provided more detailed information regard- that demonstrated a significantly increased risk of ORN in
ing the extraction sites and the interval between extractions association with post-RT tooth extractions, with HRs that
and ORN or RT. Despite the homogeneity in terms of the ranged from 2.63 (p = 0.0045) [33] to 3.60 (p = 0.004) [16].
HNC patients who underwent RT, contrasting results were Other studies identified dental extractions as the main cause
found regarding when tooth extractions were performed. of ORN and found an increased prevalence of ORN among
Beech et al. (2017) [9] and Chang et al. (2007) [13] found patients who underwent post-RT extractions compared to
a significant association between the development of ORN those who underwent pre-RT extractions [8, 26, 30, 39].
and pre-RT extractions (p = 0.001 and p = 0.0004, respec- Chopra et al. (2011) [27] also found a higher prevalence
tively), thereby suggesting that pre-RT dental extractions of stage III ORN among patients who underwent post-RT
do not seem to reduce the risk of ORN. Similarly, Sulaiman extractions (60%), although the difference was not statisti-
et al. (2003) [18] performed descriptive frequency analysis cally significant (p = 0.13). In contrast, some studies found
and found that a higher prevalence of ORN was associated that all or nearly all patients who developed ORN underwent
with pre-RT extractions, which indicates that pre-RT extrac- tooth extraction, but no conclusions could be made in terms
tions may present a greater risk than post-RT extractions, of the period of extraction, as no differences were observed
although the assumption was based on small samples and for this outcome [25, 28, 34, 37].
low prevalence numbers. In agreement, other studies did
not find a difference in ORN development between tooth Results of syntheses
extraction before and after RT, thereby demonstrating that
the risk of ORN is present regardless of when the extractions Due to the heterogeneity of the results of the qualitative
are performed, which precludes the possibility of determin- synthesis, a quantitative analysis was conducted to aid in
ing that the best time to extract teeth is prior to the start of answering the question of interest in this meta-analysis,
RT [15, 17, 19, 35, 36, 38]. In contrast, recent studies have which was not posed by individual studies, and to settle
shown that patients who underwent post-RT extractions had controversies that arose from the conflicting results.

13
8754 Supportive Care in Cancer (2022) 30:8745–8759

Fig. 3  Forest plot of studies that assessed the proportion of osteora- The analysis shows a significant association between post-RT tooth
dionecrosis (ORN) in irradiated head and neck cancer patients who extraction and ORN (p = 0.01). Abbreviations: CI, confidence inter-
underwent tooth extractions before and after radiotherapy (RT). val; and M-H, Mantel–Haenszel

Fig. 4  Forest plot of studies that assessed the frequency of tooth (ORN). The analysis shows a lack of a significant association between
extractions before and after radiotherapy (RT) in irradiated head and post-RT or pre-RT tooth extraction and ORN (p = 0.08). Abbrevia-
neck cancer patients who were diagnosed with osteoradionecrosis tions: CI, confidence interval; and M-H, Mantel–Haenszel

Seventeen studies that assessed HNC patients who under- a significant association of post-RT dental extractions with
went dental extractions either before or after RT and that the risk of ORN development (OR: 1.98; 95% CI: 1.17–3.35;
reported the number of ORN cases could be pooled in an p = 0.01), although considerable heterogeneity was observed
odds-ratio meta-analysis (Fig. 3). A total of 33,522 patients (I2 = 93%). Studies that only included HNC patients who
underwent extractions before RT, while 11,466 patients were diagnosed with ORN were also pooled in an associa-
underwent dental extractions after RT. The analysis revealed tion meta-analysis, as they reported the numbers of patients

