Clinical Oral Investigations https://doi.org/10.1007/s00784-018-2465-3

ORIGINAL ARTICLE

The association with xerostomia from sialadenitis and the jaw osteonecrosis in head and neck cancer population: a nationwide cohort study Yi-Fang Huang 1,2 & Chih-Hsin Muo 3 & Chun-Hao Tsai 4,5 & Shih-Ping Liu 6,7,8 & Chung-Ta Chang 2,9 Received: 28 July 2017 / Accepted: 19 April 2018 # Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract Objectives Determine the association between sialadenitis and osteonecrosis of the jaw (ONJ) in head and neck cancer (HNC) patients with varying severity of sialadenitis, treatment modalities, and cancer locations. Materials and methods A total of 40,168 HNC patients, including 1907 ONJ subjects and 7559 matched comparisons, were enrolled from a Longitudinal Health Insurance Database for Catastrophic Illness Patients of Taiwan between 2000 and 2006. The association with sialadenitis and ONJ was estimated by logic regression and presented as the odds ratio (OR) and 95% confidence intervals (CIs). Results The occurrence of sialadenitis increased the risk of ONJ by 2.55-fold in HNC patients (95% CI = 2.20–2.95). The ONJ incidence was proportion to sialadenitis severity (OR = 2.53 to 4.43). Irradiated HNC patients had a higher tendency to develop jaw necrosis (osteoradionecrosis, ORN) (OR = 5.05, 95% CI = 4.39–5.80). When combined with irradiation exposure, sialadenitis significantly induced the occurrence of ORN (OR = 8.94, 95% CI = 7.40–10.8), especially in oral cancer patients (OR = 15.9 95% CI = 12.5–20.3). The risk of ONJ increased with radiotherapy dosage and duration, except for nasopharyngeal cancer (NPC) patients. Conclusions There was a close association between sialadenitis and ONJ in the HNC population. The severity of sialadenitis was positive correlated to ONJ risk. Radiotherapy combined with sialadenitis significantly raised ORN incidence in HNC patients except for NPC patients. Clinical relevance HNC patients complained that xerostomia from sialadenitis might increase the risk to develop ONJ, especially among those who received radiotherapy. Keywords Sialadenitis . Osteonecrosis of the jaw . Osteoradionecrosis . Radiotherapy . Head and neck cancer . Population-based

* Shih-Ping Liu [email protected] * Chung-Ta Chang [email protected] 1

2

Department of General Dentistry, Chang Gung Memorial Hospital, Linkou 33305, Taiwan School of Dentistry, College of Oral Medicine, Taipei Medical University, Taipei 11031, Taiwan

3

Management Office for Health Data, China Medical University Hospital, Taichung 40402, Taiwan

4

Department of Orthopedics, China Medical University Hospital, Taichung 40402, Taiwan

5

Graduate Institute of Clinical Medicine, China Medical University, Taichung 40402, Taiwan

6

Graduate Institute of Basic Medical Science, China Medical University, No. 91, Hsueh-Shih Road, Taichung City 40402, Taiwan

7

Center for Neuropsychiatry, China Medical University Hospital, Taichung 40402, Taiwan

8

Department of Social Work, Asia University, Taichung 41354, Taiwan

9

Department of Emergency Medicine, Far Eastern Memorial Hospital, No. 21, Sec. 2, Nanya S. Rd., Banciao Dist, New Taipei City 22056, Taiwan

Clin Oral Invest

Introduction

Patients and methods

Osteonecrosis of the jaw (ONJ) is defined as a painful lesion with exposed bone that fails to heal within 8 weeks after appropriate treatment [1]. ONJ often presents as an asymptomatic bone exposure without significant soft tissue infection, so this silent disease is hard to diagnose in its early stages. ONJ includes an infected necrotic maxillofacial lesion with ulcerated swollen oral mucosa and chronic sinus tracts which deteriorate the patient’s oral functions [2]. There is no superior and appropriate strategy for this intractable illness. This investigations of the complex pathogenesis of ONJ to seek for better treatments of this entity are becoming noteworthy [3]. The contributing factors related to ONJ are still a controversial issue. The exposed necrotic bone in the oral cavity is usually attributed to various pathological conditions [4]. There are two common ONJ forms including osteoradionecrosis of the jaw (ORN-radiation-related necrosis) and bisphosphonaterelated osteonecrosis of the jaws (BRONJ-jaw necrosis attributed to bisphosphonate therapy) [5]. Poor oral hygiene, dental caries, and periodontal disease are associated with ONJ [6]. Our previous research also indicated that ONJ is a multifactorial disease, and the oral condition plays significant influence on ONJ occurrence [7]. Saliva contains enzymes, mucins, proteins, and immunoglobulins as a defense role against oral cavities [8]. Insufficient salivary secretion will alter the normal oral microflora, which then causes dental plaque formation, jeopardizes intact oral mucosa, and decelerates wound healing [9]. Xerostomia is a subjective experience of mouth dryness when the salivary secretion reduces to half of the normal value [10]. Sialadenitis derived from acute or chronic salivary gland dysfunction is a major factor in the cause of xerostomia [11]. Xerostomia is the common clinical sign of ONJ and sialadenitis [12, 13]. Approximately 91% of head and neck cancer (HNC) patients complained of xerostomia during their time in the clinic [14], and ORN is the most common complication in HNC patients with radiotherapy [15]. The relationship between sialadenitis and ONJ in the HNC population had never been previously mentioned but it is worthy of investigation. Due to the complex and controversial pathogenesis of ONJ, as well as a few cases suffering both sialadenitis and ONJ, it is difficult to conduct a long-term and large-scale prospective observation to evaluate the role of sialadenitis in ONJ development. The subjects of this population-based study from the Longitudinal Health Insurance Database for Catastrophic Illness Patients (LHID-CIP) of Taiwan could overcome the limited amount of ONJ data. The aim of this study is not only to investing the association with sialadenitis and ONJ in HNC patients, but to also assess the relationship under different levels of sialadenitis severity, treatment modalities, and cancer locations.

