Introduction

Human saliva as a bodily fluid—similar to blood—is utilized for diagnostic purposes. However, unlike blood sampling, collecting saliva is non-invasive, inexpensive, readily accessible, and stress-free [1,2,3,4]. The exocrine contribution from each of the three major couple salivary glands (i.e., parotid saliva (PS), sublingual saliva (SLS), and submandibular saliva (SMS)) along with the saliva secreted from numerous minor salivary glands, compose the whole mouth saliva (WMS) [5, 6]. In addition, WMS contains non-exocrine components as well (e.g., micro-organisms, leukocytes, desquamated oral epithelial cells, gingival (crevicular) fluid, and the serum-like fluid derived from the epithelial mucosa) [7, 8]. In gratitude towards the contribution of the mucosal and gingival fluids, transported substances in the circulatory system are also present in the WMS [9]. Therefore, WMS meets all the requirements for its use as a diagnostic bodily fluid [10, 11]. Given the many potentials of WMS, it can replace some of the blood samplings in patients who have difficulties with blood collection (e.g., toddlers, and seniles), or in patients who have to take blood samples weekly or even daily (e.g., diabetic patients, and patients who take drugs with serious side effects such as methotrexate and warfarin) [12].

Since the end of 2019/start of 2020, the COVID-19 pandemic led to a variety of invasive and non-invasive diagnostic tests to be taken every day from millions of people [13, 14]. The COVID-19 pandemic highlighted the speed, accuracy, and feasibility of non-invasive bodily fluid sampling (e.g., saliva sampling, and collecting specimen from oropharyngeal and nasopharyngeal mucosa) for viral infection screenings in large populations [15,16,17].

Human saliva like any other bodily fluid utilized for diagnostic purposes, requires proper collection/sampling methods and devices, precise sampling time, appropriate handling and transportation conditions, and eventually, established storage considerations until further analysis of samples [18, 19]. The endogenous and exogenous enzymes accompanied by an unforeseeable activity and configuration are responsible for vigorous and continuous modifications of specimen [1, 18]. Moreover, contributions of different salivary glands to the composition of WMS changes in accordance to the circadian rhythms [20, 21]. Therefore, the time of sampling varies depending on the purpose of the experiment [22, 23].

Over the years, a variety of different stimulating and non-stimulating saliva sampling methods have been introduced and experimented [24,25,26]. Passive drooling and spitting have been the most assessed non-stimulating methods [27]. While Salivette®, Parafilm® wax and paraffin wax have been assessed as stimulating methods [28]. Some scientists believe that a non-stimulated passive drooling of saliva provides the most unmanipulated and authentic sample for further analysis [29]. On the other hand, some believe that a highly sensitive device with collective absorption abilities results in fewer redundant and inessential nano and microparticles in the samples, and consequently faster and more accurate laboratory tests [30, 31]. Nonetheless, there are still no guidelines as to whether devices are necessary for some experiments, and if necessary which devices are preferred for each test [32,33,34]. Moreover, the superiority or inferiority of stimulated samples compared to non-stimulated samples have not been investigated in many studies [35, 36]. From leaving samples in room temperature and analyzing them without any storage immediately after sampling, to storing samples at -80 °C for months before analysis, there are numerous handling, transportation and storage methods, each employed for different analytic purposes [37,38,39]. Similar to sampling methods and devices, there are no established guidelines in regards to the transportation and storage conditions of human saliva samples [40,41,42].

Given the various diagnostic abilities of WMS and numerous features to potentially replace blood sampling in many categories of tests, WMS has gained remarkable trust as a reliable diagnostic bodily fluid [43,44,45]. In the past decade a special attention has been put upon creating more convenient and accurate sampling methods/devices assessed in fitting sampling times, along with proper transportation and storage conditions, depending on the tested DNA, hormone, molecule or nanoparticle [46,47,48]. To the best of our knowledge, there are no previously published systematic reviews on the different collection, transportation, preparation, and storage methods for human WMS in the literature, which is the main research gap of this study. The main goal for this systematic review was to gather all of the human clinical descriptive studies that have experimented different collection, transportation, preparation, and restoration techniques of human WMS. Hopefully, the extracted data reported in this review will guide clinicians and researchers in a more cohesive and accurate path in choosing the appropriate methods and devices for human WMS sampling. For a better understanding of the objectives and main purpose of this systematic review, a conceptual framework of the study has been prepared (Fig. 1).

