Introduction

Chronic kidney disease (CKD), a chronic structural and functional degeneration of the kidneys [1], has a worldwide prevalence between 11 and 13% [2] and it is considered a global health problem, representing a major morbidity and mortality factor in non-communicable chronic diseases [3, 4]. According to the Glomerular Filtration Rate (GFR) and the degree of proteinuria (albumin/creatinine ratio) [1] CKD severity can be classified into various stages.

End-stage renal disease patients require therapies that restore renal function [5], such as dialysis treatments (hemodialysis and peritoneal dialysis) and kidney transplant [1]. Hemodialysis is the most used [6], and it focuses both on filtering plasma electrolytes at high concentrations, especially hypocalcemia and metabolic acidosis [7], and on normalizing the circulatory volume by eliminating excess fluids from the organism [5].

About 90% of CKD patients have oral signs and symptoms [8, 9] such as xerostomia, halitosis, metallic taste, uremic stomatitis, mucositis, glossitis, among others [10,11,12]. Additionally, CKD patients often have changes in salivary flow rate and composition [13, 14].

CKD patients usually have a lower salivary flow rate [10], and their saliva has higher concentrations of urea, sodium, phosphate, thiocyanate, and potassium; reduced calcium levels; increased pH; and decreased buffering capacity [6, 15,16,17] in comparison to healthy individuals. Altogether, these salivary changes lead to oral dysfunctions in the remineralization process [18], and to the formation of dental calculus, which can ultimately influence the occurrence of dental caries and periodontal diseases, respectively [19, 20]. Understanding the impact of CKD in the salivary flow rate, pH, and ionic composition can be relevant for dental treatment planning, as it can assist in establishing diagnoses, medication prescriptions, and adequate oral health instructions tailored to CKD patients [9, 21].

It is of the utmost importance to provide high-quality evidence about the relationship between CKD and salivary changes in patients diagnosed with the disease and undergoing hemodialysis, as well as to verify the action of the dialytic treatment in the saliva of these patients right after the procedure. Thus, the present systematic review aims to assess changes in the flow rate, pH, and levels of calcium, phosphate, and phosphorus in the saliva of CKD patients and to investigate the influence of hemodialysis on these parameters.

Materials and methods

Protocol registration

The protocol utilized in this systematic review was described according to the PRISMA-P guidelines (Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols) [22] and registered in the PROSPERO database under number CRD42021231129. This systematic review was reported according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyzes) guidelines [23] and performed according to the norms of the JBI manual for Evidence Synthesis [24].

Research question and eligibility criteria

This review aimed to answer two guiding questions, which were designed according to the PECO acronym (Population, Exposure, Comparator, and Outcome): (1) “Will end-stage chronic kidney disease patients undergoing hemodialysis present changes in salivary flow rate, pH, and levels of calcium, phosphate, and phosphorus when compared to healthy individuals?” and (2) “Can a single hemodialysis session change the salivary flow rate, pH, and levels of phosphate, phosphorus, and calcium the in saliva of CKD patients?”.

Inclusion criteria

Question 1:

  • Population: Adults (> 18 years old);

  • Exposure: Chronic kidney disease and hemodialysis (GFR < 15 ml/min/1.73 m2);

  • Control group: Healthy individuals;

  • Outcome: Salivary flow rate, pH, and levels of phosphate, phosphorus, and calcium in whole saliva;

  • Study design: Observational studies (case–control or controlled cross-sectional).

Question 2:

  • Population: Adult patients (> 18 years old) with chronic kidney disease;

  • Exposure: Hemodialysis (GFR < 15 ml/min/1.73 m2);

  • Control Group: Salivary parameters (flow rate, pH, and levels of phosphate and calcium) before dialysis;

  • Outcome: Salivary flow rate, pH, and levels of phosphate, phosphorus, and calcium after dialysis;

  • Study design: Pre/post-test studies with or without a control group with healthy participants.

There were no restrictions of publication language or year to any of the guiding questions.

Exclusion criteria

  • Literature reviews, letters to the editor/editorials, personal opinions, books/book chapters, case reports/case series, pilot studies, conference abstracts, and patents;

  • Studies with overlapping results/samples;

  • Studies with pediatric patients;

  • Studies that did not collect whole saliva;

  • Studies including patients with renal diseases other than CKD.