Data source We implemented this study through the LHID-CIP of Taiwan. This database was a part of National Health Insurance Research Database released by National Health Insurance Administration Ministry of Health and Welfare (NHIAMHW), started in 1995 by the Taiwan Bureau of National Health Insurance (TBNHI); over 99.5% of Taiwan’s population has joined this program [7]. Thirty diseases have been defined as catastrophic illness, e.g., malignancies, end-stage renal disease, systemic lupus erythematous, and hemophilia (http://www. nhi.gov.tw/english/index.aspx). The catastrophic illness patients are registered in this database under the confirmation of several related diplomates based on the guidelines of NHIAMHW. The database utilized in this observation was a cohort of two million individuals that were randomly sampled TBNHI insurants in 2000. This sample has been substantiated to represent sex, age, health-care costs, and geographic distribution of the overall population of the insurants. In addition, this is an administrative database obtained by decoding the identification of beneficiaries. This study was certified under the Research Ethics Committee of China Medical University and Hospital, Taiwan. The LHID-CIP comprised all medical records (such as registration entries, prescriptions, inpatient care claims, and ambulatory care claims) from 1996 to 2011. The definition of diseases in the database was according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM).

Study subject During the period 2000 to 2006, we obtained 40,168 HNC (ICD-9-CM 140–148) patients from the LHID-CIP of Taiwan. The NHC patients with ONJ history (ONJ, ICD9-CM 526.89, 526.5, 730.0, and 730.1) were excluded (n = 460). The patients with ONJ diagnosis in at least one inpatient claim or three ambulatory claims by the diplomates of otolaryngology, plastic surgery, or dentistry were enrolled. A total of 1907 subjects with ONJ development after the NHC diagnosis were defined as the ONJ group. The index date was defined on the date of the first time ONJ diagnosis. The control group was selected from the HNC patients without ONJ diagnosis before the end of 2011. About four controls for each case were frequency matched with age (every 5-year stratum), gender, HNC type, HNC-year, and index-year. The study algorithm is shown in Fig. 1.

Clin Oral Invest Fig. 1 Flow chart of study subjects (LHID-CIP, longitudinal health insurance database for catastrophic illness patients; NHC, head and neck cancer; ONJ, osteonecrosis of the jaw)

LHID-CIP N = 20,346,484

NHC patients in 2000-2006 N = 40,168 Patients with ONJ history, N = 460

Patients without ONJ development N = 37,801

Patients with ONJ development N = 1,907 As ONJ group

Frequency matched with age, gender, HNC type, HNC-year and index-year at a ratio about 4:1 Patients without ONJ development N = 7,559 As comparison group

Risk factors and HNC types Risk factors contained several comorbidities and medical treatments. Comorbidities covered diabetes mellitus (ICD-9CM 250), hypertension (ICD-9-CM 401–405), and sialadenitis (ICD-9-CM 527.2, 527.3, 527.5–527.7, 537.9, and 710.2). Comorbidity was defined according to at least three occasions of ambulatory medical visits or a one-time inpatient record. Medical treatments encompassed oral and IV form of bisphosphonates (BPs), pharmaceuticals (namely, etidronate, clodronate, pamidronate, alendronate, tiludronate, ibandronae, risedronate, and zoledronate), and radiotherapy. Patients with BP prescription within 180 days prior to the index date were defined as the users. Patients with radiotherapy on head and neck region prior to the index date were defined as the users. Other comorbidities were defined before the index date. HNC participants were stratified into three groups based on their malignancy locations, namely oral cancer (ICD-9-CM 140–145), nasopharyngeal cancer (NPC, ICD-9-CM 147), and others (ICD-9-CM 148–149). The patients with sialadenitis that were diagnosed before the index date were defined as the sialadenitis group. Different treatment modalities were used to differentiate the severity of sialadenitis (Table 2). Those sialadenitis patients only treated with pilocarpine or cevimeline medication were defined as the mild sialadenitis population; those who received surgery includ ing sa livary d uct irrig ation, sialolithoto m y, sialoadenect omy, salivary gland catheterization,

sequestrectomy, saucerization, and resection of the jaw were regarded as the severe group.

