Introduction

Common salivary gland diseases generally originate from infections, sialolithiasis, congenital anomalies, and of course neoplastic diseases [1]. Sialoadenitis is often treated conservatively with antibiotics, salivary massage, and hydration [2]. Sometimes relieving the underlying obstruction is mandatory with the help of minimally invasive surgical procedures [3]. However, when the obstructive disease is too severe, or the salivary gland disorder is neoplastic in nature, a sialoadenectomy is traditionally recommended [4].

Little is known regarding the long-term adverse effects of a sialoadenectomy. Most studies focused on surgical complications with this procedure such as nerve damage and paralysis, remnant ductal tissue problems, or postoperative hemorrhage [5–7]. The risk of developing xerostomia under excessive gland removal was also mentioned [8]. Wase et al. first mentioned the possible effect of a sialoadenectomy on thyroid activity [9]. Navarivera et al. further reported the effect of a partial sialoadenectomy on thyroid gland function and structure, and proposed hormonal interrelationships of the salivary glands with other systems [10]. Removal of the submandibular gland was reported to be associated with lowered sperm production parameters in animals [11]. More recently, the importance of salivary-derived growth factors, including epidermal growth factor (EGF), that play a role in helping maintain levels of oral health by promoting wound healing and maintaining mucosal integrity, was addressed [12–14]. With such complex secretory components containing digestive enzymes, immunoglobulins, growth factors, electrolytes, and buffers, it seems that the actual long-term changes to the human body once the salivary glands have been removed remain largely unanswered [15, 16].

To date, increasing evidences indicated that salivary glands are important in nitrate transport [17]. In addition, the vasoprotective effects were considered to be associated with the activity of nitrite converted from ingested nitrate [18]. Thus, it was plausible that sialoadenectomy procedure might affect the enterosalivary conversion and further contribute to an elevated risk of CHD. The purpose of this study was to provide an estimation of risk of developing coronary heart disease (CHD) among patients receiving the sialoadenectomy procedure by utilizing a cohort study based on a population-based database in Taiwan.

Methods

Results

Of the 2432 total sampled patients in this cohort study, the mean age was 48.5±14.9 years (range 18–87 years). Table 1 shows the distributions of demographic characteristics and comorbidities stratified by the presence or absence of a sialoadenectomy. After being matched for propensity score, Table 1 reveals that there was no significant difference in sex (p>0.999), age (p>0.999), urbanization level (p = 0.443), monthly income (p = 0.274), and geographic region (p = 0.921) between patients who underwent a sialoadenectomy and comparison patients. As to comorbidities, we also failed to observe a significant difference in hypertension (p = 0.798), hyperlipidemia (p>0.999), diabetes (p = 0.395), stroke (p>0.999), obesity (p = 0.806), and tobacco use disorder (p = 0.215) between patients who did and those who did not undergo a sialoadenectomy.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic characteristics of sampled subjects (N = 2432).

https://doi.org/10.1371/journal.pone.0199135.t001

The incidence of CHD during the 3-year follow-up period is presented in Table 2. We found that respective incidence rates of CHD during the 3-year follow-up period were 3.87 (95% CI: 3.01–4.91) and 1.79 (95% CI: 1.45–2.18) per 100 person-years for patients who did and those who did not undergo a sialoadenectomy. The log-rank test suggests that patients who underwent a sialoadenectomy had a greater tendency to have CHD than comparison patients (p<0.001).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Hazard ratio (HR) for coronary heart disease among sampled subjects during the 3-year follow-up period.

https://doi.org/10.1371/journal.pone.0199135.t002

Furthermore, Table 2 shows the HR for CHD between patients who did and those who did not undergo a sialoadenectomy. The stratified Cox proportional analysis (stratified by propensity score and the year of the index date) revealed that the HR of CHD during the 3-year follow-up period was 2.43 (95% CI = 1.77–3.33) for patients who underwent a sialoadenectomy compared to comparison patients.

Table 3 analyzed the HRs of CHD between patients who did and those who did not undergo a sialoadenectomy according to sex. We found that both male and female patients who underwent a sialoadenectomy had a higher following risk of CHD than comparison patients (with an HR of 2.32 for males and 2.62 for females).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Hazard ratios (HRs) for coronary heart disease among sampled subjects during the 3-year follow-up period by sex.

https://doi.org/10.1371/journal.pone.0199135.t003

Discussion

In this study, we demonstrated that sialoadenectomy was significantly associated with subsequent incidence of CHD regardless of sex. Our findings support the salivary gland possibly having important roles that are interrelated with different systems across the human body, and the conventional thinking that the salivary glands are one of the relatively dispensable organs in the body should be re-evaluated.