Sources of information and search

Electronic searches were performed in MedLine (via PubMed), Scopus, LILACS, SciELO, Web of Science, Embase, and LIVIVO databases. OpenThesis and OpenGrey databases were used to partially capture the “grey literature”. MedLine search was updated constantly by PubMed through warnings, until July 2021. MeSH (Medical Subject Headings), DeCS (Health Sciences Descriptors), and Emtree (Embase Subject Headings) resources were used to select the search descriptors. Moreover, synonyms and free words composed the search. The Boolean operators “AND” and “OR” were used to improve the research strategy with several combinations. The search strategies in each database were made according to their respective syntax rules (Table 1). The results obtained in the primary databases were initially exported to EndNote Web™ software (Thomson Reuters, Toronto, Canada) for cataloging and for the removal of duplicates. “Grey literature” results were exported to Microsoft Word (Microsoft™, Ltd, Washington, USA) for the removal of duplicates.

Table 1 Strategies for database search
Full size table

Study selection

Before selecting the studies, the reviewers discussed the eligibility criteria and applied them to a sample of 20% of the studies retrieved to determine the inter-examiner agreement. After reaching a Kappa ≥ 0.81, the reviewers (MTCV and RCPBR) performed a title analysis of the studies (first phase), and those not related to the topic were eliminated. In the second phase, the abstracts of the studies were assessed with the initial application of the eligibility criteria. Titles that met the objectives of the study but did not have abstracts available were fully analyzed in the next phase. In the third phase, a full analysis of the texts of all eligible studies was carried out. Full texts published in languages other than English or Portuguese were translated to allow the application of the eligibility criteria. All phases were performed independently by two reviewers, and, in case of uncertainty or disagreement, a third reviewer (LRP) was consulted to make a final decision.

Data collection

Data collection from eligible articles was performed independently by two authors (RPCBR and MTCV). To ensure consistency of data extraction by the two authors, a third reviewer (WAV) conducted a calibration exercise with three eligible articles. In this calibration exercise, the authors were introduced to the information that should be collected from each eligible study and how it should be presented. At the end of the calibration exercise, the authors proceeded with data collection for the remainder of the eligible articles. Divergence in the information collected by the two authors was resolved by a third reviewer (WAV).

The following data were extracted from the articles: study identification (author, year, country, location, and application of ethical criteria), sample characteristics (number of CKD patients, number of healthy patients for studies requiring a control group, CKD stage of the patients, distribution by sex, and average age), collection and processing characteristics (method utilized for saliva collection, time of saliva collection, type of salivary analysis, and type of statistical analysis), and the main results (mean and standard deviation rate values of the flow rate, pH, and levels of phosphate and calcium of saliva, and main outcomes of each study). In cases where incomplete or insufficient information was present, the corresponding author was contacted via e-mail.

Risk of bias assessment

Two independent authors (RPCBR and WAV) assessed the risk of bias and individual quality of the eligible studies with the JBI Critical Appraisal Tools for use in the JBI Critical Appraisal Checklist for Analytical Cross-Sectional Studies and the JBI Critical Appraisal Checklist for Analytical Quasi-experimental Studies, according to the respective study design [24]. For calibration, the authors analyzed an eligible study jointly in the presence of a third reviewer (LRP) responsible for solving the divergences in case of uncertainty.

Each study was categorized according to the percentage of positive answers to the questions corresponding to the assessment instrument. Each question could be answered as follows: “Yes”, if the study did not present biases for the domain assessed in the question; “No”, if the study presented biases for the domain assessed in the question; “Uncertain”, if the study did not provide sufficient information to assess the bias of the question; or “Not Applicable”, if the question did not fit in the study. The risk of bias was considered High when the study obtained 49% or less of “yes” answers, Moderate when the study obtained 50% to 69% of “yes” answers, and Low when the study reached 70% or more of “yes” answers [25].

Data synthesis and meta-analysis

The data collected were organized and described descriptively/narratively (qualitative synthesis) according to the findings presented in each study selected. Moreover, the quantitative data presented in each study were extracted, organized in tables, and then imported into the Stata 16.1 statistical software (StataCorp LLC, College Station, TX, USA). Only studies with specific data from patients in end-stage renal failure (stage 5) were included in the analysis.

To answer research question number 1, a meta-analysis was performed comparing the mean value of each indicator of interest (calcium concentration, salivary flow rate, phosphate concentration, phosphorus concentration, and salivary pH) between patients undergoing hemodialysis and healthy individuals. To answer research question number 2, the mean values of the same five indicators of interest, before and after hemodialysis, were compared. Although there were five indicators of interest, a meta-analysis was only considered if there were at least three studies for each indicator, otherwise, a combined measure was not estimated due to the limited number of studies.