Statistical analysis The number and proportion were presented for categorical variables, such as age group (< 50, 50–64, and 65+ years), gender, comorbidity, medical treatment, and cancer type. The mean and standard deviation (SD) shows the age distribution between the two groups. The chi-square test and Student’s t test were used to test the differences of categorical and continuous variables between the two groups. Logistic regression was used to estimate the odds ratio (OR) and the 95% confidence interval (CI) for ONJ in the sialadenitis patients compared with the non-sialadenitis participants. The association between ONJ and the severity of sialadenitis was also assessed among different HNC locations. The combined effect between sialadenitis and radiotherapy was estimated among different HNC locations. In further analyses, the joint effects for ONJ between sialadenitis and radiotherapy dosage (none, < 50 Gy, and ≥ 50 Gy) and sialadenitis and radiotherapy duration (none, < 60 days, and ≥ 60 days) were also estimated based on the study by Kuo et al. [16]. The trend test was also measured by logistic regression. SAS software version 9.4 (SAS Institute, Cary, NC) was used for these aforementioned analyses and the two-tailed test p < 0.05 was considered to be statistically significant.

Clin Oral Invest

Results All 1907 ONJ subjects and 7559 age, gender, and HNC location-matched comparisons were selected in this casecontrolled study. There were no significant differences of age and gender between these two groups. ONJ patients had a mean age of 54.8 years old (SD = 11.6) and more were men (86.8 vs. 13.2%) (Table 1). In the ONJ patients group, the HNC location was highest in oral (69.4%), followed by NPC (19.6%), and others (11.0%). ONJ patients had a higher prevalence of sialadenitis (17.2 vs. 7.54%) and radiotherapy (85.3 vs. 51.7%). More details are exhibited in Table 1. Compared to non-sialadenitis participants, sialadenitis patients presented a 2.55-fold increased risk of ONJ (95% CI = 2.20–2.95) in age and gender-adjusted logistic regression (Table 2). There were similar trends in NPC, oral cancer, and others (OR = 1.98, 3.20, and 1.67, 95% CI = 1.49–2.62, 2.65–3.88, and 1.08–2.95, respectively). The risk of ONJ increased with the severity of sialadenitis increasing from 2.53 in the mild group to 4.43 in the severe group (95% CI = 2.18– 2.93 and 1.55–12.6, p for trend < 0.0001). The combined effect of sialadenitis and radiotherapy in the process of ONJ was displayed in Table 3. Compared to HNC patients without sialadenitis and radiotherapy, patients with both sialadenitis and radiotherapy had the highest ONJ risk (OR = 8.94, 95% CI = 7.40–10.8), followed by patients with radiotherapy only (OR = 5.05, 95% CI = 4.39–5.80), and with sialadenitis only (OR = 1.35, 95% CI = 0.70–2.62). Table 4 presents the combined effects for ONJ between sialadenitis and radiotherapy dosage and sialadenitis and radiotherapy duration. The ONJ risk increased with radiotherapy dosage and radiotherapy duration, except for NPC patients. Compared to non-sialadenitis patients without radiotherapy, the highest ONJ occurrence was in sialadenitis patients with ≥ 50 Gy radiotherapy (OR = 9.43, 95% CI = 4.73–11.5), followed by sialadenitis patients with < 50 Gy radiotherapy (OR = 6.84, 95% CI = 4.71–9.94), non-sialadenitis patients with ≥ 50 Gy radiotherapy (OR = 5.77, 95% CI = 4.99– 6.68), non-sialadenitis patients with < 50 Gy radiotherapy (OR = 3.85, 95% CI = 3.25–4.55), and sialadenitis patients without radiotherapy (OR = 1.35, 95% CI = 0.69–2.61). There were similar trend between sialadenitis and radiotherapy duration.

Discussion As past studies have indicated, ONJ is a multifactorial disease, and the oral condition is closely related to ONJ risk [7]. The demographic analysis indicated that sialadenitis and radiotherapy were highly correlated to ONJ occurrence in this study (p < 0.0001) (Table 1). This result was coincident with the previous consensus that head and neck irradiation is regarded

Table 1 Distribution of age, gender, comorbidity, and medical treatment between ONJ and comparison group ONJ N = 1907

Comparison N = 7559

n

%

n

%

< 50

699

36.7

2759

36.5

50–64 65+

830 378

43.5 19.8

3317 1483

43.9 19.7

Mean (SD) Gender

54.8

(11.6)

54.7

(11.3)

Women Men

251 1656

13.2 86.8

961 6598

12.7 87.3

Comorbidity Sialadenitis

Age, year-old

p value

0.96

0.81 0.60

328

17.2

570

7.54

< 0.0001

Diabetes Hypertension Medical treatment

343 618

18.0 32.4

1320 2471

17.5 32.7

0.59 0.81

Bisphosphonates use Radiotherapy Cancer type NPC Oral Others

8 1627

0.42 85.3

58 3905

0.77 51.7

0.10 < 0.0001 0.89

374 1323 210

19.6 69.4 11.0

1490 5265 804

19.7 69.7 10.6

Chi-square test and t test

as the major risk factor of ONJ in HNC patients [17]. Although BPs are considered as to have an important role in ONJ occurrence in most research [18], it was not significantly related ONJ in the HNC population (p = 0.17). The few BRONJ cases might be attribute to BPs usage not being the major strategy for HNC therapy. The close association with sialadentitis and ONJ occurrence was the major finding of this study. Sialadenitis patients usually have the sensation of xerostomia due to reduced saliva secretion and the impaired salivary defense system in the oral cavity leading to a high incidence of several dental diseases [19]. Although this relationship has never been mentioned in the past, this result is similar to the consensus we proposed in 2015 [7]. The dental status is highly correlated to the risk of ONJ. Our results revealed HNC patients with sialadenitis increased their ONJ risk by 2.55-fold (Table 2). This discovery was a precursor in clinics, and a few studies have discussed whether the different severity of sialadenitis is correlated to ONJ risk among different HNC locations. Based on the duration of symptoms, sialadenitis is divided into acute suppurative sialadenitis and chronic or recurrent sialadenitis [20]. Most of the patients in the acute sialadenitis stage can be cured successfully with medications [21]. The chronic or recurrent sialadenitis patients are usually treated with surgical