13
Supportive Care in Cancer (2022) 30:8745–8759 8755

Fig. 5  Forest plots of studies that assessed the proportion of osteo- between pre-RT or post-RT tooth extraction and ORN compared to
radionecrosis (ORN) in irradiated head and neck cancer patients extractions that were performed during RT (p = 0.66 and p = 0.55,
who underwent tooth extractions before (A) and after (B) radiother- respectively). Abbreviations: CI, confidence interval; and M-H, Man-
apy (RT) compared to the proportion in those who underwent den- tel–Haenszel
tal extractions during RT. The analysis shows a lack of association

who had undergone extractions before and after RT (Fig. 4). to serious inconsistencies, which downgraded the certainty
This analysis of eleven studies demonstrated no association of the evidence, thereby suggesting that further research may
between ORN risk and tooth extraction either before or after have an important impact on the confidence in the estimate of
RT (OR: 1.70; 95% CI: 0.94–3.07; I2 = 76%; p = 0.08). How- the effect and may change it. For the outcome assessing the
ever, in a sensitivity analysis from which studies with a high numbers of patients who underwent dental extractions before
risk of bias were excluded, an association between post-RT and after RT and developed ORN, low certainty of evidence
extraction and ORN risk was observed (OR: 2.02; 95% CI: was demonstrated. This result also relies on the inconsist-
1.02–3.98; I2 = 78%; p = 0.04) (Appendix 4). Thus, regard- ency of the analysis, in addition to the lack of association
less of the overall population that was included in the study, that was revealed in the meta-analysis. Thus, further research
namely HNC patients in general or only ORN patients, tooth will likely have an important impact on the confidence in the
extraction that is performed after RT was found to be asso- estimate of the effect.
ciated with a higher risk of ORN. Tooth extractions during
RT were reported by only five studies, and no increased risk
of ORN was associated with extractions either before (OR: Discussion
0.73; 95% CI: 0.18–2.96; I2 = 49%; p = 0.66) or after RT
(OR: 1.40; 95% CI: 0.47–4.16; I2 = 29%; p = 0.55) (Fig. 5). The present study evaluated the current literature that
addresses the risk of ORN development with tooth extrac-
Certainty of evidence tions that are performed before, during, or after head and
neck RT. Our analysis demonstrated that this issue is still
The certainty of the evidence for the outcomes that were controversial, which suggests that the timing of dental
assessed by the meta-analysis was analyzed using the GRADE extractions in relation to the start of RT may not be the
system by dividing the outcomes according to the main study most important risk factor, as has been accepted univer-
sample. For the first outcome, in which the numbers of ORN sally. A recently published systematic review also aimed
cases in HNC patients who underwent dental extractions to ascertain if there is a safe period for performing dental
before and after RT were assessed, moderate certainty of evi- extractions to reduce ORN incidence, although they could
dence was observed (Appendix 5). This result was due mainly not define an optimal time frame due to lack of information