The importance of the salivary glands and the extent of the sacrifice they can impart on the human body are truly debatable issues. Studies reporting various complications such as nerve damage and paralysis, remnant ductal tissue problems, and postoperative hemorrhage were limited in scope to immediate or short-term surgical procedure-related problems [5–7]. What about functional aspects? The most straightforward thinking about consequences after a sialoadenectomy is a reduction in the saliva amount. Dry mouth, xerostomia, and even oral candidiasis were largely addressed in various animal studies [19, 20]. However, little has been reported regarding these issues in humans; probably the problem largely depends on highly variable, age-dependent resting salivary flow, which outweighs the effect of a sialoadenectomy [21]. Similarly, little has been reported regarding the potentially decreased immunoglobulin amount in saliva after a sialoadenectomy and its consequences. What would be more interesting would be to examine those substances that were previously underlooked but were proven to have roles in normal physiology. Previously, we mentioned the reported impacts on thyroid activity and hormonal interrelationships after a sialoadenectomy [9, 10]. More recently, contributions of epidermal growth factor (EGF) by the salivary gland were emphasized. The importance of this growth factor was revealed through some animal studies decades ago [22–24]. This salivary gland-excreted growth factor is known to be associated with the development of gastric lesions [25]. Recently Ainola et al. reported that diminished salivary EGF secretion could be a link between Sjögren’s syndrome and autoimmune gastritis [26]. Azuma et al. also reported that the salivary EGF levels could play a role in the pathogenesis of refractory intraoral manifestations in Sjögren’s syndrome patients [27].

The sialochemistry behind the possible link between a sialoadenectomy and the development of CHD is likely to be mediated through nitrite metabolism. The importance of nitrite in the saliva has been known since 1974, when Tannenebaum et al. reported the possible relationship of salivary nitrite with nitrosamine formation [28]. Tenovuo et al. reported the biochemistry of nitrates, nitrites, and nitrosamines in human saliva [29]. Nitrate, after its absorption in the upper gastrointestinal tract, reaches the salivary glands via the blood circulation, where it is secreted into the oral cavity and partially reduced to nitrite by the oral microflora [30]. In addition to nitrate or nitrite under conditions that result in endogenous nitrosation being classified as "probably carcinogenic to humans", nitrite is also important in biochemistry as a source of nitric oxide, a potent vasodilator [31]. Nitrate and nitrite, through ultraviolet light exposure to skin, mammalian nitrate/nitrite reductases in tissues, and nitric oxide synthase enzymes, are converted to a diverse group of metabolites including nitric oxide, a powerful vasodilator, and potentially influence the risk of developing cardiovascular diseases [32]. Tripp et al. found that in sialoadenectomized rats, inhibition of nitric oxide synthase activity was exacerbated, implying that factors from the salivary glands influence gastric nitric oxide formation [33]. Björne et al. reported that nitrite-containing human saliva given luminally increases gastric mucosal blood flow, and these nitrite-mediated effects were associated with the generation of nitric oxide [34]. Webb et al. reported that blood pressure-lowering, vasoprotective, and antiplatelet properties can even be achieved by dietary nitrates, and interruption of the enterosalivary conversion of nitrate to nitrite (facilitated by bacterial anaerobes situated on the surface of the tongue) prevented the rise in plasma nitrite, blocked the decrease in blood pressure, and abolished the inhibitory effects on platelet aggregation [18]. Those authors stated that the vasoprotective effects were attributable to the activity of nitrite converted from ingested nitrate. Because the salivary glands are very important in nitrate transport and nitric oxide homeostasis, the sialoadenectomy procedure might affect the enterosalivary conversion pathway and ultimately lead to an increased risk of CHD [17].

Nevertheless, one should be reminded that this is an observational study, and the findings in this study must be further investigated before a true hypothesis is made.

Also, like much health insurance database analytical research, it comes with the possibility of surveillance bias. Patients receiving a sialoadenectomy are more likely to visit outpatient clinics for follow-up, and this might lead to an increased detection of CHD due to the increased exposure to medical services. Furthermore, another important limitation is that the LHID2005 provides no information on body mass index (BMI). Nevertheless, in order to eliminate the potential effects of BMI on the association between sialoadenectomy and CHD, we have used obesity in place of BMI in this study.

Although this was an observational study, the result that a sialoadenectomy might be associated with an increased risk of developing CHD is worth further investigation. Clinically, it is suggested that physicians might take the findings in this study into considerations in determining the risks and benefits of the sialoadenectomy procedure and even consider different approaches such as a partial sialoadenectomy according to the patient’s gender and other comorbidities [35].