Clin Oral Invest Table 2 Association between ONJ and sialadenitis severity in logistic regression after controlling age and gender among cancer type Sialadenitis severity

N

ONJ no.

%

OR (95% CI)

All Comparison

8568

Sialadenitis 898 Sialadenitis severity

1579

18.4

1.00

328

36.5

2.55 (2.20–2.95)***

Mild

884

321

36.3

2.53 (2.18–2.93)***

Severe p for trend

14

7

50.0

4.43 (1.55–12.6)** < 0.0001

NPC Comparison

1556

286

18.4

1.00

289

88

30.5

1.98 (1.49–2.62)***

201 0

88 0

30.5 0.00

1.98 (1.49–2.62)*** NA < 0.0001

1116 207

18.4 41.7

1.00 3.20 (2.65–3.88)***

483 13

200 7

41.4 58.9

3.16 (2.60–3.83)*** 5.22 (1.75–15.6)** < 0.0001

Comparison 916 Sialadenitis 113 Sialadenitis severity

177 33

49.3 29.2

1.00 1.67 (1.08–2.59)*

33 0

29.5 0.00

1.69 (1.09–2.62)* NA 0.027

Sialadenitis Sialadenitis severity Mild Severe p for trend Oral

Comparison 6076 Sialadenitis 496 Sialadenitis severity Mild Severe p for trend Others

Mild Severe p for trend

112 1

*p < 0.05, **p < 0.01, ***p < 0.001

interventions, such as salivary duct dilation, sialodochoplasty, and gland excision, when the initial conservative treatment has failed [22]. We took the lead to stratify the sialadenitis subjects into different severity subgroup according to surgery performance. The results revealed that the severe sialadenitis group had an increased 4.43-fold risk to develop ONJ, and the severity of sialadentitis was positive correlated to ONJ occurrence in all HNC populations (p for trend < 0.0001) (Table 2). As mentioned by Sreebny and Vissink in 2010 [23], individuals with chronic sialadenitis are particularly liable to tooth demineralization attributable to slow salivary clearance. Whether the impaired saliva buffer system leads to infectious oral and dental diseases that increase the ONJ risk in HNC patients is worth further investigation. Salivary glands are very sensitive to irradiation. The salivary function was frequently impaired after HNC patients received radiotherapy, and these patients are liable to several forms of sialadenitis and subsequent xerostomia [9]. To clarify

the joint effect of sialadenitis and irradiation on ONJ occurrence, the ONJ population was divided into four groups according to their sialadenitis and irradiation exposure status. The necrotic jaw derived from irradiation so-called ORN is the common complication of HNC patients with radiotherapy [24]. Compared to HNC patients without radiotherapy, the irradiation exposure increases the 5.05-fold risk to an 8.94fold risk of developing ORN (Table 3). However, past studies rarely mentioned the effect of variable irradiation exposure region on the occurrence of ORN. Stratifying the HNC population into different groups based on their HNC locations was another strength of this study. Although radiotherapy is usually regarded as the major factor to contribute the necrotic jaw in the previous study [25], these results indicated that 7.06% of HNC patients that did not receive radiotherapy still suffered from ONJ. Based on the previous study, ONJ caused by a number of factors including oral status and malignancy played a significant role in ONJ occurrence [7, 26], which could explain the possibility that HNC patients still have tendency to develop ONJ even without radiotherapy. Moreover, compared to those patients without radiotherapy, the irradiation exposure did not increase the jaw necrosis risk in NPC population. However, this study also revealed the variable ratio from 17.1 to 60.0% for developing jaw necrosis in the NPC population, and the potential pathologic factors of ONJ is worth further investigation. Based on these results, we might be able to deduce that the radiotherapy is not the unique risk factor in developing ONJ. With the exception of NPC patients, the sialadenitis occurrence when combined with radiotherapy significantly raises the risk of ORN in the HNC population (p < 0.0001). The irradiation exposure in NPC patients is usually away from the oral cavity, so there is no significant association with sialadenitis, radiotherapy, and ONJ. In addition, the noteworthy increasing risk of ONJ occurrence was up by 15-fold among oral cancer patients with sialadenitis that received radiotherapy. It could be attributed to the major and minor salivary glands may be injured while performing wide tumor excision for oral cancer lesions [27]. These causative factors such as sialadenitis, major oral surgery, and direct radiation exposure to the malignance in the oral cavity possibly alter the amount and composition of saliva, and then the patient develops ONJ. Although sialadenitis could mildly elevate the tendency of ONJ among the non-radiotherapy population, there is no statistical significance in the odds ratio. This result reconfirmed that ONJ is a multiple pathogenetic disease. In most of the HNC population except NPC patients, if they received radiotherapy, the sialadenitis occurrence could be regarded as synthetic role to increase the ONJ risk. It is a value issue to investigate whether the bacterial related to sialadenitis could lead to ONJ occurrence in further well-controlled research. Most investigations propose that if the radiotherapy dosage used to treat HNC is over 50 Gy, the ORN risk is significantly

Clin Oral Invest Table 3 Joint effect for ONJ between sialadenitis and radiotherapy in logistic regression after controlling age and gender among cancer type

Sialadenitis

Radiotherapy

N

ONJ no.