13
8756 Supportive Care in Cancer (2022) 30:8745–8759

and detailed reports about the timing of extractions relative usually ranging from 2 to 15%.[44]. Dental extractions are
to RT [10]. Therefore, we completed a systematic review generally recommended prior to RT, although it has already
of studies that included groups of patients who underwent been confirmed that dental extractions before RT may also
pre- or post-RT dental extractions, or even dental extractions represent a risk factor for ORN [45]. However, dental extrac-
during RT, which enabled an association meta-analyses that tions before RT have been associated with a reduced risk of
could possibly resolve the divergences. ORN compared to extractions during or after head and neck
Among all included studies in this review, heterogene- radiotherapy treatment [29, 32, 40]. The conflicting findings
ity regarding the global sample was observed. While some led us to explore the available literature regarding recom-
papers included patients with HNC and assessed the preva- mendations to extract or to not extract teeth before head and
lence of ORN, other studies included only patients who were neck radiotherapy.
diagnosed with ORN. Additionally, most studies focused The most suitable study design for answering this ques-
specifically on the possible association of tooth extraction tion would be a randomized clinical trial that allocates HNC
with ORN development, while other studies assessed overall patients to groups in which dental extraction is performed
risk factors, including tooth extraction. This heterogeneity prior to, during, or after RT and determines the frequency
regarding sample and analysis methods led to the perfor- of ORN among the groups. However, such a study design
mance of two association meta-analyses and the assessment would be ethically challenging and require a multicenter
of the certainty of evidence for two distinct outcomes. Fur- design with many patients and sufficient follow-up, which
thermore, the studies that assessed general risk factors did may explain the lack of studies that address dental extrac-
not provide precise descriptions of the extraction site or the tions in the pre-radiation setting. A Cochrane review did not
interval between extractions and radiotherapy or ORN devel- find randomized clinical trials that compared tooth extrac-
opment, as these factors were not the focus of the studies tion prior to RT with leaving teeth in place during RT [46].
but were included due to reports of dental extraction data. Therefore, the currently available evidence for answering
Thus, considering that tooth extraction is one of the most this question is based on observational studies and mainly
important risk factors for the development of ORN, studies cross-sectional studies, which enabled the performance of
should better detail all the characteristics that are related to this association meta-analysis. The meta-analyses pooled the
tooth extraction, including which teeth were extracted, the data of the included studies; only one study was excluded
time interval between extraction and initiation or termina- because quantitative assessment was not possible due to
tion of radiotherapy, the time of ORN onset in relation to the absence of ORN cases in the sample [35]. From these
extraction and RT, and whether the ORN occurred at the results, we were able to confirm through a meta-analytical
extraction site. approach that tooth extractions prior to radiotherapy treat-
Although most studies identified confounding factors, ment result in a lower risk of developing ORN, with a mod-
many did not report strategies for dealing with these fac- erate certainty of evidence.
tors, which directly impacted the risk of bias assessment. It It is of paramount importance to determine not only the
is imperative not only to identify such confounding factors best timing for performing dental extractions in relation to
but also to address them by adjusting the sample to mini- RT but also to establish a decision-making process for these
mize its effects. Among these factors, chronic use of tobacco extractions. Several survey-based studies have been per-
and alcohol, tumor site, tumor diagnosis, clinical stage, and formed to examine the major indications for tooth extraction
comorbidities have been associated with an increased risk prior to head and neck RT and the fundamental reasons for
of ORN and may impact the exact determination of whether the decisions that were made by dentists regarding whether
the ORN occurred due to tooth extraction [14, 28, 37]. Thus, to extract or not extract a tooth [47–49]. The findings dem-
further studies are needed to better identify all confounding onstrate that pre-RT dental extraction decisions may be
factors that could impact ORN development. based mainly on clinical experience and opinions rather than
Patients who develop ORN may experience a signifi- on evidence-based clinical guidelines, thereby lacking solid
cant reduction in quality of life due to pain, dysphagia, and scientific support [48]. Additionally, substantial heterogene-
changes in diet, nutrition, speech, and appearance [42]. Fur- ity in terms of oral health care protocols has been observed
thermore, the overall treatment cost for ORN patients is con- regarding the management of irradiated patients within dif-
siderably higher than that for non-ORN patients [43]. There- ferent hospitals [47]. Therefore, the need to standardize the
fore, it is extremely important that preventive measures for conditioning protocols for patients who will undergo head
ORN be implemented throughout the cancer treatment con- and neck radiotherapy is evident. In this sense, this system-
tinuum. Hence, meticulous dental care is usually performed atic review intends to facilitate decision-making by confirm-
before, during, and after head and neck radiation therapy, as ing that teeth in need of extraction should be extracted prior
dental extractions are reported to be among the most com- to RT to reduce the risk of ORN.
mon risk factors for ORN development, with prevalence