In both questions, the calculated effect measure was the standardized mean difference. This measure represents the difference, in standard deviations, between the samples compared. On research question number 1, the effect measure was calculated directly with a meta-analysis for each indicator of interest. For research question number 2, considering that the samples compared were not independent of each other (repeated measures of the patients at two time points—before and after hemodialysis), a combined mean value was estimated for each indicator of interest at each time point. Finally, the standardized mean difference between the two groups was estimated, preventing biases in the effect measures calculated [26].

All meta-analyses were performed considering random effects, proposed by DerSimonian-Laird. Three heterogeneity measures were considered in the analyses: (1) I2 statistics; (2) τ2 statistics; (3) H2 statistics. The I2 represents the rate of variability among the studies caused by heterogeneity, excluding sample errors. The τ2 represents the variance size among the studies, while H2 represents the degree of heterogeneity among the studies, in which H2 = 1 represents homogeneity. Meta-regressions were adjusted considering the method of saliva collection (stimulated or non-stimulated) and the average age of the participants of each study, to identify potential moderators. All analyses considered a 5% significance level.

Results

Study selection

During the first phase of the study selection, 4574 results were found distributed in nine electronic databases, including the “grey literature”. After removing the duplicates, 3283 results remained for analysis. A careful analysis of the titles excluded 3078 results. Two hundred and five studies remained for abstract reviews. From those, 155 studies were excluded after applying the eligibility criteria. The 50 studies remaining were assessed. Seventeen studies were excluded and the reasons for exclusion were registered separately (Supplementary Table 1). Finally, thirty-three studies [6, 10, 13,14,15,16, 21, 27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52] were included in the qualitative analysis. Figure 1 presents the details of the process of search, identification, inclusion, and exclusion of studies.

Fig. 1
figure 1

Flowchart describing the search process and selection of eligible studies

Full size image

Characteristics of eligible studies

Studies were published between 1999 and 2021 and conducted in 17 different countries, with 19 studies carried out in Asia, eight in Europe, three in South America, two in Africa, and one in Oceania. The total number of samples included 3147 participants, with 1969 CKD patients (cases) and 1178 healthy individuals (controls). The average age varied between 34.7 and 69.7 years among CKD patients and between 30.5 and 60.1 among healthy individuals. Men composed approximately 39% of the total sample among the studies that presented data related to participants’ sex.

Five eligible studies [6, 21, 33, 36, 47] collected saliva of CKD patients at pre-dialysis and dialysis stages. Samples from all the other eligible studies presented patients with stage-5 CKD, which are patients who had undergone dialysis treatment, especially hemodialysis. Only three studies [32, 33, 43] reported patients who underwent peritoneal dialysis. For saliva collection, the eligible studies reported using non-stimulated and stimulated methods (with different stimulation practices). The time of saliva collection to which the patients were subjected varied between one and 15 min or the necessary time to reach a certain amount of saliva, which was previously established. Table 2 details the most relevant information of each eligible study and Table 3 presents the quantitative results and the main outcomes of the eligible studies in detail.

Table 2 Summary of the main characteristics of the eligible studies
Full size table
Table 3 Main results and outcomes of eligible studies investigating salivary parameters (salivary flow, pH, calcium, phosphate, and phosphorus levels)
Full size table

Risk of individual bias of the studies

In the risk of bias analysis of the cross-sectional eligible studies, 10 studies [6, 21, 31, 33, 36, 37, 39, 40, 45, 48] presented a low risk of bias, 15 studies [10, 15, 27,28,29, 32, 34, 35, 41, 42, 44, 46, 47, 49, 52] presented a moderate risk of bias, and one study [13] presented a high risk of bias or low methodological quality. In the analysis of the risk of bias of the quasi-experimental studies, all of them [14, 16, 30, 38, 43, 50, 51] presented a low risk of bias. Table 4 presents detailed information on the risk of bias of the eligible studies.

Table 4 Risk of bias assessed by the Joanna Briggs Institute Critical Appraisal Tools for use in the JBI Critical Appraisal Checklist for Analytical Cross Sectional Studies (Moola et al. 2020) and the JBI Critical Appraisal Checklist for Analytical Quasi-experimental Studies (Tufanaru et al. 2020)
Full size table

Synthesis of results and meta-analysis

The study by Thorman et al. [33] did not separate the quantitative data of patients undergoing hemodialysis or peritoneal dialysis, and the study by Savica et al. [31] did not present the quantitative data of the control group. Thus, only 31 studies were included in the meta- analyses.