%

OR (95% CI)

All No

No

3826

270

7.06

1.00

Yes

No

108

10

9.26

1.35 (0.70–2.62)

No Yes

Yes Yes

4742 790

1309 318

27.6 40.3

5.05 (4.39–5.80)*** 8.94 (7.40–10.8)***

< 0.0001

NPC

0.63

No Yes

No No

236 5

57 3

24.2 60.0

1.00 4.71 (0.77–28.9)

No Yes

Yes Yes

1339 284

229 85

17.1 22.7

0.65 (0.46–0.90)** 1.34 (0.91–1.98)

No

No

3337

200

5.93

1.00

Yes No Yes

No Yes Yes

97 2717 399

7 916 200

7.22 33.7 50.1

1.22 (0.56–2.66) 8.09 (6.87–9.53)*** 15.9 (12.5–20.3)***

No No Yes Yes

215 6 686 1070

13 0 164 33

6.05 0.00 23.9 30.8

1.00 NA 4.89 (2.71–8.80)*** 6.86 (3.42–13.7)***

Oral

< 0.0001

Others No Yes No Yes

p for trend

< 0.0001

**p < 0.01, ***p < 0.001

increased [28]. It has also been noted the dosage and duration are known to impair the function of salivary glands [29]. It is worthwhile to further clarify the association with the dosage and duration of radiotherapy, sialadenitis, and ONJ risk. Fifty gray was regarded as the cut-off point to distinguish the level of radiation dose in this study. Moreover, permanent salivary gland damage could occur under exceeding accumulative irradiation exposure [30, 31], so the impact of duration of radiotherapy on ORN risk is of concern in this study. These findings showed that radiotherapy would significantly increase the ORN risk in all HNC population except for NPC patients. (Table 4) That result was coincident to the common consensus that radiotherapy plays a significant influence on ORN development [32]. The innovation of this research was that the impact of the radiotherapy dose and duration on ORN risk depends on the location of the cancer. The results indicated the irradiation dose and duration were positive-correlated to ORN development in the oral cancer population regardless of whether they had sialadenitis or not. (p < 0.0001) The risk of ORN raised accompanied with the increasing dose and duration of irradiation exposure in the oral cancer cases. Moreover, compared to the similar level of irradiation dose and duration in oral cancer population, those who suffered from sialadenitis had a higher risk of developing ORN. The ORN risk of NPC cases with sialadenitis are greatly elevated when compared to those who never received radiotherapy, if the irradiation dose was exceeding 50 Gy or the duration of

radiotherapy was over 60 days. From this discovery, we might infer that the irradiation exposure area is related to cancer location, and significantly impacts the risk of ORN. In most HNC populations except for NPC patients, the dose and duration of radiotherapy is proportional to ORN, and the sialadenitis played a synergistic role on ORN occurrence. The subjects surveyed from large and nationwide population of ONJ patients increases the validity and realism of this research. It had also overcome the limited number of ONJ cases which were usually criticized in the previous research [33]. The large-scale sample size and long-term follow-up period compensate for unfavorable errors or the flaws of the healthcare declaration. This is the first report determining the association between sialadenitis and ONJ in different cancer locations among HNC patients. The positive correlation of sialadenitis severity and ONJ risk in HNC population is the preliminary discovery. We believe that both sialadenitis and radiotherapy might be the most influential factors associated with ONJ occurrence. The irradiation exposure combined with sialadenitis could greatly raise the risk of ORN in oral cancer patients. Although these results are robust, there were still many limitations that need to be overcome in the future. The real state of oral health and oral hygiene performance were influential factors of ONJ risk and are worth further consideration; dental examination including the plaque index, probing depth, caries index, and radiographic imaging, which represent oral conditions, were not available in this

Clin Oral Invest Table 4 Joint effect for ONJ between sialadenitis and radiotherapy dosage, and sialadenitis and radiotherapy duration in logistic regression after controlling age and gender among cancer type Radiotherapy

Comparison N

Sialadenitis ONJ no.

%

OR (95% CI)

N

ONJ no.