13
Supportive Care in Cancer (2022) 30:8745–8759 8757

The interval between tooth extraction and ORN devel- Other information
opment was reported by only one study [11]; thus, it was
difficult to determine the most suitable period of healing fol- Protocol and registration
lowing surgery and beginning RT. Additionally, the interval
between tooth extraction and commencement of radiother- The methods of this systematic review were established
apy was reported by only 13 studies, and the values varied before starting the review, and the resulting protocol was
widely between patients and among studies, ranging from based on PRISMA-P [51, 53], which was registered into
3 to 210 days, with most studies reporting at least 1 week the International Prospective Register of Systematic
before RT with a mean of 24.7 days. Hence, concern with Reviews (PROSPERO) database under registration number
an adequate healing time prior to the start of radiotherapy CRD42021241631 [54]. Additionally, the present systematic
is evident, although this information is still underreported. review was reported according to the Preferred Reporting
According to the National Comprehensive Cancer Network Items for Systematic Reviews and Meta-Analyses (PRISMA)
(NCCN) guidelines, pre-RT dental extractions should be checklist [55].
completed at least 2 weeks prior to the start of RT, although
this timing remains controversial, as tooth extraction within Supplementary Information The online version contains supplemen-
7 days before RT has been reported not to increase the ORN tary material available at [Link] rg/​10.​1007/s​ 00520-​022-0​ 7215-y.
risk [50, 51]. Thus, more well-designed studies should be
Acknowledgements The authors would like to gratefully acknowledge
carried out to determine whether prolonging the wound heal- the Oral Care Study Group of the Multinational Association of Sup-
ing time is advisable, as it may compromise oncological portive Care in Cancer (MASCC) for the careful manuscript review,
treatment and prognosis, or may be beneficial for preventing especially Dr. Derek K. Smith from the Vanderbilt University Medical
ORN. Center, Nashville TN, USA, Dr. Cherry L. Estilo from the Memorial
Sloan Kettering Cancer Center, New York, NY, USA, and Dr. Erin
Watson from Princess Margaret Cancer Centre, Toronto, ON, Canada.
Also, the authors would like to thank the São Paulo State Research
Limitations Foundation (FAPESP 2019/26676-7, 2019/09692-9, and 2018/02233-
6) and the Brazilian National Council for Scientific and Technological
Development (CNPq) for Dr. Alan R. Santos-Silva funding.
Various limitations of the studies that were included in the Consortium data
present systematic review were identified. First, the data Consortium name
were described in miscellaneous ways, especially the data Oral Care Study Group of the Multinational Association of Sup-
on treatment modalities and radiation doses and the statisti- portive Care in Cancer (MASCC). Website: [Link]
oral-​care
cal analysis data. Only seven included studies performed Contact details (consortia representative)
univariate and multivariate Cox regression and reported data Dr. Derek K. Smith, Vanderbilt University Medical Center, Nash-
on hazard ratios, which impacted the qualitative association ville TN, USA
synthesis. Future studies could use standardized recording 2525 West End Avenue. Room/Suite 1141, Nashville, TN, 37203
Telephone: (615) 322-1895
forms to extract data regarding RT, dental extractions, and E-mail: [Link]@[Link]
ORN to minimize this heterogeneity. Additionally, in some Consortium members
studies, it was not possible to determine if ORN developed Cherry L. ­Estilo1, David Y
­ ang2, Deborah S­ aunders3, Erin ­Watson4
1
in the extraction field, as this information was not well Memorial Sloan Kettering Cancer Center, New York, NY, USA;
2
Provincial Practice Leader in Oral and Maxillofacial Surgery for the
reported. Finally, various insights regarding treatments and BC Cancer Agency, Canada; 3Health Sciences North, Northeast Cancer
prevention strategies such as photobiomodulation could also Centre, Sudbury, Ontario, Canada and the Northern Ontario School
be investigated in association with dental extractions [52]. of Medicine, Sudbury, Ontario, Canada; 4Princess Margaret Cancer
Centre, Toronto, ON, Canada
Consortium additional details
The members of the consortium previously requested to not have
Conclusions author status; therefore, the group members were not included in the
author’s list. The authors thanked the consortium for its contribution
Despite the heterogeneity of the findings, extracting compro- in the acknowledgements section.
mised teeth prior to head and neck radiotherapy is associated
Author contribution All authors contributed to the study conception
with a lower risk of developing ORN compared to perform- and design. Study design was performed by MAL, ACR, TBB, ACPR,
ing dental extractions after or during RT, with a moderate and LAMG. Material preparation was performed by JBE and CAM,
certainty of evidence. data collection was performed by AGCN and MEP, and analysis was
performed by AGCN and ENSG. The first draft of the manuscript was
written by AGCN and ARSS and all authors commented on previous
versions of the manuscript. All authors read and approved the final
manuscript.