CKD patients versus healthy individuals

A total of ten studies compared the mean values of calcium between the groups, showing high heterogeneity (I2 = 95.8% and H2 = 23.7). The mean values of salivary calcium concentration were similar when comparing CKD patients and healthy individuals (SMD = 0.39; 95% CI = − 0.37; 1.16; p = 0.310) (Fig. 2).

Fig. 2
figure 2

Meta-analysis of calcium concentration levels between patients undergoing hemodialysis and healthy individuals

Full size image

Regarding salivary flow rate, 17 studies were included in the analysis (Fig. 3). Similar to the concentrations of calcium, there was high heterogeneity among the studies (I2 = 94.3% and H2 = 17.6). Moreover, 90% of the studies included in this analysis had the same direction of effect, showing a reduced salivary flow rate in patients. The combined measure identified that salivary flow rate in CKD patients was 1.73 standard deviations (95% CI = − 2.14; − 1.31; p < 0.001) lower than healthy individuals.

Fig. 3
figure 3

Meta-analysis of salivary flow rate between patients undergoing hemodialysis and healthy individuals

Full size image

The concentrations of phosphorus in the saliva of patients and healthy individuals were compared in four studies. There was no heterogeneity among the studies included (I2 = 0.0% and H2 = 1.0). The standardized mean difference was 0.86 standard deviations (95% CI = 0.63; 1.09; p < 0.001) higher in CKD patients than healthy individuals (Fig. 4).

Fig. 4
figure 4

Meta-analysis of phosphorus concentration levels between patients undergoing hemodialysis and healthy individuals

Full size image

Finally, the pH levels were also compared in 12 studies. The heterogeneity among studies was considered high (I2 = 91.9% and H2 = 12.4) but 78.5% of the studies included showed the same direction of effect (higher pH levels in patients). The combined measure estimated by the meta-analysis showed that the pH was 1.57 standard deviations (95% CI = 1.11; 2.03; p < 0.001) higher than healthy individuals (Fig. 5). The levels of phosphate were not compared because only two studies considered this indicator.

Fig. 5
figure 5

Meta-analysis of pH concentration levels between patients undergoing hemodialysis and healthy individuals

Full size image

Before and after hemodialysis

The levels of calcium were similar before and after hemodialysis. It was not possible to compare the concentrations of phosphorus and phosphate in saliva. Conversely, the salivary flow rate was 0.53 standard deviations (95% CI = 0.25; 0.81) higher after hemodialysis, while the pH was 0.53 standard deviations (95% CI = − 0.88; − 0.19) lower after hemodialysis (Table 5).

Table 5 Combined means and standardized mean differences after the meta-analysis of the concentration levels of calcium, pH, and salivary flow at two time points: before and after hemodialysis
Full size table

Discussion

This study investigated whether chronic kidney disease patients undergoing hemodialysis present changes in ionic composition, salivary flow rate, and salivary pH in comparison to healthy individuals. Additionally, it assessed the ability of one hemodialysis session to reverse these parameters. The meta-analyses showed that CKD patients had reduced salivary flow rate, increased salivary pH, and increased concentration of salivary phosphate, however, salivary flow rate alone was reversed after one hemodialysis session.

Maintaining an adequate salivary flow rate is important to prevent oral diseases such as caries and periodontitis, considering that low salivary flow rate is related to the maintenance of bacterial plaque and biofilm in the oral cavity [53,54,55]. The eligible studies that are present in this review and the results of the meta-analysis showed a significant reduction in the salivary flow rate of CKD patients. This finding can be related to oral manifestations often found in CKD patients, such as xerostomia, taste changes, and pale mucosa [13, 42, 47], although such association is not fully explained in the literature [30, 42, 47]. The reduction of salivary flow rate in CKD can be related to molecular changes in salivary glands due to medications used by CKD patients [29] or a reduction in the extracellular liquid volume of salivary glands, which relates to their hydration condition and hydroelectrolytic balance [6]. However, these phenomena are not well established.