%

OR (95% CI)

Dosage All None < 50 Gy

3826 1665

270 375

7.06 22.5

1.00 3.85 (3.25–4.55)***

108 138

10 47

9.26 34.1

1.35 (0.69–2.61) 6.84 (4.71–9.94)***

≥ 50 Gy p for trend

3077

934

30.4

5.77 (4.99–6.68)*** < 0.0001

652

271

41.6

9.43 (7.73–11.5)*** < 0.0001

None < 50 Gy

236 367

57 61

24.2 16.6

1.00 0.63 (0.42–0.94)

5 31

3 2

60.0 6.45

4.70 (0.77–28.8) 0.22 (0.05–0.94)

≥ 50 Gy p for trend

972

168

17.3

0.66 (0.47–0.92) 0.047

253

83

32.8

1.53 (1.03–2.28)* 0.15

3375 1046

200 263

5.93 25.1

1.00 5.33 (4.36–6.51)***

97 90

7 40

7.22 44.4

1.20 (0.55–2.64) 12.5 (8.02–19.4)***

1671

653

39.1

10.2 (8.61–12.2)*** < 0.0001

309

160

51.8

17.0 (13.1–22.2)*** < 0.0001

215 252

13 51

6.05 20.2

1.00 3.95 (2.09–7.50)***

6 17

0 5

0.00 29.4

NA 6.59 (2.01–21.6)**

434

113

26.0

5.48 (3.00–10.0)*** < 0.0001

90

28

31.1

6.93 (3.38–14.2)*** 0.19

NPC

Oral None < 50 Gy ≥ 50 Gy p for trend Others None < 50 Gy ≥ 50 Gy p for trend

Duration All None < 60 days ≥ 60 days p for trend NPC

3826 2750 1992

270 701 608

7.06 25.5 30.5

1.00 4.53 (3.90–5.27)*** 5.82 (4.98–6.80)*** < 0.0001

108 333 457

10 105 213

9.26 31.5 46.6

1.35 (0.70–2.62) 6.09 (4.69–7.93)*** 11.6 (9.28–14.5)*** < 0.0001

None < 60 days ≥ 60 days p for trend Oral None < 60 days ≥ 60 days p for trend

236 731 608

57 116 113

24.2 15.9 18.6

1.00 0.59 (0.41–0.85)** 0.72 (0.50–1.03) 0.28

5 104 180

3 18 67

60.0 17.3 37.2

4.70 (0.77–28.9) 0.66 (0.37–1.19) 1.86 (1.22–2.84)** 0.01

3375 1621 1096

200 503 413

5.93 31.0 37.7

1.00 7.15 (5.99–8.54)*** 9.64 (7.99–11.6)*** < 0.0001

97 177 222

7 73 127

7.22 41.2 57.2

1.21 (0.55–2.65) 11.8 (7.95–15.4)*** 21.2 (15.6–28.6)*** < 0.0001

215 398 288

13 82 82

6.05 20.6 28.5

1.00 4.03 (2.19–7.43)*** 6.23 (3.36–11.6)*** < 0.0001

6 52 55

0 14 19

0.00 26.9 34.6

NA 5.66 (2.47–13.0)*** 8.16 (3.70–18.0)*** 0.11

Others None < 60 days ≥ 60 days p for trend

*p < 0.05, **p < 0.01, ***p < 0.001

investigation. Smoking and alcoholism that lead to poor oral hygiene and vasoconstriction are identified as risk factors for

ONJ. These lifestyles were not recorded in this study. Moreover, both the increasing tumor size and the tumor

Clin Oral Invest

invasion of adjacent bones will raise the ONJ incidence [34]. Unrevealed data of tumor size and cancer stage was another shortcoming in this study. Some investigators propose that tumor location is more relevant than the tumor size [35], because tumor location represents the size, shape, and irradiation of the exposure area. In this study, the tumor location that was used to stratify the population into subgroups and so could compensate for this limitation. In addition, BPs prescription within 180 days prior to the index date might limit the number BPs users, so it is probable to affect the incidence of BRONJ in this study. In the further research, we should include the different durations of BPs administration to investigate the relationship of sialadenitis and BRONJ under different dosage of BPs. Although the results showed a significant relationship between sialadenitis and ONJ, more molecular pathophysiological evidence and large-scale prospective observations are required to confirm the association between sialadenitis and ONJ in the future. In clinical practice, early identification and prevention of related risk factors to avoid ONJ are imperative. Based on our previous research, maintaining a good oral condition, including pre-radiotherapy dental prophylaxis, is able to effectively reduce the risk of ORN [15]. Since xerostomia is shown out to be one of the major symptoms after radiotherapy [36], saliva plays a vital part in teeth cleaning and keeping periodontal tissue intact [37]. The close association with sialadenitis and ONJ provides multiple clinical issues of significance. When the HNC patients prior to or after radiotherapy complain about xerostomia, clinicians have to be vigilant as the patients have high tendency to develop ONJ. Once oral cancer patients receiving radiotherapy have the clinical symptoms or signs related to sialadenitis, dentists must pay more attention to the oral hygiene care to reduce the oral infection risk that develops into ONJ. Based on this result, the severity of sialadenitis might be proportional to the risk of ONJ in HNC patients. Whether the evaluation of salivary secretory volume could be a tool to predict the risk of ONJ in HNC patients is worthy of exploring in depth. In conclusion, there was a close association with sialadenitis and ONJ in HNC patients, and sialadenitis severity was positively correlated to ONJ risk. In the most HNC populations except NPC patients, radiotherapy combined with sialadenitis significantly increased the incidence of ORN. The radiotherapy dose and duration were proportional to ORN risk in oral cancer patients, and sialadenitis had an additive effect on ORN development. Clinically, HNC patients that com pl ained about xerostomia from sialadenitis might be predicted to have higher risk in developing ONJ, especially when they have received radiotherapy.