Fig. 1
figure 1

Conceptual framework of the study

Materials and methods

This study has been prepared and organized according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) 2020 guidelines [49].This systematic review has been registered at PROSPERO (Registration ID: CRD42023415384). The study question according to the PICO format was as followed: Comparison of the performance (C) of different saliva sampling, handling, transportation, and storage techniques and methods (I), assessed for analyzing stimulated or unstimulated human saliva (P and O).

Eligibility criteria

Types of studies

Randomized or non-randomized descriptive clinical human studies that have investigated any saliva sampling technique.

Population

Human participants: no exclusions regarding age, race, or gender.

Intervention

Collecting human saliva using stimulating or unstimulating techniques. There were no restrictions on the type of saliva (e.g., parotid saliva, submandibular saliva, and sublingual saliva). All techniques were included whether they used a specific device or not.

Types of outcome measures

Studies that analyzed the following outcomes were included: 1) the efficiency of the experimented stimulated and unstimulated saliva sampling techniques for each of the tested elements in the saliva (e.g., salivary flow rate, saliva DNA quality and quantity, salivary hormone levels, etc.); 2) different preparation and transportation techniques and conditions; 3) comparison of different saliva sampling times in the day; 4) patients’ preparation before and during sampling (e.g., prohibition of drinking, eating, and smoking before sampling, etc.).

Information sources and search strategy

An electronic search was executed in Scopus, Google Scholar, and Medline via PubMed to identify eligible studies only in English language. The search was included of articles up to September 1, 2023. Search queries mentioned in Table 1 were considered for electronic search.

Table 1 Search queries

Study selection and data collection

Two reviewers (AY and HY) independently screened the titles and abstracts of articles and excluded articles based on exclusion criteria mentioned above. Selected articles were then fully read to see if they passed our inclusion criteria. In case of any disagreement a third reviewer (HM) was consulted. The data and outcomes from selected studies were then extracted and tabulated. The same reviewers performed the data extraction and any conflicts were solved by a third expert (HM).

Data items

The collected items were as followed; (1) authors’ name; (2) year of publishment; (3) study type; (4) type of saliva; (5) sampling time; (6) number of participants; (7) participants’ gender; (8) participants’ age range and mean average age; (9) participants’ preparation before and during sampling; (10) study variables; (11) collection methods/devices; (12) sampling duration; (13) transportation conditions; (14) restoring conditions; (15) sample analysis; (16) outcomes.

Synthesis methods

Based on the extracted data, different stimulated and unstimulated methods/techniques with or without sampling devices were widely diversified. Hence, it was not possible to perform a meta-analysis. Descriptive analysis of the data extracted from clinical studies, along with narrative and graphical synthesis was performed.

Risk of bias assessments

The JBI Critical Appraisal Tool for risk of bias assessment in cross-sectional studies was applied for both non-randomized and randomized studies to assess their risk of bias. Two reviewers (AY and HY) independently analyzed each study using the prefabricated questions of the JBI Critical Appraisal Tool for risk of bias assessment in cross-sectional studies. In case of any dissimilarity in the results, a third reviewer (HM) was consulted.

Results

Study selection

Database screening was performed and a total of 7637 articles were initially identified and 314 of them were assessed for eligibility (Fig. 2). A total of 291 studies were excluded for the following reasons; in vitro, in vivo, and ex vivo studies (n = 144) and unrelated subjects (n = 147) (Fig. 2). Hence, a total of 23 descriptive clinical human studies were included. Studies came from 13 different countries: USA (n = 5) [50,51,52,53,54], Brazil (n = 5) [55,56,57,58,59], China (n = 3) [60,61,62], Hong Kong (n = 1) [63], Kuwait (n = 1) [64], Singapore (n = 1) [65], Australia (n = 1) [66], Slovakia (n = 1) [67], Germany (n = 1) [68], Japan (n = 1) [69], France (n = 1) [70], Argentina (n = 1) [71], and Sweden (n = 1) [72]. Studies were published between 1995 and 2022: 1995 (n = 1), 2004 (n = 1), 2012 (n = 1), 2013 (n = 1), 2015 (n = 1), 2017 (n = 9), 2018 (n = 5), 2021 (n = 2), and 2022 (n = 2).