13
8758 Supportive Care in Cancer (2022) 30:8745–8759

Funding This work was supported by Sao Paulo Research Foundation 9. Beech NM, Porceddu S, Batstone MD (2017) Radiotherapy-
(FAPESP, Brazil) processes number 2019/26676–7 and 2019/09692–9 associated dental extractions and osteoradionecrosis. Head Neck
(PhD scholarships granted to Ana Gabriela C. Normando), and 39:128–132. [Link]
2018/02233–6 (research grants given to Dr. Alan Roger Santos-Silva). 10. Beaumont S, Bhatia N, McDowell L et al (2021) Timing of dental
The Brazilian National Council for Scientific and Technological Devel- extractions in patients undergoing radiotherapy and the incidence
opment (CNPq, Brazil) supports Dr. Alan R. Santos-Silva, Eliete of osteoradionecrosis: a systematic review and meta-analysis. Br
Guerra, and Marcio Lopes as research grantees. None of the authors J Oral Maxillofac Surg 59:511–523
received research support from companies or similar organizations that 11. Epstein JB, Rea G, Wong FLW et al (1987) Osteonecrosis: study of
may gain or lose financially through publication of this manuscript. the relationship of dental extractions in patients receiving radiotherapy.
Data availability Head Neck Surg 10:48–54. [Link]
All authors confirm that all data and materials as well as software 12. Murray HJ, Daly TE, Zimmerman S (1980) Radiation necrosis of
application or custom code support their published claims and comply the mandible: a 10 year study. Part II. Dental factors; onset, dura-
with field standards. The authors are aware of the journal’s individual tion and management of necrosis. Int J Radiat Oncol Biol Phys
policies on sharing research data in concordance with disciplinary 6:549–553. [Link]
norms and expectations. 13. Chang DT, Sandow PR, Morris CG et al (2007) Do pre-irradiation
dental extractions reduce the risk of osteoradionecrosis of the man-
Code availability The authors state that no codes or algorithms were dible? Head Neck 29:528–536. [Link] 0538
used in the manuscript under consideration. 14. Caparrotti F, Huang SH, Lu L et al (2017) Osteoradionecrosis of
the mandible in patients with oropharyngeal carcinoma treated
with intensity-modulated radiotherapy. Cancer 123:3691–3700.
Declarations [Link]
15. Widmark G, Sagne S, Heikel P (1989) Osteonecrosis of the jaws.
Ethics approval This is a systematic review of the literature. The Int J Oral Maxillofac Surg 18:302–306
Research Ethics Committee from Piracicaba Dental School, University 16. Willaert R, Nevens D, Laenen A et al (2019) Does intensity-mod-
of Campinas, Brazil, has confirmed that no ethical approval is required. ulated radiation therapy lower the risk of osteoradionecrosis of the
jaw? A long-term comparative analysis. Int J Oral Maxillofac Surg
Consent to participate This is a systematic review of the literature; 48:1387–1393. [Link]
therefore, no human subjects were included, which justifies the absence 17. See Toh YL, Soong YL, Chim YX et al (2018) Dental extractions
of informed consent to patients to participate in the study. for preradiation dental clearance and incidence of osteoradione-
crosis in patients with nasopharyngeal carcinoma treated with
Consent for publication This is a systematic review of the literature; intensity-modulated radiotherapy. J Investig Clin Dent 9:e12295.
therefore, no human subjects were included, which justifies the absence [Link]