Funding This study is supported in part by Taiwan Ministry of Health and Welfare Clinical Trial and Research Center of Excellence (MOHW105-TDU-B-212-133019), China Medical University Hospital, Academia Sinica Taiwan Biobank Stroke Biosignature Project (BM10501010037), NRPB Stroke Clinical Trial Consortium (MOST 104-2325-B-039-005), Chang Gung Memorial Hospital (CMRPG3C1791 and CMRPG3E1341), Far Eastern Memorial Hospital (FEMH-2016-C-036), Tseng-Lien Lin Foundation, Taichung, Taiwan, Taiwan Brain Disease Foundation, Taipei, Taiwan, and Katsuzo and Kiyo Aoshima Memorial Funds, Japan.

Compliance with ethical standards Conflict of interest The authors declare that they have no conflict of interest. Ethical approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the Research Ethics Committee of China Medical University and Hospital, Taiwan, and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent For this type of study, formal consent is not required.

References 1.

Khan AA, Morrison A, Hanley DA, Felsenberg D, LK MC, O'Ryan F, Reid IR, Ruggiero SL, Taguchi A, Tetradis S, Watts NB, Brandi ML, Peters E, Guise T, Eastell R, Cheung AM, Morin SN, Masri B, Cooper C, Morgan SL, Obermayer-Pietsch B, Langdahl BL, Al Dabagh R, Davison KS, Kendler DL, Sandor GK, Josse RG, Bhandari M, El Rabbany M, Pierroz DD, Sulimani R, Saunders DP, Brown JP, Compston J and International Task Force on Osteonecrosis of the J (2015) Diagnosis and management of osteonecrosis of the jaw: a systematic review and international consensus. J Bone Miner Res 30:3–23. https://doi.org/10.1002/jbmr. 2405 2. Hinchy NV, Jayaprakash V, Rossitto RA, Anders PL, Korff KC, Canallatos P, Sullivan MA (2013) Osteonecrosis of the jaw—prevention and treatment strategies for oral health professionals. Oral Oncol 49:878–886. https://doi.org/10.1016/j.oraloncology.2013. 06.008 3. Loukota RA (2012) Osteonecrosis of the jaw is becoming more common. BMJ 345:e5988; author reply e5990. https://doi.org/10. 1136/bmj.e5988 4. Gaudin E, Seidel L, Bacevic M, Rompen E, Lambert F (2015) Occurrence and risk indicators of medication-related osteonecrosis of the jaw after dental extraction: a systematic review and metaanalysis. J Clin Periodontol 42:922–932. https://doi.org/10.1111/ jcpe.12455 5. Almazrooa SA, Woo S-B (2009) Bisphosphonate and nonbisphosphonate-associated osteonecrosis of the jaw. J Am Dent Assoc 140:864–875. https://doi.org/10.14219/jada.archive. 2009.0280 6. Schiodt M, Hermund NU (2002) Management of oral disease prior to radiation therapy. Support Care Cancer 10:40–43 7. Huang YF, Chang CT, Muo CH, Tsai CH, Shen YF, Wu CZ (2015) Impact of bisphosphonate-related osteonecrosis of the jaw on osteoporotic patients after dental extraction: a population-based cohort study. PLoS One 10:e0120756. https://doi.org/10.1371/journal. pone.0120756

Clin Oral Invest 8. 9. 10.

11.

12.

13.

14.

15.

16.

17.

18.

19. 20. 21. 22. 23. 24.