Fig. 2
figure 2

The PRISMA 2020 flow diagram of the screening and selection process

The 23 included studies were published in the following journals: Oral Diseases (n = 2) [57, 63], Developmental Psychobiology (n = 2) [52, 68], Journal of Analytical Toxicology (n = 2) [58, 70], Archives of Oral Biology (n = 2) [59, 64], Clinica Chimica Acta (n = 1) [65], Clinical Oral Investigations (n = 1) [60], Steroids (n = 1) [67], Clinical therapeutics (n = 1) [51], Scandinavian Journal of Clinical Laboratory and Investigation (n = 1) [55], Scientific Reports (n = 1) [66], Journal of Applied Oral Sciences (n = 1) [56], Amino Acids (n = 1) [69], Forensic Science International: Genetics (n = 1) [54], Annals of Human Biology (n = 1) [50], Laryngoscope (n = 1) [53], The Journal of Contemporary Dental Practice (n = 1) [72], Clinical Nutrition (n = 1) [62], International Journal of Environmental Research and Public Health (n = 1) [61], and Acta Odontológica Latinoamericana (n = 1) [71].

Eighteen of the included studies were funded by either public organizations or university grants [50,51,52,53,54,55,56,57,58, 60,61,62, 64, 67,68,69,70,71], two of the studies had no external funds for their experiments [59, 72], and three of the studies did not mention their funding/support status [63, 65, 66].

Results of individual studies

The tabulated data of each study, their participants’ demographics, their experimented methods and their outcome are all detailed in Table 2.

Table 2 Different types of saliva samples, times of the day for sampling, collection techniques, collection devices, patients’ preparations, sampling duration, transporting conditions, and restoring conditions

Study characteristics

Study design

All of the studies were observational cross-sectional studies and none of them had any intervention on patients.

Demographics

Nine of the studies did not report the gender ratios of their participants. In the remaining 14 studies, 322 of the participants were females and 367 of them were males. Two of the studies did not indicate the age range or mean average age of their participants. Sixteen of the studies reported the age range of their participants and in total it ranged from 2 months to 94 years (Table 2).

Types of saliva

In total there were 4 kinds of investigated saliva: whole mouth saliva (WMS), parotid saliva (PS), sublingual saliva (SLS), and submandibular saliva (SMS). Each of these saliva samples were collected either stimulated or unstimulated (Table 2).

Sampling time

Eleven studies out of all the included studies reported their sampling times (Table 2). Only 3 of those studies compared the outcome differences of different sampling times. Sampling time varied from 6:00 AM to 20:00 PM (Table 2).

Patient preparations before and during sampling

Most studies asked participants to not drink, eat, or smoke 30 min to 60 min before sampling (Table 2).

Study variables

Studies investigated a variety of different variables in human saliva: total saliva quantity, salivary flow rate, salivary total protein, saliva pH and buffering pH, salivary minerals (e.g., calcium, potassium, iodine, etc.), salivary hormones (e.g., cortisol, testosterone, DHEA, etc.), and salivary DNA quality and quantity (Table 2).

Collection methods/devices

In total, 22 sampling methods/devices were assessed amongst studies (Tables 2 and 3). Fourteen of these methods/devices were used to collect unstimulated samples and the rest were used for stimulated samples (Table 3).

Table 3 All different stimulating and unstimulating saliva sampling methods/devices

Sampling duration

Some studies asked participants to fill a certain amount of saliva regardless of how much time it took. On the other hand some studies asked patients to use/chew on the experimented device, paraffin wax or the Parafilm® wax for a certain amount of time regardless of the total amount of collected saliva (Table 2).

Transportation, sample analysis and restoring conditions

Only 1 of the studies did not indicate their transportation or restoring conditions. The rest of the studies had a variety of different experimented conditions (Table 2).

Reported outcomes

Sampling methods/devices

Overall, none of the 22 collection methods employed in the 23 included studies (Table 3) led to underwhelming outcomes for further laboratorial analysis. However, some of the methods outshined the rest in the studies that more than 1 method was utilized for saliva collection.