of a consent for publication. 18. Sulaiman F, Huryn JM, Zlotolow IM (2003) Dental extractions
in the irradiated head and neck patient: a retrospective analysis
of memorial sloan-kettering cancer center protocols, criteria, and
Conflict of interest The authors declare no competing interests.
end results. J Oral Maxillofac Surg 61:1123–1131. [Link] oi.o​ rg/​
10.​1016/​S0278-​2391(03)​00669-4
19. Oh H, Chambers MS, Garden AS et al (2004) Risk of osteoradi-
References onecrosis after extraction of impacted third molars in irradiated
head and neck cancer patients. J Oral Maxillofac Surg 62:139–
144. [Link]
1. Chow LQM (2020) Head and neck cancer. N Engl J Med 382:60–
20. Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A (2016)
72. [Link]
Rayyan-a web and mobile app for systematic reviews. Syst
2. Bray F, Ferlay J, Soerjomataram I et al (2018) Global cancer sta-
Rev 5(1):210. [Link]
tistics 2018: GLOBOCAN estimates of incidence and mortality
21. Moola S, Munn Z, Tufanaru C, Aromataris E, Sears K, Sfetcu R,
worldwide for 36 cancers in 185 countries. CA: A Cancer J Clini-
Currie M, Lisy K, Qureshi R, Mattis P, Mu P (2020) Chapter 7: Sys-
cians 68:394–424. [Link]
tematic reviews of etiology and risk. In: Aromataris E, Munn Z (eds)
3. El-Naggar AK, Chan JKC, Grandis JR, Takata T, Slootweg PJ
JBI manual for evidence synthesis. JBI. Available from [Link]
(2017) World Health Organization classification of head and neck
esism​anual.​jbi.​global. [Link]
tumours, 4th edn. IARC, Lyon, pp 60–61, 126–127
22. Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page
4. Alterio D, Marvaso G, Ferrari A et al (2019) Modern radiotherapy
MJ, Welch VA (eds) (2022) Cochrane Handbook for Systematic
for head and neck cancer. Semin Oncol 46:233–245
Reviews of Interventions version 6.3 (updated February 2022).
5. Chronopoulos A, Zarra T, Ehrenfeld M, Otto S (2018) Osteoradi-
Cochrane. Available from www.​train​ing.​cochr​ane.​org/​handb​ook
onecrosis of the jaws: definition, epidemiology, staging and clinical
23. Schünemann H, Brożek J, Guyatt G, Oxman A (eds) (2013)
and radiological findings. A concise review. Int Dental J 68:22–30
GRADE handbook for grading quality of evidence and strength
6. Oh H, Chambers M, Martin J (2009) Osteoradionecrosis of the
of recommendations. Updated October 2013. The GRADE Work-
mandible: treatment outcomes and factors influencing the progress
ing Group. Available from guide​lined​evelo​pment.​org/​handb​ook
of osteoradionecrosis. J Oral Maxillofac Surg 67:1378–1386
24. GRADEpro GDT (2022) GRADEpro Guideline Development
7. Jacobson BD, Hu K, Urken ML (2010) Paradigm shifts in the man-
Tool [Software]. McMaster University and Evidence Prime.
agement of osteoradionecrosis of the mandible. Oral Oncol 46:795–
Available from grade​pro.​org
801. [Link]
25 Acharya S, Pai KM, Acharya S (2020) Risk assessment for osteo-
8. Reuther T, Schuster T, Mende U, Kübler AC (2003) Osteoradione-
radionecrosis of the jaws in patients with head and neck cancer.
crosis of the jaws as a side effect of radiotherapy of head and neck
Med Pharm Rep 93:195–199. [Link]
tumour patients - a report of a thirty year retrospective review. Int
26. Beumer J, Harrison R, Sanders B, Kurrasch M (1984) Osteora-
J Oral Maxillofac Surg 32:289–295
dionecrosis: predisposing factors and outcomes of therapy. Head
Neck Surg 6:819–827. [Link]