Zolotukhin S (2013) Metabolic hormones in saliva: origins and functions. Oral Dis 19:219–229. https://doi.org/10.1111/odi.12015 Proctor GB (2016) The physiology of salivary secretion. Periodontol 2000 70:11–25. https://doi.org/10.1111/prd.12116 Dawes C (1987) Physiological factors affecting salivary flow rate, oral sugar clearance, and the sensation of dry mouth in man. J Dent Res 66 Spec No:648–653 Bhayani MK, Acharya V, Kongkiatkamon S, Farah S, Roberts DB, Sterba J, Chambers MS, Lai SY (2015) Sialendoscopy for patients with radioiodine-induced sialadenitis and xerostomia. Thyroid 25: 834–838. https://doi.org/10.1089/thy.2014.0572 Mandel ID, Bennick A (1983) Quantitation of human salivary acidic proline-rich proteins in oral diseases. J Dent Res 62:943–945. https://doi.org/10.1177/00220345830620090501 Vestergaard P, Schwartz K, Rejnmark L, Mosekilde L, Pinholt EM (2012) Oral bisphosphonate use increases the risk for inflammatory jaw disease: a cohort study. J Oral Maxillofac Surg 70:821–829. https://doi.org/10.1016/j.joms.2011.02.093 Epstein JB, Emerton S, Kolbinson DA, Le ND, Phillips N, Stevenson-Moore P, Osoba D (1999) Quality of life and oral function following radiotherapy for head and neck cancer. Head Neck 21:1–11 Chang CT, Liu SP, Muo CH, Tsai CH, Huang YF (2017) Dental prophylaxis and osteoradionecrosis. J Dent Res 22034516687282. https://doi.org/10.1177/0022034516687282 Kuo TJ, Leung CM, Chang HS, Wu CN, Chen WL, Chen GJ, Lai YC, Huang WC (2016) Jaw osteoradionecrosis and dental extraction after head and neck radiotherapy: a nationwide populationbased retrospective study in Taiwan. Oral Oncol 56:71–77. https://doi.org/10.1016/j.oraloncology.2016.03.005 Fellows JL, Rindal DB, Barasch A, Gullion CM, Rush W, Pihlstrom DJ, Richman J, Group DC (2011) ONJ in two dental practice-based research network regions. J Dent Res 90:433–438. https://doi.org/10.1177/0022034510387795 Mawardi H, Giro G, Kajiya M, Ohta K, Almazrooa S, Alshwaimi E, Woo SB, Nishimura I, Kawai T (2011) A role of oral bacteria in bisphosphonate-induced osteonecrosis of the jaw. J Dent Res 90: 1339–1345. https://doi.org/10.1177/0022034511420430 Ship JA (2002) Diagnosing, managing, and preventing salivary gland disorders. Oral Dis 8:77–89 Work WP (1981) Non-neoplastic disorders of the parotid gland. J Otolaryngol 10:35–40 McQuone SJ (1999) Acute viral and bacterial infections of the salivary glands. Otolaryngol Clin N Am 32:793–811 Rogers JM, McCaffrey TV (2010) Cummings otolaryngology head and neck surgery, 5th edn. Mosby Elsevier, Philadelphia Sreebny LM, Vissink A (2010) Dry mouth. The malevolent symptom: a clinical guide. Wiley-Blackwell Xu J, Zheng Z, Fang D, Gao R, Liu Y, Fan ZP, Zhang CM, Wang SL (2012) Early-stage pathogenic sequence of jaw osteoradionecrosis in vivo. J Dent Res 91:702–708. https://doi. org/10.1177/0022034512448661

25.

Omolehinwa TT, Akintoye SO (2016) Chemical and radiationassociated jaw lesions. Dent Clin N Am 60:265–277. https://doi. org/10.1016/j.cden.2015.08.009 26. Corraini P, Heide-Jorgensen U, Schiodt M, Norholt SE, Acquavella J, Sorensen HT, Ehrenstein V (2017) Osteonecrosis of the jaw and survival of patients with cancer: a nationwide cohort study in Denmark. Cancer Med 6:2271–2277. https://doi.org/10.1002/ cam4.1173 27. Cotrim AP, Sowers A, Mitchell JB, Baum BJ (2007) Prevention of irradiation-induced salivary hypofunction by microvessel protection in mouse salivary glands. Mol Ther 15:2101–2106. https:// doi.org/10.1038/sj.mt.6300296 28. Tsai CJ, Hofstede TM, Sturgis EM, Garden AS, Lindberg ME, Wei Q, Tucker SL, Dong L (2013) Osteoradionecrosis and radiation dose to the mandible in patients with oropharyngeal cancer. Int J Radiat Oncol Biol Phys 85:415–420. https://doi.org/10.1016/j. ijrobp.2012.05.032 29. Strigari L, Benassi M, Arcangeli G, Bruzzaniti V, Giovinazzo G, Marucci L (2010) A novel dose constraint to reduce xerostomia in head-and-neck cancer patients treated with intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys 77:269–276. https://doi. org/10.1016/j.ijrobp.2009.07.1734 30. Beumer J 3rd, Silverman S Jr, Benak SB Jr (1972) Hard and soft tissue necroses following radiation therapy for oral cancer. J Prosthet Dent 27:640–644 31. Cassolato SF, Turnbull RS (2003) Xerostomia: clinical aspects and treatment. Gerodontology 20:64–77 32. Rivero JA, Shamji O, Kolokythas A (2017) Osteoradionecrosis: a review of pathophysiology, prevention and pharmacologic management using pentoxifylline, alpha-tocopherol, and clodronate. Oral Surg Oral Med Oral Pathol Oral Radiol 124:464–471. https://doi. org/10.1016/j.oooo.2017.08.004 33. Barasch A, Cunha-Cruz J, Curro F, DeRouen T, Gilbert GH, Hujoel P, Safford MM, Vena DA, Voinea-Griffin AE, Wu H (2013) Dental risk factors for osteonecrosis of the jaws: a CONDOR case-control study. Clin Oral Investig 17:1839–1845. https://doi.org/10.1007/ s00784-012-0880-4 34. Curi MM, Dib LL (1997) Osteoradionecrosis of the jaws: a retrospective study of the background factors and treatment in 104 cases. J Oral Maxillofac Surg 55:540–544 discussion 545-6 35. Hutchinson I (1996) Complications of radiotherapy in the head and neck: an orofacial surgeon’s view. In: Current radiation oncology. Arnold, London, pp 144–177 36. Marx RE, Johnson RP (1987) Studies in the radiobiology of osteoradionecrosis and their clinical significance. Oral Surg Oral Med Oral Pathol 64:379–390 37. Katsura K, Sasai K, Sato K, Saito M, Hoshina H, Hayashi T (2008) Relationship between oral health status and development of osteoradionecrosis of the mandible: a retrospective longitudinal study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 105: 731–738. https://doi.org/10.1016/j.tripleo.2007.10.011