Salivary flow rate

In total, 8 methods/devices were assessed in this category of laboratorial tests (i.e., passive drooling, spitting, non-coated Salivette®, citric-acid-coated Salivette®, dry MicroFLOQ®, wet MicroFLOQ®, chewing mint-flavored gum, and chewing Parafilm) in 5 of the included studies [52, 54, 55, 59, 61]. Passive drooling and spitting both led to average/conventional results as unstimulated techniques. The dry MicroFLOQ® traces were the only stimulated method that had modest results. All of the 5 remaining stimulated techniques (i.e., wet MicroFLOQ®, chewing Parafilm®, chewing mint-flavored gum, citric-acid-coated Salivette®, and non-coated Salivette®) resulted in remarkable outcomes.

Total saliva quantity

In total, 5 methods/devices were assessed (i.e., chewing paraffin wax, non-coated Salivette®, polypropylene-coated Salivette®, Salivac®, and Salimetrics® SalivaBio®’s children’s swab) in 2 of the included studies [63, 72]. Salimetrics® SalivaBio®’s children’s swab and Salivac® were both unstimulated techniques with moderate results. Non-coated Salivette® and polypropylene-coated Salivette® were stimulated techniques with conventional results, while chewing Parafilm® wax led to remarkable outcomes.

Saliva pH and Salivary Buffering pH

In total, 4 methods were assessed for this category of tests (i.e., chewing paraffin wax (stimulated), passive drooling (unstimulated), polypropylene-coated Salivette® (stimulated), and non-coated Salivette® (stimulated)) in 2 of the included studies [59, 63]. Whilst chewing paraffin wax had noteworthy outcomes, the other 3 managed to lead to decent yet average laboratory results.

Salivary total protein

Passive drooling (unstimulated), spitting (unstimulated), non-coated Salivette® (stimulated), and chewing on Parafilm® wax (stimulated) all had respectable results [55, 59]. While chewing mint-flavored gum (stimulated) was the only method that resulted into significant outcomes [55, 59].

Salivary DNA quantity/concentration

Nine methods/devices were assessed in total for this variable (i.e., chewing paraffin wax (stimulated), Whatman FTA® cards (unstimulated), DNA-SAL kit (unstimulated), non-coated Salivette® (unstimulated), dry MicroFLOQ® (unstimulated), wet MicroFLOQ® (unstimulated), Oragene® self-collection kit (unstimulated), spitting (unstimulated), and passive drooling (unstimulated)) in 6 of the included studies [54, 56, 57, 60, 64, 66]. Out of these 9 methods/devices, only wet MicroFLOQ® traces led to exceptional outcomes while the rest all had conventional and accepted outcomes.

Salivary DNA quality/purity

All 3 methods/devices assessed in this category (i.e., spitting, passive drooling, and Oragene® self-collection kit) were unstimulated techniques and all had standard outcomes [56, 60, 65, 66].

Salivary cortisol

Four methods/devices were assessed for this category (i.e., smell of freshly-baked bacon (stimulated), Maxissal® (lozenge-form) (stimulated), Salimetrics® (unstimulated), and passive drooling (stimulated)). All 4 of these methods/devices had standard outcomes [50, 51, 68].

Salivary testosterone

Four methods/devices were assessed for this category (i.e., smell of freshly-baked bacon (stimulated), Maxissal® (lozenge-form) (stimulated), citric-acid-coated cotton swab (stimulated), and passive drooling (unstimulated)). All 4 of these methods/devices had standard and conventional outcomes [51, 67].

Sampling time

Only 3 out of the 23 studies had investigated the outcome differences of different sampling times during the day [51, 62, 69]. The presence of oral cancer metabolites was at its peak in samples taken between 7:30 AM and 9:00 AM [69]. Salivary cortisol, testosterone, and DHEA levels were significantly higher in samples taken between 10:30 AM and 11:00 AM [51]. Salivary iodine level was at its peak in samples taken between 14:00 PM and 20:00 PM [62].