Additionally, the results of the meta-analysis showed that one hemodialysis session can increase the salivary flow rate of CKD patients, reversing the changes established by CKD. One hypothesis for this finding is that the kidney ultrafiltration mechanism promoted by the hemodialysis procedure increases the intravascular volume leading to higher gland perfusion and consequently a higher production of saliva [51]. Considering that hemodialysis can reestablish the concentrations of sodium, potassium, bicarbonate, creatinine, and urea ions to values closer to the physiological level [56], it is suggested that the reestablishment of the hydroelectrolytic balance is related to the increase in salivary secretion after one hemodialysis session. It is known that increases in blood pressure are associated with increased sympathetic activity toward the salivary glands, which promotes increased glandular vasoconstriction and can reduce salivary flow rate [57]. Thus, the decrease in blood pressure, which is frequent in most hypertensive patients undergoing hemodialysis [58], can be related to the reestablishment of the sympathetic activity for salivary glands and salivary secretion after one hemodialysis session. This finding emphasizes the importance of hemodialysis not only for filtering toxic molecules but also for improving oral health [14, 51].

The salivary pH was another major aspect investigated in several eligible studies. The results of the meta-analysis showed that salivary pH is higher in CKD patients when compared to healthy individuals. This may be justified by the fact that CKD patients present high levels of urea in their salivary composition [43], which is metabolized by the oral microflora into carbon dioxide and ammonia [59], thus increasing the salivary pH [42]. In this context, considering that an acidic pH favors the demineralization process of the dental structure [60], it can be correlated to a lower prevalence of caries in CKD patients, even though salivary flow rate is also often reduced [61]. Considering that a low salivary pH provides an acidogenic environment for the growth of aciduric bacteria that in turn leads to dental caries [60], a higher pH can work as a protective factor for enamel demineralization.

Although hemodialysis can reduce the concentration of urea in saliva [25], the results of the present meta-analysis showed that this isolated factor was not sufficient to reduce the salivary pH immediately after hemodialysis. This result can be justified by the fact that the studies collected saliva immediately after the dialysis procedures, which can represent a short time to assess the changes in the metabolism of urea in intracellular mechanisms of acinar cells from the salivary gland, and consequently a reduction in salivary pH. Further studies should be performed to observe the influence of hemodialysis in salivary pH for longer intervals after the hemodialysis session.

The level of salivary calcium is also another important biomarker in CKD patients. It is believed that higher concentrations of calcium in saliva can not only work as a protective factor for the development of caries but also as a risk factor for the formation of calculus [62, 63]. Although some eligible studies in this review highlighted the association of CKD patients with higher chances of presenting dental calculus [15, 34, 42, 49], there is no consistent evidence confirming this association and it was not the objective of the present review. Moreover, the results of this meta-analysis did not show significant changes in the levels of calcium in CKD patients when compared to healthy patients, despite the high inconsistency found in the individual results of the studies. We believe there is no standardized methodology in the studies, which used different laboratory methods to calculate the levels of calcium in saliva, besides the population differences. This prevents drawing more categorical conclusions about such a variable.

The levels of salivary phosphorus and phosphate just like calcium are important biomarkers that can be affected by CKD and are described as potential ways of assessing the need for hemodialysis [40]. The present meta-analysis evidenced increased levels of salivary phosphorus in CKD patients in comparison to healthy individuals, which can be justified by the pathophysiological changes in the kidneys that increase this component in plasma and saliva.

We emphasize that the results of these meta-analyses should be interpreted with caution due to the high heterogeneity of analyses that could not be explained by meta-regression analyses, which can be justified by the different populations and methodologies used to assess this variable and the clinical characteristics of patients that could not be assessed individually, such as comorbidities associated with CKD, dialysis vintage, age group, and medication use. However, to the best of our knowledge, this systematic review of the literature is the first to investigate the salivary changes of CKD patients and assess the effects of hemodialysis in saliva. Moreover, an extensive literature search without restrictions of publication year or language ensured the inclusion of a maximum number of published studies.

From the dental clinic standpoint, this systematic review draws attention to major salivary changes that should be considered when planning dental treatment for patients with kidney failure, such as xerostomia, taste changes caused by reduced salivary flow rate, monitoring of the periodontal condition, and the appearance of oral lesions, in an effort to reduce the impact of these oral manifestations on the quality of life of those patients.

Conclusion

Chronic kidney disease patients present major changes in salivary properties and composition, such as reduced flow rate and increased salivary pH, as well as higher levels of phosphorus, but with no differences in calcium and phosphate levels when compared to healthy individuals. It was also verified that hemodialysis can increase the salivary flow rate of these patients. These findings can assist in the understanding of the oral manifestations of chronic kidney disease patients and raise awareness of the need for customized clinical planning of dental care in this group of patients.