13
Supportive Care in Cancer (2022) 30:8745–8759 8759

27. Chopra S, Kamdar D, Tulunay Ugur OE et al (2011) Factors pre- of the mandible. Br J Oral Maxillofac Surg 53:854–857. [Link]
dictive of severity of osteoradionecrosis of the mandible. Head doi.​org/​10.​1016/j.​bjoms.​2015.​07.​008
Neck 33:1600–1605. [Link] 43. Park L, Lilic N, Addison B, Patel R (2017) Cost analysis of osteo-
28. Chronopoulos A, Zarra T, Tröltzsch M et al (2015) Osteoradione- radionecrosis. J Laryngol Otol 131:303–308. [Link]
crosis of the mandible: a ten year single-center retrospective study. 1017/​S0022​21511​60099​56
J Cranio-Maxillofac Surg 43:837–846. [Link] 44. Ben-David M, Diamante M, Radawski J et al (2007) Lack of osteo-
jcms.​2015.​03.​024 radionecrosis of the mandible after intensity-modulated radiother-
29. Dumoulin S, van Maanen A, Magremanne M (2021) Dental pre- apy for head and neck cancer: likely contributions of both dental
vention of maxillo-mandibular osteoradionecrosis: a ten-year ret- care and improved dose distributions. Int J Radiat Oncol Biol Phys
rospective study. J Stomatol Oral Maxillofac Surg 122:127–134. 68:396–402. [Link]
[Link] 45. Lajolo C, Gioco G, Rupe C, Troiano G, Cordaro M, Lucchese
30. Habib S, Sassoon I, Thompson I, Patel V (2021) Risk factors A, Paludetti G, Giuliani M (2021) Tooth extraction before radio-
associated with osteoradionecrosis. Oral Surg 14:227–235. [Link] therapy is a risk factor for developing osteoradionecrosis of the
doi.​org/​10.​1111/​ors.​12597 jaws: a systematic review. Oral Dis 27(7):1595–1605. [Link]
31. Koga DH, Salvajoli JV, Kowalski LP, Nishimoto IN, Alves FA org/​10.​1111/​odi.​13485
(2008) Dental extractions related to head and neck radiotherapy: 46. Eliyas S, Al-Khayatt A, Porter RW, Briggs P (2013) Dental
ten-year experience of a single institution. Oral Surg Oral Med extractions prior to radiotherapy to the jaws for reducing post-
Oral Pathol Oral Radiol Endod 105(5):e1-6. [Link] radiotherapy dental complications. Cochrane Database Syst
1016/j.​tripl​eo.​2008.​01.​007 Rev 2013(2):CD008857. [Link] oi.​o rg/​1 0.​1 002/​1 4651​8 58.​
32. Kojima Y, Yanamoto S, Umeda M et al (2017) Relationship CD008​857.​pub2
between dental status and development of osteoradionecrosis of 47. Lanzós I, Herrera D, Lanzós E, Sanz M (2015) A critical assess-
the jaw: a multicenter retrospective study. Oral Surg Oral Med ment of oral care protocols for patients under radiation therapy
Oral Pathol Oral Radiol 124:139–145. [Link] in the regional University Hospital Network of Madrid (Spain). J
oooo.​2017.​04.​012 Clin Exp Dent 7:613–634. [Link]
33. Kubota H, Miyawaki D, Mukumoto N, Ishihara T, Matsumura M, 48. Bruins HH, Jolly DE, Koole R (1999) Preradiation dental extrac-
Hasegawa T, Akashi M, Kiyota N, Shinomiya H, Teshima M, Nibu tion decisions in patients with head and neck cancer. Oral Surg
KI, Sasaki R (2021) Risk factors for osteoradionecrosis of the jaw Oral Med Oral Pathol Oral Radiol Endod 88(4):406–412. [Link]
in patients with head and neck squamous cell carcinoma. Radiat doi.​org/​10.​1016/​s1079-​2104(99)​70053-3
Oncol 16(1):1. [Link] 49. Watson E, Dorna Mojdami Z, Oladega A, Hope A, Glogauer M;
34. Liao PH, Chu CH, Hung YM et al (2021) Tumor subsites and Canadian Dental Oncology Network Consensus Group (2021)
risk of osteoradionecrosis of the jaw in patients with oral cavity Clinical practice guidelines for dental management prior to radia-
cancer: a national-based cohort study. Eur Arch Otorhinolaryngol. tion for head and neck cancer. Oral Oncol 123:105604. [Link] oi.​
[Link] org/​10.​1016/j.​oralo​ncolo​gy.​2021.​105604