Transportation, preparation, and storage conditions

All of the varied preparation and storage conditions were categorized into 6 groups. Figure 3 details all 6 methods’ descriptions (i.e., P + S 1, P + S 2, P + S 3, P + S 4, P + S 5, and P + S 6) and showcases the frequency of assessments for each method (Fig. 3). The “P + S” abbreviation used in tables and figures indicates the preparation and storage (P + S) conditions of samples before further analysis (Fig. 3, Table 4). Centrifuging samples before storing them at -70 °C to -80 °C (P + S 2) was the most assessed method (Fig. 3). Out of the 23 included studies, 5 of them compared the outcome differences of different preparation/storage methods [59, 60, 64, 65, 67]. Table 4 displays the results of all of the comparisons, along with the variables that these methods were assessed for (Table 4).

Fig. 3
figure 3

All 6 different preparation and storage (P + S) conditions, and their generality amongst included studies

Table 4 Evaluative comparison amongst different preparation and storage conditions

Risk of bias assessments

The results of the risk of bias assessments using the JBI Critical Appraisal Tool for risk of bias assessment in cross-sectional studies are showcased in Fig. 4. Out of the 23 included studies, 8 studies had low risks of bias [51, 55, 59, 60, 62, 63, 65, 69], while the rest all had a moderate status in overall risk of bias (Fig. 4).

Fig. 4
figure 4

Risk of bias assessment results using the JBI Critical Appraisal Tool for risk of bias assessment in cross-sectional studies

Discussion

Saliva as a diagnostic bodily fluid has gained tremendous respect and trust from clinicians and scientists in regards to experiments that were only feasible through blood samplings up until couple decades ago [61, 62, 71]. Saliva is collected to analyze the oral and systematic health of patients, and has been conspicuously called “mirror of the body’s health” [73]. Saliva as an exocrine solution, intercommunicates in both intracellular and extracellular manners with the oral cavity, and is a remarkable factor in determining and ascertaining the prevalence of dental caries [74, 75]. Human WMS comprises of numerous proteins, peptides and enzymes of clinical relevance [48]. About 30% of all blood proteins are present in WMS [76]. Saliva sampling compared to blood sampling is less complicated, has a shorter sampling time, is non-invasive, and it significantly reduces costs [77,78,79]. There are numerous saliva sampling techniques along with varied handling, transportation, and storage methods [80, 81]. This systematic review was conducted to gather all of the clinical human descriptive studies that have investigated different collection, transportation, preparation and storage methods and techniques for WMS in different times of the day for various experiments.

Foddai et al. designed and executed a systematic review on the reliability of saliva sampling instead of blood sampling for laboratorial analysis on human autoantibodies [82]. They concluded that even though in many cases saliva sampling can be an appealing alternative to serum-based testing, standardization of the saliva sampling techniques, maintenance and detection methods must be fully investigated and addressed, which only further proves the importance and the necessity of this systematic review.

Sampling time

WMS is commonly collected in the morning in order to have relatively equal contributions from parotid, submandibular and sublingual glands [83]. However, as mentioned before, there are various times of the day that saliva sampling could be performed depending on the type of hormone, mineral, nucleic product, or micro-/nanoparticles that are the main focus of each test [26]. For instance, if the main focus of the tests is to have high concentrations of parotid-secreted proteins (e.g., basic proline-rich proteins (bPRPs)), an early afternoon sampling is highly recommended [41]. Whilst, if scientists are mainly interested in sublingual- and submandibular-secreted proteins (e.g., salivary cystatins (type S)), then an early morning sampling is more appropriate [84, 85].

Out of the 23 included studies, only 3 of them had investigated the outcome differences amongst different sampling times (Table 2). Reported outcomes of Ishikawa et al.’s 2017 study suggest that 7:30 AM – 9:00 AM is the period of time with optimum features regarding the salivary oral cancer metabolites analyzes, while the 9:00 AM – 11:30 AM span had average results [69]. Peres et al. reported that 10:30 AM – 11:00 AM resulted into significantly higher levels of salivary cortisol, testosterone, and DHEA, while 9:00 AM – 10:30 AM showed lower levels [70]. Guo et al. disclosed that the salivary iodine is at its peak from 14:00 PM till 20:00 PM, while the 6:00 AM – 13:30 PM period had average iodine levels [62]. Since only 3 studies have reported comparative outcomes of different sampling times, and each study has focused on a different group of hormones and minerals, their reported outcomes could not be compared with each other. In order to have a comprehensive evaluation of different sampling time points/periods, there must be at least a couple of similar studies in each category of biomarkers, who have investigated the outcome differences of various sampling time points/periods. Unfortunately, that is not the case and it cannot be concluded if these reported outcomes are valid or not.

Sampling methods and devices

Over the past four decades a variety of different stimulating (stimulated) and unstimulating (unstimulated) methods and devices have been introduced for saliva sampling [26, 48, 78, 86]. There are some on-site direct sampling techniques (e.g., SalivaDirect™) that are designed for pandemics (e.g., the COVID-19 pandemic) and other urgent situations that require collecting and analyzing numerous saliva samples from huge populations. However, our main focus in this study was methods and devices that are used by clinicians and researchers on a daily basis and not just in special and urgent occasions. Included studies utilized a total of 22 different methods (Table 3). Passive drooling, spitting, Salivette®, Salimetrics®, and chewing paraffin wax were the most assessed techniques, while the rest of the methods were only assessed in a single study.

Passive drooling is the oldest and most accessible sampling method that has been utilized as the main sampling technique for the past decades [26]. Passive drooling (n = 9) was the most utilized technique for saliva sampling in the included studies (Table 3). When assessed for salivary flow rate, pH, buffering pH, total protein, DNA quantity, DNA quality, cortisol, and testosterone, passive drooling did not show any remarkable results and was average compared to other unstimulated and stimulated methods. There was not a single category of tests where passive drooling caused significant outcomes. Even though passive drooling is still the most utilized sampling method in the literature, results suggest that stimulating techniques on general do a much better job. In a review of literature executed by Almukainzi et al. in 2022, it was suggested that passive drooling is a reliable substitute with significant amounts of accumulated WMS [87]. Even though passive drooling may not have the most desirable laboratorial outcomes compared to some stimulated sampling techniques (e.g., non-coated Salivette®), it still leads to promising results in cases where stimulated sampling techniques/devices such as Salivette® are not available.

Spitting is next to passive drooling as the most assessed method in saliva sampling in the past decades [42, 45, 88]. Spitting was utilized in a total of 3 studies [55, 57, 66] for 4 categories of outcomes (i.e., salivary flow rate, total protein, DNA quantity, and DNA quality), which led to average outcomes in all 4 of them (Table 3). Patients were asked to chew paraffin wax to stimulate saliva in 2 studies [63, 64]. Chewing paraffin wax led to significantly better results than other methods when samples were analyzed for total salivary quantity and salivary pH and buffering pH. However, chewing paraffin wax resulted in average results for DNA quantity analysis [63, 64].

Salivette® is a cylindrical cotton roll that has been assessed in both stimulated and unstimulated samplings [63, 76, 89,90,91,92]. Salivette® was the most assessed device (n = 4) amongst the included studies (Table 3). Salivette® was assessed in 3 different forms: non-coated, polypropylene-coated, and citric-acid-coated [55, 57, 61, 63]. Salivette® non-coated resulted in significantly better outcomes compared to other methods. Whilst Salivette® non-coated was only average for saliva total quantity, pH, buffering pH, total protein, and DNA quantity. Salivette® citric-acid-coated was only assessed for the analysis of salivary flow rate, and resulted into significantly better outcomes than other methods (Table 3). Salivette® polypropylene-coated was only utilized for the testing of total saliva quantity and only had average results. Salimetrics® was used in 2 studies and for 2 purposes only: saliva quantity and cortisol [68, 72] (Table 3). Salimetrics® led to average outcomes in both categories of experiments.

Overall, since the number of studies that each method was utilized for, and the categories that they were used for are vastly different and varied, a true evaluative comparison is not feasible with the current published studies.

In 2018, MacLean et al. conducted an in vivo study on the outcome differences of Salivette®, SalivaBio® Children’s swab, citric acid and passive drooling as sampling techniques for analyzing salivary oxytocin in domestic dogs [93]. They reported that SalivaBio® outperformed Salivett®, but they both had significantly better outcomes and yielded remarkably higher concentrations of oxytocin compared to passive drooling [93]. Stimulating the secretion of saliva through the taste of citric acid was also a successful method in their in vivo study [93]. Unfortunately, to the reviewer’s knowledge there is not a single descriptive human study that has tested these 4 methods in comparison with each other. However, the reported outcomes of MacLean et al. are still complied and in favor with our results that stimulating sampling techniques lead to remarkably better laboratorial outcomes.