35. Makkonen TA, Kiminki A, Makkonen TK, Nordman E (1987) 50. Liao PH, Chu CH, Tang PL et al (2020) Preradiation tooth extrac-
Dental extractions in relation to radiation therapy of 224 patients. tion and jaw osteoradionecrosis: nationwide population-based ret-
Int J Oral Maxillofac Surg 16(1):56–64. [Link] rospective study in Taiwan. Clin Otolaryngol 45:896–903. [Link]
s0901-​5027(87)​80031-0 doi.​org/​10.​1111/​coa.​13624
36. Muraki Y, Akashi M, Ejima Y et al (2019) Dental interven- 51. National Comprehensive Cancer Network (2021) Head and neck
tion against osteoradionecrosis of the jaws in irradiated patients cancers (Version 3.2021). [Link]
with head and neck malignancy: a single-arm prospective study. physi​cian_​gls/​pdf/​head-​and-​neck.​pdf. Accessed 25 Aug 2021
Oral Maxillofac Surg 23:297–305. [Link] oi.​o rg/​1 0.​1 007/​ 52. da Silva TMV, Melo TS, de Alencar RC et al (2021) Photobio-
s10006-​019-​00783-0 modulation for mucosal repair in patients submitted to dental
37. Owosho AA, Tsai CJ, Lee RS et al (2017) The prevalence and risk extraction after head and neck radiation therapy: a double-blind
factors associated with osteoradionecrosis of the jaw in oral and randomized pilot study. Supportive care in cancer : official journal
oropharyngeal cancer patients treated with intensity-modulated of the Multinational Association of Supportive Care in Cancer
radiation therapy (IMRT): The Memorial Sloan Kettering Can- 29:1347–1354. [Link]
cer Center experience. Oral Oncol 64:44–51. [Link] 53. Shamseer L, Moher D, Clarke M et al (2015) Preferred report-
1016/j.​oralo​ncolo​gy.​2016.​11.​015 ing items for systematic review and meta-analysis protocols
38. Regezi JA, Courtney RM, Kerr DA (1976) Dental management of (PRISMA-P) 2015: elaboration and explanation OPEN ACCESS.
patients irradiated for oral cancer. Cancer 38:994–1000. [Link] BMJ. [Link]
doi.​org/​10.​1002/​1097-​0142(197608)​38:2%​3c994::​AID-​CNCR2​ 54. Pérez-de-Oliveira, Normando AGC, Santos-Silva AR (2021) To
82038​0251%​3e3.0.​CO;2-B extract or not to extract teeth prior to head and neck radiotherapy?
39. Thorn JJ, Hansen HS, Specht L, Bastholt L (2000) Osteoradione- A systematic review and meta-analysis of conditioning protocols.
crosis of the jaws: clinical characteristics and relation to the field In: PROSPERO CRD42021241631 Available from: [Link]
of irradiation. J Oral Maxillofac Surg 58:1088–1093. [Link] crd.​york.​ac.​uk/​prosp​ero/​displ​ay_​record.​php?​ID=​CRD42​02124​
org/​10.​1053/​joms.​2000.​9562 1631. PROSPERO CRD42021241631 Available from: [Link]
40. Wang TH, Liu CJ, Chao TF et al (2017) Risk factors for and www.​crd.​york.​ac.​uk/​prosp​ero/​displ​ay_​record.​php?​ID=​CRD42​
the role of dental extractions in osteoradionecrosis of the jaws: a 02124​1631. Accessed 25 Aug 2021
national-based cohort study. Head Neck 39:1313–1321. [Link] 55. Page MJ, Moher D, Bossuyt PM et al (2021) PRISMA 2020
doi.​org/​10.​1002/​hed.​24761 explanation and elaboration: updated guidance and exemplars
41. Wanifuchi S, Akashi M, Ejima Y et al (2016) Cause and occur- for reporting systematic reviews. BMJ (Clinical research ed)
rence timing of osteoradionecrosis of the jaw: a retrospective 372:n160–n160. [Link]
study focusing on prophylactic tooth extraction. Oral Maxillofac
Surg 20:337–342. [Link] Publisher's note Springer Nature remains neutral with regard to
42. Rogers SN, D’Souza JJ, Lowe D, Kanatas A (2015) Longitudinal jurisdictional claims in published maps and institutional affiliations.
evaluation of health-related quality of life after osteoradionecrosis

13