Handling, transportation and storage

Even though varied handling, transportation, and storage methods and techniques have been experimented in saliva sampling studies, there are still no guidelines indicating the methods with optimum outcomes [94]. All of the transportation and storage procedures assessed in the included studies of this review were categorized into 6 groups (Fig. 3 and Table 4). Reported outcomes show that centrifuging samples and storing them at -70°C to -80°C (T2) was the most assessed method (38%) (Fig. 3). Centrifuging samples and storing them at -20°C (T1) (22%), and immediately analyzing samples without centrifuging or storage (T3) (22%), were at second place in terms of assessment and utilization (Fig. 3). Storing samples at 4°C without centrifuging (T4) (6%), storing at 37°C without centrifuging (T5) (6%), and analyzing immediately after centrifuging without storage (T6) (6%), were the rest of the experimented methods (Fig. 3). A proper and evaluative comparison of all 6 of these methods, would have been feasible if all of these methods were assessed all together in a couple of single studies. However, 5 of the included studies in this review have compared some of these methods against each other [59, 60, 64, 65, 67] (Table 4). Since the compared methods, their category of utilization and their outcomes are notably varied and different, a conclusion cannot be drawn out (Table 4 and Fig. 3).

Out of the 23 included studies, only 3 of them had investigated the outcome differences of varied sampling times. And those 3 studies had experimented 3 completely different categories of salivary biomarkers. In order to have a clear conclusion on to which periods of time have the optimum capabilities for each category of salivary biomarkers, hormones, nucleic products, and minerals, a decent number of descriptive clinical human studies must be executed in the future so that their results can properly be evaluatively compared.

Only 5 of the experimented methods and devices were assessed in more than 1 study. Hence, the results of the remaining 17 methods and devices cannot be properly evaluated amongst different studies. Only 5 of the included studies had investigated the outcome differences of different sample transportation, handling, and storage techniques.

As mentioned before, one of the main challenges in the execution of this systematic review, was the lack of previously-published similar studies. Additionally, most descriptive human studies did not have their main focus on the outcome differences of different saliva sampling techniques. In general, most of the tested and investigated saliva sampling, transportation, and storage techniques and methods are relatively newly introduced to the field. Therefore, for valuable and reliable comparisons of their results, these 23 studies are simply not enough and there is a clear and urgent need for clinicians and scientists to utilize these varied methods and report their outcomes. Ideally, scientists can design and execute descriptive clinical human studies by utilizing multiple sampling, transportation, and storage techniques and methods, in order to compare their outcome differences. Doing so, a lot of the unanswered questions regarding the best saliva sampling, transportation, and storage methods and devices, can hopefully be answered. Scientists and clinicians can also investigate the outcome differences of various sampling times of the day, for each category of salivary biomarkers (e.g., minerals, hormones, nucleic acid products, glucose, etc.), different viruses, and bacteria.

Conclusion

Passive drooling, non-coated Salivette® and spitting were the most utilized salivary collection methods/devices amongst the included studies. Non-coated Salivette®, citric-acid-coated Salivette®, and chewing paraffin wax, were the sampling methods with the most desirable outcomes in salivary flow rate, saliva total quantity, salivary pH and buffering pH, and salivary total protein. Sampling times with optimum capabilities for cortisol, iodine, and oral cancer metabolites are suggested to be 7:30 AM to 9:00 AM, 10:30 AM to 11:00 AM, and 14:00 PM to 20:00 PM, respectively. For DNA quantity and quality, analyzing samples immediately after collection without centrifuging or storage, outperformed centrifuging samples and storing them at -70 °C to -80 °C. Using non-coated Salivette® led to exceptional laboratorial outcomes for analyzing salivary flow rate. However, it is highly suggested that authors take aid from the categorized outcomes of descriptive studies reported in this systematic review and design their study questions based on the current voids for each method and device.